GDM logs out / disconnects user randomly

[moboticdes]

Binary package hint: gdm

On Ubuntu 9.10 I am randomly disconneted fro my laptop and back to Login screen with no reason, programs are closed when I log again.
Maibe interesting these lines in dmesg:

[ 45.643371] phy0: Selected rate control algorithm 'minstrel'
[ 45.644226] ath5k phy0: Atheros AR2425 chip found (MAC: 0xe2, PHY: 0x70)
[ 45.644397] cfg80211: Calling CRDA for country: AM
[ 45.654122] cfg80211: Regulatory domain changed to country: AM
[ 45.654128] (start_freq - end_freq @ bandwidth), (max_antenna_gain, max_eirp)
[ 45.654131] (2402000 KHz - 2482000 KHz @ 40000 KHz), (N/A, 2000 mBm)
[ 45.654134] (5170000 KHz - 5250000 KHz @ 20000 KHz), (N/A, 1800 mBm)
[ 45.654137] (5250000 KHz - 5330000 KHz @ 20000 KHz), (N/A, 1800 mBm)
[ 47.677489] HDA Intel 0000:00:14.2: PCI INT A -> GSI 16 (level, low) -> IRQ 16
[ 47.813262] hda_codec: Unknown model for ALC861-VD, trying auto-probe from BIOS...
[ 47.813363] input: HDA Digital PCBeep as /devices/pci0000:00/0000:00:14.2/input/input8
[ 48.225921] __ratelimit: 6 callbacks suppressed
[ 48.225925] type=1505 audit(1264151027.755:14): operation="profile_replace" pid=928 name=/usr/share/gdm/guest-session/Xsession
[ 48.228939] type=1505 audit(1264151027.759:15): operation="profile_replace" pid=929 name=/sbin/dhclient3
[ 48.229701] type=1505 audit(1264151027.759:16): operation="profile_replace" pid=929 name=/usr/lib/NetworkManager/nm-dhcp-client.action
[ 48.230089] type=1505 audit(1264151027.759:17): operation="profile_replace" pid=929 name=/usr/lib/connman/scripts/dhclient-script
[ 48.237152] type=1505 audit(1264151027.767:18): operation="profile_replace" pid=930 name=/usr/bin/evince
[ 48.250262] type=1505 audit(1264151027.779:19): operation="profile_replace" pid=930 name=/usr/bin/evince-previewer
[ 48.260887] type=1505 audit(1264151027.791:20): operation="profile_replace" pid=930 name=/usr/bin/evince-thumbnailer
[ 48.271480] type=1505 audit(1264151027.799:21): operation="profile_replace" pid=932 name=/usr/lib/cups/backend/cups-pdf
[ 48.272385] type=1505 audit(1264151027.803:22): operation="profile_replace" pid=932 name=/usr/sbin/cupsd
[ 48.276021] type=1505 audit(1264151027.803:23): operation="profile_replace" pid=933 name=/usr/sbin/mysqld

ProblemType: Bug
Architecture: i386
Date: Fri Jan 22 10:33:28 2010
DistroRelease: Ubuntu 9.10
Package: gdm 2.28.1-0ubuntu2.1 [modified: var/lib/gdm/.gconf.defaults/%gconf-tree.xml]
ProcEnviron:
 LANG=en_US.UTF-8
 SHELL=/bin/bash
ProcVersionSignature: Ubuntu 2.6.31-17.54-generic
SourcePackage: gdm
Uname: Linux 2.6.31-17-generic i686

[Sebastien Bacher (seb128)]

Thank you for your bug report, could you add you Xorg.0.log and Xorg.0.log.old there? it could be an xorg crash

[Ralf (shusai99)]

I think I have the same problem. This did not occur with 9.04, it started with 9.10.

I'll upload all log-files which were changed at the disconnect-time, if anything else is needed, please tell me!

[Ralf (shusai99)]

Here are the log files.

[itsjustarumour (itsjustarumour-gmail-deactivatedaccount-deactivatedaccount)]

I'm getting the same problem.

Distro Version - Ubuntu Karmic 9.10, 32-bit, default fresh install with latest updates
Desktop Environment - GNOME 2.28.1 (Ubuntu 2009-11-03)
Kernel - Linux 2.6.31-19-generic
System - Athlon 64 3400 2.2GHz processor, NVidia 6800LE graphics, 1.5MB RAM

[itsjustarumour (itsjustarumour-gmail-deactivatedaccount-deactivatedaccount)]

Xorg.0.log and Xorg.0.log files attached.

[Sebastien Bacher (seb128)]

there is nothing really interesting in those log, what do you do when getting the issue?

[itsjustarumour (itsjustarumour-gmail-deactivatedaccount-deactivatedaccount)]

Hello Sebastien, thanks for looking at this bug.

In my case, this always happens when surfing the web with Firefox - although thats what I'm doing 95% of the time anyway. It doesn't happen often - perhaps only once per week - but it happens. On the last two occasions, once I was simply reading online and on the other occasion it happened when I started typing an address into the URL bar. And thats all I can really say I'm afraid. If you can suggest any other log files to include or any other way of monitoring this problem, please let me know as I'd be happy to help!

[Ralf (ralfwin)]

The last two times it happened when I was copying text with Ctrl-c. Right after pressing the keys, the log-off occurred.

Xorg.0.log ends with these lines:

Backtrace:
0: /usr/bin/X(xorg_backtrace+0x3b) [0x8133d6b]
1: /usr/bin/X(xf86SigHandler+0x55) [0x80c7d35]
2: [0xaf3400]
3: /usr/bin/X(xf86AccResFromOS+0x45) [0x80b8c55]
4: /usr/bin/X(xf86ResourceBrokerInit+0x2f) [0x80be7bf]
5: /usr/bin/X(InitOutput+0x19a) [0x80b026a]
6: /usr/bin/X(main+0x1cb) [0x807234b]
7: /lib/tls/i686/cmov/libc.so.6(__libc_start_main+0xe6) [0x249b56]
8: /usr/bin/X [0x80719c1]
Saw signal 11. Server aborting.
 ddxSigGiveUp: Closing log

Something similar is going on here: https://bugs.launchpad.net/ubuntu/+source/nvidia-graphics-drivers-180/+bug/506069

[itsjustarumour (itsjustarumour-gmail-deactivatedaccount-deactivatedaccount)]

This just happened to me again, only 2 minutes ago. I opened KMail, and it downloaded 4 new messages via POP3. I right-mouse-clicked on the Inbox, sleected "Mark All Messages As Read", andthen I got the crash and I was logged out of my PC and back to the logon box.

[itsjustarumour (itsjustarumour-gmail-deactivatedaccount-deactivatedaccount)]

I've just checked the log files. My Xorg.0.log.old file ends with:

Backtrace:
0: /usr/bin/X(xorg_backtrace+0x3b) [0x8133d6b]
1: /usr/bin/X(xf86SigHandler+0x55) [0x80c7d35]
2: [0xa25400]
Saw signal 11. Server aborting.
(II) Power Button: Close
(II) UnloadModule: "evdev"
(II) AT Translated Set 2 keyboard: Close
(II) UnloadModule: "evdev"
(II) Power Button: Close
(II) UnloadModule: "evdev"
(II) Macintosh mouse button emulation: Close
(II) UnloadModule: "evdev"
(II) USB Optical Mouse USB Optical Mouse: Close
(II) UnloadModule: "evdev"
 ddxSigGiveUp: Closing log

[itsjustarumour (itsjustarumour-gmail-deactivatedaccount-deactivatedaccount)]

Just checked that bug report that Ralf refers to above. Yup - sure looks like what we're getting!

[itsjustarumour (itsjustarumour-gmail-deactivatedaccount-deactivatedaccount)]

Several users reporting same problem on this thread on the official Ubuntu Forum:

http://ubuntuforums.org/showthread.php?t=1387438

[Pascal Vandeputte (pascal-vdp)]

Are you sure this is gdm-related? I just experienced the same problem for the first time today on my Kubuntu 9.10 (hence with kdm instead). On an HP EliteBook 8530p with a "ATI Technologies Inc Mobility Radeon HD 3650" (fglrx driver).

X crashed and kdm should have restarted (but AFAIK on this hardware the system then halts completely). The log may show some more stuff at the end because I tried some SysRq combinations to reboot cleanly (which failed as well).

[...]
(II) fglrx(0): EDID vendor "AUO", prod id 6523
(II) fglrx(0): Printing DDC gathered Modelines:
(II) fglrx(0): Modeline "1680x1050"x0.0 131.50 1680 1808 1920 2030 1050 1053 1059 1080 -hsync -vsync (64.8 kHz)

Backtrace:
0: /usr/bin/X(xorg_backtrace+0x26) [0x4f00c6]
1: /usr/bin/X(xf86SigHandler+0x41) [0x488281]
2: /lib/libc.so.6 [0x7f858d835530]
3: /usr/bin/X(FindGlyphRef+0x2c) [0x5275ec]
4: /usr/bin/X(FindGlyphByHash+0x28) [0x527c58]
5: /usr/bin/X [0x533f49]
6: /usr/bin/X(Dispatch+0x394) [0x44e174]
7: /usr/bin/X(main+0x3b5) [0x434085]
8: /lib/libc.so.6(__libc_start_main+0xfd) [0x7f858d820abd]
9: /usr/bin/X [0x433509]
Saw signal 11. Server aborting.
(II) AT Translated Set 2 keyboard: Close
(II) UnloadModule: "evdev"
(II) Video Bus: Close
(II) UnloadModule: "evdev"
(II) CKA7216: Close
(II) UnloadModule: "evdev"
(II) Sleep Button: Close
(II) UnloadModule: "evdev"
(II) Power Button: Close
(II) UnloadModule: "evdev"
(II) Macintosh mouse button emulation: Close
(II) UnloadModule: "evdev"
(II) UnloadModule: "synaptics"
(II) Microsoft Compact Optical Mouse 500: Close
(II) UnloadModule: "evdev"
(II) PS/2 Generic Mouse: Close
(II) UnloadModule: "evdev"
(II) AIGLX: Suspending AIGLX clients for VT switch
(II) fglrx(0): Backup framebuffer data.
(II) fglrx(0): Backup complete.
(II) fglrx(0): IRQ 32 Disabled.
 ddxSigGiveUp: Closing log

[Ralf (ralfwin)]

No, I'm not sure if it is gdm related.

Meanwhile I think it's something graphic-driver problem, looks as if only affects notebooks.

I had an automatic update today, no log-off since but now there is a new "feature": my notebook suddenly suspends. Happened already a few times during the last hour.

[Sebastien Bacher (seb128)]

the bugs with those lines

"Backtrace:
0: /usr/bin/X..."

indicate that xorg is crashing and not gdm issue, they should be reported against xorg-server

[Pascal Vandeputte (pascal-vdp)]

Funny you mention that sudden suspend issue... I've had that twice the past month. :-|
For now I'm going back to the x.org "radeon" driver instead, I'm not using 3D anyway. I'm curious if it'll help.

[Pascal Vandeputte (pascal-vdp)]

Sigh, back to fglrx anyway, because I'd forgotten about the "laptop overheat + shutdown" issue I experience otherwise.

Cfr. https://bugs.launchpad.net/bugs/370173
Using fglrx + running the command "sudo aticonfig --acpi-services=off" fixes that problem for me. With the x.org radeon driver I never found a workaround.

[itsjustarumour (itsjustarumour-gmail-deactivatedaccount-deactivatedaccount)]

Changed affected package to "xorg-server".

[itsjustarumour (itsjustarumour-gmail-deactivatedaccount-deactivatedaccount)]

OK, this is getting silly - I've had bug this happen 4 times so today (so far).

I am getting this on all 3 of my machines - 2 NVidia-based desktop PCs and also my Intel-based laptop (HP Compaq 6720s).

On 2 occasions I lost work when when this crash happened. This is really not very good.

I would suggest that this bug is more accurately described as "medium" importance.

[Ralf (ralfwin)]

Adding this PPA seems to have helped me:
https://launchpad.net/~ubuntu-x-swat/+archive/x-updates

No more errors after installing those updates.

[itsjustarumour (itsjustarumour-gmail-deactivatedaccount-deactivatedaccount)]

@Ralf - alas, those updates made no difference to me. Still getting the same crashes.

[Ralf (ralfwin)]

Today I had a strange looking icon in the notification tray, when clicking it said something like "if you can read this, there is a problem with your xorg-server, see http://blogs.gnome.org/hughsie/2009/8/17/ for details".

[Bryce Harrington (bryce)]

[This is an automatic notification.]

Hi moboticdes,

This bug was reported against an earlier version of Ubuntu, can you
test if it still occurs on Lucid?

Please note we also provide technical support for older versions of
Ubuntu, but not in the bug tracker. Instead, to raise the issue through
normal support channels, please see:

    http://www.ubuntu.com/support

If you are the original reporter and can still reproduce the issue on
Lucid, please run the following command to refresh the report:

  apport-collect 511095

If you are not the original reporter, please file a new bug report, so
we can work with you as the original reporter instead (you can reference
bug 511095 in your report if you think it may be related):

  ubuntu-bug xorg

If by chance you can no longer reproduce the issue on Lucid or if you
feel it is no longer relevant, please mark the bug report 'Fix Released'
or 'Invalid' as appropriate, at the following URL:

  https://bugs.launchpad.net/ubuntu/+bug/511095

[Rayne-William (iamgoldeneagle)]

 I am having a similar, if not the same, issue happening to me.

In most cases I am just browsing the internet with Firefox and the next thing I am needing to do is log back in.

I just upgraded, from Xubuntu 9.10 to Xubuntu 10.04, a few days ago with my old Dell C510/C610

[leoboy (leonard-wauters)]

I'm having a similar problem for several months, hitting ctrl+c sometimes shuts down X and sends me back to gdm logon screen, without any warning...

However, I dug a little bit the problem on my own, and I could notice the following :
-X always shuts down because I hit Crtl+c - for instance when I want to interrupt a script on a console, or when I want to copy-paste data.
-The "crtl+c" bug sometimes works as soon as I enter my login / pass on GDM ! Just after logging in, and just before seeing gnome desktop appearing, if I hit ctrl+c, I instantly get back to gdm. And I can see gdm reloading, by the way.
-When the "ctrl+c bug" does not work just after logging in, it happends randomly after around more or less 20 minutes, when hitting ctrl-c.
-I found a crazy way to get around the crtl+c bug that always work :
Just after logging in (ie : entering my username and password), and before gnome appearing, I press ctrl+c. If it crashes gdm (or whatever program), I log in again. If it does nothing, as soon as gnome has finished loading, I log out normally and log in again. Then I can work safely, without fearing the deathly ctrl-c keystroke.

Crazy, heh ? With this kind of bugs, it seems like ubuntu is walking on MS Windows path.... :) Black magic and voodoo power is now necessary on Unix world, what a pity....

BTW, I'm using Ubuntu 9.10 on a Dell XPS M1330 with nvidia video card.

Br,

Léo.

[leoboy (leonard-wauters)]

Hello,

By checking deeper (as I had an X shutdown after a ctrl+c this morning), I found that on /var/log/gdm/xsplash-gdm.log :

[09:31:47:2071228439 - MSG] received signal: gdm
[09:31:47:2071228499 - MSG] gdm mode: fading out
[09:31:48:2070062061 - MSG] fade finished
[09:31:48:2070077025 - MSG] exiting...

does it helps ?

[Roland Giesler (lifeboy)]

This doesn't appear to be nvidia related, since we have an Intel display notebook that also does this "ctrl-c" X-server crash:

00:00.0 Host bridge: Intel Corporation Mobile 4 Series Chipset Memory Controller Hub (rev 07)
 Subsystem: QUANTA Computer Inc Device 0773
 Flags: bus master, fast devsel, latency 0
 Capabilities: [e0] Vendor Specific Information <?>
 Kernel driver in use: agpgart-intel
 Kernel modules: intel-agp

00:02.0 VGA compatible controller: Intel Corporation Mobile 4 Series Chipset Integrated Graphics Controller (rev 07)
 Subsystem: QUANTA Computer Inc Device 0773
 Flags: bus master, fast devsel, latency 0, IRQ 27
 Memory at 90000000 (64-bit, non-prefetchable) [size=4M]
 Memory at 80000000 (64-bit, prefetchable) [size=256M]
 I/O ports at 5140 [size=8]
 Capabilities: [90] Message Signalled Interrupts: Mask- 64bit- Queue=0/0 Enable+
 Capabilities: [d0] Power Management version 3
 Kernel driver in use: i915
 Kernel modules: i915

00:02.1 Display controller: Intel Corporation Mobile 4 Series Chipset Integrated Graphics Controller (rev 07)
 Subsystem: QUANTA Computer Inc Device 0773
 Flags: bus master, fast devsel, latency 0
 Memory at 93500000 (64-bit, non-prefetchable) [size=1M]
 Capabilities: [d0] Power Management version 3

[Roland Giesler (lifeboy)]

Apologies, I forgot to add my software release:

Linux ASLAN1 2.6.31-21-generic #59-Ubuntu SMP Wed Mar 24 07:28:56 UTC 2010 i686 GNU/Linux

[leoboy (leonard-wauters)]

Found out some other things on /var/log/gdm/:1-slave.log.1 :

gdm-session-worker[2082]: pam_ck_connector(gdm:session): nox11 mode, ignoring PAM_TTY :1
xsplash: Fatal IO error 11 (Resource temporarily unavailable) on X server :1.0.
gdm-session-worker[2082]: pam_unix(gdm:session): session closed for user xxx

And in /var/log/gdm/xsplash-gdm.log :

[10:29:40:-1902445403 - MSG] throbber started (50 frames)
[10:29:46:-1908808413 - MSG] received signal: gdm
[10:29:46:-1908808369 - MSG] gdm mode: fading out
[10:29:47:-1909986407 - MSG] fade finished
[10:29:47:-1909978177 - MSG] exiting...

[Akif Tariq (akiftariq)]

I am getting this problem in Lucid.

[Bryce Harrington (bryce)]

@shusai99 - thanks for providing that info, but we would like to see it from the original reporter. We may want to send this bug report upstream, but upstream has a hard and fast rule that logs from someone other than the original reporter are not acceptable. The reason for that is that quite often with X.org bug reports, two people may *think* they have the same issue based on commonality of symptoms and/or hardware, but in truth they have separate issues and mixing the report and the files can cause the developers some confusion.

If this bug report seems not to be getting attention, I would encourage you to file a new bug report with this data, and simply reference this bug report in your bug description.

[Michael (null-pointer-us)]

> Bryce Harrington:
>
> If you are not the original reporter, please file a
> new bug report, so we can work with you as the
> original reporter instead (you can reference bug
> 511095 in your report if you think it may be related):
>
> ubuntu-bug xorg

My new bug report (lucid):
https://bugs.launchpad.net/ubuntu/+source/xorg/+bug/585870

[Launchpad Janitor (janitor)]

[Expired for xorg-server (Ubuntu) because there has been no activity for 60 days.]