2 · Kubernetes Internals

2. Kubernetes Pod Lifecycle & Probes

Liveness, readiness, and startup probes, graceful shutdowns, and horizontal pod autoscaling.

10 min read · 2 MCQs

Differentiating Probes

Startup Probe: Protects legacy or slow-initializing apps from being killed before boot. Liveness Probe: Detects deadlocks and restarts unresponsive containers. Readiness Probe: Evaluates if the pod can accept client traffic; failing readiness removes the pod from endpoints without restarting it.

Graceful Shutdown (SIGTERM vs SIGKILL)

When terminating, Kubernetes removes the pod from endpoints, sends `SIGTERM`, waits for `terminationGracePeriodSeconds` (default 30s) for connections to drain, and finally issues `SIGKILL`.

Chapter quiz

2 questions · pass mark 75%

1. What occurs when a pod's Readiness Probe fails?
2. Why should applications handle SIGTERM signals?