Replacing a Failing (But still functional) drive

Have a 3TB drive that’s throwing an increasing number bad sectors so going to replace it, since I have spare drive bays available am I right in thinking the best way to replace it is to install the new drive first rather than just pull it?

Running BTRFS raid5

Do I just issue a btrfs replace command from the CLI/SSH or is there a way I’m supposed to do it from within the GUI so the rockstor services know it’s been replaced?

If you are comfortable with CLI and btrfs tools, you can do so.

From the GUI, there’s no atomic replace feature. You can add a new drive and then remove the failing drive as two separate steps. Each of them will trigger a balance, so may be replace is better/faster if you are comfortable with CLI.

BTRFS replace seems to be the recommended way to do things according to the BTRFS docs so if it’s not going to upset rockstor then I’ll do it that way.

Looks like a case of shut-down Rockstor , pass through the new drive in the proxmox config for the Rockstor VM then run BTRFS replace

Also did find it amusing that despite getting multiple warnings of bad sectors and the drive failing it’s extended smart test twice with read errors it still shows smart status = Passed.when queried using smartctl on the proxmox host

1 Like