Explain the architecture of a Kubernetes cluster. What are the main components of the control plane and worker nodes?

Question

Accepted Answer

A Kubernetes cluster consists of two main parts: the control plane and the worker nodes. Each serves a distinct purpose in managing and running containerized workloads.

The control plane is the brain of the cluster. It makes global decisions about scheduling, responds to cluster events, and maintains the desired state. The key components include:

The API Server, or kube-apiserver, is the front door to Kubernetes. Every command you run with kubectl goes through this component. It validates and processes REST requests, then updates the corresponding objects in etcd.

Etcd is a distributed key-value store that holds all cluster data. It stores configuration, state, and metadata. If etcd fails, the entire cluster loses its memory. This is why production clusters typically run etcd in a highly available configuration with multiple replicas.

The Scheduler, or kube-scheduler, watches for newly created pods that have no node assigned. It evaluates resource requirements, affinity rules, taints, and tolerations to decide which node should run each pod.

The Controller Manager runs controller processes in the background. These include the Node Controller, which notices when nodes go down; the Replication Controller, which maintains the correct number of pods; and the Endpoints Controller, which populates the endpoints object.

Worker nodes are the machines that run your application containers. Each worker node contains:

Kubelet is an agent that runs on every node. It ensures containers are running in a pod according to the pod specification. It communicates with the API server and manages the containers on its node.

Kube-proxy maintains network rules on each node. It enables communication to pods from inside or outside the cluster. It uses iptables or IPVS to handle packet forwarding.

The Container Runtime is the software responsible for running containers. Kubernetes supports any runtime that implements the Container Runtime Interface, such as containerd or CRI-O.

In production, you typically have multiple control plane nodes for high availability and many worker nodes to distribute workloads.