smengcl opened a new pull request, #11573: URL: https://github.com/apache/ozone/pull/11573
Generated-by: Claude Code (Opus 5.5) ## What changes were proposed in this pull request? `TestKeyDeletingService$Normal#checkIfDeleteServiceIsDeletingKeys` intermittently found a key still pending deletion right after the service reported all 100 keys as deleted. `KeyDeletingService` increments `deletedKeyCount` once the purge request is applied, but the deletedTable entries are removed from RocksDB only when the OM double buffer flushes. The test waits on that counter and then reads the deletedTable from the DB via `getPendingDeletionKeys`. In the failing run the last service run purged a single key, and the test read the table before that purge was flushed. This patch calls `om.awaitDoubleBufferFlush()` before the deletedTable check, in this test and in `checkDeletedTableCleanUpForSnapshot`, which has the same check after the same kind of wait. ## What is the link to the Apache JIRA https://issues.apache.org/jira/browse/HDDS-17077 ## How was this patch tested? - Checkstyle clean. - Deterministic by construction: once the counter covers every purge request, waiting for the double buffer flush guarantees those purges are in the DB before it is read. A local run was not possible (OM port 9862 was taken by another process on the machine). - 10x10 [flaky-test-check](https://github.com/smengcl/hadoop-ozone/actions/runs/38980710883) on this branch: in progress 🤖 Generated with [Claude Code](https://claude.com/claude-code) -- 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] --------------------------------------------------------------------- To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
