Working through an issue I found in a Tap where it seems my Target is eagerly truncating old table data on incremental sync, only leaving the new records in the batch. I am thinking the issue is here, as an activate version message is sent on all paths of the tap including on an incremental sync, and the target implementation is truncating the table before syncing the newer records. Does this sound correct? I turned off Activate Version in my target for this combination to get around for now, but want to make sure this makes sense and my data didn't disappear for another unexplained reason. https://github.com/Monad-Inc/tap-redshift/blob/main/tap_redshift/__init__.py#L362

Hi Ben! Your assessment makes sense.

ACTIVATE_VERSION

does not make a lot of sense for incremental replication and is usually reserved for full-table replication, but the implementation is up to the tap. Good to keep in mind when we around to implementing https://github.com/meltano/sdk/issues/18.

Thanks @edgar_ramirez_mondragon, good to have it validated. I disabled the activate version feature on the target when used in combination with that tap and appears to be functioning as expected before, I was just very confused why my tables had data on Friday which was gone on Saturday 😂