Kubernetes Troubleshooting Toolkit

Production-grade diagnostic framework for Kubernetes clusters. Covers every common failure mode: CrashLoopBackOff, ImagePullBackOff, OOMKilled, stuck Pending pods, node NotReady, DNS failures, ingress misconfigurations, PVC binding issues, and deployment rollout problems. Includes a triage decision tree that classifies symptoms by layer (container, scheduling, networking, storage, infrastructure), exit code cheat sheets, ready-to-paste kubectl commands for each scenario, network debugging workflows with ephemeral test pods, resource pressure detection, health check probe debugging, and cluster cleanup/maintenance routines. 12KB of battle-tested patterns — not theory, but the exact commands and decision trees you need when production is down at 3 AM.