Skip to content

Lab 2a: Introduction to Kubernetes

In this lab, you will learn about Kubernetes and its networking model. We will use Minikube to setup a Kubernetes cluster on your computer.

Info

Kubernetes is a complex software with a large number of configurations and resources. For the sake of this lab, we will only make use of the resources discussed in class. If you want to learn more about Kubernetes model and resources, a good way to start is to read the documentation.

Installing Minikube

You can install Minikube following this guide.

Tip

We suggest to use the docker driver for MacOS and Linux; Hyper-V for Windows. In any case the Virtualbox driver should work in any OS.

Starting Minikube

Once Minikube is installed, you can start your Kubernetes cluster with the following command: 

minikube start --cni=calico
This command will download the necessary Kubernetes components and start a single-node cluster with Calico as the CNI.

Info

try to pass the --nodes=3 argument to minikube if you have a powerful laptop. This command will generate a 3 nodes cluster (think about each node as a VM).

A correct output should look like: 

😄  minikube v1.32.0 on Darwin 15.1.1
🎉  minikube 1.35.0 is available! Download it: https://github.com/kubernetes/minikube/releases/tag/v1.35.0
💡  To disable this notice, run: 'minikube config set WantUpdateNotification false'

✨  Using the docker driver based on existing profile
👍  Starting control plane node minikube in cluster minikube
🚜  Pulling base image ...
🤷  docker "minikube" container is missing, will recreate.
🔥  Creating docker container (CPUs=2, Memory=4000MB) ...
🐳  Preparing Kubernetes v1.28.3 on Docker 24.0.7 ...
    ▪ Generating certificates and keys ...
    ▪ Booting up control plane ...
    ▪ Configuring RBAC rules ...
🔗  Configuring Calico (Container Networking Interface) ...
    ▪ Using image gcr.io/k8s-minikube/storage-provisioner:v5
🔎  Verifying Kubernetes components...
🌟  Enabled addons: storage-provisioner, default-storageclass
🏄  Done! kubectl is now configured to use "minikube" cluster and "default" namespace by default

Interacting with Your Cluster

You can interact with your Minikube cluster using kubectl. For example, to get the status of your nodes, you can run: 

kubectl get nodes
Since we have only one node, the output should show only have one entry: 

NAME       STATUS   ROLES           AGE     VERSION
minikube   Ready    control-plane   4m36s   v1.28.3

You can also see which Pods are running in the cluster by running: 

# -A will tell the API server to return pods in every namespace
kubectl get pods -A

A correct output should look like: 

NAMESPACE     NAME                                       READY   STATUS    RESTARTS       AGE
kube-system   calico-kube-controllers-558d465845-fj4vf   1/1     Running   2 (13d ago)    13d
kube-system   calico-node-v94c9                          1/1     Running   1 (13d ago)    13d
kube-system   coredns-5dd5756b68-bqnks                   1/1     Running   4 (13d ago)    13d
kube-system   etcd-minikube                              1/1     Running   1 (13d ago)    13d
kube-system   kube-apiserver-minikube                    1/1     Running   1 (5m1s ago)   13d
kube-system   kube-controller-manager-minikube           1/1     Running   1 (13d ago)    13d
kube-system   kube-proxy-nf7df                           1/1     Running   1 (13d ago)    13d
kube-system   kube-scheduler-minikube                    1/1     Running   1 (13d ago)    13d
kube-system   storage-provisioner                        1/1     Running   5 (115s ago)   13d

If you like UIs

The content of this lab will mainly use the Kubernetes CLI kubectl. There are a number of projects that bring UIs to Kubernetes. If you prefer using them, we suggest headlamp.

Deploying an Application

To deploy an application, you can create a deployment file in YAML. For this example, we will deploy a Wiki server. A basic deployment (deployment.yaml) file should look like:

apiVersion: apps/v1
kind: Deployment
metadata:
  name: wikijs
  labels:
    app: wikijs
spec:
  replicas: 3
  selector:
    matchLabels:
      app: wikijs
  template:
    metadata:
      labels:
        app: wikijs
    spec:
      containers:
      - name: wikijs-container
        image: lscr.io/linuxserver/wikijs:latest
        ports:
        - containerPort: 3000

Verify the status of a deployment

You can install the application in the cluster by running: 

kubectl apply -f deployment.yaml

To verify that the app is running, you can check the output of the command: 

# In this case we do not specify -A because we want to know only pods in the default namespace
kubectl get pods

Tip

You will see more information if you use the -o wide option

A correct output should look like: 

NAME                                READY   STATUS    RESTARTS   AGE
wikijs-5ccdc5f64c-62m9s             1/1     Running   0          57s
wikijs-5ccdc5f64c-c6bks             1/1     Running   0          57s
wikijs-5ccdc5f64c-snsh6             1/1     Running   0          57s

Expose an application via NodePort

Interacting with an application in Kubernetes inevitably requires using a Service resource. In this case we will use a NodePort service to expose the Wikijs application outside the cluster. The corresponding service.yaml file will look like:

apiVersion: v1
kind: Service
metadata:
  name: wiki-svc 
spec:
  type: NodePort
  selector:
    app: wikijs
  ports:
    - protocol: TCP
      port: 3000
      targetPort: 3000
      nodePort: 30080

Apply the service:

kubectl apply -f service.yaml

Warning

The following instruction is only needed for minikube as it runs as a VM or container inside your computer. It will return an URL that you will paste to your browser 

minikube service wiki-svc --url

Opening the service url, you should see the following website:

Image title
WikiJS initial setup

Questions 1-2

The WikiJS installation process will fail at some point, after providing the initial username and password.

  • Can you find out why?
  • How does a correct YAML deployment look like for this application? Provide the updated YAML file(s).

Service communication

Pods usually communicate with each other using services. Service differs for IP address and communication patterns. In this part, you will analyze how services work and how to communicate with them.

Create a Pod with network analysis tools

apiVersion: v1
kind: Pod
metadata:
  name: test-container
  labels:
    app: test
spec:
  containers:
  - name: ubuntu
    image: ubuntu:22.04
    command: ["sleep", "infinity"]

After verifying the service runs, get a shell:

kubectl exec -it test-container -- bash
Then install nslookup so that we can make DNS requests to Kubernetes:

 apt update && apt install dnsutils iputils-ping -y

Finally, verify the IP address of the service: 

nslookup wiki-svc

Server:     10.96.0.10
Address:    10.96.0.10#53

Name:   wiki-svc.default.svc.cluster.local
Address: 10.103.43.191

Questions 3-4

Question 3

Which Pod is acting as a DNS server in the cluster? To which CIDR does the service IP belong? How can you access the wiki-svc from the ubuntu container?

Question 4

Create an headless and a clusterIP service for WikiJS. Which IP addresses do each service get? Can you access them from outside the cluster?

Deleting Resources

You can clean up resources either by passing the associated declarative file:

kubectl delete -f deployment.yaml

Alternatively, you can delete all resources in the default namespace with:

kubectl delete all --all

This command will delete all deployments, services, pods, and other resources in the default namespace.

Warning

Be careful when using the delete all --all command as it will remove all resources in the specified namespace.

Stopping Minikube

To stop your Minikube cluster, you can run: 

minikube stop

Deleting Minikube

To delete your Minikube cluster, you can run:

minikube delete