Spring Boot application with tests and CI/CD workflow

The main purpose of this repository is to demonstrate how to set up a CI/CD workflow using GitHub Actions.

The application used is a simple Spring Boot application with a REST controller and a test class.

Workflows

There are two workflows in this repository:

• Validate PR
• CI-CD

Validate PR

This workflow validates the pull request title.

It uses an action from amannn/action-semantic-pull-request to check if the pull request title follows the semantic versioning pattern.

The workflow is triggered when a pull request is opened, edited or synchronized.

The workflow file is located at .github/workflows/validate-pr.yml`.

CI-CD

This workflow is responsible for:

• Building and testing the application.
• Building a Docker image and pushing it to Docker Hub.
• Deploying the application to an environment.

The workflow is triggered when one of the following events occurs:

• A Pull Request is opened, synchronized, reopened or closed.
• A commit is pushed to a Pull Request.
• A Pull Request is merged to the branch main.
• A new release is published.

The workflow file is located at .github/workflows/ci-cd.yml.

The workflow is divided into the following jobs:

Prepare

Build and test

The code coverage is executed with JaCoCo maven plugin and the report is generated with JaCoCo Report.

Build image and push

Deploy

The application is deployed in an AWS EC2 instance using SSH.

In order to authenticate the SSH connection, a private key is needed and so, it was created in AWS and stored as a secret in the repository. The secrets used in the workflow are:

• DOCKER_USERNAME and DOCKER_PASSWORD to authenticate in Docker Hub.
• AWS_EC2_USERNAME, AWS_EC2_IPADDRESS and AWS_EC2_PRIVATE_KEY to connect to the EC2 instance.

Note: Each time the instance is stopped and started, the EC2 instance public IP address changes and we need to update the AWS_EC2_IPADDRESS secret. To avoid this, an Elastic IP could have been used but for now, it was not implemented.

The deploy job has the following steps:

• Login to Docker Hub.
• Sets permissions to the private key file that is going to be used to authenticate SSH connection.
• The following steps are then executed inside the EC2 instance (connected via SSH):
  • Pull the new image.
  • Stop the running container (by container name).
  • Remove the container (by container name).
  • Run the new container.

Containers are named with the following pattern: spring-boot-tests-cicd-<environment>.

The environment variable value is added to the container as an environment variable in the run command. It can be STAGING or PRODUCTION.

The application will load the environment variable and use the value in the /api/greeting endpoint to present a message with the environment name.

The application port is set to 8080 and the container port is mapped to the host port 8080 for the STAGING environment and 8081 for the PRODUCTION environment.