SEZ9 commented on PR #12862: URL: https://github.com/apache/seatunnel/pull/12862#issuecomment-6568168288
Following up on the approval-versus-rebasing question. The head is unchanged at `1ce4e26cdbdcac4cf2d901c807463fe82cdf2822`. The latest comparison against `dev` is `diverged`, ahead 3 / behind 21, with merge state `UNKNOWN` / `REVIEW_REQUIRED`. As noted earlier, that drift is recorded metadata only; it does not by itself request a sync, and no conflicts have been reported. If conflicts do surface, resolve them before the final approval/queue step. On CI: the blocking job remains unit-test (17, ubuntu-latest), run 40828465494 / job 123297627304, where `PhysicalVertexTest` setup reports a node failing to start with port-binding errors; the Build at 123363531370 is `FAILURE`, and the connector shard was cancelled. The merged `d2fc2079035` changes HTTP-port allocation in other tests, not the `PhysicalVertexTest` fixture, so a simple sync is not a verified cure. I am not attributing a cause yet. Concrete asks: 1. After confirming no newer Build is active, rerun the failed unit-test job and the cancelled connector shard. 2. If `PhysicalVertexTest` still fails, paste the new port/startup diagnostics here so we can attribute it precisely. 3. Keep this PR limited to the Spark decimal change; no unrelated engine or test-fixture edits. The decimal source changes remain clear from my side. This is a CI/merge-gate clarification, not an effective approval; a write-capable maintainer will do the final approval once the gates are green or the remaining failure is confirmed unrelated. <!-- streview-comment:2216 --> -- 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]
