On Mon, August 1, 2005 10:36 am, Farkas Levente said:
> hi,
> as i already wrote mail to this list after we upgrade from centos-3 to
> centos-4 with kernel-unsupported we has a once a week kernel crash.
> we're not able to manage to find the real reason, but we've got this
> once a week and it's a very big problem. we need novel fs support so we
> can't use the standard kernel. i can't report it redhat because they
> refuse to look into it since it's centos' kernel and not rhel's one
> (what's more the unsupported version, so even the config are modified).
> anybody else send back to centos, but centos is just recompile the rh
> one:-( so what do you think what is the right place to report this error
> to be able to get some kind of help to find the problem (or even more to
> fix it:-)
> i'm just attached the today's crash' netdump log file:
> [Fwd: crash netdump black]
> thank you for your help in advance.
> yours.
>
> --
> Levente "Si vis pacem para bellum!"
> _______________________________________________
> CentOS mailing list
> CentOS@xxxxxxxxxx
> http://lists.centos.org/mailman/listinfo/centos
>
Let me preference this by saying ... I am not a kernel programmer :)
But, it seems to me that the issue that you have is caused by 1 of 2
components...
Either the 3w_xxxx driver module ... or the ACPI module and power saving.
To try and fix this I would try the following:
1. Make sure you have the latest BIOS for the motherboard (new BIOSes
frequently update the ACPI code for the board).
2. Make sure you have the latest BIOS for the RAID controller card.
3. Turn off / disable all power saving features in the BIOS.
4. See if the RAID manufacturer provides a driver for the RAID that is
newer than the one provided by the 2.6 kernel.
If it still happens, add this to your kernel line:
acpi=HT
(this will turn off several ACPI features, but still let the machine power
off)
5. If that doesn't work, try this on the kernel line (instead of acpi=HT):
noacpi
--
Johnny Hughes
<http://www.HughesJR.com/>
