Chapter 17: Kubernetes Internals
⏱️ Total chapter time: ~75 min (45 min reading + 30 min exploration)
After this chapter, you will be able to: Explain exactly what happens inside Kubernetes when you run
kubectl apply, understand how etcd, the CRI, CNI, and CSI fit together, and read cluster events with a mental model of the full control-plane lifecycle.
What’s Inside
| Section | Topic | Time |
|---|---|---|
| 17.1 | How a Pod Gets Created — The Full Journey | ~15 min |
| 17.2 | etcd — The Cluster Brain | ~8 min |
| 17.3 | Container Runtimes and the CRI | ~8 min |
| 17.4 | CNI — Container Network Interface | ~8 min |
| 17.5 | CSI — Container Storage Interface | ~6 min |
Prerequisites
- Completed Chapters 1–16
- Minikube cluster running (
minikube status) - Familiarity with the control-plane components from Chapter 1.3
Why Internals Matter
You’ve been telling Kubernetes what to do for 16 chapters. Now we look at how it actually does it.
This isn’t academic. Understanding internals helps you:
- Debug faster — you know which component to blame when something breaks
- Trust the system — you’ll stop worrying “but what if K8s drops my request?”
- Design better — you know the costs and constraints of certain patterns
📝 Note: This chapter is observation-heavy. We won’t deploy apps — we’ll watch the machinery run.