Production Deployment

Deploy to Production

Deploy the Weather Platform to production using AWS Amplify Hosting with CloudFormation-managed infrastructure.

Prerequisites

• Sandbox environment tested and working
• All hardcoded bucket names updated in backend code
• Git repository pushed to GitHub/GitLab
• AWS profile ws1-amplify configured with deployment permissions

Step 1: Prepare for Production

Build and test locally:

# Build the Next.js application
pnpm build

# Test the production build locally
pnpm start

Commit all changes:

# Ensure all backend configuration changes are committed
git add .
git commit -m "Update backend configuration for production deployment"
git push origin main

Step 2: Deploy via Amplify Console

Method A: AWS Amplify Console (Recommended)

1. Go to AWS Amplify Console

2. Click “New app” → “Host web app”

3. Connect your Git repository:

   • Choose GitHub as source
   • Authorize AWS Amplify to access your repository
   • Select the Weather Platform repository
   • Choose prod branch

4. Configure build settings (auto-detected):

version: 1
backend:
  phases:
    preBuild:
      commands:
        - export NODE_OPTIONS="--max-old-space-size=3072"
    build:
      commands:
        - corepack enable
        - pnpm install --frozen-lockfile
        - pnpm exec ampx pipeline-deploy --branch $AWS_BRANCH --app-id
          $AWS_APP_ID
frontend:
  phases:
    preBuild:
      commands:
        - export NODE_OPTIONS="--max-old-space-size=4096"
    build:
      commands:
        - pnpm run build
  artifacts:
    baseDirectory: .next
    files:
      - "**/*"

5. Environment variables (if needed):

   • NODE_ENV: production
   • Any custom environment variables

6. Deploy: Click “Save and deploy”

Step 3: Production Resources

Amplify automatically creates:

• Production CloudFormation stacks for all backend resources
• CloudFront distribution for global content delivery
• SSL certificates with automatic renewal
• Custom domain support (optional)
• Branch-based deployments for different environments
• Atomic deployments with instant rollback capability

Backend resources deployed:

• Production versions of all Lambda functions
• S3 buckets with production naming (weather-dataset-{account-id}-main)
• Cognito User Pools and Identity Pools
• AWS Glue databases, crawlers, and ETL jobs
• CloudFront distribution for processed datasets
• EventBridge rules for scheduled processing

Step 4: Post-Deployment Configuration

IoT Policy Attachment

For each user, manually attach IoT policies (see Section 5.7):

# Get user's Cognito Identity ID and attach policy
aws iot attach-policy --policy-name WeatherPlatformPubSubPolicy \
  --target "us-east-1:12345678-1234-1234-1234-123456789012" \
  --profile ws1-amplify

Custom Domain (Optional)

1. In Amplify Console: Go to “Domain management”
2. Add custom domain: weather.yourdomain.com
3. Configure DNS: Add CNAME record as instructed
4. SSL certificate: Automatically provisioned by AWS

Step 5: Verify Production Deployment

Frontend verification:

• Website accessible via Amplify domain
• Authentication working (sign up/sign in)
• Dashboard loading correctly
• Real-time IoT data streaming

Backend verification:

• Lambda functions responding
• S3 buckets created with correct names (Test run by activating Step functions)
• Glue ETL pipeline running
• CloudFront serving processed data

Monitoring and Maintenance

CloudWatch dashboards:

• Application performance metrics
• Error rates and latency
• Cost tracking and optimization

Amplify Console features:

• Build history and logs
• Performance monitoring
• User analytics

Troubleshooting Production Issues

Common deployment issues:

1. Build failures:

   # Check build logs in Amplify Console
   # Verify dependencies in package.json
   

2. Backend deployment errors:

   # Check CloudFormation events
   # Verify IAM permissions
   

3. Domain issues:

   # Verify DNS configuration
   # Check SSL certificate status
   

Next Steps

• Monitor application performance
• Set up production monitoring and alerts
• Configure backup and disaster recovery
• Plan for resource cleanup when done