nzw921rx opened a new pull request, #10971: URL: https://github.com/apache/seatunnel/pull/10971
### Purpose of this pull request Fix Kudu E2E flaky test (`KuduIT.java`) — multiple bugs causing intermittent failures: - Fix `readData()` row boundary bug: `row` list was created at batch level, causing all rows within a scanner batch to be concatenated into one list; moved `row` creation inside the per-row loop - Fix `readData()` BINARY column `break` bug: `break` on BINARY column skipped all subsequent columns; changed to `continue` - Fix `readData()` timezone-sensitive timestamp formatting: replaced `Timestamp.toString()` (uses JVM local timezone) with explicit UTC `SimpleDateFormat` for `UNIXTIME_MICROS` columns, preventing 8-hour offset failures in non-UTC environments (e.g. Asia/Shanghai) - Fix ordering assertion: Kudu hash-partition scan order is non-deterministic; added sort-by-id in `readData()` return value so `assertIterableEquals` comparisons are stable - Fix unreliable table cleanup: wrapped all test methods with `try-finally` to guarantee `dropTableIfExists` runs even on mid-test failures; added `dropTableIfExists` helper that swallows "table not found" exceptions - Fix `batchInsertData()` session resource leak: added `kuduSession.flush()` and `kuduSession.close()` in both overloads - Fix `tearDown()` missing `toxiProxy.close()`: Docker container resource leak on test completion - Fix `getKuduClient()` client leak on retry: close previous `kuduClient` before creating a new one during Awaitility retry loop ### Does this PR introduce _any_ user-facing change? No. ### How was this patch tested? ```bash ./mvnw -pl seatunnel-e2e/seatunnel-connector-v2-e2e/connector-kudu-e2e -DskipUT -DskipIT=false verify ``` -- 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]
