Yesterday I got my first oops ever. Running 2.0.35. Dual pent/200. HP.
Slackware. Buslogic Scsi. PCnet32 ethernet. The machine runs squid, apache
and other apps. Not a very heavy load. 110Meg ram. The system kept going,
and restarted Squid. I happened to be in the same room when it happened,
and heard the disk activity as squid reloaded, and saw the screen.
general protection: 0000
CPU: 0
EIP: 0010:[<00141f8e>]
EFLAGS: 00010207
eax: f00010a9 ebx: 01d6d018 ecx: 00000000 edx: 00000001
esi: 01d6d0d8 edi: 00000000 ebp: 00000002 esp: 0531ff28
ds: 0018 es: 0018 fs: 002b gs: 002b ss: 0018
Process squid (pid: 11704, process nr: 31, stackpage=0531f000)
Stack: 01d6d018 0014eceb 01d6d018 01d6d018 02203190 02203190 001585f9
01d6d018
00000000 02203100 00000000 0013fad4 02203190 00000000 02203100
02203100
02203100 bffff9dc 0013fd0d 02203190 05c988c4 0012aaf4 02203100
05c988c4
Call Trace: [<0014eceb>] [<001585f9>] [<0013fad4>] [<0013fd0d>]
[<0012aaf4>] [<0012ab74>] [<07807000>]
[<0012abf0>] [<0010ae0b>]
Code: 89 10 8b 93 cc 00 00 00 8b 43 74 50 6a 01 8b 41 3c 50 0f b7
Aiee, killing interrupt handler
On 29 Jul 98, at 21:56, Matthew Hawkins wrote:
Date sent: Wed, 29 Jul 1998 21:56:58 +1000
From: Matthew Hawkins <matt@mail.goldweb.com.au>
To: "Stephen C. Tweedie" <sct@redhat.com>
Copies to: linux-kernel@vger.rutgers.edu
Subject: Re: 2.0.35 gp/oops
> On Mon, 27 Jul 1998, Stephen C. Tweedie wrote:
> > Don't bother playing around with software too much --- this is almost
> > certainly hardware. Check the CPU fan, see if it runs with cache
> > disabled, but be sure it's your box if you're getting those kfree and
> > freelist errors.
>
> CPU fan is okay. Even swapped CPU's, the newer, faster CPU just made
> it crash in a shorter time period ;)
> I turned off the L2 cache in the BIOS, and funnily enough, it hasn't
> fallen over yet (and I'd expect it to have by now). I'm giving it a
> week - that's about the longest I've ever seen squid running without
> dying. The freelist error I think was the outcome of tripping the
> route cache bug in the 2.0.34-pre patches, I haven't seen it yet with
> 2.0.35. The oops I posted was a once off, looks scsi-related to me,
> probably something to do with the aic7xxx driver which always seems
> to come to grief frequently on this list.
>
> There's probably some timing issue related to why squid gets SIGSEGV.
> It's freaky how the cache_mem setting is set to exactly the amount of
> RAM that the L2 cache can cache.
>
> --
> Matthew Hawkins <matt@goldweb.com.au> |
> WWW: http://www.goldweb.com.au/~matt/ | "Do not taunt happy fun troll."
> UID 0 @ Goldweb Internet +61262530059 |
> PGP: 1024/273E35E1 - 01 8D 6C 62 4C D1 05 3D 0F 59 5B E3 81 9F 59 B9
>
> -
> To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
> the body of a message to majordomo@vger.rutgers.edu
> Please read the FAQ at http://www.altern.org/andrebalsa/doc/lkml-faq.html
------------------Mailed via Pegasus 2.53 & Mercury 1.30---------------
A.J.Scott@casdn.neu.edu Fax (617)373-2942
Andrew Scott Tel (617)373-5278 _
Northeastern University--138 Meserve Hall / \ /
College of Arts & Sciences-Deans Office / \ \ /
Boston, Ma. 02115 http://www.casdn.neu.edu / \_/
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.altern.org/andrebalsa/doc/lkml-faq.html