Deep dives: how it breaks
The certification guides tell you what to know for each exam. These chapters go deeper: each one explains how a part of Kubernetes works, then breaks it on purpose and walks through the output, the root cause, and the fix. Every study guide links to the chapter that covers its topic.
You need a laptop with Docker or Podman, kubectl, and kind. Chapters 7 and 8 need an extra component (a Gateway API implementation, and a CNI that enforces NetworkPolicy), and say how.
All chapters are drafts: they haven’t yet been run end to end on a live cluster. Each one says so at the top until it has.
| Chapter | You’ll be able to | |
|---|---|---|
| 1 | How a pod gets from YAML to running | Tell which component stopped making progress when a pod is stuck |
| 2 | Probes and resource limits | Explain every OOM kill, throttle, and probe restart |
| 3 | Services, DNS, and kube-proxy | Find which of four hops broke when “the service is down” |
| 4 | ConfigMaps and Secrets | Know when config changes arrive, and what a Secret protects |
| 5 | Storage and CSI | Diagnose provision, attach, and mount failures |
| 6 | RBAC and service accounts | Read a Forbidden error and grant exactly what’s missing |
| 7 | Ingress and Gateway API | Walk a 404, 502, or 503 back from the edge to the pod |
| 8 | Network policies | Segment a namespace without breaking DNS |
| 9 | Deployments and rollouts | Ship, stall, and roll back safely |
| 10 | Node pressure and eviction | Tell an eviction from an OOM kill, and stop both |
| 11 | Control plane failures | Keep a cluster changeable when etcd, certificates, or webhooks fail |
| 12 | Upgrades | Upgrade one minor version at a time without an outage |
When something is broken right now, start from the error index instead.