Deployment Patterns

Deployment Patterns

Master deployment automation with Claude Code. From generating deployment configurations to orchestrating complex release processes, this guide covers practical patterns for modern deployment workflows.

Deployment Automation Overview

Notatka

Claude Code can generate deployment configurations, create rollback procedures, and help orchestrate complex multi-environment deployments with consistency and reliability.

Key Deployment Areas

Configuration Generation

• Infrastructure as Code
• Container orchestration
• CI/CD pipelines
• Environment configs

Release Management

• Deployment strategies
• Rollback procedures
• Health checks
• Migration scripts

Monitoring Setup

• Observability configs
• Alert definitions
• Dashboard creation
• Log aggregation

Infrastructure as Code

Terraform Generation

1. Generate base infrastructure

   claude "Create Terraform configuration for:
   - AWS VPC with public/private subnets
   - EKS cluster with 3 node groups
   - RDS PostgreSQL with read replicas
   - ElastiCache Redis cluster
   - Application Load Balancer
   Follow AWS Well-Architected Framework" \
   --output infrastructure/terraform/

2. Create environment variations

   # Development environment
   claude "Create Terraform tfvars for development:
   - Smaller instance sizes
   - Single AZ deployment
   - Minimal redundancy
   - Cost optimization focus" \
   --output infrastructure/terraform/environments/dev.tfvars
   
   # Production environment
   claude "Create Terraform tfvars for production:
   - High availability across 3 AZs
   - Auto-scaling configurations
   - Enhanced monitoring
   - Backup strategies" \
   --output infrastructure/terraform/environments/prod.tfvars

3. Generate modules

   claude "Create reusable Terraform modules for:
   - Standard web application setup
   - Database provisioning with backups
   - Container cluster with monitoring
   - Networking with security groups
   Include examples and documentation" \
   --output infrastructure/terraform/modules/

CloudFormation Templates

Application Stack

claude "Generate CloudFormation template for:
- Elastic Beanstalk application
- Auto-scaling configuration
- CloudFront distribution
- S3 buckets for static assets
- Route53 DNS configuration
Include parameters for environment customization" \
--output cloudformation/app-stack.yaml

Serverless Stack

claude "Create CloudFormation for serverless app:
- Lambda functions with layers
- API Gateway with custom domain
- DynamoDB tables with streams
- Step Functions for workflows
- EventBridge for event routing" \
--output cloudformation/serverless-stack.yaml

Container Stack

claude "Generate CloudFormation for containers:
- ECS Fargate cluster
- Task definitions with secrets
- Application Load Balancer
- Service auto-scaling
- CloudWatch Container Insights" \
--output cloudformation/container-stack.yaml

Kubernetes Deployments

Manifest Generation

Complete K8s Application

# Generate comprehensive Kubernetes manifests
claude "Create Kubernetes manifests for microservices app:

Services:
- API Gateway (nginx ingress)
- Auth Service (Node.js)
- User Service (Python)
- Payment Service (Go)
- Frontend (React)

Include:
- Deployments with resource limits
- Services for internal communication
- ConfigMaps for configuration
- Secrets for sensitive data
- Ingress rules with TLS
- HorizontalPodAutoscaler
- PodDisruptionBudgets
- NetworkPolicies for security
- ServiceMonitor for Prometheus

Use best practices for production" \
--output k8s/manifests/

Helm Chart Creation

1. Initialize Helm chart

   claude "Create Helm chart for the microservices app:
   - Chart.yaml with proper metadata
   - Flexible values.yaml with environments
   - Templates for all resources
   - Helpers for common patterns
   - NOTES.txt with usage instructions
   - README.md with examples" \
   --output helm/microservices-app/

2. Environment-specific values

   # Development values
   claude "Create Helm values for development:
   - Single replicas
   - Minimal resources
   - Local storage
   - Debug logging
   - No TLS" \
   --output helm/microservices-app/values-dev.yaml
   
   # Staging values
   claude "Create Helm values for staging:
   - 2 replicas
   - Moderate resources
   - Persistent storage
   - Info logging
   - Let's Encrypt TLS" \
   --output helm/microservices-app/values-staging.yaml
   
   # Production values
   claude "Create Helm values for production:
   - 3+ replicas with pod anti-affinity
   - Production-grade resources
   - Multi-AZ storage
   - Structured logging
   - Commercial TLS certificates" \
   --output helm/microservices-app/values-prod.yaml

Kustomize Configuration

Base Configuration

claude "Create Kustomize base configuration:
- Base manifests for all resources
- Common labels and annotations
- Shared ConfigMaps
- Cross-cutting policies" \
--output k8s/kustomize/base/

Environment Overlays

claude "Create Kustomize overlays for:
- Development (local)
- Staging (AWS)
- Production (AWS)
- DR site (Azure)
Each with specific patches and transforms" \
--output k8s/kustomize/overlays/

CI/CD Pipeline Generation

GitHub Actions Workflows

Complete CI/CD Pipeline

# Generated by Claude Code
name: Deploy to Production

on:
  push:
    tags:
      - 'v*'

env:
  AWS_REGION: us-east-1
  ECR_REPOSITORY: myapp
  EKS_CLUSTER: production-cluster

jobs:
  test:
    runs-on: ubuntu-latest
    steps:
      - uses: actions/checkout@v4

      - name: Run Tests
        run: |
          npm ci
          npm test
          npm run test:integration

      - name: Security Scan
        run: |
          npm audit
          docker run --rm -v "$PWD":/src \
            aquasec/trivy fs --severity HIGH,CRITICAL /src

  build:
    needs: test
    runs-on: ubuntu-latest
    outputs:
      image-tag: ${{ steps.meta.outputs.tags }}

    steps:
      - uses: actions/checkout@v4

      - name: Configure AWS
        uses: aws-actions/configure-aws-credentials@v4
        with:
          aws-access-key-id: ${{ secrets.AWS_ACCESS_KEY_ID }}
          aws-secret-access-key: ${{ secrets.AWS_SECRET_ACCESS_KEY }}
          aws-region: ${{ env.AWS_REGION }}

      - name: Login to ECR
        id: login-ecr
        uses: aws-actions/amazon-ecr-login@v2

      - name: Build and Push
        id: meta
        uses: docker/metadata-action@v5
        with:
          images: ${{ steps.login-ecr.outputs.registry }}/${{ env.ECR_REPOSITORY }}
          tags: |
            type=semver,pattern={{version}}
            type=semver,pattern={{major}}.{{minor}}
            type=sha

      - name: Build Docker image
        uses: docker/build-push-action@v5
        with:
          context: .
          push: true
          tags: ${{ steps.meta.outputs.tags }}
          cache-from: type=gha
          cache-to: type=gha,mode=max

  deploy:
    needs: build
    runs-on: ubuntu-latest
    environment: production

    steps:
      - uses: actions/checkout@v4

      - name: Configure kubectl
        run: |
          aws eks update-kubeconfig \
            --region ${{ env.AWS_REGION }} \
            --name ${{ env.EKS_CLUSTER }}

      - name: Deploy with Helm
        run: |
          helm upgrade --install myapp ./helm/myapp \
            --namespace production \
            --create-namespace \
            --values helm/myapp/values-prod.yaml \
            --set image.tag=${{ needs.build.outputs.image-tag }} \
            --wait \
            --timeout 10m

      - name: Verify Deployment
        run: |
          kubectl rollout status deployment/myapp -n production
          kubectl get pods -n production

      - name: Run Smoke Tests
        run: |
          npm run test:smoke -- --env=production

GitLab CI Pipeline

Docker Build

.gitlab-ci.yml

stages:
  - test
  - build
  - deploy

variables:
  DOCKER_DRIVER: overlay2
  DOCKER_TLS_CERTDIR: "/certs"

test:
  stage: test
  image: node:18
  script:
    - npm ci
    - npm test
    - npm run lint
  coverage: '/Coverage: \d+\.\d+%/'

build:
  stage: build
  image: docker:latest
  services:
    - docker:dind
  before_script:
    - docker login -u $CI_REGISTRY_USER -p $CI_REGISTRY_PASSWORD $CI_REGISTRY
  script:
    - docker build -t $CI_REGISTRY_IMAGE:$CI_COMMIT_SHA .
    - docker tag $CI_REGISTRY_IMAGE:$CI_COMMIT_SHA $CI_REGISTRY_IMAGE:latest
    - docker push $CI_REGISTRY_IMAGE:$CI_COMMIT_SHA
    - docker push $CI_REGISTRY_IMAGE:latest

deploy-staging:
  stage: deploy
  image: bitnami/kubectl:latest
  script:
    - kubectl config use-context $K8S_CONTEXT_STAGING
    - kubectl set image deployment/myapp myapp=$CI_REGISTRY_IMAGE:$CI_COMMIT_SHA -n staging
    - kubectl rollout status deployment/myapp -n staging
  environment:
    name: staging
    url: https://staging.example.com
  only:
    - develop

deploy-production:
  stage: deploy
  image: bitnami/kubectl:latest
  script:
    - kubectl config use-context $K8S_CONTEXT_PRODUCTION
    - kubectl set image deployment/myapp myapp=$CI_REGISTRY_IMAGE:$CI_COMMIT_SHA -n production
    - kubectl rollout status deployment/myapp -n production
  environment:
    name: production
    url: https://example.com
  only:
    - tags
  when: manual

Terraform Deploy

# Infrastructure deployment
terraform-plan:
  stage: plan
  image: hashicorp/terraform:latest
  script:
    - cd infrastructure/terraform
    - terraform init
    - terraform plan -out=tfplan
  artifacts:
    paths:
      - infrastructure/terraform/tfplan

terraform-apply:
  stage: deploy
  image: hashicorp/terraform:latest
  script:
    - cd infrastructure/terraform
    - terraform init
    - terraform apply tfplan
  dependencies:
    - terraform-plan
  when: manual
  only:
    - main

Deployment Strategies

Blue-Green Deployment

1. Generate blue-green scripts

   claude "Create blue-green deployment scripts:
   - Health check verification
   - Traffic switching logic
   - Rollback procedures
   - Database migration handling
   For both Kubernetes and AWS ECS" \
   --output deployment/blue-green/

2. Traffic management

   claude "Generate traffic switching configuration:
   - AWS ALB target group switching
   - Kubernetes service selector updates
   - Istio VirtualService for canary
   - CloudFlare load balancer rules" \
   --output deployment/traffic-management/

Canary Deployment

Progressive Rollout

# Generate Flagger configuration for canary deployments
claude "Create Flagger canary deployment config:
- Progressive traffic shifting (10%, 25%, 50%, 100%)
- Automated rollback on failures
- Custom metrics for business KPIs
- Slack notifications
- Load testing during canary
Include for both Istio and AWS App Mesh" \
--output deployment/canary/flagger-config.yaml

# Generate custom metrics
claude "Create Prometheus queries for canary analysis:
- Request success rate
- P95 latency
- Error rate by status code
- Business metrics (orders, signups)
Format for Flagger MetricTemplate" \
--output deployment/canary/metrics.yaml

Database Migrations

Migration Strategies

Zero-Downtime Migrations

claude "Create zero-downtime migration plan:
- Backward compatible schema changes
- Data migration scripts
- Rollback procedures
- Verification queries
For PostgreSQL with millions of records" \
--output migrations/zero-downtime/

Multi-Version Support

claude "Generate migration strategy for:
- Supporting old and new schema
- Feature flags for gradual rollout
- Data sync between versions
- Cleanup procedures
Include example code" \
--output migrations/multi-version/

Monitoring and Rollback

Deployment Monitoring

1. Create monitoring stack

   claude "Generate monitoring configuration:
   - Prometheus scrape configs
   - Grafana dashboards for deployments
   - Alert rules for failures
   - Custom metrics collection
   - Log aggregation queries" \
   --output monitoring/deployment/

2. Automated rollback

   claude "Create automated rollback system:
   - Health check definitions
   - Failure detection logic
   - Rollback triggers
   - State preservation
   - Notification system
   Support Kubernetes, ECS, and Lambda" \
   --output deployment/rollback/

Health Checks

Comprehensive Health Checks

# Generated health check system
import asyncio
import aiohttp
from typing import List, Dict
import json

class HealthChecker:
    def __init__(self, config_file: str):
        with open(config_file) as f:
            self.config = json.load(f)
        self.results = {}

    async def check_endpoint(self, endpoint: Dict) -> Dict:
        """Check individual endpoint health"""
        try:
            async with aiohttp.ClientSession() as session:
                async with session.get(
                    endpoint['url'],
                    timeout=aiohttp.ClientTimeout(total=endpoint.get('timeout', 30))
                ) as response:

                    # Basic health
                    health = {
                        'url': endpoint['url'],
                        'status_code': response.status,
                        'response_time': response.headers.get('X-Response-Time'),
                        'healthy': response.status == endpoint.get('expected_status', 200)
                    }

                    # Custom checks
                    if 'expected_response' in endpoint:
                        body = await response.json()
                        health['matches_expected'] = (
                            body == endpoint['expected_response']
                        )

                    return health

        except Exception as e:
            return {
                'url': endpoint['url'],
                'healthy': False,
                'error': str(e)
            }

    async def check_all(self) -> Dict:
        """Run all health checks in parallel"""
        tasks = [
            self.check_endpoint(endpoint)
            for endpoint in self.config['endpoints']
        ]

        results = await asyncio.gather(*tasks)

        # Aggregate results
        return {
            'timestamp': datetime.now().isoformat(),
            'overall_health': all(r['healthy'] for r in results),
            'endpoints': results,
            'summary': {
                'total': len(results),
                'healthy': sum(1 for r in results if r['healthy']),
                'unhealthy': sum(1 for r in results if not r['healthy'])
            }
        }

    async def continuous_monitoring(self, interval: int = 60):
        """Continuously monitor health"""
        while True:
            results = await self.check_all()

            # Store results
            self.results[results['timestamp']] = results

            # Alert if unhealthy
            if not results['overall_health']:
                await self.send_alert(results)

            await asyncio.sleep(interval)

    async def send_alert(self, results: Dict):
        """Send alerts for failures"""
        # Implement Slack, PagerDuty, email alerts
        pass

# Configuration file
health_config = {
    "endpoints": [
        {
            "name": "API Gateway",
            "url": "https://api.example.com/health",
            "expected_status": 200,
            "timeout": 30
        },
        {
            "name": "Auth Service",
            "url": "https://auth.example.com/health",
            "expected_status": 200,
            "expected_response": {"status": "healthy"}
        },
        {
            "name": "Database",
            "url": "https://api.example.com/health/db",
            "expected_status": 200,
            "critical": true
        }
    ],
    "alerts": {
        "slack_webhook": "https://hooks.slack.com/...",
        "pagerduty_key": "..."
    }
}

Environment Management

Configuration Generation

Kubernetes ConfigMaps

# Generate environment configs
for env in dev staging prod; do
  claude "Create ConfigMap for $env environment:
  - API endpoints
  - Feature flags
  - Cache settings
  - Log levels
  - Third-party service URLs
  Use template variables for secrets" \
  --output k8s/configs/configmap-$env.yaml
done

AWS Parameter Store

claude "Generate AWS Systems Manager parameters:
- Hierarchical structure (/app/env/component/key)
- SecureString for sensitive values
- Standard tier for configs
- CloudFormation template
- Terraform configuration
Include access policies" \
--output infrastructure/parameter-store/

.env Files

claude "Create environment file templates:
- .env.example with all variables
- .env.development with local values
- .env.test for testing
- Documentation for each variable
- Validation script
Never include real secrets" \
--output environments/

Deployment Automation Scripts

Orchestration Scripts

Master Deployment Script

#!/bin/bash
# deploy.sh - Generated by Claude Code

set -euo pipefail

# Configuration
ENVIRONMENT="${1:-staging}"
VERSION="${2:-latest}"
DRY_RUN="${DRY_RUN:-false}"

# Colors for output
RED='\033[0;31m'
GREEN='\033[0;32m'
YELLOW='\033[1;33m'
NC='\033[0m'

log() {
    echo -e "${GREEN}[$(date +'%Y-%m-%d %H:%M:%S')]${NC} $1"
}

error() {
    echo -e "${RED}[ERROR]${NC} $1" >&2
    exit 1
}

warn() {
    echo -e "${YELLOW}[WARN]${NC} $1"
}

# Pre-deployment checks
pre_deploy_checks() {
    log "Running pre-deployment checks..."

    # Check cluster connectivity
    if ! kubectl cluster-info &>/dev/null; then
        error "Cannot connect to Kubernetes cluster"
    fi

    # Verify namespace exists
    if ! kubectl get namespace "$ENVIRONMENT" &>/dev/null; then
        warn "Namespace $ENVIRONMENT does not exist, creating..."
        kubectl create namespace "$ENVIRONMENT"
    fi

    # Check image exists
    if ! docker manifest inspect "$IMAGE:$VERSION" &>/dev/null; then
        error "Docker image $IMAGE:$VERSION not found"
    fi

    # Run security scan
    log "Running security scan..."
    trivy image "$IMAGE:$VERSION" --severity HIGH,CRITICAL

    log "Pre-deployment checks passed ✓"
}

# Deploy application
deploy() {
    log "Deploying version $VERSION to $ENVIRONMENT..."

    if [[ "$DRY_RUN" == "true" ]]; then
        log "DRY RUN - would execute:"
        echo "helm upgrade --install myapp ./helm/myapp \\"
        echo "  --namespace $ENVIRONMENT \\"
        echo "  --values helm/myapp/values-$ENVIRONMENT.yaml \\"
        echo "  --set image.tag=$VERSION"
        return 0
    fi

    # Backup current state
    kubectl get all -n "$ENVIRONMENT" -o yaml > "backup-$ENVIRONMENT-$(date +%s).yaml"

    # Deploy with Helm
    helm upgrade --install myapp ./helm/myapp \
        --namespace "$ENVIRONMENT" \
        --values "helm/myapp/values-$ENVIRONMENT.yaml" \
        --set image.tag="$VERSION" \
        --wait \
        --timeout 10m \
        --atomic

    log "Deployment completed ✓"
}

# Post-deployment verification
verify_deployment() {
    log "Verifying deployment..."

    # Wait for rollout
    kubectl rollout status deployment/myapp -n "$ENVIRONMENT"

    # Check pod status
    READY_PODS=$(kubectl get pods -n "$ENVIRONMENT" -l app=myapp \
        -o jsonpath='{.items[?(@.status.phase=="Running")].metadata.name}' | wc -w)

    if [[ "$READY_PODS" -lt 1 ]]; then
        error "No running pods found"
    fi

    # Run health checks
    log "Running health checks..."
    ./scripts/health-check.sh "$ENVIRONMENT"

    # Run smoke tests
    log "Running smoke tests..."
    npm run test:smoke -- --env="$ENVIRONMENT"

    log "Verification completed ✓"
}

# Main execution
main() {
    log "Starting deployment process"
    log "Environment: $ENVIRONMENT"
    log "Version: $VERSION"

    pre_deploy_checks
    deploy
    verify_deployment

    log "Deployment successful! 🚀"
}

# Run main function
main "$@"

Best Practices

Deployment Best Practices

1. Automate everything - Manual steps lead to errors
2. Test in staging - Always validate before production
3. Monitor actively - Know immediately if something breaks
4. Plan rollbacks - Have a quick way to revert
5. Document processes - Keep runbooks updated
6. Use GitOps - Git as single source of truth
7. Implement gradual rollouts - Reduce blast radius

Security Considerations

Uwaga

• Never commit secrets to repositories
• Use proper secret management (Vault, AWS Secrets Manager)
• Implement least-privilege access
• Audit deployment permissions regularly
• Encrypt sensitive data in transit and at rest

Troubleshooting Deployments

Common Issues

Failed Health Checks

claude "Create troubleshooting guide for:
- Analyzing health check failures
- Common causes and fixes
- Debug commands
- Log locations
Format as runbook" \
--output docs/troubleshooting/health-checks.md

Resource Constraints

claude "Generate resource debugging scripts:
- Check cluster capacity
- Identify resource bottlenecks
- Recommend scaling solutions
- Cost optimization tips" \
--output scripts/debug-resources.sh

Next Steps

Continue improving your deployment workflows:

• GitHub Actions Integration for CI/CD
• Containerization for Docker workflows
• IDE Coordination for development integration

Remember: Good deployment practices are about reliability, repeatability, and rapid recovery. Use Claude Code to generate robust deployment configurations that handle edge cases and failures gracefully.