Deploy a React app to DigitalOcean using Github Actions and Docker

At Pindo, we recently automated all our deployment processes by setting up continuous delivery pipelines for our repositories. This setup helped us reduce the number of errors that would otherwise occur because of the repetitive steps of Continuous Delivery (CD).
In this tutorial, you will learn how to set up a continuous delivery of a React app using tools like Docker and Github Actions. We will use an Ubuntu (18.04 LTS) droplet on DigitalOcean to host our app.

Prerequisites

Here are the prerequisites required for this tutorial.

• A Github account Github
• A Docker Droplet (Ubuntu 18.04 LTS) on DigitalOcean. Sign up with my Referral Link and get $100 in credit for over 60 days.

Create your app

Use the officially supported create-react-app.dev to create a single-page React application. It offers a modern build setup with no configuration.

• Install create-react-app

npm install -g create-react-app

• Quick Start

npx create-react-app my-app && cd my-app
npm start

Dockerize your app.

Add a Dockerfile to the project root:

FROM node:13.1-alpine

WORKDIR /usr/src/app
COPY package*.json ./
RUN yarn cache clean && yarn --update-checksums
COPY . ./
EXPOSE 3000
CMD ["yarn", "start"]

yarn cache clean running this command will clear the global cache.
yarn --update-checksums lock lockfile if there's a mismatch between them and their package's checksum.

Let's build and tag our docker image

docker build -t my-app:dev .

Run the container once the build is done

docker run -it -p 3000:3000 my-app:dev 

Boom 💥! Our app is running on http://localhost:3000

Let's create another Dockerfile-prod to the project root. You will use this file in production.

Dockerfile-prod:

FROM node:13.1-alpine as build

WORKDIR /usr/src/app
COPY package*.json ./
RUN yarn cache clean && yarn --update-checksums
COPY . ./
RUN yarn && yarn build

# Stage - Production
FROM nginx:1.17-alpine
COPY --from=build /usr/src/app/build /usr/share/nginx/html
EXPOSE 80
CMD ["nginx", "-g", "daemon off;"]

In this Dockerfile-prod we create a production build for our app and then copy the build file to the Nginx HTML directory.

Next, let's build and run our production image.

docker build -f Dockerfile-prod -t my-app:prod .

docker run -itd -p 80:80 --rm my-app:prod

Our app is now running on port 80. In the next segment, we will publish the image to Github Packages.

Publish Your Image to Github Packages.

Github Packages gives you the option to publish and consume packages within your business or worldwide. To realize this, we will create a Github Action which will publish our package to the Github Packages Registry. Before we deploy our production image to the registry, we need to make sure that our code is production-ready.

deploy.yml

Let's create our first deployment action in our project.

mkdir .github && cd .github && mkdir workflows && cd workflows && touch deploy.yml

The command above creates a workflow folder and a deploy.yml file. You can replace yarn with npm in the code below.

name: build

on:
  push:
    branches: 
      - master

jobs:
  build:
    runs-on: ubuntu-latest
    steps:
    - uses: actions/checkout@v1
    - name: Use Node.js 13.10
      uses: actions/setup-node@v1
      with:
        node-version: '13.10'
    - name: Install yarn and run unittest
      run: |
        yarn
        yarn test
      env:
        CI: true
    - name: Publish to Github Packages Registry
      uses: elgohr/Publish-Docker-Github-Action@master
      with:
        name: my_github_username/my_repository_name/my_image_name
        registry: docker.pkg.github.com
        username: ${{ secrets.GITHUB_USERNAME }}
        password: ${{ secrets.GITHUB_TOKEN }}
        dockerfile: Dockerfile-prod
        tags: latest

Note that Github Actions automatically provides you with GITHUB_TOKEN secrets.

Repository

Add repository secrets

What are Secrets? They are encrypted environment variables that you create in a repository for use with GitHub Actions.

Next, let's add our GITHUB_USERNAME to the secrets.

Remember that GITHUB provides you with GITHUB_TOKEN by default.

Push to master

Let's recap. We completed setting up our secrets, created our remote repository, and set remote origins to our local repository. We are now ready to go ahead and push our changes to our remote repository.

git add -A
git commit -m "Initial commit"
git push origin master

If you click on actions, you will notice the start of the deployment workflow. Wait and see your image being published on your Github Packages Registry.

You can find your published docker image in your repository on the package tab.

We successfully published our docker app image on the Github Package Registry. We are going to order a Docker Droplet on DigitalOcean and set up a flow to deploy and our app image on DigitalOcean.

Deploy.

For deployment, we are going to create a Docker Droplet on DigitalOcean. Please do not forget to sign up with my Referral Link and get $100 in credit for over 60 days.

For this example, we access our droplet with a username and a password, please choose a one-time password over an SSH key.

After configuring and resetting your droplet password let's now add your droplet secrets to your repository.

• HOST: Droplet IP_ADDRESS
• PASSWORD: Droplet PASSWORD
• PORT: Droplet SSH port (22)
• USERNAME: Droplet USERNAME

Update deploy.yml file.

You have succeeded in setting up your droplet secrets to your repository. You will now add another code block to deploy your package and run it in our droplet using ssh-action. It's GitHub Actions for executing remote ssh commands.

name: build

on:
  push:
    branches: 
      - master

jobs:
  build:
    runs-on: ubuntu-latest
    steps:
    - uses: actions/checkout@v1
    - name: Use Node.js 13.10
      uses: actions/setup-node@v1
      with:
        node-version: '13.10'
    - name: Install yarn and run unittest
      run: |
        yarn
        yarn test
      env:
        CI: true
    - name: Publish to Github Packages Registry
      uses: elgohr/Publish-Docker-Github-Action@master
      with:
        name: my_github_username/my_repository_name/my_image_name
        registry: docker.pkg.github.com
        username: ${{ secrets.GITHUB_USERNAME }}
        password: ${{ secrets.GITHUB_TOKEN }}
        dockerfile: Dockerfile-prod
        tags: latest
    - name: Deploy package to digitalocean
      uses: appleboy/ssh-action@master
      env:
          GITHUB_USERNAME: ${{ secrets.GITHUB_USERNAME }}
          GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}
      with:
        host: ${{ secrets.HOST }}
        username: ${{ secrets.USERNAME }}
        password: ${{ secrets.PASSWORD }}
        port: ${{ secrets.PORT }}
        envs: GITHUB_USERNAME, GITHUB_TOKEN
        script: |
          docker stop $(docker ps -a -q)
          docker login docker.pkg.github.com -u $GITHUB_USERNAME -p $GITHUB_TOKEN
          docker pull docker.pkg.github.com/my_github_username/my_repository_name/my_image_name:latest
          docker run -dit -p 80:80 docker.pkg.github.com/my_github_username/my_repository_name/my_image_name:latest

We previously published the app image to the Github Package Registry by signing in with the Github Credentials (GITHUB_USERNAME and GITHUB_TOKEN ). To pull the image from the registry we must login to archive so.

Let's commit and push our changes to master.

git add -A
git commit -m "deploy to digitalocean"
git push origin master

We're using the ssh-action to remotely access our droplet from our repository.

• docker stop $(docker ps -a -q) stops all the previous running containers.
• docker run -dit -p 80:80 my_github_username/my_repository_name/my_image_name:tag pull the lastest image and run it on port 80.

As you can see below the workflow is passing.

Congratulations 🎉! You can now access your react-app on your droplet IP_ADDRESS or DOMAIN_NAME.

Mine is running on http://167.172.51.225/

You can find the code in the react-with-actions repo. I also want you to keep in mind that pushing to master is a bad practice. I did it for this tutorial. You should create a branch and make a pull request. You can follow some of git best practices here.

Should you have any questions, please do not hesitate to reach out to me on kenessajr on Twitter. Comment below if you have other feedback.