Event-Driven Microservices System

An end-to-end Java / Spring Boot project that demonstrates a realistic event-driven microservices architecture with JWT auth, gRPC, Kafka, API Gateway, Docker, and AWS Infrastructure as Code - deployed locally using LocalStack.

---

Demo (GIF)

End-to-end flow:
JWT → Customer Created → gRPC call → Kafka event → Billing + Analytics

📹 Full demo video (MP4): docs/demo.mp4
(You can download it directly from the repository.)

---

Architecture

This system is composed of multiple microservices communicating via synchronous gRPC and asynchronous Kafka events, behind an API Gateway, and secured with JWT.

Services

• customer-service: REST API for customer CRUD + Kafka producer + gRPC client
• billing-service: gRPC server that handles billing-related actions
• analytics-service: Kafka consumer for customer events
• auth-service (identity): login + JWT validation (Spring Security)
• api-gateway: single entry point, routes requests to services
• infrastructure: CloudFormation templates to deploy the stack on LocalStack

---

Tech Stack

Backend

• Java, Spring Boot
• Spring Security (JWT)
• gRPC + Protobuf
• Kafka (MSK on LocalStack)

Platform

• Docker / Docker Compose
• API Gateway (Spring Cloud routing)
• OpenAPI / Swagger
• Integration tests

Infrastructure

• AWS LocalStack
• CloudFormation
• ECS, RDS, MSK, ALB (emulated locally)

---

Key Features

• JWT Authentication

  • /login issues a token
  • /validate verifies JWT tokens

• Customer API

  • GET /customers
  • POST /customers
  • POST /customers/{id} (update)
  • DELETE /customers/{id}

• gRPC Customer → Billing

  • Customer creation triggers a sync gRPC request to Billing

• Kafka Event-Driven Analytics

  • Customer service publishes a CUSTOMER_CREATED event
  • Analytics service consumes and processes it

• Production-style local deployment

  • Infrastructure defined using CloudFormation
  • Deployed locally with LocalStack (ECS/RDS/MSK/ALB)

---

Repository Microservices Structure

.
├─ api-gateway/
├─ auth-service/
├─ customer-service/
├─ billing-service/
├─ analytics-service/
├─ infrastructure/
├─ integration-tests/
└─ docs/
   ├─ demo.gif
   └─ demo.mp4

---

Running Locally

Prerequisites

• Highly Recommended: IntellijIdea Ultimate
• Java 21+
• Docker + Docker Compose
• LocalStack Desktop (recommended)
• AWS CLI (configured for LocalStack)
• Maven

This project is designed to run locally using LocalStack to emulate AWS services.

Quick Start (Local)

1) Build services, databases and Kafka images

Build each microservice image (run this inside each service directory):

docker build -t <service-name>:latest .

• Recommended image names (example):

# From ./customer-service
docker build -t customer-service:latest .

# From ./billing-service
docker build -t billing-service:latest .

# From ./analytics-service
docker build -t analytics-service:latest .

# From ./auth-service
docker build -t auth-service:latest .

# From ./api-gateway
docker build -t api-gateway:latest .

Pull the Postgres image (used for the service databases):

docker pull postgres:latest

Pull the Kafka image (broker/runtime image):

docker pull ubuntu/kafka:latest

• Tip: you can verify images with the following command:

docker images

2) Start LocalStack

• Open LocalStack Desktop and start it

3) Deploy infrastructure (CloudFormation)

From the infrastructure/ directory, deploy the stack running the following command:

./localstack-deploy.sh

4) Verify end-to-end

To test the complete flow, you can use the pre-configured .http files located in the /requests directory. These are compatible with the IntelliJ HTTP Client or the REST Client extension for VS Code.

1. Obtain JWT (Login) Execute the request found in requests/api-requests/auth-service/login.http. This will retrieve the token required for all protected services.

2. LocalStack Endpoint: ??? - refers to your personal LocalStack url. (Note: The script within the file automatically saves the token to the {{token}} global variable).

   POST [http://lb-???.elb.localhost.localstack.cloud:5004/auth/login]
   Content-Type: application/json
   
   {
     "email": "test@gmail.com",
     "password": "password123"
   }

3. Create a Customer (Full Flow): Use the file requests/api-requests/customer-service/create-customer.http. Ensure your request includes the Authorization header:

   Authorization: Bearer {{token}}

4. Confirm the Flow in Logs After creating a customer, verify the following steps in your terminal or Docker console:

   • HTTP 200/201 response from the API Gateway.
   • Sync Communication: Check billing-service logs to see the gRPC call creating the billing account.
   • Async Event: Check analytics-service logs to confirm the consumption of the CUSTOMER_CREATED Kafka event.

---

API Documentation (Swagger)

• Customer Service API: available via Swagger/OpenAPI
• Identity (Auth) Service API: available via Swagger/OpenAPI

---

Testing

Integration tests cover:

• Login + JWT issuance
• Unauthorized access checks
• Protected customer endpoints

Run integration tests from the integration-tests/ module:

cd integration-tests
mvn test

---

What I Learned

• Designing service boundaries and integration patterns (sync gRPC vs async Kafka)
• Implementing security flows with JWT + Spring Security
• Building an event-driven pipeline with Kafka producers/consumers
• Reproducible deployments using CloudFormation + LocalStack
• Writing meaningful integration tests for backend APIs

---

Notes

This project uses LocalStack to emulate AWS services locally.

For a real AWS deployment, the same architecture principles apply, but you'll need real AWS networking, IAM, and managed services configuration.

---