Replication failed with "destination has snapshots"

[seanm]

I periodically physically connect a backup disk and use Tasks > Replication to run a backup. Has worked great generally.

Yesterday though I did what I usually do but it failed with:

Code:

Replication "offsite2" failed: Full ZFS replication failed to transfer snapshot tank1@OFFSITE-2022-06-22_11-51 as a whole. The error was: Warning: offsite2/tank1/zvol1: property 'sharenfs' does not apply to datasets of this type Warning: offsite2/tank1/zvol1: property 'mountpoint' does not apply to datasets of this type Warning: offsite2/tank1/zvol1: property 'sharesmb' does not apply to datasets of this type Warning: offsite2/tank1/zvol2: property 'sharenfs' does not apply to datasets of this type Warning: offsite2/tank1/zvol2: property 'mountpoint' does not apply to datasets of this type Warning: offsite2/tank1/zvol2: property 'sharesmb' does not apply to datasets of this type cannot receive new filesystem stream: destination has snapshots (eg. offsite2/tank1/zvol3@OFFSITE-2022-05-02_11-45) must destroy them to overwrite it warning: cannot send 'tank1/zvol3@OFFSITE-2022-06-22_11-51': signal received. The snapshot offsite2/tank1/.system@OFFSITE-2022-06-22_11-51 was not transferred. Please run `zfs destroy offsite2/tank1@OFFSITE-2022-06-22_11-51` on the target system and run replication again..

I ran the zfs destroy command it suggested and tried again, but it failed again with basically the same message.

Anyone know what the issue is here?

Thanks.

 

[sretalla]

Are you running recursive snapshot jobs on the remote side? if you add snapshots to the replica/remote side, it's always going to bork your syncs.

 

[seanm]

I never create snapshots on the remote side. The disk sits in a drawer, once a month I plug in into the NAS, import the pool, start the replication task, then unmount the pool, and disconnect the drive. The procedure has worked for a year or more, but this time I got those errors.

Maybe if I delete the last snapshots created by the replication procedure I'd be able to start it again?

 

[seanm]

Well, I deleted the snapshots on the destination and ran replication again, then it succeeded. Phew.