* tracking parity mismatches to a failing disk @ 2012-05-09 19:16 Michael Peterson 2012-06-04 4:17 ` NeilBrown 0 siblings, 1 reply; 2+ messages in thread From: Michael Peterson @ 2012-05-09 19:16 UTC (permalink / raw) To: linux-raid I'm trying to identify the source of a non-zero mismatch_cnt on a six-disk RAID6 md array. I think I've identified it as /dev/sdf since that is the only device with a non-zero value in /sys/block/md0/md/dev-sd*/errors. Is that the right way to track parity mismatches back to a disk? OS is CentOS 5.8 -> mdadm 2.6.9 Thanks, Michael ^ permalink raw reply [flat|nested] 2+ messages in thread
* Re: tracking parity mismatches to a failing disk 2012-05-09 19:16 tracking parity mismatches to a failing disk Michael Peterson @ 2012-06-04 4:17 ` NeilBrown 0 siblings, 0 replies; 2+ messages in thread From: NeilBrown @ 2012-06-04 4:17 UTC (permalink / raw) To: Michael Peterson; +Cc: linux-raid [-- Attachment #1: Type: text/plain, Size: 543 bytes --] On Wed, 09 May 2012 15:16:08 -0400 Michael Peterson <Michael.Peterson@duke.edu> wrote: > I'm trying to identify the source of a non-zero mismatch_cnt on a > six-disk RAID6 md array. I think I've identified it as /dev/sdf since > that is the only device with a non-zero value in > /sys/block/md0/md/dev-sd*/errors. Is that the right way to track parity > mismatches back to a disk? > > OS is CentOS 5.8 -> mdadm 2.6.9 There is a program in the mdadm source distro called "raid6check" which can possibly help you. NeilBrown [-- Attachment #2: signature.asc --] [-- Type: application/pgp-signature, Size: 828 bytes --] ^ permalink raw reply [flat|nested] 2+ messages in thread
end of thread, other threads:[~2012-06-04 4:17 UTC | newest] Thread overview: 2+ messages (download: mbox.gz follow: Atom feed -- links below jump to the message on this page -- 2012-05-09 19:16 tracking parity mismatches to a failing disk Michael Peterson 2012-06-04 4:17 ` NeilBrown
This is a public inbox, see mirroring instructions for how to clone and mirror all data and code used for this inbox