CryoThrust opened a new pull request, #12149: URL: https://github.com/apache/seatunnel/pull/12149
## Motivation When a streaming job is cancelled, `CheckpointCoordinator.cleanPendingCheckpoint` shuts down its scheduler and immediately creates a replacement scheduled executor. That replacement can leave an idle `checkpoint-coordinator-*` thread alive after the job reaches `CANCELED`, which is detected by the e2e thread-leak check (see #12147). ## Changes - Centralize scheduler creation and preserve the remove-on-cancel policy. - Recreate the scheduler only for `CHECKPOINT_COORDINATOR_RESET`, where the coordinator is expected to restart after master failover. - Keep terminal cleanup fully shutdown. - Add a regression test asserting cancellation leaves the scheduler shut down. Fixes #12147 ## Verification - `git diff --check` passed. - `spotless:apply` completed successfully. - Module compile is blocked by unrelated baseline errors in JettyService: HttpConfig#getUploadMaxFileSizeMb and getUploadMaxRequestSizeMb are missing on origin/dev. -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: [email protected] For queries about this service, please contact Infrastructure at: [email protected]
