rkhachatryan commented on PR #28544: URL: https://github.com/apache/flink/pull/28544#issuecomment-4815525939
@1996fanrui I re-worked CDR in #28554 and don't see the data loss there: https://github.com/apache/flink/pull/28554#issuecomment-4815514890 **Validation** — downscale data-loss reproducer from #28544 (`UnalignedCheckpointRescaleITCase`, `keyed_broadcast 7→2`, ~0.1–0.3%/run): | Branch | Reproducer result | |---|---| | this branch | **0 losses / 1498 runs** | | `d9fc48` (pre-rewrite baseline) | reproduces — 439 records lost / ~590 runs | So the reproducer is valid and the loss does not occur on this branch. Note: the loss is already present at `d9fc48`, i.e. it **predates the async-recovery rewrite**, so attribution to a single commit isn't isolated. -- 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]
