CKAD — Liveness Probes

Liveness Probe — Keep Apps Running

A liveness probe detects when an application is stuck. When the probe fails, Kubernetes restarts the container. This is your safety net against deadlocks, infinite loops, and unresponsive processes.

HTTP Liveness Probe

The most common pattern — check a health endpoint:

apiVersion: v1
kind: Pod
metadata:
  name: web-liveness
spec:
  containers:
  - name: web
    image: nginx:1.27
    ports:
    - containerPort: 80
    livenessProbe:
      httpGet:
        path: /
        port: 80
      initialDelaySeconds: 5
      periodSeconds: 10
      failureThreshold: 3

Kubernetes sends a GET request to http://localhost:80/ every 10 seconds. If it gets 3 consecutive non-2xx/3xx responses, the container is restarted.

Exec Liveness Probe

Use a command to check health — useful for non-HTTP applications:

apiVersion: v1
kind: Pod
metadata:
  name: liveness-exec
spec:
  containers:
  - name: app
    image: busybox:1.36
    command: ["/bin/sh", "-c"]
    args:
    - |
      touch /tmp/healthy
      sleep 30
      rm /tmp/healthy
      sleep 600
    livenessProbe:
      exec:
        command: ["cat", "/tmp/healthy"]
      initialDelaySeconds: 5
      periodSeconds: 5

kubectl apply -f liveness-exec.yaml

# Watch the Pod — it will restart after ~35 seconds
kubectl get pod liveness-exec -w
# After 30s, /tmp/healthy is deleted → probe fails → container restarts

What happens step by step:

1. Container starts and creates /tmp/healthy
2. Liveness probe runs cat /tmp/healthy — succeeds (exit code 0)
3. After 30 seconds, the file is deleted
4. Probe runs cat /tmp/healthy — fails (exit code 1)
5. After failureThreshold consecutive failures, Kubernetes restarts the container

---

Next: Readiness Probes