https://bugzilla.novell.com/show_bug.cgi?id=216149 ------- Comment #2 from rschulz@sonic.net 2006-10-29 11:12 MST ------- Created an attachment (id=102966) --> (https://bugzilla.novell.com/attachment.cgi?id=102966&action=view) /var/log/Xorg.log.0.old following X server crash -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug, or are watching someone who is.

https://bugzilla.novell.com/show_bug.cgi?id=216149 ------- Comment #4 from rschulz@sonic.net 2006-10-29 13:20 MST ------- Created an attachment (id=102977) --> (https://bugzilla.novell.com/attachment.cgi?id=102977&action=view) xorg.conf in use when crash occurred per request from sndirsch@novell.com -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug, or are watching someone who is.

https://bugzilla.novell.com/show_bug.cgi?id=216149 sndirsch@novell.com changed: What |Removed |Added ---------------------------------------------------------------------------- Attachment #102977|0 |1 is obsolete| | -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug, or are watching someone who is.

https://bugzilla.novell.com/show_bug.cgi?id=216149 rschulz@sonic.net changed: What |Removed |Added ---------------------------------------------------------------------------- CC| |rschulz@sonic.net ------- Comment #6 from rschulz@sonic.net 2006-10-30 09:27 MST ------- I wanted to add that this symptom is repeatable, but occurs only after Firefox has been running for a while, usually just idling. In the cases I've witnessed, Firefox does not get killed until I switch to the virtual desktop on which its displayed, at which time it is killed immediately without even getting a chance to redraw anything. After this, other applications start getting killed, too. I've noticed GKrellM killed, but not Konsole. And once this occurs, it's never long before the entire X login session is terminated and I'm presented with the login screen / desktop manager. It may or may not be relevant that the time elapsed has always been enough for both the screen saver and then DPMS to activate. RRS -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug, or are watching someone who is.

https://bugzilla.novell.com/show_bug.cgi?id=216149 ------- Comment #8 from rschulz@sonic.net 2006-10-30 14:38 MST ------- I strongly doubt that's the case. The system has 4GB of physical ram (and _is_ running a bigsmp kernel) and in the cases I've observed it so far, not even half of the system's RAM had be used for anything. After the first couple of occurrences, I stopped using the system for much of anything. I log in, which starts a default complement of KDE desktop applications and processes plus GKrellM, Konsole and a couple of local Konqueror windows and then I manually launch Firefox. That (and time) is all it takes to reliably trigger the symptom. RRS -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug, or are watching someone who is.

https://bugzilla.novell.com/show_bug.cgi?id=216149 ------- Comment #10 from sndirsch@novell.com 2006-10-31 05:15 MST ------- Ok. I suggest to attach the kernel log lines of /var/log/messages when this happens. I hope this gives us a better understanding. -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug, or are watching someone who is.

https://bugzilla.novell.com/show_bug.cgi?id=216149 mhopf@novell.com changed: What |Removed |Added ---------------------------------------------------------------------------- CC| |lfriedman@nvidia.com ------- Comment #12 from mhopf@novell.com 2006-10-31 05:27 MST ------- Hm. Ok, you obviously had the binary only driver installed some time ago (because the driver is named 'nvidia' in the config). Can you do this once again, so we know whether this is a general Xserver issue, or an nv driver problem? Adding NVIDIA to this bug. Maybe they have an idea (at least if it is driver specific). -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug, or are watching someone who is.

https://bugzilla.novell.com/show_bug.cgi?id=216149 ------- Comment #14 from sndirsch@novell.com 2006-10-31 07:30 MST -------

... excerpt of the kernel messages file ... I attached originally included everything pertinent. Cannot find this information in this bugreport. :-(

-- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug, or are watching someone who is.

https://bugzilla.novell.com/show_bug.cgi?id=216149 ------- Comment #16 from sndirsch@novell.com 2006-10-31 08:07 MST ------- You mentioned "resource-exhausted errors". Therefore I thought you've seen a similar kernel message. Well, when there is nothing logged during X clients are killed and Xserver crashes ... I thought there is, but it seems I was wrong. I'll try to reproduce ... -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug, or are watching someone who is.

https://bugzilla.novell.com/show_bug.cgi?id=216149 rschulz@sonic.net changed: What |Removed |Added ---------------------------------------------------------------------------- Status|NEEDINFO |ASSIGNED Info Provider|rschulz@sonic.net | ------- Comment #18 from rschulz@sonic.net 2006-10-31 09:43 MST ------- OK. I just reproduced the problem again, and no output at all was produced in /var/log/messages. RRS -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug, or are watching someone who is.

https://bugzilla.novell.com/show_bug.cgi?id=216149 ------- Comment #20 from rschulz@sonic.net 2006-10-31 09:51 MST ------- Here are the messages in ~/.xsession-errors that result from individual applications that termined due to X server errors: The program 'Gecko' received an X Window System error. This probably reflects a bug in the program. The error was 'BadAlloc (insufficient resources for operation)'. (Details: serial 27089 error_code 11 request_code 53 minor_code 0) (Note to programmers: normally, X errors are reported asynchronously; that is, you will receive the error a while after causing it. To debug your program, run it with the --sync command line option to change this behavior. You can then get a meaningful backtrace from your debugger if you break on the gdk_x_error() function.) The program 'gkrellm' received an X Window System error. This probably reflects a bug in the program. The error was 'BadAlloc (insufficient resources for operation)'. (Details: serial 2657852 error_code 11 request_code 53 minor_code 0) (Note to programmers: normally, X errors are reported asynchronously; that is, you will receive the error a while after causing it. To debug your program, run it with the --sync command line option to change this behavior. You can then get a meaningful backtrace from your debugger if you break on the gdk_x_error() function.) RRS -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug, or are watching someone who is.

https://bugzilla.novell.com/show_bug.cgi?id=216149 ------- Comment #22 from rschulz@sonic.net 2006-10-31 10:06 MST ------- I just reinstalled the nVidia driver, rebooted and logged in again. If the bug reproduces here, it will probably take a couple of hours to manifest. RRS -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug, or are watching someone who is.

https://bugzilla.novell.com/show_bug.cgi?id=216149 ------- Comment #24 from lfriedman@nvidia.com 2006-10-31 12:09 MST ------- This seems like an X and/or kernel bug at this point, although that mtrr type mismatch leads me to wonder if there's a BIOS problem too. -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug, or are watching someone who is.

https://bugzilla.novell.com/show_bug.cgi?id=216149 ------- Comment #26 from rschulz@sonic.net 2006-10-31 16:07 MST ------- P.S. If the change summaries that describe BIOS updates between version 0507 (what I was running before Lonni suggested a BIOS problem) and version 0804 (what I'm running now) might be instructive, they can be found at http://support.asus.com/download/download.aspx?SLanguage=en-us&model=P5B%20Deluxe RRS -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug, or are watching someone who is.

https://bugzilla.novell.com/show_bug.cgi?id=216149 ------- Comment #28 from rschulz@sonic.net 2006-10-31 17:49 MST ------- Another update: With the nVidia proprietary driver version 8774 the symptom occurs in the same manner and after approximately the same amount of elapsed time as with the open driver. The diagnostics in ~/.xsession-errors are the same as with the nv driver. Furthermore, those same diagnostics are recorded when trying to launch applications once the first application crash occurs. To wit: The program 'galculator' received an X Window System error. This probably reflects a bug in the program. The error was 'BadAlloc (insufficient resources for operation)'. (Details: serial 960 error_code 11 request_code 53 minor_code 0) (Note to programmers: normally, X errors are reported asynchronously; that is, you will receive the error a while after causing it. To debug your program, run it with the --sync command line option to change this behavior. You can then get a meaningful backtrace from your debugger if you break on the gdk_x_error() function.) Another thing I'll note is that it seems that already running applications die upon switching from one virtual desktop _to_ the one on which they're displayed. That certainly characterizes the GKrellM and Firefox crashes. The "mtrr" diagnostic does not occur when applications die. This time I could not induce the entire session to die, but KWin crashed. When I finally ended the session, /var/log/messages did record this message: Oct 31 16:47:03 smiley kernel: mtrr: type mismatch for c0000000,10000000 old: uncachable new: write-combining RRS -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug, or are watching someone who is.

https://bugzilla.novell.com/show_bug.cgi?id=216149 ------- Comment #30 from mhopf@novell.com 2006-11-02 03:51 MST ------- (In reply to comment #24)

This seems like an X and/or kernel bug at this point, although that mtrr type mismatch leads me to wonder if there's a BIOS problem too.

Probably yes. I doubt it is a kernel bug, the kernel isn't much involved when the nv driver is used. If it was a kernel agp module issue, the crash would probably occur at startup time. -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug, or are watching someone who is.

https://bugzilla.novell.com/show_bug.cgi?id=216149 ------- Comment #32 from mhopf@novell.com 2006-11-02 08:58 MST ------- The mtrr report is only printed on shutdown according to comment #28, so i doubt this is an BIOS issue. So it probably is an Xorg server bug. -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug, or are watching someone who is.

https://bugzilla.novell.com/show_bug.cgi?id=216149 ------- Comment #34 from rschulz@sonic.net 2006-11-02 10:09 MST ------- Will do. For the record, it appears the correct URL is: ftp://ftp.suse.com/pub/people/sndirsch/RPMS/bug216149 Given the typical time-frame for this symptom to manifest, testing both could be complete within a few hours (if they both fail promptly) or may take the rest of the day (it's mid-morning here). RRS -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug, or are watching someone who is.

https://bugzilla.novell.com/show_bug.cgi?id=216149 sndirsch@novell.com changed: What |Removed |Added ---------------------------------------------------------------------------- Status|ASSIGNED |NEEDINFO Info Provider| |rschulz@sonic.net -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug, or are watching someone who is.

https://bugzilla.novell.com/show_bug.cgi?id=216149 rschulz@sonic.net changed: What |Removed |Added ---------------------------------------------------------------------------- Status|NEEDINFO |ASSIGNED Info Provider|rschulz@sonic.net | ------- Comment #37 from rschulz@sonic.net 2006-11-02 17:08 MST ------- I have now been running 71before72RC1 for four hours without problem, which is at least twice as long as the failing versions have ever survived. % init 3 % rpm -U --oldpackage xorg-x11-server-71before72RC1.i586.rpm Starting SuSEconfig, the SuSE Configuration Tool... Running module permissions only Reading /etc/sysconfig and updating the system... Executing /sbin/conf.d/SuSEconfig.permissions... Checking permissions and ownerships - using the permissions files /etc/permissions.d/postfix /etc/permissions.d/squid /etc/permissions.d/susehelp /etc/permissions.d/tetex /etc/permissions.d/xorg-x11-server.easy /etc/permissions /etc/permissions.easy /etc/permissions.local setting /usr/bin/Xorg to root:root 4711. (wrong permissions 0711) Finished. % rpm -qa |egrep -i xorg-x11-server xorg-x11-server-7.1-42 xorg-x11-server-sdk-7.2-9 % init 5 ... RRS -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug, or are watching someone who is.

https://bugzilla.novell.com/show_bug.cgi?id=216149 ------- Comment #39 from sndirsch@novell.com 2006-11-02 19:44 MST ------- (In reply to comment #38)

Ok. This would mean we had a Xserver regression right after updating from xorg-server-1.1.99.3.tar.bz2 (2006-07-30) to xorg-server-1.1.99.901.tar.bz2 (7.2RC1, 2006-10.14).

This one in xorg-x11-server.changes: ------------------------------------------------------------------- Sat Oct 14 06:32:52 CEST 2006 - sndirsch@suse.de - updated to X.Org 7.2RC1 -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug, or are watching someone who is.

https://bugzilla.novell.com/show_bug.cgi?id=216149 rschulz@sonic.net changed: What |Removed |Added ---------------------------------------------------------------------------- Severity|Major |Critical ------- Comment #41 from rschulz@sonic.net 2006-11-12 13:41 MST ------- Elevating to "Critical." -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug, or are watching someone who is.

https://bugzilla.novell.com/show_bug.cgi?id=216149 trey@novell.com changed: What |Removed |Added ---------------------------------------------------------------------------- CC| |trey@novell.com ------- Comment #43 from trey@novell.com 2006-11-15 11:19 MST ------- I'm seeing similar behavior with X locking after a period of a few hours with openSUSE 10.2 beta 1 but also with my openSUSE 10.1 install as well. In fact, this is where I *first* experienced it. Sometimes it would be X restarting in the middle of me writing an e-mail or browsing a web page, taking me back to gdm. Recently, it's been a lot of X just 'locking up'. Apps won't respond and in most cases keyboard response might go as well. On 10.1 this is with xorg-x11-*-6.9.0 packages and an ATI Radeon 9550 (RV350-4153) video card. -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug, or are watching someone who is.

https://bugzilla.novell.com/show_bug.cgi?id=216149 rschulz@sonic.net changed: What |Removed |Added ---------------------------------------------------------------------------- Status|NEEDINFO |ASSIGNED Info Provider|rschulz@sonic.net | ------- Comment #46 from rschulz@sonic.net 2006-11-24 12:39 MST ------- It appears the problem has been resolved in RC1. I stayed logged in since the first boot after upgrading to RC1. On the hunch that having Firefox running might have been a factor in leading to the X server crash, I didn't start it right away. There was no failure for about 11 hours like that. Then I started Firefox and it has been over four hours since then and still it hasn't crashed. Usually one and a half to two hours was enough for it to crash. I think this bug can tentatively be considered fixed. RRS -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug, or are watching someone who is.