I have two 6.1.9 unraid servers.  One is a backup/test server.  Had a drive fail on the main server, identical to the drive on the main server.  What would be the proper procedure to pull the drive from the test server and replace the failed drive.  Do I need to do a preclear first, THEN bring it online, or can I just format it (seems like that would work).  Or will unraid just recognize it as a replacement drive and do what it needs to do?

 

Thanks in advance!

Formatting is never part of the rebuild process. Format means "write an empty filesystem to this disk". That is what it has always meant on every operating system you have ever used. If you format a disk after you have assigned it to the array it will have an empty filesystem and parity will be updated to agree that the disk has an empty filesystem.

 

There is no need to preclear if you trust the disk since every bit of it will be overwritten by the data of the old disk.

 

Stop, shutdown, replace disk, boot, assign new disk to slot of old disk, start and unRAID will rebuild.

 

Also you will have to new config and rebuild parity on the test server as a result of removing a disk from it.

I'm not sure I completely understood what you are hoping to do, so ignore me if I'm wrong!

 

But a couple of thoughts, when a drive fails, its data is at greater risk than normal, so any backup of that data should be especially protected, until the risk returns to normal.  If you were thinking of using the backup copy for rebuilding on, I wouldn't, because you would lose it as a backup, raising the risk to that data even higher.  If you were thinking of slotting it into the failed drive's place, then the main parity would be invalid, and you would have to rebuild parity, plus you would be giving up the emulated copy of that drive and its data.

 

In my view, the best course would be to not touch the backup at all, "just in case", and add a NEW drive as a replacement for the failed drive, and rebuild onto it as trurl said.

Here's what I ended up doing and it worked quite well.  The Failed drive was on Server A, I had a spare server (Server B) with an identical drive (no data on the drive).  I pulled the drive from Server B, replaced it in the slot on Server A.  The system booted, recognized the drive was missing.  I stopped the array, then I selected the drive from the list (only one) and started the array back up.  The system started a parity rebuild and a few hours later, we're back and running. 

 

Interestingly enough, I put the 'failed' drive in Server B, and it rebuilt the drive with no issues (it was the parity drive).  I'm going to run smart tests and such, and there's little data on that server.  But it is an interesting test.  I might take it offline and run a preclear on it...