In the previous phase, we successfully set up Continuous Integration (CI) using Jenkins to automate the build and testing stages of our application. Now, we take the pipeline forward — by incorporating Docker to containerize our application artifacts, pushing the images to Amazon ECR, and finally implementing Continuous Deployment (CD) by hosting those images via AWS ECS.

This write-up builds directly on the earlier Jenkins CI pipeline and details how that workflow was extended into a full CI/CD pipeline with Docker and AWS.

CI with Docker: Creating and Publishing Container Images

We began by integrating Docker into the CI workflow to containerize the application. This allowed the application, along with its environment and dependencies, to be packaged into a portable and consistent format.

Objective:

• Build a Docker image from the application code
• Push this image to Amazon Elastic Container Registry (ECR) for later use in deployment

Other registries like Docker Hub, Google Container Registry, or Azure Container Registry could have been used, but AWS ECR was chosen for its seamless integration with other AWS services.

Pipeline Steps:

1. AWS Setup:
   • Created an IAM User with permissions:
     • AmazonEC2ContainerRegistryFullAccess
     • (Later) AmazonECS_FullAccess for deployment
   • Created an ECR Repository to host the container images
2. Jenkins Configuration:

   • Installed relevant plugins:

     • Docker
     • Docker Pipeline
     • Amazon ECR

   • Stored AWS credentials securely within Jenkins

   • Installed Docker Engine on the Jenkins host

   • Verified Docker installation and permissions by:

     • Adding the jenkins user to the docker group

     

     • Rebooting the instance
     • Confirming Docker image listing via docker images
3. Local Testing and AWS CLI Setup:
   • Installed AWS CLI (classic version via Snap)
   • Verified Docker and AWS CLI integration

Result:

The Jenkins job successfully:

• Built the Docker image using the Dockerfile from the application repo
• Tagged the image
• Authenticated with ECR
• Pushed the image to ECR