0

I am attempting to salvage as much data as possible off a failed 12 TB RAID-5 volume comprising 4 x Western Digital WD4000FYYZ 4TB hard drives in an OWC RAID enclosure that is based on the Oxford Semiconductor OXUFS936QSE "Universal Interface to Quad SATA RAID Controller" chip. The volume was formatted with a single MacOS hfs+ filesystem.

It's unclear how the array has gotten to a completely failed state (the enclosure beeps on power up, indicates a "hardware error" LED, and fails to mount the volume on the OS), but I have determined by experimentation the enclosure is OK (by populating it with a different set of 4 hard drives and successfully formatting them into a RAID-5 volume), and also, the 4 original hard drives are OK (by running the manufacturer's low-level diagnostic software on each one and getting a PASS). My guess is that somehow 2 drives got marked offline and so now the enclosure does not attempt a rebuild or even present a degraded volume to the OS.

I searched the internet looking for the RAID implementation details for the OXUFS936QSE controller but found nothing. Is it because such implementations are proprietary? Are there whitepapers available explaining how RAID 5 is implemented by hardware controller to a level of detail that would facilitate data recovery by a competent programmer?

Given I have all 4 original hard drives in a fully functional state, and believe no data has been (purposely) erased, I am hopeful some kind of approach to recovering the data is possible. Thank you for sharing any advice or experience.

Nico
  • 1
  • 1
    Why not just recreate the array and restore your data from backup? – joeqwerty Nov 21 '19 at 12:44
  • The array I'm trying to recover is in fact my Backups volume, which is used as a network backup store for about 6 machines and has a decent history. So while I can afford to lose it all (and recreate baseline backups), it would be a really "nice to have" to be able to recover and keep that backup history. – Nico Nov 22 '19 at 02:01

0 Answers0