Deploy Node.js Application to AWS EC2 using GitLab CI

Integration Ninjas — Fri, 14 Feb 2025 10:22:23 +0000

Introduction

Automating deployments with GitLab CI/CD is a powerful way to streamline your development workflow. In this guide, we'll walk through the process of deploying a Node.js application to an AWS EC2 instance using GitLab CI/CD and Docker. This setup ensures that every new commit is automatically built, pushed to a container registry, and deployed to your server.

Prerequisites

Node.js application: Ensure you have a Node.js application ready for deployment.
GitLab repository: Create or use an existing repository to host your application code.(Example Repo)
Docker installed: Install Docker locally(if running self-hosted runner) to build and manage container images. If you're running workflow with GitHub hosted runners then Docker is already installed there.
AWS account: Sign up for a free AWS account if you haven't already.
EC2 instance: Create an EC2 instance in AWS with appropriate specifications for your application. Configure security groups to allow inbound traffic on the port your application uses (e.g., port 3000).
SSH Client: Install an SSH client like PuTTY or Terminal (on macOS/Linux) to connect to your EC2 instance.

AWS EC2 Setup

Use below commands to setup Docker:

sudo apt-get update
sudo apt-get install docker.io -y
sudo systemctl start docker
sudo docker run hello-world
docker ps
sudo chmod 666 /var/run/docker.sock
sudo systemctl enable docker
docker --version

Use below reference to setup self-hosted runner on EC2:

Install GitLab Runner manually on GNU/Linux | GitLab

GitLab product documentation.

docs.gitlab.com

Once you create runner, use this reference to register the GitLab runner on that EC2 machine.

Flow Chart

Now, follow below steps to create required Docker file and .gitlab-ci.yml file to deploy you Node.js code on AWS EC2:

Step 1: Dockerizing the Node.js Application

First, let's create a Dockerfile to containerize our application:
Within your application directory, create a file named Dockerfile with the following content, replacing <IMAGE_NAME> with your desired image name and <PORT> with your application's port:

FROM node:20-alpine3.18
WORKDIR /app
COPY package.json ./
RUN npm install
COPY . .
EXPOSE 5000
CMD ["npm","run","start"]

This Dockerfile:

Uses Node.js 20 on Alpine Linux for a lightweight image
Sets the working directory to /app
Installs dependencies
Copies the application files
Exposes port 5000 (adjust if needed)
Runs the application using npm run start

Build and test your Docker image locally:

docker build -t nodejs-app .
docker run -p 5000:5000 nodejs-app

If everything works, move to the next step!

Step 2: Setting Up GitLab CI/CD Pipeline

To automate deployment, create a .gitlab-ci.yml file in your project root:

default:
  image: docker:24.0.5
  services:
    - docker:24.0.5-dind
  before_script:
    - docker info

stages:
  - build
  - deploy

build_job:
  stage: build
  script: |
    echo "Building the application.."
    docker login -u $DOCKER_USERNAME -p $DOCKER_PASSWORD
    docker build -t integrationninjas/nodejs-app:$CI_COMMIT_SHA .
    docker push integrationninjas/nodejs-app:$CI_COMMIT_SHA

deploy_job:
  stage: deploy
  script: |
    docker pull integrationninjas/nodejs-app:$CI_COMMIT_SHA
    docker rm -f nodejs-app-container
    docker run -d -p 5000:5000 --name nodejs-app-container integrationninjas/nodejs-app:$CI_COMMIT_SHA
  tags:
    - ec2-runner
  only:
    - deploy-to-ec2

Understanding the Pipeline

Default Section:
- Uses Docker-in-Docker (dind) to enable Docker builds inside the GitLab Runner.
Stages Section:
- Defines two stages – build and deploy.
build_job:
- Logs into Docker Hub.
- Builds the Docker image with the commit SHA as the tag.
- Pushes the image to Docker Hub.
deploy_job:
- Pulls the latest image from Docker Hub.
- Removes the old container (if running).
- Runs the new container on port 5000.
- Uses the GitLab Runner tag ec2-runner to execute on an AWS EC2 instance.
- Triggers only when pushed to the deploy-to-ec2 branch.

Now, when you push changes to the deploy-to-ec2 branch, GitLab CI/CD will automatically build and deploy your app! 🚀

Step 4: Testing the Deployment

Once the pipeline completes, check if your app is running on EC2:

curl http://your-ec2-ip:5000

You should see your application response! 🎉

Conclusion

With this setup, you've successfully automated the deployment of a Node.js application to AWS EC2 using GitLab CI/CD and Docker. Now, every commit to the deploy-to-ec2 branch will trigger an automatic build and deployment process. This approach ensures:

✅ Faster and reliable deployments
✅ Consistency across environments
✅ Less manual effort, thanks to automation.

💬 Questions? Drop them below! 🚀

DEV Community: Integration Ninjas