I posted this at the SF discussion forum. Unsure if that place is even read anymore, I am also posting here.
I find Webmin’s “Filesystem Backup” module very powerful. I know it is currently possible to use it with ZFS as source directory. Webmin doesn’t care what the underlying filesystem is. While that works fine, it’s not taking advantage of ZFS snapshot and send/receive features.
This is a feature request to add in the ability to schedule a snapshot of an entire pool or specific dataset (zfs snapshot -r Pool/Dataset@snapshotname). We should be able to specify a retention period for a snapshot (that Webmin will go and delete after the specified date) and optionally have Webmin send (pipe) it somewhere (zfs send -r ). “Somewhere” can be another ZFS filesystem (to create a duplicate pool or dataset) or a compressed archive. The destination is either local or remote. Remote is usually piped through SSH.
For some prior art and implementation examples you may be interested in look at these projects: