porcupine-md/jonggrang/aws-deployment

Context

Deploying {{project_name}} ({{project_type}}) to Amazon Web Services (AWS). You will follow a strict 6-phase Deployment Lifecycle Contract.

Instructions

Execute the following phases in order:

Phase 1: Authentication

1. Verify the presence of AWS credentials (AWS_ACCESS_KEY_ID and AWS_SECRET_ACCESS_KEY, or AWS_PROFILE).
2. Run aws sts get-caller-identity to confirm authentication.
3. If using an IAM role (e.g., in GitHub Actions via OIDC), ensure the role is successfully assumed.

Phase 2: Build

1. Prepare the artifacts for deployment.
2. If containerized (ECS, AppRunner, EKS): run docker build -t <image-name> .
3. If static site (S3/CloudFront): run your framework's build command (e.g., npm run build).
4. If Node.js/Python on EC2: bundle the application code.

Phase 3: Install / Provisioning

1. Ensure the target AWS resources exist.
2. If using S3: run aws s3 ls s3://<bucket-name> to verify bucket presence. Create it if it doesn't exist.
3. If using ECR: authenticate Docker to ECR (aws ecr get-login-password --region <region> | docker login --username AWS --password-stdin <account-id>.dkr.ecr.<region>.amazonaws.com).
4. If using EC2: ensure you have the correct key pairs or Session Manager access.

Phase 4: Deploy

1. Ship the artifact to AWS.
2. S3/CloudFront: aws s3 sync dist/ s3://<bucket-name> --delete. Then invalidate cache: aws cloudfront create-invalidation --distribution-id <id> --paths "/*".
3. ECS: Push image to ECR, then update the ECS service (aws ecs update-service --cluster <cluster> --service <service> --force-new-deployment).
4. EC2 (File-based): Use scp to copy files, SSH in, install dependencies, and restart PM2/Systemd.

Phase 5: Checking

1. Verify the deployment was successful.
2. Curl the application's public URL or load balancer to check for an HTTP 200 OK status.
3. Check AWS logs via CloudWatch (aws logs tail ...) if you encounter errors.
4. For ECS, check aws ecs describe-services to ensure running count matches desired count.

Phase 6: Update / Rollback

1. If Phase 5 fails, immediately initiate a rollback.
2. S3/CloudFront: Restore previous files (requires versioning enabled on S3) or sync from a previous build artifact.
3. ECS: Update the service to use the previous task definition revision.
4. Note the failure in the progress log.

Validation

• [ ] AWS authentication succeeds.
• [ ] Build succeeds.
• [ ] Artifacts are transferred.
• [ ] Health check (curl) returns 200 OK.