Bodo Bauer, Michael Burghart, Roland Dyroff, Karl Eichwalder, Werner Fink, Klaus Franken, Ju"rgen Geck, Carsten Ho"ger, Dirk Hohndel, Richard Jelinek, Florian La Roche, Volker Lendecke, Hans Lermen, Hubert Mantel, Martin Scherbaum, Burchard Steinbild, Klaus Wagner, S.u.S.E. Linux 5.2 Installation, Configuration and First Steps Your registration ID for technical support: Support contact see: page 9 S.u.S.E., LLC 458 Santa Clara Avenue Oakland, CA 94610 U.S.A. Toll free phone number within the US and Canada: 1-8888-UR-LINUX (1-888-875-7873) Phone.: +1-510-TEL-SUSE (+1-510-835-7873) Fax.: +1-510-835-7875 email: info@suse.com orders@suse.com support@suse.com webmaster@suse.com WWW: http://www.suse.com Europe: S.u.S.E. GmbH Gebhardtstr. 2 D-90762 F_rth Germany Tel.: +49-911-740 53 31 Fax.: +49-911-741 77 55 E-Mail: suse@suse.de WWW: http://www.suse.de Bodo Bauer, Michael Burghart, Roland Dyroff, Karl Eichwalder, Werner Fink, Klaus Franken, Ju"rgen Geck, Carsten Ho"ger, Dirk Hohndel, Richard Jelinek, Florian La Roche, Volker Lendecke, Hans Lermen, Hubert Mantel, Martin Scherbaum, Burchard Steinbild, Klaus Wagner, Installation, Configuration and first steps with S.u.S.E. Linux 5.2 10th edition 1998 S.u.S.E. GmbH Copyright This work is copyrighted by S.u.S.E. GmbH and is placed under conditions of the GNU General Public License. You may copy it in whole or in part as long as the copies retain this copyright statement. Layout: ALTE X 2" Cover layout: Sabine Zieseniss - using a picture by Stephan Endrass. Printing: Graphische Betriebe F. Willmy GmbH, Nu"rnberg. MS-DOS , Windows , Windows 95 and Windows NT are registered trademarks of Microsoft Corporation. XFree86TM is a registered trademark of The XFree86 Project, Inc. FlagShip is a trademark multisoft Datentechnik GmbH. UNIX is a registered trademark of X/Open Company Limited. Other trademarks and registered trademarks are: Clipper of Computer Associates, dBASE of Borland, Foxbase of Microsoft, Compuserve, T-Online of Deutsche Telekom, S.u.S.E. Linux und YaST von S.u.S.E. GmbH. Other products mentioned in this manual might be trademarks of the respective manufacturer. Contents 1 Introduction 1 1.1 What is "Linux"? . . . . . . . 2 1.2 Purpose of this book . . . . . . . 2 1.3 Legend_or what does it mean: "root@earth: # ls" 3 1.4 The documentation . . . . . . . 4 1.4.1 Hypertext help . . . . . . 4 1.4.2 Texinfo . . . . . . . . 4 1.4.3 Unix Manuals . . . . . . . 4 1.4.4 FAQ, HowTo, and README files . . . 5 1.4.5 Free Books . . . . . . . 5 1.5 Laurels . . . . . . . . . 5 2 Support and services of S.u.S.E. 9 2.1 Installation support . . . . . . . 9 2.1.1 Registration . . . . . . . 9 2.1.2 Scope of our installation support . . . 10 2.1.3 What we need to know for helping you . . 12 2.1.4 How to contact the S.u.S.E. support team . . 15 2.2 Further services . . . . . . . . 15 3 Installation 17 3.1 Don't panic! You can do it! . . . . . . 17 3.2 Linux in 30 minutes - the quick install . . . 17 3.3 This installation manual . . . . . . 18 3.3.1 Finding your way through this installation guide: 19 3.4 Basic principles of Linux installation . . . . 19 3.5 S1: Partitioning . . . . . . . . 21 3.6 E1: Can I use the S.u.S.E.- boot-disk? . . . . 24 3.7 E2: Does Linux support my CD-ROM? . . . 24 * * i Contents _________________________________________________________________________________________________________________* *_______________________________________@ 3.8 S2: Copying packages onto the HD in the case CD-ROM drive not supported . . . . . . . 25 3.9 S3: Bringing Windows 95 into DOS mode . . . 27 3.10 S4: Invoking setup and first steps with setup . . 27 3.11 E3: Which way do I want to boot base-Linux? . . 28 3.12 S5: Creating boot-disks with setup . . . . 29 3.13 S6: Installing loadlin and starting your base-Linux . 30 3.14 S7: Creating boot disks with Unix . . . . 32 3.15 S8: Starting linuxrc . . . . . . . * * 32 3.16 S9: Invoke YaST . . . . . . . . 34 3.17 S10: Designing partitions . . . . . . 35 3.18 S11: Assigning filesystems for partitions . . . 37 3.19 S12: Choosing base selection of software . . . 39 3.20 S13: Quitting YaST and rebooting . . . . 41 3.21 S14: Starting YaST again and installing software . . 43 3.22 S15: Starting YaST for the last time and finishing up instal- lation . . . . . . . . . . * * 44 3.23 E4: How do you plan to start your Linux in future? . 46 3.24 S16: Starting via loadlin . . . . . . 46 3.25 After your system is installed . . . . . 52 3.26 Problem description . . . . . . . 53 3.26.1 P1: Files cannot be moved . . . . 53 3.26.2 P2: No English keymaps in DOS-mode . . 53 3.26.3 P3: No CD-ROM driver in DOS-mode . . 53 3.26.4 P4: CD is damaged . . . . . . 53 3.26.5 P5: ATAPI-CD-ROM hangs while reading . 53 3.26.6 P6: Mitsumi driver announces: mcd: init failed 54 3.26.7 P7: "Unable to mount root" using Mitsumi CD- ROM-drives (Not ATAPI-drives) . . . 54 3.26.8 P8: Loadlin doesn't have enough memory to load the kernel . . . . . . . * * 55 3.26.9 P9: Loadlin doesn't work . . . . 55 3.26.10 P10: Error with mke2fs . . . . . 55 3.26.11 P11: DOS runs in protected mode . . . 55 3.26.12 P12: The 3,5 floppy drive is connected as B: and not bootable . . . . . . . . * * 55 3.26.13 P13: Label of CD-ROM drive has changed . 56 3.27 Infoblocks . . . . . . . . . * * 56 3.27.1 IB1: Selecting a kernel . . . . . 56 3.27.2 IB2: Kernel parameters . . . . . 57 ii ______________________________________________________________________________________________________________________Conten* *ts____________________ 3.27.3 IB3: Does Linux support my CD-ROM? . . 57 3.27.4 IB4: Partition types on PC . . . . 58 3.27.5 IB5: Partitions for the Linux system . . 58 3.27.6 IB6: The live file system . . . . . 59 3.27.7 IB7: Demo mode . . . . . . 59 3.28 Partitioning for novices . . . . . . 60 3.29 Partitioning for experts . . . . . . 61 3.29.1 Size of swap partition . . . . . 61 3.29.2 Appointment of the machine as stand-alone machine 62 3.29.3 Optimizations . . . . . . . 63 4 YaST - Yet another Setup Tool 67 4.1 Control and keyboard mapping . . . . . 67 4.2 Installation media . . . . . . . 67 4.3 Installation from CD-ROM . . . . . . 67 4.4 Installation via HD partition . . . . . 69 4.5 Installation via NFS . . . . . . . 70 4.5.1 Installation via reachable directory . . . 71 4.5.2 Select the installation mode . . . . 71 4.5.3 The main menu . . . . . . 72 4.5.4 Settings . . . . . . . . 72 4.5.5 Partitioning your HD . . . . . 73 4.6 Assigning file systems . . . . . . . 75 4.6.1 Mountpoint . . . . . . . 76 4.6.2 Inodes density . . . . . . . 76 4.6.3 Formatting . . . . . . . 77 4.6.4 Reading fstab . . . . . . 77 4.6.5 Size of installation . . . . . . 77 4.6.6 Loading an existing configuration . . . 78 4.6.7 Saving your configuration . . . . 78 4.6.8 Changing your configuration . . . . 78 4.6.9 What if.. . . . . . . . . 80 4.6.10 Installation . . . . . . . 80 4.6.11 Deleting packages . . . . . . 80 4.6.12 Search packages . . . . . . 81 4.6.13 Install packages . . . . . . 81 4.7 Administration . . . . . . . . 82 4.7.1 Integrating hardware into your system . . 82 4.7.2 Kernel and boot configuration . . . . 84 4.7.3 Configuring LILO . . . . . . 84 * * iii Contents _________________________________________________________________________________________________________________* *_______________________________________@ 4.7.4 Configuring your network . . . . 85 4.7.5 Integrating / Releasing CD-Live-File-System . 85 4.7.6 User management . . . . . . 86 4.7.7 Changing configuration file . . . . 87 4.7.8 Creating backups . . . . . . 87 5 Updating the system and package management 91 5.1 Updating S.u.S.E. Linux . . . . . . 91 5.1.1 Updating base system . . . . . 91 5.1.2 Update configuration . . . . . 92 5.1.3 Updating of single packages . . . . 92 5.2 From version to version . . . . . . 92 5.2.1 From an ealier version to 4.x . . . . 92 5.2.2 From 4.x to 5.0 . . . . . . 92 5.2.3 From 5.0 to 5.1 . . . . . . 93 5.3 RPM - The package manager . . . . . 93 5.3.1 Managing packages: install, update and de-install 93 5.3.2 RPM queries . . . . . . . 94 5.3.3 Tools for RPM archives . . . . . 96 6 Notebooks and PCMCIA cards 99 6.1 The concept . . . . . . . . 99 6.2 Installation . . . . . . . . . * *100 7 Kernel parameters 101 7.1 Drivers in the kernel . . . . . . . 101 7.2 Some hints . . . . . . . . . 101 7.3 The parameters . . . . . . . . 102 7.3.1 Notation and meaning . . . . . 102 7.3.2 LILO-parameters . . . . . . 102 7.3.3 insmod -Parameter . . . . . . 111 8 LILO - what it does, how it works 121 8.1 Booting a PC . . . . . . . . 121 8.1.1 Master Boot Record . . . . . 122 8.1.2 Boot sectors . . . . . . . 122 8.1.3 Booting DOS or Windows 95 . . . . 122 8.1.4 LILO -Here you go... . . . . . 122 8.2 Different boot concepts . . . . . . 123 8.2.1 Booting another OS from a floppy disk . . 123 8.2.2 Boot chaining of additional systems . . 123 iv ______________________________________________________________________________________________________________________Conten* *ts____________________ 8.2.3 Installing a bootmanager . . . . . 123 8.3 LILO custom fit . . . . . . . . 124 8.3.1 What is LILO? . . . . . . 124 8.3.2 How to boot with LILO? . . . . . 124 8.3.3 The components of LILO . . . . 124 8.3.4 Where LILO can be installed . . . . 125 8.3.5 Configuring LILO . . . . . . 126 8.4 Installation and deinstallation of LILO . . . . 130 8.4.1 Installation after changing the configuration . 130 8.4.2 Installation recompiling a kernel . . . 131 8.4.3 Removing LILO . . . . . . 131 8.5 Creating a Linux-bootdisk . . . . . . 132 8.5.1 Boot-disk without LILO . . . . . 132 8.5.2 Boot-disk with LILO . . . . . 133 8.6 Sample configurations . . . . . . . 133 8.6.1 DOS/Windows 95 and Linux . . . . 133 8.6.2 Windows NT and Linux on one hard disk . 134 8.6.3 OS/2 and Linux . . . . . . 135 8.6.4 DOS, OS/2 and Linux . . . . . 136 8.7 LILO problems . . . . . . . . 136 8.7.1 Some guidelines . . . . . . 136 8.7.2 Diagnosis of errors: LILO start-messages . 137 8.7.3 The 1024-cylinder limit . . . . . 138 8.7.4 LILO and DOS-diskmanager (OnTrack-Manager, EZ- Drive) . . . . . . . . 140 8.7.5 Special boot-problems with kernel from 2.0 onwards 141 9 Printers 143 9.1 Overview: Interfaces, Queues (Spooling) . . . 143 9.1.1 The parallel ports . . . . . . 143 9.1.2 Spooling mode, Printer queues . . . 144 9.2 Printer queues: running and Configuring . . . 144 9.2.1 lpr: Please line up! . . . . . . 144 9.2.2 Additional control tools for the user: Overview 145 9.2.3 lpd: the print manager in background . . 145 9.2.4 Filter: the workaholics . . . . . 146 9.2.5 /etc/printcap: Configuration of the queue . . 146 9.3 Printer filter. The apsfilter . . . . . . 147 9.3.1 What is a filter and how does it work? . . 147 9.3.2 apsfilter . . . . . . . . 147 * * v Contents _________________________________________________________________________________________________________________* *_______________________________________@ 9.4 Some words on Ghostscript . . . . . . 150 9.5 List of supported printers . . . . . . 150 9.6 Printer checklist: apsfilter . . . . . . 153 10 Linux-machines and hardware 155 10.1 Preliminary notes . . . . . . . 155 10.2 ISA "Plug and Play" hardware . . . . . 155 10.2.1 General . . . . . . . . * *155 10.2.2 Configuration . . . . . . . 155 10.3 Changeable media . . . . . . . 155 10.3.1 General . . . . . . . . * *155 10.3.2 Disk drives . . . . . . . 1* *56 10.3.3 LS 120 drives . . . . . . . 156 10.3.4 ZIP drives . . . . . . . 1* *56 10.3.5 SCSI changeable media drives . . . 156 10.4 Scanner . . . . . . . . . * *156 10.4.1 General . . . . . . . . * *156 10.4.2 Hardware: Flatbedscanner, Cameras, automatic doc- ument feed : : : . . . . . . 157 10.4.3 Software . . . . . . . . * *157 10.4.4 Problems . . . . . . . 1* *58 11 Special features of S.u.S.E. Linux 159 11.1 Keyboard layout . . . . . . . . 159 11.2 Differences in the structure of the file system . . 159 11.3 linuxrc . . . . . . . . . * * 160 11.3.1 Main menu . . . . . . . 160 11.3.2 System information . . . . . . 161 11.3.3 Loading of modules . . . . . 162 11.3.4 Passing parameters . . . . . . 163 11.3.5 System / Start installation . . . . 164 11.4 The S.u.S.E. rescue system . . . . . . 164 11.4.1 Overview . . . . . . . 164 11.4.2 Launching the rescue system . . . . 166 11.4.3 Working with the rescue system . . . 166 11.5 init . . . . . . . . . . * * 167 11.6 Changes made to software packages . . . . 168 11.6.1 package curses . . . . . . 168 11.6.2 manpages . . . . . . . 168 11.6.3 sendmail . . . . . . . * * 168 11.7 Misc . . . . . . . . . . * * 168 vi ______________________________________________________________________________________________________________________Conten* *ts____________________ 12 The S.u.S.E. boot concept 169 12.1 The init program . . . . . . . . 169 12.2 Runlevels . . . . . . . . . 170 12.3 Changing of runlevels . . . . . . . 171 12.4 Init scripts . . . . . . . . . 171 12.5 /etc/rc.config and /sbin/SuSEconfig . . . 173 12.6 Options of configuration . . . . . . 174 13 The kernel 181 13.1 Kernel sources . . . . . . . . 181 13.2 Kernel modules . . . . . . . . 181 13.2.1 Handling of modules . . . . . 182 13.2.2 The kernel daemon . . . . . . 182 13.3 Kernel configuration . . . . . . . 183 13.3.1 Configuring on the command line . . . 183 13.3.2 Configuring in text mode . . . . . 183 13.3.3 Configuring under the X Window System . 183 13.4 Settings in kernel configuration . . . . . 183 13.4.1 SMP . . . . . . . . 183 13.4.2 Experimental drivers . . . . . 184 13.4.3 Modul support . . . . . . 184 13.4.4 General configuration . . . . . 184 13.4.5 Disks, (E)IDE and other block devices . . 187 13.4.6 Network options . . . . . . 190 13.4.7 SCSI . . . . . . . . 192 13.4.8 Networking cards . . . . . . 193 13.4.9 ISDN subsystem . . . . . . 195 13.4.10 Proprietary CD-ROM drives . . . . 196 13.4.11 File systems . . . . . . . 198 13.4.12 Character devices . . . . . . 201 13.5 Compiling the kernel . . . . . . . 204 13.6 Install kernel . . . . . . . . 204 13.7 Creating a boot disk . . . . . . . 205 13.8 Cleaning your disk after compilation . . . . 205 14 The X Window System 207 14.1 Configuration using XF86Setup . . . . . 208 14.1.1 Mouse . . . . . . . . 208 14.1.2 Keyboard . . . . . . . 209 14.1.3 Graphics device . . . . . . 209 * * vii Contents _________________________________________________________________________________________________________________* *_______________________________________@ 14.1.4 Monitors . . . . . . . . 2* *10 14.2 Configuration using xf86config . . . . . 210 14.2.1 Mouse . . . . . . . . * *211 14.2.2 Keyboard . . . . . . . 211 14.2.3 Monitor . . . . . . . . * *212 14.2.4 Graphics cards / X-Server . . . . 213 14.2.5 Saving your configuration . . . . 217 14.3 Optimizing the X Window System . . . . 217 14.3.1 Screen section . . . . . . . 218 14.3.2 Device section . . . . . . 220 14.3.3 Monitor section . . . . . . 220 15 The windowmanager - Window to your machine 223 15.1 Some theory . . . . . . . . 223 15.1.1 General . . . . . . . . * *223 15.1.2 What does a windowmanager manage? . . 225 15.1.3 Starting different Windowmanagers . . . 226 15.2 The fvwm2 window manager . . . . . 227 15.2.1 General . . . . . . . . * *227 15.3 Fvwm2 settings . . . . . . . . 229 15.3.1 General . . . . . . . . * *229 15.4 KDE - the K Desktop Environment . . . . 233 15.4.1 General . . . . . . . . * *233 15.4.2 kdm - graphical login . . . . . 234 15.4.3 So, what's so special with KDE? . . . 235 15.5 Configuring the windowmanager using susewm . . 236 15.6 General . . . . . . . . . * *241 15.6.1 X11 defaults for applications . . . . 242 16 Networking Linux 245 16.1 Preliminary words : : : . . . . . . 245 16.1.1 Good neighbourhood - connecting to a LAN . 245 16.1.2 Requirements, preliminary work . . . 245 16.2 Configuring using YaST . . . . . . 247 16.3 Manual network configuration - where do I find what? 248 16.3.1 Configuration files . . . . . . 248 16.3.2 Startup scripts . . . . . . . 2* *51 16.4 Routing under S.u.S.E. Linux . . . . . 252 16.4.1 Preliminaries . . . . . . . 252 16.4.2 How to commence and use it . . . . 252 viii ______________________________________________________________________________________________________________________Conten* *ts____________________ 16.5 NIS, yellow pages in a LAN . . . . . 253 16.5.1 NIS, what it is . . . . . . . 253 16.5.2 Installing a NIS-client . . . . . 253 16.5.3 NIS master and slave server . . . . 254 16.6 NFS - distributed filesystems . . . . . 255 16.6.1 Importing filesystems . . . . . 255 16.6.2 Exporting filesystems . . . . . 255 16.7 Firewall and Masquerading - Friends and "Friends" on the Internet . . . . . . . . . 256 16.7.1 What is a firewall? . . . . . . 257 16.7.2 Requirements for installing firewall and masquerad- ing . . . . . . . . . 257 16.7.3 How a firewall works . . . . . 258 16.7.4 Example firewall constellation . . . . 262 16.7.5 Principle of masquerading . . . . 264 16.7.6 Example for masquerading . . . . 266 17 Connecting to the world: UUCP, SLIP, and PPP 267 17.1 Connecting a modem . . . . . . . 267 17.1.1 minicom . . . . . . . . 267 17.2 PPP . . . . . . . . . . 268 17.2.1 Requirements for using PPP . . . . 268 17.2.2 Customizing PPP . . . . . . 268 17.2.3 Configuration of an PPP server . . . 271 17.2.4 Further information on PPP . . . . 272 17.3 SLIP . . . . . . . . . . 272 17.3.1 Establishing connections with dip . . . 273 17.3.2 Configuring a SLIP-server . . . . 275 17.4 UUCP . . . . . . . . . 277 17.4.1 Configuring Taylor-UUCP . . . . 278 17.4.2 Testing the configuration . . . . . 284 17.5 ISDN Configuration . . . . . . . 284 17.5.1 Overview . . . . . . . 285 17.5.2 Configure ISDN hardware . . . . 285 17.5.3 Testing ISDN on our S.u.S.E. host . . . 290 17.5.4 Setup configuration for your ISP . . . 293 17.5.5 Reference of the S.u.S.E. ISDN account . . 294 17.6 Let's write - configuration of email . . . . 295 17.7 Brandnew messages - C News . . . . . 297 17.7.1 Broadcasting the news . . . . . 297 * * ix Contents _________________________________________________________________________________________________________________* *_______________________________________@ 17.7.2 Installing a local news-system . . . . 298 17.7.3 Overview of configuration files . . . 299 17.7.4 Reading news . . . . . . . 300 17.8 Faxing with Linux . . . . . . . 300 17.8.1 SuSEFax - an HylaFAX fax client . . . 301 17.8.2 Automatic generation of the fax cover . . 308 17.8.3 Fax spooling on UNIX / Linux . . . 309 17.8.4 HylaFAX - distributed faxes . . . . 310 18 Samba PC Server 317 18.1 Introduction . . . . . . . . 317 18.1.1 NetBIOS . . . . . . . 317 18.1.2 SMB . . . . . . . . * *318 18.1.3 Clients . . . . . . . . * * 318 18.1.4 Permissions . . . . . . . 319 18.2 Installation of the server . . . . . . 319 18.2.1 smb.conf . . . . . . . . 3* *20 18.3 Installation of clients . . . . . . . 321 18.3.1 DOS and Windows 3.1 . . . . . 321 18.3.2 Windows for Workgroups . . . . 322 18.3.3 Windows 95 . . . . . . . 323 19 First steps with Linux 325 19.1 Logging in, "root", adding users . . . . . 325 19.2 Shutting down and booting . . . . . . 326 19.3 Virtual consoles . . . . . . . . 326 19.4 Adding and deleting users . . . . . . 327 19.5 Directories and filenames . . . . . . 327 19.6 Working with directories . . . . . . 328 19.7 Working with files . . . . . . . 328 19.7.1 Information on files . . . . . . 328 19.7.2 Wildcards . . . . . . . 3* *29 19.7.3 Contents of files . . . . . . 330 19.7.4 Hidden files . . . . . . . 3* *30 19.7.5 Copying, renaming and deleting of files . . 330 19.7.6 Searching and grepping files . . . . 331 19.7.7 Symbolic links . . . . . . 331 19.7.8 Archiving data and saving . . . . 332 19.8 Permissions . . . . . . . . 332 19.8.1 Changing permissions . . . . . 333 x ______________________________________________________________________________________________________________________Conten* *ts____________________ 19.9 Manual pages . . . . . . . . 333 19.10 System information . . . . . . . 335 19.10.1 The df command . . . . . . 335 19.10.2 The free command . . . . . 335 19.10.3 The w command . . . . . . 335 19.10.4 The du command . . . . . . 335 19.10.5 The kill command . . . . . 336 19.10.6 The ps command . . . . . . 336 19.10.7 The pstree command . . . . . 336 19.10.8 The top command . . . . . . 336 19.11 Filesystem types under Linux . . . . . 336 19.12 Mount and unmount filesystems . . . . . 338 19.12.1 umount . . . . . . . . 338 19.13 The mtools . . . . . . . . . 339 19.14 Overview . . . . . . . . . 340 19.15 And now? . . . . . . . . . 341 19.16 Literature recommendations . . . . . 341 20 Emulators 343 20.1 The DOS-emulator . . . . . . . 343 20.1.1 Introduction . . . . . . . 343 20.1.2 Installation of binary package . . . . 344 20.1.3 Compiling . . . . . . . 344 20.1.4 Configuration . . . . . . . 344 20.1.5 Booting DOS under Linux . . . . 345 20.1.6 Booting from a virtual drive . . . . 348 20.1.7 Virtual disks . . . . . . . 349 20.1.8 User management and security precautions . 349 20.1.9 Further configuration options . . . . 350 20.1.10 DEXE, directly executable DOS applications . 351 20.1.11 DOS-emulator and Windows 95 . . . 351 20.1.12 STonX - An ATARI ST emulator . . . 352 20.1.13 UAE - Unix Amiga Emulator . . . . 354 20.2 Z80 emulators . . . . . . . . 356 20.2.1 Z81 - Sinclair ZX81 emulator . . . . 356 20.2.2 XZX - ZX Spectrum Emulator . . . 357 20.2.3 Adamem - Coleco Vision/ADAM Emulator . 358 20.3 CBM-8Bit-Emulation - Vice . . . . . 359 20.3.1 Vice-installation . . . . . . 359 20.3.2 C64-emulation . . . . . . 360 * * xi Contents _________________________________________________________________________________________________________________* *_______________________________________@ 20.3.3 C128-emulation . . . . . . 360 20.3.4 VIC-20-emulation . . . . . . 361 20.3.5 PET-emulation . . . . . . 361 20.4 Emulators for game consoles . . . . . 361 20.4.1 Stella - ATARI VCS 2600 emulator . . . 361 20.4.2 VGB - A virtual gameboy . . . . 362 20.4.3 iNES - Nintendo Entertainment System Emulator 363 20.4.4 Where to get ROMs? . . . . . 363 20.5 ei - Installation of emulator-firmware . . . . 364 20.5.1 Invoking ei . . . . . . . * *364 21 The vi editor 367 22 emacs - more than just an editor 369 22.1 Basics . . . . . . . . . . * * 369 22.1.1 Invoking and handling . . . . . 369 22.1.2 Entering commands . . . . . 370 22.2 Basic commands . . . . . . . . 370 22.2.1 On-line help . . . . . . . 370 22.2.2 Cancel and quit . . . . . . 370 22.2.3 Undo . . . . . . . . * * 370 22.2.4 Cursor control, kill and yank . . . . 370 22.3 Loading and saving texts . . . . . . 371 22.3.1 Backup-files . . . . . . . 371 22.4 Buffer and windows . . . . . . . 371 22.5 Search and replace . . . . . . . 371 22.6 Keyboard macros . . . . . . . 372 22.6.1 Executing and saving macros . . . . 372 22.7 Info . . . . . . . . . . * * 373 22.7.1 More emacs . . . . . . . * * 373 22.8 Literature . . . . . . . . . * *373 22.9 Command overview (selection) . . . . . 373 23 WhyALTE X? 377 23.1 Maintenance of TEX . . . . . . . 377 23.2 Creating a document . . . . . . . 378 23.3 Compiling the document . . . . . . 380 23.4 Viewing the document . . . . . . . 380 23.5 Printing the document . . . . . . . 380 A Important keys 383 xii ______________________________________________________________________________________________________________________Conten* *ts____________________ B Glossary 385 C The directory tree 397 C.1 Overview . . . . . . . . . 397 C.2 Important directories . . . . . . . 397 D Important files 399 D.1 Device files in the /dev directory . . . . . 399 D.1.1 CDROM drives . . . . . . 399 D.1.2 Tapes . . . . . . . . 400 D.1.3 Mice (bus and PS/2) . . . . . 400 D.1.4 Modem . . . . . . . . 400 D.1.5 Serial interfaces . . . . . . 401 D.1.6 Parallel ports . . . . . . . 401 D.1.7 Special devices . . . . . . 401 D.2 Configuration files in /etc . . . . . . 401 D.3 Hidden configuration files in home . . . . 402 E Manual page of e2fsck 405 F The GNU General Public License 409 * * xiii Contents _________________________________________________________________________________________________________________* *_______________________________________@ xiv Chapter 1 Introduction Since Linus Torvalds started working on a UNIX-like operating system for Intel-based PCs lots has been done. Linux has evolved from a "hacker's toy" to a fully operational operating system whose power and efficency can compete with other operating systems (that became slow and heavy in the meantime). Our joy concerning this fact does not only derive from mercantilism but from its inherited anarchic creativity. Even after long terms of development this idealism does still exist and is increasing. The more Linux tries to get hold of places where other operating system dominate the better the development succeeds. Up to now it is a serious alternative to these existing systems. For many years, owners or users of high-end workstations were the only Unix TM users. This might be the reason for the fact that there was an air of mystery about Unix. Companies would offer training courses for their employees after changing to a Unix operating system (despite the fact that several feet of user manuals would usually come with the system). Now, Linux offers the PC user the chance to learn a multi-user-multi-tasking oper- ating system at a reasonable price1 . An increasing number of users will enter the Unix world through Linux. There are many arguments in favour of Linux. First, a complete system is available for less than $50. When used commercially, an arbitrary number of machines can be run under Linux because there are no license fees. Fur- thermore there are many PC users who are turning to "fascination Linux" nowadays and want to adjust their amount of work individually instead of having to visit a three days crash course on Linux/Unix. There are a lot of arguments in favour of Linux' commercial side. Provided by Linux' state of being free one can configure quite some machines with a powerful system instead of paying sums on end for proprietary Unix systems. Linux is an ideal development platform because of source code compatibility with most Unix systems and the availability of the standard OSF Motif user interface. Apart from that, there is no cheaper alternative for an X-Terminal; because even a cheap, old 386 machine is sufficient for this task. The strongest argument, however, is the availability of the complete source code. Apart from the fact that you can modify the system to suit your needs, you can find and correct bugs even if they are in the operating system itself. 1 ____________________________________________________ The first PC Unix system was BSD-Unix ; it was born in the eighties * * 1 1. Introduction _________________________________________________________________________________________________________________* *_______________________________________@ Instead of saying "we have to live with that" (as is the case with other popular operating systems), you can say "we have to change that". This is what makes the development of Linux so fast and leads to surprisingly robust versions. Linux is also very cooperative; it is no problem to install Linux together with other systems on your machine, and there are even ways to exchange data between the different operating systems. 1.1 What is "Linux"? By now there are many different Linux distributions and versions. Therefore we would like to explain a few of the terms used here. If we are talking about "Linux", we have to define what exactly we mean by that phrase. The "real Linux" is the kernel, the "heart" of every Unix operating system. But a kernel alone does not make an operating system. Especially for Unix, there are several free software packages, and almost all known Unix tools are available under Linux. In most cases, these are the GNU versions of the well-known programs, most of which offer additional features (the GNU-C/C++ compiler , which some people think is one of the better com- pilers currently available, is probably the most famous program, perhaps with the exception of Emacs). What makes Linux complete is XFree86 TM (current version 3.3.2). This is an X Window System (Release 6.3) for PC-based Unix systems. This port is part of the official X11 R6.3 distribution of the X Consortium, Inc., and therefore fully compatible with that standard. All these components, together with many tools and goodies (e. g. , games), make up the system known as Linux. As mentioned earlier, there is a large amount of free software available, and there are many ways to put a Linux system together. This is why there are so many different Linux-distributions on the market. At last selection of a distribution is a question of philosophy of life which divides the Linux community into several ideologies. 1.2 Purpose of this book The purpose of this book is to help you install Linux on your machine. It is not and cannot be a replacement for existing literature on Unix and the world of high performance computing. It is difficult to write manuals for Linux. Because of the highly dynamic development of Linux (a new kernel patch every day is not uncommon) it is difficult to keep the manual up to date. But of course we try our best. To install Linux for the first time, read and follow the simple instructions in chapter 3. We strongly recommend not to try to install Linux without studying this book first, unless you are an experienced Linux user who knows his way around. And even then, it might prove valuable to at least glance at the instructions to make sure things go smoothly. 2 _______________________________________1.3.__Legend_or_what_does_it_mean:_"root@earth:____#__ls"____________________________* *______________________ Linux' dynamic - sometimes in younger days, there were several kernel patches a day - makes it rather difficult to write a book that keeps pace and is adequate for beginners. This book now tries to fulfill this claim. Instead of claiming that this is all complete we emphasize on helping new Linux users of finding their way and to support them while installing their system. Installation This section escorts the Linux novice with a documented exam- ple installation (chapter 3). It shows error prone commands and sources and provides concrete help if encountering problems. Here also the S.u.S.E. installation- and administration tool YaST TM is covered (chapter 4). Configuration If the base system is up and running there are several other things that need to be done. This might be special hardware, e. g. sound cards (kernel parameter), xfree TM GUI or connection to the internet. This mainly is done from chapter 7 to chapter 18. Crash Course Basic Linux commands are discussed in chapter 19, page 325. Special topics For those of you who are familiar with Linux/Unix there are items that might even interest you. According to the actual version there are Linux specific topics such as compiling a kernel or function of the LILO bootloader. Furthermore there is an example configuration of an X11 window manager. Appendix Besides a complete register of the most common errors/problems there is a list of important commands and keys as well as numerous ex- ample configuration files. Index and glossary If you missed something you should consider having a look at the index or/and glossary. 1.3 Legend_or what does it mean: "root@earth: # ls" In the installation guide we try to obey a general text layout; the same things should typographically look the same; the most important markup strategies are explained in table 1.1. ______________________________________________________________________________________________________________________ Text_layout____________________________________Meaning____________________________________________________________________ Linus Torvalds important persons GNU Emacs (emacs ) das program GNU Emacs , to be invoked with the command emacs Applixware the product Applixware /etc/passwd file or directory name name of the variable PATH 192.168.1.2 value of a variable ls the command ls `news' the user `news' ______________________________________________________________________________________________________________________ Table 1.1: to be continued... * * 3 1. Introduction _________________________________________________________________________________________________________________* *_______________________________________@ _______________________________________________________________________________________________________* *_________ root@earth:/tmp # help `root' shell in directory /tmp, type the command help newbie@earth:/tmp > ls shell of the user `newbie' in directory /tmp, type the command ls C:\> fdisk DOS prompt, type the command fdisk Alt key to press, here the key "Alt" Strg + Alt + Entf with `+' are "added" all keys to be pressed at the same time; keys to be pressed sequentially are separated by a space only "Permission denied" system messagess `System update' menu entry `System update' Du"sentrieb company "Du"sentrieb" + Reference to the glossary (cf. chapter B) _______________________________________________________________________________________________________* *_________ Table 1.1: Legend of the text layout 1.4 The documentation Instead of trying to cover everything, we've concentrated on helping begin- ners to install Linux and explore its fascinating possibilities. As the range of available software for Linux endlessly grows towards infinity, unfortu- nately it's impossible to describe everything in printed form. Nevertheless, we should not forget to mention the on-line documentation that is available. 1.4.1 Hypertext help A major part of the documentation is available in hypertext form. To start the hypertext system, run help . If you are running X Window an additional program to read the documentation will be started. Additional options to the help system can be displayed by invoking newbie@earth:/home/newbie > help --help You can find the help system, in package susehelp, series docDocumentation. 1.4.2 Texinfo Some program packages include documentation in Texinfo format, which is another hypertext variant. These files can be read with Infoviewer (info ) or by using the Emacs (emacs ) info-mode. The most compfortable way, of course is to use tkInfo (tkinfo ) or the old fashioned xinfo (under X11). 1.4.3 Unix Manuals The usual way to get information about programs or commands is to invoke the command man . newbie@earth:/home/newbie > man 4 _________________________________________________________________________________________________________________1.5.__Laure* *ls____________________ displays the manual page for the command, which usually lists all command options and explains the command's usage. 1.4.4 FAQ, HowTo, and README files The directory /usr/doc contains subdirectories with information about the corresponding packages. There you will often find the missing command option, the name of the configuration file you could not find anywhere else, or the name of the developer's dog. In any case, it is worth looking there before dumping the software. Especially important is the directory /usr/doc/faq which holds answers to frequently asked questions about the most common problems. The directory /usr/doc/howto contains "recipes" exlaining how to install certain pack- ages or what to do when you encounter problems. In docu on the first CD there are the latest (by time of press of the CD) versions of FAQ and HOWTO files. It might be of interest to look there! Moreover these files are not compressed so they can be viewed in advance (e. g. , with type under DOS). Under Linux one uses less 2 to read text files: newbie@earth:/home/newbie > cd /usr/doc/howto newbie@earth:/usr/doc/howto > less DOS-to-Linux-HOWTO.gz Kernel specific questions and answers are to be obtained from /usr/src/linux/Documentation; this, of course, is only available if you have installed the kernel sources (package linux resp. package lx__suse), which is highly recommended anyway. Furthermore there are many useful hints in the sources subdirectories (e. g. , for the sound driver); and for the brave even the kernel sources themselves. If you have a question which is not answered in this book, please look at these sources for additional information. The scope of this book is limited. This printed document can be up to date for only a very limited time, because the development of Linux proceeds so fast, 1.4.5 Free Books package books, series doc (Documentation) has some books in PostScript format; you can view these books with package gsview, series gra (Grafic) or package gv, series gra (Grafic)_if you don't care about trees you can print them as well. Before printing you should question yourself whether it is more economic to buy such a book. 1.5 Laurels Besides everyone who contributed to the large success of Linux, we would like to especially thank Florian La Roche whose experience is of unpre- dictable value to us. His experience and knowledge helped us building a Linux system in the first place. Thanks to his long experience with his own jurix distribution (which was a real "tip" on the internet) he could participate in the creation of S.u.S.E. Linux. Without his help this distribution certainly would not have been possible. 2 ____________________________________________________ Yes, you are right our less is smart enough to handle even compressed files ;-) * * 5 1. Introduction _________________________________________________________________________________________________________________* *_______________________________________@ Also many thanks to Dirk Hohndel and Harald Ko"nig of the XFree86TM team, who have given us valuable tips and advice concerning the X Window System and to Eberhard Mo"nkeberg who was very cooperative as CD-ROM drivers are concerned. Many thanks go to Hans Lermen, author of the well-known loadlin.exe , which allows Linux to be started from DOS, and the DOS setup.exe of this Linux system. Many thanks of course go to the dauntless beta testers who risked their running system: Christian Hu"ttermann, Dirk Ulbrich, Eberhard Mo"nkeberg, Frank Hofmann, Harald Wieland, Ingo Schro"ck, Karlo Gross, Karsten Keil, Martin Schulze, Matthias Urlichs, Norbert Eicker, Ralf Geschke, Stefan Bliesener, Ulrich Goebel, Ulrich Windl und Wolfgang Barth . Further thanks go to Stefan Endrass, who computed the cover picture of the CD. For the mathematically interested: The Steiner roman surface The picture on the titlepage exhibits the so called Steiner roman surface, which is defined by the equation x2 y2 + x2 z2 + y2 z2 xyz = 0: It is named after the swiss mathematician Jakob Steiner (1796-1863, http:: //www.www-groups.dcs.st-and.ac.uk/~history/Mathematicians/ Steiner.html). This surface is an image of the projective plane (which in turn is just a plane with coordinates x and y plus some "points at infin- ity"). One can see that the Steiner roman surface admits selfintersections along three lines. In fact it has been proven that such selfintersections occur whenever one tries to produce an image of the projective plane inside the threedimensional space with coordinates x, y and z. More information on the Steiner roman surface ca be found at http://www. ipfw.edu/math/Coffman/steinersurface.html. * * * * The global team of Linux developers is still heavily working on Linux - most of them on a voluntary basis. We want to thank them for their effort - this CD would not exist without them. Our work aims at making their work accessible to a large number of interested users. Last but not least, special thanks to Patrick Volkerding for supporting our work and, of course, many thanks to Linus Torvalds! And Angela, Barbara, Bouchra, Marcus, Marko, Mohammed, Rudolf, Si- mone, Udo, Ulrich, Virgilio und Wolfgang. And of course: Alan, Alexan- der, Birgit, Bodo, Burchard, Camelia, Carsten, Christian, Christiane, Dim- itrij, Dirk, Doris, Florian, Frank, Gerda, Gerlinde, Helmut, Holger, Hubert, James, Jan, Ju"rgen, Karl, Klaus B., Klaus F., Klaus W., Lars, Marcus, Mar- git, Marius, Martin, Melanie, Michael A., Michael B., Michael S., Reinhard 6 _________________________________________________________________________________________________________________1.5.__Laure* *ls____________________ K., Reinhard M., Reinhold, Remo, Richard, Roland, Rolf, Ru"diger B., Ru"di- ger O., Sabine, Simon, Stefan D., Stefan W., Thomas, Tilman, Uli, Werner, Winfried, and last but not least: Frank Zappa und Pawar! Fu"rth, 31st March 1998 Have a lot of fun! Your S.u.S.E. Team * * 7 1. Introduction _________________________________________________________________________________________________________________* *_______________________________________@ 8 Chapter 2 Support and services Certain problems occur over and over again while installing Linux. Many of these experiences made their way into this book. So if you encounter problems please make sure the solution is not already mentioned in this book or in our support data base (section 2.2, page 15) be- fore contacting our support team. Furthermore you will find a lots of answers in the README files on the first CD. Of course it is not possible to cover all possibilities in this book. Adding every known problem to the book is not economical and makes the book unreadable. There are always new questions and problems anyway. Should you however be confronted with a situation in which all documen- tations including the manual and the support database can not help you out, please feel free to to contact the free Installation support. The installation supportshould be able to solve you problems with a high probability. Our support team is made up out of specialist, which are able to handle even the most complicated problems. We hope to have straightened out most installational problems. So if you either do not find your way or you have a "tricky" question in Linux, do not hesitate to contact us. Our technical support team is there to help you :-) S.u.S.E. offers two kinds of support: o 60 days of installation support which is included in the price of this S.u.S.E. Linux, and o Business support, commercial support which extends beyond installation support. Please call us or see our web site for details Apart from this we offer free of charge information on Linux and information about our products online via the internet. These offers will be covered in subsequent sections. 2.1 Installation support 2.1.1 Registration We would like to help you fast and unbureacratic within the free 60-day Requirements intallation support service. * * 9 2. Support and services_of_S.u.S.E._______________________________________________________________________________________________* *_______________________________________@ Please assist us - and yourself - in providing a fast and relyfull service by including your registration code and your full name in your request. In case you have not registered yet, it can also be done via our support rep- resentative. Additional to the registration code and your full name we need your complete address. Please also submit your fone, fax and electronic mail address. We prefer to assist you in questions concerning technical problems rather than in spending time to loacte your registration data. The submiting of these datas in every support request (independent from the medium used) is the base for handling the request. In following we will shortly describe the ways on how to register as a S.u.S.E. Linux user. Registration There is a registration card that came with your S.u.S.E. Linux. There is code a registration code printed on this card; this code is unique and serves for verifying that you purchased a genuine S.u.S.E. Linux. This code matches the code printed on the first page of you reference manual. Registration If you send in your registration card or register via our WWW-server card http://www.suse.com/Customer/index.html you will become a regis- Online tered S.u.S.E. Linux user. Then you are entitled to get installation support. Registration is obligatory if you want to receive support. Since this code is product-based we request you to register every S.u.S.E. Linux products you purchase, even if it is an update or if you purchased S.u.S.E. Linux directly. Customers After having sent in your registration card you will be assigned a customer number number. We will not send you this number unless you ask (e. g. ) for support. If you receive our leaflets there is the customer number printed next to your address. If you order directly from S.u.S.E., LLC the number is printed on the bill. We are obliged to help you fast and in an unburocratical manner. Please help us by filling in any "formalities", these are your customers number and your registration code (electronic mail in subject). If you do not know your customers number we still ask you to send in your registration code and your complete name and address (where you are registered, e. g., the company's name). Our employee will tell you your new customer number. 2.1.2 Scope of our installation support So, what is this "installation support"? We want to outline what the item installation support stands for. The installation support is for getting your S.u.S.E. Linux up and running. Our definition of installation support includes the following items: o Installing the S.u.S.E. Linux system from CD o Basic configuration of a standalone machine o Basic configuration of the X11 GUI o Connecting your standalone machine to a mini-network 10 ______________________________________________________________________________________________2.1.__Installation_support____* *______________________ o Basic installation of network services, electronic mail and printers for one machine. We gladly help you to connect your machine to your internet provider. Please mae sure however that before contacting us that the problems that occur don't refer to the provider itself. Please understand that the following topics are not included in our installation support: o Configuration of large or commercially used networks o Configuring servers (i. e. mail-server, news-server, NFS-server, Samba- server, ISDN-server, Modem-server etc.); o Support of DOS, Windows (3.1, 3.11, '95, NT), OS/2, SCO and any other system. o Support for commercial Linux programs like Applixware, Adabas D, Maple, Netscape, etc., also in case if they' re distributed by us. o Support for other Linux distributions (Caldera, Debian, DLD, Red Hat, etc.); o Error analysis or support for homemade or self-compiled programs. Our installation support should be regarded as help for installing Linux, it is not a complete lecture or a course on Linux. So please contact it only in case of installational purposes and not for background information. For more detailed or complex requests, please feel free to contact our business-support service. Generally speaking our support team takes any effort to help you. We assume that you have exploited all information you could gather (this manual and the on-line help on a running system). However we can not exclude the case the S.u.S.E. Linux does not run at all or not completely on a given hardware platform. We ask for your comprehen- sion that we can not give you a 100 % guarantee for success concerning our installation support. Support for software packages Besides the basic Linux kernel system we offer a wider variety (at the moment app. 1200) software packages on our S.u.S.E. Linux CD. We try to keep these packages up to date. Anyway it is not possible for us to know each and every package's documentation by heart nor to programm additional features. Please understand if we refer you to the supplied information (manpages, README files and on-line help). Linux Snapshot!, Linux Decathlon, Internet Besides all this installation support is only available for tools and packages that came along with the S.u.S.E. Linux. Excluded are e. g. questions on other packages or on "hacker kernels". Also excluded are packages that can be found on the internet or on our other products Linux Snapshot! or Linux Decathlon. * * 11 2. Support and services_of_S.u.S.E._______________________________________________________________________________________________* *_______________________________________@ Feedback We always appreciate tips, hints and problem descriptions. We will help you if the encountered problem is a basic problem or if we already have some help at hand. Anyway your solution provides us with useful information for avoiding this problem in our next release, thus helping other S.u.S.E. Linux customers via our WWW-server or the SupportDataBase. On the other hand we try to accustom our S.u.S.E. Linux product to the wishes and ideas of our customers. Therefore we highly appreciate any criticism on our CD or on the manual. In our opinion this is the best way to rectify vital errors and to keep up to our high quality standard. Feedback can be sent any time to feedback@suse.de via electronic mail; alternatively you can send us a letter or a fax. Period of time for installation support The period of time for installation support is restricted to 60 days from the day of purchase; but no longer than 60 days after the follow-up release has been released. 2.1.3 What we need to know for helping you For an effective and fast support we need to know a lot of your machine, the software running and the problem that ocurred. Further indices as error messages and configuration files will help too. Before you contact via fone, electronic mail, fax or snail mail the S.u.S.E.support, please check the following list. The more information we get in advance, the faster and effective we can help you. In lots of cases, the error can be be solved by yourself, if you try to describe it as explicit as possible for the support. ________________________________________________________________________________________________________* *________|| | There are many solutions to well-known problems already avaiable. * * |||| | * * ||| | These are presented in our SupportDataBase at http://www.suse.de/ * * ||| | Support/sdb\_e/index.html. We recommend having a look * * ||| | * * ||| | here first! * * ||| | * * ||| | Besides the on-line SupportDataBase the complete content of the Support- * * ||| | * * ||| | DataBase can also be found in package sdb, series doc on your S.u.S.E. * * ||| | Linux CD. Those articles can be read using any WWW browser (arena , * * ||| | * * ||| | lynx , netscape ). * * ||| |_______________________________________________________________________________________________________* *________|||_ Now we list what we need to know from your or from your system respec- tively: Registration 1. Always hand in your registration data as described in section 2.1.1 Describe 2. Please describe exactly o what happened (exact error messages, screen phenomena, disk-noises etc.); please submit us the original message not an interpretation! o when did it happen (while booting from disk, while accessing the CD- ROM, while moving the mouse etc.) 12 ______________________________________________________________________________________________2.1.__Installation_support_____* *_____________________ o can it be reproduced? o what you have done to solve the problem 3. Try to remember whether you have changed anything (even if it might seem irrelevant to you) before in the system configuration before the error ocurred. This might have happened the day before or before booting for the previous time. 4. Many errors and phenomena that occur are due to faulty or incorrectly Hardware working hardware. Most often there is no visible connection between the error and the hardware components. So, for being sure we need detailed background information of your system. Please check the list below and hand us as many pieces of information as possible. o Processor: manufacturer, type, frequency, mask stepping. o Motherboard: manufacturer, name, BIOS version, onboard controller. o Memory: RAM-size, number of modules, type of RAM (FP, EDO, BEDO, SDRAM), access time (60 ns, 70 ns), cache-size, type of cache. o Hard disk controller: SCSI, IDE, EIDE, onboard or module, name, BIOS-version, chipset. o Hard disks: Type, manufacturer, name, size, bussystem (SCSI, IDE, EIDE, SSA), partitoning. o CD-ROM drive: Type, manufacturer, name, size, speed, internal or external, SCSI EIDE (ATAPI), parallel port. o Other media: CD-writer, streamer (type, manufacturer, SCSI or EIDE, floppy-streamer, capacity), ZIP-drives, Jazz-drives, MO-drives, flopti- cals. o Graphics device: Type, manufacturer, chipset, bus (PCI, VL-bus, ISA), memory size. memory type (DRAM, VRAM, WRAM, SGRAM, EDO). o LAN card: type, manufacturer, chipset, bus (PCI, VLBus, ISA), LAN type (Ethernet, Token Ring etc.), Connection type (Coax-Cable, Twisted- Pair), portaddress, irq. o Other peripheral devices: Printer, parallel port, serial port, modem, ISDN card, scanner, soundcard. o 5. If you encounter difficulties while installing the XFree86 TM system the XFree86TM following information might come in handy: o Graphics device: Type, manufacturer, chipset, bus (PCI, VL-bus, ISA), memory size, memory type (DRAM, VRAM, WRAM, SGRAM, EDO). o Monitor: Type, manufacturer, max. horizontal frequency, max. verti- cal frequency, band with. o Mouse: Type, manufacturer, connection (serial, PS/2-Port), date of purchase o Installed X-server: Package name, if there are multiple servers in- stalled please tell us the order of installation. * * 13 2. Support and services_of_S.u.S.E._______________________________________________________________________________________________* *_______________________________________@ o How did you configure X? (e. g. using XF86Setup or xf86config ). o The file \tld/.X.err which is to be found in the home directory of the user who invoked a startx . o If you are running xdm (runlevel 3): - /var/lib/xdm/xdm-errors: error messages of the X-server - \tld/.xsession-errors Important files and screen messages The information just mentioned can be gathered by consulting the manual or the following files or screen messages: o loadable modules /etc/conf.modules /var/log/messages o Kernel /usr/src/linux/.config o Boot messages /var/log/messages /var/log/boot.msg ________________________________________________________________________________________________________* *________ | The last lines of /var/log/messages are sufficient. Please do not send * * ||| | * * ||| | us the complete /var/log/messages! * * ||| |_______________________________________________________________________________________________________* *________|||_ Screen messages: o PCI devices: cat /proc/pci o Interrupts: cat /proc/interrupts o Port addresses: cat /proc/ioports o Memory: cat /proc/meminfo o managed devices: cat /proc/devices o Networking devices: cat /proc/net/dev o SCSI devices: cat /proc/scsi/scsi o SCSI controller: cat /proc/scsi//0 These can be redirected into a file. Example: root@earth: # cat /proc/pci > /tmp/pcidevices Now you can send this file /tmp/pcidevices to our support team. ________________________________________________________________________________________________________* *________ | The "files" in /proc can only be viewed using cat , less , or more , but * * |||| | * * ||| | can not be edited! Well, nowadays even this is possible, but do not rely on * * ||| | it! * * ||| |_______________________________________________________________________________________________________* *________|||_ If you do not manage to get a Linux system up and running you can gather the information mentioned above via linuxrc . Please refer to section 3.15, page 32. 14 ___________________________________________________________________________________________________2.2.__Further_services___* *______________________ 2.1.4 How to contact the S.u.S.E. support team You can reach our support team via electronic mail, fax, regular mail und telephone. Contacting S.u.S.E. via the fone is on the first look the fastest way. The spelling of error messages from your side and the spelling of command entries or file names from ourside on the fone is in the majority of cases quite difficult and boring ;-) Relating to our experience, it is in most cases the fastest way to describe your request in a fax, electronic mailor letter. On the one hand you've the chance to formulate your request more precisely, and on the other our employee can tell in most cases on the first sight where the problem is located. In addition, we handle requests via electronic mail, fax or letter 7 days a week, while the hotline can only be reached 2 times a week. For the fast and easy composition of a support request, we supply you with an electronic support request questionaire on 2 different locations: o on the first CD of the installation set o on the installed system The easiest is to send us an electronic mail. You can reach our support time via: o via electronic mail Address: support@suse.com Address (Europe): support@suse.de o via fax Fax: +1-510-835-7875 Fax (Europe): +49-911-741\,77\,55 o via fon Fon: +1-510-835-7873 Mo & Th 11:00am - 5pm (PST) 2.2 Further services Additionally we would like to draw your attention to our free of charge ser- vices that are available round the clock: o S.u.S.E.'s web site http://www.suse.com o S.u.S.E. mailinglists (information and discussions via electronic mail): - suse-announce-e@suse.de - announcements concerning S.u.S.E. Linux (english) - suse-announce@suse.de - announcements concerning S.u.S.E. Linux (german) - suse-linux-e@suse.de - all about S.u.S.E. Linux (english) - suse-linux@suse.de - all about S.u.S.E. Linux (german) * * 15 2. Support and services_of_S.u.S.E._______________________________________________________________________________________________* *_______________________________________@ - suse-isdn@suse.de - ISDN and S.u.S.E. Linux (mainly in german) - suse-applix@suse.de - the Applixware office suite (mainly in ger- man) - suse-adabas@suse.de - about Adabas D (mainly in german) - suse-informix@suse.de - about Informix (mainly in german) Send an electronic mail message to majordomo@suse.de with subscribe in the body: subscribe , the subject does not count; example: subscribe suse-announce-e to receive all announcements. To unsubscribe a list write again an elec- tronic mail message to majordomo@suse.de; this time with the body: unsubscribe suse-announce-e Please note, that unsubscribe has to be done from the same electronic mail account. o S.u.S.E.'s ftp site ftp.suse.com Latest information, updates and bug fixes Login as `ftp'. 16 Chapter 3 Installation 3.1 Don't panic! You can do it! It is not easy to find a good middle ground for an installation manual. If you go into any gory details or if you describe problems and their solutions there is a risk of scaring neophytes away. On the other hand, if you leave out all the details, experienced Linux freaks will lay the book aside. If you sacrifice these details for clarity, the manual might not be useful in case you do encounter any problems. That is why we decided to split this manual into two parts: o A quick-install section that should work for most users (section 3.2). o A detailed installation manual where we cover background and pitfalls. But, as we go into detail, this section obviously will get quite voluminous (section 3.3). In case of doubt try the quick install; this will work in almost any case. If this fails you should try the detailed installation. 3.2 Linux in 30 minutes - the quick install There are several prerequisites to doing a quick install. First you have to make sure that you have either extra space on your hard disk for Linux, a free partition you might want to use, or another operating system you want to remove. If this is not the case you should prepare your hard disk in advance. fips on our first CD might come in handy; additional tips can be found at section S1, page 21 in section 3.28, page 60 and section 3.29, page 61. Then you perform the following steps. The output of the several steps is displayed on screen. Additionally you may scroll through the help texts. o Insert the S.u.S.E. boot disk included with this product and boot your machine. If the greeting appears just press _- . If you have a suitable BIOS you can even boot from CD. o linuxrc starts. Select language, screen and keyboard. o Load necessary hard disk kernel modules (You do not need modules if you only have (E)IDE disks and/or CD-ROM drives installed!) and for the installation media (via CD-ROM, network card or PCMCIA). This either works automatically or has to be done manually. For some mod- ules you can add parameters. Information on parameters can be found in section 7.3.2, page 102 * * 17 3. Installation _________________________________________________________________________________________________________________* *_______________________________________@ o Select `Start installation' for invoking YaST. The source media normally is either `CD-ROM' or `Network'. o YaST starts and requests a disk. Since we have not created one yet, we answer `No'. - Online help is always available via F1 . o Now you assign your Linux partition. To do so move to `Settings for installation' and `Partition hard disk'. Do not forget to assign a swap partition. The type of the swap partition has to be set manually! Additional information on the philosophy of partitioning is found in section 3.28, page 60 and section 3.29, page 61 o Now you have to assign mountpoints with `Settings for installation' and `Assign partitions'. o Now leave YaST. Save the data on a "save disk" (not the S.u.S.E. boot disk!)1 . Now answer `yes' and the system will reboot. o Do not forget to re-insert the S.u.S.E. boot disk! o Press _- at linuxrc 's greeting. Now again you have to choose lan- guage, screen and keyboard. Load the necessary kernel modules and se- lect `Start installation' for invoking YaST as you have already done before. Nothing new. o Now you can answer `Yes' if asked for a disk_if you have saved the data on the disk successfully. Insert "save-disk" created earlier. YaST now formats your Linux partitions. o Confirm the swap partition and allow it to be initialized o Via `Start/Setup installation' `Change/create configuration' you select which packages you want to install and start the installation procedure afterwards with `Start installation'. The selected pack- ages are installed (first only those that are on CD 1). o Now quit YaST pressing `Finish up installation' and choose the kernel you want to install for future use. o Do create a boot disk for booting Linux. This might not only be handy if there is an emergency. Here you will need to select your CD-ROM once again. o You can now install LILO, the bootmanager. o Now you give a name to your machine and select the network you have. Afterwards the machine boots again and you can log in to your freshly created Linux. o Now YaST starts again and all packages that were not on CD 1 are in- stalled. Now you configure all those tiny details of your system. o Now you can login as user `root'. You can start YaST and administer your system with `Administrating your system'. You are done! 3.3 This installation manual There are many roads leading to a correct and stress-free Linux installation. But some of them are more complicated than others. We at S.u.S.E. have tried to determine how less-experienced Linux users can be supported better than before.______________________________________ 1 Some times saving on a disk doesn't work; in this case just skip this step. 18 ____________________________________________________________________3.4.__Basic_principles_of_Linux_installation____________* *______________________ It's quite difficult for the beginner to have an overview and of course to al- ways choose the correct alternative. Therefore we separated the installation part from the rest of our book and tried to focus on background and coherence. Moreover, since we excluded everything about system management and the like this material becomes more compact. If you are an advanced Linux user you might consider some explanations as useless or too long. Please bring back to mind your first computer, happily taking advantage of any informa- tion you could get. This installation guide is not considered as an instruction for the S.u.S.E. tool YaST nor for other programs and does not describe Linux completely. Only those things considered helpful for the installation and un- derstanding are included here. 3.3.1 Finding your way through this installation guide: The chart on page 20 shows you the myriad of possible installation options. You should start at the top and work through right to the bottom, until you (hopefully successful) reach the end of the installation. Of course you should not skip steps in this plan. Steps marked as D1 to Dx are designate decisions, whereas S1 to Sx are working steps. All decisions and working steps will be thoroughly explained later on in this installation with all necessary background information. Please read this care- fully and follow the instructions. If you encounter any problems you might find the links to problem descriptions which are mentioned together with the working steps, of some use. Problem descriptions are named P1 through Px which we describe in a separate chapter. Concentrated information on a certain point you can find in info blocks I1 to Ix . You'll find links to the Infoblocks in various places; this makes life easier as it reduces the amount of redundant data. Furthermore there are links, if necessary, to other chapters in this book. If you have mastered your installation you should have a running base Linux on your system. If this fails for any reason you should contact S.u.S.E. tech- nical support. You'll find our contact information in chapter 2, page 9. 3.4 Basic principles of Linux installation We are going to describe the real Linux installation which requires you to create one (or more) partitions and a separate filesystem; the demo mode which is based on MS-DOS filesystems and provides less performance and features is not mentioned here. Linux installation mainly consists of the necessary preparation and three steps: o create space (partitions) for your Linux on your hard disk. o start an independent + Base-Linux. Here you have three different possi- bilities: - You boot with the supplied S.u.S.E. boot disk (for really obscure hard- ware you might have to create your own own boot-disk) and tell the * * 19 | * * | 3. Installation _________________________________________________________________________________________________________________* *_______________________________________@ | * * | ________________________________________________________________________________________________________* *________|||| | * * | || * * || | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | |_______________________________________________________________________________________________________* *________ | Figure 3.1: Scheme of installation kernel to load the rest of the installation system from CD. If you have suitable hardware and BIOS, you can even boot from CD. - You prepare an installation disk; boot with the supplied S.u.S.E. boot- disk (for really obscure hardware you might have to create your own boot-disk) and the prepared install-disk containing the root-image and the installation tools is loaded into RAM disk. - The Linux kernel is loaded directly from CD (under DOS) via loadlin ; the root image and the installation tools are loaded from CD or HD into a RAM-disk. o Now you can log in to your + Base-linux. You now load the necessary 20 ____________________________________________________________________________________________________3.5.__S1:_Partitioning__* *______________________ drivers using linuxrc . With YaST you customize your system, partition and format your HDs and install the base packages for your Linux system. o Now YaST creates partitions and filesystems, the selected packages are installed from CD or from an HPFS-partition. Now a real Linux is run- ning on your machine. Now you have a real Linux running; but for you becoming the system admin- istrator there are still a lot of things that need to be done. The X-Window system, networking, accessing online services, compiling of your own kernel and much more : : : These items are far beyond the aim of a basic installation and are therefore covered in separate chapters. 3.5 S1: Partitioning What's it about? This step prepares your HD for Linux partitions. Requirements You do not want a demo installation (you would not need a new partition for demo mode). You have some HD space, time, S.u.S.E. Linux, perhaps some disks or tapes for backups and a boot disk of your former OS. Summary By partitioning, your HD can be separated into several independent parts. One reason for partitioning your HD might be the coexistence of several different operating systems with different file systems on your HD . HDs are divided into cylinders, one cylinder always contains exactly the same amount of tracks, cylinder 0 being the innermost of them. How do you get space for your Linux partition? o Single partitions can be deleted but, of course, all data which resides on these partitions will be lost. This space is now available for new partitions (e. g. Linux). Of course you can also split this partitions into two or more partitions. o Single partitions can be assigned to a different file system. As with delet- ing, all data on these partitions will be lost. o If you are using MS-DOS or Windows you can minimize the last partition without losing your data. If all data resides on low-numbered cylinders, special programs (e. g. fips ) can decrease the upper cylinder limit to re- duce the size of your partition leading to the possibility of creating a new partition for Linux. Using a defragmentation program ensures you that all data is moved to the beginning of the partition. o The easiest method of course is buying a new HD. This of course will cost you some money. * * 21 3. Installation _________________________________________________________________________________________________________________* *_______________________________________@ ________________________________________________________________________________________________________* *________|| | Every change of partition tables must be done with extreme caution! Please * * |||| || read the documentation of the corresponding programs! Severe problems * * |||| | can occur while handling partition tables; you can even lose all your data. * * |||| | We at S.u.S.E. cannot be held responsible for this! It is strongly recom- * * ||| | * * ||| | mended to have at least a boot disk and a backup of your important data. * * ||| |_______________________________________________________________________________________________________* *________|||_ Step by step This is how to proceed to partition your HD: 1. If you don't know already, you have to find out how many partitions your HD consists of and, of course, the size of the partitions. Use fdisk of your OS for this task. 2. Check how many partitions you really need and which size you're going to assign to them. If you are unsure about the correct size we recommend 3 partitions (/ (root), /home , and swap ). The root partition should be at least 50 MB (depending on how many packages you are going to install) up to 200 MB (if you're planning a bit more). The size of your /home partitions depends on the number of users you are going to "install". Swap partition should be about 20 MB of space. 3. Write down the partition data; you're going to need it later on during the installation process. 4. Now a backup of your HD is highly recommended! If you don't have a tape or streamer installed and don't want to backup everything on flop- pies (which is rather tedious work) at least make a backup of the most important data and files (e. g. autoexec.bat , config.sys , *.ini etc.) Create a boot disk for your OS and check whether you can boot with it! There are several programs that might come in handy on your boot disk; those are programs as fdisk, an editor, a format program and of course your backup program. 5. The next step depends on your system configuration DOS/Windows 95, one partition on your HD, and no backup You have to reduce the size of your partition without deleting data. Move all data to the beginning of your partition using a tool such as defrag (DOS/Windows). Defragmentation programs normally don't move hid- den files. There might be such hidden files on your HD (e.g. copy protec- tion). You must ensure that those files don't have to be at a certain loca- tion on your HD. If you are sure that there aren't any of those hidden files you can change the attributes of those files from hidden to not hidden . Some defragmentation programs offer the option of deciding whether you want to set this flag or not. Please keep in mind that even the windows- swap file is a hidden file. If it's in your way you have to unselect it under "disable virtual memory". Now that you have accomplished the first task of gaining enough space you should move to the directory \dosutils\fips.exe . There you will find fips.exe . Please study the included documentation about fips carefully since handling partition tables can be rather tricky! The program 22 ____________________________________________________________________________________________________3.5.__S1:_Partitioning__* *______________________ fips.exe only runs on DOS, not on Windows; you have to exit windows or (if using Windows 95) bring it into DOS mode (see section S3, page 27). After having run fips.exe you will have a second partition which later on will be divided into your Linux partitions. DOS/Windows 95 and several partitions or you have a complete backup You can delete your DOS partitions and afterwards install them reduced in size again. Here you are going to lose all data on those partitions. You have to have a complete backup at hand (file backup, no image backup!). Using fdisk you delete all partitions and install the new ones. Next you format those partitions, install your OS and copy all data from your backup back to disk (remember that you have your backup program on this boot disk). OS/2 Here you have the following possibilities: o You reduce your OS/2 partitions. Therefore you must backup all data on your OS/2 partition, delete those partitions using OS/2's fdisk and reinstall them reduced in size. The space now available will be used further on for your Linux partitions. o You plan to use Linux instead of OS/2. Then you have to do nothing but change the partition type later on to Linux. Or you delete all OS/2 partitions and reinstall them later as Linux partitions. o You buy a new HD. Formatting and partitioning it will be done later on with YaST. UNIX/Linux You already have suitable partitions, meaning that you don't need to bother with repartitioning your disks. Setting the partition table will be done by YaST. 6. Reboot your computer. 7. Check whether your old system is still running! Linux partitions on the freshly created new space will be installed later on using YaST. Problems The following problems can occur: o Partition might not be adequately reduced due to some un removable, see files section P1, page 53. o Under DOS/Windows your CD-ROM had another HD Identifier; under Windows the computer hangs, see section P13, page 56. * * 23 3. Installation _________________________________________________________________________________________________________________* *_______________________________________@ 3.6 E1: Can I use the S.u.S.E.- boot-disk? What's it about Are you able to use the supplied S.u.S.E.-boot-disk? This mainly depends on your system. The alternative is booting via a "hand-made" bootdisk or via DOS and loadlin . You'll find an overview of all schemes on page 20. Summary The supplied S.u.S.E.bootdisk covers most existing hardware. All necessary drivers are loaded as modules via a menu. Thus making the S.u.S.E. boot disk a rather flexible installation tool. You cannot use the S.u.S.E. boot disk if your system is supplied with exotic hardware or your CD-ROM is not supported by Linux (here you need a second disk or you will have to copy files to disk. For these possibilities an alternate way is described in the diagram on page 20). If you're working with DOS or Windows 95 and plan to start Linux via loadlin (a rather convenient way) it is recommended not to use S.u.S.E. boot disk but to work with loadlin instead. This saves you the installation of loadlin and the selection of a suitable kernel. Recommendation Use S.u.S.E. boot disk if: o you don't have exotic hardware installed (in case of doubt, just try any- way). o you are using Linux, Unix, Windows NT, or OS/2. o for any reason you don't want to use loadlin (although this is a recom- mended method). 3.7 E2: Does Linux support my CD-ROM? What's it about? Whether your CD-ROM is supported by Linux is important for the instal- lation. If it is not, you will have to copy all necessary data for your Linux installation onto your HD. Summary Linux supports almost every CD-ROM. ATAPI and SCSI CD-ROMs, and even exotic CD-ROMs will work. For further information please see In- foblock section IB3, page 57. Recommendation You can assume that Linux supports your CD-ROM unless you definitely know the opposite. If you encounter problems, the most you have to do is to start all over again and choose a different installation procedure. 24 ______3.8.__S2:_Copying_packages_onto_the_HD_in_the_case_CD-ROM_drive_not_supported_________________________________________* *______________________ 3.8 S2: Copying packages onto the HD in the case CD- ROM drive not supported What's it about In the case that our standard kernels do not support your CD-ROM drive or your CD-ROM drive is not supported by Linux at all, you have to copy Linux packages onto your HD using your current non-Linux OS. This will allow you to install Linux directly from your HD, circumventing the need for the CD-ROM drive. Recommendation You are using DOS, Windows or OS/2 and know (perhaps from an failed installation) that Linux does not support your CD-ROM. Moreover you have plenty of space on your DOS, OS/2, or Windows partition. Furthermore you should have had a look in package descriptions which pack- ages you would like to install. Otherwise we will just install the base system. Summary There are two places where you can replace CD-ROM access by DOS-file- access. o If your + base-Linux is loaded using loadlin the system gathers the con- tents of the root-disk, which is loaded directly into RAM disk, and the contents of install-disk from CD. If your CD-ROM is not supported by the standard kernels on CD offered by setup , data can easily be loaded from a DOS or HPFS file system. Naturally those files have to be installed there first. o If your base Linux is already running and you want to install the chosen packages using YaST this is normally done via CD. If Linux does not sup- port your CD-ROM you have to copy the files using DOS, Windows, or OS/2 onto your HD. YaST offers you the possibility to choose a directory to install from. You might you have to install the system step by step if there is not enough space to install it all at once. After having a base system running you might find a kernel (or generate your own) which recognizes and supports your CD-ROM. On CD you'll find three types of files related to kernels. Files without extensions are considered ker- nels only for installation purposes. These kernels only support hardware that is needed for installation (no streamers or mice). Files with .ikr extension are considered normal installable kernels. There is one more file ending in .inf . This obviously is a text file containing a number which YaST uses to present information about a given kernel. You should always copy a suitable kernel even if you plan to start via loadlin or floppy. YaST will create a boot disk using this kernel at the end of the installation procedure. * * 25 3. Installation _________________________________________________________________________________________________________________* *_______________________________________@ Step by step Here's how to copy the chosen packages onto your HD. 1. First create a directory you want to copy the files to. The name of this directory is unimportant, as long as you don't forget it. In our example we will refer to it as \emil . Additionally you have to create a subdirectory \emil\suse 2. Under the \emil\suse tree you will need at least three directories. Let's call them a1 , images and setup . These are necessary for the base instal- lation.Therefore you should create these directories. Diagram figure 4.3, page 69 shows the complete file tree needed in an overview. 3. Copy all files from \suse\a1 of CD 1 into \emil\suse\a1 on your HD. 4. Under \suse\images select a kernel which supports your hardware. More detailed information about which kernel supports which hardware can be found in \suse\images\readme.dos . Copy this kernel to \emil\suse\images . A kernel consists of three files, one without an extension and those with .ikr and inf extensions. If space on your HD is not an issue you can of course copy all kernels into this directory. You can then choose your kernel later on. More informa- tion about this can be found under Infoblock section 3.27.1, page 56. 5. Copy \suse\images\root \suse\images\initdisk.gz and \suse\images\TRANS.TBL to \emil\suse\images 6. Copy \suse\setup\loadlin.exe to \emil\suse\setup 7. Furthermore copy file \suse\setup\inst-img to \emil\suse\setup . This file is relatively big, but it is only needed while doing the first instal- lation. If the base system is up and running, you will always be able to install additional packages from a DOS partition; then the file inst-img could be removed. 8. Create the directory \emil\suse\setup\descr and copy all files from \suse\setup\descr into it. 9. If there is some space left on your HD you can copy \suse\setup\du to \emil\suse\setup\du , which, of course, you have to create in advance. These files are not really necessary but will help you later on for showing you how much space there is left and how much is already occupied. If you have enough space you might find that this comes in handy. 10. Now you have everything that is absolutely necessary for installing Linux on your HD. All other software is still missing. Since your CD-ROM is not supported by Linux you have to install everything step by step, mean- ing that you have to copy everything you want onto HD and install it from there using YaST. You don't have to do that right now, but if you already know what you're going to use, you can start now: Just create the cor- responding directory under \emil\suse and copy everything from CD (meaning the corresponding directory) into this newly created directory. You can find all packages (their contents) in the online documentation or in the package descriptions. 26 ____________________________________________________________3.9.__S3:_Bringing_Windows_95_into_DOS_mode_____________________* *______________________ Problems There should not be any problems so far. If there is not much free space on your DOS, HPFS file system please keep in mind that you have to have root to install the above mentioned base system, otherwise you have to say goodbye (just for some time) to your DOS/Windows/OS/2 software. 3.9 S3: Bringing Windows 95 into DOS mode What's it about? You have to bring your computer into real mode (under DOS) to proceed with installation. Requirements You work under Windows 95 and have completed the preparations. Installa- tion of Linux can commence. Summary The installation program setup.exe is an MS-DOS program which is only able to load the Linux kernel of your base Linux into memory if either the CPU runs in real mode or a VCPI server2 is active. The DOS box of Windows 95 runs in virtual 8086 mode but does not offer an VCPI server. This is why setup does not work here. Step by step There are two alternatives: Switching to DOS mode from Windows 95 or booting your computer and selecting command line input (DOS). If Win- dows 95 is already up and running click on `Start'-`Shut down'-`Restart the computer in MS-DOS mode'. Or you can just press F8 while booting and choose `command line input'. Problems There should not be any problems so far. 3.10 S4: Invoking setup and first steps with setup What's it about Setup.exe starts your + base Linux which enables you to proceed with real Linux installation. 2 ____________________________________________________ e. g.provided by emm386.exe . * * 27 3. Installation _________________________________________________________________________________________________________________* *_______________________________________@ Requirements You have started MS-DOS or an MS-DOS box (not in protected mode), CD 1 is in your CD-ROM drive and you can access it. Summary In this step you will continue until you reach the point where you have to decide between one of two alternatives, namely booting with boot disks or with loadlin from CD. Step by step Here's how to do your first steps with setup: 1. Start setup.exe in the root directory of your CD. 2. Choose your favourite language; for `English' installation just press _- . 3. Enter your CD-ROM drive (e. g. E : on DOS). Possibly this has changed due to an additional DOS partition. 4. Setup.exe welcomes you; we reply to such a nice gesture by pressing _- . 5. The following sections will explain the booting of the base-Linux. sec- tion E3, page 28 (below) will help you decide what booting method to choose, and subsequent sections will guide you through the respective process for each method. Problems There should not be any problems so far. 3.11 E3: Which way do I want to boot base-Linux? What is it about? There are two possibilities to boot your base Linux: either from disks or directly from CD-ROM by means of loadlin . Now you will choose the best option. Summary Certainly, the easiest way is to start your base Linux directly from CD using loadlin.exe . It runs on DOS and loads a kernel image from CD into RAM and starts executing kernel code. Some things must be fixed for this to work properly though. The computer has to run in real mode or in virtual VCPI mode3 . OS/2 DOS box or Windows NT doesn't work. Booting via disks is almost always possible but is a little more tricky and requires an extra amount of time. We are talking about the disks created by setup . The easiest way is of course to use the S.u.S.E. boot disk (or booting from CD). We will explain these options later, but for let's stick to the decision about loading via setup or via loadlin . _____________________________________________3 as provided by, e. g. emm386.exe. 28 _____________________________________________________________________3.12.__S5:_Creating_boot-disks_with_setup______________* *______________________ Recommendation Take the "loadlin way" whenever possible as long as your CD-ROM is sup- ported and you are not working under OS/2 or Windows NT. When in doubt, just try it. If this fails you have to start over again at section S4, page 27 and choose the "floppy-way". If you're installing from your HD, you must chose to boot via the S.u.S.E. boot disk or the disks you'll create with setup . 3.12 S5: Creating boot-disks with setup What it's about Here you create one or two disks for loading your Linux kernel and start up your + base Linux. Requirements You need one or two 3.5"-HD-disks and of course a bootable 3.5" floppy drive. If there is an additional 3.5"-floppy drive installed that functions as boot drive, see section P12, page 55). If you are working with Windows 95 you have not launched setup from within a DOS box but in MS-DOS mode instead. Summary For booting the base Linux via disks you need at least a disk with a kernel. The S.u.S.E. Linux CD contains several disk images. Such an image can be copied onto disk with setup and is thereafter called a boot disk. Included in those images are the linux loader (LILO) and the program linuxrc . It is possible to choose a kernel at boot time and to add parameters as needed. By the way, LILO need not be installed on disk but should be your choice if you plan to install more than one operating system and want to select at boot time which to boot. Installation of LILO is not the subject of this chapter but is covered in chapter 8, page 121 in great detail. linuxrc is your assistant for loading kernel modules adapted to your hardware and to start the installation. Even the included S.u.S.E. boot disk can be used as boot disk in a case of emergency. You have to create your own boot disk only if you have exotic hardware installed that is not supported by the modular kernel on this disk as explained below In the latter case selection of a proper kernel is important. It has to sup- port your hardware. There is more information on kernels in section 3.27.1, page 56. Then you boot either with disks only or with 1 disk and a CD. Also note the following regarding boot disks on CD. These kernels are not made for being installed permanently on your system. They cover a wide range of hardware and make installation as simple as possible. The kernel should not be too big since there has to be some space left in the RAM disk. If your Linux system is up and running it is recommended you build * * 29 3. Installation _________________________________________________________________________________________________________________* *_______________________________________@ your own kernel which perfectly suits your system. Thus you gain optimal performance and a minimum amount of required memory. How to build a kernel is described in chapter 13, page 181. Step by step Here's how to create a set of disks: 1. In 3.10 you started setup.exe and now you're to the point where you must decide to install via loadlin or floppy. 2. Select `floppy' and press _- 3. Now you have to select a suitable kernel that supports your SCSI adapter if you have one. setup shows the vital part of kernel messages. If you need further information you can look it up in \disks\readme.dos or in section 3.27.1, page 56. Remember your kernel's name. You will need it later on. Now press _- . If you find out that the chosen kernel was not the best choice you can come back here and select another one. 4. Now you're ready to create the boot disks. Insert the first (DOS formatted) disk into the 3.5" drive and select which disk to create. o Start with the boot disk (only recommended, not necessary). Move the cursor onto `Boot' and press _- . o setup requests you to confirm that you have inserted a disk. Press _- and the disk is written. o When this is done press _- . o Now select `Done' to exit this screen and setup . 3.13 S6: Installing loadlin and starting your base-Linux What's it about This step shows you how to install a Linux kernel, to load a base Linux using loadlin.exe and bring up your + base Linux Requirements If you have made it this far and you are ready for your first Linux prompt everything is prepared. Summary Which kernel to choose mainly depends on your CD-ROM drive or your HD (meaning the controller). Later on you can easily choose another kernel or, which is recommended, build your own kernel which exactly fits all your needs. More information about this topic is to be found under section 3.27.1, page 56. Setup now creates a directory \loadlin on your DOS partition and copies setup.exe , loadlin.exe , linux.bat and the selected kernel (renaming it to zImage ). After having finished the installation you can start your Linux by 30 ____________________________________________3.13.__S6:_Installing_loadlin_and_starting_your_base-Linux______________________* *______________________ invoking linux.bat (all you have to do is to supply the command with the root partition of your Linux; this will be thoroughly explained in section S11, page 37). After this step has been completed the kernel will be loaded and started. Step by step Here's how to install loadlin and + base Linux. 1. Select option `loadlin' and press _- . 2. Now you'll be presented with the amount of RAM installed on your com- puter. In most cases this will be correct and you confirm by pressing _- . However, if it does not match you should correct the value. 3. Now you have to decide whether Linux supports your CD-ROM. You should have done this in section 3.7, page 24. Just give exactly the same answer. o If Linux supports the CD-ROM just press _- and you are set. o If Linux doesn't support your CD-ROM you should have copied the necessary files to your HD in section S2, page 24. Now select the `HD' option and press _- . Now you have to specify the complete path to your directory suse . In our given example we used \emil meaning that you have to fill in \emil . 4. Now you should choose a suitable kernel if you have not done so yet. If you decided in section 3.7, page 24 that Linux does not support your CD-ROM, hopefully you copied a kernel onto your HD in section S2, page 24. Now you press _- . Help on your decision about which kernel to choose can be found in Infoblock section 3.27.1, page 56. 5. The next screen requires you to decide whether to install loadlin . Answer `yes'. Setup now creates \loadlin and copies the necessary files into that directory. 6. With the option `Starting Linux' we start our + base Linux. Here you will notice about two pages scrolling down rather quickly. If everything went right you will be welcomed by S.u.S.E.. You can easily have a look at the kernel output by scrolling the text back using Shift + Page " or Shift + Page # , respectively. Problems There can be problems at two points. loadlin is not able to load the selected kernel or the kernel has problems with the specified hardware. o loadlin doesn't have enough free memory to load the kernel: see sec- tion P8, page 55 o loadlin cannot load the kernel because the machine runs in virtual 8086 mode but no VCPI server is available: see section P11, page 55 o loadlin does not work: see section P9, page 55 o Your SCSI adapter is not recognized. o The CD is damaged: see section P4, page 53 * * 31 3. Installation _________________________________________________________________________________________________________________* *_______________________________________@ 3.14 S7: Creating boot disks with Unix What's it about? You have a Unix already installed and want to switch to Linux. Requirements Your CD-ROM is supported by Linux and you have no way of installing Linux via a DOS partition. Step by step This is how you create your boot-disk: 1. If you have to format the disk: fdformat /dev/fd0h1440 2. Mount CD1(e. g. to /mnt ): cd / ; mkdir mnt mount -tiso9660 /dev/cdrom /mnt 3. Change to the disks directory on CD: cd /mnt/disks 4. Now create your boot-disk dd if= of=/dev/fd0 README file in disks directory and Infoblock section 3.27.1, page 56 contain specific information about the kernels. 3.15 S8: Starting linuxrc What's it about linuxrc loads every needed driver as kernel modules. Thereafter YaST starts as well as configuration of your system. Requirements For using the S.u.S.E. boot disk (or the creation of your own) together with CD you need a CD-ROM drive supported by Linux. Furthermore you need a bootable 3.5" drive (if you have an additional 5.25" inch drive which serves as boot-drive have a look at section P12, page 55). If your CD-ROM is not supported you should have created the installation- disk with setup . For booting directly from CD this feature has to be supported by your ma- chine. Another alternative: with loadlin you have already started your Linux. 32 ____________________________________________________________________________________________3.15.__S8:_Starting_linuxrc_____* *______________________ Summary The extreme flexibility of the kernel is achieved via kernel modules. Which modules are needed is entered by the user in linuxrc . This also applies for the S.u.S.E. boot disk, your own boot disks or the kernel started via loadlin . Even drivers loaded as modules might be supplied with parameters. The item kernel parameter is discussed thoroughly in section IB2, page 57 and section 7.3.3, page 111. Furthermore you can use the boot disk for starting the installed system at a later date as well, not only for installation (as an emergency "rescue disk" so to speak). Step by step Here's how to start Linux with a boot disk: 1. If you start with a boot disk: (a) Insert the boot disk into your drive and boot the machine. (b) A welcome screen appears, just press _- or wait a while : : : (c) The message "Loading Linux..." appears and after a while the ker- nel boots. 2. Now linuxrc is running. Now you may select a language. " and # moves you up and down _ and ! choose responses; here it's simply `Ok'. _- sends the command. 3. Now select color or monochrome monitor; `OK' moves you on. 4. Now the menus get much prettier. `English' is already set as default, so just press _- 5. Now we are in the main menu. Here are the options available: Help Select if necessary Settings Here you can change language, screen and keyboard. System information Here a lot of interesting information about your machine can be obtained. Kernel modules Here is where you go to load modules. Exception: If your hard disk and your CD-ROM are connected to an (E)IDE controller you do not need to select this item. (E)IDE sup- port is integrated into the kernel. You can skip this section and start installation; how this is done is described below. Installation /Start system Here the system boots. More on this is given below. A detailed description of linuxrc can be found in section 11.3, page 160. 6. Select loading of kernel modules using `Kernel modules'. 7. The next submenu asks which modules should (or must) be loaded. Nor- mally these are: o If you have a SCSI hard disk or a SCSI CD-ROM: a SCSI module. o A CD-ROM module if your CD-ROM is attached neither to a SCSI nor to an (E)IDE controller. * * 33 3. Installation _________________________________________________________________________________________________________________* *_______________________________________@ o A networking module if you want to install via network (NFS) o A PCMCIA module if you want to access the installation system via a PCMCIA card 8. This is how to proceed for loading kernel modules: o For selection you press the corresponding submenu and select the ap- propriate type. After a _- there is an entry to enter parameters. Here typical settings are set as default. More on kernel parameters is in section IB2, page 57 and section 7.3.3, page 111. o You can also select `Automatic loading of modules'; most of- ten this is sufficient. If unnecessary modules are loaded you can dein- stall them using `Remove loaded modules'. o The result of loading is presented. If it failed, either the driver does not suit the hardware or the kernel parameters are wrong or missing. o With `Show loaded modules' you can check the result of your ac- tivities. You will see all loaded modules and you can identify acci- dently loaded modules. o Accidentally loaded modules can be removed with `Remove loaded modules' 9. If you have selected any needed modules, go to `Installation/Start system' in main menu. In the submenu popping up you indicate the installation system to take. Of course you have inserted the first CD. Now the installation tool YaST starts. Problems Problems can occur if: o Your SCSI adapter is not found. Use a kernel with integrated SCSI driver. o The ATAPI CD-ROM hangs. See section 3.26.5, page 53 3.16 S9: Invoke YaST What's it about? Now you start installing the real Linux. You do this with the installation tool YaST which you start either from CD, installation disk, or a reachable directory (depending on your installation). Then you set some general items on YaST and your Linux system. Requirements You have launched the + base Linux and YaST. 34 ____________________________________________________________________________________3.17.__S10:_Designing_partitions________* *______________________ Summary By providing YaST (YaST stands for Yet another Setup Tool which is a play on yacc yet another compiler compiler) S.u.S.E. offers you a means of easily installing the Linux system. You don't have to edit configuration files but rather decide what YaST should do for you. In most cases YaST "edits" those configuration files for you. Apart from such basic things as designing a proper keymap, YaST supports you while partitioning your disks, installing software and maintaining your system. Of course you can do everything by hand, but this is not the aim of this chapter. For special questions there are upcoming chapters and, of course, references to further reading. Here we only want to do a "first-time" Linux installation; we don't explore all of YaST's abilities, but merely a few. If you want to get to know YaST more thoroughly or you have special needs refer to the YaST chapter (chapter 4, page 67). Step by step This is what to do to start YaST: 1. You are asked whether you want to install demo mode. Demo mode is based on DOS' filesystem (see section 4.5.2, page 71 ) and is remarkably slower than a normal installation. We choose `normal', meaning just press _- . 2. Now you are back in YaST's main menu (see section 4.5.3, page 72). 3.17 S10: Designing partitions What's it about? This step shows you how to create partitions or to change the existing parti- tions to Linux partitions. Requirements You have started YaST and have completed the previous steps correctly. Summary Depending on your HD you will either reduce existing partitions in size or create new smaller partitions for your OS, buy a new HD or decided to use existing partitions for your Linux. Then you will allocate space for Linux. Therefore you create Linux partitions and add several specifications for your Linux file system on your HD. You can find information about partition types in the Infoblocks. After terminating YaST all partition types will be written to your partition section IB4, page 58 table. When writing partition tables you have to reboot your computer to let the changes take effect. Partition tables are, by the way, independent of the Linux system. There is only one on your HD and every OS installed takes its information from this table. Partition tables can be manipulated by any OS, and that's the reason why you have the system specific fdisk program. * * 35 3. Installation _________________________________________________________________________________________________________________* *_______________________________________@ If you already have an activated swap partition you should not change it in this step! How to partition is a very philosophic decision: how many partitions, how big and mounted to which directories etc. For further information please refer to Infoblock section IB5, page 58. Step by step This is how you assign your partitions: 1. In YaST choose `Adjustments of installation' and herein the item `Configure hard disk partitions' (see section 4.5.4, page 72). YaST presents a screen with different sections: At the top there are the parameters of your HD. Below that there might be warnings or error messages from the fdisk program. You can inspect those messages more thoroughly by pressing F6 . In the lower part are those partitions that fdisk has found on your system. Here you can see, for example, partitions of your former operating system (e. g. MS-DOS). If you already have assigned a swap partition this will be listed here as well. 2. If you want to delete existing partitions you should do that first. ________________________________________________________________________________________________________* *________ | If you delete a partition all data on this partition will be lost! * * ||| |_______________________________________________________________________________________________________* *________|||_ 3. Go to the partition you want to delete by using " or # respectively. Ensure that you are on the right partition; look at the partition type as well! Then you press F4 and confirm with _- . If you plan to delete multiple partitions you should do that now. 4. If you want to use existing partitions (another OS) for use with Linux you can do this by just changing the partition type. ________________________________________________________________________________________________________* *________ | If you change the partition type all data will be lost! * * ||| |_______________________________________________________________________________________________________* *________|||_ Go to the partition whose type you want to change using the " and # keys. If you're absolutely certain you don't need this partition, just press F3 . You will now see a selection window for the partition types. Choose between normal Linux partition or a "swap partition" and confirm by pressing _- . 5. If you want to assign new partitions to Linux (this probably is the "nor- mal" case) you should assign them one at a time. So just press F5 . If nothing happens you have no more space left on your device. You should remove unnecessary partitions first. A selection window ap- pears where you can decide what type of partition to create. Please se- lect `primary partition', `extended partition', or `logical partition'; confirm with _- . Please remember: you can assign no more than 4 primary partitions. If you need more than 4 partitions, the 36 _____________________________________________________________3.18.__S11:_Assigning_filesystems_for_partitions_______________* *______________________ last partition has to be assigned an extended partition; within this extended partition you can assign logical partitions. You'll find more information is to be found under section IB4, page 58. If you have assigned a primary or logical partition you have to specify the device name of the partition. YaST offers you all available free device- names. Just select the uppermost name (e. g. `/dev/hda2') and press _- . The next step allows you to set the size of your new partition. YaST offers the first free cylinder as start cylinder. Normally you just accept this offer by pressing _- . Now you enter the end of the partition. Here there are three possibilities: the number of the end cylinder (e. g. 976 ) number of cylinders for this partition (e. g. +66 ) or size in megabytes (e. g. +100MB ). Pressing _- leads you to `more'; please confirm with _- . Now there is the freshly created partition in the lower window. If this is not as you wanted it you can delete it easily, just press F4 . 6. One of your partitions has to be assigned a swap partition. If you have not done so already you can choose a suitable partition using " and # ( a reasonably good size for a swap partition is approximately 20 MB); please have a look at section IB4, page 58 for further information. Press F3 and choose `Linux-Swap-Partition' and confirm with _- . 7. If all partitions have been set (your HD should be completely used; hope you have not forgotten the swap partition) you move the cursor to `More' by pressing TAB . Pressing _- leads you back to the main menu. 3.18 S11: Assigning filesystems for partitions What's it about You have created all necessary partitions and these partitions have been added to the partition table (but they are still not written to disk!). There are some vital parameters still missing which will be given in this step. Requirements You have created the necessary partitions. Summary In this step you enter information for your Linux partitions (even for OS/2 partitions). This information is going to be written to /etc/fstab (this is the filesystem table). This file contains all constant values concerning your file system, such as device name, mount point in your Linux tree, or just the type of the partition. YaST needs this data for creating Linux filesystems on your freshly formatted partitions. Swap partitions remain untouched in this step since they have already been configured in section S10, page 35 and only have one degree of freedom. As opposed to the partition table, the steps performed in this section are just concerning Linux and do not concern other operating systems located on other partitions. * * 37 3. Installation _________________________________________________________________________________________________________________* *_______________________________________@ If you want to know more about these mechanisms see the various resources in the bibliography Also noteworthy: o In Linux all filesystems are linked together on one tree (look at figure C.1, page 397). You must declare each filesystem a branch of the whole tree; this is going to be its + Mountpoint. Even DOS or HPFS filesystems can be linked into your Linux tree. o Memory of filesystems is administrated under Linux by usage of + in- odes. An inode just points to the corresponding data in your files. The number of inodes is assigned while creating a filesystem. If you want to create lots of small files you will need many inodes (which of course need some space for their own as well); for filesystems containing many large files, less inodes are necessary. More about this in chapter section 4.6.2, page 76. Step by step This is how you proceed to assign filesystems to your newly created partitions Go to the submenu, item `Set target partitions / filesystems' under `Adjustments of installation': 1. First some information about what is to be done: o For your DOS HPFS filesystems (in DOS HPFS partitions) you can create a + Mountpoint. o For each of your Linux partitions: - you must select a mount point (if none is entered). - you can change the suggested + inodesize. - you can change the suggested formatting mode. o Functions `setting type' and `reading fstab' are not needed normally not needed. 2. If you want to access a DOS or HPFS partition under Linux you have to select it and press F4 . Now a window pops up and you can enter a directory. Under this directory you will later on find your DOS HPFS file system. e. g.enter /dosc (don't forget the slash!). Confirm this action by pressing _- . If you have selected a DOS partition another window appears; here you can choose which way to take for accessing your DOS under Linux (see section 4.6.1, page 76). Generally you want to use your DOS indepen- dently but just want to open a "window" under Linux for exchanging files. Please choose `DOS'. UNIX-file attributes and long filenames as offered by the + UMSDOS file system are only necessary for demo-mode. 3. Next you enter the + Mountpoint of your Linux partitions. Just select your Linux partitions one after the other and press F4 . A window will pop up letting you decide where to mount this partition on your filesystem tree. Give in the complete path beginning with the "slash" (/). Which partitions you give as mount point depends on how you plan to organize your partitions. In any case, you need a root-directory / which 38 ____________________________________________________________3.19.__S12:_Choosing_base_selection_of_software_________________* *______________________ is the "root" to the whole filesystem. All other filesystems and their mount points are more or less due to your preferences. You should create a mount point for each Linux filesystem. You will not be able to access unmounted filesystems. _______________________________________________________________________________________________________________________|| |||||||| Directories which are needed at boot-time must be directly under / ; at this ||| | |||||||| || | |||||||| stage the separate branches are not yet put together (mounted). That's the || | |||||||| reason why you are not allowed to assign /bin , /dev , /lib , /etc , and || | |||||||| || | |||||||| /sbin to other partitions! || | ||||||||_______________________________________________________________________________________________________________||_ 4. Now you set the i-node density for your Linux partitions. YaST sets a default value depending on the partition size. Generally: take an + i-node density of 4096 bytes per i-node with two exceptions: You have lots of little files (or you want to create them later on) or you want to use the live file system. In both the cases 1024 or 2048 are a better choice. Information on the Live-filesystem is in Infoblock section IB6, page 59. Now go ahead and choose those Linux file systems whose i-node density you want to change and press F5 . A window with all possible alterna- tives appears. Choose the one you want and press _- . 5. Now you should decide whether (and how) you want to partition your Linux partitions. Since we created those partitions newly in section S10, page 35 they have to be formatted. If you own a modern HD `normal formatting' must do; if your storage media is not that new perhaps you should choose `Formatting and checking'. Select the Linux partitions one after another press F6 to select the convenient formatting. Your partitions are not formatted instantly as your partitions were just designed. In this case YaST "remembers" this until the next boot. Then it will format your partitions accordingly, and the file systems will be created. 6. Select `More' to go back to YaST's main menu. 3.19 S12: Choosing base selection of software What's it about Your HD is now prepared. Now you can decide which software to install. First you will chose the software you want and then the actual installation will occur later, after rebooting. Requirements You have made it so far and have an empty disk on hand. If Linux doesn't support your CD-ROM you have already selected soft- ware packages (in section S2, page 24) and copied them onto your DOS HPFS partition. Furthermore you have selected section 3.16, page 34 in `Installation via HD partition' as installation source. * * 39 3. Installation _________________________________________________________________________________________________________________* *_______________________________________@ Summary Let's summarize: The partition table is changed but has not yet been written to disk and you have decided how your future file system is going to look. You cannot yet install software but it is quite convenient to select it now. You might find (while selecting packages) that your partitions are too small, too big or the relation is not the best. If this happened just go back to YaST(section S10, page 35 and section S11, page 37) and fix your settings until everything is to your satisfaction. The software you choose for installation has to be saved somewhere. Since there is no HD already formatted and installed this is not possible. Therefore you need a disk for saving your settings. This disk is going to be formatted and used for saving your settings. Now you have a whole CD full of compressed Linux software! What you choose is almost up to you, but there are certain packages that you need (series a). Series series base (a)is the "'base" of your Linux system. Other packages can be installed and removed as you like. Step by step Here's how to proceed to select your of base software installation: 1. In YaST select the item `Choose and install packages' and con- firm with _- . 2. YaST now asks for a disk. Insert a 3.5" disk into floppy drive A:, which, of course, should not contain any important data. If you don't insert a disk you are not able to save your settings, thus having to repeat the selection of packages later on once again. If the disk is inserted confirm with _- . YaST now creates a filesystem on this disk; if necessary the disk is formatted first. Now you've got to be patient. Data has to be read from CD (or from HD if you've chosen this installation media). Afterwards there will be the installation menu (section 4.6.5, page 77). 3. Now you select `Change /create configuration'. 4. Now you are in series selection (section 4.6.8, page 78). In the uppermost window all series are listed with a short description. On the right hand side there is the amount in megabytes that you've already selected. 5. Now you can go through your series and packages, choose packages you would like to have, but have an eye on your free disk space. Don't forget: you urgently need every package from series a. How selection of series and packages work is described below. 6. What can I do in series selection? o Pressing _- leads you to package selection within the chosen series. o In the lower window you can view how your Linux partition gets filled. o Using F3 YaST shows you your partition in a large window. 40 ______________________________________________________________________3.20.__S13:_Quitting_YaST_and_rebooting______________* *_______________________ o F5 lets YaST check for dependencies of the selected packages. De- pendencies take into consideration that certain packages require other packages to be installed, and sometimes packages conflict with each other. o F10 brings you back to the installation menu. 7. Now choose a series that you are interested in. Press _- and the package selection list of that series appears (section 4.6.8, page 78). In the upper left corner of the screen there is a list of the packages, in the upper right corner there is the free space left on device and below a description of the actual selected package. 8. What can I do in package selection? o _- selects or unselects the highlighted package respectively. [X] means that this package is going to be installed. o F2 shows a detailed description of the respective package, _- leads you back to package selection. o F3 gives a detailed view of your partition's free space. o F10 leads you back to package selection. o In the upper righthand corner you can check your partition's space. 9. If you're done with series and packages and everything is selected as you like (and everything fits on your HD) then you can switch back to series selection with F10 and, by pressing it a second time you will be brought back to the installation menu. 10. The menu entry `What if' shows you what is going to happen (which packages are going to be installed). You don't have to select this menu entry, it's just for informational purpose. This item is very useful if you plan (later on) to remove packages to make sure nothing unwanted is about to be removed. 11. Now your selection has to be written to disk. Therefore you just select `Save configuration'. In the box that appears you can now enter a name for your configuration. The entry below serves that as a simple description entry; here you can enter a short explanation of your config- uration. This might come in handy if you plan to use this configuration later on. Go to `Save configuration' using TAB and press _- . Your selection is now written to disk. 12. Next you go back into YaST's main menu by choosing `Main menu' and pressing _- . If you hit `Start installation' you will be shown an error box since this is not possible yet. 3.20 S13: Quitting YaST and rebooting What's it about Anything that could be done without a usable hard disk has been done. Now you have to shut down and reboot your Linux system for making the Linux partitions available. * * 41 3. Installation _________________________________________________________________________________________________________________* *_______________________________________@ Requirements If you have made it that far all requirements are fulfilled. Summary We have already mentioned that your HD is not ready to be used yet. By terminating YaST you write this new partition table to HD. Now you have to reboot immediately! You don't have to hurry but you should not start any other programs before having rebooted. You shut down Linux with the fol- lowing command: shutdown -r now . Don't just switch off your machine! Now your machine should boot and you have to proceed exactly as you did before (booting via CD-ROM section S6, page 30 or boot disks section 3.12, page 29) After starting from CD lilo is already installed but you cannot use it right now (nor the batch file linux.bat ); you have to start via setup on CD 1 again. Step by step That's how to proceed to terminate YaST and to reboot: 1. Choose `Finish installation and exit YaST' and confirm with _- . Now there is a message that your partition table will be written; Eventually YaST reminds you of rebooting your machine immediately. Confirm this with _- . Now you should see the system prompt. 2. Enter shutdown -r now and press _- . After a couple of seconds booting commences. 3. You have to boot just like you did at the first time. If you boot via disk insert the boot disk (just like first-time: either S.u.S.E. boot disk or your self created boot disk). If you boot using two disks you don't need to create installation-disk since you already have it. If you have booted via loadlin remove your disk from the floppy drive. After booting has finished start the program setup from CD. You have to go through the whole program again; the only thing that differs is that you don't have to install loadlin again since this has already been done in section S6, page 30. Here is a quick reference of what you have to enter in setup . If you encounter problems please go back to section S4, page 27 and section S6, page 30 o Choose `English' (or German if you prefer...) o Enter your CD-ROM drive (letter). o Confirm the welcome screen with ENTER o Choose `loadlin'. o Now select `CD' or `HD' just like in section S6, page 30. o Select a suitable kernel like in section S6, page 30. o Select your CD-ROM drive just as in section S6, page 30. o Select `Loading Linux'. o Now you have a Linux prompt again; ready for logging in. 42 _______________________________________________3.21.__S14:_Starting_YaST_again_and_installing_software______________________* *______________________ 3.21 S14: Starting YaST again and installing software What's it about Now we are up to formatting the partitions and assigning file systems. There- after Linux software will be installed. Requirements You have booted and have got a Linux prompt on your screen, plus you have your disk at hand on which your software selections have been saved. Summary Now your partition table has been written but all freshly assigned partitions have not been formatted yet. They don't even contain any filesystems yet. If you login "RAM Linux" runs again. After starting YaST first your partitions are formatted, then filesystems as- signed. Next your swap partition is activated. Next you can start installing your selected software packages. Step by step Here's how to start YaST and to install software packages: 1. You now have a Linux prompt; log in using user name `root'. 2. Now there is the "welcome screen" with lots of good hints and tricks; since you've already read them previously you can skip it. 3. Enter yast and YaST starts. 4. Now you have to answer some questions you already know. Here is a quick overview: o Choose `English' (or German if you prefer). o Now you have a disk containing your software selection (you'll re- member we created it in section S12, page 39). Insert this disk and confirm with `Yes'. o Now YaST takes care of the freshly created partitions. Select `Formatting' and watch YaST formatting and assigning filesystems on your drives. o Confirm your swap partition by entering _- . Then your swap parti- tion is activated. o Choose `Normal' if the system asks you whether to install in demo mode or normal mode. 5. Now you are taken back to YaST's main menu. You can now finally install your selected software. You have saved your software selection to disk. Select `Setting up installation/starting' and wait a while until all descriptions have been read. Then you will see the installation menu. * * 43 3. Installation _________________________________________________________________________________________________________________* *_______________________________________@ 6. Now you choose `Load configuration'; a window with a couple of configurations appears. Most of them have been predefined by S.u.S.E.. Your own configuration is located at the end of the list. Select it (or do you want to try one of S.u.S.E.'s configurations?) and confirm loading. After a while you will see the installation menu again. Please notify that your configuration is now set at the top line. 7. If you like you can now adapt your software configuration to your per- sonal needs. This is achieved easily by moving to `Changing/creating configuration' and confirming it using _- . This changed configu- ration of course can be saved to disk using `Saving configuration'. This task has already been described in section S12, page 39. 8. Now you start installation with `Starting installation'. YaST presents you with a list of packages that are to be installed and which packages already have been installed. 9. Go back to main menu by selecting `Main menu'. 10. Select `Finishing installation and leaving YaST' 11. Select a kernel. 12. Now you have to confirm to create a boot disk. Since your disk containing the software installation is still in your drive YaST takes this and turns it into a boot-disk. 13. Next you'll see the question whether you want to configure lilo; this ques- tion is to be answered with `No' since this is explained in system man- agement later on. 14. YaST now informs you that you have to boot your machine afterwards; you should confirm this. 15. Now you have the Linux prompt back. Enter shutdown -r now for starting your system properly. Problems There is one problem that can occur: o The error message "mke2fs failed" or similar messages might be dis- played on the screen; see section P10, page 55 3.22 S15: Starting YaST for the last time and finishing up installation What's it about This step is for customizing your Linux system. Requirements You have installed the base software packages. 44 _______________________3.22.__S15:_Starting_YaST_for_the_last_time_and_finishing_up_installation___________________________* *_______________________ Summary The machine does not reboot, but continues with loading your installed Linux. YaST starts automatically. YaST now asks you a couple of questions for configuring additional devices as mice etc. You do not have to configure these things right now since you can invoke any time and configure your system. If you are unsure just skip this item and do it later on. Since access to the CD now is possible YaST may install even software that is not on CD 1. Step by step Here's how to customize your Linux system: 1. YaST now requests a CD change and automatically installs the selected software that has not been installed yet. 2. YaST now asks whether you want to test some screen fonts. If you do so, just answer `yes'. Choose the font you want from the list and select `test' for testing a font, or `cancel' if you want to stop testing. 3. If you have selected `test' the screen switches to the new font. Check if all special keys (such as German umlauts if you need them) are accessible! If you do not like the font just select `Other font' and the game starts all over again. If you decide to keep it, select `Standardfont' to keep it. 4. YaST now enquires if you want to configure your modem. If there is a modem installed you can do so right now or come back here later on. If you say `yes' you are asked for the serial device of the modem. 5. YaST now wants to know if you want to configure your mouse, If you want to do so right now select `yes'. Select your mouse-type. If it is a serial mouse YaST needs to know the serial interface where it is connected. 6. Please confirm the start of gpm at boot-up. If you encounter problems with it just come back here and remove it. 7. Now enter a time-zone. Here again there are yards of entries. 8. YaST now creates a login for you since you should not work with the sys- tem as `root' but with your own login. Think of a nice little nickname (without spaces). A password can be entered later on when you log in for the first time. This also applies to `root'. 9. Now you are back to YaST's main menu. You have done it! 10. Now you can select any menu item. Most of them should already look familiar to you. You may as well leave YaST and log in using your new user name. By the way there are far more consoles than one. Press Alt + F2 to Alt + F6 for switching consoles. 11. If you have logged in using your new user account you should enter a password by giving passwd . Then you are asked for a password which has to be entered twice. 12. Now you may work. You can enter (e. g. ) ls -a for listing all files in your home directory. * * 45 3. Installation _________________________________________________________________________________________________________________* *_______________________________________@ 3.23 E4: How do you plan to start your Linux in future? What's it about Your Linux system is now almost completely installed. Now you can decide how to start your Linux on a regular basis. Summary In principle there are three ways of starting your Linux: 1. You start Linux via loadlin . This is the most comfortable method but requires certain steps to be taken. o Your machine must run in real-mode under MS-DOS, or in virtual 8086-mode4 with an VCPI server available (in other words: under Unix, OS/2 or DOS-box of Windows 95 this does not work). It works fine under DOS-prompt or in Windows 95 DOS-mode. o You must have enough memory below 640 kB. Here you can encounter problems if too many drivers are loaded. 2. You start Linux via boot-disk. This always works and needs no work to be done, as we have created this boot-disk in section S14, page 42; of course you can use S.u.S.E. boot disk as well. Boot disk also is a rather good alternative if you have problems using the other two methods. 3. The most advanced method, of course, is use of the boot-loader LILO. LILO lets you decide which system to boot (before the system actually boots). Configuration of lilo is a little tricky. It has to be installed in the boot sector of your HD, which means a couple of problems can occur. Therefore configuration and installation of lilo is discussed in a chapter of its own and not here in the installation section. Recommendation The best thing to do if you're working with DOS, Windows 3.x, or Win- dows 95 is the "loadlin" method. This will work in almost any case. To- gether with OS/2 or other operating systems working in protected-mode, or if loadlin doesn't work use the boot-disk. 3.24 S16: Starting via loadlin What's it about If you plan to use loadlin for starting Linux there are a couple of things you have to prepare in advance. Depending on your system there are some files which have to be adapted. Requirements You want and are able to use loadlin (at least you hope so.;)). _____________________________________________4 offered e. g.by emm386.exe 46 _____________________________________________________________________________________3.24.__S16:_Starting_via_loadlin_______* *______________________ Summary In principle there are two ways of activating loadlin. One is to switch between various systems via boot menu or to start Linux via loadlin from another running system. Both methods have advantages and disadvantages: o A boot menu saves you the detour via another operating system. o You can add other configurations to your boot menu for creating an uni- versal starting mechanism. o A boot menu is the more flexible way and you will be able to solve prob- lems if they occur. o You have to modify start files to build a boot menu which you will have to do by trial & error. o Changing to Linux from the DOS prompt is very simple. o Linux start can be very nicely integrated into a Windows 95 session. Dou- ble clicking an icon will start Linux (Windows 95 contains DOS 7.0). Recommendation You should use a boot menu from DOS or Windows 3.x. If using Windows 95 you can start from the running system. Start menus under Windows 95 are a very complex matter so please under- stand that we are only able to give you some hints on them. Necessary steps for all loadlin users This is what you have to do whether you decide to use a boot menu or if you start from a running system (under DOS, Windows 3.x, or Windows 95): 1. Probably you already installed loadlin (this was done in section S6, page 30); if not, you have to do so via setup . 2. Change to c:\loadlin (under MS-DOS). There you will find a file called linux.par . Create a file named startlin.bat (or any other name). Now insert the following line as described in file contents 3.24.1, page 47: _______________________________________________________________________________________________________________________|| |||||||| | |||||||| c:\loadlin\loadlin @@c:\loadlin\linux.par | |||||||| | ||||||||_______________________________________________________________________________________________________________ | File contents 3.24.1: Example of a batch file for starting Linux Now, please edit the file linux.par (file contents 3.24.2, page 48). Instead of xxx you have to enter your root partition's device name (you wrote down this name in section S11, page 37). startlin.bat starts your Linux. The file linux.par is used by startlin.bat as well as by config.sys and contains essential parameters. Later on, after getting more familiar with Linux you can of course add/replace param- eters in here. If you have built a kernel of your own just copy it to c:\loadlin\vmlinuz and it will be booted from now on. * * 47 3. Installation _________________________________________________________________________________________________________________* *_______________________________________@ ________________________________________________________________________________________________________* *________|| | * * | | c:\loadlin\vmlinuz # first value must be * * | | * * | | # the filename of the Linux-kernel * * | | * * | | * * | | root=/dev/xxx # the device which gets mounted as root FS * * | | * * | | * * | | ro # mount root read-only * * | | * * | |_______________________________________________________________________________________________________* *________ | File contents 3.24.2: Example of the file linux.par with customized values Using a boot menu if you have DOS or Windows 3.x Here's how you configure a boot menu under DOS or Windows 3.x respec- tively: 1. First you have to define a boot menu in c:\config.sys . Enter something like file contents 3.24.3, page 48 ________________________________________________________________________________________________________* *________|| | * * | | [Menu] * * | | menuitem=Win, starting Windows,... * * | | * * | | menuitem=DOS, starting DOS, MS-DOS... * * | | menuitem=Linux, starting Linux... * * | | * * | | menucolor=15,1 * * | | menudefault=Win,5 * * | | * * | |_______________________________________________________________________________________________________* *________ | File contents 3.24.3: Example for config.sys (first part) for the Linux start Under the label [Menu] you define entries in the boot menu, its color and after how many seconds which item is started automatically. 2. Below you enter the labels [Common], [Win], [DOS], and [Linux]. Things written in [Common] are always executed, all other entries are system specific. Use the lines that are in your current config.sys as our ex- ample certainly does not suit everyone's (Diagram file contents 3.24.4, page 49). Now save the file. 3. Now open c:\autoexec.bat . Here you have to put exactly the same labels and assign entries to labels. Notation differs slightly. Linux case is mentioned first. Which label has been selected is to be found in vari- able %config%. Perhaps you write something similar to diagram file con- tents 3.24.5, page 50. 4. If you boot your machine now the boot-menu appears and you have five seconds to choose an operating system. After 5 seconds Windows starts automatically. If you select `Linux' Linux starts and awaits your login. Starting Linux from within Windows 95 Here's how you create a start icon for Linux for booting Linux from within a running Windows 95 session: 48 _____________________________________________________________________________________3.24.__S16:_Starting_via_loadlin_______* *______________________ |||||||| | _______________________________________________________________________________________________________________________|||| |||||||| | ||||||||| [Common]device=c:\dos\himem.sys /testmem:off || |||||||| | |||||||| device=c:\dos\emm386.exe noems I=E000-F4FF | |||||||| dos=high,umb | |||||||| | |||||||| files=30 | |||||||| buffers=10 | |||||||| | |||||||| shell=c:\dos\command.com | |||||||| | |||||||| [Win] | |||||||| | |||||||| devicehigh=c:\dos\dblspace.sys /move | |||||||| devicehigh=c:\cd\slcd.sys /D:SONY_000 /B:340 /M:P /V /C | |||||||| | |||||||| | |||||||| [DOS] | |||||||| | |||||||| devicehigh=c:\dos\dblspace.sys /move | |||||||| devicehigh=c:\cd\slcd.sys /D:SONY_000 /B:340 /M:P /V /C | |||||||| | |||||||| | |||||||| [Linux] | |||||||| | |||||||| shell=c:\loadlin\loadlin.exe @@c:\loadlin\linux.par | |||||||| | |||||||| [Common] | |||||||| | |||||||| rem Remains blank | |||||||| | ||||||||_______________________________________________________________________________________________________________ | File contents 3.24.4: Example for config.sys (second part) for the Linux start 1. Open the explorer change to c:\loadlin , drag the file startlin.bat onto your desktop. 2. Go to a file or onto the desktop (depending on where you want your Linux icon) and press the right mouse-button and select `Create link'. 3. Mark the link just entered, press the right mouse-button and select `Properties'. Go to register `programs' click on the button `extended'. Cross entry `MS-mode'. Confirm with `OK'. 4. Via `Change icon' you can select a nice icon; give the link a suitable name; Voila! 5. Double clicking this new symbol should bring up a mask telling you that Windows 95 is about to switch to DOS-mode. If this mask bothers you: turn it off in the properties menu. An alternative: a boot menu with Windows 95 Here's how to proceed to install a boot menu for Windows 95: 1. First you have to edit c:\msdos.sys Make the file visible by entering: C:> attrib -R -S -H c:\msdos.sys This is a text file where you have to enter some lines for deactivating Windows 95' start menu. Label [Options] should resemble file con- tents 3.24.6, page 50 The parameter Logo=0 is optional and avoids switching to graphics mode before Windows 95 is started; booting is much faster and you avoid lots of trouble if you plan to use the DOS emulator on Linux later on. * * 49 3. Installation _________________________________________________________________________________________________________________* *_______________________________________@ || * * || ________________________________________________________________________________________________________* *________|||| | * * | | @@echo off * * | | * * | | * * | | rem Entries for all Configurations * * | | switches= /f * * | | * * | | set comspec=c:\dos\command.com * * | | prompt $p$g * * | | * * | | loadhigh c:\dos\keyb gr,,c:\dos\keyboard.sys * * | | loadhigh c:\dos\doskey * * | | set temp=c:\temp * * | | * * | | loadhigh c:\dos\mscdex.exe /D:SONY_000 /E /V /L:H * * | | c:\logimaus\mouse.exe * * | | * * | | * * | | goto %config% * * | | * * | | * * | | :Win * * | | * * | | c:\dos\smartdrv.exe a- b- c+ 2048 1024 * * | | path c:.;d:.;c:\windows;c:\dos;c:\util; * * | | * * | | win : * * | | c:\dos\smartdrv /C * * | | goto ende * * | | * * | | * * | | :DOS * * | | * * | | path c:.;d:.;c:\dos;c:\util; * * | | goto ende * * | | * * | | * * | | :ende * * | | * * | | echo * Goodbye * * * | | * * | |_______________________________________________________________________________________________________* *________ | File contents 3.24.5: Example for autoexec.bat for Linux start ________________________________________________________________________________________________________* *________|| | * * | | [Options] * * | | BootGUI=0 * * | | * * | | BootDelay=0 * * | | BootMenu=0 * * | | * * | | Logo=0 * * | | * * | |_______________________________________________________________________________________________________* *________ | File contents 3.24.6: msdos.sys for Linux start 50 _____________________________________________________________________________________3.24.__S16:_Starting_via_loadlin_______* *______________________ The parameter BootGUI=0 is for booting Windows 95 directly into DOS mode. For starting Windows out of this you have to enter: C:> win but this is already done by our example c:\autoexec.bat if you have selected Win95 from our menu. 2. Then you have to define your own boot menu in c:\config.sys . There- fore you should enter (diagram file contents 3.24.7, page 51): _______________________________________________________________________________________________________________________|| |||||||| | |||||||| [Menu] | |||||||| menuitem=Win95, start Windows 95... | |||||||| | |||||||| menuitem=DOS, start MS-DOS... | |||||||| menuitem=Linux, start Linux... | |||||||| | |||||||| menudefault=Win95,5 | |||||||| | ||||||||_______________________________________________________________________________________________________________ | File contents 3.24.7: Example config.sys (part one) for starting Linux un- der Windows 95 Under the Label [Menu] you define entries of the boot menu, its color and the delay for starting up automatically. 3. Further down you enter labels [Win95] , [DOS] , [Linux] , and [Common] . Common is for entries that apply every time (this will very rarely be the case under Windows 95); all other labels are for the corresponding sys- tem. Use those lines that are already written in your config.sys ; the example in our diagram file contents 3.24.8, page 51 should only be re- garded|as|a hint. | _______________________________________________________________________________________________________________________|||| |||||||| | |||||||| [Win95] | |||||||| dos=high,umb | |||||||| | |||||||| device=c:\windows\himem.sys /testmem:off | |||||||| | |||||||| | |||||||| [DOS] device=c:\plugplay\drivers\dos\dwcfgmg.sys | |||||||| dos=high,umb | |||||||| | |||||||| device=c:\windows\himem.sys /testmem:off | |||||||| device=c:\windows\emm386.exe noems I=B000-B7FF | |||||||| | |||||||| devicehigh=c:\cdrom\torisan.sys /D:TSYCD3 /P:SM | |||||||| | |||||||| | |||||||| [Linux] | |||||||| shell=c:\loadlin\loadlin.exe @@c:\loadlin\linux.par | |||||||| | |||||||| | |||||||| [Common] | |||||||| accdate=C+ D+ H+ | |||||||| | |||||||| switches= /F buffers=20 | |||||||| | ||||||||_______________________________________________________________________________________________________________ | File contents 3.24.8: Example config.sys (part two) for starting Linux- Start under Windows 95 Now save this file. 4. Now open c:\autoexec.bat . Here you have to enter exactly the same labels. Notation slightly differs. Which label has been selected is written * * 51 3. Installation _________________________________________________________________________________________________________________* *_______________________________________@ to variable %config%. Perhaps you should write something similar to file | contents 3.24.9, page 52 * * | ________________________________________________________________________________________________________* *________|||| | * * | | @@echo off * * | | loadhigh keyb gr,,c:\windows\command\keyboard.sys * * | | * * | | goto %config% * * | | * * | | :Win95 * * | | * * | | win * * | | goto ende * * | | * * | | * * | | :DOS * * | | * * | | path c:.;d:.;c:\windows\command;c:\util; * * | | loadhigh c:\windows\command\mscdex.exe /D:TSYCD3 /L:x * * | | * * | | loadhigh c:\windows\command\doskey * * | | c:\windows\command\mouse.exe * * | | * * | | goto ende * * | | * * | | :ende * * | | * * | | echo * And now? * * * | | * * | |_______________________________________________________________________________________________________* *________ | File contents 3.24.9: Example autoexec.bat for starting Linux under Win- dows 95 5. If you now boot your machine there should be the Windows 95 boot menu should appear, giving you two seconds to select an entry. If you choose Command line your own boot menu will be displayed. Now you have five seconds to select a system. After this delay Windows 95 starts automatically. If you select `Linux', Linux starts and awaits your login. 3.25 After your system is installed Congratulations! We have achieved a lot! You have a working Linux with lots of software installed on your HD. Don't be afraid, it's not going to be dull now. As the system administrator there still are lots of tasks to perform. Most of it is explained in our S.u.S.E. Linux; more help is available on the CD-ROMs. Here are some forthcoming topics: o Installing the X-Window System o Installing LILO (if you want to boot via LILO) o Building a kernel of your own o Installing aps filter (it prints almost everything) o Connecting to a LAN o Mail o Connecting to the internet o Configuring the DOS-emulator 52 ___________________________________________________________________________________________3.26.__Problem_description_______* *______________________ 3.26 Problem description 3.26.1 P1: Files cannot be moved Files with the system or hidden attribute set can't be moved by defragmen- tation programs. If you want to find out where the problem lies just type: attrib n*.* /s > Now you can see a whole list of your HD in . Here you can identify the files that cause problems and change them using: attrib -S -H Here you should proceed with care not to delete copy-protection files, per- manent swap files, or other system specific files. After defragmentation you should undo your changes of attributes. If this doesn't work, then you really have to repartition your HD. This means that you have to save all your important files for restoring them after you have finished repartitioning your drive. The alternative is, of course, buying a new HD. Since prices are going down steadily this might be a good solution. 3.26.2 P2: No English keymaps in DOS-mode In DOS-mode you enter the following line: oloadhigh keyb gr,,c:\windows\command\keyboard.sys or you simply add this line to your autoexec.bat . If your Windows is located elsewhere then of course you have to adjust the path accordingly. 3.26.3 P3: No CD-ROM driver in DOS-mode In MS-DOS mode only those drivers are available that have been loaded in config.sys and autoexec.bat . It makes sense not to put your CD-ROM driver into those files since Windows 95 has its own drivers. For using those drivers in DOS mode you have to create a link to an MS-DOS command line and insert properties `program' under `extended' to create own startup files. There you have to enter your drivers for using those drivers in a DOS box. 3.26.4 P4: CD is damaged This possibility is extremely unusual but who knows, it can happen. 3.26.5 P5: ATAPI-CD-ROM hangs while reading If your + ATAPI CD-ROM is not recognized or it hangs while reading this mostly is due to incorrectly connected hardware. Generally all devices should be connected to the EIDE controller in the correct order. The first device is master on the first controller, the second device is slave on first controller. The third device should be master on the second controller and so forth. Now it often occurs that there is only a CD-ROM besides the first device. The CD-ROM drive sometimes is connected as master to the second con- troller (secondary IDE controller). This is wrong and can cause Linux not to * * 53 3. Installation _________________________________________________________________________________________________________________* *_______________________________________@ know what to do with this "gap". You can try to fix this by passing the ap- propriate parameter to the kernel (hdc=cdrom ) (see section 7.3.2, page 102). Nevertheless connecting both devices to the same controller is much better. Sometimes one of the devices is just "mis-jumpered". This means it is jumpered as slave but is connected as master or vice versa. In case of doubt just check your hardware settings and correct them. Finally there exist a couple of faulty EIDE-chipsets, most of which have been recognized and the code of the kernel has been built to handle them. There is a special kernel for these cases. The parameters are described more thoroughly in section 7.3.2, page 102 and chapter 13, page 181. 3.26.6 P6: Mitsumi driver announces: mcd: init failed This does not be have to be anything serious, provided you have a Mitsumi CD-ROM drive and it is not an + ATAPI device. o There is no Mitsumi CD-ROM drive in your machine but your kernel contains the Mitsumi driver. This is alright. o You have a Mitsumi ATAPI drive installed (e. g. FX001DE, FX200, FX300, FX400). You should have used a kernel with EIDE drivers not a kernel with the mcd driver. o You really have a Mitsumi CD-ROM that's not an ATAPI CD-ROM (e. g. Lu005, Lu005S, FX001, FX001S, FX001D). Then your drive is not set to the default values 0x300, 10. You should give the correct parameters to the kernel at boot time. You can assist your driver using the following line: mcd=,[,] is the hexadecimal port address (e. g. 0x300) stands for the used interrupt. The last parameter can be omitted first (see section P7, page 54). More information is to be found in section 7.3.2, page 102. 3.26.7 P7: "Unable to mount root" using Mitsumi CD-ROM-drives (Not ATAPI-drives) This problem may occur while installing via a Mitsumi (non-ATAPI) drive. The kernel recognizes the drive but claims due to timing problems: "No Disk in Drive", although the CD is inserted. Most of the times, but unfortunately not always it helps setting the last parameter to 5. mcd=,,5 . stands for hexadecimal port address (e. g. 0x300), stands for used interrupt More information about kernel parameters is to be found under section IB2, page 57 and section 3.26.6, page 54. If this all doesn't work you have to install via disks. You'll have to go back to section S4, page 27 and select disks in section E3, page 28. 54 ___________________________________________________________________________________________3.26.__Problem_description_______* *______________________ Our experience has shown that this generally works. Obviously the DOS CD-ROM driver puts the drive into a wrong state, thus making it impossible for Linux to recognize the drive correctly. Therefore you should reset your machine before trying to install via disks or switch it off for a moment. 3.26.8 P8: Loadlin doesn't have enough memory to load the kernel You don't have enough free memory below 640 kB. Try to remove drivers from your startup files or shift them to high memory. If you use compressed drives under Windows 95 and shifting the driver to high memory doesn't work you have to decompress those drives. 3.26.9 P9: Loadlin doesn't work If you encounter any problems using setup or loadlin you should start setup like this: setup -DLOADLIN__TEST Now use the button `Run loadlin -t' for creating an diagnostic file: c:\loadlin\susesave.dbg . Please send this file to technical support. 3.26.10 P10: Error with mke2fs This annoying problem rarely occurs. Unfortunately it is only to be solved by moving the end of the partition forwards or backwards one cylinder. Of course the beginning of the next partition has to be adjusted accordingly. 3.26.11 P11: DOS runs in protected mode loadlin can only boot the kernel if the machine isn't running in protected mode, or if a VCPI server is running. If you're working under Windows 95 you have to switch to MS-DOS-mode. o Either you go via the `Start'-Button to `Shut down', then `Restart computer in MS-DOS-mode' or o you create a link to the MS-DOS command line and change the properties accordingly: In register `program' you select `extended' and mark with an "X" on `MS-DOS-mode'. If you now restart your machine it will be placed in MS-DOS mode. 3.26.12 P12: The 3,5 floppy drive is connected as B: and not bootable PCs can only be booted from the first physical floppy drive (this is A: under MS-DOS). Moreover this drive has to be marked active in BIOS otherwise you can only boot via HD. If you have accessed your 3.5 floppy as "B:" you have to change the connec- tions of those drives (5.25 and 3.5): o Turn off your machine and open the case. * * 55 3. Installation _________________________________________________________________________________________________________________* *_______________________________________@ o Look for the broad data cable which connects floppy drives to the con- troller. o There normally are two pairs of plugs (one for each floppy). Only one plug from each pair is used (3.5 and 5.25 have different plugs). You now have to exchange both connections using the appropriate plugs. If this doesn't work due to short cables you have to swap the drives' positions as well. o If your cable only has two single plugs (no pairs) you have to buy a new data cable. o Close the case again and go to the setup-menu after switching it on. How you can access setup depends on your BIOS. A "nice" BIOS gives in- structions on screen. If not look int your manual or try the Del key. o Change position of your floppy drives in the setup menu: Drive A: 1.44 MB, 3,5-floppy Drive B: 1.2 MB, 5,25-floppy. o Now mark this floppy drive as active. Here you have to enter submenu `Advanced C-MOS setup' (or similar). You should see a line `System boot-up sequence'; here you designate A:as the first drive in the boot sequence. o Save these settings and leave setup. 3.26.13 P13: Label of CD-ROM drive has changed If you have created an additional partition using fips this is a DOS-partition. Therefore all other drives are shifted back one step and your CD-ROM is no longer D: but E: . After changing partition types with YaST this changes again; your CD-ROM drive should now be D: . If you are not able to open the Windows explorer or desktop under Win- dows 95 this means that it still tries to access `E:' (using the above example). You have to use the system management tool to change your CD-ROM label to `D:'. 3.27 Infoblocks 3.27.1 IB1: Selecting a kernel As installation proceeds you have to select a kernel for driving your system during installation and perhaps afterwards as well. The Linux kernel contains drivers for certain drives (networks, HDs, CD-ROM drives). The kernel must be the right one for your machine. On our CD there are lots of different kernels either as disk images for boot disks or as files for loadlin. Those kernels are temporary since they must support a lot of hardware but can't be too large. Later on you should build your own kernel, specifically adapted to your system. In most cases you can use our supplied boot disk for installation. Only if this kernel supplied with our boot disk doesn't recognize your hardware properly 56 ___________________________________________________________________________________________________________3.27.__Infoblocks* *______________________ will you have to try another kernel. You can get more information about kernel configuration online via setup or in files /disks or /suse/images on CD 1, respectively. For covering almost any hardware configuration there is a modular kernel in- stalled on your system which contains only the absolutely essential drivers necessary for booting. So there are two kernels for EIDE-machines and sev- eral for SCSI-machines. All other drivers are loaded by the kernel daemon as modules. If you encounter problems make sure your hardware is correctly specified in /etc/conf.modules . The other option is to build a monolithic kernel which includes all drivers (some driver still cause problems if loaded as modules). or you build a monolithic kernel (some drivers still cause problems if loaded as modules). 3.27.2 IB2: Kernel parameters If you have chosen a suitable kernel you have to decide which kernel param- eters are needed. Kernel parameters are your way of "talking" to the kernels drivers. Using kernel parameters you can achieve several things: o You tell the kernel which hardware you use and how it is accessed (ad- dress, interrupt etc.). Sometimes the driver can find out by itself but it is much safer and faster if you do this manually. o You can disable drivers. This might be helpful if there are several CD- ROM drivers provided in the kernel and each of them tests its hardware in a very complicated manner (auto-probing). This only is relevant if you use a monolithic kernel which includes all CD-ROM drivers. o In case of problems you can probably achieve a running Linux by using specific kernel parameters. You can enter kernel parameters either if you boot via loadlin or if you boot via lilo . Even with S.u.S.E.boot disk which supplies the relevant drivers as modules you can enter kernel parameters. In this case the notation of kernel parameters differs slightly. You have to enter those parameters every time you boot. The kernel doesn't remember. After installation you can put kernel parameters into /etc/lilo.conf or /etc/conf.modules (if you use a modularized kernel) respectively where they are read automatically by lilo or modprobe , respectively. More about parameters, which other parameters exist and how you have to enter them is explained in chapter section 7.3.2, page 102. 3.27.3 IB3: Does Linux support my CD-ROM? Generally it can said that almost every CD-ROM drive is now supported by Linux. o Using + ATAPI-drives (those drives connected to an EIDE-controller) there should be no problems at all. * * 57 3. Installation _________________________________________________________________________________________________________________* *_______________________________________@ o Using SCSI CD-ROM drives it is only important whether the SCSI con- troller is supported by Linux (supported SCSI controllers are listed in an appendix). If your SCSI controller is not supported and there is a HD connected to it you will have a problem anyway. o Lots of proprietary CD-ROM drives are supported under Linux (see Ap- pendix). Here it is most likely that problems occur. If your drive is not mentioned you could try with a similar type. 3.27.4 IB4: Partition types on PC Each HD contains a partition table which has space for four entries. Each entry can either be a primary partition or an extended partition where only one can be an extended partition. Primary partitions are very simple; they are a continuous range of cylinders which are assigned to an OS. Using primary partitions you can only assign four partitions per HD. More won't fit into the partition table. Here the extended partition principle comes into use. The extended partition is one that can be subdivided into several logical drives which don't need a separate entry in the partition table. The extended partition is more or less a container which contains all logical partitions. If you need more than four partitions you can simply make the last partition an extended partition and give it the remaining free cylinders (space). Here you can now create as many logical partitions as you like. Linux doesn't care where it is installed (primary and/or logical). 3.27.5 IB5: Partitions for the Linux system The philosophy of Unix file systems (therefore Linux as well) differs from those used by MS-DOS file systems. There are no different drives but only one directory tree. Its branches can be on different drives but this is usually not seen by the user. Division into several partitions includes advantages and disadvantages. Be- cause partitions are independent of each other, possible problems are re- stricted to the respective partition itself. But since the partition sizes are fixed, your system can't "breathe". If the partition is full you need to repartition or to add another partition to your tree. How many partitions should you create? Minimum should be three partitions: o A root partition which contains the system and all other commonly used files. This is the root of your tree and is designated by / . o A home partition which contains all user-specific data. This one is added (mounted) as /home . o A swap partition which is not included in the file system tree. This is used by the system to enlarge available RAM. Of course you can divide up your system into more partitions. It is recom- mended that you make the root partition rather small for minimizing access to this HD and thus improving system security. 58 ___________________________________________________________________________________________________________3.27.__Infoblocks* *______________________ In case of doubt or if you are unsure of what to do and you had a DOS or Windows 95 partition you should add three primary partitions. 3.27.6 IB6: The live file system Our feature live file system integrates programs from CD (labeled live system into an existing Linux directory tree. YaST creates symbolic links which point to programs on CD. So you can use all programs from CD without having to install them and thus minimizing necessary HD space. The advantage of the live file system is its minimal space requirements. But there are disadvantages as well: o Accessing a live file system takes longer since accessing CD-ROMs is much much slower than accessing HDs. o Since many symbolic links have to be created and each of them uses one inode you must have enough free + Inodes available. The amount of inodes can be defined while creating file systems; you can only change them afterwards by repartitioning your HDs. o For integrating the live file system your partition has to be at least 50 MB in size. 3.27.7 IB7: Demo mode Since a lot of first-time users want to "play" with Linux before repartitioning the whole disk we have included the possibility of installing a "demo mode" onto an existing MS-DOS file system. A minimal Linux system will be installed in a directory in your existing MS-DOS file system which needs approximately 20 MB. Access to the 600 MB live file system is reached via symbolic links which point to certain programs on CD. Dynamic integration of the live file system is not implemented in this mode since about 40000 links have to be created. As the current implementation of the + UMSDOS-filesystem with HDs of 8k-clusters would use up to 300 MB this, of course, would make the term "demo-mode" absurd. Nevertheless you are free to perform a "normal" Linux installation on top of a DOS partition. But you should realize that performance and system security are less optimal. _______________________________________________________________________________________________________________________|| |||||||| On the DOS partition where you want to install the "minimal system" there ||| | |||||||| should not be any file named linux . This special directory will take up || | |||||||| || | |||||||| your base-system! || | ||||||||_______________________________________________________________________________________________________________||_ For booting your demo system later on there are two options. Either you start it via our boot disk or you can use loadlin which enables you to start Linux from a running DOS system. This tool was used during first-time installation provided that you have DOS installed and you have a DOS-boot configuration that runs your machine in real mode not in protected mode. * * 59 3. Installation _________________________________________________________________________________________________________________* *_______________________________________@ 3.28 Partitioning for novices Is Linux and its filesystem completely new to you? You probably ask your- self questions like: How much space should I assign to Linux? What's the minimum I need? What's best for my needs? How should I divide up the available space? Let's start with the minimum: 80 MB. This only works if you use the machine for a simple purpose: You only work on a text console (no X-Window sys- tem). If you want to peek into X and start a couple of applications: 200 MB. Both values include swap. What's an extensive installation? 500 MB. In this world of giga-byte hard disks this is rather modest. Swap included and there is no upper limit. What's the best for your needs? That depends on what you want to do: o X and modern applications like Applixware or Netscape : 700 MB to 1 GB. o For creating your own little applications on X you also need 700 MB to 1 GB. o Compile your own X servers, write your own CDs and the items men- tioned above: 4 GB. o Set up an Internet/FTP server: 400 MB minimum. How should you divide up the hard disk? This simple question cannot be answered easily. If you do not want to partition these simple rules apply: o up to approx. 550 MB: swap partition and a root (/ ). o more than approx. 600 MB: swap partition, root (100 MB) and the rest for programs (/usr ). o more than 750 MB: swap, root (100 MB), home (/home ) app. 100 MB and the rest for programs (/usr ). ________________________________________________________________________________________________________* *________|| | If you plan to start Linux directly from the hard drive, you will need a * * |||| | Linux partition below the "1024 cylinder limit" as a boot partition (cf. sec- * * ||| | * * ||| | tion 8.3.4, page 125 and section 8.7.3, page 138). This doesn't concern * * ||| | you if you will start Linux from DOS/Windows with loadlin . Most of the * * ||| | * * ||| | time this boot partition will be the same as the root partition. * * ||| |_______________________________________________________________________________________________________* *________|||_ Furthermore you should consider that some programs (mostly commercial programs) install their data under/opt ; just in case, please, either provide for /opt an own partition or make the root partition big enough. Examples: o KDE (30 MB) o Wabi (10 MB) o Netscape (35 MB) o Applixware (300 MB) o Staroffice (130 MB) If you don't want a separate partition /opt , you may use this "work around": 60 _______________________________________________________________________________________3.29.__Partitioning_for_experts______* *______________________ _______________________________________________________________________________________________________________________|| |||||||| There's enough space anywhere else in your system, e. g. under /usr ; ||| | ||||||||| create a directory (e. g. opt ) and make a symbolic link: || || |||||||| root@earth: # cd / ||| | |||||||| || | |||||||| root@earth:/ # mkdir /usr/opt || | |||||||| root@earth:/ # ln -s /usr/opt /opt || | ||||||||_______________________________________________________________________________________________________________||_ If you plan a bigger configuration than 1 GB it is recommended that you read further. For swap partitions please see section 3.29.1. 3.29 Partitioning for experts Partitioning has already been outlined in section S1 and section S10. This section should provide more detailed information for tailoring a system that best suits your needs. This section mainly is important for those of you who want an optimized system as far as security and performance are concerned. It is necessary to have a profound knowledge of the functions of a Unix filesystem. The topics + mountpoint, physical, extended and logical partitions should not sound Chinese to you. First: There is no golden mean for all but lots of tiny golden means for anybody. Do not panic we will give you some real numbers as hint. First gather the following information: o What is the purpose of the machine? (Fileserver, compute-server, stand- alone machine) o How many persons are going to work with this machine? (simultaneous logins). o How many hard disks are installed? How big are they and which system? (EIDE, SCSI or even RAID controllers). 3.29.1 Size of swap partition Quite often you will read: "+ Swap should at least be quite as large as physical RAM". This is a relic of times when 8 MB was regarded as a lot of RAM. Well, times sure have changed. Those who purchase machines with less than 16 MB of RAM have not been counselled well. Lets come back to the statement mentioned above. It was our aim that the machine will have app. 30 to 40 MB of virtual + memory. Using modern applications that need quite an amount of memory these values have shifted up a bit. Generally 64 MB of virtual swap should be sufficient. Do not be stingy. If you compile a kernel under X and want to have a look at the manual pages using Netscape and have an emacs running will already take up all your 64 MB. For being safe, opt for at least 96 MB virtual memory. One thing you should never do is not to assign swap space at all! Even on a machine with 256 MB RAM there should be a swap partition. The reasons are described in sec- tion 3.29.3 * * 61 3. Installation _________________________________________________________________________________________________________________* *_______________________________________@ You plan to run extensive simulations and need some giga-bytes of mem- ory(!). In case of doubt as to whether Linux suits your needs please read section 3.29.2 (Compute-Server). 3.29.2 Appointment of the machine as stand-alone machine The most common use for a Linux machine is a stand-alone machine. In order to make decisions for you as easy as possible we now provide you with some concrete figures which you can use at home or at your company. In table 3.1 there is a overview of different installation volumes for a Linux system. ________________________________________________________________________________________________________* *________ __Installation________Needed_disk_space________________________________________________________________* *________ minimum 80 MB up to 200 MB small 200 MB up to 500 MB medium 500 MB up to 1 GB big 1 GB up to 3 GB ________________________________________________________________________________________________________* *________ Table 3.1: Examples for installation Of course these values increase as you want to save additional data volumes. Example: Printer server/router Let's assume that you do not want to throw your old 386 SX 20 with is 80 MB hard disk away. Separate the disk into a 16 MB swap partition and the rest (64 MB) for root / . This little machine may serve as a firewall or as a gateway to the internet. Of course it should only be used by root. Example: stand-alone machine (small) You have app. 500 MB spare hard disk that should be used for Linux.: 100 MB for root / , a 32 - 40 MB swap partition and the rest on user /usr . Example: stand-alone machine (average) There is 1 GB available for Linux. As above 100 MB for root / , 64 MB for swap,100 MB for /home and the rest for /usr . Note, that the RPM database under /var occupies several MBs (cf. section 5.3.2, page 96)! Example: stand-alone machine (luxury) If you have more than 1 GB available there is no overall solution. Please read section 3.29.3. 62 _______________________________________________________________________________________3.29.__Partitioning_for_experts______* *______________________ Example: File server Here hard disk performance really is crucial. You should use SCSI devices if possible. Please also keep in mind the performance of the disk and the controller. A file server is used for centrally saving data. These might be + home direc- tories, a data base or other archives. The advantage is an easy administration. If the file server should serve a huge net (from 2o users onwards) optimizing hard disk access is essential. Supposed you want to provide a file server for 25 users (their home directo- ries). You know that every user will have 80 MB of personal space. If not all of them compile the whole day a 2 GB disk which is mounted under home will do. If there are 40 users mathematically a 4 GB disk is necessary. Anyway it is better to split home onto 2 2 GB disks as now they share the load (and access!). As a compute server: A compute server generally is a powerful machine that carries out extensive calculations over the net. Normally such a machine is equipped with exten- sive main memory (from 256 RAM onwards). The only point where fast disks are needed is for the swap space. Same applies here: separate the swap partitions onto separate disks. Furthermore keep in mind that Linux can only handle 128 MB swap partitions but anyway can handle 8 of those5 3.29.3 Optimizations Normally the disks are the bottle-neck. For avoiding this there are two possi- bilities which should be used together: o Separate the load on multiple disks o Equip your file server with enough memory (at least 64 MB) Parallelising multiple disks This needs some further discussion. The total amount of time needed for transferring data can be separated into four parts: o Time elapsed until the request reaches the controller o Time elapsed until this request is send to the disk o Time elapsed until the hard disks manages to set its head o Time elapsed until the media has turned to the right sector o Time elapsed for transferring data Item 1 depends on the network connection and has to be regulated elsewhere. We do not to cover this right here. Item 2 can can be ignored; this depends on the controller. Item 3 is the vital part. The time is counted in milli seconds. Relative to the access time of main memory (that is measured in ns) this is 5 ____________________________________________________ slight modifications provided even 64 * * 63 3. Installation _________________________________________________________________________________________________________________* *_______________________________________@ a factor of 1 million! Item 4 depends on the disk rotational speed, item 5 depends on this speed and the number of heads and of the actual position (inside or outside). For optimized performance one should consider item 3. Here the SCSI fea- tures "disconnect" comes into action. Lets have a look at what happens: The controller sends the command (in this case to the hard disk) "Go to track x, sector y" to the device. Now the mechanic of the disk has to start- up. If this is an intelligent disk (if it supports disconnect) and the driver itself also is able of doing disconnect, the controller sends a disconnect and the disk separates itself from the SCSI bus. Now other SCSI devices can do their task. After a while (depending on the strategy or load on the SCSI bus respectively) a connection to the disk is reestablished. Normally the device now has reached the demanded track. On a multitasking-multiuser system like Linux there are lots of optimizations that can be done here. Let's have a look at an output of the command df (comp. screen output 3.29.1). ________________________________________________________________________________________________________* *________|| | Filesystem 1024-blocks Used Available Capacity Mounted on * * | | * * | | /dev/sda2 45835 27063 16152 63% / * * | | /dev/sdb1 992994 749694 192000 80% /usr * * | | /dev/sdc1 695076 530926 133412 80% /usr/lib * * | | * * | |_______________________________________________________________________________________________________* *________ | Screen output 3.29.1: Example of a partitioning: Output using command df . So, what benefits does parallelizing bring us? Provided we enter in /usr/src : root@earth:/usr/src/ > tar xzf paket.tgz -C /usr/lib Here paket.tgz should be installed onto /usr/lib/paket . For doing so the shell launches tar und gzip (these are located in /bin thus on /dev/sda ), then paket.tgz in /usr/src is read (is on /dev/sdb ). At last the extracted data is written to /usr/lib (under /dev/sdc ). Now, positioning as well as read/write of the disks internal buffers can be activated at the same time. This is only one example; there are many more. As a rule of thumb, if there are many disks (with the same speed), /usr and /usr/lib should physically be placed on different disks. Here /usr/lib should have approximately 70% of the capacity of /usr . / due to its access it should be placed on the disk containing /usr/lib . From a certain number of SCSI disks onwards ( 4-5) one should consider buying a RAID controller. Thus operations on the disks are not only quasi- parallel but parallel. Fault tolerance is one of its famous side-products. Processing speed and size of main memory We want to point out that size of main memory is more important under Linux than the processor itself. One reason6 is Linux' ability of dynamically creating buffers of hard disk data. Here Linux uses lots of tricks as "read ahead" (gets sectors in advance) and "delayed write" (saves writes until there _____________________________________________6 if not the main reason 64 _______________________________________________________________________________________3.29.__Partitioning_for_experts______* *______________________ is a bundle to write). The latter is the reason why you should not switch off your Linux machine. Both items are the reason why Linux is so fast and why the memory seems to fill rapidly. Linux works with shared libraries, that is lots of programs and applications share the same library. This leeds to the fact that a library call has only to be allocated once. Of course this also means that your memory fills up rather quickly. So if you do not know whether to purchase another main board or some more memory we recommend to do the latter as it increases the speed of Linux. _______________________________________________________________________________________________________________________|| |||||||| total used free shared buffers cached | |||||||| | |||||||| Mem: 63304 62312 992 15920 38692 4200 | |||||||| -/+ buffers: 19420 43884 | |||||||| Swap: 199508 14548 184960 | |||||||| | ||||||||_______________________________________________________________________________________________________________ | Screen output 3.29.2: Output of free If you want to find out what memory is used up to now just enter: root@earth:/root > free This will give you an overview of used memory and buffers. The screen output 3.29.2 shows that some 38 MB are saved in buffers. If you want to access data that already is buffered this data almost immediately is available. * * 65 3. Installation _________________________________________________________________________________________________________________* *_______________________________________@ 66 Chapter 4 YaST - Yet another Setup Tool YaST is a main part of S.u.S.E. Linux. It helps you to install and maintain your software (installation, deinstallation) and to administrate your system. This chapter will cover YaST's main features to make the installation as easy as possible. _______________________________________________________________________________________________________________________ |||||||| There is a relative new YaST on S.u.S.E. Linux 5.2. Thus the screenshots ||| | |||||||| || | |||||||| in this book may differ from those you see. Furthermore some of YaST's || | |||||||| new functions are not mentioned. || | ||||||||_______________________________________________________________________________________________________________||_ 4.1 Control and keyboard mapping Control of YaST is mainly achieved by the cursor and tab keys. You can move throughout the list using the cursor-keys as well as Page " and Page # and accept your selection with Enter or _- respectively. When encountering Yes/No or entry-fields you can jump to and fro using 7 ! . Esc always leaves the current menu. If you want to keep the settings you made you need to press F10 . On some rare occasions (e. g. when starting YaST from a non-Linux terminal) you are not able to use the function keys since the codes are misinterpreted by the terminal. In this case you should use Strg + f for simulating the function keys F . F10 can be used instead of Ctrl + f 0 . F1 and F12 are not used under YaST. 4.2 Installation media This menu (figure 4.1) lets you choose the installation media. Normally you would select `Installation via CD-ROM'. `Installing from HD' allows you to install even if your CD-ROM is not recognized by Linux. (section 4.4) `Installation via NFS' enables you to install Linux over a network, meaning your computer doesn't have a CD-ROM of its own but is connected to a machine (via + ethernet) which has a supported CD- ROM. Please refer to section 4.5 for further information. 4.3 Installation from CD-ROM If you plan to install via your CD-ROM please specify your CD-ROM. * * 67 4. YaST - Yet another_Setup_Tool__________________________________________________________________________________________________* *_______________________________________@ | * * | ________________________________________________________________________________________________________* *________|||| | * * | || * * || | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | |_______________________________________________________________________________________________________* *________ | Figure 4.1: Selection of installation media in YaST | * * | ________________________________________________________________________________________________________* *________|||| | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | |_______________________________________________________________________________________________________* *________ | Figure 4.2: Selecting an CD-ROM drive ________________________________________________________________________________________________________* *________|| | There are different drivers for Mitsumi drives! Those mentioned in the * * |||| | above list as "mitsumi drivers" are designed for the "old" Mitsumi CD- * * ||| | * * ||| | ROMs which are connected to a separate controller (e.g. LU-500 or FX- * * ||| | 001). New Mitsumi drives (e.g. FX-400) are ATAPI drives and therefore * * ||| | * * ||| | you must select ATAPI EIDE. Same applies to Sony and Aztech drives. * * ||| |_______________________________________________________________________________________________________* *________|||_ Mitsumis mcdx driver differs in so far as it is able to read multisession CDs. Therefore it is rather meaningless for installation which one you choose. We have decided upon putting both for selection because there might be cases where one of them works and the other doesn't. So now you can choose 68 ________________________________________________________________________________4.4.__Installation_via_HD_partition_________* *______________________ either of the two. 4.4 Installation via HD partition If your CD-ROM is not supported by Linux, you still can install Linux on your computer by choosing another installation media. YaST expects a cer- tain file structure which you have to "mirror" on your MS-DOS - OS/2 (can be HPFS (+ filesystem)) drive. Since you almost certainly don't have enough space to mirror the whole CD you have to decide which packages you want to install in each step. You can install as many packages as fit on your HD tem- porarily. After installation of your base system you can add other packages, one after|the|other.| | _______________________________________________________________________________________________________________________|||| |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | ||||||||_______________________________________________________________________________________________________________ | Figure 4.3: Directory tree for installation Our example shows you how to install a base system via an MS-DOS HD partition. Here directories and file structures are given and cannot be changed. All directories belonging to series A (a1) are needed as well as directory setup including all subdirectories. Directory images has to contain the ker- nel that you plan to use (double! one with the .ikr extension and one the ker- nel itself; file with the .inf extension contains a text number enabling YaST to show a description of this specified kernel and should be copied along with the other files. Configuration of a specific kernel can be found in the "README" file which resides in the same directory). If you want to start from MS-DOS you additionally need the root file under the subdirectory images. If you plan to start your installation system directly from MS-DOS, the direc- tory /suse/ is obligatory. Assuming that you have created a directory suse under the directory C:\emil , then you have to fill in the path C:\emil in * * 69 4. YaST - Yet another_Setup_Tool__________________________________________________________________________________________________* *_______________________________________@ setup (without the `\' at the end!). Later on when YaST asks you what the source directory is you type /emil/suse. 4.5 Installation via NFS Installation via network provides a means of easily installing multiple com- puters for experienced Linux users, even if only one of them is equipped with a CD-ROM or the sources are reachable over an + NFS mounted harddisk. You should be familiar with configuring an NFS server before you try in- stalling via NFS. Even installation of a notebook which is equipped with a PCMCIA networking card is possible with this feature. | * * | ________________________________________________________________________________________________________* *________|||| | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | |_______________________________________________________________________________________________________* *________ | Figure 4.4: Entering network data for NFS installation For installation via NFS computers need not to be linked via ethernet, but with others (e. g. , parallel interface) as well. This feature mainly is interesting for laptops and notebooks. If you plan such an installation you should, of course, select a kernel with PLIP support, and you have to configure your PLIP interface before being able to start installation. If you're directly connected to the NFS server the address being used matches the address of the PLIP partner and is exactly the same of your NFS server. The PLIP interface in most cases is just plip1; even the hardware parameters should not differ from standard for most cases. Please be aware that while using your parallel interface as a PLIP interface you're not able to connect a printer to the same parallel interface. In most cases a printer connected to this interface will cause a permanent reset as soon as the driver starts! The next mask asks you to give in the IP-address of your NFS server and the source directory. Of course the server has to export the named directory! 70 ______________________________________________________________________________________________4.5.__Installation_via_NFS____* *______________________ |||||||| | _______________________________________________________________________________________________________________________|||| |||||||| | ||||||||| || |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | ||||||||_______________________________________________________________________________________________________________ | Figure 4.5: PLIP configuration 4.5.1 Installation via reachable directory This option is generally used to install software once you have your Linux system running. In addition, it lets you install from drives for which no drivers existed when this CD was produced. Now, to access this drive you start YaST. After having done so, change to another console (e. g. using Alt + F2 ) and login as `root'. Now you have to + mount your CD-ROM by hand to a directory, this can be done by (e. g. ): root@earth: # mount -tiso9660 /dev/cdrom /cdrom See section 19.12. Now you can perform a normal installation; just fill in (figure 4.6) the direc- tory of the sources, which is the path where the CD-ROM is mounted and add /suse to it. So, if you have mounted your CD-ROM (or your HD partition) to /cdrom you enter (as in figure 4.6) /cdrom/suse. /cdrom/suse 4.5.2 Select the installation mode Here you can choose between either "demo" or "normal" Linux installation. If you use "demo" mode Linux will be installed on top of an existing MS- DOS partition; of course this is only possible if there is at least one MS-DOS partition on your system! Use of demo-mode is only available if there is at least one DOS partition on your system! * * 71 4. YaST - Yet another_Setup_Tool__________________________________________________________________________________________________* *_______________________________________@ | * * | ________________________________________________________________________________________________________* *________|||| | * * | || * * || | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | |_______________________________________________________________________________________________________* *________ | | Figure 4.6: Entering the source media * * | ________________________________________________________________________________________________________* *________|||| | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | |_______________________________________________________________________________________________________* *________ | Figure 4.7: Select installation mode in YaST 4.5.3 The main menu After having entered all the data absolutely necessary for the installation, you will find yourself in YaST's main menu. All previous settings can be changed at any time. 4.5.4 Settings All installation relevant information is done in `Setting up installation'. Installation can only proceed if you tell YaST where to mount your partitions. If your computer has very little RAM it is recommended to disable `Reading DU-information' (as it consumes rather a big amount of memory). This 72 ______________________________________________________________________________________________4.5.__Installation_via_NFS____* *______________________ |||||||| | _______________________________________________________________________________________________________________________|||| |||||||| | ||||||||| || |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | ||||||||_______________________________________________________________________________________________________________ | |||||||| Figure 4.8: YaST's main menu | _______________________________________________________________________________________________________________________|||| |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | ||||||||_______________________________________________________________________________________________________________ | Figure 4.9: Menu `Setting up installation' also is a good choice if your computer is supplied with 8 MB RAM but you have not activated a swap partition yet. 4.5.5 Partitioning your HD The most critical and error prone point while installing a new OS is parti- tioning your HD(s). Normally each OS requires you to use a partition of its own. Linux offers you the possibility of installing a demo mode on top of an existing MS-DOS filesystem but you should only take this into consid- eration if you plan to "experiment" with Linux. The performance of demo mode is much less efficient compared to a normal installation. Furthermore * * 73 4. YaST - Yet another_Setup_Tool__________________________________________________________________________________________________* *_______________________________________@ the system is not as stable as if given a partition of its own since there are no filecheck programs under MS-DOS and your Linux can be influenced by MS-DOS during runtime. So if you want to give Linux some partitions of its own you should split your HD into a couple of partitions. Generally a relatively small partition is used as root partition thus reducing write access and resulting in a more stable and not so error prone system and quite large partitions for containing the rest of the system. So /usr on most Unix systems is a partition of its own. It can be convenient to use a separate partition for /var or /tmp. After all partitioning is more a philosophic question so there is no "ideal solution". There is one partition that you should create; this is a swap partition to in- crease your amount of virtual memory (+ memory). It is possible to create a swap file but since every access to this file is done via the filesystem this reduces performance; this is important if you have little memory installed. Here a swap file is no alternative at all to a swap partition! If there is more than one HD installed you will be asked which HD to partition leading|to a menu where all your partitions (of this HD) are presented. * * | ________________________________________________________________________________________________________* *________|||| | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | |_______________________________________________________________________________________________________* *________ | Figure 4.10: Partitioning with YaST Using the cursor ( " and # ) keys you can move up and down through the partition list. F3 changes partition type, F4 deletes a partition and F5 creates a new partition. Changing of partition type is necessary for creating a swap partition. Perhaps no partitioning is required if you have decided to use partitions momentarily used by another OS for Linux. In this case just move the cursor on to the desired partition and change the type by pressing F3 . Be aware that Linux has no restrictions of where to put your system. It is possible to install Linux on a primary partition as well as on a logical within an extended partition. 74 ___________________________________________________________________________________________4.6.__Assigning_file_systems_____* *______________________ Logical partitions have been introduced because in the partition table there is only room for four partitions; if you need more than four partitions at least the last partition has to be assigned an extended partition. Now you can create as many logical partitions in this extended partition as you like.1 After partitioning your computer has not to be booted at once but you can set the amount of software you want to install later on. Suppose you find out that your freshly created partitions are not big enough to hold everything you decided to install, you can go back and change the partition table until it is to your satisfaction. Only after quitting YaST the partition table will be written. Then you really have to reboot your computer immediately! 4.6 Assigning file systems After partitioning your HD you have to link your freshly created partitions to directories in your Linux tree. Select `Set up your target partitions/filesystems' under the topic Installation|settings.| | _______________________________________________________________________________________________________________________|||| |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | ||||||||_______________________________________________________________________________________________________________ | Figure 4.11: Assigning file systems In our example (figure 4.11) you see the partitions of a system containing one HD. You can now decide for each partition if and how you want it to be formatted and of course the mount point in your Linux tree. You have to assign one partition as root partition! This is called root as it is the root of all directories. That's why it is assigned the mount point (directory) /. Manipulating your filesystems can be achieved by using the corresponding function keys. Needless to say that you can only manipulate Linux partitions. If the cursor is positioned on another filesystem than Linux all functions except mounting point are disabled. 1 ____________________________________________________ Extended as well as logical partitions are known to DOS as extended DOS partition or logical drive although this concept applies to any operating system * * 75 4. YaST - Yet another_Setup_Tool__________________________________________________________________________________________________* *_______________________________________@ 4.6.1 Mountpoint F4 lets you choose where to mount your partitions. You have to assign one partition as root-partition ((/)). You can mount your MS-DOS partitions to directories where you can find them later on easily (e. g. /dosc for the first MS-DOS partition or /dosd for your second MS- DOS partition. ________________________________________________________________________________________________________* *________|| | Be aware that all mount points have to be given using absolute pathnames * * |||| | and that special keys for those directories are not allowed. You may not use * * ||| | * * ||| | /etc, /bin, /sbin, /lib, and /dev for your own partitions since those directories * * ||| | are reserved for system commands, system files, and contain the programs * * ||| | * * ||| | needed to mount other partitions! * * ||| |_______________________________________________________________________________________________________* *________|||_ There is one special thing about MS-DOS partitions! You can mount MS- DOS partitions in two different ways. o One of them is the "normal" MS-DOS partition where all MS-DOS restric- tions apply and o the other is a UMSDOS partition. UMSDOS allows you, e. g. usage of long filenames, furthermore all Linux-specific properties such as owner, per- missions and creating of links are possible under UMSDOS. Since MS- DOS doesn't provide you with the use of long filenames those are saved in special files which are located in each directory. Those files are named "- linux-" and should not be deleted at all, since otherwise Linux is not able to access its files! While running linux you will not see those files since they only contain internal information about the UMSDOS filesystem. Since the DOS filesystem does not allow setting such special information they are located in special files which are located in each subdirectory. These files are called --linux-.--- and should not be deleted under DOS, since in that case Linux will not be able to access them any more! Under Linux these files are invisible since they only contain internal information of the UMSDOS filesystem. 4.6.2 Inodes density + Inodedensity gives the average size of a file on a particular partition. Number of inodes determines how many files can be created on a certain partition. If this number is too small it can happen that a partition seems to be full although there are still some blocks left. If you, e. g. , take 4096 bytes per inode it means that on average each file has a size of 4 KB; now imagine you put only small files on this partition (about 1 kB) you can only use a quarter of your HD since your partition seems to be "full". 4 kB per file has proven to be an adequate size (more inodes per partition of course mean that there is less "netto-data-space" since inode tables also have to be written on this partition). If you plan to use one partition as spool directory , e. g. for netnews, a size of 2048 is enough since those articles normally are rather small. Another reason for choosing a large number of inodes is our Live-File-System . When using 76 ___________________________________________________________________________________________4.6.__Assigning_file_systems_____* *______________________ our live- filesystem approximately 40.000 files will be incorporated which use one inode per file. 4.6.3 Formatting Pressing F6 you can tell if and how a partition should be formatted. With modern SCSI HDs checking for bad sectors is not really necessary but if you want to be sure you can select this option. Formatting and checking for errors of course takes quite an amount longer. If you didn't change the partition table in this YaST session you can format your HDs immediately. Otherwise YaST saves your selections and formats after them (after you have rebooted). Partitioning takes some time depending on the size of the partition to be formatted. 4.6.4 Reading fstab If you want to read an existing fstab you have to press F7 . You will be shown even those entries which belong to the HD's filesystem (swap, proc, nfs-mount etc.). Those entries are greyed out and can't be changed. Of course they will be kept if you decide to save your changed fstab partitions. This feature is needed if you do an update of your existing Linux system, since YaST needs to know where the system resides. It is possible to have several versions of Linux running parallel on your computer. 4.6.5 Size of installation After you have finished configuration of your filesystem please go to `Start / setup/installation' to choose the amount of packages you are going to install. YaST provides a means of saving your own installation-profile or to load|an|existing|one. | _______________________________________________________________________________________________________________________|||| |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | ||||||||_______________________________________________________________________________________________________________ | Figure 4.12: Assigning amount of installation * * 77 4. YaST - Yet another_Setup_Tool__________________________________________________________________________________________________* *_______________________________________@ Furthermore you can start installation or make sure what would happen if you install the actual configuration. 4.6.6 Loading an existing configuration There are several predefined configurations under `Load configuration'; you can choose any of them. Under several others there is a base system which you should use if you start YaST from disk. A minimal system gives you the opportunity of starting Linux from HD and to install more packages afterwards. Since YaST isn't loaded from disk, handling and performance is much more convenient. This applies mainly to computers with little RAM. If you have designed (and saved) your own profile you can, of course, find it here and load it. This might come in handy if you're installing an identical configuration on several different computers. If you have already installed a system you should be careful when you load a different configuration since all packages not belonging to this particular con- figuration will be marked for deletion. If you want to keep those preinstalled packages you merely answer no if presented the question whether you want to delete or not. Otherwise you have to deselect all packages marked with `[D]'; afterwards there should be an `[i]'. 4.6.7 Saving your configuration Here you can save your own configuration. If YaST is started from disk this disk is used for saving your configuration; if you have booted directly from CD you will be asked to insert a pre-formatted disk; YaST will write your configuration to this disk. 4.6.8 Changing your configuration If you select this option you will be presented a list; all packages on our CD are listed here. You can decide which package to install plus you can deselect previous installed packages and thus let them be deleted. You can move up and down by using the cursor-keys or # and " , as well as Page " and Page # ;. _- opens the series for further details on packages belonging to this series. In the lower window there is a description of the package under the actual cursor position; the right window shows you the actual percentage of your partitions being used. You can leave this mask by pressing F10 ; then you'll be back in the config- uration menu. If you press _- on a series you'll be put into package selection of this series. If you have loaded a configuration before, the packages belonging to this configuration are marked with a cross. Here you can see a short description of the package in the window underneath of the actual package (the one under the cursor). The right window shows you the percentage of your partitions. Those values are being updated while selecting/deselecting a package. In some cases there is not enough room to show all necessary information. By 78 ___________________________________________________________________________________________4.6.__Assigning_file_systems_____* *______________________ |||||||| | _______________________________________________________________________________________________________________________|||| |||||||| | ||||||||| || |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | ||||||||_______________________________________________________________________________________________________________ | |||||||| Figure 4.13: Selecting series in YaST | _______________________________________________________________________________________________________________________|||| |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | ||||||||_______________________________________________________________________________________________________________ | Figure 4.14: Selection of packages, series a1base system pressing F2 you can enlarge the lower window, by pressing F3 you can enlarge the right window. Each package name is preceeded by it's current state: `[ ]' indicates that this package is not yet installed `[X]' indicates that this package will be installed `[i]' indicates that this package is installed `[D]' indicates that this package will be deleted `[R]' indicates that this package will be updated (replaced) * * 79 4. YaST - Yet another_Setup_Tool__________________________________________________________________________________________________* *_______________________________________@ By pressing _- you can switch either between `[ ]' and `[X]' or be- tween `[i]' `[R]' and `[D]'. Pressing F10 leads you back to the series selection menu; pressing ESC leaves without changes. If you, by any chance encounter problems with space of your partitions you can easily switch back to the main menu and repartition your disk. The actual partitioning will be written if you leave YaST. If you're running YaST from disk, only install a minimal configuration, boot your Linux and install the remaining packages. Of course you can try by selecting your "dream"-configuration to see how much space is needed. 4.6.9 What if.. YaST can also be used for deinstallation, therefore we created this menu so that you can see what will happen if you start installation. Here you can make sure that you didn't mark packages for deletion which are necessary for your Linux system to work properly. 4.6.10 Installation This is the real installation. All marked packages will be read from CD, uncompressed and copied to their directories. If you have changed partition tables during this YaST session, installation is not yet possible since after changing partition tables your computer needs to be rebooted. Furthermore this modified data has not yet been written to partition table, this only happens after leaving YaST. During installation status messages will be shown in the lower window. The upper window informs you about the actual package to install. After installation you could jump back to the protocol window by pressing 7 ! . That's for making sure everything went right. Here you can see if problems occurred. 4.6.11 Deleting packages Selecting these item gives you an overview of the packages that are actu- ally installed. This list includes "foreign" packages, these are packages not included on the S.u.S.E. CD. YaST can neither update these packages nor check for any dependencies. The easiest way is to replace these packages by packages on your S.u.S.E. CD. Just select the packages you need to replace. A short description is displayed if you press F2 (as there is no complete documentation available). F10 deletes the marked packages. Thereafter you may install the corresponding packages from CD. Check package dependencies YaST checks the dependencies of packages already installed and packages that are to be installed, minus the packages that are to be removed; or phrased differently, the packages that would have been installed if you had already selected the `Start installation' option. The following types of dependencies exist: The package dependencies which were found are then displayed in a list. 80 ___________________________________________________________________________________________4.6.__Assigning_file_systems_____* *______________________ _______________________________________________________________________________________________________________________ AND Means that if this packages is installed, all listed packages should be installed too. For example, if you install the compiler, you'll also need the include and lib files. OR Means that if this package is installed, at least one of the listed packages should be installed as well. EXCL Means that if this package is installed, then none of the listed packages should be installed. _______________________________________________________________________________________________________________________ Table 4.1: Package dependencies Index of all series and packages This option simply provides a list of all packages included on the CD. Those packages marked with an `*' are either already installed or are selected for installation. This option is helpful and allows you to get a quick overview. 4.6.12 Search packages Sometimes you might wonder where the file is that you are looking for. You try to find out where it is located on your S.u.S.E. Linux CD. There is a "package information" that makes it quite a snap to look for pack- ages and/or files. You may either search through the installation medium or through the entire system (or both). YaST looks for this file and presents you a list of hits. 4.6.13 Install packages This item is for installing packages. These might be either new or bugfgixed versions which may be downloaded from our FTP site ftp.suse.com. Well, you may even install packages you downloaded from the Internet or built yourself. A couple of formats are supported: Tar archives (.tgz) and RPM packages (.rpm, .spm and .src.rpm) as well as special designed patch files (.pat) that are available (if needed) on our FTP site. This installation procedure is divided into three steps. (a complete instruction may be get by pressing F1 ): o Selection of installation medium o Selection of the packages offered o Installation of the given packages You may now select the package medium by entering `Source:' and press- ing _- . Here the following items are available: `Directory', `FTP', `Installation medium' as well as `Floppy'. You may need to change the default path (for directory and/or FTP). YaST will connect to the source medium as soon as you hit _- . Thereafter you will get a list of the avail- able packages. This makes it possible to install packages via the Internet. Just select `FTP'. * * 81 4. YaST - Yet another_Setup_Tool__________________________________________________________________________________________________* *_______________________________________@ | * * | ________________________________________________________________________________________________________* *________|||| | * * | || * * || | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | |_______________________________________________________________________________________________________* *________ | Figure 4.15: Install packages via FTP . The default address is: ftp.suse.com:/pub/SuSE-Linux/suse_update (vgl. figure 4.15, page 82). ________________________________________________________________________________________________________* *________ | If you get an error message as "530 User ftp access denied." this * * ||| | * * ||| | means that there are too many users currently online. Just try later * * ||| |_______________________________________________________________________________________________________* *________|||_ You need to work your way through to the package you want to install. Now you can install this package as usual by marking it using the Space key. F10 accepts your selection. The package itself is copied to /tmp/ftp . Thus if something goes wrong you may install it manually. (see section 5.3.1, page 93). 4.7 Administration Besides pure installational work YaST also helps you maintaining your sys- tem now that you are a brand new + system administrator. After installation (installing every package) has been done, there are still a couple of things to be done for adjusting your system to your computer and your personal needs. You can reach system administration via `System administration'. 4.7.1 Integrating hardware into your system This is designed for exactly specifying your hardware. In most cases a sym- bolic link (+ Link) from a standard device to your personal hardware will be created allowing you to access your hardware whenever you want without having to know the device's name. Adjustment of mice and CD-ROMs is quite easy, whereas configuring your printer can be rather tricky. Configuring a printer is much more demanding. 82 _____________________________________________________________________________________________________4.7.__Administration___* *______________________ |||||||| | _______________________________________________________________________________________________________________________|||| |||||||| | ||||||||| || |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | ||||||||_______________________________________________________________________________________________________________ | Figure 4.16: Administering your system Configuring your printer Accessing a printer under Linux is not as trivial as it seems to be. Luckily with apsfilter there is a powerful filter which automatically detects the file type, converts it and sends it to the printer. General (like all other Unix) PostScript plays a central role under Linux. Printing a PostScript file on a PostScript printer is rather easy but since those printers are quite expensive only a few users call one of them their own. That's what we use GhostScript (gs ) for, a free software that interprets a PostScript document and converts it into a form readable by the printer. apsfilter even converts + ASCII files into ghostscript files for printing them directly on a PostScript printer or converting them into a format readable by the printer (programmGhostScript). YaST offers the possibility of easily installing and configuring apsfilter . If your printer is capable of printing colours you just activate the correspond- ing flag. In `Printer type' you type in the type of your printer and whether it is a PostScript printer or not. `Printer name' shows you a list of the most common (and supported) printers. You can scroll through the list and choose a printer. If your printer is not mentioned here you can try a similar one or the ancestor model (e. g. if you have LaserJet 5L just take LaserJet 4). `Paper size' usually is A4 . If your printer is connected to a serial + interface you can adjust the baud-rate as well. In general printers are connected to the parallel port, meaning that you have to enter the user interface. For most cases this will be /dev/lp1; if your printer is connected to the first parallel interface. /dev/lp2 is only for * * 83 4. YaST - Yet another_Setup_Tool__________________________________________________________________________________________________* *_______________________________________@ | * * | ________________________________________________________________________________________________________* *________|||| | * * | || * * || | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | |_______________________________________________________________________________________________________* *________ | Figure 4.17: Hardware configuration printers on the second parallel port. /dev/lp0 is the parallel port on a Hercules graphic device . 4.7.2 Kernel and boot configuration Here you can adjust and configure your system's boot setup. You can install either of the precompiled kernels (from S.u.S.E. CD), if you notice your first choice wasn't optimal. Anyway it is recommended to build a customized kernel which is perfectly adapted to your system; this kernel (see chapter 13)will be much smaller and faster. Moreover you will be given the possibility of creating a boot-disk for your system if you haven't done so already. This is highly recommended since you will be able to boot your entire system even if something goes wrong while installing LILO or you can't boot your system for another reason. ________________________________________________________________________________________________________* *________|| | If you have Linux already installed and want now to install Windows 95 * * |||| | * * ||| | you really must create a Linux boot-disk! Windows 95 assumes that it has * * ||| | the computer to itself and therefore overwrites the Master Boot Record * * ||| | * * ||| | without any hints or warnings! * * ||| |_______________________________________________________________________________________________________* *________|||_ 4.7.3 Configuring LILO YaST also serves as a frontend for correctly configuring LILO. A detailed description of all entries and options is to be found in chapter 8. We recom- mend reserving an additional entry in /etc/lilo.conf (e. g. linux.old) and to activate the entry kernel optional . If you recompile your kernel (see 84 _____________________________________________________________________________________________________4.7.__Administration___* *______________________ |||||||| | _______________________________________________________________________________________________________________________|||| |||||||| | ||||||||| || |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | ||||||||_______________________________________________________________________________________________________________ | Figure 4.18: Setup apsfilter with YaST chapter 13) a backup of your old kernel will be saved allowing you to always start the older kernel if anything should happen to the new one. 4.7.4 Configuring your network General networking configurations can be done by YaST. Do configure your network in any case since there are a lot of programs and applications that require you to have a correctly working network; even if you're not connected to a real network at all. General configuration of a network mainly consists of assigning an + IP- Address and a name; furthermore it is possible to add one or more name- servers and to install a sendmail configuration file which should be adequate for most cases. During configuration you will be asked whether you only want loopback installed or a complete TCP/IP network, If you don't have an ethernet card installed you should answer `yes'; this also spares you further unnecessary configuration. More information about configuring networks is found in chapter 16. 4.7.5 Integrating / Releasing CD-Live-File-System If you integrate Live-File-System links will be created for every package not already installed on your system. These programs can be easily started from CD without using disk-space. You should keep in mind since your CD-ROM will be mounted every time you boot, you have to leave the CD in your CD- ROM drive (mount point is /S.u.S.E. ). If you want to mount another CD first boot with our live-filesystem and then unmount it using the command: root@earth: # umount /S.u.S.E. * * 85 4. YaST - Yet another_Setup_Tool__________________________________________________________________________________________________* *_______________________________________@ | * * | ________________________________________________________________________________________________________* *________|||| | * * | || * * || | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | |_______________________________________________________________________________________________________* *________ | Figure 4.19: Kernel and boot configuration and mount the other CD afterwards wherever you want to. In some cases, even base programs are integrated from CD. In this special case those programs are constantly running and therefore you are not able to unmount the CD. Our base system is designed so that it is able to run without CD so just boot your computer with the CD removed. Linux will display a couple of error messages which you can ignore. All programs for accessing a CD are on your system. If you receive an error message "no space left on device" this could be because the + inode density on your system is too small. With very small partitions this is the most common error. Unfortunately this can only be solved by reformatting and decreasing inode size (e. g. 1 kB /inode instead of 4 kB /inode- for further information look at section 4.6.2). If your HD is used up to extend (no space is left) you can't even start YaST for disintegrating live-filesystem as YaST can't create the files it needs for start- ing. Just delete some of the symbolic links pointing to /S.u.S.E. , hereafter starting YaST again. 4.7.6 User management YaST provides a comfortable means of maintaining, adding, and deleting your users. If you are an experienced Unix user it might be too much work to start YaST for a simple useradd ; needless to say that you can use the standard programs for achieving the same result, such as useradd and userdel . 86 _____________________________________________________________________________________________________4.7.__Administration___* *______________________ |||||||| | _______________________________________________________________________________________________________________________|||| |||||||| | ||||||||| || |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | ||||||||_______________________________________________________________________________________________________________ | Figure 4.20: User management with YaST 4.7.7 Changing configuration file S.u.S.E. Linux is maintained by a central configuration file (/etc/rc.config). This file is read at boot time by the bootscripts which configure your system accordingly. There is an editor supplied with YaST to edit and change all these entries by hand, without having to know the exact syntax and what all these entries mean. Move the cursor onto the entry that you want to change, press _- or F3 to change its value. If you have edited this file by hand (without using YaST) you should run the script SuSEconfig to update the changes. This script reads /etc/rc.config and adjusts all program specific changes in its configura- tion files. A detailed description of all configuration files is to be found in section 12.6. 4.7.8 Creating backups It can be quite helpful to backup all modified and new files and packages to a file or to a tape- these in most cases are configuration files. That's what this function has been designed for. This dialog consists of three parts: 1. Adjusting the amount of files being backuped Here you can tell YaST which directories should not(!) be updated (should be excluded). Predefined are /tmp /dev and /proc; you should add mounted CD-ROMS or NFS-mounted filesystems to this list. The less you want to be backuped the faster it runs since unnecessary compar- isons with package lists are omitted. Using + and - you can add new directories or remove them. Pressing F10 leads to the next step. * * 87 4. YaST - Yet another_Setup_Tool__________________________________________________________________________________________________* *_______________________________________@ | * * | ________________________________________________________________________________________________________* *________|||| | * * | || * * || | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | |_______________________________________________________________________________________________________* *________ | Figure 4.21: Editing the configuration file with YaST 2. Searching Now YaST searches for files which should be backuped. The amount and size of packages found are updated while searching. After this has been done there will be a list with all the files that have been found. Here you still can deselect files by using the SPACE key. 3. Entering the command Here you decide how those files are going to be saved. You can give archive names, options and more. Of course this backing-up mechanism can only work if the dates of the files have not been changed otherwise. Furthermore this function requires quite an amount of RAM. Filenames of an ordinary CD take up to 6 MB of RAM; of course you need enough free disk space to save the archive. Compressing the archive will lead to a file reduced in size (approximately half of the original). The best way to do backups is a tape anyway. 88 _____________________________________________________________________________________________________4.7.__Administration__________* *_______________ |||||||| | _______________________________________________________________________________________________________________________|||| |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | ||||||||_______________________________________________________________________________________________________________ | Figure 4.22: Backup with YaST 89 4. YaST - Yet another_Setup_Tool__________________________________________________________________________________________________* *_______________________________________@ 90 Chapter 5 Updating the system and package management 5.1 Updating S.u.S.E. Linux S.u.S.E. Linux offers you the option of updating an existing system without having to reinstall everything. Here you have to distinguish between an up- date of one or more packages or of the whole system (perhaps because your older system still is a.out ). 5.1.1 Updating base system Since updating changes all central parts (e. g. , libraries) of the system this can't be done while the system is up and running. That in particular means that you have to start an update like a first-time installation (see section 3.15). Under `Installation settings' you choose `Assigning partitions / filesystems' and specify your root partition; YaST reads the existing /etc/fstab and mounts your filesystems. After going back to the main menu you should choose the item `Updating your system' (section section 4.5.3). Now your old system is analyzed by YaST. If it still is an a.out system YaST automatically converts it into + ELF. After that all central configuration files of the system are updated; of course YaST creates backup files of those files that have been changed since the last installation. If the base system has been updated you are lead to YaST's special update- mode where you can decide which packages to update. YaST offers two lists; first, those packages YaST has recognized and decided that an update could be useful. The second list shows you a list of packages where this is not as simple (perhaps the old package did work without any information about their version). In those two lists you can freely decide which package should be updated. If you start the update all selected packages are replaced by the new ones saving all files that haven been changed since the last installation. After having completed this task you should proceed as with a normal instal- lation; perhaps you should install a newer kernel as the older one might not recognize the ELF format. * * 91 5. Updating the system_and_package_management_____________________________________________________________________________________* *_______________________________________@ 5.1.2 Update configuration S.u.S.E. Linux uses an enhanced concept for configuring and updating your system (see chapter 12 und section 12.5); this is far more complex (than Slackware ) but extremely powerful and makes future updates really easy to handle. As automatically converting old configuration files is not possible, you have do some adjustments after having updated. You will be rewarded for doing so with a very flexible configuration system which enables you to change system wide administrations without having to reboot your machine. Furthermore a base for future updates is generated, as all settings relevant to the system are put into one single file (/etc/rc.config). 5.1.3 Updating of single packages With S.u.S.E. Linux you can update single packages whenever you want. In YaST's package list (see section 4.6.8) you can move around as you please. If you select a vital package which is necessary for the system to run you will be warned by YaST; such packages should be replaced only in update-mode. For example, lots of packages contain shared libraries which might be in use when you want to run an update. Updating a running system will inevitably lead to a malfunctioning system. 5.2 From version to version In the following sections problematic details are listed which have been changed from release to release. In this overview changes concerning con- figuration files will appear or whether well know programs will show a sur- prising behavior. But only those things are listed which might arise problems for the administrator or for the user at their daily work. This list is probably incomplete; please, also consult the SDB - to be found in package sdb, series doc (cf. section 2.1.3, page 12). 5.2.1 From an ealier version to 4.x The init scripts in /etc/rc.d which used to bring up the system have been replaced by a system that is able to organize the different runlevels in accor- dance to System V . Since the file system standard does not allow executables under /etc, the init -scripts have been moved to /sbin/init.d. Automatic conversion of old configuration files isn't possible in all case; therefore you probably will have to edit some files by hand. The goal ist, that from now on system relevant configurations could be done without booting. Since all vital settings will be stored in a system wide config file (/etc/rc. config) coming updates will be easy to perform. 5.2.2 From 4.x to 5.0 Problems and special issues: 92 _______________________________________________________________________________5.3.__RPM_-_The_package_manager______________* *______________________ o Package management changed from TGZ to RPM (cf. section 5.3). o News Bash . o startx is not started in the background any more. o To start Samba set the variable START\_SMB=yes in /etc/rc.config. o System relevant cron jobs are listed in /etc/crontab now. o New group `dialout'; put all users in this group, who are allowed so execute "dialout" programs (minicom , pppd , etc.). o The file system of the rescue system is modelled like usual layout of the running system. 5.2.3 From 5.0 to 5.1 Problems and special issues: Nothing known right now. o You will find the package description on the first CD in directory /docu. 5.3 RPM - The package manager With S.u.S.E. Linux 5.0 RPM (rpm ) ("Red Hat Package Manager") was intro- duced. This way life will become easier for all involved people: for the user, for the system administrator, and last, but not least for the package builders. With the mighty RPM database detailed information about the installed soft- ware packages is comfortably available. rpm mainly operates in three different modes: compile installable packages from socalled pristine sources, install those binary packages resp. de-install or update them in a clean manner, query the RPM database or an RPM package. In the following sections the second and the third mode are described only1 . Installable RPM archives are packed in a special binary format; these archives consist of (program) files to install and certain meta-information, which will be used by rpm to configure the software package or which will be stored in the database for documentation purposes. RPM archive normally have the extension .rpm. 5.3.1 Managing packages: install, update and de-install Normally, installing an RPM archive is incredibly easy: root@earth: # rpm -i .rpm With this standard command, the package will be installed only, if its depen- dency requirements are met and if it does not conflict with another package. With an error message rpm "cries" for packages which you also have to in- stall to fulfill the dependencies. In the background the database takes care that no conflict will arise: a file can belong to one package only. With dif- ferent options you can force the RPM system to ignore these defaults_but please, make sure that you know what you do: otherwise you will risc the integrity of the system and you might loose the feature to update the system straightforwardly! 1 ____________________________________________________ Concernig the first issue (build mode), cf. [Bailey, 1997 ] or manpage of rpm (man rpm ). * * 93 5. Updating the system_and_package_management_____________________________________________________________________________________* *_______________________________________@ Use the option -U resp. --upgrade to update a package. This way rpm removes the files of the old version and immediately installs the new files; at the same time it tries to handle the configuration files with care_slightly simplified: o If a configuration file was not changed by the system administrator, rpm will replace the file silently; no action by the system administrator is necessary. o If a configuration file was changed by the system administrator before the update, rpm will save the changed file with the extension .rpmorig or .rpmsave (backup file) and install the version from the new package; but only then, when the originally installed file and the new version are dif- ferent. If this happens, it's probably necessary that you have to compare the backup file (.rpmorig or .rpmsave) with the newly installed file and to make your changed again. After doing so, please delete all .rpmorig and .rpmsave files, to avoid touble if the next update will take place in the future : : :2 Obviously, the switch -U is more than just de-installing (-e ) and installing (-i ); please, use -U as often as possible. ________________________________________________________________________________________________________* *________|| | After every update you have to check all backup files created by rpm ; these * * |||| | are your old configuration files. If necessary take over your customiza- * * ||| | * * ||| | tions from the old .rpmorig or .rpmsave files to the new configuration * * ||| | files; after this process delete the files with the extension .rpmorig resp. * * ||| | * * ||| | .rpmsave. * * ||| |_______________________________________________________________________________________________________* *________|||_ To remove a package enter the command: root@earth: # rpm -e rpm will only erase the package if there is no open dependency any more; this way e. g. , it theoretically isn't possible to de-install an old libc as long as another program still needs it to work properly_the RPM database watches over it. 5.3.2 RPM queries With option -q ("query") rpm initiates queries. With this option it is pos- sible to inspect RPM archive (add the option -p ) and to query the installed database. Several switches are available to specify the wanted information (cf. table 5.1). _______________________________________________________________________________________________________* *_________ -i Summary information _______________________________________________________________________________________________________* *_________ ____________________________________Table_5.1:_to be continued... 2 rpm will choose .rpmorig, if the file was unknown for the RPM database until now_ otherwise .rpmsave; in other words: .rpmorig files will be created while updating from a foreign format to RPM and .rpmsave while updating one RPM package with another RPM package. 94 _______________________________________________________________________________5.3.__RPM_-_The_package_manager______________* *______________________ ______________________________________________________________________________________________________________________ -l File list -f + Query a package owning (specify with full path!) -s File list with state information (implies -l ) -d Documentation files (implies -l ) -c Configuration files (implies -l ) --dump File list with complete details ( to be used with -l , -c or -d benutzen!) --provides Capabilities the package provides --requires , -R Capabilities the package requires --scripts (De-)installation scripts (pre/post install/uninstall) ______________________________________________________________________________________________________________________ Table 5.1: The most important query options (-q [-p ] : : :) For example, the command root@earth: # rpm -q -i rpm displays the following information: Name : rpm Distribution: S.u.S.E Linux Version : 2.4.1 Vendor: S.u.S.E. GmbH Release : 1 Build Date: Wed Jun 18 14:46:53 1997 Install date: Sat Jun 21 12:01:21 1997 Build Host: Fibonacci.suse.de Group : Source RPM: rpm-2.4.1-1.src.rpm Size : 1365662 Packager : feedback@suse.de URL : (none) Summary : Red Hat Package Manager Description : RPM is a powerful package manager, which can be used to build, install, query, verify, update, and uninstall individual software packages. A package consists of an archive of files, and package information, including name, version, and description. Option -f does only work, if you specify the complete file name (full path); specify as many file name as you want; e. g. : rpm -q -f /bin/rpm /usr/bin/wget rpm-2.4.1-1 wget-1.4.5-2 If you do know only a part of the file name, you will have to use a short shell script like this: #! /bin/sh for i in `rpm -q -a -l | grep $1 `; do echo \"$i\" is in package: rpm -q -f $i echo "" done With the help of the database you can do verify checks; these checks will be initiated with the option -V (the same as -y or --verify ). With this option * * 95 5. Updating the system_and_package_management_____________________________________________________________________________________* *_______________________________________@ rpm will show all files of a package which are changed since the first instal- lation. rpm knows eight character symbols to give some hints concerning the kind of the change: _______________________________________________________________________________________________________* *_________ 5 MD5 check sum S File size L Symbolic link T Modification Time D Major and minor device numbers U Owner G Group M Mode (permissions and file type) _______________________________________________________________________________________________________* *_________ Table 5.2: Verify options In the case of configuration files the character c will be printed; example, if you have changed /etc/wgetrc from the package wget: root@earth: # rpm -V wget S.5....T c /etc/wgetrc The files of the RPM database are placed under /var/lib/rpm. If the par- tition /usr has a size of 500 MB, the database can occupy nearly 20 MB; especially after a complete update. If the database is much bigger than ex- pected, it should help to rebuild the database with the option --rebuilddb ; before rebuilding make a backup copy of the old database. Furthermore the cron script cron.daily makes gzipped copies of the database and stores them under /var/adm/backup/rpmdb; the number of copies is controlled by the variable (default 5 ) in /etc/rc.config. The size of one backup is approximately 2 MB (this value is valid for a 500 MB large /usr partition). Please, take this "waste" of space into account, if you make your decision concerning the size of the root par- tition; if you have an own /var partition, you don't have to worry about all these things. 5.3.3 Tools for RPM archives The Midnight Commander (mc ) is able to "browse" RPM archives and to operate on parts of them. This tool works on an RPM archive as if the archive would be a regular file system. This way you can view HEADER information with F3 and you can copy parts of an archive with F5 . - By the way, right now Emacs doesn't have the facility to do similar operations ;-) xrpm is the name of a new graphical RPM manager; this tool is realized in the Python programming language. xrpm supports actions via FTP. 96 _______________________________________________________________________________5.3.__RPM_-_The_package_manager______________* *______________________ KDE comprehends the tool krpm , a graphical interface under the X Window System to deal with RPM. Looks nice, but at this moment is in the shape of an early development stage. With Alien (alien ) it is possible to convert or install an "alien" binary pack- age. One can try to convert "old" TGZ archives to RPM before installing, this way the RPM database can keep track about such an package while installing. But attention: alien is a perl script and still is "alpha" software, as its author says, nevertheless it has reached version 5.17 (at the moment ). Last, not least, there is YaST : : : * * 97 5. Updating the system_and_package_management_____________________________________________________________________________________* *_______________________________________@ 98 Chapter 6 Notebooks and PCMCIA cards There are several PCMCIA adapters supported by Linux which makes it pos- sible to use PCMCIA adapters on notebooks running Linux. All common cards are supported, such as Intel, Cirrus, Vadem, VLSI, Ricoh and Databook chips. Even those special adapters found in IBM and Toshiba laptops and PCMCIA adapters found in desktops work. For installation PCMCIA cards are only relevant if o you install via + NFS using the PCMCIA networking card. o you install via CD-ROM and the CD-ROM is connected via PCMCIA. o you install onto a SCSI disk which is connected to a SCSI-PCMCIA adapter. 6.1 The concept The PCMCIA support is a kernel-independent subsystem and therefore has not yet been included in the kernel itself. Further development aims at this; most probably PCMCIA support will be included in kernel version 2.2. PCMCIA support is only available via modules (see section 13.2). For main- taining these modules there is a special program, Card-Manager (cardmgr ), whose task it is to supervise the PCMCIA sockets and to load the appropriate modules on demand, or removing it respectively. From this point of view the Card-Manager can be considered a second kernel daemon (kerneld , see section 13.2.2). This concepts also enables you to change cards in a running system without any special precautions. But it also has its disadvantages. Since it is not possible to build monolithic kernels with PCMCIA support there are some restrictions as far as hardware is concerned. Well, it is not possible to install Linux entirely on a SCSI disk which is connected to a net laptop via such an SCSI controller. To be precise: Installation is possible but booting the system is rather tedious since the SCSI driver's adapter can only be loaded after booting. Use of JAZ-Drives on parallel port is not such a problem, since this driver can be compiled into the kernel. Well, you will not be able to use the parallel port for (e. g. ) a printer (it's our aim to make this possible in future versions of S.u.S.E. Linux). The only solution at the moment is a special form of RAM disk which enables the system to launch programs before booting the system (see section 13.4.5). * * 99 6. Notebooks and PCMCIA_cards_____________________________________________________________________________________________________* *_______________________________________@ For addressing PCMCIA hardware you can load the appropriate modules before the "real" booting. Realisation is a rather complex matter and can not be discussed in greater details in this chapter1 . The special S.u.S.E.boot-disk uses this kind of ramdisk; so if you want to install the system onto a SCSI disk connected via an PCMCIA-SCSI adapter booting with this disk is the only choice you have. Since this is rather dis- turbing and system tasks like upgrading a kernel are an immense waste of time we highly recommend to install at least the root partition of an existing (E)IDE disk; the amount of space needed is approximately 20 MB. 6.2 Installation as described above the S.u.S.E. boot disk (from S.u.S.E. Linux 4.4.1 or later) is able to load the Card-Manager in advance and to load the required modules. Hereby automatically recognizing the PCMCIA chipset. For installing Linux on a laptop, this should be your first choice. In linuxrc you select `Kernel modules (hardware driver)' and then `Load PCMCIA modules'. linuxrc now tries to recognize the PCMCIA chipset, loads the PCMCIA kernel modules and starts the Card-Manager. The software now takes some seconds for analyzing the hardware and for loading the necessary modules (please be aware that for memory reasons only those modules are on disk that really are relevant for an installation; modules for modem cards are not included). As soon as linuxrc announces that it has successfully launched the Card- Manager you can load the installation system directly from CD and com- mence "as usual" by starting YaST (as described in section 3.16). On the other consoles you can have a look whether all cards haven been installed correctly (e. g. using Alt + F3 or Alt + F3 ). _____________________________________________1 If you are interested: Information can be found at ramdisk.txt and initrd.txt in the directory: /usr/src/linux/Documentation. 100 Chapter 7 Kernel parameters 7.1 Drivers in the kernel There is a wide variety of PC hardware. Now, to access this hardware you will need the appropriate driver. The Linux kernel has to cooperate with this driver and your hardware. Generally there are two ways of integrating drivers into your system: o These drivers can be compiled directly into the kernel. These kernels are referred to as monolithic kernels. Monolithic kernels e. g. are on our CD for creating boot-disks for rather exotic hardware. Some drivers have to be compiled directly into the kernel for accessing them at all. o Drivers can be loaded into the system at runtime (on demand). This should be used since it reduces the amount of code being loaded at a given time; thus it saves a lot of memory. Our S.u.S.E. boot-disk kernel makes use of modules, thus being able to serve most hardware configurations. Some drivers do not exist as modules yet. Those are all drivers for EIDE- controllers, which therefore are compiled into our kernels on S.u.S.E.- boot-disk. Nevertheless it can happen that a hardware component is not recognized. In such a case you have the advantage of passing parameters to the kernel for "helping" it. With monolithic kernels parameters have to be passed by LILO (or loadlin ); that's why they are normally referred to as LILO parameters. Modular drivers get their parameters via insmod resp. modprobe and start simultaneously. Unfortunately the format of LILO parameters differs to loadlin parameters. That's why they are parted in two groups and shown separately. Some mod- ules have generalized parameters (CD-ROM drives) where you can use the same parameters either with LILO or with loadlin 7.2 Some hints Before we present you the list of parameters some hints on recognizing hard- ware of the driver, passing parameters and booting using the S.u.S.E.boot- disk: o Most drivers can do an autoprobing, i. e. they test different standard ad- dresses where this hardware usually can be found. By doing so it can * * 101 7. Kernel parameters______________________________________________________________________________________________________________* *_______________________________________@ happen that the driver hits an address that is not his but initializes it any- way. Then most certainly the machine will hang. o Some modules can even be loaded successfully if their hardware is not installed at all (this mainly applies to 3Com networking card drivers. Anyway, you should try autoprobing first. Unused drivers can be removed easily. Hardware that is not automatically recognized must then be started using the appropriate parameters. o Some driver are not available as modules. These are (e. g. ) all EIDE drivers. So, if you only have EIDE components (hard disk or CD-ROM) you do not need to load modules when booting with S.u.S.E. boot-disk. o At last there are some hardware components that do not only have one driver but plenty ( NCR 53C810 , Ultrastor ). As far as we know there is no significant difference between these two Ultrastor drivers. The BSD driver for NCR53C810 also supports other NCR53C8xx-products (e. g. 53C875), whereas the old NCR driver is the only one that recognizes CD writers. Just try which one suits your needs. 7.3 The parameters 7.3.1 Notation and meaning Below you will see an alphabetical list of kernel parameters and the corre- sponding devices. The following parameters always have the same meaning: _______________________________________________________________________________________________________* *_________ hexadecimal portaddress, (e. g. 0x300) Interrupt to access the device (e. g. 7) DMA channel of the device (e. g. 1) , hexadecimal memory allocation for shared memory _______________________________________________________________________________________________________* *_________ Table 7.1: Frequently used variable names for kernel parameters Now we mainly describe the parameters that are necessary for a successful installation. But there are other parameters for special aims and purposes. Please keep in mind that upper cases are significant! A complete introduction to possible kernel parameters is to be found (after installation) under BootPrompt-HOWTO in /usr/doc/howto. 7.3.2 LILO-parameters These listed parameters can only be entered at the LILO prompt (or via loadlin ) to the kernel. If you plan to install this driver as a module please have a look at parameters in the next section (section 7.3.3). 102 ____________________________________________________________________________________________________7.3.__The_parameters______* *____________________ _______________________________________________________________________________________________________________________ |||||||| All parameters have to be entered directly one after another, separated by ||| | ||||||||| commas! There should not be a blank (whitespace) between the parame- || || |||||||| ters! ||| | ||||||||_______________________________________________________________________________________________________________||_ o Adaptec AHA-1520 / 1522 / 1510 / 1515 / 1505 SCSI-Host adapter aha152x=,,[,[,]] _________________________________________________________________________________________ ||__Variable____________________________________________|__Values_/_Meaning__|____________ || (SCSI-ID of the host adapter) | 0 , 1 | || (reconnect) | 0 , 1 | ||__________(parity)_________________________________|0_,_1_______________________|_ This driver serves for running many low-cost SCSI controllers. All sound- cards with a SCSI controller (except for Pro Audio Spectrum ) have an adaptec chip installed and thus can be accessed. For all not-original 152x the fourth parameter (RECONNECT ) seems to be necessary. For most types it has to be set to `0'; only the AHA2825 needs `1'. Example: aha152x=0x300,10,7 o Adaptec AHA-1540 / 1542 SCSI-Host adapter aha1542=[,,[,]] ___________________________________________________________ ||__Variable_____________|___Values_/_Meaning__|____________ || |2..15 | || |1..64 | ||_______________|5,6,7,8,10________________|__ Example: aha1542=0x300 o Adaptec AHA-274x / 284x / 294x Host adapter aic7xxx=, ___________________________________________________________________________________________ ||__Variable_________________|__Values_/_Meaning_______________________________|____________ || e|xtended activates translation mode | || n|o__reset disables resetting | || (|engl. reset) the SCSI-bus while | ||______________________________i|nitializing_the_hostadapter__________________|___________ Example: aic7xxx=no__reset , If the machine hangs at SCSI reset. The SCSI hostadapter Adaptec 2920 is supported by the Future Domain driver (section 7.3.2, page 107). o AdvanSys SCSI-Host adapter advansys=,,..., Example: advansys=0x110,0x210 This example tells the kernel to search for the AdvanSys host adapter at the given addresses. * * 103 7. Kernel parameters______________________________________________________________________________________________________________* *_______________________________________@ o AM53/79C974 SCSI-Host adapter AM53C974=,,, ____________________________________________________________________________________ |__Variable_____________|_____Values_/_Meaning_____________________________|_________ | | the host adapters SCSI-ID (often 7 ) | | | the devices SCSI-ID (often 0..7 ) | | | 3,5,10 MHz/s max. transfer rate | |___________________|_Transfer_mode;_0__=_asynchron__________|______________ If the host adapter seems to "swallow" some packages, reduce the max- imum transfer rate for this device (e. g. the first CD-ROM SCSI drive /dev/scd0 with ID 5) on the SCSI bus with: Example: AM53C974=7,5,3,0 o Amount of SCSI devices per ID max__scsi__luns= ____________________________________________________ |__Variable_______|_____Values_/_Meaning__|__________ |______________|1..8_______________________|_ Example: If the first LUN (logical unit number) should be used explicitly you have to set the parameter max__scsi__lun=1 o ATAPI-CD-ROMam (E)IDE-Controller hd=cdrom hd=serialize _______________________________________________ |__Variable__|_____Values_/_Meaning__|__________ |______________|a,_b,_c,_d_______________|__ where: ______________________________________________________________ |__Variable__|_____Values_/_Meaning_________________|__________ | a |Master on 1. IDE controller | | b |Slave on 1. IDE controller | | c |Master on 2. IDE controller | |__d______________|Slave_on_2._IDE_controller_____|___________ Example: An ATAPI CD-ROM as master on secondary IDE controller is set by hdc=cdrom . o Aztech CDA268-01 CD-ROM aztcd=[,0x79] The value 0x79 only has to be set with unknown Firmware-version. Example: aztcd=0x320 Since probing this driver takes extremely long it is recommended to pass aztcd=0 if it is not needed. With our installation kernels disabling also can be done by aztcd=off . 104 ____________________________________________________________________________________________________7.3.__The_parameters______* *____________________ o BusLogic SCSI-Host adapter BusLogic= Example: BusLogic=0x300 This host adapter can be configured by far more parameters but these only serve for fine tuning the device and are described in /usr/src/linux/drivers/scsi/BusLogic.c. o EIDE controller chipsets Some EIDE controllers do have faulty chipsets or cause problems if the secondary controller should be used. In the meantime there exists support for many of those chipsets in the kernel (section 13.4.5, page 188) which additional have to be activated by a kernel parameter. The following chipsets can be configured: _____________________________________________________________________________________________________________ CMD 640 This chipset can be found on many mother boards. Since it contains many bugs the kernel offers a special support which recognizes the chip and bypasses the problems. Moreover, in some cases use of the secondary controller is only able by using this special code. In PCI systems this chip automatically is recognized; for VLB systems the following parameter is needed: ide0=cmd640__vlb RZ 1000 This chip is used on many mother boards that use the Neptun chipset and it is buggy. If sup- port for this chip is activated the system works a little slower but reliable. An additional acti- vation via kernel parameter is not necessary. DTC-2278 Only activating this driver via ide0=dtc2278 makes it possible to use the secondary con- troller. Holtek HT6560B For activating the secondary controller the fol- lowing parameter is needed: ide0=ht6560b QDI QD6580 If this driver is activated it enables a higher speed: ide0=qd6580 UMC 8672 For activating the secondary controller the fol- lowing parameter is needed: ide0=umc8672 ALI M1439/M1445 For activating the secondary controller the fol- lowing parameter is needed: ide0=ali14xx PROMISE DC4030 For activating the secondary controller the fol- lowing parameter is needed: ide0=dc4030 . CD-ROM and tapes on secondary controller are not supported yet. _____________________________________________________________________________________________________________ Table 7.2: Special EIDE chipsets * * 105 7. Kernel parameters______________________________________________________________________________________________________________* *_______________________________________@ o Ethernet networking cards ether=,[,[,...]], The different parameters for to do have different meanings for several drivers. Mostly only two parameters are needed, where the first is the start address and the second is the end addressed of shared memory. The first non-numerical argument is treated as the name. ____________________________________________________________________________________________________* *______ Used interrupt; 0 for autoprobing Port address; 0 for autoprobing Start address for shared memory; some drivers use the 4 lowest bits for the debug level; the Lance uses them for its DMA channel. End address for shared memory; the 3COM 3c503 driver uses this parameter to differ between internal and external transceivers. Cabletron E21XX card uses the lowest four bits, for selecting a media. The interface's name (normally eth0 ) ____________________________________________________________________________________________________* *______ Table 7.3: Variable names for ethernet networking cards The main reason for passing this parameter is to let the kernel recognize more than one networking card since only the first card is searched by default. This can easily be done by: ether=0,0,eth1 Please notice that by passing 0 for IRQ and address the driver is explic- itly told to launch autoprobing which means probing several addresses independently. o Floppy disk drives floppy=,,cmos _____________________________________________________________________ |__Variable_____|_____Values_/_Meaning______________________|_________ | |0, 1, 2, 3 | | |0 - unknown or not recognized | | |1 - 5 1/4" DD, 360 KB | | |2 - 5 1/4" HD, 1,2 MB | | |3 - 3 1/2" DD, 720 KB | | |4 - 3 1/2" HD, 1,44 MB | | |5 - 3 1/2" ED, 2,88 MB | |____________________|6__-_3_1/2"_ED,_2,88_MB_____________|__________ floppy= For the following values may be set (table 7.4): 106 ____________________________________________________________________________________________________7.3.__The_parameters______* *____________________ _____________________________________________________________________________________________________________ all__drives more than two floppy disk drives asus__pci denies access to 3. and 4. floppy disk drive daring only with reliable controllers - en- hances performance 0,daring Opposite of daring ,two__fdc If the value is omitted while using a secondary floppy controller a port address of 0x370 is set. thinkpad IBM Thinkpad -machines 0,thinkpad no Thinkpad-machine unexpected__interrupts Show warnings if something unex- pected happens. no__unexpected__interrupts and... L40SX This value reaches the opposite of unexpected__interrupts _____________________________________________________________________________________________________________ Table 7.4: Kernel parameter values for floppy o Future Domain TMC-16x0-SCSI-Host adapter fdomain=,[,] ______________________________________________________________________ ||__Variable__|___Values_/_Meaning_________________________|___________ ||________|___The_host_adapters_SCSI-ID_0..7______|_______________ This driver also servers the Adaptec 2920 host adapter. Example: fdomain=0x140,11,7 o Future Domain TMC-885/950-Hostadapter tmc8xx=, Example: tmc8xx=0xca000,5 o Goldstar R420-CD-ROM drive gscd= o Iomega ZIP Drive on parallel port ppa=[,[,[,]]] ________________________________________________________________________________________________________ ||__Variable_______|___Values_/_Meaning______________________________________________________|___________ || P|ort address | || W|aiting time while transferring data (micro seconds) | || W|aiting time for other operations | ||_____________i|f_1__use_4_bit_mode____________________________________________________|_______ * * 107 7. Kernel parameters______________________________________________________________________________________________________________* *_______________________________________@ o Logitech bus mouse bmouse= o Mitsumi-CD-ROM drive mcd=,[,] ____________________________________________________________________________ |__Variable____|____Values_/_Meaning______________________________|__________ |____________|Value_for_waiting_time_while_starting__|________________ You can vary the parameter between 0 and 10 if the CD-ROM drive does not react fast enough on system requests ("timeout") and therefore does not find the root image while installing. Example: mcd=0x300,10,5 o Mitsumi-CD-ROM drive (Multisession) mcdx=, Example: mcd=0x300,10 o Mozart interface isp16=,,, _______________________________________________________________________ |__Variable____|____Values_/_Meaning__________________________|_________ |____________|Sanyo,_Panasonic,_Sony,_Mitsumi__|_________________ This driver is responsible for CD-ROM drives connected to either of ISP16 , MAD16 or Mozart . The value of follows the inter- face plug where the CD-ROM is connected to the sound card. Example: isp16=0x340,10,3,Sony o NCR 5380 SCSI host adapter family ncr5380=,, ncr5380=0x340,10,3 o NCR 53c406a SCSI host adapter family ncr53c406a=[,[,]] ___________________________________________________________________________ |__Variable_________|_____Values_/_Meaning________________________|_________ |______________|_0,_if_no_fast_PIO_mode_is_wanted__|______________ Example: ncr53c406a=0x340,10,0 o Optics Storage 8000 AT CD-ROM drive optcd= Example: optcd=0x340 o Parallel-Port CD-ROM drives bpcd= Example: bpcd=0x37f 108 ____________________________________________________________________________________________________7.3.__The_parameters______* *____________________ o Philips CM206 CD-ROM drives cm206=, Example: cm206=0x340,10 o Pro Audio Spectrum 16 - SCSI host adapter pas16=, On Pro Audio Spectrum 16 there is an host adapter whose settings can propagated to the kernel by this parameter. Example: pas16=0x340,10 o reboot mode(at leaving Linux) reboot= here you have the following values for ; ________________________________________________________________________ ||__Variable__|___Values_/_Meaning__________________________|____________ || warm W|arm reboot (no memory check) | || cold C|old reboot (with memory check) | || bios B|IOS reboot | ||__hard__________C|PU_crash_reboot_(triple_fault)_______|______________ Example: reboot=cold boots the machine after shutdown as if the reset button has been pressed. o Protecting memory segments(Reserve) reserve=,,...,, By means of this parameter you can reserve memory space for hardware which does not like being autoprobed and (e. g. ) reacts with a system crash. Example: A "sensitive" networking card can be protected from being autoprobed and initialized by entering: reserve=0x330,32 ether=5,0x330,eth0 In this example the networking card has a 32 bit data bus which starts at address 0x330 and uses interrupt 5 . For the full description of networking card parameters see section 7.3.2, page 106. o Passing root partitions root= ____________________________________________________________________________ ||__Variable_____________|___Values_/_Meaning____________________|___________ ||________________|e._g./dev/hda1,_/dev/sdb5__|__________________ Example: root=/dev/hda5 boots the kernel and tries to load the root partition from the first logical drive in the extended partition on the first (E)IDE hard disk. * * 109 7. Kernel parameters______________________________________________________________________________________________________________* *_______________________________________@ o Sanyo CD-ROM drive sjcd= Example: sjcd=0x340 o SCSI streamer(Streamer) st=,[,] _________________________________________________________________________________________________ |__Variable_____________|_____Values_/_Meaning__________________________________________|_________ | | Size of buffer (Amount of blocks of 1 KB) | | | Write threshold (Amount of blocks of 1 KB) | |______________________|_Maximal_amount_of_buffers;_optional_(e._g._2)___|__________________ Example: o Seagate ST01/02 SCSI host adapter st0x=, Example: st0x=0xc8000,5 o Sony CDU 31/33 A cdu31a=, Autoprobing of this driver has been removed from the kernel, thus you have to set any parameters explicitly Example: cdu31a=0x340,5 If there is no interrupt assigned to the drive and if you only can access it via polling mode you have to enter 0 for the IRQ. Example: cdu31a=0x340,0 If this drive is (e. g. ) connected to a Pro Audio Spectrum card this could resemble the following line: Example: cdu31a=0x1f88,0,PAS o Sony CDU 535 sonycd535=, Example: sonycd535=0x340,10 o Size of main memory (RAM) mem= Size of memory can be entered in three different ways; either in bytes, kilobytes or megabytes. These examples show the different ways using a size of 96 MB. Examples: mem=96M mem=96000k mem=96000000 110 ____________________________________________________________________________________________________7.3.__The_parameters____* *______________________ o Soundblaster Pro 16 MultiCD sbpcd=, ___________________________________________________________________________ ||__Variable____|__Values_/_Meaning____________________________|____________ ||___________L|aserMate_______,_SPEA____,_SoundBlaster_____________|_ Example: sbpcd=0x340,10 Since probing this driver takes extremely long, it is recommended to dis- able it using the command: sbpcd=0 On S.u.S.E. boot-disks this is also possible using sbpcd=off o Trantor T128/128F/228 SCSI host adapter t128=, Example: t128=0x340,10 o Trantor T130B SCSI host adapter ncr53c400=, Example: ncr53c400=0x340,10 o XT h rd disk controllerparm.xt xd=,,, 7.3.3 insmod -Parameter This section now describes those parameters that can be loaded as modules. If you encounter difficulties while loading a driver (although you have entered the parameters) or if there is no section describing the parameter you have to integrate this driver into a monolithic kernel. Some drivers do not exist as a module yet; some of them recognize your hardware properly only if they are compiled into the kernel. Anyway, it is recommended to try the "module variant" first. If a driver is loaded as a module each and every variable used can be over- written on command line. There is e. g.the variable io in NE2000 driver which specifies the used I/O range. Thus, the correct command for loading this module is (cf. the kernel chapter, section 13.2, page 181): root@earth:/ # insmod ne io=0x300 irq=10 or better with modprobe : root@earth:/ # modprobe ne io=0x300 irq=10 Please be aware that there should be no spaces before or after the "equal". Moreover, hexadecimal values have to be set in the given form (leading `0x'). * * 111 7. Kernel parameters______________________________________________________________________________________________________________* *_______________________________________@ ________________________________________________________________________________________________________* *________|| | If you want to enter more than one parameter they have to be separated * * |||| || by blanks! This is the main difference to entering parameters at the LILO * * |||| | prompt where no blanks should be used within parameters for one driver! * * |||| |_______________________________________________________________________________________________________* *________|||_ The parameters you enter here can additionally be integrated in /etc/conf.modules. Here many parameters can be assigned to a certain module. This is done in one line per module. The line should resemble the following: options = ... where: _________________________________________________________________________________________________ |__Variable_________________|____Values_/_Meaning______________________________________|__________ | |Name of the module without extension .o | | |Parameter 1 | |________________________|Value_that_is_assigned_to_parameter_1__________|________________ An entry for the example NE2000 card might look like: options ne io=0x300 irq=10 Now, we want to list the most important parameters for most modules used. o 3Com 3c501 / 3c503 / 3c505 / 3c507 Networking cards Name of module: 3c501.o,3c503.o,3c505.o,3c507.o _____________________________________ |__Parameter__|______Value_______|____ | io | | |__irq______________|___________| Example: modprobe 3c509 io=0x300 irq=10 o 3Com 3c509 / 3c579 Networking cards Name of module: 3c509.o,3c579.o ___________________________________________________ |__Parameter__|______Value______________________|___ | io | | | irq | | |__xcvr_____________|0_:_intern;_1_:_extern__|_____ Example: modprobe 3c509 io=0x300 irq=10 xcvr=0 o Adaptec AHA-1520 / 1522 / 1510 / 1515 / 1505 SCSI host adapter Name of module: aha152x.o aha152x=,,[,[,]] _____________________________________________________________________ |__Variable__|_____Values_/_Meaning________________________|__________ | |SCSI-ID of the hostadapter; 0 , 1 | | |reconnect; 0 , 1 | |____________|parity;_0_,_1__________________________________|__ Example: modprobe aha152x aha152x=0x300,10,7,1,1 112 ____________________________________________________________________________________________________7.3.__The_parameters______* *____________________ o AdvanSys SCSI host adapter Name of module: at1700.o ______________________________________ ||__Parameter__|____Value_______|______ || io <|addr> | ||__irq_____________<|irq>__________|_ Example: modprobe at1700 io=0x300 irq=10 o Aztech CDA268-01 CD-ROM drive Name of module: aztcd.o ______________________________________ ||__Parameter__|____Value_______|______ ||__aztcd___________<|addr>_________|_ Example: modprobe aztcd aztcd=0x300 o Cabletron E21xx Networking card Name of module: e2100.o ____________________________________________________ ||__Parameter__|____Value______________________|_____ || io <|addr> | || irq <|irq> | || mem <|addr> | ||__xcvr____________0|:_intern;_1_:_extern__|_______ Example: modprobe e2100 io=0x300 irq=10 mem=0xd000 o Digital DE425 / 434 / 435 / 450 / 500 Networking cards Name of module: de4x5.o io=0x _______________________________________________________________________________ ||__Parameter_____|____Value______________________________________________|_____ || bus N|umber of the PCI bus, normally 0 | ||__device-id__________N|umber_of_the_PCI_device________________|______________ This data is shown with new PCI BIOSes at boot time, or you can view them under Linux respectively using: root@earth: # cat /proc/pci Example: modprobe de4x5 io=0x007 o DECchip Tulip (dc21x4x) Networking cards Name of module: tulip.o __________________________________________ ||__Parameter__|____Value___________|______ || io <|addr> | ||__if__port________<|medium>___________|_ * * 113 7. Kernel parameters______________________________________________________________________________________________________________* *_______________________________________@ where can be one of the following: _______________________________________________ |__Variable__|_____Values_/_Meaning__|__________ | -1 | auto | | 0 | 10TP | | 1 | 100TP | |__2_____________|_BNC____________________|____ Example: modprobe tulip io=0x300 if__port=-1 o Digital DEPCA / DE10x / DE20(012) / DE42, EtherWORKS Networking cards Name of module: depca.o _____________________________________ |__Parameter__|______Value_______|____ | io | | |__irq______________|___________| Example: modprobe depca io=0x300 irq=10 o D-Link DE620 pocket adaptor Networking card Name of module: de620.o ___________________________________________ |__Parameter__|______Value_____________|____ | io | | | irq | | | bnc |1 if BNC I/O | |__utp______________|1__if_UTP_I/O___|_____ Example: modprobe de620 io=0x300 irq=10 bnc=1 utp=0 o EtherWORKS 3 (DE203, DE204, DE205) Networking card Name of module: ewrk3.o _____________________________________ |__Parameter__|______Value_______|____ | io | | |__irq______________|___________| Example: modprobe ewrk3 io=0x300 irq=10 o Intel EtherExpress 16 Networking card Name of module: eexpress.o _____________________________________ |__Parameter__|______Value_______|____ | io | | |__irq______________|___________| Example: modprobe eexpress io=0x300 irq=10 114 ____________________________________________________________________________________________________7.3.__The_parameters______* *____________________ o Intel EtherExpressPro Networking card Name of module: eepro.o ______________________________________ ||__Parameter__|____Value_______|______ || io <|addr> | || irq <|irq> | ||__mem_____________<|addr>_________|_ Example: modprobe eepro io=0x300 irq=10 mem=0xd000 o Fujitsu FMV-181/182/183/184 Networking card Name of module: fmv18x.o ______________________________________ ||__Parameter__|____Value_______|______ || io <|addr> | ||__irq_____________<|irq>__________|_ Example: modprobe fmv18x io=0x300 irq=10 o Future Domain TMC-16x0 controller Name of module: fdomain.o ______________________________________________ ||__Parameter__________|____Value_______|______ || setup__called 1| | ||__port__base______________<|addr>_________|_ Example: modprobe fdomain setup__called=1 port__base=0x300 o Goldstar R420-CD-ROM drive Name of module: gscd.o ______________________________________ ||__Parameter__|____Value_______|______ ||__gscd____________<|addr>_________|_ Example: modprobe gscd gscd=0x300 o HP PCLAN+ (27247B and 27252A) Networking card Name of module: hp-plus.o ______________________________________ ||__Parameter__|____Value_______|______ || io <|addr> | ||__irq_____________<|irq>__________|_ Example: modprobe hp-plus io=0x300 irq=10 o HP PCLAN (27245 / 27xxx) Networking card Name of module: hp.o * * 115 7. Kernel parameters______________________________________________________________________________________________________________* *_______________________________________@ _____________________________________ |__Parameter__|______Value_______|____ | io | | |__irq______________|___________| Example: modprobe hp io=0x300 irq=10 o HP 10/100 VG-AnyLAN (ISA, EISA, PCI) Networking cards Name of module: hp100.o _________________________________________ |__Parameter______|______Value_______|____ |__hp100__port__________|__________| Example: modprobe hp100 hp100__port=0x300 o IBM Tropic chipset Token Ring Networking card Name of module: ibmtr.o _____________________________________ |__Parameter__|______Value_______|____ |__io_______________|__________| Example: modprobe ibmtr io=0x300 o ICL EtherTeam 16i / 32 Networking cards Name of module: eth16i.o _____________________________________ |__Parameter__|______Value_______|____ | io | | |__irq______________|___________| Example: modprobe eth16i io=0x300 irq=10 o Iomega ZIP Drive on parallel port Name of module: ppa.o _____________________________________ |__Parameter__|_______Value_______|___ |__ppa__base________|__________| Example: modprobe ppa ppa__base=0x300 o Mitsumi CD-ROM drive Name of module: mcd.o mcd=,[,] _____________________________________________________ |__Variable____|____Values_/_Meaning_______|__________ |____________|Delay_at_startup_time__|_________ You can vary the from 0 to 10 if the CD-ROM drive does not react fast enough ("timeout") thus probably not finding the root image during installation. Example: modprobe mcd mcd=0x300,10,5 116 ____________________________________________________________________________________________________7.3.__The_parameters______* *____________________ o Mitsumi-CD-ROM drive (Multisession) Name of module: mcdx.o modprobe mcdx mcdx=, o Mozart-Soundcard with CD-ROM drive interface Name of module: isp16.o isp16=,,, ________________________________________________________________________ ||__Variable____|__Values_/_Meaning__________________________|___________ ||___________S|anyo,_Panasonic,_Sony,_Mitsumi__|__________________ This driver is responsible for CD-ROM drives connected to either of ISP16 , MAD16 or Mozart . The value of follows the inter- face plug where the CD-ROM is connected to the sound card. Example: modprobe isp16 isp16=0x300,10,1,sony o Novell NE2000 / NE1000 Networking cards Name of module: ne.o ______________________________________ ||__Parameter__|____Value_______|______ || io <|addr> | ||__irq_____________<|irq>__________|_ Example: modprobe ne io=0x300 irq=10 o NI6510 (AM7990 "lance" Chip) Networking card Name of module: ni6510.o _______________________________________ ||__Parameter___|____Value_______|______ || io <|addr> | || irq <|irq> | || memstart <|addr> | ||__memend___________<|addr>________|__ Example: modprobe ni6510 io=0x300 irq=10 o Optics Storage 8000 AT CD-ROM drive Name of module: optcd.o ______________________________________ ||__Parameter__|____Value_______|______ ||__optcd___________<|addr>_________|_ Example: modprobe optcd optcd=0x300 o Parallel-Port CD-ROM drive Name of module: bpcd.o * * 117 7. Kernel parameters______________________________________________________________________________________________________________* *_______________________________________@ ______________________________________ |__Parameter____|______Value______|____ | bp__base | | |__bp__nybble_________|1______________| Example: modprobe bpcd bp__base=0x300 o Parallel Port IP (PLIP) Name of module: plip.o _____________________________________ |__Parameter__|______Value_______|____ | io | | |__irq______________|___________| Example: modprobe plip io=0x300 irq=10 o Philips CM206 CD-ROM drive Name of module: cm206.o cm206=, Example: modprobe cm206 cm206=0x300,irq=10 o Sanyo CD-ROM drive Name of module: sjcd.o _____________________________________ |__Parameter__|______Value_______|____ |__sjcd_____________|__________| Example: modprobe sjcd sjcd=0x300 o SMC Ultra Networking card Name of module: smc-ultra.o _____________________________________ |__Parameter__|______Value_______|____ | io | | |__irq______________|___________| Example: modprobe smc-ultra io=0x300 irq=10 o SMC 9194 Networking card Name of module: smc9194.o _________________________________________ |__Parameter__|______Value___________|____ | io | | | irq | | |__if__port_________|____________| where can be one of the following: 118 ____________________________________________________________________________________________________7.3.__The_parameters______* *____________________ ________________________________________________ ||__Variable__|___Values_/_Meaning__|____________ || 0 |auto | || 1 |TP | ||__2_____________|AUI,_10base2________|________ Example: modprobe smc9194 io=0x300 irq=10 if__port=2 o Sony CDU 31/33 A Name of module: cdu31a.o ____________________________________________ ||__Parameter________|____Value_______|______ || cdu31a__port <|addr> | ||__cdu31a__irq___________<|irq>__________|_ Example: modprobe cdu31a cdu31a__port=0x300 cdu31a__irq=10 o Sony CDU 535 Name of module: sonycd535.o _________________________________________ ||__Parameter_____|____Value_______|______ ||__sonycd535__________<|addr>________|__ Example: modprobe sonycd535 sonycd535=0x300 o Soundblaster Pro 16 MultiCD Name of module: sbpcd.o sbpcd=, where can be one of the following: ________________________________________________ ||__Variable__|___Values_/_Meaning__|____________ || 0 |LaserMate | || 1 |SoundBlaster | || 2 |SoundScape | ||__3_____________|Teac16bit_________________|__ Example: modprobe sbpcd sbpcd=0x300,1 o Western Digital WD80x3 Networking card Name of module: wd.o ______________________________________ ||__Parameter__|____Value_______|______ || io <|addr> | ||__irq_____________<|irq>__________|_ Example: modprobe wd io=0x300 irq=10 * * 119 7. Kernel parameters______________________________________________________________________________________________________________* *_______________________________________@ 120 Chapter 8 LILO - what it does, how it works One of the packages that are part of your Linux system is LILO, the LInux LOader (bootmanager). Former DOS users at first glance might think this is just one more of the all too many "tools and goodies" of the system; but upon closer inspection you will find that LILO plays a crucial role in making working with Linux comfortable. Besides providing the ability to pass potentially needed additional informa- tion about your hardware to the kernel at boot time, the bootloader allows you to have multiple operating systems installed on a single machine and lets you selectively boot one of them from the hard disk. This chapter describes how a bootloader works in general and how to install LILO on your system. 8.1 Booting a PC After turning on your computer, the first thing that happens is that the BIOS (Basic Input Output System) takes control, initializes the screen and key- board, and tests the main memory. Until this task is completed, no external devices or external storage media are known to the system. Once the basic system has finished its internal setup it starts to verify the hardware around it. Date, time, and information about some of the most important external devices is read from the CMOS ( CMOS setup ) settings. After reading the CMOS, the BIOS should recognize the first hard disk (in- cluding details such as its geometry). It can then start to load the operating system (OS) from there. To load the OS, the system loads a 512-byte data segment from the first MBR hard disk into main memory and executes the code stored at the beginning of this segment. The instructions contained there determine the rest of the boot process. This is the reason why the first 512 bytes of the hard disk are often called Master Boot Record (MBR). Even though the whole process is rather complicated and this description is fairly simplistic, it should be clear that up to this point (loading of the MBR), the boot sequence is independent of the installed operating system. * * 121 8. LILO - what it_does,_how_it_works______________________________________________________________________________________________* *_______________________________________@ 8.1.1 Master Boot Record The layout of the MBR has been certified by an independent convention. The first 446 byte are reserved for program code1 . The next 64 byte offer space for a partition table for up to four partitions2 . The last 2 bytes have to contain a special "magic number" (AA55); any MBR which replaces this number by a different number is rejected. The layout of a boot sector follows a certain standard. The first 446 bytes contain executable code, the next data section that contains space for a par- tition table.3 The last two bytes of the 512 byte segment are used to mark a sector as valid boot sector. 8.1.2 Boot sectors Boot sectors are the first sectors on a hard-disk partition4 . They offer 512 bytes of space and are designed to contain code which is able to launch an oper- ating system on this partition. Boot sectors of formatted DOS, Windows, or OS/2 partitions do so (and furthermore contain more basic instructions of the filesystem). On the contrary Linux boot partitions are empty at the very start. A Linux partition can not be started although it might contain a kernel and a valid root filesystem. A valid boot sector follows the traditions and enters the "magic number" of the MBR into the last two bytes. 8.1.3 Booting DOS or Windows 95 The MBR contains information that determines which partition of a hard disk is "active", i. e., which partition should be searched for the operating system to be booted. 5 The executable code in the MBR (first stage of the boot loader) tests whether the marked partition contains a valid boot sector. If this is the case, the second stage of the boot loader can be started from there. DOS system programs can now be loaded and you will see the usual DOS prompt. Under DOS, only primary partitions can be marked active. Therefore you can not use logical partitions inside an extended partition as bootable DOS partitions. 8.1.4 LILO -Here you go... The Linux boot loader is usually installed in the MBR. (Details in section 8.3.4 nd section 8.4). When started, LILO already has access to both real-mode hard disks and due to its installation is able to write all data to the raw-device6 without information on partitioning. Because of this, operating systems can _____________________________________________1 The code itself _ and its capabilities _ do rely on the system that created the MBR 2 Without this partition table no filesystem (MS-DOS: drives) can exist; i. e.you cannot use the hard disk 3 You cannot use a hard disk without a partition table in its MBR under DOS. 4 except for the extended partition which serves as a "container" for other partitions 5 This implies that DOS has to be installed on the first hard drive. 6 A raw device is a device that is accessed directly without using a filesystem 122 _________________________________________________________________________________________8.2.__Different_boot_concepts______* *______________________ be booted from the first as well as from the second hard drive. The entries in the partition table that the standard DOS MBR uses to mark the active partition are ignored when using LILO in MBR. An important difference to the standard DOS boot sequence is that you can select either of the installed systems at boot time when using LILO. After loading the MBR into memory, LILO is started and asks the user to select one of the installed operating systems (section 8.3.2). 8.2 Different boot concepts The most simple boot concept affects only one machine with one operating system installed. A widely spread configuration is DOS or Windows 95 as the only system installed. The boot sequence for this case has already been outlined. A similar concept can be used for Linux as well7 ; that is, if Linux is the only operating system being used. In this case one could theoretically skip the installation of LILO. The big disadvantage of doing this is that you can't pass additional parameters to the system kernel at boot time. As soon as there is more than one operating system installed there are a couple of boot concepts. 8.2.1 Booting another OS from a floppy disk The first OS can be booted from the hard disk; by use of boot-disks other operating system can be loaded alternatively o Requirements:The floppy drive is bootable o Advantage: You can skip the potentially tricky boot-loader installation o Disadvantage: You have to make sure that you are not running out of working boot-disks. 8.2.2 Boot chaining of additional systems The same OS is always booted and others can optionally be started from within the first OS. o Requirements:Adequate programs for chain booting of operating systems are available. o An example is the loading of Linux from DOS via loadlin or starting a NetWare server from DOS with server.exe . 8.2.3 Installing a bootmanager Theoretically, this allows you to use an arbitrary number of operating systems on a single machine. The choice of systems is done at boot time. Changing operating systems requires a reboot. o Requirements:The chosen boot manager works smoothly with all installed operating systems. 7 ____________________________________________________ You would have to write the Linux kernel directly onto a "raw" partition and launch this from MBR; this is rather uncommon, however * * 123 8. LILO - what it_does,_how_it_works______________________________________________________________________________________________* *_______________________________________@ o An example: for (at least under certain circumstances) an cooperating bootmanager are OS/28 or the DOS boot loader boot.sys . The following chapter describes installation and configuration of a boot man- ager, using the Linux boot manager LILOas an example. A complete descrip- tion of LILO's abilities can be found in [Almesberger, 1994 ]9 8.3 LILO custom fit 8.3.1 What is LILO? LILO is a versatile boot manager; it can load and launch the following com- mands at system startup: o Boot sectors of partitions (starting an operating system from this partition) _ same as with other bootmanagers_, o Linux kernel (starting Linux); this can not be done by most boot man- agers. Furthermore it offers the important feature of passing a command line to the kernel. For security reasons this can be protected totally or partially by a password. 8.3.2 How to boot with LILO? When LILO is launched it displays the text LILO and a greeting message (the latter you have entered during installation). Thereafter a command prompt appears boot: Here you select your operating system by entering its name, which is directly started up afterwards. The name of the operating systems also has been set by you during installation. You can now pass a parameter line to the Linux kernel. You can get a list of all names (of operating systems) by pressing 7 ! . 8.3.3 The components of LILO The whole LILO-machinery includes the following components10 : o The beginning _ the "first step" _ of the LILO code in a boot sector (to be described later) which activates the system start. o the LILOcode (its "heart"). Localised in /boot/boot.b. _____________________________________________8 more in section 8.6.3 9 this file can be printed by entering root@earth:/usr/doc/packages/lilo # touch user.tex root@earth:/usr/doc/packages/lilo # make all root@earth:/usr/doc/packages/lilo # lpr user.dvi 10 Besides: the boot sectors installed by LILO contains a typical virus boot sequence. Therefore it is not astonishing that DOS-virus scanners claim to have found a AIRCOP boot-sector-virus in files as /boot/any_b.b or /boot/any_d.b 124 ____________________________________________________________________________________________________8.3.__LILO_custom_fit___* *______________________ o a map file where LILO enters the location of Linux kernels and other data during its installation. Normally: /boot/map o The different Linux kernel and boot sectors that LILOshould offer. _______________________________________________________________________________________________________________________|| |||||||| Any write access (even through file movements) on any of these files turns ||| | |||||||| the map file corrupt thus requiring you to reinstall LILO (section 8.4.1). || | |||||||| || | |||||||| This mainly applies if you change to another kernel. || | ||||||||_______________________________________________________________________________________________________________||_ 8.3.4 Where LILO can be installed This is referring to the mentioned above "first step" of LILO. Before going into detail we would like to show the general restrictions: _______________________________________________________________________________________________________________________ |||||||| Any part of LILO has to be on the first 1024 cylinders on hard disks! || | ||||||||_______________________________________________________________________________________________________________||_ This is because only these are cylinders are available at BIOS system start. With older BIOSes and IDE disks one can be restricted to either of the first two hard disks (/dev/hda and /dev/hdb). This should be considered before first time installation (section S1) - it is too late afterwards and causes you lots of additional work. More detailed information please find in section 8.7.3. For the "first steps" there are three possibilities: o On a disk This is the most secure but also the slowest alternative for booting with LILO(see section 8.5.2). For those of you who do not want to change boot sectors please (now) choose this alternative. o In boot sector of a primary Linux partition on the first hard disk. This leaves the MBR untouched. Before being bootable it has to be marked active with fdisk . If Linux is fully installed on logical drives or partitions on the second hard disk there is only the boot sector of the second drive left (if there is one). Linux fdisk can even activate such a partition. If you like to boot multiple systems from hard disk this is quite annoying; every time you want to boot you have to activate the corresponding boot sector beforehand. The next two variants are much more comfortable. o In Master Boot Record This variant offers the highest flexibility. More- over this is the only alternative if all of your Linux partitions reside on the second hard disk and there is no extended partition on the first disk. Every setting of the MBR should be edited with extreme care since errors may have severe consequences. Safety aspects are described in section 8.4 o If there is another bootmanager already installed and running : : : and you want to continue using it further there are some variations de- pending on the flexibility and power of it. A common case: You have a primary Linux partition on the second hard disk where you start your Linux from. Your bootmanager is able to start this partition via a boot sector. Then you can activate your Linux partition by installing LILO into its boot sector and tell your boot manager that it is active. * * 125 8. LILO - what it_does,_how_it_works______________________________________________________________________________________________* *_______________________________________@ ________________________________________________________________________________________________________* *________|| | Be careful while trying to make a logical Linux partition bootable by in- * * |||| || stalling LILO onto it: success is not guaranteed!, even if your other boot * * |||| | manager could launch logical partitions (mai 1997). * * |||| |_______________________________________________________________________________________________________* *________|||_ Well, you can try it; the best way is to use a tiny Linux installation for making sure. Probably you are lucky. Anyway, creating a primary and bootable Linux partition is recommended. 8.3.5 Configuring LILO Being a rather flexible boot manager LILO offers many ways of adapting a configuration to one's needs. The most important options and meanings are described below if you want to go into more detail consider having a look at [Almesberger, 1994 ]. Configuration of LILO is done in /etc/lilo.conf. If you are installing LILO for the first time we recommend to use YaST to configure LILO.. An adaptation of /etc/lilo.conf can be done later on. It is recommended to keep any existing old (working) configuration file and to keep it in a save place. Your settings only take place if you reinstall LILO after having changed /etc/lilo.conf (see section 8.4)! Construction of lilo.conf /etc/lilo.conf starts with a global section followed by one or more system sections for each operating system LILO should start. A new section is started by either of image= or other= . The order of entries in /etc/lilo.conf only matters in as much as the first one in the list is booted by default if no user interaction is taken _ this can be set to delay= and timeout= . file contents 8.3.1 (page page 127) shows a sample configuration on a ma- chine with DOS and Linux. There are two Linux kernels (an older and a newer one) on the momentary root partition (primary on second disk) as well as MS-DOS (or Windows 85) on /dev/hda1. Anything between a `#' and the end of line is regarded as a comment. Spaces and comments are ignored by LILO and can be used to improve readability. Now we go through the most important lines step by step, the other options will follow. o Global section (Parameter part) - boot= The device on whose first sector LILO should be installed. may be: a floppy disk drive (/dev/fd0), a partition (e. g. /dev/hdb3), or a whole disk (e. g. /dev/hda); the latter means installing LILO in MBR. Default: if this option is missing LILO is installed on the actual root partition. - prompt Forces sending the LILO prompt. Default is: no prompt! (compare to delay further down). This is recommended if LILO should manage 126 ____________________________________________________________________________________________________8.3.__LILO_custom_fit__________* *_______________ |||||||| | |||||||| | |||||||| | |||||||| | _______________________________________________________________________________________________________________________|||| |||||||| | |||||||| # LILO Configuration file | |||||||| # Start LILO global Section | |||||||| boot=/dev/hda # LILO Installation target | |||||||| | |||||||| backup=/boot/MBR.hda.970428 # Backup file for the old MBR | |||||||| # Apr 28 1997 | |||||||| | |||||||| #compact # faster, but won't work on all systems. | |||||||| linear | |||||||| | |||||||| message=/boot/greetings # LILO's Greeting | |||||||| prompt | |||||||| | |||||||| password = q99iwr4 # General LILO password | |||||||| timeout=100 # wait at prompt for 10 s before default | |||||||| | |||||||| # is booted | |||||||| vga = normal # normal text mode (80x25 characters) | |||||||| | |||||||| # End LILO global section | |||||||| | |||||||| # Linux bootable partition config begins | |||||||| | |||||||| image = /vmlinuz # Setting | |||||||| root = /dev/hdb3 # Root partition for kernel | |||||||| | |||||||| read-only | |||||||| label = Linux | |||||||| | |||||||| # Linux bootable partition config ends | |||||||| | |||||||| | |||||||| # Second Linux bootable partition config | |||||||| image = /vmlinuz.old | |||||||| | |||||||| root = /dev/hdb3 | |||||||| read-only | |||||||| label = Linux.old | |||||||| | |||||||| # 2nd Linux bootable partition config ends | |||||||| | |||||||| | |||||||| # DOS bootable partition config begins | |||||||| other = /dev/hda1 | |||||||| | |||||||| label = DOS | |||||||| loader = /boot/chain.b | |||||||| | |||||||| table = /dev/hda | |||||||| # DOS bootable partition config ends | |||||||| | ||||||||_______________________________________________________________________________________________________________ | File contents 8.3.1: Sample configuration in /etc/lilo.conf 127 8. LILO - what it_does,_how_it_works______________________________________________________________________________________________* *_______________________________________@ more than one system. Additional timeout should be set for guaran- teeing an automatic reboot if nothing is entered at the prompt. - timeout= Sets a time-out for the prompt option, thus enabling an automatic re- boot if no entry occurs in the given time. is the re- maining time in 0.1 s. Pressing Shift * starts the time-out over. De- fault: infinite, i. e.no automatic reboot! o Linux section - image= Here has to be the name of the kernel. With your new system this most probably is /vmlinuz or /zImage. - label= This name has to be unique in /etc/lilo.conf otherwise you can freely choose a name for the system (e. g. Linux ). Maximum length is 15 characters; you should consider using letters, numbers and under- score _no blanks_, special characters as German umlauts11 . Default is: the filename of the kernel image (e. g. vmlinuz). By entering this name you select which system to boot at boot time at the LILO prompt. It is recommended if there are many systems installed to keep track of them in a special message file ( message= ). - root= This is for giving the kernel the name of the root partition (e. g. /dev/ hda2) of your Linux system. This is recommended for security rea- sons! If this option is omitted the kernel takes its own root partition12 . o Other systems - other= other tells LILO to start partitions of other systems (e. g. /dev/ hda1). - loader= For loading a boot sector that belongs to another operating system, a certainLILO constructs an "Pseudo MBR" in its map file (at boot time LILO first starts this "Pseudo MBR" and this starts the foreign boot sector. This option sets the file where to get the code for the Pseudo MBR. Default: /boot/chain.b (generally this is ok)13 . But even this has to be set if the start partition is neither on the first floppy nor on the first hard disk. In our given example this is not necessary and only set for illustrational purposes. - table= has to set the source device for the partition table that should be written into Pseudo MBR (normally /dev/hda oder /dev/sda). 1_____________________________________________1 The specific rules on which characters to use can be found at [Almesberger, 1994 ], 3.2.1. 12 it can be seen with the command rdev 13 some mor loaders for special purposes are any_b.b, any_d.b and os2_d.b. 128 ____________________________________________________________________________________________________8.3.__LILO_custom_fit__* *_______________________ - label= Name (free choice) for the system. Recommended. The default _ the raw device name _ is not that explanatory. Optional options for configuration (selection) The last sector covered the neccesary entries in /etc/lilo.conf. Some additional and useful options should be discussed here. Those options that are marked as image options belong to the appropriate section of the operating system. The others are considered for the global section of /etc/lilo.conf. o backup= (optional) The file where LILO backups the boot sector. The default is /boot/ boot.xxxx (where xxxx is the internal device number of the installation partition)14 . We recommend not using a cryptic name, as in our example with device name and date. _______________________________________________________________________________________________________________________|| |||||||| If the backup file exists LILO does not create a new one! Please make sure ||| | |||||||| to use a name not in use. || | ||||||||_______________________________________________________________________________________________________________||_ o compact This option is recommended if you want to install LILO onto a disk. If enabled LILO tries to read more sectors at a time and thus might boot faster. This does not work on every machine. You should not set this as it is safer and after all, it is only one or two seconds. o linear This option causes all references to sectors to be written as logical instead of physical addresses. This option might come in handy if LILO does not recognize the geometry of the hard disk correctly. Still it does not make the 1024 cylinders border obsolete. This is scarceley ever needed. o message= (optional) Points to a textfile that should be shown on screen at system boot up. It should not contain more than 24 lines and can present an overview of the LILO boot selection. Recommended! _______________________________________________________________________________________________________________________ |||||||| If this option is set the message file belongs to the LILO boot machinery || | |||||||| || | |||||||| and after every change of this file LILO has to be reinstalled (section 8.4)! || | ||||||||_______________________________________________________________________________________________________________||_ o password= (optional) May be located either in a global or system specific section. Locks LILO services and booting with a password. If you take this seriously you should remove the password from lilo.conf after first use. As `root' you can set a new password for LILO any time you like (just reinstall it). 14 ____________________________________________________ To be found in kernel sources under /usr/src/linux/init/main.c, function parse_ root_dev() * * 129 8. LILO - what it_does,_how_it_works______________________________________________________________________________________________* *_______________________________________@ o read-only This option tells the kernel to initially mount the root partition read-only as commom to all Linux systems. If this is omitted the kernel uses its internal settings15 o delay= If the prompt is not explicitly set the user can order a prompt by typing ( Shift * , Strg , Alt ) . The delay= option sets to time to elapse before LILOboots the first system in its list. Default is: 0, i. e.no waiting. The delay option is obsolete if a prompt is ordered by rompt anyway. o vga= Selects VGA mode at startup. Valid s are normal (80x25), ext (80x50) or ask (well, ask). o append="" Image option for Linux kernel. Enables passing kernel parameters and passing hardware components at the LILO prompt. The kernel first gets the append line then the prompt; that's why prompt commands predomi- nate at collisions; For example: append="mcd=0x300,10" ! list of kernel parameters (Page chapter 7, page 101) 8.4 Installation and deinstallation of LILO During a new installation YaST leads you through all the steps interactively. Additional action in general is not necessary. Well, here we assume that LILO should be integrated with certain special options. ________________________________________________________________________________________________________* *________|| | The installation of a boot manager is rather tricky! Please ensure in ad- * * |||| | * * ||| | vance that you are 100% able to boot your Linux or other systems respec- * * ||| | tively. You must have fdisk installed on this disk otherwise you might find * * ||| | * * ||| | yourself in the bad situation of not being able to access your hard disk at * * ||| | all! * * ||| |_______________________________________________________________________________________________________* *________|||_ 8.4.1 Installation after changing the configuration If any of the LILO components have changed or you have modified your configuration in /etc/lilo.conf you will have to reinstall LILO. This is easily done by launching the "Map Installer". root@earth: # /sbin/lilo What happens now is that LILO writes a backup of the target-boot-sector, writes its "first step" into it and creates a new map file (comp. section 8.3.3). LILO now announces each installed system _ in our case (see screen out- put 8.4.1): After correct installation the machine can be rebooted. root@earth: # shutdown -r now 1_____________________________________________5 This can be seen using the command rdev-R . Installation and freshly com- piled kernels have read-only set by default (check!); therefore you normally do not need this option. 130 ________________________________________________________________8.4.__Installation_and_deinstallation_of_LILO_______________* *______________________ _______________________________________________________________________________________________________________________|| |||||||| Added Linux* | ||||||||| Added Linux.old || |||||||| Added DOS | |||||||| | ||||||||_______________________________________________________________________________________________________________ | Screen output 8.4.1: Output after launching LILO First the BIOS performs its system test and directly afterwards you will see LILO and its command prompt where you can enter parameters and select a bootimage (the recently installed configurations). 7 ! shows you a list of all installed systems. 8.4.2 Installation recompiling a kernel If you want to include a freshly created kernel into your boot concept there is a comfortable alternative apart from the LILO installation: Organization of commands needed for configuring and compiling a kernel is written to /usr/src/linux/Makefile. There is a target named zlilo in Makefile which automatically copies /vmlinuz to /vmlinuz.old and the new kernel to /vmlinuz and reinstalls LILO. This can be done by the command: root@earth:/usr/src/linux # make zlilo This, of course, only is useful if you have adjusted /etc/lilo.conf in ad- vance. The new as well as the old kernel should now be listed. file con- tents 8.3.1 shows an example. At the LILO prompt you can launch either of the two kernels. That improves safety because you can boot even if one of your kernels fails. ! creating a new kernel (Page chapter 13, page 181) 8.4.3 Removing LILO _______________________________________________________________________________________________________________________|| |||||||| Removing a boot manager is rather tricky! Please ensure in advance that ||| | |||||||| || | |||||||| you are 100% able to boot your Linux or other systems respectively. You || | |||||||| must have fdisk installed on this disk otherwise you might find yourself in || | |||||||| || | |||||||| the bad situation of not being able to access your hard disk at all! || | ||||||||_______________________________________________________________________________________________________________||_ Perhaps one day it will be necessary for you to deinstall LILO:-( . This is achieved by writing back the target boot sector where LILO has been in- stalled. This is no problem under Linux if there is a valid backup (comp. section 8.3.5, Option backup ). _______________________________________________________________________________________________________________________|| |||||||| A boot sector backup is turned invalid if the partition has got a new filesys- ||| | |||||||| || | |||||||| tem (For DOS users: being formatted). The partition table of an MBR || | |||||||| backup becomes invalid if the hard disk in question has been repartitioned || | |||||||| || | |||||||| in the meantime. Such "backups" are time-bombs; the best is to delete || | |||||||| them as soon as possible! Unpacking old and invalid backups into system || | |||||||| || | |||||||| sectors is a rather straight way to data loss! || | ||||||||_______________________________________________________________________________________________________________||_ * * 131 8. LILO - what it_does,_how_it_works______________________________________________________________________________________________* *_______________________________________@ It is very simple to get back a DOS, Windows 95 or OS/2 MBR. Just enter the MS-DOS command (available since 5.0) C:\> FDISK /MBR or on OS/2 C:\> FDISK /NEWMBR These commands only write the first 446 bytes (the boot code) into MBR and leave partitions untouched16 . Otherwise make a backup of the LILO sector in question _ the safer the better. Now you should check (at least twice :-) ) whether your old backup file is the correct one and if it is exactly 512 bytes in size! At last you play it back but do not confuse if= and of=! o If LILO resides in partition yyyy (e. g. hda1, hda2, : : :): root@earth: # dd if=/dev/yyyy of=New-File bs=512 count=1 root@earth: # dd if=Backup-Datei of=/dev/yyyy o If LILO resides in MBR of zzz (e. g. hda, sda): root@earth: # dd if=/dev/zzz of=New-File bs=512 count=1 root@earth: # dd if=Backup-Datei of=/dev/zzz bs=446 count=1 The last command is cautious and does not write partition table. By the way, note how easy and fast a boot sector backup is done! It is recommended to do this frequently! 8.5 Creating a Linux-bootdisk A Linux boot-disk consists (a little simplified) of one or more Linux kernels, possibly managed by LILO. It serves for starting up your system even if booting directly from hard disk is not possible (possible reasons: overwritten MBR, misconfigured bootmanager, errors while installing and many more...). Such a boot-disk only loads the kernel; everything else (system programs, init start scripts) has to be provided by the system and has to work. The connection is established via the kernel. The kernel gets the root device of your machine. Do not confuse this with S.u.S.E.-boot-disk for installation and emergencies (which you can create according to disks anytime). The latter load a fully in- dependent Linux system supplying the ability of accessing the existing Linux partitions from "the outside" (e. g. mounting). This is necessary if you en- counter errors on your hard disk (damaged filesystems, forgotten `root' password etc.) and resembles booting DOS or OS/2 via boot-disk. 8.5.1 Boot-disk without LILO If you are in the lucky situation that your kernel does not need any hardware parameters the easiest way of creating a boot-disk is to just write the actual kernel onto a raw disk and to adjust the root device (if not done before). 1_____________________________________________6 provided that the MBR (section 8.1.1) has valid code! If not it is considered invalid and the partition table is moved to "null" : : : 132 __________________________________________________________________________________________8.6.__Sample_configurations_______* *______________________ root@earth: # dd if=Ihr_Kernel of=/dev/fd0 bs=18k root@earth: # rdev /dev/fd0 Your_Root_Partition root@earth: # rdev -R /dev/fd0 1 The last command makes sure that the kernel initially mounts root read-only (as it is expected; the startup scripts rely on it). 8.5.2 Boot-disk with LILO A much more comfortable boot-disk with greeting, prompt, kernel parameters and other LILO goodies is created by transferring a complete start machinery onto the disk (comp. section 8.3.3). For this the disk needs a filesystem; Minix suits this best. If you only want to install a kernel and you do not want the greeting YaST can perform this task for you (in `Administrating the system'). _______________________________________________________________________________________________________________________|| |||||||| Your kernel has to follow the old Linux tradition and should be labeled ||| | |||||||| vmlinuz otherwise YaST will not find it! || | ||||||||_______________________________________________________________________________________________________________||_ Otherwise you proceed as follows: o Assign a minix filesystem on a new and empty disk _force an error check for safety reasons_; after this is done mount the disk to, e. g. , /mnt: root@earth: # /sbin/mkfs.minix -c /dev/fd0 1440 root@earth: # /bin/mount /dev/fd0 /mnt o now copy your kernel files and the LILO file /boot/boot.b to /mnt (i. e.onto the disk). o optional: create a message file /mnt/message o Create an own lilo.conf in /mnt. Of course you have to adapt this to your needs (give the correct name of the kernel etc.): o install LILO with this lilo.conf: root@earth: # /sbin/lilo -C /mnt/lilo.conf o Unmount floppy; you are done! root@earth: # /bin/umount /mnt o Do not forget to check your boot-disk at the next system start as to whether it works or not :-) 8.6 Sample configurations If your Linux is all alone on your machine there is no immediate action to be taken since everything needed has already be done by YaST. 8.6.1 DOS/Windows 95 and Linux Required: There is at least a primary partition for each DOS/Windows 95 and Linux which is below the 1024 cylinders limit (section 8.3.4). For this case we already know a configuration (file contents 8.3.1) _ only the settings for root= , image= and other= have to be adapted. LILO is installed * * 133 8. LILO - what it_does,_how_it_works______________________________________________________________________________________________* *_______________________________________@ | * * | ________________________________________________________________________________________________________* *________|||| | # LILO Configuration file bootdisk * * | || # Start LILO global Section * * || | boot=/dev/fd0 # Installation: Floppy * * | | * * | | install=/mnt/boot.b # Of course LILO and * * | | map=/mnt/map # map file onto floppy! * * | | * * | | message=/mnt/message # optional * * | | prompt * * | | * * | | timeout=100 # Wait at prompt: 10 s * * | | vga = normal # * * | | # End LILO global section * * | | * * | | # * * | | # Linux bootable partition config begins * * | | * * | | image = /mnt/First_Kernel # default * * | | root = /dev/Your_Root_Device # Here your root partition! * * | | * * | | label = linux * * | | # Linux bootable partition config ends * * | | * * | | # * * | | # System sections for more kernel here: * * | | * * | |_______________________________________________________________________________________________________* *________ | File contents 8.5.1: lilo.conf for a bootdisk in MBR. Of course you should omit the Linux.old if you do not have an old kernel installed. Save your /etc/lilo.conf and a Linux boot-disk. Especially Windows 95 feels inclined17 to eliminate "foreign" MBRs. If you are able to boot Linux using your boot-disk this problem is solved rather quickly. root@earth: # /sbin/lilo 8.6.2 Windows NT and Linux on one hard disk 1. For booting the NT bootmanager is used. This can start images or the boot sector itself. Execution of the following steps prepares everything for a friendly coexistence of Linux and Windows NT: o Installation of NT o Partition the NT disks (using FAT so that Linux can write on it). o Install Linux "as usual" (in our example the root partition is on /dev/ sda3). Mount DOS data disk (e. g. on /dos). o Install LILO but install it in Linux' root partition (/dev/sda3), not in MBR (/dev/sda)! You still have the possibility of configuring a selection of Linux kernel for LILO. Example-lilo.conf: see file contents 8.6.1. o Copy the LILO boot sector where NT can find it; e. g. dd if=/dev/sda3 bs=512 count=1 of=/dosen/bootsek.lin o Boot NT. Copy from data disk to main directory of NT's system drive (if it is not already there). 1_____________________________________________7 A complete list is not known : : : 134 __________________________________________________________________________________________8.6.__Sample_configurations______* *_______________________ |||||||| | _______________________________________________________________________________________________________________________|| |||||||| | |||||||| # LILO Configuration file | |||||||| | |||||||| # Start LILO global Section | |||||||| boot=/dev/sda3 # Target of installation | |||||||| | |||||||| backup=/boot/boot.sda3.970428 # Backup previous boot sector; | |||||||| # 28. Apr 1997 | |||||||| prompt | |||||||| | |||||||| timeout=100 # Wait at prompt: 10 s | |||||||| vga = normal # force sane state | |||||||| | |||||||| # End LILO global section | |||||||| # Linux bootable partition config begins | |||||||| | |||||||| image = /vmlinuz # default image to boot | |||||||| root = /dev/sda? # Here the root partition! | |||||||| | |||||||| label = Linux | |||||||| # Linux bootable partition config ends | |||||||| | ||||||||_______________________________________________________________________________________________________________ | File contents 8.6.1: lilo.conf for making a Linux root partition startable o In boot.ini (setting attributes) supplement at the end: c:\bootsek.lin="Linux" o After next booting (if everything went right) there should be an entry in NT's bootmanager! 2. Other possibility: install LILO in MBR and claim that it's DOS to Win- dows NT (as in our previous example). _______________________________________________________________________________________________________________________ |||||||| || | |||||||| NT 3.5* does not recognize Linux' partition types 82 and 83! Please make || | |||||||| sure that no NT programm tries to "repair" your partition table. Data loss || | |||||||| || | |||||||| would be the result! Always have valid backups of the LILOboot sector at || | |||||||| hand. || | ||||||||_______________________________________________________________________________________________________________||_ 8.6.3 OS/2 and Linux 1. Possibility: The OS/2 bootmanager is used for booting. It can launch unlimited primary and logical partitions provided they are below the 1024 cylinders limit. The user is responsible for the partitions This boot- manager is configured by OS/2 fdisk . Preparation on Linux side: is just making a partition startable (usually this is the root partition) with LILO. You can use the same lilo.conf as in the Windows NT example, but ther is one thing that you have to consider in advance : : : Preparation on OS/2 side: OS/2 not only uses the conventional and un- derstandable entries for existing partitions (in MBRs on hard disks) but uses "waste" space in this sectors for additional infos18 . If these are in- consistent, OS/2's fdisk considers this partitions to be faulty and refuses bootmanager services. The fdisk s of other systems of course do not know how to handle these additional partitions : : :Conflicts are rather natural. 18 ____________________________________________________ A new Support DataBase article is about to be written; keyword "OS/2" * * 135 8. LILO - what it_does,_how_it_works______________________________________________________________________________________________* *_______________________________________@ Therefore: Before installing Linux load OS/2 (the installation system is sufficient) and create the Linux partitions with OS/2's fdisk , at least the logical drives. This initially creates additional OS/2 partitions which can be in the way. Solution: Immediately after creating the partitions load the Linux installation system (or the rescue disk from S.u.S.E. LinuxCD) and change partition types to 83 (Linux native). Now these partitions will be ignored by OS/2. 2. Possibility: Install LILO as a main bootmanager on a primary partition on the first disk19 . This special case is considered in our next example where DOS is involved additionally. 8.6.4 DOS, OS/2 and Linux 1. If you have used the OS/2 bootmanager for DOS and OS/2 and want to continue using it you simply add Linux to its start menu as described in the previous example. 2. Possibility: LILO is installed as the main bootmanager on a primary par- tition of the first hard disk. The following, purposely complicated example for lilo.conf (file con- tents 8.6.2, page page 137) assumes that DOS' and Linux' startpartition are primary and on the first hard disk whereas OS/2 (logical drive) resides on the second hard disk _ all of them below 1024 cylinders. OS/2 is on the second disk. That's why a special loader (/boot/os2_b.b instead of/boot/chain.b) is used. The MBR code might be either from DOS or OS/2 (of no consequence). The LILO startpartition (/dev/sda4) has to be marked active with any fdisk . 8.7 LILO problems 8.7.1 Some guidelines Some simple guidelines at the beginning to avoid most LILO problems in advance (this is from the LILO documentation [Almesberger, 1994 ]): o Do not panic! If anything does not work try to find the error and/or the cause first; check the diagnosis and then commence with fixing the problem. o Always have an up to date and tested boot-disk at hand. o S.u.S.E. Linux 5.0 contains a full Linux system on its boot-disk and in- stallation CD (For rescue system see section 11.4) for enabling you to reach all your Linux partitions. Included are tools for repairing (almost) any problems that can occur. o Read this documentation. Especially if the system does not do what you want it to do. o Check /etc/lilo.conf before installing the map installer (/sbin/ lilo). o Be attentive if using big, or more than one disks. Be aware of the 1024- cylinders limit! 1_____________________________________________9 Not that good an idea to install in MBR as repartitioning with another fdisk could erase the MBR and thus remove LILO. 136 ____________________________________________________________________________________________________8.7.__LILO_problems_____* *______________________|||| |||||||| | _______________________________________________________________________________________________________________________|| |||||||| | |||||||| # LILO Configuration file | |||||||| | |||||||| # Start LILO global Section | |||||||| boot = /dev/sda4 # LILO in Linux root partition | |||||||| | |||||||| backup = /boot/boot.sda4.970428 | |||||||| message = /boot/message # Greeting message | |||||||| | |||||||| prompt | |||||||| delay = 100 | |||||||| vga = normal | |||||||| | |||||||| # | |||||||| # Linux bootable partition config begins | |||||||| | |||||||| image = /vmlinuz | |||||||| label = linux | |||||||| | |||||||| root = /dev/sda4 | |||||||| # Linux bootable partition config ends | |||||||| | |||||||| # | |||||||| # OS/2 bootable partition config begins | |||||||| | |||||||| other = /dev/sdb5 | |||||||| table = /dev/sdb | |||||||| label = os2 | |||||||| | |||||||| loader = /boot/os2_b.b | |||||||| # OS/2 bootable partition config ends | |||||||| | |||||||| # | |||||||| # DOS bootable partition config begins | |||||||| | |||||||| other = /dev/sda1 | |||||||| table = /dev/sda | |||||||| | |||||||| label = dos | |||||||| # DOS bootable partition config ends | |||||||| | ||||||||_______________________________________________________________________________________________________________ | File contents 8.6.2: LILOwith DOS, OS/2 and Linux on two hard disks 8.7.2 Diagnosis of errors: LILO start-messages This mainly is section 5.2.1 from [Almesberger, 1994 ]. If LILO is launched it displays the word `LILO'. Every letter signifies a certain state. If LILO cannot be launched the output letters give a hint on where the error occured. none No part of LILO has been loaded. Either LILO is not installed at all or the partition with LILO's boot sector has not been started. `L' error : : : The "first step" has been loaded and started but the second step (/boot/boot.b) could not be loaded. This normally points to a physical error of the boot device or a faulty disk-geometry. `LI' Second step has been invoked but could not be started. This can be due to a faulty disk-geometry or to moving /boot/boot.b without rein- stalling LILO. `LIL' Second step of LILO has been started but could not allocate needed data (pointer etc.) from its map file. This typically is due to a physical error of the boot device or a faulty disk-geometry. * * 137 8. LILO - what it_does,_how_it_works______________________________________________________________________________________________* *_______________________________________@ `LIL?' The second step has been loaded at a wrong memory address. This can be due to a subtile error in disk-geometry or to moving /boot/boot. b without reinstalling LILO. `LIL-' Data in the map file are invalid. This can be due to a faulty disk- geometry or to moving /boot/boot.b without reinstalling LILO. `LILO' Every part of LILO has been loaded successfully. The most common causes for geometry errors are not physical defects or invalid partition tables but errors in LILO installation. o Disregard of the 1024 cylinders limit (see next section); o Unsuccessfull attempt at starting LILO from a logical partition. 8.7.3 The 1024-cylinder limit As emphasized a couple of times before (section 8.3.4) the whole LILO ma- chinery (i. e.any data that is needed for starting) has to be accessible by BIOS calls. The sections that can be used (we call theses allowed sections) have already been highlighted. Which possibilities (besides the restrictions) are still left? Quite a couple! Keep in mind that only the start-up machinery is concerned. There is now law that says it has to be on the Linux root-partition; it even is possible (but quite dangerous too) to put the start-machinery onto partitions of other operating systems as long as Linux has read- and write access. ________________________________________________________________________________________________________* *________ | Never install the LILO boot sector on an unknown partition because you * * ||| | * * ||| | will severly damage their filesystem! * * ||| |_______________________________________________________________________________________________________* *________|||_ o The best way, of course, is to install a primary partition (within the al- lowed section) and to install all LILO files (including the LILO boot sec- tor) onto this partition. This mostly will be the Linux root partition. You can also add it to /boot with YaST. The only condition is, that there has to be enough space for - boot.b, map, message, - the Linux kernes LILO should boot. Some mega-bytes will be sufficient. It does not matter where you put the rest of your partitions. There are no more restrictions; as soon as the kernel runs you have unrestricted access to all installed disks. But what to do if there is no space for such a partition? If you neither want to repartition your disk, nor to upgrade to SCSI, nor want to purchase a new BIOS version, there still are two (makeshift) possibilities: o Use a boot-disk instead of LILO on the hard disk, or if you are addition- ally running MS-DOS, you can use loadlin as well. o Install LILO start-machinery onto a none Linux partition which is in the allowed section and where Linux has write access (e. g. a FAT/VFAT drive). Of course we cannot put the LILO boot sector there! So there are only two places where to put it. Either the start of an extended partition 138 ____________________________________________________________________________________________________8.7.__LILO_problems___________* *________________ on the first disk _ as long as it is beneath the 1024 cylinder limit _ or the MBR. Consider that the partition in question is mounted to mnt. LILO should be installed in MBR (/dev/hda) and should additionally boot DOS from /dev/hda1. You should proceed as follows: - Create a new directory (e. g. /mnt/LINUX) and copy the LILO files mentioned above thereto: boot.b, map, message, as well as the Chain-Loader of other operating systems (normally chain.b) and of course the Linux kernel LILO should boot. -|||Create| a /mnt/LINUX/lilo.conf where all paths point to /mnt/ | |||LINUX.|| || _______________________________________________________________________________________________________________________|| |||||||| | |||||||| # LILO Configuration file | |||||||| | |||||||| # Start LILO global Section | |||||||| boot=/dev/hda # Installation target | |||||||| | |||||||| backup=/mnt/LINUX/hda.xxxx # backup of old MBR | |||||||| install=/mnt/LINUX/boot.b # Of course LILO and | |||||||| | |||||||| map=/mnt/LINUX/map # map file are in /mnt/LINUX! | |||||||| message=/mnt/LINUX/message # optional | |||||||| prompt | |||||||| | |||||||| timeout=100 # Wait at prompt: 10 s | |||||||| vga = normal # | |||||||| | |||||||| # End LILO global section | |||||||| # | |||||||| | |||||||| # Linux bootable partition config begins | |||||||| image = /mnt/LINUX/Erster_Kernel # default | |||||||| | |||||||| root = /dev/Ihr_Root_Device # Root partition! | |||||||| label = linux | |||||||| | |||||||| # Linux bootable partition config ends | |||||||| # | |||||||| # System section for other kernels: | |||||||| | |||||||| # | |||||||| # Ende Linux | |||||||| | |||||||| # DOS bootable partition config begins | |||||||| other = /dev/hda1 # MSDOS system drive | |||||||| | |||||||| label = dos | |||||||| loader = /mnt/LINUX/chain.b | |||||||| | |||||||| table = /dev/hda | |||||||| # DOS bootable partition config ends | |||||||| | ||||||||_______________________________________________________________________________________________________________ | File contents 8.7.1: lilo.conf for other partitions - Install LILO with this lilo.conf: root@earth: # /sbin/lilo -C /mnt/LINUX/lilo.conf After that LILO should run. Boot MS-DOS and protect the LILO files as well as possible against write access (Just to remind you: any write access disables LILO!). Anyway you should assign to all files under X:\LINUX (wher the `X' is the DOS drive mounted to /mnt) the DOS attributes system and hide. 139 8. LILO - what it_does,_how_it_works______________________________________________________________________________________________* *_______________________________________@ As conclusion we want to point out to two HOWTOs in /usr/doc/howto/ mini/, these are LILO.gz and Large-Disk.gz . 8.7.4 LILO and DOS-diskmanager (OnTrack-Manager, EZ-Drive) There are particular details on this topic in the above mentioned mini-HOWTO /usr/doc/howto/mini/Large-Disk.gz . General issues on the DOS diskmanagers A DOS diskmanager is most often installed in the MBR of IDE drives. Since the BIOS and DOS can only handle 1024 cylinders the manager replaces the BIOS functions for accessing the disk with its own. It recalculates the existing geometry with a recalculation routine to a pseudo one below 1024 cylinders. The problem is that this pseudo geometry is entered into the partition table. (This is no problem to SCSI or EIDE disks (connected to an EIDE controller) since this recalculation (if needed) is done by the hardware. Linux can handle this without any problems.) Well, Linux is not compatible to diskmanagers since it does not access the disk with BIOS calls! There will be severe problems as Linux tries to read these "psudo" geometries and takes them seriously! From kernel 1.3.29 onwards Linux can detect these OnTrack managers and EZ drives and support them. But this is only a makeshift! The best solution however, is to disable the manager and let DOS only see the first 504 MB. Then Linux can be installed (except LILO) on the back part of the disk. Furthermore you can think about installing an EIDE controller since then the manager is no longer needed. OnTrack-Manager: Deinstallation The following hint has kindly been given to us by Matthias Wimpf. This description relies on version 7.04 of the OnTrack-manager. For other products this procedure can differ. 1. IMPORTANT! Do make a backup of the disk! All data will get lost while removing the diskmanager and have to be unpacked later on. For those of you who additionally install an EIDE controller this might not be relevant (see 3. and conclusion). Anyway, a backup never hurts. 2. Create a boot-disk using the OnTrack-Manager (dm.exe). 3. Start the OnTrack-DiskManager from floppy drive A: C:\> dm /m /y- The parameter `/m' causes the manager to start in manual mode. Now there are some more functions available. The parameter `/y-' causes not to read the Dynamic Drive Overlay. The `-' is important since otherwise the OnTrack-Manager will start anyway. o Choose Maintenance Menu 140 ____________________________________________________________________________________________________8.7.__LILO_problems_____* *______________________ o Choose UnInstall Disk Manager o Now a message appears telling us that all data will be lost (that's what we have the backup for). If an EIDE controller has been installed this message might not appear, so try (having a backup makes you feel safe!). o Go back to the main menu; the manager has been removed 4. Entering the disk parameters into CMOS. If you do not have an EIDE controller or the BIOS only supports up to 1024 cylinders you can only use 504 MB for DOS however big your disk might be (this number varies depending on the calculation; the limit is 1024 cylinders). Now there is a way of letting DOS use the maximum of 504 MB and to assign the rest of the disk to Linux. Linux does not use BIOS functions for recognizing the disk's geometry and therefore knows the exact size. Configure the DOS partition: Generally one wants to use the maximum amount available for DOS (504 MB). So you enter the maximum amount of cylinders (1024) and all other parameters into the CMOS. If you do not know the other parameters (there should be a label on the hard disk) you can use the Ontrack-Manager to find out. Just enter 305 cylinders, 4 heads and 17 sectors/track. Then start the DiskManager as described above and choose the item CMOS setup. Then you should select your disk and Automatic CMOS Selection. If the values in CMOS are not optimal you will get a message with the correct values. These values have to be added to CMOS by hand. Now you can format and partition the disk with the DiskManager. This can be done by DOS' fdisk as well but the DiskManager provides higher flexibility and is faster. Anyway, do not forget to pass the `/y-' as the OnTrack-Manager will be reinstalled! 5. Install Linux. Now Linux can be installed as usual with setup and YaST. Linux' fdisk automatically recognizes the geometry and you can assign partitions as usual. Conclusion All of the steps mentioned above have been performed using a Maxtor 7850AV (850 MB) and everything ran without problems. The most important thing is not to forget the backup and a boot-disk with the OnTrack-Manager. Still, time needed for a backup and a (DOS) restriction to 504 MB is immense. On the other hand you get rid of the OnTrack-Manager, which accelerates booting and disk accesses enormously. Generally buying an EIDE controller or BIOS is the better solution. Perhaps not even a backup of the disk is needed, as removal of the DiskManager might be possible without loss of data. Anyway I have not tried this yet. 8.7.5 Special boot-problems with kernel from 2.0 onwards Problems Booting with LILO (from either floppy or hard disk) problems might occur after having loaded quite an amount of the kernel (e. g. S.u.S.E. installation kernel).:-( * * 141 8. LILO - what it_does,_how_it_works______________________________________________________________________________________________* *_______________________________________@ You can select a kernel at the LILO prompt and this kernel is loaded (some dots are output onto screen) but starting the kernel fails. Before reaching "uncompressing Linux" the system crashes with different behaviours. Possible error messages: o System reboots o Sytem just hangs o "crc-error" o "no free space" o "Error 0x00" o "Error 0x01" o "incomplete literal tree" Thereafter access on the floppy is attempted but the system hangs. Cause The cause lies in an combination of big kernels, LILO and faulty hardware. Roughly this concerns 1% of all machines. We assume that this is due to a faulty BIOS that has problems with the fast memory access. This problem does not occur, if: o the machine is booted via loadlin o the kernel has been (e. g. ) copied to a floppy dd if=/vmlinuz of=/dev/fd0 and booted from it o a smaller kernel that has been created with make zImage (e. g. an older 1.2.13 kernel) and is booted via LILO. The following BIOS settings did not cause any problems either: o Disable Internal Cache o DRAM Precharge Wait State value: 1 and o DRAM Wait Burst Timing value: 0x3333 Solution First of all a you should be able to install a system. So, if you can neither boot via loadlin nor via setup you should use an old 1.2.13 boot-disk for installation. If you do not have such a disk at hand, you should change the BIOS settings accordingly. After a successfull installation there is the question of how you want to boot in future. First you should use the same media as during installation. loadlin from DOS should not cause any problems. With a boot-disk you should enter the following parameters: load__ramdisk=0 root=/dev/??? where ??? is your root partition (e. g. /dev/hda1). Thereafter you should build your own kernel since this can be booted by LILO. 142 Chapter 9 Printers While installing Linux with YaST you probably have already done some printer configuration. This chapter now covers the essentials that stand be- hind the processes. It does not show each aspect but enough to understand the principle that acts behind the printer system. 9.1 Overview: Interfaces, Queues (Spooling) 9.1.1 The parallel ports Connecting a printer to a Linux system is normally done via a parallel port. These ports are accessible via device files (as usual under Unix). Here are their properties: File ___________|___major________minor__|_____DOS_name__|__________IRQ_______Port______________________ /dev/lp1 | 6 1 | LPT1 | (7) 0x378 /dev/lp2 | 6 2 | LPT2 | (5) 0x278 /dev/lp0 | 6 0 | LPT3 | (5)- 0x3bc As a standard, those parallel ports are set to polling mode (the CPU can reg- ulary has to ask them, if there is a data transfer in progress). The systemad- ministrator `root' can Of course you can only use those devices that really are connected to a physically existing device. At system start up the driver lp.c announces which ports have been found. You can have a look at these messages in /var/log/boot.msg or by entering dmesg . _______________________________________________________________________________________________________________________|| |||||||| You can not use a kernel with PLIP (IP over parallel line) and the /dev/ ||| | |||||||| lp? devices simultaneously. The same applies to other devices connected || | |||||||| || | |||||||| to the parallel port such as CD-ROMS or ZIP drives. || | ||||||||_______________________________________________________________________________________________________________||_ The PLIP and the lp driver would "fight" for the port (a fix for this problem is currently under construction). These ports use polling mode by default (CPU has to interrogate regularly if there is a data transfer). The system administrator (`root') can switch this to interrupt mode with tunelp and perform additional tasks for increasing performance. Usually IRQ 7 is used for lp1 and IRQ 5 for lp2. For testing the interface you can send the file directly to the printer using: root@earth: # cat text-file >/dev/lp1 * * 143 9. Printers _________________________________________________________________________________________________________________* *_______________________________________@ (only being `root'). For every day's work this is more than useless since there can be many print jobs at once and the printer would collapse. Therefore print jobs are performed via a queue. 9.1.2 Spooling mode, Printer queues Printers are accessed via a spooling mechanism as usual under multitasking systems as Linux; i. e. printer jobs are temporarily saved in a file and are processed one after another by a job control program called a daemon. This ensures that many users can send printer jobs simultaneously to the printer without risking a conflict. After having sent a printer job the user can com- mence as usual with his work. A printer queue consists of: o an entry in /etc/printcap; here it is defined; o a directory normally under /var/spool; here there is a data file and a control file for each printer job. Each queue is processed by exactly one printer. There can be many printing queues for one printer. With S.u.S.E. Linux there comes the BSD-spooling-system from University of California at Berkeley which is included in package lprold. Traditionally it is designed for use over a network and assumes that TCP/IP has been configured and is up and running. Whether queues are set local or remote makes little difference to the configuration. Processing a print job takes place in three steps: o The data file is copied to the queues directory and is completed together with a freshly created control file. o This data file now has to pass a filter program which converts it to the printer specific format (e. g. PostScript). This conversion can take up sev- eral individual steps. o The printer-specific file now is queued and printed in order. 9.2 Printer queues: running and Configuring Let's have a look at the stations a print job has to pass before being printed. And let's also have a look at the programs that take care of the job. Many of the details we have to skip can be found in the individual manpages of lpr(1) , lpd(8) and printcap(5) . 9.2.1 lpr: Please line up! lpr is your "interface" to the printer queues on your machine. Generally a job is started using: newbie@earth: > lpr [-P queue] text-file If you omit the option -P ("Printer", for the queues name) the default is taken from $PRINTER. If this is not set either the traditional name lp is used. This also applies to lpq , lprm and lpc (see below). 144 _____________________________________________________________9.2.__Printer_queues:_running_and_Configuring__________________* *______________________ lpr checks whether this queue exists by reading /etc/printcap. If so, it creates a control file (cf file) and a copy of the data file in the spool directory. If it is a huge job it can be practical if you just link the file instead of copying it. This can be done by lpr -Option -s . Of course you should not write the file until the job has been processed. lpr allows to set a certain filter (printer driver) by hand (this is hardly ever needed). More can be found in lpr and in section 9.3.1 9.2.2 Additional control tools for the user: Overview Any of these commands (same as with lpr ) accept the -P option for passing the queue. All details that have to be skipped in this chapter are explained in the corresponding man pages. o lpq shows your own jobs in the queue: For example: newbie@earth: > lpq -P queue _______________________________________________________________________________________________________________________|| |||||||| | |||||||| queue is ready and printing | |||||||| Rank Owner Job Files Total Size | |||||||| | |||||||| active newbie 676 Hello.txt 259420 bytes | |||||||| | |||||||| 1st newbie 677 letter.dvi 11578 bytes | |||||||| 2nd newbie 683 picture.gif 37464 bytes | |||||||| | ||||||||_______________________________________________________________________________________________________________ | o lprm removes own jobs from a queue. For example: newbie@earth: > lprm -P queue 676 _______________________________________________________________________________________________________________________|| |||||||| | |||||||| dfA676Aa05005 dequeued | |||||||| cfA676Aa05005 dequeued | |||||||| | ||||||||_______________________________________________________________________________________________________________ | If no job is given, the current active job is removed if it is your own job. o lpc control commands for queues. 9.2.3 lpd: the print manager in background lpd is launched by /sbin/init.d/lpd and runs in the background if the corresponding option START__LPD=yes has been set in /etc/rc.config. lpd checks /etc/printcap to see which printer queues are defined. Its job is to organize the execution of spooled jobs: o it manages local printer queues: it sends every job's data file through the appropriate filter_which is assigned by the entry in /etc/printcap and by an explicit setting in the control file_and to the printer's interface afterwards; o it takes care of the order of the jobs; o it checks the queues and printers state (status) in the local spool direc- tories; * * 145 9. Printers _________________________________________________________________________________________________________________* *_______________________________________@ o it sends jobs to an lpd on a remote machine; o it accepts requests from remote machines for local queues or refuses them if authorization fails. Authorization: Only requests from hosts who are listed in /etc/hosts.lpd are executed (it is also possible to enter the name in /etc/hosts.equiv but this has wide ranged consequences as far as access and security are concerned and therefore should be avoided). Additional definitions of the queue can restrict access to certain users or users with an account on the local machine. 9.2.4 Filter: the workaholics The job of the filter programs is to transfer the jobs into a printer specific format. They have to do almost everything. They resemble more or less the printer drivers on other systems such as Windows or OS/2. But unlike "drivers" they are also used to count jobs (amount, used up resources : : : ). See section 9.3. 9.2.5 /etc/printcap: Configuration of the queue In /etc/printcap each queue is defined by an entry of its own (line feed ends the entry). The entry starts with one or more names for the queue (separator | ) followed by a list of specifications1 as shorthand= (separator : ). "Empty" lines or those starting with an `#' are regarded as comments. There are a couple of example entries (commented) in /etc/printcap. A simple entry might look like this: ________________________________________________________________________________________________________* *________ | * * | | ascii|deskjet:lp=/dev/lp1:sd=/var/spool/ascii:sh:mx#10240 * * | | * * | | * * | |_______________________________________________________________________________________________________* *________ | File contents 9.2.1: /etc/printcap : simple local queue This queue can be accessed either by ascii or by deskjet. Its spool direc- tory is /var/spool/ascii, its printer /dev/lp1. It does not show a header page at the beginning (sh = suppress header) and accepts jobs up to a size of 10240 bytes. Here now is an example for a queue on a remote machine: ________________________________________________________________________________________________________* *________ | * * | | lp1|HP-4P:\ * * | | * * | | :rm=Galois.suse.de:rp=HP:\ * * | | :sd=/var/spool/lpd/lp1:\ * * | | * * | | :mx#0:sh * * | | * * | |_______________________________________________________________________________________________________* *________ | File contents 9.2.2: /etc/printcap : simple remote queue (The backslashes `\' hide the linefeed _ as with shell commands _ this is now regarded as only one line.) _____________________________________________1 all shorthands, their meaning and default settings are described in printcaps man page. 146 ____________________________________________________________________________________9.3.__Printer_filter._The_apsfilter_____* *______________________ The name of the queue is lp1 and HP-4P. Instead of the device (lp= ) it is referred to the remote host Galois.suse.de and its queue HP. Setting a filter is not necessary since the queue on Galois.suse.de is re- sponsible for the job; so only the spool directory /var/spool/lpd/lp1 and the size (mx#0 means: none) is defined. If you install apsfilter with YaST it enters three new queues in /etc/printcap. Further information under section 9.3.2. 9.3 Printer filter. The apsfilter 9.3.1 What is a filter and how does it work? As mentioned above the job of a filter is to convert the data file of a job into a suitable format. Only the papersize, login name, host name (local), name of the file for the count is passed to lpd . Being a Unix filter it gets the data file via standardin and sends the printable output through standardout. Obviously the filter has to know the format of the file (ASCII, DVI, PostScript : : :) for converting it. There are two ways of letting it know: o The filter is "intelligent" so it checks the file and recognizes the format. It is put together of a "pre-filter" for checking the format and many other programs who actually do the conversion. This, very roughly, is how apsfilter works which is installed by default with your S.u.S.E. Linux. o Different filters for different file formats. These filters are assigned to a certain queue by entering them in /etc/printcap. The user has to give the proper selection if calling lpr . There are eight options: _printcap_Entry__|____________if=______cf=______df=______gf=_______nf=______rf=______tf=______vf=______ lpr Option | -c -d -g -n -f -t -v Traditionally these options2 are assigned to special file formats (only set- ting of lpr s option is obligatory not the tradition3 . if= refers to the stan- dard filter whereas df= refers to DVIfiles (that come from TEX andALTEX). You can set an own DVI filter for the HP Laserjet 4 by entering: _______________________________________________________________________________________________________________________|| |||||||| | |||||||| df="/usr/lib/teTeX/bin/i386-linux/dvilj4 -e- -" | |||||||| | ||||||||_______________________________________________________________________________________________________________ | File contents 9.3.1: DVI filter entry in /etc/printcap 9.3.2 apsfilter A rather comfortable filter, apsfilter is included in the package aps. apsfilter exploits the following circumstances: 2 ____________________________________________________ see man page of lpr 3 Thus a so called "null-filter entry" cf=/bin/cat is allowed and is used with a lpr -c * * 147 9. Printers _________________________________________________________________________________________________________________* *_______________________________________@ o The standard format for printable files under Unix is PostScript. o There are many tools that turn other formats, such as tex and pictures into PostScript files (e. g. , dvips for DVI files and a2ps for ASCII files). o Then there is Ghostscript which is capable of transferring PostScript files to many non-PostScript formats for the printer (so to speak a collection of printer drivers). apsfilter now concludes almost any of these tools including the tools of rec- ognizing a certain format (file ) and decompression tools to one. It sends the result of the format recognition through the appropriate tools, and, at last, to the printer. In /var/lib/apsfilter4 , the central shell script the following file formats and compression types are available: o ASCII, DVI, PS, Data (PCL, : : :), GIF, TIFF, PBM, Sun Raster, X11 Bitmap o Compression types: compress, gzip, freeze apsfilter can either be configured by YaST or with its own menu driven con- figuration tool SETUP . Configuration using YaST This should suit your needs in the vast majority of cases. This configuration is done in YaST `Administration of the system' `Integrate hardware into system' `Configure printers' and already has been described in section 4.7.1, page 83. YaST creates three queues in /etc/printcap. file contents 9.3.2, page 149 shows this using a Canon BubbleJet 800 as an example. The three entries are constructed the same, that's why we shortened them. Spool directory (sd= ), log file (sd= ), count file (af= ) and standard filter (if= ) are assigned. To be precise all filters are symbolic links to /var/lib/apsfilter/bin/apsfilter . These queues are designed for use as: o lp standard queue: for any file format o ascii for printing files as ascii text, even if the spooling systems supposes another format5 . Output is done in a format defined in apsfilter . o raw for printing files that are already in a printable format; here no con- version is made. Configuration by hand A detailed configuration of apsfilter can be done by apsfilter 's SETUP script, which we will cover now in greater detail. o Install the package aps with YaST. _____________________________________________4 This used to be /usr/lib/apsfilter 5 e. g. if there are German umlauts 148 ____________________________________________________________________________________9.3.__Printer_filter._The_apsfilter_____* *______________________|| _______________________________________________________________________________________________________________________|||| |||||||| | ||||||||| # LABEL apsfilter || |||||||| # apsfilter setup Fri May 16 12:17:54 MEST 1997 | |||||||| # | |||||||| # APS_BASEDIR:/var/lib/apsfilter | |||||||| | |||||||| # | |||||||| lp|lp2|bjc800-a4-auto-mono|bjc800 auto mono:\ | |||||||| | |||||||| :lp=/dev/lp1:\ | |||||||| :sd=/var/spool/bjc800-a4-auto-mono:\ | |||||||| | |||||||| :lf=/var/spool/bjc800-a4-auto-mono/log:\ | |||||||| :af=/var/spool/bjc800-a4-auto-mono/acct:\ | |||||||| | |||||||| :if=/var/lib/apsfilter/filter/aps-bjc800-a4-auto-mono:\ | |||||||| :mx#0:\ | |||||||| | |||||||| :sh: | |||||||| # | |||||||| ascii|lp1|bjc800-a4-ascii-mono|bjc800 ascii mono:\ | |||||||| | |||||||| # [ ... abbreviated ... ] | |||||||| # | |||||||| | |||||||| raw|lp3|bjc800-a4-raw|bjc800 auto raw:\ | |||||||| # [ ... abbreviated ... ] | |||||||| | |||||||| # | |||||||| | ||||||||_______________________________________________________________________________________________________________ | File contents 9.3.2: apsfilter Queues in /etc/printcap o Change to /var/lib/apsfilter (used to be /usr/lib/apsfilter). o Launch the configuration script by SETUP . root@earth: > cd /var/lib/apsfilter root@earth:/var/lib/apsfilter > ./SETUP The greeting message now prints out some information, telling you that it: o can enter entries for parallel, serial and remote printers into /etc/printcap, o will create spool directories under /var/spool/lpd, o filter scripts will be linked to /var/lib/apsfilter/bin, o a global configuration file /etc/apsfilterrc and printer specific files /etc/apsfilterrc. are created. Now enter _- to proceed. For adding a new printer you should choose `ENTRY Add/Overwrite/Delete an apsfilter entry'. Now you should go through the menu items in or- der. o `DEVICE Which printer interface' lets you choose the interface for the printer and is added in the appropriate submenu. Normally the parallel port is /dev/lp1 (/dev/lp2 only applies if there are two parallel ports; /dev/lp0 defines the parallel port on a Hercules-graphics card. Compare section 9.1.1). While configuring a serial port you should make sure not to use the mouses device; furthermore you have to enter the baud rate. o `PRINTER Which printer driver' lets you select either of a couple of printers: PostScript, DeskJets, others and remote. See section 9.5. * * 149 9. Printers _________________________________________________________________________________________________________________* *_______________________________________@ Now we want to refer you to the abundant configuration file /etc/apsfilterrc which also contains some special settings as control sequences for printers, individual DVI filters, Ghostscript resolutions and many more. 9.4 Some words on Ghostscript If you don't own a PostScript printer, Ghostscript is the most popular choice for a filter. Ghostscript accepts PostScript files and contains many printer drivers for conversion. Ghostscript (gs(1) ) is a rather extensive tool with a lot of command line options. If you invoke Ghostscript directly it presents you its own dialog GS after having done the command line that can be left by entering quit. Unfortunately there is not enough space for covering all the gory details6 . Anyway, we want to point out the very helpful help command newbie@earth: > gs -h | less which lists all possible options as well as _ important! _ version number and the current list of supported devices. This list (by time of going to press of this book) is to be found under section 9.5. If you encounter any difficulties it might be helpful to invoke Ghostscript with a printable PostScript (.ps) file and to send the constructed printer spe- cific data directly to the device. A summary of suitable PostScript files can (e. g. ) be found under /usr/share/ghostscript//examples or /var/lib/apsfilter/test. The command for invoking Ghostscript e. g. for the included printer driver necp6 using a 360 360 resolution, and sending it to the printer might look like: newbie@earth: > gs -q -dNOPAUSE -dSAFER -sDEVICE=necp6 \ -r360x360 -sOutputFile=testfile.lpr testfile.ps GS> quit newbie@earth: > su root@earth: # cat testfile.lpr >/dev/lp1 Hint: in /usr/share/ghostscript//doc helpful documen- tation can be found (e. g. in devices.txt). 9.5 List of supported printers Besides PostScript and HP-Deskjet printers lots of other printers are sup- ported by Ghostscript (version 4.03): _______________________________________________________________________________________________________* *_________ Apple _______________________________________________________________________________________________________* *_________ ____________________________________Table_9.1:_to be continued... 6 A first overview can be found in Ghostscript's man page. Fortunately apsfilter releases you from the tedious construction of command line options. 150 _____________________________________________________________________________________9.5.__List_of_supported_printers__________* *___________________ ______________________________________________________________________________________________________________________ appledmp Apple Dot Matrix Printer (even Imagewriter) iwhi Apple Imagewriter, high resolution iwlo Apple Imagewriter, low resolution iwlq Apple Imagewriter, 320x216 dpi Canon bj10e Canon BubbleJet 10e bj200 Canon BubbleJet 200 bjc600 Canon BubbleJet 600c, 4000c (Colour) bjc800 Canon BubbleJet 800c (Colour) lbp8 Canon LBP-8II lips3 Canon LIPS III DEC declj250 DEC LJ 250 la50 DEC LA50 la70 DEC LA70 la75 DEC LA75 la75plus DEC LA75 Plus lj250 DEC LJ250 ln03 DEC LN03 Epson ap3250 Epson AP3250 eps9high Epson FX-80-compatible, 240 dpi eps9mid Epson FX-80-compatible, 120 dpi epson Epson FX-80-compatible 9 or 24 dot-matrix printer epsonc Epson LQ-2550, Fujitsu 1200/2400/3400, Colour prints epscp2 Epson ESC/P2 st800 Epson Stylus 800 stcolor Epson Stylus Color Hewlett-Packard cdeskjet HP DeskJet 500C, black cdj500 HP DeskJet 500C, 540C cdj550 HP DeskJet 550C, 560C cdjcolor HP DeskJet 500C, colour cdjmono HP DeskJet 500C, black deskjet HP DeskJet, HP DeskJet Plus ______________________________________________________________________________________________________________________ Table 9.1: to be continued... * *151 9. Printers _________________________________________________________________________________________________________________* *_______________________________________@ _______________________________________________________________________________________________________* *_________ djet500 HP DeskJet 500 djet500c HP DeskJet 500c dnj650c HP DesignJet 650C laserjet HP LaserJet ljet2p HP LaserJet IIp ljet3 HP LaserJet III ljet3d HP LaserJet IIID ljet4 HP LaserJet IV ljetplus HP LaserJet Plus lj4dith HP LaserJet IV, ditherd lp2563 HP 2563B LinePrinter cp50 HP PaintJet 300XL paintjet HP PaintJet Colour printer pj HP PaintJet XL, Alternative pjetxl HP PaintJet 300XL pjxl HP PaintJet 300XL pjxl300 HP PaintJet 300XL, HP DeskJet 1200C IBM ibmpro IBM Proprinter, 9 dot-matrix printer jetp3852 IBM-Jetprinter 3852 OKI oki182 OKI MicroLine 182 okiibm OKI MicroLine, IBM -compatible Tektronix t4693d2 Tektronix 4693d, colour t4693d4 Tektronix 4693d, colour t4693d8 Tektronix 4693d, colour tek4696 Tektronix 4695/4696 Andere imagen Imagen ImPress m8510 C.Itoh M8510 necp6 NEC P6, P6+, P60 bei 360 dpi nwp533 Sony Microsystems NWP533 Laser printer (only Sony) oce9050 OCE 9050 r4081 Ricoh 4081 Laser printer sj48 StarJet 48 _______________________________________________________________________________________________________* *_________ Table 9.1: to be continued... 152 ____________________________________________________________________________________9.6.__Printer_checklist:_apsfilter______* *______________________ ______________________________________________________________________________________________________________________ sparc SPARCprinter xes Xerox XES (2700, 3700, 4045) ______________________________________________________________________________________________________________________ Table 9.1: Supported printers (Ghostscript v. 4.03) Of course any HP4 compatible PostScript printers are supported too. During installation of apsfilter or during configuration using YaST the line given in the left column has to be set as printer device (e. g. djet500). If the printer is not supported this does not necessarily mean that the apsfilter can not be used at all. In this case try a similar printer. Quite often this driver can be used. A quite common error is setting the wrong resolution. If this is the case the correct value (the DPI rate the printer accepts) can be set in $GS__RESOL in /etc/apsfilterrc (e. g. GS__RESOL=360x360 . Some entries (commented) do already exist). 9.6 Printer checklist: apsfilter o Has package aps, series been installed for apsfilter ? o Has apsfilter been configured by YaST or by /var/lib/apsfilter/ SETUP o Does the kernel contain TCP/IP support? (check with dmesg ) o Have package nkita and package nkitb been installed? o If the printer does not support the default setting of 300 300 dpi: has the correct resolution been set according to your printers manual by YaST or been entered in /etc/apsfilterrc o What does lpc -status say? o Try lpc up all o At least the raw queue should work on any printer and enabling you to print ASCII files. o Did you install the correct port (comp. section 9.1.1)? o The kernel should not contain a PLIP driver (dmesg ). * * 153 9. Printers _________________________________________________________________________________________________________________* *_______________________________________@ 154 Chapter 10 Linux-machines and hardware 10.1 Preliminary notes Meanwhile it is possible to integrate most PC components more or less easy into a Linux system. How to handle this in detail und which software is avail- able can be found in this chapter1 . Solutions for problems with "printings" ref. chapter 9. Some hints concerning PCMCIA-Hardware are to be found in chapter 6. 10.2 ISA "Plug and Play" hardware 10.2.1 General In the meantime so called "PnP" (Plug and Play) are available (some peo- ple refer to them as "Plug and Pray"). Anyway. These adapters have been designed fora automatic recognition. The user should not bother with con- figuration. This might apply in the "Windows world". It used to be quite a problem for Linux to access these cards unless you were able to disable PnP (which adapter does have jumpers?) 10.2.2 Configuration There are two tools for configure these adapters: o pnpdump o and isapnp The first one is for creating the configuration file /etc/isapnp.conf. The latter reads this configuration file and sets up the PnP device. 10.3 Changeable media 10.3.1 General Lots of "changeable media" may be used on Linux. These in particular are: disk drives, ZIP drives, JAZ drives, SyQuest drives, MO (magneto optical) drives - everything is possible. 1 ____________________________________________________ At least this is a try ;-) * * 155 10. Linux-machines_and_hardware___________________________________________________________________________________________________* *_______________________________________@ 10.3.2 Disk drives For accessing disks you need to enable "floppy" support in the kernel (see section 13.4.5, page 187). You may access MS-DOS disks via the mtools . Usually one assigns a minix on disks. These may be mounted as usual. 10.3.3 LS 120 drives LS 120 are ATAPI devices and are supported by the IDE driver . These devices may be integrated just as (E)IDE hard drives. 10.3.4 ZIP drives There are different ZIP drives for parallel ports, ATAP as well as SCSI. If you plan to use the parallel port versions you will need SCSI hard disk support (see section 13.4.7, page 192) as well as the ppa driver. The ppa may be found under the "low-level" SCSI drivers. For avoiding conflicts with the printer you should activate the ppa as a mod- ule. Thus it may be loaded on demand using modprobe (see section 13.2, page 181). Now you may access ZIP disks as well as SCSI hard drives. For the ATAPI version there applies section 10.3.3, page 156. For SCSI versions section 10.3.5, page 156. 10.3.5 SCSI changeable media drives Magneto optical drives (MO), SyQuest drives, ZIP and JAZ drives attached to a SCSI interface are considered as normal SCSI devices. 10.4 Scanner 10.4.1 General Meanwhile both software and drivers are available for addressing scanners under Linux. For being able to use this software, the scanner should be connected to the SCSI port and the proprietary delivered controllers should be avoided. ________________________________________________________________________________________________________* *________ | The software SANE - Scanner Access Now Easy is still only available in * * ||| | * * ||| | a beta phase; for that reason system crashes and more can not be excluded. * * ||| |_______________________________________________________________________________________________________* *________|||_ From S.u.S.E. Linux5.2 on you will be able to install a scanner with YaST. Be aware that the scanner uses the generic SCSI devices. That is it directly accesses the scanner on a low level format. So, if you are not familiar with these devices make sure you read the instructions and READMEs before you change the settings! 156 ______________________________________________________________________________________________________________10.4.__Scanner* *______________________ 10.4.2 Hardware: Flatbedscanner, Cameras, automatic document feed : : : For using your scanner under Linux, a matching "Scanner Device File" needs to installed. Under Linux, scanners are addressed as "Generic SCSI Devices" using the device drivers /dev/sg0 to /dev/sg6; under S.u.S.E. Linux how- ever only the generic SCSI drivers 0 -3 come as a standard. In case the scanner should be linked to the generic device /dev/sg4 or higher, you have create your own device. Please refer manpage of sane-scsi (man sane-scsi ) and manpage of mknod (man mknod ). Set your SCSI-Scanner to the end of the SCSI-chain. In that case it is easy to determine the device - just count all the SCSI devices you have connected to the chain. Example: You have connected an SCSI-harddrive, a CD-ROM-drive and an SCSI-scanner to the SCSI-bus. This makes up three SCSI-device. The con- clusion is, since the scanner is linked up at the end of the chain, that the right device driver is /dev/sg2 (Attention: the count starts at 0 ), the third generic SCSI-device. Now place a link from that device to /dev/scanner: root@earth: # ln -s /dev/sg2 /dev/scanner Afterwards you have to place the properties for the generic SCSI-Device. SANE also expects write properties: root@earth: # chmod -m 777 /dev/sg2 Another option is to create a group `scanner' (e. g. mit YaST) and to locate only those users to the group which are allowed to use the scanner. Of course you have to execute: root@earth: # chown root.scanner /dev/sg2 at the end. In any case you should be able to access the scanner right now. 10.4.3 Software Install the package sane2 . After the installation process, please make sure to study the README-files located in /usr/doc/packages/sane! You then get an executable program xscanimage . When you start this program, it should be able to recognize your scanner if it belongs to the supported ones. The graphics package "the GIMP" (package gra, series gimp) is designed that way - and realized with the same library - that SANE is able to recog- nize it as "plug-in". For managing this,you have to create the following link after the installation of GIMP (the path \tld/.gimp must exist already; it is automatically created at the first start of gimp ): newbie@earth: > cd ~/.gimp/plug-ins newbie@earth:~/.gimp/plug-ins > ln -s /usr/bin/xscanimage . Now the SANE-Tool is located automatically and is displayed in the menu `Xtns' of "the GIMP". If you now scan an image, it is loaded directly into GIMP and you can access and work with it. 2 ____________________________________________________ see. ftp://ftp.mostang.com/pub/sane/. * * 157 10. Linux-machines_and_hardware___________________________________________________________________________________________________* *_______________________________________@ 10.4.4 Problems In case it is not possible to locate the scanner, please check if it was located by the SCSI-controller. This can be done e. g. via: root@earth: # cat /proc/scsi/scsi and then you should see line similar to: Attached devices: Host: scsi0 Channel: 00 Id: 00 Lun: 00 Vendor: SEAGATE Model: ST32550N Rev: 0016 Type: Direct-Access ANSI SCSI revision: 02 Host: scsi0 Channel: 00 Id: 01 Lun: 00 Vendor: PIONEER Model: CD-ROM DR-U10X Rev: 1.07 Type: CD-ROM ANSI SCSI revision: 02 Host: scsi0 Channel: 00 Id: 02 Lun: 00 Vendor: QUANTUM Model: FIREBALL__TM3200S Rev: 300N Type: Direct-Access ANSI SCSI revision: 02 Host: scsi0 Channel: 00 Id: 06 Lun: 00 Vendor: SCANNER Model: Rev: 2.00 Type: Scanner ANSI SCSI revision: 01 CCS If you don't get a prompt at this point, the SCSI-controller has not been located. In this case, please read in the manual section 13.4.7 on how to integrate your controller. Should besides the existence of a scanner xscanimage not able to recognize your scanner, the reason for the problem can be that o the scanner is not getting support yet or o that you have to transfer certain parameters to xscanimage. Both cases are handled in the README files that came with the package. Gen- erally one can say that most HP-scanners meanwhile are supported; Mustek- scanners however not in all cases. Some of them perform quite well, others not at all. Of the newer Umax-scanners the majority is supported. We're currently working on the support for older Umax-scanners. 158 Chapter 11 Special features of S.u.S.E. Linux 11.1 Keyboard layout For unifying keyboard mapping of certain programs the following files had to be adapted: /usr/lib/X11/Xmodmap /etc/inputrc /etc/skel/.exrc /etc/skel/.less /etc/skel/.lesskey /etc/csh.cshrc /etc/termcap /usr/lib/terminfo/x/xterm /usr/lib/X11/app-defaults/XTerm /usr/share/emacs/19.34/site-lisp/term/*.el /usr/lib/joerc These changes only apply to applications that make use of terminfo entries or change their configuration files (vi , less , etc.). Every "none" S.u.S.E. application should be adapted. 11.2 Differences in the structure of the file system S.u.S.E. Linux tries to be File System Standard conform (FSSTD), therefore paths of some files had to be changed: o Lock files are to be found under /var/lock. o Log files (boot messages, warnings, output of pppd and many more) have been moved to /var/spool/uucp/Log as well as to /var/spool/ uucp/Stats. o The UUCP log files are know located under /var/spool/uucp/Log as well as under /var/spool/uucp/Stats. o The directory /usr/data has been replaced by /usr/share; all pictures and sounds have been shifted there. o cron tables are now located under /var/cron/tabs instead of /var/ lib/cron. Since the introduction of S.u.S.E.Linux 5.0 , the file /etc/crontab has been configured as a systemwide library; additionally you must insert after the time setting, under which user the specific order should be han- dled. * * 159 11. Special features_of_S.u.S.E._Linux____________________________________________________________________________________________* *_______________________________________@ 11.3 linuxrc linuxrc is started during the boot-up of the kernel, before the "real" booting commences. 1 . This allows you to boot a small modularized kernel and to load the required drivers as modules. It is (at the moment) still possible, to include all drivers which the kernel supports und which are needed for the installation (including PCMCIA) on one disk. linuxrc is your assistant for loading all relevant hardware drivers. You can even use linuxrc as a boot-disk for an already installed system (e. g. as a rescue-disk). You can even start a totally independent RAM-disk based rescue system (e. g. , if something severe happened to your hard disk or you have simply forgotten your root password. More in section 11.4. 11.3.1 Main menu After you have selected language, screen and keyboard you find yourself in linuxrc| 's main menu. * * | ________________________________________________________________________________________________________* *________|||| | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | |_______________________________________________________________________________________________________* *________ | Figure 11.1: Main menu of linuxrc You start the installation with `Installation / start-up system'. Whether you can go there directly mainly depends on your hardware: If all components that are needed for installation have already been recog- nized by the kernel you do not need to load additional drivers. This mainly applies to those that only have (E)IDE adapters (and of course only (E)IDE hard disks and CD-ROM). If there is a SCSI adapter installed which is necessary for installation2 you have to load the corresponding SCSI module. The same applies if you want _____________________________________________1 Of course the kernel has to be configured (section 13.4.5) 2 An adapter with only a scanner connected to it is not necessary at boot time 160 ________________________________________________________________________________________________________________11.3.__linux* *rc____________________ to install via an existing network. Here the suitable module has to be loaded first. Furthermore there are a lot of older CD-ROMs that are driven by proprietary controllers and which therefore need their own kernel module. If PCMCIA devices are connected to a laptop you need modules as well. 11.3.2 System information If you are not sure about your hardware the boot messages might help you, You can check some system information under `System information'. Here you can check the used interrupts, I/O ports, used main memory and recognized|PCI|devices as detected by Linux. | _______________________________________________________________________________________________________________________|||| |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | ||||||||_______________________________________________________________________________________________________________ | Figure 11.2: System information The next lines show how a hard disk and a CD-ROM connected to an (E)IDE controller announce their start. In this case you do not need to load additional modules: hda: ST32140A, 2015MB w/128kB Cache, LBA, CHS=1023/64/63 hdb: CD-ROM CDR-S1G, ATAPI CDROM drive Partition check: hda: hda1 hda2 hda3 < hda5 > If you booted a kernel that already has a SCSI driver integrated you of course do not need this SCSI driver as a module either. Quite typical announcements when loading SCSI adapters and connected devices might resemble: scsi : 1 host. Started kswapd v 1.4.2.2 scsi0 : target 0 accepting period 100ns offset 8 10.00MHz FAST SCSI-II * * 161 11. Special features_of_S.u.S.E._Linux____________________________________________________________________________________________* *_______________________________________@ scsi0 : setting target 0 to period 100ns offset 8 10.00MHz FAST SCSI-II Vendor:TyQUANTUMpe: Model:DiVP32210rect-Access Rev: 81H8 ANSI SCSI revision* *: 02 Detected scsi disk sda at scsi0, channel 0, id 0, lun 0 scsi0 : target 2 accepting period 236ns offset 8 4.23MHz synchronous SCSI scsi0 : setting target 2 to period 248ns offset 8 4.03MHz synchronous SCSI Vendor: TOSHIBA Model: CD-ROM XM-3401TA Rev: 0283 Type: CD-ROM ANSI SCSI revision: 02 scsi : detected 1 SCSI disk total. SCSI device sda: hdwr sector= 512 bytes. Sectors= 4308352 [2103 MB] [2.1 GB] Partition check: sda: sda1 sda2 sda3 sda4 < sda5 sda6 sda7 sda8 > | * * | ________________________________________________________________________________________________________* *________|||| | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | |_______________________________________________________________________________________________________* *________ | Figure 11.3: Kernel messages 11.3.3 Loading of modules You select which kind of modules you need. If booted via disk the corre- sponding data has to be read by linuxrc and displayed in a list. If you have booted from CD or from DOS (via loadlin ) these modules are already set in linuxrc . This saves the tedious loading but needs additional memory. If your machine is supplied with less than 8 MB of RAM you have to boot from disk. linuxrc offers you a list of available drivers. On the left there is the name of the module and on the right you can see a short message telling you what it can be used for. For some components there are various drivers at your disposition (or newer alpha drivers). These are offered as well. 162 ________________________________________________________________________________________________________________11.3.__linux* *rc____________________ |||||||| | _______________________________________________________________________________________________________________________|||| |||||||| | ||||||||| || |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | ||||||||_______________________________________________________________________________________________________________ | Figure 11.4: Load modules |||||||| | _______________________________________________________________________________________________________________________|||| |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | ||||||||_______________________________________________________________________________________________________________ | Figure 11.5: Selection of SCSI-drivers 11.3.4 Passing parameters If you have found a suitable driver, move to it with the cursor and press _- . Now there is a mask where you can add additional parameters for this module. More on module-parameters can be found in section 7.3.3, page 111. We would just like to point out that on the contrary to the LILO prompt parameters for the same module have to be separated by blanks. In most cases it is not necessary to specify the hardware in detail; most drivers find their components automatically. Most networking cards and proprietary CD-ROM drives however need parameters. In case of doubt just try _- . * * 163 11. Special features_of_S.u.S.E._Linux____________________________________________________________________________________________* *_______________________________________@ | * * | ________________________________________________________________________________________________________* *________|||| | * * | || * * || | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | |_______________________________________________________________________________________________________* *________ | Figure 11.6: Entering parameters for loading a module Recognizing and initializing of certain hardware can take quite some time. Switching to console 4 ( Alt + F4 ) lets you watch the kernel messages while loading. SCSI drivers need quite some time as they have to wait for each device to load. If loading succeeded the messages are displayed by linuxrc just for make sure everything ran smoothly. Otherwise (if it fails) the messages might give you a hint why it failed. 11.3.5 System / Start installation If you achieved hardware support via modules you can switch to `System / Start installation'. There are many items to choose from for installation and/or rescue system: 11.4 The S.u.S.E. rescue system 11.4.1 Overview Since version 4.2, S.u.S.E. Linux contains a completely independent rescue system which enables you to access your system ("from outside") in a case of emergency. This system consists out of a special selection of system tools. They should be sufficient to solve most hardware and configuration errors and problems. The rescue system consists of a boot-disk (the same as for the new-installation) and an installation CD. Since this boot-disk can be created at any time by means of the CD (the boot-disk can be found in disks) this serves as a rather secure and easy to handle rescue system. If you copy this file using the com- mand root@earth: # /sbin/badblocks -v /dev/fd0 1440 root@earth: # dd if=/cdrom/disks/rescue of=/dev/fd0 bs=18k 164 __________________________________________________________________________________11.4.__The_S.u.S.E._rescue_system_________* *______________________ |||||||| | _______________________________________________________________________________________________________________________|||| |||||||| | ||||||||| || |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | ||||||||_______________________________________________________________________________________________________________ | |||||||| Figure 11.7: Target of linuxrc | _______________________________________________________________________________________________________________________|||| |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | ||||||||_______________________________________________________________________________________________________________ | Figure 11.8: Selecting source media in linuxrc or using the DOS command (provided Q is your CD-ROM under DOS) Q:\dosutils> rawrite.exe onto a 2nd error-free disk ("rescue-disk") you also can launch the rescue system from the boot-disk and this rescue disk. Please be aware that you cannot mount the rescue disk because it does not contain a file-system but a compressed image (the uncompressed image would take up to 3.5 MB which would not fit on a disk). If you want to have a look at it you have to decompress it in advance and then mount it (the name has to be `root'). Provided that your Linux kernel supports the loop-device you enter: * * 165 11. Special features_of_S.u.S.E._Linux____________________________________________________________________________________________* *_______________________________________@ root@earth: # /bin/cp cdrom/disks/rescue /root/rescue.gz root@earth: # /bin/gunzip /root/rescue.gz root@earth: # /bin/mount -t minix -o loop /root/rescue /mnt Now you can have a look at it under /mnt. ________________________________________________________________________________________________________* *________|| | Always have some boot-disks at hand! Creating such a disk does not take * * |||| | * * ||| | long; well, anyway, it takes less time than searching for a disk in case of * * ||| | emergency (in such a situation you can be sure that your CD-ROM does * * ||| | * * ||| | not work either (Murphy's law...))! * * ||| |_______________________________________________________________________________________________________* *________|||_ 11.4.2 Launching the rescue system The rescue system, like as the normal installation, is launched using the S.u.S.E. boot-disk. Step by step: o Requirements: The floppy drive is bootable (if not so, please run CMOS setup to modify the settings). o Launch system with the S.u.S.E. boot-disk. o Enter language, keyboard etc. until you get to the main menu. o Now select `Installation/Start system'. o Insert the CD or the disk containing the compressed image of the rescue system. o In `Installation/Start system' you should choose either, depend- ing on the source media: `Load rescue system from CD' , or. `Load rescue system from disk'. The rescue system is now being decompressed and loaded into a RAM-disk as new root-file system, mounted and started. Now it is ready for use. 11.4.3 Working with the rescue system The rescue system provides three virtual consoles on keys Alt + F1 to Alt + F3 . Here `root' may log in without a password. Alt + F4 brings you to the system console where you can view the kernel and syslog messages. A shell and lots of other useful utilities (net-tools) can be found under /bin. In sbin you cand find efsck which is very useful for checking and repairing file systems. In sbin there are the important binaries for system maintenance, such as fdisk , mkfs , mkswap , mount , mount , init , shutdown , as well as ifconf* *ig , route , netstat for maintaining the network. As editor, vi is located under /usr/bin; also additional tools like (grep ,find ,less , etc.) and, most important of all, telnet are avaiable. Example: Accessing your normal system For mounting your Linux system, you should use mountpoint /mnt. Of course you can also use or generate another directory. $PATH already contains the standard directories and subdirectories as /mnt/ bin and /mnt/usr/bin. Consider the following example: Provided your 166 ______________________________________________________________________________________________________________________11.5._* *_init_________________ _______________________________________________________________________________________________________________________ |||||||| | |||||||| /dev/sdb5 swap swap defaults 0 0 | |||||||| | |||||||| /dev/sdb3 / ext2 defaults 1 1 | |||||||| /dev/sdb6 /usr ext2 defaults 1 2 | |||||||| | ||||||||_______________________________________________________________________________________________________________ | File contents 11.4.1: Example: /etc/fstab system is structured (according to /etc/fstab) as given in file contents 11.4.1, page 167: Then you may mount it step by step to /mnt; just execute the following commands in order: root@earth:/ # mount /dev/sdb3 /mnt root@earth:/ # mount /dev/sdb6 /mnt/usr Now you can access your whole system and (e. g. ) correct mistakes in con- figuration files (as /etc/fstab, /etc/passwd, /etc/inittab). Of course those files now are located under /mnt/etc instead of /etc! _______________________________________________________________________________________________________________________|| |||||||| Every experienced Linux user does a hard copy of /etc/fstab and of the ||| | |||||||| output of fdisk-l to his "files". Even completely corrupt partitions can || | |||||||| || | |||||||| be reassigned if the exact geometry and the parameters are known! || | ||||||||_______________________________________________________________________________________________________________||_ Example: Repairing file-systems Damaged file-systems are tricky tasks for the rescue system. This might happen after a failed shutdown (current down) or a system crash. Generally, file-systems can not be repaired on a running system. If you encounter really severe problems you might not even be able to mount your root-file-system and the system start ends up in a "kernel panic". Here the only chance is to repair the system from the "outside" using a rescue system. The S.u.S.E. Linux rescue system contains the utilities e2fsck and, for the diagnosis, dumpe2fs . These should cover most problem. In an emergency there normally are no man pages available, that is why we included them in this manual under appendix E, page 405. Example: If mounting a file-system fails due to an invalid superblock, then e2fsck would most probably fail too. If this is the case your superblock might be corrupted too. There are copies of the superblock located every 8192 blocks (8193, 16385 : : :) If your superblock got corrupted you might want to try one of the copies instead. This is achieved by entering the command: root@earth: # e2fsck -f -b 8193 /dev/damaged__partition The -f option forces the file-system check and overrides e2fsck 's error so that _ since the superblock is intact _ everything is fine. 11.5 init The init scripts /etc/rc.d, which so far have regulated the start up of the system, have been substituted through a system which allows an orderly * * 167 11. Special features_of_S.u.S.E._Linux____________________________________________________________________________________________* *_______________________________________@ administration of the different Runlevels similary to System V . Since the filesystem-standard does not allow exectutable files under /etc, the init - scipts have been located under /sbin/init.d. 11.6 Changes made to software packages 11.6.1 package curses On our S.u.S.E. Linux CD we newly included the package package ncurses. The corresponding libraries are named libncurses.so.. This means that some Makefiles have to be adapted concerning the linker. Packages of your own should be linked with the command -lncurses instead of -lcurses . For those of you who want to do that anyway try -I/usr/include/termcap -I/usr/include/curses -L/usr/lib/termcap -L/usr/lib/curses 11.6.2 manpages Some man pages (e. g. tar ) are not maintained anymore. They have been replaced by info files. Info (info ) is GNU's hypertext system. Typing man info gives you first help. info can be launched via emacs -f info or standalone: info . The most comfortable way is xinfo . 11.6.3 sendmail Since GNU dbm has been replaced by BSD db the commands for installing and creating mail tables have slightly changed. Instead of entering makemap dbm mailertable you have to type makemap hash mailertable 11.7 Misc The tasks of /etc/securetty have been transferred to /etc/login.defs. 168 Chapter 12 The S.u.S.E. boot concept Booting and initialisation of a Unix system challenges even an experienced system administrator. This chapter gives you an short overview of the S.u.S.E. Linux' boot concept. This concept is much more complex but also more flexible than other Linux' concepts. It is based on a modern boot concept of a System V workstation as described in [Frisch, 1993 ]. The simple words "Uncompressing Linux..." show that the kernel, takes over control over all your hardware. It checks and sets your console1 for reading BIOS settings and to initialize basic hardware interfaces. Next your drivers "probe" existing hardware for initializing it accordingly. After parti- tion check and mounting of root-filesystem 2 the kernel starts /sbin/init which starts the main system with all its programs and configurations. The kernel will maintain the whole system, including CPU time of programs and hard- ware access. 12.1 The init program The program /sbin/init is responsible for a correct initialization for all sys- tem processes; thus being the father of all processes in the whole system. Init takes up a special position: Init is directly started by the kernel and is resistant against signal 9 which normally enables you to kill processes. All further programs are either started directly by init or from one of its "child" processes. Init is centrally configured via /etc/inittab file; here the so-called "run- levels" are defined (more about run-levels in the next chapter); here is set what should happen in the several levels. Depending on entries in etc/inittab several scripts are started by init which, for clarity, all reside in the same directory /sbin/init.d. The whole starting up of the system (of course the shutting down as well) is therefore maintained by init ; from this point of view the kernel can be considered as a "background process" whose task it is to maintain all other processes and to adjust CPU time and hardware access according to requests of other programs. 1 ____________________________________________________ precisely BIOS registers of graphic cards and output format. 2 assigning root partition to " / " * * 169 12. The S.u.S.E. boot_concept_____________________________________________________________________________________________________* *_______________________________________@ 12.2 Runlevels Under Linux there are so-called run levels which define how the system is going to be started up. After booting the system starts as defined in /etc/inittab in line initdefault . An alternative is assigning a special runlevel at boot time (e. g. at the LILO-prompt): the kernel passes parameters which it doesn't need directly to init . For changing run-levels while the sys- tem is up and running you can just enter init with the appropriate number. e. g. root@earth:/ > init S brings you to single user mode which is for maintenance and administration of your system. After having finished work in S mode the system administra- tor can change run-level to 2 again by typing root@earth:/ > init 2 Now all essential programs are started and users can log in and work with the system. The table below gives you an overview about available run levels 3 : ________________________________________________________________________________________________________* *________ __Runlevel_________Meaning_____________________________________________________________________________* *________ 0 Halt S Single User Mode 1 Multi User without network 2 Multi User with network (Standard) 3 Multi User with network and xdm 4 Unused 5 Unused 6 Reboot ________________________________________________________________________________________________________* *________ Table 12.1: List of Linux' valid runlevels Thus leading to the fact that you can halt the system using root@earth:/ > init 0 or restart it with root@earth:/ > init 6 respectively. If you have already installed and configured the X-Window System properly (Section section 14.1) and want the users to log in via a graphical user in- terface you can easily change runlevel to 3 . You should give it a try first by typing root@earth:/ > init 3 to see whether the system works as you expected. _____________________________________________3 Runlevel 1 should not be used on a system whose /usr-partition is mounted via NFS! 170 _________________________________________________________________________________________12.3.__Changing_of_runlevels_______* *______________________ _______________________________________________________________________________________________________________________|| |||||||| With a damaged /etc/inittab you can end up in a system which can ||| | ||||||||| not be brought up properly! Therefore be extremely careful while editing || || |||||||| /etc/inittab! ||| | ||||||||_______________________________________________________________________________________________________________||_ 12.3 Changing of runlevels Generally there are a couple of things that happen if you change runlevels. First, so-called stop-scripts of the current runlevel are launched (closing down some programs which are essential for the current runlevel); thereafter start scripts of the new runlevel are started. Here, in most cases, some programs will be started. For more clarity we will show you a change from runlevel 2 to 3: o The administrator (`root')4 tells init to change runlevels: root@earth:/ > init 3 o init now consults its configuration file (/etc/inittab) and realizes that it should start /sbin/init.d/rc with the new runlevel as parameter. o now rc calls all the stopscripts of the current runlevel; but only for those where there is no start script in the selected new runlevel. In our example these are all scripts which reside in /sbin/init.d/rc2.d (old runlevel was 2) and which start with a `K' 5 . The number following `K' guaran- tees a certain order to start as there are some dependencies which have to be taken into consideration. o The last thing to start are the start scripts of the new runlevel; those are (in our example) under /sbin/init.d/rc3.d and begin with an `S'. The same procedure regarding the order in which they are started is applied here. If you want to change into the same runlevel which you are already in, init only checks /etc/inittab for changes and starts the appropriate steps (e. g. starting a getty on another interface). 12.4 Init scripts Scripts under /sbin/init.d are divided into two sections: o Scripts which are executed directly by init : this only applies while boot- ing as well while shutting down the system immediately (power failure or pressing Ctrl Alt Del by the user). o Scripts which are started indirectly by init : This happens while changing the runlevel; here generally /sbin/init.d/rc is executed which guarantees the correct order of the relevant scripts. All scripts are located under /sbin/init.d. Scripts for changing the run- level are also to be found there but are called via symbolic link from one of the subdirectories /sbin/init.d/rc0.d to /sbin/init.d/rc6.d. This 4 ____________________________________________________ of course changing of runlevels can only be done by `root' 5 names of stop scripts always start with an `K' whereas start scripts always start with an `S' * * 171 12. The S.u.S.E. boot_concept_____________________________________________________________________________________________________* *_______________________________________@ is just for clarity reasons and avoids duplicate scripts (e. g. if they are used in several runlevels). Since every script can be executed as start as well as stop-script those scripts have to "understand" parameters "start" and "stop". Example An example: While leaving runlevel 2 /sbin/init.d/rc2.d/K40network is executed among others; this results in /sbin/init.d/network being ex- ecuted from /sbin/init.d/rc with the "stop" parameter. When entering runlevel 3 the same script is started but with the "start" parameter instead of "stop". Links in those runlevel-specific subdirectories simply serve for allocation pur- poses of the scripts to a certain runlevel. Here we give an overview of all scripts being used with a short description: Boot and Here we give you a short introduction on the boot and stop scripts that are shutdown launched first (or last respectively) as well as an explanation of the maintain- ing script. o boot Is executed while starting the system directly by init . It is independent from the chosen runlevel and is only executed once. Here filesystems are checked, the kernel daemon is launched some unnecessary files under /var/lock are deleted and the network is being configured for loopback device (if it is selected in /etc/rc.config). Furthermore the system time is setup and Plug and Play hardware is initialized by the isapnp tools . If an error occurs while automatically checking and repairing the filesys- tem the system administrator can intervene after having entered the root password. Directory /sbin/init.d/boot.d is assigned to this script; all scripts in this directory are executed while bringing up the system. This is the right place for your own personal extensions which should be executed only once. Last to be executed is the script boot.local o boot.local Here you can enter additional commands to be executed at boot time before changing into a runlevel; it can be compared to AUTOEXEC.BAT on DOS systems. o boot.setup General settings that have to be performed while changing from single user mode to another runlevel. Here keyboard maps are loaded and the kerneldaemon is started which takes care of the automatic loading of modules. o halt This script is only executed while changing into runlevel 0 or 6. Here it is executed either as halt or as reboot. Whether the system shuts down or reboots depends on how halt is called. o rc The script of overriding importance which is executed at each change of 172 ______________________________________________________________12.5.__/etc/rc.config______________and_/sbin/SuSEconfig_______* *______________________ runlevels. It calls the appropriate stop scripts of the current runlevel and the start scripts of the new selected runlevel. You may add necessary scripts to this skeleton very easily. A template Creating my be found under /sbin/init.d/skeleton. For enabling a script via scripts /etc/rc.config it is recommended to create a variable in this file. Additional parameters should only be added if really needed. (see the /sbin/init.d/gpm script for reference). Now you need to create the links from the correspondent rc to your script. This is for making sure that it is launched when changing into this given runlevel (see above section 12.3, page 171 for script names etc.). manpage of init.d (man 7 init.d ) gives you all the needed technical background. 12.5 /etc/rc.config and /sbin/SuSEconfig Almost every configuration of S.u.S.E. Linuxcan be done via a central con- figuration file called /etc/rc.config. Here a couple of environment vari- ables are set which are (among others) checked by the init scripts; each of the scripts under /sbin/init.d executes /etc/rc.config as first steps for reading the values of those variables. Moreover very many configuration files can be generated with /etc/rc. config; this is the task for /sbin/SuSEconfig . If you change the network configuration (e. g. ) the file /etc/resolv.conf will be re-generated as it depends on the configuration you have made. So, if you change /etc/rc.config manually you should invoke /sbin/SuSEconfig afterwards to make sure all changes are made at the cor- rect places. If you change configuration with YaST you don't have to bother; YaST automatically executes /sbin/SuSEconfig and updates your configura- tion files. This concept enables you to make basic changes with your configuration without having to reboot your system. Since some changes are rather complex some programs probably have to be restarted for the changes to take effect. This procedure has been explained in network configuration (see section 16.3) where those programs are forced to be restarted using the command: root@earth:/ > /sbin/init.d/network stop root@earth:/ > /sbin/init.d/network start As you can see you can easily start init scripts by hand. Generally we recommend the following steps for configuring your system: o Bring the system into single user mode: root@earth:/ > init S As an alternative you can select runlevel 1; here you have the possibility of login in on several consoles: root@earth:/ > init 1 o Change configuration file /etc/rc.config as needed. This can be done by an editor of your choice or by YaST under `Changing configuration file' (look at section 12.6). * * 173 12. The S.u.S.E. boot_concept_____________________________________________________________________________________________________* *_______________________________________@ o Execute /sbin/SuSEconfig to make the changes take effect. If you have changed /etc/rc.config via YaST this is done automatically. o Bring your system back into the previous runlevel: root@earth:/ > init 2 This procedure mainly is relevant if you have changed system wide settings (e. g. network configuration); it is not necessary for "small" changes to go into single user mode; but it ensures you that all relevant programs are correctly restarted. 12.6 Options of configuration Now all parameters of the system are described including their default set- tings. If you don't use YaST for changing /etc/rc.config make sure to set "empty" parameters as two quotation marks (e. g. KEYTABLE="" ) and to surround parameters that contain a blank by quotation marks (parameters con- sisting of only one word do not have to be quoted). In our description each parameter is given a value for making their settings as clear as possible: o ENABLE__SUSECONFIG=yes With this entry you can disable SuSEconfig completely. Please don't contact our support if you have trouble configuring your system after having disabled SuSEconfig ;-) o MAIL__REPORTS__TO=newbie SuSEconfig can mail reports (created by YaST or included in packages) to you. Here you can set the address. If you don't want it, simply set it to "" . o MOUSE=/dev/ttyS2 Interface where mouse is connected to. YaST (SuSEconfig respectively) creates a link from /dev/mouse to the given device. o MODEM=/dev/ttyS1 Interface where the modem is connected to. YaST (SuSEconfig respec- tively) creates a link from /dev/modem to the given device. o KEYTABLE=de-latin1-nodeadkeys Defines keymaps. o KBD__NUMLOCK=no NumLock on/off. o KBD__CAPSLOCK=no CapsLock on/off. o KBD__RATE=30 Sets the automatic keyboard frequency. Possible values are from twice to 30 times per second. For this change to take effect keyboard delay has to be set as well. o KBD__DELAY=250 This is the delay whereafter automatic repetition of the pressed key com- mences. This value is in milliseconds but isn't very accurate. You have to assign KBD__RATE as well. 174 _____________________________________________________________________________________12.6.__Options_of_configuration________* *______________________ o FONT=mr.fnt This is the console font. Not all fonts support German umlauts. YaST provides a little window where you can test all fonts and choose the one you like best. o GMT=-u If your hardware is set to GMT (Greenwich Mean Time) you should set this variable to -u 6 ; otherwise you leave it "empty". This setting is relevant for automatic changing to summer or winter time respectively. o TIMEZONE=Europe/Berlin Your timezone. Is important for automatic switching to summer or winter time respectively o START__LOOPBACK=yes Sets if sort of an "mini" network is created by configuring Loopback- device. Since rather an amount of programs rely on this functionality it should be set. 7 . o CHECK__ETC__HOSTS=yes SuSEconfig can do some checks and modifications for /etc/hosts. o SETUPDUMMYDEV=yes Setup dummy network device. This is useful for non permanent network connections (e. g. SLIP or PPP). o CREATE__HOSTCONF=yes SuSEconfig can create and check /etc/host.conf. o CREATE__RESOLVCONF=yes SuSEconfig can maintain /etc/resolv.conf. If set to yes and one of SEARCHLIST and NAMESERVER is empty, it is assumed, that no DNS is wanted and /etc/resolv.conf will be deleted. no simply lets /etc/ resolv.conf untouched. o NETCONFIG=__0 Tells how many networking cards (or other net devices) are installed. This is an example for one networking card (they start with 0); for a system with two cards installed it should resemble NETCONFIG="__0 __1" . For a system without network it should not be set. o IPADDR__0=193.141.17.202 IP-address of the first networking card. o NETDEV__0=eth0 Name of the first network device (normally an ethernet card therefore eth0 ). Other possible settings are str1 or plip1 . If there is more than one card installed those additional cards are supplied with the variables NETDEV__1 to NETDEV__3. o IFCONFIG__0="193.141.17.205 broadcast 193.141.17.255 netmask 255.255.255.192" Configuration command for the first networking device installed. Those settings can easily be assigned using YaST. If you have more than one card installed just enter the corresponding values in the appropriate variables. 6 ____________________________________________________ 7 -u is an abbreviation for universal time. Of course your kernel must have be compiled with networking support! * * 175 12. The S.u.S.E. boot_concept_____________________________________________________________________________________________________* *_______________________________________@ o NETWORK__0="-net 193.141.17.0" Network address for your first card. For use with Point-to-Point con- nections (e. g. PLIP) YaST will create an entry which resembles the line "-host 193.141.17.202" . The address of the PPP partner is set. o GATEWAY__0=193.141.17.200 Gateway address of the first networking card. All IP packages which can't be delivered locally are sent to this address. This typically is the system which realizes internet connectivity. o CLOSE__CONNECTIONS=false If true and if the "runlevel" is 0 or 6 (halt or reboot), /sbin/init.d/route sends SIGTERM to the processes. o FQHOSTNAME=earth.cosmos.com Fully qualified hostname of your machine. o SEARCHLIST=cosmos.com This entry is used for completing a not fully qualified hostname. If, e. g. you give in venus it is checked whether venus.cosmos.com is a valid address. This variable has to be set if you plan to use DNS! At least enter your domain name here. You can enter up to three entries which should be separated by blanks. o NAMESERVER=helios.cosmos.com Address of name server which is to be interrogated if a hostname has to be transposed to an IP address, You can enter up to three nameservers which should be separated by blanks. If you plan to use a nameserver SEARCHLIST has to be set! o ORGANIZATION="S.u.S.E. GmbH, Fuerth, Germany" This text appears in every newsposting you send. o NNTPSERVER=helios Address of your news server; if you receive your news via UUCP and they are saved locally you should enter localhost . o IRCSERVER=helios This is the place for your IRC server (Internet Relay Chat). Names of the servers should be separated by blanks. o START__INETD=yes Controls whether the inetd -super daemon should be activated. This dae- mon reacts on calls from other hosts and starts (depending on the port) the appropriate service. You need it if you want to log in via telnet or rlogin . o SENDMAIL__TYPE=uucp-only Sets how email is sent. The example is configured for a host which re- ceives (and sends) email only via UUCP (section 17.4). Possible entries are: uucp-only , smtp and no . The latter is if you don't want to receive or send email at all. o SENDMAIL__SMARTHOST=uucp-dom:Galois Entry for a host which transports all of his mail via UUCP. All mail is send to smarthost; the local system relies on the smarthost to deliver its mail. o SENDMAIL__LOCALHOST=localhost Here you can enter a couple of hosts which are considered local hosts. 176 _____________________________________________________________________________________12.6.__Options_of_configuration__________* *____________________ electronic mailto one of these hosts is delivered locally. You can enter up to three hosts here. o SENDMAIL__RELAY=helios If this variable is set all incoming and outgoing mail is delivered via this host. No email is delivered locally. This might come in handy if certain hosts only occasionally run under Linux. o SMTP=no Sets if a sendmail daemon should be activated. If you receive your email exclusive via UUCP you don't need it, provided you call sendmail -q after each polling; (rmail started by UUCP just puts mail into a queue but doesn't deliver it). If mail-spool-directories are mounted via NFS (e. g. a network) and the single host has got only outgoing mails this could be set to no as well; same applies with use of relay hosts. o FROM__HEADER=cosmos.com This address appears as sender in your electronic mail. If this is not set the fully qualified hostname is used instead which is not desirable in a net using a central mail-server. o START__KERNELD=yes This variable sets whether kerneldaemon should be started automatically at boot time. This daemon is responsible for automatic loading kernel modules on demand. A short description of the module concept and func- tions of kerneld are to be found in chapter section 13.2. o START__PORTMAP=no Assigns whether to start portmapper or not. You need portmapper if you plan to use your host as NFS-server (see section 16.6); without this dae- mon rpc.mountd and rpc.nfsd can't run! It also is necessary for NIS (see section 16.6). o NFS__SERVER=no If the host is going to be used as NFS-server this variable has to be set to yes . This initializes the start of rpc.nfsd and rpc.mountd . A forthgoing description of an NFSserver (e. g. definition of directories to be exported) is described in chapter section 16.6. o REEXPORT__NFS=no yes , if imported NFS should be re-exported. o START__ROUTED=no Whether to start the route daemon. o START__NAMED=no Whether to start the name daemon. o CREATE__YP__CONF=yes Sets up if SuSEconfig should create the necessary files for YP; this de- pends on the next two entries as well (look at section 16.5). Furthermore /etc/passwd and /etc/group will be adjusted as well. o YP__DOMAINNAME=cosmos.com YP-domainname of hosts. For detailed information please refer to chapter section 16.5. o YP__SERVER=helios.cosmos.com Name of the NIS server. * * 177 12. The S.u.S.E. boot_concept_____________________________________________________________________________________________________* *_______________________________________@ o START__GPM=yes Sets whether mouse console support is to be started; this enables you to exchange text between consoles using the mouse. Gpm can cause problems in connection with certain bus mice; if you encounter problems while starting X you should disable gpm . The other alternative is to start xdm since gpm is not started in runlevel 3. o GPM__PARAM=" -t logi -m /dev/mouse" Initialization parameters for gpm ; those are normally set via YaST. o RUN__UPDATEDB=yes Decides whether database for locate should be updated once a night; locate serves for quick finding of files. o CREATE__INFO__DIR=yes Decides whether to create /usr/info/dir which serves as a general index of all indexpages. This is useful after installing a package which contains infopages. Please keep in mind that perl needs to be installed for this to work. o CHECK__PERMISSIONS=set : : : o PERMISSION__SECURITY="easy local" : : : o RPMDB__BACKUP__DIR=/var/adm/backup/rpmdb : : : o MAX__RPMDB__BACKUPS=5 Maximum number of the backup files of the RPM database. o DELETE__OLD__CORE=yes Corefiles are memory images of programs which have been killed due to a segmentation fault; those images are very useful for debugging. If this is enabled a regular search and deletion of old corefiles is launched. o MAX__DAYS__FOR__CORE=7 Assigns how old corefiles should be before they are deleted. o MAX__DAYS__FOR__LOG__FILES=365 If a log file (mainly under /var/log) reaches a distinct size it will be au- tomatically compressed and archived; `root' is informed via electronic mail. This parameter sets how long those files are to be kept on disk before being deleted automatically. If you set this value to 0 no compressing and no archiving will be done; files will be written eternally and can reach a remarkable size! Compressed logfiles can be viewed anytime using zless . o MAX__DAYS__IN__TMP=30 /tmp-directories are checked daily whether they have been touched during the last days (this variable is set in days). If this is true they will be deleted automatically. o TMP__DIRS__TO__CLEAR="/tmp /var/tmp" Enter all directories which are to be searched for old files (as given above) here. o OWNER__TO__KEEP__IN__TMP="root bs" Files of system users given here should not be deleted even if they have not been touched within the given time. 178 _____________________________________________________________________________________12.6.__Options_of_configuration__________* *____________________ o START__LPD=yes : : : o START__NNTPD=yes : : : o START__ATD=yes : : : o START__HTTPD=yes : : : o START__SQUID=no : : : o START__MARSNWE=no : : : o SUSEWM__UPDATE =yes : : : o CRON=yes Sets the start and stop of cron daemon. This daemon lets you start certain programs at a given time. This daemon is only started in runlevel 2 or 3 respectively! It is highly recommended to activate this daemon mainly if your computer runs all time. _______________________________________________________________________________________________________________________|| |||||||| There are a lot of options which require you to regularly run certain com- ||| | |||||||| mands and programs therefore crondaemon should be activated on every || | |||||||| || | |||||||| system || | ||||||||_______________________________________________________________________________________________________________||_ * * 179 12. The S.u.S.E. boot_concept_____________________________________________________________________________________________________* *_______________________________________@ 180 Chapter 13 The kernel The kernel which is written to disk after installation (and which is found in a correctly installed system under /) is configured to support a wide range of hardware. Therefore this certainly is not the best choice since it's rather tedious to wait until all (unnecessary) drivers are installed. Moreover some kB are wasted since lots of unnecessary drivers are loaded. This leads to the great advantage of building an kernel of your own. Further- more creating an own kernel in some cases provides access to exotic hardware which is not supported by our standard-kernels (e. g. rather exotic bus-mice or sound devices). What is more, kernel configuring provides a view into the inside of the actual status quo of kernel development. For compiling a kernel there already are a couple of Makefiles for the C compiler. These Makefiles almost automatically generate everything for you. The only thing you have to do by hand is to give in the list of your hardware. 13.1 Kernel sources For building a kernel obviously the kernel sources and the C compiler have to be installed. Both of them are located in series D (series D (Development)). Generally it is highly recommended to install the C compiler since C lan- guage is inseparable from Unix operating systems. Kernel sources are to be found under /usr/src/linux. If you plan to exper- iment with different kernel sources it is obvious that you can unpack them in different directories and access the actual sources via a symbolic link. That's what YaST does automatically. For clarity: Kernel sources can be put and compiled (theoretically) in any directory. Since there is a lot of software that relies on the sources being under /usr/src/linux you should keep this directory providing an error free compilation of system programs. 13.2 Kernel modules Many drivers and features do not have to be compiled directly into the kernel anymore but can be loaded at runtime via kernel modules. Which drivers are to be compiled into the kernel and which are loaded as runtime modules is defined in kernel configuration. * * 181 13. The kernel _________________________________________________________________________________________________________________* *_______________________________________@ Kernel modules are located at /lib/modules/ where is the actual kernel version. 13.2.1 Handling of modules The following commands are at your disposition: o insmod insmod loads the given module. The module is searched in a subdirectory of /lib/modules/ . o rmmod Unloads the corresponding module. This obviously is only pos- sible if this module is not needed any more. It is not possible to unload isofs module while a CD is still mounted. o depmod This command creates in /lib/modules/ the file modules.dep where dependencies of all of the modules are defined. This is necessary to ensure that all dependent modules are loaded along with the selected ones. If kernel daemon is set in /etc/rc.config this file is created each time the system is started. o modprobe Loading or unloading of a given module while considering dependencies of this module. This command is extremely powerful and can be used for a lot of things (e. g. probing of all modules of a given type until one is successfully loaded). Contrary to insmod modprobe checks /etc/ conf.modules and should be used for loading modules. For detailed information on this topic please refer to the corresponding manual page. o lsmod Shows you which modules are currently loaded and by how many other modules they are being used. Modules started by kernel-daemon are marked by (autoclean) which shows you that those modules will be removed automatically if they reach their idle-time. Loading of modules is further influenced by /etc/conf.modules. Here you can add parameters to those modules that directly access your hardware (e. g. CD-ROM drivers..). Those parameters given here are mainly identical to those given at the LILO-prompt (see chapter 7), but they differ in quite some cases. If loading of a module failed you can try to specify your hardware in this file. 13.2.2 The kernel daemon The most elegant way to use modules is the kernel-daemon anyway. This process runs in the background and takes care of necessary modules to be loaded as soon as this is requested by the kernel. Moreover, unused modules are unloaded automatically after a certain idle time (default: 1 minute). For using kernel-daemon you have to set the corresponding variable in /etc/rc.config (see section 12.6). ________________________________________________________________________________________________________* *________|| | Those drivers that are needed for accessing the root-filesystem have to be * * |||| | compiled directly into the kernel! So you should not load your SCSI-driver * * ||| | * * ||| | or your filesystem (normally: ext2 ) via modules! * * ||| |_______________________________________________________________________________________________________* *________|||_ 182 ___________________________________________________________________________________________13.3.__Kernel_configuration______* *______________________ 13.3 Kernel configuration Configuring your kernel can be done in three different ways: 1. On command line 2. In a menu in textmode 3. In a menu under the X Window System Here we give a short overview on these three possibilities. 13.3.1 Configuring on the command line For configuring your kernel just change to /usr/src/linux and enter: root@earth:/usr/src/linux # make config Now you are asked a couple of options which should be supported by your freshly to be created kernel. Here there are two or three possible answers. Either y or n or y n or m . `m' means that this device is not compiled directly into the kernel but as a module instead. It is obvious that any driver that is needed to bring up the system properly should not be loaded as a module. If you press any other key you get a short help text on this subject. 13.3.2 Configuring in text mode There is a much more comfortable way by typing root@earth:/usr/src/linux # make menuconfig Moreover if you see that you have mistyped something by accident you do not have to fiddle through all the questions. 13.3.3 Configuring under the X Window System If you have installed the X-Window System (package xf86) Tcl/Tk (package tcl and package tk) you can use root@earth:/usr/src/linux # make xconfig as an alternative. You will be presented an GUI (Graphical User Interface) which makes kernel configuration very comfortable. You should have started the X-Window System as `root' or you have to take additional steps into consideration (e. g. gathering the display from another user). All options are roughly previewed in the following chapter. Please keep newest in mind that several options might differ from version to version. The documentation brand-new documentation is always to be found under /usr/src/linux/ Documentation. 13.4 Settings in kernel configuration 13.4.1 SMP In general a kernel for a one-processor machine is built. Compiling an SMP version is not included in standard configuration, so, if you plan to compile an SMP version you have to edit the top-level Makefile by hand. Just remove the `#' from line 14 (SMP now has the value 1 ) and you are done. * * 183 13. The kernel _________________________________________________________________________________________________________________* *_______________________________________@ 13.4.2 Experimental drivers + Experimental drivers Config Question: Prompt for development code/drivers Config Variable: CONFIG__EXPERIMENTAL Allows you to choose from brand new drivers which are not fully tested yet. If you answer `No', a couple of drivers will not be present. 13.4.3 Modul support + Support of loadable modules Config Question: Enable loadable module support Config Variable: CONFIG__MODULES Recommended Enables use of loadable modules. Those are drivers that are not compiled directly into the kernel but are loaded via insmod at runtime. If you enable this option there will be an additional select option `m' (module); otherwise you will only be presented `y' and `n'. More information is located under /usr/src/linux/Documentation/modules.txt + Version information on all symbols Config Question: Version information on all symbols Config Variable: CONFIG__MODVERSIONS Your modules are more or less independent from the kernel version if you enable this feature (normally any modules have to be recompiled if you install a newer kernel). For novices it is not recommended to use this feature. Kernel and modules have to be compiled the same! If you install an MODVERSIONS kernel you have to install the modules exactly the same! + Kernel daemon support Config Question: Kernel daemon support Config Variable: CONFIG__KERNELD Recommended Enables automatic loading of modules at runtime. It is recommended to only use this feature for drivers that are not urgently needed (e. g. addi- tional filesystems or similar). 13.4.4 General configuration + Kernel math emulation Config Question: Kernel math emulation Config Variable: CONFIG__MATH__EMULATION Only set this option if you plan to install this kernels on machine without FPU (e. g. 386, 486SX). ________________________________________________________________________________________________________* *________ | If you omit this option the kernel will not run on a machine without FPU! * * ||| | * * ||| | An emulation on a machine with a FPU does not hurt but uses up memory. * * ||| |_______________________________________________________________________________________________________* *________|||_ 184 __________________________________________________________________________13.4.__Settings_in_kernel_configuration____________* *_____________________ + Networking support Config Question: Networking support Config Variable: CONFIG__NET This is obligatory if you plan to install a network (normal TCP/IP or Recommended! loopback ). Disabling the network support should only be done if there is little memory available. Drivers for certain networking devices are specified later on. + Limit memory to low 16 MB Config Question: Limit memory to low 16 MB Config Variable: CONFIG__MAX__16M Activating this option makes only use of the first 16 MB of RAM. This might be necessary on machines that can not handle caching of memory beyond 16 MB. In most cases this 'loss' of memory leads to a higher performance than loss of cache. + PCI bios support Config Question: PCI bios support Config Variable: CONFIG__PCI Enables 32 bit calls on PCI boards. Some older boards have bugs in their BIOS and can cause trouble. Consider PCI-HOWTO for more help. There should not be any problem on newer machines. + PCI bridge optimisation (experimental) Config Question: PCI bridge optimisation (experimental) Config Variable: CONFIG__PCI__OPTIMIZE If you have a PCI motherboard enabling this option can enhance access to some devices. + System V Inter Process Communication Config Question: System V IPC Config Variable: CONFIG__SYSVIPC Inter Process Communication are services such as messages, shared Recommended! memory and semaphores which are provided by the kernel and which are used by many programs (e. g. the DOS-emulator). + Kernel support for a.out binaries Config Question: Kernel support for a.out binaries Config Variable: CONFIG__BINFMT__AOUT a.out is the old binary format under Linux. If you still have some of those old programs built with this binary format you have to activate this option. Since there still are (mainly commercial products) that use this format it is recommended to load it as a module. + Kernel support for ELF binaries Config Question: Kernel support for ELF binaries * * 185 13. The kernel _________________________________________________________________________________________________________________* *_______________________________________@ Config Variable: CONFIG__BINFMT__ELF Necessary! ELF is Linux' current binary format for libraries and programs. Since almost any software has been ported to ELF you have to activate this option! ________________________________________________________________________________________________________* *________ | If you do not select this file or select it as a module you will not be able to * * ||| | * * ||| | boot your system since all relevant files use ELF! * * ||| |_______________________________________________________________________________________________________* *________|||_ + Kernel support for JAVA binaries Config Question: Kernel support for JAVA binaries Config Variable: CONFIG__BINFMT__JAVA Enables direct starting of JAVA applets by the kernel itself. For bene- fit of this option the Java Developers Kit (JDK) has to be installed on your system. For gathering more information please refer to /usr/src/ linux/Documentation/java.txt. Loaded as a module it will not hurt anyway. + Support of different binary formats Config Question: Kernel support for MISC binaries Config Variable: CONFIG__BINFMT__MISC Enables directly launching different binary formats by the kernel. It is mainly used for DOSEMU but may also replace the JAVA-loader. Please be aware that this is an experimental feature that is not included in the official kernel version. For using it you have to take the kernel extended by S.u.S.E.. A complete description of this feature can be found under /usr/src/linux/Documentation/binfmt_misc.txt. + Compile kernel as ELF Config Question: Compile kernel as ELF Config Variable: CONFIG__KERNEL__ELF Assigns the kernel to use the ELF format. You should answer this ques- tion in the affirmative. + Processor type Config Question: Processor type Config Variable: 386, 486, Pentium, PPro Enables you to optimize your code according to the processor you use. If the kernel should run on different processors you have to optimize it for 386. If you set your processor to Pentium really aggressive optimization is done (a real fast memcopy routine is used which really exploits the pro- cessor). There are hints that some processors can cause problems with these routines ( Cyrix , Intel 100 MHz B-Step). If your system does not run reliably you should not optimize your kernel for Pentium for exclud- ing this cause. 186 __________________________________________________________________________13.4.__Settings_in_kernel_configuration___________* *______________________ _______________________________________________________________________________________________________________________ |||||||| A 486 or Pentium kernel will not run on an 386! || | ||||||||_______________________________________________________________________________________________________________||_ 13.4.5 Disks, (E)IDE and other block devices + Normal floppy disk support Config Question: Normal floppy disk support Config Variable: CONFIG__BLK__DEV__FD Necessary for using floppy drives under Linux. Disabling this option for security reasons might be useful. + Enhanced IDE/MFM/RLL support Config Question: Enhanced IDE/MFM/RLL support Config Variable: CONFIG__BLK__DEV__IDE Is needed if you have non-SCSI devices installed (harddisk, ATAPI CD- ROM etc.). If your machine does not contain an EIDE disk, or is driven via network, or without any hard disks you can switch off this option. If you enable this option you may define configuration for these drivers later on. Some drivers themselves require additional arguments for fully exploiting the drivers capabilities (section 13.4.5, page 188). + Use old disk-only driver for primary i/f Config Question: Use old disk-only driver for primary i/f Config Variable: CONFIG__BLK__DEV__HD Only IDE hard disk support without CD-ROM support. In this special case no enhanced IDE disks are supported. Normally this option is obso- lete. + Include IDE/ATAPI CDROM support Config Question: Include IDE/ATAPI CDROM support Config Variable: CONFIG__BLK__DEV__IDECD Necessary for using CD-ROM devices connected to an EIDE controller (ATAPI CD-ROMs) + Support for ATAPI streamer Config Question: Include IDE/ATAPI TAPE support Config Variable: CONFIG__BLK__DEV__IDETAPE Support for ATAPI streamers. + Partial support for PCMCIA Config Question: Support removeable IDE interfaces Config Variable: CONFIG__BLK__DEV__IDE__PCMCIA Support for devices connected to a PCMCIA port. PCMCIA is not part of the official kernel but has to be compiled separately. Is is required to install the sources of the PCMCIA subsystem (package pcmcia__s). Please have a look at chapter 6. * * 187 13. The kernel _________________________________________________________________________________________________________________* *_______________________________________@ + Support for CMD640 EIDE chipsets Config Question: CMD640 chipset bugfix/support Config Variable: CONFIG__BLK__DEV__CMD640 This chipset used on many EIDE controllers is buggy. If you call one your own you have to activate this option for an error free system. + Optimization for CMD640 EIDE Chipset Config Question: CMD640 enhanced support Config Variable: CONFIG__BLK__DEV__CMD640__ENHANCED Enables auto tuning of some PIO-modes of this special chipset. + Support for RZ1000 EIDE chipset Config Question: RZ1000 chipset bugfix/support Config Variable: CONFIG__BLK__DEV__RZ1000 This chipset used on many EIDE controllers is buggy. If you call one your own you have to activate this option for an error free system. + Support for Mainboard chipset Intel 82371 PIIX Config Question: Intel 82371 PIIX (Triton I/II) chipset Config Variable: CONFIG__BLK__DEV__TRITON Enables DMA capabilities if you have one installed. This reduces CPU accesses on your system. + Other IDE chipset support Config Question: Other IDE chipset support Config Variable: CONFIG__IDE__CHIPSETS Support for lots of other chipsets can be set here. Some or them might not even let you handle your secondary controller without this option (most of them have a CD-ROM connected to this controller). Most of these drivers have to be activated by a command line parameter as well (section 7.3.2)! + ALI M14xx support Config Question: ALI M14xx support Config Variable: CONFIG__BLK__DEV__ALI14XX This driver is activated by the kernel parameter ide0=ali14xx . This enables support for the secondary EIDE controller as well as it in- creases I/O speed. + DTC-2278 support Config Question: DTC-2278 support Config Variable: CONFIG__BLK__DEV__DTC2278 This driver is activated by the kernel parameter ide0=ht6560b This enables support for the secondary EIDE controller as well as it in- creases I/O speed. This driver is famous for creating timeouts and retries at an high I/O load. CD-ROMs and tapes are not supported yet. 188 __________________________________________________________________________13.4.__Settings_in_kernel_configuration____________* *_____________________ + QDI QD6580 support Config Question: QDI QD6580 support Config Variable: CONFIG__BLK__DEV__QD6580 This driver is activated by the kernel parameter ide0=qd6580 . This enhances I/O speed. + UMC 8672 support Config Question: UMC 8672 support Config Variable: CONFIG__BLK__DEV__UMC8672 This driver is activated by the kernel parameter ide0=umc8672 . This enhances I/O speed and supports the secondary EIDE controller. + Loopback device support Config Question: Loopback device support Config Variable: CONFIG__BLK__DEV__LOOP Do not confuse this wit networks loopback mode. Loopback mode allows you to treat a file the same way as a filesystem (e. g. you can mount it). Selfmade CD images or disk images can be tested that way. Moreover encryption is possible using this feature. + Multiple devices driver support Config Question: Multiple devices driver support Config Variable: CONFIG__BLK__DEV__MD Enables you to treat many physical devices as one logical device. This enables software linear and RAID-0. Only for experts. + Linear (append) mode Config Question: Linear (append) mode Config Variable: CONFIG__MD__LINEAR Here partitions are bundled together logically by the multiple device driver thus leading to a large mass storage media. + Software Raid-0 (Stripping) Config Question: RAID-0 (stripping) mode Config Variable: CONFIG__MD__STRIPPED This enables software RAID-0. Enhances performance if the bundled partitions reside on different disks. + Providing RAMDISK devices Config Question: RAM disk support Config Variable: CONFIG__BLK__DEV__RAM Allows use of a RAM-disk which can be accessed via /dev/ram. It is not necessary to assign the size in advance (e. g. at booting). Memory is not allocated and released dynamically. If you plan to use a RAM disk please have a look at /usr/src/linux/Documentation/ramdisk.txt. * * 189 13. The kernel _________________________________________________________________________________________________________________* *_______________________________________@ + Initial RAM disk (initrd) support Config Question: Initial RAM disk (initrd) support Config Variable: CONFIG__BLK__DEV__INITRD A special kind of RAM-disk which is dealt with by the loader. This feature is not necessary for every days work. It is used (e. g. ) in our installation system for loading the root image from CD. + Support for old XT hard disks Config Question: XT harddisk support Config Variable: CONFIG__BLK__DEV__XD This driver only is needed if you plan to use an XT-controller (8 bit). Normally this is not the case. 13.4.6 Network options + Network firewalls Config Question: Network firewalls Config Variable: CONFIG__FIREWALL Enables firewalling in kernel for letting you run your Linux machine as a firewall. A firewall serves for protecting your machine against illegal attacks from the Internet or another network by supervising the network traffic. If you plan to use your machine as a firewall you should enable CONFIG__IP__FORWARD too. + Network aliasing Config Question: Network aliasing Config Variable: CONFIG__NET__ALIAS Supports multiple IP addresses on one physical network device. This nor- mally is for services that behave differently on different addresses (e. g. the Apache server). + Support of TCP/IP protocol Config Question: TCP/IP networking Config Variable: CONFIG__INET Very much TCP/IP are the most important network protocols in Unix/Internet. Since recommended! many services use these protocols you should enable this feature even if there is no networking device installed in your machine. At least a so called loopback device is provided which enables network access to your own machine. Without this option printing is impossible! + IP forwarding/gatewaying Config Question: IP forwarding/gatewaying Config Variable: CONFIG__IP__FORWARD This option is only needed if the machine should be installed as a router in a local network, or should be connected to the Internet or should serve as a SLIP server. 190 __________________________________________________________________________13.4.__Settings_in_kernel_configuration____________* *_____________________ + IP multicasting Config Question: IP multicasting Config Variable: CONFIG__IP__MULTICAST Enables simultaneously addressing multiple networked machines (Fan out). gated needs this to be activated. + IP accounting Config Question: IP accounting Config Variable: CONFIG__IP__ACCT Serves for logging of IP packages and is not needed normally. Some programs on the other hand (e. g. xosview ) can not show network traffic without this option being set. + PC/TCP compatibility mode Config Question: PC/TCP compatibility mode Config Variable: CONFIG__INET__PCTCP If you encounter problems while addressing Linux via PC/TCP software you should answer `y'. + Reverse ARP Config Question: Reverse ARP Config Variable: CONFIG__INET__RARP Machines that must boot without network do know their ethernet hard- ware address at booting but not their IP address. For gathering their ad- dress they send a Reverse Address Resolution Protocol over the net. If your Linux machine should answer such requests you have to enable this option. + Disable Path MTU Discovery Config Question: Disable Path MTU Discovery Config Variable: CONFIG__NO__PATH__MTU__DISCOVERY Normally the MTU (maximum transfer unit) is assigned automatically. This starts with huge packages going down to smaller ones if necessary. If you encounter problems using this feature you should disable it. + Drop source routed frames Config Question: Drop source routed frames Config Variable: CONFIG__IP__NOSR Packages whose routing is already assigned by the sender should be ig- nored. For security reasons you should answer yes unless you are pretty sure of what you are doing. + Allow large windows Config Question: Allow large windows Config Variable: CONFIG__SKB__LARGE * * 191 13. The kernel _________________________________________________________________________________________________________________* *_______________________________________@ Enables a higher network performance by use of big sized buffers taken from RAM. If you have less than 16 MB installed you should omit this option. + The IPX protocol Config Question: The IPX protocol Config Variable: CONFIG__IPX Support of the IPX protocol of Novell. It enables you to access an Novell Netware server from either within ncpfs or from the DOS-em- ulator. + Appletalk DDP Config Question: Appletalk DDP Config Variable: CONFIG__ATALK Enables integration of a Linux machine into an Appletalk network. + Amateur Radio AX.25 Level 2 Config Question: Amateur Radio AX.25 Level 2 Config Variable: CONFIG__AX25 Enables data transfer via CB. You will need additional software which can be found on our CD under /unsorted/Amateur-Funk (see also /usr/src/linux/Documentation/networking/ax25.txt). + Bridging Config Question: Bridging Config Variable: CONFIG__BRIDGE Enables you to use your Linux machine as an ethernet bridge. This means that the segments the machine is connected to seem like one big network to the other users. Bridging code is still under test and should only be used by experienced users. + Kernel/User network link driver Config Question: Kernel/User network link driver Config Variable: ONFIG__NETLINK This driver enables communication between the kernel and user processes in both directions. 13.4.7 SCSI + SCSI Support Config Question: SCSI support Config Variable: CONFIG__SCSI If you have SCSI devices installed activate this option. + Scsi disk support Config Question: Scsi disk support 192 __________________________________________________________________________13.4.__Settings_in_kernel_configuration___________* *______________________ Config Variable: CONFIG__BLK__DEV__SD Activates the SCSI hard disk driver. Do not select this option as a module if you have to boot from a SCSI device! Please be aware that some SCSI controllers can cause problems when loaded as modules! + Support of SCSI tapes Config Question: Scsi tape support Config Variable: CONFIG__CHR__DEV__ST Activates SCSI streamer driver. The tape can be addressed via the device /dev/st0. + Scsi CDROM support Config Question: Scsi CDROM support Config Variable: CONFIG__BLK__DEV__SR Activates SCSI-CD-ROM driver. + Scsi generic support Config Question: Scsi generic support Config Variable: CONFIG__CHR__DEV__SG Enables sending generic SCSI commands to the appropriate devices. This enables you to use special SCSI devices as scanners, synthesizers or SCSI CD-ROMs, provided you have the proper software. A direct kernel sup- port for such devices is not implemented yet. + Probe all LUNs on each SCSI device Config Question: Probe all LUNs on each SCSI device Config Variable: CONFIG__SCSI__MULTI__LUN Some SCSI devices support LUNs (logical unit number) like some CD juke-boxes . Activate this kernel support if you plan to use one of these. + Verbose SCSI error reporting Config Question: Verbose SCSI error reporting Config Variable: CONFIG__SCSI__CONSTANTS Extensive SCSI error reports instead of cryptical hex-codes need app. 12 KB of kernel size. Only necessary if you have problems and want to research. In this section you can select drivers for different SCSI hostadapters. 13.4.8 Networking cards + Network device support? Config Question: Network device support? Config Variable: CONFIG__NETDEVICES If there is a networking card installed or you plan to use SLIP (Serial Line Internet Protocol) you should answer yes. * * 193 13. The kernel _________________________________________________________________________________________________________________* *_______________________________________@ + Dummy net driver support Config Question: Dummy net driver support Config Variable: CONFIG__DUMMY The dummy driver comes in handy for simulating network connections as well as for configuration and use of SLIP and PPP. + EQL (serial line load balancing) Config Question: EQL (serial line load balancing) Config Variable: CONFIG__EQUALIZER If you call two serial lines your own (e. g. two modems) and use SLIP or PPP you can double your speed by separating the stream on both lines simultaneously. Please also have a look at the README file in /usr/ src/linux/drivers/net/README.eql. + FRAD (Frame Relay Access Device) support Config Question: FRAD (Frame Relay Access Device) support Config Variable: CONFIG__DLCI Frame Relay is a fast and cheap way for connecting to one's Internet provider or for maintaining a private wide area network. Via a physical connection multiple logical connections are transmitted. You need special hardware to run Frame Relay. It is recommended to look at /usr/src/ linux/Documentation/framerelay.txt. + Max open DLCI Config Question: Max open DLCI Config Variable: CONFIG__DLCI__COUNT Maximum number of point-to-point connections the driver can maintain. + Max DLCI per device Config Question: Max DLCI per device Config Variable: CONFIG__DLCI__MAX Maximum number of point-to-point connections per device. + Sangoma S502A FRAD support Config Question: Sangoma S502A FRAD support Config Variable: CONFIG__SDLA Support for Sangoma S502A, S502E and S508 Frame Relay Access De- vices. + PLIP (parallel port) support Config Question: PLIP (parallel port) support Config Variable: CONFIG__PLIP PLIP (Parallel Line Internet Protocol) enables use of the Internet protocol via the parallel interface. A special cable is needed for this. /usr/src/ linux/drivers/net/plip.c describes the pins of this cable. Please keep in mind that you are not able to connect a printer if the kernel is compiled using this option! In this case you should not activate CONFIG__PRINTER (section 13.4.12, page 202). 194 __________________________________________________________________________13.4.__Settings_in_kernel_configuration___________* *______________________ + PPP (point-to-point) support Config Question: PPP (point-to-point) support Config Variable: CONFIG__PPP PPP is a more modern protocol than SLIP but in fact contains the same features. Please have a look at PPP-HOWTO and the documentation un- der /usr/src/linux/Documentation/networking/ppp.txt. If you have set CONFIG__MODVERSIONS you cannot add the PPP device directly into the kernel but have to compile it as a module. + SLIP (serial line) support Config Question: SLIP (serial line) support Config Variable: CONFIG__SLIP SLIP (Serial Line Internet Protocol) enables the use of the internet proto- col via a serial line (e. g. a modem). Also if you plan to use the machine as a SLIP server you need to activate this option. + CSLIP compressed headers Config Question: CSLIP compressed headers Config Variable: CONFIG__SLIP__COMPRESSED Use of compressed headers enhances SLIP's performance but is only pos- sible if used and configured on both sides. + Keepalive and linefill Config Question: Keepalive and linefill Config Variable: CONFIG__SLIP__SMART Enhancements mainly for using SLIP on interrupted analog lines. Now you can select drivers for different networking cards. The kernel even contains Ethernet pocket adapter support. Please consult Ethernet-HOWTO. 13.4.9 ISDN subsystem + ISDN support Config Question: ISDN support Config Variable: CONFIG__ISDN ISDN4linux (package i4l) has become integrated into the standard ker- nel in the meantime. We recommend to use the modularized version. + Support synchronous PPP Config Question: Support synchronous PPP Config Variable: CONFIG__ISDN__PPP Enables synchronous PPP via ISDN used e. g. by Cisco and Sun. A com- plete documentation can be found under /usr/src/linux/Documentation/isdn/README.syncppp. + Support audio via ISDN * * 195 13. The kernel _________________________________________________________________________________________________________________* *_______________________________________@ Config Question: Support audio via ISDN Config Variable: CONFIG__ISDN__AUDIO Enables you to make telephone calls over the internet. The only low- level driver that supports this feature is Teles. Moreover this function is necessary if you want to use the answering machine vbox . + ICN 2B and 4B support Config Question: ICN 2B and 4B support Config Variable: CONFIG__ISDN__DRV__ICN Support for ICN's active ISDN cards. Documentation concerning this topic may be found at /usr/src/linux/Documentation/isdn/ in README and README.icn. + PCBIT-D support Config Question: PCBIT-D support Config Variable: CONFIG__ISDN__DRV__PCBIT Support of Octal's ISDN cards. For using this card you need additional firmware which has to be loaded into your card. More information can be found under /usr/src/linux/Documentation/isdn/ in README and README.pcbit. + (old)Teles/NICCY1016PC/Creatix support Config Question: Teles/NICCY1016PC/Creatix support Config Variable: CONFIG__ISDN__DRV__TELES Support of Teles S0-16.0 , Teles S0-16.3 , Teles S0-8 , and compatible ISDN cards. Please look at /usr/src/linux/Documentation/isdn/ README. You should use the newer HiSax anyway. 13.4.10 Proprietary CD-ROM drives + Aztech/Orchid/Okano/Wearnes CDROM Config Question: Aztech/Orchid/Okano/Wearnes CDROM Config Variable: CONFIG__AZTCD Drivers for these devices with an proprietary CD-ROM controller. AT- API drives are specified via CONFIG__BLK__DEV__IDECD (section 13.4.5, page 187). + Goldstar R420 CDROM support Config Question: Goldstar R420 CDROM support Config Variable: CONFIG__GSCD Driver for Goldstar CD-ROM drive R 420. + Matsushita/Panasonic CDROM driver support Config Question: Matsushita/Panasonic CDROM driver support Config Variable: CONFIG__SBPCD 196 __________________________________________________________________________13.4.__Settings_in_kernel_configuration____________* *_____________________ Driver for all CD-ROM drives (even Creative Labs, Kotobuki,...) that are connected to a SoundBlaster card. This driver can drive up to 4 de- vices. Please read CDROM-HOWTO. + MicroSolutions backpack CDROM support Config Question: MicroSolutions backpack CDROM support Config Variable: CONFIG__BPCD This is a beta driver for CD-ROMs connected to a parallel port. Since this driver is not implemented in the standard kernel this option only is present if you have installed an appropriately patched kernel. If you use this driver your parallel port cannot be used for anything other that is why it is recommended to load this device as a module. + Support of proprietary Mitusmi CD-ROMs Config Question: Mitsumi (not IDE/ATAPI) CDROM driver Config Variable: CONFIG__MCD Driver for proprietary Mitsumi drives with their own controller. This does not apply to ATAPI CD-ROMs e. g. an FX-400 . + Mitsumi [XA/MultiSession] CDROM support Config Question: Mitsumi [XA/MultiSession] CDROM support Config Variable: CONFIG__MCDX Driver for proprietary Mitsumi drives with their own controller which additionally supports XA and multisession (Photo CDs). + Optics Storage DOLPHIN 8000AT CDROM Config Question: Optics Storage DOLPHIN 8000AT CDROM Config Variable: CONFIG__OPTCD Driver for the Optics Storage CD-ROM drive DOLPHIN 8000AT . + Philips/LMS CM206 CDROM support Config Question: Philips/LMS CM206 CDROM support Config Variable: CONFIG__CM206 Driver for Philips CM206 . + Sanyo CDR-H94A CDROM support Config Question: Sanyo CDR-H94A CDROM support Config Variable: CONFIG__SJCD Driver for Sanyo CDR-H94A . + Soft configurable cdrom interface card Config Question: Soft configurable cdrom interface card Config Variable: CONFIG__CDI__INIT Supports software configured drives which have to be initialised at boot time. At the moment there is support for ISP16/MAD16/Mozart . * * 197 13. The kernel _________________________________________________________________________________________________________________* *_______________________________________@ + Sony CDU31A/CDU33A CDROM driver support Config Question: Sony CDU31A/CDU33A CDROM driver support Config Variable: CONFIG__CDU31A Driver for Sony CDU31A or. CDU33A. Please be aware that the driver does no longer autoprobe. Now you have to give the necessary parameters at the boot prompt. Therefore you should add these values to your LILO configuration file in append line (chapter 7). + Sony CDU535 CDROM driver support Config Question: Sony CDU535 CDROM driver support Config Variable: CONFIG__CDU535 Driver for Sony CDU535. 13.4.11 File systems + Quota support Config Question: Quota support Config Variable: CONFIG__QUOTA Enables you to assign disk space to the several users on your system (diskquotas). This only is useful for several users and only works with ext2 file system. + Mandatory lock support Config Question: Mandatory lock support Config Variable: CONFIG__LOCK__MANDATORY Mandatory locking is used by some System V database applications. For using this feature you will need newer NFS daemons as well as new SAMBA,new Nettalk, new Mars-nwe and more. At the moment none of them is available Therefore you should only activate this option if you really need it and if you are absolutely sure of what you are doing. + Standard (minix) fs support Config Question: Standard (minix) fs support Config Variable: CONFIG__MINIX__FS Minix was the first file system available for Linux. Due to its restrictions (maximum size of a file system is 64 MB, filenames only 14 characters long filenames) it is more or less only used for installation disks. + Extended fs support Config Question: Extended fs support Config Variable: CONFIG__EXT__FS Extended file system was the first Linux-specific file system which got rid of Minix' restrictions. It is not used any more and has only nostalgic value. + Second extended fs support 198 __________________________________________________________________________13.4.__Settings_in_kernel_configuration____________* *_____________________ Config Question: Second extended fs support Config Variable: CONFIG__EXT2__FS The Second Extended Filesystem in the meantime has become Linux' standard file system. _______________________________________________________________________________________________________________________ |||||||| This file system has to be compiled into the kernel since it is root's standard || | |||||||| || | |||||||| file system! Use as loadable module therefore is not possible! || | ||||||||_______________________________________________________________________________________________________________||_ + xiafs filesystem support Config Question: xiafs filesystem support Config Variable: CONFIG__XIA__FS Xia file system was the "opponent" of the Second Extended File System in Linux' early days. Nowadays it is not important anymore. + DOS FAT fs support Config Question: DOS FAT fs support Config Variable: CONFIG__FAT__FS Basic FAT support for FAT file systems such as MS-DOS and VFAT. + MSDOS fs support Config Question: MSDOS fs support Config Variable: CONFIG__MSDOS__FS If you enable this option you will be able to access your MS-DOS par- titions from within Linux. Many installation packages for Linux come with disks in DOS format. If you want to use them you need this option enabled. + VFAT (Windows-95) fs support Config Question: VFAT (Windows-95) fs support Config Variable: CONFIG__VFAT__FS Enables accessing Windows 95 partitions with long filenames. This file system can not be used as a root partition. Take the UMSDOS filesystem instead. You can mount a Windows 95 partition either as DOS partition or as VFAT partition; you will only get the long filenames with the latter. + umsdos: Unix like fs on top of FAT Config Question: umsdos: Unix like fs on top of FAT Config Variable: CONFIG__UMSDOS__FS UMSDOS is an add-on for the MS-DOS file system which enables you to Demo mode create files on normal DOS partitions with all properties and permissions of Unix files. These information is kept in one special file which will be created in any directory of the used partition and which is kept up to date by the UMSDOS file system. Furthermore this file system enables you to install a Linux demo-mode on an existing DOS partition. Of course, accessing this file system is much slower than the Second Extended File System. * * 199 13. The kernel _________________________________________________________________________________________________________________* *_______________________________________@ + /proc filesystem support Config Question: /proc filesystem support Config Variable: CONFIG__PROC__FS Recommended The process-file-system represents an easy to handle means for receiving process information and kernel addresses. This in fact is a virtual file system, thus it does not occupy any disk space. This file system is needed for gathering information on processes (e. g. ps , free and more) and is therefore really important and necessary. + NFS filesystem support Config Question: NFS filesystem support Config Variable: CONFIG__NFS__FS Enables transparent mounting of remote partitions via a network in one's own directory tree. + Root file system on NFS Config Question: Root file system on NFS Config Variable: CONFIG__ROOT__NFS Allows you to mount even the root directory via NFA. This is for installing diskless workstations. + SMB filesystem support Config Question: SMB filesystem support Config Variable: CONFIG__SMB__FS If you need to mount disks running under Windows 95 or NT you need this file system. If you want the Linux machine to serve as a SMB server for the appropriate clients this file system is not needed. More information concerning this topic can be found under /usr/src/linux/Documentation/filesystems/smbfs.txt and in chapter 18. + NCP filesystem support Config Question: NCP filesystem support Config Variable: CONFIG__NCP__FS Enables you to connect volumes of a Novell NetWare server in your di- rectory tree, thus gaining full access on it. It is required that the kernel supports IPX (see section 13.4.6, page 192). Please have also a look at /usr/src/linux/Documentation/filesystems/ncpfs.txt. + ISO9660 cdrom filesystem support Config Question: ISO9660 cdrom filesystem support Config Variable: CONFIG__ISO9660__FS The file system used for CD-ROMs. If this is disabled no access onto CD-ROM drives is possible. + OS/2 HPFS filesystem (read only) 200 __________________________________________________________________________13.4.__Settings_in_kernel_configuration___________* *______________________ Config Question: OS/2 HPFS filesystem (read only) Config Variable: CONFIG__HPFS__FS OS/2's own High Performance File System. Since there is no information on this file system and the Linux implementation mainly relies on guesses there is only read access to it. + System V/Coherent filesystem support Config Question: System V/Coherent filesystem support Config Variable: CONFIG__SYSV__FS SCO , Coherent and Xenix all are commercial Unix systems for PCs and use this file system. If you plan to access one of these disks or floppies you have to activate this option. If you only plan to access these file systems via + NFS of course you do not need it in your own kernel. + Amiga FFS filesystem support Config Question: Amiga FFS filesystem support Config Variable: CONFIG__AFFS__FS The Fast File System (FFS) is used by Amiga machines since version AmigaOS 1.3 (34.20). If you plan to access one of these partitions you have to enable this option. Unfortunately access of Amiga disks is not possible since their disk controllers are not compatible. + UFS filesystem support (read only) Config Question: UFS filesystem support (read only) Config Variable: CONFIG__UFS__FS BSD, some Unix derives ( SunOS , FreeBSD , NetBSD , NeXTstep ) use this file system. Activating this option enables read only access on these file systems. 13.4.12 Character devices + Standard/generic serial support Config Question: Standard/generic serial support Config Variable: CONFIG__SERIAL Support for serial ports. Is needed if you plan to install a mouse. Recommended! + Digiboard PC/Xx Support Config Question: Digiboard PC/Xx Support Config Variable: CONFIG__DIGI Drivers of Digiboard's multi-serial cards. + Cyclades async mux support Config Question: Cyclades async mux support Config Variable: CONFIG__CYCLADES Cyclades multi-serial drivers. * * 201 13. The kernel _________________________________________________________________________________________________________________* *_______________________________________@ + Stallion multiport serial support Config Question: Stallion multiport serial support Config Variable: CONFIG__STALDRV Stallion multi-serial driver. + Parallel printer support Config Question: Parallel printer support Config Variable: CONFIG__PRINTER Driver for connecting a printer to the parallel port. If you want to use PLIP (CONFIG__PLIP , section 13.4.8) you should not activate this option. If you want to benefit from both you must either patch the kernel or compile both drivers as modules (see section 13.2) and load them on demand. More information can be found under Printing-HOWTO. + Mouse Support (not serial mice) Config Question: Mouse Support (not serial mice) Config Variable: CONFIG__MOUSE If you do not use a serial mouse but a mouse with its own interface, or a PS/2 mouse you have to set this option. Selecting a busmouse is done like this. The interrupt used by the mouse (default is IRQ 5) is set in /usr/src/linux/include/linux/busmouse.h. Either you change this value or you parse the appropriate parameter at boot time. + ATIXL busmouse support Config Question: ATIXL busmouse support Config Variable: CONFIG__ATIXL__BUSMOUSE Driver for ATIXL busmice. (on ATI graphics cards) + Logitech busmouse support Config Question: Logitech busmouse support Config Variable: CONFIG__BUSMOUSE Driver for Logitech busmice. + Microsoft busmouse support Config Question: Microsoft busmouse support Config Variable: CONFIG__MS__BUSMOUSE Driver for Microsoft mice. + PS/2 mouse (auxiliary device) support Config Question: PS/2 mouse (auxiliary device) support Config Variable: CONFIG__PSMOUSE Driver for PS/2 mouse. In almost any notebook the integrated mouse is touchpad, trackpoint and PS/2 compatible. + Support for user misc device modules Config Question: Support for user misc device modules 202 __________________________________________________________________________13.4.__Settings_in_kernel_configuration____________* *_____________________ Config Variable: CONFIG__UMISC Support for some modules which are not part of the kernel and which should be loaded on demand (e. g. touch screens). If you intend to design such a driver or just want to test you should say yes, all others say no. + QIC-02 tape support Config Question: QIC-02 tape support Config Variable: CONFIG__QIC02__TAPE Driver for streamers which are neither connected to a SCSI nor to an IDE bus (floppy tape). + Ftape (QIC-80/Travan) support Config Question: Ftape (QIC-80/Travan) support Config Variable: CONFIG__FTAPE Driver floppy tape streamer. + number of ftape buffers Config Question: number of ftape buffers Config Variable: NR__FTAPE__BUFFERS [3]) Amount of buffers for the FTape driver. + Advanced Power Management BIOS support Config Question: Advanced Power Management BIOS support Config Variable: CONFIG__APM Enables power management with machines that fulfil the standard. This mainly is of interest for notebook owners. This feature does not work with TI 4000M TravelMate and ACER 486/DX4-75 since they do not follow the standard. + Watchdog Timer Support Config Question: Watchdog Timer Support Config Variable: CONFIG__WATCHDOG If you enable this you release a "watchdog". If a certain file is not written for more than one minute (/dev/watchdog) the machine is booted au- tomatically. This might come in handy for network servers that want to keep the time-down as short as possible. + Enhanced Real Time Clock Support Config Question: Enhanced Real Time Clock Support Config Variable: CONFIG__RTC Allows access onto the real time hardware clock. Further information can be found under /usr/src/linux/Documentation/rtc.txt. + Sound card support Config Question: Sound card support Config Variable: CONFIG__SOUND * * 203 13. The kernel _________________________________________________________________________________________________________________* *_______________________________________@ Driver for different sound cards. They are configured in a separate step. The Linux kernel in the meantime supports almost any well-known sound card. More information is to be found under drivers/sound. + Kernel profiling support Config Question: Kernel profiling support Config Variable: CONFIG__PROFILE This is for kernel hackers who want to know what the kernel does and how long it takes. 13.5 Compiling the kernel After having adopted the kernel to your needs you start compilation by enter- ing: root@earth:/usr/src/linux # make dep root@earth:/usr/src/linux # make clean root@earth:/usr/src/linux # make zImage These three commands can be entered in one line as well. The are started one after the other. This might be useful if you want to (e. g. ) compile a kernel over night. Just enter: root@earth:/usr/src/linux # make dep clean zImage Dependent on your systems power it now takes app. 4 minutes (fast PentiumPro 1 up to several hours (386 with 8 MB) until you get your own kernel. While compiling you can of course examine the system on one of the other consoles. After a successful compilation you will find the kernel under /usr/src/linux/arch/i386/boot. The kernel image (the file containing the kernel) is known as zImage. If this file does not exist most definitely an error occurred. This might got lost in all the output. Verify it by entering: root@earth:/usr/src/linux # make zImage And watch out for error messages. But do not panic: Errors are very rare! If you have configured some parts as modules you have to start compiling these modules as well. This is done by entering root@earth:/usr/src/linux # make modules If the modules have been compiled successfully you can install them into the appropriate directories (/lib/modules/ ) by entering root@earth:/usr/src/linux # make modules__install 13.6 Install kernel After having compiled a kernel you have to make sure that it is used from now on. If you use LILOyou have to reinstall it. Normally you just copy the new kernel to /vmlinuz and invoke LILO root@earth:/ # lilo _____________________________________________1 A very popular test for hard and software s to compile the kernel with the make -j . You will need a quite an amount of RAM (more than 100 MB). A compiler is launched for each source file 204 ___________________________________________________________________________________________13.7.__Creating_a_boot_disk______* *______________________ For reassuring that nothing unexpected happens it is recommended to keep the old kernel just in case of an emergency. Enter an additional label /vmlinuz.old into /etc/lilo.conf as boot im- age and rename the old kernel to /vmlinuz.old thus making sure that you can boot the old one if the new one happens to fail. This is thoroughly de- scribed in chapter 8. If you have adopted /etc/lilo.conf to your needs you can enter root@earth:/usr/src/linux # make zlilo for installing LILO after having compiled the kernel automatically. If you boot Linux via DOS using linux.bat (loadlin) you have to copy the kernel to /dosc/loadlin/zimage2 for it to become active at next booting. Furthermore there is the following to keep in mind: The file /System.map contains kernel symbols which are needed by the modules for launching ker- nel functions correctly. This file depends on the current kernel. Therefore you should copy /usr/src/linux/System.map3 to the root directory (/ ). If you create your kernels using make zlilo this task is automatically done for you. If you get an error message like "System.map does not match actual kernel" then probably System.map has not been copied. 13.7 Creating a boot disk If you like to create a boot disk with the new kernel you can use the following command: root@earth:/usr/src/linux # make zdisk 13.8 Cleaning your disk after compilation You can delete the object files created during compilation if you do not have enough space on your disk by root@earth: # cd /usr/src/linux root@earth:/usr/src/linux # make clean If there is plenty of space and you plan to compile a kernel more often you might skip the last step. A new compilation will be faster as only those parts of the system are re-compiled that have changed. 2 ____________________________________________________ or to the directory where you have installed loadlin respectively 3 this file is created every time you create a new kernel * * 205 13. The kernel _________________________________________________________________________________________________________________* *_______________________________________@ 206 Chapter 14 The X Window System The X Window System is a de facto standard GUI for Unix. X11 was de- veloped as an enterprise of DEC (Digital Equipment Corporation) and the project Athena at MIT (Massachusetts Institute of Technology). The first re- lease of X11R1 was released in september 1987. Since release 6 the X Con- sortium, Inc. is responsible for the development of the X Window System . Up to now the X Window System supports a wide variety of machines. X11 is a network oriented product. Applications running on one machine earth can display their results on another machine provided the machines are con- nected via a network. This might be a local LAN or another machine reach- able via Internet. XFree86TM is a freely available implementation of X servers for PC systems. It was developed by a handful of ambitious programmers who founded the XFree86 team in 1992. This team lead to the foundation of the The XFree86 Project in 1994 whose aim it is to continue research and development on X11 and to provide it to public. We would like to thank the XFree86 team for their help and for the permission to include beta servers on our CDs,1 because without these realisation of this CD would have been a lot more work (if it would have been possible at all). The current release, XFree86TM 3.3.2, is a realisation of an X11R6.3 system for PC based Unix systems. The next chapters cover the configuration of your X server. XF86Setup is presented which enables you to do this rather easily. For really exploiting your hardware (graphic adaptor, monitor) there exists an option for optimiz- ing the configuration. Detailed information on configuring the X Window System can be found under /usr/doc/packages/xf86. _______________________________________________________________________________________________________________________||| |||||||| || | |||||||| Please configure your X Window System with extreme care! Never start || | |||||||| the X Window System until the configuration has finished. A faulty con- || | |||||||| || | |||||||| figured system can lead to irreparable damages to your hardware (this spe- || | |||||||| cially applies for fixed frequencies monitors). The authors of this book and || | |||||||| || | |||||||| S.u.S.E. cannot be held responsible for damages. This text has been written || | |||||||| with extreme care, but this does not guarantee that all methods presented || | |||||||| || | |||||||| in this book are correct and cannot damage your hardware. || | ||||||||_______________________________________________________________________________________________________________||_ 1 ____________________________________________________ Parts of this documentation are taken from chapter XFree86konfigurierenfrom[Hetze et al., 1995 ] which was kindly given to us by Dirk Hohndel * * 207 14. The X Window System___________________________________________________________________________________________________________* *_______________________________________@ Besides xf86config S.u.S.E. Linux contains another one: XF86Setup . On the contrary to xf86config XF86Setup directly works together with your server and is mouse driven. Install the package xfsetup with YaST (and all dependencies) as well as the appropriate X server. 14.1 Configuration using XF86Setup XF86Setup serves for a menu driven easy configuration of the XFree86TM system. Here, on the contrary to xf86config , configuration is almost done in graphical mode. Nevertheless XF86Setup does not try to replace xf86config (see page 210). There are known situations where configuring using XF86Setup fails (e. g. if you have a graphics card that is not 100% VGA compatible, or you use a fixed-frequency monitor). In these cases you should switch back to xf86config . For configuration the following data should be held at hand: o Type of mouse and port where the mouse is connected to. o Vendor's name of your graphics device, better, the technical specifica- tions. o Your monitor's data (mainly horizontal and vertical frequencies, etc.) If these data are known, or at hand you can start configuring. This, of course, can only be done by `root'. The configuration is launched using the command: root@earth:/root # /usr/X11R6/bin/XF86Setup First the program checks whether all required files and packages are installed. If there already exists an /etc/XF86Config on your machine you are asked if you would like to use it as default. Then you should press Enter for entering graphical mode. After having pressed Enter it can take up to two minutes (depending on the load of your machine) until the graphic screen appears and you can see the welcome mes- sage of XF86Setup . A short help screen is presented which tells you how to proceed. Now just press Enter ; a help screen appears that tells you the keyboard settings. This page can be viewed anytime by pressing ? or Alt + H . 14.1.1 Mouse After pressing Enter once again you are on the mouse configuration page. Please do not move the mouse right now but use p for selecting the appro- priate protocol instead. For most modern serial mice Microsoft or LogiTech is the right choice. Most bus mice are PS/2 compatible. After having selected a protocol you should press Enter for selecting the interface. First serial interface is /dev/ttyS0 , the second is /dev/ttyS1 . Bus mice are connected to /dev/psaux (nor- mally). By means of the cursor keys you can scroll the list of interfaces to 208 __________________________________________________________________________14.1.__Configuration_using_XF86Setup______________* *______________________ find the correct one. If you have already entered a port for your mouse at installation you should just enter /dev/mouse . After having set these items, press a (for apply) for writing the settings. Now you should try to move your mouse. The cursor should follow the mouse and a mouse-click should cause the pictogram on the right to blink accordingly. If this does not work you should select another protocol (and eventually another interface) by pressing p and n . So, if the middle mouse button does not behave as it should you could try to activate ChordMiddle (either click it or select it by c ). Quite similar you can select e (or click on Emulate3Buttons ) for emulating the third button by pressing button one and two simultaneously. 14.1.2 Keyboard If the configuration of you mouse is done you can go to keyboard configura- tion either by clicking on Keyboard or by pressing Alt + k . 14.1.3 Graphics device Now on it goes with specification of your graphics device. Either you click on Card or you press Alt + c . Now there is a long list of supported graphics cards. You can scroll through this list with the scrollbars on the right side. If you find your card in the list just click on it. Most probably this is all you have to do. You should not blindly accept a card with a similar name since even on cards with similar names there can be slight variations in clock-chips and ramdac2 ! If your card is not on this list configuration becomes slightly more com- plicated. Anyway you should select Detailed Setup and decide yourself which server and which options to choose. First you should select the X-server you are going to use. Here Mono stands for the simplest server of the XFree86server family. which should run on any VGA card. On the other hand it does not exploit the card. VGA16 is a similar generic server which should also work on most graphics cards. This server at least offers 16 colours. SVGA -server is the server of your choice for most SVGA cards that do not have a special server. The other servers are only interesting for a small group of cards. 8514 for older 8514 compatible cards, AGX for XGA cards and compatible, I128 for cards based upon Number Nine's I128 chip, Mach8 , Mach32 und Mach64 for the corresponding ATI cards. P9000 is designed for Weitek P9000 chipset only, they can not be used with P9100 cards. S3 suits every S3 cards except ViRGE and ViRGE/VX. S3V server supports these cards. Finally the W32 server only supports Tseng ET4000/W32 and ET6000. These cards are sup- ported by SVGA as well. Which one to choose mainly depends on your card. Some W32 cards only run with the W32 server, others only run with SVGA . 2 ____________________________________________________ Random Access Memory Digital-to-Analogue Converter * * 209 14. The X Window System___________________________________________________________________________________________________________* *_______________________________________@ And only the latter supports higher colour depths with ET6000 cards. W32 cards are only supported (by both servers) with 256 colours. All other settings should only be done by experienced users and only if the default settings fail. In this case you are obliged to find out which chipset and which ramdac is installed on your card (if this is not part of the graphic chipset). Studying the online documentation and the man pages is inevitable. 14.1.4 Monitors Next you should select your monitor. Click on Monitor or you press Alt + n . Now you can enter your monitor's settings. There are many default configu- rations listed. If none of them suits you you have to enter the correct values in the above entries. Now configuration is almost done. The options that are sorted under Other usually do not have to be changed. Now click on Done or press Alt + d and confirm by pressing Okay . XF86Setup now tries to launch the selected server. Normally this should not fail. If you receive an error message or if the system crashes you should restart configuration and try to solve the problems by means of the online documentation as well as the man pages. If the server started as expected you will be presented a screen where you can finally write your configuration. Normally there is no reason why you should change the path /etc/XF86Config, so just click on Save the configuration and exit and confirm by pressing Okay . Finally you are asked whether you want to create a link X to the selected server. Please answer yes. The X Window System now can be launched by any user using the startx command3 All error messages that occur can be found in ~/.X.err 14.2 Configuration using xf86config In most cases XF86Setup is superior to xf86config as a simple configuration tool. There are nevertheless some rare cases where XF86Setup fails, in this case just use xf86config which works in almost any cases. Please have the following information at hand. o Mouse type, port where the mouse is connected to and baud rate (the latter normally is optional). o Specification of the graphics card o Monitor data (Frequencies etc.) If these settings are known, or you have your manuals at hand you can start configuring. This can only be done by `root'! The configuration is launched by: root@earth:/root # /usr/X11R6/bin/xf86config _____________________________________________3 A pre-configured X-Window GUI (using the fvwm windowmanager) is provided along with the `sample user'; it is recommended to start startx as the `sample user' instead of `root' 210 ___________________________________________________________________________14.2.__Configuration_using_xf86config____________* *______________________ 14.2.1 Mouse After the welcome screen you are asked your mouse type. There are the following selections: _______________________________________________________________________________________________________________________|| |||||||| | |||||||| 1. Microsoft compatible (2-button protocol) | |||||||| 2. Mouse Systems (3-button protocol) | |||||||| 3. Bus Mouse | |||||||| | |||||||| 4. PS/2 Mouse | |||||||| 5. Logitech Mouse (serial, old type, Logitech protocol) | |||||||| 6. Logitech MouseMan (Microsoft compatible) | |||||||| 7. MM Series | |||||||| | |||||||| 8. MM HitTablet | |||||||| | ||||||||_______________________________________________________________________________________________________________ | Screen output 14.2.1: Selection of mouse for X While assigning mice you should consider that many of the new Logitech mice are Microsoft compatible or use the MouseMan protocol. The selection Bus Mouse refers to any bus mice, even Logitech! The suitable type is selected by giving its number. There might be a request whether "ChorMiddle" should be activated. This is necessary for some Log- itech mice or trackballs to activate the second mouse button (the middle one). Please answer the following question with either 'y' or 'n'. Do you want to enable ChordMiddle? If you use a two-button mouse you can emulate the third button by answering `y' to the next question. Please answer the following question with either 'y' or 'n'. Do you want to enable Emulate3Buttons? The middle button is emulated by simultaneously pressing the two mouse buttons. Next you have to specify the mouses interface: Now give the full device name that the mouse is connected to, for example /dev/tty00. Just pressing enter will use the default, /dev/mouse. Mouse device: If you already entered a port for your mouse at system installation you should enter it (/dev/mouse ) 14.2.2 Keyboard Now you are asked whether to assign Meta (ESC) to the left Alt-key and to assign ModeShift (AltGr) for the right. Please answer the following question with either 'y' or 'n'. Do you want to enable these bindings for the Alt keys? You should answer `y' for being able to access all keys via AltGr and the left Alt can serve as Meta-key4 . 4 ____________________________________________________ e. g.in Emacs * * 211 14. The X Window System___________________________________________________________________________________________________________* *_______________________________________@ 14.2.3 Monitor Next you have to specify your monitor. You should be extreme careful with vertical and horizontal frequencies! These can be found in your monitors handbook. ________________________________________________________________________________________________________* *________|| | Setting frequencies wrongly can lead to irreparable damages to your mon- * * |||| | * * ||| | itor! The X-Window system only addresses video modes which drive the * * ||| | monitor in the given frequency range. Entering frequencies which the * * ||| | * * ||| | monitor was not designed for can severely damage it! * * ||| |_______________________________________________________________________________________________________* *________|||_ Some monitors are listed under /usr/X11R6/lib/X11/doc/Monitors5 . For entering horizontal frequency the following selection is presented: ________________________________________________________________________________________________________* *________|||| | * * | | hsync in kHz; monitor type with characteristic modes * * | | 1 31.5; Standard VGA, 640x480 @ 60 Hz * * | | 2 31.5 - 35.1; Super VGA, 800x600 @ 56 Hz * * | | 3 31.5, 35.5; 8514 Compatible, 1024x768 @ 87 Hz interl. * * | | * * | | (no 800x600) * * | | 4 31.5, 35.15, 35.5; Super VGA, 1024x768 @ 87 Hz il., * * | | 800x600 @ 56 Hz * * | | 5 31.5 - 37.9; Extended Super VGA, 800x600 @ 60 Hz, * * | | * * | | 640x480 @ 72 Hz * * | | 6 31.5 - 48.5; Non-Interlaced SVGA, 1024x768 @ 60 Hz, * * | | 800x600 @ 72 Hz * * | | 7 31.5 - 57.0; High Frequency SVGA, 1024x768 @ 70 Hz * * | | * * | | 8 31.5 - 64.3; Monitor that can do 1280x1024 @ 60 Hz * * | | 9 31.5 - 79.0; Monitor that can do 1280x1024 @ 74 Hz * * | | 10 Enter your own horizontal sync range * * | | Enter your choice (1-10): * * | | * * | | * * | |_______________________________________________________________________________________________________* *________ | Screen output 14.2.2: Entry for the monitor's horizontal frequency Only if the settings of your monitor are unknown you should take one of the predefined modes. Selection `10' lets you enter the correct frequencies. ________________________________________________________________________________________________________* *________|| | * * | | 1 50-70 * * | | 2 50-90 * * | | 3 50-100 * * | | * * | | 4 40-150 * * | | 5 Enter your own vertical sync range * * | | * * | | Enter your choice (1-5): * * | | * * | | * * | |_______________________________________________________________________________________________________* *________ | Screen output 14.2.3: vertical frequencies in detail Again, using the known values should be preferred to one of the items `1' to `4' Next you have to enter a name for the monitor: _____________________________________________5 Of course these values are not in our liability! 212 ___________________________________________________________________________14.2.__Configuration_using_xf86config____________* *______________________ Enter an identifier for your monitor definition: the vendor Enter the vendor name of your monitor: and the model Enter the model name of your monitor: Here a name can be given or you select the predefined values by pressing Enter . Setting up your monitor is now done. 14.2.4 Graphics cards / X-Server Now setup commences with specifying your graphics card: Do you want to look at the card database? If you enter `y' a selection of predefined cards is presented. Here you can select your card by pressing the corresponding number. Do not trust this list blindly since there can be differences in clock-chip and ramdac6 settings! That is why later on there is a menu item to select a ramdac and a clock-chip although you have entered it already. Anyway, the predefined settings for this card are set as defaults. The card definitions contain information on clock-chips, ramdac and the X- Server to be used. Furthermore some valuable information concerning the card is written to the device section in XF86Config. If your card is not listed, do not panic. In this case switch back to the menu by selecting `q'. Please do only select one of the defined cards if it matches your card exactly! Selecting a card with a similar name is not recommended. Similar names do not necessarily refer to similar hardware : : : Further information on how to configure your card are described in chapter section 14.3. After having specified your card, the X-Server is next. 1 A server for monochrome (b/w) monitors. This should run on any VGA compatible graphics card and at least offer a resolution of 640x480. 2 16 colours server XF86__VGA16. Should run with any VGA compatible card. 3 SVGA server XF86__SVGA. This server supports a wide variety of SVGA cards. Graphic acceleration is used with some Cirrus or WD cards. The 16, or 32 bit colour mode can be activated with some Cirrus cards. 4 Server for accelerated cards. Here there are some server at your disposition (see below). 5 This item only exists if you have entered a card definition in the previous selection. Here the server is selected (default) that suits the selected card. If a server has been selected you are asked whether to create a symbolic link to /usr/X11R6/bin/X; if this is answered with `y' you are asked whether to put it in /var/X11R6/bin/X 6 ____________________________________________________ Random Access Memory Digital-to-Analogue Converter * * 213 14. The X Window System___________________________________________________________________________________________________________* *_______________________________________@ ________________________________________________________________________________________________________* *________|||| | 1 The XF86_Mono server. This a monochrome server that should work on * * | || any VGA-compatible card, in 640x480 (more on some SVGA chipsets). * * || | 2 The XF86_VGA16 server. This is a 16-color VGA server that should * * | | * * | | work on any VGA-compatible card. * * | | 3 The XF86_SVGA server. This is a 256 color SVGA server that supports * * | | a number of SVGA chipsets. It is accelerated on some Cirrus and WD * * | | chipsets; it supports 16/32-bit color on certain Cirrus * * | | * * | | configurations. * * | | 4 The accelerated servers. These include XF86_S3, XF86_Mach32, * * | | XF86_Mach8, XF86_8514, XF86_P9000, XF86_AGX, XF86_W32 and * * | | XF86_Mach64. * * | | * * | | * * | | These four server types correspond to the four different "Screen" * * | | sections in XF86Config (vga2, vga16, svga, accel). * * | | * * | | * * | | 5 Choose the server from the card definition, XF86_S3. * * | | * * | | Which one of these four screen types do you intend to run * * | | by default (1-4)? * * | | * * | | * * | |_______________________________________________________________________________________________________* *________ | Screen output 14.2.4: Selecting a X-Server Do you want to set it in /var/X11R6/bin? Please reply in the affirmative, since it might be not possible to write to /usr7 Afterwards (if you have pressed `4' in the previous selection) there is a menu with all available accelerated X-servers. ________________________________________________________________________________________________________* *________|||| | * * | | Select an accel server: * * | | * * | | 1 XF86_S3 * * | | * * | | 2 XF86_Mach32 * * | | 3 XF86_Mach8 * * | | 4 XF86_8514 * * | | 5 XF86_P9000 * * | | * * | | 6 XF86_AGX * * | | 7 XF86_W32 * * | | 8 XF86_MACH64 * * | | * * | | * * | | Which accel server: * * | | * * | |_______________________________________________________________________________________________________* *________ | Screen output 14.2.5: Accelerated X-server These servers support the corresponding cards. Assigning this link requires a suitable X-server to be be installed, i. e.the correct X-server has been selected. After selecting the X-server the graphics card has to be adapted. First you have to enter the memory installed: Afterwards the name, vendor name and type of your card is requested. If a graphic card has been selected, pressing Enter will do. _____________________________________________7 e. g.at CD installation 214 ___________________________________________________________________________14.2.__Configuration_using_xf86config____________* *______________________ _______________________________________________________________________________________________________________________|| |||||||| | |||||||| How much video memory do you have on your video card: | |||||||| | |||||||| | |||||||| 1 256K | |||||||| 2 512K | |||||||| 3 1024K | |||||||| 4 2048K | |||||||| | |||||||| 5 4096K | |||||||| 6 Other | |||||||| | |||||||| Enter your choice: | |||||||| | |||||||| | ||||||||_______________________________________________________________________________________________________________ | Screen output 14.2.6: Selecting memory Enter an identifier for your video card definition: Enter the vendor name of your video card: Enter the model (board) name of your video card: If an accelerated X-server has been selected you have to enter the ramdac settings. These are only relevant for S3 and AGX servers. _______________________________________________________________________________________________________________________|||| |||||||| | |||||||| 1 AT&T 20C490 (S3 server) att20c490 | |||||||| 2 AT&T 20C498/21C498/22C498 (S3) att20c498 | |||||||| 3 AT&T 20C505 (S3) att20c505 | |||||||| | |||||||| 4 BrookTree BT481 (AGX) bt481 | |||||||| 5 BrookTree BT482 (AGX) bt482 | |||||||| 6 BrookTree BT485/9485 (S3) bt485 | |||||||| 7 Sierra SC15025 (S3, AGX) sc15025 | |||||||| | |||||||| 8 S3 GenDAC (86C708) (autodetected) s3gendac | |||||||| 9 S3 SDAC (86C716) (autodetected) sdac | |||||||| 10 STG-1700 (S3) stg1700 | |||||||| 11 TI 3020 (S3) ti3020 | |||||||| | |||||||| 12 TI 3025 (S3) ti3025 | |||||||| | ||||||||_______________________________________________________________________________________________________________ | Screen output 14.2.7: Setting a RAMDAC In most cases a simple Enter will do. If your graphics card has been selected that supports a certain ramdac this should be chosen. After having answered this question you can enter a clock-chip for accelerated cards (if there is one). Giving a clock-chips avoids clock lines as the clocks needed can be programmed. If a card without clock-chip is selected, a simple Enter will do (thus not selecting a clock-chip). If a card has been selected the clock-chip is set as default (if there is any). If no clock-chip has been set xf86config suggests running X -probeonly for determining the supported clock timings. These are automatically written in XF86Config in a separate Clocks line. Here we have to put straight why the automatically defined settings can be really dangerous: If the card has a programmable clock-chip the X-server can not distinguish between the different clocks and only recognizes clocks * * 215 14. The X Window System___________________________________________________________________________________________________________* *_______________________________________@ ________________________________________________________________________________________________________* *________|||| | 1 AT&T 20C490 (S3 server) att20c490 * * | || 2 AT&T 20C498/21C498/22C498 (S3) att20c498 * * || | 3 AT&T 20C505 (S3) att20c505 * * | | 4 BrookTree BT481 (AGX) bt481 * * | | * * | | 5 BrookTree BT482 (AGX) bt482 * * | | 6 BrookTree BT485/9485 (S3) bt485 * * | | 7 Sierra SC15025 (S3, AGX) sc15025 * * | | 8 S3 GenDAC (86C708) (autodetected) s3gendac * * | | * * | | 9 S3 SDAC (86C716) (autodetected) s3_sdac * * | | 10 STG-1700 (S3) stg1700 * * | | 11 TI 3020 (S3) ti3020 * * | | 12 TI 3025 (S3) ti3025 * * | | * * | | 13 Normal DAC normal * * | | * * | |_______________________________________________________________________________________________________* *________ | Screen output 14.2.8: Setting of clock-chip 0, 1, and sometimes 2. All other values are more or less random numbers (normally clocks 0, 1, and 2 repeat and are replace by zeros.) Well, any clock (despite 0 and 1) depends on the programming of the clock- chip. Thus clock 2 can have a different setting at probing as at runtime (the value entered in XF86Config). Then all timings are wrong and the monitor can be damaged severely! A good hint on a programmable clock-chip (and the problems that follow) are many zeros or repeating timing values. Never ever write such values to XF86Config! For determining the clock-chips commence as follows: o The best way is to enter an existing (programmable) clock-chip (if there is one). Now it is programmed accordingly; your XF86Config will not contain clock lines. You can compare chips on the card with the chips offered in the menu. Most modern S3 cards do have a programmable clock-chip. o If you do not have a programmable clock-chip you should launch X -probeonly and compare these values with those of the manual. If these values corre- spond ( 2) you should enter them in XF86Config. If there are no hints in the manual you can determine the values by running X -probeonly (best on an unloaded machine). Check if the values are correct since clock values can not be determined on every card (many zeros or repeat- ing values are a sign for not valid settings). Enter these valid values into XF86Config. Do not omit values, do not try to rearrange them nor change them in any way. The values have to be entered in the exact order. If the P9000 server is used order is irrelevant; just enter the modes for the desired clock in the Clocks-line. o In general: If there is a programmable clock-chip there should be no clocks-line in XF86Config (exception P9000). For cards without a programmable clock-chip there should be a Clocks- line in XF86Config. This avoids the tedious (and sometimes even dan- gerous) testing at each startup. Furthermore for cards with unreadable values there are no invalid values and there is no risk for your monitor. 216 ______________________________________________________________________14.3.__Optimizing_the_X_Window_System_________________* *______________________ So, if (after having read the previous section) you want to let clocks be rec- ognized automatically, just enter `y' to the following question: Do you want me to run 'X -probeonly' now? Now the screen turns black and then a list of determined clocks is presented; or a message appears that no clocks could be found. If you have selected a clock-chip this question does not appear since clocks are programmed auto- matically. In this case this section is skipped. _______________________________________________________________________________________________________________________|| |||||||| If the previous question has been answered with `y' and the screen re- ||| | |||||||| || | |||||||| mains black for more than 30 seconds you should cancel testing immedi- || | |||||||| ately by Ctrl + Alt + _ , or Ctrl + c respectively! If this does not || | |||||||| || | |||||||| work switch off monitor and machine so that the hardware is not damaged! || | ||||||||_______________________________________________________________________________________________________________||_ 14.2.5 Saving your configuration So far for installation. Now the configuration file has to be written. It is rec- ommended to do so in XF86Config in directory /etc. This is to ensure that even in a networking environment each machine has its own configuration file even if they share the /usr filesystem. xf86config first suggests writing XF86Config to the current directory. This should be answered in the negative: Do you want it written to the current directory as 'XF86Config'? Then you are asked where to save your configuration: Please give a path+filename to write to: Here you have to enter: "/etc/XF86Config"! Now xf86config is done and configuration of your X-Window system is over. 14.3 Optimizing the X Window System xf86config creates /etc/XF86Config. This is the primary configuration file for the X Window System . Here you can find all settings concerning your graphics card, mouse and monitor. XF86Config is divided into several sections (Sections ) which serve for configuring of a certain topic. A section always resembles: Section entry 1 entry 2 entry n EndSection There exist the following types of sections: * * 217 14. The X Window System___________________________________________________________________________________________________________* *_______________________________________@ Files This section describes all paths used and the RGB colour table. ServerFlags Here general switches are set. Keyboard Servers for describing and setting up your keyboard. In our case this can only be Device "Standard" Pointer Assigns all necessary references to the appropriate pointer. In most cases this will be a mouse. Further pos- sibilities are LightPens or GraphicsBoards. Important set- tings are the Protocol and Device . Monitor Describes the monitor in use. Elements of this sections are a name that is referred to by defining Screen , as well as bandwidth (bandwidth ) and sync frequencies (HorizSync and VertRefresh ). Settings are given in MHz, kHz and Hz. Normally the server refuses any mod- eline that does not correspond with the specification of the monitor. This is to prevent that too high frequencies are sent to the monitor by accident. Device This section defines a certain graphics card. Its referenced by its name. Screen This section now puts together a Driver (e. g. vga2 ) a monitor and a Device thus forming all necessary set- tings for XFree86. Subsection Display lets you assign a size to the virtual screen (Virtual of ViewPort using Modes ). We now take a closer look at Monitor Device and Screen . Information on the other sections can be found under [The XFree86TM -Team, 1996 ]. There can be many Monitor sections in XF86Config. Even multiple Screen sections are possible; which one is started depends on the launched server. 14.3.1 Screen section First we have a closer look at the screen section. This, as mentioned above combines a monitor and a device section and determines which resolution using which colour depth is used. A screen section can, e. g. , resemble the following diagram: Driver determines which X-server it applies to. Servers listed on page 214 are accessed via the keywords . _______________________________________________________________________________________________________* *_________ Accel For special accelerated servers Mono Not VGA 1 and 4 bit server SVGA Super VGA server VGA2 1 Bit (monochrome-)VGA server VGA16 4 Bit VGA server _______________________________________________________________________________________________________* *_________ Table 14.1: Keywords for X server in /etc/XF86Config 218 ______________________________________________________________________14.3.__Optimizing_the_X_Window_System_________________* *______________________|| _______________________________________________________________________________________________________________________|||| |||||||| Section "Screen" | ||||||||| Driver "accel" || |||||||| Device "Miro Crystal 40SV" | |||||||| | |||||||| Monitor "EIZO T563-T" | |||||||| DefaultColorDepth 16 | |||||||| Subsection "Display" | |||||||| Depth 8 | |||||||| | |||||||| Modes "1024x768" "800x600" "640x480" | |||||||| ViewPort 0 0 | |||||||| Virtual 1024 768 | |||||||| EndSubsection | |||||||| | |||||||| Subsection "Display" | |||||||| Depth 16 | |||||||| Modes "1280x960" "1152x864" "1024x768" "800x600" | |||||||| ViewPort 0 0 | |||||||| | |||||||| Virtual 1280 960 | |||||||| EndSubsection | |||||||| Subsection "Display" | |||||||| Depth 32 | |||||||| | |||||||| Modes "1024x768" "800x600" "640x480" | |||||||| ViewPort 0 0 | |||||||| Virtual 1024 768 | |||||||| EndSubsection | |||||||| | |||||||| EndSection | |||||||| | ||||||||_______________________________________________________________________________________________________________ | File contents 14.3.1: Screen section of /etc/XF86Config There can be a screen section each in XF86Config which will be used if the corresponding server is started. The next two lines, Device and Monitor , set the graphics card and the mon- itor which belong to this definition. These are just points to the Device- and Monitor sections with the corresponding names. This section is discussed further on more thoroughly. Using ColourDepth one can set the colour mode to start if it is started with- out explicitly setting the mode. Now there is a Display subsection for each colour depth. Depth assigns the colour depth of this subsection. Possible values for Depth are: 8, 15, 16 24 and 32. Not every X-Server supports all these modes. For most systems 24 and 32 are equal, some others take 24 for packed-pixel 24bpp mode whereas others choose 32 for padded-pixel mode. * * 219 14. The X Window System___________________________________________________________________________________________________________* *_______________________________________@ After the colour depth a list of resolutions is set (Modes ). This list is checked by the server coming from the left to the right. For each resolution a suitable Modeline is searched for which has to correspond with one of the given clock rates or a clock rate to program the card. The first one that is found is the so-called Default mode . Using Strg + Alt + grey + one can switch to the right, using Strg + Alt + grey - to the left. Thus enabling you to vary the resolution at runtime. The last two lines of this subsection refer to the size and anchor of the virtual screen . Size depends on the amount of memory installed on your card and the desired colour depth, not on the maximum resolution of the monitor. If the card has 1 MB video RAM installed the virtual screen can be up to 1024x1024 by 8 bit colour depth. Especially for accelerated cards it is not recommended to use up all your memory for the virtual screen since this memory on the card is used for several font and graphic caches. Size of the virtual screen is assigned by Virtual . Viewport assigns a viewport. This is the point where the upper left corner of the physical screen is reflected in the virtual screen. Assigning 0 0 means that the upper left corners overlap at startup time. The visible section is moved across the virtual screen by moving the mouse to the edge. Thus the viewport is only important at startup time and even then only if size of the virtual screen differs from the size of the real screen. 14.3.2 Device section A device section describes a certain graphics card. There can be as many device entries in XF86Config as you like, as long as the names of them differ. We do not go into greater details of the device section. Instead we recom- mend to use the extensive documentation (/usr/X11/lib/X11/doc and the manual pages [The XFree86TM -Team, 1996 ]). This chapter mainly shows where to get predefined device sections and sets which cards are supported by XFree86. In /usr/X11R6/lib/X11/doc/Devices there is a collection of device sec- tions. If you used a card mentioned in this file you should set the appropriate section in /etc/XF86Config and adopt the screen section by entering the specific device. If the card is not listed this does not necessarily mean that this card is not supported by XFree86! This mainly means that no adequate device section has been sent to the XFree86team. A list of supported cards can be found under /usr/X11/lib/X11/doc/README, in /usr/X11/lib/X11/ doc/AccelCards there is a list of accelerated cards. 14.3.3 Monitor section Monitor sections describe a monitor analogous to the device sections. Again, there can be as many Monitor sections as desired in XF86Config. Screen section sets which monitor section is relevant. 220 ______________________________________________________________________14.3.__Optimizing_the_X_Window_System_________________* *______________________ For monitor definition the same applies: it should only be set by experienced users. A vital part of the monitor section are the so called modelines which set horizontal and vertical timings for the appropriate resolution. _______________________________________________________________________________________________________________________|| |||||||| Without a profound knowledge of the monitors and graphics cards func- ||| | |||||||| tions nothing should be changed in the modelines since this can lead to || | |||||||| || | |||||||| severe damage to your monitor! || | ||||||||_______________________________________________________________________________________________________________||_ For those of you who might want to develop their own monitor descriptions the documentation in /usr/X11/lib/X11/doc might come in handy. It is recommended to have a look at [Fang et al., 1993 ] where function, hardware and the creation of modelines are explained in great detail. Predefined monitor sections that are taken over in XF86Config are to be found in /usr/X11/lib/X11/Monitors. A monitor not listed there should be driven with VESA standard timings such as set by xf86config . It is really important, that horizontal and vertical sync frequencies are set correctly! If you have a tested configuration, which was not mentioned here, we S.u.S.E. or the XFree86 team will include it in the list. * * 221 14. The X Window System___________________________________________________________________________________________________________* *_______________________________________@ 222 Chapter 15 The windowmanager - Window to your machine Once the configuration of the X-Server is realized, it is evident that the typical user wants to implement a color full desktop with windows, menus and loads of other stuff that an adequate desktop needs. This chapter deals with window managers1 . Here the following topics are available: o The window manager and its tasks. o fvwm2 - the window manager. o fvwm2 - the K Desktop Environment as an alternative. o susewm - a very elegant way to your own configuration file. o Real life: Adapting and configuring your personal desktop. Even if you want to go ahead right now, you still need some theory to begin with, so hold on! 15.1 Some theory 15.1.1 General On contrary to the monolithic graphic user desktops used e. g. in Windows or OS/2, the various functional layers used by Unix or Linux are separated into independent sections. This at first glance makes the system more comples. On the other hand it gains flexibility and the system is able to handle more complex tasks. The first layer is the operating system, which handles the "trivial" tasks like the memory management for example. On the next layer is located the + X Server (X Window System), which cor- responds to the "graphics device driver" used in other systems. The X Server serves provides a transparency layer below the GUI itself. Thus you are able to use these services via a whole network (including the Internet). In a nutshell: The "only" things the X-server can not are: o correspond with the graphics card, 1 ____________________________________________________ Window Manager is abbreviated as WM * * 223 15. The windowmanager_-_Window_to_your_machine____________________________________________________________________________________* *_______________________________________@ o draw dots, lines, rectangles and text, and o distribute services over the net or on the local host. Most users run a local X Server anyway but even they may exploit the services it provides. You may launch applications on a server, e. g. in the office, and redirect the output to your local screen. Another example. You don't need to sit in the same root with your noisy, roaring server , just connect a small and smart workstation via Ethernet and you can work on the server remotely. Now, for being able to display all these graphics stuff such as rectangles and things known as "windows" the services of a window manager are essential. Windows are rather important for clarity; windows let you start applications in windows next to each other. Menus make the use of your system easy at your fingertips. The windowmanager is an additional layer between the X-server, your appli- cation programs and the user2 . ____________________________________________________|||| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | |___________________________________________________ | Figure 15.1: Layers of GUIs under Linux Under Linux there is a wide variety of window managers, e. g. : o fvwm , or fvwm2 , respectively (the window manager) o fvwm95 (Win 95 clone) o bowman (look & feel of NeXTSTEP ) o ctwm o mwm ( Motif window manager) o olvwm ( OpenLook Virtual window manager) o CDE - Comon Desktop Environment (commercial) o Kwm - window manager of the K Desktop Environments (KDE) Besides there are many other windowmanagers available. There are among others: Wm2 , Mlvwm , Qvwm , Enlightenment , 9wm , Twm , Icewm , sc* *wm , :_:_:________________________________________ 2 Of course X application developers may access the server directly; just for clarity 224 _______________________________________________________________________________________________________15.1.__Some_theory___* *______________________ Which window manager you use mainly depends on your wishes and the supplied functions. There are remarkable differences in memory exhaust be- tween the window managers. More over the setup and the flexibility towards new features and updates play an important part. In figure 15.2, figure 15.3 and figure 15.4 you see three examples of window decoration used by Fvwm, Fvwm95 and AfterStep. Figure 15.2: Fvwm2 window decoration Figure 15.3: Fvwm95 window decoration Figure 15.4: AfterStep window decoration Of course you may test and install the various window managers. If you have decided on one window manager you can adapt it and your personal desktop to your needs. Most of the WM mentioned locate their configuration files and related data in subdirectories in path /usr/X11R6/lib/X11. Please feel free to rummage around! _______________________________________________________________________________________________________________________ |||||||| A hint: Most of the information in this chapter refers to fvwm2 , or KDE || | |||||||| || | |||||||| respectively so if you are unsure, we recommend using either of them! || | ||||||||_______________________________________________________________________________________________________________||_ 15.1.2 What does a windowmanager manage? Here we give you a short (and incomplete) list what properties of your desktop you can change via the use of a windowmanager: o Appearance of your windows - Width and height, colours, 3D effects of the window-frame - Window-controls, e. g. for moving, enlarging , title, fonts (window decoration), headlines and fonts o Overlapping of windows - Raising of windows (e. g. AutoRaise). * * 225 15. The windowmanager_-_Window_to_your_machine____________________________________________________________________________________* *_______________________________________@ - Pinning of windows o Focusing of windows by: - Clicking - "Touching" with the mouse pointer o Popup menus - Look&feel of menus (colours, fonts) - Behaviour of menus and submenus o Screen background o Virtual desktop (multiple desktops/screens) o Icon management o Linking sounds to actions on your desktop Unfortunately only a few windomanagers provide a means for configuring themselves. There is no configuration tool as you might expect. Some of them do have a GUI for setup. These are: AfterStep, Olvwm, CDE und KDE. Most WM read one or more configuration files at startup. The behaviour of the WMs may be set in these files. The syntax itself is more or less mnemonic. It's a pity that all of the WMs use a different format for saving their configuration files. You are obliged to read the appropriate manpages. At last you will find "your" WM. You will pay attention to it and configure its configuration files to suit all your needs. The rest of the time should be "work" not only configuration..;) 15.1.3 Starting different Windowmanagers For starting a WM, S.u.S.E. Linux provides different possibilities depending on how you start your X Window System3 . Starting via the susewm-Menu susewm provides a list of menus that may be launched directly from the task bar. You may switch to and fro between the WMs. Normally the windows on your screen will remain untouched (well the decoration changes). This feature is not available along with ctwm, mwm, kwm and CDE (the developers just forgot to implement it). The variable $WINDOWMANAGER On the long run, it is of course annoying if one wants to use for example Fvwm 95 and has to start the pre-set Fvwm2 all the time just for changing over to Fvwm 95 via the menu. For that reason the environment variable $WINDOWMANAGER was in- troduced for starting a specific WM right at the beginning. The value of the variable $WINDOWMANAGER is used for starting a certain WM. _____________________________________________3 either via the XDM or a text console 226 ______________________________________________________________________________15.2.__The_fvwm2_______window_manager_________* *______________________ Starting with startx First we want to provide you with a way on how to start the X Window System right out of the text console (via startx ). You can do it for example easily via the following command: newbie@earth: > startx fvwm95 for starting Fvwm95 directly. This works for most of the WMs included in S.u.S.E. Linux. This command may be extended by the colour depth of you want to start the rather colour intensive AfterStep (afterstep ): newbie@earth: > startx afterstep -- -bpp 16 starts the X Window System in 16 bit color mode (65536 colors) using the AfterStep WM. In these examples we always use the name of the WM binary. If you don't want to enter the name of an WM all the time at the beginning, or if you rather want to use another WM than Fvwm2 as standard, you can insert resp. change the following line in the file ~/.bashrc in path $HOME: export WINDOWMANAGER=fvwm95 Here you need to set the WM's binary. You might need to give in the complete pathname especially if /usr/X11R6/bin is not included in $PATH. You may as well insert this entry in the systemwide /etc/profile for set- ting it systemwide as default. As any user might overwrite this setting in his/her personal ~/.bashrc file we at S.u.S.E. did not set it (e.g. with YaST). Start via xdm Setting an environment variable ~/.bashrc is the best way to use your favourite WM if you plan to work with the XDM. _______________________________________________________________________________________________________________________|| |||||||| If you start your X Window System via KDM (KDE GUI) you may not set ||| | |||||||| the environment variable $WINDOWMANAGER but select your WM || | |||||||| || | |||||||| from the KDM pulldown menu. See section 15.4. || | ||||||||_______________________________________________________________________________________________________________||_ 15.2 The fvwm2 window manager 15.2.1 General Now we want to deal with Fvwm2 (fvwm2 ), a windowmanager that soon became the windowmanager under Linux, or XFree86TM rrespectively. Fvwm2 4 is the successor of the old fvwm1. It now needs much more memory than the older version but provides lots of new functionalities and configura- tion possibilities. Besides those usual functions for managing windows (sic!) and look & feel of the buttons (desktop) it now provides background menus and modular ap- plications which can be loaded at runtime. Those features provide interesting functions as, e. g. , a button-bar. 4 ____________________________________________________ In S.u.S.E. Linux fvwm2 is packed under package fvwm, series xwm; the previous version fvwm in package package fvwm1, series xwm * * 227 15. The windowmanager_-_Window_to_your_machine____________________________________________________________________________________* *_______________________________________@ More information about functions, starting and configuration of fvwm2 and its modules is reached via the corresponding manual pages (manpage of fvwm2 (man fvwm2 ), manpage of FvwmAudio (man FvwmAudio ), manpage of FvwmButtons (man FvwmButtons ), etc.) or under /usr/doc/packages/ fvwm which automatically is installed when installing the fvwm package. In any case of doubt you should consider looking at those documents first. ________________________________________________________________________________________________________* *________|| | As the original packages of both Fvwm and Fvwm2 use the same place for * * |||| | * * ||| | their manual pages (manpages for fvwm2 can be viewed as usual), there is * * ||| | * * ||| | a special action to be taken to view them. You can read the manpages * * ||| | of Fvwm2 as usual with the read command. The manpages of Fvwm * * ||| | * * ||| | (version 1) and Fvwm95 have been located in different paths under S.u.S.E. * * ||| | Linux. * * ||| |_______________________________________________________________________________________________________* *________|||_ For reading an Fvwm (version 1) manpage (e. g. ) you should proceed as given: fvwmman FvwmButtons for reading a Fvwm95 manpage (e. g. ) you should enter: fvwm95man FvwmButtons Then you will get access to the manpage to FvwmButtons of the correspond- ing package. You also can view the manpages of the different WM right out of the menu. You can find this "special" manpages (if available) in the menu `Windowmanager', submenu `Manpages'. If you don't want to start from scratch you may generate a "frame" with susewm and tailor it to your needs afterwards. susewm is described in sec- tion 15.5. Configuration of fvwm2 The Fvwm2 can be configured via two files: o A system wide configuration file which should be available at any time: This is .fvwm2rc and lies under /usr/X11R6/lib/X11/fvwm2. o A user specific file which is not really necessary (but recommended). This file is called ~/.fvwm2rc and lies in the home directory of the user. The package susewm (described later on) ensures the existence of a system wide configuration file for fvwm2 . fvwm2 reads these configuration files at startup time5 First fvwm2 tries to read the user's configuration file; if this doesn't exist it reads the system wide file. All fvwm2 modules read this file too, but they only use the statements corre- sponding to themselves. It is recommended for every user to create his/her own configuration file which he/she can change and adapt to his/her personal needs. _____________________________________________5 In general the windowmanager is either started from /usr/X11R6/lib/X11/xinit/xinitrc or from the user's private configuration file ~/.xinitrc. The environment variable $WIN- DOWMANAGER should be set to the full pathname of the corresponding window manager. 228 ___________________________________________________________________________________________________15.3.__Fvwm2_settings____* *______________________ _______________________________________________________________________________________________________________________ |||||||| After changing configuration files the WM has to be restarted for the || | |||||||| || | |||||||| changes to take effect. || | ||||||||_______________________________________________________________________________________________________________||_ For restart there is a menu entry (supplied with pre-installed S.u.S.E. Linux configuration) in `Working menu' called `windowmanager'. You could as well restart the X-server. It is also possible to restart the WM from within an xterm or a text console. Just enter: newbie@earth: > killall -10 fvwm2 Interactive changing of configurations in fvwm2 is only available via module FvwmConfig , this tool however only offers very limited capabilities concern- ing the configuration since those modifications can not be stored. Besides this one there exist two more modules - FvmSave and FvwmSaveDesk - which in fact can store the actual state of the WM. However, their files can not be read automatically at the next start of the WM. If you're interested, please read the Manpages of the respective module. 15.3 Fvwm2 settings 15.3.1 General Now we shall deal with your personal Fvwm2 configuration file. Of course you should have created one in advance. The howto is listed in section 15.5. Besides the file created by susewm you may also use the configuration file provided by the authors of Fvwm2. This may be found under /usr/doc/ packages/fvwm/system.fvwm2rc. Now start up your favourite editor and load the file ~/.fvwm2rc. We will have a look at some of the options. What happens when fvwm2 starts Scroll through the text until you reach the following comment: _______________________________________________________________________________________________________________________|| |||||||| | |||||||| ######################################################### | |||||||| | |||||||| # # | |||||||| # initialization function head # | |||||||| # common to all wms # | |||||||| | |||||||| # # | |||||||| ######################################################### | |||||||| | ||||||||_______________________________________________________________________________________________________________ | File contents 15.3.1: InitFunction in /.fvwm2rc Hereafter you will find anything that is launched at a restart of fvwm2 . Here the banner is loaded (FvwmBanner , a couple of xterms and an xpmroot is launched. This program simply serves for putting images onto your root windows . For that purpose you can take any program that is capable of writing onto the root window (e. g. xli , xv , etc.). Here a sample with xv : + "I" Exec xv -quit -root -owncmap -maxpect ~/pics/bild13.gif * * 229 15. The windowmanager_-_Window_to_your_machine____________________________________________________________________________________* *_______________________________________@ You should list this background picture also in the sections, which are exe- cuted with the restart of fvwm2 , which means you should insert them within Restart Function . This is located under: ________________________________________________________________________________________________________* *________|| | * * | | ######################################################### * * | | * * | | # # * * | | # restart function # * * | | # common to all wms # * * | | * * | | # # * * | | ######################################################### * * | | * * | |_______________________________________________________________________________________________________* *________ | File contents 15.3.2: InitFunction in /.fvwm2rc Often, the two sections (InitFunction and RestartFunction ) look the same, since they both deal with starting the WM. In the RestartFunction function one normally does not include fvwm's banner. Newer versions of fvwm (Fvwm, Fvwm2, Fvwm95, Bowman and AfterStep in S.u.S.E. Linux5.0 and later) have an additional function ExitFunction (besides the usual InitFunction and RestartFunction ). This function enables you to set programs that have to be started before the WM is launched, or before you exit the WM. So you can (e. g. ) remove a background picture before restarting the WM sets a new one. Colours and fonts The settings for colours and fonts can be found in the following section: ________________________________________________________________________________________________________* *________|| | * * | | ######################################################### * * | | # # * * | | # colors and fonts # * * | | * * | | # # * * | | ######################################################### * * | | * * | |_______________________________________________________________________________________________________* *________ | File contents 15.3.3: Colour and keyboard settings in /.fvwm2rc Here you can do whatever you like. Select those colours that you like best. You can use any installed colour. Which colours are installed mostly de- pends on your graphics card and the colour depth. Press the right mouse button in the root menu and go to the item `system'; here you change to `information'. Far at the end you should see an icon `colours'. Start it and it will show you all colours known to your system. All names of these colours can be found under /usr/X11R6/lib/X11/rgb.txt. A few lines below are lines that resemble the following one: WindowFont -misc-fixed-bold-r-normal-*-13-*-75-75-c-80-iso8859-1 This is the systematic description of a font under X11. Every font is classified in this style. It would lead far beyond the scope of this book to explain all those items. For making life easier there are a couple of well known aliases. Fonts reside (e. g. ) under /usr/X11R6/lib/X11/fonts/misc. In 230 ___________________________________________________________________________________________________15.3.__Fvwm2_settings____* *______________________ this directory there is the file fonts.alias which contains all font aliases for this font directory. Some most common fonts: variable -*-helvetica-bold-r-normal-*-*-120-*-*-*-*-iso8859-1 5x7 -misc-fixed-medium-r-normal--7-70-75-75-c-50-iso8859-1 It's much easier to remember these names : : ::-) Icons Icons are dealt with in the following environment: _______________________________________________________________________________________________________________________|| |||||||| | |||||||| # | |||||||| # others | |||||||| # | |||||||| | |||||||| | |||||||| Style "xterm" Icon Terminal.xpm | |||||||| Style "xosview" NoTitle, Sticky | |||||||| Style "xosview" UsePPosition | |||||||| | ||||||||_______________________________________________________________________________________________________________ | File contents 15.3.4: Icons for certain windows Here you can assign icons to the corresponding applications. You only have to be sure that these icons are in the IconPath (which is set almost at the beginning of ~/.fvwm2rc). In principle you can take any icon that resides under /usr/X11R6/include/X11/3dpixmaps/ (that's an example, there are more). All other icons need their absolute path name. Now, let's assume you want to assign a ghost icon to ghostview . Look in the directory mentioned above and you will see an icon ghostbusters.xpm. Just insert the following line: Style "ghostview" Icon ghostbusters.xpm That's it. You can give icons to almost all applications in the same way. Most applications do have their default icons (remember application defaults?). Anyway, you have to know the exact name of the application6 since an "xTerm" instead of "xterm" would lead to no success. The name of a certain window can be identified from within (`Windowmanager', `Module')by se- lecting `Ident' (Program name FvwmIdent ) and clicking on the window of interest. Cursor Even the shape and colour of the mouse cursor can be set. Here you have the tool xsetroot (which can be used for setting the root window as well in a rather simple way). It is invoked as: newbie@earth: > xsetroot -cursor Here bitmapfile stands for any bitmap file. The bitmaps are located under /usr/X11R6/include/X11/bitmaps/ by default. Select a suitable one or create your own, e. g. with bitmap . 6 ____________________________________________________ To be precise: the exact name of the application since the WM takes care of your windows. * * 231 15. The windowmanager_-_Window_to_your_machine____________________________________________________________________________________* *_______________________________________@ Focus An extremely popular feature of the Fvwm-WM-Family is, that one change the behaviour of the windows while changing the active window while the system is up an running. One can change the focus & raising policy of the windows in the configuration file. With focus we define the properties of the WM, which submit the actual en- tries and mouse clicks to a certain window. Basically, there are 3 possibilities: o You have to click on a window for, i.e., doing key entries to the process active in the window. In English, this behaviour is called Click to focus. It's a widely spread standard and can also be found under Windows or OS/2. o You point with the mouse cursor on a certain window and the window receives automatically the focus. This reaction is called Focus follows mouse, If the mouse Pointer leaves the window, also the focus leaves the window, even in the case the cursor is located over the background or on another window. o A more advanced version is the Sloppy Focus behaviour, which can be configured with Fvwm2 and Fvwm95. In this case, the focus shows the same behaviour as Focus follows mouse, but with the exception that the focus stays with the focused window as long as one does not focus on another window. The focus even stays with window touched if you move the mouse cursor over the background. With the Fvwm2, you can set the focus behaviour individually for every single window. Like many other settings, the configuration of the focus is done with a style-command: Style "*" ClickToFocus adjusts the focus setting ClickToFocus for all windows. The same may be done for the settings SloppyFocus and FocusFollowsMouse , the latter being the default of Fvwm2. Autoraise Autoraising of windows is an interesting feature. Normally the windows keep their position while the focus changes until you click onto the title bar. Autoraise raises the focused windows to the top of the stack. Obviously this only makes sense along with FocusFollowsMouse or SloppyFocus . There is no effect if you use SloppyFocus . For being capable to use AutoRaising, you can either start the Fvwm mod- ule AutoRaising out of the menu (Menu `Windowmanager', `Modules', `AutoRaise On/Off', or, for installing it permanently, you need to insert an entry into Fvwm's configuration file ((~/.fvwm2rc) into the functions InitFunction and RestartFunction : Function InitFunction + "I" Module FvwmAuto 200 Function RestartFunction + "I" Module FvwmAuto 200 232 ____________________________________________________________________15.4.__KDE_-_the_K_Desktop_Environment__________________* *______________________ The value 200 sets the delay to 200 milliseconds, before a window is popped to the foreground. If you don't have a delay set every touch of a window leads to a "movement" of windows which will result in permanent flicker. 15.4 KDE - the K Desktop Environment KDE offers a means of GUI that is easy to configure and standardized. Lots of applications use the Look&Feel of KDE. KDE stands for "K Desktop Environment" and is a project that was founded in 1996. KDE comes along with a windowmanager (kwm), a filemanager (kfm, which plays a central role), a systemwide helpsystem (kdehelp). This helpsystem supports HTML manpages as well as GNU info. Lots of applications are available either (among others these are mailers news readers, games, system info tools and many more). Well, please be aware that KDE is still considered beta (as seen in the ver- sion). It may happen that it crashes. Luckily this doesn't happen all that often : : : KDE is totally + URL based. This in particular means that every path and links to files use a standard form (e. g. links to an HTML page, a file in your filesystem, a help page or an FTP site). Thus it is possible to view different files and formats with the same viewer. Furthermore KDE is capable of lots of Drag&Drop features (e. g. copying a file from an FTP site to your local system). A rather powerful application is the KDE helpdesk. All authors are obliged to provide their help pages as HTML pages. Besides the help system itself (this may be launched from the applications itself via the `Help' button) it is capable of browsing even GNU info pages or UNIX manpages. One of KDE's interesting features (among others) is its integrity. It is rather simple to create icons and place them onto your desktop. This behaviour of KDE is realized by means of the QT widget set. This, similar to Motif, is a set of graphical libraries that enable you to create GUIs on the X Window System. Please have a look at the licence conditions of QT since they differ from GPL (/usr/doc/packages/qt/LICENSE). 15.4.1 General Here we give you a short overview on details of the KDE installation. This refers mainly to path settings location of files and configuration options. By default KDE is written to /opt/kde. Every KDE application (file) may be found below this subdirectory. Setting an environment variable $KDEDIR makes it easier to access them (in /etc/profile). The KDE binary path /opt/kde/bin is automatically added to your $PATH variable when you install KDE. There are many subdirectories under /opt/kde. Here in table 15.1 we only discuss the most important ones. * * 233 15. The windowmanager_-_Window_to_your_machine____________________________________________________________________________________* *_______________________________________@ ________________________________________________________________________________________________________* *________ /opt/kde/bin every KDE programs (binaries) /opt/kde/share/config systemwide configuration files /opt/kde/share/applnk Appliaction links (Menu) /opt/kde/share/apps Files of KDE programs /opt/kde/share/doc the online help system /usr/doc/packages/kde additional information on KDE ________________________________________________________________________________________________________* *________ Table 15.1: KDE - important directories 15.4.2 kdm - graphical login The KDM displaymanager that comes along with the KDE system is a nice enhancement or your Linux system. This tool replaces the default xdm which realizes a graphical login. KDM's default configuration with S.u.S.E. Linux is shown in figure 15.5. Figure 15.5: The display manager kdm There are buttons for selecting the desired window manager (`Sessiontype') or the language (`Language'). An interesting feature (especially for stan- dalone workstations) is the ability to shutdown the machine via the `Shutdown' button. The kdm may be adapted to your needs via either a configuration file in /opt/kde/share/config/kdmrc or (since Beta3) from a configuration di- alog which may be launched from the KDE menu. S.u.S.E. Linux provides another enhanced technique. This refers to the con- figuration of the WMs themselves and the startup mechanism of the kdm. Here you should create two variables:7 o DISPLAYMANAGER Assigns whether the user wants to log into a text console (console or empty string this is "runlevel 2") or via kdm or xdm respectively ("runlevel ______3".____________________________________ 7 These variables are described in page 174 234 ____________________________________________________________________15.4.__KDE_-_the_K_Desktop_Environment__________________* *______________________ o KDM__SHUTDOWN Assigns which user is permitted to shutdown the machine from within kdm. Here you may enter either `root' (root ) every user (all ), nobody (none ) or a user on the local machine (local ). These values are written to /opt/kde/share/config/kdmrc by SuSEconfig and are available with the next start of kdm (you may want to restart the server by entering Ctrl Alt 7 !). If you plan to make changes your- self please make them in kdmrc.in as /opt/kde/share/config/kdmrc is created from this file by SuSEconfig . _______________________________________________________________________________________________________________________|| |||||||| If you want to get rid of this mechanism just delete or remove /opt/ ||| | |||||||| kde/share/config/kdmrc.in. SuSEconfig will not touch your kdmrc || | |||||||| || | |||||||| anymore || | ||||||||_______________________________________________________________________________________________________________||_ 15.4.3 So, what's so special with KDE? Now that we are able to use kdm's graphical login it is time to go into the details of KDE. We just describe the behaviour of KDE after you have started the server by entering startx or via the kdm. Shortly after the actual login a couple of dialog windows appear announcing the missing of certain files and directories. This is not an error. KDE creates similar to the tree under /opt/kde its configuration directories in your home directory. These in particular are: $HOME/.kde and $HOME/Desktop. The first is for configuration file and the latter for links. Thus you may just hit _- on the dialog boxes. They won't appear at the next KDE session. Thereafter the kfm (K filemanager) is launched. The kfm is a basic com- ponent of the KDE system. As mentioned above it is able to handle lots of different file formats. The behaviour is totally transparent to the user; be it a archive from an FTP site(URL prefix: ftp:) or a manpage (URL prefix man:), a GNU info page (URL prefix info:) or a HTML page from a WWW site (URL prefix http:). It is even possible to view images (via an external tool). A typical KDE desktop is shown in figure 15.6. The user will get familiar with these features within minutes. Clicking on a symbol with the right mouse button opens a configuration tool for this application. Every setting of the program itself is written to a file with the .kdelnk ex- tension. Changing of settings is done via an property window as shown in figure 15.7. This file is an editable ASCII file. It is recommended to have a look at these files that are located under $HOME/Desktop (or un- der /opt/kde/share/applnk for the systemwide configuration files). The contents are standard to every KDE applications. So if you know how to write one of them you are set. The last program to be launched is KPanel the menu bar. Here you may find some elements for the virtual desktops. A special one is the "K" button. Here all available KDE applications are grouped together (see kdelnk files!). Here you may also find the KDE system settings via the KDE control centre and many more. * * 235 15. The windowmanager_-_Window_to_your_machine____________________________________________________________________________________* *_______________________________________@ Figure 15.6: The filemanager kfm Well, that's not all KPanel has to offer. You may copy application icons via Drag&Drop onto it and launch them directly from the panel. Moreover KPanel may "swallow" applications. These might be system information applications such as the kwmpager or an xosview . These now run within the KPanel. Another one of KDE's programs the kdisplay is shown in figure 15.8. Of course we cannot list each and every KDE application in this manual. One of KDE's aims is to make the machine more usable and cooperative. Please have a look around at the various applications that come along with KDE. There is help available with most of KDE's applications. You may reach the kdehelp by clicking onto the background (right mouse button). But please keep in mind that some of the KDE packages are still considered alpha. This in particular means that they may crash or other strange things may happen. In general this should not happen anymore. Due to the complexity of KDE and the steady development theses packages are not yet included in our installation support. Anyway we try to provide you with as much information as we can gather on our Support Database. Furthermore there are a couple of mailing lists available on the KDE WWW site (http://www.kde.org). Here you may as well find actual information on many topics as well as the latest release of KDE (and applications). We at S.u.S.E. provide you with specially adapted KDE packages that may be downloaded from our FTP site ftp://ftp.suse.com/pub/suse_update/ KDE. These packages may be installed by YaST. Please have a look at the README files that are located on the same URL. 15.5 Configuring the windowmanager using susewm 236 _________________________________________________15.5.__Configuring_the_windowmanager_using_susewm__________________________* *______________________ Figure 15.7: The property dialog What is susewm ? susewm serves for making the handling of the supported WMs (Fvwm , Fvwm2 , Fvwm95 , Bowman , AfterStep (afterstep ), Ctwm und Mwm 8 ) easier9 , kwm . Since fvwm , bowman and afterstep resp. fvwm2 and fvwm95 are mainly based on the same WM where fvwm2 is the descendant of fvwm , those five window manager are mainly configured the same way and supply almost identical features. On the other hand there are considerable differences in configuring those WM's. For letting the user benefit of all common tasks of those WM without having to maintain five different configuration files, susewm combines their configuration via a rather abstract macro language. 8 ____________________________________________________ MWM is part of the commercial Metrolink Motif or Metrolink Motif Runtime Libraries respectively. 9 The pseudo-WM CDEsim (cdesim ) should be excluded here. More information on this topic can be found in package cdesim, series xwm, in directory /usr/doc/packages/cdesim. * * 237 15. The windowmanager_-_Window_to_your_machine____________________________________________________________________________________* *_______________________________________@ Figure 15.8: Setting up your display with kdisplay And even more, the totally different WM ctwm ,mwm and kwm are config- ured with susewm as well. However this only covers the automatically gen- erated menu entries (more in the sections below). Differences between those single WM's can be taken into consideration through WM specific statements. One of the most interesting features susewm offers is that it creates menus and supported modules depending on the installed software packages. There are no menu entries for packages which are not installed. If you select a menu item you can be sure that the corresponding program is installed and can be launched. Here are some more of susewm 's features: o Configuration of eight windowmanagers: Fvwm , Fvwm2 , Fvwm95 , Bowman , Afterstep , Ctwm , Mwm , kwm o Central library for all WM's; unified macros for different configuration files. o Additive configuration files for taking specialities and peculiarities of the WM's into consideration. o Loading of additional source files10 in your system via /etc/rc.config in automatic configuration. 1_____________________________________________0 Source files are WM configuration files in general. For using them with susewm , files in susewm - format are particular suitable 238 _________________________________________________15.5.__Configuring_the_windowmanager_using_susewm__________________________* *______________________ o Considering source files of the respective user (in HOME directory) dur- ing automatic configuration. o Creation of user-specific configuration files respecting almost any changes the user has made (in older configuration files). o Unified commands for all eight WM's. o Integrating of widely used commercial applications, which are not part of S.u.S.E. Linux, while generating menus. o Switching to and fro between windowmanagers (only for the supported ones) without having to change, e. g. , startup scripts. o Clearly designed m4 macros. Technically susewm is a set of several shell scripts and m4-macros as well as a database. Details are not covered on this section. If you to know more about susewm , please refer to the online-documentation, like described in section 15.5. That's how to use susewm susewm is used in two different cases: o You don't have a configuration of your own file yet but want to adapt your desktop accordingly. o You already have a configuration file but software has been installed using YaST and the menus are no longer up to date. Requirements Here the following is assumed to be installed: o Your Linux system and the X Window System XFree86TM are installed, configured properly and running. o You use the fvwm2 WM (which is the default with S.u.S.E. Linux), but perhaps have another supported WM installed. o susewm is installed on your system (standardsettings). o You are logged in as normal user (e. g. the sample user defined with the standard installation of YaST), not as `root'. o You have set the language to be English . The same applies for german menus. This is how it is done. If you click onto the root window (which actually is the "background") with your left mouse button a menu appears. This is called the `Working menu'. Here the most often used programs needed during your work life are listed (depends on the user actually). The last entry of this menu (`Window manager') contains a submenu with items on functions and configurations of the actual running WM. If you want to configure another WM just switch to it using menu item `Other window managers' and configure it accordingly in the `Configuration' item: You'll find several different items: * * 239 15. The windowmanager_-_Window_to_your_machine____________________________________________________________________________________* *_______________________________________@ o The items of the first window are for reading the documentation on susewm . There you can find lots of details that are not handled in this section. o The items of the second window allow you to change the configuration file of your PC with an + editor o The items of the third window help in creating a user specific configura- tion file in German. o Also located in the third window is an item for creating a user specific configuration file in English. This|can|be seen in figure 15.9 with the `Working menu' of fvwm2 * * || ________________________________________________________________________________________________________* *________|||| | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | |_______________________________________________________________________________________________________* *________ | Figure 15.9: Menus for WM configuration If you select the second or third item a window with mksusewmrc (this pro- gram is a part of susewm ) appears. Just read the text and press the appropriate buttons if asked for it. That's it. Now you can change and edit this freshly created config file in your HOME directory. susewm will take over almost everything in the new configuration file. ________________________________________________________________________________________________________* *________|| | If you want to change the settings of the desktops slightly you have to * * |||| | * * ||| | create an configuration file of your own first. How this is done is mentioned * * ||| | above. Only after doing so, you can change the file. If you plan to write a * * ||| | * * ||| | completely different configuration file, don't it by using susewm . * * ||| |_______________________________________________________________________________________________________* *________|||_ 240 ______________________________________________________________________________________________________________15.6.__General* *______________________ An example A coloured desktop is nice, but what's the use if the hardware (memory, pro- cessor, hard disk) is not fast enough? It's not the point of a WM to use up all resources for animating icons and more.A hint what to do if you think that your machine is too slow if you start up fvwm95 or fvwm2 . Most prob- lems are due to the hundreds of icons that are loaded. There are now two possibilities: o Deinstall the package package 3dpixms, or package 3dpixm if you don't want to have the big icons either. o Set in the YaST menu `System Administration', submenu `Change the config file' the variable SUSEWM__XPM=no , which results in the same effect as mentioned before. In case you have changed the variable directly in the file /etc/rc.config, please don't forget after doing so to start SuSEconfig . o Remove icons from configuration file. The first way is rather simple. Just start YaST, deinstall the mentioned pack- ages and feel fine. YaST in conclusion with susewm ensures that the system wide WM configuration file is adapted. If you have a user specific configu- ration file you have to adapt it explicitly. Use the WM menu as described in section 15.5. The second way doesn't need to be commented further. The third way: Let susewm create a configuration file for you as described in section 15.5. Then you have to replace all entries that contain icon statements: _______________________________________________________________________________________________________________________|| |||||||| | |||||||| AddToMenu thiswmpopup "Fvwm2" Title | |||||||| | |||||||| | |||||||| + "Other windowmanagerr%small.warning_3d.xpm%" Popup otherwmpopup | |||||||| + "Configuration%small.checklist2_3d.xpm%" Popup susewmpopup | |||||||| + "" Nop | |||||||| + "Fvwm2 Restartt%small.restart_suse_3d.xpm%" Restart fvwm2 | |||||||| | |||||||| + "Exit Fvwm2 and%small.exit.xpm%" Function QuitSave | |||||||| | |||||||| # end popup thiswmpopup | |||||||| | ||||||||_______________________________________________________________________________________________________________ | File contents 15.5.1: .fvwm2rc with icons for menu entries will become: Now fvwm95 or fvwm2 should start even faster. And, as mentioned before, all personal changes you have applied to your configuration file should persist even if you invoke susewm a second time. 15.6 General As promised above we are now going to tailor the look and functions of your personal desktop - why buy something predefined if we can tailor it ourselves. First some general information about configuring the desktop - there are two places where you can fiddle: * * 241 15. The windowmanager_-_Window_to_your_machine____________________________________________________________________________________* *_______________________________________@ ________________________________________________________________________________________________________* *________|| | * * | | AddToMenu thiswmpopup "Fvwm2" Title * * | | * * | | * * | | + "Other window manager" Popup otherwmpopup * * | | + "Configuration" Popup susewmpopup * * | | + "" Nop * * | | + "Fvwm2 Restart" Restart fvwm2 * * | | * * | | + "Exit Fvwm2 and X" Function QuitSave * * | | * * | | # end popup thiswmpopup * * | | * * | |_______________________________________________________________________________________________________* *________ | File contents 15.5.2: .fvwm2rc without icons for menu entries o The pre-seetings for the applications of the X Window System o The WM configuration file(s) as already mentioned in section 15.3 15.6.1 X11 defaults for applications Global settings Almost any application under X11 has their own default configuration. This configuration will be copied to /usr/X11R6/lib/X11/app-defaults at installation time.11 . Here files as Xarchie can be found. As the name already tells us this must be the central configuration file for the application xarchie . Just have a look at this file (e. g. with less Xarchie ). Here you can find (among others) the following lines: Xarchie.color*background: powder blue Don't bother with the sometimes cryptical names; you don't have to under- stand them right now. Every program under X11 is built out of "Widgets"12 . Here there is a main widget, which is the root window (it is called first). All other programs are child widgets of this parent widget (more or less). That means that every child widget has got exact one parent widget and one or more child widgets. Each of these widgets can be addressed with an unique name. Windows and widgets should not be confused. A scrollbar, e. g. , is a window of its own (a window which lacks any decoration). A widget can contain a scrollbar, a text field and more. That leads to the conclusion that any window and its resources can be ad- dressed by a unique name. For our example this means: o The first word to the first dot (Xarchie ) is the name of the toplevel wid- gets of the xarchie program (it is a rule to write names of application defaults beginning with an uppercase letter). o After the dot there is a `color'. This, of course, is the colour. But which colour? o Then there is a `.'. This dot could have been an asterisk (`*' as well). - A `.' means that this is a direct hierarchy. 1_____________________________________________1 It is quite helpful to assign an alias to such abstruse and long names 12 Think of a "Widget" as a sort of "brick"; it's a shortage for "windows" and "gadget" 242 ______________________________________________________________________________________________________________15.6.__General* *______________________ - An `*' is a wild card. It indicates that between those two windows may lay 1 or more additional windows. o The word "background" now shows us which colour is going to be set. Here you can enter a defined colour. User specific settings Obviously there is the possibility for every user to make specific settings. For that reason you have a ~/.Xresources file located in the $HOME directory. The `.' at the beginning means that this is a hidden file. In this file, all user specific settings are done. For example you can define now that every window should have a yellow background except the root window which should be red. Coming back to the example given above this means that you can over- write system wide settings (app-defaults files) with those in your ~/ .Xresources file. So, if you set the following in your ~/.Xresources file: Xarchie.color*background: gold the application xarchie is started with a golden background. Of course you can not only change colours but almost any decoration or setting of your windows. A rather useful program in this context is editres (editres stands for EDitRESources). This little tool lets you show all resources of a given program and to change them at runtime (so you can see what you've done). Here some examples of what you can set: Xarchie.color*background: powder blue Xarchie.color*SimpleMenu*background: wheat Xarchie.color*Command*background: wheat Xarchie.color*MenuButton*background: wheat Xarchie.color*Text*background: wheat Setting start parameters There is a third alternative of influencing the application. You can give pa- rameter to it at start time. Of course you can enter those parameters in your WM's configuration file if you start them from there. For example you can explicitly launch a program with another font and back- ground colour (bg = background, fg = foreground): xterm -bg darkblue -fg white You get a blue xterm with white letters. How do the configuration possibilities "work"? Principally, the systemwide settings are set at the start of the X Window System. The settings themselves are administrated by the X-server in the X Resource Database, xrdb. If you want that the changed settings become effective systemwide, it is necessary the read in the resource files once over manually. This can be done with the command: The various settings for applications are handled in the following order during the build-up of a an intern X-server Resource Database at the X-server start: * * 243 15. The windowmanager_-_Window_to_your_machine____________________________________________________________________________________* *_______________________________________@ Hierarchy of configurations o First all system wide settings from /usr/X11R6/lib/X11/app-defaults are read. o If you have special settings in your ~/.Xresources file these will over- write the settings above. o If you start an application with certain parameters these settings overwrite the last one. They have the highest priority. ________________________________________________________________________________________________________* *________ | In case this explicit changes have no influence of the behaviour of the * * |||| | * * ||| | application it might help to vary capitalization of the respective widget * * ||| | name * * ||| |_______________________________________________________________________________________________________* *________|||_ 244 Chapter 16 Networking Linux 16.1 Preliminary words : : : In the age of communication the number of linked machines grows day by day. Networking becomes a "way of life". Either on a local + network (+ LAN), on the + internet or simply connected to a mailbox; one has to commu- nicate. + electronic mail , news, and participating in world wide discussion forums now becomes a must for all who want to be up to date. Linux, being a real child of the internet, offers you all preconditions which are necessary to connect to the multiple networks. All common networking tools are available under Linux thus letting Linux-machines participate in every network. Here we will give you an overview of all tools needed for installing and maintaining your network connections; the most important configuration files will be presented and discussed and some of the more important tools are shown. First of all we want to show how a Linux-machine can be integrated in an existing + LAN, or how to create a network of Linux-machines respectively. Thereafter connections to other hosts via modems are explained. We intro- duce UUCP, show how you can install an internet connectivity via SLIP or PPP; furthermore configuration of mail and news system is discussed thor- oughly. Almost everything to do with networks can be done by YaST. (see section 4.7, page 82 and section 12.6, page 174); on the other hand configuration of a net- work can become very complex, that's why we only show basic mechanisms and all relevant files. 16.1.1 Good neighbourhood - connecting to a LAN Connecting a Linux-machine to other Unix-machines (of course there can be other Linux-machines among them as well) via a LAN doesn't impose a problem. There have to be (some) preconditions given, which do not restrict the use of a Linux-machine in a networking environment. 16.1.2 Requirements, preliminary work At the meantime Linux supports almost any networking cards (Ethernet, Arc- net, TokenRing) and knows almost any known networking protocols (TCP/IP, AppleTalk, IPX). Taking every possible configuration into consideration is, * * 245 16. Networking Linux______________________________________________________________________________________________________________* *_______________________________________@ of course, far beyond the aim of this chapter, therefore we will discuss the generic case (integrating Linux into a TCP/IP network via an ethernet card). The latest information about other networks is to be found under Documentation at your kernel sources directory; moreover the kernel's help- function supplies you with very useful additional information. The following requirements have to be set: o Your machine has to have a supported card; whether your card has been started correctly is to be seen through the following command: root@earth:/ # cat /proc/net/dev There should be a line beginning with eth0: . o The kernel has to be configured for the net to be used correctly (see sec- tion 13.1, page 181 for more details). ________________________________________________________________________________________________________* *________||| | If the kernel support is realized by modules (see section 13.2, page 181) * * |||| | * * ||| | (all S.u.S.E. kernels are configured that way) the name of the module has * * ||| | to be entered in /etc/conf.modules. For your first Ethernet card this * * ||| | * * ||| | might look like: * * ||| | alias eth0 tulip * * ||| | * * ||| | This is done automatically if you selected the network module in Linuxrc * * ||| | at the installation. Later this may be done with YaST. (see section 4.7.1, * * ||| | * * ||| | page 82). * * ||| |_______________________________________________________________________________________________________* *________|||_ If those preconditions are given there are some things to be thought of before starting configuring your machine: _______________________________________________________________________________________________________* *_________ Hostname The name of your machine in the network. Hostname should not exceed 8 characters and should not already be used in the local net. Domainname Name of the domain your machine belongs to. Do- mains serve for structuring big networks (e. g. the internet). A host is addressed via its full name which is assembled of hostname and domainname. Gauss.Suse.de is the machine Gauss in the domain Suse.de . _______________________________________________________________________________________________________* *_________ Table 16.1: to be continued... 246 _____________________________________________________________________________________16.2.__Configuring_using_YaST__________* *______________________ ______________________________________________________________________________________________________________________ IP-address Address of your machine in the net. Every machine at least has got one IP-address . This address is a 32 bit word, and in most times is given as a quadruple (e. g. 193.141.17.1 ). If you choose an IP-address you should consider whether you plan to connect to the internet in the near future. If this is true it is recommended to use regis- tered IP-addresses from the very start. If you only have a private local network there are some address ranges defined by the pseudo- standard RFC1597 which ensures you that even if one of those addresses "escapes" to the inter- net it won't hurt anybody or cause any damage. 10.0.0.0 ) 10.255.255.255 (Class A) 172.16.0.0 ) 172.31.255.255 (Class B) 192.168.0.0 ) 192.168.255.255 (Class C) Gateway ad- If there is a Gateway computer in your network dress (i. e.machine that is connected to more than one net- work and which transports packages from one net- work to the other) you can enter the appropriate ad- dress while configuring your network. Netmask By means of the netmask it is clear which network the machine belongs to. The IP-address is added to the netmask via a logical AND thus the "host" part is cut out and gives the network's address as result. Address of the Nameservers provide DNS (DomainNameService) nameserver for turning hostnames into IP-addresses. If there is a reachable nameserver on the net and you want to use it by default its IP-address can be entered while configuring the network. ______________________________________________________________________________________________________________________ Table 16.1: Values for network configuration 16.2 Configuring using YaST If all points from section 16.1.2, page 245 are fulfilled you can start configur- ing your network via YaST. Select menu item `System administration' `Network configuration'. YaST first asks you about the hostname and the domainname. Thereafter it wants to know whether only the loopback-device should be used. If this is true all other questions are irrelevant, but this also means that you can't use the machine in a network! After having entered the IP-address you are asked for a gateway address. If there isn't any gateway installed in your network you should enter your own * * 247 16. Networking Linux______________________________________________________________________________________________________________* *_______________________________________@ IP-address. Next the netmask has to be set. For a class-C network (up to 254 machines in a subnet) this typically is 255.255.255.0 . After all that you can enter the address of a nameserver, and if it exists enter the appropriate address. Moreover you can enter a lot of settings directly via our central configuration file; even here YaST supports you (see section 12.6, page 174). So far for network configuration, really that's all. YaST enters all given settings into the corresponding files (see section 16.3) and quits thereafter. For the changes to take effect you have to reconfigure the changed files and restart the daemons. This can easily be done by: root@earth:/ # /sbin/init.d/network stop root@earth:/ # /sbin/init.d/network start (see chapter 12, page 169). 16.3 Manual network configuration - where do I find what? You should use YaST for configuring your network but since (unfortunately) it doesn't cover all parts of networking configuration yet, there might be some work to be done by hand. Generally all settings should be done in /etc/rc.config (see section 12.5, page 173). If you do change this file using YaST you don't have to bother. If you change this file manually don't forget to launch SuSEconfig each time you've changed it for the configuration changes to take effect! 16.3.1 Configuration files This section gives you an overview of the most important configuration files for your network and roughly explains their functions and format. /etc/rc.config In this central configuration file almost everything concerning your net- work can be set. After having changed it with YaST or after having launched SuSEconfig most of the necessary files are generated automati- cally. Even boot-scripts are built via settings in this file. /etc/hosts Here machines (see file contents 16.3.1, page 249) are assigned IP- addresses. If no nameserver is used every machine you want to be able to reach has to be set. There is one line per machine. This line consists of the IP-address, the fully qualified host name (full name), and the unof- ficial name (nickname). The IP-address has to be at the beginning of the line, entries are separated by blanks, or tab-stops respectively. Comments are started by a `#'. /etc/networks Here network names are converted to network addresses. The format of this file resembles hosts file; here in fact network names are put in front of the addresses (see file contents 16.3.2, page 249): 248 ______________________________________16.3.__Manual_network_configuration_-_where_do_I_find_what?___________________________* *______________________ _______________________________________________________________________________________________________________________|| |||||||| | |||||||| # | |||||||| # hosts This file describes a number of hostname-to-address | |||||||| | |||||||| # mappings for the TCP/IP subsystem. It is mostly | |||||||| # used at boot time, when no name servers are running. | |||||||| # On small systems, this file can be used instead of a | |||||||| # "named" name server. Just add the names, addresses | |||||||| | |||||||| # and any aliases to this file... | |||||||| # | |||||||| 127.0.0.1 localhost | |||||||| 193.141.17.1 gauss.suse.de gauss | |||||||| | |||||||| 193.141.17.42 sofa.suse.de sofa | |||||||| # End of hosts | |||||||| | ||||||||_______________________________________________________________________________________________________________ | File contents 16.3.1: /etc/hosts _______________________________________________________________________________________________________________________|| |||||||| | |||||||| # | |||||||| # networks This file describes a number of netname-to-address | |||||||| | |||||||| # mappings for the TCP/IP subsystem. It is mostly | |||||||| # used at boot time, when no name servers are running. | |||||||| # | |||||||| loopback 127.0.0.0 | |||||||| | |||||||| localnet 193.141.17.0 | |||||||| # End of networks. | |||||||| | ||||||||_______________________________________________________________________________________________________________ | File contents 16.3.2: /etc/networks /etc/host.conf This file is for resolving host names. Here you enter hostnames (or net- works); resolving itself is down by the resolver library. Comments begin with an `#'. The following parameters can be set: _____________________________________________________________________________________________________________ order bind hosts Order in which services for resolving a host name are called. Possible arguments are: bind: Using a nameserver hosts: Searching /etc/hosts multi on off Determines if a machine in /etc/hosts is allowed to have multiple IP-addresses. nospoof on alert on off Just influences spoofing of the nameserver whithout any other consequences. _____________________________________________________________________________________________________________ Table 16.2: to be continued... * * 249 16. Networking Linux______________________________________________________________________________________________________________* *_______________________________________@ ____________________________________________________________________________________________________* *______ trim The given domain name is cut off from its hostname before being resolved (provided this machinename contains the given do- mainname). This might come in handy if in /etc/hosts there are only local domains which should be recognized even with the full domainname attached. ____________________________________________________________________________________________________* *______ Table 16.2: Parameter fu"r /etc/host.conf An example for /etc/host.conf can be seen under file contents 16.3.3, page 250. ________________________________________________________________________________________________________* *________|| | * * | | # * * | | # /etc/host.conf * * | | * * | | # * * | | # We have named running * * | | * * | | order hosts bind * * | | # Allow multiple addrs * * | | multi on * * | | * * | | # End of host.conf * * | | * * | |_______________________________________________________________________________________________________* *________ | File contents 16.3.3: /etc/host.conf /etc/resolv.conf Same as /etc/host.conf this file plays a vital role for resolving ma- chine names via the resolver-library. Here the domain for a given machine is set (keyword search ) and how the address of the nameserver is to be called. There can be several entries for domainnames in here.1 If a not fully qualified name is resolved entries in search are appended one after the other to create a fully qualified name. Several nameservers can be given here, each in a line of its own; those entries have to start with nameserver . Comments are entered as usual using `#'. YaST (see section 16.2, page 247) enters the given nameserver here! /etc/HOSTNAME Here the complete name of the machine is given including its domain- name (this is called the fully qualified domain name. This file is read by a couple of scripts at startup time. It should only contain one line with the machine's name! This file is also automatically generated via settings in /etc/rc.config. _____________________________________________1 The more entries there are the slower resolving of a name will be! 250 ______________________________________16.3.__Manual_network_configuration_-_where_do_I_find_what?___________________________* *______________________ _______________________________________________________________________________________________________________________|||| |||||||| | ||||||||| ## /etc/resolv.conf || |||||||| | |||||||| # Our domain | |||||||| search suse.de | |||||||| | |||||||| # | |||||||| # We use Gauss (193.141.17.1) as nameserver | |||||||| nameserver 193.141.17.1 | |||||||| | |||||||| # End of resolv.conf | |||||||| | ||||||||_______________________________________________________________________________________________________________ | File contents 16.3.4: /etc/resolv.conf 16.3.2 Startup scripts Besides those described configuration files there are a couple of scripts that start networking programs at boot-time. These scripts are started as soon as the machine switches to one of the multiuser-runlevels. ______________________________________________________________________________________________________________________ /sbin/init.d/network This script is responsible for configuring your (network) hard- and software at boot time; it also evaluates IP-address, network address, netmask, and gateway given in /etc/rc.config (created by YaST see section 16.2, page 247). /sbin/init.d/inetd Starts inetd if defined in /etc/rc. config. This is necessary if you want to login into this machine via network. /sbin/init.d/rpc Serves for starting the various RPC- servers which are needed (e. g. ) if filesys- tems are exported to other machines via NFS (NFS-server). /sbin/init.d/sendmail Controls sendmail process in dependency of settings in /etc/rc.config. ______________________________________________________________________________________________________________________ Table 16.3: The network startup scripts * * 251 16. Networking Linux______________________________________________________________________________________________________________* *_______________________________________@ 16.4 Routing under S.u.S.E. Linux 16.4.1 Preliminaries Setting of the routing table under S.u.S.E. Linux is not done via variables in the central configuration file /etc/rc.config but via a special script in /sbin/init.d. and another configuration file in /etc. After the network has been initialized by the boot-scripts under /sbin/init.d/ network, i4l_hardware and possibly additional boot- scripts /etc/route.conf are searched by /sbin/init.d/route routing-table. Then this table is configured for the system. Any static routes that are needed may be added to /etc/route.conf: Routes to a host, routes to a host via a gateway and routes to a network. Another possibility is to use the program routed . Well, this configuration is a bit more tricky. Here we refer to the manpage of routed . 16.4.2 How to commence and use it The rules that apply for /etc/route.conf are adapted from the output of route . If route is called without parameters the routing table in use is presented. Except for Flags, Metric, Ref und Use the entries in /etc/ route.conf are identical. Here we give you the rules that apply for /etc/route.conf: o Lines beginning with # or blank lines are regarded as comments. An entry consists of one line and 2 up to 4 columns. o The first column gives the target of a route. Here the IP address of a host, a network, or a reachable nameserver may be given. Even the full name is allowed (Fully Qualified Domain Name). o The keyword default marks the default gateway. Please do not use 0.0.0.0. o The second columns either contains a separator (0.0.0.0) or the IP ad- dress (or the FQDN) of a host. This host may be the default gateway or a gateway behind a host or network. o The third column serves for entering the netmask for networks or hosts behind a gateway. For hosts behind a gateway this is 255.255.255.255. o The last column only is important for networks (loopback, ethernet, ISDN, PPP, Dummy-device : : : ) connected to the local machine. Here the device has to be given. A simple example of /etc/route.conf is given in figure 16.4.1. If new entries are added to /etc/route.conf entering root@earth:/ > /sbin/init.d/route stop and thereafter root@earth:/ > /sbin/init.d/route start sets the routing table with the new entries. 252 ________________________________________________________________________________16.5.__NIS,_yellow_pages_in_a_LAN___________* *______________________ |||||||| | _______________________________________________________________________________________________________________________|| |||||||| | |||||||| # Destination Dummy/Gateway Netmask Device | ||||||||| # || |||||||| # Net devices | |||||||| # | |||||||| 127.0.0.0 0.0.0.0 255.255.255.0 lo | |||||||| | |||||||| 204.127.235.0 0.0.0.0 255.255.255.0 eth0 | |||||||| # | |||||||| # Gateway | |||||||| # | |||||||| | |||||||| default 204.127.235.41 | |||||||| # | |||||||| # Host behind Gateway | |||||||| # | |||||||| | |||||||| 207.68.156.51 207.68.145.45 255.255.255.255 | |||||||| # | |||||||| # Net behind a Gateway | |||||||| # | |||||||| | |||||||| 192.168.0.0 207.68.156.51 255.255.0.0 | |||||||| | ||||||||_______________________________________________________________________________________________________________ | File contents 16.4.1: Simple example of /etc/route.conf 16.5 NIS, yellow pages in a LAN 16.5.1 NIS, what it is As soon as multiple Unix-systems in a network want to access common re- sources you have to make sure that all users, and groups are identical. The network should be transparent to the user; no matter where he/she is working he will always find exactly the same environment. This is possible via NIS and NFS services. NFS serves for distributing filesystems over a network and is discussed in section 16.6, page 255. NIS (Network Information Service) 2 is considered to be a database service which enables access to /etc/passwd,/etc/shadow and /etc/groups over the net. NIS can be used for other, more specialised tasks (e. g. fu"r /etc/hosts oder /etc/services)which are not mentioned here. 16.5.2 Installing a NIS-client S.u.S.E. Linux contains all necessary packages for installing an NIS-client. These tools are bundled in package ypclient, series n. For installing an NIS-client you should proceed as follows: o Setting the NIS-domain at startup time. All you have to do is to set YP__DOMAINNAME in /etc/rc.config. If switching to a (networking) runlevel /sbin/init.d/network evaluates those settings and sets the name accordingly. NIS-domainname should not be confused with DNS-domainname ; They are not the same even if they do have the same name! 2 ____________________________________________________ NIS is commonly referred to as YP; this comes from "yellow pages", the "yellow pages" in the net * * 253 16. Networking Linux______________________________________________________________________________________________________________* *_______________________________________@ o Assigning the NIS-servers. The NIS-server is set via /etc/rc.config in the variable YP__SERVER . SuSEconfig now writes the correct values to /etc/yp.conf (vgl. file contents 16.5.1, page 254) If you have setup this variable with YaST this step is performed automatically. In this file there has to be a line starting with ypserver followed by the name of the NIS-server. An example of /etc/yp.conf can be seen in ________________________________________________________________________________________________________* *________|||| | * * | | # * * | | # yp.conf * * | | # * * | | * * | | # Legal entries are: * * | | # * * | | * * | | # ypserver Define which host to contact * * | | # for YP service. * * | | * * | | # * * | | ypserver galois.suse.de * * | | * * | | # End of yp.conf * * | | * * | |_______________________________________________________________________________________________________* *________ | File contents 16.5.1: /etc/yp.conf o helios.cosmos.com and heliosII.cosmos.com have to be set in /etc/ hosts for being resolved. o Making sure the RPC-portmapper is started. NIS is realized via RPC (Remote Procedure Calls) therefore starting of the RPC-portmapper is obligatory. This server is started by /sbin/init.d/rpc. This as well is done automatically if you have configured it in /etc/rc.config. o Completing entries in /etc/passwd and /etc/groups . For asking the NIS-server about those files after having searched the lo- cal ones, a line only containing an `+' has to be added. NIS-allows you to set a couple of other informations (e. g. netgroups or local overwriting of NIS-entries. Info about those settings is to be found under the corre- sponding README file3 o Starting ypbind . The last step for activating the NIS-server is to launch ypbind; this in fact really starts the NIS-client4 . Even this program is launched automatically if you have configured your network with YaST. o Activating your changes. Either you restart your system or you enter: root@earth:/ # /sbin/init.d/network stop root@earth:/ # /sbin/init.d/network start 16.5.3 NIS master and slave server You need to install package ypserver, series n. How to proceed is given in /usr/doc/packages/yp/HOWTO. _____________________________________________3 /usr/doc/packages/ypclient/yp-clients-2.2 4 Strictly speaking this is no longer necessary but it guarantees the reconnect if, e. g., the NIS- server has been booted 254 ______________________________________________________________________________16.6.__NFS_-_distributed_filesystems__________* *______________________ 16.6 NFS - distributed filesystems As mentioned in chapter section 16.5, page 253 NFS serves (together with NIS) for making a network transparent to the user. Via NFS it is possible to distribute filesystems over the network. It doesn't matter at which terminal the user is logged in; he will always find the same environment. Same as NIS, NFS is an asymmetric service. There are NFS-servers and NFS-clients. Well, a machine can be both of them (it can supply filesystems over the network (export) and mount filesystems from other hosts (import)). Generally there are servers with enormous HD capacity whose filesystems are mounted by other clients. 16.6.1 Importing filesystems Importing filesystems from an NFS-server is rather simple. The only thing required is that the RPC-portmapper has been started. Starting of this server has already been explained in section 16.5, page 253 in connection with NIS (see page section 16.5.2, page 254). If this is the case other filesystems can be mounted (as long as they are exported by the server) just as easily as local filesystems using the program mount . This is the syntax: mount -t nfs : So, if user directories from machine Gauss.suse.de (e. g. ) should be im- ported this can be achieved via the following command: root@earth:/ # mount -t nfs helios:/home /home 16.6.2 Exporting filesystems A machine that exports filesystems is called an NFS-server . On an NFS- server there are a couple of tools that need to be started: o RPC-portmapper (rpc.portmap) o RPC-mount-daemon (rpc.mountd) o RPC-NFS-daemon (rpc.nfsd) Those are started by /sbin/init.d/rpc at startup time. Starting of the RPC- portmapper has already been discussed in section 16.5, page 253. Besides starting those daemons it has to be decided which filesystems should be exported. This is done in /etc/exports . One line for each filesystem that is going to be exported is needed; here you can enter which machine may import this filesystem and how (all subdirec- tories of an exported filesystem are exported as well!). All entitled machines are usually given with their full name (including domainname) but it is possi- ble to use wildcards like `*' or `?' as well. If no machine is given here any machine is allowed to import this filesystem with the given permissions. Permissions of the filesystem to be exported are given in brackets after the machine name. The most important options are: * * 255 16. Networking Linux______________________________________________________________________________________________________________* *_______________________________________@ _______________________________________________________________________________________________________* *_________ ro Filesystem is only exported with read-permissions (de- fault). rw Filesystem is exported with read-write permissions. root__squash This makes sure that the user `root' of the given machine doesn't have `root' specific permissions on this filesystem. This is achieved by assigning user- id 65534 to users with user-id 0 (root). This user-id should be set to `nobody' no__root__squash doesn't assign user-id 0 to user-id 65534 (default). link__relative Transposing absolute links (those beginning with `/') in a sequence of `../'. This is only useful if the whole filesystem of a machine is mounted (default). link__absolute Symbolic links remain untouched. map__identity User-ids are exactly the same on client and server. (de- fault) map-daemon Client and server don't have matching user-ids. This tells nfsd to create a conversion table for user- ids.ugidd is required for this to work. _______________________________________________________________________________________________________* *_________ Table 16.4: Permission for exported filesystems Your exports-file might look like file contents 16.6.1, page 256. ________________________________________________________________________________________________________* *________|||| | # * * | | * * | | # /etc/exports * * | | # * * | | * * | | /home helios(rw) venus(rw) * * | | /usr/X11 helios(ro) venus(ro) * * | | * * | | /usr/lib/texmf helios(ro) venus(rw) * * | | / earth(ro,root_squash) * * | | * * | | /home/ftp (ro) * * | | # End of exports * * | | * * | |_______________________________________________________________________________________________________* *________ | File contents 16.6.1: /etc/exports File /etc/exports is read by mountd . So, if you have changed anything in this file make sure to restart mountd and nfsd for making your changes take effect. This can easily be done by: root@earth:/ # /sbin/init.d/nfsserver stop root@earth:/ # /sbin/init.d/nfsserver start 16.7 Firewall and Masquerading - Friends and "Friends" on the Internet This section mainly consists of parts of the "S.u.S.E. Firewall and Masquerad- ing Mini-Howto" from Bodo Bauer (if package firewall package is in- 256 _________________16.7.__Firewall_and_Masquerading_-_Friends_and_"Friends"_on_the_Internet___________________________________* *______________________ stalled; see /usr/doc/packages/firewall/README). Please note, that for installing a firewall you can't refer to the installation support; (vgl. section 2.1.2, page 10); however we warmly welcome to pro- vide help within the conditions of the business support (vgl. section 2.1.2, page 11). _______________________________________________________________________________________________________________________|| |||||||| There is no warranty that the procedures given here work and do no harm. ||| | |||||||| || | |||||||| So, if hacker conquer your system although you did follow the rules, do || | |||||||| not blame the authors. We would highly appreciate if you have could send || | |||||||| || | |||||||| any feedback@suse.de instead! || | ||||||||_______________________________________________________________________________________________________________||_ 16.7.1 What is a firewall? A "firewall" serves for protecting a local net from attacks from the inter- net. This might be achieved by connecting the local net to the internet via a firewall. This configuration is described here. More details can be found in "Firewalling and Proxy Server HOWTO" in /usr/doc/howto/ Firewall-HOWTO.gz. 16.7.2 Requirements for installing firewall and masquerading package firewall, series n1 (network) You need package firewall, series n (network) which can be installed by YaST. It contains the following files: /sbin/init.d/masquerade Setting up masquerading /sbin/init.d/firewall Setting up the firewall /sbin/redirect Redirecting packages with this daemon /sbin/fcii Redirecting IPs with this daemon /etc/fw-ssh List of hosts that have ssh - permissions /etc/fw-friends List of "friends" /etc/fw-inout List of hosts that have access to the Internet /usr/doc/packages/firewall/* Documentation on the firewall package And some links that provide starting up everything at boot time. Requirements for the kernel Firewalling and masquerading has to be supported by the kernel. Our S.u.S.E. standard kernels are already configured properly ! Therefore it is not neces- sary to build a kernel of your own. Should it however become necessary to build your own kernel, you have to configure the network the right way. In the following, we list the fea- tures that have to be enabled. These items refer to kernel configuration us- ing make menuconfig . For firewalls please also see screen output 16.7.1. The "transparent proxy" may only be activated if the option `Prompt for * * 257 16. Networking Linux______________________________________________________________________________________________________________* *_______________________________________@ development and/or incomplete code / drivers' has been enabled. Here we give you a list of the network options needed: | * * | ________________________________________________________________________________________________________* *________|||| | * * | | Networking options ---> * * | | [*] Network firewalls * * | | [ ] Network aliasing * * | | * * | | [*] TCP/IP networking * * | | [*] IP: forwarding/gatewaying * * | | [ ] IP: multicasting * * | | [*] IP: firewalling * * | | * * | | [ ] IP: firewall packet logging * * | | [ ] IP: transparent proxy support (EXPERIMENTAL) * * | | [ ] IP: always defragment * * | | * * | | [*] IP: accounting * * | | [ ] IP: optimize as router not host * * | | < > IP: tunneling * * | | [ ] IP: multicast routing (EXPERIMENTAL) * * | | * * | | < > IP: aliasing support * * | | --- (it is safe to leave these untouched) * * | | [ ] IP: PC/TCP compatibility mode * * | | < > IP: Reverse ARP * * | | * * | | [ ] IP: Disable Path MTU Discovery (normally enabled) * * | | [*] IP: Drop source routed frames * * | | [ ] IP: Allow large windows (not recommended if <16Mb of memory) * * | | * * | | * * | |_______________________________________________________________________________________________________* *________ | Screen output 16.7.1: Kernel configuration for firewall If these options are set you can compile your kernel as usual (see chapter 13, page 181). 16.7.3 How a firewall works The firewall should protect the local net from intruders. This requires at least two networking cards being installed. One is for the local net and the other for the outside traffic. This device can be any networking device such as an ethernet card, an ISDN interface or a SLIP / PPP connection. The only important thing is that it has to be static, not dynamic - the IP address has to be fix. Only the traffic running via the specified device in and going out is checked, the local net remains untouched by the firewall. Here we give you an example network. The firewall is located between the LAN and the router which enables the connection to the Internet. All exam- ples refer to this example network. 258 _________________16.7.__Firewall_and_Masquerading_-_Friends_and_"Friends"_on_the_Internet___________________________________* *______________________ I N T E R N E T ^| Router | ...1 | Transfernetz (193.141.17.0/30) | | WWW/FTP/ Mail/News | ...2 DNS-Server und DNS-Server Firewall WWW-Proxy | | ..65 |..66 |..67 <---+--------+--------+--------+------->lokales Netz | ..68 |..69 |..70 |..71 (193.141.17.64/25) Host A Host B Host C Host D Only packages that leave the local network via the gateway are controlled. The internal traffic is not controlled by the firewall. Which variables do I need for firewalls? The firewall is controlled by a couple of variables. These have the prefix FW__ and follow the same format. They contain a list of IP addresses or host names separated by blanks. Exceptions are described separately. You have to enter the IPs here not the hostnames! During setting up of the firewall there are no nameserver requests as every connection is closed. These variables are set in /etc/rc.config (see table 16.5, page 261). ______________________________________________________________________________________________________________________ The firewall is only started if this variable is set to yes List of local networks. They are pro- tected. only friends may access it. Adresses of FTP sites that are acces- sible from the outside. Adresses of WWW sites that are ac- cessible from the outside. Adresses Secure-Socket WWW sites that are accessible from the outside. Port where the SSL server expect re- quests. Here you may only enter one number! Adresses of SMTP sites that are ac- cessible from the outside. Adresses of DNS sites that are acces- sible from the outside. Adresses of NNTP sites that are ac- cessible by news feeds. Adresses of news feeds that are al- lowed to conect to the NNTP servers. ______________________________________________________________________________________________________________________ Table 16.5: to be continued... * * 259 16. Networking Linux______________________________________________________________________________________________________________* *_______________________________________@ _______________________________________________________________________________________________________* *_________ Device that should be protected. Here you may enter only one device. TCP portnumbers that should be locked Here you may enter a range that consists of pairs of numbers sep- arated by a colon. Example: 1:6 8:1023 The ports 1 to 6 and 8 to 1023 are locked. Device for the internal network. Connection to the outside are con- trolled by this device. If this is set to yes all violations of the firewall-deny rules are logged to /var/log/messages. This means that every try of breaking the firewall is logged. If this is set to yes all packages that suit the firewall-accept rules are logged to /var/log/messages. This means that each packages that passes the firewall (allowed) is logged. Adress of the Internet router. This should only be set if the routers address lies in the range given in . If this is set to yes /etc/fw-inout is read. Otherwise every machine on the local network full access to the Internet. Here you may enter a list of ports and IP addresses for redirecting pack- ages "OnTheFly" to local ports. (see section 16.7.3). This is for redirect- ing incoming packages. Same as above but for outgoing con- nections. This is for redirecting local ports to ports on other machines. This nev- ertheless is experimental and should not be used! (see section 16.7.3). _______________________________________________________________________________________________________* *_________ Table 16.5: to be continued... 260 _________________16.7.__Firewall_and_Masquerading_-_Friends_and_"Friends"_on_the_Internet___________________________________* *______________________ ______________________________________________________________________________________________________________________ If this is set to yes the file /etc/ friends is read. Otherwise no ma- chine on the local net has permis- sions to connect to the Internet. This for activating the SSH port (port 22) for those hosts listed in /etc/ fw-ssh. UDP portnumbers that should be locked Syntax is the same as with the TCP ports. It is recommended to set this top 1:1023 i. e. all reserved ports are locked. ______________________________________________________________________________________________________________________ Table 16.5: Variables for the firewall File /etc/fw-friends Machines that have uncontrolled access to the local net are entered in /etc/ fw-friends. One line a machine; comments are marked with an `#'. This file will only be read if is set to yes . Otherwise no ma- chine on the local net has access to the Internet. File /etc/fw-inout Only the hosts listed here have direct access to the Internet. Every machine not included in this list is blocked. Comments are marked with an # (as usual). This file is only read if is set to yes . Otherwise any machine of the local network may access the Internet. Die Datei /etc/fw-ssh Just as above this file contains a list of hostnames (or IP addresses). If is set to yes in /etc/rc.config all listed machines have ac- cess to port 22. This means they may access the secure shell daemon (sshd ). i. e., sie haben Zugriff auf den Secure Shell Daemon (sshd ). Transparent proxy and IP redirection This serves for redirecting IP packages. These may be local ports on the firewall machine itself (Transparent Proxy) or ports on other hosts. By setting incoming IP traffic may be redirected to local ports. It contains a list of quadrupels separated by blanks: Source IP,Target IP,Target Port,Local Port This in particular means that any package that comes from a host with "Source IP" and is aimed for a machine "Target IP" on port "Target port" are redirected to the "Local Port". has the exact same meaning for outgoing traffic. The difference is that _OUT on * * 261 16. Networking Linux______________________________________________________________________________________________________________* *_______________________________________@ the device given in filters whereas _IN does the same on . There may be a daemon on the local port taking care of arriving packages or the may be routed to another host. That's what is needed for. This variable contains a list of triples that resemble the following: Local port,Target IP,Remote port Every package that arrive at "Local port" are redirected to the host given by "Target IP" and "Remote port". ________________________________________________________________________________________________________* *________ | IP redirection is still in an experimental state and should not be used. * * ||| |_______________________________________________________________________________________________________* *________|||_ Starting and stopping the firewall The firewall script accepts four parameters (see table 16.6, page 262). ________________________________________________________________________________________________________* *________ start The firewall is activated stop The firewall is shut down; anybody can go anywhere. block Nothing in nothing out. Resembles unplugging the cable to the internet list The firewall rules are shown ________________________________________________________________________________________________________* *________ Table 16.6: Parameters of the firewall script 16.7.4 Example firewall constellation In section 16.7.3, page 259 you see our example net. There is a local net- work (193.141.17.64/25 ) as well as a transfer net (193.141.17.0/30 ) which connects the firewall to the router. There are a couple of machines connected to the local network either. Well, only A und B shoudl be permitted to access the Internet. Machines that are permitted to access the local net are host.suse.de as well as kiste.info.de . News.provider.de . is the news feed machine. The WWW traffic should be redirected to the local proxy (it listens on port 3128 ). Thus we need to setup a transparent proxy which captures every package that goes to the outside via port 80 and sends it to a local port; this port is redirected to port 3128 of the WWW proxy. Well provided this example network the variables have to be set as given in file contents 16.7.1 needs to be set for letting machines from the local net accessing the router. If this is not needed you should set this variable to an empty string. The route is not protected by the firewall, so if it is hacked it's rather easy to access the local network. In /etc/fw-friends there should be the following lines: (s. file con- tents 16.7.2). Und in /etc/fw-inout sind die Hosts A und B eingetragen (s. file con- tents 16.7.3). 262 _________________16.7.__Firewall_and_Masquerading_-_Friends_and_"Friends"_on_the_Internet__________________________________________* *_______________ |||||||| | _______________________________________________________________________________________________________________________|||| |||||||| | |||||||| FW_START="yes | |||||||| | |||||||| FW_LOCALNETS="193.141.17.0/30 193.141.17.64/25" | |||||||| FW_FTPSERVER="193.141.17.3" | |||||||| FW_WWWSERVER="193.141.17.3" | |||||||| | |||||||| FW_SSLSERVER="" | |||||||| FW_SSLPORT="" | |||||||| | |||||||| FW_MAILSERVER="193.141.17.4" | |||||||| FW_DNSSERVER="193.141.17.3" | |||||||| | |||||||| FW_NNTPSERVER="193.141.17.4" | |||||||| FW_NEWSFEED="134.222.90.2" | |||||||| | |||||||| FW_WORLD_DEV="eth1" | |||||||| FW_INT_DEV="eth0" | |||||||| | |||||||| FW_LOG_ACCEPT="no" | |||||||| FW_LOG_DENY="yes" | |||||||| | |||||||| FW_ROUTER="193.141.17.1" | |||||||| FW_FRIENDS="yes" | |||||||| FW_INOUT="yes" | |||||||| | |||||||| FW_TRANSPROXY_OUT="193.141.17.64/25,0/0,80" | |||||||| FW_REDIRECT="80,193.141.17.66:3128" | |||||||| | |||||||| FW_TCP_LOCKED_PORTS="1:1023" | |||||||| FW_UDP_LOCKED_PORTS="1:1023" | |||||||| | ||||||||_______________________________________________________________________________________________________________ | File contents 16.7.1: Example configuration in /etc/rc.config _______________________________________________________________________________________________________________________|| |||||||| | |||||||| # /etc/fw-friends | |||||||| host.suse.de | |||||||| kiste.info.de | |||||||| | ||||||||_______________________________________________________________________________________________________________ | File contents 16.7.2: Example configuration for /etc/fw-friends _______________________________________________________________________________________________________________________|| |||||||| | |||||||| # /etc/fw-inout | |||||||| 193.141.17.68 # Host A | |||||||| 193.141.17.69 # Host B | |||||||| | ||||||||_______________________________________________________________________________________________________________ | File contents 16.7.3: Example configuration for /etc/fw-inout 263 16. Networking Linux______________________________________________________________________________________________________________* *_______________________________________@ Now you may activate the firewall by entering: root@earth: # /sbin/init.d/firewall start 16.7.5 Principle of masquerading Masquerading means that local addresses are hidden. To the outside there is only one address; the one of your masquerade machine. This serves for hiding the LAN's architecture or for simply assigning internet access to machines without official IP address. Same as with the firewall, there is a device where masquerading takes place. Kernel configuration for masquerading Even masquerading needs to be supported by the kernel see screen out- put|16.7.2.| * * || ________________________________________________________________________________________________________* *________|||| | * * | | Code maturity level options ---> * * | | [*] Prompt for development and/or incomplete code/drivers * * | | * * | | Networking options ---> * * | | * * | | [ ] Network firewalls * * | | [ ] Network aliasing * * | | [*] TCP/IP networking * * | | [*] IP: forwarding/gatewaying * * | | * * | | [ ] IP: multicasting * * | | [ ] IP: syn cookies * * | | [ ] IP: rst cookies * * | | [ ] IP: firewalling * * | | * * | | [ ] IP: firewall packet logging * * | | [*] IP: masquerading * * | | [ ] IP: transparent proxy support (EXPERIMENTAL) * * | | * * | | [*] IP: always defragment * * | | [*] IP: accounting * * | | [*] IP: optimize as router not host * * | | < > IP: tunneling * * | | * * | | --- (it is safe to leave these untouched) * * | | [ ] IP: PC/TCP compatibility mode * * | | < > IP: Reverse ARP * * | | [ ] IP: Disable Path MTU Discovery (normally enabled) * * | | * * | | [*] IP: Drop source routed frames * * | | [*] IP: Allow large windows (not recommended if <16Mb of * * | | memory) * * | | * * | | * * | |_______________________________________________________________________________________________________* *________ | Screen output 16.7.2: Kernel configuration for masquerading Which variables do i need for masquerading? Even masquerading is controlled by a set of variables. The principle is similar to the firewall script, well, there are less variables and they have an MSQ__ as prefix. These are also set in /etc/rc.config (see table 16.7, page 265). 264 _________________16.7.__Firewall_and_Masquerading_-_Friends_and_"Friends"_on_the_Internet___________________________________* *______________________ ______________________________________________________________________________________________________________________ Masquerading is only started if this is set to yes Device where masquerading takes place. List of local networks that should be masqueraded. Modules that are needed for mas- querading: There are three modules that may be given: ip__masq__ftp (enable FTP). ip__masq__irc (enable IRC). ip__masq__raudio (enable). ______________________________________________________________________________________________________________________ Table 16.7: Variables for masquerading Special modules Not all of the protocols accept the replacement of IP addresses in the pack- ages. So, if the server tries to connect to the client this is not possible via a masqueraded router. This is why you may include certain modules for certain protocols (see table 16.8). All modules given in are loaded into the kernel during the start of the masquerading. ______________________________________________________________________________________________________________________ Modul Protokoll ip_masq_cuseeme CU-See-Me ip_masq_ftp FTP ip_masq_irc IRC ip_masq_quake Quake ip_masq_raudio Real Audio ip_masq_vdolive VDO Live ______________________________________________________________________________________________________________________ Table 16.8: Special modules for masquerading Starting and stopping masquerading The masquerade script accepts two parameters (see table 16.9, page 265). ______________________________________________________________________________________________________________________ start Activate masquerading stop Deactivate masquerading list Display masquerading rules ______________________________________________________________________________________________________________________ Table 16.9: The parameters of the masquerading script * * 265 16. Networking Linux______________________________________________________________________________________________________________* *_______________________________________@ Three parameters are known to the masquerade script (see table 16.9, page 265) 16.7.6 Example for masquerading Here we give an example network. The IP addresses used internally are from a class C network 192.168.100.0 . There is only on official IP address available 193.141.17.1 . FTP as well as IRC should be possible. I N T E R N E T ^ | | 193.141.17.1 MSQ router | ..1 local network (192.168.100.0/24) <---+--------+--------+--------+-------> | ..65 |..66 |..67 |..68 Host A Host B Host C Host D Now you need to set the masquerading variables in /etc/rc.config as given in file contents 16.7.4 zu setzen. ________________________________________________________________________________________________________* *________|| | * * | | MSQ_START="yes" * * | | MSQ_DEV="isdn0" * * | | * * | | MSQ_NETWORKS="192.168.0.0/255.255.255.0" * * | | MSQ_MODULES="ip_masq_ftp ip_masq_irc" * * | | * * | |_______________________________________________________________________________________________________* *________ | File contents 16.7.4: Example configuration in /etc/rc.config Now you may activate the masquerading by entering: root@earth: # /sbin/init.d/masquerade start 266 Chapter 17 Connecting to the world: UUCP, SLIP, and PPP Apart from the network connections to remote networks (Wide Area Net- works, WANs) the use of mailboxes might be of interest. There are two standards known in the Unix world, UUCP or TCP/IP (via modem or ISDN). While UUCP (Unix to Unix Copy) is mainly designed for transporting mail and news, TCP/IP supplies a real network connection which supplies all services from a LAN (which is the basic of TCP/IP). If TCP/IP is run over a modem most of the time SLIP (Serial Line Internet Protocol) or PPP (Point to Point Protocol). How to connect to a WAN is the topic of the next section. Connection via a modem is explained as well as data transports via UUCP; furthermore we will show you how to configure a SLIP, or PPP connection, respectively and the configuration of a SLIP or PPP server. 17.1 Connecting a modem Connecting a modem to your machine is exactly how you would expect it to be; it's the same as with other operating systems. The modem is connected to your machine via a serial cable. In YaST you specify which interface to use (see section 12.6, page 174). A link from the modem-device is created to /dev/modem thus letting you access your modem no matter where it is connected to. Normal terminal programs are either minicom or seyon (of course there are others as well). 17.1.1 minicom Minicom is a very easy to handle terminal program which resembles the DOS-program Telix . This is not an introduction to minicom but a short overview of how to configure. All users that want to use minicom have to be entered into /etc/minicom. users. Here is given which users can access which modem with what kind of permissions. You configure minicom (being `root') as follows: * * 267 17. Connecting to_the_world:_UUCP,_SLIP,_and_PPP__________________________________________________________________________________* *_______________________________________@ root@earth:/ # minicom -s Settings are self-explanatory and don't differ to other operating systems. 17.2 PPP PPP (Point to Point protocol) enables you to establish a TCP/IP connection via the serial line. PPP offers far more flexibility and options then SLIP. Client and server may communicate while establishing their connection and setup their needed parameters accordingly. The server may configure an IP address as well as a name to the client. Furthermore PPP is a standard protocol (which SLIP isn't) and is normally the only protocol offered by ISPs. The pppd which is used for communicating with the devices plays a vital and central role in PPP connections. This daemon may either serve as a server or client respectively. The connection itself is done via the program chat . If you plan to setup an "dial on demand" connection you will need the diald (dial daemon) as well. This daemon starts a connection as soon as you request services that need a connection to the Internet (such as reading news from another NNTP server). In a nutshell: it establishes a connection as soon as TCP/IP packages need to be send via TCP/IP. 17.2.1 Requirements for using PPP There are certain things that are required if you want to use PPP on S.u.S.E. Linux: o The kernel needs to support TCP/IP and PPP. Please make sure if you compile a kernel. (see section 13.4.8, page 195). o If you want to make use of the diald (package diald, series n) make sure that you have SLIP support compiled into your kernel. (see section 13.4.8, page 195). o The networking packages need to be installed. The packages that are required are package nkita and package nkitb, series a. o The PPP base package package ppp, series n. The pppd as well as the chat script are included in this package. o The package suseppp, series n. This offers you a wide list of pre con- figured data of some ISPs. This is for automatically setting up PPP with YaST. o Of course you need to know your login and password at the PPP site. 17.2.2 Customizing PPP Configuration may be rather easy if the data of your ISP is listed in our configuration scripts. In package suseppp there is a list of known scripts (see /usr/share/suseppp) which enables YaST to setup PPP according to this ISP. If we don't know the details of your ISP you may start using the default settings (generic ) and adapt them accordingly. If you have managed to establish a connection to an unknown ISP (by using the generic settings) we would be very grateful if you let us know. This 268 ____________________________________________________________________________________________________________________17.2.__P* *PP____________________ enables us to provide the script in our database making live for future users much easier. You may gather all necessary commands for S.u.S.E. (without the password) by launching the following command: root@earth: # /sbin/conf.d/SuSEconfig.ppp -create-cf see README in /usr/doc/packages/suseppp/. _______________________________________________________________________________________________________________________ |||||||| This description assumes that the modem has already been setup as de- || | |||||||| || | |||||||| scribed in section 17.1. It only applies to analogue modems. || | ||||||||_______________________________________________________________________________________________________________||_ YaSTmenu In YaST please select `System administration' as given in section 4.7 and select `Network configuration'. First you need to setup the PPP networking device in `Network base configuration'. Three steps are needed: 1. Select a device with F5 (`Modem PPP'). 2. Configure the device with F6 . 3. Activate the device with F4 . |||||||| | _______________________________________________________________________________________________________________________|||| |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | ||||||||_______________________________________________________________________________________________________________ | Figure 17.1: YaST: PPP Configuration If you press F6 you find yourself in `PPP configuration'. Here you need to enter your values. You may jump between the entries using 7 ! . o In Provider you get a list of the known ISPs. If your provider is not listed please use generic . Country filters only ISPs of this country. o Enter your login and password in PPP Login und PPP Password (you got them from your ISP). o If you want to connect a standalone host to the Internet via PPP you may activate Setup Default route * * 269 17. Connecting to_the_world:_UUCP,_SLIP,_and_PPP__________________________________________________________________________________* *_______________________________________@ o Activate use diald (`X') if you like to use the diald . This enables you to use your PPP on demand. o It is recommended to enable Debug for your first steps and tests (`X'). o Normally you should not change anything in the lower part. If you se- lected the generic template you need to setup this part. - Telephone number of ISP : Enter the telephone number; at the mo- ment you may enter only one number. - Authentification : Here you may select either of PAP , CHAP or TERMINAL . TERMINAL is needed if you the remote machine is a "Ter- minal server". If this is the case you need to adapt /etc/suseppp/ chat-secrets later on as well. - Additional options for pppd my be entered in PPP options . Exam- ple: mru=1500 -vj -pc Options as the one given above might be necessary if you have a bad connection. Details are covered in manpage of pppd (man 8 pppd ). - Normally you should set Provider nameserver as default (`X'). - The variable Provider News server should be configured as given by your ISP. If this is set as default (`X') an PPP connection may be established as soon as you invoke a "newsreader" for reading news. The diald is required for this to work. If you plan to read news etc. locally you do not have to set this variable. - In WWW Proxy there should be something resembling: wwwproxy.provider.de: has to be set to the given port. | * * | ________________________________________________________________________________________________________* *________|||| | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | |_______________________________________________________________________________________________________* *________ | Figure 17.2: YaST: Selection of network (PPP active) 270 ____________________________________________________________________________________________________________________17.2.__P* *PP____________________ If you leave this menu figure 17.2 by entering F10 a couple of files will be generated in /etc/suseppp. These are: .cf , .chat , .options and chat-secrets as well as chap-secrets (or pap-secrets) in /etc/ppp. /etc/ppp/options only is a "Dummy" for making pppd happy and satis- fied. Now SuSEconfig is launched for writing these variables to /etc/rc. config (actually this is done by SuSEconfig.ppp). There is a documenta- tion on SuSEconfig.ppp in /usr/doc/packages/suseppp. Anyway ev- ery information on package suseppp is located in this directory. Configuration of diald If the dial on demand doesn't run out of the box the diald needs some addi- tional configuration. This is done in /etc/suseppp/diald/.diald . How to setup the diald may be seen in manpage of diald (man 8 diald ). Here you may (e. g. ) set "Timeouts" or the number of tries if a connection fails. Check the PPP connection Now you may test your connection by entering: root@earth: # /etc/suseppp/scripts/ppp-up Of course you have to set by the name given in Provider . The error messages are not displayed on screen but written to /var/log/ messages instead. by entering root@earth: # tail -f /var/log/messages on a second console you will be able to watch the output. Thus you will see essential warnings and messages. root@earth: # /etc/suseppp/scripts/ppp-down closes the connection. If everything runs smoothly you should disable the Debug as otherwise /var/log/messages will grow rapidly. 17.2.3 Configuration of an PPP server There is no installation support for setting up an PPP server (see section 2.1.2, page 10). Anyway we offer you this service via our Business-Supports (this nevertheless is not free of charge) (for detailed information please see sec- tion 2.1.2, page 11). Controlling the serial interface on an PPP server is quite comfortable if you use package mgetty, series n for this task. It is recommended to use mgetty even for casual dial ins. You need to setup one account a PPP client. This can be done by invoking the command useradd (more in manpage of useradd (man 8 useradd )). The far easier way is to use YaST for this task (detailed information in sec- tion 4.7.6). Invoking passwd ppp sets a password for the PPP account. The PPP login shell is a small script which in turn launches the PPP daemon: * * 271 17. Connecting to_the_world:_UUCP,_SLIP,_and_PPP__________________________________________________________________________________* *_______________________________________@ ________________________________________________________________________________________________________* *________|| | * * | || #!/bin/sh * * || | # * * | | # /etc/ppp/ppplogin * * | | # * * | | * * | | * * | | exec /usr/sbin/pppd modem passive : * * | | * * | |_______________________________________________________________________________________________________* *________ | File contents 17.2.1: /etc/ppp/ppplogin is used for the server whereas is used for the client. The PPP daemon is launched as soon as the user `ppp' logs in and terminated as soon as he closes the connection. 17.2.4 Further information on PPP PPP offers far more options and capabilities than we could list here. It would be beyond the scope of this book to mention them all here. If you need more capabilities or options please refer to the corresponding manuals and manpages: NET-2-HOWTO.gz and PPP-HOWTO.gz in /usr/doc/howto as well as in the documentation in /usr/doc/packages/ppp. More detailed information on PPP and its protocols may be found in the corresponding RFCs: o RFC1144: Jacobson, V. "Compressing TCP/IP headers for low-speed se- rial links." 1990 February; o RFC1321: Rivest, R. "The MD5 Message-Digest Algorithm." 1992 April; o RFC1332: McGregor, G. "PPP Internet Protocol Control Protocol (IPCP)." 1992 May; o RFC1334: Lloyd, B. Simpson, W.A. PPP authentication protocols."' 1992 October; o RFC1548: Simpson, W.A. "The Point-to-Point Protocol (PPP)." 1993 December; o RFC1549: Simpson, W.A. "PPP in HDLC Framing." 1993 December; 17.3 SLIP SLIP enables you to use TCP/IP via a serial line. This line can be a modem connection to a remote server or a nullmodem as well . SLIP transports IP-packages via the serial port. These interfaces are config- ured as SLIP-devices (SLIP-devices are named sl0, sl1 and are identified over ttyS0 and ttyS1) For identifying, or for switching them there are two tools; one of them is dip the other one slattach . We want to present the more comfortable one dip . The program slattach serves for configuration of SLIP on an existing connection perhaps a permanent line or nullmodem. More information about slattach is to be found in [Dawson, 1995 ]. Before configuring it, some words about requirements: 272 ___________________________________________________________________________________________________________________17.3.__SL* *IP____________________ o The kernel has to support TCP/IP and SLIP (perhaps CSLIP) . To be certain just build a new kernel; this is explained in chapter 13, page 181. o Networking packages have to be installed; absolutely necessary are pack- age nkita and package netkitb. o You have to know your login and password for the SLIP-server o IP-address of the server and your own IP (for static addressing) have to be known. o Which SLIP (SLIP or CSLIP) variation should be used? What's your MTU ( Maximum Transmission Unit)? In case of doubt you can use an MTU of 296. 17.3.1 Establishing connections with dip Program dip (Dial up IP) provides an easy means of establishing and au- tomatizing a connection to a SLIP-server . Program dip allows to set se- rial interfaces to a certain speed, calling a server via modem, doing all login stuff, checking all messages from the server (e. g. for getting to know ones IP- address) and last but not least to switch to the desired SLIP-mode. All those functions can be put together into one script as shown in file contents 17.3.1, page 276. There are three modes known to dip. o dip -t puts you into command mode. This is for interactively setting up and configuring a connection. This is mainly designed for testing purposes. More about command mode is to be found in the corresponding manual page or by entering help in command mode. o Dialin-mode is started with dip -i or through diplogin . This is for build- ing a SLIP-server. More about SLIP-server in section 17.3.2, page 275. o Dialout-mode. A script is executed which describes how to establish a SLIP-connection. A typical dip-script is shown in file contents 17.3.1, page 276 1 Establishing a SLIP-connection needs several actions to be taken: 1. Giving the SLIP-server's IP-address. get $remote sets the server's address . You can either enter the name in quadruples or use the server's name. If you supply the name make sure that it is entered correctly in /etc/hosts (see section 16.3.1, page 248). 2. Initializing the serial line port ttyS1 tells dip to establish the connec- tion via /dev/ttyS1. Here the port where the modem is connected to is given. speed 38400 sets the speed of the interface to 38400 bps. reset initializes the line. This can cause trouble in some cases. If this happens (i. e.the script hangs) you should omit this command. 3. Initializing of modem and selection of server. The command send sends a string to the in port given device. You can wait for an input by wait ". If the expected string 1 ____________________________________________________ adapting this sample script to your personal needs is described below * * 273 17. Connecting to_the_world:_UUCP,_SLIP,_and_PPP__________________________________________________________________________________* *_______________________________________@ () is read within the given time (), variable errlvl is set to the value 0. The modem is initialized by sending ATQ0V1E1S1 to the interface. If it doesn't answer OK within 2 seconds an error is assumed and it jumps to error . If initialization was successful the server is selected by dial . Command dial puts to a value according to table 17.1: ________________________________________________________________________________________________________* *________ OK 0 CONNECT 1 ERROR 2 BUSY 3 NO CARRIER 4 ________________________________________________________________________________________________________* *________ Table 17.1: Answer of modem ad value of Thus only value 1 for errlvl guarantees a successful connection. 4. Logging in at the SLIP-server After connection is up login is to follow. That's why the server is given some time sleep to start the login-program. The modem then sends two _- and awaits the prompt "ogin".2 If this has happened (loginname is send) the password-prompt ("ord") is awaited. Reading of "ord" is answered by sending the password. 5. Passing the machine's IP-address, setting of MTU, routing. After both machines are connected, the serial line has to be switched to SLIP-mode. Before this happens MTU and IP-addresses have to be set. For your IP- address there are two possibilities. Either it already exists (static IP) or it is assigned after the connection is up (dynamical IP). Whether your IP-address is static or not depends on the SLIP-server. Small server with small clients normally provide static IPs for their "members" whereas bigger servers search a pool of IPs while the other system is trying to connect. o Static allocation of IP-address. With static allocation the IP-address is already known. It therefore can be put directly into your dip-script. This is analogue to assigning a server address with get $local . Here the same applies that your hostname has to be entered in /etc/hosts or you have to use the IP-address instead of the server's name. o Dynamic allocation of IP-address If your IP is assigned dynamically the SLIP-server outputs this address on its "welcome" screen after the login. This means that your machine has to "read" the IP-address from the login message. dip supports this by providing get $locip remote 30 which reads the local IP-address and sets the variable local accordingly _____________________________________________2 it awaits "ogin" so that you can type either "Login" or "login" 274 ___________________________________________________________________________________________________________________17.3.__SL* *IP____________________ MTU is set via get $mtu . It is important that both sides use the same MTU. You should agree on a certain MTU with your SLIP provider! If the local machine is not connected to a LAN, you can be sure that all network traffic is done via SLIP you should put this into the kernel's routing table as default-route. This is done by: default .3 6. Switching to SLIP/CSLIP mode. Last command executed by dip is switching the interface to SLIP mode. This happens via mode SLIP . If the server offers CSLIP instead you have to invoke it as mode CSLIP . After having run this script your TCP/IP connection is established. Now you can perform all common tasks within this net (like in a LAN), such as TCP/IP services (e.g. telnet , ftp , WWW : : :) This script can be used as a template for your scripts. Of course you have to adapt your server-address, interface and modem initialization string, tele- phone number, login name, password, your MTU and your IP-address as long as you don't have a dynamical one. _______________________________________________________________________________________________________________________ |||||||| As this file contains unprotected passwords it should not be readable for || | |||||||| || | |||||||| anybody except `root'! || | ||||||||_______________________________________________________________________________________________________________||_ The script is invoked by dip . If you supply it with the param- eter -v (say dip -v ) dip verbosely tells you what it is doing. This can be helpful for debugging. Shutting down SLIP-connection A connection established with dip can be shut down by dip -k . The alter- native is to send a SIGINT to the dip-process. 17.3.2 Configuring a SLIP-server A SLIP-connection always is established between two machines. Using dip, always one of the two is the server whereas the other is the client. How to configure the client side has been said in the previous chapter, but how to configure a SLIP-server? Building a SLIP-server can be done in multiple ways. Our mentioned method is rather simplistic and therefore certainly only useful for small servers. Only the base configuration is described, provided that everything else including the network (as far as there is one : : : ) is set up correctly. Aspects of expe- rienced network configuration are viewed in brief in chapter section 16.1.1, page 245 and more thoroughly in [Kirch, 1995 ]. 3 ____________________________________________________ default does exactly the same that route add -net default does on command line. * * 275 | * * | | * * | | * * | | * * | 17. Connecting to_the_world:_UUCP,_SLIP,_and_PPP__________________________________________________________________________________* *_______________________________________@ | * * | ________________________________________________________________________________________________________* *________|||| | # sample.dip * * | || # This script should be an example of how establishing * * || | # a connection with DIP. * * | | * * | | main: * * | | * * | | * * | | # Announcing IP-address of the server (Galois.suse.de) * * | | get $remote galois.suse.de * * | | * * | | * * | | # Initializing serial line. * * | | port ttyS1 * * | | speed 38400 * * | | * * | | reset * * | | * * | | # Initialising modem and calling the server. * * | | send ATQ0V1E1X1\r * * | | * * | | wait OK 2 * * | | if $errlvl != 0 goto error * * | | * * | | dial 49911-3247122 * * | | * * | | if $errlvl != 1 goto error * * | | * * | | login: * * | | * * | | * * | | # Login in SLIP-server. * * | | sleep 3 * * | | send \r\n\r\n * * | | wait ogin: 10 * * | | * * | | if $errlvl != 0 goto error * * | | * * | | # Send login * * | | send LOGINNAME\n * * | | * * | | wait ord: 5 * * | | if $errlvl != 0 goto error * * | | * * | | # Send password * * | | * * | | send PASSWORT\n * * | | * * | | loggedin: * * | | * * | | * * | | # Setting of SLIP-parameters * * | | get $mtu 296 * * | | * * | | * * | | # Local machine is hal.suse.de * * | | # (static allocation) * * | | get $local hal.suse.de * * | | * * | | * * | | # (dynamic allocation) * * | | # get $locip remote 30 * * | | * * | | # Setting SLIP connection as default-route * * | | * * | | default * * | | * * | | done: * * | | * * | | * * | | # Well, now we can start... * * | | print CONNECTED to $remote with address $rmtip * * | | mode CSLIP * * | | goto exit * * | | * * | | * * | | error: * * | | print SLIP to $remote failed. * * | | * * | 276 | * * | | exit: * * | | * * | |_______________________________________________________________________________________________________* *________ | File contents 17.3.1: Example dip script _________________________________________________________________________________________________________________17.4.__UUCP* *______________________ Dip as SLIP-server Our well-known program dip can easily be used to configure a SLIP-server as well. We use the dialin mode of dip . Assigning and adapting clients is done via /etc/diphosts. o Configuring a SLIP-server. For each client a pseudo-user has to be created to login at the server. The only thing special about this user is that it hasn't an interactive login shell (as e. g. /bin/bash or /bin/tcsh) but the program /usr/sbin/diplogin. 4 It is important that the appropriate user is allowed to execute /usr/sbin/diplogin; and he has to be in the appropriate group ('dialout'). _______________________________________________________________________________________________________________________|| |||||||| For serving as SLIP-server diplogin needs root permissions. For security ||| | |||||||| reasons diplogin of S.u.S.E. Linuxdoesn't have this permission! If you || | |||||||| || | |||||||| want to design a SLIP-server you have to assign the necessary permissions || | |||||||| yourself using the following command: || | |||||||| || | |||||||| || | |||||||| root@earth:/root # chmod 4710 /usr/sbin/dip || | ||||||||_______________________________________________________________________________________________________________||_ A new user (here it is earth) can be added with useradd : root@earth:/ # useradd -u 6000 -g 300 -c "SLIP by earth" -d /tmp -s /sbin/diplogin slearth Now your /etc/passwd should contain this entry: slearth::6000:300:SLIP by hal:/tmp:/sbin/diplogin If a password is given to the user (using passwd ) the first step is complete. o Entering clients into /etc/diphosts. After having assigned a login for the client a relation between the login- name and the server name has to be set. This is done via /etc/diphosts. One entry in /etc/diphosts consists of one line containing seven fields separated by colons: user:password:remote:local:netmask:comments:prot,MTU Diagram file contents 17.3.2, page 278 shows an entry for slearth . Name of the client is earth, name of the server is earth. Netmask is set to "255.255.255.0" und SLIP is driven at MTU 296. For either client or server or both applies that you have to specify either the name or the IP-address. That's all that's needed for creating a SLIP-server. A client now can commu- nicate with your SLIP-server via TCP/IP. 17.4 UUCP UUCP has been designed in the late seventies by Mike Lesk at Bell Labo- ratories for realizing a simple Dial-up-network via telephone lines. UUCP works using the store-and-forward principle. Messages for another host are 4 ____________________________________________________ diplogin just is a symbolic link to dip * * 277 17. Connecting to_the_world:_UUCP,_SLIP,_and_PPP__________________________________________________________________________________* *_______________________________________@ ________________________________________________________________________________________________________* *________ user Login name for the client. password This is for assigning another password; generally this is kept empty. remote host Name or IP of the client. local host Name or IP of the server.(could be empty) netmask Netmask for the SLIP device (could be empty.) comments Comment. protocol, Assigning a protocol (possible values are SLIP or MTU CSLIP ) and MTU). ________________________________________________________________________________________________________* *________ Table 17.2: Contents of /etc/diphosts ________________________________________________________________________________________________________* *________|| | * * | | # * * | | # diphosts This file assigns Login names to IP-addresses * * | | * * | | # for the dip program. * * | | * * | | slerde::erde:galois:255.255.255.0:SLIP by erde:SLIP,296 * * | | * * | | * * | |_______________________________________________________________________________________________________* *________ | File contents 17.3.2: /etc/diphosts kept until a connection is established. If there is a connection data is trans- fered and its processing is launched. In the same way data is received and processed. Processing in this context means that mail is delivered and news are sorted, but you can copy any files from one host to the other. Both hosts being directly connected is not required but data is transported via a queue of other hosts before they reach their destination. 17.4.1 Configuring Taylor-UUCP Along with S.u.S.E. Linux you get the UUCP implementation of Ian Taylor, so called Taylor-UUCP . The corresponding configuration files are located under /var/lib/uucp/taylor_config. Configuration of the UUCP sys- tems is achieved via following files: _______________________________________________________________________________________________________* *_________ config: Central configuration file sys: Information on systems for communicating port: Description of available interfaces dial: Description of available modems _______________________________________________________________________________________________________* *_________ Table 17.3: to be continued... 278 _________________________________________________________________________________________________________________17.4.__UUCP* *______________________ ______________________________________________________________________________________________________________________ call: Logins and passwords ______________________________________________________________________________________________________________________ Table 17.3: Configuration files Taylor-UUCP As Taylor-UUCP is rather flexible and there are a lot of keywords available it would be beyond the aim of this chapter to explain all of them. Here we just explain the vital options that are necessary for configuring a UUCP- connection. In our example the name of the local host is uuearth which wants to create a UUCP-connection to helios.helios is regarded the host which delivers news and mails to uuearth. config This is the central configuration file (see file contents 17.3, page 279). Editing this file can overwrite all hardcoded settings. In general those can be used so that you just have to enter the name of the machine. This is done via the keyword nodename . _______________________________________________________________________________________________________________________|| |||||||| | |||||||| # | |||||||| # config - main UUCP configuration file | |||||||| # | |||||||| | |||||||| | |||||||| # UUCP name of the host | |||||||| nodename uuvenus | |||||||| | ||||||||_______________________________________________________________________________________________________________ | Figure 17.3: /var/lib/uucp/taylor__config/config sys File sys sets which remote hosts are known to the UUCP-system. Each host description takes its own line which should contain the keyword system followed by the host name. All settings until the next system line are regarded as lines for the defined system. Definitions that are made before the separate system lines are regarded as common entries which apply to all systems, as long as they are not overwritten in the corresponding system line. Meaning of the used keywords: _____________________________________________________________________________________________________________ commands Allowed commands. command-path Path where commands are searched. call-login Setting of login name. Setting `*' makes UUCP look in call for the appropriate logins. _____________________________________________________________________________________________________________ Table 17.4: to be continued... * * 279 17. Connecting to_the_world:_UUCP,_SLIP,_and_PPP__________________________________________________________________________________* *_______________________________________@ ____________________________________________________________________________________________________* *______ call-password Setting of password. Setting `*'" makes UUCP look in call for the appropriate passwords. time This string is to define when the system (or sys- tems, depending where it is defined) are allowed to be called. This string is an order of substrings which are separated by a vertical bar `|' or a comma `,'. Each of those substrings has to start with either of `Su', `Mo', `Tu', `We', `Th', `Fr', `Sa', `Wk' or `Any'. The day may be followed by a period of time. This period is given with two times separated by a dash `-'. Three examples: Wk2305-0855,Sa,Su2305-1655 ) working days before 8h55 or after 23h05, sat- urdays anytime, and sundays before 16h55 or af- ter 23h05. Wk0955-2205,Su1705-2255 ) working day between 9h55 and 22h05 and sundays between 17h05 and 22h55. Any ) No restrictions. system Name of the remote system. phone Telephone numbers, where the remote system is reached. port Port to use for the call. References to an entry in port (see below). ____________________________________________________________________________________________________* *______ Table 17.4: Parameter in /var/lib/uucp/taylor__config/sys An example for /var/lib/uucp/taylor_config/sys can be seen at file contents 17.4.1, page 282.. port , This is where the available interfaces (ports) are defined. The structure of this file resembles sys. Each interface description starts with with the keyword port ; global definitions are set before the first port - line. Since almost any time there is only one interface available for UUCP, there are only a few entries necessary in port. The definitions we used mean: ____________________________________________________________________________________________________* *______ port Name of the described port. Is referenced in sys! device Path to special-device-file; if you have configured your modem with YaST you should enter /dev/modem. ____________________________________________________________________________________________________* *______ Table 17.5: to be continued... 280 _________________________________________________________________________________________________________________17.4.__UUCP* *______________________ _____________________________________________________________________________________________________________ speed Speed in Bps (Bits per second), for the interface. dialer Name of modem that is connected to the interface. Is ref- erenced in dial. _____________________________________________________________________________________________________________ Table 17.5: Parameter in /var/lib/uucp/taylor__config/port An example for /var/lib/uucp/taylor_config/port can be seen at file contents 17.4.2, page 282. dial In dial all available modems are defined. Same applies here; all settings before the specific modem settings apply to all entries, which, in this case is the line dialer followed by a name. Besides the name you have to set how your modem is to be initialized, how to call a remote system, which error-codes can occur, and how it should be initialized after a shutdown (or hang-up). This is done via the following keywords: _____________________________________________________________________________________________________________ dialer Name of modem definition; is referenced in port. chat Initializing your modem and setting a certain tele- phone number. Here a couple of strings are given; the first one is sent to the modem while the second one is received by the modem. An empty line is given as "", strings are separated by blanks. Within those strings the following replacements occur: \T Telephone number \r Carriage return \c Suppressing carriage return \d Pausing 1 to 2 seconds \s An empty string chat-fail Answer from the modem in case of an error. There can be as many lines as you like. abort Modem initialization after connection is established (complete ), or aborted respectively (abort ). Syntax of this line is identical to chat -line. _____________________________________________________________________________________________________________ Table 17.6: Parameter in /var/lib/uucp/taylor__config/dial An example for /var/lib/uucp/taylor_config/dial can be seen at file contents 17.4.3, page 283. * * 281 17. Connecting to_the_world:_UUCP,_SLIP,_and_PPP__________________________________________________________________________________* *_______________________________________@ | * * | | * * | ________________________________________________________________________________________________________* *________|||| | * * | | # * * | | # sys - Desription of known systems * * | | * * | | # * * | | * * | | * * | | # Globale settings for all systems * * | | commands rmail rnews * * | | command-path /usr/lib/news/bin /usr/bin * * | | * * | | * * | | # Read login name and password from 'call' * * | | * * | | call-login * * * | | call-password * * * | | * * | | * * | | #No time restrictions * * | | * * | | time any * * | | * * | | * * | | # System specific settings * * | | * * | | # System 'hal' * * | | * * | | system hal * * | | * * | | * * | | # Telephone number * * | | phone 0123-123456 * * | | * * | | * * | | # Portdefinition; which port to take * * | | * * | | port serial1 * * | | * * | |_______________________________________________________________________________________________________* *________ | File contents 17.4.1: /var/lib/uucp/taylor__config/sys | * * | ________________________________________________________________________________________________________* *________|| | * * | | # * * | | * * | | # port - Description of interfaces * * | | # * * | | * * | | * * | | # Name of the interface * * | | * * | | port serial1 * * | | * * | | # Device for this interface * * | | * * | | device /dev/modem * * | | * * | | * * | | # Speed * * | | speed 38400 * * | | * * | | * * | | # Name of your modem (references to 'dial') * * | | * * | | dialer generic * * | | * * | |_______________________________________________________________________________________________________* *________ | File contents 17.4.2: /var/lib/uucp/taylor__config/port 282 _________________________________________________________________________________________________________________17.4.__UUCP* *______________________ |||||||| | _______________________________________________________________________________________________________________________|||| |||||||| # | ||||||||| # dial - Description of available modems || |||||||| | |||||||| # | |||||||| # Name of modem (is referenced in 'port') | |||||||| dialer generic | |||||||| | |||||||| | |||||||| # | |||||||| | |||||||| # Initalization strings | |||||||| # | |||||||| | |||||||| chat "" ATZ OK ATDT\T\r\c CONNECT | |||||||| | |||||||| | |||||||| # Error codes the modem can send | |||||||| chat-fail BUSY | |||||||| | |||||||| chat-fail NO\sDIALTONE | |||||||| chat-fail NO\sCARRIER | |||||||| | |||||||| | |||||||| # Modemreset after normal shutdown | |||||||| complete \d\d+++\d\dATH0Z\r\c | |||||||| | |||||||| | |||||||| # Modemreset after unexpected shutdown | |||||||| | |||||||| abort \d\d+++\d\dATH0Z\r\c | |||||||| | ||||||||_______________________________________________________________________________________________________________ | File contents 17.4.3: /var/lib/uucp/taylor__config/dial call Last needed information; this is the name and password for logging into the remote system. This is located in call. Each line defines a system and has the following syntax: An example for /var/lib/uucp/taylor_config/call can be seen in file contents 17.4.4, page 283. _______________________________________________________________________________________________________________________|||| |||||||| | |||||||| # | |||||||| # call - Login informations | |||||||| | |||||||| # | |||||||| | |||||||| | |||||||| # | |||||||| # Login name und Password for the given systems | |||||||| | |||||||| # | |||||||| # | |||||||| hal uusofa hempel | |||||||| | |||||||| | |||||||| | ||||||||_______________________________________________________________________________________________________________ | File contents 17.4.4: /var/lib/uucp/taylor__config/call * * 283 17. Connecting to_the_world:_UUCP,_SLIP,_and_PPP__________________________________________________________________________________* *_______________________________________@ ________________________________________________________________________________________________________* *________ | Since there are unprotected passwords in this file you should ensure that * * ||| | * * ||| | nobody except UUCP has read permissions to this file! * * ||| |_______________________________________________________________________________________________________* *________|||_ 17.4.2 Testing the configuration For testing your configuration just copy a file to the remote system using the command uucp . That's how this is done: root@earth:/ # uucp ! (In bash , csh , or tcsh respectively the `!' has to be 'masked' with a `\'!) Here is to be replaced by the filename to be copied and is the name of the remote system. This system has to be defined in sys (uuname outputs a list of all machines known to the UUCP-system). uustat shows which processes are waiting for being executed: root@earth:/ # uucp testfile erde!/testfile root@earth:/ # uustat -a halN0002 hal bb 10-24 16:11 Sending /home/user/testfile (276 bytes) to "/testfile Establishing a connection and transferring data is done by uucico . The parameter -S tells which system is to be called. uucico starts a new process which maintains the data transfer. Parameters -x <0-9> are for different debug-levels. Passing 0 means that no debug information is wanted, whereas 9 means that any package transferred is protocolled. Default is 2 . Debug information is written to /var/spool/uucp/.Log/uux/ and /var/spool/uucp/Log as well as /var/spool/uucp/Stats 17.5 ISDN Configuration ________________________________________________________________________________________________________* *________ | At the moment only the Euro-ISDN is supported by isdn4linux , well, to * * ||| | * * ||| | be precise the passive cards definitely will not work! * * ||| |_______________________________________________________________________________________________________* *________|||_ Besides its "normal" network capabilities Linux is capable of connecting to your ISP via ISDN. As most of this may be setup with YaST ISDN configu- ration with S.u.S.E. Linux is rather straight forward and easy. Our chapter mainly refers to a standard connection to your ISP via ISDN. Of course there is lots more that can be done by isdn4linux . Please keep in mind that some of the procedures mentioned below might be "illegal". Every active cards and their firmware are certified. Passive ISDN controllers are certified if run with the software of the manufacturer. For those who need a certification please use either an active ISDN card or connect the controller to your PBX. 284 ____________________________________________________________________________________________17.5.__ISDN_Configuration_______* *______________________ There is one great difference to a modem connection. Once the network has been setup and configured no additional commands are needed. This is called "on demand", as soon as you launch an e. g. telnet session the connection will be established. This normally takes up to three seconds. Thus it is possible to let "normal" users establish a connection. You may set the idle time, this is the period of time when the connections should be cancelled automatically if all processes that depend on the connection are idle. During configuration of your ISDN system it is recommended to carefully watch messages in /var/log/messages. Just start another xterm or login onto another console and enter: root@earth: # tail -f /var/log/messages Now you will see every line added to /var/log/messages 17.5.1 Overview S.u.S.E. Linux includes package isdn4linux. It includes hardware drivers, network interfaces as well as modem emulation (only digital modems). Fur- thermore it includes software for e. g. an answer phone. The ISDN hardware driver is launched by /sbin/init.d/i4l_hardware (see chapter 12). Configuration of the ISDN part is done via isdnctrl (man- page of isdnctrl (man isdnctrl )). The network interfaces are config- ured just as standard ethernet interfaces by ifconfig (manpage of ifconfig (man ifconfig )) and route (manpage of route (man route )). On S.u.S.E. Linux /sbin/init.d/i4l performs this task (see chapter 12). The actions taken are based upon entries in /etc/rc.config. These names resemble (where possible) the options of isdnctrl . /sbin/init.d/route sets up the routing of the devices given in /etc/ route.conf. Establishing an connection is done by isdnctrl , or bzw. /sbin/init.d/i4l respectively using settings from /etc/rc.config. These parameters may be listed by entering: root@earth: # isdnctrl list all as soon as somebody requests an ISDN service (this may be either a user or an application) the connection is established. 17.5.2 Configure ISDN hardware Requirements For successfully creating a connection on S.u.S.E. Linux you will need: 1. an ISDN connection 2. a supported ISDN controller 3. S.u.S.E. Linux installed 4. one of S.u.S.E. Linux standard kernels (on CD). _______________________________________________________________________________________________________________________ |||||||| You don't need to compile a kernel! If you want to compile a kernel || | |||||||| || | |||||||| anyway make sure you use the sources from package lx__suse, series d! || | ||||||||_______________________________________________________________________________________________________________||_ * * 285 17. Connecting to_the_world:_UUCP,_SLIP,_and_PPP__________________________________________________________________________________* *_______________________________________@ 5. the package kernmod, series a 6. the package i4l, series n 7. documentation may be found in package i4ldoc, series doc (recom- mended). What you need to know: o Type of your ISDN controller o Controller settings: IRQ, Port address, etc. (depends on the type) o The ISDN protocol you use: - 1TR6 : (old) national ISDN - DSS1 : Euro-ISDN ________________________________________________________________________________________________________* *________|| | Some of the PBX (as opposed in the documentation) still use 1TR6 instead * * |||| | of DSS1 * * ||| |_______________________________________________________________________________________________________* *________|||_ What is a MSN/EAZ? With Euro ISDN you get a MSN (Multiple-Subscribe-Number) which nor- mally is your telephone number without the prefix. If you just subscribed for ISDN you will receive three different numbers. Any of them may be used for your ISDN connection even if you use the same number as telephone number as they may be distinguished by their service indicator. Normally the ISDN controller is directly attached to an NTBA. You might as well connect another S0 bus to the PBX. If you use Euro ISDN on your PBX the MSN normally is the extension (direct call number). For 1TR6 you have an EAZ (German: "Endger"ate Auswahl Ziffer = end user selection number") otherwise they are treated the same. The EAZ is a single number. Just select one in the range of 1 to 7 . Don't forget the 0 ! Configure ISDN hardware with YaST The driver itself is provided by a loadable kernel module. You don't need to reboot your system. Standard ISDN controllers are supported by the HiSax driver. Some controllers e. g. ICN or AVM-B1 (as well as PnP cards) may not be configured with YaST yet. They need a special treatment. Please have a look at the settings for ISDN controllers further down. Here is how to proceed step by step: 1. Login as user `root' 2. launch YaST 3. Now select `System administration', `Integrate hardware into system', `Configure ISDN hardware'. The menu structure is shown in figure 17.4. 4. Please enter the following parameters: o Start I4L ISDN is only launched if this is activated. Thus you can make sure whether ISDN is started automatically at boot up. 286 ____________________________________________________________________________________________17.5.__ISDN_Configuration________* *_____________________ |||||||| | _______________________________________________________________________________________________________________________|||| |||||||| | ||||||||| || |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | ||||||||_______________________________________________________________________________________________________________ | Figure 17.4: Menu structure for ISDN configuration in YaST o ISDN protocol Here you may choose either the old (national) German ISDN (1TR6) or the default Euro ISDN (EDSS1). Please keep in mind that connections via a PBX often still use 1TR6. o ISDN controller type Select the supported ISDN controller. Please have a look at /usr/ doc/packages/i4l/README.SuSE. for PnP and/or PCMCIA con- trollers. o Controller ID You should leave this untouched to Tel0 . o Interrupt Memory base address IO port ISAC HSCX Dependent on the card in use some additional settings might be nec- essary. Only the parameters available for the device are enabled the others are disabled. o ISDN options This should be empty! Pressing F1 will give you additional help. figure 17.5 shows you the menu mask. 5. Now please confirm by pressing `Start'. This is a test: The module will be loaded and the messages in the window tell you whether the card has been setup correctly. * * 287 17. Connecting to_the_world:_UUCP,_SLIP,_and_PPP__________________________________________________________________________________* *_______________________________________@ | * * | ________________________________________________________________________________________________________* *________|||| | * * | || * * || | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | | * * | |_______________________________________________________________________________________________________* *________ | Figure 17.5: Menu for ISDN configuration with YaST If OK: Confirm by pressing `Save'. Now your settings are written to /etc/rc.config permanently. They will remain active unless you change them. After being tested the driver remains loaded. If it fails: Check and change the parameters and don't forget to have a look at /var/log/messages (you remember you opened it, didn't you?.:)). Possible problems may be: o On some boards the IRQs 12 or 15 cannot be used. o The given address or IRQ are already in use. Remove (just for test) all controllers that are not urgently needed for checking (e. g. sound or networking devices). o The module has already been loaded. To remove it please change to another console and enter: root@earth: # rmmod hisax o The card you use is an PnP device please have a look at /usr/doc/ packages/i4l/README.SuSE. o Your card is not supported by HiSax (e. g. ICN, AVM-B1). Please refer to /usr/doc/packages/i4l/README.SuSE. 6. Exit YaST. 7. configure isdnlog You should configure isdnlog before launching the modules. Its task is it to supervise every activity on the S0 bus system. Please adapt the following files to your needs: o /etc/isdn/isdn.conf: 288 ____________________________________________________________________________________________17.5.__ISDN_Configuration________* *_____________________ The first parameter is for specifying your country where you use your isdn4linux. If this is Germany you should set them as given in file contents 17.5.1, page 289: _______________________________________________________________________________________________________________________|| |||||||| | |||||||| # /etc/isdn/isdn.conf | |||||||| | |||||||| | |||||||| [GLOBAL] | |||||||| COUNTRYPREFIX = + | |||||||| | |||||||| COUNTRYCODE = 49 | |||||||| AREAPREFIX = 0 | |||||||| | ||||||||_______________________________________________________________________________________________________________ | File contents 17.5.1: /etc/isdn/isdn.conf Here (in the GLOBAL section) you also need to enter your area code AREACODE (the dialing prefix) without the leading zero. So if your area code is 0911 e. g. you will need to enter: AREACODE = 911. In Germany this is the only part that needs to be adapted. CHARGEMAX = 20.00 lets you set the maximum amount of money (in Deutsch Marks) that you accept a day. Please do not rely on this feature! o /etc/isdn/callerid.conf: Here you may enter every known telephone number. You will see their "names" instead of the MSNs in /var/log/messages and if you launch isdnrep . See our example in file contents 17.5.2, page 289; Your number is 4711 , the number of your ISP is 4712 . _______________________________________________________________________________________________________________________|||| |||||||| | |||||||| # /etc/isdn/callerid.conf | |||||||| | |||||||| | |||||||| [MSN] | |||||||| NUMBER = 4711 | |||||||| | |||||||| SI = 1 | |||||||| ALIAS = myself | |||||||| | |||||||| ZONE = 1 | |||||||| | |||||||| | |||||||| [MSN] | |||||||| NUMBER = 4712 | |||||||| SI = 1 | |||||||| | |||||||| ALIAS = ISP | |||||||| ZONE = 1 | |||||||| | ||||||||_______________________________________________________________________________________________________________ | File contents 17.5.2: /etc/isdn/callerid.conf o /etc/isdn/isdnlog.isdnctrl0.options: Here you may enter options for isdnlog this normally is not necessary. 8. Now enter the commands: root@earth: # init 1 root@earth: # init 2 * * 289 17. Connecting to_the_world:_UUCP,_SLIP,_and_PPP__________________________________________________________________________________* *_______________________________________@ thus restarting all network services. You may as well activate ISDN with YaST or reboot if you prefer. 17.5.3 Testing ISDN on our S.u.S.E. host The S.u.S.E.ISDN server This server has been setup for S.u.S.E. Linux users to check your ISDN con- figuration by logging in using test logins. In the near future we plan to offer more capabilities, this being access on our Support Database (update pack- ages etc). Please have a look at /home/suse/README on the the "test" server. ________________________________________________________________________________________________________* *________ | You will not be able to access the Internet via this server! * * ||| |_______________________________________________________________________________________________________* *________|||_ Selection of a protocol This ISDN host offers three different modes: o Terminal-Login using X.75 o rawip-HDLC o syncPPP Of course you may check any of these protocols. Anyway it is recommended to check the protocol offered by your ISP. Requirements The ISDN hardware configuration ran smoothly and the ISDN driver has been loaded. Furthermore you need to have your MSN/EAZ at your hand. Select a protocol ( rawip , syncPPP ). Of course you may setup different connections. Our example refers to an syncPPP connection. rawip is almost the same but much easier. Well, here are the steps: 1. Please invoke YaST and change to `System administration', `Network configuration', `Basic network configuration'. figure 17.6 shows you a screen shot. 2. Now use a number not already used e. g. 4 . 3. Now select the device `ISDN SyncPPP' by pressing F5 . 4. Hit F6 (`IP address') and enter ein: o IP address of your host: 192.168.0.99 o IP address of the PointToPoint partner: 192.168.0.1 5. Leave this window by pressing `Continue'. 6. Hitting F4 activates the network device if you haven't done so already. 7. F8 (`ISDN')lets you enter some ISDN specific options. This is shown in: figure 17.7 . Now please enter the following values: o Now enter Your telephone number (MSN) 290 ____________________________________________________________________________________________17.5.__ISDN_Configuration_____________* *________________ |||||||| | _______________________________________________________________________________________________________________________|||| |||||||| | ||||||||| || |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | ||||||||_______________________________________________________________________________________________________________ | |||||||| Figure 17.6: Network configuration with YaST | _______________________________________________________________________________________________________________________|||| |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | |||||||| | ||||||||_______________________________________________________________________________________________________________ | Figure 17.7: Setting of ISDN network parameters with YaST o The number to be dialled: 09113206726 09113206726 is the number of the S.u.S.E. test host for syncPPP . _______________________________________________________________________________________________________________________ |||||||| You might need to dial a leading "0" for some PBXs. || | ||||||||_______________________________________________________________________________________________________________||_ o Numbers, that are allowed to dialin: Only needed for dialin servers o Only given numbers are allowed: Please set this for making sure nobody starts and unwanted connection to your machine! 291 17. Connecting to_the_world:_UUCP,_SLIP,_and_PPP__________________________________________________________________________________* *_______________________________________@ o Idle time: After a period of time (idle time) the connection is closed if it is cur- rently not in use: o Name of PPP login: For our S.u.S.E. syncPPP test the login name is `suse'. o Password of PPP login: For our S.u.S.E. syncPPP test server the password is `linux'. You will not see the actual password while entering it (bus asterisks in- stead). It is written to /etc/ppp/pap-secrets. Hit F1 to get additional help. 8. Confirm `Start'. Now a network test is launched. You will get an error message if anything went wrong. There should not be problems so far. If OK: Please confirm by hitting `Save'. Now your settings will be written permanently to /etc/rc.config. They will remain active unless you change the settings. The settings remain active after the "test" start. If it fails: Eventually the ISDN modules were not loaded. Don't forget to have a look at /var/log/messages. 9. Now switch to another xterm or another console. You don't need to be user `root' to enter the following: 10. Establish an Internet with syncPPP mit: root@earth: # telnet 192.168.0.1 or if you use rawip root@earth: # telnet 192.168.0.2 After approximately 2-3 seconds you should see the S.u.S.E. server login prompt. Now you may enter with your account and password (`suse', `linux'). Thereafter have a look at README. e. g. by entering less README logout lets you close the connection. After the idle time that was setup previously the connection will be dropped anyway. Every activity may be watched in /var/log/messages. If anything fails: o Check /var/log/messages for "suspicious" messages o Try the rawip connection o Is the MSN/EAZ setup correctly? o Perhaps you need a leading 0? There are further hints in our Support Database. It may be found on our WWW site at http://www.suse.de/Support/sdb or by launching the S.u.S.E. help system (enter: hilfe or select it from the menu; package susehilf, series doc and package sdb, series doc needs to be installed) 292 ____________________________________________________________________________________________17.5.__ISDN_Configuration______* *_______________________ 11. Presumably the connection was established. Switch to YaST select `Save' and exit YaST. 12. Now please adopt /etc/hosts. All IP addresses used should be listed here (see file contents 17.5.3, page 293). _______________________________________________________________________________________________________________________|| |||||||| | |||||||| 192.168.0.1 pppserver.suse.de pppserver | |||||||| 192.168.0.99 pppclient.suse.de pppclient | |||||||| | |||||||| 192.168.0.2 rawipserver.suse.de rawipserver | |||||||| 192.168.0.98 rawipclient.suse.de rawipclient | |||||||| | ||||||||_______________________________________________________________________________________________________________ | File contents 17.5.3: Entries in /etc/hosts for connecting to the S.u.S.E. ISDN test server 17.5.4 Setup configuration for your ISP If you have successfully managed the "test" connection you may connect to your ISP in exact the same manner. This might be tricky but only if the ISP uses protocols and switches that are not usual. Some instructions and vital hints may be found along with some scripts on our SDB (Support Database) (see section 2.1.3, page 12). The method mentioned above connects using either rawip or synchronous PP using PAP as authentification method. This normally is "standard". Use the settings for the test server to start with and change the parameters listed below: o ISP telephone number o User name and password (for syncPPP) o IP numbers (if known, please see below) o Enter a nameserver This can be done interactively from YaST in `System administration', `Configure network', `Name server' o Setup your routing section 16.4 Howto use dynamic IP numbers with syncPPP Your ISP only provides dynamic IP addresses; what do I need to setup? If you don't have an IP (dynamic IP only for PPP) you should add a private address for yourself and your ISP (192.168.*.*). Thus you may just leave the entries as given in our example. After the connection has been established invoking ifconfig or having a look at /var/log/messages will give you the IP addresses used. Your IP address is not important at all. The IP address of your PointToPoint partner should be a fixed one. Now you may enter this address and thus adjust your gateway. Each time a connection is established the ipppd launches an ifconfig (man- page of ifconfig (man ifconfig )). All routing information goes into the bin as soon as you restart ifconfig for this device. So if you like to * * 293 17. Connecting to_the_world:_UUCP,_SLIP,_and_PPP__________________________________________________________________________________* *_______________________________________@ have your "default-route" after the connection is down you need to set it in /etc/ppp/ip-down. There is an example of this file in /usr/doc/packages/i4l/pppsample. 17.5.5 Reference of the S.u.S.E. ISDN account _______________________________________________________________________ |__telnet______-Login_______________________________________________|___ |__User_name_________________________|`suse'___________________________| |__Password__________________________|`linux'__________________________| _______________________________________________________________________ |__rawip_____________________________________________________________|__ |__Telephone_number__________________|09113206728______________________| |__Client_IP_address_________________|192.168.0.98_____________________| |__Server_IP_address_________________|192.168.0.2______________________| |__Network_device____________________|isdn0____________________________| _______________________________________________________________________ |__syncPPP________________________________________________________|_____ |__Telephone_number__________________|09113206726______________________| |__Client_IP_address_________________|192.168.0.99_____________________| |__Server_IP_address_________________|192.168.0.1______________________| |__Network_device____________________|ippp0____________________________| |__User_name_________________________|`suse'___________________________| |__Password__________________________|`linux'__________________________| |__Authentification__________________|PAP______________________________| _______________________________________________________________________ |__Modem_________________________________________________________|______ |__Telephone_number__________________|09113247114______________________| Initializing the modem: ATZ ATS14=0&E123 ATD09113247114 In our example 123 is your MSN. Additional information Further information on how to establish a connection to the S.u.S.E. ISDN server may be found in: o File /usr/doc/packages/i4l/README.SuSE o Support Database: http://www.suse.de/Support/sdb o In package i4ldoc (e. g. the ISDN-FAQ in: /usr/doc/packages/i4ldoc/ i4l-faq) o /usr/doc/inetcfg (package inetcfg): e. g. : T-Online via ISDN 294 ____________________________________________________________________17.6.__Let's_write_-_configuration_of_email_____________* *______________________ 17.6 Let's write - configuration of email If a connection to the wide world has been established, let it be it via UUCP, SLIP, or ISDN, it should be used to do something. A rather typical application is + electronic mail, electronic mail. This section describes configuration of sendmail 5 . sendmail has to decide how to deliver incoming (and outgoing) electronic mail. This might either be via a TCP/IP network using the SMTP protocol or to use another transport agent such as UUCP. sendmail 's main configuration file is called /etc/sendmail.cf. If you have a normal and simple configuration you may setup the necessary parameters using YaST. YaST then creates a valid /etc/sendmail.cf for you. All settings are written to /etc/rc.config and YaST will create the file /etc/ sendmail.cf for you using the parameters given there. As sendmail configuration is rather complex S.u.S.E. provides two precon- figured settings that should be sufficient in most cases: If you plan to use sendmail within a TCP/IP network make sure to have a valid DNS server. Here you need to setup one entry ("MX record" mail exchange record) per name. The current settings may be checked with the host command (in package package bind): root@earth: # host helios.cosmos.com helios.cosmos.com address 192.168.0.1 helios.cosmos.com mail is handled (pri=10) by helios.cosmos.com helios.cosmos.com mail is handled (pri=100) by mail-relay.cosmos.com If there is no entry for mail you should ask your system administrator for help. The following electronic mail variables may be set by YaST in /etc/rc. config (see section 12.6, page 174): o SENDMAIL__TYPE="yes" This variable has to be set to yes if the sendmail configuration file shall be created using the values in /etc/rc.config. If you want to create a /etc/sendmail.cf yourself no is the answer. o SENDMAIL__LOCALHOST="localhost helios.cosmos.com www.cosmos.com" sendmail needs to know which electronic mail should be stored locally and which needs to be delivered to another host. Only electronic mail to the local host itself is saved locally by default. By givng a list in SENDMAIL__LOCALHOST you can configure other names that should be con- sidered a local host. Example: the name of the machine is helios.cosmos.com it serves as WWW site for www.cosmos.com. For accepting electronic mail that goes to www.cosmos.com you need to enter the following line: SENDMAIL__LOCALHOST="localhost www.cosmos.com" . 5 ____________________________________________________ There is an alternative to sendmail called smail or qmail , well these are not covered in this chapter. * * 295 17. Connecting to_the_world:_UUCP,_SLIP,_and_PPP__________________________________________________________________________________* *_______________________________________@ o FROM__HEADER=cosmos.com Normally the local machine's name is used as the from header. This may be set to an indifferent name: Example: The machine's name is earth.cosmos.com. Anyway you'd like to send electronic mail as newbie@ cosmos.comT`his may be achieved by the parameter: FROM__HEADER=cosmos.com . o SENDMAIL__SMARTHOST=mail-server.provider.de sendmail asks for the DNS names of every mail that is not delivered lo- cally and tries to send the electronic mail via the SMTP protocol. This host might be anywhere on the Internet and eventually has a rather slow connection to our local host. Setting this parameter lets you setup an inter- mediate host that gets all our outgoing mail. Then this host is responsible of delivering your electronic mail. First example: This is for a dialup connection. Thus you deliver all of your electronic mail directly to your ISP: SENDMAIL__SMARTHOST=smtp:mail-server.provider.de . Second example: If you are connected via UUCP you may send all electronic mail that is not local to your UUCP server: SENDMAIL__SMARTHOST=uucp-dom:uucp.cosmos.com . o SENDMAIL__NOCANONIFY=no sendmail tries to resolve each and every email address from the mail header and replaces the name with the "Fully Qualified Domain Names" (FQDN). So if there is no DNS server available (due to a dialup connec- tion perhaps) and you enter the name correctly you may switch this of by setting it to yes . o SENDMAIL__ARGS="-bd -q30m -om" That's how sendmail is invoket at bootup time. -q30m tells sendmail to check /var/mqueue every 30 minutes if there is electronic mail wait- ing. -bd starts sendmail in "daemon mode" this enables you to accept electronic mail via the TCP/IP network. So if you only have a dialup connection you may omit the -q30m and invoke sendmail directly with sendmail -q . This might be done via a crontag entry once or twice a day. Additionally you may enter the start of sendmail -q into your scripts that establish the connection to your ISP. This lets you exchange elec- tronic mail every time you connect to the net. o SENDMAIL__EXPENSIVE=no Normally sendmail tries to deliver the electronic mail immediately via SMTP. So if you are only temporarily connected this might not be what you need as a connection is established each time you write an electronic mail. If setting this to yes the mail is queued in /var/mqueue and deliv- ered as soon as you launch a sendmail -q . 296 ____________________________________________________________________________17.7.__Brandnew_messages_-_C_News_______________* *______________________ All locally delivered electronic mail is handled and saved to the local elec- tronic mail folder /var/spool/mail/ by procmail . Please have a look at manpage of procmailrc (man procmailrc ), manpage of procmailex (man procmailex ) as well as manpage of procmail (man procmail ) for a description of this extreme versatile tool. If you do not deliver remote electronic mail immediately it is saved to the queue directory /var/mqueue and delivered at the next run of sendmail . You may as well launch sendmail directly by entering sendmail -q . There are further settings that can be made, e. g. in /etc/aliases and some other files in /etc/mail/. There are commented examples included. Some files need to be translated to databases using the makemap tool. This is invoked automatically if you start SuSEconfig or when you leave YaST. If you need a more complex configuration of sendmail you should disable the automatic setup of /etc/sendmail.cf by setting SENDMAIL__TYPE=no . Then you may use /etc/mail/linux.mc as a template for your configura- tion. linux.mc is written using m4 commands. root@earth: # m4 /etc/mail/linux.mc > /etc/sendmail.cf creates a valid sendmail configuration by using the macros in /usr/share/ sendmail Further documentation my be found in /etc/mail, /usr/share/sendmail as well as /usr/doc/packages/sendmail There is a web site at http: //www.sendmail.org/. If you need to solve more complex tasks you will need the sendmail boot from O'Reilly6 This describes sendmail in any gory detail. 17.7 Brandnew messages - C News One of the most important services that is provided by the internet is the transport and delivery of news that are sorted in different groups; this part of the internet is often referred to as the usenet. Only through this media development it was possible to develop Linux at all, and only by means of this high-efficient form of communication a rapid development and removing of bugs was and is possible. Since a complete description of the news-system with all its thousands of possibilities (e. g. passing news to other machines) would be far beyond the scope of this book; only a basic local system based on the widely spread C News -packages is described. 17.7.1 Broadcasting the news The machine that wants to receive news generally is not on-line all the time (due to telephone costs which are immense for the private user). Usually UUCP (chapter section 17.4, page 277) serves for exchanging news over the net. Those machines that deliver news are called newsfeed. On those machines there is a program running several times a day which packs all news that have to be transferred to other machines and makes them ready to 6 ____________________________________________________ Vgl. [Costales et al., 1993 ]. * * 297 17. Connecting to_the_world:_UUCP,_SLIP,_and_PPP__________________________________________________________________________________* *_______________________________________@ send via UUCP. Those machines that want to receive news have to poll the newsfeed; they call it up once or several times a day and let those prepared packages be sent. Since this is a regular job it is quite obvious to let cron handle this task; thus you don't have to start the job by hand each time you want to poll the news. This can be done by editing the cron table via crontab -e being `root'; you should enter the following line: ________________________________________________________________________________________________________* *________|| | 7 05 * * * /usr/lib/uucp/uucico -f -s erde * * | | * * | | * * | |_______________________________________________________________________________________________________* *________ | File contents 17.7.1: Example cron entry A machine with this entry in crontab would call the newsfeed earth every day at 5h07 for polling news. Of course it is required that UUCP is configured correctly and the machine is running at the given time. 17.7.2 Installing a local news-system For processing those received files it is necessary that package cnews is in- stalled properly. Since permissions are very important for cnews to work properly all configu- ration should be done as user ` news'. This can be achieved if you login as `root' and change to news by su - news . Before starting the configuration you should take the amount of news to be received into consideration. News are saved under /var/spool/news; since those files are relatively small you could consider giving it a partition with a small + inode density (app. 2048 bytes per inode). If the amount is relatively small this is not necessary. But you should ensure that there is enough space for your news. Now you have to create directories for all these newsgroups; for maintain- ing news there are several tools available under /usr/lib/news/bin. In directory maint there is the script addgroup ; it is designed for adding a newsgroup. Invoking it like addgroup de.comp.os.linux.x y will create the following hierarchy under /var/spool/news: de/comp/os/ linux/x. Herein all future news belonging to that special group are sorted. ________________________________________________________________________________________________________* *________ | Remember to become user `news' before adding newsgroups! Otherwise * * ||| | * * ||| | news will not have write permissions for these directories! * * ||| |_______________________________________________________________________________________________________* *________|||_ After having assigned all relevant groups you have to make sure that those packages received by UUCP are processed correctly and the articles are put into their directories. Typically this also is done via a cron-job. Your cron-tab, which can be seen by typing crontab -l , for `news' should resemble file contents 17.7.2, page 299. 298 ____________________________________________________________________________17.7.__Brandnew_messages_-_C_News_______________* *______________________|| |||||||| | _______________________________________________________________________________________________________________________|| |||||||| | |||||||| # | |||||||| # Example crontab for the C news Cleanup Release | |||||||| | |||||||| # | |||||||| # Sort new news once every 15 minutes | |||||||| # | |||||||| 0,15,30,45 * * * * touch /tmp/newsrun; /usr/lib/news/bin/input/newsrun | |||||||| | |||||||| # | |||||||| # Prepare new news once an hour | |||||||| # | |||||||| | |||||||| 40 * * * * /usr/lib/news/bin/batch/sendbatches | |||||||| # | |||||||| # Delete old news once a day | |||||||| # | |||||||| | |||||||| 59 0 * * * /usr/lib/news/bin/expire/doexpire | |||||||| # | |||||||| # Run newsdaily once a day at 08:10 | |||||||| # | |||||||| | |||||||| 10 8 * * * /usr/lib/news/bin/maint/newsdaily | |||||||| # | |||||||| # Run newswatch once a day at 13:00 | |||||||| # | |||||||| | |||||||| 00 13 * * * /usr/lib/news/bin/maint/newswatch | mail usenet | |||||||| | ||||||||_______________________________________________________________________________________________________________ | File contents 17.7.2: A news crontab file The first entry sets that all newly arrived news-packages are sorted every 15 minutes. Of course this can be changed if there is only one user who polls only once a day. In this case it is most recommended to launch newsrun directly after polling. sendbatches is responsible for the local news to be packed correctly and to be passed to the newsfeed at next polling. In our given example file con- tents 17.7.2 this would be done once an hour; even here you can do this only once a day just before polling. Since all those articles get older and thus occupy useless disk pace on your system those articles that have reached a certain "age" have to be expired; that's doexpire 's task whose actions are set via /var/lib/news/explist (see manpage of expire (man expire )). newsdaily and newswatch solely serve for supervision purposes. As almost all news programs run in background those tools make sure that nothing hap- pens; if something happens NEWSMASTER is informed. If something worth mentioning to the newsmaster is found a report is sent via email to `news', `newsmaster', or to `usenet'. There should be alias entries for those pseudo-users in your mail-system for those mails to reach the user in question (see manpage of newalias (man newalias )). 17.7.3 Overview of configuration files In table 17.7, page 300 we give a short overview of those files used by cnews ; those files are located under the home directory of the pseudo-user `news', this normally is /var/lib/news. All files contain useful defaults and should only be changed if you really know what you are doing. * * 299 17. Connecting to_the_world:_UUCP,_SLIP,_and_PPP__________________________________________________________________________________* *_______________________________________@ ________________________________________________________________________________________________________* *________ | All changes to those files have to be done as user `news'! * * ||| |_______________________________________________________________________________________________________* *________|||_ ________________________________________________________________________________________________________* *________ active Contains names of local newsgroups and information about existing articles. This file is kept on the latest up- date all the time and is read among others by the news- readers for finding out which newsgroups are available. batchlog Log file for packing local news. More generations of this file are kept. An .o extension is added to older versions. batchparms Determines how locally written news are packed and pre- pared for sending. crontab Example crontab table for user `news'; this table can easily be installed via crontab . errlog Here all error-messages are logged. Older versions are kept a while before being deleted. explist Determines the behaviour or expire . Here you can spec- ify which articles are to be deleted after how many days, whether articles should be archived and more. log Here sorting of all articles is logged. newsgroups List of existing newsgroups including description. organization Should contain the name of your organization (e. g. . your company's name). This field appears in every article wrote on this machine. whoami Name of the local machine ________________________________________________________________________________________________________* *________ Table 17.7: CNews files 17.7.4 Reading news For reading news there are a couple of tools available (e. g. nn , tin , or trn ). Choice of a newsreader is, same as with editors, a personal decision. All those newsreaders can be configured to use a nameserver or to access local spool-directory. There are pre-configured newsreaders in our S.u.S.E. Linux package (series e ). If you want trn to access the local spool directory you should install the package trn-spl.tgz; for use with a nameserver take package trn.tgz package. In the latter /etc/nntpserver has to contain the name where the NNTP-server runs. 17.8 Faxing with Linux There are two alternatives if you plan to use your Linux machine as a faxma- chine. o Either you use mgetty together with sendfax , o or you install the HylaFAX fax server. Here you have a frontend SuSEFax which is designed in Java. 300 ______________________________________________________________________________________________17.8.__Faxing_with_Linux______* *______________________ _______________________________________________________________________________________________________________________|| |||||||| FromS.u.S.E. Linux version 5.0 the package mgetty has been separated in ||| | ||||||||| two packages (package mgetty and package sendfax) as some commands || || |||||||| of package hylafax and package sendfax are identical. ||| | ||||||||_______________________________________________________________________________________________________________||_ The next two sections describe how to setup and configure hylafax and SuSEFax on S.u.S.E. Linux. 17.8.1 SuSEFax - an HylaFAX fax client As already mentioned SuSEFax has been designed in Java. This in particular means that you need to install the "Java Developers Kit" along with package susefax. If you want to use SuSEFax on another platform you should in- stall package susefax and copy everything from /usr/lib/SuSEFax to the destination directory. The wrapper itself The program is launched via a small script. This is called a "wrapper" . It in turn invokes the Java interpreter and sets all needed parameters and checks for settings etc. This script is located in /usr/X11/bin/susefax. You may setup this program system independently (System Properties); ta- ble 17.8 shows all properties that have been setup their meaning, default val- ues and more. If you launch the interpreter without any additional options these settings are used. There is only one parameter (normally) that needs to be set. This is susefax.images. If you run the program on an operat- ing system with multiuser capabilities this is not needed. This in particular means that each user is assigned a home directory ( UNIX systems as well as Windows NT not OS/2 ). If this is not possible on your system you should set the values for susefax.setup.path, susefax.setup.file as well as susefax.phonebook.file as the tool might not work otherwise. Well, it has not been tested yet! _____________________________________________________________________________________________________________________________ ||||||||__Property________________________Defaul|t_value__|________Meaning____________________________________|_____________* *__ |||||||| susefax.setup.path | $HOME | The directory where the con- | |||||||| | | figuration files and the tele- | |||||||| | | | |||||||| | | phone database should re- | |||||||| | | | ||||||||______________________________|_____________________|______side_______________________________________|______________ |||||||| susefax.setup.file | .susefaxrc | Name of the configuration | |||||||| | | | ||||||||______________________________|_________________|__________file_______________________________________|_____________* *_|||| |||||||| susefax.phonebook.file | .susephone | Name of the address | |||||||| | | database (telephone num- | |||||||| | | | ||||||||__________________________|_____________________|__________bers)______________________________________|______________ |||||||| susefax.images |./images | Directory where all neces- | |||||||| | | | ||||||||__________________________________|________________|_______sary_images_are_stored_____________________|______________ Table 17.8: Die System Properties of SuSEFax If you want to change some of these parameters just remove the comments behind the variable names (see file contents 17.8.1). * * 301 17. Connecting to_the_world:_UUCP,_SLIP,_and_PPP__________________________________________________________________________________* *_______________________________________@ ________________________________________________________________________________________________________* *________|| | * * | || # if you want to store the settings other than * * || | # $HOME/.susefaxrc, then you may place another path and/or * * | | # filename here * * | | * * | | * * | | SETUPDIR= # -Dsusefax.setup.path=/where/ever/you/want * * | | SETUPFILE= # -Dsusefax.setup.file=/what/ever/you/want * * | | * * | | * * | | # even the phonebook can be renamed to whatever * * | | * * | | * * | | PHONEBOOK= # -Dsusefax.phonebook.file=asyoulikeit * * | | * * | |_______________________________________________________________________________________________________* *________ | File contents 17.8.1: Section of the wrapperscript: /usr/X11/bin/susefax Handling - the instructions for use This is the main application window after the `Send queue' has been ac- tivated and `Fetch state' has been invoked. `Fetch state' tells you about jobs that are still queued i. e. faxes that were send within the last min- utes. If you did activate the `Receive queue' button all faxes that were received within the last days are listed. You may setup the automatic display by clicking on `Update information' (enter an interval in the box below and confirm by pressing _- ). There is a jobs list that tells you what actions are performed on a double click. This depends on which button (`Receive queue' or `Send queue') has been activated. In `Send queue' you may decide what should happen on a double click; either to get information on the job or to delete it. In `Extras' you may select the language (English or German ). Let's set it up first Well, before you will get any feedback from the server or launch any activities you need to configure the application. This is done via `Settings' in the `Programm' menu . All settings will be saved. If you exit the program all settings will be saved either. Global settings Here we give you a list of the entries and their meaning: User name: Here you may enter the name of the user. This is needed for creating fax covers. EMail: All messages of the fax server is send to this electronic mail address. e. g. if a fax has been removed without having been sent. User account: The fax server is capable of distinguishing between certain users. You may allow or deny access. That's why you need to enter the account name that is known to the server. You may even assign a password to the user. Hostname of the faxserver: This is the name of the host where your fax server runs. 302 ______________________________________________________________________________________________17.8.__Faxing_with_Linux______* *______________________ Figure 17.8: Send queue Automatic sending of faxes: If this button is activated the fax server checks the file given in `Spool file' every couple of seconds. If it has changed since you will get a `Send fax' automatically. This comes in handy if you print to a file from another application. Thus each application is ca- pable of sending faxes provided it is capable of converting it to Postscript (see section 17.8.3). Spool file: Here you need to setup the complete path to the spool file if you want to make use of the "automatic fax sending" feature. Clicking on `Search' offers a file browser where you may setup this file. Fax cover: For automatical generation if a fax cover a special PostScript file is needed. Here you need to enter the complete path. Time zone: This should be the same that you setup on the fax host. Country: Date and time settings depend on this entry. e. g. on your fax cover. Job settings After the global settings have been done you may send your first fax. A fax, or a file respectively is converted to a "job" which waits for being send (perhaps * * 303 17. Connecting to_the_world:_UUCP,_SLIP,_and_PPP__________________________________________________________________________________* *_______________________________________@ Figure 17.9: Dialog for global settings along with some other jobs). There are a couple of job parameters a job. These may be set prior or after queueing the job. Prior to sending the job this may be done in `Job settings' which is accessible via `Extras'. Here we give a short overview of their respective meaning: Notification scheme: Here you may set when the fax server should send a notify message to the user whose electronic mail address was set previ- ously. Here there are four different schemes: o Never (only errors): A message to the user is send if an error occurs that sending of a job failed. o After sending: The user is informed after each successful fax sending. o After an `Re queue': The user is send a message if a fax fails if the remote machine is busy e. g. . o After `Re queue' and `Send:' This is a combination of the last two schemes. Generally the first scheme applies even if you select scheme two. Resolution: Here you give the resolution of the fax to be send. It is given in lines per inch (lpi). Priority: Priority of a job in the queue. The default value is set to 127. This is actualizes if, e. g. , a job fails (due to busy line e. g. ) Maximum amount of tries: Here you give how often the server should try to send a fax. Maximum amount of rings: Here you may give how often the server should try to connect to a remote machine (e. g. if the line is busy). 304 ______________________________________________________________________________________________17.8.__Faxing_with_Linux______* *______________________ Figure 17.10: Dialog for job parameters Paper size: At the moment at the moment there are three supported formats: A4, A3 and "North American Letter". This depends on the format of the PostScript document to be send. If you have activated `Modify job parameters' in the main window you will get a dialog window on a double click. Here you may set the parame- ters `Notification scheme', `Maximum amount of tries' as well as `Maximum amount of rings'. External viewer If the job list shows the received faxes you may view the fax using an external viewer. The fax itself uses the tiffg3 -Format thus your viewer should be ca- pable of displaying this format. This format enables you to combine multiple images within one file. There is a small script that comes with SuSEFax that creates a PostScript file (it uses fax2ps from TIFFSoftware by Sam Leffler [Leffler, 1996b ] (package tiff)). This is handed to a PostScript viewer. The tool looks for gv (Johannes Plass (package gv)) if this is not found it uses GhostView (by Timothy O. Theisen (package gs__x11)) This tool is called docview and may be found in /usr/lib/SuSEFax * * 305 17. Connecting to_the_world:_UUCP,_SLIP,_and_PPP__________________________________________________________________________________* *_______________________________________@ Here are the options for the external viewer Path to temporary files: Here SuSEFax stores the tiffg3 from the server and replaces the $F with the images complete pathname. ________________________________________________________________________________________________________* *________ | The user who invoked SuSEFax needs to have read and write permissions * * ||| | * * ||| | in this directory! * * ||| |_______________________________________________________________________________________________________* *________|||_ Invoking the viewer: Here you need to enter the full pathname to the pro- gram that should display the file. It has to accept the files name as a parameter. Sending a fax Well, after having set the most important settings you should make sure whether you are able to fetch the status settings of the fax server (see page 302). If not you will not be able to send a fax. You might need to check the con- figuration of SuSEFax and/or HylaFAX . Anyway, if everything was setup correctly you should see a dialog when clicking on `Fax send'. Here are the different items and their meaning: Telephone number of the remote host: The telephone number of the re- mote machine. You may select an item from the telephone book if you click on `From telephone book' (of course you will need to assign them first) Document to be send: Here there should be the full pathname to the docu- ment you'd like to send. If you select `Search' you may select a file in the browser. Use fax cover: This may only be activated if you you have setup a fax cover file in the global settings. A fax cover is generated and send. Do not send immediately: If this is activated a dialog window pops up where you may enter the time for the fax to be sent. If you leave this dialog by clicking `Cancel' the fax will be sent immediately after hav- ing left with `Send fax'. For time and date settings please consider the following: o If you have made a selection please confirm by pressing _- . Now the program checks whether this is an existing value. If not so it is converted to a valid date. Name of sender: Here the name that has been set in global settings is used. Name of recipient: No comment... Regarding: Regarding.. An Firma: This should be obvious... Comment: No comment... Selection of `Poll fax' assumes that you have setup a telephone number of the fax polling server. Of course you need to give the name of a file that should be sent. 306 ______________________________________________________________________________________________17.8.__Faxing_with_Linux______* *______________________ Meaning of the job list As already mentioned above you may switch between the contents of the send queue as well as the contents of the received queue. Here we give a short overview on the items: Reception queue Figure 17.11: Example of an reception queue Well, from left to right: first there are the permissions, the size (in bytes) number of pages, the TSI as well as the name of the received fax. The TSI (engl. Transmission Subscriber Identification) is an identification format the user has setup on his machine. This must not be a telephone number. It might be a company's name instead. _______________________________________________________________________________________________________________________|| |||||||| Faxes received may only be viewed by a double click if the fax ||| | |||||||| || | |||||||| server has been setup for every user to read them. For this || | |||||||| to work you need to set the value 0644 for RecvFileMode: in|| | |||||||| || | |||||||| /var/spool/fax/etc/config.device (see section 17.8.4) || | ||||||||_______________________________________________________________________________________________________________||_ Send queue Figure 17.12: Example of an send queue From left to right: first there is the job ID, the priority of the job, the user account, the target telephone number, the time and/or number of tries and the maximum number of rings. If an error occurred you will see the correspon- dent error message. The job ID is assigned automatically by the fax server. You may set a priority a user. Anyway the server may correct this setting. The user account tells you which user has launched the job. He is the only one who can delete the job or change its parameters. The telephone book The telephone book (see figure 17.13)is for administering and maintaining your personal phone numbers. You may list them sort them according to names, surnames, fax numbers as well as company. Double clicking an entry will launch the editor. You may now select another item and its entries will be presented. `Accept entry' add the item to your list required you filled out the following entries: `Surname', `Name' und `Telephone number'. * * 307 17. Connecting to_the_world:_UUCP,_SLIP,_and_PPP__________________________________________________________________________________* *_______________________________________@ Figure 17.13: The telephone book If you have activated the `View entry' mode double clicking on `Send fax' will invoke the `Send fax dialogue' using the entries `Surname', `Name', `Telephone number', `Company' and `Comment' as defaults for the cover page (see page 306). Of course the cover will only be send if you have activated the item `Use fax cover'. If you have launched an activity that disables the selection bar (e. g. by `Sort') you may reactivate it afterwards by selecting an entry. A new entry is added as soon as you click `Add'. `Delete' removes the currently selected entry. The list is only sorted if you explicitly acti- vate the `Sort' button. The order is given in `Sort order' (order is in- creasing alphanumeric). The buttons `Save changes' and `Save & Exit' are disabled as soon as you invoke `Accept entry' in either `Sort' or `Delete'. The telephone book may be used standalone too. Just enter susephone and the wrapper will be called. Obviously you may not send faxes with this tool. ________________________________________________________________________________________________________* *________ | Do not invoke the telephone boot twice a user. If so, please make sure not * * ||| | * * ||| | to save it simultaneously. * * ||| |_______________________________________________________________________________________________________* *________|||_ The faxed form letter The form fax dialog enables you to generate a serial fax list from the tele- phone list. For adding (or removing) an item it has to be marked. You may select either by mouse of by one of the toggle buttons. If `(Toggle' is activated all selected entries will be marked as disabled and vice versa. Same applies to the list. `Send faxes' causes all faxes in the list to be send. You cannot create fax cover for serial faxes. 17.8.2 Automatic generation of the fax cover As mentioned above you will need a PostScript template for automatic gen- eration of fax covers. This by itself is no PostScript file but a template that 308 ______________________________________________________________________________________________17.8.__Faxing_with_Linux______* *______________________ includes certain holders that are inserted automatically during creation of the cover. Creation of a template may be time consuming. If you are familiar withALTEX you may use the package latex-cover7 . Here there is an easy to handle TEX style for creating covers with TEX. The cover of used for this package and for package hylafax has been created with this versatile tool. If you don't want to use it you will have to create a "normal" PostScript file and insert the entries by hand. Which options are known to SuSEFax ? If you're up to modify the TEX document please be aware that SuSEFax only replaces the following macros: __________________________ |__\toperson______|_______ |__\from______________|___ |__\regarding____|________ |__\tocompany____|________ |__\todaysdate__|_________ |__\comments______|_______ If you want to check your self-made template you may use the faxcover tool which is included in package hylafax. This will create a PostScript file out of the template. Now you may print it or view it. You may as well use the Java binary FaxCovergen.class . from package susefax Just enter the following: newbie@earth:/home/newbie > java -classpath /usr/lib/java/lib/classes.zip:/usr/lib SuSEFax.FaxCovergen Now you should see: _______________________________________________________________________________________________________________________ |||||||| | |||||||| Command: FaxCovergen sourcecover.ps docname.ps targetcover.ps | |||||||| | |||||||| | ||||||||_______________________________________________________________________________________________________________ | The source cover is your template. docname.ps is the document that will be send. It will be saved under targetcover.ps. Now you may view either of them. 17.8.3 Fax spooling on UNIX / Linux The spooling mechanism of SuSEFax has originally been designed for the use with Windows . Anyway you may use it on Linux as well. For this to work you will need to install package faxprint, series n (Network support). Now, if you convert /etc/passwd to an PostScript file by entering the command a2ps -nP /etc/passwd | lpr -Pfax there should be a file in /tmp called fax__accountname.ps . accountname is just your login. If this file exists you may enter it as a spool file as given in section 17.8.1 and acti- vate the `Automatic fax' button. 7 ____________________________________________________ It will be installed in /usr/doc/packages/hylafax when installing package hylafax * * 309 17. Connecting to_the_world:_UUCP,_SLIP,_and_PPP__________________________________________________________________________________* *_______________________________________@ ________________________________________________________________________________________________________* *________|| | The spooling mechanism only works if SuSEFax is up and running. If * * |||| || this is the case it checks the time stamp Lastmodified of the spool file * * |||| | regularly and opened if it has been changed. * * |||| |_______________________________________________________________________________________________________* *________|||_ 17.8.4 HylaFAX - distributed faxes Function Installation and configuration of HylaFAX is not covered by our installation support (see section 2.1.2, page 10); you may contact our business support team. This service is not free of charge (see section 2.1.2, page 11). figure 17.14 shows how the fax server works and how to interact with the client. As you may see there are three different ways for communicating with the server. The protocol that is used on port 4557 is still used due to compatibility reasons with older versions of HylaFAX . WinFlex by Peter Bentley e. g. runs on Windows and still uses this protocol. New clients should stack to the new protocol behind Port 4559. This protocol is based on File Transfer Protocol, RFC959. The third available protocol is SNPP (Simple Network Paging Protocol, RFC1861). Figure 17.14: Functioning of HylaFAX -Servers The server itself consists of three different daemons. Each of them is respon- sible for a certain task: 310 ______________________________________________________________________________________________17.8.__Faxing_with_Linux______* *______________________ hfaxd This is the protocol server. It is responsible for the communication between the client and the server. It may be launched standalone e. g. via the Init process or by the Inetd . It uses a file "FIFO special file" with the faxq process itself. faxq This is the so called "Queueing Agent". It is responsible for maintain- ing in and outgoing faxes as well as the job queue. This process runs all the time. Make sure there is only one of them running. faxgetty This tool is responsible for the communication between the server and the modem. As an alternative for faxgetty you may use the faxmodem tool (if you want to send but not receive faxes!). Furthermore you might administrate it via a FIFO file8 . Directory structure The server itself runs in a "Change-Root" mode. By default this Server-Root lies under /var/spool/fax. All processes and the Server-Root directory itself belong to `uucp'. table 17.9 shows a list of all directories that may be found under Server-Root and gives a short introduction. ______________________________________________________________________________________________________________________ archive Here jobs are archived if job archival support was activated. bin Here all scripts that are used by: faxq, faxsend, pagesend and faxgetty. client This is for the FIFO files that serve for communicating with faxq. config Configuration, permissions and user accounts may be found here. etc. dev Since the whole system runs in chroot you will find all char- acter devices needed (null, socksys und tcp). docq This as well as tmp are used for pre checking jobs e. g. .. doneq Jobs that were done but neither archived nor deleted. etc see config. info This is for general information on hosts that are already known to HylaFAX . log Here you will find the logs of sender and reception pollq This is for documents that will be get by polling the server. recvq Outgoing faxes... sendq Incoming faxex. status Status information of the server itself. tmp see docq. ______________________________________________________________________________________________________________________ Table 17.9: The Server-Root directories and their function In addition there are some FIFO files. The file /var/spool/fax/FIFO itself as well as one /var/spool/fax/FIFO.devname per modem that is main- 8 ____________________________________________________ A FIFO file (FIFO = First In First Out) * * 311 17. Connecting to_the_world:_UUCP,_SLIP,_and_PPP__________________________________________________________________________________* *_______________________________________@ tained by faxgetty . devname stands for the device where the modem is con- nected too. Configuration The configuration itself is divided into 2 up to 2+n configuration files. Here `n' gives the number of modems used. In /var/spool/fax/etc you will find the files config and config.device . The latter configures the modem attached to device. So, if there is a modem connected to /dev/ttyS0 the name would be config.ttyS0. General settings are entered in config. These are needed by the scheduler process faxq itself. Modem specific setup may be found in config.device . Settings for the queueing service may be found in config. The files for the protocol server are in /usr/lib/fax/hfaxd.conf. These configuration files are created automatically if you invoke faxsetup after installation. Example configuration Here you see an example session with faxsetup . We will use the data men- tioned in table 17.10: ________________________________________________________________________________________________________* *________ Telephone num- (0)49(0)911-32067289 ber Modem Fax-Class 2.0 ________________________________________________________________________________________________________* *________ Table 17.10: The bold letters are user entries. Configuration of the scheduler o Should an entry be added to /etc/inetd.conf [no]? _- o Country code [1]? 49 o Area code []? 911 o Long distance dialing prefix [1]? 0 o International dialing prefix [011]? 00 o Dial string rules file (relative to /var/spool/fax) ["etc/dialrules"]? _- o Tracing during normal server operation [1]? 527 o Default tracing during send and receive sessions [0xffffffff]? 527 o Continuation cover page (relative to /var/spool/fax) []? etc/cover.templ o Timeout when converting PostScript documents (secs) [180]? _- o Maximum number of concurrent jobs to a destination [1]? _- o Define a class of modems []? "any" _____________________________________________9 Of course you should use your telephone number and enter it correctly in faxsetup . 312 ______________________________________________________________________________________________17.8.__Faxing_with_Linux______* *______________________ o Time of day restrictions for outbound jobs ["Any"]? _- o Pathname of destination controls file (relative to /var/spool/fax) []? _- o Timeout before purging a stale UUCP lock file (secs) [30]? _- o Max number of pages to permit in an outbound job [0xffffffff]? 30 o Syslog facility name for ServerTracing messages [daemon]? _- After the data has been entered you will get a conclusion of what you just entered as given in screen output 17.8.1. _______________________________________________________________________________________________________________________|||| |||||||| | |||||||| The non-default scheduler parameters are: | |||||||| | |||||||| | |||||||| CountryCode: 49 | |||||||| AreaCode: 911 | |||||||| | |||||||| LongDistancePrefix: 0 | |||||||| InternationalPrefix: 00 | |||||||| | |||||||| ServerTracing: 527 | |||||||| ContCoverPage: etc/cover.templ | |||||||| | |||||||| MaxSendPages: 30 | |||||||| ModemClass: "any" | |||||||| SessionTracing: 527 | |||||||| | |||||||| | ||||||||_______________________________________________________________________________________________________________ | Screen output 17.8.1: Configuration of the server After the scheduler has been configured faxsetup requests if you want to setup your modem using faxaddmodem . Obviously you should reply yes . Now the serial line has to be entered (without entering the full pathname) just modem if it is located in /dev/modem. o Country code [49]? _- o Area code [911]? _- o Phone number of fax modem [+1.999.555.1212]? +49.911.3206728 o Local identification string (for TSI/CIG) ["NothingSetup"]? "S.u.S.E. GmbH" o Long distance dialing prefix [0]? _- o International dialing prefix [00]? _- o Dial string rules file (relative to /var/spool/fax) ["etc/dialrules"]? _- o Tracing during normal server operation [1]? 527 o Tracing during send and receive sessions [11]? 527 o Protection mode for received facsimile [0600]? 0644 * * 313 17. Connecting to_the_world:_UUCP,_SLIP,_and_PPP__________________________________________________________________________________* *_______________________________________@ o Protection mode for session logs [0600]? _- o Protection mode for modem [0600]? 0666 o Rings to wait before answering [1]? _- o Modem speaker volume [off]? _- o Command line arguments to getty program ["-h %l dx_%s"]? "-r -b -s %s %l" o Pathname of TSI access control list file (relative to /var/spool/fax) [""]? _- o Pathname of Caller-ID access control list file (relative to /var/spool/fax) [""]? _- o Tag line font file (relative to /var/spool/fax) [etc/lutRS18.pcf]? _- o Tag line format string ["From %%l|%c|Page %%p of %%t"]? _- o Time before purging a stale UUCP lock file (secs) [30]? _- o Hold UUCP lockfile during inbound data calls [Yes]? _- o Hold UUCP lockfile during inbound voice calls [Yes]? _- o Percent good lines to accept during copy quality checking [95]? _- o Max consecutive bad lines to accept during copy quality checking [5]? _- o Max number of pages to accept in a received facsimile [25]? _- o Syslog facility name for ServerTracing messages [daemon]? _- o Set UID to 0 to manipulate CLOCAL [""]? _- The summary given in screen output 17.8.2 is created according to your set- tings entered above. So far for configuration of the scheduler and the server. Now you may set if you want to start faxmodem for each of the configured modems. This serves as an alternative to faxgetty which is only capable of "send only". You may choose either of them. Adaptive Answer Support A rather handy feature of the fax server (faxgetty ) is the so called "Adaptive Answer Support". It enables the server to launch any getty depending on the type of incoming call. This might be a data call as well. For this to work make sure the entry in file contents 17.8.2 is configured (see page 312). Here %s is a substitute for the DTE/DCE between machine and modem. This is set to 38400 bps (bits per second) by default. Some types manu- factured by USRobotics cannot support this rate (see docu [Leffler, 1996a ]). Thus transceiving errors may occur. This may be avoided by decreasing this 314 ______________________________________________________________________________________________17.8.__Faxing_with_Linux_____* *_______________________|| |||||||| | _______________________________________________________________________________________________________________________|| |||||||| | |||||||| The non-default server configuration parameters are: | |||||||| | |||||||| | |||||||| CountryCode: 49 | |||||||| | |||||||| AreaCode: 911 | |||||||| FAXNumber: +49.911.3206728 | |||||||| | |||||||| LongDistancePrefix: 0 | |||||||| InternationalPrefix: 00 | |||||||| DialStringRules: "etc/dialrules" | |||||||| | |||||||| ServerTracing: 527 | |||||||| SessionTracing: 527 | |||||||| | |||||||| RecvFileMode: 0644 | |||||||| DeviceMode: 0666 | |||||||| | |||||||| RingsBeforeAnswer: 1 | |||||||| SpeakerVolume: off | |||||||| | |||||||| GettyArgs: "-r -b -s %s %l" | |||||||| LocalIdentifier: "S.u.S.E. GmbH" | |||||||| | |||||||| TagLineFont: etc/lutRS18.pcf | |||||||| TagLineFormat: "From %%l|%c|Page %%p of %%t" | |||||||| MaxRecvPages: 25 | |||||||| | |||||||| | ||||||||_______________________________________________________________________________________________________________ | Screen output 17.8.2: _______________________________________________________________________________________________________________________ |||||||| | |||||||| GettyArgs: "-r -b -s %s %l" | |||||||| | |||||||| | ||||||||_______________________________________________________________________________________________________________ | File contents 17.8.2: Entry for the Adaptive Answer Support to 19200 in the appropriate modem configuration file. The default getty is mgetty (package mgetty, series n (Network support)). Well for this to work you need to adapt your mgetty configuration file ((/etc/ mgetty+sendfax/mgetty.config)) by the entries given in file contents 17.8.3: _______________________________________________________________________________________________________________________ |||||||| | |||||||| port modem | |||||||| | |||||||| direct y | |||||||| toggle-dtr n | |||||||| | ||||||||_______________________________________________________________________________________________________________ | File contents 17.8.3: Entry in mgetty configuration file The key modem 10 . This is your modem's device name. Make sure that faxgetty as well as mgetty use the same device. Fax Dispatching An interesting feature is the redirection of incoming faxes to an given elec- tronic mail address. You will need to create a file etc/FaxDispatch in the Server-Root directory. file contents 17.8.4 shows an example configuration: 10 ____________________________________________________ If it is /dev/modem refers to a link to /dev/ttySx * * 315 17. Connecting to_the_world:_UUCP,_SLIP,_and_PPP__________________________________________________________________________________* *_______________________________________@ ________________________________________________________________________________________________________* *________|| | * * | || case "$SENDER" in * * || | *0815*) SENDTO=newbie;; * * | | *) SENDTO=FaxMaster;; * * | | esac * * | | * * | |_______________________________________________________________________________________________________* *________ | File contents 17.8.4: Example of etc/FaxDispatch The incoming faxes are identified by their TSI. In our example every fax with 0815 in their TSI will be automatically forwarded to the user `newbie' via electronic mail(as PostScript attachment)11 . Furthermore every incoming fax is redirected to `FaxMaster'. ________________________________________________________________________________________________________* *________ | * * ||| | If you encounter any problems while configuring HylaFAX please have a * * ||| | look at our SupportDatabase (Help system package susehilf, series doc * * ||| | * * ||| | (Documentation)). Enter the keyword "fax"and you will find lots of useful * * ||| | information. * * ||| |_______________________________________________________________________________________________________* *________|||_ 1_____________________________________________1 Of course you may change the file format by modifying bin/faxrcvd 316 Chapter 18 Samba PC Server Using samba developed by Andrew Tridgell] from Australia one can con- vert any Unix machine into a powerful file and print server for DOS and Win- dows machines. Since its start in 1991 Samba has proven to be a stable and reliable product which made its way into companies and serves as supplement or even as a replacement for NetWare - or Windows NT servers. In the meantime Samba has become a rather complex product. We can not cover all details here in our manual but only an overview on its functionality. In /usr/doc/packages/samba many a documents can be found that help you building complex network solutions with Samba. Samba's reference file /etc/smb.conf can be found on an extra man page (manpage of smb.conf (man smb.conf )) 18.1 Introduction Concepts of PCs and Unix differ considerably. That is why we want to give a short overview of PCs architecture with NetBIOS . 18.1.1 NetBIOS NetBIOS is a software interface (API) which has been designed for commu- nication between machines. Here a so-called name service is provided. This enables machines connected to the net to reserve names for themselves. After reservation these machines can be addressed by their names. There is no cen- tral process that checks names. Any machine on the net can reserve as many names as it wants provided the name is not already in use. This dynamic architecture has its origin in PC networks where installation of a new network node had to be made as easy as possible. Configuration of a machine was reduced to giving it a name. Problems with unique names, maximum length of names is 16 characters, could be ignored as the networks were not so extended. Besides the name service there are services for communication itself. There are secure and insecure data streams. These might be compared to TCP and UDP protocols in the Unix world. The higher protocols such as the SMB protocol are the layer on top of them. The NetBIOS interface now can be implemented for different network archi- tectures. An implementation works relative close with network hardware and is called NetBEUI which is often referred to as NetBIOS * * 317 18. Samba PC Server_______________________________________________________________________________________________________________* *_______________________________________@ For addressing the single packages NetBEUI works together with the hard- ware address of the adapter. On the contrary to IPX or IP addresses you can not gather routing information from it. Thus it is not possible to transfer NetBEUI packages via routers and a net running NetBEUI is reduced to the range that can be reached by repeaters and bridges. Further network protocols that have been implemented with NetBIOS are IPX from Novell and TCP/IP. The protocol to lay NetBIOS onto TCP/IP is de- scribed in RFC 1001 and 1002. RFC 1001 furthermore contains a rather good and understandable introduction in NetBIOS' concepts which helps while try- ing to understand services as WINS 1 . The NetBIOS names that are given away via TCP/IP have nothing in common with names as in /etc/hosts or are given via + DNS. NetBIOS is a name space of its own. Well, it is recommended to use names that correspond to DNS hostnames for making administration easier. This is the default Samba uses. 18.1.2 SMB The SMB protocol (Server Message Block) serves for making file and print services in the Windows and LAN Manager available. Then SMB protocol is based on NetBIOS services thus it is comparable to NFS. Here there is no difference to other protocols as NetWare Core protocol. Amazingly Microsoft has released the specifications of the SMB protocols so that others can now support SMB as well. Microsoft has extended the SMB protocol recently and renamed it to CIFS (Common Internet File System). Microsoft now tries to establish CIFS as the internet standard for gaining more support for this protocol2 Well, Samba is a server that implements the SMB protocol under Unix. Samba can turn any Unix machine into a file and print server for most PC operating systems. Samba has been ported to any "intelligent" operating sys- tem. It runs on the real-time QNX as well as on a Cray. Novel has ported Samba to NetWare 4.1 and calls it "Migration Toolkit" providing an easy means for Windows NT users to switch to NetWare. 18.1.3 Clients Except for DOS and Windows 3.1 any current PC operating system supports the SMB protocol for importing and exporting. Windows for Workgroups also supports SMB; in its standard installation only via IPX and NetBEUI. For using Samba which can only provide SMB via TCP/IP additional software has to be installed which (if needed) may be obtained from ftp://ftp. microsoft.com/bussys/Clients. SMB servers provide hardware space to their clients by means of so called shares . Here a share includes a directory and its subdirectories. It is exported _____________________________________________1 WINS is nothing more than an extended NetBIOS name server and not an idea of Microsoft; only the name is new! 2 For those of you who are interested in CIFS specifications have a look at ftp://ftp. microsoft.com/developr/drg/CIFS/cifs; in /developr/drg/CIFS there are the older specifications of SMB 318 _____________________________________________________________________________________18.2.__Installation_of_the_server______* *______________________ by a name and can be accessed by its name. Of course the Share name can be set to any name. It does not have to be the name of the to export directory. A printer also is assigned a name; clients can access the printer by its name. 18.1.4 Permissions An NFS server is configured via /etc/exports. Additional restrictions are only possible a single machine. In NFS this makes sense since it was de- signed for Unix workstations that themselves check permissions and authen- tification. Nowadays in times of Windows and OS where any user can gather "root" permissions the NFS protocol is no longer suitable. NFS clients for DOS must be regarded as immense security gaps! The SMB protocol comes from within the DOS world and directly covers the security issues. Any access to a share can be protected by a password. SMB now has two alternatives for providing this: o Share Level Security A password is directly assigned to any share. Anybody who knows this password can access the share. o User Level Security This variant introduces the user concept in SMB. Every user has to login at the server using his login and password. After successful login the server can now grant access dependent on the users permissions. Distinction between share and user level security has to be set for the server as a whole. It is not possible to export some shares as share level security whereas others are exported as user level security. Per default Samba is set to share level security. Here the home directories of the users are protected by the normal user password. For other shares a user has to be entered by giving the user option whose password is used as protection. The parameter security may be given as security = user in smb.conf. Then users are validated as usual under Unix using /etc/ passwd and /etc/group. Samba offers also a third possibility by providing security = server . If this option is activated Samba validates the user on another (NT) server which has to be set by the option password server . 18.2 Installation of the server Almost anything that can be configured is done in smb.conf. This file re- sembles a Windows .INI file. It is separated in different sections which by themselves contain certain parameters. Generally a share is described per sec- tion whose name is given by the sections name. Furthermore there are three special sections as well. These are [globals] , [homes] and [printers] . In [globals] parameters are set that are not specific to a certain share. If the option [homes] is set any user on the server can access his/her home di- rectory without having to define a home-share for each user. Same applies to printers; all printers in /etc/printcap are accessible without setting them separately. * * 319 18. Samba PC Server_______________________________________________________________________________________________________________* *_______________________________________@ 18.2.1 smb.conf A|simple example file can be seen in file contents 18.2.1. * * | | * * | ________________________________________________________________________________________________________* *________|||| | * * | | [global] * * | | workgroup = workgoup * * | | * * | | guest account = nobody * * | | keep alive = 30 * * | | os level = 2 * * | | * * | | security = share * * | | printing = bsd * * | | * * | | printcap name = /etc/printcap * * | | load printers = yes * * | | * * | | * * | | [homes] * * | | * * | | comment = homedirectory * * | | browseable = no * * | | * * | | read only = no * * | | create mode = 0750 * * | | * * | | * * | | [printers] * * | | comment = All Printers * * | | * * | | browseable = no * * | | printable = yes * * | | * * | | public = no * * | | read only = yes * * | | * * | | create mode = 0700 * * | | directory = /tmp * * | | * * | |_______________________________________________________________________________________________________* *________ | File contents 18.2.1: Example for /etc/smb.conf This /etc/smb.conf provides access to the home directories of the users as well as all printers listed in /etc/printcap. o workgroup = workgroup Any Windows machine Samba is assigned just like a work group where it can be seen in the "network environment". "Work group" is the default setting for Windows for Workgroups. o guest account = nobody Samba needs a user that is listed in /etc/password that only has limited permissions for certain tasks. If public shares (Parameter public = yes ) are defined all operations are executed with this user ID. Even if no public share is defined the account guest account has to be defined for the Samba machine to appear in the networking environment. o keep alive = 30 Windows machines tend to crash from time to time. If they leave open connections behind it could happen that the server recognizes this very much later. If you do not want Samba to waste resources you can tell it to have a look whether the client is still alive by giving the option keep alive = 30 . 320 __________________________________________________________________________________________18.3.__Installation_of_clients____* *______________________ o os level = 2 This parameter os level = 2 assigns that Samba provides browser ser- vices to WfW and Windows 95. If there is a NT machine on the net Samba will not provide these services to it but use the NT machine itself. o security = share See section in permissions. The following three parameters serve for reading /etc/printcap and to export any mentioned printer. The section [homes] assigns parameters for the home directories. These directories are reachable via the users name. o comment = homedirectory Any share can be given a comment with SMB servers which classifies the share. o browsable = no This setting prevents the share homes being visible in the networking environment. Any user can reach its home directory via its name. o read only = no Per default Samba prohibits write access on exported shares. Logged in users should have the possibility to write in their home directories anyway, thus read only = no has to be set. o create mode = 750 Windows machines do not know the concept of Unix permissions. That is why the cannot assign permissions while creating a file. The parameter create mode assigns which permissions should be used while creating a new file. 18.3 Installation of clients I consider DOS, Windows for Workgroups, and Windows 95 to be important clients. First of all it should be mentioned that clients can access Samba only via TCP/IP. NetBEUI and NetBIOS via IPX are not available at the moment. Since TCP/IP becomes more and more "popular" even on Novell and Mi- crosoft it is not certain whether this is going to change in the near future. 18.3.1 DOS and Windows 3.1 DOS has not been designed for networks. For making DOS able to access Samba a protocol stack from card drivers to network redirect has to be loaded. This protocol stack has been written for "naked" DOS and makes plenty use of memory below 1 MB. There is not very much memory left for applications. For those of you who want to continue using DOS programs it is recom- mended to switch to Windows 95 or Windows for Workgroups and run these programs in a DOS box. These two operating systems enable you to use TCP/IP without reducing DOS' conventional memory. This is achieved by a client software that makes use of the protected mode and directly accesses memory above the 1 MB limit. * * 321 18. Samba PC Server_______________________________________________________________________________________________________________* *_______________________________________@ If this is not possible either as a 80386 processors (or less) is used or Windows is not available or the applications do not run in a DOS box, mars__nwe from Novell and its client software should be used. The Novell client software has been designed not to use an immense amount of memory. You can get the NetWare client either from an existing NetWare installation or get a 2 users NetWare 4.1 test license from a bookstore. Many books on NetWare 4.1 also contain this CD. But even DOS cooperates with Samba. The software required can be obtained from ftp://ftp.microsoft.com/bussys/Clients/MSCLIENT. These are two self extracting archives that have to be unpacked on a DOS disk each. On the first disk there is the setup program for automatically initializing the protocol stack. This setup program first selects NetBIOS vis IPX as protocol. This has to be replaced by adding the TCP/IP protocol first and thereafter remove IPX. The setup program configures the networking software for usage of an DHCP server which might not be available under Linux. Therefore you have to set the parameters explicitly in the TCP/IP protocol settings and disable auto- matic configuration by setting Disable Automatic Configuration to 1 . If automatic configuration is not disabled the machine will hang at the next reboot while waiting for the DHCP server. You should keep in mind that MSCLIENT does not give IP addresses as usual (as 192.168.0.20) but as 192 168 0 20 (without the dots). 18.3.2 Windows for Workgroups Windows for Workgroups already has the SMP support built in. This is called a client for Microsoft networks and normally runs on NetBEUI or IPX. TCP/IP connection has to be installed additionally. The TCP/IP protocol stack is freely available at ftp://ftp.microsoft.com/bussys/clients/ WFW. TCP32B.EXE is a self-extracting archive that has to be unpacked onto a disk in advance. Now for the installation. First you have to enter `not listed or actualised protocol' in `drivers' (network setup) which is on disk. After the contents of the disk has been copied the TCP/IP parameters have to be set which is done under `Settings'. Here you need (same as with Linux) the IP address of the machine, a netmask and perhaps an existing gate- way. More options can be set under `Extended'. Thereafter you should set TCP/IP as standard protocol or remove IPX/SPX and NetBEUI. This reduces the amount of memory being used and accelerates the start of Windows. If there are other machines that rely on this protocol you may, of course, not remove them! After installation a restart is obligatory. Thereafter you can access the Linux machine via Samba the same as with other machines. For using a printer on the Samba server one should install the general postscript printer driver of Windows for Workgroups and connect it to the Linux queue which includes the apsfilter recognition possibilities. 322 __________________________________________________________________________________________18.3.__Installation_of_clients____* *______________________ 18.3.3 Windows 95 Windows 95 has the TCP/IP support already built in. Same as with Windows for Workgroups it is not installed by default. For adding TCP/IP you have to choose `Add' (in `System'), `Protocols' TCP/IP from Microsoft. For the rest the same applies as for Windows for Workgroups. * * 323 18. Samba PC Server_______________________________________________________________________________________________________________* *_______________________________________@ 324 Chapter 19 First steps with Linux Since Unix is rather a complex system we can only cover the most important aspects. This book is not aimed at replacing existing literature on Linux (or UNIX)_ this is impossible and this book is not regarded as replacement. Anyway we recommend that ex DOS-users consider buying a book (see page 341). If you call some Unix literature your own (not Linux specific) this might be enough to start, since most of Unix' features are known under Linux as well. Some books that are published under GPL are located on CD 3 as .dvi files in directory /usr/doc/books. You can view those files using xdvi (under X11) and print them using dvips . (Hints can be found in the manual pages). or in ! WhyALTEX (Page page 380)). Well, until you have your book at hand this chapter should give you a short overview and as well support you while "experimenting" with Linux for the first time. Being a Unix novice you should log in with your `user name' (not as root!). First, there are a lot of settings that already have been dealt with for you and second, you're only responsible for your private home directory. This only serves for security purposes; to avoid deletion of system-relevant data. There is no undelete under Linux (as under DOS); so, if you delete system files by accident you probably have to reinstall the whole system. The first steps are just a little complicated since you are the system adminis- trator as well, which is not an easy task for a novice user. 19.1 Logging in, "root", adding users Since Linux is a multiuser system you have to login on the + console anytime you want to use your system. This is called login. You enter your username (e. g. `newbie') and enter your password (only Enter at the first login). login: newbie _- If this was succesful you will find yourself in your "own" home directory (e. g. /home/newbie for the user `newbie'). If you want to leave, you log out using logout ; that's all. `Root' is the System administrator (+ sysadmin)he is allowed to do every- thing. All important system information files can only be changed by `root'. * * 325 19. First steps with_Linux________________________________________________________________________________________________________* *_______________________________________@ Therefore you should only log in as `root' if you plan to configure some- thing or to execute system specific tasks. Never log in as `root' if it is not necessary (most novices forget this)! Thus you protect your system against unwanted damages! Some things only `root' can do: o + mounting filesystems (such as CDs, floppies, installation of software etc). This right can be given to certain users by adding the option user for the corresponding device in /etc/fstab. o Adding and deleting of users. o Installation of a new kernel. o Configuring the system. o Shutting down the system. 19.2 Shutting down and booting ________________________________________________________________________________________________________* *________ | * * ||| | IMPORTANT! You should never turn off the machine while it is running * * ||| | nor press the reset button (+ reset) If you switch it off without bringing * * ||| | * * ||| | it down properly loss of data and damage of your + filesystem is almost * * ||| | certain! * * ||| |_______________________________________________________________________________________________________* *________|||_ For shutting down your system properly there are the commands showed in table 19.1 : ________________________________________________________________________________________________________* *________ shutdown -h now halts the system and (when you see the output: "the system is halted" you can switch off your machine) shutdown -r now reboots the system immediately ________________________________________________________________________________________________________* *________ Table 19.1: Commands for halting your Linux system Commands shutdown can only be invoked by `root'. For shutting down your system log in as `root' and enter the commands shutdown -h now or shutdown -r now . Alternatively you can use Strg + Alt + Entf ( Ctrl + Alt + Del ) (which you might know from booting DOS). This combination does not work if you are currently running X11; nevertheless this method can be used by any user on any virtual console. 19.3 Virtual consoles Linux is a multiuser and + multitasking- system. You will appreciate these features even on one-PC systems: 326 __________________________________________________________________________________19.4.__Adding_and_deleting_users__________* *______________________ In textmode there are 6 virtual + consoles available; you can switch to either of them by using Alt + F1 to Alt + F6 . The seventh console is reserved for X11.1 If you want to switch to a console from X11 without leaving X11 you should use either of: Ctrl + Alt + F1 to Ctrl + Alt + F6 . Alt + F7 brings you back to X11. 19.4 Adding and deleting users You can add users by using the useradd command. Again, this can only be done by `root'. A good alternative for invoking this command by hand is to use YaST for adding users (see page 86). Example: root@earth:/ # useradd -u 300 -g users -d /home/newbie \ -s /bin/bash -m newbie adds user `newbie' with its home directory /home/newbie , user-id 300 . `newbie' belongs to group `users' and uses bash . Now root can give a password to Emil by: root@earth:/ # passwd newbie `root' gives a password to the user `newbie' which he uses for logging in. `newbie' can (and should!) change his password after he is logged in for the first time. The `root' password is given in exact the same manner. If you are adding a user all files from /etc/skel are copied into the user's home directory providing minimal system-wide configuration to all users. Of course each user can adapt those files to his/her personal needs. For comfortably maintaining users it is recommended to use YaST. 19.5 Directories and filenames The character for separating directories (`\' under DOS) is a `/' under Unix. Thus a path being a string where all directories are separated by `/'. One single `/' stands for the topmost directory, the + "root directory". Upper- and lower case are significant under Unix, meaning that Emil names another filename as emil; separating a filename into its name and its exten- sion is not necessary, but there are some programs that expect a certain name (e. g.ALTEX). A nice feature is the globbing feature of the (bash ) shell; just enter the first significant word of a file or command and press 7 ! . This string is now completed by the shell. Pressing 7 ! twice shows all possibilities (if it was not significant). 1 ____________________________________________________ you can assign more consoles via /etc/inittab * * 327 19. First steps with_Linux________________________________________________________________________________________________________* *_______________________________________@ 19.6 Working with directories After having logged in you'll find yourself in your home directory. You can check this by typing pwd (print working directory): newbie@earth:/home/newbie > pwd /home/newbie For changing into another directory there is the cd command (same as under DOS). Typing newbie@earth:/home/newbie > cd /usr/bin newbie@earth:/usr/bin > changes into directory /usr/bin, newbie@earth:/home/newbie > cd latex newbie@earth:/home/newbie/latex > changes to latex provided there is a directory /home/newbie/latex in `newbie'. If you invoke cd without any argument you will be brought back home. This can be reached by typing a tilde (`~') as well. So, typing newbie@earth:/home/newbie > cd ~/latex changes to latex in your current home directory. Same as under DOS, `.' signifies the current directory whereas `..' stands for the parent directory. You can create new directories with mkdir (make directory); The command newbie@earth:/home/newbie > mkdir texts creates text under the current directory. Empty directories can be removed using the command rmdir (remove directory). 19.7 Working with files Until they are replaced by objects or symbols (in future days) files play a vital and central role while working with a computer. Therefore a huge variety of file commands exist under Linux. 19.7.1 Information on files The command ls (list) outputs the contents of your current directory. A list of all filenames and directories is presented. Directory names end with a `/'. Of course you can see any directory if you enter a directory: newbie@earth: > ls /usr/bin Executable programs do not have any extensions as .exe or .com, but an asterisk `*'. You don't have to set it, it is set by the attribute "executable" behind the filename. (see ! Infos on files (Page page 332)) A rather nice option is -l . This gives you a long list of filenames including their owner, permissions, groups and size. newbie@earth: > ls -l This will create an output as screen output 19.7.1. The meaning of the entries in screen output 19.7.1 is explained in table 19.2 328 _______________________________________________________________________________________________19.7.__Working_with_files____* *______________________ _______________________________________________________________________________________________________________________|||| |||||||| drwxr-xr-x 6 newbie users 1024 Mar 21 12:39 ./ | ||||||||| drwxr-xr-x 4 newbie users 1024 Mar 21 17:13 ../ || |||||||| drwxr-xr-x 2 newbie users 1024 Nov 6 16:19 bin/ | |||||||| | |||||||| -rwxr-xr-x 1 newbie users 4160 Mar 21 12:38 check* | |||||||| drwxr-xr-x 2 newbie users 1024 Nov 6 16:23 etc/ | |||||||| | |||||||| drwxr-xr-x 2 newbie users 1024 Nov 6 16:19 sbin/ | |||||||| drwxr-xr-x 12 newbie users 1024 Nov 6 18:20 usr/ | |||||||| | |||||||| -rw-r--r-- 1 newbie users 185050 Mar 15 12:33 xvi.tgz | |||||||| -rw-r--r-- 1 newbie users 98444 Mar 14 12:30 xvnews.tgz | |||||||| | ||||||||_______________________________________________________________________________________________________________ | Screen output 19.7.1: Output of ls -l ______________________________________________________________________________________________________________________ Permissions The first character indicates the file type. `d' stands for directory, `l' for a link and `-' is a normal file. The next 9 characters indicate permissions for the user, the group and all other users (three characters each). `r' stands for read, `w' for write, and `x' for execution. For example, `-rw-r--r--' refers to a file which can be read by the owner, the group and all others, whereas it can only be changed by the owner. See manpage of chmod (man chmod ). Owner The owner of the file. See manpage of chown (man chown ). Group The group the file belongs to. See manpage of chgrp (man chgrp ). Size The file's size in bytes. last change Date when the file has been changed last. Files that have been changed more than a year ago are marked with the year instead of hours:minutes. Name The file or directory's name. ______________________________________________________________________________________________________________________ Table 19.2: Explanation of UNIX file attributes 19.7.2 Wildcards Unlike DOS those + wildcard interpreting options of the shell (e. g. bash ) are fascinating. Wildcards are not only reduced to just `*' and `?'. For example: ls *a???.? gives all files in the current directory having an `a' as the sixth from the last and the second last one being a `.'. Instead of a single character you can give a whole range of different characters. For example letters `a', `b', `c', `d', `e', `f'. This would be done like: ls *[a-f]???.? You can even search in non alphabetical order: ls *[1,3-5,M-P,a,k]???.? * * 329 19. First steps with_Linux________________________________________________________________________________________________________* *_______________________________________@ 19.7.3 Contents of files You can view the contents of a file with less and more page by page. more lets you only scroll downwards. Scroll forward with SPACE and backwards with b . A quite similar command is cat (concatenate) which takes multiple files as arguments and writes the contents directly to standard out (normally your screen). Since most files are bigger than just one screen cat is mainly used to concatenate files via piping standard out. Entering newbie@earth: > cat one two > oneandtwo does not show the contents of one and two but writes the result into the file oneandtwo 19.7.4 Hidden files A special sort of files are the hidden files. Those files begin with a dot and are only seen if you pass the shell the special option -a . In your home directory just enter ls -a ; now you should see all files, even these hidden ones (like .profile or .xinitrc). Hidden files are protected from an otherwise haz- ardous rm * (see section 19.7.5) those files have to deleted separately with rm <.filename> . ________________________________________________________________________________________________________* *________|| | Entering rm .* deletes all hidden files of the current directory! If the * * |||| | option -r is added (recursive; rm -r .* ) all files of the parent directory * * ||| | * * ||| | are deleted as well (they are named `../bla' which is represented by * * ||| | `.*' as well)! * * ||| | * * ||| | So be extremely careful with -r ! * * ||| |_______________________________________________________________________________________________________* *________|||_ 19.7.5 Copying, renaming and deleting of files The command for copying files under Linux is cp : cp source target Now, for copying file /etc/XF86Config in one's own home directory you would use: newbie@earth: > cp /etc/XF86Config ~ Files can be removed using the rm command. A very handy option is -r which deletes recursively (removing all subdirectories and their files as well; this is rather comparable to DOS' deltree ). Entering: newbie@earth: > rm -r bin deletes directory bin and all files and subdirectories within. Please use this option with extreme precaution, since there is absolutely no means of recov- ering (restoring) a deleted file! You can move files by typing mv ; syntax is identical to cp : newbie@earth: > mv xvnews.tgz XVNEWS.tgz Moves file xvnews.tgz to XVNEWS.tgz which is not more than just renaming it. It's going to be more interesting if you move whole directory trees: newbie@earth: > mv bin ~/latex This moves the directory bin (if there is one) to ~/latex. All files that used to be under bin are now to be found under ~/latex. Even this command 330 _______________________________________________________________________________________________19.7.__Working_with_files____* *______________________ should be used with care since it is very easy to move whole trees to places that you do not remember afterwards : : : Moving of a complete directory tree only is possible within one filesystem (one partition). 19.7.6 Searching and grepping files This leads to another useful command: find . For searching all subdirectories of the current directory for the file emil you should enter: newbie@earth: > find . -name "emil" First argument gives the directory where the search should commence. The option name expects a searchstring (wildcards are allowed). Thus, for search- ing all files containing the string `emil' you would have to change the line to: newbie@earth: > find . -name "*emil*" As with all commands we suggest you look at the manual pages for more information. There is a fast way anyway; try finding files with locate . Have a look at the corresponding manual pages too. If you do not want to search for a file but for a string inside a file instead, you should use the grep command. This commandline searches for the string `detective' in the file emil: newbie@earth: > grep "detective" emil Thus you can search vast amounts of data for certain strings very quickly. Any number of filenames can be entered, even searching using wildcards and regular expressions can be used. grep sends its results to standard output; it gives you every location where the string was found. grep knows many options; please have a look at manpage of grep (man grep ). 19.7.7 Symbolic links By means of symbolic links you can give another name to a file. This name "points" to the corresponding file. Imagine that you want to keep different versions of a file but the version actually used should be obtainable by the same name (always). The solution is called symbolic linking which points to the currently in use version. Symbolic links behave just like the file they link too, thus being executable if the "source" file is executable. The entry newbie@earth: > ln -s check.2.4 check creates a symbolic link check which points to check.2.4.1. In your direc- tory this looks like figure 19.7.2: _______________________________________________________________________________________________________________________ |||||||| | |||||||| lrwxrwxrwx 1 newbie users 1024 Mar 21 17:13 check -> check.2.4* | |||||||| | |||||||| | ||||||||_______________________________________________________________________________________________________________ | Screen output 19.7.2: A symbolic link using ls -l Links can be removed by rm . _______________________________________________________________________________________________________________________ |||||||| Here only the link is removed not the source file! || | ||||||||_______________________________________________________________________________________________________________||_ * * 331 19. First steps with_Linux________________________________________________________________________________________________________* *_______________________________________@ 19.7.8 Archiving data and saving For creating and unpacking archives there is the command ( tape archive). Normally compressed archives have the extension z or tar.gz; uncom- pressed archives the extension .tar. Most important uses of tar are: 1. Unpacking archives (e. g. from CDs) newbie@earth: > tar xvfz archive-file.tgz tar unpacks (x ) the compressed (z ) archive archivdatei.tgz (f ) and assigns necessary subdirectories as well. Each file that is extracted is put to standard output (v ). 2. Creating archives newbie@earth: > tar cvfz archive-file.tgz file1 dir1 tar creates (c ) the compressed (z ) archive archive file.tgz (f ), where file1 and all files in dir1 are contained as well. Each file that is packed is put to standard output (v ). 3. Viewing the archive newbie@earth: > tar tfz archive file.tgz tar shows a table of contents (t ) of the compressed (z ) archive archiv file.tgz (f ). Flag z tells tar to create/unpack compressed archives using GNU zip (gzip ). newbie@earth: > tar xvf archive file.tar unpacks the uncompressed archivefile.tar. More information can be found by using newbie@earth: > info tar 19.8 Permissions Only the user `root', being the system administrator, has got unrestricted permissions to all files. Permissions are structured as follows: o Owner permissions o Group permissions o "All others" permissions Each of this categories is represented by three characters; in conclusion with the first character (file type: d, l, or -) this results in a total length of 10. Each flag is represented by a certain character. The possible flags for all three categories are the same. Reading `r', Writing `w' and Executing `x'. If a flag is not set there is a `-' instead of the flag. Let's have a look at the file linux.info: -rw-r-xr-- 1 newbie users 29524 Jun 29 13:11 linux.info Well, this file can be read and written by the owner (newbie); all members of the group users are allowed to read and execute it whereas all other users are only allowed to read it. `-' indicates that this is a "normal" + file. 332 _____________________________________________________________________________________________________19.9.__Manual_pages____* *______________________ Figure 19.1: Overview on file permissions The same applies to directories; if the given file is a directory there is a `d' in front of those 9 characters (d stands for directory). This might look like: drwxr-xr-- 3 newbie users 1024 Jun 29 13:11 info/ `x' in this case means that one is allowed to cd into this directory. 19.8.1 Changing permissions You can change permissions with the chmod command (change mode). The main options chmod needs are the permissions to be changed and a filename. o permissions that are to be changed o a file name. Those three categories are represented by `u' (user), `g' (group, and `o' (others) followed by the corresponding permissions. A `+' or a `-' means adding or removing the corresponding flag. Our example sets permissions of the file linux.info to readable, changeable and executable for group- members: newbie@earth: > chmod g+rwx linux.info If permissions for all categories should be set, then giving the permissions to be changed is sufficient. This command sets permissions to pathlinux.info so nobody has write permissions: newbie@earth: > chmod -w linux.info Those permissions concerning reading and executing are not involved. You can give permissions and remove them within a single command line. This sets the permissions of linux.info to executable, not readable, and not changeable: newbie@earth: > chmod u+x-rw linux.info If you look at the result: newbie@earth: > ls -l linux.info ---xr-xr-- 1 newbie users 29524 Jun 29 13:11 linux.info This is an interesting command in connection with chmod are chgrp (change group) and chown (change owner). 19.9 Manual pages Information about commands, configuration files, and C-libraries can always be found by using the corresponding table 19.3. o man -k searches the headword and gives a list of manual pages found. o man -f searches all sections of manual pages for and lists the manual pages found. * * 333 19. First steps with_Linux________________________________________________________________________________________________________* *_______________________________________@ o man invokes the corresponding manual page for . o man
invokes the manual page for from section . Thus the command man 1 man invokes man 's manual page. o If you have S.u.S.E. help installed, all manuals are available here too; if you are using the X-Window System you might find the program xman of some use. Anyway man is much faster. _______________________________________________________________________________________________________* *_________ man -k searches for and lists the man page found. man -f searches for in all man page sections and lists the man pages found. man launches the man page for . man
launches the man page for from
(e. g. man from section 1 ). _______________________________________________________________________________________________________* *_________ Table 19.3: Invoking the man command In the following sections you can find: _______________________________________________________________________________________________________* *_________ 1 Describes user commands. Some of them are built in commands (bash - und tcsh ); i. e.manpages of bash or tcsh give you help. 2 System calls of libraries. 3 C-library functions. 4 Description of configuration files. 5 Syntax of important files. 6 Description of games. 7 Anything that covers text, text formats etc. 8 System Administrators commands. 9 Description of Linux kernel routines. n n supposed to derive from "new", here other man pages are listed that belong to another section but have been placed here by reasons of tradition or which do not fit in one of the other sections. _______________________________________________________________________________________________________* *_________ Table 19.4: Man page sections 334 ___________________________________________________________________________________________19.10.__System_information_______* *______________________ Please notify that there is not a manual page for every command. Perhaps you can find more information under /usr/doc (e. g. /usr/doc/howto, /usr/ doc/howto/mini, or /usr/doc/packages)) 19.10 System information Sometimes it is important to gather information about the system's state. The commands df, free, top, ps should help. 19.10.1 The df command df outputs information on the used and occupied hard disk space. Output is given as shown in screen output 19.10.1 _______________________________________________________________________________________________________________________|| |||||||| | |||||||| Filesystem 1024-blocks Used Available Capacity Mounted on | |||||||| | |||||||| | |||||||| /dev/sda4 699392 659258 5165 99% /home | |||||||| /dev/sda1 102384 23955 73310 25% / | |||||||| | |||||||| /dev/sdb1 2097136 2070485 26651 99% /usr | |||||||| /dev/sda3 126976 106908 20068 84% /opt | |||||||| | ||||||||_______________________________________________________________________________________________________________ | Screen output 19.10.1: Output of df 19.10.2 The free command free gives information on usage of RAM and swap. _______________________________________________________________________________________________________________________|| |||||||| | |||||||| total used free shared buffers cached | |||||||| Mem: 30900 29272 1628 25608 1012 6412 | |||||||| | |||||||| -/+ buffers: 28260 2640 | |||||||| Swap: 66524 176 66348 | |||||||| | ||||||||_______________________________________________________________________________________________________________ | Screen output 19.10.2: Output of free 19.10.3 The w command w shows all currently active users on your system. This command outputs a rather informative list on certain issues. It shows you all users, how long the system has been up and running, its load and the user's current action. 19.10.4 The du command du gives information on the space that is used by subdirectories and single files. More can be found under manpage of du (man du ). * * 335 19. First steps with_Linux________________________________________________________________________________________________________* *_______________________________________@ ________________________________________________________________________________________________________* *________|||| | 11:19pm up 9 days, 11:13, 13 users, load average: 3.26, 2.80, 2.67 * * | || User tty from login@ idle JCPU PCPU what * * || | root tty2 2:09pm 9:10 -bash * * | | root ttyp1 :0.0 2:11pm 2 4 2 xdvi -s 3 Li * * | | * * | | root ttypa :0.0 11:19pm w * * | | root tty1 2:07pm 9:08 50 (startx) * * | | newbie ttyp0 earth.cosmos.com 11:37am 11 2 2 -bash * * | | root ttyp3 :0.0 3:24pm 4 rlogin helios * * | | * * | | newbie ttyp2 earth.cosmos.com 3:22pm 1 46 2 -bash * * | | root ttyp4 :0.0 3:27pm 1:48 8 bash * * | | root ttyp5 :0.0 5:40pm 5 1 1 telnet earth * * | | newbie ttyp6 venus.cosmos.com 3:53pm 3 5 5 -bash * * | | * * | | root ttyp7 :0.0 4:25pm 6:05 bash * * | | newbie ttyp8 helios.cosmos.com 9:37pm 1:30 1 telnet earth * * | | newbie ttyp9 helios.cosmos.com 9:50pm 1:27 -bash * * | | * * | |_______________________________________________________________________________________________________* *________ | Screen output 19.10.3: Output of w 19.10.5 The kill command Sends signals to currently active processes (+ prozess) Expects a process number (PID) as an argument. This PID can be obtained by ps (see sec- tion 19.10.6). kill is invoked: root@earth: # kill If the corresponding process catches the signal it can be killed by giving the optional parameter -9 Entering root@earth: # kill -9 kills PID definitively. 19.10.6 The ps command ps (process status) shows the processes started by the user. More information on this command can be obtained in manpage of ps (man ps ). ps shows in- formation on running processes of the other users as well. Giving the process id (1. row) it is possible to kill running processes (see section 19.10.5). 19.10.7 The pstree command pstree shows the whole process tree. This is shown in screen output 19.10.4. 19.10.8 The top command Outputs all running processes and their loads and much more. This list is being updated within a given frequency. You can end it using q . 19.11 Filesystem types under Linux There are a couple of + file systems available under Linux. This is shown in table 19.5. 336 ___________________________________________________________________________19.11.__Filesystem_types_under_Linux________________* *___________________|| |||||||| | _______________________________________________________________________________________________________________________|| |||||||| | |||||||| init-+-bash---startx---xinit-+-X | ||||||||| | `-sh---fvwm-+-FvwmPager || |||||||| | |-FvwmWinList | |||||||| | |||||||| | |-GoodStuff | |||||||| | |-xclock | |||||||| | |||||||| | `-xeyes | |||||||| |-color_xterm---bash---xdvi.bin---gs | |||||||| | |||||||| |-2*[color_xterm---bash---vi] | |||||||| |-color_xterm---bash---pstree | |||||||| |-coolmail---coolmail---xterm---pine | |||||||| | |||||||| |-cron | |||||||| |-gpm | |||||||| | |||||||| |-inetd | |||||||| |-kflushd | |||||||| | |||||||| |-klogd | |||||||| |-kswapd | |||||||| | |||||||| |-5*[mingetty] | |||||||| |-4*[nfsiod] | |||||||| | |||||||| |-sh---master---slipto | |||||||| |-syslogd | |||||||| |-update | |||||||| | |||||||| |-xload | |||||||| `-xosview | |||||||| | ||||||||_______________________________________________________________________________________________________________ | Screen output 19.10.4: Output of pstree ______________________________________________________________________________________________________________________ ext Extended Filesystem: Ancestor of the ext2 file systems and without any significant meaning nowadays . ext2 Second extended Filesystem. Linux' standard filesystem. fat File Allocation Table usually this is DOS' filesystem but is used by a couple of operating systems nowadays. ffs Fast Filesystem a filesystem used on Amigas. hpfs High Performance Filesystem this is IBM OS/2 's standard filesystem. It is only supported in read-only mode. iso9660 Standard file system on CDROMs. minix This filesystems has its origin in academical projects on op- erating systems. It was the first file system used for Linux. Nowadays it is used as filesystem for floppy disks. ncpfs Filesystem for mounting Novell-volumes. nfs Network Filesystem. Here data can be stored on any machine in a network and access may be granted via a network. proc Process filesystem (virtual) smb Server Message Block is used by products as WfW , Windows NT und Lan Manager and servers for making files accessible over a network. ______________________________________________________________________________________________________________________ Table 19.5: to be continued... * *337 19. First steps with_Linux________________________________________________________________________________________________________* *_______________________________________@ _______________________________________________________________________________________________________* *_________ sysv Used on SCO-Unix , Xenix and Coherent (commercial Unix systems for PCs) . ufs Used by BSD , SunOS and NeXTstep . Is only supported in read-only mode. umsdos Unix on MSDOS Is applied on top of a normal fat filesys- tem. It achieves Unix functionality (permissions, links, long filenames) by creating special files. It is rather slow. It is required for the demo-mode. vfat Virtual FAT Extension of the fat file system (supports long filenames). xiafs An old filesystem that is hardly used anymore. _______________________________________________________________________________________________________* *_________ Table 19.5: Filesystem types under Linux 19.12 Mount and unmount filesystems Via the command mount (which can only be invoked by `root'), a stor- age media can be linked into the Linux filesystem tree. Two arguments are required by mount ; o Name of the device (corresponds to the device's name; e. g. /dev/hda3) o And, of course, where it should be attached (mounted). Option -t passes the type of the filesystem (see ta- ble 19.5). root@earth:/ # mount -t msdos /dev/hda2 /dosa makes DOS partition /dev/hda2 available under /dosa. Passing the -r option mounts it read-only. Now writing on this filesystem is not allowed. mount protocols the mounted partitions. This protocol can be found in /etc/ mtab; if mount is invoked without any argument the contents of this list is printed to screen. This corresponds to all mounted filesystems. 19.12.1 umount umount removes a partition from Linux' filesystem2 . You can pass either the name of the device or the name of the directory where it is mounted as an argument to umount . So, for removing /dev/hda2 which is mounted under /dosa you can enter either of: root@earth: # umount /dosa root@earth: # umount /dev/hda2 _____________________________________________2 This command used to be called unmount but the `n' got lost somewhere in Unix history. 338 ________________________________________________________________________________________________________19.13.__The_mtools__* *______________________ _______________________________________________________________________________________________________________________|| |||||||| It is important (if you have mounted a disk) to execute umount before you ||| | ||||||||| remove the disk since not all files are necessarily written to disk at this || || |||||||| time; you can lose your data that way! ||| | |||||||| || | |||||||| If there are opened files on this device umount will try to write them || | |||||||| || | |||||||| otherwise it produces an error message. Furthermore you have to leave || | |||||||| the directory you want to unmount. || | |||||||| || | |||||||| If there are still users working (e. g. in a shell) on this path the filesystem || | |||||||| can not be unmounted either. || | ||||||||_______________________________________________________________________________________________________________||_ 19.13 The mtools Especially for using MS-DOS filesystems either on disk or HD mtools (in package mtools, series ap )are at your disposal. All of those little pro- grams try to emulate the corresponding DOS-original as good as possible. All mtools commands are named after their DOS pendant. _______________________________________________________________________________________________________________________ |||||||| You can only use mtools commands if the disk (or HD) is not mounted! || | ||||||||_______________________________________________________________________________________________________________||_ DOS filenames are normally constructed by a device-letter followed by a colon, a subdirectory (evtl .), and the respective filename. For separating only the `\' is used. For separating directories you can either use the `/' as well as the `\'. If you use `\' or + wildcards they have to be put into quotation marks since otherwise they will be misinterpreted by the commandline interpreter (e. g. bash ). An asterisk `*' corresponds to DOS' `*.*'. Instead of using a `/' for passing parameters you have to use the `-'. The standard device for mtools is `A:'. If another one should be used please change to it by typing mcd . Please do not forget to go back to the "root" directory of the device since otherwise no directory tree can be read. Up to now the following commands are supported: ______________________________________________________________________________________________________________________ mattrib Change DOS file attributes (hidden , system , etc.). mcd Change to another directory. mcopy Copy from DOS to Unix. Remember not to forget the target. mdel Deleting a DOS file. mdir Listing a DOS directory. mformat Assigning a DOS filesystem to a low level-formatted disk. (Low-level formatting is done via the command fdformat ) mlabel Renaming a DOS device. mmd Creating a DOS subdirectory. mrd Deleting a DOS subdirectory. mread Reading (low-level) a DOS file into Unix. mren Renaming an existing DOS file. ______________________________________________________________________________________________________________________ Table 19.6: to be continued... * * 339 19. First steps with_Linux________________________________________________________________________________________________________* *_______________________________________@ _______________________________________________________________________________________________________* *_________ mtype Shows contents of a DOS file. mwrite Low-level copy of a Unix file to a DOS filesystem. _______________________________________________________________________________________________________* *_________ Table 19.6: package mtools's commands Device A: is set to a 3.5" device, and device B: to 5.25". Diskformats that are supported are: 720 kB, 1.44 MB, or 360 kB, and 1.2 MB, respectively. Those default settings can be changed in /etc/mtools. They are structured into one line as follows: o Name of the drive (under DOS), e. g. a: o Device file (Linux), e. g. /dev/fd0 o ID (12 floppy disk drives, 16 hard disk drives), o Number of tracks, heads and sectors. By changing those parameters you can use two 5.25" disk devices. You should not enter the same DOS device-letter or device-name twice. 19.14 Overview _______________________________________________________________________________________________________* *_________ cd directory changes to given directory cd .. changes to parent directory cd /directory changes to directory cd ~ Changes to home-directory of the user. cp file__from file__to copies files ln [-s] source linkname creates the symbolic link linkname in the cur- rent directory to source; linkname gives the path where the file can be found. Only sym- bolic links can be created across different filesys- tems. Even directories can be linked via "sym- bolic links" . ls [directory] lists contents of directory (short) ls -l [directory] lists contents of directory (long) ls -a [directory] lists even hidden files. (e. g. .xinitrc in your home-directory) mkdir new__directory creates a new directory less file pages through a file (page up using space , page down using b ) mv file__from file__to moves or renames a file rm file deletes file (including links !); rm -r directory recursively deletes directory. rmdir directory deletes directory (if empty) _______________________________________________________________________________________________________* *_________ Table 19.7: Overview of the most important UNIX commands 340 _________________________________________________________________________________________________________19.15.__And_now?___* *______________________ ______________________________________________________________________________________________________________________ `find . -name' "file" searches all subdirectories of the actual directory for file. `find . -name' searches all files containing the string "*emil*" `emil'. `man command' gives you the manual page of command . `grep string files' searches all files for the given `string' which can, of course, contain + regular expressions as well (see section 19.7.2 or man regexp ). ______________________________________________________________________________________________________________________ Table 19.8: Overview of "search" commands 19.15 And now? The general directories for executables under Linux are: o /bin o /sbin o /usr/bin o /usr/sbin o /usr/X11R6/bin More commands are to be found under The command man gives you detailed information on the given command (see section 19.9) and programs (provided a manual page exists). Output is directed to standard output (which normally is your screen); you can redirect output by means of pipes (`|') to a printer or to a file. This is achieved by entering: newbie@earth: > man -t ls | lpr Thematically sorted introductions about manual pages are found in hypertext- help system of your S.u.S.E. Linux system. Just enter help . Here other links to manual pages can also be found (hypertext). 19.16 Literature recommendations AE. Frisch Essential System Administration Sebastopol Calif., O'Reilly & Ass. 1991 Gilly, D., et.al. UNIX in a nutshell, Sebastopol, Calif., O'Reilly & Ass. 1992 H. Herold Unix-Grundlagen, Addison-Wesley 1992 Kofler, M. Linux, Bonn, Addison-Wesley 1995 G. Perlman Unix For Software Developers, Prentice Hall 1994 K. Pugh Unix For The MS-DOS User, Prentice-Hall 1994 G. Todino, J. Strang, J. Peek Learning the UNIX operating system, Sebastopol, Calif., O'Reilly & Associates 1993 S. Hetze, D. Hohndel, M. Mu"ller, O. Kirch, et. al. Linux Anwenderhandbuch, LunetIX Softfair 1994 * * 341 19. First steps with_Linux________________________________________________________________________________________________________* *_______________________________________@ 342 Chapter 20 Emulators 20.1 The DOS-emulator 20.1.1 Introduction Having a DOS-emulator which enables you to run MS-DOS- programs under Linux is not only of interest for former DOS-users. This short introduction should give you the ability of installing the emulator on your system and give you an overview of how it can be configured. This project is still being developed and the frequency of the patches can grab hold with those of the kernel. We encourage you to experiment with it but please remember that some features are not yet implemented and others are still buggy. An interesting way has been chosen to implement the DOS-emulator; some words of introduction are appropriate. The DOS-emulator is not really an emulator but a BIOS-emulator which en- ables you to really "boot" DOS under Linux. That's why the user needs a DOS-license since the emulator just offers you a virtual machine. This en- ables you to run any DOS that runs on a PC under Linux as long as it doesn't switch to protected mode. Windows 3.0 only runs in real-mode. Although the DOSEMU (since version 0.66.4) includes the FreeDOS kernel it has not reached the reliability and compatibility yet you might expect. Pro- grams running under protected mode under DOS can be run if they make use of the DPMI. VCPI based DOS extender cannot be used up to now. Win- dows 4.1 applications use this VCPI whereas Windows 3.1 itself does not. That is why Windows 3.0 only runs in real mode under DOSEMU. The possi- bility of running the OS/2 Windows kernel (described in certain news groups and mailing lists) is only recommended under reservation (license problems). _______________________________________________________________________________________________________________________|| |||||||| DOS-emulator make extensive use of kernel functions! Using it exces- ||| | |||||||| || | |||||||| sively can cause a freeze of the whole system! Don't use the emulator on || | |||||||| really important machines, only use it if you're absolutely sure nothing can || | |||||||| || | |||||||| happen! || | ||||||||_______________________________________________________________________________________________________________||_ This description is based upon 2.0.30 and the DOS-emulator version 0.66.5 as they are on your CD. Due to the lightning-fast development of this package there might as well be another version on CD which was not available by time of going to press of this book. * * 343 20. Emulators _________________________________________________________________________________________________________________* *_______________________________________@ 20.1.2 Installation of binary package If you have installed the binary-package via YaST you can skip the compiling section. Please be aware that the DOS-emulator recently needed a function which was not available in kernel code, so there was a special module which supplied the necessary functions. Since kernel version 2.0.28 this is not needed anymore, as the kernel itself now provides this functionality. Furthermore it is no longer needed to recompile the DOS-emulator after a kernel reinstall. Now please go on reading section configuration. 20.1.3 Compiling Install the sources onto your system with YaST. If you want to compile log in as `root' and change to the source directory of DOS-emulator (if you have installed the package using YaST this is /usr/ src/packages/dosemu/dosemu-0.64.3.1). You start compilation by entering: root@earth: # cd /usr/src/packages/dosemu root@earth:.../dosemu-0.66.5 # pkgmake extract dosemu-0.66.5 root@earth:.../dosemu-0.66.5 # pkgmake compile dosemu-0.66.5 root@earth:.../dosemu-0.66.5 # pkgmake install dosemu-0.66.5 Please keep in mind that some modules require quite an amount of memory (you should have at least 16-20 MB of swap; RAM and swap-partition). Compilation should be over after a couple of minutes. Please note that not the version of the currently running kernel is of interest but the contents of /usr/src/linux/include, since /usr/src/linux/ include defines the kernel interface during compilation. So make sure that the DOS-emulator suits the actual installed sources. ________________________________________________________________________________________________________* *________ | For this version at least kernel version 2.0.28. Older versions do not work! * * ||| |_______________________________________________________________________________________________________* *________|||_ Furthermore the programs bison and flex have to be installed. 20.1.4 Configuration For being able to use the emulator it has to be configured; this is achieved by two central configuration files, /etc/dosemu.conf and /etc/dosemu. users. Personal settings for each user can be put into local .dosrc. Fur- thermore it is possible to configuration statements via commandline (option -I). For your orientation there are are sample files dosemu.conf and dosemu. users.* under /var/lib/dosemu/etc. You should copy this file to /etc and create a link from dosemu.users.easy to dosemu.users. root@earth: # cd /var/lib/dosemu/etc/ root@earth: # cp dosemu.config dosemu.users.* /etc root@earth: # cd /etc root@earth: # ln -sf dosemu.users.easy dosemu.users 344 _____________________________________________________________________________________________20.1.__The_DOS-emulator________* *______________________ Now you have a good position to start with. If you have installed the binary- package everything will already be in the proper place. It is recommended to leave this file untouched until you definitely know what you do (especially video-parameters). Please do not use the /usr/src/dosemu/Configure script; the current ver- sion is faulty! 20.1.5 Booting DOS under Linux For starting the DOS-emulator for the first time you will need a running DOS on one of your partitions or an bootable disk (apart from system only command.com is needed). Now setup-hdimage creates a bootable HDim- age (virtual drive) of your DOS1 root@earth: # cd /var/lib/dosemu root@earth: # ./setup-hdimage First the script needs to know where to get DOS from. All partitions are checked for DOS and a list of suitable partitions is displayed. _______________________________________________________________________________________________________________________|||| |||||||| checking you partitions, wait... | |||||||| | |||||||| ...done | |||||||| I've seen the following bootable DOS partitions: | |||||||| | |||||||| | |||||||| /dev/hda1 | |||||||| | |||||||| /dev/hdc1 | |||||||| | |||||||| | |||||||| Which one do you want to use as input? | |||||||| Type in the full device name or ENTER to cancel | |||||||| | ||||||||_______________________________________________________________________________________________________________ | Screen output 20.1.1: DOSEMU searches for bootable DOS-partitions Please enter the device of the partition (or /dev/fd0 for floppy). /dev/hda1 The script now starts detecting your DOS and recognizes the following types (manufacturer): o MS-DOS (MicroSoft) o Windows 95 (MicroSoft) o PC-DOS (IBM) o DR-DOS (Novell, Digital Research) o OpenDos (Caldera) o FreeDos (FreeDos Team) If Windows 95 is selected command.com is taken from \windows otherwise from the root directory. If neither of the DOS versions mentioned above was found you are asked to enter the name of the DOS images io.sys and msdos.sys. This is shown in screen output 20.1.2. Please enter, e. g. : * * 345 20. Emulators _________________________________________________________________________________________________________________* *_______________________________________@ ________________________________________________________________________________________________________* *________|||| | Your /dev/hda1 seems not to contain one of the DOSes that * * | || I know. * * || | * * | | * * | | A normal MSDOS system has io.sys and msdos.sys, these are * * | | the files that are needed to boot your system. These * * | | * * | | apparently are different for your system, but I need to * * | | know their names. * * | | * * | | * * | | What is the name of the file that corresponds IO.SYS? * * | | * * | |_______________________________________________________________________________________________________* *________ | Screen output 20.1.2: DOSEMU searches for io.sys ________________________________________________________________________________________________________* *________|| | What is the name of the file that corresponds MSDOS.SYS? * * | | * * | | * * | |_______________________________________________________________________________________________________* *________ | Screen output 20.1.3: DOSEMU searches for msdos.sys ibmbio.com Please enter, e. g. : ibmdos.com Some DOS users use command interpreters (e. g. 4dos.com) that differ from command.com, that is the background for the following question: ________________________________________________________________________________________________________* *________|| | * * | | A normal MSDOS system has command.com as shell. * * | | If you have a different one, please enter the name, * * | | else just type ENTER * * | | * * | | * * | |_______________________________________________________________________________________________________* *________ | Screen output 20.1.4: DOSEMU searches for DOS-interpreters So, if you want to use command.com just press _- , otherwise enter the name of your favourite interpreter. If the script is done you will have an preliminary hdimage file /var/lib/dosemu/hdimage.first; this file is a virtual disk for booting DOS. Configuration of DOS is rather simplistic and does only contain pro- grams that belong to DOSEMU. Among others these are drivers for EMS CD-ROM, mice and many more. This disk is activated by: disk image "/var/lib/dosemu/hdimage.first" in the file /etc/dosemu.conf. Later on if the emulation runs smoothly and you got used to the configuration you might as well try drivers you are used to under DOS. For editing the configuration files config.sys and autoexec.bat on this virtual disk there is a small editor available (ezedit) for doing so. _____________________________________________1 The package mtools, series ap has to be installed 346 _____________________________________________________________________________________________20.1.__The_DOS-emulator________* *______________________ You may even copy files from or to the hdimage by use of the mtools thus enables you to edit files under Linux. For accessing this hdimage use N:. This is already configured in /etc/mtools.conf. Please have a look at manpage of mtools (man mtools ). Now start the emulator by entering: root@earth:/ # dos After a while you should see the DOS bootprompt on your screen. Please be patient, this might take longer than on normal DOS machines (without emulator). Do not be afraid (or switch off the machine) if the BIOS messages of your graphics device appear; after all DOSEMU is an BIOS emulator and starts at a the lowest level. If booting fails just switch to one of the other consoles by entering either of Ctrl + Alt + F1 to F6 and enter: root@earth:/ # dosdebug You should see the debugger screen output 20.1.5 _______________________________________________________________________________________________________________________|||| |||||||| | |||||||| DOSEMU Debugger V0.3 connected | |||||||| - type ? to get help on commands - | |||||||| | |||||||| Reading map file | |||||||| 1900 symbol(s) processed | |||||||| | |||||||| highest address 0814ac90(_end) | |||||||| system state: running | |||||||| AX=0100 BX=00c0 CX=0000 DX=1814 SI=010d DI=a8b3 SP=0a96 BP=0000 | |||||||| | |||||||| DS=0133 ES=0133 FS=0000 GS=0000 FL=7246 | |||||||| CS:IP=fdc8:84f0 SS:SP=0133:0a96 | |||||||| | |||||||| | |||||||| fdc8:84f0 C3 ret | |||||||| | |||||||| | |||||||| | ||||||||_______________________________________________________________________________________________________________ | Screen output 20.1.5: The DOS debugger Enter "kill" and both processes (debugger and the emulator) are interrupted. Just repeat the start-up process. _______________________________________________________________________________________________________________________||| |||||||| || | |||||||| The DOS-emulator expects a resolution of 25 lines by 80 columns. If || | |||||||| you have started you Linux in another text-mode (e. g. vga=extended ) || | |||||||| || | |||||||| perhaps beginning of lines are shifted diagonally. Especially while boot- || | |||||||| ing DOS 7.0 (Windows 95) the logo (in 320x300 graphic mode) de- || | |||||||| || | |||||||| stroys screen geometry set by DOS-emulator. In this case you can acti- || | |||||||| vate graphic mode in /etc/dosemu.conf later on leading to the correct || | |||||||| || | |||||||| graphic mode being set. || | ||||||||_______________________________________________________________________________________________________________||_ Leaving the DOSEMU is done by entering: C:> exitemu There also is a "rude" method for leaving the emulator: just enter Ctrl + Alt + Page # . * * 347 20. Emulators _________________________________________________________________________________________________________________* *_______________________________________@ ________________________________________________________________________________________________________* *________ | Be careful! This can lead to a blocked keyboard and screen. First try * * |||| || exitemu , then the debugger and as a last resort the Ctrl + Alt + Page # * * |||| | method. * * |||| |_______________________________________________________________________________________________________* *________|||_ If you want to switch to another console you should use Ctrl + Alt + F1 to F6 as the "normal" switching Alt + F1 is only used for the emulation itself2 not for switching consoles. If invoking the DOS emulator should fail you might pass parameters so that it presents you with its debug messages. Just invoke it using: root@earth:/ # dos -D+a 2> /tmp/debug After the emulator has finished you may analyze the file /tmp/debug for finding out what went wrong. Normally this will not happen so that you can start adapting DOSEMU to your system. 20.1.6 Booting from a virtual drive After having made your virtual drive bootable you again start the emulator with: root@earth:/ # dos -D-a 2> /dev/null If you change to drive D: (yes, you are right, this exists) you will notice that you find yourself in your Linux' root directory. Well, those long filenames look quite "funny". Nevertheless, if you have mounted your real DOS drive (C:) to /dos under Linux you can access this drive at D:\DOS and start programs. Using an editor you can now edit config.sys and autoexec.bat. It is recommended to replace the virtual drive by the real-mounted drive in autoexec.bat (using LREDIR ). You will need two identical copies of autoexec.bat; one on your virtual drive and one on your real drive otherwise command.com will not find the next line while switching. Here is an example: ________________________________________________________________________________________________________* *________ | * * | | c:\lredir c: linux\fs\dos * * | | * * | | c:\lredir del d: * * | | c:\lredir d: linux\fs\dos-drive-D * * | | * * | |_______________________________________________________________________________________________________* *________ | File contents 20.1.1: Accessing DOS-drives from the emulator If you, e. g. , want to map /usr/dos to F: you enter: c:\lredir F: linux\fs\usr\dos Normal / are to be replaced by DOS-typical backslashes (`\') . Filenames are case-insensitive. Now you have similar conditions as under "native" DOS: C: is C: and D: is D:; including the advantage that either Linux or DOS can access your drive. _____________________________________________2 If rawkeyboard is set in the configuration file for using any combination while working with the emulator. 348 _____________________________________________________________________________________________20.1.__The_DOS-emulator________* *______________________ 20.1.7 Virtual disks An alternative for booting your drive is to virtualize the boot-floppy. Change to /var/lib/dosemu and enter the command: root@earth:/var/lib/dosemu # dd if=/dev/fd0 of=diskimage This writes a original copy of your disk to diskimage3 Now you have to "tell" the DOS-emulator that you plan to boot this virtual disk. In your /etc/dosemu.conf there are (directly after the greeting mes- sage) the following lines: _______________________________________________________________________________________________________________________ |||||||| | |||||||| ## comment out the below line, if you want to boot from a | |||||||| | |||||||| ## virtual floppy | |||||||| #define vbootfloppy | |||||||| | ||||||||_______________________________________________________________________________________________________________ | File contents 20.1.2: Defining a virtual floppy Just remove the `#' in front of the statement define vbootfloppy ; now the necessary statements in /etc/dosemu.conf are activated. For testing you may now start the emulator again. This should be possible without the inserted disk. Please do not hand diskimage to anybody since it contains a bootable DOS! For using your boot-drive as a real drive you should run BOOTOFF under DOS. The counterpart BOOTON sets your virtual disk to on again. 20.1.8 User management and security precautions From version 0.66.5 onwards the DOSEMU inherits an enhanced configura- tion system which improves security and provides quite some flexibility. Con- figuration variables may be checked in /etc/dosemu.conf which are set in /etc/dosemu.users as well. The file /etc/dosemu.users.secure, e. g. , defines a more secure environment than the beginners configuration described above (/etc/dosemu.users.easy. _______________________________________________________________________________________________________________________|| |||||||| | |||||||| root c_all | |||||||| nobody guest | |||||||| | |||||||| guest guest | |||||||| all restricted | |||||||| | ||||||||_______________________________________________________________________________________________________________ | File contents 20.1.3: Secure DOSEMU provided by /etc/dosemu.users.secure The first word in every line defines a user (login); all other words in a line are considered configuration variables and are set. The user `all' refers not to a login but to all users. The first line that matches is used. So if you have set all at the bottom you can handle all other special users above. 3 ____________________________________________________ Command dd is especially useful if you want to create 1 : 1 copies of a disk. * * 349 20. Emulators _________________________________________________________________________________________________________________* *_______________________________________@ Configuration statements are combined in classes (Prefix `c__') that forbid setting not allowed statements outside of dosemu.conf (e. g. in .dosrc of the user). You can decide what is allowed by defining the variables in /etc/dosemu. users. In our example the user `root' is allowed to perform everything, all other users are restricted. Further information can be found in /usr/doc/packages/dosemu/README.config. For a normal user (besides root) to be able to access files on DOS partitions you have to make sure the permissions are already set while mounting the DOS partition. Since DOS does not know about permissions they are globally defined for the whole partition. That is why S.u.S.E. Linux defined a group `dosemu' and creates the ap- propriate entries in /etc/fstab if you have assigned a DOS partition to a mountpoint in your Linux system. If you want to add additional partitions this should be done like: /dev/hda1 /dosc msdos defaults,gid=30,umask=002 0 0 All users that have permissions set to access this partitions in /etc/dosemu. users have to be added to `dosemu' as well. This may be done by YaST in `Group administration' in `Administrating the system'. After having entered the group name, `dosemu', add the users into the list Group member list 20.1.9 Further configuration options Besides linking a DOS partition as virtual drive it is possible to give the DOS- emulator read/write permissions to your drive. You have to uncomment the appropriate line in /etc/dosemu.conf. e. g. : disk partition "/dev/hda1" Generally speaking the order in /etc/dosemu.conf is important. This is how drives are specified. ( C:, D:, E:, etc.). Thus a bootable disk has to be listed first! Use of one partition (/dev/hda1) is preferable to (/dev/hda, use of a wholedisk); if you want to access compressed drives ( Stacker , Dblspace ) you have to use "wholedisk". ________________________________________________________________________________________________________* *________|| | Attention! Never use both methods of linking a DOS-partition at the same * * |||| | time! If you assign a DOS-partition as virtual drive you should not give * * ||| | * * ||| | the DOS-emulator read/write access to this partition since this undergoes * * ||| | kernel filesystems and a loss of data will be inevitable if you write to disk * * ||| | * * ||| | by two different methods! * * ||| |_______________________________________________________________________________________________________* *________|||_ If you have made it so far you should have got a feeling for the emulator's configuration and be able to adjust further configuration yourself. For an in- depth description of the emulator a book would be necessary so that we give only a few more "hints & tricks". By correct configuring video subsection at the beginning of dosemu.conf it is possible to run programs which need direct access to your graphic device; 350 _____________________________________________________________________________________________20.1.__The_DOS-emulator________* *______________________ even high-resolution modes and higher colour-depths are supported. Use of a mouse and connectivity to an existing Novell network are supplied. You can even start DOS under X11; start the emulator with xdos . If other users (except `root') should be allowed to use the DOS-emulator they have to be entered into /etc/dosemu.users. Enter one line per user; each user not mentioned here is not allowed to use the DOS-emulator. Furthermore there are many features provided or implemented, such as use of EMS and XMS, direct access to certain hardware components, and the use of the normal printer spooling from within the emulator. Even dual monitor con- figurations are possible (for CAD-programs or debuggers). There is detailed information in /etc/dosemu.conf about all of these options. Furthermore you are directed to /usr/doc/dosemu for detailed information. /var/lib/hdimage contains a complete DOS-image. It is possible that you will need a new version (e. g. if you plan to install a new DOS version). If this is the case just invoke the script setup-hdimage (make sure you made a backup of hdimage.first). All necessary files are still under /var/lib/ dosemu/. 20.1.10 DEXE, directly executable DOS applications From version 0.66.5 onwards special pre-configured DOS applications can be started directly. These are commands that contain a virtual drive where every- thing the DOS application needs is installed, but nothing more. Additionally there are configuration parameters (not accessible from DOS) that enables you to adapt the application. One of the advantages is that it can be launched rather easily (the S.u.S.E. ker- nels recognize this format and launch the DOSEMU directly). Another is the security: The application runs isolated on the hdimage. Even viruses cannot burst into the frame of DEXE files. Thus your Linux is protected against such intruders. Well, packing of DEXE files is only for the experienced DOSEMU user. We only want to point out to the installed binary packed tetris clone fallout . It only runs under X and is invoked by entering: root@earth: # fallout If you have an (non-S.u.S.E.) kernel installed that has not been configured for binfmt ___misc (section 13.4.4) you can invoke the DEXE as follows: root@earth: # dosexec fallout In /usr/doc/packages/dosemu/README.dexe there is additional infor- mation on DEXE. Here it is described how to create such files. We at S.u.S.E., LLC in cooperation with the DOSEMU team plan to establish an DEXE archive that will contain quite some of the good old DOS public-domain soft- ware that is not ported to Linux. 20.1.11 DOS-emulator and Windows 95 To put one thing straight: Under DOS-emulator you cannot run Windows 95 GUI, only DOS-mode (DOS 7.0). For not letting Windows 95 automatically start the GUI you should proceed as follows: * * 351 20. Emulators _________________________________________________________________________________________________________________* *_______________________________________@ Under Windows 95 you click on `Start', `Shut down', `Restart in MS-DOS-mode'. If you have left the GUI and got presented the DOS prompt you should make msdos.sys (this is a text file) "visible" using the command attrib and replace BootGUI=1 by BootGUI=0 . Furthermore you should ap- pend a line BootDelay=0 . You can attach the key F8 to the DOSEMU to get into the Windows,95 boot-menu. This is done as follows: dos -D-a -I "keystroke \F8;" Otherwise booting under DOS 7.0 using DOSEMU is exactly the same as described above. 20.1.12 STonX - An ATARI ST emulator Figure 20.1: The ATARI ST desktop; TOS 2.06 Overview STonX is a software emulator which runs on Unix workstations under the X-Window system. Same as with the DOS-emulator STonX is designed to use already existing programs furthermore. The emulators concept is to enable _ up to a certain degree _ direct hardware access. The following system components are emulated: o M68000 CPU o 4 MB or 14 MB Main memory o Blitter o MFP (partially) o IKBD (partially) o ST-Shifter (STE partially) o YM2149 Sound Chip o Serial and parallel interfaces Installation Installing the STonX binary package is done automatically via YaST. Well, the emulator still needs the ATARI operating system4 , which is not avail- able for free. If you have an ATARI with the adequate TOS you can use the _____________________________________________4 Version 2.05 oder 2.06 (1.x only after recompiling, 3.x and 4.x do not run yet. 352 _____________________________________________________________________________________________20.1.__The_DOS-emulator________* *______________________ program tosdump.prg to save the contents of your ROM into a file. This pro- gram is located on our CD in dosutils/atari. Under ftp://pc42-c603. uibk.ac.at/pub/TOS there is a TOS 2.06. Please keep in mind that you are only allowed to use it if you have a registered original TOS 2.06. If you invoke this program on an ATARI you will get a file tos.img. This file now has to be copied onto disk using mcopy 5 root@earth: > ei -e stonx -d a: reads this image and installs STonX accordingly. Assume that tos.img is located under /tmp this has to be done as: root@earth: > ei -e stonx -p /tmp. A detailed instruction on ei can be found in section 20.5. The emulator itself now is installed. Now you can start STonX by typing stonx . On our CD additional ATARI software can be found. This is included in package stonxsft and is automatically installed to /var/lib/stonx/ software. This software can be found on C: of the ATARI desktop. Configuration STonX has lots of options that can be configured. An overview on these options is listed in table 20.1. An example: newbie@earth: > stonx -disk a:file -color -noaudio would use /dev/fd0 (floppy a) as drive A:, boot in colour mode and not emulate ATARIs sound. ______________________________________________________________________________________________________________________ -q Quiet (n output) -size x emulated resolution: Width x Heigth -disk : Use file as drive : -fs : Use directory as drive : -colo[u]r Colour emulation -mono Monochrome emulation -usec sets the microsecond interval timer -vbl Sets * usec VBL Interval -timer-c Sets * usec / 2 Timer C Interval -refresh Sets * VBL refresh Interval -para as parallel interface -serial as serial interface -noaudio No sound emulation -noshm Do not use XShm extensions -warmboot Simulated warm reboot -vdi Enable Xlib VDI drivers -private Use own colourmap -help This List:-) ______________________________________________________________________________________________________________________ Table 20.1: List of valid Commandline parameter for STonX 5 ____________________________________________________ IN package package mtools * * 353 20. Emulators _________________________________________________________________________________________________________________* *_______________________________________@ You can also add all these parameters to .stonxrc in your home directory. This file is automatically created for the sample user. Please keep in mind that the "-" prefixes are no longer needed (so, only fs instead of -fs). ________________________________________________________________________________________________________* *________|| | The size parameter is not fully implemented yet. At first it is recom- * * |||| | mended only to try 800x600 resolution. In .stonxrc noaudio and noshm * * ||| | * * ||| | are set; if you already have a kernel with sound support you can remove * * ||| | noaudio. For test purposes you can remove the noshm as well as this speeds * * ||| | * * ||| | up the emulator; if the system crashes, well, then insert it again! * * ||| |_______________________________________________________________________________________________________* *________|||_ Also be aware that STonX only runs with 8bit colour depth and under X11. More documentation on the package can be found in /usr/doc/packages/ stonx. Brand-new information on STonX actual state of development can be get via the WWW from http://www.complang.tuwien.ac.at/nino/ stemu.html. 20.1.13 UAE - Unix Amiga Emulator Figure 20.2: Amiga Workbench Overview UAE emulates an A5000 almost perfect. All kickstart ROM versions are supported, provided they have not been compiled for use on a 68020 CPU. The following system components are supported: o max 11,8 MB RAM (2 MB Chip- and up to 9,5 MB FastRAM) o complete OCS functionality. o partially ECS. Installation For running the Amiga emulator you need a Kickstart ROM as file kick. rom. For copyright reasons this file can not be shipped together with S.u.S.E. Linux. It has to be created using the tool transrom . This program can be found on CD in dosutils/amiga. Comfortable installation is provided by ei . _____________________________________________ 354 _____________________________________________________________________________________________20.1.__The_DOS-emulator________* *______________________ Figure 20.3: GUI of UAE for controlling media. Configuration There are many configuration options or UAE. table 20.2 gives you an overview. ______________________________________________________________________________________________________________________ -h Overview of parameters. -m Mount filesystem at as AmigaDOS drive LW: LW: -M like -m, but read-only LW: -a Do not mount hard disk file automatically; Comes handy for test purposes. -s n Emulate n*256 KB slow-RAM at 0xC00000 -c n Emulate n*512 KB chip-RAM at 0x000000 -F n Emulate n*1 MB fast-RAM at 0x200000 -f n Set number of frames. -D Launch the included debugger -i Announce invalid memory access -[0123] Use instead of df[0123].adf as disk. -r file Use as ROM-image instead of kick.rom. -J Emulate joystick via number block. -S Disable sound-support (if configured) -x Use Cross-hair cursor -l lang Keyboard language DE, SE, US, FR or IT -p command Command for sending output to the printer -I device Name of serial interface (/dev/ttyS1) -d mode Select resolution (see table 20.3). -H mode Number of colours. -C Use correct proportions for pages. ______________________________________________________________________________________________________________________ Table 20.2: List of valid commandline parameters for UAE * * 355 20. Emulators _________________________________________________________________________________________________________________* *_______________________________________@ ________________________________________________________________________________________________________* *________ Valid resolutions 0 (320x200); 1 (320x240); 2 (320x400); 3 (640x480); 4 (800x600) Valid colour 0 (256 colours); 1 (32768 colours); modes 2 (65536 colours) 3 (256 colours using dither) 4 (16 colours dithered); 5 (16 millions of colours) ________________________________________________________________________________________________________* *________ Table 20.3: List of valid graphic modes for UAE ________________________________________________________________________________________________________* *________ | Please be aware that UAE ignores some resolution/colours options since * * ||| | * * ||| | the video hardware is not designed for any combinations. * * ||| |_______________________________________________________________________________________________________* *________|||_ 20.2 Z80 emulators Z80 of Zilog is the best-selling 8 bit micro processor of the world . On June, 3rd 1996 the ZX80 celebrated its 20th birthday which makes it more than a methusalem in computer history. Anyway it is still used nowadays in a couple of devices and even new realisations of products are planned with the Z80 . In the mid-eighties it dominated the market of 8 bit home computers and is still used nowadays in "toys" as the gameboy. Emulator based on this system on S.u.S.E. Linux are: o ZX81 (see section 20.2.1) o ZX Spectrum (see section 20.2.2) o Coleco Vision (see section 20.2.3) o Gameboy (see section 20.4.2) 20.2.1 Z81 - Sinclair ZX81 emulator The Sinclair ZX81 (in the US Timex TS1000) is the predecessor of Sinclair ZX Spectrum. In 1981 the ZX81 Sinclair was released either assembled or as kit. Its look resembles an ancient (prehistoric) computer species: 1 KB RAM, foil-keyboard, black/white block graphics. Nevertheless this machine initiated the "home computer area" and lots of programs were created for this machine. These programs now can be viewed under Linux as well. You can also enter your own programs or try the early ZX80 programming (using its assembler programs). The underlying operating systems has kindly been given to us by Amstrad. Amstrad still holds the copyright for this software. 356 ____________________________________________________________________________________________________20.2.__Z80_emulators____* *______________________ Figure 20.4: One of over 4000 games for ZX Spectrum 20.2.2 XZX - ZX Spectrum Emulator Overview XZX is the emulator for the ZX Spectrum computer of Sinclair. A Sinclair ZX80 emulator. A Spectrum 48 K, 128 K and a +3 are emulated. The following system components are supported either: o Z80 CPU o 16 K, 48 K, 128 K RAM o Interface 1 o Sound (Beep) o Kempston Joystick Installation XZz needs _ dependent on its mode _ certain ROM files. These are in- cluded, thanks to the permission by Amstrad, in the package. Anyway Amstrad keeps the copyright of these files. On http://www.ludd.luth.se/users/alfredo/spectrum.html there are some of the "game classics" in SNA format. Please be aware that those files might be copyrighted and only may be used if you are a registered cus- tomer of their product. _______________________________________________________________________________________________________________________|| |||||||| The files on this FTP server are protected by copyright! Any use without ||| | |||||||| permission of the manufacturer and/or possession of the original software || | |||||||| || | |||||||| is considered illegal! || | ||||||||_______________________________________________________________________________________________________________||_ Configuration ______________________________________________________________________________________________________________________ -help this table ______________________________________________________________________________________________________________________ Table 20.4: to be continued... * * 357 20. Emulators _________________________________________________________________________________________________________________* *_______________________________________@ _______________________________________________________________________________________________________* *_________ -version actual version -visual Use given X-display -font Use given X Font -geometry Give window position -romName Use as 48K ROM image -rom128[01] Use as 128K ROM image -rompl3[0-3] Use as +3 ROM image -libDir Assign library paths -mitshm Use MIT-SHM -flashing Activate flashing [true] -rrShm Refresh screen every interrupt -rrNoshm Refresh screen every interrupt -mono Use monochrome -private Private colour map -intStyle [signal_counter] Art of interrupt implementation -iFreq Interrupts every n commands -kempstUp Kempston up key [q] -kempstDown Kempston down key [a] -kempstLeft Kempston left key [o] -kempstRight Kempston right key [p] -kempstFire Kempston fire key [space] -sound Sound on/off [false] -if1 Interface 1 on/off [true] -if1Rom Give interface 1 ROM -crlf CR/LF translation for RS232 input -strcr Remove CR at RS232 output -m Microdrive cartridge file in _______________________________________________________________________________________________________* *_________ Table 20.4: List of valid commandline parameters for XZX 20.2.3 Adamem - Coleco Vision/ADAM Emulator The company Coleco and their products ColecoADAM and ColecoVision presented a game console as well as a PC like successor (using 64 KB RAM and a Z80 processor). The Linux emulator understands both modes Even nowadays the "Adam" and its 5.25" disks and 20 - 40 MB hard disks is used by freaks. Since its introduction in 1981 lots of programs were written that now can be used under Linux. Installation Adamem needs _ depending on its mode _ different ROM files which nor- mally are included in the source package. Since copyright of these programs could not be figured out you have to get those from the original package.ei is a great help provided you have internet connectivity. Here the archive with the original sources from http://www.komkon.org/\~{}dekogel/adamem. 358 _______________________________________________________________________________20.3.__CBM-8Bit-Emulation_-_Vice_____________* *______________________ Figure 20.5: One of thousands of games available for the Coleco System html. The necessary ROM files are extracted and installed on your system. This process is launched by entering root@earth:/root > ei -e adamem -s Software \footnotesizehttp://www.sponsor.net/\~{}gchance/ColecovisionStuff/ Docs/docs.html gives an overview of available software for this system. 20.3 CBM-8Bit-Emulation - Vice Commodore Business Machines (CBM) has produces several 6 bit ma- chines since the early eighties. The most famous one certainly is the Com- modore C64. Now, vice is four emulators in one. It emulates C64, C128, VIC20 and Commodore PET. 20.3.1 Vice-installation Of course you need the appropriate ROM for the different types. These ROMs can be found on ftp://sunsite.unc.edu/pub/Linux/system/emulators/ commodore or a mirror. There the vice-roms.tar.gz can be found. ei expects the file for installation. Please be aware that the precompiled versions only run under X and only support a 8 bpp colour depth. The different options can be gathered by 8 bit colour newbie@earth:/home/newbie > -help depth All options can be abbreviated as long as they remain unique. Meaning of all these options can be looked up in package vice's documentation which lies under /usr/doc/packages/vice. On an average machine these emulators reach 100% of the original machines. If required you can speed them up (see figure 20.6). * * 359 20. Emulators _________________________________________________________________________________________________________________* *_______________________________________@ Figure 20.6: Elite on C64 running in "mach 2" :-) 20.3.2 C64-emulation After a correct installation of package vice and installation of ROMs (see above) the program can be invoked by x64 . The emulator is started using newbie@earth:/home/newbie > x64 The screen shot shown under figure 20.6 has been achieved as follows: You have to get a "disk-file". This may be done by manually transferring the original disk (or cassette; see vice documentation) or by access on certain internet pages. This file should be app. 174 KB in size6 . The start command for achieving the above result is: newbie@earth:/home/newbie > x64 -floppy ELITE.D64 -speed 200 20.3.3 C128-emulation The C128 was the C64 compatible successor. The emulator emulates "only" the C128 specific features, since there is the x64 if you want to emulate an C64. Launching GO 64 therefore is not possible. The emulator is started via newbie@earth:/home/newbie > x128 Figure 20.7: Message screen of C128. You must see the copyright... There is a wide variety of additional information, programs and data that can be_found_in_the_internet._Some_URLs_of_interest:_ 6 mostly with an .D64 extension 360 ______________________________________________________________________________20.4.__Emulators_for_game_consoles____________* *______________________ o http://www.cucug.org/c64128.html The C64/128 Web Directory. The most well-known pate on C64 and C128. Contains numerous lins to other pages. o http://www.hut.fi/\~{}msmakela/cbm/ The Commodore 8-bit WWW server. Becomes the main source for infor- mation on this architecture. o ftp://ftp.funet.fi/pub/cbm Finnish FTP server with "tons" of software for CBM's 8 bit computers. 20.3.4 VIC-20-emulation This machine was the CBM's first real "home computer" since all other ma- chines (PET) were designed for commercial purposes and were not affordable to the public. The VIC-20 is the predecessor of C64 and C16. A vast amount of information on this model can be found under http://www.mds.mdh.se/\~{}dal95acn/vic20.html. The emulator is launched as newbie@earth:/home/newbie > xvic Figure 20.8: Welcome screen of VIC20. 80 characters /line were a rarity on PCs . 20.3.5 PET-emulation CBM's first computer7 . For lots of computer freaks this was the first contact with a PC. Further information on this type can be obtained under http: //www.ncsc.dni.us/fun/user/tcc/cmuseum/pet.htm. The emulator is invoked using newbie@earth:/home/newbie > xpet It emulates 40 as well as 80 characters (see options xpet -help ). 20.4 Emulators for game consoles 20.4.1 Stella - ATARI VCS 2600 emulator Before ATARI started creating "real" computers they were famous for their game consoles. A quite early and well-known type was Video Console System VCS 2600 which came out in 1977. All games for this console were only available as ROM cartridges. The successors were 5200 und 7800; at the moment there is no emulation availabe for these machines. 7 ____________________________________________________ if KIM-1 does not count * * 361 20. Emulators _________________________________________________________________________________________________________________* *_______________________________________@ Figure 20.9: One of the multiple games for the VCS 2600 system Installation The original console does not contain an operating system. After installation of the emulator it can be started immediately. Under /usr/lib/stella/ ROMS there is a little test program (TEST.BIN). by entering newbie@earth:/home/newbie > cd /usr/lib/stella/ROMS newbie@earth:/usr/lib/stella/ROMS > xstella TEST.BIN it can be launched. It shows psychedelic waves :-) Software There are more than 450 games that run under the VCS 2600, which can be run with Stella. These ROM images can also be found in the internet. ________________________________________________________________________________________________________* *________ | These ROM images are protected by copyright. Using withoud permission * * ||| | * * ||| | is regarded illegal! * * ||| |_______________________________________________________________________________________________________* *________|||_ ei knows all the URLs of software that belong to this emulator. Provided the URL still exists _ and you are connected to the internet _ the software can get by invoking ei like: root@earth:/ > ei -e stella -s We have to point out that you have to delete files if you do not own the original files! 20.4.2 VGB - A virtual gameboy Gameboy of Nintendo is not a very powerful but widely distributed game console. Now there is an emulator available for Linux. This emulator provides high comatibility. According to the author any known game is supported. Vgb does not provide network and sound support up to now. Well, the first is actually being worked on. Interesting pages in connection with this program are e. g. http://www.freeflight.com/fms/ VGB/ as well as a beta Linux port (with sound and joystick-Support) on http://www.geocities.com/SiliconValley/Vista/8340/. 362 ______________________________________________________________________________20.4.__Emulators_for_game_consoles____________* *______________________ Figure 20.10: A gameboy that runs under Linux Figure 20.11: Nintendos Entertainment System under Linux 20.4.3 iNES - Nintendo Entertainment System Emulator Nintendo Entertainment System (NES) is a little game console for home use and is designed for attaching it to a TV. iNES provides this for Linux in a window : : : The emulator can be launched immediately. This means that you do not need an operating system. Only the ROM images (in form f *.NES files) have to be there. iNES is invoked as newbie@earth:/home/newbie > ines Here is the filename of the *.NES file; the games ROM image. 20.4.4 Where to get ROMs? This is the most common question. Furthermore it also is the most incon- venient question that can be asked by a user of emulator software. It does not matter where you ask, be it the S.u.S.E. support or a news-group. Do not * * 363 20. Emulators _________________________________________________________________________________________________________________* *_______________________________________@ forget that ROMs are proteced by copyright and nobody wants to commit a crime by sending you copies. Here some information on how to proceed: o Internet search engines The first (and often the only) choice for finding files. A clever request to one of the big engines ( Lycos , Excite etc.) often gives a useful result (e. g. Kickstart AND Amiga). A couple of search engines is given on http://home.netscape.com/escapes/search/. o Suitable WWW server We already gave some hints where certain ROM files can be found. Some web-servers are a good start for this search. The acknoledged Node 99 can be found on http://206.101.78.248/node99/. What you should avoid; what you should never ever do: o Ask the author of the emulator for ROMs. The authors will never fulfill this request and (at best) ignore your question. Some authors announce you to an "abusers" list where your electronic mail address is presented to public. It already happened that authors stopped writing an emulation because thousands of people pledged them for a ROM. o Ask S.u.S.E. support for ROMs. 20.5 ei - Installation of emulator-firmware The instalation script ei serves as an interface for installing the needed ROM im- ages of the emulators. These images normally contain the operating systems of the emulated machine. Since we can not provide these ROM s due to copyright restriction you have to get these files your own. Most of the data can be found onthe internet. We point out that this is only for those customers that have a registered license of the software (because it is not easy to port 15 year old software from proprietary machines onto modern machines.) This documentatin as well as ei (emu-install) should help you installing ev- erything needed easily. 20.5.1 Invoking ei Whenever you launch ei without parameters or with the -h option this mes- sage appears: The options in detail: o ei -e uae -d a: Expects kick.rom in drive a: incl. the AMIGA Kickstart ROM. o ei -e stonx -p /tmp Expects ATARI TOS (tos.img) in /tmp. Please keep in mind that STonX has been precompiled for versions 2.0.x. o ei -e vice -d a: Expects vice.tgz in floppy drive A: which contains ROM files for CB machines. 364 _________________________________________________________________20.5.__ei__-_Installation_of_emulator-firmware_______________* *____________________ _______________________________________________________________________________________________________________________|||| |||||||| | ||||||||| Emuinstall - anCoinstallationpyprogramriforghvarioust emulators(cunder) linux1996 S.u.S.E GmbH, Fuerth |Germa* *ny.| |||||||| | |||||||| Written by Richard Jelinek (rj@suse.de) | |||||||| | |||||||| Arguments: | |||||||| ei [-h] -e MODE -p EI_PATH | -d EI_DISK | |||||||| | |||||||| -d,--disk use disk instead of list of paths. | |||||||| DISK is either a: or b: | |||||||| -h,--help this help | |||||||| -e,--emu which ROM to install. Possible values for MODE: | |||||||| | |||||||| amiga, uae, atarist, stonx, commodore, vice | |||||||| spectrum, xzx, all | |||||||| -p|--path directory where to search for ROM-images | |||||||| | |||||||| | ||||||||_______________________________________________________________________________________________________________ | Screen output 20.5.1: ei help o ei -e adamem -s If there is a connection to the internet this gets the ROMs needed for the Coleco-emulator. * * 365 20. Emulators _________________________________________________________________________________________________________________* *_______________________________________@ 366 Chapter 21 The vi editor Using vi is a little difficult to get accustomed to. It is preferred to other edi- tors since it is installed on every standard Unix system and comes along with Linux' standard installation. On the other hand its use is well defined, thus this can't lead to misunderstandings. This little help shall enable you to edit your configuration files using vi . See [Lamb, 1990 ]. Concept: There are three modes available in vi : o Command mode Every pressed key is interpreted as part of a command. o Insert mode Pressed keys are interpreted as text. o Last line mode For more complex commands that have to be edited in the last line. The most important commands of the command mode are: ______________________________________________________________________________________________________________________ `i' Changes to input mode (Characters are entered at the actual cursor position) `a' Changes to input mode (Characters are entered after__ the actual cursor position) `A' Changes to input mode (Characters are appended at the end of line) `R' Changes to input mode (overwrites old text) `r' Changes to input mode for overwriting one___ charac- ter. `s' Changes to input mode (character the cursor is own will be overwritten) `C' Changes to input mode (rest of line is replaced by the new text.) ______________________________________________________________________________________________________________________ Table 21.1: to be continued... * * 367 21. The vi editor_________________________________________________________________________________________________________________* *_______________________________________@ _______________________________________________________________________________________________________* *_________ `o' Changes to input mode (after the actual line a new line is inserted) `O' Changes to input mode (before the current line a new line is inserted) `x' Deletes the current character `dd' Deletes current line `dw' Deletes to the end of the current word `cw' Changes to input mode (rest of the actual word is overwritten by the input) `u' Undoes the last command `J' Appends line to follow actual line `.' Repeats the last command `:' Changes to last line mode _______________________________________________________________________________________________________* *_________ Table 21.1: Simple commands of vi All commands can be preceeded by a number; this number sets how many times the given command should be executed; thus entering `3dw' deletes three words at once and `10x' deletes 10 characters, `20dd' deletes 20 lines respectively. The most important commands in last line mode are: _______________________________________________________________________________________________________* *_________ `:q!' quits vi without saving changes `:w [file]' saves in [file] `:x' saves changed file and quits `:e [file]' edits [file] `:u' undoes the last editing command _______________________________________________________________________________________________________* *_________ Table 21.2: Complex commands of vi Typing ESC in input mode changes to command mode. 368 Chapter 22 emacs - more than just an editor The choice of an editor can become a religious question. There are mainly two opposite groups: the stubborn vi users and the increasing amount of emacs adherents. Since vi was created when there were no graphical in- terfaces at all it is mostly used by users that keep sticking to this age of com- puting. Well, some vi addicts nowadays confess that emacs is really easy o handle (which for itself is nothing special whatsoever) and much more pow- erful. 22.1 Basics That emacs is more than a simple editor is proven by people that execute any task available on their machine with this program. You can compile, read and maintain mails, use ftp very easily, read news with emacs and many more. There are extensive commands at your disposition (search and replace using regular expressions). Moreover emacs can be programmed using its own lan- guage (Elisp: a Lisp dialect). Quite an amount of emacs itself is programmed in Lisp. Thus you can easily adapt emacs to suit your personal needs. Due to this versatility there are plenty of emacs modes for any language which makes handling of them very easy. After all this praise we do not want to hide the disadvantages of this famous tool. The system resources it eats are quite big (the emacs binary itself needs 1.3 MB). Since a great amount has been programmed in Lisp it is not as fast as other (smaller) editors. The first item gas been removed in the meantime since there are fast and small emacs clones (e. g. jove and jed ). For those who just want to make a small change to a configuration file, but do not want to miss emacs handling, they can use one of the smaller ones instead without having to load GNU's monster emacs . 22.1.1 Invoking and handling emacs is invoked (under X) using: user@earth:/home/user > emacs If you supply emacs with one or more names at start-up time these files will all be loaded into different buffers. Most of the handling in emacs is done via keyboard. The most important commands can alternatively be started via mouse. We just cover keyboard handling since mouse handling should be self-explanatory and because most users do not use the mouse. It is rather * * 369 22. emacs - more_than_just_an_editor_________________________________________________________________________________________* *_______________________________________@ tedious to change from mouse to keyboard all the time at writing. After launching emacs the copyright messages and some helpful commands are presented. The second most line is the status line. Here the name of the buffer (*scratch* , since the buffer does not have a name yet) and the mode where the buffer is in are presented ( in this case this is Lisp Interaction). The line below the status line is the mini-buffer where commands are entered. 22.1.2 Entering commands Most emacs commands are launched via the Ctrl or Esc key. In Emacs they are spelled as C-h ; this means simultaneously pressing Ctrl and h . M-h means press and release the Esc key followed by h . After having pressed and released Esc the message "ESC" appears in the mini-buffer. If X-Windows has been configured to assign the left Alt key to ESC commands as M-h can be entered as Alt + h too. 22.2 Basic commands 22.2.1 On-line help The most important command for the emacs novice is C-h . This is for start- ing the on-line help system. Entering ? gives you a list of available options that can be scrolled using space. For getting accustomed to emacs it is recom- mended to call help-with-tutorial (by entering t ). This is an interactive tutorial showing you how to handle everything using examples. The tutorial file is copied and can be changed in any way by the novice user to learn han- dling. After having worked through the tutorial you should know all the basic commands of emacs . 22.2.2 Cancel and quit The most important command besides starting is ending: C-x C-c . For can- celling commands (e. g. if an entry is requested in the mini-buffer and the cursor automatically jumps there) you use C-g . Repeatedly pressing C-g and C-x C-c should enable you to leave emacs . 22.2.3 Undo Command C-x u (undo) lets you undo the last command. Pressing C-x u repeatedly lets you undo step by step. The variable sets how many bytes can be undone (default is 20000). 22.2.4 Cursor control, kill and yank Those of you who read the manual thoroughly might have recognized that any cursor movement can be performed via keyboard commands. This of course is still useful on terminals without cursor keys. Anyway mainly the cursor is moved around by the mouse or the keyboard. The keys 7 !(Backspace), Del (Delete) oder C-d also can be used for deleting single characters. The expert might press C-d faster than Del . This also is useful if you are handling emacs via a network connections where key-mapping is different (or not fully 370 ____________________________________________________________________________________22.3.__Loading_and_saving_texts_________* *______________________ configured). C-v and M-v or Page " and Page # can be used for page-up and page-down . C-k deletes all characters to the right of the cursor and saves them in the kill- ring. They can be reentered anywhere by C-y (yank). Frequently pressing C-k saves all strings in the kill-ring. By pressing C-y one can insert the text back again. M-w saves all text between the mark and the cursor-position into the kill-ring without deleting the text. 22.3 Loading and saving texts C-x C-f (find) loads a file into a buffer of its own. The cursor jumps to the mini-buffer expecting a file name. If a non-existent file is entered an empty buffer is opened. If you save the buffer this file is created. If you want to load an existing file use the 7 ! key for globbing. C-x i (insert) lets you insert a file into the current buffer at the current cursor position. Files are saved using C-x C-s (save). If you want to save the file under another name type C-x C-w (write). 22.3.1 Backup-files emacs automatically creates a backup-file. If the file is deleted by accident or another disaster happens you can rely to this file. The backup-file has got exactly the same name but with an `~' attached at the end. Thus the backup name of filename is filename". So do not be astonished if every file exists twice while working with emacs . 22.4 Buffer and windows Any text in emacs is in a buffer of its own. You can work with as many texts as you like at the same time. C-x b lets you jump between buffers. C-x C-b shows a list of all existing buffers. In emacs you can use "text"-windows. Each of these windows contains a buffer. C-x 2 splits the window horizontally. So you can edit or view dif- ferent (or the same) buffers at the same time. Quite helpful is C-x o (other) which lets you jump to the other window. C-x 3 splits the window hori- zontally. C-x 1 goes back to one window and C-x 0 removes the current window without closing the buffer. 22.5 Search and replace C-s (search) lets you search a string. After having entered the command itself all other keystrokes are regarded as text to be searched. As you enter the letter the cursor jumps to the next occurrence of the given string. Pressing C-s (search) again jumps to the next suitable string. For deleting reversely you enter C-r (reverse). This search is an incremental search. It even is possible to delete characters from the search string. The search can be cancelled by either C-g or a cursor key. * * 371 22. emacs - more_than_just_an_editor_________________________________________________________________________________________* *_______________________________________@ Start of the search is marked for you. So if you want to jump back to where you were before just enter C-x C-x . Search and replace is done via the commands M-x replace-string , M-x query-replace , M-% respectively. The latter differs in so far as it lets you decide whether you want to replace or not (so if you are unsure you should use this command). Both commands are invoked via M-x . As you have seen some commands have key combinations as well. After having entered M-x the cursor jumps in the mini-buffer and expects the name of an executable command. Most interactive commands can be launched that way either. Quite often it is useful to use M-x for finding out whether a certain command exists. Globbing is done, same as with files, by 7 ! . 22.6 Keyboard macros By means of keyboard macros key-combinations can be saved and executed any time. Making (intelligent) use of macros can save you quite an amount of (programming) time. To start recording a macro type C-x ( Everything entered after this command is saved into the macro. C-x ) lets you stop recording. and C-g cancels it. 22.6.1 Executing and saving macros Macros that have been defined can be executed by C-x e (execute). This command always launches the macro that was defined last. If you want to use more than one macro you have to give them names. M-x name-last-kbd-macro gives a name to the last macro; M-x lets you execute this macro. For further use it is necessary to save these macros. For saving macros you use insert-kbd-macro . This inserts the macro definition at the cursor position. It is recommended to create a file of its own for macros; all macro definitions are written in Elisp. This Elisp file can be loaded by M-x load-file . Im- mediately after having load this file you are able to use your macros. The other idea is to save it into ,/.emacs 1. Thus these programs are available immediately. A quite useful combination in connection with macros is C-u for repeating a command -times. For example C-u 5 l writes the letter `l' 5 times. For repeating it 20 times it must be C-u 20 l . ________________________________________________________________________________________________________* *________||| | * * ||| | On S.u.S.E. Linux there is a ,/.gnu-emacs and a ,/.xemacs-options * * ||| | besides the usual ,/.emacs . This unfortunately is necessary as the Lisp * * ||| | * * ||| | syntax of GNU-emacs differs from X-emacs and the precompiled lisp * * ||| | files of the two editors have an incompatible binary format. So if there is * * ||| | * * ||| | no ,/.gnu-emacs in your HOME after an update you should save your * * ||| | ,/.emacs and copy /etc/skel/.emacs and /etc/skel/.gnu-emacs * * ||| | * * ||| | to your HOME. * * ||| |_______________________________________________________________________________________________________* *________|||_ _____________________________________________1 $\sim$/.emacs and files loaded by it contains emacs commands that are launched when emacs is invoked. 372 _____________________________________________________________________________________________________________________22.7.__* *Info__________________ 22.7 Info Info is an hypertext documentation system integrated into emacs . It is in- voked by entering M-x info or C-h i . h now gives you help on info. For selecting an item in info you have to move the cursor to the corresponding line and press _- . m is for directly selecting an item. Alternatively it can be selected by m and entry of the menu item. u moves you up one node upwards, p shows the previous node, l the last (which not necessarily has to be the pre- vious). Selecting an reference node and pressing _- moves you that point. d always brings you back to the toplevel menu and last but not least q quits info . Many GNU-programs (including emacs and info ) are documented in info. 22.7.1 More emacs In info there are lots of functions and settings described for emacs which can not be discussed in this book. For those who are interested just make use of the real good documentation in info . We recommend having a look at the following topics: Dired, VM, GNUS and Ange-FTP. Furthermore in /usr/share/emacs/19.34/etc/refcard.tex.ps there is a emacs reference-card that might be of use to you. 22.8 Literature Some books on emacs : Cameron, D. / Rosenblatt, B. Learning GNU Emacs, Sebastopol, Cal., O'Reilly & Associates 1991 Schoonhover, M. / Bowie, J. GNU Emacs, Reading, Mass., Addison-Wesley 1992 Grieger, W. Wer hat Angst vorm Emacs?, Bonn, Addison-Wesley 1994 22.9 Command overview (selection) ______________________________________________________________________________________________________________________ Stop, Undo, Redo C-g Cancels a command C-x u Undoes the last command C-u Repeats - times Help C-h ? Help menu M-x info Hypertext documentation system C-h i Hypertext documentation system M-x apropos Online search a keyword ______________________________________________________________________________________________________________________ Table 22.1: to be continued... * * 373 22. emacs - more_than_just_an_editor_________________________________________________________________________________________* *_______________________________________@ _______________________________________________________________________________________________________* *_________ C-h k Explains the function of Cursor movement C-v Page down M-v Page up C-M-v Other window page up M-f Word forward M-b Wort backward M-< Cursor at beginning of text M-> Cursor at end of text C-a Cursor at beginning of line C-e Cursor at end of line M-a Cursor at beginning of paragraph M-e Cursor at end of paragraph M-m Cursor onto first character that is no white space C-x C-x Change cursor and mark C-l Centre cursor position in window Deleting and moving C-k Deletes to line end C-y Inserts deleted text C- space sets a mark C-x h Marks the whole buffer C-w Deletes from cursor to mark M-w Deletes from cursor to mark into kill- ring M- space Removes blanks + Tabs except for one C-x C-o Deletes blank lines except for one M-d Deletes to end of word M- Entf Deletes to beginning of word M-k Deletes to end of phrase M-z Deletes to (including) File commands C-x C-f Load file C-x C-r Load file read only C-x C-s Save file C-x C-w Save file under another name C-x i Insert file into text Search and replace C-s Search forwards C-r Search backwards _______________________________________________________________________________________________________* *_________ Table 22.1: to be continued... 374 ___________________________________________________________________________22.9.__Command_overview_(selection)_________________* *___________________ ______________________________________________________________________________________________________________________ C-x C-x Go back where search started M-x replace-string Replace string M-x query-replace Query replacement of string M-% Query replacement of string Buffer and windows C-x b Changes to another buffer C-x C-b Shows a list of all buffers C-x k Removes current buffer C-x C-q read only switch for current buffer C-x 2 Splits current window horizontally C-x 3 Splits current window vertically C-x 1 Removes all window except for the current one C-x 0 Removes the current window C-x o Changes to another window C-x > Scrolls horizontal to the right C-x < Scrolls horizontal to the left C-x ^ Enlarges current window vertically C-x + Adjusts all windows C-x - Reduces window according to its con- tents C-x { Reduces window horizontally C-x } Enlarges window horizontally Transform M-l The following word to lower-case M-u The following word to upper-case C-x C-l Marked region to lower-case C-x C-u Marked region to upper-case M-c First character upper-case, rest lower- case Miscellaneous Shift * - F5 Shows line number C-t Change characters at cursor C-x C-t Change lines at cursor position C-q Insert character, can be octal C-x = Info on character at cursor position C-x f Column for carriage-return at cursor position C-x TAB Indent the marked region M-i Fills with blanks to next tab stop M-! Executes shell commands ______________________________________________________________________________________________________________________ Table 22.1: to be continued... * *375 22. emacs - more_than_just_an_editor_________________________________________________________________________________________* *_______________________________________@ _______________________________________________________________________________________________________* *_________ Esc j Executes shell command, uses se- lected region as input M-= Cunts lines and charachters in marked region Keyboard macros C-x ( Starts recording a Aufnahme macro C-x ) Stops recording a Aufnahme macro C-x e Executes the last defined macro M-x name-last-kbd-macro Gives a name to the last macro M-x insert-kbd-macro Inserts a macro in current buffer _______________________________________________________________________________________________________* *_________ Table 22.1: Selected emacs commands 376 Chapter 23 WhyA LTE X? TEX andALTEX have become a de-Facto standard in science as well as in pub- lishing houses. The reason lies in its enormous capabilities of manipulating and formatting texts. There are many fonts available at highest PostScript quality (mathematic, Cyrillic, arabica etc.). Various graphic formats can easily be integrated. And, although at the beginningALTEX might look strange or cryptical, really, it is not. The longer you work with it the more you benefit from it and use it as an easy means of designing your texts. ALTEX itself provides some macro packages for text formatting (such as ar- ticles, books, letters etc.); this reduces layout problems and ensures that all documents use the same format. This e. g. is useful if many authors cooperate. . Normally you just enter plain ASCII text without taking care of formats. You can create your own macros if you encounter formats that you want to use more often. This can reduce a great amount of work. Macros are just typed into the document as if they were normal text. 23.1 Maintenance of TE X The TEX-package that comes with your S.u.S.E. Linux is a widely spread package in the UnixTM world: teTEXby Thomas Esser. Along with teTEX there comes a couple of documentation under /usr/lib/teTeX/texmf/ doc/ on TEX, ALTEX, and some layout styles. Besides the extensive TEX- system including BibTEX, makeindex , dvips , xdvi , MetaFont and various others there is a well-designed maintenance tool. texconfig enables you to maintain and configure the whole teTEX-system. For example, if you invoke texconfig and enter `MODE' you can select the default font that suits your printer and much more. It really is recommended to read the manual page of texconfig . Just launch texconfig and let yourself be guided through the menus. There is another tool, texhash . When invoked it creates a hash- table (ls-R(5) ) which contains the whole directory under /usr/lib/teTeX/ texmf/. Most of teTEX's tools consider this hash-table to find files under /usr/lib/teTeX/texmf/. After installation of new styles, classes, fonts, or graphics you should always start texhash afterwards. * * 377 23. WhyALTE X? _________________________________________________________________________________________________________________* *_______________________________________@ 23.2 Creating a document With this little example we can only cover a small part ofALTEX's capabilities. There already are plenty of good books on this subject [Goossens et al., 1993 ] and [Lamport, 1994 ]; we will not add one but give you an every-day's use of TEX. We concentrate on the latestALTEX release, LaTeX. This version is the predecessor ofALTEX 2.09 and the predecessor of the developer's version ALTEX3. Before we go into the gory details of the letter layout some words on TEX, and ALTEX. TEX is an interpreter for word processing. It has been devel- oped by Donald E. Knuth and compiles a text using control and macro commands. TeX, or better ALTEX is a complex TEX macro package which avoids the cryptical language of "plain" TEX. One word on compatibility to the experiencedALTEX 2.09 user: ALTEX contains anALTEX 2.0.9 mode which smoothly works together withALTEX 2.0.9 macro calls. Problems might be caused byALTEX 2.08 styles as they access internal macros ofALTEX 2.0.9 and those differ from the styles used byALTEX. A rather important part is writing letters. Included in our teTEX-packages there is theALTEX documentation class letter . We take this as an example to show you howALTEX works. file contents 23.2.1, page 379 shows a letter using letter document class. A detailed description of the letter 's capabilities can be found in the teTEX- package or under [Lamport, 1994 ]. Besides this letter there are many other classes; among these are book , report oder article . There are certain font-sizes that belong to these classes which can be selected via options. The classes book , report and article have three different fonts: 10 points (Option 10pt ), 11 points (Op- tion 11pt ) and 12 points (Option 12pt ). \usepackage is used to load other layout styles or special macro packages. All TEX commands start with a `\'. Environments are structured using \begin{} and \end{} \begin{document} starts the document and \end{document} ends it. \begin{textbf} : : :\end{texbf} turns all to bold whereas \textbf{ : : : } is considered for small, bold paragraphs. \bfseries switches to bold-mode. Besides this there are font families: \textrm{ : : : } or \rmfamily , \textsf{ : : : } or \sffamily , \texttt{ : : : } or \ttfamily and different styles \textup{ : : : } or \upshape , \textit{ : : : } or \itshape , \textsl{ : : : } or \slshape . 378 __________________________________________________________________________________________23.2.__Creating_a_document_______________* *_______________ |||||||| | |||||||| | ||||||||| || |||||||| | _______________________________________________________________________________________________________________________|||| |||||||| 1 %% | |||||||| | |||||||| 2 %% Created with Emacs and the AucTeX package from `e-site' | |||||||| 3 %% package. | |||||||| 4 %% Uses: LaTeX2e from `tetex' package. | |||||||| | |||||||| 5 % | |||||||| 6 % Document class `letter.cls' | |||||||| 7 % Option `norm' for size | |||||||| 8 % | |||||||| | |||||||| 9 \documentclass[12pt]{letter} | |||||||| 10 % A document :-) | |||||||| 11 % | |||||||| 12 \begin{document} | |||||||| | |||||||| 13 % | |||||||| 14 % By whom? | |||||||| 15 % | |||||||| 16 \address{Gladstone Ganter\\ | |||||||| | |||||||| 17 10 Luckyavenue.\ 100\par | |||||||| 18 08015 Duckcity} | |||||||| 19 % | |||||||| 20 \subject{\textbf{Your fort-knox}} | |||||||| | |||||||| 21 % | |||||||| 22 \encl{A Dollar bill} | |||||||| 23 % | |||||||| 24 \backaddress{Heinz Hugo, Poor-road\ 100, 08015 Duckcity} | |||||||| | |||||||| 25 % | |||||||| 26 \signature{Heinz Hugo} | |||||||| 27 % | |||||||| 28 \date{Duckcity /06/12/1995} | |||||||| | |||||||| 29 % | |||||||| 30 % To whom? | |||||||| 31 % | |||||||| 32 \begin{letter}{Scrooge McDuck \\ | |||||||| | |||||||| 33 1 Dollaravenue\ 1\par | |||||||| 34 04711 Duckcity} | |||||||| 35 % | |||||||| | |||||||| 36 % Opening | |||||||| 37 % | |||||||| 38 \opening{Dear Mr Duck,} | |||||||| 39 | |||||||| | |||||||| 40 Your fort-knox. ... | |||||||| 41 | |||||||| 42 % | |||||||| 43 % Closing | |||||||| | |||||||| 44 % | |||||||| 45 \closing{yours sincerley} | |||||||| 46 % | |||||||| 47 \end{letter} | |||||||| | |||||||| 48 \end{document} | |||||||| | ||||||||_______________________________________________________________________________________________________________ | File contents 23.2.1: Example letter of the letter document class 379 23. WhyALTE X? _________________________________________________________________________________________________________________* *_______________________________________@ 23.3 Compiling the document Using the command latex .tex the letter is compiled. This should run without any problems (if you did not change the structure of the file). If an error occurs the compiler stops and requests you to correct the error. Since this normally is not so easy you can cancel compiling by pressing Strg + d . Last output of TEX gives you a hint where you can find the error. 23.4 Viewing the document Now the document can be viewed (under X): xdvi .dvi This can take a while as some fonts have to be generated automatically. Po- sitioning of the new window should be somewhere in the upper left corner (it is quite big; you can fix it using the left mouse button). There is a zoom in xdvi . Pressing a number lower or higher than the standard 3 followed by s the screen is zoomed out (or in). The graphic can be scrolled sidewards using the cursor keys; moving page by page is done with Page " and Page # . If your document contains more pages you can switch to a certain page by entering the page number and g . Of course you can use the mousebuttons on the right hand side as well. More information can be obtained via manual pages. 23.5 Printing the document Printing of the .dvi file is done by: dvips texbrief[.dvi] [Optionen] . You do not have to give the .dvi extension. dvips is a rather powerful tool that supports various options for printing PostScript; you have to read the manual pages. Here just some options: _______________________________________________________________________________________________________* *_________ -r Pages are printed in reverse. -A Only odd numbers are printed. -B Only even numbers are printed. In connection with the previ- ous option this serves printing both sides of a page. -c Each page is printed times. -C The whole document is printed times. -o Output directly goes into another file. If no new filename is selected the file keeps its name but with an .ps extension. Without this option a name given in config.ps is used. -pp Printing certain defined pages. Single pages are separated by commas; A range of pages can be given by a dash (e. g. A-Z). _______________________________________________________________________________________________________* *_________ 380 Table 23.1: Options of dvips ________________________________________________________________________________________23.5.__Printing_the_document________* *______________________ Example: dvips Linux_manual -r -A -pp 2,5,24-32 -C 2 Pages 5, 25, 27, 29, 31 would be printed in reverse and twice. Page 2 would not be printed at all. * * 381 23. WhyALTE X? _________________________________________________________________________________________________________________* *_______________________________________@ 382 Appendix A Important keys A short summary of helpful and important keys. _______________________________________________________________________________________________________________________ Enter Executing a command on commandline, other- wise linefeed. Alt + F1 - F6 Change to another console (in text mode). Ctrl + Alt + F1 - F6 Under X11 and Dosemu: Change to one of the text consoles. Alt + F7 Switching back to X11. Ctrl + Alt + Backspace Exit X11. Shift * + Page " Shows previous screens on console Works until consoles are switched. Shift * + Page # Opposite of Shift * + Page " . " Shows the last command in command line. # Just the opposite of " (in shell). Shift * + Alt + Cursor X11: Change to the next virtual desktop (direc- tion with the cursor) Amount of virtual desk- tops is set in .fvwm[2]rc with the parameter "DeskTopSize". _______________________________________________________________________________________________________________________ Table A.1: Ctrl + Alt + (grey) Changes resolution according to entries in Xconfig. Ctrl + D Log out; corresponds to Befehl exit . Entry in /etc/profile: ignoreeof=x . x how often to repeat until the command is active. * * 383 A. Important keys_________________________________________________________________________________________________________________* *_______________________________________@ 384 Appendix B Glossary We have limited our glossary to UNIX or Linux specific terms respectively, since a complete introduction to electronic data processing is beyond the scope of this book. The reader of this glossary should be familiar with terms as bits and bytes. ATAPI ATAPI is a type of CD-ROM drive that is connected to an (E)IDE con- troller. Apart from ATAPI drives there are SCSI- CD-ROM drives, that are handled by a SCSI controller and proprietary CD-ROM drives that use their own controller or are connected to a sound card. Account Combination of login and password. In general the account is created by the + system administrator. He also assigns one or more groups to the user, as well as the resulting permissions. Creating an user account normally includes assigning a + home directory and the delivery of + email. Acronym Quite often abbreviations are called acronyms. + Linux, + FTP and + GNU are well-known acronyms. Alias This expression often is used in connection with + shells. An alias lets you abbreviate often used commands. Please have a look in chapter on shells. Background process background process If your + Shell (apparently) is only occupied with one process one speaks of a foreground process. Moreover most shells offer the ability of running processes in the background. If you want to run a process in the background you have to attach a `&' at the end. The only thing that is required for such a behaviour is a + multitasking operating system. Backup Backups should be done regularly, especially the important files! Even * * 385 B. Glossary _________________________________________________________________________________________________________________* *_______________________________________@ some configuration files that have been created with difficulty are worth being back-upped. The Linux specific backup command is tar . It back- ups the files to a + device. Quite often tar is used in combination with gzip . + GNU Base-Linux When you install Linux for the first time the base-Linux has to be brought up first. It works without the hard disk which is not accessible at this moment. Its kernel is on the boot-disk or on CD-ROM, the root-image is loaded into a RAM-disk (also on CD-ROM or on boot-disk). The other programs (e. g. YaST) are also loaded into RAM-disk. After the first login one starts YaST and prepares for installation of the "real" Linux. BIOS Basic Input Output System Any PC contains a little memory area that contains the "BIOS". This is a program for executing basic operations as memory check and recognizing hard disks. Under Linux the BIOS is not active since it runs in real mode and is switched off by the kernel at boot time. Anyway the Linux + Kernel provides much better capabilities than the BIOS. Booting Booting is the sequence from the very start until the system is ready for use. Under Linux this is booting the kernel which can be seen by the mes- sage "uncompressing linux..." and which ends with the `login:' prompt Buffer A buffer is a sort of intermediate memory which enables you to speed up access to data stored in itself. Under Linux there are many buffers. Command line Working with Unix is command line oriented. This in particular means that any process you enter in a + shell can have its own command line (e. g. the command ls can take a lot of options to change its behaviour. Console In former times this was a synonym for terminal. Nowadays under Linux you have virtual consoles. This enables you to use one screen for many independent running sessions. In standard + runlevel 2 you have 6 virtual consoles which can be reached by pressing Alt + F1 to Alt + F6 . From within a running X-Window System1 you reach the text consoles by pressing Strg + Alt + F1 to Strg + Alt + F6 . CPU Central Processing Unit The processor. Intel x86 processors can be run in many different modes; here we want to distinguish between two of them: _____________________________________________1 which uses either console 7 (runlevel 2) or console 3 (runlevel 3) 386 _________________________________________________________________________________________________________________B.__Glossary_* *____________________ o real Mode: The "original" mode using a segmented memory protocol. This is slow and antiquated ("16 bit software"). o protected Mode: (avaliable from 386 onwards). Mode with linear memory model. Only this mode uses the full power of the CPU. Linux only runs in protected mode. By time of writing of this book Linux exists for the following processor architectures: Intel x86, DEC alpha, Motorola m68k, Sparc, PowerPC, MIPS, ARM. Cursor The cursor normally is a little block-character which marks the input. Under Linux this term occurs in: o Shell/Editor: A spot marked with a rectangle of a (flashing) line where the next entry will occur. In the shell the cursor is right to the + Prompt. o Mouse cursor under X: The mouse cursor under X. Depending on the background it changes its shape. For example an arrow in a xterm, X-shaped on the root window or I-shaped in an + Editor (buffer). o GPM cursor (Console): A one character block which can be positioned on the console by means of the mouse. It can be used for Cut & Paste (+ Selection). Directories Build a + file-systems structure. In a directory files or directories are listed. You say a + file `x' lies in directory `y' if its name is listed in this directory. Because there are branches of subdirectories within other directories this is often referred to as + directory tree. If you want to see another directory you can change into it. Files are considered the leaves of the directory tree which cannot (quite logically) contain any more subdirectories (branches). Directories follow the same restrictions as files. The special directories `.' and `..' refer to the directory itself and to its parent directory in file-system hierarchy. Editor Editors are programs for changing text (e. g. by entering text). Well-known editors under Linux are GNU Emacs (emacs ) or the UNIX editor vi . Email electronic mail Means of transporting electronic mail between registered users via a net- work, Same as with "normal" mail (often referred to as "snail mail") the address has to be entered as "email address". electronic mail not only lets you send text, but sound-documents or pictures. electronic mail has got plenty of advantages: first its quite cheap and second the mails reach the address normally within minutes. Environment A + shell normally provides some kind of environment where the user temporarily can set options such as paths of programs, the user name, * * 387 B. Glossary _________________________________________________________________________________________________________________* *_______________________________________@ the current path, prompt, shape and more. This data is stored in a + environment variable. These variables can be assigned e. g. by the shells configuration files. Environment variable A storage place in the + environment of the + shell. Every variable consists of a name (which usually is written in uppercase) and a value (e. g. pathname). If you use the bash this is done like: root@earth:/ > export EDITOR=emacs Invoking env you can get a list of the variables set. If a variable is needed (e. g. in a shell-script) it is dereferenced by attaching a $ at the beginning. Important environment variables are $HOME (path to user directory), $SHELL (shell path), $USER (user name), $PATH (path for executa- bles), $MANPATH (path for man pages). EXT2 second extended File-system is the file-system used by Linux. It offers a high throughput, long file names, permissions, and error tolerance as well. File Under Linux a file is the central concept for handling data. As with other systems files serve for writing data onto a mass storage media. The file name has to be unique in its directory, By means of a + file-system these files can be hierarchically structured. Please have a look at other topics on files in this glossary. Additionally there are special files under Linux as well. See + link, + device and + proc. Filesystem A file-system is a system for structuring files. There are many file-systems available which differ (sometimes extremely) in their performance and power. Some file-systems are strictly tied to certain media. One cannot easily say "Linux uses file-system 'X"'. A list of supported file-systems can be found in section 19.11. Focus A control item, e. g. the command line of a + terminal, has the focus if the actual input is directed to it. Mostly the focus is connected with the + cursor. The way a window manager manages the focus is called focus policy. You can differentiate between a focus that follows the mouse or a focus that has to be set by clicking the mouse. FTP file transfer protocol FTP is a means under UNIX for transferring files from one machine to another. There is the FTP server (the machine which sends the files) on the one side, and the FTP client (which receives the packages) on the other. 388 _________________________________________________________________________________________________________________B.__Glossary_* *____________________ GNU GNU stands for GNU is Not Unix and is a product of the Free Software Foundation (FSF) whose aim it is to provide a completely free (source code available by no costs) Unix compatible operating system. In this progress any Unix tools are completely rewritten and new functionality is added. Linux profits from these tools but should not be confused with it. Graphical User Interface A GUI is the graphical form of a normal desktop. Whereas you lay dif- ferent papers onto a normal desktop, here these "papers" are called + windows. You can put as many of these windows on your desktop as you like. All your different processes run in a separate window. A GUI normally is controlled by + control elements via a mouse, track- ball or something similar. Well-known GUIs are e. g. + X Window System, Apple MacIntosh System 7, Digital Research GEM , MS-Windows . Home directory The home directory is the starting point for most user activities on the machine. He can put his private files into his home directory. Besides the + system administrator he is the only one who has access to this directory. The location of the home directory in the file-system under UNIX is given by the + environment variable $HOME. Interface Generally speaking, an interface is a device where different systems ex- change information in any way. Such an interface e. g. is a keyboard where men and machine meet. This is rather abstract, some other interfaces are much more exact: o Hardware interface: for connecting peripherical devices as a parallel port, SCSI or serial interfaces. o Software interface: Assigns how programs should communicate. See + protocol. o User interface: Here persons and machine exchange data. e. g.mouse, monitor and keyboard. Internet The internet is a worldwide heterogenuos (i. e. it consists of lots of dif- ferent machines) + network. Machines are accessed via their + IP ad- dresses (IP = internet protocol) which are unique. Those IP addresses are structured in an hierarchical manner. There are national, domains, sub-domains and the machines address. Besides the numerical IP ad- dress (like: 192.168.0.1) there are aliases (like: helios.cosmos.com) which simplify the names for the user. It is not only the + hardware layer that keeps the internet up but also a couple of + protocols (e. g. ftp, http, tcp) on certain logical layers. Well-known services of the internet are + email or the WorldWideWeb (WWW or W3). A real important keyword * * 389 B. Glossary _________________________________________________________________________________________________________________* *_______________________________________@ in connection with the internet is the "Netiquette" which makes sure that everything runs in order. This can be regarded as a voluntar self-control. Inode The + EXT2 file-system uses inodes for organizing information on files. Moreover inodes contain information such as owner of its file, permis- sions and more : : : LAN local area network A LAN is a local + network which means that is is rather small and gener- ally supervised by one + system administrator. LANs are normally con- nected to other LANs via a gateway thus forming a + WAN. Main memory (RAM) This is often referred to as + RAM. RAM access is very fast in comparison to hard disk access. Under Linux this + memory is often referred to as physical memory. Mass storage media A collection of different media for store data. Typical mass storage media are disks, hard disks, tapes, CD-ROMs, magneto-optical disks, holographic media and many more. Memory The memory is the brain of your machine. Under Linux one often refers to two different types of memory: o physical memory: This is the typical memory (RAM) chip. Typical size is from 8 MB up to 128 MB. o virtual memory: By means of virtual memory the system can consider certain ranges of the disk/floppy etc. as memory (+ swap). Menu On + graphical user interfaces functions and actions of an application can be reached via a menu. Menus offer all available commands and options (just in a restaurant), so the user can choose either of them. Normally there is a menu bar with sub-menus. Moreover there are popup-menus which can be invoked e. g. with a button or a mouse-click. Mountpoint A mountpoint is the directory where a partition or another device is at- tached to the Linux file-system. Multiprocessing If your machine works with more than one + processor this is called a multiprocessor or multiprocessing system. Under Linux you will en- counter the term SMP which stands for symmetric multi processing and is a special form of multiprocessing. 390 _________________________________________________________________________________________________________________B.__Glossary_* *____________________ Multitasking Operating systems that can invoke more than one + program simultane- ously are called multitasking systems. There are two forms of multitask- ing: o preemptive multitasking: The operating system is responsible for shar- ing cpu-time between the processes. o cooperative multitasking: The processes themselves give back cpu- time. As you can see the first variant is the better one, since no process can block the whole CPU. Linux offers you real preemptive multitasking. Multiuser + Multitasking is required if more than one person wants to work on the same machine. An operating system that offers this capability is called a multiuser system. Network A network is a connection of different machines. There are different types of networks, that depends on how the machines are connected: ring, star,bus, tree and more. Well-known hardware standards for networks are e. g. ethernet and Token-Ring or ISDN. Important terms on software protocols are (on different layers) e. g. TCP, UDP, IPX and more. Operating system The operating system is a process running permanently and in background which enables basic operation with the machine. Tasks of an operat- ing system are the management of all machine specific resources. Un- der Linux this is done by the + kernel and perhaps some modules. Well known operating systems are Linux , AmigaOS , MacOS , OS/2 , UNIX , Windows NT and many more : : : Path Via the path a + file unambiguous is assigned to a + file-system. In Unix the different layers are separated by a `/', a "slash". There are two sorts of paths: o relative path: here the position of a file or a directory is relative to the actual (current) path. o absolute path: here the position is described relative to the + root di- rectory. Pipe A pipe stands for connecting the standard-out of a process (+ program) with the standard-in (+ standard in/out) of a successor process. Thus avoiding having to write temporary files for further processing. In shell mode the processes to be "piped" are entered one after another separated by a pipe `|' (ASCII 124). * * 391 B. Glossary _________________________________________________________________________________________________________________* *_______________________________________@ Process A process is a program or an executables files (+ Shell) "living" variant. Often it is referred to as task. Prompt Within a text + Shell the place where users can type commands to the + operating system is referred to as prompt. The prompt might contain the users name, the machines name, current time and certain others. Most of the time the + cursor stands directly behind the prompt. If the prompt reappears this shows that the system is ready (for yourinput). Protocol Protocols organize either on hardware or on software side the communi- cation between the different machines in + networks. They assign how the format of the transferred data has to be, which machines controls a machine etc. Well-known protocols are e. g. FTP, UDP, TCP, HTTP, etc. ROM Read-only Memory A CD is a good example for a ROM. Reset If the machine hangs and does not react anymore to keystrokes it might be trapped in an endless loop. The only solution to such a problem is to bring the machine back to a defined state. This is called a reset. After a reset the machine is in exactly the same state as when it is started. A reset should always be preferred as this is better to the machinery. IMPORTANT!: As you might guess after a reset all data that resided in + RAM of the machine is irreversibly lost! Rlogin Using a remote login one can log into a remote machine via the + internet as if you were sitting at its + console. If there is a + X-Server running on both machines you can even display the output of an X-application on the local display provided the $DISPLAY variable is set correctly. Root See + System administrator Root directory The toplevel directory of a + file system. On the contrary to all other directories there is no parent directory of the root directory. `..' points to itself. Under UNIX the root directory is known as `/'. Runlevel A runlevel describes a certain state of your system. In other words: The system behaves differently on different runlevels. There is a runlevel for system administration (S ), as well as a runlevel for the GUI of the "X- Display Manager" (xdm; 3 ). 392 _________________________________________________________________________________________________________________B.__Glossary_* *____________________ Selection Selection is a mechanism of the X-Server. You can mark text characters with the mouse (by moving over them holding down the left mouse but- ton) and "move" them to e. g. another application (by moving the cursor in the appropriate window and pressing the middle mouse button). This is called "Cut & Paste". Server A server generally is a quite powerful machine which serves other ma- chines connected via a + network and provides services or data. Besides the machines there are programs that are called servers as well. These programs offer (serve) services too. An example for a software server is e. g. the + X-Server. Shell The shell is the fundamentary interface to the + operating system (Ker- nel). By means of the shell commands can be entered. The shell provides a command line. To make processes run automatically most shells provide an scripting language of their own. These programs (called shell-scripts) can be considered intelligent batch files. Examples for shells are: bash , sh , tcsh Standard- in/out Every process inherits three channels where it can read or output data. These are standard-in (stdin), standard-out (stdout) and the standard-error (stderr). These channels are set to certain devices as default. Standard-in is set to your keyboard, standard-out and standard-err to the screen. By means of the shell these channels can be redirected. If the characters are not read from keyboard but from a file instead this is called a redirection of standard-in. In shell-mode redirection is invoked by entering `<' (std- in), `>' (std-out) and `2>' (std-err) . See + Pipe Switch Switches can change the (default) behaviour of programs. The so called + command line consists of the programs name and some (optional) switches. System administrator The person who is responsible for maintaining and supervising a complex system or network. Generally the system administrator is the only one who has access to any corner of the system (root permissions). Task See + Process Telnet Telnet creates a connection to a (remote) host and gives you a login on this machine, provided you have an account. * * 393 B. Glossary _________________________________________________________________________________________________________________* *_______________________________________@ Terminal A combination of a screen and a keyboard without computing capabilities. Also used on workstations that emulate a real terminal. UNIX is an operating system that is widely spread on workstations. UNIX supports vital concept such as running different machines in a network. UNIX consists of a kernel, a + shell and applications. Since mid-nineties there is a + pc freeware version available: Linux. WAN wide area network As opposed to a + LAN this + network is widely spread. Wildcards The characters `*' and `?' are generic signs and are used as joker or wildcards (as in a card game). The `?' stands for exactly on charac- ter, `*' replaces any number of characters even no characters. Wild- cards are often used in regular expressions. For example: the command ls -l bild* lists all files in the actual directory that start with a "bild" and end with any character(s) (even only bild). Window Windows are rectangle shaped screen segments which normally are deco- rated by a frame. This frame normally contains decorations for changing the size, for moving the window and more. For being able to work with windows a + X-server and a + Window manager have to run. Window manager A window manager is responsible (among other things) for decorating your windows, provide you with certain functionality as resizing, moving and destroying windows. Furthermore it is responsible for the "loo&feel" of your system. X-Server Machines that run a X-Server can use + graphical user interfaces ser- vices provided by the + X Window System. An important task of the X- Server is to manage the displays. Normally each terminal has exact one display. Name of this display is set by the + environment variable $DIS- PLAY. The format of this variable is: : ZB earth:0 . You need to know your displays name for + rlogin. X Window System A collection of programs, protocols and routines for organizing and main- taining a + graphical user interface. The X-Window System (short: X) has been developed as project Athena at MIT (Massachusetts Institute of Technology). Some of the big advantages of X-Window versus other systems (e. g. MS-Windows or GEM) are its networking capabilities and its flexibility. It is, e. g. possible to run programs on other machines and 394 _________________________________________________________________________________________________________________B.__Glossary_____* *________________ redirect their output to your display while free adapting the look&feel of your system. 395 B. Glossary _________________________________________________________________________________________________________________* *_______________________________________@ 396 Appendix C The directory tree C.1 Overview The following|figure|shows a small portion of the Linux directory tree: * * | ____________________________________________________________________________________________________________________________* *______________________|||| |||||||| * * | |||||||| * * | |||||||| * * | |||||||| * * | |||||||| * * | |||||||| * * | |||||||| * * | |||||||| * * | |||||||| * * | |||||||| * * | |||||||| * * | |||||||| * * | |||||||| * * | |||||||| * * | |||||||| * * | |||||||| * * | |||||||| * * | |||||||| * * | |||||||| * * | |||||||| * * | |||||||| * * | |||||||| * * | |||||||| * * | |||||||| * * | |||||||| * * | |||||||| * * | |||||||| * * | |||||||| * * | |||||||| * * | ||||||||____________________________________________________________________________________________________________________* *______________________ | C.2 Important directories The directory tree of your Linux system is well organized. Some important directories are: * * 397 C. The directory tree_____________________________________________________________________________________________________________* *_______________________________________@ ________________________________________________________________________________________________________* *________ / "root" directory, start of the directory tree /home the (private) directories of the users /dev files, that represent the system hardware (device files) /etc important system configuration files /usr/bin publicly accessible commands /bin commands needed to start the system /usr/sbin commands reserved for the superuser /sbin commands reserved for the superuser and needed for system start /sbin/init.d scripts for system start /usr/include header files for the C compiler /usr/include/g++ header files for the C++ compiler /usr/doc documentation files /usr/man manual pages /usr/src source code for the system software /usr/src/linux the kernel sources /tmp temporary files /var/tmp big temporary files /usr user commands and applications, configuration files, can be mounted readonly /var configuration files (linked from /usr). /lib shared libraries (for dynamically linked pro- grams) /proc the process file system /usr/local local extensions, independent from the distribu- tion /opt optional software, large systems (e. g. KDE) ________________________________________________________________________________________________________* *________ Table C.1: Overview of important directories. 398 Appendix D Important files The most important file in your system is the kernel itself. It can be found in the root directory in the file /vmlinuz. D.1 Device files in the /dev directory Disks and hard disks: 1 ______________________________________________________________________________________________________________________ /dev/fd0 first floppy drive /dev/fd1 second floppy drive /dev/hda first AT bus HD /dev/hda1 - /dev/hda15 partitions of first AT bus HD /dev/sda first SCSI HD /dev/sda1 - /dev/sda15 partitions of first SCSI HD /dev/sdb second SCSI HD /dev/sdc third SCSI HD ______________________________________________________________________________________________________________________ Table D.1: Overview of device file of mass storage media D.1.1 CDROM drives ______________________________________________________________________________________________________________________ /dev/cdrom Link to the used CD-ROM drive; one of the following files. Is assigned by YaST) /dev/aztcd Aztech CDA268-01 CD-ROM /dev/cdu535 Sony CDU-535 CD-ROM /dev/cm206cd Philips CM206 /dev/gscd0 Goldstar R420 CD-ROM ______________________________________________________________________________________________________________________ Table D.2: to be continued... 1 ____________________________________________________ Besides the listed device files, you may create additional ones. Information: Manual Pages for "mknod". * * 399 D. Important files________________________________________________________________________________________________________________* *_______________________________________@ _______________________________________________________________________________________________________* *_________ /dev/hda to ATAPI IDE CD-ROM /dev/hdd /dev/lmscd Philips CM 205/250/206/260 CD-ROM /dev/mcd Mitsumi CD-ROM /dev/sbpcd0 to /dev/sbpcd3 CD-ROM on SoundBlaster /dev/scd0 to SCSI CD-ROM drives /dev/scd1 /dev/sonycd Sony CDU 31a CD-ROM /dev/sjcd Sanyo CD-ROM /dev/optcd Optics Storage CD-ROM _______________________________________________________________________________________________________* *_________ Table D.2: Overview of device files of CD-ROM drives D.1.2 Tapes _______________________________________________________________________________________________________* *_________ /dev/rmt0 1. SCSI-Streamer "`rewinding"' (rewinds automati- cally) /dev/nrmt0 1. SCSI-Streamer "`non rewinding"' /dev/ftape Floppy-Streamer "`rewinding"' (rewinds automati- cally) /dev/nftape Floppy-Streamer "`non rewinding"' _______________________________________________________________________________________________________* *_________ Table D.3: Overview of device files of streamers D.1.3 Mice (bus and PS/2) _______________________________________________________________________________________________________* *_________ /dev/mouse Link to the interface used by the mouse (an pseudo file for bus mice) a serial interface for others. Is as- signed by YaST). /dev/atibm ATI graphics card bus mouse. /dev/logibm Logitech bus mouse. /dev/inportbm PS/2 bus mouse. _______________________________________________________________________________________________________* *_________ Table D.4: Overview of device file of mice D.1.4 Modem 400 ___________________________________________________________________________________D.2.__Configuration_files_in_/etc________* *______________________ ______________________________________________________________________________________________________________________ /dev/modem Link to the COM port where the modem is connected to. I assigned by YaST). ______________________________________________________________________________________________________________________ Table D.5: Overview of device files of modems D.1.5 Serial interfaces ______________________________________________________________________________________________________________________ /dev/ttyS0 Serial interfaces 0 to 3 to /dev/ttyS3 /dev/cua0 Serial interfaces 0 to 3 (for outgoing modem connec- to tions). /dev/cua3 ______________________________________________________________________________________________________________________ Table D.6: Overview of device files of serial interfaces D.1.6 Parallel ports ______________________________________________________________________________________________________________________ ______________________________________________________________________________________________________________________ Table D.7: Overview of device files of parallel ports D.1.7 Special devices ______________________________________________________________________________________________________________________ /dev/null "swallows" any data (data bin) /dev/tty1 Virtual consoles to /dev/tty8 /dev/zero Outputs any number of null bytes. ______________________________________________________________________________________________________________________ Table D.8: Overview of device files of special (virtual) devices D.2 Configuration files in /etc * * 401 D. Important files________________________________________________________________________________________________________________* *_______________________________________@ _______________________________________________________________________________________________________* *_________ /etc/rc.config The systems central configuration file; is cre- ated by YaST and read by the bootscripts and SuSEconfig . /etc/inittab Configuration file for the init process. /etc/lilo.conf Configuration of LILO /etc/conf. Configuration of kernel modules modules /etc/DIR\_COLORS Assigning colours for ls /etc/XF86Config Configuration of the X-Window System /etc/fstab Table of file systems that are automatically mounted at boot time /etc/profile The shells login script /etc/passwd User database: user names, home directory, login-shell, user number /etc/shadow Passwords /etc/group User groups /etc/printcap Description of installed printers. Is used by the lpd printer daemon. page 143 /etc/hosts Assigning hostnames to IP addresses; necessary if no nameserver is installed. /etc/inetd.conf Definition of configured IP services (telnet , finger , ftp and many more). /etc/syslogd.conf Configuration file for the syslog daemon; it servers for reporting certain system messages. _______________________________________________________________________________________________________* *_________ Table D.9: Configuration files in / etc D.3 Hidden configuration files in home In the home directory of a user there are several configuration files which for practical reasons are "hidden". The are touched rather rarely. Each file becomes hidden if you attach a dot at the beginning of the file. One can see those files by typing ls -a . Some examples are to be found in table D.10. _______________________________________________________________________________________________________* *_________ .profile the users private login script (for bash) .bashrc Bash configuration .exrc Vi configuration .xinitrc X-Window system startup script .fvwmrc Configuration of fvwm windowmanager .ctwmrc Configuration of ctwm windowmanager .openwin-menu Configuration of olvwm and olwm windowmanager _______________________________________________________________________________________________________* *_________ Table D.10: Invisible files in the user's home directory 402 _______________________________________________________________________D.3.__Hidden_configuration_files_in_home_____________* *______________________ These files are copied from /etc/skel when creating a new user. * * 403 D. Important files________________________________________________________________________________________________________________* *_______________________________________@ 404 Appendix E Manual page of e2fsck |||||||| | |||||||| | |||||||| | _______________________________________________________________________________________________________________________|||| |||||||| | |||||||| E2FSCK(8) E2FSCK(8) | |||||||| | |||||||| | |||||||| | |||||||| NAME | |||||||| e2fsck - check a Linux second extended file system | |||||||| | |||||||| | |||||||| SYNOPSIS | |||||||| e2fsck [ -pacnyrdfvstFSV ] [ -b superblock ] [ -B block- | |||||||| size ] [ -l|-L bad_blocks_file ] device | |||||||| | |||||||| | |||||||| DESCRIPTION | |||||||| e2fsck is used to check a Linux second extended file | |||||||| system. | |||||||| | |||||||| | |||||||| device is the special file corresponding to the device | |||||||| (e.g /dev/hdXX). | |||||||| | |||||||| | |||||||| OPTIONS | |||||||| -a This option does the same thing as the -p option. | |||||||| It is provided for backwards compatibility only; it | |||||||| is suggested that people use -p option whenever | |||||||| | |||||||| possible. | |||||||| | |||||||| -b superblock | |||||||| Instead of using the normal superblock, use the | |||||||| | |||||||| alternative superblock specified by superblock. | |||||||| | |||||||| -B blocksize | |||||||| Normally, e2fsck will search for the superblock at | |||||||| | |||||||| various different block sizes in an attempt to find | |||||||| the appropriate block size. This search can be | |||||||| fooled in some cases. This option forces e2fsck to | |||||||| only try locating the superblock at a particular | |||||||| | |||||||| blocksize. If the superblock is not found, e2fsck | |||||||| will terminate with a fatal error. | |||||||| | |||||||| | |||||||| -c This option causes e2fsck to run the badblocks(8) | |||||||| program to find any blocks which are bad on the | |||||||| filesystem, and then marks them as bad by adding | |||||||| | ||||||||_______________________________________________________________________________________________________________ | _ to be continued * * 405 | * * | | * * | E. Manual page of_e2fsck__________________________________________________________________________________________________________* *_______________________________________@ | * * | ________________________________________________________________________________________________________* *________|||| | * * | || them to the bad block inode. * * || | * * | | -d Print debugging output (useless unless you are * * | | debugging e2fsck ). * * | | * * | | -f Force checking even if the file system seems clean. * * | | * * | | * * | | -F Flush the filesystem device's buffer caches before * * | | beginning. Only really useful for doing e2fsck * * | | time trials. * * | | * * | | * * | | -l filename * * | | Add the blocks listed in the file specified by * * | | filename to the list of bad blocks. * * | | * * | | * * | | -L filename * * | | Set the bad blocks list to be the list of blocks * * | | * * | | specified by filename. (This option is the same as * * | | the -l option, except the bad blocks list is * * | | cleared before the blocks listed in the file are * * | | added to the bad blocks list.) * * | | * * | | * * | | -n Open the filesystem read-only, and assume an answer * * | | of ``no'' to all questions. Allows e2fsck to be * * | | used non-interactively. (Note: if the -c, -l, or * * | | * * | | -L options are specified in addition to the -n * * | | option, then the filesystem will be opened read- * * | | write, to permit the bad-blocks list to be updated. * * | | However, no other changes will be made to the * * | | * * | | filesystem.) * * | | * * | | -p Automatically repair ("preen") the file system * * | | without any questions. * * | | * * | | * * | | -r This option does nothing at all; it is provided * * | | only for backwards compatibility. * * | | * * | | * * | | -s This option will byte-swap the filesystem so * * | | that it is using the normalized, standard byte- * * | | order (which is i386 or little endian). If the * * | | filesystem is already in the standard byte-order, * * | | * * | | e2fsck will take no action. * * | | * * | | -S This option will byte-swap the filesystem, regard- * * | | less of its current byte-order. * * | | * * | | * * | | -t Print timing statistics for e2fsck. If this option * * | | is used twice, additional timing statistics are * * | | * * | | printed on a pass by pass basis. * * | | * * | | -v Verbose mode. * * | | * * | | * * | | -V Print version information and exit. * * | | * * | | -y Assume an answer of ``yes'' to all questions; * * | | * * | |_______________________________________________________________________________________________________* *________ | _ to be continued 406 |||||||| | |||||||| | ____________________________________________________________________________________________E.__Manual_page_of_e2fsck______________* *_______________|||| |||||||| | _______________________________________________________________________________________________________________________|| |||||||| | |||||||| | |||||||| allows e2fsck to be used non-interactively. | |||||||| | |||||||| EXIT CODE | |||||||| | |||||||| The exit code returned by e2fsck is the sum of the | |||||||| following conditions: | |||||||| 0 - No errors | |||||||| 1 - File system errors corrected | |||||||| | |||||||| 2 - File system errors corrected, system should | |||||||| be rebooted if file system was mounted | |||||||| 4 - File system errors left uncorrected | |||||||| 8 - Operational error | |||||||| | |||||||| 16 - Usage or syntax error | |||||||| 128 - Shared library error | |||||||| | |||||||| REPORTING BUGS | |||||||| | |||||||| Almost any piece of software will have bugs. If you | |||||||| manage to find a filesystem which causes e2fsck to crash, | |||||||| or which e2fsck is unable to repair, please report it to | |||||||| | |||||||| the author. | |||||||| | |||||||| Please include as much information as possible in your bug | |||||||| report. Ideally, include a complete transcript of the | |||||||| | |||||||| e2fsck run, so I can see exactly what error messages are | |||||||| displayed. If you have a writeable filesystem where the | |||||||| transcript can be stored, the script(1) program is a handy | |||||||| way to save the output of to a file. | |||||||| | |||||||| | |||||||| It is also useful to send the output of dumpe2fs(8). If a | |||||||| specific inode or inodes seems to be giving e2fsck | |||||||| trouble, try running the debugfs(8) command and send the | |||||||| | |||||||| output of the stat command run on the relevant inode(s). | |||||||| If the inode is a directory, the debugfs dump command | |||||||| will allow you to extract the contents of the directory | |||||||| inode, which can sent to me after being first run through | |||||||| | |||||||| uuencode(1). | |||||||| | |||||||| Always include the full version string which e2fsck | |||||||| displays when it is run, so I know which version you are | |||||||| | |||||||| running. | |||||||| | |||||||| AUTHOR | |||||||| This version of e2fsck is written by Theodore Ts'o | |||||||| | |||||||| . | |||||||| | |||||||| SEE ALSO | |||||||| mke2fs(8), tune2fs(8), dumpe2fs(8), debugfs(8) | |||||||| | |||||||| | |||||||| | |||||||| E2fsprogs version 1.06 October 1996 | |||||||| | |||||||| | ||||||||_______________________________________________________________________________________________________________ | 407 E. Manual page of_e2fsck__________________________________________________________________________________________________________* *_______________________________________@ 408 Appendix F The GNU General Public License GNU GENERAL PUBLIC LICENSE Version 2, June 1991 Copyright (C) 1989, 1991 Free Software Foundation, Inc. 675 Mass Ave, Cambridge, MA 02139, USA Everyone is permitted to copy and distribute verbatim copies of this license document, but changing it is not allowed. Preamble The licenses for most software are designed to take away your freedom to share and change it. By contrast, the GNU General Public License is intended to guarantee your freedom to share and change free software-to make sure the software is free for all its users. This General Public License applies to most of the Free Software Foundation's software and to any other program whose authors commit to using it. (Some other Free Software Foundation software is covered by the GNU Library General Public License instead.) You can apply it to your programs, too. When we speak of free software, we are referring to freedom, not price. Our General Public Licenses are designed to make sure that you have the freedom to distribute copies of free software (and charge for this service if you wish), that you receive source code or can get it if you want it, that you can change the software or use pieces of it in new free programs; and that you know you can do these things. To protect your rights, we need to make restrictions that forbid anyone to deny you these rights or to ask you to surrender the rights. These restrictions translate to certain responsibilities for you if you distribute copies of the software, or if you modify it. For example, if you distribute copies of such a program, whether gratis or for a fee, you must give the recipients all the rights that you have. You must make sure that they, too, receive or can get the source code. And you must show them these terms so they know their rights. We protect your rights with two steps: (1) copyright the software, and (2) offer you this license which gives you legal permission to copy, distribute and/or modify the software. Also, for each author's protection and ours, we want to make certain that everyone understands that there is no warranty for this free software. If the * * 409 F. The GNU General_Public_License_________________________________________________________________________________________________* *_______________________________________@ software is modified by someone else and passed on, we want its recipients to know that what they have is not the original, so that any problems introduced by others will not reflect on the original authors' reputations. Finally, any free program is threatened constantly by software patents. We wish to avoid the danger that redistributors of a free program will individually obtain patent licenses, in effect making the program proprietary. To prevent this, we have made it clear that any patent must be licensed for everyone's free use or not licensed at all. The precise terms and conditions for copying, distribution and modification follow. GNU GENERAL PUBLIC LICENSE TERMS AND CONDITIONS FOR COPYING, DISTRIBUTION AND MODIFICATION 0. This License applies to any program or other work which contains a no- tice placed by the copyright holder saying it may be distributed under the terms of this General Public License. The "Program", below, refers to any such program or work, and a "work based on the Program" means either the Program or any derivative work under copyright law: that is to say, a work containing the Program or a portion of it, either verbatim or with modifications and/or translated into another language. (Hereinafter, trans- lation is included without limitation in the term "modification".) Each licensee is addressed as "you". Activities other than copying, distribution and modification are not cov- ered by this License; they are outside its scope. The act of running the Program is not restricted, and the output from the Program is covered only if its contents constitute a work based on the Program (independent of having been made by running the Program). Whether that is true de- pends on what the Program does. 1. You may copy and distribute verbatim copies of the Program's source code as you receive it, in any medium, provided that you conspicuously and appropriately publish on each copy an appropriate copyright notice and disclaimer of warranty; keep intact all the notices that refer to this License and to the absence of any warranty; and give any other recipients of the Program a copy of this License along with the Program. You may charge a fee for the physical act of transferring a copy, and you may at your option offer warranty protection in exchange for a fee. 2. You may modify your copy or copies of the Program or any portion of it, thus forming a work based on the Program, and copy and distribute such modifications or work under the terms of Section 1 above, provided that you also meet all of these conditions: a) You must cause the modified files to carry prominent notices stating that you changed the files and the date of any change. b) You must cause any work that you distribute or publish, that in whole or in part contains or is derived from the Program or any part thereof, to be licensed as a whole at no charge to all third parties under the terms of this License. 410 ____________________________________________________________________________F.__The_GNU_General_Public_License______________* *______________________ c) If the modified program normally reads commands interactively when run, you must cause it, when started running for such interactive use in the most ordinary way, to print or display an announcement including an appropriate copyright notice and a notice that there is no warranty (or else, saying that you provide a warranty) and that users may redis- tribute the program under these conditions, and telling the user how to view a copy of this License. (Exception: if the Program itself is inter- active but does not normally print such an announcement, your work based on the Program is not required to print an announcement.) These requirements apply to the modified work as a whole. If identifiable sections of that work are not derived from the Program, and can be rea- sonably considered independent and separate works in themselves, then this License, and its terms, do not apply to those sections when you dis- tribute them as separate works. But when you distribute the same sections as part of a whole which is a work based on the Program, the distribution of the whole must be on the terms of this License, whose permissions for other licensees extend to the entire whole, and thus to each and every part regardless of who wrote it. Thus, it is not the intent of this section to claim rights or contest your rights to work written entirely by you; rather, the intent is to exercise the right to control the distribution of derivative or collective works based on the Program. In addition, mere aggregation of another work not based on the Program with the Program (or with a work based on the Program) on a volume of a storage or distribution medium does not bring the other work under the scope of this License. 3. You may copy and distribute the Program (or a work based on it, under Section 2) in object code or executable form under the terms of Sections 1 and 2 above provided that you also do one of the following: a) Accompany it with the complete corresponding machine-readable source code, which must be distributed under the terms of Sections 1 and 2 above on a medium customarily used for software interchange; or, b) Accompany it with a written offer, valid for at least three years, to give any third party, for a charge no more than your cost of physically performing source distribution, a complete machine-readable copy of the corresponding source code, to be distributed under the terms of Sections 1 and 2 above on a medium customarily used for software interchange; or, c) Accompany it with the information you received as to the offer to distribute corresponding source code. (This alternative is allowed only for noncommercial distribution and only if you received the program in object code or executable form with such an offer, in accord with Subsection b above.) The source code for a work means the preferred form of the work for making modifications to it. For an executable work, complete source code means all the source code for all modules it contains, plus any associated * * 411 F. The GNU General_Public_License_________________________________________________________________________________________________* *_______________________________________@ interface definition files, plus the scripts used to control compilation and installation of the executable. However, as a special exception, the source code distributed need not include anything that is normally distributed (in either source or binary form) with the major components (compiler, kernel, and so on) of the operating system on which the executable runs, unless that component itself accompanies the executable. If distribution of executable or object code is made by offering access to copy from a designated place, then offering equivalent access to copy the source code from the same place counts as distribution of the source code, even though third parties are not compelled to copy the source along with the object code. 4. You may not copy, modify, sublicense, or distribute the Program except as expressly provided under this License. Any attempt otherwise to copy, modify, sublicense or distribute the Program is void, and will automati- cally terminate your rights under this License. However, parties who have received copies, or rights, from you under this License will not have their licenses terminated so long as such parties remain in full compliance. 5. You are not required to accept this License, since you have not signed it. However, nothing else grants you permission to modify or distribute the Program or its derivative works. These actions are prohibited by law if you do not accept this License. Therefore, by modifying or distribut- ing the Program (or any work based on the Program), you indicate your acceptance of this License to do so, and all its terms and conditions for copying, distributing or modifying the Program or works based on it. 6. Each time you redistribute the Program (or any work based on the Pro- gram), the recipient automatically receives a license from the original li- censor to copy, distribute or modify the Program subject to these terms and conditions. You may not impose any further restrictions on the recip- ients' exercise of the rights granted herein. You are not responsible for enforcing compliance by third parties to this License. 7. If, as a consequence of a court judgment or allegation of patent infringe- ment or for any other reason (not limited to patent issues), conditions are imposed on you (whether by court order, agreement or otherwise) that contradict the conditions of this License, they do not excuse you from the conditions of this License. If you cannot distribute so as to satisfy si- multaneously your obligations under this License and any other pertinent obligations, then as a consequence you may not distribute the Program at all. For example, if a patent license would not permit royalty-free re- distribution of the Program by all those who receive copies directly or indirectly through you, then the only way you could satisfy both it and this License would be to refrain entirely from distribution of the Program. If any portion of this section is held invalid or unenforceable under any particular circumstance, the balance of the section is intended to apply and the section as a whole is intended to apply in other circumstances. It is not the purpose of this section to induce you to infringe any patents or other property right claims or to contest validity of any such claims; this section has the sole purpose of protecting the integrity of the free soft- ware distribution system, which is implemented by public license prac- 412 ____________________________________________________________________________F.__The_GNU_General_Public_License______________* *______________________ tices. Many people have made generous contributions to the wide range of software distributed through that system in reliance on consistent ap- plication of that system; it is up to the author/donor to decide if he or she is willing to distribute software through any other system and a licensee cannot impose that choice. This section is intended to make thoroughly clear what is believed to be a consequence of the rest of this License. 8. If the distribution and/or use of the Program is restricted in certain coun- tries either by patents or by copyrighted interfaces, the original copyright holder who places the Program under this License may add an explicit ge- ographical distribution limitation excluding those countries, so that distri- bution is permitted only in or among countries not thus excluded. In such case, this License incorporates the limitation as if written in the body of this License. 9. The Free Software Foundation may publish revised and/or new versions of the General Public License from time to time. Such new versions will be similar in spirit to the present version, but may differ in detail to address new problems or concerns. Each version is given a distinguishing version number. If the Program specifies a version number of this License which applies to it and "any later version", you have the option of following the terms and conditions either of that version or of any later version published by the Free Soft- ware Foundation. If the Program does not specify a version number of this License, you may choose any version ever published by the Free Software Foundation. 10. If you wish to incorporate parts of the Program into other free programs whose distribution conditions are different, write to the author to ask for permission. For software which is copyrighted by the Free Software Foundation, write to the Free Software Foundation; we sometimes make exceptions for this. Our decision will be guided by the two goals of pre- serving the free status of all derivatives of our free software and of pro- moting the sharing and reuse of software generally. NO WARRANTY 11. BECAUSE THE PROGRAM IS LICENSED FREE OF CHARGE, THERE IS NO WARRANTY FOR THE PROGRAM, TO THE EXTENT PER- MITTED BY APPLICABLE LAW. EXCEPT WHEN OTHERWISE STATED IN WRITING THE COPYRIGHT HOLDERS AND/OR OTHER PAR- TIES PROVIDE THE PROGRAM "AS IS" WITHOUT WARRANTY OF ANY KIND, EITHER EXPRESSED OR IMPLIED, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MER- CHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE. THE ENTIRE RISK AS TO THE QUALITY AND PERFORMANCE OF THE PROGRAM IS WITH YOU. SHOULD THE PROGRAM PROVE DEFECTIVE, YOU ASSUME THE COST OF ALL NECESSARY SER- VICING, REPAIR OR CORRECTION. 12. IN NO EVENT UNLESS REQUIRED BY APPLICABLE LAW OR AGREED TO IN WRITING WILL ANY COPYRIGHT HOLDER, OR * * 413 F. The GNU General_Public_License_________________________________________________________________________________________________* *_______________________________________@ ANY OTHER PARTY WHO MAY MODIFY AND/OR REDISTRIBUTE THE PROGRAM AS PERMITTED ABOVE, BE LIABLE TO YOU FOR DAMAGES, INCLUDING ANY GENERAL, SPECIAL, INCI- DENTAL OR CONSEQUENTIAL DAMAGES ARISING OUT OF THE USE OR INABILITY TO USE THE PROGRAM (INCLUDING BUT NOT LIMITED TO LOSS OF DATA OR DATA BEING RENDERED INACCURATE OR LOSSES SUSTAINED BY YOU OR THIRD PAR- TIES OR A FAILURE OF THE PROGRAM TO OPERATE WITH ANY OTHER PROGRAMS), EVEN IF SUCH HOLDER OR OTHER PARTY HAS BEEN ADVISED OF THE POSSIBILITY OF SUCH DAMAGES. END OF TERMS AND CONDITIONS Appendix: How to Apply These Terms to Your New Programs If you develop a new program, and you want it to be of the greatest possible use to the public, the best way to achieve this is to make it free software which everyone can redistribute and change under these terms. To do so, attach the following notices to the program. It is safest to attach them to the start of each source file to most effectively convey the exclusion of warranty; and each file should have at least the "copyright" line and a pointer to where the full notice is found. < one line to give the program's name and a brief idea of what it does.> Copyright (C) 19yy < name of author> This program is free software; you can redistribute it and/or modify it under the terms of the GNU General Public License as published by the Free Soft- ware Foundation; either version 2 of the License, or (at your option) any later version. This program is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABIL- ITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for more details. You should have received a copy of the GNU General Public License along with this program; if not, write to the Free Software Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA. Also add information on how to contact you by electronic and paper mail. If the program is interactive, make it output a short notice like this when it starts in an interactive mode: Gnomovision version 69, Copyright (C) 19yy name of author Gnomovision comes with ABSOLUTELY NO WARRANTY; for details type `show w'. This is free software, and you are welcome to redistribute it under certain conditions; type `show c' for details. The hypothetical commands `show w' and `show c' should show the appro- priate parts of the General Public License. Of course, the commands you use may be called something other than `show w' and `show c'; they could even be mouse-clicks or menu items-whatever suits your program. You should also get your employer (if you work as a programmer) or your school, if any, to sign a "copyright disclaimer" for the program, if necessary. Here is a sample; alter the names: 414 ____________________________________________________________________________F.__The_GNU_General_Public_License______________* *______________________ Yoyodyne, Inc., hereby disclaims all copyright interest in the program `Gnomo- vision' (which makes passes at compilers) written by James Hacker. < signature of Ty Coon>, 1 April 1989 Ty Coon, President of Vice This General Public License does not permit incorporating your program into proprietary programs. If your program is a subroutine library, you may consider it more useful to permit linking proprietary applications with the library. If this is what you want to do, use the GNU Library General Public License instead of this License. * * 415 F. The GNU General_Public_License_________________________________________________________________________________________________* *_______________________________________@ 416 Bibliography [Almesberger, 1994] Almesberger, W. (1994). LILO User's guide. (siehe Datei /usr/doc/ lilo/user.dvi). [Bailey, 1997] Bailey, E. C. (1997). Maximum RPM. Red Hat. (ISBN 1-888172-78-9). [Cameron and Rosenblatt, 1991] Cameron, D. and Rosenblatt, B. (1991). Learning GNU Emacs. O'Reilly & Associates, Inc. (ISBN 0 937175-84-6). [Costales et al., 1993] Costales, B., Allman, E., and Rickert, N. (1993). sendmail. O'Reilly & Associates, Inc. (ISBN 1-56592-056-2). [Dawson, 1995] Dawson, T. (v2.8, 07 Jan 1995). Linux NET-2/NET-3 HOWTO. (siehe Datei /usr/doc/howto/NET-2-HOWTO). [Fang et al., 1993] Fang, C., Crosson, B., and Raymond, E. S. (1993). The Hitchhiker's Guide to X386/XFree86 Video Timing (or, Tweaking your Monitor for Fun and Profit). (siehe Datei /usr/X11/lib/X11/doc/VideoModes.doc). [Frisch, 1993] Frisch, AE. (1993). Essential System Administration. O'Reilly & Associates, Inc. (ISBN 0-937175-80-3). [Garfinkel and Spafford, 1993] Garfinkel, S. and Spafford, G. (1993). Pratical UNIX Security. O'Reilly & Associates, Inc. (ISBN 0-937175-72-2). [Gilly, 1992] Gilly, D. (1992). UNIX in a nutshell: System V Edition. O'Reilly & Associates, Inc. (ISBN 1-56592-001-5). [Goossens et al., 1993] Goossens, M., Mittelbach, F., and Samarin, A. (1993). TheALTEX Com- panion. Addison Wesley GmbH. (ISBN 3-54199-8). [Grieger, 1994] Grieger, W. (1994). Wer hat Angst vorm Emacs? Addison Wesley GmbH. (ISBN 3-89319-620-X). [Herold, 1992] Herold, H. (1992). UNIX Grundlagen. Addison Wesley GmbH. (ISBN 3-89319- 542-8). [Hetze et al., 1995] Hetze, S., Hohndel, D., Mu"ller, M., and Kirch, O. (1995). Linux Anwender- handbuch. LunetIX Softfair, 5.s edition. (ISBN 3-929764-04-0). [Hunt, 1995] Hunt, C. (1995). TCP/IP Netzwerk Administration. O'Reilly & Associates, Inc. (ISBN 3-930673-02-9). [Kirch, 1995] Kirch, O. (1995). LINUX Network Administrator's Guide. O'Reilly & Associates, Inc. (ISBN 1-56592-087-2). [Kofler, 1995] Kofler, M. (1995). Linux. Addison Wesley GmbH. (ISBN 3-89319-796-6). [Kopka, 1994] Kopka, H. (1994). ALTEX-Einfu"hrung. Addison Wesley GmbH. (ISBN 3-89319- 664-1). [Lamb, 1990] Lamb, L. (1990). Learning the vi Editor. O'Reilly & Associates, Inc. (ISBN 0-937175-67-6). * * 417 Bibliography _________________________________________________________________________________________________________________* *_______________________________________@ [Lamport, 1994] Lamport, L. (1994). ALTEX User's Guide and Reference Manual. Addison Wes- ley GmbH. (ISBN 0-201-52983-1). [Leffler, 1996a] Leffler, S. (1996a). HylaFAX Home Page. http://www.vix.com/hylafax/. [Leffler, 1996b] Leffler, S. (1996b). TIFF Software. http://www-mipl.jpl.nasa.gov/ ~ndr/tiff/. [O'Reilly and Todino, 1992] O'Reilly, T. and Todino, G. (1992). Manging UUCP and Usenet. O'Reilly & Associates, Inc. (ISBN 0-937175-93-5). [Perlman, 1994] Perlman, G. (1994). Unix For Software Developers. Prentice-Hall. (ISBN 13-932997-8). [Pugh, 1994] Pugh, K. (1994). UNIX For The MS-DOS User. Prentice-Hall. (ISBN 13-146077- 3). [Schoonover and Bowie, 1992] Schoonover, M. and Bowie, J. (1992). GNU Emacs. Addison Wesley GmbH. (ISBN 0-201-56345-2). [The XFree86TM -Team, 1996] The XFree86TM -Team (1996). XF86Config(4/5) - Configuration File for Xfree86TM . Manual-Page zu XFree86TM . [Todino et al., 1993] Todino, G., Strang, J., and Peek, J. (1993). Learning the UNIX operating system. O'Reilly & Associates, Inc. (ISBN 1-56592-060-0). [Welsh, 1994] Welsh, M. (1994). Linux Installation and Getting Started. S.u.S.E. GmbH, 2. edition. (ISBN 3-930419-03-3). [Welsh and Kaufman, 1995] Welsh, M. and Kaufman, L. (1995). Running Linux. O'Reilly & Associates, Inc. (ISBN 1-56592-100-3). 418 Index Symbols LILO Problems . . . . . . . 138 YaST . . . . . . . . . . . . * *. . . . . . 82 .susefaxrc 1TR6 . . . . . . . . . . . . . . . . . . 286 AdvanSys . . . . . . . . . . 103* *, 113 SuSEFax . . . . . . . . . . . . . 301 3COM 3c501 . . . . . . . . . . . 112 afterstep . . . . . . . . . . . * *. . . . . 237 .susephone 3COM 3c503 . . . . . . . 106, 112 afterstep . . . . . . . .* * 227, 237 SuSEFax . . . . . . . . . . . . . 301 3COM 3c505 . . . . . . . . . . . 112 Afterstep . . . . . . . . . . . * *. . . . 238 .tar . . . . . . . . . . . . . . . . . . . . . 3323COM 3c507 . . . . . . . . . . . 112 AfterStep . . . . . . 223, 227, * *237 .tar.gz . . . . . . . . . . . . . . . . . . 332 3COM 3c509 . . . . . . . . . . . 112 AIRCOP . . . . . . . . . . . . * *. . . 124 .tgz . . . . . . . . . . . . . . . . . . . . 3323COM 3c579 . . . . . . . . . . . 112 Alias . . . . . . . . . . . . . * *. . . . . . 385 / . . . . . . . . . . . . . . . . . . . . . . . 3483dpixm . . . . . . . . . . . . . . . . 241 alien . . . . . . . . . . . * *. . . . . . . . 97 /bin . . . . . . . . . . . . . . . . . . . . . 643dpixms . . . . . . . . . . . . . . . 241 Alien . . . . . . . . . . . . . * *. . . . . . . 97 /etc/crontab . . . . . . . . . . . . . 159 9wm . . . . . . . . . . . . . . . . . . . 224 AM53/79C974 . . . . . . . . . . * *104 /etc/fstab . . . . . . . . . . . . . 77 AMD 53/79C974 . . . . . . . . 104 /etc/printcap . . . . . . . . . . . . 146 A AmigaOS . . . . . . . . . . . .* * . . 391 /etc/rc.config . . . . . . . . . . . . 174 a.out . . . . . . . . . . . . . . . . 91, 185 Amstrad . . . . . . . . . . . 3* *56, 357 /etc/rc.config . . . . . . . . 87 a2ps . . . . . . . . . . . . . . . . . . . 148 Andrew Tridgell] . . . . . . . .* * 317 /sbin/SuSEconfig vii, 173, 174 abort . . . . . . . . . . . . . . . . . . . 281 Apache . . . . . . . . . . . . .* * . . . . 190 /sbin/init . . . . . . . . . . . . . . . 169 Account . . . . . . . . . . . . . . . . 385 APM . . . . . . . . . . . . . . * *. . . . . 203 /sbin/init.d/rc . . . . . . . . . . . 171 ACER 486/DX4-75 . . . . . 203 Apple MacIntosh . . . . . . . . * *389 /sbin/init.d/rpc . . . . . . . . . . 255 Acronym . . . . . . . . . . . . . . . 385 Appletalk . . . . . . . . . . . * *. . . . 192 /usr/dos . . . . . . . . . . . . . . . . 348 Adamem . . . . . . . . . . . . . . . 358 Applications /var/lib/apsfilter/bin/apsfilter . . Adaptec Configuration files . . . . * *242 148 AHA-152x/151x/1505 . 103 aps . . . . . . . . . . . 147, 1* *48, 153 Applixware . . . . . . . . . . . 3, 60 AHA-154x . . . . . . . . . . . 103 apsfilter 83, 85, 147-150, 153 LILO . . . . . . . . . . . . . . . . . . . 84 AHA-274x . . . . . . . . . . . 103 Check list . . . . . . . . .* * . . . 153 and DOS-Diskmanager 140 AHA-284x . . . . . . . . . . . 103 Archiving data . . . . . . . . .* * . 332 and EZ-Drives . . . . . . . . 140 AHA-294x . . . . . . . . . . . 103 arena . . . . . . . . . . . . . * *. . . . . . . 12 and OnTrack-Manager . 140 Adaptec 2920 . . . . . . . . . . . 107 ATAPI . . . . . . . . . . . . .* * . . . . 385 /opt . . . . . . . . . . . . . . . . . . . .A60daptec ATAPI-CD-ROM hangs . . . 53 /etc/host.conf . . . . . . . 175 AHA-152x/151x/1505 . . . ATARI . . . . . . . . . . . . . * *. . . . 361 /etc/hosts . . . . . . . . . . . . 175 112 atd . . . . . . . . . . . . . . * *. . . . . . . 179 /etc/resolv.conf . . . . 175 Adaptive Answer Support attrib . . . . . . . . . . . . .* * . . . . . . 352 S.u.S.E. GmbH . . . . . . . . . . . . 2 HylaFAX . . . . . . . . . . . . . 314 autoexec.bat . . . . . . . . . * *. . . . 46 XFree86 . . . . . . . . . . . . . . 2, 13 add AVM-B1 . . . . . . . . . . . . * *. . . 286 XFree86TM . . . . . . . . . . . . . . . 2 User . . . . . . . . . . . . . . . . . 327 Aztech CD-ROM . . . 113, 196 YaST addgroup . . . . . . . . . . . . . . . 298 Firsttime installation . . . . 32 adding B YaST . . . . . . . . . . . . . . . . . . 2, 3 Users . . . . . . . . . . . . . . . . 325 Background picture . . . . . . 2* *29 YaST Adding DOS partitions . . . . 76 Background process . . . . . 385 Installation . . . . . . . . . . . . 34 Administration Backup . . . . . . . . . . . . * *. 87, 385 1024 Cylindesr user . . . . . . . . . . . . . . . . . . . 86Base-Linux . . . . . . . . . . .* * . . 386 * * 419 Index _________________________________________________________________________________________________________________* *_______________________________________@ baseLinux Buffer . . . . . . . . . . . . . . . . . . 386 Sony CDU31A . . 110, 119, Boot methods . . . . . . . . . . 28 BusLogic . . . . . . . . . . . . . . . 105 198 bash . 284, 327, 329, 339, 393 Busmice Sony CDU33A . . 110, 119, Bash . . . . . . . . . . . . . . . . . . . . 93 Logitech . . . . . . . . . . . . . 202 198 Basic commands . . . . . . . . 327 Busmouse . . . . . . . . . 202, 211 Sony CDU535 . . . 110, 119, Bell Laboratories . . . . . . . . 277 ATIXL . . . . . . . . . . . . . . . 202 198 bin/faxrcvd . . . . . . . . . . . . . 316 Logitech . . . . . . . . . . . . . 108 SoundBlaster Pro 16 . . . 119 bind . . . . . . . . . . . . . . . . . . . 249 Microsoft . . . . . . . . . . . . . 202 Wearnes . . . . . . . . . . .* * . . . 196 bind . . . . . . . . . . . . . . . . . . . 295 PS/2 . . . . . . . . . . . . . . . . . 202 CD-ROM drives BIOS . . . . . . . . . . . . . . . . . . 386 Support by Linux . . . . . . * *. 57 bison . . . . . . . . . . . . . . . . . . . 344 C CD-ROM driver is missing 53 bitmap . . . . . . . . . . . . . . . . . 231 C . . . . . . . . . . . . . . . . . . . . . . 181CD-ROM-drive hangs . . . . 53 Bodo Bauer . . . . . . . . . . . . . 256 C News . . . . . . . . . . . . . . . . 297 CDE . . . . . . . . . . . . . . * *. . . . . 224 books . . . . . . . . . . . . . . . . . . . . 5 C128 . . . . . . . . . . . . . . . . . . . 360 cdesim . . . . . . . . . . .* * . . . . . 237 boot . . . . . . . . . . . . . . . . . . . 121 C64 . . . . . . . . . . . . . . . . . . . . 360 cdesim . . . . . . . . . . . . * *. . . . 237 additional systems . . . . . 123 Cabletron E21xx . . . . . . . . 113 CDEsim . . . . . . . . . . . . .* * . . . 237 booting DOS . . . . . . . . . 122 Cabletron E21XX . . . . . . . 106 Change job parameters disk . . . . . . . . . . . . . . . . . . 123 call . . . . . . . . . . . . . . . . . . . . 283 SuSEFax . . . . . . . . . . .* * . . 305 loader . . . . . . . . . . . . . . . . 121 call-login . . . . . . . . . . . . . . . 279 chat . . . . . . . . . . . . . * *. . 268, 281 manager . . . . . . . . . . . . . . 121 call-password . . . . . . . . . . . 280 chat-fail . . . . . . . . . . . * *. . . . . 281 sector . . . . . . . . . . . . . . . . 122 Card-Manager . . . . . . . . . . . 99 Check . . . . . . . . . . . . . * *. . . . . 405 Boot cardmgr . . . . . . . . . . . . . . . . 99 Check package dependencies . concepts . . . . . . . . . . . . . . 123 cat . . . . . . . . . . . . . . . . . . . . . 330 80 boot disk . . . . . . . . . . . . . . . 123 CBM . . . . . . . . . . . . . . . . . . 361 chmod . . . . . . . . . . . . .* * . . . . 333 Boot sector . . . . . . . . . . . . . 122 CD damaged . . . . . . . . . . . . . 53 Choosing packages . . . . . . . * *39 boot-disk . . . . . . . . . . . . . . . 125 CD-Live-System . . . . . . . . . 85 Choosing software . . . . . . . * *. 39 Boot-disk CD-ROM . . . . . . . . . . . . . . . 196 CIFS . . . . . . . . . . . . . .* * . . . . . 318 create . . . . . . . . . . . . . . . . 132 ATAPI . . . . . . . . . . . . . . . 104 Cirrus . . . . . . . . . . . . .* * . . . . . . 99 Boot-disks Aztech . . . . . . 104, 113, 196 Clipper . . . . . . . . . . . . * *. . . . . . . 2 Create using setup . . . . . . 29 Changer . . . . . . . . . . . . . . 193 Clock-Chip . . . . . . . . 209,* * 213 boot.local . . . . . . . . . . . . . . 172 Creative Labs . . . . . . . . . 197 CMOS setup . . . . . . . . . . .* * . 121 boot.sys . . . . . . . . . . . . . . . . 124 Doplhin . . . . . . . . . . . . . . 197 cnews . . . . . . . . . . . . . * *. . . . . 299 Booten . . . . . . . . . . . . . . . . . 405 EIDE . . . . . . . . . . . . . . . . 104 cnews . . . . . . . . . . . . . * *. . . . . 298 Booting . . . . . . . 169, 326, 386 Goldstar . . . . . 107, 115, 196 Cnews . . . . . . . . . . . . .* * . . . . 297 base-Linux . . . . . . . . . . . . 28 Jukebox . . . . . . . . . . . . . . 193 CNews . . . . . . . . . . . . . * *. . . . 300 Bootmanager . . . . . . . . . 123 Kotobuki . . . . . . . . . . . . . 197 Coherent . . . . . . . . . . . 2* *01, 338 Concepts . . . . . . . . . . . . . 123 Matsushita . . . . . . . . . . . 197 Coleco . . . . . . . . . . . . .* * . . . . 358 from disks . . . . . . . . . . . . . 29 Mitsumi . . . . . . . . . 108, 116 Coleco Vision . . . . . . . . . * *. . 358 Methods . . . . . . . . . . . . . . . 46 Mitsumi FX-001(D) . . 108, ColecoADAM . . . . . . . . . . * *358 Root partition . . . . . . . . . 109 116 ColecoVision . . . . . . . . . * *. . 358 Starting via loadlin . . . . . 46 Mitsumi Multisession . 108, Colour depth . . . . . . . . . .* * . . 220 Bootmanager 117 Colours . . . . . . . . . . . .* * . . . . 230 boot.sys . . . . . . . . . . . . . . 123 Mozart . . . . . . 108, 117, 197 Command LILO . . . . . . . . . . . . . . . . 123 Okano . . . . . . . . . . . . . . . 196 df . . . . . . . . . . . . .* * . . . . . . 335 OS/2 . . . . . . . . . . . . . . . . . 123 Optics Storage . . . 108, 117, du . . . . . . . . . . . . .* * . . . . . . 335 Windows NT . . . . . . . . . 123 197 free . . . . . . . . . . .* * . . . . . . 335 Bootmenu . . . . . . . . . . . . . . . 46 Orchid . . . . . . . . . . . . . . . 196 kill . . . . . . . . . . .* * . . . . . . 336 Borland . . . . . . . . . . . . . . . . . . 2 Panasonic . . . 111, 119, 197 ps . . . . . . . . . . . . .* * . . . . . . 336 bowman . . . . . . . . . . . 224, 237 Parallelport . . . . . . 108, 117 pstree . . . . . . . . . * *. . . . . 336 Bowman . . . . . . 223, 237, 238 Philips CM206 . . . 109, 118, top . . . . . . . . . . . .* * . . . . . . 336 BSD . . . . . . . . . . . . . . . . . . . 338 197 w . . . . . . . . . . . . . * *. . . . . . . 335 BSD db . . . . . . . . . . . . . . . . 168 Sanyo . . . . . . . 110, 118, 197 Command line . . . . . . . . . .* * 386 BSD-Unix . . . . . . . . . . . . . . . . 1 Sony . . . . . . . . . . . . . . . . . 198 command.com . . . . . . . . . . * *348 420 ___________________________________________________________________________________________________________________________I* *ndex__________________ Commands . . . see Commands DE204 . . . . . . . . . . . . . . . . . 114 starting . . . . . . . . . . * *. . . . . 275 basic . . . . . . . . . . . . . . . . . 327 DE205 . . . . . . . . . . . . . . . . . 114 wait . . . . . . . . . . . . * *. . . . . . 273 Commodore Business DEC . . . . . . . . . . . . . . . . . . . 207 diphosts . . . . . . . . . . . * *277, 278 Machines . . . . . . . . . . . 359 Decisions comments . . . . . . . . . .* * . . 278 Compiling Installation . . . . . . . . . . . . 19 local host . . . . . . . . . * *. . . . 278 DOS-emulator . . . . . . . . 344 defrag . . . . . . . . . . . . . . . . . . . 22 mtu . . . . . . . . . . . . .* * . . . . . 278 Compuserve . . . . . . . . . . . . . . 2 delete netmask . . . . . . . . . . .* * . . . 278 Computer Associates . . . . . . 2 User . . . . . . . . . . . . . . . . . 327 password . . . . . . . . . . * *. . . 278 config.sys . . . . . . . . . . . . . . . 46 Demo-mode . . . . . . . . . . . . . 59 protocol . . . . . . . . . . * *. . . . 278 Configuration depmod . . . . . . . . . . . . . . . . 182 remote host . . . . . . . . .* * . . 278 Change . . . . . . . . . . . . . . . . 78 desktop . . . . . . . . . . . . . . . . . . 56 user . . . . . . . . . . . . * *. . . . . . 278 changing . . . . . . . . . . . . . 173 Desktop diplogin . . . . . . . . . . . * *273, 277 E-Mail . . . . . . . . . . . . . . . 295 background . . . . . . . . . . . 229 directories LILO . . . . . . . . . . . . . . . . 126 colours . . . . . . . . . . . . . . . 230 important . . . . . . . . . .* * . . . 397 Load . . . . . . . . . . . . . . . . . . 78 Configuration . . . . . . . . . 242 Directories . . . . . . . . . .* * . . . 387 Network . . . . . . . . . . . . . . 247 cursor . . . . . . . . . . . . . . . . 231 Directory Taylor-UUCP . . . . . . . . . 278 fonts . . . . . . . . . . . . . . . . . 230 Change . . . . . . . . . . . * *. . . . 328 X11 . . . . . . . . . . . . . 208, 210 icons . . . . . . . . . . . . . . . . . 231 Creating . . . . . . . . . . * *. . . . 328 Configuration after Update 92 Deutsche Telekom . . . . . . . . . 2 Remove . . . . . . . . . . . * *. . . 328 Configuration file . . . . . . . 174 device . . . . . . . . . . . . . . . . . . 280 directory tree . . . . . . . . * *. . . 397 Configuration files . . . . . . . 248 Device . . . . . . . . . . . . . . . . . 218 Dirk Hohndel . . . . . . . . . 6* *, 207 Connecting a modem . . . . 267 Device-Section . . . . . . . . . . 218 Disk drives . . . . . . . . . .* * . . . 106 Console . . . . . . . . . . . . . . . . 386 devices change . . . . . . . . . . .* * . . . . . 55 virtual . . . . . . . . . . . . . . . . 386 floppy drive . . . . . . . . . . . 399 making bootable . . . . . . * *. 55 Consoles, virtual . . . . . . . . 326 IDE hard drive . . . . . . . . 399 Diskmanager (for DOS) . . 140 cp . . . . . . . . . . . . . . . . . . . . . 330 SCSI hard drive . . . . . . . 399 $DISPLAY . . . . . . . . 392, 394 CPU . . . . . . . . . . . . . . . . . . . 386 df . . . . . . . . . . . . . . . . . . . . . 335DNS . . . . . . . . . . . . .* * . 176, 295 Crash . . . . . . . . . . . . . . . . . . 405 DHCP . . . . . . . . . . . . . . . . . 322 DNS-domain . . . . . . . . . . * *. 253 CrashCourse . . . . . . . . . . . . 325 dial . . . . . . . . . . . . . . . . . . . . 281Doaminname . . . . . . . . . .* * . 247 Creating diald . . . . . . . . . . 268, 270, 271 documentation . . . . . . . . .* * . . . 4 Directory . . . . . . . . . . . . . 328 diald . . . . . . . . . . . . . . . . . . 268 read . . . . . . . . . . . . * *. . . . . . . . 5 Creative Labs . . . . . . . . . . . 197 dialer . . . . . . . . . . . . . . . . . . 281 docview Creative Labs CD-ROM . 197 Differences SuSEFax . . . . . . . . . . .* * . . 305 cron . . . . . . . . . . . . . . . 159, 298 Slackware . . . . . . . . . . . . 159 doexpire . . . . . . . . . . . .* * . . . . 299 Cron daemon . . . . . . . . . . . 179 Digiboard . . . . . . . . . . . . . . 201 DOLPHIN 8000AT . . . . . . 197 crontag . . . . . . . . . . . . . . . . . 296 Digital . . . . . . . . . . . . . . . . . 113 Dolphin CD-ROM . . . . . . 197 csh . . . . . . . . . . . . . . . . . . . . 284Digital DEPCA . . . . . . . . . 114 Domain . . . . . . . . . . . . * *. . . . 250 CSLIP . . . . . . . . . . . . . 273, 275 Digital Research . . . . . . . . 389 Domainname . . . . . . . . . . * *. 246 ctwm . . . . . . . . . . . . . . 224, 238 dip . . . . . . . . . . . . 272-275, 277 Donald E. Knuth . . . . . . . . * *378 Ctwm . . . . . . . . . 223, 237, 238 default . . . . . . . . . . . . . . . 275 DOS curses . . . . . . . . . . . . . vi, 168 dial . . . . . . . . . . . . . . . . . . 274 Assigning Linux partitions . Cursor . . . . . . . . . . . . . 231, 387 errlvl . . . . . . . . . . . . . . . . . 274 22, 23 Cyrix . . . . . . . . . . . . . . . . . . 186 Example script . . . 276, 278 Bootmenu . . . . . . . . . .* * . . . 46 get . . . . . . . . . . . . . . . . . . . 273DOS commands . . see mtools D local . . . . . . . . . . . . . . . . . 274 DOS disks D-Link DE620 . . . . . . . . . . 114 locip . . . . . . . . . . . . . . . . . 274 Access under Linux . . . 339 Du"sentrieb . . . . . . . . . . . . . . . . 4 mtu . . . . . . . . . . . . . . . . . . 275 DOS mode . . see Windows 95 Data backup . . . . . . . . . . . . 332 port . . . . . . . . . . . . . . . . . . 273 DOS setup . . . . . . . . . . . * *. . . . 27 Databook . . . . . . . . . . . . . . . . 99 reset . . . . . . . . . . . . . . . . . 273 DOS-emulator . . . . . . . . . * *. 343 dBASE . . . . . . . . . . . . . . . . . . . 2 send . . . . . . . . . . . . . . . . . 273 Compiling . . . . . . . . . .* * . . 344 Dblspace . . . . . . . . . . . . . . . 350 shutting down . . . . . . . . . 275 Starting . . . . . . . . . .* * . . . . 345 dd . . . . . . . . . . . . . . . . . . . . . 349 sleep . . . . . . . . . . . . . . . . . 274 Virtual disks . . . . . . . .* * . . 349 DE203 . . . . . . . . . . . . . . . . . 114 speed . . . . . . . . . . . . . . . . 273 Virtual drive . . . . . . . .* * . . 348 * * 421 Index _________________________________________________________________________________________________________________* *_______________________________________@ DOS-Emulator C64 . . . . . . . . . . . . . . . . . . 360 faxmodem . . . . . . . . . . . .* * . . 311 Quitting . . . . . . . . . . . . . . 348 CBM . . . . . . . . . . . . . . . . 359 faxprint . . . . . . . . . . . .* * . . 309 Drive label has changed . . . 56 Coleco Vision . . . . . . . . . 358 faxq . . . . . . . . . . . . . * *. . . . . . 311 Drivespace . . . . . . . . . . . . . . . 55 iNES . . . . . . . . . . . . . . . . . 363 fdformat . . . . . . . . . . . .* * . . . . 339 DSS1 . . . . . . . . . . . . . . . . . . 286 Installation . . . . . . . . . . . 364 fdisk 22, 23, 35, 36, 125, 130, du . . . . . . . . . . . . . . . . . . . . . 335 PET . . . . . . . . . . . . . . . . . 361 131, 135, 136, 141, 166 Dummy device . . . . . . . . . . 175 ROM . . . . . . . . . . . . . . . . 363 FIFO Datei . . . . . . . . . . .* * . . 311 dumpe2fs . . . . . . . . . . . . . . . 167 Spectrum . . . . . . . . . . . . . 357 file . . . . . . . . . . . . . .* * . . . . . . . 148 dvips . . . . . 148, 325, 377, 380 vcs2600 . . . . . . . . . . . . . . 361 File . . . . . . . . . . . . . .* * . . . . . . 388 vgb . . . . . . . . . . . . . . . . . . 362 Link . . . . . . . . . . . .* * . . . . . 331 E VIC-20 . . . . . . . . . . . . . . . 361 File attributes . . 328, 330, 332 E-Mail . . . . . . . . . . . . . . . . . 295 vice . . . . . . . . . . . . . . . . . . 359 File systems . . . . . . . . . * *. . . 198 Configuration . . . . . . . . . 295 ZX81 . . . . . . . . . . . . . . . . 356 File Transfer Protocol . . . . 3* *10 e2fsck . . . . . . . . . . . . . . . . . . 167 Enlightenment . . . . . . . . . . 224 files Manual-Page . . . . . . . . . 405 Entry-Editor devices . . . . . . . . . . .* * . . . . 399 2fsck . . . . . . . . . . . . . . . . . . 405 SuSEFax . . . . . . . . . . . . . 307 Files . . . . . . . . . . 218, 3* *28, 330 Eberhard Mo"nkeberg . . . . . . 6 Environment . . . . . . . . . . . . 387 Commands . . . . . . . . . .* * . 328 Editor . . . . . . . . . . . . . . . . . . 387 Environment variable . . . . 388 copy . . . . . . . . . . . . * *. . . . . 330 Editors EtherTeam 16i/32 . . . . . . . 116 delete . . . . . . . . . . . * *. . . . . 330 emacs . . . . . . . . . . . . . . . . 369 Excite . . . . . . . . . . . . . . . . . . 364 grepping . . . . . . . . . .* * . . . 331 vi . . . . . . . . . . . . . . . . . . . . 367exitemu . . . . . . . . . . . . . . . . 348 hidden . . . . . . . . . . . * *330, 402 editres . . . . . . . . . . . . . . . . . 243 expire . . . . . . . . . . . . . . . . . . 300 rename . . . . . . . . . . . * *. . . . 330 efsck . . . . . . . . . . . . . . . . . . . 166 Explanation of commands 333 search . . . . . . . . . . . * *. . . . . . 81 ei xii, 353, 354, 358, 359, 362, explorer . . . . . . . . . . . . . . . . . 49 searching . . . . . . . . . .* * . . . 331 364 exporting . . . . . . . . . . . . . . . 255 Files not movable . . . . . . .* * . 53 EIDE chipsets . . . . . . . . . . 105 exports . . . . . . . . . . . . . . . . . 255 Files-Section . . . . . . . . . * *. . . 218 ELF . . . . . . . . . . . . . . . . . . . 186 EXT2 . . . . . . . . . . . . . . . . . . 388 Filesystem . . . . . . . . . . .* * . . . 388 emacs . . . . . . xii, 61, 369-373 Extended partitions . . . . . . . 75 Ext2 . . . . . . . . . . . .* * . . . . . 199 emacs . . . . . . . . . . . . . 3, 4, 387 external Viewer UMSDOS . . . . . . . . . . .* * . . 76 emacs . . . . . . . . . . . . . . . . . . 369 SuSEFax . . . . . . . . . . . . . 305 Filesystems Backup-files . . . . . . . . . . 371 EZ-Drive . . . . . . . . . . . . . . . 140 assigning . . . . . . . . . .* * . . . . 37 Command overview . . . 373 Coherent . . . . . . . . . .* * . . . 201 Cursor control . . . . . . . . 370 F Ext . . . . . . . . . . . . * *. . . . . . 198 Execute macros . . . . . . . 372 F: . . . . . . . . . . . . . . . . . . . . . . 348 HPFS . . . . . . . . . . . . * *. . . . 201 Info mode . . . . . . . . . . . . 373 fallout . . . . . . . . . . . . . . . . . . 351 ISO9660 . . . . . . . . . . * *. . . 200 Load texts . . . . . . . . . . . . 371 FAQs . . . . . . . . . . . see manuals Minix . . . . . . . . . . . .* * . . . . 198 Macros . . . . . . . . . . . . . . . 372 Fax cover msdos . . . . . . . . . . . .* * . . . . 199 On-line help . . . . . . . . . . 370 SuSEFax . . . . . . . . . . . . . 306 NetWare Volumes . . . . . 200 Quit . . . . . . . . . . . . . . . . . 370 Fax Server proc . . . . . . . . . . . .* * . . . . . 200 Save macros . . . . . . . . . . 372 HylaFAX . . . . . . . . . . . . . 310 Quota . . . . . . . . . . . .* * . . . . 198 Search and replace . . . . 371 Fax cover System V . . . . . . . . . . * *. . . 201 Tutorial . . . . . . . . . . . . . . 370 SuSEFax . . . . . . . . . . . . . 308 umsdos . . . . . . . . . . . * *. . . . 199 Undo . . . . . . . . . . . . . . . . 370 Fax Send at VFAT . . . . . . . . . . . . * *. . . . 199 Windows . . . . . . . . . . . . . 371 SuSEFax . . . . . . . . . . . . . 306 Xia . . . . . . . . . . . . * *. . . . . . 199 Emacs . . . . . . . . . . . . . . . . 4, 96 fax2ps . . . . . . . . . . . . . . . . . 305 Filter Email . . . . . . . . . . . . . . 267, 387 faxcover . . . . . . . . . . . . . . . . 309 apsfilter . . . . . . . . . * *. . . . . 147 emm386.exe . . . . . . . . . . 27, 46 faxcover . . . . . . . . . . . . . . 309 Printer filter . . . . . . 14* *6, 147 Emuinstall . . . . . . . . . . . . . . 364 Faxcover find . . . . . . . . . . . . . * *. . 166, 331 Emulator . . . . . . . . . . . . . . . 343 SuSEFax . . . . . . . . . . . . . 308 finger . . . . . . . . . . . . * *. . . . . . 402 Adam . . . . . . . . . . . . . . . . 358 FaxCovergen.class . . . . . . 309 fips . . . . . . . . . . . . . .* * . 17, 21, 56 Amiga . . . . . . . . . . . . . . . 354 FaxCovergen.class fips.exe . . . . . . . . . . . * *. . . 22, 23 Atari . . . . . . . . . . . . . . . . . 352 SuSEFax . . . . . . . . . . . . . 309 firewall . . . . . . . . . 256,* * 257 C128 . . . . . . . . . . . . . . . . . 360 faxgetty . . 311, 312, 314, 315 Firewall . . . . . . . . . . . .* * . . . . 256 422 ___________________________________________________________________________________________________________________________I* *ndex__________________ Firsttime installation setting . . . . . . . . . . . . . . . . 231 gsview . . . . . . . . . . . . .* * . . . . . . 5 via HD . . . . . . . . . . . . . . . . 25 start . . . . . . . . . . . . . . . . . . 229gv . . . . . . . . . . . . . . * *. . . . . . . 305 Firsttime installation fvwm1 . . . . . . . . . . . . . . . . . . 227 gv . . . . . . . . . . . . . .* * . . . . . . . 305 Assigning file systems . . 37 fvwm2 . . . . . . . . viii, 223-225, gv . . . . . . . . . . . . . . .* * . . . . 5, 305 Boot methods . . . . . . . . . . 28 227-230, 237, 239-241 gzip . . . . . . . . . . . . . .* * . . . . . . . 64 copying packages . . . . . . 25 fvwm2 . . . . . . . . . . . . . . . . . . 227gzip . . . . . . . . . . . . * *. . . . . . . 332 Creating boot-disks with Fvwm2 . . . 223, 227, 237, 238 Unix . . . . . . . . . . . . . . . . 32 Fvwm2 . . . . . . . . . . . . . . 227 H Designing partitions . . . . 35 Configuration files . . . . 228 halt . . . . . . . . . . . . . * *. . . . . . . 172 Setup . . . . . . . . . . . . . . . . . 27 Configure . . . . . . . . . . . . 229 Harald Ko"nig . . . . . . . . .* * . . . . 6 Starting YaST . . . . . . . . . . 32 General . . . . . . . . . . . . . . 227 Hardware FlagShip . . . . . . . . . . . . . . . . . . 2 Settings . . . . . . . . . . . . . . 229 Changegeable media . . . 155 flex . . . . . . . . . . . . . . . . . . . . 344Fvwm2 . . . . . . . . . . . . . . . . 229 Disk drives . . . . . . . . * *. . . 156 Floppy . . . . . . . . . . . . . . . . . 106 fvwm95 . . . . . . . 224, 237, 241 Integration . . . . . . . . .* * . . . 155 Floppy-Streamer see Streamer Fvwm95 . . . . . . 223, 237, 238 ISApnp . . . . . . . see ISA* *pnp Florian La Roche . . . . . . . . . . 5 FvwmBanner . . . . . . . . . . . 229 LS120 drives . . . . . . . .* * . 156 Focus . . . . . . . . . . . . . . . . . . 388 FvwmButtons . . . . . . . . . . . 228 Plug-and-Play . . . . . . . * *. 172 Fonts . . . . . . . . . . . . . . . . . . 230 FvwmIdent . . . . . . . . . . . . . 231 PlugAndPlay . . . . . . . . * *. 155 Form fax FX-400 . . . . . . . . . . . . . . . . 197 Scanner . . . . . . . . . . * *. . . . 156 SuSEFax . . . . . . . . . . . . . 308 ZIP drives . . . . . . . . . * *. . . 156 create form fax list G help . . . . . . . . . . . . . .* * . . . . . . . . 4 SuSEFax . . . . . . . . . . . . . 308 Gameboy . . . . . . . . . . . . . . . 362 books . . . . . . . . . . . * *. . . . . . . 5 Formatting gated . . . . . . . . . . . . . . . . . . . 191 Hercules graphic device . . . 84 partition . . . . . . . . . . . . . . . 77 Gateway . . . . . . . . . . . 247, 251 Hercules-graphics card. . . 149 Foxbase . . . . . . . . . . . . . . . . . . 2 Gateway Address . . . . . . . 247 hfaxd . . . . . . . . . . . . .* * . . . . . 311 free . . . . . . . . . . . . . . . . . . . . 200Gateway-address . . . . . . . . 248 hiddenHfilesiS.a.x. .............* *......133096, 286, 288 free . . . . . . . . . . . . . . . . . . . 335GEM . . . . . . . . . . . . . . . . . . 389 $HOME . 227, 235, 243, 388, Free Software Foundation getty . . . . . . . . . . . . . . 314, 315 389 (FSF) . . . . . . . . . . . . . . 389 Ghostscript . . . . . . . . 148, 150 Home directory . . . . . . . . * *. 389 FreeBSD . . . . . . . . . . . . . . . 201 GhostScript . . . . . . . . . . . . . . 83 Horizontal frequency . . . . 212 ftp . . . . . . . . . . . . . . . . 275, 402 ghostview . . . . . . . . . . . . . . 231 host.conf . . . . . . . . . . .* * . . . . 249 FTP . . . . . . . . . . . . . . . . . . . . 388 GhostView . . . . . . . . . . . . . 305 alert . . . . . . . . . . . * *. . . . . . 249 Client . . . . . . . . . . . . . . . . 388 GhostView . . . . . . . . . . . . . 305 multi . . . . . . . . . . . .* * . . . . . 249 Server . . . . . . . . . . . . . . . . 388 GNU . . . 2, 168, 369, 373, 389 nospoof . . . . . . . . . . * *. . . . 249 Fujitsu FMV-181/182/183/184 GNU dbm . . . . . . . . . . . . . . 168 order . . . . . . . . . . . .* * . . . . . 249 115 GNU Emacs . . . . . . . . . . 3, 387 trim . . . . . . . . . . . . * *. . . . . . 250 fun . . . . . . . . . . . . . . . . . . . . . . .G7NU zip . . . . . . . . . . . . . . . 332 Hostname . . . . . . . . . . 246* *, 247 Function GNU-C/C++ compiler . . . . . 2 HOSTNAME . . . . . . . . . . . 2* *50 HylaFAX . . . . . . . . . . . . . 310 Goldstar CD-ROM . . 115, 196 hosts . . . . . . . . . . 248, 2* *49, 273 Future Domain . . . . . 107, 115 Goldstar-CD-ROM . . . . . . 107 Hotline . . . . . . . . . . . . * *. . . . . . 15 fvwm . . . . . . . . . 224, 227, 237 GPL . . . . . . . . . . . . . . . 325, 409 HowTo files . . . . . see manuals fvwm . . . . . . . . . . . . . . . . . . . 227 gpm . . . . . . . . . . . . . . . . 45, 178 HP 10/100 VG-AnyLAN . 116 Fvwm . . . . . . . . . 223, 237, 238 Gpm . . . . . . . . . . . . . . . . . . . 178 HP PCLAN . . . . . . . . . . . .* * . 115 Fvwm . . . . . . . . . . . . . . . 227 GPM . . . . . . . . . . . . . . . . . . . 178 HP PCLAN+ . . . . . . . . . . .* * 115 Background picture gra . . . . . . . . . . . . . . . . . . . . 157 httpd . . . . . . . . . . . . . * *. . . . . . 179 setting . . . . . . . . . . . . . . . . 229 Graphical User Interface . 389 hylafax . . . . . . . . . . . . * *. . . . . 301 colours . . . . . . . . . . . . . . . 230 grep . . . . . . . . . . . . . . . 166, 331 hylafax . . . . . . . . . . . 30* *1, 309 setting . . . . . . . . . . . . . . . . 230 groups . . . . . . . . . . . . . . . . . 254 HylaFAX . . . x, 300, 301, 306, cursor . . . . . . . . . . . . . . . . 231 gs . . . . . . . . . . . see Ghostscript 310, 310__, 311, 316 setting . . . . . . . . . . . . . . . . 231 gs . . . . . . . . . . . . . . . . . . . . . .h83ypertext-help . . see susehe* *lp fonts . . . . . . . . . . . . . . . . . 230 gs(1) . . . . . . . . . . . . . . . . . . . 150 setting . . . . . . . . . . . . . . . . 230 $GS_RESOL . . . . . . . . . . . 153 I icons . . . . . . . . . . . . . . . . . 231 gs__x11 . . . . . . . . . . . . . . . . . 305 i4l . . . . . . . . . . . . . .* * . 195, 286 * * 423 Index _________________________________________________________________________________________________________________* *_______________________________________@ i4ldoc . . . . . . . . . . . . 286, 294 via HD partition . . . . . . . . 69 kdm . . . . . . . . . . . . . .* * . . . . . 234 Ian Taylor . . . . . . . . . . . . . . 278 via NFS . . . . . . . . . . . . . . . 70 Kernel . . . . . . . . . . . . * *. . . . . 181 IBM . . . . . . . . . . . . . . . . . . . . 99 Windows 95 Block devices . . . . . . . .* * . 187 IBM OS/2 . . . . . . . . . . . . . . 337 DOS mode . . . . . . . . . . . . . 27 CD-ROM . . . . . . . . . . .* * . 196 IBM Thinkpad . . . . . . . . . . 107 Writing partition table . . 41 Compiling . . . . . . . . . .* * . . 181 Icewm . . . . . . . . . . . . . . . . . 224 Installation media . . . . . . . . 67 Configuration . . . . . . . .* * . 183 ICL EtherTeam . . . . . . . . . 116 Intel . . . . . . . . . . . . . . . . 99, 186 File systems . . . . . . . .* * . . 198 ICN . . . . . . . . . . . . . . . 196, 286 Intel EtherExpress 16 . . . . 114 ISDN . . . . . . . . . . . .* * . . . . 195 Icons . . . . . . . . . . . . . . . . . . . 231 Intel EtherExpressPro . . . . 115 Misc. . . . . . . . . . . . * *. . . . . 201 Identifier . . . . . . . . . . . . . . . 220 Interface . . . . . . . . . . . . . . . 389 Module . . . . . . . . . . .* * . . . 184 ifconfig . . . . . . . . . . . . 166, 175 Internet . . . . . . . . . . . . . . . . 389 Modules . . . . . . . . . . .* * . . . 181 Important keys . . . . . . . . . . 383 Iomega ZIP drive . . . . . . . . 116 Network . . . . . . . . . . .* * . . . 190 importing . . . . . . . . . . . . . . . 255 Iomega ZIP Drive . . . . . . . 107 Networking cards . . . . . 1* *93 Index of all series and IP-address . . . . . . . . . 247, 248 Parameter . . . . . . . . . * *. . . 101 packages . . . . . . . . . . . . 81 IP-Address . . . . . . . . . . . . . 251 Parameters . . . . . . . . . * *. . . . 57 iNES . . . . . . . . . . . . . . . . . . . 363 IP-allocation SCSI . . . . . . . . . . . . * *. . . . . 192 inetcfg . . . . . . . . . . . . . . . 294 dynamic . . . . . . . . . . . . . . 274 Selection . . . . . . . . . * *. . . . . 56 inetd . . . . . . . . . . . . . . 176, 251 static . . . . . . . . . . . . . . . . . 274 Kernel daemon . . . . . . . . . * *. 172 Inetd . . . . . . . . . . . . . . . . . . . 311 ipppd . . . . . . . . . . . . . . . . . . 293 Kernel profiling . . . . . . . .* * . 204 info . . . . . . . . . . . . . . . . . . 4, 373 IPX . . . . . . . . . . . . . . . . . . . . 318 Kernel-Parameters . . . . . . . * *. 57 info . . . . . . . . . . . . . . . . 4, 168 IRC . . . . . . . . . . . . . . . . . . . . 176 kerneld . . . . . . . . . 99, 17* *7, 182 Info . . . . . . . . . . . . . . . . . . . . 168isapnp . . . . . . . . . . . . . . . . . . 155 Kerneldaemon . . . . . . 177, 182 Infoviewer . . . . . . . . . . . . . . . . 4 ISApnp . . . . . . . . . . . . . . . . 155 kernmod . . . . . . . . . . . .* * . . . 286 init . . . . . . vi, vii, 92, 166-172 ISDN . . . . . . . . . . . . . . 195, 284 Keyboard . . . . . . . . . . . .* * . . . 218 Scripts . . . . . . . . . . . . . . . 171 Configuration . . . . . . . . . 285 CapsLock . . . . . . . . . .* * . . 174 Init . . . . . . . . . . . . . . . . 169, 311 YaST . . . . . . . . . . . . . . . . 286 Delay . . . . . . . . . . . .* * . . . . 174 Inode . . . . . . . . . . . . . . . 76__, 390 isdn4linux . . . . . . . . . . . . . . 284 Map . . . . . . . . . . . . * *. . . . . 174 Density . . . . . . . . . . . . . . . . 85 isdn4linux . . . . . . . . . . . . 285 NumLock . . . . . . . . . . * *. . 174 insmod . . . . . . . . . . . . 101, 182 ISDN4linux . . . . . . . . . . . . 195 repeat . . . . . . . . . . . * *. . . . . 174 Parameter . . . . . . . . . . . . 101 isdnctrl . . . . . . . . . . . . . . . . . 285 Keyboard-Section . . . . . . . 2* *18 installation isdnlog . . . . . . . . . . . . . . . . . 288 Keymap wrong in DOS-mode demo-mode . . . . . . . . . . . . 71 ISP16 . . . . . . . . . . . . . 108, 117 53 from directory . . . . . . . . . 71 ISP16/MAD16/Mozart . . . 197 kfm . . . . . . . . . . . . . .* * . 235, 236 Installation . . . . . . . . . . . . . . 17 kill . . . . . . . . . . . . * *. . . . . . . 336 Boot-disks . . . . . . . . . . . . . 29 J Kotobuki . . . . . . . . . . . .* * . . . 197 Choosing software . . . . . . 39 JAVA . . . . . . . . . . . . . . . . . . 186 Kotobuki CD-ROM . . . . . 197 de-install packages . . . . . 93 Java Developers Kit . . . . . 186 krpm . . . . . . . . . . . . . .* * . . . . . . 97 Decisions . . . . . . . . . . . . . . 19 JAZ-Drives . . . . . . . . . . . . . . 99 kwm . . . . . . . . . . . . . .* * 237, 238 Finish . . . . . . . . . . . . . . . . . 44 jed . . . . . . . . . . . . . . . . . . . . . 369Kwm . . . . . . . . . . . . . * *. . . . . 224 Formatting your HD . . . . 43 Job (Fax-) from CD-ROM . . . . . . . . . 67 SuSEFax . . . . . . . . . . . . . 303 L future boot-methods . . . . 46 Job parameters LAN . . . . . . . . . . 208, 245* *, 390 install packages . . . . . . . . 93 SuSEFax . . . . . . . . . . . . . 304 Lan Manager . . . . . . . . . .* * . 337 Installing loadlin . . . . . . . 30 Job priority LAN Manager . . . . . . . . . .* * 318 Installing packages . . . . . 43 SuSEFax . . . . . . . . . . . . . 304 Lance . . . . . . . . . . . . . * *. . . . . 106 LILO . . . . . . . . . . . . . . . . 130 Job-ID . . . . . . . . . . . . . . . . . 307 LaTeX . . . . . . . . . . . . .* * . . . . 377 Media . . . . . . . . . . . . . . . . . 67 jove . . . . . . . . . . . . . . . . . . . . 369latex-cover . . . . . . . .* * . . . . 309 Partitioning . . . . . . . . . . . . 21 jurix . . . . . . . . . . . . . . . . . . . . .l5atex-cover . . . . . . . . . * *. 309 Principle . . . . . . . . . . . . . . 19 Laurels . . . . . . . . . . . . * *. . . . . . . 5 Problemdescription . . . . . 53 K less . . . . . . . . . . . . . * *5, 166, 330 Size of . . . . . . . . . . . . . . . . 77 KDE . . . . . . . . . . . . 60, 97, 225 Letter with LATEX . . . . . . . * *. 377 Starting via loadlin . . . . . 46 $KDEDIR . . . . . . . . . . . . . 233 License . . . . . . . . . . . .* * . . . . 409 Steps . . . . . . . . . . . . . . . . . . 19 kdisplay . . . . . . . . . . . 236, 238 lilo . . . . . . . . . . . . . .* * . . . 57, 121 424 ___________________________________________________________________________________________________________________________I* *ndex__________________ Components . . . . . . . . . . 124 logout . . . . . . . . . . . . . . . . . . 325 mdir . . . . . . . . . . . . . * *. . . . . . 340 Configuration . . . . . . . . . 126 loopback . . . . . . . . . . . . . . . . 85 Memory . . . . . . . . . . . . .* * . . . 390 Interface . . . . . . . . . . . . . 124 Loopback . . . . . . . . . . 175, 247 protect . . . . . . . . . . * *. . . . . 109 Introduction . . . . . . . . . . 122 lpc . . . . . . . . . . . . . . . . . . . . . 145 reserve . . . . . . . . . . .* * . . . . 109 What is it . . . . . . . . . . . . . 124 lpd . . . . . . . . . . . . . . . . 145-147 Memory check . . . . . . . . . .* * 109 where to install . . . . . . . 125 lpd(8) . . . . . . . . . . . . . . . . . . 144 Menu . . . . . . . . . . . . . * *. . . . . 390 LILO lpq . . . . . . . . . . . . . . . . . . . . . 145MetaFont . . . . . . . . . . . * *. . . . 377 Boot-disk . . . . . . . . . . . . . 132 lpr . . . . . . . . . . . . 144, 145, 147 mformat . . . . . . . . . . . . * *. . . . 340 DOS booting . . . . . . . . . 136 lpr(1) . . . . . . . . . . . . . . . . . . 144 mgetty . . . . . . . . 271, 300* *, 315 DOS/Boot Win 95 . . . . . 133 lprm . . . . . . . . . . . . . . . . . . . 145 mgetty . . . . . . . 271, 301, * *315 Installation . . . . . . . . . . . 130 lprold . . . . . . . . . . . . . . . . 144 Microsoft . . . . . . . . . . .* * . 2, 318 OS/2 booting . . . . . 135, 136 ls . . . . . . . . . . . . . . . . . 328, 402 Microsoft Corporation . . . . .* * 2 Parameter . . . . . . . . . . . . 101 ls-R(5) . . . . . . . . . . . . . . . . . 377 Midnight Commander . . . . 96 Problems . . . . . . . . . . . . . 136 lsmod . . . . . . . . . . . . . . . . . . 182 Mike Lesk . . . . . . . . . . . * *. . . 277 1024 Cylinders . . . . . . . . . 138 LUN . . . . . . . . . . . . . . . . . . . 104 minicom . . . . . . . . . . . . * *93, 267 Diagnosis . . . . . . . . . . . . . 137 lx__suse . . . . . . . . . . . . . 5, 285 Minicom . . . . . . . . . . . .* * . . . 267 Kernel from 2.0 onwards 141 Lycos . . . . . . . . . . . . . . . . . . 364 MIT . . . . . . . . . . . . . .* * . . . . . 207 Start messages . . . . . . . . . 137 lynx . . . . . . . . . . . . . . . . . . . . . 12Mitsumi . . . . . . . . . . . .* * . . . . 197 Remove . . . . . . . . . . . . . . 131 Mitsumi CD-ROM . . 108, 116 Sample configurations . 133 M Mitsumi FX-001(D) 108, 116 Windows NT booting . . 134 m4 . . . . . . . . . . . . . . . . 239, 297 mke2fs . . . . . . . . . . . . .* * . . . . . 55 lilo.conf . . . . . . . . . . . . . . . . 126 MacOS . . . . . . . . . . . . . . . . . 391 mkfs . . . . . . . . . . . . . .* * . . . . . 166 Link MAD16 . . . . . . . . . . . 108, 117 mksusewmrc . . . . . . . . . . .* * . 240 Symbolic . . . . . . . . . . . . . 331 Main memory mkswap . . . . . . . . . . . . .* * . . . 166 Linus Torvalds . . . . . . . . 1, 3, 6 Size . . . . . . . . . . . . . . . . . . 110 mlabel . . . . . . . . . . . . * *. . . . . 340 linux . . . . . . . . . . . . . . . . . . . . 5 Main memory (RAM) . . . 390 Mlvwm . . . . . . . . . . . . .* * . . . 224 Linux . . . . . . . . . . . . . . 309, 391 makeindex . . . . . . . . . . . . . . 377 mmd . . . . . . . . . . . . . . * *. . . . . 340 Start . . . . . . . . . . . . . . . . . 325 makemap . . . . . . . . . . . . . . . 297 Modeline . . . . . . . . . . 21* *8, 221 Linux installation . . . . . . . . . 17 Management Modem . . . . . . . . . . . . .* * . . . 174 linux.bat . . . . . . . . . . 30, 31, 42 user . . . . . . . . . . . . . . . . . . . 86modprobe . . 57, 101, 111, 182 linuxrc . . ii, 14, 17, 18, 21, 29, $MANPATH . . . . . . . . . . . 388 Parameter . . . . . . . . . * *. . . 101 32, 33, 100, 160, 162, 164, Manual . . . . . . . . . . . . . . . . . 333 Module . . . . . . . . . . . . * *. . . . 184 165 Manual pages . . . . . . . . . . . 333 Filesystems . . . . . . . . .* * . . 198 Linuxrc . . . . . . . . . . . . . . . . 246 manuals . . . . . . . . . . . . . . . . . . 4 handling . . . . . . . . . . * *. . . . 182 Live-Filesystem . . . . . . . . . . 85 FAQs . . . . . . . . . . . . . . . . . . . 5 Laden . . . . . . . . . . . .* * . . . . 162 Live-System . . . . . . . . . . . . . 59 HowTo files . . . . . . . . . . . . . 5 Parameter . . . . . . . . . * *. . . 163 Live-File-System . . . . . . . . . 76 man . . . . . . . . . . . . . . . . . . . .M4odules . . . . . . . . . . . * *. . . . . 181 ln . . . . . . . . . . . . . . . . . . . . . 331 README files . . . . . . . . . . 5 compile . . . . . . . . . . * *. . . . 204 loadlin . 20, 24, 25, 28, 30-33, mars_nwe . . . . . . . . . . . . . . 322 Monitor . . . . . . . . . . . . * *. . . . 218 42, 46, 47, 55, 57, 59, 60, Masquerading . . . . . . . . . . . 256 Monitor-Section . . . . . . . . * *. 218 101, 102, 123, 138, 142, Mass storage media . . . . . 390 Monitors . . . . . . . . . . . 2* *10, 212 162 Master Boot Record . . 84, 121 more . . . . . . . . . . . . . .* * . . . . . 330 Loadlin . . . . . . . . . . . . . . . . . . 30 Matsushita CD-ROM . . . . 197 Motif . . . . . . . . . . . . .* * . . . . . 224 Loadlin doesnt start . . . . . . . 55 Matthias Wimpf . . . . . . . . . 140 mount . . . . . . . . . 166, 255* *, 338 Loadlin doesnt work . . . . . . 55 mattrib . . . . . . . . . . . . . . . . . 340 mountd . . . . . . . . . . . . .* * . . . . 256 loadlin.exe . . . . . . . . . 6, 28, 30 MBR . . . . . 121, 122, 125, see Mountpoint . . . . . . . . . . .* * . . 390 Local Area Network see LAN Master Boot Record Mouse . . . . . . . . . . . . .* * . . . . 174 locate . . . . . . . . . . . . . . . . . . 178 seeMaster Boot Record 122 Bus . . . . . . . . . . . . * *. 208, 211 Log files . . . . . . . . . . . . . . . 178 mc . . . . . . . . . . . . . . . . . . . . . . 96GlidePoint . . . . . . . . . * *. . . 208 Logging in . . . . . . . . . . . . . 325 mcd . . . . . . . . . . . . . . . . . . . . 340 HiTablet . . . . . . . . . 20* *8, 211 Logical partitions . . . . . . . . . 75 mcd: init failed . . . . . . . . . . . 54 Logitech . . . . . . . . . 20* *8, 211 Logitech . . . . . . . . . . . . . . . 211 mcopy . . . . . . . . . . . . . . . . . 340 Logitech (MouseMan) 208, Logitech Busmouse . . . . . 108 mdel . . . . . . . . . . . . . . . . . . . 340 211 * * 425 Index _________________________________________________________________________________________________________________* *_______________________________________@ Microsoft . . . . . . . . 208, 211 Configuration . . . . . . . . . 247 Lance . . . . . . . . . . . .* * . . . . 117 MM-Series . . . . . . . 208, 211 Configuration files . . . . 248 Novell NE1000/2000 . . 117 Mouse Systems . . . 208, 211 configuration with YaST 85 SMC 9194 . . . . . . . . . . * *. . 118 PS/2 . . . . . . . . . . . . . 208, 211 Dummy device . . . . . . . . 175 SMC Ultra . . . . . . . . . * *. . 118 Mouse type . . . . . . . . . . . . . 211 Parallel Port . . . . . . . . . . 118 Token ring . . . . . . . . . * *. . . 116 Mousebuttons . . . . . . . . . . . 211 Network address . . . . . . . . 251 Tulip . . . . . . . . . . . .* * . . . . . 113 Mozart . . . . . . . . . . . . 108, 117 Network File System see NFS WD80x3 . . . . . . . . . . .* * . . 119 Mozart CD-ROM . . 108, 117, Network Information Service . Western Digital . . . . . . * *. 119 197 see NIS Networking cards . . . . . . . * *106 mrd . . . . . . . . . . . . . . . . . . . . 340 Networking card networks . . . . . . . . . . . * *. . . . 248 mread . . . . . . . . . . . . . . . . . . 340 3COM 3c501 . . . . . . . . . 112 News . . . . . . . . . . . . . * *. . . . . 267 mren . . . . . . . . . . . . . . . . . . . 340 3COM 3c503 . . . . . . . . . 112 newsdaily . . . . . . . . . . .* * . . . 299 MS-DOS . . . . . . . . . . . . . . . . . 2 3COM 3c505 . . . . . . . . . 112 newsrun . . . . . . . . . . . . * *. . . . 299 MS-Windows . . . . . . . . . . . 389 3COM 3c507 . . . . . . . . . 112 newswatch . . . . . . . . . . .* * . . 299 msdos.sys . . . . . . . . . . . . . . . 46 3COM 3c509 . . . . . . . . . 112 NeXTstep . . . . . . . . . . 201* *, 338 mtools . . . . . . . . 156, 339, 347 3COM 3c579 . . . . . . . . . 112 NeXTSTEP . . . . . . . . . . . .* * . 224 mtools . . . 339, 340, 346, 353 AM7990 Chipset . . . . . . 117 NFS . . . . see Filesystems, 255 mtype . . . . . . . . . . . . . . . . . . 340 Cabletron . . . . . . . . . . . . . 113 Server . . . . . . . . . . . * *. . . . . 177 Multiprocessing . . . . . . . . . 390 D-Link DE620 . . . . . . . . 114 NFS-Client . . . . . . . . . . .* * . . 255 Multiprocessor . . . . . . . . . . 183 DE10x . . . . . . . . . . . . . . . 114 NFS-server . . . . . . . . . . .* * . . 255 Multisession CD-ROM . . 108, DE20 . . . . . . . . . . . . . . . . 114 NFS-Server . . . . . . . . . . .* * . . 255 117 DE203 . . . . . . . . . . . . . . . 114 nfsd . . . . . . . . . . . . . .* * . . . . . . 256 multisoft Datentechnik GmbH DE204 . . . . . . . . . . . . . . . 114 NI6510 . . . . . . . . . . . . .* * . . . . 117 2 DE205 . . . . . . . . . . . . . . . 114 Nintendo . . . . . . . . . . . 3* *62, 363 Multitasking . . . . . . . . . . . . 391 DE42 . . . . . . . . . . . . . . . . 114 Nintendo Entertainment Multiuser . . . . . . . . . . . . . . . 391 DE425 . . . . . . . . . . . . . . . 113 System . . . . . . . . . .* * . . . 363 mv . . . . . . . . . . . . . . . . . . . . . 330 DE434 . . . . . . . . . . . . . . . 113 NIS . . . . . . . . . . . . . . * *. . . . . . 253 mwm . . . . . . . . . . . . . . . . . . 238 DE435 . . . . . . . . . . . . . . . 113 NIS-domain . . . . . . . . . . * *. . 253 Mwm . . . . . . . . . 223, 237, 238 DE450 . . . . . . . . . . . . . . . 113 NIS-server . . . . . . . . . . .* * . . . 254 mwm . . . . . . . . . . . . . . . . . . 224 DE500 . . . . . . . . . . . . . . . 113 nkita . . . . . . . . 153, 268,* * 273 mwrite . . . . . . . . . . . . . . . . . 340 DEC 21x4x . . . . . . . . . . . 113 nkitb . . . . . . . . . . . . . * *153, 268 DEC EtherWORKS . . . 114 nn . . . . . . . . . . . . . . * *. . . . . . . 300 N Digital . . . . . . . . . . . . . . . 113 NNTP-Server . . . . . . . . . . * *. 176 Nameserver . . . 176, 247-249 Digital DEPCA . . . . . . . 114 Notification scheme ncpfs . . . . . . . . . . . . . . . . . . . 192 E21xx . . . . . . . . . . . . . . . . 113 SuSEFax . . . . . . . . . . .* * . . 304 NCR 5380 . . . . . . . . . . . . . . 108 EtherBlaster . . . . . . . . . . 117 Novell . . . . . . . . . . . . .* * 192, 318 NCR 53c406a . . . . . . . . . . . 108 EtherTeam 16i/32 . . . . . 116 Novell NE1000/2000 . . . . 117 NCR 53C810 . . . . . . . . . . . 102 EtherWORKS 3 . . . . . . . 114 Novell Netware . . . . . . . . * *. 192 ncurses . . . . . . . . . . . . . . . 168 Fujitsu Novell-Server-Emulation . 179 NE1000/2000 . . . . . . . . . . . 117 FMV-181/182/183/184 . . Nullmodem . . . . . . . . . . . * *. . 272 NE2000 . . . . . . . . . . . . . . . . 111 115 NetBEUI . . . . . . . . . . . . . . . 317 HP 10/100 VG-AnyLAN . . . O NetBIOS . . . . . . . . . . . . . . . 317 116 Octal . . . . . . . . . . . . . * *. . . . . . 196 NetBSD . . . . . . . . . . . . . . . . 201 HP 27245 . . . . . . . . . . . . 115 Okano CD-ROM . . . . . . . . 196 netgroups . . . . . . . . . . . . . . . 254 HP 27247B . . . . . . . . . . . 115 olvwm . . . . . . . . . . . . .* * . . . . 224 netkitb . . . . . . . . . . . . . . . 273 HP 27252A . . . . . . . . . . . 115 Olvwm . . . . . . . . . . . . . * *. . . . 223 Netmask . . . . . . 247, 248, 251 HP 27xxx . . . . . . . . . . . . 115 Online manual . . . . . . . . . * *. 327 netscape . . . . . . . . . . . . . . . . . 12 HP PCLAN . . . . . . . . . . . 115 OnTrack-Manager . . . . . . . 140 Netscape . . . . . . . . . . . . . 60, 61 HP PCLAN+ . . . . . . . . . 115 OpenLook . . . . . . . . . . . .* * . . 224 netstat . . . . . . . . . . . . . . . . . . 166 IBM Token ring . . . . . . . 116 Operating system . . . . . . . .* * 391 NetWare . . . . . . . . . . . . . . . 317 ICL EtherTeam . . . . . . . 116 Optics Storage . . . . . . . . .* * . 197 NetWare 4.1 . . . . . . . . . . . . 322 Intel EtherExpress 16 . . 114 Optics Storage CD-ROM 108, Network . . . . . . . . . . . 245, 391 Intel EtherExpress Pro . 115 117, 197 426 ___________________________________________________________________________________________________________________________I* *ndex__________________ Orchid CD-ROM . . . . . . . . 196 tcl . . . . . . . . . . . . . . . . . . 183 relative . . . . . . . . . . * *. . . . . 391 OS/2 . . . . . . . . . . . . . . 301, 391 tiff . . . . . . . . . . . . . . . . . 305 $PATH . . . 166, 227, 233, 388 Assigning Linux partitions . tk . . . . . . . . . . . . . . . . . . . 183 Patrick Volkerding . . . . . . .* * . . 6 23 trn-spl.tgz . . . . . . . . 300 PCMCIA . . . . . . . . . . . . .* * . . . 99 Bootmanager . . . . . . . . . 123 trn.tgz . . . . . . . . . . . . . 300 pcmcia__s . . . . . . . . . . .* * . . . 187 OSF Motif . . . . . . . . . . . . . . . . 1 vice . . . . . . . . . . . . 359, 360 PentiumPro . . . . . . . . . . .* * . . 204 wget . . . . . . . . . . . . . . . . . . 96 perl . . . . . . . . . . . . . .* * . . . . . . 178 P xf86 . . . . . . . . . . . . . . . . . 183 Permissions . . . 328, 330, 332 Pachkages xfsetup . . . . . . . . . . . . . 208 Samba . . . . . . . . . . . * *. . . . 319 search . . . . . . . . . . . . . . . . . 81 ypclient . . . . . . . . . . . . 253 PET . . . . . . . . . . . . . .* * . . . . . 361 package ypserver . . . . . . . . . . . . 254 Philips CM206 . 109, 118, 197 3dpixm . . . . . . . . . . . . . . 241 Package Philips CM206 CD-ROM 197 3dpixms . . . . . . . . . . . . . 241 de-installation . . . . . . . . . . 93 phone . . . . . . . . . . . . . * *. . . . . 280 aps . . . . . . . . . 147, 148, 153 installation . . . . . . . . . . . . . 93 Pipe . . . . . . . . . . . . . * *. . . . . . 391 bind . . . . . . . . . . . . . . . . . 295 Package description . . . . . . 79 PLIP . . . . . . . . . . . . . .* * . . . . . 118 books . . . . . . . . . . . . . . . . . . 5 package format . . . . . . . . . . . 93 PlugAndPlay . . . . . . . . . .* * . 155 cdesim . . . . . . . . . . . . . . 237 Package information . . . . . . 79 pnpdump . . . . . . . . . . . . * *. . . 155 cnews . . . . . . . . . . . . . . . . 298 Package installation . . . . . . 79 Pointer . . . . . . . . . . . . * *. . . . . 218 curses . . . . . . . . . . . vi, 168 package manager . . . . . . . . . 93 Pointer-Section . . . . . . . .* * . 218 diald . . . . . . . . . . . . . . . . 268 package replace . . . . . . . . . . 80 port . . . . . . . . . . . . . .* * . . . . . . 280 faxprint . . . . . . . . . . . . 309 package update . . . . . . . . . . . 80 Portmapper . . . . . . . . . . .* * . . 177 firewall . . . . . . . 256, 257 Packages ports . . . . . . . . . . . . . * *. . . . . . 280 fvwm . . . . . . . . . . . . . . . . . 227 Check dependencies . . . . 80 Ports fvwm1 . . . . . . . . . . . . . . . . 227 Configurations . . . . . . . . . 78 parallel . . . . . . . . . . * *. . . . . 143 gra . . . . . . . . . . . . . . . . . . 157 Delete . . . . . . . . . . . . . . . . . 80 PostScript template gs x____11 . . . . . . . . . . . . . . . 305 Index . . . . . . . . . . . . . . . . . 81 SuSEFax . . . . . . . . . . .* * . . 308 gsview . . . . . . . . . . . . . . . . . 5 install . . . . . . . . . . . . . . . . . 81 Powermanagement . . . . . . 203 gv . . . . . . . . . . . . . . . . . 5, 305 Selection . . . . . . . . . . . . . . 78 ppp . . . . . . . . . . . . . . * *. . . . . . 268 hylafax . . . . . . . . . 301, 309 paket.tgz . . . . . . . . . . . . . . . . 64 PPP . . . . . . . . . . . . . . * *. . . . . . 268 i4l . . . . . . . . . . . . . 195, 286 Panasonic CD-ROM 111, 119, pppd . . 93, 159, 268, 270, 271 i4ldoc . . . . . . . . . . 286, 294 197 ppplogin . . . . . . . . . . . * *. . . . 271 inetcfg . . . . . . . . . . . . . 294 Paper size Primary partitions . . . . . . * *. . 75 isdn4linux . . . . . . . . . . 285 SuSEFax . . . . . . . . . . . . . 305 Print manager(lpd) . . . . . . * *145 kernmod . . . . . . . . . . . . . 286 Parallelport-CD-ROM . . . 108, Print system see Spool system latex-cover . . . . . . . . 309 117 printcap . . . . . . . . . . . .* * . . . . 146 linux . . . . . . . . . . . . . . . . . . 5 Partition printcap(5) . . . . . . . . . .* * . . . 144 lprold . . . . . . . . . . . . . . 144 formatting . . . . . . . . . . . . . 77 Printer . . . . . . . . . . . . * *. . 83, 202 lx s____use . . . . . . . . . . . 5, 285 Swap . . . . . . . . . . . . . . . . . . 61 Check list . . . . . . . . .* * . . . 153 mgetty . . . . . 271, 301, 315 Partitioning Daemon . . . . . . . . . . . * *. . . 145 mtools . 339, 340, 346, 353 Experts . . . . . . . . . . . . . . . . 61 supported . . . . . . . . . .* * . . . 150 ncurses . . . . . . . . . . . . . 168 Novices . . . . . . . . . . . . . . . 60 $PRINTER . . . . . . . . . . . * *. 144 netkitb . . . . . . . . . . . . . 273 Partitions Printer drivers . . . . . . . . * *. . . . . . . nkita . . . . . . 153, 268, 273 Designing . . . . . . . . . . . . . 35 seePrinter filter146 nkitb . . . . . . . . . . . 153, 268 Distributing . . . . . . . . . . . . 58 Printer filter . . . . . . . . 1* *46, 147 pcmcia__s . . . . . . . . . . . . 187 extended . . . . . . . . . . . . . . 75 apsfilter . . . . . . . . . * *. . . . . 147 ppp . . . . . . . . . . . . . . . . . . 268 Formatting . . . . . . . . . . . . . 43 Printer interface . . . . . . . * *. . 118 sane . . . . . . . . . . . . . . . . . 157 logical . . . . . . . . . . . . . . . . 75 Printer queue sdb . . . . . . . . . . . 12, 92, 292 primary . . . . . . . . . . . . . . . 75 Terms . . . . . . . . . . . .* * . . . . 144 sendfax . . . . . . . . . . . . . 301 Types . . . . . . . . . . . . . . . . . 58 Printer queues susefax . . . . . . . . . 301, 309 Writing partition table . . 41 Run . . . . . . . . . . . . .* * . . . . . 144 susehelp . . . . . . . . . . . . . . 4 passwd . . . . . . . . . . . . 254, 277 Printers susehilf . . . . . . . 292, 316 Path . . . . . . . . . . . . . . . . . . . . 391 Overview . . . . . . . . . . * *. . . 143 suseppp . . . . . . . . . 268, 271 absolute . . . . . . . . . . . . . . 391 Pro Audio Spectrum 103, 109 * * 427 Index _________________________________________________________________________________________________________________* *_______________________________________@ Pro Audio Spectrum 16 . . 109 Root partition . . . . . . . . . . . 109 In kernel . . . . . . . . . * *. . . . 192 Problems route . . . . . . . . . . . . . . . . . . . 166 LUN . . . . . . . . . . . . .* * . . . . 104 at installation . . . . . . . . . . 53 route.conf . . . . . . . . . . . . 252 NCR 5380 . . . . . . . . . . * *. . 108 Process . . . . . . . . . . . . 385, 392 routed . . . . . . . . . . . . . . . . . . 252 NCR 53c406a . . . . . . . . .* * 108 procmail . . . . . . . . . . . . . . . 297 Routing Seagate ST01/02 . . . . . . 1* *10 Prompt . . . . . . . . . . . . . . . . . 392 route.conf . . . . . . . . . . 252 Streamer . . . . . . . . . .* * . . . 110 Protect I/O address . . . . . . 109 RPC-mount-daemon . . . . . 255 TMC-16x0 . . . . . . . 107, 1* *15 Protected mode . . . . . . . . . . 55 RPC-NFS-daemon . . . . . . 255 TMC-885/950 . . . . . . . . * *107 Protocol . . . . . . . . . . . . . . . . 392 RPC-portmapper . . . 254, 255 Trantor T128/128F/228 111 ps . . . . . . . . . . . . . . . . . . . . . 200rpc.mountd . . . . . . . . . 177, 255 Trantor T130B . . . . . . . .* * 111 ps . . . . . . . . . . . . . . . . . . . . . 336rpc.nfsd . . . . . . . . . . . . 177, 255 scwm . . . . . . . . . . . . . * *. . . . . 224 pstree . . . . . . . . . . . . . . . . 336 rpc.portmap . . . . . . . . . . . . 255 sdb . . . . . . . . . . . . . 1* *2, 92, 292 rpm . . . . . . . . . . . . . . . . . . . . . 93Seagate ST01/02 . . . . . . . * *. 110 Q RPM . . . . . . . . . . . . . . . . . . . . 93 Searchlist . . . . . . . . . . .* * . . . . 176 qmail . . . . . . . . . . . . . . . . . . 295 database . . . . . . . . . . . . . . 178 Secure Shell Daemon . . . . 261 Queueing Agent RPM (rpm ) . . . . . . . . . . . . . . 93 secure shell daemon . . . . . 261 HylaFAX . . . . . . . . 311, 312 rpmorig . . . . . . . . . . . . . . 93 Security Qvwm . . . . . . . . . . . . . . . . . 224 rpmsave . . . . . . . . . . . . . . 93 Firewall . . . . . . . . . . * *. . . . 256 Runlevel . . . . . . . . . . . 170, 392 Selection . . . . . . . . . . . * *. . . . 393 R changing . . . . . . . . . . . . . 171 Send queue RAM SuSEFax . . . . . . . . . . .* * . . 302 Size . . . . . . . . . . . . . . . . . . 110 S Send Queue . . . . . . . . . . * *. . 307 Ramdac . . . . . . . . . . . . 209, 213 S.u.S.E. sendbatches . . . . . . . . . . * *. . . 299 rawip . . . . . . . . . . . . . . 290, 292 Services . . . . . . . . . . . . . . . 15 sendfax . . . . . . . . . . . .* * . . . . 300 rawip-HDLC . . . . . . . . . . . 290 S.u.S.E. . . . . . . . . . . . . . . . . . 15 sendfax . . . . . . . . . . . .* * . . . 301 rc . . . . . . . . . . . . . . . . . . . . . . 171samba . . . . . . . . . . . . . . . . . . 317 rdev . . . . . . . . . . . . . . . . . . . . 128Samba . . . . . . . . . . . . . 200, 317 sendmail . . . . vi, 85, 168, 25* *1, README files . . see manuals Permissions . . . . . . . . . . 319 295-297 reboot . . . . . . . . . . . . . . . . . . 326 sane . . . . . . . . . . . . . . . . . . . 157 sendmail -q . . . . . . . . 177* *, 296 Reboot . . . . . . . . . . . . . . . . . 109 SANE - Scanner Access Now series Receive queue . . . . . . . . . . 307 Easy . . . . . . . . . . . . . . . 156 a . . . . . . . . . . . . . .* * . . 268, 286 SuSEFax . . . . . . . . . . . . . 302 Sanyo CD-ROM . . . 110, 118, a1 . . . . . . . . . . . . .* * . . . . . . . 79 Reducing size of a partition 21 197 ap . . . . . . . . . . . . . * *. . 339, 346 Remove Sanyo CDR-H94A . . . . . . 197 b . . . . . . . . . . . . . * *. . . . . . . 153 Directory . . . . . . . . . . . . . 328 Scanner . . . . . . . . . . . . . . . . 156 base . . . . . . . . . . . . * *. . . . . . 40 LILO . . . . . . . . . . . . . . . . 131 Flat bed . . . . . . . . . . . . . . 157 d . . . . . . . . . . . . . * *. . . . . . . 285 Rescue system . . . . . . . . . . 164 SCO . . . . . . . . . . . . . . . . . . . 201 D . . . . . . . . . . . . . * *. . . . . . . 181 launch . . . . . . . . . . . . . . . 166 SCO-Unix . . . . . . . . . . . . . . 338 doc . 4, 5, 12, 92, 286, 292, Use . . . . . . . . . . . . . . . . . . 166 Screen . . . . . . . . . . . . . . . . . 218 316 Reset . . . . . . . . . . . . . . 109, 392 Screen resolution . . . . . . . . 220 gimp . . . . . . . . . . . . * *. . . . . 157 resolv.conf . . . . . . . . . . . . . 250 Screen-Section . . . . . . . . . . 218 gra . . . . . . . . . . . . * *. . . . . . . . 5 Restart Function . . . . . . . . 230 SCSI n . . 253, 254, 257, 268, 271, RFC1861 . . . . . . . . . . . . . . . 310 Adaptec 286, 309, 315 RFC959 . . . . . . . . . . . . . . . . 310 AHA-152x/151x/1505 . . . n1 . . . . . . . . . . . . . * *. . . . . . 257 Ricoh . . . . . . . . . . . . . . . . . . . 99 112 xwm . . . . . . . . . . . . * *. 227, 237 rlogin . . . . . . . . . . . . . . . . . . 176 AdvanSys . . . . . . . . 103, 113 Series Rlogin . . . . . . . . . . . . . . . . . 392 AHA-152x/151x/1505 . 103 Index . . . . . . . . . . . * *. . . . . . 81 rm . . . . . . . . . . . . . . . . . . . . . 330 AHA-154x . . . . . . . . . . . 103 search . . . . . . . . . . . * *. . . . . . 81 rmail . . . . . . . . . . . . . . . . . . . 177 AHA-274x . . . . . . . . . . . 103 Series selection . . . . . . . * *. . . 78 rmmod . . . . . . . . . . . . . . . . . 182 AHA-284x . . . . . . . . . . . 103 Server . . . . . . . . . . . . .* * . . . . . 393 ROM . . . . . . . . . . . . . . 363, 392 AHA-294x . . . . . . . . . . . 103 server.exe . . . . . . . . . . * *. . . . 123 Root . . . . . . . . . . . . . . . . . . . 392 AM53/79C974 . . . . . . . . 104 ServerFlags . . . . . . . . . . * *. . . 218 Root directory . . . . . . . . . . 392 Future Domain . . . 107, 115 Services . . . . . . . . . . . .* * . . . . . 15 428 ___________________________________________________________________________________________________________________________I* *ndex__________________ setup ii, 25, 28-30, 32, 42, 47, sshd . . . . . . . . . . . . . . . . . . . 261 General . . . . . . . . . . * *. . . . 237 55, 57, 70, 141, 142 Stacker . . . . . . . . . . . . . . . . . 350 susewm . . . . . . . . . . * *. . . . . . 237 Setup . . . . . . . . . . . . . . . . . . . 30 Stallion . . . . . . . . . . . . . . . . 202 Susewm SETUP . . . . . . . . . . . . . . . . . 148 Standard- in/out . . . . . . . . . 393 Usage . . . . . . . . . . . .* * . . . . 239 setup tool . . . . . . . . . . . . . . . . 67 Staroffice . . . . . . . . . . . . . . . . 60 Swap partition . . . . . . . . 6* *1, 74 setup.exe . . . . . . . . . . . 6, 27, 30 Start . . . . . . . . . . . . . . . . . . . 325Switch . . . . . . . . . . . . * *. . . . . 393 Setup.exe . . . . . . . . . . . . 27, 28 Startup-scripts . . . . . . . . . . 251 Symbolic link . . . . . . . . . * *. . 331 seyon . . . . . . . . . . . . . . . . . . 267 startx . . . . . . . . . . . . 14, 93, 210 syncPPP . . . . . . . . . . . 29* *0-292 sh . . . . . . . . . . . . . . . . . . . . . 393Stefan Endrass . . . . . . . . . . . . . 6 system . . . . . . . . . . . . * *. . . . . 280 Share . . . . . . . . . . . . . . . . . . 318 Stella . . . . . . . . . . . . . . . . . . 361 System . . . . . . . . . . . . .* * . . . . 335 Shell . . . . . . . . . . . . . . . . . . . 393 Steps System Properties $SHELL . . . . . . . . . . . . . . . 388 Installation . . . . . . . . . . . . 19 SuSEFax . . . . . . . . . . .* * . . 301 shutdown . . . . . . . . . . 166, 326 STonX . . . . . . . . . . . . . . . . . 352 System administrator . . . . 393 Simple Network Paging Streamer System configuration . 87, 174 Protocol . . . . . . . . . . . . 310 QIC-02 . . . . . . . . . . . . . . . 203 System information . . . . . . 1* *61 Sinclair . . . . . . . . . . . . . . . . 357 QIC-80 . . . . . . . . . . . . . . . 203 System V . . . . . . . . . . . * *92, 168 Slackware . . . . . . . . . . . . . . . 92 SCSI . . . . . . . . . . . . . . . . . 110 System V . . . . . . . . . . . .* * . . . 169 Differences . . . . . . . . . . . 159 Su.S.E. slattach . . . . . . . . . . . . . . . . . 272 Services . . . . . . . . . . . . . . . 15 T SLIP . . . . . . . . . . 267, 272, 275 S.u.S.E. . . . . . . . . . . . . . . . . . 15 T-Online . . . . . . . . . . . .* * . . . . . . 2 SLIP-server . . . . . . . . 273, 275 SunOS . . . . . . . . . . . . . 201, 338 tar . . . . . . . . . . . . . 6* *4, 168, 332 smail . . . . . . . . . . . . . . . . . . . 295 Support Task . . . . . . . . . . . . . * *. . . . . . 393 Smarthost . . . . . . . . . . 176, 296 Configuration settings . . 12 Taylor-UUCP . . . . . . 278, 279 SMB . . . . . . . . . . . . . . 200, 317 Hotline . . . . . . . . . . . . . . . . 15 Taylor-UUCP . . . . . . . . ix, * *278 SMC 9194 . . . . . . . . . . . . . . 118 Installation . . . . . . . . . . . . . . 9 tcl . . . . . . . . . . . . . . * *. . . . . . 183 SMC Ultra . . . . . . . . . . . . . 118 Services . . . . . . . . . . . . . . . 15 tcsh . . . . . . . . . . . . . * *. . 284, 393 SMP . . . . . . . . . . . . . . . . . . . 183 Times . . . . . . . . . . . . . . . . . 15 Telephone book SMTP . . . . . . . . . . . . . 177, 295 suse SuSEFax . . . . . . . . . . .* * . . 307 SNPP . . . . . . . . . . . . . . . . . . 310 Launch rescue system . 166 Teles S0-16.0 . . . . . . . . . * *. . 196 Sony CD-ROM . . . . . . . . . 198 Rescue system . . . . . . . . 164 Teles S0-16.3 . . . . . . . . . * *. . 196 Sony CDU31A . . . . . 110, 119 S.u.S.E. . . . . . . . . . . . . . . . . 164 Teles S0-8 . . . . . . . . . . .* * . . . 196 Sony CDU33A . . . . . 110, 119 SuSEconfig 87, 174, 175, 177, Telix . . . . . . . . . . . . . * *. . . . . . 267 Sony CDU535 CD-ROM 110, 235, 241, 248, 254, 271, telnet . . . . . 166, 176, 275, * *402 119, 198 297, 402 Telnet . . . . . . . . . . . . .* * . . . . . 393 Soundblaster Pro CD-ROM . . SuSEconfig . . . . . . . . . . . . 174 Temporary files 197 susefax . . . . . . . . . . . 301, 309 Delete . . . . . . . . . . . * *. . . . . 178 Soundcard SuSEFax . . . x, 300, 301, 301__, Terminal . . . . . . . . . . . * *. . . . 394 Pro Audio Spectrum . . . 109 305, 306, 309, 310 Terminal programs . . . . . . 2* *67 Soundcards . . . . . . . . . . . . . 204 susefax.images texconfig . . . . . . . . . . . * *. . . . 377 speed . . . . . . . . . . . . . . . . . . 281 SuSEFax . . . . . . . . . . . . . 301 texhash . . . . . . . . . . . . * *. . . . . 377 Spool system . . . . . . . . . . . 143 susefax.phonebook.file Texinfo-files . . . . . see tex* *info apsfilter . . . . . . . . . . . . . . 147 SuSEFax . . . . . . . . . . . . . 301 text files Control . . . . . . . . . . . . . . . 145 susefax.setup.file read . . . . . . . . . . . . * *. . . . . . . . 5 Daemon . . . . . . . . . . . . . . 145 SuSEFax . . . . . . . . . . . . . 301 The XFree86 Project . . . . 207 Filter . . . . . . . . . . . . 146, 147 susefax.setup.path The XFree86 Project, Inc. . . 2 Parts . . . . . . . . . . . . . . . . . 144 SuSEFax . . . . . . . . . . . . . 301 Thomas Esser . . . . . . . . . .* * . 377 Queues . . . . . . . . . . . . . . . 146 susehelp . . . . . . . . . . . . . . . . 4 TI 4000M TravelMate . . . 203 Spooling susehilf . . . . . . . . . 292, 316 tiff . . . . . . . . . . . . . .* * . . . . . 305 Terms . . . . . . . . . . . . . . . . 144 suselinux TIFF Software . . . . . . . . . * *. 305 Spooling mechanism Features . . . . . . . . . . . . . . 159 tiffg3 . . . . . . . . . . . . .* * . . . . . . 305 SuSEFax . . . . . . . . 303, 309 suseppp . . . . . . . . . . . 268, 271 TIFFSoftware . . . . . . . . . .* * . 305 squid . . . . . . . . . . . . . . . . . . . 179 susewm . . viii, 223, 226, 228, time . . . . . . . . . . . . . * *. . . . . . 280 ssh . . . . . . . . . . . . . . . . . . . . . 257 229, 236-241 Timezone . . . . . . . . . . . .* * . . . 175 * * 429 Index _________________________________________________________________________________________________________________* *_______________________________________@ tin . . . . . . . . . . . . . . . . . . . . . 300UUCP ix, 176, 277, 284, 295, Configuration . . . . . . . .* * . 242 tk . . . . . . . . . . . . . . . . . . . . . 183 298 Fvwm2 . . . . . . . . . * *. . . . . 227 tkinfo . . . . . . . . . . . . . . . . . . . 4 Start . . . . . . . . . . . * *. . . . . . 228 tkInfo . . . . . . . . . . . . . . . . . . . . 4V Tasks . . . . . . . . . . . * *. . . . . 225 TMC-16x0 . . . . . . . . . 107, 115 Vadem . . . . . . . . . . . . . . . . . . 99 $WINDOWMANAGER . . . . TMC-885/950 . . . . . . . . . . 107 Variable 226-228 Token ring . . . . . . . . . . . . . . 116 $DISPLAY . . . . . . 392, 394 Windows . . . . . . . . 2, 309,* * 310 top . . . . . . . . . . . . . . . . . . . . 336 $GS_RESOL . . . . . . . . . 153 SMB . . . . . . . . . . . . .* * . . . . 317 TOS . . . . . . . . . . . . . . . 352, 353 $HOME . . . 227, 235, 243, Windows . . . . . . . . .* * . . . . . 317 tosdump.prg . . . . . . . . . . . . 353 388, 389 Windows explorer . . . . . . . .* * 56 Toshiba . . . . . . . . . . . . . . . . . . 99 $KDEDIR . . . . . . . . . . . 233 Windows NT . . . . . . . 301, 317 Touchpad . . . . . . . . . . . . . . . 202 $MANPATH . . . . . . . . . 388 Bootmanager . . . . . . . . * *. 123 Trackball . . . . . . . . . . . . . . . 202 $PATH . 166, 227, 233, 388 Windows 95 . . . . . . . . . . * *. . . . 2 Trackpoint . . . . . . . . . . . . . . 202 $PRINTER . . . . . . . . . . 144 Assigning Linux partitions . Transmission Subscriber $SHELL . . . . . . . . . . . . . 388 22, 23 Identification . . . . . . . 307 $USER . . . . . . . . . . . . . . 388 Bootmenu . . . . . . . . . .* * . . . 46 transrom . . . . . . . . . . . . . . . 354 $WINDOWMANAGER . . DOS mode . . . . . . . . . .* * . . 27 Trantor T128/128F/228 . . 111 226-228 Windows NT . . . . . . . . . 2, * *391 Trantor T130B . . . . . . . . . . 111 vbox . . . . . . . . . . . . . . . . . . . 196 WinFlex . . . . . . . . . . . . * *. . . . 310 Travan . . . . . . . . . . . . . . . . . 203 VCS 2600 . . . . . . . . . . . . . . 361 WinFlex . . . . . . . . . .* * . . . . . 310 trn . . . . . . . . . . . . . . . . . . . . . 300Vertical frequency . . 210, 212 WINS . . . . . . . . . . . . . .* * . . . . 318 trn-spl.tgz . . . . . . . . . . 300 VESA . . . . . . . . . . . . . . . . . . 221 Wm2 . . . . . . . . . . . . . .* * . . . . 224 trn.tgz . . . . . . . . . . . . . . . 300 VG-AnyLAN . . . . . . . . . . . 116 Wrapper TSI . . . . . . . . . . . . . . . . 307, 316 Vgb . . . . . . . . . . . . . . . . . . . . 362 SuSEFax . . . . . . . . . . .* * . . 301 Tulip . . . . . . . . . . . . . . . . . . . 113 VGB . . . . . . . . . . . . . . . . . . . 362 WWW . . . . . . . . . . . . . . * *. . . 275 tunelp . . . . . . . . . . . . . . . . . . 143 vi . . . . . . . . 166, 367-369, 387 Twm . . . . . . . . . . . . . . . . . . . 224 VIC-20 . . . . . . . . . . . . . . . . . 361 X vice . . . . . . . . . . . . . . . . . . . . 359X . . . . . . . . . . . . . . . * *. . . . . . . 210 U vice . . . . . . . . . . . . . . 359, 360 X -probeonly . . . . . . . 215, * *216 UAE . . . . . . . . . . . . . . . . . . . 354 Video Console System . . . 361 X Consortium, Inc. . . . . 2, 207 ugidd . . . . . . . . . . . . . . . . . . 256 virtual consoles . . . . . . . . . 326 X Window System 2, 207, see Ultrastor . . . . . . . . . . . . . . . 102 virtual screen . . . . . . . . . . . 220 X11, 217, 394 umount . . . . . . . . . . . . . . . . . 339 virus . . . . . . . . . . . . . . . . . . . 124 X-Server . . . . . . . . . . . * *. . . . 394 UMSDOS . . . . . . . . . . . . . . . 76 VLSI . . . . . . . . . . . . . . . . . . . . 99 X.75 . . . . . . . . . . . . . .* * . . . . . 290 Unable to mount root . . . . . 54 vmlinuz . . . . . . . . . . . . . . . . 399 X/Open Company Limited . 2 Unix . . . . . . . . . . . . . . . . . 1, 325 X11 . . . . . . . . . . . . . . * *. . . . . . 207 UNIX . 2, 301, 309, 385, 389, W .Xresources . . . . . . . . .* * . . 243 391, 394 w . . . . . . . . . . . . . . . . . . . . . . 335 Configuration . . . . 208, 2* *10 Update . . . . . . . . . . . . . . . . . . 91 Wabi . . . . . . . . . . . . . . . . . . . . 60 Keyboard . . . . . . . . . 209,* * 211 Configuration . . . . . . . . . . 92 WAN . . . . . . . . . . . . . . 267, 394 Mice . . . . . . . . . . . . .* * 208, 211 Usage of printers . . . . . . . . 143 Watchdog . . . . . . . . . . . . . . 203 Monitors . . . . . . . . . . 21* *0, 212 Usenet . . . . . . . . . . . . . . . . . 297 WD80x3 . . . . . . . . . . . . . . . 119 X-server . . . . . . . . . . . * *. . . . 213 User Wearnes CD-ROM . . . . . . 196 X-Server . . . . . . . . . . .* * . . . 209 add . . . . . . . . . . . . . . . . . . 327 Western Digital WD80x3 119 Defaults . . . . . . . . . . * *. . . . 242 delete . . . . . . . . . . . . . . . . 327 WfW . . . . . . . . . . . . . . . . . . 337 Graphic cards . . . . . . . .* * . 213 management . . . . . . . . . . . 86 wget . . . . . . . . . . . . . . . . . . . . 96 Graphics cards . . . . . . . * *. 209 $USER . . . . . . . . . . . . . . . . 388 Wide Area Network see WAN Optimization . . . . . . . .* * . 217 useradd . . . . 86, 271, 277, 327 Widget . . . . . . . . . . . . . . . . . 242 User settings . . . . . . . .* * . . 243 userdel . . . . . . . . . . . . . . . . . . 86 Wildcards . . . . . . . . . . 329, 394 X11R1 . . . . . . . . . . . . . * *. . . . 207 Users Windows NT . . . . . . . . . . . 337 X11R6.3 . . . . . . . . . . . .* * . . . 208 adding . . . . . . . . . . . . . . . 325 Window . . . . . . . . . . . 242, 394 xarchie . . . . . . . . . . . .* * 242, 243 USRobotics . . . . . . . . . . . . . 314 Window manager . . . . . . . 394 xdm . . . . . . . . . . . . 14, * *178, 234 uucico . . . . . . . . . . . . . . . . . 284 Windowmanager . . . . . . . . 223 xdos . . . . . . . . . . . . . * *. . . . . . 351 430 ___________________________________________________________________________________________________________________________I* *ndex__________________ xdvi . . . . . . . . . . . . . . . 325, 377 xv . . . . . . . . . . . . . . . . . . . . . 229 Network . . . . . . . . . . .* * . . . . 85 Xenix . . . . . . . . . . . . . 201, 338 xzx . . . . . . . . . . . . . . . . . . . . 357 package replace . . . . . . .* * . 80 xf86 . . . . . . . . . . . . . . . . . . . 183 Package selection . . . . . * *. 78 xf86config . viii, 14, 208, 210, Y package update . . . . . . . * *. . 80 215, 217, 221 yast partitioning . . . . . . . . * *. . . . 73 XF86Config . . . . . . . . . . . . 210 ISDN . . . . . . . . . . . . . . . . 286 fstab . . . . . . . . . .* * . . . . . . . 77 Clocks . . . . . . . . . . . . . . . 220 YaST . . . . . . . . . . . . . . . . . . . . 67 Reading fstab . . . . . . . .* * . . 77 Depth . . . . . . . . . . . . . . . . 219 LILO . . . . . . . . . . . . . . . . . 84 Saving your configuration . . Device . . . . . . . . . . . . . . . 219 Administration . . . . . . . . . 82 78 Device-Section . . . . . . . . 220 Backup . . . . . . . . . . . . . . . . 87 Series selection . . . . . .* * . . 78 Driver . . . . . . . . . . . . . . . . 218 Boot configuration . . . . . 84 settings . . . . . . . . . . * *. . . . . . 72 modeline . . . . . . . . . . . . . 220 Boot kernel . . . . . . . . . . . . 84 Size of installation . . . . * *. . 77 Modes . . . . . . . . . . . . . . . 220 CD-ROM . . . . . . . . . . . . . . 82 User management . . . . . . * *86 Monitor . . . . . . . . . . . . . . 219 Check package What if . . . . . . . . . . .* * . . . . . 80 Monitor-Section . . . . . . 220 dependencies . . . . . . . . 80 YP Screen-Section . . . . . . . . 218 Configuration . . . . . . . . . . 78 Server . . . . . . . . . . . * *. . . . . 177 Subsection Configuration file . . . . . . . 87 yp.conf . . . . . . . . . . . . * *. . . . . 254 Display . . . . . . . . . . . . . . . 219 Configurations . . . . . . . . . 78 ypbind . . . . . . . . . . . . * *. . . . . 254 Virtual . . . . . . . . . . . . . . . 220 Configure printer . . . . . . . 83 ypclient . . . . . . . . . . . .* * . . 253 XF86Setup . vii, 14, 207, 208, Delete packages . . . . . . . . 80 ypserver . . . . . . . . . . . .* * . . . . 254 210 file systems . . . . . . . . . . . . 75 ypserver . . . . . . . . . . . .* * . . 254 xfree . . . . . . . . . . . . . . . . . . . . . 3 Formatting . . . . . . . . . . . . . 77 XFree86TM . . . . . . . . . . . . . 208 Index of all series and xfsetup . . . . . . . . . . . . . . . 208 packages . . . . . . . . . . . . 81 Z xinfo . . . . . . . . . . . . . . . . . . . . . 4 Inodes density . . . . . . . . . 76 Z80 . . . . . . . . . . . . . . * *. . . . . . 356 xli . . . . . . . . . . . . . . . . . . . . . 229 install packages . . . . . . . . 81 Zilog . . . . . . . . . . . . . * *. . . . . . 356 xosview . . . . . . . . . . . 191, 236 Installation . . . . . . . . . . . . 80 zImage . . . . . . . . . . . . .* * . . . . . 30 xpmroot . . . . . . . . . . . . . . . . 229 Installation media . . . . . . 67 ZIP drive xrpm . . . . . . . . . . . . . . . . . . . . 96 Integrating hardware . . . . 82 Parallel . . . . . . . . . . * *. . . . . 116 xscanimage . . . . . . . . 157, 158 main menu . . . . . . . . . . . . 72 ZIP Drive xsetroot . . . . . . . . . . . . . . . . 231 Modem . . . . . . . . . . . . . . . . 82 Parallel . . . . . . . . . . * *. . . . . 107 XT hard disk controller . . 111 Mountpoint . . . . . . . . . . . . 76 zless . . . . . . . . . . . . . * *. . . . . . 178 xterm . . . . . . . . . . . . . . 285, 292 Mouse . . . . . . . . . . . . . . . . 82 ZX81 . . . . . . . . . . . . . * *. . . . . 356 * * 431