Francesco Ciulla

Posted on Apr 18, 2023 • Updated on Apr 21, 2023

Kubernetes quick tutorial

#kubernetes #docker #javascript #node

👋 Introduction

Let's make the simplest but understandable Kubernetes example ever.

Video version:

We will create our first example with Kubernetes starting from scratch, using:

JavaScript
Node.js
Docker
Docker Compose
Kubernetes

📃 Requirements

Node.js
Docker (with Kubernetes enabled)

🚀 Getting started

Let's start in a new VS Code window (or any other editor of your choice).

1. Create a new project

npm init -y

2. Install the dependencies

npm i express

3. Create a new file called `index.js`

touch index.js

Let's create a simple Express server that listens on port 3000 and returns a response.

//simple express server
const express = require('express');
const app = express();

app.get('/', (req, res) => {
  res.send('Hello World!');
});

app.listen(3000, () => {
  console.log('Example app listening on port 3000!');
});

Run this by typing:

node index.js

Now it's time to Dockerize our application.

🐳 Dockerizing our application

We will dockerize the application by creating three files:

.dockignore
Dockerfile
docker-compose.yml

touch .dockignore Dockerfile docker-compose.yml

❌ .dockignore

Add node_modules to the file

This is to ignore the local node_modules folder when building the image.

🐳 Dockerfile

Open the Dockerfile and add the following:

FROM node:19-alpine

# Create app directory
WORKDIR /app

# Copy package.json and package-lock.json
COPY package*.json ./

# Install dependencies
RUN npm install

# Copy source code
COPY . .

# Expose port 3000
EXPOSE 3000

# Run the app
CMD [ "node", "index.js" ]

To build the app, let's populate the docker-compose.yml file with the following:

version: '3.9'

services:
  myapp:
    container_name: myapp
    image: myapp:1.0.0
    build: .
    ports:
      - 3000:3000

Explanation of the docker-compose.yml file:

version: '3.9' - the version of the docker-compose file
services: - the services (containers) we want to run. In this case, just one.
myapp: - the name of the service
container_name: myapp - the name of the container
image: myapp:1.0.0 - the name of the image
build: . - the path to the Dockerfile. In this case, the current directory.
ports: - the ports that we want to expose. In this case, port 3000 of the container will be exposed on port 3000 of the host machine.

Before we test it with Docker, let's modify the index.js file by changing Hello World! to Hello Docker!

//simple express server
const express = require('express');
const app = express();

app.get('/', (req, res) => {
  res.send('Hello Docker!');
});

app.listen(3000, () => {
  console.log('Example app listening on port 3000!');
});

Now let's run the container by typing this in the terminal:

docker compose up

Nice! Now it's the turn of Kubernetes.

But before we do that, let's change the Hello Docker! to Hello Kubernetes!

//simple express server
const express = require('express');
const app = express();

app.get('/', (req, res) => {
  res.send('Hello Kubernetes!');
});

app.listen(3000, () => {
  console.log('Example app listening on port 3000!');
});

Then rebuild the image by typing:

docker compose build

ℹ️ This is not really necessary, but good to understand how it works.

☸️ Kubernetes

Before we start, be sure Kubernetes is enabled in Docker Desktop.

Pods, Deployments, and Services

Kubernetes is a container orchestration tool. It is used to manage multiple containers on multiple hosts. In our case, we will use our machine as a single host.

We will keep it as minimal as possible by creating a deployment and a service.

Just three definitions to get started as soon as possible.

pod: the smallest deployable unit of computing that can be created and managed in Kubernetes.

deployment: a Kubernetes object that manages a set of pods. It is responsible for creating and updating pods.

service: an abstraction that defines a logical set of pods and a policy by which to access them. They are used to expose a set of pods to external traffic.

📝 Create the deployment

We will create a deployment in less than 20 lines of code.

At the root of the project, create a new file called deployment.yaml and add the following:

apiVersion: apps/v1
kind: Deployment
metadata:
  name: myapp
spec:
  replicas: 3
  selector:
    matchLabels:
      app: myapp
  template:
    metadata:
      labels:
        app: myapp
    spec:
      containers:
        - name: myapp
          image: myapp:1.0
          ports:
            - containerPort: 3000

⚠️ You might see a warning in the last five lines, but it's only because this configuration might eat a lot of resources. But let's keep it simple for now.

Explanation of the deployment.yaml file:

apiVersion: apps/v1 - the version of the Kubernetes API
kind: Deployment - the kind of object we want to create. In this case, a deployment.
name: myapp - the name of the deployment
replicas: 3 - the number of replicas of the deployment
image: myapp:1.0 - the image of the container. We defined this in the docker-compose.yml file
ports: - the ports of the container. 3000 in this case.

Before we create the deployment, let's check if there are any deployments in the cluster:

kubectl get deployments

Let's create the deployment by typing:

kubectl apply -f deployment.yaml

Then let's try to get the deployments again:

kubectl get deployments

We can also get a detail of the pods:

kubectl get pods

As you can see, we have 3/3 pods running (we defined replicas: 3 in the deployment.yaml file).

But creating a deployment is not enough. We need to expose the deployment to the outside world!

🌐 Create the service

We will create a service in 11 lines of code.

At the root of the project, create a new file called service.yaml and add the following:

apiVersion: v1
kind: Service
metadata:
  name: myapp
spec:
  selector:
    app: myapp
  ports:
    - port: 80
      targetPort: 3000
  type: LoadBalancer

This is enough to expose the deployment to the outside world (port 80)

Before we apply this configuration, let's check if there are any services in the cluster:

kubectl get services

There is one called Kubernetes but that's the default service that is created when you install Kubernetes.

Let's create the service by typing:

kubectl apply -f service.yaml

Then let's try to get the services again:

kubectl get services

Finally, visit localhost:80 in your browser and you should see Hello Kubernetes!

🧹 Clean up

To clean up, delete the deployment and the service:

kubectl delete -f deployment.yaml
kubectl delete -f service.yaml

🏁 Conclusion

We did it! We created our first example with Kubernetes starting from scratch, using:

JavaScript
Node.js
Docker
Docker Compose
Kubernetes

Video Version:

All the code is available in the GitHub repository (link in the video description): https://youtu.be/noJmOirhHKM

That's all.

If you have any questions, drop a comment below.

Francesco

Top comments (11)

Shashwat Tiwari • Apr 18 '23

Hi @francescoxx thank you writing a detailed post on kubernetes for beginners like me. I'm facing some issues with few steps mentioned in the article.

can you please explain the output of docker compose build?
and visit localhost:80 in your browser step doesn't work for me. The browser shows This page isn’t working. localhost didn’t send any data. ERR_EMPTY_RESPONSE. Please help here.

Francesco Ciulla • Apr 19 '23

docker compose build builds all the images that have a build key in the docker compose

I think you might have that port alrady in use. try to change the 80 with another one in all the files and try again

Brian Kirkpatrick • Apr 21 '23

Probably goes without saying, but the above also works seamlessly with Rancher Desktop for your local Docker & K8s provider.

Brian Kirkpatrick • Apr 22 '23

Actually, my mistake--you can't use a local LoadBalancer to expose your service from Rancher Desktop; instead you may want to define an Ingress resource.