Startup, readiness and liveness probes

Without probes Kubernetes knows only “the process runs”. Probes tell it whether the application works.

Probe	Question	On failure
startupProbe	has it finished booting?	the other probes wait; after `failureThreshold` -> restart
readinessProbe	can it take traffic now?	removed from Service endpoints - no restart
livenessProbe	is it hung?	container restarted

graph LR S[container starts] --> SP{startup OK?} SP -->|no, too long| R[restart] SP -->|yes| L[liveness + readiness run] L -->|readiness fails| NR[not Ready - no traffic] L -->|liveness fails| R

containers:
  - name: app
    image: nginx:1.27
    startupProbe:
      httpGet: { path: /, port: 80 }
      failureThreshold: 30        # 30 x 2 s = up to 60 s to start
      periodSeconds: 2
    readinessProbe:
      httpGet: { path: /ready, port: 80 }
      periodSeconds: 5
    livenessProbe:
      httpGet: { path: /healthz, port: 80 }
      periodSeconds: 10
      failureThreshold: 3

Handlers: httpGet (2xx/3xx = OK), tcpSocket, exec (exit 0 = OK), grpc. Tuning: initialDelaySeconds, periodSeconds, timeoutSeconds (1 s!), failureThreshold, successThreshold.

Probes and rollouts

A rolling update waits for readiness. A new version that never becomes ready stalls the rollout - the old Pods keep serving. No readiness probe means “ready as soon as the process starts” - traffic hits an app that is still warming up.

Pitfalls

Liveness must not check dependencies. Database down -> every Pod restarted -> outage made worse. Put dependency checks in readiness.
Liveness too strict or timeoutSeconds too low -> restart loops under load.
Slow starters: use a startupProbe, not a huge initialDelaySeconds.

References

Configure probes