CodePipeline and CodeBuild

AWS CodePipeline orchestrates CI/CD workflows with stages, each containing actions.

A typical pipeline:

Source stage: Pulls code from a repository. Use CodeCommit for AWS-hosted repos, or CodeStar Connections for GitHub/GitLab.

Build stage: Compiles code and runs tests using CodeBuild. The build produces artifacts stored in S3.

Test stage: Runs additional tests (integration, security, performance) as separate actions.

Deploy stage: Deploys artifacts to target environments using CodeDeploy, ECS, CloudFormation, or Lambda.

Approval stage: Manual approval before production deployment. Sends an SNS notification and waits for approval via the console or CLI.

aws pipeline create-pipeline --cli-input-json file://pipeline.json

Pipeline execution history shows each run with timestamps, status, and change details. CloudWatch Events triggers notifications on pipeline state changes.

CodeBuild compiles code, runs tests, and produces artifacts. It uses a buildspec.yml file to define the build process.

version: 0.2

phases:
  install:
    runtime-versions:
      nodejs: 20
    commands:
      - npm ci
  pre_build:
    commands:
      - npm run lint
      - npm test
  build:
    commands:
      - npm run build
  post_build:
    commands:
      - echo Build completed

artifacts:
  files:
    - dist/**/*
    - package.json

cache:
  paths:
    - node_modules/**/*

Environments define the build runtime: compute type (BUILD_GENERAL1_SMALL), image (aws/codebuild/standard:7), and service role. Use custom Docker images for specialized build tools.

Artifacts are the build outputs stored in S3. CodePipeline passes artifacts between stages. Use artifact identifiers for multiple artifact groups.

Caching speeds up builds by storing dependencies between runs. CodeBuild supports S3 and local caching.

CodePipeline supports multiple source providers.

CodeCommit: AWS-managed Git repositories. IAM-based access, triggers for CloudWatch Events, and integration with CodePipeline and CodeBuild.

GitHub via CodeStar Connections: Connect to GitHub repos using OAuth. Create a connection in the CodeStar console, then reference it in your pipeline. Supports GitHub Enterprise Cloud.

{
  "actionTypeId": {
    "category": "Source",
    "owner": "AWS",
    "provider": "CodeStarSourceConnection",
    "version": "1"
  },
  "configuration": {
    "ConnectionArn": "arn:aws:codestar-connections:us-east-1:123456789012:connection/abc",
    "FullRepositoryId": "myorg/myrepo",
    "BranchName": "main"
  }
}

S3: Store source artifacts in S3. Useful for packaging code before pipeline execution.

Source polling vs events: Pipelines can poll for changes (every few minutes) or use CloudWatch Events for immediate triggering. Events-based triggering is preferred for faster feedback.

CodePipeline supports advanced features for complex deployments.

Pipeline variables pass dynamic values between stages. Variables can come from CloudWatch Events, manual values, or variables defined in the pipeline.

Cross-region deployment deploys to resources in different AWS regions. Define stages that target Lambda functions or CloudFormation stacks in other regions.

Cross-account deployment deploys to resources in different AWS accounts. Use IAM roles in target accounts and assume them from the pipeline account.

Conditional execution uses conditions to skip stages based on branch, file changes, or manual input. Deploy to staging only on main branch, deploy to production only on release tags.

Parallel actions run multiple actions within a stage simultaneously. Run unit tests and linting in parallel during the build stage.

Stage transitions can be disabled manually to pause the pipeline. Use this during maintenance windows or incidents.