A monitoring cleanup normalizes labels and later dashboard panels for deployments and statefulsets go empty despite healthy scrapes.
A kube-state-metrics dashboard goes blank because relabeling removed owner_kind and the panels still join workloads on that field
Metrics arrive normally, yet higher-level workload views disappear because the join key used by the dashboard was stripped out upstream.
Scenario
What to check first
- Identify the primary failure signal in the The Metric Survived Scrape and Lost the Label the Dashboard Needed to Join scenario.
- Separate visible symptoms from the underlying technical dependency.
- Describe the safest recovery path and the follow-up prevention work.
Checking checklist
- Summarize the current impact and the last known change.
- Collect direct evidence from logs, runtime state, and configuration before changing anything.
- Separate immediate recovery from permanent prevention work.
Recovery and prevention
Inspect the final label set before editing the panel query or scrape configuration.
Questions worth viewing together
Community-field Kubernetes problem inspired by Grafana and Prometheus threads where owner_kind relabel cleanup blanked dashboards. Dashboard emptiness can be a post-scrape label problem even when collection is healthy.
Teams often blame kube-state-metrics instability when the relabel step removed the join label.
Relabel changes should be regression-tested against dashboard joins and templates, not only target discovery.
Similar cases seen in the field