Home / BeaverDeck / Docs / Insights Guide / Node Insights / Metrics Pipeline

Metrics Pipeline

BeaverDeck uses this check to identify a specific node condition that may need operator review.

Permissions: viewing checks requires insights: view. Opening a linked object or logs requires the corresponding resource permission, and the BeaverDeck ServiceAccount must be allowed to read the Kubernetes resources used by the check. Suppressing a finding requires insights: edit and affects all users.
Check typemetrics-pipeline
Insights sectionNode Insights
Alert severityWarning; critical if both metrics paths fail

When It Reports A Finding

The metrics.k8s.io/v1beta1 API is unavailable. The finding is a warning when BeaverDeck can use direct kubelet /metrics/resource scraping, and critical when that fallback also fails.

Why This Is A Problem

Without a working metrics path, current CPU and memory usage cannot be shown reliably. Capacity and right-sizing decisions can then be based on incomplete data, and a failed fallback can hide active resource pressure.

Recommended Response

  1. Check the metrics-server Deployment, Pods, APIService status, certificates, and logs.
  2. Verify network access and Kubernetes RBAC for the metrics API. If relying on the fallback, verify access to nodes/proxy and the kubelet resource metrics endpoint.
  3. Repair the preferred metrics path, wait for fresh samples, and refresh Node Insights.

Scope And Limitations

The direct kubelet fallback is degraded rather than equivalent: CPU usage derived from counters needs a previous scrape before it stabilizes. This check reports pipeline availability, not the accuracy of every individual metric.

After remediation: refresh Node Insights and verify the underlying resource or metric. Suppress the finding only when the condition is intentional and its risk is accepted.