Minimal Demo: Kafka Schema Registry with Spring Boot (Java 21, Avro)

This repo shows a tiny producer and consumer using Avro with Confluent Schema Registry.

Quickstart

1. Start infra: Postgrsql : Run Dockerfile and the docker compose file

docker compose up -d
# Kafka at localhost:29092, Schema Registry at http://localhost:8081

it will create database schema with record

2. Build all:

./mvnw -q -DskipTests package || mvn -q -DskipTests package

4. Run consumer:

cd consumer-app
../mvnw spring-boot:run || mvn spring-boot:run

5. Run producer (new terminal):

cd producer-app
../mvnw spring-boot:run || mvn spring-boot:run

6. Send a message:

curl -X POST http://localhost:8080/users   -H "Content-Type: application/json"   -d '{"id":"u-20","email":"[email protected]","phone":"2109456738","firstName":"Manthos","lastName":"Staurou","isActive":true,"age":35}'

You should see the consumer log the received User record.

---

Evolution test

• Create account https://confluent.cloud/
• Update common-schemas/src/main/avro/User.avsc (e.g., add optional field with default).
• Use BACKWARD compatibility in confluent cloud and register the new version.
• Apps have auto.register.schemas=false and use.latest.version=true to force CI-driven registration.

Use with Confluent Cloud (bring your own keys)

Create a .env file or export env vars:

export CLOUD_BOOTSTRAP_SERVERS="pkc-xxxxx.us-central1.gcp.confluent.cloud:9092"
export CLOUD_API_KEY="******"
export CLOUD_API_SECRET="******"
export SR_URL="https://xxxxx.us-central1.gcp.confluent.cloud"
export SR_API_KEY="******"
export SR_API_SECRET="******"

Then run with the cloud profile:

# Consumer
cd consumer-app
../mvnw spring-boot:run -Dspring-boot.run.profiles=cloud

# Producer
cd ../producer-app
../mvnw spring-boot:run -Dspring-boot.run.profiles=cloud