Your RAID is failing you (Re: [zfs-discuss] Re: Question regarding resizeing/expanding a zpool)
gregor at kopka.net
Sun Apr 15 05:53:43 EDT 2012
Am 15.04.2012 01:26, schrieb peram:
> The status of the spool is this :
> peram at ubuntu:/$ sudo zpool status NAS
> pool: NAS
> state: ONLINE
> status: One or more devices has experienced an error resulting in data
> corruption. Applications may be affected.
> action: Restore the file in question if possible. Otherwise restore
> entire pool from backup.
> see: http://zfsonlinux.org/msg/ZFS-8000-8A
> scan: scrub repaired 5.50K in 9h58m with 1 errors on Sat Apr 14
> 04:54:42 2012
> NAME STATE READ WRITE CKSUM
> NAS ONLINE 0 0 1
> sdc1 ONLINE 0 0 3
> errors: 1 data errors, use '-v' for a list
As you can see from this zpool status: *your RAID has failed, some of
your data is lost.
Caused by your setup ZFS sees only one device, hence it's self-healing
capacities are severely crippled.
And you already have been bitten by it:
The RAID happily returned 3 blocks of garbage, for two of them ZFS was
able to correct the damage delivered by the underlying storage system
(must have been metadata which is always redundant in ZFS). But the
contents of one data block hadn't had redundancy in ZFS, thus couldn't
be repaired and _is lost_ (use 'zpool status NAS -v' to see what
file/zvol was damaged).
Do yourself a favour and let ZFS handle redundancy, your RAID clearly
isn't able to do it.
* backup the complete pool (zfs send/recv can do the trick nicely
* break up the RAID into JBOD
* expose the individual disks (using /dev/disk/by-id paths)
* build a new pool on them with mirrors or raidz for redundancy
> I've tried to export and then import the pool again, but the allocated
> size remains the same.
> So to sum it up, my disk device is bigger than what I have available
> in my zpool. What have I forgotten to do ?
zpool offline -t /dev/sdc1 # take disk offline
zpool online -e /dev/sdc1 # online it and expand
Better plan still is to do youself the favour as above.
-------------- next part --------------
An HTML attachment was scrubbed...
More information about the zfs-discuss