How best to handle remote repository after a zfs rollback?

[kevdogg]

I was doing some testing this afternoon on my ZFS machines:

1. Arch linux VM with ZOL
2. FreeNAS

I made a number of snapshosts on the client (arch linux) using:

znapzend --debug --runonce tank/data/home

which then set snapshot to FreeNAS. I then performed a rollback on the client and was curious how znapsend would handle this situation.

I got the following error complaining about no common found on source and destination:

[Sat Feb  1 15:53:58 2020] [debug] sending snapshots from tank/data/home/kevdog to root@10.0.1.197:tank/backups/zfs_backup/archZFSProxy/home/kevdog
# zfs list -H -o name -t snapshot -s creation -d 1 tank/data/home/kevdog
# ssh -o batchMode=yes -o ConnectTimeout=30 root@10.0.1.197 zfs list -H -o name -t snapshot -s creation -d 1 tank/backups/zfs_backup/archZFSProxy/home/kevdog
[Sat Feb  1 15:53:58 2020] [warn] ERROR: snapshot(s) exist on destination, but no common found on source and destination clean up destination root@10.0.1.197:tank/backups/zfs_backup/archZFSProxy/home/kevdog (i.e. destroy existing snapshots)

What's the best work around for this problem?

[oetiker]

you have to remove the excess snapshots on the destination by hand