The GitHub Actions job "CI" on rocketmq-dashboard.git/fix/proxy-discovery-grading has failed. Run started by GitHub user Frun1na (triggered by Frun1na).
Head commit for run: 44f988608d6ddeb5672e00c155e75dc40e9f6cea / Apulupie <[email protected]> fix(cluster): grade a failed heartbeat-syncer proxy discovery fix(cluster): grade a failed heartbeat-syncer proxy discovery `discoverProxiesViaHeartbeatSyncer` turned every failure of the heartbeat-syncer read into an empty Proxy list, which is also what a cluster without a Proxy returns: an unreachable broker or NameServer was reported as "this cluster runs no Proxy". A caller could not tell the two apart - the instance path answers 404 "Proxy not found" for it and the rmq.proxy.list tool reports an empty inventory. The read is now graded like the other discovery calls in the repository: CONSUMER_NOT_ONLINE / TOPIC_NOT_EXIST (a cluster that currently runs no Proxy) keeps returning an empty list, anything else is a 502 naming the cause. ProxyConsumerResolver already grades the same call this way, and discoverClustersAt / refreshClusterDetail in the same class throw 502 on failure. RocketMQClusterProviderTest: 20 tests green, 0 checkstyle violations. The failure case fails on the unfixed code, and the absent-group case pins the empty business result that must keep working. Report URL: https://github.com/apache/rocketmq-dashboard/actions/runs/38038542508 With regards, GitHub Actions via GitBox
