The Silent Killer Probe
Pods keep restarting but never reach Ready. The liveness probe is killing them before startup completes.
The Scenario
A routine deployment to the payment gateway service went out 20 minutes ago. Since then, pods cycle through Running and Terminating but never reach Ready state. The HPA keeps scaling up replicas, but each new pod dies the same way. Error rates are climbing and the team is starting to panic. The liveness probe is configured to hit /health every 5 seconds with no initialDelaySeconds - meaning it fires before the JVM even finishes loading.
What You'll Learn
How liveness probes can kill pods during slow startup sequences
Reading kubectl describe pod events to trace probe failure timing
The difference between liveness, readiness, and startup probes
Setting initialDelaySeconds and failureThreshold to match real startup curves
Tools You'll Use
Real-World Context
Misconfigured liveness probes are one of the top five causes of mysterious K8s restarts in production. Any service with a slow JVM, database pool warmup, or ML model load is at risk.
Ready to debug this?
Free account required - sign up with GitHub or Google in 10 seconds
Play The Silent Killer Probe