Hello! We have been running Postgres RDS-> Redshift CDC syncs for a while now, using the 1.0.1 version of Postgres connector as newer versions don't pick up any changes for us. Things were great for a while, but we've been noticing unexpected behaviour, where instead of an incremental sync we get a full one at least once a day. These full syncs take a lot longer than incremental syncs, so this is very inconvenient for our end user. One line in the log

Saved offset is before Replication slot's confirmed_flush_lsn, Airbyte will trigger sync from scratch

points at why full syncs are being initiated. My Q is why could this be happening? Is it an error from the driver's side - it's not storing the latest lsn offset? Is the database not acknowledging this offset? Has anyone encountered this before?

I didn’t see this before @Kristina Ushakova. Do you mind creating a Github issue, maybe sharing logs and screenshots to help the discussion