Rangsh commented on PR #12081: URL: https://github.com/apache/seatunnel/pull/12081#issuecomment-5944443124
@SEZ9 thank you for the clear guidance — and for confirming that the merge-only sync does not need another F1–F8 pass. Here is the outcome of fork Build [`36520988425`](https://github.com/Rangsh/seatunnel/actions/runs/36520988425) on head `1cbac0086`: - The run completed with **failure** (not green overall). - Specifically for the case you asked about: **`sink_file_text_to_assert.conf` / the `MIN_ROW` assertion (expected `>=100`, got `80`) did not reproduce**. `CheckpointEnableIT` passed in the uploaded `engine-v2-it` test reports (both JDK 8 and 11). - `engine-v2-it` itself is still red, but on different failures this time: - **JDK 8:** `BackpressureSlowSinkIT` (checkpoint progress under sustained backpressure); `CheckpointCoordinatorFailoverIT` (exact row count expected `610`, was `620`) - **JDK 11:** the same `CheckpointCoordinatorFailoverIT` (`610` vs `620`); `SavepointBusySourceBarrierIT` (`Checkpoint expired before completing`) - A few other jobs also failed (`all-connectors-it-1` / `all-connectors-it-7` on JDK 11, `doris-connector-it` on JDK 8), which look less related to this PR’s WAL/checkpoint-storage path. Happy to dig into the remaining `engine-v2-it` failures next if that is what you would like before merge, or to re-dispatch a fresh Build if you prefer to treat them as flaky / infra first. Please let me know which direction you prefer — and thank you again for shepherding this. Error / dead-worker remains tracked in #12492 and out of scope here, as agreed. -- 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]
