Mastering CI/CD Integration: Connecting Version Control to Application Environments

This guide explores the foundational skills required for the AWS Certified DevOps Engineer – Professional (DOP-C02) exam, focusing on how version control systems serve as the trigger and source for automated release pipelines.

---

Learning Objectives

After studying this guide, you should be able to:

• Configure AWS CodePipeline to trigger automatically from version control revisions.
• Distinguish between Stages, Actions, and Transitions within a pipeline.
• Implement cross-region and cross-account deployment patterns.
• Manage Artifacts and Revisions as they flow from source to production environments.

Key Terms & Glossary

• Revision: A specific change made to the source code (e.g., a Git commit or an S3 object version) that triggers a pipeline execution.
• Artifact: A file or set of files (like compiled code or deployment manifests) produced by one action and consumed by another.
• Stage: A logical grouping of actions (e.g., "Build", "Test", "Deploy"). Pipelines must have at least two stages.
• Transition: The link between two stages. Transitions can be disabled to prevent changes from progressing to the next stage (manual gating).
• Action: A single task performed on a revision within a stage (e.g., running a build command or deploying to Lambda).

The "Big Idea"

In a modern DevOps culture, Version Control is the Single Source of Truth. The pipeline acts as a bridge, ensuring that every code change is automatically validated and systematically moved into application environments. By integrating version control directly with pipelines, teams eliminate manual hand-offs, reduce human error, and achieve a predictable release cadence.

Formula / Concept Box

Hierarchical Outline

1. Source Integration (Version Control)
   • Supported Sources: AWS CodeCommit, GitHub, Bitbucket, Amazon S3.
   • Trigger Mechanisms: Webhooks (GitHub/Bitbucket) vs. Polling (Legacy) vs. Event-based (Amazon EventBridge).
2. Pipeline Structure
   • Stages: Grouping actions logically.
   • Actions: Serial (one after another) or Parallel (simultaneous).
   • Action Types: Source, Build, Test, Deploy, Approval, Invoke.
3. Artifact Management
   • Input/Output Artifacts: How data passes between stages.
   • Encryption: AWS KMS is used to encrypt artifacts in the S3 bucket.
4. Environment Integration
   • Deployment Providers: CodeDeploy, Elastic Beanstalk, ECS, CloudFormation, Lambda, S3.
   • Multi-Region Actions: Actions can reside in different regions than the pipeline itself.

Visual Anchors

Pipeline Flowchart

Logic of Stages and Actions

Definition-Example Pairs

• Action Type: Invoke
  • Definition: An action that triggers an AWS Lambda function to perform custom logic within the pipeline.
  • Example: A pipeline invokes a Lambda function to check if a third-party security scan is complete before allowing a production deployment.
• Action Type: Approval
  • Definition: A stage that stops the pipeline until a user manually grants permission via the console or API.
  • Example: A DevOps lead must click "Approve" in the AWS Console after reviewing the staging environment results before the code reaches Production.

Worked Examples

Scenario: Integrating a GitHub Repository

Goal: Configure a pipeline that triggers whenever a developer pushes to the main branch.

1. Source Connection: Use AWS CodeStar Connections to authorize AWS to access your GitHub account.
2. Pipeline Creation: Define a Source action in the first stage.
   • Provider: GitHub (Version 2).
   • Repo: my-org/my-app.
   • Branch: main.
   • OutputArtifact: SourceArtifact.
3. Build Integration: Add a second stage using AWS CodeBuild.
   • InputArtifact: SourceArtifact.
   • OutputArtifact: BuildArtifact.
4. Observation: Every git push now creates a new Revision ID, which initiates the pipeline execution automatically.

Checkpoint Questions

1. What is the minimum number of stages required for a valid AWS CodePipeline?
2. Where does AWS CodePipeline store the artifacts used between stages?
3. How can you prevent a pipeline from moving to the next stage without deleting the stage itself?
4. True or False: A pipeline action can be located in a different AWS region than the pipeline itself.

▶Click to reveal answers

1. Two stages.
2. In an S3 bucket (the Artifact Store).
3. By disabling the Transition between the two stages.
4. True (Cross-region actions are supported).

Muddy Points & Cross-Refs

• Serial vs. Parallel Actions: Remember that actions within a stage can run simultaneously (Parallel) to save time, but stages always run sequentially.
• Cross-Account Deployments: This requires specific IAM Role configurations (AssumeRole) and S3 bucket policy updates. Look into Unit 6 (Security) for more on IAM delegation.
• Artifact Encryption: If you use a custom S3 bucket for artifacts, you must manage the KMS keys carefully, or CodePipeline won't be able to read the files.

Comparison Tables

Source Provider Comparison