Introduction & Your Cluster

👋 Welcome to Kubernetes 101! You already know how to build and run containers with Docker. Kubernetes is what you reach for when one container on one machine isn't enough - when you need to keep your app running, scale it, update it without downtime, and expose it to the world.

By the end of this lab you will have deployed a real application to a real Kubernetes cluster and learned the core building blocks:

📦 Pods - how Kubernetes runs your containers
🚀 Deployments - keeping your app healthy and self-healing
🔌 Services - stable networking and load balancing
📈 Scaling & rolling updates - growing and upgrading with zero downtime
🌐 Ingress - exposing your app at a real URL

Run the interactive Labspace

This lab is also available as a self-contained, click-to-run Labspace at github.com/ajeetraina/labspace-kubernetes-101. It ships a live 3-node Kubernetes cluster, a terminal, an in-cluster registry, and the sample app in one repo:

git clone https://github.com/ajeetraina/labspace-kubernetes-101
cd labspace-kubernetes-101
docker compose up

Then open the Labspace in your browser at http://localhost:3030. The pages that follow mirror that Labspace content, adapted for this workshop.

Note

The same Kubernetes you'll use here is the one built into Docker Desktop. Everything you learn transfers directly to the cluster on your own machine - and to production clusters in the cloud.

🎁 The cluster you've been given

There's no cluster setup to do. The labspace already includes a fully running k3s cluster (a lightweight, certified Kubernetes distribution) right beside your terminal, with kubectl pre-configured to talk to it.

kubectl is the command-line tool you use to talk to any Kubernetes cluster. Let's confirm everything is alive.

Ask the cluster for its nodes - the machines that run your workloads:
```
kubectl get nodes
```
You should see three nodes with a STATUS of Ready - one control-plane node (which also runs workloads) and two worker nodes (<none> role). Together they form your cluster for this lab, so you'll get to see how Kubernetes spreads work across multiple machines.
Look at the pods already running across all namespaces. These keep Kubernetes itself working:
```
kubectl get pods -A
```
You'll see system components like CoreDNS (cluster DNS) and Traefik (the ingress controller you'll use in the last section).
Get a one-line summary of the cluster:
```
kubectl cluster-info
```

Tip

Typing kubectl all day gets old. A short k alias is pre-configured, so k get nodes works too. This lab spells out kubectl for clarity, but use whichever you prefer.

🗺️ How you'll talk to Kubernetes

Kubernetes is declarative. Instead of running imperative commands like docker run, you describe the desired state of your system in YAML files, hand them to the cluster, and Kubernetes works continuously to make reality match your description.

graph LR
    A[You write YAML] --> B[kubectl apply]
    B --> C[Kubernetes control plane]
    C --> D[Reconciles actual state<br/>to match desired state]
    D --> C

That reconciliation loop is the heart of Kubernetes - and the reason your app heals itself when things go wrong. You'll see it in action very soon.

When kubectl get nodes shows a Ready node, you're ready. Head to the next section to run your first Pod. 🚀