mdadm / RAID issues

    This isn't directly related to OMV, but I was really hoping someone here could help me out. I originally created this array with OMV 0.5, however I needed to upgrade to the latest Debian as I was having issues with my network adapter (which had known issues with Debian 6). After installing Debian 8 and reinitialising the array everything was going smoothly until I tried to grow the array. I'd be very grateful if anyone had any ideas.


    My RAID 6 array consisted of 5x 3TB drives, and I tried to grow it with another 3TB drive, however after 12 hours the reshape didn't appear to be making any progress; reshape = 0.0% (0/2930265088). Unfortunately at some point I suffered power loss and upon powering on the array wouldn't start. After googling I've read that the first moments of a reshape are critical, however as the reshape seemed to be stuck I'm hoping there's a way to recover from this? I also read that you can define a back up file, unfortunately the guide I was following for the reshape didn't mention this, so I don't have one. All 5 are reporting clean and correct checksums.


    At the moment I don't care about trying to grow the array, I'd just like to try and recover data if possible.


    Here's the output of mdadm --examine for each of the old drives: http://pastebin.com/raw/DRQdjWef (had to pastebin because post was too long)


    Output of mdadm -D /dev/md127


    When trying to force assemble:

    Code
    root@nas:~# mdadm --assemble --force /dev/md127 /dev/sda /dev/sdb /dev/sdc /dev/sdd /dev/sde
mdadm: Failed to restore critical section for reshape, sorry.
Possibly you needed to specify the --backup-file


    And with "--invalid-backup"


    And finally /proc/mdstat & blkid & fdisk -l


    Thanks very much!

    mdadm --assemble --verbose --invalid-backup --force /dev/md127 /dev/sd[abcde]


    Could be that it mounts read-only. Check with cat /proc/mdstat  If it is read-only, make it read-write with mdadm --readwrite and check with cat /proc/mdstat  again.


    Please post the outputs of every step...

    Hi WastlJ, thanks for taking a look.


    I had forgotten while trying to debug last night I had removed the new hard drive I was attempting to grow with, here's the above commands ran with both the new drive connected and disconnected.


    Disconnected:


    Connected (with the new/grow drive being /dev/sdc):

    Code
    root@nas:~# mdadm -A /dev/md127 -f --update=summaries /dev/sd[abcdef]
mdadm: --update=summaries not understood for 1.x metadata
root@nas:~# cat /etc/mdadm.conf
DEVICE /dev/sda
DEVICE /dev/sdb
DEVICE /dev/sdc
DEVICE /dev/sdd
DEVICE /dev/sde
ARRAY /dev/md/Storage  metadata=1.2 UUID=b4812ae8:804e182d:f813a5e1:e9ee6da4 name=nas:Storage


    Thanks again

    Stephen from freedatarecovery.us replied to one of my threads on another forum, and was able to restore the array with no data loss. Definitely recommend his service if anyone is ever in a similar situation and everything else had failed.

Jetzt mitmachen!

Sie haben noch kein Benutzerkonto auf unserer Seite? Registrieren Sie sich kostenlos und nehmen Sie an unserer Community teil!

Benutzerkonto erstellen Anmelden