SEZ9 commented on PR #13154: URL: https://github.com/apache/seatunnel/pull/13154#issuecomment-6790834822
CI result on head `8701e6434e`, as promised in my previous comment. ### Apache-side `Build`: **success** ### Fork run [42966332196](https://github.com/SEZ9/seatunnel/actions/runs/42966332196) Attempt 1: 83 passed, 12 skipped, 1 failed. Attempt 2 (rerun of the one failed job, no code change): **success**. The verification you asked about is in `unit-test (11, ubuntu-latest)`: ``` [INFO] Tests run: 11, Failures: 0, Errors: 0, Skipped: 0, Time elapsed: 1.241 s - in org.apache.seatunnel.engine.server.dag.physical.SubPlanTest [INFO] Tests run: 7, Failures: 0, Errors: 0, Skipped: 0, Time elapsed: 0.013 s - in org.apache.seatunnel.engine.server.diagnostic.JobRuntimeDiagnosticsTest ``` All four `unit-test` jobs passed (Java 11/17 x ubuntu/windows). Incidentally this also measures Issue C: you clocked `SubPlanTest` at about 4.1 s for 11 tests, and it is now 1.241 s for the same 11, which is the 3 s `job.retry.interval.seconds` sleep going away. ### Two failures on the way here, neither in this diff **1. `connector-file-sftp-it` on both JDKs, on the previous head.** `SftpFileIT` could not pull its image: ``` Status 404: failed to resolve reference "docker.io/atmoz/sftp:alpine-3.7": not found ``` This is #13202, which was already fixed on `dev` before I hit it — `atmoz/sftp` now publishes only `latest`, `alpine` and `debian`, and `dev` has moved to `atmoz/sftp:alpine`. My branch simply predated the fix. So I merged current `dev` into the branch rather than editing `SftpFileIT` myself, which would have duplicated a merged fix. `git diff dev..HEAD` is still exactly the same six files, 333 insertions, 0 deletions, so the sync changed nothing about this PR's own diff. Both sftp jobs pass on `8701e6434e`. **2. `all-connectors-it-5 (11, ubuntu-latest)`, on this head.** `TiDBCDCIT.startUp` -> `TiDBTestBase.initializeTidbTable:213`: ``` java.io.EOFException: Can not read response from server. Expected to read 4 bytes, read 0 bytes before connection was unexpectedly lost ``` The log prints `Container pingcap/tidb:v6.1.0 started in PT21.870928S` immediately before the first JDBC connect, so the container is reported up while its MySQL port is not serving yet. Evidence that it is nondeterministic rather than a real break: - on this same commit, `all-connectors-it-5 (17, ubuntu-latest)` **passed** while `(11)` failed — same shard, same tests, same code - the polarity is inverted from the occurrence you noted on `3fdac536`, where 11 passed and 17 failed - the other 13 `all-connectors-it` shards passed on both JDKs - rerunning only that job, with no code change, passed ### Also in this push `a2f59c7` removed a duplicated `@SuppressWarnings("unchecked")` on `setErrorByPhysicalVertex` that my previous push had left behind. `@SuppressWarnings` is not `@Repeatable`, so it was a compile error that would have failed all four `unit-test` jobs. I caught it by re-reading my own diff, not from CI, but it should not have been pushed; I have no JDK in this environment, so a mechanical re-read of the diff is the only check available to me and I had skipped it. Issues A, B, C and D are all addressed and CI is green. @DanielLeens, whenever you have a moment for the re-review. -- 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]
