You are not logged in.
Yesterday I updated packages and saw that nvidia driver was updated along with some others. I am usually using kde and thought crashes were caused by the newest kde-base (release 3), but then as crashes continued I used windowmaker for 2-3 hours today and the same kind of crash happened.
Only abnormal thing that can be seen in xorg log is "Caught signal 11. Server aborting." Where could I get the older nvidia driver to downgrade it?
Offline
to downgrade: pacman -U /var/cache/pacman/pkg/<pkg>.pkg.tar.gz
Offline
the problem is I have cleaned the pacman cache many times (during my problems with kde crashes before). I hope that some Arch repositories have older packages - at least that nluug.nl used to have older ones...
Anyway, I have some more info about the crash - it seems to be reated to mozilla (and perhaps also to firefox) somehow - perhaps something to do with qt programs. It has almost always happened when I have clicked some widget on the browser window.
EDIT: nluug has a version of nvidia driver from november,2004 - http://ftp.nluug.nl/os/Linux/distr/arch … pkg.tar.gz - I'll try this and see how it goes then.
EDIT: nluug's ftp site seems to have even 11.06. version of nvidia driver: ftp://ftp.nluug.nl/pub/metalab/distribu … a/os/i686/
- 11.06. version suddenly crashed,too. That earlier driver from november didn't start up properly so I used nv instead in xorg.conf that did not crash while I tested it
=> Now I'll try version 1.0.7174-1 from 01.04. That at least worked fine before.
EDIT: 1.0.7174 was not compatible with my self-made kernel and the module could not be inserted. I think my customized kernel could be the reason for crashes. I also had set nosuid set for /tmp - and now removed it.
EDIT: still crashed - now I removed the nvidia driver totally and I am using just xorg's default nv - driver.
Offline
I've been using xorg's nv driver now for some hours and almost forgot that crash. Only downside is that OpenGL doesn't work, but at least I can keep on working.
Offline
almost always happened when I have clicked some widget .
Do you have composite or render on in your xorg.conf? most of these problems come from that. try turning them off if they're on.
Offline
Now nv crashed,too. Option "RenderAccel" was commented out - now I set it to 0. I also set this to the end:
Section "Extensions"
Option "Composite" "Disable"
EndSection
I'll try again with those newest drivers from extra repo.
Offline
those things did not help - still got random crash. Now I installed current Arch stock kernel 2.6 - let's see how it goes. I left those options as before in xorg.conf.
Offline
That did not help either - again crashed to shell I was listening to music and a webpage was loading - didn't click on anything. Now what could I do - Xorg.log doesn't seem to have much information. I am starting to think that this is harware failure.
Offline
Does it always happen when you browse webpages and/or playing music ?
What if you leave it for some time without any applications working?
BTW I can crash like this (back to shell or even a reboot) while browsing for example wikipedia homepage with opera - it's realated to fonts and many different languages they use on their webpage (works fine with firefox).
Offline
Yes, it has always happened when I browsed webpages - like clicking on links. I have a hunch that this has something to do with gtk apps. Now I have been using Konqueror instead of mozilla and has worked fine for 2-3 hours.
I usually have computer always on - and it works fine when I don't do something. This fonts thing seems interesting - I don't know how to check it out though.
Offline
Yes, it has always happened when I browsed webpages - like clicking on links. I have a hunch that this has something to do with gtk apps. Now I have been using Konqueror instead of mozilla and has worked fine for 2-3 hours.
I usually have computer always on - and it works fine when I don't do something. This fonts thing seems interesting - I don't know how to check it out though.
Try turning off Anti-aliased fonts and see if that works. The RenderAccel option has somthing to do with how the fonts are drawn, iirc. You can try explicitly setting RenderAccel to "False" instead of "0" as well.
Hope that helps.
Offline
Now I had Xorg working inactively for 10 hours overnight. I only had KDE and TK/TCL programs on.
I had not set anti-aliasing at all and I am only using Smoothansi fonts for everything in KDE. Now I'll try setting RenderAccel to false - and then use mozilla again as I used to.
Offline
Seems like this problem of mine has widened - when I tried to start up X, everything seemed very slow and when I killed X, all partitions got set as faulty on one disk in raid-1 array I have. I removed them all and now using just one one disk - let's see how it works now. Perhaps this whole thing has been caused by one faulty disk or disk getting faulty.... I rememeber that during this all I had that problem before, too. Then I just re-attached those partitions to raid-1 and they were accepted. But not anymore - the partitions were not accepted anymore and said to be faulty.
Now I started smartmontools long tests (smartctl /dev/hdc -t long)
EDIT: it completed and here is the output of smartctl /dev/hdc -a I also tested the current harddrive (3 year-old Samsung).
Offline
crash again I'll try to get myself a new graphics card - Palit Radeon 9250 That nvidia I have been using is really old, bought 2001 and even tv-out has stopped working.
EDIT: still trying another thing: I pacman -Rd'd xorg and then reinstalled it. If it is something fonts - related, xorg seemed to update fonts cache, too. Hopefully this would fix this crash.
Also checked logs for that faulty hdc:
Jun 30 08:40:44 athena end_request: I/O error, dev hdc, sector 47102333
Jun 30 08:40:44 athena md: write_disk_sb failed for device hdc9
Jun 30 08:40:44 athena end_request: I/O error, dev hdc, sector 47102333
Jun 30 08:40:44 athena md: write_disk_sb failed for device hdc9
Jun 30 08:40:44 athena end_request: I/O error, dev hdc, sector 47102333
Jun 30 08:40:44 athena md: write_disk_sb failed for device hdc9
Jun 30 08:40:44 athena end_request: I/O error, dev hdc, sector 47102333
Jun 30 08:40:44 athena md: write_disk_sb failed for device hdc9
Jun 30 08:40:44 athena end_request: I/O error, dev hdc, sector 47102333
Jun 30 08:40:44 athena md: write_disk_sb failed for device hdc9
Jun 30 08:40:44 athena end_request: I/O error, dev hdc, sector 7190629
Jun 30 08:40:44 athena Operation continuing on 1 devices
Jun 30 08:40:44 athena end_request: I/O error, dev hdc, sector 7190637
Jun 30 08:40:44 athena RAID1 conf printout:
There were lots of similar log entries for different mds.
Spotted also something interesting in logs:
Jun 30 00:06:31 athena hdc: dma_timer_expiry: dma status == 0x21
Jun 30 00:07:13 athena hdc: DMA timeout error
Jun 30 00:07:13 athena hdc: dma timeout error: status=0xd0 { Busy }
Jun 30 00:07:13 athena hdc: DMA disabled
Jun 30 08:00:33 athena hdc: dma_timer_expiry: dma status == 0x21
Jun 30 08:25:22 athena md: running: <hdc1><hda1>
Jun 30 08:26:23 athena hdc: irq timeout: status=0xd0 { Busy }
Jun 30 08:33:24 athena hdc: dma_timer_expiry: dma status == 0x21
Jun 30 08:33:34 athena hdc: DMA timeout error
Jun 30 08:33:34 athena hdc: dma timeout error: status=0xd0 { Busy }
Jun 30 08:33:34 athena hdc: DMA disabled
Jun 30 08:34:09 athena hdc: status timeout: status=0x80 { Busy }
Jun 30 08:34:39 athena end_request: I/O error, dev hdc, sector 284095
Jun 30 08:34:39 athena end_request: I/O error, dev hdc, sector 284103
Jun 30 08:34:39 athena end_request: I/O error, dev hdc, sector 284111
Jun 30 08:34:39 athena end_request: I/O error, dev hdc, sector 284119
Jun 30 08:34:39 athena end_request: I/O error, dev hdc, sector 284127
Jun 30 08:34:39 athena end_request: I/O error, dev hdc, sector 284135
Jun 30 08:34:39 athena end_request: I/O error, dev hdc, sector 284143
Jun 30 08:34:39 athena end_request: I/O error, dev hdc, sector 284151
Jun 30 08:34:39 athena end_request: I/O error, dev hdc, sector 444207
Jun 30 08:34:39 athena end_request: I/O error, dev hdc, sector 76399411
Jun 30 08:34:39 athena end_request: I/O error, dev hdc, sector 6156561
Jun 30 08:34:39 athena md: write_disk_sb failed for device hdc10
Jun 30 08:34:39 athena disk 0, wo:1, o:0, dev:hdc1
Jun 30 08:34:39 athena disk 0, wo:1, o:0, dev:hdc5
Jun 30 08:34:39 athena disk 0, wo:1, o:0, dev:hdc7
Jun 30 08:34:39 athena end_request: I/O error, dev hdc, sector 75972211
Jun 30 08:34:39 athena end_request: I/O error, dev hdc, sector 76399411
Jun 30 08:34:39 athena md: write_disk_sb failed for device hdc10
Jun 30 08:34:39 athena end_request: I/O error, dev hdc, sector 47137979
Jun 30 08:34:39 athena disk 1, wo:1, o:0, dev:hdc10
Jun 30 08:37:16 athena md: unbind<hdc1>
Jun 30 08:37:16 athena md: export_rdev(hdc1)
Jun 30 08:37:33 athena md: unbind<hdc5>
Then errors.log for today:
Jun 30 08:40:44 athena md: errors occurred during superblock update, repeating
Jun 30 08:40:44 athena md: excessive errors occurred during superblock update, exiting
Jun 30 08:40:47 athena devfsd[1160]: action_compat: error unlinking: "vcs1" No such file or directory
Jun 30 08:40:47 athena devfsd[1160]: action_compat: error unlinking: "vcsa1" No such file or directory
Jun 30 08:40:49 athena dhcpcd[1506]: terminating on signal 15
Jun 30 08:42:06 athena devfsd[1156]: action_compat: error unlinking: "vcs1" No such file or directory
Jun 30 08:42:06 athena devfsd[1156]: action_compat: error unlinking: "vcsa1" No such file or directory
Jun 30 16:52:33 athena dhcpcd[1502]: terminating on signal 15
Jun 30 16:54:29 athena devfsd[1160]: action_compat: error unlinking: "vcs1" No such file or directory
Jun 30 16:54:29 athena devfsd[1160]: action_compat: error unlinking: "vcsa1" No such file or directory
and Xorg.0.log
Offline
Still the same - clicked on link and a few moments later back to shell again. Nothing new in error.log and in Xorg.0.log this "Caught signal 11. Server aborting". In everything.log there is just this:
Jun 30 19:58:50 athena (sven-4661): Received signal 15, shutting down cleanly
Jun 30 19:58:50 athena (sven-4661): Exiting
Offline
sven, before buying any new videocard, please consider that I have exactly the same problem with an ati 8500 using the stock x11 dri drivers.
I already noticed firefox just crashing randomly lately. But since this week I also get x11 crashes randomly. With the exact same entry in the log:
Fatal server error:
Caught signal 11. Server aborting
Offline
ok,skr,thanks for the info. I am now using Arch on a laptop that has Ati Radeon 9100 IGP and I got no crashes at all, that's why I thought this is just an nvidia issue. Well, I'll to use this laptop for more that just a couple of hours now and let's see if it is ok after all. I had that random crash occur on the other computer at least once during 2-3 hours.
Now I have been comparing the two computers - the packages they have:
xorg: xorg 11R6.8.2-2 on both
kde: only kdebase version differs - the faulty one has kdebase 3.4.1-3 while the laptop has kdebase 3.4.1-2 (but that probably isn't the issue because even when I used window maker, that crash occured).
I haven't updated laptop for some time after that crash occured, so pacman -Syu offers these:
bin86-0.16.17-1 dirmngr-0.9.2-1 file-4.14-1
freetype2-2.1.10-1 fribidi-0.10.5-1 gimp-2.2.7-1 gnome-vfs-2.10.1-4
gnupg2-1.9.17-1 ivman-0.6.4-1 kdebase-3.4.1-3 lesstif-0.94.4-2
libassuan-0.6.10-1 libgpg-error-1.1-1 man-pages-2.05-1
openoffice2-1.9.109-1 pacman-2.9.6-2 pcre-6.1-1 procps-3.2.5-2
The stuff I use on both computers is pretty much the same
Offline
I've been searching for this signal 11 error on Google and will collect here the links I've found:
On Redhat EDIT: A Redhatter seems to have got the TWM to work for almost two days, but TWM doesn't look too good
xorg doesn't like gcc 3.4.3 optimizations
Offline
I could not resist to buy a cheap Radeon 9200SE I had read that 2D image would be sharper than with nvidia cards -but did not believe it would be _that_ much.
I set 'radeon' as driver in xorg.conf and could start KDE without any glitch. Now let's see if it still crashes...
EDIT: one capacitor on that old gf2mx was almost popping open - probably one of those faulty ones that were common on motherboards 2-3 years ago. *thumbs up* - hopefully that gf2mx was still the reason for crashes
EDIT: now I started to use ATI's proprietary driver instead. I used the installer and all went well, TV-out works already.
Offline
I have used the computer for many hours - doing surfing and shell scripting and all the usual stuff. I also left the computer on overnight. Not a single crash has occured since I switched the cards - so I believe now the reason was that faulty old nvidia. Thanks to everyone who was with me
Offline