davidzollo commented on PR #11569:
URL: https://github.com/apache/seatunnel/pull/11569#issuecomment-5396756323
CI failure triage for the last red Build (fork run 32582046522, both
attempts, head `0d9f281e054a`):
1. **`kafka-connector-it` — deterministic on the stale branch, not caused by
this PR.** Both attempts failed on the identical test, line and container:
`KafkaIT.testKafkaToKafkaExactlyOnceOnStreaming{TestContainer}[2]` timing out
in `awaitStreamingPipelineReady` (KafkaIT.java:1327), on java 8 and java 11.
The branch was based on 2026-08-13 dev (95 commits behind) and misses the Kafka
exactly-once fixes merged since: #11541 (exactly-once sink losing the first
record on checkpoint), #11231 (wait for exactly-once test sends), plus
#11572/#10925 e2e stabilizations. This PR only touches the JDBC XA path, so the
exactly-once Kafka failure is stale-branch fallout.
2. **`unit-test (11, windows-latest)` — known environmental flake**:
`IllegalStateException: Node failed to start!` in `seatunnel-engine-server`
(`CheckpointPlanTest`, 311s). The same Hazelcast node-start flake has hit
unrelated PRs on windows-latest with a rotating test class. The java-8 twins
were fail-fast matrix cancellations of the same event; the branch also predates
#11873, which set `fail-fast: false`.
**Action taken:** merged current dev (`43fe63b1fca`) into the branch as a
maintainer edit — no conflicts; the diff vs dev is still exactly the PR's XA
changes plus docs. `spotless:apply -pl seatunnel-connectors-v2/connector-jdbc`
clean. A fresh Build run is in progress on head `403a0b09f88d`. If the Windows
Hazelcast flake strikes again it only needs a job-level rerun, not code changes.
--
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]
