The Kubernetes API and kubectl

When you run kubectl get pods, a table of running containers appears. When you run kubectl apply -f deployment.yaml, workloads start across a cluster. Both commands do exactly the same thing under the hood: send an HTTP request to the Kubernetes API server. That single fact explains more about how Kubernetes works than any architecture diagram.

The API server is the single gateway into the cluster. Every component, from kubectl on your laptop to the kubelet on each node to the controllers running in the control plane, communicates exclusively through the REST API. There is no back channel. Understanding how this API is structured, how requests flow through it, and how kubectl interacts with it gives you a mental model that applies to every Kubernetes interaction. That understanding is also a core part of the KCNA Kubernetes Fundamentals domain, which carries 46% of the exam.