The GitHub Actions job "CI" on rocketmq-dashboard.git/codex/instance-load-failure-states has succeeded. Run started by GitHub user iloveeyjafjalla (triggered by iloveeyjafjalla).
Head commit for run: 8577b192174e05472fc3915bdf166cb36c84469e / iloveeyjafjalla <[email protected]> fix(web): report a failed instance-list load on the two remaining console pages The alert-rule dialog (ops/alerts.tsx) and the broker-cluster page (studio/BrokerCluster.tsx) both answered a failed GET /api/instances with an empty list plus a transient toast: once the toast faded the instance picker looked like a deployment without instances, and nothing offered a retry. The broker-cluster page also discarded the topology it had already loaded, although the failure says nothing about the cluster. Both pages now keep the previous options, mark the Select as errored and render the failure with a retry that re-runs the load, matching the shape already used on the four pages #6078 covers (dashboard, data sources, producer, lite topic) and the session-details fix folded into #6144. Tests: AlertsPage.test.tsx gains a case that opens the rule dialog on a rejecting listInstances, asserts the failure text and a retry that reaches the wire twice and clears the message; BrokerCluster.test.tsx asserts the same plus that the retry recovers the broker topology. Report URL: https://github.com/apache/rocketmq-dashboard/actions/runs/38735967602 With regards, GitHub Actions via GitBox
