Re: Default Window Manager
Angel Leyva wrote: > > I have X installed and running fine, using xinit. When I start the X > environment, I get no window manager at all. > > How can I get X to start a window manager by default? See man xinit. You´ll have to make up a .xsession file starting all of the applications you want in the background, and starting the window manager as the last set in the foreground. Once you exit the window manager, this script terminates and ends your X session. Ciao, -- Thomas Baetzler, [EMAIL PROTECTED], [EMAIL PROTECTED] http://home.pages.de/~thb/";>thb's Homepage -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word "unsubscribe" to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED]
Suggestion
Hi, all! I want to make a suggestion. Many persons like me have only a 14400 modem, without CDROM and/or sound-card. Here from Brazil, the connection speed to ftp.debian.org is below 0.4 kbps. And the worst is that the other mirrors I tried, like farofa.ime.usp.br (wich is about 10 km from here) are more slow (!). When I have to do a download of a 2.0 Mb updated debian package, my connection bill increases tremendously! :-) So, I think that a good thing would be a bin-diff distribution. When a package is updated from, say, 2.0.3-2 to 2.0.3-5, a bin-diff distribution can decrease very much that bill. Instead of downloading a 2.0 Mb new package, I probably want to make a 100 kb download. The time to execute the undiff is very less than that of download. I saw that there is a diff distribution for sources. Another to the binaries isn't much dificult, specially in rex, is it? That's all. Some suggestion? Alexander Gieg =-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-= By: Alexander Gieg E-mail: [EMAIL PROTECTED] URL: http://www.geocities.com/TimesSquare/3222 IRC: AlexG =-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-= -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word "unsubscribe" to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED]
boot disk for new PC
Hi, I'm trying to put debian on a new PC with a Buslogic Flashpoint LT SCSI card in it. This card wasn't supported by the boot floppies I had so I patched my kernel (2.0.27) and made a custom rescue disk with buslogic, ramdisk, and initrd support compiled in. The boot disk recognizes the card fine but right after the partition check finds "sda: sda1", I get the following messages: RAMDISK: Couldn't find valid ramdisk image starting at 0. Kernel Panic: VFS: Unable to mount root fs on 01:00 >From the previous boot disks I've downloaded, I expected to be asked to insert further disks to load the ramdisk image but the PC now seems to be checking on the rescue disk itself un-successfully. Does anyone know what I'm doing incorrectly? Thanks much... J. Goldman -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word "unsubscribe" to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED]
Re: A.out binaries in debian 1.2 ??? nah.... (fwd)
( Re Message From: Bruce Perens ) > > From: [EMAIL PROTECTED] (Danny ter Haar) > > I am very suprised to find a.out binaries in the rex (and bo) > > I suspect any packages with a.outs in them are orphaned packages. > Of course 1.2 is meant to be ELF-only. > > Thanks > > Bruce > -- > Bruce Perens K6BP [EMAIL PROTECTED] I hope, however, that a.out libraries will continue to be available: they are important when certain software can only be obtained as pre-compiled a.out binary executables. Ted.([EMAIL PROTECTED]) -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word "unsubscribe" to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED]
Re: General clueless upgrade question (fwd)
Bruce Perens wrote: > > From: Ed Down <[EMAIL PROTECTED]> > > I went to the rex directory - not the one in the Debian-1.2 directory, > > so I assume I'm in the 1.1 tree - and downloaded a new Packages file and > > the packages I require. > > "rex" and "Debian-1.2" are two different names for the same directory. > We always use code-names during development, and "rex" was the code-name > for 1.2 . > > Debian 1.1 is, I think, still available on debian.crosslink.net, but not > for long. It's code-name is "buzz". You may want to reload the Packages > file from there. > > > I ran dselect and it seems to not let me just install the new > > packages - it wants me to upgrade loads of other packages which the new > > ones are not dependent on. > > It will want to upgrade all of your currently-installed packages to the new > versions. > > > I've tried to ask deselect to just leave the > > old packages alone, but then it insists on wanting to delete all the > > packages dependent upon these. For instance, I try to leave libc5 as the > > old version, so it insists on wanting to delete cron! > > Pressing "-" doesn't mean leave it alone, it means delete the package. > You can select "=" (hold in present state) for all of the packages you > want to keep the same, but it would be better to load a Packages file > for 1.1 than to do this for 800 packages. > You don't need the installation floppies to upgrade. Just put a Debian 1.2 > CD in your system and let dselect perform the upgrade. You can also have > dselect do it via FTP. You will need the latest version of the "dpkg" and > "dpkg-ftp" packages installed first - install these by hand using "dpkg", > and then "dselect" can handle the rest of the upgrade. I had a problem with perl not being replaced by perl-base during my upgrade. Would this be due to me not upgrading dpkg-ftp and dselect first? -- Bernard LeachAustralian Business Access Pty Ltd. [EMAIL PROTECTED] http://www.aba.net.au/ -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word "unsubscribe" to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED]
Re: 1.2 Upgrade Experience
On Mon, 23 Dec 1996, Scott J. Geertgens wrote: > > While upgrading, a number of packages (about 10 of the over 100) > > failed to install. The result of this was that after the first >When I first installed Debian, this really spooked me as well. I > managed to watch the messages as they flew by, and was able to note that > all of the packages that failed depended on perl being installed. The way > the files were selected though, perl is one of the later things to be > installed, so everything installed before it that depends on it will fail. > How's that for a sentence? Anyway, I found that by simply doing a 'dpkg -i > perl' and then a 'dpkg -i dpkg...', everything worked once again and > all other failed packages could then be installed and set up by dselect. >My suggestion to the debian team would be to add a feature to dselect > that would flag certain packages as 'on hold', install as much as > possible, then go back and try to install the failed packages again. The > dependencies are all correct, but order of installing makes a difference > as well. I vaguely assumed that "pre-depends" was a mechanism to handle this situation. Is that not so? If it is, then are all those packages missing the pre-depends flag? ...RickM... -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word "unsubscribe" to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED]
Re: Is it crucial to use cfdisk to repartition
You don't have to use cfdisk, but you _do_ have to wipe out the filesystem. Various people have reported some success with installing Debian _over_ Slackware, but if you do that there will be files left behind from the old system that will never be cleaned up, etc. We don't support upgrading on top of another Linux distribution without first wiping out the filesystem because we can't guarantee it will work well. Thanks Bruce -- Bruce Perens K6BP [EMAIL PROTECTED] Finger [EMAIL PROTECTED] for PGP public key. PGP fingerprint = 88 6A 15 D0 65 D4 A3 A6 1F 89 6A 76 95 24 87 B3 -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word "unsubscribe" to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED]
Re: Is it crucial to use cfdisk to repartition
On Mon, 23 Dec 1996, Walter Tautz wrote: Hi Walter, > by slackware. I take it that one can skip the `initialization' step and > simply use the existing partitions or is it advised that I repartition > using cfdisk? There is no immediate need for repartitioning with cfdisk. If you are happy with your current layout, just keep it! Yours, -- martin -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word "unsubscribe" to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED]
Re: 1.2 Upgrade Experience
[EMAIL PROTECTED] wrote: > I've just completed an upgrade to Debian 1.2, and would like to > relate some details of the experience in the hopes that it might > help others who get stuck as I did. I also have some suggestions > for the maintainers at the end of this message. But before I get > to any of that, I want to say "thanks". I just did the same last night and I'd have to say ditto! The upgrade went pretty well and "thanks"! > [...] > > While upgrading, a number of packages (about 10 of the over 100) > failed to install. The result of this was that after the first > round of attempted installations of the new/updated packages, the > ftp access method for dselect no longer worked. I had similar problems, but dpkg-ftp wasnt working properly. For me it was just a matter of re-running the configure process (dpkg --configure --pending) until things started working again. > I really didn't want to restore from backup, and start over again, > so I took a shot at using the "mounted" access method to access > the package files that the "ftp" method had just downloaded. A > little "find"-ing found the .deb files in > > /var/lib/dpkg/methods/ftp/debian/Debian-1.2/binary-i386/..., You could have just used dpkg directly with dpkg --install *.deb :) > [...] > The following is some suggestions for the Debian maintainers. > > In the above I would have liked to be more specific about what failed > and how, but I did not keep a written log, and apparently dselect didn't > either. I'd like to suggest that dselect keep a transcript of what > it attempts to do and whatever messages that attempt produces. > > I noted that many packages printed warning messages and advice as > they were installed or configured. This works fine when your just > upgrading a single package, but when your upgrading 100 or so packages, > these messages just scroll off the screen never to be seen again. The > aforementioned transcript would help here too, but I'd rather see such > advice and recommendations put into a readme.debian file in a > /use/doc/packageX/ file so it can be reread later easily. What would really be nice is a transcript log that could be played back through dselect to reproduce the same results. OS/2 had a feature similar to that which would allow you to install one machine and then reproduce the same system on other machines by using the transcript log. -- Bernard LeachAustralian Business Access Pty Ltd. [EMAIL PROTECTED] http://www.aba.net.au/ -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word "unsubscribe" to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED]
Re: 1.2 Upgrade Experience
> > While upgrading, a number of packages (about 10 of the over 100) > failed to install. The result of this was that after the first > round of attempted installations of the new/updated packages, the > ftp access method for dselect no longer worked. > > This was very disconcerting! > > -gavin... When I first installed Debian, this really spooked me as well. I managed to watch the messages as they flew by, and was able to note that all of the packages that failed depended on perl being installed. The way the files were selected though, perl is one of the later things to be installed, so everything installed before it that depends on it will fail. How's that for a sentence? Anyway, I found that by simply doing a 'dpkg -i perl' and then a 'dpkg -i dpkg...', everything worked once again and all other failed packages could then be installed and set up by dselect. My suggestion to the debian team would be to add a feature to dselect that would flag certain packages as 'on hold', install as much as possible, then go back and try to install the failed packages again. The dependencies are all correct, but order of installing makes a difference as well. SJG -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word "unsubscribe" to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED]
Is it crucial to use cfdisk to repartition
I am thinking about installing debian over slackware, but I don't want to repartion the drive. The drive was partitioned with the program used by slackware. I take it that one can skip the `initialization' step and simply use the existing partitions or is it advised that I repartition using cfdisk? I want to retain certain files simply for convenience and because I don't have a tape backup system. Note: I HAVE BACKED UP CRITICAL FILES VIA FLOPPY TO ANOTHER SYSTEM. -Walter -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word "unsubscribe" to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED]
1.2 Upgrade Experience
I've just completed an upgrade to Debian 1.2, and would like to relate some details of the experience in the hopes that it might help others who get stuck as I did. I also have some suggestions for the maintainers at the end of this message. But before I get to any of that, I want to say "thanks". I've been using Debian since the early 0.93 days, and while it's never been perfect, it has always been better than anything else out there. I have installed both Windows and Debian a number of times and I'm still shocked that Debian is easier. Thanks to everyone who has contributed to this! I have been using the dselect ftp access method to add or upgrade packages to my system for some time now, and in general have had no problems with it. So I decided to use it for the upgrade to 1.2. While upgrading, a number of packages (about 10 of the over 100) failed to install. The result of this was that after the first round of attempted installations of the new/updated packages, the ftp access method for dselect no longer worked. This was very disconcerting! I really didn't want to restore from backup, and start over again, so I took a shot at using the "mounted" access method to access the package files that the "ftp" method had just downloaded. A little "find"-ing found the .deb files in /var/lib/dpkg/methods/ftp/debian/Debian-1.2/binary-i386/..., and it only took a couple of attempts to properly answer "mounted"'s querys for the location of the files it needed. (Thanks to whoever decided that the files downloaded by the ftp access method should be stored in a way acceptable to the mounted access method.) The only hitch was that I had to add a link from /var/lib/dpkg/methods/ftp/debian/stable to Debian-1.2. My assumption was that running install again over the same .deb files would produce the same results, but at least I could look at the errors, and make some guesses about how to fix them. I was wrong. About 3/4 of the packages that failed to install the first time, installed the second time. Running install yet again got the last few installed. While I'm happy with the result, this seems to me to indicate some missing dependencies somewhere. I now have a working 1.2 system, and am happily writing this message there on. I hope the previous paragraphs are helpful to anyone who gets stuck as I did. The following is some suggestions for the Debian maintainers. In the above I would have liked to be more specific about what failed and how, but I did not keep a written log, and apparently dselect didn't either. I'd like to suggest that dselect keep a transcript of what it attempts to do and whatever messages that attempt produces. I noted that many packages printed warning messages and advice as they were installed or configured. This works fine when your just upgrading a single package, but when your upgrading 100 or so packages, these messages just scroll off the screen never to be seen again. The aforementioned transcript would help here too, but I'd rather see such advice and recommendations put into a readme.debian file in a /use/doc/packageX/ file so it can be reread later easily. Again, thanks for your efforts. -gavin... -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word "unsubscribe" to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED]
New 1.2 boot floppy uploaded
A new boot floppy for 1.2 is in ftp://ftp.i-connect.net//pub/Linux/Debian/rex/disks-i386/1996-12-8/new-resq1440.bin If your system hangs while booting, please try this. Various device drivers have been removed from the kernel. Nothing else has been changed. We will be making a point release (1.2.1) shortly after new years with many of the current list of installation bugs fixed and a new boot disk set. I'd make it sooner, but I can't make that release until various key staff people return from Xmas vacation. As always, if you have 1.2 (or if you have already ordered a CD) you will be able to upgrade to 1.2.1 easily via FTP. Thanks Bruce -- Bruce Perens K6BP [EMAIL PROTECTED] Finger [EMAIL PROTECTED] for PGP public key. PGP fingerprint = 88 6A 15 D0 65 D4 A3 A6 1F 89 6A 76 95 24 87 B3 -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word "unsubscribe" to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED]
Re[2]: DNS Request
Me too... more than happy... Ryan __ Reply Separator _ Subject: Re: DNS Request Author: debian-user@lists.debian.org at Internet Date:12/23/96 3:38 PM I'll take the job on. On Sat, 21 Dec 1996, Angel Leyva wrote: > I am trying to set up DNS on a local network at home and things are > not going too smoothly. I wonder if some one is willing to correspond > with me via e-mail so that I can get this thing running. -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word "unsubscribe" to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED] -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word "unsubscribe" to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED]
Re: A.out binaries in debian 1.2 ??? nah....
From: [EMAIL PROTECTED] (Danny ter Haar) > I am very suprised to find a.out binaries in the rex (and bo) I suspect any packages with a.outs in them are orphaned packages. Of course 1.2 is meant to be ELF-only. Thanks Bruce -- Bruce Perens K6BP [EMAIL PROTECTED] Finger [EMAIL PROTECTED] for PGP public key. PGP fingerprint = 88 6A 15 D0 65 D4 A3 A6 1F 89 6A 76 95 24 87 B3 -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word "unsubscribe" to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED]
Re: SCSI cdrom
Edit /etc/makedev.cfg to remove the "omit { }". It's a syntax error.
We'll make a point release after New Years with this fixed.
Bruce
--
Bruce Perens K6BP [EMAIL PROTECTED]
Finger [EMAIL PROTECTED] for PGP public key.
PGP fingerprint = 88 6A 15 D0 65 D4 A3 A6 1F 89 6A 76 95 24 87 B3
--
TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word "unsubscribe" to
[EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED]
Re: Debian Boot Problems on a Thinkpad
Try adding "floppy=thinkpad" to the boot parameters, as in: boot: linux floppy=thinkpad You also might need to add hd=,, . Bruce -- Bruce Perens K6BP [EMAIL PROTECTED] Finger [EMAIL PROTECTED] for PGP public key. PGP fingerprint = 88 6A 15 D0 65 D4 A3 A6 1F 89 6A 76 95 24 87 B3 -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word "unsubscribe" to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED]
A.out binaries in debian 1.2 ??? nah....
I am very suprised to find a.out binaries in the rex (and bo) debian distributions. In the package /debian/bo/binary-i386/graphics/ucbmpeg_1r2-2.deb which installs: /usr/bin/ppmtoeyuv /usr/bin/eyuvtoppm /usr/bin/eyuvtojpeg /usr/bin/jmovie2jpeg /usr/bin/mpeg_demux /usr/bin/mpeg_bits /usr/bin/mpeg_blocks /usr/bin/mpeg_encode /usr/bin/mpeg_play /usr/bin/mpeg_stat etc mpeg_bits: Linux/i386 demand-paged executable (QMAGIC), stripped mpeg_blocks: Linux/i386 demand-paged executable (QMAGIC), stripped mpeg_demux: Linux/i386 demand-paged executable (QMAGIC), stripped mpeg_encode: Linux/i386 demand-paged executable (QMAGIC), stripped mpeg_play: Linux/i386 demand-paged executable (QMAGIC), stripped mpeg_stat: Linux/i386 demand-paged executable (QMAGIC), stripped I thought 1.1 and 1.2 would be ELF only ??? A happy Debian user. ;) Danny -- Danny ter Haar | Cistron Internet Services | Unix & Internet [EMAIL PROTECTED] | finger [EMAIL PROTECTED] for PGP-key |specialists ---+31-172-419445SP6-- -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word "unsubscribe" to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED]
SCSI cdrom
I'm trying to install Debian 1.2 to my PPRO with the SCSI hard disk and CD-ROM. I have successfully installed the kernel and the base disks. Now, I want to install all packages from the CDROM. However, there is no /dev/sr0 or /dev/scd0, which means I cannot mount the CD-ROM. I try to run MAKEDEV in /dev, it conplains about some unexpected "}" in makedec.cfg. Can anybody please help? By the way, the SCSI-cdrom is recognized at the booting time. Xiaoming Ding -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word "unsubscribe" to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED]
Debian Boot Problems on a Thinkpad
I hope that I am not missing a FAQ anywhere, but I am having trouble getting the Debian 1.2 floppies to work on my IBM THinkpad 365XD. Booting the rescue floppy just gets to 'Loading Linux' and then it hangs. I have been forced to install Slackware 3.0 on the machine in the interim (and that boot/root floppy combination worked perfectly). What can I do to get this working? TIA, PAul -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word "unsubscribe" to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED]
Re: DNS Request
I'll take the job on. On Sat, 21 Dec 1996, Angel Leyva wrote: > I am trying to set up DNS on a local network at home and things are > not going too smoothly. I wonder if some one is willing to correspond > with me via e-mail so that I can get this thing running. -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word "unsubscribe" to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED]
Another general clueless upgrade question (fwd)
On Mon, 23 Dec 1996, Don Prezioso wrote: > On Mon, 23 Dec 1996, Ed Down wrote: > > > OK, I instaled Debian 1.1 a while ago and it all went fine. Now I want to > > add a couple more packages, so I ftp'd to my local mirror. I went to the > I have Debian 1.1.4 and it came with kernel 2.0.6. Now I want to upgrade to kernel 2.0.27 without upgrading entirely to Debian 1.2 yet. Neither do I want to download the gzipped 2.0.27 source, I just want to patch my old source tree. Is there a Debian way to do this? Thanks in advance, Joost -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word "unsubscribe" to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED]
Re: IP Routing Problems....
[EMAIL PROTECTED] wrote: > > -BEGIN PGP SIGNED MESSAGE- > > On Sat, 21 Dec 1996, Kevin Traas wrote: > > > > > The router at the ISP is pretty non-configurable. It only supports > > dynamic routing, so the only thing that can be configured on it is the > > IP address of the port I'm connecting to. All routing information is > > supposed to be "learned" via RIP. But, it's not learning > > > > If this is my problem, do you have any ideas why? > > > > I suppose that your ISP has granted you a subdoamin for your local > network. But if your problem is in configuring the ISP router (which you > didn't have access to) you could set up a firewall on box_a. That could be > done rather transperant, with the only dissadvantage that you can't run > internet services on the same port on two machines on you local network. > But as long as the ports differ, there should be no problem setting up the > firewall to route the packages to the right box. > I missed the original message, but I'll assume you're only net (internet or otherwise) connection is through your dialup to your ISP. If this is the case the only route you need is a default route to the other end of the PPP link. You can write an "ip-up" script which pppd will run and set the route there. pppd will pass the IP address which you'll need for the 'route add' as a parameter to the script. Jens B. Jorgensen [EMAIL PROTECTED] -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word "unsubscribe" to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED]
Re: Bug or feature?
> > If I understand correctly, as soon as I set the 'user' flag in /etc/fstab, > > any user can mount/unmount a partition. But - why can't I use 'mount > > -atnfs' when all nfs partitions have the user flag set (error message is > > 'mount: only root can do that')? > I suspect you also need mount permission from the nfs hosts. > Otherwise, IMHO, this would be a very handy trick to hack into other > perople's computers ?? I didn't state so, but mounting each of the nfs partitions separately works, i.e. 'mount /Firefranc' fstab entry is: firefranc:/ /Firefranc nfs user,exec,dev just the 'mount -atnfs' comes up with an error... Benedikt signoff --- Benedikt Eric Heinen - Muehlemattstrasse 53 - CH3007 Bern - SWITZERLAND email: [EMAIL PROTECTED]phone: ++41.79.3547891 RIOT, n. A popular entertainment given to the military by innocent bystanders. Ambrose Bierce ``The Devil's Dictionary'' -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word "unsubscribe" to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED]
Re: InfoMagic Debian 1.2 Installation
On Sun, 22 Dec 1996, Randall E. Price wrote: > Bruce Perens wrote: > > > > I think you might also have to install "xext". > > > > SO far adding the modules for pex and xie has solved the first problem, > but now X11 starts up fine and appears to be working well and then it > just exits without an error message. I installed xext, well sort of, > there is no xext module in the directory with pex5.so and xie.so, so I > don't really understand what is supposed to happen with "xext". So I > still do not have it working. Have you added /usr/X11R6/lib to /etc/ld.conf? It might be the old ld.so-xlib problem in a disguise. The program started last, probably your window manager, in your .xsession file looks for the xlibraries, can't find it, and exits with an error immediately, upon which the xserver exits normally. If not the xlibraries, there might be some other configuration error for the w-manager. --prashanth -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word "unsubscribe" to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED]
dpkg-ftp, SOCKS, elf dynamic linking
Someone recently posted a question about how they could use the ftp access method for dselect (dpkg-ftp) across a SOCKS 4.x firewall. I've wanted to do this for quite a while myself, and it seemed to me the shorted path would be to first create a perl package called 'RSocket' with the same functions/syntax as Socket and then just replace the 'use Socket' with 'use RSocket'. I'm positive this would work, but then I looked at the utilities for created dynalink (???) packages to object libraries and decided I'd have to wait until I had time to spend a few hours learning how to do this. Since that time I've had it in the back of my mind and then the other day I thought "Hey, wait a minute! Doesn't ELF just use the object symbols to perform dynamic linking? Couldn't I just make a SOCKS shared libraries with its socket calls renamed to be the same as the regular socket calls and put this in the front of my LD_LIBRARY_PATH before I run dpkg-ftp?" My suspicion is that this would not work unless I made a version of the entire C library to replace the one I have. Am I wrong? Has anyone made a SOCKS package for perl? Isn't there a better way? Jens B. Jorgensen [EMAIL PROTECTED] ps. Yes, I know if I were a *real* man I'd make that perl package and then make a new dpkg-ftp (dpkg-socks-ftp?) and share it with the rest of my debian brothers/sisters--I just don't have the time. -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word "unsubscribe" to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED]
Re: General clueless upgrade question (fwd)
From: Ed Down <[EMAIL PROTECTED]> > I went to the rex directory - not the one in the Debian-1.2 directory, > so I assume I'm in the 1.1 tree - and downloaded a new Packages file and > the packages I require. "rex" and "Debian-1.2" are two different names for the same directory. We always use code-names during development, and "rex" was the code-name for 1.2 . Debian 1.1 is, I think, still available on debian.crosslink.net, but not for long. It's code-name is "buzz". You may want to reload the Packages file from there. > I ran dselect and it seems to not let me just install the new > packages - it wants me to upgrade loads of other packages which the new > ones are not dependent on. It will want to upgrade all of your currently-installed packages to the new versions. > I've tried to ask deselect to just leave the > old packages alone, but then it insists on wanting to delete all the > packages dependent upon these. For instance, I try to leave libc5 as the > old version, so it insists on wanting to delete cron! Pressing "-" doesn't mean leave it alone, it means delete the package. You can select "=" (hold in present state) for all of the packages you want to keep the same, but it would be better to load a Packages file for 1.1 than to do this for 800 packages. > On a different track, I want to upgrade to 1.2 at some point. Is there a > doc on this? If I use the new installation disks does it give me the > option just to upgrade, without the disk format, or can I just upgrade the > kernel and all the other packages I use to the latest vewrsions and that > will be the same as a 1.2 system installed from scratch? You don't need the installation floppies to upgrade. Just put a Debian 1.2 CD in your system and let dselect perform the upgrade. You can also have dselect do it via FTP. You will need the latest version of the "dpkg" and "dpkg-ftp" packages installed first - install these by hand using "dpkg", and then "dselect" can handle the rest of the upgrade. Thanks Bruce -- Bruce Perens K6BP [EMAIL PROTECTED] Finger [EMAIL PROTECTED] for PGP public key. PGP fingerprint = 88 6A 15 D0 65 D4 A3 A6 1F 89 6A 76 95 24 87 B3 -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word "unsubscribe" to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED]
Re: General clueless upgrade question (fwd)
On Mon, 23 Dec 1996, Ed Down wrote: > > > -- Forwarded message -- > Date: Sun, 22 Dec 1996 18:58:54 + (GMT) > From: Ed Down <[EMAIL PROTECTED]> > To: debian-user@lists.debian.org > Subject: General clueless upgrade question > > > OK, I instaled Debian 1.1 a while ago and it all went fine. Now I want to > add a couple more packages, so I ftp'd to my local mirror. I went to the > rex directory - not the one in the Debian-1.2 directory, so I assume I'm > in the 1.1 tree - and downloaded a new Packages file and the packages I > rex IS 1.2. I think that buzz is 1.1. There are several directories like Debian-1.2, stable, etc that point to rex (1.2) I don't know if there is anything besides buzz that has 1.1 anymore. I also have no clue as to how up-to-date it is. > On a different track, I want to upgrade to 1.2 at some point. Is there a > doc on this? If I use the new installation disks does it give me the > option just to upgrade, without the disk format, or can I just upgrade the > kernel and all the other packages I use to the latest vewrsions and that > will be the same as a 1.2 system installed from scratch? >From my 1.1 system I just used dselect using the ftp access method and things went fairly smoothly. I don't know if there is a doc about such upgrading, it would be nice if there was one posted like with the install instructions, but here is what I did. After running dselect the first time through, I got some problems with packages that pre-depended on others. It might help to try JUST updating the base packages first, then the others. Maybe it would be better to use dpkg on base, then use dselect, I don't know. In any case, I re-ran dselect a couple of times which took care of most of the problems, the only difficult one being that dselect got kind of broken in the process, and the ftp access stopped working for a while. Since I had downloaded all the packages that I needed (including dselect) I re-directed dselect to use the directory on the hard-drive that had the downloaded files. It worked fine getting finished and that stuff was OK. The only problems left were getting my Ethernet card recognized, and getting the CD-Rom module for my cd to load. I fixed the Ethernet card using modconf, pretty easy and no problems. The CD was tougher, and I finally re-made the kernal to get rid of all the probing for CD-Roms that the default 1.2 Kernal tries to do. Along the way, I found out how to get the CD working without the recompile, but I have an old, weird (cdu33a) CD so that may not be a problem for you. I suspect that there are a few simple things that could have made the upgrade easier, but all in all, it was pretty nice, and I certainly did NOT have to reformat my disk. Most of the package upgrades were even able to recognize my original configurations and so just about everything came up just as it had been. Hope this helps, - Don - Don Prezioso Ashland University Phone: (419) 289-5015 System Programmer/Analyst 401 College Avenue E-mail: [EMAIL PROTECTED] Administrative Computing Ashland, OH 44805 http://www.ashland.edu/~dprez -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word "unsubscribe" to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED]
Re: Re[2]: IP Routing Problems....
> I do have access to their router, but there is nothing, besides the > interface IP address, to configure on it - all routing is learned via RIP. > I can't establish any static routes and it's not learning them dynamically > and I can't figure out why. I don't want Box_A to be a firewall (yet) - I > just want it to route traffic through to my network. > > What kind of router is this? Sorry, can't remember and don't have access right now > Also, what kind of protocol connects you > to their network? (E.g. If you are running PPP, you must be aware > that ppp is not a broadcast protocol so RIP does little good. The > router must be aware of the subnetting scheme so it can forward > packets efficiently). And, what are you using for a router on your > end? UNIX, Cisco, Windows95 (hehehe)??? Unix all the way! Win95!!! hehehe is right! To answer your question, here's an excerpt from my original post: A <---1---> B <---2---> C <---3---> D A is an Ethernet interface at my ISP B is a Linux box C is a Linux box D is my local area network 1 is an Ethernet segment 2 is a Serial pointopoint running CSLIP 3 is another Ethernet segment Thanks for your help, Kevin Traas -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word "unsubscribe" to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED]
Default Window Manager
I have X installed and running fine, using xinit. When I start the X environment, I get no window manager at all. How can I get X to start a window manager by default? Angel Leyva ([EMAIL PROTECTED]) http://cybernex.net/~airborne -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word "unsubscribe" to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED]
Re[2]: IP Routing Problems....
I do have access to their router, but there is nothing, besides the interface IP address, to configure on it - all routing is learned via RIP. I can't establish any static routes and it's not learning them dynamically and I can't figure out why. I don't want Box_A to be a firewall (yet) - I just want it to route traffic through to my network. What kind of router is this? Also, what kind of protocol connects you to their network? (E.g. If you are running PPP, you must be aware that ppp is not a broadcast protocol so RIP does little good. The router must be aware of the subnetting scheme so it can forward packets efficiently). And, what are you using for a router on your end? UNIX, Cisco, Windows95 (hehehe)??? Ryan -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word "unsubscribe" to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED]
Re: IP Routing Problems....
> On Sat, 21 Dec 1996, Kevin Traas wrote: > > > > The router at the ISP is pretty non-configurable. It only supports > > dynamic routing, so the only thing that can be configured on it is the > > IP address of the port I'm connecting to. All routing information is > > supposed to be "learned" via RIP. But, it's not learning > > > > If this is my problem, do you have any ideas why? > > I suppose that your ISP has granted you a subdoamin for your local > network. But if your problem is in configuring the ISP router (which you > didn't have access to) you could set up a firewall on box_a. That could be > done rather transperant, with the only dissadvantage that you can't run > internet services on the same port on two machines on you local network. > But as long as the ports differ, there should be no problem setting up the > firewall to route the packages to the right box. > - -- > Name: Hakan Ardo > E-Mail: [EMAIL PROTECTED] I do have access to their router, but there is nothing, besides the interface IP address, to configure on it - all routing is learned via RIP. I can't establish any static routes and it's not learning them dynamically and I can't figure out why. I don't want Box_A to be a firewall (yet) - I just want it to route traffic through to my network. Thanks for your help. Kevin Traas -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word "unsubscribe" to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED]
Re: Socksified dpkg-ftp?
Jens B. Jorgensen wrote: > > Stefan Sunaert wrote: > > > > Hi, > > > > our debian linux box is behind a firewall operating socks4 protocol. > > The dpkg-ftp package does not seem to use our rftp client, so we cannot > > use dselect (ftp method). > > Does anyone have a solution? > > -- > > I'm in the same situation but I don't have a solution. I've thought > about > making a Perl package, but I don't know about how to do that, so I > haven't > gotten around to it. I think that would probably be easiest. There is > currently a Socket package. I would make an RSocket package and then > just > change the dpkg-ftp perl code to use that package instead. (dpkg-ftp is > written in perl in case you hadn't caught on). > > Jens B. Jorgensen > [EMAIL PROTECTED] I finally got one way to do this (what a weekend!) One way to do this is to socksify perl. There's a lot of information on http://www.socks.nec.com/. There, you will find the sourcecode for the socksserver (sockd) and socksclient software (such as rftp, and the socks library). There's an archived mailing list with information on how to build the soft on linux, and how to socksify a client; see also http://www.socks.nec.com/how2socksify.html. Our linux box is behind a socks4 firewall, so I downloaded the socks4 software. After too many trial an error attempts, and after reading the archived mailinglist on how to compile this on linux, I finally got a working rftp and a libsocks library. (I didn't get the full soft to compile cleanly - no rtelnet, no rfinger) The next step is to socksify the complete perl package (instead of, as you suggested, make a RSocket package). I downloaded the debian perl5.003 sourcecode, and modified the debian.rules, according to the "How to Socksify a client" page (http://www.socks.nec.com/how2socksify.html): this is: add the cflags: -Dconnect=Rconnect -Dgetsockname=Rgetsockname -Dbind=Rbind -Daccept=Raccept -Dlisten=Rlisten -Dselect=Rselect and add -lsocks for the libs. Then type "debian.rules binary", which will make a .deb perl package, which you can install using dpkg. This debian perl package now knowns how to deal with the socks4 server running on the firewall. Then the ftp option of dselect worked! (except for DNS; give the IP address of your local debian mirror - for DNS some more cflags are needed - see the webpage and the archived mailing list) I hope this helps. -- ~ Stefan Sunaert, MD [EMAIL PROTECTED] Department of Radiology Phone: +32 16 347754 University Hospital Gasthuisberg Fax : +32 16 343769 K.U.Leuven Herestraat 49, B-3000 Leuven (Belgium) ~ -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word "unsubscribe" to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED]
General clueless upgrade question (fwd)
-- Forwarded message -- Date: Sun, 22 Dec 1996 18:58:54 + (GMT) From: Ed Down <[EMAIL PROTECTED]> To: debian-user@lists.debian.org Subject: General clueless upgrade question OK, I instaled Debian 1.1 a while ago and it all went fine. Now I want to add a couple more packages, so I ftp'd to my local mirror. I went to the rex directory - not the one in the Debian-1.2 directory, so I assume I'm in the 1.1 tree - and downloaded a new Packages file and the packages I require. I ran dselect and it seems to not let me just install the new packages - it wants me to upgrade loads of other packages which the new ones are not dependent on. I've tried to ask deselect to just leave the old packages alone, but then it insists on wanting to delete all the packages dependent upon these. For instance, I try to leave libc5 as the old version, so it insists on wanting to delete cron! As my system works fine at the moment I do not want to risk instaling anything with dselect as it is - any tips? Also, as I have to transfer all my packages by disk from college to home I don't want to have to upgrade all of the packages if I can help it, just add the new ones (and any dependencies). On a different track, I want to upgrade to 1.2 at some point. Is there a doc on this? If I use the new installation disks does it give me the option just to upgrade, without the disk format, or can I just upgrade the kernel and all the other packages I use to the latest vewrsions and that will be the same as a 1.2 system installed from scratch? Thanks, Ed -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word "unsubscribe" to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED]
Re: install troubles
On Sat, 21 Dec 1996, Adam Wiggins wrote: > Hey there folks. New (potential) Linux user here; having trouble getting > Debian to install on my dinosaur system, and haven't been able to locate > anything in the troubleshooting guides or user docs to help out. > Upon booting from the Rescue floppy, everything works fine for a while > (lots of failed to locate cdrom messages and such) until it gets to: > > Partition check: >hda: hda1 > RAMDisk: Compressed image found at block 0 > Couldn't get a free page... > out of memoryVFS: Mounted root (minix filesystem) > init: cache '/etc/ld.so.cache' is corrupt > > At which point it goes no further. The machine is a 486-33 (intel) with > 120meg IDE drive (currently formated for dos 5.0), SVGA, a few other > random periphrials. The boot disk is a 3 1/2 floppy and I've tried making > several just in case it was a data error of some sort. My first guess is > that it's trying to make a RAM disk and then running out of memory in my > paltry 4 megs of ram; however I was unable to locate any boot options for > disabling this, and I have no idea if that's really it being that I know > next to nothing about Linux. It is version 2.0.27 of Debian. Any > thoughts or perhaps pointers to documentation which adresses this > appreciated. > Adam, I don't know how much help this will be but since I didn't see any replys to your message, here goes... Back a while ago, I loaded Debian 1.1 on an aging 386 with with 4 Meg of RAM (seems like ages ago, but it was less than 6 months...), and one of the first things I had to do was configure a swap partition. If I remember correctly, after loading the boot and root disks (I know the procedure is different now, I think the rescue disk is all you need), and before I got the nice menu screen, there was a plain terminal-like screen about creating and initializing a swap partition. My memory of this is sort of vague, but essentially it allowed me to use fdisk (or cfdisk, I'm not sure) to partition my hard drive and create a swap partition and activate it, all before actually starting the real install program. I think the way the screen was worded, made it sound sort of 'optional' but on a 4 Meg system it is REQUIRED. Hopefully, Bruce has gotten back to you outside of the list. I know he has done quite a bit with 'low memory' systems. If however you are getting this all before you have the oportunity to create a swap partition you may have another problem where your system doesn't get to see all 4 Meg of memory. I remember seeing something on this list recently about some systems that were configured with 'Shadow RAM' enabled having problems. Check out: ftp://ftp.debian.org/debian/Debian-1.2/disks-i386/current/install.html (The current Debian 1.2 installation instructions, 2.0.27 is the Linux kernal version) There is a section under 'Configuring Your System' about Shadow RAM. Hope this helps, - Don - Don Prezioso Ashland University Phone: (419) 289-5015 System Programmer/Analyst 401 College Avenue E-mail: [EMAIL PROTECTED] Administrative Computing Ashland, OH 44805 http://www.ashland.edu/~dprez -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word "unsubscribe" to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED]
Re: dpkg --configure texbin ... fails
On Mon, 23 Dec 1996, Zenon Fortuna wrote: > My original Debian installation was done from the InfoMagic > CD-ROM from Sept 1996. > To update to "Debian 1.2" I have downloaded the *.deb packages a week ago, > and today have run the "dselect" to upgrade all packages. > > The most of the packages updated cleanly (I was really impressed !!). > Only a few got some DPKG errors. It seems, that most (if not all) > problems were caused by failure of the "texbin" update. > > Below I add the "dpkg --configure texbin" messages: > > Any hints what to do ? > TIA, > Zenon > > ... > # dpkg --configure texbin > Setting up texbin (3.1415-5) ... > kpathsea: Running MakeTeXTFM manfnt.tfm > Running MakeTeXPK manfnt.tfm > mf \mode:=nullmode; mag:=1; scrollmode; input manfnt \ mf: can't load library 'libXt.so.6' > Metafont failed for some reason on manfnt.tfm > kpathsea: Appending font creation commands to missfont.log. > dpkg: error processing texbin (--configure): > subprocess post-installation script returned error exit status 1 > Errors were encountered while processing: > texbin > # > ... > > Actually, there exist the following libraries: > /usr/X11R6/lib/i486-linuxaout/libXt.so.6 > /usr/X11R6/lib/libXt.so.6 > > So, what to do ? > The same happened to me. This must be a bug in appropriate instalation script responsible for adding X11R6 to the system. Because of this no one of your X apps should work, not only TeX staff. You can repaire it by hand as follows: 1) Locate all the libX* libraries (may be you have more then ones mentioned in your letter. 2) Add their pathnames to /etc/ld.so.conf 3) Run 'ldconfig'. This worked for me. Jerzy Kakol Institute of Microbiology Wroclaw University, Poland -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word "unsubscribe" to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED]
Re: Taking down PPP at a certain hour of the day
On Mon, 23 Dec 1996, Matthew Stone wrote: > I am wondering.. how I can do this.. I would like to create a script that > run in the background and check the time.. if it is 8am it do a > wall announcement saying the system is going down, kill all the incoming > telnet connections, and disconnect my PPP connection.. if it's not then > it will just keep on running until is.. Anywhere on how I would attempt > to do this.. I don't know how to shell program so I wouldn't know where to > start.. Hello Matthew, A shell command that you can use to kill pppd is: kill -HUP `cat /var/run/ppp0.pid` The back-tics will replace the cat command with its output which is the pid that is contained in the file /var/run/ppp0.pid. (You may want to check that this is indeed the filename used on your system by taking a look in the /var/run directory when pppd is running). Note that you will have to be root for this command to work. As far as running a command to warn your users that pppd is going down, I'm not quite sure what will work there. I know that for people that are on your system via a terminal session, you can use wall. I don't think that will work for a user that is visiting your site via netscape or ftp. Perhaps someone else can advise you on this. Finally, to run programs at specific times, be they shell commands or otherwise you can use the cron facility. Cron's man page is quite well written and has a number of illustrative examples. Hope this helps. Best regards, Nick -- Nick Busigin [EMAIL PROTECTED] To obtain my pgp public key, email me with the subject: "get pgp-key" -- -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word "unsubscribe" to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED]
Re: Bug or feature?
> If I understand correctly, as soon as I set the 'user' flag in /etc/fstab, > any user can mount/unmount a partition. But - why can't I use 'mount > -atnfs' when all nfs partitions have the user flag set (error message is > 'mount: only root can do that')? > > Benedikt I suspect you also need mount permission from the nfs hosts. Otherwise, IMHO, this would be a very handy trick to hack into other perople's computers ?? H.C.Lai -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word "unsubscribe" to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED]
Iomega Drives
> "sivart" == sivart <[EMAIL PROTECTED]> writes: sivart> Hello all. Do any of the latest versions of linux contain sivart> drivers for IOMEGA ZIP drives??? Sure would be nice to use sivart> 100MB portable storage with linux... I have one, it works great! I've got a parallel port zip drive. When you configure the kernel, with 'make xconfig', select for scsi support, scsi disk support and then under 'low level drivers', select to build the IOMEGA driver. I build all of those as modules, rather than building them into the kernel, and run 'kerneld' at boot, by putting "auto" into /etc/modules. (this is Debian specific.) I found that the Zip driver, "ppa", works BEST as a module, since when the module is loaded, either automagicly by 'kerneld' and 'modprobe', or with an explicit 'insmod', it initializes the device. If it is compiled in, that only happens at boot time, and if the drive is not plugged into the port, you cannot mount one later without rebooting the computer. Whereas, as a module, 'ppa' will initialize on loading the driver, allowing you to move the drive around at will. In order to make it so that 'kerneld' can auto-load 'ppa', I had to hand-edit /lib/modules/`uname -r'/modules.dep, after running 'depmod -a' one time. (being certain that it is not run every boot.) For some reason, a required dependancy was not added by the 'depmod' program. The 'ppa.o' driver depends on 'sd_mod.o' and 'scsi_mod.o' These are the relevant changes from the modules.dep file, YMMV: 8<->8 161c161 < /lib/modules/2.0.27/scsi/ppa.o: /lib/modules/2.0.27/scsi/scsi_mod.o --- > /lib/modules/2.0.27/scsi/ppa.o: /lib/modules/2.0.27/scsi/sd_mod.o 8<->8 (... The line for sd_mod already contains the depend for scsi_mod, and so it ref-cascades just fine.) ... And then, in '/etc/fstab', I have these lines: 8<->8 /dev/sda1 /zip ext2 defaults,noauto,user,exec 0 1 /dev/sda4 /dzip msdos defaults,noauto,user 0 0 8<->8 ... Since Zip disks come from the factory partitioned as partition 4, I assume here that DOS disks will retain that. When I make a disk into a Linux disk, I run 'fdisk' on it and make the partition be partition 1, then run 'mke2fs' on it. (Someone mentioned that it is possible to use '/dev/sda', using the "whole device", without partitioning it... I've never tried that.) Now all you do is plug the drive into the parallel port, put the disk in the drive, and then 'mount' it. Use 'mount /zip' for a Linux disk, 'mount /dzip' for a DOS one. You could add mount points for "vfat" (Win95) filesystems, and whatever else as well. While the disk's filesystem is mounted, the eject button is disabled, and pressing it has no effect. When you 'umount' the disk, it gets ejected if you have pressed the button, but not if you haven't. After you've umounted the disk, 'kerneld' will unload the driver for you after a minute of disuse. There is a program, called 'ziptool', that allows you to set the write protection on a Zip disk. It will also eject a disk under software control. When you change the write protect, it ejects the disk in order to make sure that things will get re-initialized by the drive before you remount its filesystem. I seem to remember that I found 'ziptool' on Sunsite, but if you cannot find it, let me know, and I'll mail or 'sendfile' you a copy of it; I don't think the author will mind that. It's documantation is the comment header; it is 'c' source, and compiles in a jiff. -- __ _Karl M. Hegbloom <[EMAIL PROTECTED]> / /(_)_ __ _ ___ __ http://www.inetarena.com/~karlheg / / | | '_ \| | | \ \/ / Portland, OR, USA / /__| | | | | |_| |> < Proudly running Linux 2.0.27 transname \/_|_| |_|\__,_/_/\_\ and Debian GNU public software! -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word "unsubscribe" to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED]
LD_PRELOAD still solves tar -M problems
Hi, didn't thought that my private mail archive will serve me that much, as I recorded a message from Marek Duszynski at 07 Dec 1996. It is still valid, depart from [TAB] doesnt work, it must be: "LD_PRELOAD=/usr/lib/libgnumalloc.so.5" (still can't lprm from normal user because of permission denied, alias suid-problems!) Thanks, Marek! Andreas. : Date: Sat, 07 Dec 1996 20:28:52 + : From: Marek Duszynski <[EMAIL PROTECTED]> : Subject: tar multivolume archive - problem solved : : Hi, : : Thanks to all who responded to my problem of not being able : to use the tar command with the 'M' option for multivolume : archive. : : The machine was capable of creating such archives and then : reading its contents, but every attempt of extracting it : with the command: : tar xvMf /dev/fd0 : ended up in segfault message. : : Things are sorted now and all the credit goes to Arrigo. : If anybody experiences the same problem on Debian1.1, the : solution is to define the environment variable LP_PRELOAD: : LD_PRELOAD=/usr/lib/libgnumalloc[TAB]; export LD_PRELOAD : The [TAB] will give you your system version of libgnumalloc. : : I would like to know one thing, though. Should this be set in the : 1.2 release of Debian, or was it set already and my installation of : Debian 1.1 was not quite up to scratch :-) ? : : -- : : : Marek -- Uni Wuppertal, FB Elektrotechnik, Tel/Fax: (0202) 439 - 3009 Dr. Andreas Wehler; [EMAIL PROTECTED] -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word "unsubscribe" to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED]
Re: Taking down PPP at a certain hour of the day
> I am wondering.. how I can do this.. I would like to create a script that > run in the background and check the time.. if it is 8am it do a cron provides the ability to run specified programs at a certain time. See the man page for more details. You should be able to write a small script that just does the things you want if you just want to fake a shutdown. If you want to run a real shutdown, just run shutdown. Ciao, Thomas -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word "unsubscribe" to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED]
dpkg --configure texbin ... fails
My original Debian installation was done from the InfoMagic CD-ROM from Sept 1996. To update to "Debian 1.2" I have downloaded the *.deb packages a week ago, and today have run the "dselect" to upgrade all packages. The most of the packages updated cleanly (I was really impressed !!). Only a few got some DPKG errors. It seems, that most (if not all) problems were caused by failure of the "texbin" update. Below I add the "dpkg --configure texbin" messages: Any hints what to do ? TIA, Zenon ... # dpkg --configure texbin Setting up texbin (3.1415-5) ... kpathsea: Running MakeTeXTFM manfnt.tfm Running MakeTeXPK manfnt.tfm mf \mode:=nullmode; mag:=1; scrollmode; input manfnt \
Taking down PPP at a certain hour of the day
I am wondering.. how I can do this.. I would like to create a script that run in the background and check the time.. if it is 8am it do a wall announcement saying the system is going down, kill all the incoming telnet connections, and disconnect my PPP connection.. if it's not then it will just keep on running until is.. Anywhere on how I would attempt to do this.. I don't know how to shell program so I wouldn't know where to start.. Thanks Matthew - Matthew Stone, 23 Roosevelt Drive, Richmond Hill, Ontario, Canada L4C6V1 Pager (416)339-9052 -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word "unsubscribe" to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED]
ftp://master.debian.org/debian is missing.
Shouldn't there be a symlink there? I think the Mirrors file gives the URL as ftp://master.debian.org/debian and not the true one deeper into the directory tree. -- __ _Karl M. Hegbloom <[EMAIL PROTECTED]> / /(_)_ __ _ ___ __ http://www.inetarena.com/~karlheg / / | | '_ \| | | \ \/ / Portland, OR, USA / /__| | | | | |_| |> < Proudly running Linux 2.0.27 transname \/_|_| |_|\__,_/_/\_\ and Debian GNU public software! -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word "unsubscribe" to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED]
Re: ppp install problem
>In article <[EMAIL PROTECTED]>, Lawrence Chim wrote: >>I have problem to upgrade PPP and it seems that it is a chown problem. >>Any helps? >> >>Preparing to replace ppp 2.2.0f-2 (using ppp_2.2.0f-19.deb) ... >>Unpacking replacement ppp ... >>Setting up ppp (2.2.0f-19) ... >> [snip] >>Installing new version of config file /etc/ppp/pap-secrets ... >>chown: root.dip: invalid group > >For some reason there is no 'dip' group in your system. Add an entry by >appending the line 'dip:*:30:' to '/etc/group'. > >Make sure that group id 30 is not used by other group (search for ':30:' in >the file) in such case change it to any non used group id number you like. > >Once this is done, you can safely reconfigure the ppp package. Oups forgot something important. Actually you should start a new root login session when reconfigurating the package, otherwise you might think that nothing changes :) (i.e. logout, then login) >As root execute: > >dpkg --configure ppp -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word "unsubscribe" to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED]
Tips for updating from Red Hat?
I've been using Red Hat for the year or so, and I'd like to switch to Debian. Can anyone who has made this switch tell me what problems/changes I have to watch out for? Thanks. Scott -- Scott Hanson <[EMAIL PROTECTED]> Jesteburg-Lüllau, Germany work: Inter-Research Science Publisher <[EMAIL PROTECTED]> http://www.int-res.com -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word "unsubscribe" to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED]
Re: ppp install problem
In article <[EMAIL PROTECTED]>, Lawrence Chim wrote: >I have problem to upgrade PPP and it seems that it is a chown problem. >Any helps? > >Preparing to replace ppp 2.2.0f-2 (using ppp_2.2.0f-19.deb) ... >Unpacking replacement ppp ... >Setting up ppp (2.2.0f-19) ... > [snip] >Installing new version of config file /etc/ppp/pap-secrets ... >chown: root.dip: invalid group For some reason there is no 'dip' group in your system. Add an entry by appending the line 'dip:*:30:' to '/etc/group'. Make sure that group id 30 is not used by other group (search for ':30:' in the file) in such case change it to any non used group id number you like. Once this is done, you can safely reconfigure the ppp package. As root execute: dpkg --configure ppp -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word "unsubscribe" to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED]
Re: just starting
> I am a rank beginner and I know this question will display ignorance > (perhaps terminal) but --- > Should I forget the whole program and > stay fat, dumb and happy in WIN95? > > Thanks, ([EMAIL PROTECTED]) >From someone who's been there, prolly still is there, and will be there for some time. ;-) Stick with Linux. It is well worth the effort. The battle is much easier if you stick with Debian, but feel free to check out what others say about all of the distributions. I credit Win95 for making it easy enough for this newbie to get totally addicted. I thank Linux for keeping me motivated, and willing to learn more. Jan '96 -PC with win95 enters home oh, I had a vic20 when I was a teen, but that is about the extent of my computer background. I read like a bastard for about a year and a half before I got my PC, so I wasn't totally clueless, but close. April '96- totally addicted to my computer and the Internet, enter Linux. Still reading everything I can get my hands on, but its tough to make up for years of not learning. ;-) ~Oct'96-enter Debian. ahhh, it sure makes life easier, now I can actually use my Linux system instead of trying to configure every damn little thing. I don't have much disk space allocated, but I've got XWin, Apache, Smail, Netscape, Emacs, some games(quake!), learning some perl, msql, and so much more than I'd bore you with here. :-) Theres nothing like watching the output from tcpdump in one window, while you work on a webpage with your local server in another window, while you upgrade your version of smail, while you test those new perl scripts, continue until you choke your machine. Point being, this has lead me change my career ambitions. It may not do the same for you, but you're sure to learn a hell of alot, have a little fun, take some pride in something not everyone can do, and meet some damn fine people via these lists and the Linux community at large. Stick with it! Best Wishes, and many thanks to everyone on the Debian Development team for a fine job! (lookin' for the job of his dreams) Rich M [EMAIL PROTECTED] as for your trouble with making disks. Under win 95 Get rawrite, get the floppy images on your HD. drag and drop the images onto rawrite with a floppy in the drive, it should prompt you which disk you want to write to, a or b whichever applies. Hope that helps. -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word "unsubscribe" to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED]
Re: printing pdf files
Hello Lars! On 21 Dec 1996, lars wrote: > I know how to view pdf files now, and like to use gs... which happens > to be the main filter in apsfilter... however, when I try to print pdf > files with something like "lpr .pdf" i get an error > message. I want to modify apsfilter to recognize PDF files, but can't > figure it out. Can someone please point me in the right direction? > Thanks I noticed a few days ago that there is a program called "ps2pdf" on my hard drive. I just did a quick check, and lo and behold, there is a "pdf2ps" too! :) I guess they both come with your GhostScript Debian package. I am using Aladdin GhostScript 4.03, so I don't know whether the GNU GhostScript v3.x has it yet or not. But anyway, to print your .pdf file, I think you must first run the "pdf2ps" utility (there is a manpage for it too), and then you can print the resulting .ps file. Of course, if you know the PDF file format, you may add your own filter for your apsfilter to do pdf2ps conversion automatically, so you can print your .pdf files without having to convert the files manually. :) (Sorry, I haven't typed e-mail for a while, so I know I was really wordy and I kept repeating myself. :) Merry Christmas! :) Anthony -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word "unsubscribe" to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED]
Re: just starting
Thank you all for the help regarding downloading and creating installation disks. I now have the full set and will attempt to install as soon as I can work up the courage to repartition my hard drive. Has anyone tried to set up Debian on an Iomega Jaz drive? Gary Gifford ([EMAIL PROTECTED]) -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word "unsubscribe" to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED]
ppp install problem
I have problem to upgrade PPP and it seems that it is a chown problem. Any helps? Preparing to replace ppp 2.2.0f-2 (using ppp_2.2.0f-19.deb) ... Unpacking replacement ppp ... Setting up ppp (2.2.0f-19) ... Installing new version of config file /etc/init.d/ppp ... Configuration file `/etc/ppp/pap-secrets' ==> File on system created by you or by a script. ==> File also in package provided by package maintainer. What would you like to do about it ? Your options are: Y or I : install the package maintainer's version N or O : keep your currently-installed version Z : background this process to examine the situation The default action is to keep your current version. *** pap-secrets (Y/I/N/O/Z) [default=N] ? Y Installing new version of config file /etc/ppp/pap-secrets ... chown: root.dip: invalid group dpkg: error processing ppp (--install): subprocess post-installation script returned error exit status 1 Errors were encountered while processing: ppp -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word "unsubscribe" to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED]
Kernel source won't build
I'm having trouble with the kernel source. I put kernel-source 2.0.27 on the other day, and I get this output when I try to do a make mainly "make dep" to start things off... Here's what happens: dst:/usr/src/linux# make dep gcc -I/usr/src/kernel-source-2.0.27/include -O2 -fomit-frame-pointer -o scripts/mkdep scripts/mkdep.c scripts/mkdep.c: In function `main': scripts/mkdep.c:267: `patth_array' undeclared (first use this function) scripts/mkdep.c:267: (Each undeclared identifier is reported only once scripts/mkdep.c:267: for each function it appears in.) scripts/mkdep.c:267: parse error before `;' scripts/mkdep.c:273: `name' undeclared (first use this function) scripts/mkdep.c: At top level: scripts/mkdep.c:290: parse error before `return' make: *** [scripts/mkdep] Error 1 dst:/usr/src/linux# what happened? ack! -- Daniel Stringfield mailto:[EMAIL PROTECTED] http://www.jax-inter.net/users/servo Send email for more information on the Jacksonville Linux Users Group! -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word "unsubscribe" to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED]
Can't see modem on /dev/cua1
This is not a Linux/Debian specific problem but I don't know whom else to ask so I beg the list's indulgence. I recently replaced my motherboard with a Tyan Tomcat III with one Pentium 133 (getting ready for SMP). The board has two on-board 16550A based serial ports. Everything appears to working fine except that I can't access the modem. The serial module is being loaded and setserial sees the ports with correct IRQs and port addresses. However, when I use minicom I see nothing, ATZ does nothing and the modem lights don't come on. I have also tried /dev/ttyS1 and various baud settings to no avail. On the other hand Windows95 can detect the modem correctly on COM2 and I can see the modem lights blinking. Hence I don't think it is a hardware or cabling problem. But neither the hyperterm "application" nor Telix can initialise COM2. The mouse (on COM1) works fine under Windows95, I haven't tried it under X yet. Thanks, Evan. -- Evan Thomas Department of Physiology University of Melbourne Parkville, 3052 ph: 9344-5849 fax: 9344-5818 -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word "unsubscribe" to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED]
Re:Continued Debian Boot Problems
I'm still confused. If I understand correctly, you are booting Debian 1.2 with a Debian 1.1 boot floppy, and it's hanging just after configuring the WD driver. The problem with this is that it's not necessarily the WD driver, but the driver after that one. I should have the new boot floppy set without the cm206 uploaded late this evening. Let's try with that disk, so that we can both be on the same page. I don't have the 2.0.6 kernel source hanging around any longer, so it's difficult for me to debug a problem with the 1.1 boot floppy. Thanks Bruce -- Bruce Perens K6BP [EMAIL PROTECTED] Finger [EMAIL PROTECTED] for PGP public key. PGP fingerprint = 88 6A 15 D0 65 D4 A3 A6 1F 89 6A 76 95 24 87 B3 -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word "unsubscribe" to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED]
Re: InfoMagic Debian 1.2 Installation
Did we go through this already? Try putting "ppp" in /etc/modules. It seems it gets loaded too slowly if kerneld loads it. Thanks Bruce -- Bruce Perens K6BP [EMAIL PROTECTED] Finger [EMAIL PROTECTED] for PGP public key. PGP fingerprint = 88 6A 15 D0 65 D4 A3 A6 1F 89 6A 76 95 24 87 B3 -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word "unsubscribe" to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED]
