Feature request or enhancement

Which use case/requirement will be addressed by the proposed feature?

When for whatever reason a large volume of operations not concerning a table being watched is written to the logical log, the connectors last committed LSN may shift out of the log. If the connector is then restarted its last committed LSN is no longer in the logical log and the connector fails to start.

Implementation ideas (optional)

Implement heartbeat.action.query functionality similar to that of the PostgreSQL connector that can execute an (ignorable?) update to a watched table, forcing an update to the last committed LSN.