DanielLeens commented on issue #11364: URL: https://github.com/apache/seatunnel/issues/11364#issuecomment-4997998491
Thanks for the update — the separation you added between worker-side reader collection and coordinator-local enumerator collection is the right correction, and it removes the biggest transport concern from my earlier review. At this point, the next thing I would keep explicit before implementation resumes is the stabilization gate: 1. keep #11354 as the feature umbrella, and keep this issue as the design source of truth 2. treat the MySQL + PostgreSQL path as validation scope first, not as a release promise yet 3. make the transition from experimental internal reports to stable public `seatunnel-api` DTOs explicit in the acceptance criteria, including what evidence is required before freezing `schemaVersion = 1` 4. when PR #11395 is reworked, keep it sliced in the order already listed here and avoid landing REST/public DTO commitments ahead of the worker/coordinator collection and checkpoint/restore wiring With those boundaries, this still looks like the right STIP path rather than a regular enhancement thread. Please continue refining the STIP here, and keep the first implementation strictly within the additive latest-snapshot scope until the contract is settled. -- 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]
