Bug report

For bug reports, provide this information, please:

What Debezium connector do you use and what version?

3

What is the connector configuration?

multi-shard

What is the captured database version and mode of deployment?

(E.g. on-premises, with a specific cloud provider, etc.)

vitess v15

What behavior do you expect?

When starting with a current for several shards for a single task, we get "Invalid GTID" errors for each shard as it flips from current to a real transaction. This is from a bug in the logic when there is a partial flip (see unit test in PR)

What behavior do you see?

The behavior should be we do not wrongfully fail on this case and handle cases where a VGTID is partially overridden and partially standard GTIDs.

Do you see the same behaviour using the latest released Debezium version?

Yes

Do you have the connector logs, ideally from start till finish?

(You might be asked later to provide DEBUG/TRACE level log)

<Your answer>

How to reproduce the issue using our tutorial deployment?

<Your answer>

Implementation ideas (optional)

Will put out a PR