I have 2 encrypted pools:
- apps (1x DISK | 1 wide)
- tank (1x RaidZ2 | 4 wide)
I want to do a daily full backup of the apps-pool.
Therefore I created a replication task like this:
As described in the Truenas docs here, I think this shouldn’t even work.
However it did the first time and the whole apps root-dataset was replicated to tank/apps-backup and still was encrypted with the original key.
I then unlocked the dataset (tank/apps-backup) and everything seamed fine.
But now the next snapshot was created and when I now run the replication I get this error:
[EFAULT] Last full ZFS replication failed to transfer all the children of the snapshot apps@auto-2026-01-15_03-30. The snapshot tank/apps-backup/.system@auto-2026-01-15_03-30 was not transferred. Please run `zfs destroy -r tank/apps-backup@auto-2026-01-15_03-30` on the target system and run replication again.
If I then run zfs destroy -r tank/apps-backup@auto-2026-01-15_03-30 like told in the error and start the replication again, Truenas crashes.
I tried reading logs, but I can’t find anything useful to me.
What would be the proper way to backup the apps pool here?
How can I debug a crash?
PS:
My initial setup were two unencrypted pools with a similar replication task and that worked. Also my Truenas System is already running for several months without any other crashes, so I don’t think it’s about hardware, like some other topics suggest.

