DanielLeens commented on PR #12081: URL: https://github.com/apache/seatunnel/pull/12081#issuecomment-5662001869
Thanks for the recap and the cc, @Rangsh — this matches what I independently verified in my own last pass (https://github.com/apache/seatunnel/pull/12081#pullrequestreview-5193499943): - `7c74bf982` — closes Issues 1–3 from my prior round. - `3d9f80f04` — narrows both `notifyCheckpointMonitor` helpers (`CheckpointCoordinator.java`, `CheckpointManager.java`) from `catch (Throwable t)` to `catch (Exception e)`, matching `WALWorkHandler.onEvent()`'s own deliberate Exception-only boundary so a genuine `Error` still propagates. Confirmed closed on my side. - `7f69d7f69` — the `dev` sync merge. I diffed the pre-merge tip against the merge commit for every file this PR owns and found only `CheckpointCoordinator.java` picked up any change, from `dev`'s unrelated `#11503` landing in a different method — zero overlap with `notifyCheckpointMonitor`, zero conflict markers, narrowed `catch (Exception e)` intact verbatim. My conclusion on the synced head is still **Ready to merge**, no blockers. On CI: the fork's run for `7f69d7f69f` is red on `engine-v2-it` (both JDKs), several connector-IT shards, and cancels `kudu-connector-it`, but I traced `engine-v2-it` specifically (the one module that actually covers this PR's changed package) against `dev`'s own HEAD-of-branch CI at the exact commit this PR merged from — it's *also* red on `engine-v2-it` on both JDKs there, on different specific test methods each run. That's the standing "flaky/environmental at HEAD-of-`dev`" signature, not a regression this merge introduced. The connector-IT failures are outside this PR's changed files entirely. I wouldn't block merge on a bare CI rerun turning fully green given that trace, but a maintainer should feel free to ask for one before merging regardless. @SEZ9 — welcome to take your pass directly against `7f69d7f69f`; happy to compare notes if you land on a different read of any of the WAL/RequestFuture items Rangsh scoped above. -- 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]
