How to best utilize three disks for resilience, when data fits in one disk?
Currently, I have one disk + one parity disk (SnapRAID). I have a spare disk and I want increase general resilience of data by using it as the third disk in the array. Data fits in one disk and will continue to do so for the foreseeable future. I’m cloud-averse due to the costs of it, plus not having it reduces the threat complexity. What would be the best way to do utilize the third disk?
**# One possible solution**
Have 4 disks, 3 of them in a SnapRAID array (2 mirrors of each other (via Drivepool) as data disks + 1 parity disk), and one off-site. Regularly bring the off-site disk back and hot-swap it with one of the (fully-synced) SnapRAID disks, then rebuild (“snapraid fix -d offsite-disk” + “snapraid sync”). Take the disconnected disk offsite (no need to wait until sync finishes). ([Alternative solutions](https://pastebin.com/PuRbaWPg))