I have been happily using zfsnap with CentOS and ZFSonLinux in a scripted manner similar to how it is described on the web site, but have run into an issue. Unfortunately backups did not run a few days ago, and now I am stuck in an odd situation. I have a series of incremental backups on my backup ZFS server from my main ZFS server and some newer snapshots on the main server. Unfortunately due to the missed backup and automatic destruction of older snapshots on the main server there is no longer a common snapshot between the two systems to perform an incremental send with. i.e. on backup there is:
backuppool/x@snap2 backuppool/x@snap3 backuppool/x@snap4
and on the main zfs server there is:
mainpool/x@snap6 mainpool/x@snap7 mainpool/x@snap8
Is there any way to get them in sync again or do I have to destroy everything on backup and perform a full(there is not enough room for two full backups)? I tried specifying a send referencing the name of snap4 on the remote backup server from the main server, but as the main server does not have snap 4 it cannot calculate the incremental. I also tried to send snap4 from the backup to the main server, but it appears that the only way to do this would be to do it as a full send(not enough room), and I would need to wipe out existing snapshots. The deleting of all your backup history if you get snapshots mismatched seems pretty inconvenient, is there any solution in this scenario?