API Service Deployment Project

1. System Architecture

This project follows a microservice-based architecture using containerized services:

• Nginx (Reverse Proxy): Handles incoming HTTP requests and routes them to PHP.
• PHP-FPM Application: Processes business logic and API endpoints.
• Prometheus: Collects metrics from the application.
• Grafana: Visualizes monitoring data.
• Docker Compose: Orchestrates multi-container setup.
• AWS EC2: Hosts the application publicly.
• Terraform: Provisions infrastructure (EC2 + Security Groups).
• (Optional) Kubernetes: Provides orchestration, scaling, and rolling updates.

---

2. Containerization Approach

• Each service runs in its own container.
• Dockerfile builds the PHP application.
• Docker Compose manages:
  • PHP containers (scaled)
  • Nginx proxy
  • Monitoring stack (Prometheus + Grafana)

Advantages:

• Isolation of services
• Easy scaling (e.g., multiple PHP containers)
• Consistent environment across deployments

---

3. Deployment Process

CI/CD Pipeline (GitHub Actions)

1. Code is pushed to GitHub
2. Workflow triggers:
   • Pull latest code
   • Build Docker images
   • Deploy to EC2 via SSH
3. Containers are updated using: docker compose up -d --build --scale php=2

Infrastructure Provisioning

• Terraform provisions:
  • EC2 instance
  • Security group (ports 80, 22)

---

4. Zero-Downtime Deployment

Zero-downtime is achieved using:

• Multiple PHP containers (--scale php=2)
• Nginx load balances traffic between them
• During deployment:
  • New containers are built and started
  • Old containers are replaced gradually

(Optional with Kubernetes):

• Rolling updates ensure no downtime
• Traffic shifts gradually to new pods

---

5. Logging & Monitoring Setup

Logging

• Application logs stored in app.log
• Docker logs available via: docker compose logs

Monitoring

• Prometheus collects metrics
• Grafana visualizes:
  • Request rate
  • Response time
  • System performance

---

6. Handling ~100 Requests/sec

System can handle high load using:

Horizontal Scaling

• Multiple PHP containers (--scale php=2 or more)
• Load distributed via Nginx

Efficient Components

• Nginx is lightweight and high-performance
• PHP-FPM handles concurrent requests efficiently

Monitoring

• Prometheus tracks performance
• Grafana helps identify bottlenecks

Future Improvements

• Add Redis caching
• Use Kubernetes HPA (auto-scaling)
• Add CDN (CloudFront)

---

7. Security & Best Practices

• No hardcoded credentials
• Use environment variables / .env
• Use AWS IAM roles (recommended)
• Restrict SSH access in production
• Secrets can be managed via:
  • AWS Secrets Manager
  • HashiCorp Vault

---

8. Public API Testing

You can test the deployed API directly from your browser:

GET Request (Browser)

http://35.154.99.46/status

---

POST Request (Using curl)

curl -X POST http://35.154.99.46/data
-H "Content-Type: application/json"
-d '{"name":"test"}'

---

9. Conclusion

This project demonstrates a full DevOps pipeline including:

• Containerization (Docker)
• CI/CD (GitHub Actions)
• Monitoring (Prometheus + Grafana)
• Infrastructure as Code (Terraform)
• Optional orchestration (Kubernetes)