1fanwang commented on PR #3683: URL: https://github.com/apache/celeborn/pull/3683#issuecomment-4432526376
The 10 CI failures here look pre-existing rather than caused by this PR. - 9 of the 10 are `celeborn spark integration test - hash-checkDiskFull` (with the root cause `IllegalStateException: After trying all the available Master Addresses ... an usable link still couldn't be created`, then `RESERVE_SLOTS_FAILED`). `spark3 (11, 3.3)` failed the same way on `main` at run [25709302327](https://github.com/apache/celeborn/actions/runs/25709302327) (2026-05-12 02:19 UTC) and run [25645931093](https://github.com/apache/celeborn/actions/runs/25645931093) (2026-05-11 01:41 UTC). - The 10th is `WordCountTestWithAuthentication` in `flink1 (11, 1.16)`, asserting `new File(fileName).length() == fileWriter.getDiskFileInfo.getFileLength` (`WordCountTest.scala:118`). That path is local-disk shuffle via `FileWriter` — this PR only touches `HdfsFlushTask`, so the assertion is independent of the change. Could a committer kick off a rerun when convenient? -- 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]
