DevStream: Cloud-native, End-to-End CI/CD

Overview

DevStream is a cloud-native, end-to-end CI/CD project built using AWS DevOps services. It demonstrates the complete lifecycle of a web application from cloud-based development setup to automated build, deployment, infrastructure provisioning, and continuous delivery using a fully orchestrated pipeline.

This project is divided into 7 structured parts, where each part focuses on one critical stage of CI/CD implementation and automation.

---

Architecture Diagram

---

Part 1: Set Up a Web App in the Cloud

Objective

Set up a cloud-based development environment and generate a Java web application using Maven.

Implementation

• Launched an Amazon EC2 instance to act as a cloud development server.
• Configured secure access using:
  • Key Pair authentication (.pem)
  • SSH connection through terminal and VS Code
• Installed required build tools:
  • Java 8 (Amazon Corretto)
  • Apache Maven
  • Git
• Generated a Maven-based Java web application using archetypes:
  • Created a standard Maven web application structure
  • Prepared the project for CI/CD integration
• Updated and verified the application UI:
  • Modified index.jsp to validate that the application structure and deployment output are correct

Key Outputs

• EC2 development instance created and accessible via SSH
• Maven and Java environment configured successfully
• Web application generated and validated in the instance

To know more Click Here

---

Part 2: Connect a GitHub Repo with AWS

Objective

Connect the web application to GitHub for version control and enable remote repository tracking.

Implementation

• Created a dedicated GitHub repository for the application codebase.
• Initialized Git inside the EC2-hosted project directory:
  • Enabled version tracking for all source files
• Connected the local repository to GitHub remote origin:
  • Configured remote URL for push and fetch operations
• Pushed application code into GitHub:
  • Staged files, committed changes, and pushed to the master branch
• Configured GitHub authentication securely using a Personal Access Token (PAT):
  • Replaced password authentication with token-based access
  • Improved repository security and compatibility with automation workflows
• Verified repository sync:
  • Ensured all project files were visible and updated in GitHub

Key Outputs

• GitHub repository linked successfully with the EC2 project directory
• Source code version history established and trackable
• PAT-based secure authentication enabled for future CI/CD stages

To know more Click Here

---

Part 3: Store Dependencies in CodeArtifact

Objective

Securely manage and consume project dependencies through a private artifact repository using AWS CodeArtifact.

Implementation

• Created an AWS CodeArtifact domain:
  • Domain name: nextwork
• Created a CodeArtifact repository:
  • Repository name: nextwork-devops-cicd
  • Package format: Maven
• Enabled a public upstream repository:
  • Connected upstream to maven-central-store
  • Ensured Maven dependencies not found in CodeArtifact could still be retrieved securely
• Created and attached an IAM policy for CodeArtifact access:
  • Enabled permissions for:
    • Token retrieval
    • Repository endpoint access
    • Read access to repository packages
• Created an IAM role for the development EC2 instance:
  • Attached CodeArtifact consumer policy to this role
  • Associated the role with the EC2 instance
• Configured Maven authentication using settings.xml:
  • Added CodeArtifact repository endpoints
  • Used CODEARTIFACT_AUTH_TOKEN for secure access
• Verified CodeArtifact integration:
  • Ran Maven compile using the custom Maven settings configuration
  • Confirmed successful build and dependency resolution

Key Outputs

• Private dependency storage established using CodeArtifact
• Maven authentication and dependency retrieval working correctly
• CodeArtifact enabled secure dependency management for the pipeline

To know more Click Here

---

Part 4: Continuous Integration with CodeBuild

Objective

Automate build and packaging of the Java web application using AWS CodeBuild.

Implementation

• Created an S3 bucket for build artifacts:
  • Used for storing build outputs generated by CodeBuild
• Created an AWS CodeBuild project:
  • Source provider: GitHub
  • Repository: nextwork-web-project
  • Environment:
    • Managed Image: Amazon Linux 2
    • Runtime: Corretto 8
  • Output artifacts stored in S3 as ZIP
• Enabled logging:
  • Integrated CloudWatch Logs for build monitoring and debugging
• Updated CodeBuild IAM Role:
  • Attached CodeArtifact consumer policy
  • Ensured build environment could retrieve dependencies securely
• Created and added buildspec.yml in the repository root:
  • Exported CodeArtifact authentication token during build
  • Ran Maven build and packaging phases automatically
• Validated build execution:
  • Triggered manual build
  • Confirmed artifact generation in S3 bucket

Key Outputs

• Automated CI build process implemented using CodeBuild
• Build artifacts packaged and stored consistently in S3
• Build logs enabled through CloudWatch for transparency

To know more Click Here

---

Part 5: Deploy an App with CodeDeploy

Objective

Automate application deployment to EC2 using AWS CodeDeploy.

Implementation

• Provisioned deployment infrastructure using CloudFormation:
  • Created a dedicated stack for a deployment EC2 instance
  • Enabled required instance tags for deployment targeting
• Created a CodeDeploy application:
  • Compute platform: EC2/On-premises
• Created a deployment group:
  • Deployment type: In-place
  • Target selection:
    • EC2 instances with tag:
      • Key: role
      • Value: webserver
  • Deployment configuration:
    • Default (single-instance deployment strategy)
• Created CodeDeploy service role:
  • Attached AWS managed policy: AWSCodeDeployRole
• Executed deployment using build artifacts:
  • Retrieved build artifact S3 URI from the CodeBuild output bucket
  • Deployed the ZIP/WAR revision package into the EC2 environment
• Verified deployment:
  • Confirmed deployment status in CodeDeploy as successful
  • Validated the web application access through Public IPv4 DNS

Key Outputs

• Automated deployment configured through CodeDeploy
• EC2 instance successfully received and deployed new application revisions
• Deployment workflow ready for pipeline orchestration

To know more Click Here

---

Part 6: Automate Your Infrastructure with CloudFormation

Objective

Convert the CI/CD infrastructure into reusable Infrastructure-as-Code using AWS CloudFormation.

Implementation

• Used CloudFormation IaC Generator to scan AWS resources:
  • Identified CI/CD resources created in previous parts
• Generated an initial CloudFormation template containing key resources such as:
  • CodeArtifact domain and repositories
  • S3 artifact bucket
  • IAM roles and policies
  • CodeDeploy application
  • GitHub connection resource
• Resolved template deployment issues:
  • Fixed missing dependency ordering by using DependsOn
  • Eliminated circular dependencies by removing redundant policy references
• Enhanced the template manually by defining missing resources:
  • CodeBuild project definition
  • CodeDeploy deployment group definition
• Improved reusability with parameters:
  • GitHub repository owner
  • Repository name
• Validated template by deploying in a fresh stack:
  • Verified stack creation and resource provisioning
  • Confirmed key services were recreated successfully from the template

Key Outputs

• Infrastructure fully reproducible using CloudFormation templates
• Manual fixes implemented to resolve dependency and ordering issues
• IaC template improved for reuse and portfolio demonstration

To know more Click Here

---

Part 7: CI/CD with CodePipeline

Objective

Build a fully automated CI/CD pipeline that connects GitHub, CodeBuild, and CodeDeploy into one continuous workflow.

Implementation

• Created a CodePipeline pipeline:
  • Pipeline name: nextwork-devops-cicd
  • Execution mode: Superseded
  • Service role: auto-generated by CodePipeline
• Configured Source stage:
  • Provider: GitHub (via GitHub App)
  • Branch monitored: master
  • Enabled webhook trigger:
    • Automatically starts pipeline on code commits
• Configured Build stage:
  • Provider: AWS CodeBuild
  • Project selected: nextwork-devops-cicd
  • Build output passed to deploy stage as artifact
• Skipped Test stage:
  • Not required for this implementation
• Configured Deploy stage:
  • Provider: AWS CodeDeploy
  • Linked deployment group and application
  • Enabled rollback configuration on stage failure
• Verified automatic execution:
  • Pushed a new change to index.jsp
  • Observed new pipeline execution triggered automatically
• Verified deployment success:
  • Confirmed application changes reflected in the web server output
• Performed rollback validation:
  • Triggered rollback manually for deployment stage
  • Verified deployment reverted to previous stable version

Key Outputs

• Fully automated CI/CD pipeline created using CodePipeline
• Webhook-based deployments achieved with no manual intervention
• Rollback capability tested successfully for reliability

To know more Click Here

---

Final Outcome

What Was Achieved

• Implemented a complete CI/CD workflow where:
  • Code changes pushed to GitHub automatically trigger pipeline execution
  • CodeBuild compiles and packages the application
  • Artifacts are stored in S3 and deployed using CodeDeploy
  • Deployments are automated and monitored through CodePipeline

Final Results

• Fully automated end-to-end CI/CD pipeline successfully deployed on AWS
• Infrastructure reproducible through CloudFormation templates
• Dependency management secured using CodeArtifact
• Rollback procedures validated for production-level reliability