SEZ9 commented on PR #11746: URL: https://github.com/apache/seatunnel/pull/11746#issuecomment-6183276794
Thanks for the detailed follow-up on `e7586c06a` and for validating the actual class ordering instead of relying on a standalone SplitIT run. On the shared-image teardown point: the SQL Server evidence looks sound to me — `JdbcSqlServerIT` running first, removing the shared image, and `JdbcSqlServerSplitIT` re-pulling and passing 2/2 under `-Dfailsafe.runOrder=alphabetical` is exactly the ordered scenario we needed. Switching both SplitITs to `PullPolicy.alwaysPull()` is a reasonable fix for the cached image-presence problem, so I consider the original image-not-found failure resolved. For Oracle, I appreciate you stating plainly that this is not a green result. Since `JdbcOracleSplitIT` did re-pull after `JdbcOracleIT` teardown, the teardown fix itself is confirmed there too; the remaining gap is the `DATABASE IS READY TO USE` timeout during first-time initialization. Two asks on that: 1. Your comment appears to be cut off at "I have not changed the image version or weake…" — could you repost the rest? I want to make sure I understand what you did and did not change for the Oracle wait. 2. When you have an Oracle run where the test methods actually execute (both `JdbcOracleIT` and `JdbcOracleSplitIT`, in the ordered configuration), please post the results here so we can close this out with real Oracle coverage rather than only the re-pull evidence. Separately, the earlier review points on the composite-key splitting itself (the SQLite NULL-component handling, the Oracle `FETCH FIRST` / 11g fallback, the STRING-key collation vs `compareArrays` ordering, the comma-based separator encoding, the Integer/Long and BigDecimal scale comparison fragility, the missing upgrade/opt-out note in `docs/en/connectors/source/Jdbc.md`, and the removed single-split-key guard affecting `FixedChunkSplitter`) are still open from my side. If `e7586c06a` touched any of those, please point me to it; otherwise I'll wait for a follow-up commit before re-reviewing them. <!-- streview-comment:1760 --> -- 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]
