Hard lock plus Caps and Scroll Lock flashing (kernel panic?)
I have an OpenSUSE 10.3 x64 install which seems to be panicking the kernel. The screen will freeze and the caps lock and scroll lock lights will flash on and off. This seems to happen sporadically, even when the computer isn't under much load.
Is there a way I can narrow down the problem? I tested the RAM so I don't think it's that.
Here's some generic questions:
Any weird motherboard, unstable kernel modules, exotic hardware or software running?
Did you check OpenSUSE mailing lists / bug tracking system for people experiencing similar problems?
Did this happen always or only after a HW addition, SW update or kernel update? (If so, what changed?)
Does a previous OpenSUSE 10.3 x64 kernel version run w/o glitches?
Does a vanilla kernel.org kernel version run w/o glitches?
If you have a vanilla kernel.org kernel running tainted with a proprietary driver like Nvidia's, does it run OK w/o that driver?
Does your syslog contain a stack trace? (If yes run ksymoops on it and save the result for reporting)
unSpawn, thanks for the reply.
I'm trying to use this box as a fileserver, so I have a 3ware 9650 se installed (using the 3w-9xxx.ko that came with the OS--I had a bit of trouble getting it to work, but now it seems to recognize it). However, I saw this problem even before I had the kernel module installed, so I don't think it's the driver's fault.
I did some Googling and searching on opensuse.org and didn't find anything.
This happened always, AFAIK. The system is very new (just built it and installed the OS about a week ago) and I've been seeing this problem since the first day I installed it.
I haven't tried any other kernels. It's hard to repro the problem (one time I left it on all night and it ran fine, only to conk later that morning when it was just sitting there running the screensaver), but I'll try some other kernels and see what develops.
AFAIK the kernel isn't tainted with proprietary drivers. I do have an ATI gfx card (nothing fancy, just a twenty-buck PCI VGA card I picked up online), but I don't think it's running any special drivers for it.
I'll check the syslog tomorrow. I also want to write down the CPU temps when it happens so I can see if overheating might be an issue.
OK, here's some data:
* The CPU isn't overheating...I reproduced the problem twice and both times the CPU was chillin' at a cool 40C. I have a pretty beefy HSF on it, so that's no surprise, but good to know.
* Nothing suspicious in /var/log/messages. But...
* I got "lucky" and had it happen while booting, during an fsck (>_<), so I could see the actual text output. It says:
Unable to handle kernel paging request at ffff80ff810137b8 RIP:
Oops: 0002  SMP
last sysfs file: /devices/pci0000:00:02.0/modalias
Modules linked in: usbhid hid ff_memless sd_mod ehci_hcd ohci_hcd usbcore 3w_9xxx reiserfs edd ext3 mbcache jbd fan sata_nv pata_amd libata scsi_mod thermal processor
Pid 0, comm: swapper Tainted: G N 18.104.22.168-0.3-default #1
[[[Whole lotta registers]]]
Process swapper (pid: 0, threadinfo [blah], task [other blah])
Stack [[Lots of addresses]]
<IRQ> [addr] default_idle+0x0/0x3d
<EDI> [addr] default_idle+0x29/0x3d
Code: [[some hex numbers]]
Kernel panic - not syncing: Attempted to kill the idle task!
I typed all that in myself from reading the screen, so apologies for any typos and for not bothering to type in all those 64-bit addresses.
Potentially relevant hardware:
AMD Athlon64 FX-55 (single-core)
Primary HD is a 60GB SATA drive plugged into the motherboard.
Any help/insights would be greatly appreciated!
Thanks for posting info. BTW, next time maybe just take a cam shot and upload it? I'm waiting for you to post more info, esp. running a vanilla kernel.org kernel and posting more details because at this point w/o leads there is no "potentially relevant" SW/HW, just SW/HW.
|All times are GMT -5. The time now is 05:21 PM.|