SEZ9 commented on issue #12058: URL: https://github.com/apache/seatunnel/issues/12058#issuecomment-5707369716
@Rangsh thanks for closing out all three items. 1. **Scope note** — good. The one-line scope note at the top of `NOISY-ALIGN-REPORT.md` (results limited to the `file:///` LocalFileSystem harness on `38a37104e`, not to be generalized to a real HDFS/DFS client) is exactly what was needed, and the updated gist at https://gist.github.com/Rangsh/3c3f49c79c6583a1ca954b7d47a49627 covers it. Nothing further here. 2. **Diagnostic property removed** — confirmed as done on your side. Your post-run check (both trees back at `CHECKPOINT_OPERATIONS_PER_INVOCATION = 100`, durable reload/validate TearDown restored, `seatunnel.benchmark.diag.skipOverviewValidate` absent from the fixture/workload sources) is sufficient; I don't need anything more on this. 3. **Production / issue status** — agreed and unchanged. No production change from this evidence pack, `#12081` stays Related-only, and this issue remains open until the historical ~32% / ~393 µs day is actually reproduced. Only remaining ask, and it's the one you already offered: if the high-CV day reappears, capture it with the same baseline, JDK, JMH defaults, `file:///` store and 0 ns `ThreadPark` JFR settings as attempt-04, and add it to the same gist (report + JSON summaries; raw `.jfr` staying local is fine) so the comparison is direct. Until then, nothing else is outstanding from this review scope. <!-- streview-comment:1102 --> -- 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]
