Jan 6, 2024
7 Min read

A Kubernetes tutorial for absolute beginners

In this tutorial, we will install a Kubernetes cluster on your laptop and deploy an Nginx Pod.

System requirements

You need a Linux, Mac(Intel or Apple Silicon), or Windows laptop with at least 6GB of memory to complete this tutorial.

Outline of steps

  1. Install Multipass
  2. Set up K0s
  3. Install and use kubectl
  4. Deploy Nginx Pods
  5. What’s next

Install Multipass

Multipass is a tool for running Ubuntu virtual machines on Linux, Mac (Intel or Apple Silicon), or Windows. We are going to install a Kubernetes cluster on a virtual machine created with Multipass on your laptop.

Install Multipass by following the installation instructions.

Set up K0s

K0s is a lightweight Kubernetes distribution optimized to be deployed on resource-constrained environments like laptops.

Launch a new virtual machine

Open the terminal on your laptop and launch a new virtual machine named k0s10.

multipass launch -d 5G -m 2G -c 2 -n k0s10 22.04

This virtual machine has 5GB storage, 2GB memory, and 2 CPU cores. We use Ubuntu Jammy as the OS.

Log in to the virtual machine.

multipass shell k0s10

This takes us to the bash terminal of the k0s10 virtual machine.

This virtual machine is the host for our K0s cluster. All next steps must be done on the terminal of this virtual machine and not the terminal of your laptop unless specifically mentioned.

Install K0s

On the virtual machine k0s10, run the K0s installation script.

The script will detect your CPU architecture (amd64 or arm64), download the latest version of K0s, and install it at /usr/local/bin.

curl -sSLf https://get.k0s.sh | sudo sh

Run k0s as a service.

sudo k0s install controller --single

Start k0s.

sudo k0s start

Check the status of k0s service with systemctl.

sudo systemctl status k0scontroller.service 

The service should be in active state.

Check k0s status with k0s status command.

 sudo k0s status
Version: v1.28.4+k0s.0
Process ID: 1887
Role: controller
Workloads: true
SingleNode: true
Kube-api probing successful: true
Kube-api probing last error: 

We have installed Kubernetes 1.28 version.

Now we have a Kubernetes cluster… up and running.

Kubernetes is an open-source project and the code is hosted on a GitHub. The Kubernetes project also releases binaries of Kubernetes which you can install by following their installation instructions. But, that’s a lenghthy process.

So we chose to install K0s instead. K0s is a Kubernetes distribution from Mirantis. K0s includes a set of installation scripts that makes it easy to install.

There are many Kubernetes distributions from vendors like Canonical, RedHat, etc. Most of those Kubernetes distributions are intended to be installed on multi-server environments. Since we are installing Kubernetes on a laptop, we opted for K0s which is more of a lightweight Kubernetes distribution.

Install and use kubectl

Our Kubernetes cluster is ready. To interact with the Kubernetes cluster, we use kubectl - the Kubernetes CLI tool.

kubectl need not be installed on the same server that runs Kubernetes. But, let’s install kubectl also in the k0s10 virtual machine in this tutorial.

The kubectl version must match the Kubernetes cluster version. Since we have installed Kubernetes 1.28, let’s install the same version of kubectl.

The kubectl binaries are available for both amd64 and arm64 architecture. Download the appropriate binaries for your architecture by replacing <arch> with either amd64 or arm64.

curl -LO https://dl.k8s.io/release/v1.28.0/bin/linux/<arch>/kubectl

If you are unsure of what architecture your Ubuntu virtual machine you can check with dpkg --print-architecture command.

Install kubectl.

sudo install -o root -g root -m 0755 kubectl /usr/local/bin/kubectl

Verify that you’ve successfully installed kubectl.

kubectl version --client
Client Version: v1.28.0
Kustomize Version: v5.0.4-0.20230601165947-6ce0bf390ce3

kubectl need to know the URL and authentication information of the cluster. By default kubectl look for these information in ~./kube/config. This file is also known as the kubeconfig file.

Let’s generate a kubeconfig file for our K0s cluster.

sudo k0s kubeconfig admin > ~/.kube/config

Now we should be able to se kubectl.

Print the nodes in the cluster with kubectl.

kubectl get nodes 
k0s10   Ready    control-plane   15h   v1.28.4+k0s

We have just one node in the cluster.

We can use kubectl to manage any resource in a Kubernetes cluster.

List the Pods.

kubectl get pods
No resources found in default namespace.

We don’t have any Pods in the default namespace, since we haven’t deployed any application on this Kubernetes cluster.

But we have Pods running in other namespaces which we can get useing -A option (all namespaces).

kubectl get pods -A
NAMESPACE     NAME                              READY   STATUS    RESTARTS   AGE
kube-system   kube-proxy-zlnnv                  1/1     Running   0          15h
kube-system   kube-router-dxl8m                 1/1     Running   0          15h
kube-system   coredns-85df575cdb-wzq96          1/1     Running   0          15h
kube-system   metrics-server-7556957bb7-4z4rj   1/1     Running   0          15h

The Pods in the kube-system namespace belong to Kubernetes control plane.

Deploy Nginx Pods

We deploy an application on Kubernetes by creating Pods. A Pod runs one or more containers. A software application running on Kubernetes is a collection of Pods.

Let’s create a Pod that runs an Nginx web server.

kubectl run nginx-app --image=nginx --port=80

This command downloads the container image Nginx from Docker Hub and runs a single Pod in the default namespace. The port parameter specified the port that the Pod is listening on.

List the Pod.

kubectl get pods
nginx-app   1/1     Running   0          97s

Our Nginx Pod is running.

To get more details in the kubectl output we can use the -o option.

kubectl get pods -o wide
nginx-app   1/1     Running   0          2m50s   k0s10   <none>           <none>

Let’s create a second Pod. Objects of the same type in Kubernetes must have unique names, so we name this Pod as nginx-app-2.

kubectl run nginx-app-2 --image=nginx --port=80 

List the Pods.

kubectl get pods -o wide
nginx-app     1/1     Running   1 (11m ago)   6h30m   k0s10   <none>           <none>
nginx-app-2   1/1     Running   0             4s   k0s10   <none>           <none>

The Pods are assigned with unique IP addresses. These IP addresses are internal to the cluster. To test the Nginx web server with curl we need to use port forwarding.

Let’s forward port 4000 in the localhost to port 80 in nginx-app Pod.

kubectl port-forward pod/nginx-app 4000:80

The port-forward command does not return immediately. So open a second terminal in the laptop and type in multipass shell k0s10 to open another termianl to the virtual machine k0s10.

Test the Nginx web app from the virtual machine terminal.

<!DOCTYPE html>
<title>Welcome to nginx!</title>
html { color-scheme: light dark; }
body { width: 35em; margin: 0 auto;
font-family: Tahoma, Verdana, Arial, sans-serif; }
<h1>Welcome to nginx!</h1>
<p>If you see this page, the nginx web server is successfully installed and
working. Further configuration is required.</p>

<p>For online documentation and support please refer to
<a href="http://nginx.org/">nginx.org</a>.<br/>
Commercial support is available at
<a href="http://nginx.com/">nginx.com</a>.</p>

<p><em>Thank you for using nginx.</em></p>

curl returns the response with the home page of our Nginx web server.

In the terminal where the port-forward command is running press Ctrl+c to stop the running command.

Delete both Pods.

kubectl delete pod nginx-app
kubectl delete pod nginx-app-2

What’s next

In this tutorial, we installed a Kubernetes cluster on our laptop and deployed an Nginx web server using kubectl. Now we are ready to explore more about Kubernetes.

Check out this article on introduction to Kubernetes to get an understanding on the architectural concept of Kubernetes.