zmuxuny opened a new pull request, #6184: URL: https://github.com/apache/rocketmq-dashboard/pull/6184
## Summary Fixes #6179. Aliyun and Tencent consumer progress APIs expose topic/group aggregate lag, not individual queue offsets or per-queue lag. Stop publishing the largest aggregate as an AVAILABLE `consumer.lag.max_queue` value. Emit an explicit UNSUPPORTED sample with null value and `CLOUD_QUEUE_LAG_UNSUPPORTED` reason instead. Keep its consumer-group labels, metric declaration and catalog unchanged, so existing rules do not receive a misleading zero or a missing sample that could clear their state. Actual consumer totals and per-topic backlog remain available; group RPC failures continue to return UNAVAILABLE. ## Validation - Red against unchanged production code: four new regression instances fail among six tests. Aliyun/Tencent topic aggregate 80 incorrectly breaches a queue threshold of 50; Aliyun total fallback reports a fake queue maximum; aggregate zero falsely clears a FIRING queue alert. - Green: `mvn -B -ntp -Dtest=CloudRocketMqBusinessMetricsCollectorTest,NativeAlertProcessorTest,AlertRuleEvaluatorTest,AlertStateMachineTest,NativeAlertMetricCatalogServiceTest checkstyle:check test` — 52 tests passed, zero Checkstyle violations. - Updated existing fixtures to use actual provider row shapes, plus coverage for empty progress and RPC failures. These are synthetic provider-shape regression tests; no live cloud account or full MySQL-backed integration suite was used. GitHub CI is pending. AI assistance was used for implementation and regression validation; the code and tests received independent 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]
