Hi all,

This is an embarrassing post but here we are!

I had one drive that was failing 3 weeks ago, but I didn't get a chance to get to it and made the mistake of just leaving things as is. While on vacation, a second drive (the one I bought at the same time...) seems to have failed.

One is marked as Disabled, but has passed its most recent SMART check. The other is seemingly unreadable, as it's listed as an unassigned device and can't be read by the linux filesystem.

Here's my current setup:

Array configuration:

- Parity: 10TB drive (working fine)

- Disk 1: 5TB drive - showing as disabled with red X but a scan shows its 99.99% fine - just a very small sector that failed on write recently

- Disk 2: 5TB drive - showing as missing

- Disk 3: 10TB drive (working fine)

- Running Unraid 6.12.8 with single parity

What I've discovered:

- Disk 1 (disabled) is physically connected and detected as /dev/sdf

- SMART shows "PASSED" but has 488 pending/uncorrectable sectors

- Disk 2 (missing) is physically connected as /dev/sdb but shows as "Unassigned Device"

- Disk 2 has severe I/O errors - can't even read sector 0

- Array won't start due to "ERROR:TOO_MANY_MISSING_DISKS"

My understanding:

Since I have 2 failed drives with only single parity, normal rebuild isn't possible. Disk 2 appears to be completely

dead (hardware failure), while Disk 1 seems readable despite the bad sectors.

Is there a way to force Unraid to use Disk 1 as valid (despite being disabled) so parity can emulate the dead Disk 2?

This seems like it would give me the best chance at data recovery since Disk 1 is at least partially readable.

I'm hesitant to use New Config without expert guidance. Any help would be greatly appreciated! Regardless - new drives will be on order and I will remember this in the future so I don't run into this situation again!

Attach Diagnostics to your NEXT post in this thread.

Apologies - i pulled them but forgot to attach.OwlServer Diagnostics May 22 2025.zip

Replace/swap both cables from this disk:

Device Model: ST5000LM000-2AN170

Serial Number: WCJ28CPN

And post new diags, but looks like it may have failed.

No dice. :(

I'm going to pick up new drives tomorrow - hopefully I can save Disk1 so i can recreate 2.OwlServer Diagnostics May 24 2025.zip

The other disk appears to have failed, you can try using ddrescue on it, if it gives as good result (>99% recovered), then use the clone to try and rebuild disk 1, if it doesn't, also use ddrescue on disk1