A DevOps + microservices project that orchestrates a distributed Spring Boot application in Kubernetes using Helm, monitored with Prometheus & Grafana, with tracing via Zipkin and asynchronous communication via RabbitMQ.
📦 The app simulates an e-commerce ecosystem where customers can order products, orders are automatically processed, and notifications are sent asynchronously through RabbitMQ.
- ✅ Fully automated deployment using Helm
- 📊 Metrics monitoring (CPU, RAM, etc.) with Prometheus + Grafana
- 📬 Message distribution via RabbitMQ (
order-service→notification-service) - 🔍 Tracing with Zipkin
- 📂 Configurable persistence (MySQL, Redis)
- 🔁 Liveness & Readiness probes
- 📊 Resource Requests, Limits & HPA (when enabled)
- API Gateway with Redis rate limiting & Circuit Breakers
- Resilient Communication with Resilience4j
- Asynchronous Messaging with Spring Cloud Stream (RabbitMQ / Kafka-ready)
- Distributed Tracing with Zipkin
- Metrics Exposure with Spring Boot Actuator
- Optimized Docker Images following multi-stage build best practices
- Spring Cloud Bus for real-time propagation of configuration changes
| Service | Local Access URL | Description |
|---|---|---|
gateway-server |
http://localhost:8080 | Routes incoming requests to microservices. Integrated with Redis for rate limiting and Resilience4j for Circuit Breakers. Exposes metrics via Spring Boot Actuator and is connected to Zipkin for distributed tracing. |
order-service |
http://localhost:8082 | Calls product-service via Feign to register orders and check stock. Integrated with Resilience4j for resilience. After processing, publishes messages via RabbitMQ to notification-service. Uses Spring Cloud Stream for broker flexibility. Exposes Actuator metrics and sends tracing data to Zipkin. |
product-service |
http://localhost:8081 | Stores product information. Exposes Actuator metrics and sends tracing data to Zipkin. |
notification-service |
http://localhost:8083 | Saves notifications in the database and logs them. Can be extended with email/SMS. Connected to RabbitMQ via Spring Cloud Stream. Exposes Actuator metrics and sends tracing data to Zipkin. |
mysql |
N/A (internal) | Relational database |
redis |
N/A (internal) | Used by gateway-server for rate limiting |
rabbitmq |
http://localhost:15672 | Message broker: used by order-service to notify notification-service |
grafana |
http://localhost:3000 | Metrics dashboard |
prometheus |
http://localhost:9090 | Metrics collector |
zipkin |
http://localhost:9411 | Distributed request tracing |
Each microservice is built into a Docker image used during Kubernetes deployment.
| Microservice | Docker Image | Role |
|---|---|---|
gateway-server |
xalx1/gateway-server:2.0.0-k8s |
Routing + Rate Limiting |
order-service |
xalx1/order-service:2.0.0-k8s |
Order processing |
product-service |
xalx1/product-service:2.0.0-k8s |
Product catalog |
notification-service |
xalx1/notification-service:2.0.0-k8s |
Async notifications |
- Docker
- Helm
- Minikube
- Java 17+
- Maven / Gradle
- Kubectl
spring-cloud-k8s/
├── gateway-server/ # API Gateway (Spring Cloud Gateway)
├── notification-service/ # Notification microservice
├── order-service/ # Order management microservice
├── product-service/ # Product catalog microservice
├── k8s-charts/ # Helm charts for each service and infrastructure
├── k8s-manifests/ # K8s manifests (optional)
├── scripts/ # Automation and helper scripts
└── README.md
# 1. Start Minikube
sh scripts/init-minikube.sh
# 2. Deploy all infrastructure and services
sh scripts/deploy-all.sh
# 3. (Optional) Port forward for local UI access
sh scripts/port-forward.sh
- Grafana: http://localhost:3000
➔ Default login:admin / admin
- Prometheus: http://localhost:9090
- Zipkin UI: http://localhost:9411
- Distributed tracing for request flow analysis
# Get all products
curl http://localhost:8081/products
# Get all orders
curl http://localhost:8082/orders
# Create new order (triggers async notification)
curl -X POST http://localhost:8082/orders \
-H "Content-Type: application/json" \
-d '{"productId": 1, "quantity": 2, "customerEmail": "test@example.com"}'
# Get notifications
curl http://localhost:8083/notifications
| Script | Description |
|---|---|
init-minikube.sh |
Starts Minikube and sets up context |
deploy-all.sh |
Full Helm-based deployment of all services |
delete-all.sh |
Removes all Helm releases from the project |
status.sh |
Shows status of K8s resources |
port-forward.sh |
Exposes local ports for UI access (Grafana, RabbitMQ, etc.) |
- Config values (resources, persistence, HPA) are adjustable via
values.yaml - HPA is enabled only when
autoscaling.enabled=truein the Helm chart - Persistent storage uses auto-generated
PersistentVolumeClaimobjects - The architecture is easily extensible with new services or DevOps tools
➡️ http://localhost:3000 (login: admin/admin)