☸

Kubernetes Probe Simulator

Lifecycle Presets Quick Setup
Startup Probe
Enabled
failureThreshold (attempts) 3
periodSeconds (interval) 5s
initialDelaySeconds 0s
Liveness Probe Restarts Pod
failureThreshold 3
periodSeconds 10s
initialDelaySeconds 10s
Readiness Probe Routes Traffic
failureThreshold 3
periodSeconds 10s
initialDelaySeconds 5s
Fault Injection Simulate Breakage
Pod

signup-app-container

Phase: Readiness | Status: Running & Ready

Pod Restarts 0
Probes Dispatched 14
Service Endpoints 1/1 IP
Pod Container State Machine Healthy In-Service
1. Container Boot
Image pulled & process started inside container namespace.
Completed
2. Startup Probe
Locks other probes until container initialization passes.
Passed (1 attempt)
3. Liveness Probe
Ensures process isn't deadlocked. Failure triggers container restart.
Passing (0 fails)
4. Readiness Probe
Checks if ready for traffic. Failure cuts Pod IP from Service.
Ready (Accepting Traffic)
Cluster Ingress → Service → Pod /health: TRAFFIC ROUTED
Pod IP is active in Endpoints
Kubelet Probe Telemetry
Tuned Deployment YAML

Kubernetes Probes: The Critical Differences

Startup Probe

Runs before liveness and readiness. Perfect for slow-starting legacy apps or cache warm-ups. If it reaches failureThreshold, Kubernetes restarts the pod. Once it succeeds once, it retires and hands off to liveness/readiness.

Liveness Probe

Answers: Is the container alive or deadlocked? If GET /health fails consecutively past failureThreshold, Kubernetes kills the container and invokes the restartPolicy.

Readiness Probe

Answers: Is the container ready for traffic? If /health returns non-200, Kubernetes removes the pod IP from the Service endpoints without restarting it. Traffic resumes as soon as it recovers.

Grounded in DevOps Kubernetes Probes guide by Shreya Sinha (Medium, Day 19 of DevOps). Explains the mechanics of initialDelaySeconds, periodSeconds, and failureThreshold across Startup, Liveness, and Readiness probes.