Event-Driven Microservices with Kotlin, Spring Boot, and Apache Kafka

Demo project for the JetBrains blog tutorial "Building Event-Driven Microservices with Kotlin and Apache Kafka".

The project follows a single e-commerce scenario — processing customer orders across three independent services — and evolves it through three branches, each corresponding to a section of the tutorial.

The Scenario

An e-commerce backend receives orders and needs to coordinate between three services:

• Order Service — accepts incoming orders via REST
• Inventory Service — reserves stock for each confirmed order
• Notification Service — sends a confirmation to the customer

The tutorial starts with the services talking to each other synchronously over HTTP, then introduces Apache Kafka to decouple them, and finally improves the Kafka integration for production-readiness.

Branch Guide

starter — Synchronous REST

POST /orders
     │
     ├─── RestClient ──► POST /inventory/reserve   (Inventory Service :8081)
     │                          ↓ success / failure
     └─── RestClient ──► POST /notifications        (Notification Service :8082)

The Order Service calls the other two services synchronously and in sequence before returning a response to the caller. This is the starting point that the tutorial uses to illustrate the problem: tight coupling, cascading failures under load, and latency that compounds as you add more downstream services.

Key files to read:

• order-service/src/main/kotlin/com/example/orderservice/service/OrderService.kt — orchestrates the synchronous calls
• order-service/src/main/kotlin/com/example/orderservice/client/InventoryClient.kt — Spring RestClient wrapper
• order-service/src/main/kotlin/com/example/orderservice/client/NotificationClient.kt — Spring RestClient wrapper

Prerequisites

Tool	Version
JDK	25+
Docker & Docker Compose	any recent version

No Kafka installation needed. Docker Compose handles it.

Running the code

The starter requires no Docker. All three services talk over plain HTTP. The other two branches require running docker compose up to start the Kafka container.

1. Check out the branch

git checkout starter

2. Start the services (three separate terminals)

# Terminal 1 — Inventory Service must start first (Order Service calls it)
./gradlew :inventory-service:bootRun

# Terminal 2
./gradlew :notification-service:bootRun

# Terminal 3
./gradlew :order-service:bootRun

3. Place a test order

curl -s -X POST http://localhost:8080/orders \
  -H "Content-Type: application/json" \
  -d '{"customerId":"CUST-001","productId":"PRODUCT-001","quantity":5}' \
  | jq .

4. Try an out-of-stock order

curl -s -X POST http://localhost:8080/orders \
  -H "Content-Type: application/json" \
  -d '{"customerId":"CUST-001","productId":"PRODUCT-001","quantity":999}' \
  | jq .