K8s — Local Kubernetes with Kind (Kubernetes-in-Docker)

Hands-on notes for spinning up local Kubernetes clusters with Kind (Kubernetes-in-Docker), installing kubectl, defining a multi-node cluster, and enabling the Kubernetes Dashboard.

This is a personal learning-notes repository — a concise, copy-paste reference for getting a local Kubernetes environment running quickly. It is not a framework or a tool; it is a set of tested commands and example manifests collected while learning and testing Kubernetes.

What is this?

A short, practical cheat sheet that covers the fastest path to a local Kubernetes cluster:

• Kind — run Kubernetes clusters inside Docker containers (nodes are containers, not VMs).
• kubectl — install and use the Kubernetes command-line client.
• Multi-node cluster — an example Kind config with one control-plane and two workers.
• Kubernetes Dashboard — enable it and expose it as a service (see Kubernetes-Dashboard.md).

Who is this for?

Developers and students who want a fast, disposable local Kubernetes cluster for testing applications, controllers, and Kubernetes itself — without the overhead of VM-based tools like minikube or microk8s.

Why Kind?

• Kind is Kubernetes-in-Docker.
• It uses Docker containers to simulate nodes.
• Boots a cluster in ~30 seconds.
• Ideal for testing Kubernetes, applications, controllers, and for training.

Prerequisites

• Docker installed and running.
• A Linux/macOS shell (commands below target Linux x86_64).

Quickstart

1. Install Kind

[ $(uname -m) = x86_64 ] && curl -Lo ./kind https://kind.sigs.k8s.io/dl/v0.19.0/kind-linux-amd64
chmod +x ./kind
sudo mv ./kind /usr/local/bin/kind

v0.19.0 is the version used in these notes. Check the Kind releases page for the latest version.

2. Create a cluster

List existing clusters:

kind get clusters

Create a cluster from a config file:

kind create cluster --config=test_config.yaml --name my_cluster

Example test_config.yaml — a three-node cluster (one control-plane, two workers):

# three node (two workers) cluster config
kind: Cluster
apiVersion: kind.x-k8s.io/v1alpha4
nodes:
- role: control-plane
- role: worker
- role: worker

Delete the cluster when done:

kind delete cluster --name my_cluster

3. Install and use kubectl

kubectl is the official Kubernetes command-line tool for managing and controlling Kubernetes resources.

sudo snap install kubectl --classic
kubectl version --client
kubectl config view

Kubernetes Dashboard

Steps to enable the Kubernetes Dashboard and expose it (via NodePort or LoadBalancer) are in Kubernetes-Dashboard.md.

FAQ

How is Kind different from minikube or microk8s? Kind runs each Kubernetes node as a Docker container instead of a virtual machine, so clusters boot faster (~30s) and are easy to create and throw away.

Can I run a multi-node cluster locally with Kind? Yes. Use a Kind config file that lists multiple nodes (see the example above) to get a control-plane plus worker nodes.

Is this a production tool? No. These are learning notes and Kind itself is intended for local development, testing, and CI — not production workloads.

References

• Kind quick start: https://kind.sigs.k8s.io/docs/user/quick-start/
• Kubernetes: https://kubernetes.io/
• kubectl: https://kubernetes.io/docs/reference/kubectl/

Companion video

Walkthrough on YouTube: watch here

License

Licensed under the Apache License 2.0.