Looking to export a server to an external drive, wipe, and copy back

achosid · October 30, 2025, 6:41pm

I had a poorly set-up server with mismatched drives and a failure of one drive has made me rethink my architecture. I’ve replaced drives and ended up with my previous pool functioning correctly with 4x12TB drives, but they’re not configured in my eventual plan of RAIDZ1. I am trying to sort out the easiest way to get everything off the server, onto a large external drive I have, get the server reoriented as RAIDZ1 so I can survive single drive failure, and restore the data back onto the system. I have done some googling and had a long failure of trying to navigate this process with ChatGPT (a bad idea). Is there a simple way to do this? I have a Windows machine that is always on with SMB access to the server, if that is of any help.

argumentum · October 30, 2025, 6:50pm

So you’ve got 3x12 TB to copy to a 1x36 TB in windows ?
I’d get me another PC to plug drives in, then when done storing elsewhere, plug those drives back to the original NAS/PC/box ?, maybe.

etorix · October 30, 2025, 6:58pm

Mind that the “large external drive” would likely have no redundancy.

USB drive I suppose? Plug it to the NAS. Create a single drive pool on it. Replicate to this pool. Export and destroy the old pool. Create your new pool. Replicate in the other direction.

achosid · October 30, 2025, 9:38pm

Mind that the “large external drive” would likely have no redundancy.

The goal is that this would be a very temporary operation, basically copying over, creating the new pool, and copying back.

I’ve never used Replication before, I put together what made sense to me as a replication task and it completed immediately, saying “No snapshots to send for replication task”

etorix · October 30, 2025, 9:49pm

Time to learn about snapshots then…

Arwen · October 31, 2025, 6:34am

Gee, I just wrote up command line pool cloning in another thread:

How to rebuild mirror/replicate a pool

zfs snapshot -r ${MY_SRC_POOL}@${MY_SNAP}
zfs send  -Rpv  ${MY_SRC_POOL}@${MY_SNAP} | \
  zfs receive -dFu ${MY_DEST_POOL}

zfs destroy -rv ${MY_SRC_POOL}@${MY_SNAP}
zfs destroy -rv ${MY_DEST_POOL}@${MY_SNAP}
Replace ${MY_SNAP} with a unique snapshot name. Then replace the pool names as well.

When done, you can export the original pool and rename the new pool to be the same as the old pool. That is covered in a Resource forum post if I remember correctly.

achosid · November 3, 2025, 8:42pm

I started a snapshot routine, tied it to a replication task, and now the size of files on the external drive matches the size on the current pool. Verifying I understand this correctly before I start hitting buttons and losing all my data: I export the current data pool, check the “destroy data” tick box, create a new pool with the existint 4x12TB drives with RAIDZ1 enabled, go to Data Protection and Replication, and set a replication task from the backup to the new pool? I see already that there are two snapshots on the external drive, so I do not expect the same errors I encountered before.

Arwen · November 3, 2025, 10:14pm

Without seeing exactly what you did and now have, it is hard to say you did it right. Sorry.

ZFS has some amazing features, but it is complex enough that casual users have been known to shoot themselves in the foot, (or worse…). Add in an evolving GUI for TrueNAS CE, and that can make things worse.