DanielLeens commented on PR #12134: URL: https://github.com/apache/seatunnel/pull/12134#issuecomment-5664748651
Per the 2026-09-14 review's Issue #1: added a Javadoc-only acknowledgement of the cascading double-failover limitation on `JobMaster#markRedriveSavepointAfterRestore()` and `CoordinatorService#redriveSavepointAfterMasterFailover(...)` in 8844f46b61dcb899fb3cfd4d5e3db5562e2a9dbd. The new paragraphs state that the redrive covers a single master failover during an in-flight savepoint: a second failover landing after `restoreJobFromMasterActiveSwitch` has re-persisted the job as `PENDING`/`RUNNING` but before the redriven `savePoint()` has re-persisted `DOING_SAVEPOINT` drops the request again, because the redrive flag lives only on the in-heap `JobMaster`. That window is narrow and no worse than the pre-fix behavior, and is left as a documented limitation rather than extended here; persisting the redrive intent itself would be the follow-up. The `CoordinatorService` Javadoc also notes the shared-executor thread cost (Issue #3). No executable code changed; the diff is 16 inserted comment lines across the two files, spotless-applied. -- 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]
