hesam-oxe commented on PR #11077: URL: https://github.com/apache/seatunnel/pull/11077#issuecomment-5874939244
Thank you @DanielLeens and @SEZ9 for the thorough triage and in-depth reviews throughout this PR. **Summary of current state (head 1bfcea62):** - **F1 (destination-key collision, HIGH):** Resolved — `DestinationKey` now folds in `sink.getClass()` + connector-supplied `getPhysicalDestinationIdentifier()`, with fallback to object identity when absent. Covered by `testSamePhysicalIdentifierDoesNotShareAcrossConnectorClasses` in `MultiTableSinkWriterTest`. - **F2 (snapshot fan-out / restore union, HIGH):** Resolved — shared writer state is emitted once via `groupByIdentity` and recorded under canonical identifier (`aliasedIdentifiers.get(0)`). Restore path `getRestoredState()` scans full alias set, so ordering is not a concern (verified `SinkIdentifier` is content-based). - **F4-F6 (docs / proxyContexts / restoreWriter contract):** Addressed with docs updates and `proxyContexts` fix. - **Quarantine path `CONTINUE_OTHER_TABLES`:** Fixed in e3e57c19518e to isolate shared writer failures. **CI:** Thanks for confirming failures are pre-existing flakes (RocketMQ double-identity #12115, Kudu hang #12132, Paimon privilege restore hang, Doris container startup #12142, FilterRowKind fixed by #11995, PostgresCDC tombstone NPE, etc.) and not related to `seatunnel-api/.../multitablesink/*` or `connector-file-base`. I've rerun failed jobs for the current head; the fork run is now in_progress (https://github.com/hesam-oxe/seatunnel/actions/runs/35497458667). Happy to address any remaining concerns. Ready for final review when you have bandwidth. -- 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]
