Trying To Upgrade ZFS RAID-1 Now Have Issues

[karlywarly]

Hi Guys,

Tried to replace my 2 hard drives in my RAID 1 mirror to larger capacity ones, followed the manual on replacing a failed drive or zil device and the first hard drive resilvered OK.

I have just replaced the second hard drive to do the same and now i'm not sure what has gone wrong.

I have a yellow alert icon saying "WARNING: The volume volume1 (ZFS) status is UNKNOWN: One or more devices has experienced an error resulting in data corruption. Applications may be affected.Restore the file in question if possible. Otherwise restore the entire pool from backup.".

All my data is still as it should be from the first hard drive i added and resilvered but now i have the following issue in the volume status from replacing the second hard drive and isn't resilvering:



Can someone please help me fix this issue as i'm not sure what has gone wrong.

Thanks :)

 

[ben]

Could you post the output of 'zpool status' in

Code:

 tags here?

Also, how long had you been running the mirror for storage, and did you scrub the volume at all during that time?

 

[fracai]

Your image shows three drives; one functional, two unavailable. Did you detach the replaced drive after the resilver completed?

The process to replace a drive should proceed: offline the old device, shutdown, remove the old drive, insert the new drive, boot, replace the old device with the new device, wait for the resilver to complete, detach the old device.

 

[karlywarly]

I had a mirror with 2x 1.5TB hard drives and i simply followed the replace a failed drive section in the manual by clicking the disks offline button (ada0) and then pulled the disk with the NAS still turned on as my hardware supports AHCI. I then put the new one in (2.0TB WD RED) and clicked on the replace button and selected the hard drive from the list and then it resilvered fine and pressed the detach button on the old drive which was yesterday.

Today I wanted to swap the remaining 1.5TGB with another 2.0TB WD RED the same way by pressing the offline button, pulling that hard disk and then clicking on the replace button and selecting from the drop down box but once selecting the correct hard drive, the nas seemed to hang for quite a while and possibly didn't like the hard drive so it got rebooted as i couldn't do anything with the GUI.

I now have 2x 1.5TB drives that were in the RAID before that have not been formatted. If there is a chance of loosing my data, is it possible to inport these 2 hard drives back into the system to start again?

I have not done any scrubs either.

Hope this helps.



Zpool status:

Code:

 pool: volume1
 state: DEGRADED
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 the
        entire pool from backup.
   see: http://www.sun.com/msg/ZFS-8000-8A
  scan: resilvered 1.18T in 8h24m with 0 errors on Tue Feb 26 03:01:00 2013
config:

        NAME                                            STATE     READ WRITE CKSUM
        volume1                                         DEGRADED     0     0     0
          mirror-0                                      DEGRADED     0     0     0
            gptid/456015bc-7f7a-11e2-bbc8-009c02975522  ONLINE       0     0     0
            replacing-1                                 UNAVAIL      0     0     0
              5876343771440100576                       OFFLINE      0     0     0  was /dev/gptid/19c24041-18a9-11e2-8a79-009c02975522
              12597927988271675085                      UNAVAIL      0     0     0  was /dev/gptid/5a93c9ee-8006-11e2-bbc8-009c02975522

errors: 1 data errors, use '-v' for a list

 

[ben]

Well, 'zpool status -v' will tell you what file got corrupted. That's worth knowing at least. Dare I ask whether you have backups?

 

[karlywarly]

Just had a look now with zpool status -v which is pointing to my torrent folder for transmission which isn't such a big deal compared to other stuff on there, hopefully there isn't anything else corrupted.

Do you know how I would be able to rebuild my mirror back to how it was?

 

[paleoN]

Do you know how I would be able to rebuild my mirror back to how it was?

Depends. If you have the 2[sup]nd[/sup] original drive, you can stick it back in and try a scrub. Assuming it wasn't the source of the errors this will repair the pool.

The process to replace a drive should proceed: offline the old device, shutdown, remove the old drive, insert the new drive, boot, replace the old device with the new device, wait for the resilver to complete, detach the old device.

Actually, for this case the following is superior:

1. Shutdown
2. • If enough available bays add all drives to the case.
   • If not then swap 1[sup]st[/sup] drive with replacement.
3. Attach 1[sup]st[/sup] new drive creating a 3-way mirror. (With one device unavailable if b.)
4. Repeat for 2[sup]nd[/sup] drive creating a 4-way mirror. (Shutdown & replace if b.)
5. Scrub pool.
6. Detach old drives.

If something goes wrong simply swap the original drives back in, assuming they aren't the source of the problem, to restore the pool.

Furthermore, for any shutdown/replace I actually wouldn't offline the old device first. Removing it and having it be UNAVAIL simplifies things if you need to put the original back in for any reason. Now if I was hot-swapping, I would always offline the drive first.

 

[fracai]

Furthermore, for any shutdown/replace I actually wouldn't offline the old device first. Removing it and having it be UNAVAIL simplifies things if you need to put the original back in for any reason.

Interesting. Thanks. Wish I'd known that for my recent capacity upgrade; it would have simplified things slightly.

 

[karlywarly]

Depends. If you have the 2[sup]nd[/sup] original drive, you can stick it back in and try a scrub. Assuming it wasn't the source of the errors this will repair the pool.

Actually, for this case the following is superior:

1. Shutdown
2. • If enough available bays add all drives to the case.
   • If not then swap 1[sup]st[/sup] drive with replacement.
3. Attach 1[sup]st[/sup] new drive creating a 3-way mirror. (With one device unavailable if b.)
4. Repeat for 2[sup]nd[/sup] drive creating a 4-way mirror. (Shutdown & replace if b.)
5. Scrub pool.
6. Detach old drives.

If something goes wrong simply swap the original drives back in, assuming they aren't the source of the problem, to restore the pool.

Furthermore, for any shutdown/replace I actually wouldn't offline the old device first. Removing it and having it be UNAVAIL simplifies things if you need to put the original back in for any reason. Now if I was hot-swapping, I would always offline the drive first.

Managed to get my nas back to how it was.

Thanks for your help :)