How to delete temp files reported as corrupt by zpool status

truenas

I have just upgraded from FreeNAS 8.0.2 to 8.3.0 Release-p1.

After the update reportedly succeeded I was going to upgrade the ZFS pool from ZFSv15 to ZFSv28. First, I checked the status of zpool. At that time I discovered that the pool has errors. Here is the output of the zpool status -v command:

[root@freenas] ~# zpool status -v
  pool: first
 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 the
    entire pool from backup.
   see: http://www.sun.com/msg/ZFS-8000-8A
  scan: scrub in progress since Sat Feb  2 17:15:20 2013
        496G scanned out of 886G at 74.9M/s, 1h28m to go
        0 repaired, 55.95% done
config:

    NAME                                            STATE     READ WRITE CKSUM
    first                                           ONLINE       0     0     2
      mirror-0                                      ONLINE       0     0     4
        gptid/5f279dd0-0ce9-11e1-919a-001871691677  ONLINE       0     0     4
        gptid/5f76d1d1-0ce9-11e1-919a-001871691677  ONLINE       0     0     4

errors: Permanent errors have been detected in the following files:

        first/tmp@auto-20130130.1100-1m:<0x198b>
        first/tmp@auto-20130130.1100-1m:<0x4b3>
        first/tmp@auto-20130130.1100-1m:<0x18ec>
        first/tmp@auto-20130130.1100-1m:<0x9fd>
        <0x1eed>:<0x4b3>
        <0x1efd>:<0x9fd>

As you can see from the output I have already started the scrub. So far it did not fix any errors but only found 2 extra checksum errors.

I read the recommendation in other forums to first delete the files with errors and then do the scrub. Question is: where to find and how to delete the files like first/tmp@auto-20130130.1100-1m:<0x9fd>.

Best Answer

I did not find where these files were located.

What I end up doing was creating a new volume, copying all the files to it, detaching the old volume which had errors, and then attaching its hard drives to the newly created volume (aka zpool).

I was going to extend my storage anyway, but initially I thought to connect new hard drives to the existing volume. Instead, I created a completely new volume, copied all the files from the existing volume to the new volume. It took me about 2 hours for 1TB of data but now I do not have annoying blinking yellow traffic light on Web GUI of the FreeNAS server. Also I needed to recreate CIFS shares so that they point to the new volume instead of the old.

Related Topic