davidzollo commented on PR #10447: URL: https://github.com/apache/seatunnel/pull/10447#issuecomment-5552450643
Pushed `caa98bfa72d` to close out the last CI blocker. **Root cause of the `Build` failure**: the only failing job was `unit-test (11, ubuntu-latest)` (run [33762900121](https://github.com/Suresh-Krishna-Kusuma/seatunnel/actions/runs/33762900121/job/100675975246)), and it's not a HubSpot-path issue. I downloaded the raw job log and the actual failure is: ``` CoordinatorServiceTest.testClearCoordinatorServiceDropsPendingJobsUnderRejectStrategy:642 Pending job scheduler did not enter preApplyResources ==> expected: <true> but was: <false> ``` This is a pre-existing test-isolation race in `CoordinatorServiceTest`'s teardown, unrelated to `connector-http-hubspot`/`connector-http-base`: `stopCoordinatorSchedulers()` called `scheduler.shutdownNow()` without waiting for the scheduler thread to actually stop, so a prior test method's background pending-job scheduler could still be running when the next test method started and asserted scheduler state — a classic cross-test bleed, not a HubSpot regression. This was independently root-caused and fixed on `dev` in #12050 ([Fix][Zeta] Stabilize CI test isolation](https://github.com/apache/seatunnel/pull/12050), merged 2026-09-04 — one day after this branch's last CI run on 2026-09-03), which makes teardown block on `awaitTermination()` before returning. This branch simply predated that fix (branch was `behind_by=19` against `dev` at the time). Since #12050 only touches test-isolation code (`CoordinatorServiceTest`, `BaseServletTest`, a handful of e2e IT classes, and two Hazelcast test-only yaml configs — nothing in `connector-http-hubspot` or `connector-http-base`), I cherry-picked just that one commit onto this branch instead of doing a full `dev` sync, to avoid re-opening the kind of Airtable/FTP merge churn we hit earlier in this PR's history. `spotless:apply` on the touched modules reported no changes needed (the commit is already dev-clean). `Build` is running on the new head now: https://github.com/apache/seatunnel/runs/101321214358 I also replied to and resolved the two open review threads (the sink-connector follow-up suggestion, and the stale Github.md doc question — confirmed via the latest source-level review that the shared `HttpSourceFactory` refactor is additive-only and needs no Github.md update). -- 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]
