Re: oops with 2.4.17 + mini-ll patch

Matt Bernstein (matt@theBachChoir.org.uk)
Thu, 10 Jan 2002 23:53:05 +0000 (GMT)


At 12:54 -0800 Andrew Morton wrote:

>> Dual PIII 1GHz, modular everything inc. ATA/IDE (VIA); SCSI (gdth.o);
>> NFSv3 (udp, client only); autofs4; ext2 only for local fs. Debian woody.
>
>It could be a timer deletion race. There are still zillions of these,
>but nobody ever encounters them

This was the second oops the kernel produced--it wasn't syncing so I
copied it from the screen. The first one I didn't record as I'd thought it
might be related to a power surge we had.

>What was the system doing at the time? Do you think a module could
>have been in the process of unloading? autofs unmount, something
>like that?

Hmm.. I don't think modules were unloading, though it's very possible that
usage counts were going down for exactly that sort of reason. The first
oops occurred within minutes of system boot; the second was a few hours. I
think the common element might have been mildly high system load--this
machine is used by a PhD student running models on a mysql database on
the SCSI system.

Anyway, it's gone back to 2.4.9-ac18 now which it was running happily
until the power problems. I do have two other machines with the exact same
2.4.17 build (but different initrds) which both auto{,u}mount a lot more
than the problem box, but they have other differences (no SCSI, different
VIA IDE chipset..)

Cheers though,

Matt

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/