Red Hat Bugzilla – Bug 52768
kernel crash during normal usage
Last modified: 2008-08-01 12:22:51 EDT
From Bugzilla Helper: User-Agent: Mozilla/4.77 [en] (X11; U; Linux 2.4.3-12 i586) Description of problem: ever since upgrading my system to redhat-7.1, i am expreience very annoying and regular system crashes. normally, this is in the X process (as seen in the attachment) but it can happen in other processes also. sometimes, i get the following message: kernel BUG at page_alloc.c:82 but normally, it is similar to what you see in the log in the attachment. this is, in fact, the third attempt at entering this bug. the system crashed the previous two times before i could complete it... Version-Release number of selected component (if applicable): How reproducible: Always Steps to Reproduce: 1. load up evolution 2. load up galeon 3. start using it Actual Results: crash Expected Results: not crashed ;-) Additional info:
can you past the output of lsmod ? this looks like the bug in the nvidia binary only driver; are you using that ?
Created attachment 29956 [details] log containing the /var/log/messages of the most recent crashes
[root@jigra /root]# lsmod Module Size Used by tuner 3952 0 (autoclean) i2c-core 12640 0 (autoclean) [tuner] emu10k1 45328 1 (autoclean) soundcore 3408 4 (autoclean) [emu10k1] nfs 70720 1 (autoclean) lockd 47888 1 (autoclean) [nfs] sunrpc 55696 1 (autoclean) [nfs lockd] ne 6336 1 (autoclean) 8390 5632 0 (autoclean) [ne] ipchains 27648 0 (unused) usb-uhci 19472 0 (unused) usbcore 46656 1 [usb-uhci]
sorry, i should have added: no, i'm not using the nvidia driver. however, i now notice that since i simply used netscape 4.77 insted of galeon to try and create this bug, i have managed to create it and it seems to be working ok. i haven't started evolution yet either. not sure if that's related or not. can it be?
It shouldn't be. Can you give a bit more information about this system ? eg cpu type, memory etc
cpu: K6/2@500 vid: matrox millenium (x2, but this was happening before i added the second one) mem: i actually have 4 configs i've been testing. i have 2 128MB sticks and a 256MB stick. normally, i have been running with one 128 and the 256, but i have tried using only, trying both the 128 and the 256 separately. still crashes in a very similar manner.
I think I found a codebug; I'm building a new kernel with a patch and I hope you're willing to test that to see if it fixes it.
you betcha i'm willing to try it. all the sensitive data is being back up hourly, just in case. i'd HATE to have to reinstall an older version... thanks!
Ok the i586 kernel is done, the i686 one is still building (there is a bit of a queue in the buildsystem and I had to build on my workstation), the kernel is versioned 2.4.3-18 and can be found at http://people.redhat.com/arjanv/testkernels/
ok, new kernel installed and currently operating. so far, it looks very good. i will keep y'all informed. couple points: 1) is it ok that i'm using a K6/2 and this is an i586 kernel? 2) there was one case that seemed to trigger a crash somewhat regularly. while using galeon, if i tried resizing and/or scrolling while the page was still loading, it seemed to crash the system more frequently. just now, while loading the bugquery page ( takes a nice relatively long time to load up and layout the form ) i tried scrolling around. i had galeon crash, but this time it just sat with a unupdated window. normally, when it crashes i get a little dialog box popup, this was the very first time i've had it just stop updating. SO, that gives me a very good feeling about this update. thank you so very much for your incredibly prompt and helpful attention to this matter. i promise i will report any thing i find as soon as it happens. now, i'm off to try and get some work done!
hmm, had a similar crash this afternoon, i will attach the relevant log separately. when this happened, the machine was doing too much. i had left it for a little while. when i came back, i had been dropped back to the gdm login screen. i am still convinced it seems better though. let me know if there is anything else i can do.
Created attachment 30174 [details] /var/log/messages for latest crash(es)
after using my machine again for less than two hours, it once again crashed. i have attached the log from a crash on thursday before i took off for the weekend, then the one from this evening. i really do need to try and be able to actually work. i'm seeing what i can do about compiling the 2.2.19 kernel from kernel.org for myself to use, as well as 2.4.9 to see if that helps. if there is still anyone working on this, i will still be willing to try new kernels if need be, i just can't be having crashes every 2-6 hours, its just too painful.
Created attachment 30909 [details] latest crash log
Thanks for the bug report. However, Red Hat no longer maintains this version of the product. Please upgrade to the latest version and open a new bug if the problem persists. The Fedora Legacy project (http://fedoralegacy.org/) maintains some older releases, and if you believe this bug is interesting to them, please report the problem in the bug tracker at: http://bugzilla.fedora.us/