Details
Description
Kafkacdc will connect to Kafka and read the latest GTID from it regardless of what is configured in the MaxScale configuration. Sometimes this is exactly what is needed but at other times, for example during replication topology changes, the GTID should not be read from Kafka.
Either this behavior needs to be made configurable so that it can be disabled or the reading of the GTID must see when either the domain changes or the stored GTID is newer than the one in Kafka.