This project demonstrates a production-like deployment of a MedusaJS v2 ecommerce backend using Docker, Jenkins CI/CD, Azure Kubernetes Service (AKS), and PostgreSQL. The backend exposes REST APIs and an admin panel for managing ecommerce operations.
🔧 Tech Stack
- Backend: MedusaJS v2
- Database: PostgreSQL (containerized)
- CI/CD: Jenkins (Pipeline-based deployment)
- Containerization: Docker & Docker Compose
- Orchestration: Kubernetes (AKS)
- Cloud: Microsoft Azure
- Others: Helm (optional), kubectl, Azure CLI
📁 Project Structure
MedusajsV2/
│
├── k8s/ # Kubernetes YAML files
│ ├── deployment.yaml
│ ├── service.yaml
│ └── postgres.yaml
│
├── jenkins/
│ └── Jenkinsfile # Jenkins CI/CD pipeline script
│
├── Dockerfile # Docker image for MedusaJS backend
├── docker-compose.yml # Local development setup
└── README.md # This file
🚀 Features
- CI/CD pipeline triggered via Jenkins
- Deployment on AKS using kubectl
- External LoadBalancer to access Medusa Admin UI
- Logs and service monitoring with
kubectl - Secure PostgreSQL with volume persistence
- Easily scalable & production-ready
🛠️ Setup & Deployment
1. Clone the Repo
git clone https://github.com/jaspreet237/MedusajsV2.git
cd MedusajsV2
2. Build Docker Images
docker build -t medusa-app . docker-compose up -d
3. Push to Docker Hub
docker tag medusa-app <your-dockerhub-username>/medusa-app docker push <your-dockerhub-username>/medusa-app
4. Create AKS Cluster (Optional: if not already created)
az aks create --resource-group rg-aks-dev --name my-aks-cluster --node-count 1 --generate-ssh-keys az aks get-credentials --resource-group rg-aks-dev --name my-aks-cluster
5. Apply Kubernetes Manifests
kubectl apply -f k8s/postgres.yaml kubectl apply -f k8s/deployment.yaml kubectl apply -f k8s/service.yaml
6. Access Application
Check external IP:
kubectl get svc medusa-service
Visit:
http://<EXTERNAL-IP>:9000/app/login
🧪 Create Admin User (Inside Pod)
kubectl exec -it <medusa-pod-name> -- yarn medusa user -e admin@mail.com -p password
🧰 Jenkins CI/CD Setup
- Pipeline includes steps to:
- Clone repo
- Build Docker image
- Push to Docker Hub
- Apply Kubernetes manifests
Example Jenkinsfile is included under jenkins/.
💰 Azure AKS Cost Estimation (For MedusaJS V2 Deployment)
Deploying MedusaJS v2 on Azure Kubernetes Service (AKS) involves multiple Azure resources that contribute to the total monthly cost. The following breakdown provides an approximate cost structure for running a single production-like MedusaJS backend with PostgreSQL.
⚙️ Resource Breakdown (1-Node Setup)
| Component | Resource Type | Estimated Monthly Cost (INR/USD) | Description |
|---|---|---|---|
| AKS Cluster Node | Standard_B2s (2 vCPU, 4 GB RAM) | ₹2,000 / $25 | Runs MedusaJS and PostgreSQL pods |
| AKS Control Plane | Managed by Azure | Free | Control plane is free for 1st cluster |
| Azure Container Registry (Optional) | Basic Tier | ₹350 / $5 | Stores Docker images for deployment |
| Managed Disk (Persistent Storage) | 20 GB Premium SSD | ₹300 / $4 | Persistent volume for PostgreSQL |
| Azure Load Balancer | Standard SKU | ₹700 / $9 | Provides external access to MedusaJS API |
| Public IP (Static) | 1 Static IP | ₹250 / $3 | Assigned to the LoadBalancer service |
| Bandwidth / Egress Data | ~50 GB outbound/month | ₹500 / $6 | Depends on app traffic and usage |
🟩 Estimated Monthly Total: ~ ₹4,100 – ₹5,000 (~$50 – $60 USD)
💡 This estimate assumes 24/7 uptime with low-to-medium traffic. For development or testing environments, cost can be reduced by using Spot instances or scaling down node count.
📈 Scaling Cost for Additional Applications or Pods
If you want to deploy multiple MedusaJS backends or microservices, the cost scales linearly based on pod and node utilization.
| Scenario | Example | Approx. Cost Impact |
|---|---|---|
| 2 MedusaJS Apps | 2 Deployments sharing same AKS Node | +₹1,000 / +$12 |
| 3–4 MedusaJS Apps | Requires 2nd AKS Node | +₹2,000 / +$25 |
| Horizontal Pod Autoscaler | Scales automatically with load | Based on CPU usage and node addition |
🔹 Each new node adds approximately ₹2,000–₹2,500 ($25–$30) per month.
🔹 Storage and bandwidth usage also increase proportionally.
🧮 How to Estimate Exact Cost
You can calculate your precise AKS expenses using the Azure Pricing Calculator:
🔗 https://azure.microsoft.com/en-us/pricing/calculator
To use it:
- Add Azure Kubernetes Service
- Select VM Size (e.g., B2s or D2s_v3) and region
- Include Storage, Bandwidth, and Public IP
- Compare between Dev/Test and Production usage levels
🪄 Cost Optimization Tips
- Use B-series VMs for development (low-cost burstable instances)
- Enable node auto-scaling to reduce idle costs
- Schedule cluster shutdown for non-production workloads
- Store Docker images in Azure Container Registry (Basic tier)
- Monitor and right-size resources using Azure Advisor
📌 Future Enhancements
- Implement Ingress Controller (Nginx or Traefik)
- Enable TLS with Let’s Encrypt
- Add Redis for caching & job queue
- Configure Horizontal Pod Autoscaler
- Enable Role-Based Access Control (RBAC)
- Integrate monitoring (Prometheus + Grafana)
🧠 What You Can Learn From This Project
- Real-world AKS deployment
- End-to-end DevOps pipeline using Jenkins
- Secure and scalable Kubernetes practices
- Practical MedusaJS backend usage
🤝 Contributing
Pull requests and stars are welcome! For major changes, please open an issue first.
Pull requests and stars are welcome! For major changes, please open an issue first.