SEZ9 commented on PR #12081: URL: https://github.com/apache/seatunnel/pull/12081#issuecomment-5964397723
Thanks for the detailed report on Build `36520988425` against head `1cbac0086` — this is exactly what I needed, and apologies for the delayed answer. The key outcome for me is that `sink_file_text_to_assert.conf` / the `MIN_ROW` assertion did not reproduce and `CheckpointEnableIT` passed on both JDK 8 and 11. That was the one case I wanted confirmed against the post-sync baseline, so I consider that item closed. As you noted, the merge-only sync does not need another F1–F8 pass from my side. On direction for the remaining `engine-v2-it` red: please treat them as suspected flaky/infra first and re-dispatch a fresh Build rather than diving into root-causing them right away. Concretely: 1. Re-dispatch once more on `1cbac0086` (no code changes) and report whether `BackpressureSlowSinkIT`, `CheckpointCoordinatorFailoverIT` (`610` vs `620`) and `SavepointBusySourceBarrierIT` (`Checkpoint expired before completing`) reproduce on the same JDKs. 2. If any of them reproduce a second time, please check the surefire logs for that test for anything coming from the imap-storage-file write path — i.e. a `TimeoutException` out of the timed `RequestFuture.get(timeout, unit)` or the `queryExecuteStatus` / `batchQueryExecuteFailsStatus` error logging covered by F3/F6/F8. If the WAL/checkpoint-storage path does not appear in those logs, I'm happy to treat them as unrelated to this PR and we can look at them separately. 3. `all-connectors-it-1` / `all-connectors-it-7` and `doris-connector-it` I agree look unrelated to the WAL/checkpoint-storage path; no action needed on those beyond seeing whether they clear on the re-run. Agreed that the Error / dead-worker case stays tracked in #12492 and out of scope here. Once the re-dispatched run is in, post the link and the pass/fail for those three ITs and I'll do the final pass on F1–F8 for merge. <!-- streview-comment:1475 --> -- 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]
