As per subject line, I've been happily running SuSE 8.2 for several months now and even managed to get it to recognise my Nvidia GeForce card. All was fine and good.

Last night in a display-unrelated session, I was tweaking some bits and bobs under Samba to share out some space on my box to allow my Windoze laptop to dump some files there. This all went well; I created a share under my home directory on the Linux box, shared it out via SMB, did my transfer, cut a CD with the Linux machine, deleted the source files and happily shut down my boxen for the night.

This morning, I fired up my Linux box to check my email and discovered, to my horror, I have no X capabilities! I have not changed anything in the configuration at all other than last night's Samba sharing/file-copy/CD-burning episode, so am at a loss as to why X has suddenly stopped working.

Looking at the "log console" (CTRL-ALT-F10), the last dozen or so messages are all related to X and not being able to start KDM.

The messages are (truncated, but in order):
modprobe: Can't locate module char-major-195
kdm[2160]: Server for display :0 terminated unexpectedly
modprobe: Can't locate module char-major-195
kdm[2220]: IO Error in XOpenDisplay
kdm[2160]: Display :0 cannot be opened
modprobe: Can't locate module char-major-195
kdm[2223]: IO Error in XOpenDisplay
kdm[2160]: Display :0 cannot be opened
modprobe: Can't locate module char-major-195
kdm[2226]: IO Error in XOpenDisplay
kdm[2160]: Display :0 cannot be opened
kdm[2160]: Display :0 is being disabled (restarting too fast)

Logging in as root and trying to fire off kde "by hand" yields:
# kde
xsetroot: unable to open display ' '
ksplash: cannot connect to X server
kdeinit: Aborting. $DISPLAY is not set.
Warning: connect() failed: : No such file or directory
ksmserver: cannot connect to X server
startkde: Shutting down...
Warning: connect() failed: : No such file or directory
Error: Can't contact kdeinit!
startkde: Running shutdown scripts...
startkde: Done.
#

Aaargh!

Okay, what if I try and run startx on its own?:
# startx
XFree86 Version 4.3.0
Release Date:27 February 2003
X Protocol Version 11, Revision 0, Release 6.6
Build Operating System: SuSE Linux [ELF] SuSE
Build Date: 11 September 2003
Before reporting problems, check http://www.XFree86.Org/ to make sure you have the latest version(Okay, I don't and I haven't but "if it ain't broke, don't fix it", right? It was working fine 24 hours ago...)
Module Loader present
Markers: (--) probed, (**) from config file, (==) default setting, (++) from command line, (!!) notice, (II) informational, (WW) warning, (EE) error, (NI) not implemented, (??) unknown.
(==) Log file: "/var/log/XFree86.0.log", Time: Wed Jan 7 08:24:40 2004
(==) Using config file: "/etc/X11/XF86Config"
NV: could not open control device /dev/nvidiactl (No such device)
(EE) NVIDIA(0): Failed to initialize the NVIDIA kernel module!
(EE) NVIDIA(0): ***Aborting***
(EE) Screen(s) found, but none have a usable configuration.

Fatal Server error:
no screens found

When reporting a problem related to a server crash, please send the full server output, not just the last messages.
This can be found in the log file "/var/log/XFree86.0.log".
Please report problems to http:/www.suse.de/feedback.

XIO: fatal IO error 104 (Connection reset by peer) on X server ":0.0" after 0 requests (0 known processed) with 0 events remaining.
#

What the heck does this all mean? It can't "see" the Nvidia card (although the card works fine for "text mode" and graphically under Windoze)?

Why is this?

How did it happen?

More importantly, how do I fix it?


Please help!

Rik.

Having read the post about the kernel vulnerabilities from yesterday and having had a bit of a "lightbulb moment" as I skimmed through it, I suspect my SuSE automatic update may have generated the problem but won't be able to confirm 'til I get home to it, tonight.

My thinking is that the machine would automatically have been patched to fix the kernel prob ('cause that's how it's configured) yet the Nvidia drivers have to be (from bitter experience) carefully configured into the kernel to make them work.


Anyone else able to confirm or deny?

Rik.

It's amazing what one can do with a little perseverance, some logic and a bit of common-sense.

I was right (or, at least, as far as I can tell, partially).

Certainly, the Nvidia drivers were borked, but I can't be 100% positive quite what went on with the kernel, or, more specifically, I can't be 100% positive how the update checker's (part of YOU) is behaving. Its installation log shows no mention of having updated the kernel, the last log entry in the "update-messages" file is for SuSEfirewall2_update on December 11.

Launching YAST Online Update, selecting the "Manually select patches" option and subsequently ticking the "show installed patches" then checking the status of kernel software, however, shows that the kernel has been updated from 2.4.20-101 to 2.4.20-102.
*ding*

Sooo, how to fix the Nvidia driver?

Well, I checked with the Nvidia site (thank heavens for Lynx!) and found there was a new driver, anyway, downloaded it (V1.0-5328), switched to run-level 3 and installed it following the instructions on SuSE's ftp site (much easier than the previous version I was using, V1.0-4496), configured XFree86 with sax2 as per the instructions, did the business on the tuning of the display, saved the config, re-booted (to make sure ) and Voila! Back to glorious GeForce-powered technicolour!

On reflection, I guess I might have been able to avoid the initial nastiness and panicking that something was seriously wrong by simply running sax2 at the outset. Next time (although I hope not to have a next time!) I'll try that, first.

Phew! Boy, am I a happy puppy!?

Rik.