hutiefang76 opened a new pull request, #12448: URL: https://github.com/apache/seatunnel/pull/12448
### Purpose of this pull request Fixes #12442. A periodic checkpoint trigger that throws unexpectedly can exit its scheduled callback before scheduling the next trigger. The pipeline then keeps running without further checkpoints. This change catches that failure at the scheduled callback, logs the checkpoint type, job ID, and pipeline ID through SeaTunnel's logger, and calls the existing coordinator error handler. The error handler marks the coordinator as failed and informs the checkpoint manager. Failing through this established path makes the fault visible to the job and avoids silently retrying a deterministic exception forever. ### Does this PR introduce _any_ user-facing change? Yes. On `dev`, an unexpected periodic trigger exception previously left checkpointing stopped without a job-level failure. The coordinator now reports a failed state and the checkpoint manager receives the error. ### How was this patch tested? - Added `CheckpointCoordinatorScheduledTriggerTest`: an injected exception used to make the coordinator wait time out before the fix; after the fix, it observes `FAILED` and verifies checkpoint-manager notification. - On JDK 11, the module reactor package passed with `-Dtest=CheckpointCoordinatorScheduledTriggerTest,CheckpointCoordinatorTest`; Surefire reports 17 tests, 0 failures, 0 errors, 0 skipped. - `mvn -q -o -pl seatunnel-engine/seatunnel-engine-server spotless:check` passed. ### Check list * [x] No new Jar binary package was added. * [x] No connector code was changed. * [x] No configuration or API incompatibility was introduced. -- 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]
