mjhjmtu

I'm experiencing a terrible fault of my raid 10 made of 4 disks of 4TB connected to a serveraid M5110 card. The VD got degraded and no way to restore it with MegaRaid SW, so I disconnected disks from the serveraid card and got on physical sata ports on mainboard to run mdadm.

The results:

# mdadm --assemble --verbose /dev/md0 /dev/sdb /dev/sdc /dev/sdd /dev/sde
mdadm: looking for devices for /dev/md0
mdadm: /dev/sdb is identified as a member of /dev/md0, slot 1.
mdadm: /dev/sdc is identified as a member of /dev/md0, slot 2.
mdadm: /dev/sdd is identified as a member of /dev/md0, slot 0.
mdadm: /dev/sde is identified as a member of /dev/md0, slot 0.
mdadm: ignoring /dev/sde as it reports /dev/sdb as failed
mdadm: no uptodate device for slot 0 of /dev/md0
mdadm: added /dev/sdc to /dev/md0 as 2 (possibly out of date)
mdadm: added /dev/sdb to /dev/md0 as 1
mdadm: Container /dev/md0 has been assembled with 1 drive (out of 2)

# mdadm --examine /dev/md0
/dev/md0:
 Magic : de11de11
 Version : 01.00.00
Controller GUID : 49424D20:20202020:53563330:32313930:38300000:3A62653A
 (IBM SV30219080)
 Container GUID : 49424D20:20202020:1000005B:1014040B:46AF0993:B423E95D
 (IBM 07/30/17 12:06:11)
 Seq : 000000b1
 Redundant hdr : yes
 Virtual Disks : 1

 VD GUID[0] : 49424D20:20202020:1000005B:1014040B:46AF0994:16E1A7A7
 (IBM 07/30/17 12:06:12)
 unit[0] : 0
 state[0] : Failed, Not Consistent
 init state[0] : Not Initialised
 access[0] : Read/Write
 Name[0] : VD_RAID10
 Raid Devices[0] : 2 (0@0K 1@0K)
 Chunk Size[0] : 128 sectors
 Raid Level[0] : RAID1
 Secondary Position[0] : 1 of 2
 Secondary Level[0] : Striped
 Device Size[0] : 3906469888
 Array Size[0] : 7812939776

 Physical Disks : 63
 Number RefNo Size Device Type/State
 0 a899eb71 3906469888K /dev/sdc active/Offline, Failed
 1 28548427 3906469888K /dev/sdb active/Online
 2 28548427 3906469888K /dev/sdb active/Online
 3 28548427 3906469888K /dev/sdb active/Online
 Physical disks not in metadata!:
 41d13ecd /dev/sdd

# mount /dev/md0 /mnt/usb/
mount: /mnt/usb: can't read superblock on /dev/md0. 

No way to mount md0?

This error message seems to belong to RAID-5; in any case it does not make sense for RAID-10.

It may be a good idea to create a small copy-on-write device for each of the disks and use dmsetup to set up a snapshot for each. That way you can make tests with the snapshot devices, write to them and easily throw away the changes.

I suggest that you recreate the MD metadata on the disks because the current metadata is obviously misunderstood by mdadm.

Copy (e.g. with dd) a sector from the three working disks in order to determine which are the RAID-1 pair. Then set up a new RAID-1 with these disks (i.e. their snapshot devices). Mind using the correct offset and metadata version. Do the same with the third drive i.e. set up a degraded RAID-1 with it.

Use mdadm --build ... --readonly to create a non-persistent RAID-0 (Mind the correct chunk size!) from the two RAID-1 arrays. Check whether you see a filesystem on the resulting device. If not then delete the new array and change the order of the lower arrays.