davidzollo commented on PR #11812:
URL: https://github.com/apache/seatunnel/pull/11812#issuecomment-5386384324

   Checked the failing `Build` check on the current head (`29f1754b`) against 
the fork's Actions run — both failing jobs are environmental, unrelated to this 
PR's classloader-release fix:
   
   1. **`all-connectors-it-1 (8, ubuntu-latest)`** — failed on `CouchbaseIT`, a 
`Connection reset` during container bootstrap. This is a known, pre-existing 
flake unrelated to any specific PR (Couchbase's `containerIsStarting` REST call 
isn't covered by the container's startup-timeout retry). There's already an 
open fix for it (adds `startupAttempts=3` + a log consumer) — this class of 
failure should stop once that lands.
   
   2. **`engine-v2-it (11, ubuntu-latest)`** — failed on 
`SinkErrorToMysqlIT.testSinkMaxErrorRatioThreshold`, an `Awaitility` timeout 
waiting for the container's JVM threads to fully drain (`HikariPool-1 
housekeeper`, `st-multi-table-sink-writer-1` still alive after 120s). I traced 
this: the check lives in the generic, shared 
`SeaTunnelContainer.doExecuteJob()` thread-leak assertion used by essentially 
every engine E2E test, and the two lingering threads belong to the sink's own 
JDBC connection pool and multi-table sink writer executor — neither is touched 
by this PR's diff, which only modifies 
`TaskExecutionService.deployTask`/`deployLocalTask`'s classloader 
acquisition/release bookkeeping. So this reads as a pre-existing 
timing-sensitive check tripped under CI load, not something introduced here.
   
   I don't have push/rerun rights on this fork (confirmed both via API 
permissions and a rejected check-run rerequest), so I can't retrigger the two 
failed jobs myself.
   
   @goutamadwant — could you re-run the two failed jobs from your fork's 
Actions tab? No code changes needed on your side. Once that's green, I don't 
see any remaining CI-side blocker.


-- 
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]

Reply via email to