RAID1 disappears after installing updates(?)

TadeuszD · 26. März 2015

Hello!
After long year of using my simple NAS server I encountered big problem. I'm not sure what happens. Yesterday I installed many updates available form 'Update Manager' menu in OMV admin panel.
My original configuration contains two 1TB HDs (sdb, sdc) organized in one RAID1 matrix. Only one partition EXT4 labeled 'data1t' was created on this matrix.
But today something breaks. I loose my data. In 'RAID Management' admin panel I see two data matrix:
/dev/md126

Code

Version : 1.2
  Creation Time : Wed Feb 26 22:03:04 2014
     Raid Level : raid1
     Array Size : 976761424 (931.51 GiB 1000.20 GB)
  Used Dev Size : 976761424 (931.51 GiB 1000.20 GB)
   Raid Devices : 2
  Total Devices : 1
    Persistence : Superblock is persistent


    Update Time : Thu Mar 26 00:03:05 2015
          State : clean, degraded
 Active Devices : 1
Working Devices : 1
 Failed Devices : 0
  Spare Devices : 0


           Name : blackbox:mirror2x1t  (local to host blackbox)
           UUID : fa291e4f:31cbfbd3:83a29e31:67e65509
         Events : 3426


    Number   Major   Minor   RaidDevice State
       0       8       16        0      active sync   /dev/sdb
       1       0        0        1      removed

Alles anzeigen

...and /dev/md127

Code

Version : 1.2
  Creation Time : Wed Feb 26 22:03:04 2014
     Raid Level : raid1
     Array Size : 976761424 (931.51 GiB 1000.20 GB)
  Used Dev Size : 976761424 (931.51 GiB 1000.20 GB)
   Raid Devices : 2
  Total Devices : 1
    Persistence : Superblock is persistent


    Update Time : Wed Mar 25 00:22:40 2015
          State : clean, degraded
 Active Devices : 1
Working Devices : 1
 Failed Devices : 0
  Spare Devices : 0


           Name : blackbox:mirror2x1t  (local to host blackbox)
           UUID : fa291e4f:31cbfbd3:83a29e31:67e65509
         Events : 8752


    Number   Major   Minor   RaidDevice State
       0       0        0        0      removed
       1       8       32        1      active sync   /dev/sdc

Alles anzeigen

In 'File systems' panel I see two partitions, both labeled 'data1t', but only first partition reports real capacity and mounted='Yes', the second reports capacity='n/a' and mounted='No'.

What happens and how can I fix it? Sorry, if it is lame question, but I'm not Linux expert and I don't want destroy everything...

TadeuszD · 27. März 2015

I found similar topics in OMV forum, for example: Single raid1 split into two degraded ones
Those topics contains usually simple solution:

Code

umount /dev/md126
umount /dev/md127
mdadm –-stop /dev/md126
mdadm --stop /dev/md127
mdadm -A /dev/md0 -f –-update=summaries /dev/sdb /dev/sdc
mdadm /dev/md0 -a /dev/sdc

Now I have additional questions:
- should I use /dev/md0 in fifth and sixth line? Or should I use location of my original array, which is located in /dev/md/mirror2x1t ?
- why should I re-add device /dev/sdc in last line? And why sdc not sdb? How can I identify it?

ryecoaaron · 27. März 2015

I would use your first four lines (they probably aren't mounted) and then this:

mdadm --assemble /dev/md127 /dev/sd[bc] --verbose --force

Whether you use md0 or md127 really doesn't matter.

TadeuszD · 28. März 2015

Thanks ryecoaaron for your suggestions!
After reboot I unmounted and stopped arrays (/dev/md127 was still mounted). Next I ran:

Code

mdadm -A /dev/md/mirror2x1t -f –-update=summaries /dev/sdb /dev/sdc

After this new array was established, but only sdc was active. So, I re-added sdb. This triggered rebuild process.
After few hours my array looks like this:

Code

Personalities : [raid1]
md127 : active raid1 sdb[0] sdc[1]
      976761424 blocks super 1.2 [2/2] [UU]


unused devices: <none>

My data are back.

Current question is: how to prevent this situation in the future?

Jetzt mitmachen!