ZFSREP automates the replication and management of ZFS volumes (raw) or filesystems (cooked) at block level using ZFS 'send' and 'recv'. Set it up and just schedule it via 'cron' to run as often as you wish.
Essentially this gives you the same functionality as Netapp SnapVault between Netapp arrays, but here we do it on the host level so it is 100% storage agnostic.
Both the source and destination pools and volumes/filesystems need to already exist. ZFSREP will take care of taking initial snapshots on the SRC host, so ideally there should be no existing snapshots on any pools/volumes/filesystems that you want ZFSREP to replicate. SSH public key auth must be enabled from SRC host to DST host.
poolname/fsname:max_src_snap_to_keep:max_dst_snaps_to_keep
or
poolname/volumename:max_src_snap_to_keep:max_dst_snaps_to_keep
mybigdatapool/data1:2:5 mybigdatapool/data2:2:5 mylogpool/archive:1:10
It is NOT permitted to have a snapshot keep count of zero(0), else incremental snapshots will NOT work!!!