FrankChen021 opened a new issue, #20604: URL: https://github.com/apache/druid/issues/20604
_This issue was generated automatically by Claude Code (Anthropic's AI coding agent) running a scheduled CI-triage routine on behalf of @FrankChen021. Analysis and suggested fixes are AI-produced; please verify before acting on them._ **Status:** Open: no fix PR **Subject:** `KafkaIndexFaultToleranceTest.test_supervisorRecovers_afterCoordinatorRestart` (embedded-tests, `unit tests / unit tests(main) (25, K*,U*,Z*,Y*,X*)`) **Failures:** 1 · **First seen:** 2026-10-14 · **Last seen:** 2026-10-14 ### Root cause `verifyAndTearDown` timed out after 240 s waiting for `ingest/rows/published` to reach 3000 after a Coordinator stop/start with transactional Kafka records. Publishing stalled, likely because the handoff of tasks running during the restart was delayed and a later task group could not publish before the earlier one finished. #20505 (merged a few hours earlier) now blocks a task group's publish until earlier groups finish, which may lengthen such stalls; watch for more occurrences. The commit (#20577) only bumps GitHub Actions versions. This class passed in attempt 1 of the same run, #20505's own run passed, and the next master commit was green. ### Suggested fix After `coordinator.start()`, wait for a Coordinator duty metric (for example `coordinator/time`) on its `latchableEmitter()` before publishing the last 1000 records, so handoff can resume. If it recurs, capture the task logs and supervisor status on timeout to check whether a task group was waiting on an earlier group's publish. ### Occurrences Failed push-triggered master jobs only. The daily triage routine adds one row per new failed job. | Date | Commit | Job | Failure log | Detail | Reported in | |---|---|---|---|---|---| | 2026-10-14 | afe71f5381 (#20577) | `unit tests / unit tests(main) (25, K*,U*,Z*,Y*,X*) / test-jdk25-[K*,U*,Z*,Y*,X*]` | [job 116011534392](https://github.com/apache/druid/actions/runs/38585343856/job/116011534392) | `ISE: Timed out waiting for event after [240,000]ms` in `verifyAndTearDown` (published rows < 3000); 1/1 attempt; run attempt 2 | | -- 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] --------------------------------------------------------------------- To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
