Microservice Recipe Application — Enterprise CI/CD Deployment

A fully automated, enterprise-style CI/CD deployment following the microservice architectural style. The application is a recipe management system composed of three independently deployable components:

Component	Technology	Purpose
Frontend (FE)	Node.js / Express / EJS	Dynamic web UI for managing recipes
Backend (BE)	Java 17 / Spring Boot 3	RESTful API for recipe CRUD operations
Database	MongoDB 7	Persistent data storage

Architecture

┌──────────────┐       HTTP/REST        ┌──────────────┐       MongoDB        ┌──────────────┐
│   Frontend   │ ────────────────────►  │   Backend    │ ───────────────────► │   MongoDB    │
│  (Node.js)   │  ◄──────────────────── │ (Spring Boot)│  ◄───────────────── │  (Database)  │
│   Port 3000  │       JSON             │   Port 8080  │      Driver         │  Port 27017  │
└──────────────┘                        └──────────────┘                      └──────────────┘

Deployment Architecture

GitHub Actions CI/CD
       │
       ├── Build & Test (per-service pipelines)
       ├── Vulnerability Scan (Trivy)
       ├── Container Build & Push (ACR)
       └── Deploy (Helm → AKS)
              │
    ┌─────────┴─────────┐
    │   Azure Cloud      │
    │                     │
    │  ┌───────────────┐  │
    │  │  AKS Cluster  │  │
    │  │  ┌──────────┐ │  │
    │  │  │ FE Pods  │ │  │
    │  │  ├──────────┤ │  │
    │  │  │ BE Pods  │ │  │
    │  │  ├──────────┤ │  │
    │  │  │ MongoDB  │ │  │
    │  │  └──────────┘ │  │
    │  └───────────────┘  │
    │  ┌───────────────┐  │
    │  │     ACR       │  │
    │  └───────────────┘  │
    │  ┌───────────────┐  │
    │  │ Azure Monitor │  │
    │  └───────────────┘  │
    └─────────────────────┘

Quick Start — Local Development

Prerequisites

• Docker & Docker Compose
• Node.js 20+
• Java 17+ & Maven
• MongoDB (or use Docker)

Run with Docker Compose

docker-compose up --build

• Frontend: http://localhost:3000
• Backend API: http://localhost:8080/api/recipes
• MongoDB: localhost:27017

Run Services Individually

Backend:

cd services/backend
mvn spring-boot:run

Frontend:

cd services/frontend
npm install
npm start

Cloud Deployment (Azure)

Prerequisites

• Azure CLI (az) authenticated
• Terraform >= 1.5
• Helm >= 3.12
• kubectl

1. Provision Infrastructure

cd infrastructure/terraform
cp terraform.tfvars.example terraform.tfvars  # Edit with your values
terraform init
terraform plan
terraform apply

2. Deploy Services

./scripts/deploy.sh

3. Destroy Infrastructure

./scripts/destroy.sh

CI/CD Pipeline

Each service has its own CI pipeline triggered on changes to its directory:

• .github/workflows/ci-frontend.yml — Lint, test, build, scan, push FE image
• .github/workflows/ci-backend.yml — Compile, test, build, scan, push BE image
• .github/workflows/cd-deploy.yml — Deploy all services to AKS via Helm
• .github/workflows/cd-destroy.yml — Tear down entire infrastructure

Deployment Strategy: Blue/Green

The pipeline maintains two environments (blue/green). New deployments target the inactive environment, traffic is switched after health checks pass, and the previous environment is retained for instant rollback.

Additional Features

1. Automated MongoDB Backup/Restore — CronJob-based backup to Azure Blob Storage
2. Horizontal Pod Autoscaling — CPU/memory-based auto-scaling for FE and BE
3. Vulnerability Scanning — Trivy scanning in CI with build failure on CRITICAL CVEs
4. Blue/Green Deployments — Zero-downtime deployments with instant rollback