kube-state-metrics

kube-state-metrics generates and exposes cluster-level metrics by listening to the Kubernetes API server. The kube-state-metrics Helm chart automates deployment and configuration of kube-state-metrics in Kubernetes clusters using Minimus images.

Prerequisites

A subscription to the Minimus images included in the chart
A running Kubernetes cluster (v1.21 or higher is recommended)
kubectl installed and configured
Helm v3.8 or higher for OCI support (Installation instructions)
Cluster Admin privileges

Try It Out

In this example we will deploy kube-state-metrics on a Kubernetes cluster using the Minimus kube-state-metrics Helm chart. kube-state-metrics will expose Prometheus-style metrics about the state of your Kubernetes objects.

Step 1: Start Cluster

If you have a cluster to work with, skip this step. Otherwise, start a Minikube cluster. Deployment instructions

Step 2: Create Secret

First, create a namespace:

kubectl create ns kube-state-metrics

Next, create a Kubernetes secret for the Minimus token so Helm will be able to pull images from the Minimus registry:

kubectl create secret docker-registry minimus-registry \
    --docker-server=reg.mini.dev \
    --docker-username=minimus \
    --docker-password={token} \
    --namespace=kube-state-metrics

Step 3: Deploy the Helm Chart

Deploy the Minimus Helm chart:

helm install my-kube-state-metrics oci://helm.mini.dev/kube-state-metrics \
  -n kube-state-metrics \
  --wait

If you want to override the chart's defaults, go to the Values tab, save a copy locally with your changes. Deploy the chart with your overrides using -f values.yaml:

helm install my-kube-state-metrics oci://helm.mini.dev/kube-state-metrics \
  -f values.yaml \
  -n kube-state-metrics \
  --wait

Step 4: Verify kube-state-metrics is Running

Confirm the chart deployed successfully and kube-state-metrics started:

kubectl get pods -n kube-state-metrics

You should see output similar to:

NAME                                          READY   STATUS    RESTARTS   AGE
my-kube-state-metrics-xxxxxxxxx-xxxxx         1/1     Running   0          2m

Step 5: Test HTTP Endpoint

Forward the kube-state-metrics service port to your local machine:

kubectl port-forward svc/my-kube-state-metrics -n kube-state-metrics 8080:8080

In another terminal, verify the health endpoint responds:

curl -sf http://localhost:8080/healthz

Then check that Prometheus metrics are being exposed:

curl -s http://localhost:8080/metrics | head -20

You should see Prometheus-formatted metrics including lines like:

# HELP kube_pod_info Information about pod.
# TYPE kube_pod_info gauge
kube_pod_info{...} 1

Step 6: Clean Up

Uninstall the chart and delete the namespace to remove all resources:

helm uninstall my-kube-state-metrics -n kube-state-metrics
kubectl delete namespace kube-state-metrics