In the process of reorganizing my attic, which is where I keep my server and other network stuff, I accidentally moved the hdd case that I had outside my server (temporarily) and one of the sata power cable partially disconnected while operating.

Immediately I received an email notification that my array had a failed device. I checked the hdds and noticed the one partially disconnected. Connected the power cable properly and rebooted the server. The message still appears but I'm reasonably confident that the disk is okay, meaning that it wasn't failing and was just a matter of time. My array is composed of 6 drives, including 2 parity drives. The failed drive is one of the parity drives.

Do you have any suggestions about how to proceed? Is my only option to recompute the parity of the entire array? or maybe this drive has like a "dirty bit" that I can clean up?

7 hours ago, Mad_cyclist said:

Is my only option to recompute the parity of the entire array?

Just for the failed parity drive, you could also do a new config and check parity is valid, but will need to run a correcting check, and it will take the same time.

Will this only work on the parity drives then? Regardless of the time, I suspect that would put less strain on the system (drives) than doing it on the entire array? If so, could you point me to some thread or part of the documentation where this process is detailed?

Edit: Is this the procedure you're referring to? https://docs.unraid.net/unraid-os/manual/storage-management/#rebuilding-a-drive-onto-itself

Edited July 11, 2025Jul 11 by Mad_cyclist

58 minutes ago, Mad_cyclist said:

I suspect that would put less strain on the system (drives) than doing it on the entire array?

Any solution will read the entirety of all drives.