JeremyXin commented on PR #11841: URL: https://github.com/apache/seatunnel/pull/11841#issuecomment-6295362237
@SEZ9 Thanks for the careful recheck. I addressed the remaining points in the latest head, 9d446cb99: 1. shouldReadLegacyPayload() now asserts both isRestoreJob() and isSavepointRestore() for a legacy payload with the savepoint flag set. 2. restoreModeShouldMatchLegacySavepointFlag() verifies the legacy flag for each RestoreMode, including after serialization round trips. The flag remains true only for SAVEPOINT. 3. I added a comment at the JobMaster.init guard clarifying that restore jobs currently skip cluster-side save-mode handling. Changing this behavior requires a separate decision and regression test. 4. The PR description now clarifies that checkpoint restores encountering pending cleanup or existing job metrics continue to follow the existing JobException path; these checks are unchanged by this PR. The existing-metrics branch is not separately covered by a new test here. 5. JobMaster.initCheckPointManager() passes isRestoreJob() || restart to CheckpointManager, which forwards it to CheckpointCoordinator. 6. I rechecked the current head and could not find the null-guard followed by an unconditional dereference in submitJob. For legacy payloads, readData() initializes restoreMode from the legacy savepoint flag before downstream code uses it. -- 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]
