Raid Gurus,
I have a need to determine whether a linux partion has been corrupted (say
by s/w running amok on a window partition). I think that it should be
possible to detect this by having two raid5 linux partitions attempt to
sync and then check whether any error was reported via a mismatching
parity. I'm using mdadm 1.6 (CentOS 4.4). Can someone tell me what
commands could be used for this?
If my linux raid5 drives are /dev/sda1 and /dev/sdb1
Would something like the following work?
#force resync
mdadm --fail /dev/md0 /dev/sdb1
mdadm --remove /dev/md0 /dev/sdb1
mdadm --add /dev/md0 /dev/sdb1
#then check for an error?
/sys/block/md0/stat
Since the raid drive is configured for auto checking during boot-up would
that set any flag in the /sys/block/md0/stat file that I could use?
I'm just trying to avoid creating a bunch of checksums... (takes about 10
minutes to create/check checksums on my 20GB drive).
Regards
-
To unsubscribe from this list: send the line "unsubscribe linux-raid" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
