X crashes on logout - takes long to come up on boot

[Izzy]

Binary package hint: xorg

Not sure whether it applies to xorg or rather some installation scripts...

After upgrade from Feisty to Hardy, X takes very long to startup after boot (user just sees the black screen with the last boot messages, ended with "No resume image, doing normal boot...", and already starts to thing something is broken). Logging out (or doing a reboot) brings X to a crash:

Backtrace:
0: /usr/bin/X(xf86SigHandler+0x7e) [0x80c780e]
1: [0xb7fa5420]
2: /usr/lib/xorg/modules/drivers//nvidia_drv.so(_nv001216X+0xe5) [0xb71b2711]
3: [0x1]

Fatal server error:
Caught signal 11. Server aborting

I suspected something with the display managers, and in fact I found: The upgrade added gdm to the rc.* folders (I use KDE). Checking inside the /etc/init.d/gdm script I found:

# To start gdm even if it is not the default display manager, change
# HEED_DEFAULT_DISPLAY_MANAGER to "false."
HEED_DEFAULT_DISPLAY_MANAGER=true

Maybe that explains at least the long startup time? What is to be done:
a) changing that line in /etc/init.d/gdm to "...=false"
b) removing the links to gdm from the /etc/rc.?/ folders?

Advice welcome!

1) System information:
lsb_release -rd
Description: Ubuntu 8.04.1
Release: 8.04

2) last updated yesterday - whatever package is guilty :)

3) starting X immediately / shutdown X without error

4) waiting at the black screen, such confusing the user ;) / Crashing with above error message

In case it matters: Update was done via the update-manager (as described in the Wiki). I updated 2 machines this way - only one is affected by described bug. Both have NVidia graphics.

[Izzy (izzy-qumran)]

Ooops... Just checked the log files of the second machine, and found the same crash there - so I just didn't notice. So it may be I also did not notice the slow startup since I haven't rebooted that machine ever after the dist-upgrade...

On the second machine, the gdm startup scripts are also linked to the runlevel directories, same as with the report above (i.e. also with "HEED_DEFAULT_DISPLAY_MANAGER=true" in the scripts). To me it is not clear why I should startup GDM when I only run KDE which was set to be the default display manager before the upgrade. Is that really needed?

[Izzy (izzy-qumran)]

Some more details on the PC with the heavy problems: The card is a GeForce 6100 nForce 430 (according to nvidia-settings). I removed the above mentioned GDM links, nothing changed. Even more: If I do a warm boot after logging out from KDE (and having no other choice, since I have only the black screen as alternative to the remaining text consoles), I finally end up with Bios beeping and a black screen (beep code indicates memory error - Award Bios, continuously beeeep, beeep, beeep... I have to turn off the computer, wait a few seconds, turn it on again. Then it comes up until after the boot splash - and hangs for about 20 seconds as described above, before X finally gets started.

I ran nvidia-xconfig again (which changed and re-arranged the xorg.conf), so I gave it another try. After the logout from KDE the same situation. This time I grabbed a console and had a look at the log files - and here's what I found in /var/log/syslog:

 Jul 23 20:15:14 nebo console-kit-daemon[6171]: WARNING: Unable to activate console: No such device or address
Jul 23 20:15:14 nebo kernel: [39286.563103] Xorg[6575]: segfault at 00000009 eip 080c671b esp bff8cce0 error 4
Jul 23 20:15:14 nebo kdm[6570]: X server terminated: [0, 0, 0]
Jul 23 20:15:14 nebo kdm[6570]: X server for display :0 terminated unexpectedly
Jul 23 20:15:14 nebo kdm[6570]: Unable to fire up local display :0; disabling.

looks like the video driver messes things up completely. Driver details (taken from "dpkg -l nvidia-*)

ii nvidia-glx 1:96.43.05+2.6.24.14-20.46 NVIDIA binary XFree86 4.x/X.Org driver
un nvidia-glx-legacy <keine> (keine Beschreibung vorhanden)
un nvidia-glx-new <keine> (keine Beschreibung vorhanden)
un nvidia-glx-src <keine> (keine Beschreibung vorhanden)
un nvidia-kernel-1.0.7184 <keine> (keine Beschreibung vorhanden)
un nvidia-kernel-1.0.7185 <keine> (keine Beschreibung vorhanden)
un nvidia-kernel-1.0.9631 <keine> (keine Beschreibung vorhanden)
un nvidia-kernel-1.0.9639 <keine> (keine Beschreibung vorhanden)
un nvidia-kernel-1.0.9755 <keine> (keine Beschreibung vorhanden)
un nvidia-kernel-100.14.19 <keine> (keine Beschreibung vorhanden)
un nvidia-kernel-169.12 <keine> (keine Beschreibung vorhanden)
un nvidia-kernel-71.86.04 <keine> (keine Beschreibung vorhanden)
un nvidia-kernel-96.43.05 <keine> (keine Beschreibung vorhanden)
ii nvidia-kernel-common 20051028+1ubuntu8 NVIDIA binary kernel module common files
un nvidia-kernel-source <keine> (keine Beschreibung vor...

[Izzy (izzy-qumran)]

Anybody here reading this?

[Izzy (izzy-qumran)]

Attaching the Xorg log file in the hope it helps on the issue.

Missing technical details:
* 32bit Ubuntu (on an AMD64) - kernel 2.6.24-19-generic
* nvidia-glx 1:96.43.05+2.6.24.14-20.46
* xorg 1:7.3+10ubuntu10.2

If I missed something else, please let me know.

[Alberto Milone (albertomilone)]

Please try this:

1) make a backup of your /boot/grub/menu.lst
cp /boot/grub/menu.lst $HOME/

2) Open your /boot/grub/menu.lst (using sudo and a text editor) and get to the following part:

## additional options to use with the default boot option, but not with the
## alternatives
## e.g. defoptions=vga=791 resume=/dev/hda5
# defoptions=quiet splash

DO NOT uncomment it, just make it look like the following:

## additional options to use with the default boot option, but not with the
## alternatives
## e.g. defoptions=vga=791 resume=/dev/hda5
# defoptions=quiet

Save the file and exit.

3) type:
sudo update-grub

then make sure that the word "splash" it's gone from your /boot/grub/menu.lst

4) Restart your computer and see if you can reproduce the problem.

[Izzy (izzy-qumran)]

Alberto,

thank you for your reply. Did what you suggested - problem stays the same. But now I can state more clearly it *must* have to do with X: It runs through the init scripts normally, and right after the "battery check" (which is told to be [OK]) it stalls for about 20..30 seconds. Next message then is "Starting KDE".

What I meanwhile also did: Reset the /etc/X11/xorg.conf to use "nv", rebooted. Removed the nvidia-glx using dpkg. Invoked jockey, which installed nvidia-glx-new (ah - different package). Rebooted - problem remained. Checking /etc/X11/xorg.conf - Ooops? Half of the stuff was missing (Keyboard etc.), so Xorg had to probe for most of it! Fixed that up and booted again, this time with your changes to grub. Problem still remained.

I checked the Xorg.0.log and saw it still probed for the nvidia card and some other things - only a few, but it could be that probing the card alone ate most of the time. Is there something wrong with my xorg.conf?

[Izzy (izzy-qumran)]

Also attaching the Xorg.0.log of the last boot - just in case

[Izzy (izzy-qumran)]

I checked all the log files now to get some indicators. All I could find out is a gap of 34 seconds between the last pre-KDM entry and the first KDM-entry - so it seems that KDE takes more than 30 seconds to come up. I also attach the latest piece from the KDM log, which may provide some interesting details. Unfortunately, there are no timestamps in that log... But it shows up a bunch of errors and warnings, so I hope this may provide a clue.

I'm not sure where the last boot-up is to be located in this file. The time-stamp is a few hours after the last boot, so at least the last lines are from after it...

Please let me know if, what and how I can provide to help locating the issue.

[Bryce Harrington (bryce)]

Hi izzy-qumran,

Thank you for taking the time to report this bug and helping to make Ubuntu better. You reported this bug a while ago and there hasn't been any activity in it recently. We were wondering is this still an issue for you? Can you try with the latest development release of Ubuntu? (ISOs are available from cdimage.ubuntu.com)

If it remains an issue, could you also attach a new /var/log/Xorg.0.log?
Thanks in advance.

The output of lspci -vvnn would also be worth having.

[cenora (cenora)]

Hi, I had a similar problem on a fresh 8.10 install.

PC is a Lenovo laptop (R61).

I have installed the restricted drivers for the NVidia graphics card (version 177).
When I select logout, the screen goes black for 4 minutes, finally, the X downgrades to a lower resolution and displays this message:

"Ubuntu is running in low graphics mode. Your screen, graphics card, and input device setting could not be detected correctly. You will need to configure these yourself."

Then, I selected to view (archive) the proglem. It saved it to /var/log/failsafeX-backup-081108114801.tar (attached).
After other warnings, I was brought back to the desktop (low resolution, with screen broken in half and got this message:

"You do not appear to be using the NVIDIA X driver. Edit your X configuration file (just run 'nvidia-xconfig' as root) and restart..."

I did just that, but the problem remains.

Regards,

[bugbot (bugbot)]

This bug report was filed against an old version of Ubuntu.
Can you confirm whether this is still an issue in natty?

If you don't mind, it would be very helpful if you could update the bug
report in launchpad to 'Fix Released' if it is no longer an issue for
you, or if it is still occurring under natty, please tag the bug 'natty'
so it's easier for us to track.

[bugbot (bugbot)]

We're closing this bug since it is has been some time with no response from the original reporter. However, if the issue still exists please feel free to reopen with the requested information. Also, if you could, please test against the latest development version of Ubuntu, since this confirms the bug is one we may be able to pass upstream for help.