andyhuangdev opened a new pull request, #11594: URL: https://github.com/apache/ozone/pull/11594
## What changes were proposed in this pull request? Recon does not refresh the suppression flag when an existing container is suppressed or unsuppressed in SCM. This causes stale alerts after suppression and missing alerts after unsuppression. Suppressed containers can also retain their previous health state because the health check skips them before updating that state. This patch: - Synchronizes suppression flags in both directions using the existing SCM suppressed-container listing API. - Completes the paginated listing before updating flags, preserving existing flags if listing fails. - Resets suppressed containers’ health state before processing so stale unhealthy records are removed. ## What is the link to the Apache JIRA [HDDS-17070](https://issues.apache.org/jira/browse/HDDS-17070) ## How was this patch tested? Five new regression tests were confirmed to fail before the fix. After the fix, all 55 tests passed across TestReconSCMContainerSyncIntegration, TestReconReplicationManager, and TestStorageContainerServiceProviderImpl. Coverage includes bidirectional synchronization, pagination, listing failures, older SCM filter behavior, RocksDB persistence, suppression and unsuppression of MISSING containers, and existing 100,000-container scenarios. The project checkstyle script and git diff --check passed. -- 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]
