SEZ9 commented on PR #10808: URL: https://github.com/apache/seatunnel/pull/10808#issuecomment-5789126447
Thanks for the pointers to the hunks in `0d4d8624c` for F3/F7 (`incompatible-changes.md` CLI member list and failover-window bullets), F4 (`telemetry.md` 1-second fixed, non-configurable flush timeout), F5 (`rest-api-v1.md` no-authentication note) and F8 (the `Collections` import in `ServerExecuteCommandTest.java`). That matches what I asked for; I'll confirm each against the diff itself before marking them resolved. Still open on the `ServerExecuteCommand.java` side: 1. **F1 / F2:** how does the "first non-lite member" fallback relate to the server-side coordinator election? If the intent is that the CLI never claims a coordinator unless the server reports one, please confirm that and point me to the hunk; if the guess is deliberately kept, a short code comment explaining why it is acceptable would be enough for me. 2. **F6:** when no active coordinator can be resolved, does the member-list output now indicate that the coordinator is unknown instead of marking every non-lite member as `MASTER`? A test covering that case would be ideal. Separately, per the update above, the `Build` run for `0d4d8624c` is reported red and the branch is reported to have a merge conflict with `dev`. Even if the doc/CLI items check out, please rebase and re-run CI so we can confirm the `TaskDeploymentPrePublicationClassLoaderLeakIT` failure is unrelated to this change before this moves forward. <!-- streview-comment:1263 --> -- 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]
