It's not possible to select a totally clean pool, just created, as the target of a Replication Task. Because this error log:
[2020/03/14 17:22:28] ERROR [replication_task__task_1] [zettarepl.replication.run] For task 'task_1' non-recoverable replication error ReplicationError("Target dataset 'tank2' does not have snapshots but has data (541632 bytes used) and replication from scratch is not allowed. Refusing to overwrite existing data.")
But, there is no data in pool! It's just created.
And two questions about the new replication system please:
1) Do I need to create a "readonly=on" as the dataset tree replication target, or the system is smart enough to handle it by itself?
2) Do I need to manually replicate the source dataset tree in the target dataset? Or the system do it by itself? Thanks.