[natty] X often crashes since upgrade to natty

Also it seems I am not lucky with natty if we speak about vga cards :) With ATI radeon it turned out I hit a drm module bug which was fixed already in upstream but not in ubuntu's kernel, with nvidia, trying nouveau I have no video signal till gdm or when I switch to text console at all (and I need libgl1-mesa-dri-experimental for any decent performance, but still CPU usage goes high sometimes), with nvidia binary driver, random crashes, and also no boot logo+text console (but there is video signal, just some barcode like ugliness on the screen). I am wondering if I should stay with maverick, it seems to be more stable both video nvidia and ATI in my experience. Also I am very open to help Ubuntu like with the ATI radeon bug too, when I did the kernel git bisecting to find the problematic commit (bug 735126) . Just please tell me what I can do to help in debugging or so, and I try to support ubuntu development with the help I can do at least ... Thanks in advance!

please try to get a backtrace from the crash, by attaching gdb to the Xorg process, and waiting until it crashes. See https://wiki.ubuntu.com/X/Backtracing for instructions. You need to install xserver-xorg-core-dbg and libpixman-1-0-dbg to get the debugging symbols for the backtrace.

Now I discovered (I haven't noticed it earlier since I rebooted the system more times per day) that it seems after a system boot, X seems to be unstable, however after an X crash (when gdm screen is redisplayed and I must relogin then) it seems to be stable then. I've tried to do the backtrace (with using NoTrapSignals in config file as it was suggested). The interesting thing that one time I got SIGINT and after I got the backtrace and I quited with gdb (it was attached to already running process with ssh'ing from another machine) X continued to work without any crash ... I re-attach gdb to X process and after some minutes I noticed the "true" problem (SIGQUIT) however the backtrace itself seems to be very similar only the signals are different.

Anyway the result is attached (the other backtrace I was talking about is almost the same just it is SIGINT instead of SIGQUIT, I have checked it out with making diff between the traces).

I am wondering if my bug report (bug 764272) is the same issue or not ...

Sorry my very last comment here was posted here as a mistake, it was attended to write as response to another browser tab ... (isn't it possible to delete my own comment somehow?).

The only difference between first and further logins after boot that in case of first one X uses tty7, furher logins use tty8. I don't now if it's normal or it counts ...

So, any idea what can I help to eliminate this bug? Thanks in advance.

Looks like a GPU lockup rather than a crash actually.

Might test if a newer kernel makes a difference. http://kernel.ubuntu.com/~kernel-ppa/mainline/

If it does, you could do a git bisect to help locate the patch that fixes it. See https://wiki.ubuntu.com/Kernel/KernelBisection

I will check the new kernel out. What is odd to me, that in case of the second login after the boot (either with logging out first, or after the crash when gdm displayed again) it does not happen, only at the first time. The only difference I could notice that in case of the first login X seems to use tty7 while it's tty8 later (as I've commented already). I had the idea that maybe it's related to my boot problem that no splash, text consoles etc can be seen only some kind of odd "bar code" like image.

Anyway I would be keen to help with bisecting however it can be hard, since the crash sometimes need a hour or so to happen, so I can't be sure exactly (at least not within a "sane" time) that the given kernel is OK or not. But I will see ...

I have meet the same crashes but with intel chipset.
Here is a snip of xorg logs:

[ 5627.344] (II) XKB: reuse xkmfile /var/lib/xkb/server-4083DBECFEED155AEB26754E3529D374C9A5ED79.xkm
[ 6389.389] (II) XKB: reuse xkmfile /var/lib/xkb/server-4083DBECFEED155AEB26754E3529D374C9A5ED79.xkm
[ 6869.605]
Backtrace:
[ 6869.606] 0: /usr/bin/X (xorg_backtrace+0x3b) [0x80eab1b]
[ 6869.606] 1: /usr/bin/X (0x8048000+0x5fac8) [0x80a7ac8]
[ 6869.606] 2: (vdso) (__kernel_rt_sigreturn+0x0) [0xb4f40c]
[ 6869.606] 3: /usr/bin/X (_CallCallbacks+0x3e) [0x8074e1e]
[ 6869.606] 4: /usr/bin/X (WriteToClient+0x267) [0x80a7607]
[ 6869.606] 5: /usr/lib/xorg/modules/extensions/libdri2.so (ProcDRI2WaitMSCReply+0x62) [0x1b5bf2]
[ 6869.606] 6: /usr/lib/xorg/modules/extensions/libdri2.so (DRI2WaitMSCComplete+0x75) [0x1b3fc5]
[ 6869.606] 7: /usr/lib/xorg/modules/drivers/intel_drv.so (0x78e000+0x226bc) [0x7b06bc]
[ 6869.606] 8: /usr/lib/xorg/modules/drivers/intel_drv.so (0x78e000+0x87d2) [0x7967d2]
[ 6869.606] 9: /lib/i386-linux-gnu/libdrm.so.2 (drmHandleEvent+0xf5) [0x1af665]
[ 6869.606] 10: /usr/lib/xorg/modules/drivers/intel_drv.so (0x78e000+0x7937) [0x795937]
[ 6869.606] 11: /usr/bin/X (WakeupHandler+0x52) [0x8074602]
[ 6869.606] 12: /usr/bin/X (WaitForSomething+0x1ba) [0x80a1f3a]
[ 6869.606] 13: /usr/bin/X (0x8048000+0x27f1e) [0x806ff1e]
[ 6869.606] 14: /usr/bin/X (0x8048000+0x1a81c) [0x806281c]
[ 6869.606] 15: /lib/i386-linux-gnu/libc.so.6 (__libc_start_main+0xe7) [0x346e37]
[ 6869.606] 16: /usr/bin/X (0x8048000+0x1a411) [0x8062411]
[ 6869.606] Segmentation fault at address 0xb5a53008
[ 6869.606]
Caught signal 11 (Segmentation fault). Server aborting
[ 6869.606]

Strange think i notice, the xorg seems to be build as a server??

ptsiamis@wintermute:/var/log$ cat Xorg.0.log.old
[ 5408.461]
X.Org X Server 1.10.1
Release Date: 2011-04-15
[ 5408.461] X Protocol Version 11, Revision 0
[ 5408.461] Build Operating System: Linux 2.6.24-29-server i686 Ubuntu
[ 5408.461] Current Operating System: Linux wintermute 2.6.38-9-generic #43-Ubuntu SMP Thu Apr 28 15:25:15 UTC 2011 i686
[ 5408.461] Kernel command line: BOOT_IMAGE=/boot/vmlinuz-2.6.38-9-generic root=UUID=a2609e31-b915-4c34-a77a-9ad70cb5b8c7 ro quiet splash vt.handoff=7
[ 5408.461] Build Date: 19 April 2011 03:33:17PM
[ 5408.461] xorg-server 2:1.10.1-1ubuntu1 (For technical support please see http://www.ubuntu.com/support)

Anyway chipset:

[ 5408.468] (II) intel: Driver for Intel Integrated Graphics Chipsets: i810,
 i810-dc100, i810e, i815, i830M, 845G, 854, 852GM/855GM, 865G, 915G,
 E7221 (i915), 915GM, 945G, 945GM, 945GME, Pineview GM, Pineview G,
 965G, G35, 965Q, 946GZ, 965GM, 965GME/GLE, G33, Q35, Q33, GM45,
 4 Series, G45/G43, Q45/Q43, G41, B43, B43, Clarkdale, Arrandale,
 Sandybridge, Sandybridge, Sandybridge, Sandybridge, Sandybridge,
 Sandybridge, Sandybridge

[ 5408.476] (II) intel(0): Integrated Graphics Chipset: Intel(R) 965GM
[ 5408.476] (--) intel(0): Chipset: "965GM"

ptsiamis@wintermute:/var/log$ modinfo i915
filename: /lib/modules/2.6.38-9-generic/kernel/drivers/gpu/drm/i915/i915.ko
license: GPL and additional rights
description: Intel Graphics
autho...

I can confirm this bug. I'm also using an Intel chipset (GM965) and my log is pretty much the same as #11.

I'm having the same problem, also with the Intel chipset (GM45 in my case).

Here are some snippets from my Xorg.0.log

[ 4820.594]
X.Org X Server 1.10.1
Release Date: 2011-04-15
[ 4820.594] X Protocol Version 11, Revision 0
[ 4820.594] Build Operating System: Linux 2.6.24-29-server i686 Ubuntu
[ 4820.594] Current Operating System: Linux david-laptop 2.6.38-8-generic-pae #42-Ubuntu SMP Mon Apr 11 05:17:09 UTC 2011 i686
[ 4820.594] Kernel command line: root=UUID=3311ce7c-f839-469c-bd24-f00efdb49c5b ro quiet splash
[ 4820.594] Build Date: 19 April 2011 03:33:17PM
[ 4820.594] xorg-server 2:1.10.1-1ubuntu1 (For technical support please see http://www.ubuntu.com/support)
[ 4820.600] (II) intel: Driver for Intel Integrated Graphics Chipsets: i810,
 i810-dc100, i810e, i815, i830M, 845G, 854, 852GM/855GM, 865G, 915G,
 E7221 (i915), 915GM, 945G, 945GM, 945GME, Pineview GM, Pineview G,
 965G, G35, 965Q, 946GZ, 965GM, 965GME/GLE, G33, Q35, Q33, GM45,
 4 Series, G45/G43, Q45/Q43, G41, B43, B43, Clarkdale, Arrandale,
 Sandybridge, Sandybridge, Sandybridge, Sandybridge, Sandybridge,
 Sandybridge, Sandybridge
[ 4820.608] (II) intel(0): Integrated Graphics Chipset: Intel(R) GM45
[ 4820.608] (--) intel(0): Chipset: "GM45"91
[ 6214.642]
Backtrace:
[ 6214.642] 0: /usr/bin/X (xorg_backtrace+0x3b) [0x80eab1b]
[ 6214.642] 1: /usr/bin/X (0x8048000+0x5fac8) [0x80a7ac8]
[ 6214.642] 2: (vdso) (__kernel_rt_sigreturn+0x0) [0xb782540c]
[ 6214.642] 3: /usr/bin/X (_CallCallbacks+0x3e) [0x8074e1e]
[ 6214.642] 4: /usr/bin/X (WriteToClient+0x267) [0x80a7607]
[ 6214.642] 5: /usr/lib/xorg/modules/extensions/libdri2.so (ProcDRI2WaitMSCReply+0x62) [0xb73d4bf2]
[ 6214.642] 6: /usr/lib/xorg/modules/extensions/libdri2.so (DRI2WaitMSCComplete+0x75) [0xb73d2fc5]
[ 6214.642] 7: /usr/lib/xorg/modules/drivers/intel_drv.so (0xb736e000+0x226bc) [0xb73906bc]
[ 6214.643] 8: /usr/lib/xorg/modules/drivers/intel_drv.so (0xb736e000+0x87d2) [0xb73767d2]
[ 6214.643] 9: /lib/i386-linux-gnu/libdrm.so.2 (drmHandleEvent+0xf5) [0xb73bf665]
[ 6214.643] 10: /usr/lib/xorg/modules/drivers/intel_drv.so (0xb736e000+0x7937) [0xb7375937]
[ 6214.643] 11: /usr/bin/X (WakeupHandler+0x52) [0x8074602]
[ 6214.643] 12: /usr/bin/X (WaitForSomething+0x1ba) [0x80a1f3a]
[ 6214.643] 13: /usr/bin/X (0x8048000+0x27f1e) [0x806ff1e]
[ 6214.643] 14: /usr/bin/X (0x8048000+0x1a81c) [0x806281c]
[ 6214.643] 15: /lib/i386-linux-gnu/libc.so.6 (__libc_start_main+0xe7) [0xb753de37]
[ 6214.643] 16: /usr/bin/X (0x8048000+0x1a411) [0x8062411]
[ 6214.643] Segmentation fault at address 0xb2b99008
[ 6214.643]
Caught signal 11 (Segmentation fault). Server aborting

The crashes on intel are different, please follow https://bugs.launchpad.net/ubuntu/+source/xserver-xorg-video-intel/+bug/774978 for that one

Don't know where to post it, but I get some weird x crashes (gdm restarts), which all seem to be produced by keypresses. I don't remember which keys are these, but it happens when I type pretty fast. Logs seem to have no segmentation faults or other anomalies, clean as nothing happened.

I also have Nvidia-173 binary driver, Microsoft Natural 4K usb keyboard, if that matters. I use lubuntu with gdm.

Interesting, as far as I can remember, it was also my case, I mean X crashed during typing always (?). Nowdays I can't reproduce this problem anymore though ... (I am not sure if there is connection, but about since then when I applied GRUB_TERMINAL=console in /etc/default/grub since otherwise I have no text console just some odd garbage on the screen if I switch back to text consoles with CTRL-ALT-F1 or so).

My wife has reported this, X always crashes when she's Facebooking. The only relevant section I can find from Xorg.0.log.old is this:

[ 50080.223] 0: /usr/bin/X (xorg_backtrace+0x26) [0x4a2656]
[ 50080.223] 1: /usr/bin/X (0x400000+0x621ca) [0x4621ca]
[ 50080.223] 2: /lib/x86_64-linux-gnu/libpthread.so.0 (0x7f34ddf83000+0xfc60) [0
x7f34ddf92c60]
[ 50080.224] 3: /lib/x86_64-linux-gnu/libc.so.6 (__select+0x13) [0x7f34dcf8c123]
[ 50080.224] 4: /usr/bin/X (WaitForSomething+0x19b) [0x45c79b]
[ 50080.224] 5: /usr/bin/X (0x400000+0x2e032) [0x42e032]
[ 50080.224] 6: /usr/bin/X (0x400000+0x21a7e) [0x421a7e]
[ 50080.224] 7: /lib/x86_64-linux-gnu/libc.so.6 (__libc_start_main+0xff) [0x7f34
dcecceff]
[ 50080.224] 8: /usr/bin/X (0x400000+0x21629) [0x421629]
[ 50080.224]
Caught signal 3 (Quit). Server aborting

I thought it might have been a video driver issue, but this has happened with two different nvidia cards, with the legacy binary driver and the current one. Very frustrating.

Set to confirmed since LGB has posted a backtrace as requested.

LGB, did the new kernel fix it for you?

Hmm I haven't seen this kind of problem a since a while, let's hope than, it's fixed :) It was hard to tell when and how the problem can be triggered, but for sure it should have happened since some days at least, and nothing since a hmm month or so? I can't tell exactly, but as I've noted above in a comment: it seems setting grub to use only terminal mode "fixed" this ... I don't know how it's related to the kernel fix you've mentioned. Without setting grub to terminal I have not even usuable console if I switch there with CTRL+ALT+F1 for example

I'm also getting a X server crash randomly.

The trace in my log files is the same as reported by Geoffrey Pursell in comment #17.
Note that I also get the crash when I'm typing in the keyboard, most times (maybe always, I'm not sure) when I press the enter key. Of course, not always I press the enter key, it only happens sometimes. I also have a NVIDIA card with proprietary drivers.

I've the same issue (with the same log output).
I'm on a Nvidia integrate card and the bug happens both with and without using Compiz.
It seems to me that it happens mostly (maybe only, I'm not sure) when I press the enter key when writing in a text box on Firefox.

I haven't done a bisect yet, but I did manually install the Linux 3.0-rc1 kernel from Oneiric and haven't seen this crash for several days. The downside is that it looks like Natty's nvidia drivers won't build on top of the new kernel, so I have to use Nouveau.

Yeah NVIDIA hasn't released drivers compatible with linux 3.0 yet.

If building drivers to do the git bisection is hampering progress on this bug, alternatively you could install and boot from the .39 series at http://kernel.ubuntu.com/~kernel-ppa/mainline/. If v2.6.39-oneiric reproduces the crash, then test the .39-rc* kernels until you find one that works; that would narrow the fix to a specific one of the -rc's.

Some of you me-too'ing on this bug might want to look at bug #774978 which is a common crash and just now fixed in oneiric. If you see xrecord in your backtrace you might have that bug instead.

I'll try that. But isn't v2.6.39-oneiric a newer kernel than the -rc ones?

Same here, with the "Caught signal 3 (Quit). Server aborting". Never had the problem in Maverick. It does seem to happen when pressing Enter, and only once per session. It didn't happen in the first month or so that I was using Natty on this laptop, though...

I can also confirm having this issue. It happens only on upgraded machine, not on directly installed natty.

There appears to be various reasons for X/GDM crashing. I got some "signal 3"s in the gdm logs, while others (in other bugs got crashes with signal 11). The old bug described in
https://bugs.launchpad.net/ubuntu/+source/gdm/+bug/396226 was consistent with the behaviour I saw: logged out while typing, and if I then switched to tty1 (with Crtl-Alt-F1) it showed some garbled text. Disabling splash appears to have fixed it - did it yesterday, and no forced logouts today. For a discussion of the actual culprit, read the bug and comments.

Regarding my setup: I have two boxed upgraded to 11.04, and one fresh install. Only one of the updated boxes had the forced logout behaviour. All of the boxed had full disk encryption, but the one with problems also had an issue regarding the splash screen after entering passphrase: I had to jump out of the the splash screen, by pressing "Esc(ape)" at the latest right after entering the passphrase, otherwise the splash screen would lock up and I would have to reboot. Today, I could boot without any "security by obscurity" keypress magic (albeit text only boot) and I have not been kicked out of this session. I have done some other tweeks as well, but chaning between "Ubuntu Classic" and Unity made no difference. I don't know what I've done to trigger #396226, but I'm happy now, knowing that it was a software issue and not hardware as I suspected. Others might shed some light on why this issue reappearing on 11.04 (for some configurations).

LGB [Gábor Lénárt], this bug was reported a while ago and there hasn't been any activity in it recently. We were wondering if this is still an issue? If so, could you please test for this with the latest development release of Ubuntu? ISO images are available from http://cdimage.ubuntu.com/daily-live/current/ .

If it remains an issue, could you please run the following command in the development release from a Terminal (Applications->Accessories->Terminal), as it will automatically gather and attach updated debug information to this report:

apport-collect -p xorg-server REPLACE-WITH-BUG-NUMBER

Please note, given you already have done this in a prior release, performing this on a release prior to Trusty would not be helpful.

Thank you for your understanding.

Helpful bug reporting tips:
https://wiki.ubuntu.com/ReportingBugs

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