Cloud & hybrid monitoring · 6 min read

Monitor a Kubernetes cluster's node and workload health

Kubernetes doesn't speak SNMP; Linivo's cluster integration reads from the Kubernetes API and metrics-server instead, the same sources kubectl top uses.

What you'll need

  • A Kubernetes cluster with metrics-server installed
  • A service account with a ClusterRole scoped to read-only access
1

Install metrics-server if it isn't already running

Node and pod resource metrics depend on metrics-server being deployed in the cluster; most managed Kubernetes offerings include it by default, but self-managed clusters may need it installed.

kubectl apply -f https://github.com/kubernetes-sigs/metrics-server/releases/latest/download/components.yaml
2

Create a read-only service account and ClusterRole

Scope the service account's permissions to get/list/watch on nodes, pods and their metrics, rather than granting a broad cluster-admin binding for what's purely a monitoring integration.

3

Add the cluster in Linivo using the generated kubeconfig

Linivo polls node and pod resource usage, restart counts and pod status, and can alert on patterns like a pod stuck in CrashLoopBackOff the same way it would alert on a flapping physical interface.