How to choose safe configuration at boot time?
I am unable to use my Debian Linux system after the major update I did
on Tuesday.
WHAT HAPPENED: I used dselect in an xterm window under XFree to update the
system after an idle period of 1 week. I update from stable, unstable,
contrib, and nonfree. I allowed one of the installers to shut down xdm,
which led to a reboot. I finished the installation by starting dselect again
from an Xless console. The next reboot led to a login prompt on a non-X
console screen, flashing and ignoring all input, except for CTL-ALT-DEL
which led to another reboot ending in the same state.
WHAT I THINK IS WRONG: I think that my configuration of xdm is broken. The
last two outputs from the boot process, before the login prompt and the
flashing, are an indication that xdm is starting, and an indication that the
nas, which started earlier, has failed. The nas problem has been with me for
weeks, presumably has to do with a misconfigured SoundBlaster, and has never
before affected my ability to run X silently. Besides the install script
that shut down xdm for some change, I accidentally allowed another install
script to change the default X server (I had intended to experiment with
several X servers before deciding whether to change the default). One way or
another, I suspect that xdm is looping infinitely on a failed attempt to
start up X. I experienced similar behavior in an earlier incarnation, when X
failed due to lack of the right mouse driver, but in that case I saw some
failed attempts at driving the video in color, and now I see only on/off
flashing.
WHAT I THINK I NEED: I think that I need to boot the system in single-user
mode, or otherwise avoid starting up xdm, so that I can seek out the bad
configuration files and fix them. I have *only* Debian Linux on the system,
and I start it with LILO. I have been able to run several versions of the
kernel from LILO, including a very primitive one which I keep on a diskette
for emergencies. They all lead to the same behavior. Probably, there is an
appropriate option to give LILO, and it's probably mentioned in the online
documentation which I am now unable to read :^{ (I hope that this emoticon
portrays embarassment). In the past, I have used CTL-ALT-F1 to bring up a
single-user state, but I have been getting no useful response to this
signal, although it resets the phase of the flashing, so something is
evidently reading and ignoring it.
I'll be grateful for all suggestions. Useful ones will help me, and useless
ones will at least distract me from despair for a while :^)
Mike O'Donnell
[EMAIL PROTECTED]
http://www.cs.uiowa.edu/~odonnell
--
TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to
[EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED]
Re: hard disk recommendation
On Wed, 20 Nov 1996, Marc A. Volovic wrote: On Tue, 19 Nov 1996, Paul Christenson wrote: In an attempt to avoid a war, I'll simply say that Silicon Graphics is using IBM SCSI drives exclusively. Given the nature of their products, I I am afraid you're wrong. I saw Seagate drives (in older stations) and, more recently, Quantum drives in High Impacts. SGI used to use Seagate exclusively (up to a year or so ago), but then they started getting customer complaints on the drives failing. They did go to Quantum for a short period, but they found that the Quantum drives have a nasty habit of recalibrating themselves from time to time, which makes the drives unavailable for a second or so, which really causes a performance hit on these machines. Quantum was unwilling to rewrite the firmware on their drives, so testing continued. IBM drives are the only ones that currently meet SGI specs. This was from a conversation with an SGI tech about three weeks ago. -- | This is OFFICIAL WRITTEN notification that I want to be REMOVED | | from ALL commercial mailing lists. EVERY message sent from this | | account has had this request posted. ALL UNSOLICITED ADVERTISEMENTS | | SENT TO THIS ACCOUNT ARE IN VIOLATION OF FEDERAL (U.S.) LAW.| -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED]
Re: Invitation for testers of frozen distribution
On Wed, 20 Nov 1996 20:38:48 +1100 Lawrence Chim ([EMAIL PROTECTED] t.edu.au) wrote: Is the 2GB partition install problem solved yet? Not yet as this is part of the boot floppies which haven't been updated yet. Phil. -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED]
Re: X is painful
On Tue, 19 Nov 1996, Karl M. Hegbloom wrote: Isn't X the GUI that started it all though? And the codes are there for all to see; so perhaps it served as the reference from which MSWindows, MAC and the others have evolved? The guys who wrote MAC and Windows must have studied X-Windows in college, right? Nah, Xerox at PARC invented the GUI, they actually made a computer that used it, but it was really bad. Both Steve Jobs (Mac guy) and Bill Gates visited PARC and saw their GUI. They both saw that this was where the computer interface should head. The rest is history. Bill And that environment is supposed to justify the complete Bill piece of crap that resulted? You're complaining about hand-me-downs. I'll bet there's some really neato in-house X stuff out there. Not just in-house stuff, CDE is pretty nice, and so is NEXTSTEP as has been mentioned. On a sidebar, has anyone noticed that all of Jobs's companies have had good UIs. Apple, Next, Shaya -- Shaya Potter [EMAIL PROTECTED] -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED]
Re: Newbie Question
On Wed, 20 Nov 1996 09:34:16 MST Wayne Richardson ([EMAIL PROTECTED] code3.com) wrote: I have just installed the Debian Linux system on a 386 machine for experimental purposes and have transferred and installed the gnuchess.deb package (I used dpkg --install gnuchess.deb because I am still trying to figure out what dselect does and how it works) . Now how do I run the darn thing. I have found the following files in the /usr/games directory: gnuchess gnuchessc gnuchessn gnuchessr gnuchessx What are all of these files and what do I do with them (Do I have to build an executable?). Obviously I come from a wintel background and am expecting a .exe file but there isn't one here (or is there?). The attributes show them to be -rwxr-xr-x (whatever that means) There is nothing such like .exe under unix/linux. The x in the `attributes' (called permissions here) just mean executable. Just try /usr/games/gnuchess, or if /usr/games is in your path, gnuchess. Is this typical with other packages? For example, I have also downloaded the ppp.deb package and installed it but where in the world did it put the executable files so that I can run it. (I am still trying to get the PPP to work so that I don't have to download from my Windows 95 machine and transfer everything over via diskette (very painful and slow and not feasible for the bigger packages). You can mount your win95/dos partitions to that they appear do be in linux's directory tree like this: mount -t msdos /dev/hda1 /mnt ^ ^ ^ DOS files will appear under there | | The name of your partition | Type of your partition (for win95 I don't remeber the name, but this should work for DOS drives). Though that it might look very difficult at the beginning, installing linux should be a rewarding experience. Good luck. Phil. -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED]
Re: Invitation for testers of frozen distribution
I put in a fix for the 2GB partition problem, but I haven't tested the boot disk. This morning I built a kernel without initrd, and wasted some time on that. I have a kernel with initrd building now (so that the root and boot disks are merged into one floppy) and will test it later. 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: Crack and cops
Pardon my ignorance but what exactly are crak and cops? Cops: security checker. Cops does some cute things. First off, it checks for some obvious things like, say, your /var/spool/cron/crontabs dir being world-writable or your hosts.equiv file being world writable, etc It's got one really *cute* feature called kuwang, I think. Basically, it's supposed to find ways that a user can gain root access through a *process*. For example, let's we've got three users on the system: A, B, and root. Let's also say that A's primary group is X but it's also in Z. B's primary group is Z and is also in the root group. Further, let us assume that B was careless enough to turn on group write permissions for his/her .profile. So, we've got something like this: % ls -l /home/B/.profile -rwxrwxr-x BZ1534 Jan 17 12:34 .profile And let us assume the same of root: % ls -l /root/.profile -rwxrwxr-x root root 2543 Feb 23 16:32.profile Well, now, it's possible for user A to gain root privledges. A will be able to write to Bs .profile and, hence, will be able to run anything as B. This means that A (while running something as B) will be able to write to roots .profile and will be able to run anything as root. I know this seems preposterous... like you need this impossible conspiracy of little misconfigurations to allow for a security hole of this nature... but it's really not that impossible. Imagine, for example, if you put a certain user in the www group to allow them to maintain a portion of your web site. Also imagine that you've added www to the root group so that certain CGI scripts will be able access some files that www doesn't normally have access to. Well, now you're more than half way there... and you got there by doing two things that, in themselves, didn't seem as all that unreasonable. So, to keep a long story from getting any longer, that is what kuwang is supposed to do. I'm not sure if it really *does*, since it's never found a hole like that on my machine yet. - Joe -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED]
Re: Mgetty
Hi, I'm having some trouble getting mgetty to work in the inittab file, if i do mgetty 38400 ttyS1 it works fine but i have to manually reset it everytime a user calls. What line do i need to add to my inittab for it to automatically do it for me as i keep getting it wrong. S1:23:respawn:/sbin/mgetty ttyS1 Notice the respawn, which restarts the process when it dies. Getty's are one-shot deals. They listen, and then they spawn a login and then they die. Now, an entry like this (athough commented out) should have been placed in your inittab when you installed the mgetty package. - Joe -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED]
Problem with filter
Is anyone aware of permission/ownership problems with /usr/bin/filter? I was trying to use the filter command that comes with the elm package and I ran across a problem. When instructing filter to save to a file or to do anything other than placing a copy in my mailbox, it does fine. When it tries to add to my mailbox, however, it fails and put the stuff in $HOME/EMERGENCY_MBOX. I tracked this down to the fact the the perms/ownerships on filter were: -rwxr-xr-x root root What needs to be changed is that filter needs to be owned by the mail *group* and then the setgid bit needs to be set, like so: -rwxr-sr-x root mail This makes filter run just swell. Has this problem already been fixed? - Joe -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED]
How can I dial out on ttyS?
I'm trying to get an e-mail-alpha-pager thing working. Everything is working except the fact that I don't have permissions to use /dev/ttyS1 When I send a page as root, the page goes out fine. When I send mail to, say, jdoe, I have /usr/bin/filter pick it up and execute the pager script, and it fails claiming it doesn't have permission. So, I added jdoe to the uucp group (which is the group owner of the ttyS's). Now, when I su to jdoe and send a page, the page goes through. When I send e-mail to jdoe, it still fails with a permission thing. I'm suspecting that it's some sort of GID vs. EGID thing. Perhaps, when I su jdoe, I'm keeping some of my root permissions... or, when filter runs, it's keeping some of it's permissions from mail or something. Ideas? -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED]
Re: Mgetty
On Wed, 20 Nov 1996, Fingers wrote: I'm having some trouble getting mgetty to work in the inittab file, if i do mgetty 38400 ttyS1 it works fine but i have to manually reset it everytime a user calls. What line do i need to add to my inittab for it to automatically do it for me as i keep getting it wrong. It sounds like you are using the incorrect runlevels in your inittab. Debian runs at level 2 (I believe). If you installed the mgetty package, then the sample should be correct. The mgetty line I'm using: s0:23:respawn:/sbin/mgetty ttyC0 (I'm using a modem attached to a Cyclades port, of course...) | This is OFFICIAL WRITTEN notification that I want to be REMOVED | | from ALL commercial mailing lists. EVERY message sent from this | | account has had this request posted. ALL UNSOLICITED ADVERTISEMENTS | | SENT TO THIS ACCOUNT ARE IN VIOLATION OF FEDERAL (U.S.) LAW.| -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED]
Re: Invitation for testers of frozen distribution
On Wed, 20 Nov 1996, Lawrence Chim wrote: Is the 2GB partition install problem solved yet? No, 'fraid not. This problem only exists on the install disks though - you can manipulate 2GB partitions fine once you've installed the base system. Hopefully this will be fixed before 1.2 - it's on our list of Release Critical Bugs! Cheers, Michael -- Michael Nonweiler [EMAIL PROTECTED]Trinity College, Cambridge, UK. http://www.cl.cam.ac.uk/users/mrn20/ -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED]
Re: syslog daemon is dying
On Wed, 20 Nov 1996, Branden Robinson wrote: I have a similar problem (but I can eventually get syslogd and klogd to both come up by ./etc/init.d/sysklogd restart over and over again). I did a test -e for every file mentioned in /etc/syslog.conf and got a 0 status every time? Does anyone have any other suggestions? Yeppers; run strace. Mine dies when it tries to do gethostname on the local machine, which almost makes sense, since it is on a location on the network different from its destination, and the DNS server is unreachable right now. But then again, why is it not getting the localhost name from /etc/hosts? I have named set up in forwarder mode, (order host, bind). $0.02 __ Todd Graham Lewis Linux! Core Engineering Mindspring Enterprises [EMAIL PROTECTED] (800) 719 4664, x2804 -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED]
Serial port IRQ strangeness
I have a Debian 1.1 machine with 4 serial ports, each on its own IRQ. I edited /etc/rc.boot/0setserial to configure the appropriate IRQs. Question #1: At boot, when /etc/rc.boot/0setserial runs, it complains no such device for each of the devices. If I do the setserial as /etc/rc2.d/S14setserial (after ), it works. Why? Question #2: I am running diald. Periodically, the following messages appear in /var/log/messages: kernel: Serial driver version 4.13 with no serial options enabled kernel: tty00 at 0x03f8 (irq = 4) is a 16550A kernel: tty01 at 0x02f8 (irq = 3) is a 16550A kernel: tty02 at 0x03e8 (irq = 4) is a 16550A kernel: tty03 at 0x02e8 (irq = 3) is a 16550A and the IRQs for cua2 and cua3 are changed from 2 and 5 back to 4 and 3! Any ideas? thanks, Steven Gaarder Network and Systems Administrator [EMAIL PROTECTED] C-MOLD, Ithaca, N.Y., USA -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED]
Re: Newbie Question
From: Wayne Richardson [EMAIL PROTECTED] I have very basic questions... Get an introduction to Unix at any good bookstore when you have a chance. It will help a lot. That system is pretty deep. I used dpkg --install gnuchess.deb because I am still trying to figure out what dselect does and how it works) . If you can FTP the file ftp://oak.scotch.wa.edu.au/pub/dselect.beginner.2.gz (use binary mode if your FTP has one), put it on your system and use zmore to look at it. The attributes show them to be -rwxr-xr-x (whatever that means) The x means executable. You can probably type /usr/games/gnuchess and get something to happen. Is this typical with other packages? For example, I have also downloaded the ppp.deb package and installed it but where in the world did it put the executable files so that I can run it. For any command, type man command-name, info command-name, and go look in /usr/doc/command-name . zmore will view those .gz files. If man command-name doesn't tell you anything, try man -k command-name to see if that suggests the name of an appropriate man page. PPP put configuration files in /etc/ppp, and man pppd (that's the PPP server, or daemon) will tell you about them. 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: Got unusable memory address 0x400cd008 (and libtermcap)
From: [EMAIL PROTECTED] [EMAIL PROTECTED] Would there be in simple reason, why the binary works in RedHat, but not in Debian? If you have upgraded libc to 5.4.7, try this: LD_PRELOAD=/usr/lib/libgnumalloc.so.5 export LD_PRELOAD or if you use csh or tcsh: setenv LD_PRELOAD /usr/lib/libgnumalloc.so.5 If that fixes the problem, there's a malloc bug in the program and the new allocator we're using makes the bug show up instead of being silent. 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]
Win NT OS loader
Go to http://www.owlnet.rice.edu/~sedmison/diretboot.html and read how to do it. I followed the instructions and got it to work with Debian, NT and 95. No problems so far. Bruce Jackson
Re: X is painful
On Wed, 13 Nov 1996, Bill Bumgarner wrote: Maybe I'm just spoiled by years of NEXTSTEP-- but, damnit, NEXTSTEP really is the most well-inntegrated user inrface *ever* built. Seriously. i suppose that's a matter of personal preference. I have to work with nextstep machines and can't stand it. it's bloated and it's horribly slow and i can't customise it anywhere near as much as i can X fvwm (and fvwm derivatives). I especially don't like the dock stealing an inch or so down the entire right-hand side of my screen. I much prefer clicking on the root window and getting a menu up: left-click for a launch menu, middle-click for window operations, and right-click for list of open windows. Functionality when i need it *WITHOUT* cluttering up my screen with crap i dont need all the time. I also intensely dislike next-style menus. I want menus inside the application window, not cluttering up the top left hand corner of my screen. I am also far from impressed by nextstep's lack of stability - damnit, unix boxes are *supposed to* stay up for months, not crash randomly for no apparent reason. It's also extremely fussy about what hardware it will run on - a big problem when motherboard models change every few months and it's almost impossible now to get the old pentium boards which nextstep will run on...newer HX VX boards either crash randomly or dont work at all. the latest compatibility problem is the adaptec 2940 driver. cant get the old 2940 cards any more...can only get 2940-AU cards. The driver notes for NS say that it works on 2940-AU cards. In practice it doesn't work at all...fails at install time with a message about can't get config space. (note, i've only really worked with intel version of nextstep, so can make no comment on stability of versions for other architectures except that black box next stations are slow by today's standards) All in all, my opinion of nextstep (as a user interface AND as a unix) can be accurately summed up as it was fabulous for 1988however it's 1996 now. Another summary: i'd rather use linux. A good user interface should be layered on top of a stable, reliable operating systemit shouldn't become an excuse for not creating that stability. Functionality stability come first. Prettiness second. I guess you like what you get used to. Craig -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED]
Where is Netscape package?
I have been looking for the netscape package under rex but can't find it. Where is it? Thanks, Mark Phillips. ([EMAIL PROTECTED]) -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED]
Re: csh * expantion problem
If you really want to echo an asterisk, use: echo '*' If you want to echo the list of files in the current directory, use: echo * If the current directory is empty, you will get the no match message. Is that what is happening? 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]
cron problem
I don't know why cron always send mail to root with the following body. Any idea? Subject: Cron [EMAIL PROTECTED] run-parts /etc/cron.daily X-Cron-Env: SHELL=/bin/sh X-Cron-Env: PATH=/usr/local/sbin:/usr/local/bin:/sbin:/bin:/usr/sbin:/usr/bin X-Cron-Env: HOME=/root X-Cron-Env: LOGNAME=root getopt: illegal option -- o lawrence, -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED]
Re: Emacs and tcsh
Oh, now I have a guess at what is going wrong. Invoke unset noglob and see if the csh problem gets better. Then you'll have to figure out what is setting it. It is generally set while running tset because the termcap strings sometimes contain shell metacharacters. 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: Dselect docs for beginners
On Wed, 20 Nov 1996, Ian Jackson wrote: I don't know whether it's really necessary to advise people not to select any large packages at this stage. Everything ought to work reasonably well if they just select what they want. Unfortunately, it doesn't. I took a 1.1.4 CD release, followed the instructions through first boot/password/adduser/dselect. With my Slackware mindset, I selected everything that I thought would be useful, not knowing how easy it is to add later. Halfway through the install, packages started failing. Required packages didn't install. Needless to say, it was a mess. I reformatted, and started over. I only installed the preselected stuff, and it went well. Re-ran dselect, selected all the stuff from before, and it went smoothly. I guess that there was a pre-dependency problem involved. | This is OFFICIAL WRITTEN notification that I want to be REMOVED | | from ALL commercial mailing lists. EVERY message sent from this | | account has had this request posted. ALL UNSOLICITED ADVERTISEMENTS | | SENT TO THIS ACCOUNT ARE IN VIOLATION OF FEDERAL (U.S.) LAW.| -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED]
Re: How to choose safe configuration at boot time?
I am unable to use my Debian Linux system after the major update I did
on Tuesday.
WHAT HAPPENED: I used dselect in an xterm window under XFree to update the
system after an idle period of 1 week. I update from stable, unstable,
contrib, and nonfree. I allowed one of the installers to shut down xdm,
which led to a reboot. I finished the installation by starting dselect again
from an Xless console. The next reboot led to a login prompt on a non-X
console screen, flashing and ignoring all input, except for CTL-ALT-DEL
which led to another reboot ending in the same state.
WHAT I THINK IS WRONG: I think that my configuration of xdm is broken. The
last two outputs from the boot process, before the login prompt and the
flashing, are an indication that xdm is starting, and an indication that the
nas, which started earlier, has failed. The nas problem has been with me for
weeks, presumably has to do with a misconfigured SoundBlaster, and has never
before affected my ability to run X silently. Besides the install script
that shut down xdm for some change, I accidentally allowed another install
script to change the default X server (I had intended to experiment with
several X servers before deciding whether to change the default). One way or
another, I suspect that xdm is looping infinitely on a failed attempt to
start up X. I experienced similar behavior in an earlier incarnation, when X
failed due to lack of the right mouse driver, but in that case I saw some
failed attempts at driving the video in color, and now I see only on/off
flashing.
WHAT I THINK I NEED: I think that I need to boot the system in single-user
mode, or otherwise avoid starting up xdm, so that I can seek out the bad
configuration files and fix them. I have *only* Debian Linux on the system,
and I start it with LILO. I have been able to run several versions of the
kernel from LILO, including a very primitive one which I keep on a diskette
for emergencies. They all lead to the same behavior. Probably, there is an
appropriate option to give LILO, and it's probably mentioned in the online
documentation which I am now unable to read :^{ (I hope that this emoticon
portrays embarassment). In the past, I have used CTL-ALT-F1 to bring up a
single-user state, but I have been getting no useful response to this
signal, although it resets the phase of the flashing, so something is
evidently reading and ignoring it.
I'll be grateful for all suggestions. Useful ones will help me, and useless
ones will at least distract me from despair for a while :^)
Mike O'Donnell
[EMAIL PROTECTED]
http://www.cs.uiowa.edu/~odonnell
--
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: Laptop and PCMCIA
I'm trying slowy to get my whole office over to Debian g, and the next machine I'm going to tackle is the TI-Extensa 560CD laptop. I installed it on an Extensa 570CDT (except for X); no problems with the base installation. The CD-ROM is a standard ATAPI on /dev/hdb. The fact that there is a reboot involved between the last floppy disk and the first CD usage made it quite simple (as the CD and FDD share one bay). The fun started when I tried to get networking started. The PCMCIA package installed just fine, except that Debian tries to set up networking BEFORE Card Services is installed. I moved Card Services into /etc/init.d/boot before the network is initialized, and added a 'sleep' command that was just long enough to allow the Ethernet card (in slot A) to initialize. I suppose that moving Card Services into the 'networking' script is the proper approach, but the bottom line was that it worked. | This is OFFICIAL WRITTEN notification that I want to be REMOVED | | from ALL commercial mailing lists. EVERY message sent from this | | account has had this request posted. ALL UNSOLICITED ADVERTISEMENTS | | SENT TO THIS ACCOUNT ARE IN VIOLATION OF FEDERAL (U.S.) LAW.| -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED]
Re: How to choose safe configuration at boot time?
On Wed, 20 Nov 1996, Mike O'Donnell wrote: WHAT I THINK I NEED: I think that I need to boot the system in single-user mode, or otherwise avoid starting up xdm, so that I can seek out the bad configuration files and fix them. While many users (self included) have some negative feelings about Slackware, there is one tool that is quite useful. Create the appropriate boot disk, and the rescue root disk. With them, you will have a reasonably complete set of tools for fixing problems such as these. | This is OFFICIAL WRITTEN notification that I want to be REMOVED | | from ALL commercial mailing lists. EVERY message sent from this | | account has had this request posted. ALL UNSOLICITED ADVERTISEMENTS | | SENT TO THIS ACCOUNT ARE IN VIOLATION OF FEDERAL (U.S.) LAW.| -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED]
Re: X is painful
On Wed, 20 Nov 1996, Shaya Potter wrote: On a sidebar, has anyone noticed that all of Jobs's companies have had good UIs. Apple, Next, But they still operate under the assumption that the user has three hands. Or at least a prehensile, well, never mind... | This is OFFICIAL WRITTEN notification that I want to be REMOVED | | from ALL commercial mailing lists. EVERY message sent from this | | account has had this request posted. ALL UNSOLICITED ADVERTISEMENTS | | SENT TO THIS ACCOUNT ARE IN VIOLATION OF FEDERAL (U.S.) LAW.| -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED]
Re: Floppy management?
Jean Orloff [EMAIL PROTECTED] writes: 1) How to prevent a user to prematurely eject a disk before sync is over? Mounting with the sync option is not enough: that's the way I do it, and still the fd light goes off before the file is truly written to disk (and umount or sync gets it on again). You can't. If we had Mac style drives that had to be asked to eject the disk, you could have some protection. This falls under the category of just don't do that. No choice, given the hardware. BTW, there's a program out there somewhere called DevAlloc that might interest you. I think the model is that you can let a group of users have the right to allocate some device for their exclusive use for some period of time (I think it's partially done with chown). So a user could do something like: allocate /dev/floppy do stuff with the floppy if this succeeds deallocate /dev/floppy I haven't actually looked at the program, so I can't say how it works, or even how well. I could imagine that something like this could be extremely useful now that things like Zip drives (or even cdroms) are really cheap. I actually had to write a pair of scripts that let a group of users (via sudo) exclusively mount and unmount a zip drive, and optionally remap users and groups on the zip disk on mount (and remember and unmap them on unmount) so that if you have different uids at home and in the lab, you still get the same results both places. (or perhaps there's some better mechanism we already have that I just don't know about.) -- Rob -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED]
Re: How to choose safe configuration at boot time?
I am unable to use my Debian Linux system after the major update I did
on Tuesday.
WHAT HAPPENED: I used dselect in an xterm window under XFree to update the
system after an idle period of 1 week. I update from stable, unstable,
contrib, and nonfree. I allowed one of the installers to shut down xdm,
which led to a reboot. I finished the installation by starting dselect again
from an Xless console. The next reboot led to a login prompt on a non-X
console screen, flashing and ignoring all input, except for CTL-ALT-DEL
which led to another reboot ending in the same state.
Something similar happened to me. I figured out what was wrong in my
case, which was similar.
In my case, because I knew that xbase required restarting X, I
intentionally stopped X before doing the install.
I also ran the new XF86Setup program, and configured it for my card, a
Cirrus 5426-based VLB video card. XF86Setup decided that probing my
dot clocks and adding a clocks line to the XF86Config file would be
a good thing. My video card doesn't need a clocks line, and in fact,
the X server won't run with it.
So at reboot, xdm starts up. It runs the X server, which finds the
problem with the configuration, and exits. So xdm starts it again.
It exits, and repeat.
I basically dealt with it by careful typing... (flash) r (blank)
(flash) o (blank) (flash) o (blank) (flash) t (blank) (flash) enter
(blank) (flash) and so on. It took me several attempts before I was
able to time -all- the needed keypresses in my password correctly
(that'll teach me to use a 11-character root password). But then I
was able to kill xdm, start X manually (and thus discover the
problem), and fix.
What happens is that while the VC is on the login prompt, it is
listening to the keyboard, but when X takes over, it switches the VC.
WHen I was able to finally kill it, the other VC contained all of my
mistimed keystrokes.
WHAT I THINK IS WRONG: I think that my configuration of xdm is broken. The
last two outputs from the boot process, before the login prompt and the
flashing, are an indication that xdm is starting, and an indication that the
nas, which started earlier, has failed.
It is probably true that it is a misconfigured X server. That is
exactly what I saw causing my similar problems.
WHAT I THINK I NEED: I think that I need to boot the system in single-user
mode, or otherwise avoid starting up xdm, so that I can seek out the bad
configuration files and fix them.
This will work, but I was also very surprised that when upgrading X on
my system, the install scripts decided that starting XDM for me was
good. Even though I was in single-user mode.
I have *only* Debian Linux on the system,
and I start it with LILO. I have been able to run several versions of the
kernel from LILO, including a very primitive one which I keep on a diskette
for emergencies. They all lead to the same behavior. Probably, there is an
appropriate option to give LILO, and it's probably mentioned in the online
documentation which I am now unable to read :^{ (I hope that this emoticon
portrays embarassment). In the past, I have used CTL-ALT-F1 to bring up a
single-user state, but I have been getting no useful response to this
signal, although it resets the phase of the flashing, so something is
evidently reading and ignoring it.
When starting with LILO, hit the shift key to get a boot: prompt, then
type linux emergency, which will start it in single user mode.
Or try timing your keystrokes, like I did. (My system has a long
uptime I didn't want to ruin even when upgrading).
I'll be grateful for all suggestions. Useful ones will help me, and useless
ones will at least distract me from despair for a while :^)
Mike O'Donnell
[EMAIL PROTECTED]
http://www.cs.uiowa.edu/~odonnell
--
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: How to choose safe configuration at boot time?
Mike O'Donnell [EMAIL PROTECTED] writes: WHAT I THINK IS WRONG: I think that my configuration of xdm is broken. OK, I think I can help as I got myself in a very similar situation a couple of times. This is probably not an xdm problem, but an X problem. Most likely, in the course of upgrading, you did something that left you with an X server that's not configured properly, or with xdm pointing to an invalid X server. So what you're seeing is xdm repeatedly launching an X server that fails, or trying to launch a non-existent server. The best way to avoid this in the future is to not reenable xdm (via rebooting) until you have tested that running the X that xdm points to in /etc/X11/xdm/Xservers succeeds after your upgrade xbase xdm, or other relevant X packages (Ctrl-Alt-Backspace will kill X for you if it does run). Now, how to solve your problem. You're running lilo, good. I think you can just hold down the shift key at boot (unfortunately on my machine I have to actually press it repeatedly just as lilo gets started; holding it down doesn't seem to work), and when you get the lilo: prompt, type linux single. This will boot linux in single user mode, without launching X (I hope), and you'll be able to check things out and fix the problem. Check that /etc/X11/xdm/Xservers points to a real server, and check that running this server directly works. The problem will hopefully be obvious at this point. Good luck, and I hope that's it. -- Rob -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED]
Re: Serial port IRQ strangeness
Hi, You are trusting that kerneld will load the module serial. However, this will happen just when kerneld is running, but it is just later in the boot sequence. Your system, just like mine, requires setserial for ttyS2 and ttyS3 to use the right interrupts. Compile a kernel with serial built in. Esa - Esa Turtiainen http://www.turtiainen.dna.fi [EMAIL PROTECTED] fax: +358-0-801 3008 On Wed, 20 Nov 1996, Steve Gaarder wrote: I have a Debian 1.1 machine with 4 serial ports, each on its own IRQ. I edited /etc/rc.boot/0setserial to configure the appropriate IRQs. Question #1: At boot, when /etc/rc.boot/0setserial runs, it complains no such device for each of the devices. If I do the setserial as /etc/rc2.d/S14setserial (after ), it works. Why? Question #2: I am running diald. Periodically, the following messages appear in /var/log/messages: kernel: Serial driver version 4.13 with no serial options enabled kernel: tty00 at 0x03f8 (irq = 4) is a 16550A kernel: tty01 at 0x02f8 (irq = 3) is a 16550A kernel: tty02 at 0x03e8 (irq = 4) is a 16550A kernel: tty03 at 0x02e8 (irq = 3) is a 16550A and the IRQs for cua2 and cua3 are changed from 2 and 5 back to 4 and 3! Any ideas? thanks, Steven Gaarder Network and Systems Administrator [EMAIL PROTECTED] C-MOLD, Ithaca, N.Y., USA -- 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: hard disk recommendation
Paul Christenson wrote: On Wed, 20 Nov 1996, Marc A. Volovic wrote: On Tue, 19 Nov 1996, Paul Christenson wrote: In an attempt to avoid a war, I'll simply say that Silicon Graphics is using IBM SCSI drives exclusively. Given the nature of their products, I I am afraid you're wrong. I saw Seagate drives (in older stations) and, more recently, Quantum drives in High Impacts. SGI used to use Seagate exclusively (up to a year or so ago), but then they started getting customer complaints on the drives failing. They did go to Quantum for a short period, but they found that the Quantum drives have a nasty habit of recalibrating themselves from time to time, which makes the drives unavailable for a second or so, which really causes a performance hit on these machines. Quantum was unwilling to rewrite the firmware on their drives, so testing continued. IBM drives are the only ones that currently meet SGI specs. This was from a conversation with an SGI tech about three weeks ago. Which Seagate/Quantum drive was using and which IBM drive is using by SGI? lawrence, -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED]
bo is full of symlinks to rex?
Prior to the freezing of rex, unstable (rex) contained all of the physical files of the distribution. Now, unstable is bo, and it has many links back to rex. At what point will bo loose the links to rex and be self-contained again? When the change occured, my mirror of unstable lost a bunch of files. Now I have to get both rex and bo. I think I can continually mirror unstable only, by careful use of the following option in mirror, but I haven't actually tried this yet: If the remote site contains symlinks that you want to flat- ten out into the corresponding files, then do this by changing the flags passed to the remote ls: flags_recursive+L or flags_nonrecursive+L First test this by trying a ls -lRatL on the remote site under the ftp command to check whether the remote file-store has any symlink loops. Anybody got a better method to mirror only the unstable tree efficiently? -- ...RickM... -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED]
Re: Win NT OS loader
At 23:35 19-11-96 -0500, Bruce Jackson wrote: Go to http://www.owlnet.rice.edu/~sedmison/diretboot.html and read how to do it. I followed the instructions and got it to work with Debian, NT and 95. No problems so far. Bruce Jackson Let me correct that. You forgot a letter. The correct name is http://www.owlnet.rice.edu/~sedmison/directboot.html Remco -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED]
Re: Invitation for testers of frozen distribution
From: Michael Nonweiler [EMAIL PROTECTED] No, 'fraid not. [no fix for 2GB fdisk problem yet] Actually, the boot-floppies package I uploaded this morning has a fix for the fdisk problem with disks over 2GB. I'll call for testers of the boot disk once I upload that. I just hacked generate_library to always include llseek if the library defines it - that should fix it. 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: Serial port IRQ strangeness
Put serial in /etc/modules . The problem will get better. Kerneld is not starting before 0setserial. 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]
where is 'gv'?
where is 'gv'? It used to be in rex but disppear recently. lawrence, -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED]
Re: How can I dial out on ttyS?
On Wed, 20 Nov 1996, CoB SysAdmin wrote: I'm trying to get an e-mail-alpha-pager thing working. Everything is working except the fact that I don't have permissions to use /dev/ttyS1 When I send mail to, say, jdoe, I have /usr/bin/filter pick it up and execute the pager script, and it fails claiming it doesn't have permission. So, I added jdoe to the uucp group (which is the group owner of the ttyS's). Now, when I su to jdoe and send a page, the page goes through. When I send e-mail to jdoe, it still fails with a permission thing. The problem is that user 'jdoe' is not who runs the pager process. My bet is that it's user 'mail' that is trying to send the page, who does not have the proper permissions. | This is OFFICIAL WRITTEN notification that I want to be REMOVED | | from ALL commercial mailing lists. EVERY message sent from this | | account has had this request posted. ALL UNSOLICITED ADVERTISEMENTS | | SENT TO THIS ACCOUNT ARE IN VIOLATION OF FEDERAL (U.S.) LAW.| -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED]
Re: hard disk recommendation
On Thu, 21 Nov 1996, Lawrence Chim wrote: Which Seagate/Quantum drive was using and which IBM drive is using by SGI? I don't have that info. You'd have to ask SGI. | This is OFFICIAL WRITTEN notification that I want to be REMOVED | | from ALL commercial mailing lists. EVERY message sent from this | | account has had this request posted. ALL UNSOLICITED ADVERTISEMENTS | | SENT TO THIS ACCOUNT ARE IN VIOLATION OF FEDERAL (U.S.) LAW.| -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED]
Re: cron problem
Rick Macdonald wrote:
Lawrence Chim wrote:
I don't know why cron always send mail to root with the following
body. Any idea?
Subject: Cron [EMAIL PROTECTED] run-parts /etc/cron.daily
X-Cron-Env: SHELL=/bin/sh
X-Cron-Env:
PATH=/usr/local/sbin:/usr/local/bin:/sbin:/bin:/usr/sbin:/usr/bin
X-Cron-Env: HOME=/root
X-Cron-Env: LOGNAME=root
getopt: illegal option -- o
You might want to post the file /etc/cron.daily so we can see
what the problem is.
Here is the files in /etc/cron.daily. Sorry if it is too many.
lawrence,
#! /bin/sh
#
# cron script to rotate process accounting file.
#
# Written by Ian A. Murdock [EMAIL PROTECTED]
cd /var/account
savelog -g adm -m 644 -u root -c 7 /var/account/pacct /dev/null
accton 2/dev/null
accton /var/account/pacct 2/dev/null
#
# This is needed because accton.c returns errno
exit 0
#! /bin/sh
#
# apacheRotate the apache logsfiles daily.
#
[ -d /var/log/apache/. ] || exit 0
umask 022
# Take care of the standard logfiles.
cd /var/log/apache
if [ -f access.log ]
then
savelog -c 7 access.log /dev/null
fi
if [ -f error.log ]
then
savelog -c 7 error.log /dev/null
fi
# Send a reload signal to the apache server.
if [ -f /var/run/apache/apache.pid ]
then
kill -1 `cat /var/run/apache/apache.pid` 2/dev/null
fi
#!/bin/sh
#
# This housekeeping shell script is installed as /etc/cron.daily/dwww.
# Cron automatically runs it once a day.
#
# This script re-builds the dwww web pages, and removes old documents
# from the cache that have not been accessed for some number of days.
# The number of days should be configured by setting the DWWW_KEEPDAYS
# variable in /etc/default/dwww. The value set in this script is a
# default; don't modify it, or an upgrade will wipe out your modification.
DWWW_KEEPDAYS=10
DWWW_QUICKFIND_DB=/var/lib/dwww/quickfind.dat
if test -r /etc/dwww.conf
then
. /etc/dwww.conf
fi
dwww-build --default /dev/null
find /var/lib/dpkg/info -name '*.list' |
dwww-quickfind --build $DWWW_QUICKFIND_DB.new
mv -f $DWWW_QUICKFIND_DB.new $DWWW_QUICKFIND_DB
find /var/spool/dwww -atime +$DWWW_KEEPDAYS -name '[^.]*' \
-maxdepth 1 -type f -print0 | xargs -0 rm -f
dwww-cache --clean
#! /bin/sh
#
# cron script to update the `find.codes' database.
#
# Written by Ian A. Murdock [EMAIL PROTECTED].
su nobody -c cd / updatedb 2/dev/null
#!/bin/sh
#
# $Id: cron.daily,v 1.4 1996/02/17 16:09:20 alvar Exp $
#
# expunge old catman pages which have not been read in a week
/usr/bin/nice find /var/catman -type f -name '*.gz' -atime +7 -print | xargs rm
-f
# expunge old catman pages which are older than one month
/usr/bin/nice find /var/catman -type f -name '*.gz' -mtime +30 -print | xargs
rm -f
exit 0
#!/bin/sh
cd /var/log/mgetty
for LOG in mg_*log; do
if [ -f $LOG ]; then
savelog -g adm -m 644 -u root -c 7 $LOG /dev/null
fi
done
#!/bin/sh
# /etc/cron.daily/net-acct: nacctd daily maintenance script
#
# Created Thu Jul 18 21:41:26 MET DST 1996 Bernd Eckenfels [EMAIL PROTECTED]
#
# See /usr/doc/net-acct/ for informations
#
test -d /var/log/net-acct || exit 0;
test -r /etc/naccttab || exit 0;
cd /var/log/net-acct
umask 022
# Stop writing out entries
kill -TSTP `cat /var/run/nacctd.pid`
# cycle logfiles (keep the last 14)
savelog -g root -m 644 -u root -c 14 net-acct.log /dev/null
# save the last debug file, for inforation only. nacctd never closes this
# file, dont delete.
# yes, here is a race, but we dont need that data, anyway
cp net-acct.debug net-acct.debug.0
date net-acct.debug
# Start writing out entries again
kill -CONT `cat /var/run/nacctd.pid`
exit 0
#!/bin/sh
# /etc/cron.daily/netbase: netbase daily maintenance script
# Written by Peter Tobias [EMAIL PROTECTED]
bak=/var/backups
umask 022
if cd $bak
then
cmp -s inetd.conf.bak /etc/inetd.conf || cp -p /etc/inetd.conf
inetd.conf.bak
fi
#!/bin/sh
if [ -f /var/log/ppp.log ]; then
savelog -g adm -m 644 -u root -c 4 /var/log/ppp.log /dev/null
/etc/init.d/sysklogd reload
fi
#!/bin/sh
# Run the Smail error checking script and rotate the logfiles.
# We have to rotate the logs as root as otherwise we can't fix the
# timestamps, and so we might as well add the checkerr script here
# too.
test -f /usr/sbin/smail || exit 0
cd /tmp
su mail -c /usr/lib/smail/checkerr
set -e
cd /var/log/smail
savelog -o root -g mail -m 644 -c 7 logfile paniclog /dev/null
#!/bin/sh
# /etc/cron.daily/standard: standard daily maintenance script
# Written by Ian A. Murdock [EMAIL PROTECTED]
# Modified by Ian Jackson [EMAIL PROTECTED]
# Modified by Steve Greenland [EMAIL PROTECTED]
bak=/var/backups
# The following three find commands are commented out do to the
# severe, easily exploited security hole introduced by 'find . _stuff_
# | xargs rm' style commands. Changing it to '-exec rm {}' doesn't
# help. If you need a program to prune back your tmp directories, try
# filereaper, also a Debian package.
# if [ -d /tmp -a ! -L /tmp ] cd /tmp
#
Re: How to choose safe configuration at boot time?
On Wed, 20 Nov 1996, Buddha M Buck wrote: (that'll teach me to use a 11-character root password). Unless you changed your system, only the first 8 have any significance. (Part of the shadow password suite allows extended length passwords.) i.e. 'oneverylongpassword' == 'oneveryl' | This is OFFICIAL WRITTEN notification that I want to be REMOVED | | from ALL commercial mailing lists. EVERY message sent from this | | account has had this request posted. ALL UNSOLICITED ADVERTISEMENTS | | SENT TO THIS ACCOUNT ARE IN VIOLATION OF FEDERAL (U.S.) LAW.| -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED]
RE: Laptop and PCMCIA
On Wed, 20 Nov 1996, Paul Christenson wrote: I'm trying slowy to get my whole office over to Debian g, and the next machine I'm going to tackle is the TI-Extensa 560CD laptop. I installed it on an Extensa 570CDT (except for X); no problems with the base installation. The CD-ROM is a standard ATAPI on /dev/hdb. The fact that there is a reboot involved between the last floppy disk and the first CD usage made it quite simple (as the CD and FDD share one bay). The fun started when I tried to get networking started. The PCMCIA package installed just fine, except that Debian tries to set up networking BEFORE Card Services is installed. I moved Card Services into /etc/init.d/boot before the network is initialized, and added a 'sleep' command that was just long enough to allow the Ethernet card (in slot A) to initialize. I suppose that moving Card Services into the 'networking' script is the proper approach, but the bottom line was that it worked. I had the same problem. I posted an message to the list a while ago on it, but it may have been missed. When first installing or configuring Debian, you get asked if you are connected to a network (or something similar). You should say no there. This is buried in the pcmcia package docs. Let the pcmcia script in init.d configure the network for you. You should put the information you normally keep in /etc/init.d/network in /etc/pcmcia/network.opts. You can just remove the entries from your /etc/network file and put them in /etc/pcmcia/network.opts now and reboot the system to get it going. The latest pcmcia card services package does not contain support for the Xircom card you mentioned. I had that card and had to switch it. Xircom has changed its non-disclosure policy a bit, so there may be support in a future release, but not yet. There is a _lot_ of information on linux on laptops available on the web. Do a search on those keywords. You can also follow links from RedHat's web site to the Linux Laptop Home Page. It has working configurations for a slew of specific laptop models, commentaries, technical links, etc. Very informative. Good luck. Richard G. Roberto [EMAIL PROTECTED] 011-81-3-3437-7967 - Tokyo, Japan -- *** Bear Stearns is not responsible for any recommendation, solicitation, offer or agreement or any information about any transaction, customer account or account activity contained in this communication. *** -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED]
Mouse troubles
I have a Logitech Pilot mouse which is giving me some trouble: the middle doesn't work correctly. Using xev it appears that at the mose is dragged with the middle button down, it bounces continuously. Using mev the button gives both up and down events at the start and the end of the drag and pretends not to be pressed in between. Also mc doesn't understand the mouse properly in the console or in X. (Do I have to explicitly tell emacs to do mouse things in the console?) Tony. -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED]
can't yp_bind: Reason: RPC failure
When I start up NIS, 'ypwhich' gets: can't yp_bind: Reason: RPC failure and 'ypcat passwd' gets: YPBINDPROC_DOMAIN: No bound server for domain lcinc.com No such map passwd.byname. Reason: Can't bind to server which serves this domain domainname is set correctly from /etc/defaultdomain, and at one point it actually created maps in /var/yp, but n longer. What am I missing here? Thanks -- __ David Phillips Lembersky Chi Incorporated Systems Engineer651 Strander BLVD [EMAIL PROTECTED] Tukwila WA 98188 206.575.2950 (voice)206.575.2717 (fax) -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED]
Re: How can I dial out on ttyS?
The problem is that user 'jdoe' is not who runs the pager process. My bet is that it's user 'mail' that is trying to send the page, who does not have the proper permissions. Well, the pager process is spawned by filter, which was spawned from the user's .forward. So it would be a bad thing(tm) for it to run as any user *except* the user. Besides, I even checked. I did: (whoami; $HOME/pagemail.pl 5551212 test message) $HOME/out and out then contains: jdoe Running pager process... ERROR: No permission to open /dev/ttyS1!!! dying - Joe -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED]
Re: How to choose safe configuration at boot time?
Mike O'Donnell [EMAIL PROTECTED] writes: WHAT I THINK I NEED: I think that I need to boot the system in single-user mode To reboot in single user mode, pass the `single' flag to LILO. (For example, if your image is called linux, type linux single). To reboot in emergency mode, pass the `emergency' flag. Single user mode is run level 1, where practically nothing is running. Emergency mode is even more basic - nothing is even mounted. The above flags are documented in init(8). Rebooting in single user mode is the easiest way to fix your xdm problem, as xdm only starts in level 2. Going to emergency mode is hardly ever needed. Guy -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED]
Re: bo is full of symlinks to rex?
Rick Macdonald [EMAIL PROTECTED] writes: Prior to the freezing of rex, unstable (rex) contained all of the physical files of the distribution. Now, unstable is bo, and it has many links back to rex. At what point will bo loose the links to rex and be self-contained again? Not until bo becomes stable. I know this setup is nonideal for people who only mirror unstable, but most do mirror the whole distribution. It's rude to have hundreds of megs of duplicate files. bo has ~100 meg of unique files right now, most of which is XFree 3.2. Can you suggest a better solution? When the change occured, my mirror of unstable lost a bunch of files. Now I have to get both rex and bo. I think I can continually mirror unstable only, by careful use of the following option in mirror, but I haven't actually tried this yet: That sounds like it work, assuming that symlinked directories aren't flattened. Otherwise you'll have to filter out the symlinked directories. Guy -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED]
Re: bo is full of symlinks to rex?
Rick Macdonald writes: When the change occured, my mirror of unstable lost a bunch of files. Now I have to get both rex and bo. I think I can continually mirror unstable only, by careful use of the following option in mirror, but I haven't actually tried this yet: [description of flags_[no]recursive+L option deleted] Anybody got a better method to mirror only the unstable tree efficiently? This is exactly what I do to mirror just unstable. It works fine. David -- David EngelOptical Data Systems, Inc. [EMAIL PROTECTED] 1001 E. Arapaho Road (972) 234-6400 Richardson, TX 75081 -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED]
Re: bo is full of symlinks to rex?
Rick Macdonald [EMAIL PROTECTED] writes: changing the flags passed to the remote ls: flags_recursive+L or flags_nonrecursive+L First test this by trying a ls -lRatL on the remote site under the ftp command to check whether the remote file-store has any symlink loops. Anybody got a better method to mirror only the unstable tree efficiently? Nope, this is what I use, and it seems to work. -- Rob -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED]
Re: Crack and cops
-BEGIN PGP SIGNED MESSAGE- On Sun, 17 Nov 1996, CoB [EMAIL PROTECTED] (Joe Emenaker) wrote: I didn't notice crack or cops listed in the Debian 1.1 package listing. I ftp'd crack and had trouble compiling it, discovered many others did, too; found the glitch and fixed it. So, it brings me to an interesting question: Is there a reason why cops and crack aren't in a package yet, other than possibly not having a maintainer? I figured that people might not like making a package like crack quite so plug-n-play, lest the baddie baddies get wind of it. qcrack is already in debian 1.2 (rex frozen), works well and has a good dictionnary. If the only impediment is that they need a maintainer, what do I need to do to enlist? (Probably check the FAQ first, huh? Duh!) Well, the FAQ about maintenance need was post lately... did you want a copy? :) - --- The trick isn't that free software are among the best, it's that commercial stuff aren't the best! - --- Fabien Ninoles aka Baffouille || Running Debian-Linux [EMAIL PROTECTED]|| Lover of MOO, mountains, http://www-edu.gel.usherb.ca/ninf01 || poetry and Freedom. - --- -BEGIN PGP SIGNATURE- Version: 2.6.3i Charset: noconv iQCVAwUBMpPPgFX6fc7jcjhFAQEEhwQAueQB/y0lJq05RPhunv5yrVyNKincER21 0ZiFVI6j4LjX1AMLg34VT7EUzMpySvQVAanfyMRIvWjog/FTlrAUNSbvQ+BZp9Rg BmqpKippKT7J7poG2XfaJy26tigu2ffZ2Snqm7Kisgtv6ahFGHEtBqSFpgax90MH 0b7YHPSHd6o= =ct7+ -END PGP SIGNATURE- -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED]
[no subject]
unsubscribe -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED]
Re: Windows NTFS 4 Partitions
-BEGIN PGP SIGNED MESSAGE- On Wed, 20 Nov 1996, Fingers wrote: Does anybody know if its possible to mount a Windows NT NTFS 4 partition on your Linux drive, baring in mind its on the same machine so i can't use samba. Chris. Fingers [EMAIL PROTECTED] Looks in the /project/experimental section of ftp.debian.org they're a still very alpha ntfs support... but remember that you would have a big leak of security with that and that ntfs aren't supposed to be mounted by anything else than NT (Although their are some good TSR for DOS in the Shareware market). For better security than NT, use cfs (same directory) :) - --- -- M.Marceau - --- Fabien Ninoles aka Baffouille || Running Debian-Linux [EMAIL PROTECTED]|| Lover of MOO, mountains, http://www-edu.gel.usherb.ca/ninf01 || poetry and Freedom. - --- -BEGIN PGP SIGNATURE- Version: 2.6.3i Charset: noconv iQCVAwUBMpPXAFX6fc7jcjhFAQHnSQP/S3bNZoTL/5HaJAkraKaSVJm1pQqgpvta d3TQEoBFLgnsXgn8nAQI1HtLk0N3Lz7w/okU7/ZjcySTQsv5AD9gQN7fnwh+2Va7 8Dhp65M6d+OiS4nXkVn4P83SbQL8MjLTFSRUmOMlqBvGITB7dft3PDuL+9nJ24/2 SOfOb0MduYw= =CFoJ -END PGP SIGNATURE- -- This message was delayed because the list mail delivery agent was down.
Re: cron problem
On Thu, 21 Nov 1996, Lawrence Chim wrote: I don't know why cron always send mail to root with the following body. Any idea? Subject: Cron [EMAIL PROTECTED] run-parts /etc/cron.daily X-Cron-Env: SHELL=/bin/sh X-Cron-Env: PATH=/usr/local/sbin:/usr/local/bin:/sbin:/bin:/usr/sbin:/usr/bin X-Cron-Env: HOME=/root X-Cron-Env: LOGNAME=root getopt: illegal option -- o lawrence, I just had the same problem the other day. The problem is in the script /etc/cron.daily/smail. The line savelog -o root ... should be savelog -u root ... I think there might even be a man page for savelog. Fix that and your error message will go away. Ciao, Shane -- ## [EMAIL PROTECTED] Shane Turner at Home [EMAIL PROTECTED] at Work -- This message was delayed because the list mail delivery agent was down.
PnP in kernel?
Just courius why the PnP is not part of the kernel and I have to patch the kernel everytime I upgrade the kernel. lawrence, -- This message was delayed because the list mail delivery agent was down.
Re: cron problem
On Thu, 21 Nov 1996, Lawrence Chim wrote: Rick Macdonald wrote: Lawrence Chim wrote: I don't know why cron always send mail to root with the following body. Any idea? Subject: Cron [EMAIL PROTECTED] run-parts /etc/cron.daily getopt: illegal option -- o You might want to post the file /etc/cron.daily so we can see what the problem is. Here is the files in /etc/cron.daily. Sorry if it is too many. The man page for savelog doesn't show any -o. My script has -u root, not -o root. Here's yours: set -e cd /var/log/smail savelog -o root -g mail -m 644 -c 7 logfile paniclog /dev/null ...RickM... -- This message was delayed because the list mail delivery agent was down.
Re: How to choose safe configuration at boot time?
to boot in single-user mode, add the word single or the word emergency to the end of the boot command line. It should look like this: boot: linux emergency single boots single-user with all daemons running. emergency just runs the shell. Paul, I have just obsoleted that Slackware rescue disk. The next boot floppies package has a one floppy combined installation root/boot disk that contains fsck and other rescue tools. 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: PnP in kernel?
We generally distribute Linus' sources. When he accepts the PnP patch into the main source thread you'd see it here. Perhaps he's already done that for 2.1 . 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: cron problem
Rick Macdonald wrote: On Thu, 21 Nov 1996, Lawrence Chim wrote: Rick Macdonald wrote: Lawrence Chim wrote: I don't know why cron always send mail to root with the following body. Any idea? Subject: Cron [EMAIL PROTECTED] run-parts /etc/cron.daily getopt: illegal option -- o You might want to post the file /etc/cron.daily so we can see what the problem is. Here is the files in /etc/cron.daily. Sorry if it is too many. The man page for savelog doesn't show any -o. My script has -u root, not -o root. Here's yours: set -e cd /var/log/smail savelog -o root -g mail -m 644 -c 7 logfile paniclog /dev/null I am surprised to know that because I never touch the cron script and it was installed by the cron package by default I think. lawrence, -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED]
Re: cron problem
Lawrence Chim [EMAIL PROTECTED] writes: [1 text/plain; us-ascii (7bit)] Rick Macdonald wrote: Lawrence Chim wrote: I don't know why cron always send mail to root with the following body. Any idea? Subject: Cron [EMAIL PROTECTED] run-parts /etc/cron.daily X-Cron-Env: SHELL=/bin/sh X-Cron-Env: PATH=/usr/local/sbin:/usr/local/bin:/sbin:/bin:/usr/sbin:/usr/bin X-Cron-Env: HOME=/root X-Cron-Env: LOGNAME=root getopt: illegal option -- o You might want to post the file /etc/cron.daily so we can see what the problem is. Here is the files in /etc/cron.daily. Sorry if it is too many. [11 smail text/plain; us-ascii (7bit)] #!/bin/sh # Run the Smail error checking script and rotate the logfiles. # We have to rotate the logs as root as otherwise we can't fix the # timestamps, and so we might as well add the checkerr script here # too. test -f /usr/sbin/smail || exit 0 cd /tmp su mail -c /usr/lib/smail/checkerr set -e cd /var/log/smail savelog -o root -g mail -m 644 -c 7 logfile paniclog /dev/null ^^ this is the culprit, change the -o to -u. I think it has already been reported as a bug against smail. Also look in /var/log/smail and remove the droppings. Andy. -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED]
Re: Crack and cops
On Sun, 17 Nov 1996, CoB SysAdmin wrote: I didn't notice crack or cops listed in the Debian 1.1 package listing. Both packages would be more than welcome to Debian. However, COPS would be more important since Debian 1.2 has qcrack (a high speed version of crack using hashing files). Currently, COPS and Crack are in my to-do list of packages to Debianize but if you feel a urgent need for them go ahead to package them. I ftp'd crack and had trouble compiling it, discovered many others did, too; found the glitch and fixed it. Yes, Crack is a pain but easily fixed. So, it brings me to an interesting question: Is there a reason why cops and crack aren't in a package yet, other than possibly not having a maintainer? I figured that people might not like making a package like crack quite so plug-n-play, lest the baddie baddies get wind of it. It not a matter of Debian/Linux not accepting them, it matter of time. Most package developers aren't paid for their time, so it takes awhile for packages such as COPS and Crack to get packaged. If the only impediment is that they need a maintainer, what do I need to do to enlist? (Probably check the FAQ first, huh? Duh!) You got it! :) And good luck should you take on this job COPS is going to be a major pain in the *ss to debianize (in my opinion). Let me know if you decide to take on either package so we don't duplicate our efforts. --- LEAR: Into her womb convey sterility! Dry up in her the organs on increase... (King Lear) --- Patrick J. Edwards [EMAIL PROTECTED] http://www.cs.usask.ca/undergrads/pje120/ http://hup1.usask.ca:8000/ finger [EMAIL PROTECTED] for my PGP Key Key fingerprint = 9F 45 7D 6E C0 A4 B4 0D 48 C7 14 CA 23 B0 B4 F8 -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED]
Re: hard disk recommendation
On Wed, 20 Nov 1996, Paul Christenson wrote: firmware on their drives, so testing continued. IBM drives are the only ones that currently meet SGI specs. This was from a conversation with an SGI tech about three weeks ago. Thanks for the correction. Not that it helped a whole lot, by the way, after the infamous patch 466 disaster with IBM Ultrastars. ---MAV (finger for PGP signature block) My opinions are my own and only my own. Standard disclaimer applies. Marc A. Volovic ([EMAIL PROTECTED]) Linguists do it cunningly -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED]
laptop, ethernet and networking
I'm an amateur trying to configure a newly bought laptop to be connected to a net using a 3COM 3C589C card. The following is part of the boot message FDC 0 is an 8272A NCR53c406a: no available ports found eata_dma: No BIOS32 extensions present. This driver release still depends on it. Skipping scan for PCI HBAs. qlogicisp : PCI bios not present PPA: unable to initialise controller at 0x378, error 2 scsi : 0 hosts. scsi : detected total. Partition check: hda: hda1 hda2 hda3 VFS: Mounted root (ext2 filesystem) readonly. Adding Swap: 29228k swap-space Linux PCMCIA Card Services 2.8.22 Intel PCIC probe: Intel i82365sl B step at 0x3e0, 2 sockets irq mask (scanned) = 0x9eb8, status change on irq 15 PCMCIA IO port probe: excluding 0x1f8-0x1ff 0x378-0x37f 0x3b8-0x3e7 0x3f8-0x3ff PCMCIA IO port probe: clean. loading device 'eth0'... eth0: 3Com 3c589, port 0x300, irq 3, 10baseT port, 00 60 97 40 B8 3F Swansea University Computer Society IPX 0.34 for NET3.035 as far as I can see it there are no errors in the automatic detection of the card? my /etc/hosts file looks like this: 127.0.0.1 localhost wall 130.226.66.83 fm_lap1 dmi66083.dmi.min.dk my /etc/resolv.conf looks like this: nameserver 130.226.64.24 nameserver 130.226.64.25 domiain dmi.min.dk then I run this (I guess it has to go in /etc/init.d/network later ?) ifconfig eth0 fm_lap1 netmask 255.255.255.0 ifconfig reports eth0 is up route add -net 130.226.66.83 with this result SIOCADDRT: Invalid argument any help will be highly appreciated... I have used the Linux Network Administration Guide to get so far.. Karsten Bolding -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED]
Re: Serial port IRQ strangeness
On Thu, 21 Nov 1996 03:20:14 +0200 Esa Turtiainen ([EMAIL PROTECTED] fi) wrote: Hi, You are trusting that kerneld will load the module serial. However, this will happen just when kerneld is running, but it is just later in the boot sequence. Compile a kernel with serial built in. Or just add a line with just the word ``auto'' to /etc/modules :-) Easier than rebuilding a kernel. Phil. -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED]
Re: Dselect docs for beginners
On Wed, 20 Nov 1996, Tim Sailer wrote: In the Access part, explain what the common cdrom devices are, what to do if you have problems with nfs (I have written up locally about the nfs.o problem, and have a ftpable copy), and offer a little more detail on the ftp method. Maybe point to some ftp sites. Are there any other public nfs sites? I gave this some thought, but as I am documenting dselect rather that the install process I have not gone into cdroms and the nfs problem will go away with 1.2. Maybe ftp should get a mention, but there again it is a different layer from dselect. I have not used it. Explain in detail what the 'EIOM' headings mean. Done. explain the '=' option in case something is broken in a package and you can't get by it (as seen in the e2fsprogs in the past few weeks) Done At this point the document is 300 lines which is about as far as I can stretch Bruce's request for a SHORT file. Look on ftp://oak.scotch.wa.edu.au/pub for dselect.beginner.3.gz. Please report errors. Lindsay As an aside here, why do messages from me to the list come up in my Pine index (in the From column) as To:debian-user rather than as from Lindsay? L -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED]
How to choose safe configuration at boot time?
Mike == Mike O'Donnell [EMAIL PROTECTED] writes:
Mike I am unable to use my Debian Linux system after the major
Mike update I did on Tuesday.
Mike WHAT HAPPENED: I used dselect in an xterm window under XFree
Mike to update the system after an idle period of 1 week. I
Mike update from stable, unstable, contrib, and nonfree. I
Mike allowed one of the installers to shut down xdm, which led to
Mike a reboot. I finished the installation by starting dselect
Mike again from an Xless console. The next reboot led to a login
Mike prompt on a non-X console screen, flashing and ignoring all
Mike input, except for CTL-ALT-DEL which led to another reboot
Mike ending in the same state.
I've seen this, usually right after a kernel compile when I had
forgotten to 'make modules_install' AND (cd /lib/modules;rm current;ln
-s 2.0.${NEW} current; depmod -a) and the lo device wasn't there... ?
What happens is that X-Windows cannot get a connection to the
loopback device, I think. The module hadn't been installed; I didn't
have it compiled into my kernel.
X needs a socket to connect to; if it can't find one it bombs, and
xdm loops and tries to start it again.
I suppose that another possiblility is that your X-server has become
misconfigured, and is not starting; if it cannot find a mode that is
within the constraints of your video hardware, as set in XF86Config,
it won't start... A missing file perhaps?
Or, a ~/.xsession that launchs all of its files , so the script ends
and X finishes. The last program run must be foreground, AFAIK. (If
you use the Debian setup, and run your startup programs all from the
fvwm2 hooks, you don't need an .xsession file at all.
See if there is a ~/.xsession-errors file, it may shed some light.
--
Karl M. Hegbloom [EMAIL PROTECTED]
http://www.teleport.com/~karlheg
(K0D) AYG-GE01 Portland, OR, USA
:) Proudly running Linux 2.0.25 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: Please do not use Qt (fwd)
On 20 Nov 1996, Arnt Gulbrandsen (Troll) wrote: As I said wrote earlier, the license doesn't limit distribution in any way which is particularly relevant to CD distribution. Quoting: You may copy this version of the Qt toolkit provided that the entire archive is distributed unchanged and as a whole, including this notice. That's what applies to FTP archive administrators, and that's what applies to CD vendors. Ok, I will forward this to the cdrom producers/Debian people. Yours, --martin // Martin Konold, Muenzgasse 7, 72070 Tuebingen, Germany // // Email: [EMAIL PROTECTED] // Linux - because reboots are for hardware upgrades -- Edwin Huffstutler [EMAIL PROTECTED] -- Just go ahead and write your own multitasking multiuser os ! Worked for me all the times. -- Linus Torvalds -- -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED]
Re: Got unusable memory address 0x400cd008 (and libtermcap)
From: [EMAIL PROTECTED] [EMAIL PROTECTED] Would there be in simple reason, why the binary works in RedHat, but not in Debian? If you have upgraded libc to 5.4.7, try this: LD_PRELOAD=/usr/lib/libgnumalloc.so.5 export LD_PRELOAD or if you use csh or tcsh: setenv LD_PRELOAD /usr/lib/libgnumalloc.so.5 If that fixes the problem, there's a malloc bug in the program and the new allocator we're using makes the bug show up instead of being silent. Bruce -- Thank you. That makes the program work(?). I have informed the programmers of the origical source. Apparently our single RedHat user does not update his system as often as I update our 10 Debian systems. Jouni RynöTel. (+358)-9-19294656 Finnish Meteorological Institute TLX 124436 EFKL FI Geophysical Research TFAX (+358)-9-19294603 P.O.BOX 503 Internet: [EMAIL PROTECTED] FIN-00101 HelsinkiX-400: /G=Jouni/S=Ryno Finland /O=il/P=il/A=mailnet/C=fi/ WWW: http://www.geo.fmi.fi/ WWW: http://www.geo.fmi.fi/~ryno/ It's just zeros and ones, it cannot be hard -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED]
Re: Strange behavior of lpr+lpd
You wrote: Which version of gs dou you use? Version 3.33 it is not a debian package, I installed it manually. If I invoke the filter with /etc/filter.ps filetoprint.ps /dev/lp1 the printer prints. Ok. So for so good. If I do lpr filetoprint.ps nothing appear. If I do lpq it says lp is ready and is printing but it is not true. This doesn't say anything. But you do see any pending jobs in the queue? A la: ([EMAIL PROTECTED]) ~$lpq stylus is ready and printing Rank Owner Job Files Total Size active david 312 demo.c 180 bytes There is the pending job. Lpc says queuing is enabled printing is enabled 1 entry in spool area etc PS: What was your success with magicfilter? No success with magicfilter no with apsfilter PPS: Have you read the Printing-HOWTO? Obviously. I begin to suspect there is a cable problem, with windows95 and directly with postscript I can print, but when lpq says that lp is ready and printing I notice that the data led on my printer lights, and sometimes the printer loads a sheet of paper. I will try another printer cable. But the question is, what handshake the lpd do with the printer? I mean, what return value lpd would receive from the printer before send another data? --- Mario Giammarco | Tel/FAX +39-545-22965 Via Calamandrei,5 |[EMAIL PROTECTED] 48022 Lugo (RA) -- ITALY | [EMAIL PROTECTED] -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED]
Emacs and tcsh
I use the following in my .cshrc: stty -icrnl -onlcr -echo susp ^Z which I believe fixes the problem of ^M. However, you may also want to use terminal mode rather than shell mode. If you say: M-x term you get a real terminal. If you type TAB, you get filename completion from the tcsh, and the terminal is in fact advanced enough to run for instance `less' in the buffer. It is not perfect, but it works well enough for me that I hardly use xterms any more. Only hassle is that you have to find out how to install the terminal definition. ---+-- Christian Lynbech | Computer Science Department, University of Aarhus Office: R0.32 | Ny Munkegade, Building 540, DK-8000 Aarhus C Phone: +45 8942 3218 | [EMAIL PROTECTED] -- www.daimi.aau.dk/~lynbech ---+-- Hit the philistines three times over the head with the Elisp reference manual. - [EMAIL PROTECTED] (Michael A. Petonic) -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED]
Re: How to choose safe configuration at boot time?
On Wed, 20 Nov 1996, Buddha M Buck wrote: (that'll teach me to use a 11-character root password). Unless you changed your system, only the first 8 have any significance. (Part of the shadow password suite allows extended length passwords.) Yes, I know that bigpassw is equivilant to bigpassword, but I still -think- of my password as bigpassword. If the shadow suite changes this behavior, I might consider switching to it. -- Buddha Buck [EMAIL PROTECTED] Just as the strength of the Internet is chaos, so the strength of our liberty depends upon the chaos and cacaphony of the unfettered speech the First Amendment protects. -- A.L.A. v. U.S. Dept. of Justice -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED]
XConfig for Thinkpad 350
We're installing Debian and X on an IBM Thinkpad notebook, and are having trouble getting X to work. The server won't start; it finds no valid modes. Does anyone have one of these running X Windows who could mail me the XF86Config file for it? (I will also look on ftp.xfree86.org) It has a vga screen; active(?) matrix, and 1M of video ram, I think. We are using the VGA16 server. -- Karl M. Hegbloom [EMAIL PROTECTED] http://www.teleport.com/~karlheg (K0D) AYG-GE01 Portland, OR, USA :) Proudly running Linux 2.0.25 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: X is painful
From: Shaya Potter [EMAIL PROTECTED] snip Nah, Xerox at PARC invented the GUI, they actually made a computer that used it, but it was really bad. Both Steve Jobs (Mac guy) and Bill Gates visited PARC and saw their GUI. They both saw that this was where the computer interface should head. The rest is history. snip Just a little aside. One of my jobs is for Xerox, and the Xerox workstations are pretty good. The UI and networking (XNS) are fully integrated, giving some very interesting and powerful side-effects. So interesting in fact that I understand that Novell licensed some it for NetWare 4. The workstations are Sparc based. Unfortunately they are dying out, as only Xerox internal development supports them, because neither the UI nor XNS became commercial products (don't ask me why). I know that Mac OS, Windows and Solaris are based on the PARC UI projects (there is even a thank you note to Xerox in the Solaris documentation), but this is not surprising seeing as Xerox developed the mouse, an early windowing environment, etc, whilst most other people hadn't even thought about a simple menu system. Thanks for listening | [EMAIL PROTECTED] | http://www.geocities.com/SiliconValley/Park/1152 | Simon Martin | Old software engineers never die, | they just fail to boot | | Any Trademarks used in this document are recognized | as Registered Trademarks of their respective owners. -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED]
Unidentified subject!
unsubscribe [EMAIL PROTECTED] -- -- Ingo Stapel RZ-Operator| Rechenzentrum TU Clausthal | E-Mail: [EMAIL PROTECTED] Erzstrasse 51 | D-38678 Clausthal-Zellerfeld | Germany| -- It's not save to hope for the best without preparing for the worst! -- -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED]
Re: cron problem
Lawrence Chim wrote: I don't know why cron always send mail to root with the following body. Any idea? I think what you are asking to related to the MAILTO variable. If you set MAILTO='' in your crontab for root the cron stuff will not be sent. An exerpt from the cron man page: When executing commands, any out put is mailed to the owner of the crontab (or to the user named in the MAILTO environment variable in the crontab, if such exists). -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED]
Re: Oleo - Any docs?
Ed Down wrote: I've used a few spreadsheets in my time, but the Oleo docs do not give me enough info to use the program. Anyone know of any user-friendly Oleo docs, or maybe an easier to use X spreadsheet program? I made a shot at installing SIAG but have not really put much effort into it. Has anyone tried it? How does it compare to the likes of NExS? SIAG: http://linus.asogy.stockholm.se/~ulric/siag/siag.html NExS: http://www.xess.com/NExS/nexs.html Incidentally, SIAG is GPL; NExS is commercial. -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED]
Re: Where is Netscape package?
On Wed, 20 Nov 1996, Mark Phillips wrote: I have been looking for the netscape package under rex but can't find it. Where is it? Thanks, Mark Phillips. ([EMAIL PROTECTED]) Looks in the contrib section, that's where is it. You have to download the .tar.gz file from netscape too. The deb package is just an wrapper/installer. (Only work with the netscape 3.0 for the moment too.) -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED]
Re: Strange behavior of lpr+lpd
You wrote: Which version of gs dou you use? Version 3.33 it is not a debian package, I installed it manually. If I invoke the filter with /etc/filter.ps filetoprint.ps /dev/lp1 the printer prints. Ok. So for so good. If I do lpr filetoprint.ps nothing appear. If I do lpq it says lp is ready and is printing but it is not true. This doesn't say anything. But you do see any pending jobs in the queue? A la: ([EMAIL PROTECTED]) ~$lpq stylus is ready and printing Rank Owner Job Files Total Size active david 312 demo.c 180 bytes There is the pending job. Lpc says queuing is enabled printing is enabled 1 entry in spool area etc PS: What was your success with magicfilter? No success with magicfilter no with apsfilter PPS: Have you read the Printing-HOWTO? Obviously. I begin to suspect there is a cable problem, with windows95 and directly with postscript I can print, but when lpq says that lp is ready and printing I notice that the data led on my printer lights, and sometimes the printer loads a sheet of paper. I will try another printer cable. But the question is, what handshake the lpd do with the printer? I mean, what return value/signal lpd would receive from the printer before send another data? --- Mario Giammarco | Tel/FAX +39-545-22965 Via Calamandrei,5 |[EMAIL PROTECTED] 48022 Lugo (RA) -- ITALY | [EMAIL PROTECTED] -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED]
Re: PPP
Are there any programs available for linux that allow you to logon to a normal telnet getty and then start up a PPP connection from the command line? I have a dialup connection to the whole network, but the PPP lines for this school are monstrously overcrowded. I'm not aware of such a tool for PPP, but there is one for SLIP: SLiRP. Basically, you dial in, make a terminal connection, preferably make that connection 8-bit clean, and start a daemon on the remote system. Check out /usr/doc/HOWTO/mini/Dip+SLiRP+CSLIP.gz (doc-linux package). Hope this helps, Ray -- PATRIOTISM A great British writer once said that if he had to choose between betraying his country and betraying a friend he hoped he would have the decency to betray his country. - The Hipcrime Vocab by Chad C. Mulligan -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED]
Re: csh * expantion problem
On Wed, 20 Nov 1996, Bruce Perens wrote: If you want to echo the list of files in the current directory, use: echo * The point is this doesn't work. Is there a working version of csh? Neal -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED]
X11 GUI Support (was Re: X11 bashing)
-BEGIN PGP SIGNED MESSAGE- Well, good discussion everyone, a pleasure to read it. The lacks of weel integrated GUI tools in X is volunteers but who knows a Windows System that run on most Unix System (including Linux, A/UX, FreeBSD, AIX, Sun, WindowsNT) and are distributed (a very painful, hard to maintain but powerful caracteristic)? X aren't to be change. It has to be complete with a mostly standard toolkit for GUI programmers. That's what the caracteristics I would like to find in this toolkit and if some already exist who has all those, it will be a good thing if Debian support it : 1) Open Free Standard. That means you can develop something absolutely free from it, with no restriction! It means too that, with all the standard already existing, I don't think we have to reinvent the wheel (to use a French expression that I don't know if I can translate ;). Other advantage of Open Standard is that you can ease software develop on other machine/OS and easily put it on Linux, a most performing system. That's help too to reduce the number of libraries on one system. X are already big enough! 2) Ease of development and maintainance. Object oriented? Not necessarely. Has someone says, C is standard and mostly use for almost software but Objects are supposed to be easier and C++ and Objective-C support C very well. 3) Good integration Why need a pretty GUI if we can't cutpaste between each windows. CORBA, OLE, etc. are some standard for IPC but first, each process has to know basically how to communicate between each other. 4) User configurable Because we prefered to (don't) have a big dock on the right side of us 13 monitor, we prefered (not) to have that or this in the pop-up root menu, we (don't) want a tool bar, a small (large) grey (blue) bordure on us windows and one (three) button to close (resize, quit) the window in its upper right corner. 5) Freeness Need to discuss about it? Well, that's the main three I can extract from the discussion... I don't want to go further because it can be subversive. Let's see now the different solutions already purpose. Motif: 1) Already an open standard mainly free altought that the only free version of the libraries are still in beta development. 2) Not yet ease to develop but a good object wrapper (like Object Windows for the WinAPI; I don't think that MFC are good objects... their have lot of bad scheme in it) can really ease the process and use the same library. But that can be apply to almost everything. Motif aren't a kind of wrapper around Xlib, is it? 3) Permit it in an entirely Motif interface. 3) Not as well as I would like but can be. 4) see 1) :) GNUStep: 1) See Motif 1)... 2) Is Objective-C... that's a plus and a weakness. A plus because Objective C are a pretty well done Object language and permit some real object oriented (loose binding) development. Weak because you are mainly restrict to write it in this language, who aren't well supported by most entreprise. But may be I'm wrong here. If someone can correct me :) 3) That's the goal of NextStep and the main domain of Jacob's expertise. 4) Well, GNUStep defined their own goal to be more open and configurable then NextStep already was. It can be good thing. 5) That's GNU stuff :) JAVA Desktop (surprised?): 1) Java are a well known standard, although young. 2) Never see a easiest and as powerful language than Java. It's supported integration of C function and can be exported in C programs. Like GNUStep, you are dependant of a Java Workspace for a well integrated system. Another point, Linux dis you would like to see Corel Suite running on a Linux Box? That can be possible. 3) Well done. 4) Still to be implemented but the java-VM are just going out the box... It can be really improved. 5) Java are kind of free. I don't think the Java license are too mush restrictive and avoid the GPL... ... I don't talk about V++, KDE or QT cause I don't know it well... For myself, the Java Desktop are my own project then I'm working on it (Learning Java are good for job, although I'm studying in `electrical' engineering.. :). If some `autonomous volunteer' want to contrib. I'll be happy to heard about you :). Pleasure to read your replies to this mail. Fab - --- Linux - because reboots are for hardware upgrades -- Edwin Huffstutler [EMAIL PROTECTED] -- - --- Fabien Ninoles aka Baffouille || Running Debian-Linux [EMAIL PROTECTED]|| Lover of MOO, mountains, http://www-edu.gel.usherb.ca/ninf01 || poetry and Freedom. - --- -BEGIN PGP SIGNATURE- Version: 2.6.3i Charset: noconv iQCVAwUBMpR1d1X6fc7jcjhFAQEr1AP/W1qp2eOOaxI5qipOmowbJ+XhcFqvSsEv 65nbJTUzT1T53So8tJwICSEqAQ624x1Zj3Qv7nPUnHhyO8bWfuulFbClEdR37W/J FdJyy0aLOrMmj+GXGoaA5CoT9JgnUTcGgrVU9OVo1QiV7EoomePMn6e3BYUNrsgX
SUMMARY: Emacs and tcsh
On Wed, 20 Nov 1996, Neal R. Dalton wrote: In emacs-19.31, with in emacs shell, the ^Ms show up. In csh they don't. However, in csh the '*' didn't expand. Is there a fix? As far as I know, the version of csh is broken. Is there a good TERM type for emacs shell, so things like passwds don't show up? passwds can be fixed with: (add-hook 'comint-output-filter-functions 'comint-watch-for-password-prompt) The answer was in the emacs FAQ for ^M in tcsh 62: How do I get rid of the ^M junk in my shell buffer? Try typing M-x shell-strip-ctrl-m RET while in shell-mode to make them go away. If that doesn't work, you have several options: For tcsh, put this in your `.cshrc' (or `.tcshrc') file: if ($?EMACS) then if ($EMACS == t) then if ($?tcsh) unset edit stty nl endif endif Or put this in your .emacs_tcsh file: unset edit stty nl Alternatively, use csh in your shell buffers instead of tcsh. One way is: (setq explicit-shell-file-name /bin/csh) and another is to do this in your .cshrc (or .tcshrc) file: setenv ESHELL /bin/csh (You must start Emacs over again with the environment variable properly set for this to take effect.) -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED]
Re: Serial port IRQ strangeness
[EMAIL PROTECTED] said: On Thu, 21 Nov 1996 03:20:14 +0200 Esa Turtiainen ([EMAIL PROTECTED] fi) wrote: ... Compile a kernel with serial built in. Or just add a line with just the word ``auto'' to /etc/modules :-) Easier than rebuilding a kernel. Unfortunately, this won't always help you if you use non-standard settings with setserial (like different IRQs). Everytime the modules is unloaded, all settings are lost. If you don't want to build a new kernel, and you use special settings, you have two options: - Load the module before using setserial - Use the post-install feature of kerneld setup your ports Cheers, Lukas --- Dr. Lukas Nellen | Email: [EMAIL PROTECTED] Depto. de Fisica Teorica, IFUNAM | Apdo. Postal 20-364 | Tel.: +52 5 622 5014 ext. 218 01000 Mexico D.F., MEXICO| Fax: +52 5 622 5015 -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED]
Re: libc
Thanks! But I installed the debian ELF system and what it put in as default is 4.6.27. I have 5.2.18 in there (as the ldconfig -v shows) but it is not being used as default. How do I go about upgrading/changing the default? Ricardo On Fri, 22 Nov 1996, Dimitri Maziuk wrote: Ricardo Kleemann wrote: Hi, I've installed the stable debian 1.1 release and noticed that my libc is defaulting to libc.so.4.6.27 Is there any reason why it's not using a later libc? Are there bugs with libc versions 4.6 (either 5.x or 4.7)? Hi. I suspect libc4 never got upgraded to libc-4.7.5 because it's an a.out library needed by only a few obscure packages which only a few people use. Debian is ELF and is based on libc5. The latest libc5 package (in rex) is libc-5.4.7, your system might have a libc-5.3.x. I suggest you re-check your /lib or just run ldd on something -- say, 'ldd /bin/bash'. Rgds Dimitri -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED]
Re: Where is Netscape package?
I have been looking for the netscape package under rex but can't find it. Where is it? The 3.0 version is in contrib. The 3.01 version still hasn't made it out of Incoming (to my knowledge). Brian ( [EMAIL PROTECTED] ) --- Debian GNU/Linux! Search it at http://insite.verisim.com/search/debian/simple -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED]
Re: Sources for rootdisk
I think the subject says it all... I'm looking for the sources of the programs found on Debian rootdisks. Can anybody help me? Thanks, Juan Cespedes Install the boot-floppies package. If you are planning to create an umsdos installation, please contact me. Click, MartinS. -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED]
Re: laptop, ethernet and networking
Karsten Bolding writes: [ A correctly working ethernet card ] - ifconfig reports eth0 is up - - route add -net 130.226.66.83 - - with this result - - SIOCADDRT: Invalid argument - Thos should be route add -net 130.226.66.0. route -net takes the network address to route to as the argument, not the interface address. If the machine has another interface with the same network, then you can just say route add -net 130.226.66.0 dev eth0 to get it to use the eth0 device. HTH -Larry -- Larry Daffner| Linux: Unleash the workstation in your PC! [EMAIL PROTECTED] / http://web2.airmail.net/vizzie/ Clarke's Third Law: Any sufficiently advanced technology is indistinguishable from magic. -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED]
Re: bo is full of symlinks to rex?
Prior to the freezing of rex, unstable (rex) contained all of the physical files of the distribution. Now, unstable is bo, and it has many links back to rex. At what point will bo loose the links to rex and be self-contained again? The plan is to leave symlinks from bo to rex until the time that bo is frozen. When the change occured, my mirror of unstable lost a bunch of files. Now I have to get both rex and bo. I think I can continually mirror unstable only, by careful use of the following option in mirror, but I haven't actually tried this yet: If the remote site contains symlinks that you want to flat- ten out into the corresponding files, then do this by changing the flags passed to the remote ls: flags_recursive+L or flags_nonrecursive+L First test this by trying a ls -lRatL on the remote site under the ftp command to check whether the remote file-store has any symlink loops. Anybody got a better method to mirror only the unstable tree efficiently? We have the same problem here. I didn't know about the options you describe. Unfortunately, that would disable all symlinks for the entire Debian distribution unless you get unstable separately. This is not too difficult to do, luckily. Brian ( [EMAIL PROTECTED] ) --- Searching for something? Look to us! http://www.verisim.com/ferret/ -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED]
Re: your mail
On Wed, 20 Nov 1996, Steven Thomas Christy wrote: unsubscribe -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED] Gotta love it... All messages going through the list have unsub info attached. I guess some people don't read. | This is OFFICIAL WRITTEN notification that I want to be REMOVED | | from ALL commercial mailing lists. EVERY message sent from this | | account has had this request posted. ALL UNSOLICITED ADVERTISEMENTS | | SENT TO THIS ACCOUNT ARE IN VIOLATION OF FEDERAL (U.S.) LAW.| -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED]
special kernels info: where?
Hello: I was looking for a special kernel (aha2940uw and also 3c905 if already available) in buzz/disks-i386/SpecialKernels and found out that the config-? files that used to sit there and described what special kernels are configured for are no more in the directory tree. According to the FAQ there are 6 special-kernel boot disks, but I have found 9 of them there. What are the three others configured for? Where can I find the information about these special kernels? I know that I'll probably use special kernel #1 - the problem is not there. I just thought that it would be nice to have them all documented somewhere at the FTP site. Thanks in advance -- [EMAIL PROTECTED] -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED]
Linux books.
Hi. I am just beginning to learn how to use and install Linux, and I was wondering if anyone had any book recommendations. I'm looking for something in the beginner to intermediate range, covering installation, system administration, configuration, and the like. Thanks. Jorge -- [EMAIL PROTECTED] [EMAIL PROTECTED] -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED]
Re: hard disk recommendation
- Received message begins Here - I looked at a new drive and the only thing I saw was a DHFS sticker from IBM. Is that a model number or something? As far as the Seagates, I saw a couple of ST43400ND drives here in the office. They appear to be 2-3 years old. I don't know of any others in use by SGI for sure, but I'd guess there are others. The 43400 is a bit too big to fit into an Indy. I don't have an Indy torn appart to look at right now. Paul Christenson wrote: On Wed, 20 Nov 1996, Marc A. Volovic wrote: On Tue, 19 Nov 1996, Paul Christenson wrote: In an attempt to avoid a war, I'll simply say that Silicon Graphics is using IBM SCSI drives exclusively. Given the nature of their products, I I am afraid you're wrong. I saw Seagate drives (in older stations) and, more recently, Quantum drives in High Impacts. SGI used to use Seagate exclusively (up to a year or so ago), but then they started getting customer complaints on the drives failing. They did go to Quantum for a short period, but they found that the Quantum drives have a nasty habit of recalibrating themselves from time to time, which makes the drives unavailable for a second or so, which really causes a performance hit on these machines. Quantum was unwilling to rewrite the firmware on their drives, so testing continued. IBM drives are the only ones that currently meet SGI specs. This was from a conversation with an SGI tech about three weeks ago. Which Seagate/Quantum drive was using and which IBM drive is using by SGI? lawrence, -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED] Jim Lynch, System Engineer, SGI/Cray Research, Inc. / ARS: K4GVO Federal Business Systems, Phone: (770) 631-2254, Email: [EMAIL PROTECTED] Suite 270, 200 Westpark Drive, Peachtree City, GA 30269 -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED]
Re: csh * expantion problem
Neal R. Dalton [EMAIL PROTECTED] writes: Under csh (not a link to tcsh), when I do a echo *, I get the following error: echo: No match. It work on other OSs. I tried echo * in an empty directory on a Solaris, SunOS, and HP/UX -- all the `csh' and `tcsh' agree -- they are unified in saying: echo: No match. Then, I said set nonomatch echo * and, lo and behold, they all echoed a `*' at me. The point is this doesn't work. Is there a working version of csh? Fix your shell script. - Hari -- Raja R Harinath -- [EMAIL PROTECTED] When all else fails, read the instructions. -- Cahn's Axiom Our policy is, when in doubt, do the right thing. -- Roy L Ash -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED]
Re: PPP
JHMD == J H M Dassen [EMAIL PROTECTED] writes: logon to a normal telnet getty and then start up a PPP connection from the command line? I have a dialup connection to the whole JHMD I'm not aware of such a tool for PPP, but there is one for JHMD SLIP: SLiRP. Basically, you dial in, make a terminal JHMD connection, preferably make that connection 8-bit clean, and JHMD start a daemon on the remote system. JHMD Check out /usr/doc/HOWTO/mini/Dip+SLiRP+CSLIP.gz (doc-linux JHMD package). I know that FreeBSD has such a tool, but I haven't been able to find such a beast for Linux.. -- \|/ \|/ Zaphod Beeblebrox in '96!! @'/ ,. \`@ /_| \__/ |_\ Michael Mattice - EE student at OK State U \__U_/ [EMAIL PROTECTED] -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED]
How's HUP'ping isdnlog?
Hi, I've got a minor problem for some time now, and I hope somebody might help me out. I have been running I4L for some time now without a hitch, so I was amazed to find out that something broke recently. Since I didn't notice it right away, there's no easy way for me to figure out what exactly caused it :-/ Anyways, as part of my isdn setup in init.d, I start isdnlog to keep an eye on my isdn traffic. I have isdnlog set up to dump data to /dev/tty8. This used to work just fine. However, recently somebody or something caused isdnlog to break on startup. The error message re-reading config file that gets dumped to the terminal is obviously caused by a SIGHUP that isdnlog receives. Now I'm wondering who goes around twiddling my processes when I'd rather have them run quietly :-) Is it due to the fact that the virtual consoles aren't available when isdnlog gets started? If that's the case, how do I get around it? Right now, isdnlog has a start/stop priority of 20. If I stop/start the script in /etc/init.d manually, everything works just fine. TIA for your opinions on that matter! Ciao, -- Thomas Baetzler, [EMAIL PROTECTED], [EMAIL PROTECTED] A HREF=http://home.pages.de/~thb/;thb's Homepage/A -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED]
Re: Dselect docs for beginners
In your email to me, Lindsay Allen, you wrote: On Wed, 20 Nov 1996, Tim Sailer wrote: In the Access part, explain what the common cdrom devices are, what to do if you have problems with nfs (I have written up locally about the nfs.o problem, and have a ftpable copy), and offer a little more detail on the ftp method. Maybe point to some ftp sites. Are there any other public nfs sites? I gave this some thought, but as I am documenting dselect rather that the install process I have not gone into cdroms and the nfs problem will go away with 1.2. Maybe ftp should get a mention, but there again it is a different layer from dselect. I have not used it. Explain in detail what the 'EIOM' headings mean. Done. explain the '=' option in case something is broken in a package and you can't get by it (as seen in the e2fsprogs in the past few weeks) Done At this point the document is 300 lines which is about as far as I can stretch Bruce's request for a SHORT file. Look on ftp://oak.scotch.wa.edu.au/pub for dselect.beginner.3.gz. Please report errors. Looks great! We should expand on this (maybe on a web page) to make a more complete documentaion of dselect. People tend to read stuff they can print out, but get flustered in an application like dselect. Tim PS: I'll make the web page and add to it if there is that need. -- (work) [EMAIL PROTECTED] / (home) [EMAIL PROTECTED] - http://www.buoy.com/~tps Too much information running through my brain, too much information, driving me insane. -- The Police ** Disclaimer: My views/comments/beliefs, as strange as they are, are my own.** -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED]
Will Caldera's WABI run on a Debian system?
Does anybody know if Caldera's WABI will run on a Debian system? Caldera's web page kind of says maybe. I'm tempted to pay the $200 just so I can run Quicken without booting up DOS/Windows. I haven't checked for awhile, but I assume that WINE is still a long ways away. I couldn't tolerate having my Quicken files corrupted! -- ...RickM... -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED]
Re: Will Caldera's WABI run on a Debian system?
Rick Macdonald [EMAIL PROTECTED] writes: Does anybody know if Caldera's WABI will run on a Debian system? Caldera's web page kind of says maybe. I'm tempted to pay the $200 just so I can run Quicken without booting up DOS/Windows. Actually I use Executor for that. It runs Macintosh Quicken 6 in an X window with no problem. I actually own both the Mac and Windows versions of Quicken (don't ask), and I prefer the Mac version. For those that don't know, Executor's a (commercial) Mac emulator that runs many Mac apps at roughly the speed of a 68040 running at half the clock speed of your pentium (YMMV). Details at www.ardi.com. -- Rob -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED]
Installation Instruction at www.debian.org
Hi, I really like the Installation Instruction presented at http://www.debian.org/. I think it is very clear and very well written. Tha authors are congratulated for a job well done. I do have a suggestion, however; that is we need more installation instruction after installing the base set. The instruction might include suggestions on what to be installed next and how. I think it will be very helpful to novice users of Debian Linux. Thanks. Le -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED]
Re: Strange behavior of lpr+lpd
I can not comment on the thread as developed so far, but let me add that since I have moved to debian1.1, lpr does not reliably print to a printer that worked for 0.93. The printer is an Apple 16/600 accessed directly via tcp/ip. If i send a printjob to a different IP address that is a Gatorbox that forwards jobs to the same 16/600, all is fine. I originally thought the problem was for printjobs sent to the printer via stdin (ie the print button xpdf, etc), but I had a couple of failures for regular files. The final straw was when my linux box apparently was flooding the printer with some sort of communication, thus choking it for hours at a time. I am afraid that since this was at work, I have not had time to pursue a rigorous debug of the situation. I have started to just use Windows for now since Word documents emailed as attachments seem to be our stock in trade. I had xbin in my old Linux but have not tracked it down again to build it for the new kernel. Just rumbling to see if this rings a bell. i can record detailed release numbers if necessary (I'm not in Linux :-( ) hunter -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED]
xterm resource not working after upgrading to XFree 3.2
Just after I upgrade to XFree 3.2, I don't why after I login from xdm, it couldn't set the resource properly, eg. in my .Xresources file, I set XTerm*foreground: white XTerm*background: black It worked before the upgrade. But now, the color is reversed. I can type 'xterm -bg black -fg white' and it works. It seems that it changes the way how we set the X resources. lawrence, -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED]
