How to intermediate · 3 min read

MCP server Kubernetes deployment

Q: MCP server Kubernetes deployment

Deploy an MCP server on Kubernetes by containerizing your Python mcp.server app using stdio_server for communication, then create a Kubernetes Deployment and Service manifest to run and expose the server. Use a Dockerfile to package your server code and apply the manifests with kubectl for scalable, managed deployment.

Quick answer

Deploy an MCP server on Kubernetes by containerizing your Python mcp.server app using stdio_server for communication, then create a Kubernetes Deployment and Service manifest to run and expose the server. Use a Dockerfile to package your server code and apply the manifests with kubectl for scalable, managed deployment.

PREREQUISITES

Python 3.8+
Docker installed
kubectl configured for your Kubernetes cluster
pip install mcp
Basic knowledge of Kubernetes manifests

Setup MCP server environment

Install the mcp Python package and prepare your MCP server script using stdio_server for standard input/output communication, which is compatible with containerized environments like Kubernetes.

Create a requirements.txt with mcp and a Python script server.py that runs the MCP server.

python

pip install mcp

# requirements.txt
mcp

# server.py
from mcp.server import Server
from mcp.server.stdio import stdio_server

class MyAgent(Server):
    def handle_request(self, request):
        # Implement your agent logic here
        return {"response": "Hello from MCP server!"}

if __name__ == "__main__":
    stdio_server(MyAgent())

Containerize MCP server with Docker

Create a Dockerfile to containerize your MCP server app. This container will run your Python MCP server using stdio_server inside Kubernetes.

python

FROM python:3.10-slim
WORKDIR /app
COPY requirements.txt ./
RUN pip install --no-cache-dir -r requirements.txt
COPY server.py ./
CMD ["python", "server.py"]

Kubernetes deployment manifest

Define a Kubernetes Deployment and Service manifest to deploy and expose your MCP server container. This example runs one replica and exposes port 8080 (adjust as needed for your setup).

yaml

apiVersion: apps/v1
kind: Deployment
metadata:
  name: mcp-server
spec:
  replicas: 1
  selector:
    matchLabels:
      app: mcp-server
  template:
    metadata:
      labels:
        app: mcp-server
    spec:
      containers:
      - name: mcp-server
        image: your-dockerhub-username/mcp-server:latest
        ports:
        - containerPort: 8080
---
apiVersion: v1
kind: Service
metadata:
  name: mcp-server-service
spec:
  selector:
    app: mcp-server
  ports:
  - protocol: TCP
    port: 80
    targetPort: 8080
  type: ClusterIP

Deploy and verify

Build and push your Docker image, then apply the Kubernetes manifests to deploy your MCP server.

Commands:

bash

docker build -t your-dockerhub-username/mcp-server:latest .
docker push your-dockerhub-username/mcp-server:latest
kubectl apply -f mcp-server-deployment.yaml
kubectl get pods
kubectl get svc mcp-server-service

output

NAME                          READY   STATUS    RESTARTS   AGE
mcp-server-xxxxxxxxxx-xxxxx   1/1     Running   0          1m

NAME                 TYPE        CLUSTER-IP      EXTERNAL-IP   PORT(S)   AGE
mcp-server-service   ClusterIP   10.96.0.123     <none>        80/TCP    1m

✅

Key Takeaways

Use stdio_server for MCP server communication in containerized environments.
Containerize your MCP server with a simple Python Dockerfile for Kubernetes deployment.
Define Kubernetes Deployment and Service manifests to manage MCP server lifecycle and networking.
Build, push, and deploy your container image with kubectl commands for production-ready MCP servers.

Verified 2026-04

Verify ↗