TL;DR: I have one server and one spare drive. Is local replication the best way to create a backup for offsite storage?
The rest of it:
I’m trying to understand the best way to create a backup for the “1” in the “3-2-1” backup strategy, while still being able to update it on occasion. I’ve read into it and come across some confusing and sometimes conflicting information. Here’s my situation:
I’m running a 16TB pool (1 three-wide mirror) on a Dell T330 (TrueNAS Scale 24.10.2). I have a 4th 16TB drive which is currently unused (we’ll call it “#4”), I’d like to use that for the main offsite backup.
The only other PCs I have that support drives of that size don’t have ECC RAM, so I don’t really have reliable hardware to run a 2nd server for rsync or anything.
My (limited) research on the topic is suggesting that local replication is the best solution here. Is that the correct assumption? If so, I just have a few questions:
-
The first time I replicate it creates a 1:1 clone of the pool, datasets, EVERYTHING -not just snapshots- to #4 correct?
-
So if the house burns down, I can buy a new server, put #4 in, import the pool to brand new drives, and essentially be right back where I left off, correct? (The encryption key and system config are safely backed up as well).
-
Let’s assume I copy some new stuff (Piano Recital) to my server. When I plug #4 back into the server and perform another replication task, will it redo the whole thing, only copy Piano Recital, or only the snapshot referring to Piano Recital?
Snapshots are where I start to get a bit confused. I’d hate to replicate over the years only to find out in an emergency that replication wasn’t the right method.
Is there anything else I should know, or any settings to enable/disable when performing a replication?
As a side-note, I’m not as concerned with snapshots since I’m only ever adding data and not deleting it. But again, I could be completely misunderstanding what snapshots are for.
Thank you for reading!