Corrupted raid-z pool
(too old to reply)
Dan Davis
2013-01-21 05:56:06 UTC
Raw Message
I've been fighting with this raid-z on pool for almost a year now on FreeBSD.

Earlier this week I managed to mount it with one corrupted/offline
disk. I tried to wipe the disk, and ended up settling with using gpart
to make a new partition on it.
It resilvered partially, but when I attempted to clear errors (that
had been present since I last tried getting the pool to work) it
restarted the resilver. After it finished the resilver I tried
clearing the errors again, and it again resilvered the pool. A scrub
after that finished without any errors. I switched disks around to
attach another pool, hoping to migrate data and swap drives out to
alleviate this issue. This pool refused to mount after a reboot,
reporting I/O errors. I exported the pool, hoping that an import might
fix any issues with zpool.cache. It currently won't import.

***@snorri:/dev # zpool import
pool: njal
id: 7502878427997551002
state: ONLINE
status: One or more devices contains corrupted data.
action: The pool can be imported using its name or numeric identifier.
see: http://illumos.org/msg/ZFS-8000-4J

raidz1-0 ONLINE
replacing-1 ONLINE
3537756537319650203 UNAVAIL corrupted data
ada1p1 ONLINE
1543306220834098718 UNAVAIL corrupted data

***@snorri:/dev # zpool import njal
cannot import 'njal': I/O error
Destroy and re-create the pool from
a backup source.

3537756537319650203 and ada1p1 are the same disk - the former was
ada1p2, but no longer has a listing under /dev

Dan Davis