[FLINK-35890][cdc] Select only the increment but still enter the SNAPSHOT lead to ckp failed

[leosanqing]

ISSUE

Hi, when I was using Flink CDC Postgre, I chose 'debezium.snapshot.mode' = 'never' to skip the snapshot phase, but the code still entered the snapshot branch, and this log was printed, 'Database snapshot phase can 't perform checkpoint, acquired Checkpoint lock.' (Since our business creates the task first and then writes the data), the task cannot perform ckp, and the status is always stuck in in progress. After the timeout, the task restarts. Normally, if you only choose incremental data synchronization, you should not enter this branch.

The failure of ckp does not mean that the first one fails. After multiple ckp are successfully performed, it will then get stuck, causing a timeout and restarting the task. This is a problem that must occur. After it appeared in production, it also recurred many times in local testing.

Then we test, if there is data flowing in, the task status will be normal.

sql and conf like this

Why do this？

I think this miss one case，when I set "snapshot = nerver", I don't want to enter DB snapshot phase, and it can resolve my problem.

[leonardBang]

Thanks @leosanqing for the contribution, could you also check other connectors? I think some start up mode of debezium should not pass to underlying debeziunm engine once the Flink CDC incremental snapshot mode enabled.

[github-actions[bot]]

This pull request has been automatically marked as stale because it has not had recent activity for 60 days. It will be closed in 30 days if no further activity occurs.

[github-actions[bot]]

This pull request has been closed because it has not had recent activity. You could reopen it if you try to continue your work, and anyone who are interested in it are encouraged to continue work on this pull request.