Re: apt-question
Bradley wrote: I have run into a problem doing apt-get updates. woody/main Packages Sub-process gzip returned an error code (1) I saw the same thing last night, using unstable. After that, I just backed slowly away from they keyboard, with my hands in the air. It may just be a problem with the files on the server. I hope so. - Brian
Weird keymap problems in NEdit
Hi When I use NEdit these days, most control characters don't do what they are supposed to. For example, if I type Ctrl-S to save, I just get what I assume is a Ctrl-S printed in the document, as br. Oddly copy and paste (Ctrl-C and Ctrl-V) work fine. I assumed at first that I had toggled some NEdit option somewhere, but I will be dd if I can find it. Can anybody help? - Brian
Freshly compiled kernel too big?
Hi I just compiled a 2.2.17 kernel. I've got plenty of RAM and disk space, so I generally chose to compile everything in rather than add modules. However, when I did make-kpkg kernel-image it complained about the image being too big. Sure it's big, but not over 5MB. What's going on, and can I work around it? - Brian
Re: Freshly compiled kernel too big?
Hi bzImage? Thanks for the suggestion. I thought that was the default, but I'll check when I get home tonight. - Brian
Re: xserver-xfree86 4.0.1-9
Charles Setting up xserver-xfree86 (4.0.1-9) ... /var/lib/dpkg/info/xserver-xfree86.config: [: -eq: unary operator expected I saw this too, but it did not end up being a problem. B
Re: How to purge and reinstall XFree86 [woody] ? SOLVED
Hi folks This was solved by Marc Wilson's answer to someone else's problem...they asked about problems that were fairly similar. Both task-x-window-system and using dexter instead of xf86cfg were useful bits of advice. - Brian Marc's respons was: You don't have all of Xf4 installed. Make sure you've done task-x-window-system... some packages have changed names and the dependencies aren't yet all they could be. Then use dexter to configure it. I wrote: I've gotten myself into a real dill of a pickle. I run woody, but around Thanksgiving I hadn't upgraded for weeks. I did apt-get upgrade which screwed up KDE and X. After apt-get dist-upgrade, uninstalling and reinstalling KDE (from 1.x to 2.x), and making a symlink from the X binary to xserver-3dlabs, I got things working. Or so I thought. Last night, I did another apt-get upgrade and X would no longer start. I tried using apt-get remove on xserver-common, and whatever other xfree86 related packages I could find. I made sure all the xfree86 version 3.x stuff was gone. Then I did apt-get install for the various version 4 packages. Finally, I did xf86cfg, which is supposed to try to autodetect my hardware, then give me configuration options. It did give me an X screen with a cursor, but it hung after that. I rebooted, and now it just seems like it keeps trying to start the X server. It shows the VGA text startup screen, blinks a minute, then goes back to that screen. I can't even log in on the console!! I'm going to bring another machine home tonight so I can ssh in to try to fix this, but can anybody give me pointers as to what I need to do? I'm tempted to try the nuclear option -- back up the homedirs, dpkg --get-selections, wipe the disk, and start fresh. But that's inelegant, and lots of work. i would really appreciate other ideas. Thanks, Brian --To UNSUBSCRIBE, email to [EMAIL PROTECTED] a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: xserver-xfree86 4.0.1-9
Setting up xserver-xfree86 (4.0.1-9) ... /var/lib/dpkg/info/xserver-xfree86.config: [: -eq: unary operator expected I've been getting that too, but without the hang. And my X Window system is completely screwed up, which I suspect may be related, so please let me know if you get the solution! Best, Brian
How to purge and reinstall XFree86 [woody] ?
Ouch I've gotten myself into a real dill of a pickle. I run woody, but around Thanksgiving I hadn't upgraded for weeks. I did apt-get upgrade which screwed up KDE and X. After apt-get dist-upgrade, uninstalling and reinstalling KDE (from 1.x to 2.x), and making a symlink from the X binary to xserver-3dlabs, I got things working. Or so I thought. Last night, I did another apt-get upgrade and X would no longer start. I tried using apt-get remove on xserver-common, and whatever other xfree86 related packages I could find. I made sure all the xfree86 version 3.x stuff was gone. Then I did apt-get install for the various version 4 packages. Finally, I did xf86cfg, which is supposed to try to autodetect my hardware, then give me configuration options. It did give me an X screen with a cursor, but it hung after that. I rebooted, and now it just seems like it keeps trying to start the X server. It shows the VGA text startup screen, blinks a minute, then goes back to that screen. I can't even log in on the console!! I'm going to bring another machine home tonight so I can ssh in to try to fix this, but can anybody give me pointers as to what I need to do? I'm tempted to try the nuclear option -- back up the homedirs, dpkg --get-selections, wipe the disk, and start fresh. But that's inelegant, and lots of work. i would really appreciate other ideas. Thanks, Brian
Re: How to convert multiple pages html doc to other formats?
but the info command - I much prefer man pages to be quite honest. Amen, brother. Whose bright idea was it to make everybody learn some mysterious set of keystrokes just to read documentation? If I'm enough of a newbie to need the man pages, then maybe, just maybe, I'm enough of a newbie to be flummoxed by the arcane info system as well! Growl. - B
Cannot ssh from woody -- slink
Hi I have a slink-based machine (running the Linux Router Project, actually) that I use as a server, and a workstation I dual-boot with Win NT and Debian woody. I can ssh to the server from (1) itself, using ssh 1.2.26 (2) my workstation in Win NT (using SecureCRT) (3) my wife's Win NT machine, using SecureCRT I can also ssh from the server to the workstation. However, using the latest binaries, I cannot ssh to the server from my workstation running woody. The IP is the same in WinNT mode as in Linux, so there's no firewall on the server preventing this. If I run ssh -v myserver, I see (as best I can recall) something like this: Connecting to myserver [192.168.1.1] port 22. Allocated local port 1023. Connection closed. -- Here are some things I have tried: (1) Both ssh-nonfree and openssh packages (2) Removing and regenerating /etc/ssh/ (3) Removing ~/.ssh (4) Attempting to ssh from a newly created user ID (5) Removing known_hosts from the server (6) Running the server's sshd in debug mode (with -d flag) When I ran sshd in debug mode, it died when I tried to connect from the workstation, with a complaint about not forking in debug mode. Fair enough, perhaps. So, I am now completely stumped! The fundamental issue, I suppose, is why can I ssh in from WinNT, but not from Debian? Has anybody else got a theory as to why this might be happening? Help! Yours, Brian
Woody: missing parse-xf86config? /etc/init.d.xdm wants it!
Hi Something blew away a program called parse-xf86config this weekend. I suspect it was when I did my apt-get upgrade. This is a problem because /etc/init.d/xdm won't run properly without it. Any clues as to how I can get it back? - Brian
Re: Woody: missing parse-xf86config? /etc/init.d.xdm wants it!
Something blew away a program called parse-xf86config this weekend. I suspect it was when I did my apt-get upgrade. This is a problem because /etc/init.d/xdm won't run properly without it. According to the docs, you don't need that any more. Just remove the line 'check-local-xserver' from your /etc/X11/xdm/xdm.options. The short explanation is somewhere in the /usr/doc/xserver-common/changelog.Debian.gz file. Reasonable, but that doesn't really help /etc/init.d/xdm. I've changed /etc/init.d/xdm by hand for now, but I doubt this is what the maintainers meant for *everyone* to do. Maybe I somehow ended up with a broken xserver-common? That would explain the bad init script, I guess. - Brian
Re: cable modem and LAN
Hi Matt The following link: http://linuxrouter.sourceforge.net/documentation/LRP-2.9.4/ has documentation on doing this, for the Debian-based Linux Router Project. For various reasons, you might want to set a machine up with that instead. - Brian
Current sources.list entries for KDE?
Hi All my usual suspects for sources.list entries capable of finding KDE .debs are stale. Even the reliable rkrusty!! Has anybody got anything current? - Brian
Re: firewall
Why don't you take a look at http://www.linuxrouter.org/? I think it should fit your purpose, and it's based on Debian :) I second the recommendation, but you should know the main website lacks the documentation. See http://linuxrouter.sourceforge.net and http://lrp.c0wz.com for the good docs. - Brian
Re: Has anyone gotten TechExplorer / MathML for Netscape working?
Hi Gerhard Well, I'm not really qualified to test the plugin. For the standalone, I put the recommended shell variables in ~/.profile, but that did no good. If I don't hear from anybody soon, I may appeal to IBM. Basically, the way the install is supposed to work is you run their little program, and it extracts everything and copies it to a new subdirectory in a spot you choose. That part went OK. But when I ran the app, it just died. I tried running in gdb, but got nothing useful out of that either. 1 - Brian not jet, just downloaded it... didn't expect any problems. please tell me (us) more if you find the problem. gerhard
Re: ssh and stuff
Andrew You can probably get the behavior you want by using the RSA login and ssh-agent. That is, if you use ssh-keygen to generate .identity and .identity.pub files locally, then on the remote machine append the contents of .identity.pub to a file named ~myuser/.ssh/authorized_keys you will enable RSA logins. Then, in your local ~/.ssh/config file append the lines: Host otherguy User myuser Finally, in any given shell session, run ssh-agent bash to start a bash shell with an agent (any shell will do). Finally, type ssh-add as your first command. Logins will now be invisible and automatic, like rsh, so scripts ought to work. There is an X version of the agent, too, but I've never gotten it working for me. Yours, Brian
Solved: Boot off non-BIOS SCSI drive - can I use LILO or SysLinux?
Hi Well, Her Majesty's empire really came through for me! Thank you to David and Frank for your help. Frank was right about booting off the non-BIOS SCSI disk...for some reason, it does work (and, as he hinted, the magic incantation to LILO seems to be linear). I did not end up needing to try loadlin, though that looks like a fine solution. Incidentally, because I share this box with others (hence my hesitation to touch the IDE drive), I wanted something a little more user-friendly in the boot loader. I had an old copy of System Commander around, so I tried installing it. It would not see the second drive even though LILO did, but it has an option to boot from a MBR saved to a file. So I used dd to save the LILO MBR to a file on the Dos/Win partition. System Commander was then able to boot Win98 directly, and to chain-boot LILO. Thanks, guys, Brian
Re: ssh pam
However, AFAIK, you'll have to hand out some $$ if you want a Windoze (95/98/NT) ssh client. TeraTerm SSH is free, and works prety well. B
Boot off non-BIOS SCSI drive - can I use LILO or SysLinux?
Hi I have a system with Windows 95 on an IDE drive, and Debian potato on a SCSI drive, hung off of an old DPT PM2021/9x controller. This controller does not appear to be mapping the SCSI disk to D: like I had hoped and expected. Presumably, I cannot use LILO to dual-boot the machine (after all, how would the machine let LILO see the partition with the kernel?). Up to now, I have been booting off of a floppy, and that is torture. But it seems SysLinux might be able to do the job. How might I use it? Can I call it from Win95, or only after an MS-DOS boot? I took a look at /usr/share/doc/syslinux/syslinux.doc and it seems that booting from DOS mode would not be a problem, so long as I keep a copy of the kernel around, but I'm not so sure of this business of messing with the MBR. Where would that get me? Brian
Re: xxx.tar.gz on Windows.
Hi EasyZip is a very nice freeware clone, and I'm pretty sure it does .tar.gz. - B Marcus wrote: Unfortunately I don't know of any freeware for windows that will read tar.gz. HJ-Zip is freeware. I'll ask the developer if he's adding xxx.tar.gz support sometime.
Re: Make a network firewall with real IPs
Hi Paulo You might want to try the Debian-based Linux Router Project, at http://www.linuxrouter.org/ The whole thing runs off a write-protected floppy, so you can't be hacked as badly as with a HD. The docs on the official site are not too good, you can find good ones at: http://lrp.c0wz.com To use real IP's, turn off the masquerading in /etc/network.conf. - Brian You wrote: Hi all, I have a computer (486 SX 33 - 16 Mb - 2 ethernet cards) that I want to put to be a firewall in my network of 5 or 6 PCs that have real IPs. What I can do to all packages pass by the 486 and still have real IPs?
Re: Linux as a router
You wrote: I have an Intel/486 and want to set it up as a router. You might want to try the Debian-based Linux Router Project, at http://www.linuxrouter.org/ The whole thing runs off a write-protected floppy, so you can't be hacked as badly as with a HD. The docs on the official site are not too good, but here's a good one: http://www.geocities.com/Athens/Ithaca/9660/lrphowto.html - Brian
Re: How to savely copy a disk?
dd --- diskdump --- is what you want. Byte for byte copying. I'm not so sure this would work around bad sectors on the hard drives. - Brian
Solved -- Re: Win NT does 1600x1200, why not X? (FireGL 1000Pro)
Hi Everybody First off, a very big thank you to nate, Brian, and Bryan for all your help. Just 3 more pillars of the Debian community. The Problem With a monitor capabale of achieving (even exceeding) 1600x1200 resolution, a Diamond FireGL 1000Pro with 8MB VRAM would not exceed 1280x1024. This video card is a GLINT Permedia2 chipset, served by the 3DLabs accelerated server. Proper modelines were present for 1600x1200, as in --- # 1600x1200 @ 70 Hz, 87.50 kHz hsync Modeline 1600x1200 1891600 1664 1856 2160 1200 1201 1204 1250 -HSync -VSync --- and the modes were defined in the server --- Subsection Display Depth 24 Modes 1600x1200 1280x1024 1024x768 ViewPort0 0 EndSubsection Subsection Display Depth 8 Modes 640x480 800x600 1024x768 1280x1024 ViewPort0 0 EndSubsection --- but Ctrl-Alt-+ would never obtain the high-res mode. The Solution The key information was bit depth. The X server defaults to depth 8, for which I had no 1600x1200 mode defined. At first, I tried to set it in depths 24 and 32, which still didn't work. The ultimate reason was that my hardware does not support 1600x1200 at these depths. Ultimately, the problem was solved by putting the line --- DefaultColorDepth 16 --- into /etc/X11/XF86Config, and by starting X with --- :0 local /usr/bin/X11/X vt7 -bpp 16 -dpi 100 --- in /etc/X11/kdm/Xservers. A little quality time with the Font Deuglification guide, at http://www.frii.com/~meldroc/Font-Deuglification.html and I am a happy guy. Thanks, Brian
Re: Win NT does 1600x1200, why not X? (FireGL 1000Pro)
OK, using startx X.log I get the following: XFree86 Version 3.3.5 / X Window System (protocol Version 11, revision 0, vendor release 6300) Release Date: August 23 1999 If the server is older than 6-12 months, or if your card is newer than the above date, look for a newer version before reporting problems. (see http://www.XFree86.Org/FAQ) Operating System: Linux 2.2.10 i686 [ELF] Configured drivers: GLINT: accelerated server for 3DLabs GLINT graphics adapters (Patchlevel 0) (using VT number 7) XF86Config: /usr/X11R6/lib/X11/XF86Config (**) stands for supplied, (--) stands for probed/default values (**) XKB: keymap: xfree86(us_microsoft) (overrides other XKB settings) (**) Mouse: type: PS/2, device: /dev/mouse, buttons: 3 (**) GLINT: Graphics device ID: 3DLabs Card (**) GLINT: Monitor ID: ViewSonic PT813 (--) GLINT: Mode 1280x1024 needs hsync freq of 107.16 kHz. Deleted. (**) FontPath set to /usr/X11R6/lib/X11/fonts/misc/,/usr/X11R6/lib/X11/fonts/75dpi/:unscaled,/usr/X11R6/lib/X11/fonts/100dpi/:unscaled,/usr/X11R6/lib/X11/fonts/Type1/,/usr/X11R6/lib/X11/fonts/Speedo/,/usr/X11R6/lib/X11/fonts/75dpi/,/usr/X11R6/lib/X11/fonts/100dpi/ (--) GLINT: found TI chip GLINT Permedia 2 at card #0 func #0 with base 0xe300 (--) GLINT: Framebuffer address at 0xe200 (**) GLINT: videoram : 8192k (--) GLINT: Using builtin RAMDAC of Permedia 2 chip (--) GLINT: Fitted Memory type is : SGRAM (--) GLINT: VGA core is : Enabled (--) GLINT: Maximum allowed dot-clock: 230.000 MHz (**) GLINT: Cannot support interlaced modes, deleting. (**) GLINT: Cannot support interlaced modes, deleting. (**) GLINT: Cannot support interlaced modes, deleting. (**) GLINT: Mode 640x480: mode clock = 45.800 (**) GLINT: Mode 800x600: mode clock = 69.650 (**) GLINT: Mode 1024x768: mode clock = 115.500 (**) GLINT: Mode 1280x1024: mode clock = 157.500 (--) GLINT: Using XAA (XFree86 Acceleration Architecture) (--) GLINT: XAA: Solid filled rectangles (--) GLINT: XAA: Screen-to-screen copy (--) GLINT: XAA: Using 10 128x128 areas for pixmap caching (--) GLINT: XAA: Caching tiles and non-transparent stipples (--) GLINT: XAA: Horizontal and vertical lines and segments System: `/usr/X11R6/lib/X11/xkb/xkbcomp -w 1 -R/usr/X11R6/lib/X11/xkb -xkm -m us_microsoft -em1 The XKEYBOARD keymap compiler (xkbcomp) reports: -emp -eml Errors from xkbcomp are not fatal to the X server keymap/xfree86 compiled/xfree86.xkm' waiting for X server to shut down
Re: Win NT does 1600x1200, why not X? (FireGL 1000Pro)
nate wrote: keys are here (**) GLINT: Cannot support interlaced modes, deleting. (**) GLINT: Cannot support interlaced modes, deleting. (**) GLINT: Cannot support interlaced modes, deleting. Hi Thanks, Nate. I worried about those messages, too, but decided it was not the problem, due to: % fgrep Interlace XF86Config Modeline 1024x76844.9 1024 1048 1208 1264 768 776 784 817 Interlace ModeLine 1152x864651152 1168 1384 1480 864 865 875 985 Interlace Modeline 1280x1024 801280 1296 1512 1568 1024 1025 1037 1165 Interlace So I reckon these are the 3 lines being thrown away. I do wonder, though, if I have a bitdepth problem. I reckoned the server would default to depth 24 since that is the first Display listed, but if it is defaulting to 8, then there's a good reason it would not go so high. I will try another modeline generator, as you suggest -- probably both XF86Setup and http://www.inria.fr/cgi-bin/nph-colas-modelines. - Brian - Brian
Re: Alsa problem - can't locate module sound-slot-0
Anyone know a fix to this and/or how this module gets generated? Whacks with a clue-bat would be appreciated, TIA. This is what eventually worked for me, with a different card: http://www.bright.net/~dlphilp/alsa-pci128.html it's part of a whole collection of ALSA driver docs at http://www.bright.net/~dlphilp/linuxsound/drivers.html - Brian
Win NT does 1600x1200, why not X? (FireGL 1000Pro)
Hi I have a Diamond Fire GL 1000 Pro 8MB AGP, which I've generally been pretty happy with. But I've spent 6 MONTHS (on and off) trying to get the darn thing to give me 1600x1200. I can do it in WinNT, and the card and monitor are both able to take it (even should be able to handle 1800x1440). But it won't work. The most recent thing I tried was to change /etc/X11/XF86Config so that the ONLY lines appearing under Section Screen, Driver accel were like this: ... Subsection Display Depth 8 Modes 1600x1200 ... However, this just caused X not to start at boot, until I restored my old XF86Config from the backup. You might think I have a modelines problem, but XF86Config has the modelines: ... # 1600x1200 @ 60Hz, 75.00 kHz hsync Modeline 1600x1200 162 1600 1664 1856 2160 1200 1201 1204 1250 +HSync +VSync # 1600x1200 @ 70 Hz, 87.50 kHz hsync Modeline 1600x1200 1891600 1664 1856 2160 1200 1201 1204 1250 -HSync -VSync # 1600x1200 @ 75 Hz, 93.75 kHz hsync Modeline 1600x1200 202.5 1600 1664 1856 2160 1200 1201 1204 1250 +HSync +VSync # 1600x1200 @ 85 Hz, 105.77 kHz hsync Modeline 1600x1200 2201600 1616 1808 2080 1200 1204 1207 1244 +HSync +VSync ... which look OK to my (unedumacated) eyes. Hitting Ctrl-Alt-+ never cycles me up to 1600x1200 -- only to 1280x1024. Obviously, I am missing something. Can anyone help? Below are the relevant potions of XF86Config: - Brian Section Monitor Identifier ViewSonic PT813 VendorName ViewSonic ModelName PT813 HorizSync 30-107 VertRefresh 50-160 # 640x400 @ 70 Hz, 31.5 kHz hsync Modeline 640x400 25.175 640 664 760 800 400 409 411 450 # 640x480 @ 60 Hz, 31.5 kHz hsync Modeline 640x480 25.175 640 664 760 800 480 491 493 525 # 800x600 @ 56 Hz, 35.15 kHz hsync ModeLine 800x600 36 800 824 896 1024 600 601 603 625 # 1024x768 @ 87 Hz interlaced, 35.5 kHz hsync Modeline 1024x76844.9 1024 1048 1208 1264 768 776 784 817 Interlace # 640x400 @ 85 Hz, 37.86 kHz hsync Modeline 640x400 31.5 640 672 736 832 400 401 404 445 -HSync +VSync # 640x480 @ 72 Hz, 36.5 kHz hsync Modeline 640x480 31.5 640 680 720 864 480 488 491 521 # 640x480 @ 75 Hz, 37.50 kHz hsync ModeLine 640x48031.5 640 656 720 840 480 481 484 500 -HSync -VSync # 800x600 @ 60 Hz, 37.8 kHz hsync Modeline 800x600 40 800 840 968 1056 600 601 605 628 +hsync +vsync # 640x480 @ 85 Hz, 43.27 kHz hsync Modeline 640x480 36 640 696 752 832 480 481 484 509 -HSync -VSync # 1152x864 @ 89 Hz interlaced, 44 kHz hsync ModeLine 1152x864651152 1168 1384 1480 864 865 875 985 Interlace # 800x600 @ 72 Hz, 48.0 kHz hsync Modeline 800x600 50 800 856 976 1040 600 637 643 666 +hsync +vsync # 1024x768 @ 60 Hz, 48.4 kHz hsync Modeline 1024x768651024 1032 1176 1344 768 771 777 806 -hsync -vsync # 640x480 @ 100 Hz, 53.01 kHz hsync Modeline 640x480 45.8 640 672 768 864 480 488 494 530 -HSync -VSync # 1152x864 @ 60 Hz, 53.5 kHz hsync Modeline 1152x864 89.9 1152 1216 1472 1680 864 868 876 892 -HSync -VSync # 800x600 @ 85 Hz, 55.84 kHz hsync Modeline 800x60060.75 800 864 928 1088 600 616 621 657 -HSync -VSync # 1024x768 @ 70 Hz, 56.5 kHz hsync Modeline 1024x768751024 1048 1184 1328 768 771 777 806 -hsync -vsync # 1280x1024 @ 87 Hz interlaced, 51 kHz hsync Modeline 1280x1024 801280 1296 1512 1568 1024 1025 1037 1165 Interlace # 800x600 @ 100 Hz, 64.02 kHz hsync Modeline 800x60069.65 800 864 928 1088 600 604 610 640 -HSync -VSync # 1024x768 @ 76 Hz, 62.5 kHz hsync Modeline 1024x768851024 1032 1152 1360 768 784 787 823 # 1152x864 @ 70 Hz, 62.4 kHz hsync Modeline 1152x864 921152 1208 1368 1474 864 865 875 895 # 1280x1024 @ 61 Hz, 64.2 kHz hsync Modeline 1280x1024 1101280 1328 1512 1712 1024 1025 1028 1054 # 1024x768 @ 85 Hz, 70.24 kHz hsync Modeline 1024x768 98.9 1024 1056 1216 1408 768 782 788 822 -HSync -VSync # 1152x864 @ 78 Hz, 70.8 kHz hsync Modeline 1152x864 110 1152 1240 1324 1552 864 864 876 908 # 1280x1024 @ 70 Hz, 74.59 kHz hsync Modeline 1280x1024 126.5 1280 1312 1472 1696 1024 1032 1040 1068 -HSync -VSync # 1600x1200 @ 60Hz, 75.00 kHz hsync Modeline 1600x1200 162 1600 1664 1856 2160 1200 1201 1204 1250 +HSync +VSync # 1152x864 @ 84 Hz, 76.0 kHz hsync Modeline 1152x864 1351152 1464 1592 1776 864 864 876 908 # 1280x1024 @ 74 Hz, 78.85 kHz hsync Modeline 1280x1024 1351280 1312 1456 1712 1024 1027 1030 1064 # 1024x768 @ 100Hz, 80.21 kHz hsync Modeline 1024x768 115.5 1024 1056 1248 1440 768 771 781 802 -HSync -VSync # 1280x1024 @ 76 Hz, 81.13 kHz hsync Modeline 1280x1024 1351280 1312 1416 1664 1024 1027 1030 1064 # 1600x1200 @ 70 Hz, 87.50 kHz hsync Modeline 1600x1200 189
Re: Win NT does 1600x1200, why not X? (FireGL 1000Pro)
aphro wrote: What is the full output of X when you start it ? use startx X.log and print the X.log to the list ..What version of X ? Dang. I knew I was leaving something out. Thanks for responding. I've had to leave the machine for the afternoon, but I can say that the X version is the latest from the potato .debs (for 3DLabs Permedia). X starts automatically for me on boot -- I end up in kdm (used to be xdm). How do I get the logfile you mention? Thanks, Brian
Re: window managers
You wrote: debs, how do i prevent twm from automatically being my window manager when i type startx? This is determined by the file: /etc/X11/window-managers. The first one is what gets used. - Brian poe % cat window-managers # /etc/X11/window-managers # # This file contains a list of available window managers. The default # Xsession file will start the first window manager that it can in this # list. See the window-managers(5) and register-window-manager(8) manual # pages for more information. /usr/bin/kde /usr/bin/X11/kde /usr/bin/X11/fvwm2 /usr/bin/X11/fvwm95 /usr/bin/X11/wmaker /usr/bin/X11/olwm /usr/bin/X11/scwm /usr/bin/X11/afterstep /usr/bin/X11/ctwm /usr/bin/X11/wm2
Re: Win NT does 1600x1200, why not X? (FireGL 1000Pro)
Through a telnet session, I followed Brian's suggestions. Basically, I have an xdm.log but no kdm.log. But kdm is what is runnning. Below is what I found. I'll try to get more tonight when I am at console. - Brian mass % psgrep kdm root 269 0.0 0.8 5552 2244 ?S17:13 0:00 /usr/bin/kdm root 277 0.1 1.5 13960 3868 ?S17:13 0:00 /usr/bin/X11/X vt7 -auth /var/state/kdm/A:0-Z265qE mass % cat xdm.log xdm error (pid 281): error 98 binding socket address 177 Fatal server error: Server is already active for display 0 If this server is no longer running, remove /tmp/.X0-lock and start again. When reporting a problem related to a server crash, please send the full server output, not just the last messages xdm error (pid 281): server unexpectedly died xdm error (pid 281): Server for display :0 can't be started, session disabled mass % ls /var/log/kd* zsh: no matches found: /var/log/kd*
Re: how to compile packages optimized for Pentium or Pentium-II?
Ingo wrote: On Thu, Nov 04, 1999 at 05:08:04PM +0100, Robert Varga wrote: How can I recompile the packages so that they be optimized for running on Pentium or Pentium-II or else? According to the docs, this Replaces gcc, cc, and g++ with scripts that build with pentium optimizations, using egcc. By default, after installing this package, the compilers will behave normally. However, if the environment variable DEBIAN_BUILDARCH=pentium is set, they will enter pentium optimized compile mode. Does that mean that gcc normally is NOT Pentium optimized? - Brian
Scrollback default in terminal emulators?
Hi I've been spoiled by the OpenStep UI, and Terminal.app. I would love to set a (large) default scrollback buffer for a decent terminal application -- something like rxvt or kvt. Is my best option here to find and compile the appropriate sources, or does someone know of a secret environment variable/config file? - Brian
Solution: SB PCI 128 sound card and ALSA 0.4.1 on 2.2.x potato
Hi
It was a terrible pain for me to get this card working with my
potato system, so I for the benefit of any future souls clever enough to
search the archives for help, here is (my understanding of) how it worked:
- Compile a new kernel with OSS native sound support. It does not matter if
you compile it with the specific support for the Ensoniq 1370/1371 or not.
For a Debian system, I got the kernel source 2.2.12 package and did
# cd /src/linux
# zcat kernel-source-2.2.12.tar.gz | tar xf -
# ln -s kernel-source-2.2.12 linux
# cd linux
# make menuconfig
{ Set lots of kernel choices, a tedious process. Do not
forget to include sound!}
# make-kpkg clean
# make-kpkg --revision=3:custom.1.0 kernel_image
{ This 3: business is an 'epoch number' that will
later turn out to be incompatible with
ALSA, but I want it for kernel installations }
- Compile the ALSA modules
# cd /src/linux
# make-kpkg --revision=foralsa.1.0 modules_image
{ We cannot use the epoch scheme, so we do a
different revision}
- Install the new kernel and modules
# cd /usr/src
# dpkg -i kernel-image-2.2.blahblahblah
# dpkg -i alsa-modules-2.2.blahblahblah
- Let ALSA configure itself
# alsaconfig
{One mistake to avoid: I was not sure if I needed ens1370
or ens1371 -- even though it autodetected ens1370
I actually thought the latter was what I wanted.
So I chose both. Big problems.}
- Reboot (we're not finished yet)
# reboot
- Get the driver going
# modprobe snd-card-ens1370
# amixer set Master unmute
# amixer set Master 100%
# amixer set CD unmute
# amixer set CD 100%
# amixer set PCM unmute
# amixer set PCM 100%
- Try playing a CD and some sounds. You'll probably want to play with those
volumes a little.
Things I am still trying to figure out:
- Why do I get constant hissing whenever the volume is at an audible
setting or above?
- Is it better to have the Master volume at high settings and the
others low, or vice versa?
- Will I need to do anything for the Joystick port?
Yours,
Brian
Solution: SB PCI 128 sound card and ALSA 0.4.1 on 2.2.x potato (revised)
(revised with a forgotten step)
Hi
It was a terrible pain for me to get this card working with my
potato system, so I for the benefit of any future souls clever enough to
search the archives for help, here is (my understanding of) how it worked:
- Compile a new kernel with OSS native sound support. It does not matter if
you compile it with the specific support for the Ensoniq 1370/1371 or not.
For a Debian system, I got the kernel source 2.2.12 package and did
# cd /src/linux
# zcat kernel-source-2.2.12.tar.gz | tar xf -
# ln -s kernel-source-2.2.12 linux
# cd linux
# make menuconfig
{ Set lots of kernel choices, a tedious process. Do not
forget to include sound!}
# make-kpkg clean
# make-kpkg --revision=3:custom.1.0 kernel_image
{ This 3: business is an 'epoch number' that will
later turn out to be incompatible with
ALSA, but I want it for kernel installations }
# make-kpkg clean
# make-kpkg --revision=foralsa.1.0 modules_image
{ We will not end up installing this one, but ALSA
will want a build with this name }
- Compile the ALSA modules
# cd /src/linux
# make-kpkg --revision=foralsa.1.0 modules_image
{ We cannot use the epoch scheme, so we do a
different revision}
- Install the new kernel and modules
# cd /usr/src
# dpkg -i kernel-image-2.2.whatever_custom.1.0_i386.deb
# dpkg -i alsa-modules-2.2.whatever_0.4.1b-1+foralsa.1.0_i386.deb
- Let ALSA configure itself
# alsaconfig
{One mistake to avoid: I was not sure if I needed ens1370
or ens1371 -- even though it autodetected ens1370
I actually thought the latter was what I wanted.
So I chose both. Big problems.}
- Reboot (we're not finished yet)
# reboot
- Get the driver going
# modprobe snd-card-ens1370
# amixer set Master unmute
# amixer set Master 100%
# amixer set CD unmute
# amixer set CD 100%
# amixer set PCM unmute
# amixer set PCM 100%
- Try playing a CD and some sounds. You'll probably want to play with those
volumes a little.
Things I am still trying to figure out:
- Why do I get constant hissing whenever the volume is at an audible
setting or above?
- Is it better to have the Master volume at high settings and the
others low, or vice versa?
- Will I need to do anything for the Joystick port?
Yours,
Brian
sources.list entry for KDE - how to construct?
Hi I'm not stupid, but I feel that way sometimes; I seem to be unable to get my sources.list right for KDE. Examining the lynx -dump command (see below), we find that the Packages.gz file resides in the same place as all the .deb files. According to my understanding, apt-get looks for the Packages.gz file in the second field (URI) of the sources.list, and the .deb files themselves in the third field (distribution). So I reckon my entry should look like this: deb ftp://ftp.us.kde.org/pub/kde/stable/latest/distribution/deb/potato/i386 / which totally does not work, or like this deb ftp://ftp.us.kde.org/pub/kde/stable/latest/distribution/deb/potato i386/ which seems happy, but then doesn't show the packages as available for dselect. Has anyone worked this out? - Brian lynx -dump ftp://ftp.us.kde.org/pub/kde/stable/latest/distribution/deb/potato/i386/ Current directory is /pub/kde/stable/latest/distribution/deb/potato/i386 [1]Up to higher level directory [2][LINK]-[3][IMAGE] Packages.gz 3 Kb Wed Sep 15 14:36:00 1999 [4][LINK]-[5][IMAGE] kdeadmin_1.1.2-199... 613 Kb Mon Sep 13 18:37:00 199
Re: Help with this simple makefile
Hi How about this? eps_files = fig1.eps fig2.eps all : $(eps_files) %.eps: %.fig fig2dev -L ps $ $@ - Brian You wrote: Hello, I have this simple makefile: eps_files = fig1.eps fig2.eps $(eps_files): %.eps: %.fig fig2dev -L ps $ $@
Backup -- what directories are important?
Hi I've got a CD-RW drive, and a hankering to back up. I'm using a few GB of space with my potato distribution, but most of that is packages that I could easliy fetch again from the web if my HD crashed. Are there any comments on the following list of directories to backup? /home/ /etc/ wherever the dpkg data is stored /usr/local/ One thing I'm wondering: if I do this, how do I restore my list of packages in such a way that dpkg will go out and refetch/install them all? - Brian
XEamcs -- how to assign keys to Cut, Copy, Paste?
Hi How can I assign keys to Cut, Copy, and Paste in XEmacs. Ideally, they would not be lots of keystrokes combined. On a related note, my frien Bill Bumgartner once gave me the included function for my .emacs file...it causes the paste command to replace the selection. How can I make it work here? ;; clear-and-paste ;; ;; Replaces current region with yank-buffer and moves point to end of ;; freshly yanked text. If no selection, simply inserts text. (defun clear-and-paste (p) Act like Text object paste. That is-- paste yank buffer, replacing current selection (if any) and move point to end (interactive *d) (let ((newPoint 0)) (save-excursion (cond ((eval mark-active) (delete-region p (mark (yank) (setq newPoint (point))) (goto-char newPoint))) ;; Replace more like text object (setq delete-selection-mode t) - B
Re: multiple-file search replace
Guilherme Soares Zahn (IEN) wrote:
I was wondering if there is a tool to help me search replace a given
keyword in all files inside a directory (and its subdirectories)... For
instance, imagine that I wanted to change, in ALL makefiles of the Linux
kernel source, the name of the compiler to use...
I like to type the following in zsh:
foreach i ( `echo **/Make*` )
replace '/bin/cc' '/usr/local/bin/cc' $i /tmp/tempfile;
mv /tmp/tempfile $i;
where replace is a perl script I slapped together that looks like below, and
the **/ construct recursively searches subdirectories. I would love to know
if someone has a better version, but this one at least keeps me from having
to remember lots of syntax all the time:
#!/usr/local/bin/perl5
###
### Created by Brian K. Boonstra
###
### Replace a
###
###
$/ = undef;
$find = @ARGV[0];
$replace = @ARGV[1];
if ($find eq ) {
print replace find-string replace-string\n\nTake input fr
om stdin, replace the find-string with\nreplace-string, and output to stdo
ut. The find and\nreplace strings should follow Perl syntax.\n\n;
exit(0);
}
$mystring = STDIN;
$mystring =~ s/$find/$replace/sg;
print $mystring;
Help! Hang on boot, with letters LI
Hi I can no longer boot into Debian! I installed IE5 in my WinNT partition, and after that Lilo would boot WinNT at the prompt, but hang on trying to boot Linux. Then after some fiddling it broke completely, giving me the letters LI (in the same position as I previously saw a whole LILO prompt), and the system hung. I reckoned Lilo was the problem, and installed a different boot loader (PQBoot from Powerquest). Now I can still boot into WinNT, but if I try to boot Linux, I get the same LI hang I had when Lilo was the boot loader. I tried booting off the CD, activating my partitions, and reinstalling LILO, but it win't install. I then tried to create a boot floppy, and I can't do that either. I have a 7.5 GB SCSI system with 4 primary partitions. In order they are: DOS, WinNT, swap, ext2. I was running potato with a custom-compiled kernel that I had set up with dpkg-kernel. - Brian
Install IE5 on WinNT partition, kill LILO
Hi On Sunday, I booted into my Windows NT partition, downloaded Internet Explorer 5.0, and installed it. Later that night, I discovered that LILO would no longer load my Linux partition. I still get the LILO boot prompt, and if I hit Shift and type WinNT for my NT partition, it loads fine. But if I type Linux (or wait) I get the message Loading Linux and everything hangs. I tried reinstalling LILO by using the Debian rescue floppy, but that just dies, complaining about the 1024 cylinder limit, or some such. My disk is an IBM SCSI disk, and has 4 partitions: 2.5 GB NTFS 128 MB Linux Swap 3.0 GB Linux this is the active partition right now 2.0 GB FAT32 Any ideas about how I can get out of this? Will another boot loader help? - Brian P.S. I am not going to blame Microsoft for this right away (though I am suspicious), since I don't know if this config *ever* workedI set up LILO so that it loaded Linux just fine, and assumed it worked with WinNT. But I think this is the first time I've used NT since I last played with LILO.
alsa starts, then disappears; play works, aplay fails
Hi I have a weird problem where ALSA appears to start, but then remains unavailable. When I try to use aplay, it doesn't work. But when I use play, (through sox) it does. Check this out: mass ~ sudo aplay Aplay: version 0.3.2 by Jaroslav Kysela [EMAIL PROTECTED] Error: audio open error: No such file or directory mass ~ amixer The ALSA sound driver was not detected in this system. mass ~ sudo /etc/init.d/alsa start ALSA driver is already running. mass ~ sudo /etc/init.d/alsa restart Shutting down sound driver: done. Starting sound driver: snd-audiopci1370 failed. mass ~ sudo /etc/init.d/alsa start ALSA driver is already running. mass ~ aplay Aplay: version 0.3.2 by Jaroslav Kysela [EMAIL PROTECTED] Error: audio open error: No such file or directory mass ~ ls /proc/asound/ cards dev devices meminfo oss-devices pcm seq timers version How can this be? Here is what I see at boot time: Aug 17 20:41:51 mass kernel: es1370: version v0.22 time 16:39:10 Aug 11 1999 Aug 17 20:41:51 mass kernel: es1370: found adapter at io 0xb000 irq 5 Aug 17 20:41:51 mass kernel: es1370: features: joystick on, line in, mic impedance 0 Aug 17 20:41:51 mass kernel: es1371: version v0.11 time 16:39:14 Aug 11 1999 Aug 17 20:57:48 mass kernel: isapnp: No Plug Play device found Aug 17 20:58:16 mass kernel: snd: Ensoniq AudioPCI soundcard #1 not found or device busy and here are the outputs from some commands that may be relevant: mass ~ cat /proc/modules snd-ens1370 5772 0 snd-pcm1 16988 0 [snd-ens1370] snd-ak4531-codec9932 0 [snd-ens1370] snd-mixer 24288 0 [snd-ens1370 snd-ak4531-codec] snd-seq-midi4292 0 snd-seq29544 0 [snd-seq-midi] snd-timer 7804 0 [snd-pcm1 snd-seq] snd-midi 12716 0 [snd-ens1370 snd-seq-midi] snd-pcm 7660 0 [snd-pcm1] snd33728 0 [snd-ens1370 snd-pcm1 snd-ak4531-codec snd-mixer snd-seq-midi snd-seq snd-timer snd-midi snd-pcm] isapnp 22697 0 sg 11756 0 (unused) parport_probe 3172 0 (autoclean) (unused) lp 5372 0 (unused) parport 7268 0 [parport_probe lp] ufs52000 0 (unused) smbfs 26712 0 (unused) mass ~ fgrep SOUND /boot/config-2.2.10 CONFIG_SOUND=y # CONFIG_SOUND_ES1370 is not set # CONFIG_SOUND_ES1371 is not set # CONFIG_SOUND_SONICVIBES is not set # CONFIG_SOUND_MSNDCLAS is not set # CONFIG_SOUND_MSNDPIN is not set # CONFIG_SOUND_OSS is not set
Copy, paste keys and clipboard like Mac/NeXT/WinNT
Hi For better or worse, over the last 15 years, I've become accustomed to having a clipboard, with key combinations to access it. I can deal with Command-C, Ctrl-C, Alt-C, whatever. But despite my most enthusiastic efforts, I can't get used to the X-windows mouse-button paradigm. I run KDE on a hybrid slink/potato system. Is there a way to get the kind of copy/paste behavior I like between all the various types of X app I run (editors, xterms, kvt, netscape, etc.), given that some have an Edit menu and others do not? What would be really cool is if I could somehow map that useless Windows key to this function. I can no longer find a keyboard with a decent touch and no Windows key! - Brian
Re: Netscape .debs conflict with perl5.004?
Everybody, Thanks to Bob and Brian for the help. Bob is right about the safety -- a late-night session with dselect over the weekend totally screwed up my system (which I had pointing to unstable in order to fetch some drivers I needed). I recommend removing unstable from your sources.list until Perl is fixed, or the same could happen to you! - Brian Bob Nielsen wrote: I agree wholeheartedly. It is probably safest at this point to do upgrades with 'apt-get upgrade' rather than with dselect. Or run 'apt-get update' and see what packages it intends to keep back, say no, go to dselect and put those packages on hold. Bob On Tue, Jul 06, 1999 at 05:55:25PM -0500, Brian Servis wrote: You have unstable in your sources.list file and unstable is VERY unstable right now with regards to perl, most likely lots of broken dependencies. Perl is being transitioned from 5.004 to 5.005 which is not as small of a change as it may appear from the version numbers. Read the debian-devel archives for the details. Basically if you are using any thing in unstable that needs perl then expect major problems for a while until everything settles down. It's called unstable for a reason. Brian
Netscape .debs conflict with perl5.004?
Hi I seem to have some incompatible dependencies in trying to obtain Netscape version 4.6 from the .debs. My problem is that, according to the dependency resolution screens (I copy and paste from various conflict readouts here): navigator-smotif-461 depends on netscape-base-4 (= 10) netscape-base-4 depends on liburi-perl liburi-perl depends on libmime-base64-perl liburi-perl depends on perl (= 5.004) libmime-base64-perl depends on perl-5.005 perl-5.005 depends on perl-5.005-base perl-5.005-base conflicts with perl so basically perl (which is v5.004) is somehow keeping me from installing. I'm no expert here, but I'm savvy enough to know my /etc/apt/sources.list may have something to do with this. Please find it below. One thing I *have* tried is the Q option in dselect to override the dependencies. This doesn't seem to work, as dselect fails to download and install the netscape .debs anyway. My question is, what is the right way to solve this dilemma? Yours, Brian sources.list: deb http://http.us.debian.org/debian stable main contrib non-free deb http://non-us.debian.org/debian-non-US stable non-US deb http://debian.midco.net/debian unstable main contrib non-free deb ftp://ftp.us.kde.org/pub/kde/stable/1.1.1/distribution/deb/binary-i386 / deb http://ftp.netgod.net/ x/ deb ftp://ftp.gnome.org/pub/GNOME/gnome-1.0/debian slink main
Re: OFFTOPIC: How about a /dev/clipboard ?
John Hasler wrote: Brian Boonstra writes: For what it's worth, my OpenStep machine at work has command line directives copy and paste, so you can do things like paste | wc or cat myfile.txt | copy A fairly simple script, if I understand you correctly. I'm not really sure. You can copy something from (say) a word processor using the mouse selection and Alt-C. And then you can paste it as shown above. So unless a script language knows how to access the Display Postscript clipboard, I assume there would have to be some nontrivial C in there. - Brian
Re: Debian hardware vendors
Two vendors I've been happy with are http://www.tdl.com/~netex/ and http://swt.com/ Both are quite Linux-aware. - Brian
Re: OFFTOPIC: How about a /dev/clipboard ?
For what it's worth, my OpenStep machine at work has command line directives copy and paste, so you can do things like paste | wc or cat myfile.txt | copy - Brian
apt through unusual FTP proxy?
Hi
I want to specify ftp sites for apt from behind my company's proxy,
but I can't get it working. Does anyone have experience with USER no PASS
proxies? The relevant portions of my sources.list and /etc/apt/apt.conf look
like the following right now:
Begin
deb http://http.us.debian.org/debian stable main contrib non-free
deb http://non-us.debian.org/debian-non-US stable non-US
deb http://debian.midco.net/debian unstable main contrib non-free
deb ftp://ftp.varesearch.com/pub/mirrors/kde stable apps contrib distribution
deb http://ftp.netgod.net/ x/
- End
and
Begin
ftp
{
Proxy ftp://147.113.146.220:21;;
/* Required script to perform proxy login. */
ProxyLogin
{
// USER $(PROXY_USER);
// PASS $(PROXY_PASS);
// USER $(SITE_USER)@$(SITE) $(SITE_PORT);
// PASS $(SITE_PASS);
USER $(SITE_USER)@$(SITE);// $(SITE_PORT);
};
Timeout 100;
/* Passive mode control, proxy, non-proxy and per-host. Pasv mode
is prefered if possible */
//Passive true;
//Proxy::Passive true;
};
- End
This doesn't work. I have tried all the suggested configutrations
in the sample apt.conf. Different ones give me different error messages, but
as an example here is the error message from the current configuration:
Begin
Err ftp://ftp.varesearch.com stable/apps Packages
TYPE failed, server said: Please login with USER and PASS.
Get:1 ftp://ftp.varesearch.com stable/apps Release
Err ftp://ftp.varesearch.com stable/apps Release
Unable to fetch file, server said 'Please login with USER and PASS. '
Hit http://debian.midco.net unstable/contrib Release
- End
Our FTP proxy works like the following (note the directions it gives
you are not the way it really works). The server can't be too strange
because, for example, the Win95 program WS_FTP can be configured to work with
the proxy by choosing type USER with no PASS:
Begin
wz5000% ftp internet
Connected to internet.cmg.fcnbd.com.
220-===First Chicago NBD FTP proxy server =
220-
220-At prompt enter your username and password.
220-
220-After authenticated to the proxy, you can connect to remote
220-
220-server using the following syntax:
220-
220-user [EMAIL PROTECTED]
220-
220 Enter ctl-D or bye to exit proxy.
Name (internet:boonstb): [EMAIL PROTECTED]
331-(GATEWAY CONNECTED TO ftp.gatekeeper.com)
331-(220 cajon.cfg.com FTP server (Version wu-2.4.2-academ[BETA-17](1) Wed
Jun 10 20:12:28 PDT 1998) ready.)
331 Guest login ok, send your complete e-mail address as password.
Password:
230 Guest login ok, access restrictions apply.
ftp quit
- End
Re: apt's sources.list
Hi there Ben Collins and Jason Gunthorpe have helped me work out how to run apt to FTP debs from unusual places. Thanks guys! I'm an old Unix hand, but way too new to Debian to figure this out. Reading the documentation made me wonder: (1) Is the required format for such targets defined, exemplified, or documented anywhere? (2) What is the deal with the TYPE field in apt.conf, sseing as how only deb and deb-src types seem to be defined? - Brian
Re: apt's sources.list
Hi All Jason wrote: man sources.list ? There are several examples at the bottom and the exact line syntax is given in the section title 'The deb type'. Yeah, it's the only available resource, but it's really too obscure to be useful unless you already know the answer! Take it from somebody who tried. I think part of the problem is that there's no clear explanation of when apt wants to look for a dists/ directory, etc, and when not, and what to do in the latter case. Take the 2nd-to-last example from the manpage: deb ftp://nonus.debian.org/debian-non-US unstable/binary-i386/ versus Ben's eventual solution to Kent's problem deb ftp://ftp.us.kde.org/pub/kde/stable/1.1.1/distribution/deb/binary-i386 / it's really unclear how you get from the first to the second. What's the criterion for splitting it that way? Why can't apt just deal with 2 fields? Best Regards, Brian
Debian installation - list of things that make it hard
Hi I'm experienced with Solaris and NextStep/Openstep -- I've even installed OpenStep from scratch on several machines. I have the official silk-screened 2.1 CD's. My system is pretty standard, with the exception of having SCSI rather than IDE. Now I've decided to check out Linux. I expected installation to go a little more smoothly than it did. As my contribution, however small, to the Debian effort, here are some of the things I found confusing, none of which were addressed in the Installation Guide. Next month, I'll be helping my mother install Debian, so maybe I can add to this list from the perspective of a non-Unix person then: - Both Disks 1 and 2 are bootable. Disk 2 should be labeled as being for special situations. - It is unclear what half or more of the device drivers do in the dbootstrap selection. Even the ones I kind of know, make me wonder. For example, do I need lp to talk to my printer? Will the serial ports on modern motherboards work only if I have serial? When do I need the standard cdrom stuff? When do I need generic scsi? Do I need ftp just to be able to ftp in to my machine? Nothing explains it. - The tasks selection is superb, but the lists of what packages the various tasks load are in random order, with no documentation strings like in dselect, and they are very long. The result is that they are useless in helping decide which tasks to load. I gave up and guessed what I needed from the task names. - After I got to dselect, I mistakenly inserted Disk 1 rather than Disk 2. It seemed logical enough. But no error message told me what I had done wrong -- instead they all complained about not finding files. I ended up reformatting the hard drive and reinstalling from the start TWICE before finally realizing my mistake. If I'm willing to admit how silly I was, how many frustrated mutes must there be? This is especially crazy since it appears one rarely needs Disk 2 anyway. Could they not be made independently functional? - Nothing says where to go from here to set up a sound card, a video card, LILO, or X. Oh, and a final question that isn't so rhetorical. Is it possible to use apt from behind a corporate firewall that does *not* seem to do IP forwarding or masquerading? I don't know much about firewalls, but I work at a big company where I can pretty much guarantee I have no influence over its configuration. Can apt be told about a proxy in the way that, say, lynx can? - Brian
Re: Debian installation - list of things that make it hard
Christensen wrote: Now I feel foolish! I had no idea I was supposed to use disk 2 for dselect. Could that be why there seem to be pieces missing from my installation, like the mouse driver? After getting a helpful email from [EMAIL PROTECTED] I added psaux to the /etc/modules file, and now my mouse works. But maybe I should just start all over again! Well, you're supposed to put in disk 2 for it to decide which Packages are available. Then it makes you switch back to Disk 1 for the actual install. So Disk 2 is in the drive for maybe half a minute. Ever. (That's what I found so crazy). Given that install worked for you my guess (only as a fellow newbie) is you did fine. You might have played a little fast and loose with the mouse setup that happens in the [C]onfig stage. In principle, if you don't know which packages' configuration you might have screwed up, I would assume only a fresh install would make sure you got them taken care of. Alternatively, (and I'm really speculating here) you could wait for the periodic package to break, and reconfigure on the fly. - Brian
