problem Attempting to rebuild Snap Server Raid

[Noan21]

My Snap Server 410 Crashed, and while I have good backups it's probably going to be 48hrs+ to get the data off of the tapes.

I've got the arrays built, but I think I'm missing something.

cat /proc/mdstat


Personalities : [linear] [multipath] [raid0] [raid1] [raid6] [raid5] [raid4] [raid10] 
md0 : inactive sde4[2] sdd4[1] sdc4[0] sdb4[3]
      2925526528 blocks super 0.91
       
md100 : inactive sdd2[1] sde2[3] sdc2[0] sdb2[2]
      2184448 blocks super 0.91
       
md101 : inactive sdc5[0] sdd5[3] sde5[2] sdb5[1]
      1092096 blocks super 0.91
       
unused devices: <none>


sudo mdadm -R /dev/md0
mdadm: failed to run array /dev/md0: Invalid argument

Thoughts? Suggestions, more / different info required? I've never had to do anything like this before.

 

[Noan21]

This might be useful as well...

mdadm --misc --examine /dev/sde4
/dev/sde4:
          Magic : a92b4efc
        Version : 00.91.00
           UUID : 53e98dd8:b661558a:1148a17c:42edb0a8
  Creation Time : Sun Mar  9 12:39:13 2008
     Raid Level : raid5
  Used Dev Size : 731381632 (697.50 GiB 748.93 GB)
     Array Size : 2194144896 (2092.50 GiB 2246.80 GB)
   Raid Devices : 4
  Total Devices : 4
Preferred Minor : 0

  Reshape pos'n : 0
      New Level : raid0
     New Layout : left-asymmetric
  New Chunksize : 0

    Update Time : Fri Jan 21 16:34:14 2011
          State : active
 Active Devices : 4
Working Devices : 4
 Failed Devices : 0
  Spare Devices : 0
       Checksum : c853dacb - correct
         Events : 56

         Layout : left-symmetric
     Chunk Size : 64K

      Number   Major   Minor   RaidDevice State
this     2       8       36        2      active sync   /dev/sdc4

   0     0       8        4        0      active sync   /dev/sda4
   1     1       8       20        1      active sync   /dev/sdb4
   2     2       8       36        2      active sync   /dev/sdc4
   3     3       8       52        3      active sync   /dev/sdd4

 

[Noan21]

My dmesg output.

[    2.034721] md: linear personality registered for level -1
[    2.036530] md: multipath personality registered for level -4
[    2.040245] md: raid0 personality registered for level 0
[    2.042806] md: raid1 personality registered for level 1
[    2.748032] md: raid6 personality registered for level 6
[    2.748033] md: raid5 personality registered for level 5
[    2.748035] md: raid4 personality registered for level 4
[    2.757498] md: raid10 personality registered for level 10
[    3.166920] md: bind<sdb2>
[    3.177209] md: bind<sdb5>
[    3.179830] md: bind<sdb4>
[    3.195434] md: bind<sdc2>
[    3.254732] md: bind<sdc5>
[    3.264345] md: bind<sdd2>
[    3.274532] md: bind<sdc4>
[    3.283590] md: bind<sdd5>
[    3.309252] md: bind<sde5>
[    3.312859] md: bind<sdd4>
[    3.314273] md: md101: raid array is not clean -- starting background reconstruction
[    3.452775] md: bind<sde2>
[    3.453448] md: md100: raid array is not clean -- starting background reconstruction
[    3.511265] md: bind<sde4>
[    3.511962] md: md0: raid array is not clean -- starting background reconstruction
[    3.513286] md: pers->run() failed ...
[  281.918254] md: pers->run() failed ...

 

[Mercutio]

Did you have a drive failure or just some random crap where one or more of the drives disconnected?

 

[Noan21]

These drives were in an Overland Storage Snap Server 410. That device is dead but the drives still work.

 

[Howell]

It is unclear what your goal is so we need to establish that first.

What we do know is:
You had an Overland Storage Snap Server 410.
The server itself died but you think the drives are good.
You will have access to the data from the tapes in 48rs.

Where do you have the drives connected now?

 

[Noan21]

Yes, I suppose that would be good information.

I have the drives connected into an ubuntu box as jbod on a promis sata card. I want to get the raid arrays rebuilt if I can and pull the data directly from them. The above is output from different mdadm commands on that system.

Here is my dmesg output as well...

[    0.625042] md: linear personality registered for level -1
[    0.690429] md: multipath personality registered for level -4
[    0.730499] md: raid0 personality registered for level 0
[    1.094841] md: raid1 personality registered for level 1
[    1.810383] md: raid6 personality registered for level 6
[    1.810387] md: raid5 personality registered for level 5
[    1.810389] md: raid4 personality registered for level 4
[    1.815656] md: bind<sda2>
[    1.819601] md: raid10 personality registered for level 10
[    1.820725] md: bind<sda5>
[    1.823752] md: bind<sda4>
[    1.830480] md: bind<sdb4>
[    1.834598] md: bind<sdd5>
[    1.840389] md: bind<sdb2>
[    1.873101] md: bind<sdd4>
[    1.878022] md: bind<sdb5>
[    1.885970] md: bind<sdc2>
[    1.929937] md: bind<sdc4>
[    1.932835] md: pers->run() failed ...
[    9.248097] md: bind<sdc5>
[    9.262597] md: bind<sdd2>

 

[Noan21]

The array's themselves are built, and as far as I can tell everything looks fine when you examine them, but they will not start.

mdadm --detail /dev/md0
/dev/md0:
        Version : 00.91
  Creation Time : Sun Mar  9 12:39:13 2008
     Raid Level : raid5
  Used Dev Size : 731381632 (697.50 GiB 748.93 GB)
   Raid Devices : 4
  Total Devices : 4
Preferred Minor : 0
    Persistence : Superblock is persistent

    Update Time : Fri Jan 21 16:34:14 2011
          State : active, Not Started
 Active Devices : 4
Working Devices : 4
 Failed Devices : 0
  Spare Devices : 0

         Layout : left-symmetric
     Chunk Size : 64K

      New Level : raid0
  New Chunksize : 0K

           UUID : 53e98dd8:b661558a:1148a17c:42edb0a8
         Events : 0.56

    Number   Major   Minor   RaidDevice State
       0       8       52        0      active sync   /dev/sdd4
       1       8       36        1      active sync   /dev/sdc4
       2       8       20        2      active sync   /dev/sdb4
       3       8        4        3      active sync   /dev/sda4

 

[paulo79]

Have you had any success in recovering your RAID array? I'm trying to recover a 4-drive RAID5 array from a Snap Server 4500. The GuardianOS hangs at start-up, so I needed to boot from an external CD drive with Ubuntu live.

I tried using mdadm to manually assemble the RAID array but got an error message of a "pending reshape" of the array. It also says it couldn't restore the "critical section" and that I should probably specify a "--backup-file". I have no such backup file.

When I run "mdadm --examine /dev/sd{a,b,c,d}4", I can see an output similar to yours. One thing that attracted my attention, in particular, is this part:

Reshape pos'n : 0
New Level : raid0
New Layout : left-asymmetric
New Chunksize : 0

What does it mean? Why have a new chunksize = 0 ? Pretty weird. Please let me know if you make any progress here. I will keep you posted as well.

Best,

~Paulo