If you are not familiar, the ReadyNAS is a network device you can buy which hosts disk space. You can use it for file shares, or as a place to back up all of your computers on your network.
I’ve been getting ominous warning e-mails every few days from my ReadyNAS that one of the drives was generating errors and that I should expect a failure soon. Today, it happened – I got two e-mails from the device:
- Volume ‘DataRAID10’ health changed from ‘Redundant’ to ‘Degraded’.
- Disk in channel ‘1’ (Internal) changed state from ONLINE to FAILED.
I logged into the web interface and sure enough, the first drive was red and showed as failed.
What about RAID?
Considering that I wanted this to be super-indestructible AND considering that disk space is crazy-cheap, I got four 4TB drives. Then, I set them up as RAID 10. This means that disk 1 and disk 2 are “striped”, or combined to make one big 8TB disk, and then those two disks are mirrored to disk 3 and 4.
This means that I could lose one or potentially two disks (if it was the right two disks) and I would not lose the logical drive.
Just as expected, with just disk 1 down, the logical drive is still available and all is well. All I need to do is pull the drive, put in a new one and it should rebuild the replacement… right?
Replacing the drive:
Physically replacing the drive is simple: you open the front “door” of the device, hit the button and pull out drive 1 – then unclip the drive from the tray. No tools required. From a previous project, I already had the exact replacement drive for this available on a shelf – so I used that. I put that replacement drive in the tray, and then re-inserted the tray.
This didn’t work the way I thought it would.
I was expecting this to be fully-automatic, but I needed to do some steps. I didn’t take screenshots, because I wasn’t expecting to do a blog post – but let me try to describe:
- In the web portal – click the replacement drive, click the gear, go into settings and “DESTROY” the existing partition – for the replacement drive.
- Once the UI updates and it is stable (in a minute), reboot the ReadyNAS.
- Now, with the empty drive available – you can’t do anything from the UI, so, I pulled the replacement drive and reboot again.
- In the web portal – it shows disk 2,3, and 4 as online and stable – and disk 1 missing.
- I plugged in disk 1 and the ReadyNAS immediately started bringing it back into the RAID set
So now, I see:
Note how drive 1 has a yellow icon, and ReadyNAS still reports the RAID volume as degraded. I expect in ~:07 hours when this is done bringing the replacement drive online, the volume will go back to being “Healthy” again.
I might have been able to do this with just one reboot: pull the drive, put the new drive in and destroy the partition table – then reboot with the drive unplugged. Plug the drive in and the rebuild should kick off automatically.
I obviously didn’t test that, but that (I’m guessing) might work. But bottom line, this only took about :05 minutes to remedy – given that you already have the hot spare available. Thanks, Netgear!