Data recovery from a single RAID1 disk

I extracted the disk from a Mybook World 2 (White Led). The device has only this 1TB SATA disk, and it is using RAID1. Yes, it is bizzarre, but WD enabled RAID even for a 1-disk configurations.

This is a report from disktype:

francesco@blackhole:~$ sudo disktype /dev/sdb
--- /dev/sdb
Block device, size 931.5 GiB (1000204886016 bytes)
DOS/MBR partition map
Partition 1: 1.869 GiB (2006974464 bytes, 3919872 sectors from 64320)
  Type 0xFD (Linux raid autodetect)
  Ext3 file system
    UUID 78C28AA5-BDFD-4BE6-9F58-90158E3B0E34 (DCE, v4)
    Volume size 1.869 GiB (2006908928 bytes, 489968 blocks of 4 KiB)
  Linux RAID disk, version 0.90.0
    RAID1 set using 2 regular -1 spare disks
    RAID set UUID 03D291C1-5D91-F02A-42F0-E4F5C3F82B1D (NCS)
Partition 2: 251.0 MiB (263159808 bytes, 513984 sectors from 3984192)
  Type 0xFD (Linux raid autodetect)
  Linux RAID disk, version 0.90.0
    RAID1 set using 2 regular -1 spare disks
    RAID set UUID A1A39027-5114-9598-6F0E-78C3AC55D2B2 (NCS)
  Linux swap, version 2, subversion 1, 4 KiB pages, little-endian
    Swap size 250.9 MiB (263053312 bytes, 64222 pages of 4 KiB)
Partition 3: 964.8 MiB (1011712000 bytes, 1976000 sectors from 4498176)
  Type 0xFD (Linux raid autodetect)
  Ext3 file system
    UUID 0AEEE999-BE23-4CC1-A5B8-71EBE9DF871E (DCE, v4)
    Volume size 964.8 MiB (1011613696 bytes, 246976 blocks of 4 KiB)
  Linux RAID disk, version 0.90.0
    RAID1 set using 2 regular -1 spare disks
    RAID set UUID 71CAFCD8-5E07-B57B-1C59-C590CF05141F (NCS)
Partition 4: 928.4 GiB (996887495168 bytes, 1947045889 sectors from 6474176)
  Type 0xFD (Linux raid autodetect)
  Linux RAID disk, version 0.90.0
    RAID1 set using 2 regular -1 spare disks
    RAID set UUID 8B3DEF8B-534C-90D1-D7BA-B2FE15673C8F (MS GUID)
  XFS file system, version 4
    Volume name ""
    UUID 100D6A32-784A-4250-9B41-8E193B998415 (DCE, v4)
    Volume size 928.4 GiB (996887363584 bytes, 243380704 blocks of 4 KiB)

The important files are in sdb4.

I tried to mount it:

sudo mount -t xfs -o ro /dev/sdb4 /media/recovery/

But it fails saying that the filesystem is wrong, or the superblock damaged.

I also tried mdadm:

sudo mdadm --assemble /dev/md4 /dev/sdb4
mdadm: no recogniseable superblock on /dev/sdb4
mdadm: /dev/sdb4 has no superblock - assembly aborted

Any idea to access the XFS filesystem?

Francesco

Posted 2012-11-01T16:07:18.277

Reputation: 111

WD sure has a strange idea of what raid 1 is, gawd. I build my own usb backup drives, pick my enclosure, pick my drives, assemble, I know what I have now, I hope you solve it. – Moab – 2012-11-01T16:10:51.140

Surely they have a strange idea of RAID. I hope someone can help me to mount this partition. – Francesco – 2012-11-01T16:13:14.270

Image the drive to a file. 2) Search the file to see how far into it is before the partition data starts. 3) Copy the file, skipping that many bytes. 4) Loopback mount the copied file.

< – David Schwartz – 2012-11-01T16:16:37.703

What is the output of mdadm --examine /dev/sdb4? – Zoredache – 2012-11-01T18:42:27.340

Uhm, I have to check but I tried a lot of switches in mdadm and I always got errors about the missing superblock. – Francesco – 2012-11-04T00:20:05.297

@DavidSchwartz: I can image the drive to a file, but how can I detect how much data to skip? – Francesco – 2012-11-04T00:21:12.567

You literally look at the bytes of data in the file to see where the partition table is or where the partition data starts. – David Schwartz – 2012-11-04T19:11:51.310

Data recovery from a single RAID1 disk

No answers