Question

A cluster administrator runs `kubectl top nodes` to investigate a performance complaint, but the command returns the error: `error: Metrics API not available`. Deployments, pods, and services all function normally, and `kubectl get pods -A` shows all system pods as Running. What is the most likely cause and the correct first step to resolve it?

Accepted Answer

The metrics-server component is not deployed (or its Deployment is unhealthy); install or verify the metrics-server Deployment in the kube-system namespace.

Answer

The kube-apiserver is down; restart the kube-apiserver static pod on the control-plane node.

Answer

The kubelet on each node has crashed; restart the kubelet service on all worker nodes.

Answer

The Horizontal Pod Autoscaler is misconfigured; delete and recreate all HPA objects to restore metrics.