Rangsh commented on PR #12081: URL: https://github.com/apache/seatunnel/pull/12081#issuecomment-5577541536
@DanielLeens thanks for catching this — and for digging into the fork CI logs rather than stopping at the check-run status. You're right: `HdfsWriterDurableFlushTest` was not modeling production. `IMapFileStorage` calls `fs.setWriteChecksum(false)` before handing the `FileSystem` to the WAL writer; the test used bare `FileSystem.getLocal(conf)`, so `ChecksumFileSystem` buffering made mid-stream read-back from a second handle see 0 records until `close()`. That matches the deterministic `expected: <4> but was: <0>` failure on JDK 8/11 `ubuntu-latest`. Pushed `$(git rev-parse --short HEAD)`: - Call `fs.setWriteChecksum(false)` before `writer.initialize(...)`, matching production (`IMapFileStorage.java:157-158`) - Added a short comment explaining why (cross-handle visibility under checksum buffering) Locally re-ran `HdfsWriterDurableFlushTest` (+ the other new regression tests) successfully after the change. Issue 2 (`deleteAll` double-publish) remains a separate follow-up as agreed. Happy to re-review once CI is green on this head. -- 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]
