Re: Oops on 2.6.9-ac16: xfs, dm and md may be involved

[Nathan Scott <nathans@sgi.com>]

On Tue, Dec 21, 2004 at 07:57:54PM +0100, Joerg Sommrey wrote:
> Hello,
> 
> last night my box died with a kernel oops.  There was a backup
> running at that time. The setup:
> - 2 SATA disks + 1 SCSI disk
> - SATA partitions build up md-raid-arrays (level 0 and 1)
> - md-raid-devices and SCSI partitions are physical volumes for dm
> - dm logical volumes are used for xfs filesystems
> - backup is done on dm-snapshots of those filesystems
> ...
> syslog:
> Dec 21 03:28:09 bear kernel: Unable to handle kernel paging request at virtual address 6b6b6b73

Hmm, looks like a memory poisoning pattern there.

> Dec 21 03:28:10 bear kernel: EIP:    0060:[__origin_write+112/608]    Tainted: P   VLI
> Dec 21 03:28:10 bear kernel: EFLAGS: 00010217   (2.6.9-ac16) 
> Dec 21 03:28:10 bear kernel: EIP is at __origin_write+0x70/0x260
> Dec 21 03:28:10 bear kernel: eax: 6b6b6b4b   ebx: ccd2f888   ecx: f0f55190   edx: ffffffff
> Dec 21 03:28:10 bear kernel: esi: 6b6b6b4b   edi: dc1eebd8   ebp: c4593920   esp: f7487d74
> Dec 21 03:28:10 bear kernel: ds: 007b   es: 007b   ss: 0068
> Dec 21 03:28:10 bear kernel: Process xfsbufd (pid: 849, threadinfo=f7486000 task=c1b9e070)

>From XFS's point of view, we are in the middle of submitting
a delayed metadata write down to the driver.  Something seems
to have gone wrong inside "__origin_write" (dm-snap.c), which
looks like its doing snapshot stuff inside device mapper?

cheers.

-- 
Nathan