Help Recovering FBSD 8 ZFS System
Drew Tomlinson
drew at mykitchentable.net
Mon Jan 2 22:15:31 UTC 2012
On 1/2/2012 12:36 PM, Daniel Staal wrote:
> --As of December 31, 2011 1:40:59 PM -0800, Drew Tomlinson is alleged
> to have said:
>
>> Thus it appears I am missing ad16 that I used to have. My data zpool
>> was
>> the bulk of my system with over 600 gig of files and things I'd like to
>> have back. I thought that by creating a raidz1 I could avoid having to
>> back up the huge drive and avoid this grief. However it appears I have
>> lost 2 disks at the same time. :(
>>
>> Any thoughts before I just give up on recovering my data pool?
>
> Ouch. All I can really say is 'Redundancy is not backup', but that's
> a bit trite...
Yes, I know redundancy doesn't protect against operator error and thus
isn't a true backup. However this is a personal system whose main
function was to store DVDs, MP3s, photos, and the like. I can recreate
most of the content and have backups of the photos up until about a year
ago (bad me).
> The one thing you haven't mentioned trying that might be worth the
> attempt is trying the recovery from a 9.0 disk. There has been work
> done on the ZFS system, and it's possible that something might work.
> But that's mostly just to be thorough...
I may try this. However I suspect before anything can work, I have to
get the missing disk(s) detected by the OS. One (ad6) is detected but
full of errors. There is another that's not even seen.
> As for what it was telling you: It was just saying it couldn't open
> the drives. ;) Which does bring up one other option: If you've got a
> different drive controller, you might try plugging the drives into
> it. (In the hopes that it's the *controller* and not the drive that's
> gone bad. Unlikely, bit it *does* happen.)
Thanks. I'll keep that in mind. However in this case, the controller
is a SATA that's integrated into the motherboard. Since two of 4 are
working, that would mean the controller is OK, right? I guess I could
swap SATA cables for a test.
> (Depending on the value of the data pool, a good data recovery service
> might be able to do something as well. But they'd have to be a very
> good service, and know what they were working with.)
>
>> And regarding my root pool, my system can't mount root and start. What
>> do I need to do to boot from my degraded root pool. Here's the current
>> status:
>>
>> # zpool status
>> pool: root
>> state: DEGRADED
>> status: One or more devices could not be opened. Sufficient replicas
>> exist for
>> the pool to continue functioning in a degraded state.
>> action: Attach the missing device and online it using 'zpool online'.
>> see: http://www.sun.com/msg/ZFS-8000-2Q
>> scrub: none requested
>> config:
>>
>> NAME STATE READ
>> WRITE CKSUM
>> root DEGRADED 0
>> 0 0
>> mirror DEGRADED 0
>> 0 0
>> gptid/5b623854-6c46-11de-ae82-001b21361de7 ONLINE 0
>> 0 0
>> 12032653780322685599 UNAVAIL 0
>> 0 0 was /dev/ad6p3
>>
>> Do I just need to do a 'zpool detach root /dev/ad6p3' to remove it from
>> the pool and get it to boot? And then once I replace the disk a 'zpool
>> attach root <new partition>' to fix?
>>
>> Thanks for your time.
>
> Personally, I'd do a 'zpool replace /dev/ad6p3 /dev/$NEWDRIVE', but
> the above should work as well. What's odd though is that you can't
> boot from it as is: Degraded should be considered functional, and it
> should let you boot. You mentioned updating the zpool to v15. Did
> you update the boot block at the same time? (Just checking the
> basics.) It'd need to be able to read the updated zpool.
I assume I upgraded the boot block since I've had no trouble booting
before the drive failures and the upgrade was a long time ago.
Thanks for your help.
Drew
--
Like card tricks?
Visit The Alchemist's Warehouse to
learn card magic secrets for free!
http://alchemistswarehouse.com
More information about the freebsd-questions
mailing list