Example Documentation: Kubernetes Health Probes
Goal
Trace startup, readiness and liveness probes on a small Deployment.
What Is Done
The lab creates a two-replica Deployment and a ClusterIP Service. All probes check the same HTTP endpoint, but each probe has a different control-plane meaning.
Typical Commands
kubectl apply -f manifests/
kubectl rollout status deployment/probe-web -n probes-lab
kubectl get endpoints -n probes-lab probe-web
kubectl describe pod -n probes-lab -l app.kubernetes.io/name=probe-web
Validate Success
- Pods become
READY 1/1. - The Service has endpoints.
- The
describeoutput shows all three probes.
Common Mistakes
- Probe port does not match the container port.
initialDelaySecondshides startup issues instead of usingstartupProbe.- Readiness and liveness are confused, causing unnecessary restarts.