Accessing Cluster Applications via NodePort Service

When internal access to a application isn't sufficiant and external access is required, a NodePort Service can be used. This service type exposes the application using a static port on each node’s IP. From outside the cluster, the service can be accessed via <NodeIP>:<NodePort>.

Here's a sample YAML definition:

kind: Service
apiVersion: v1
metadata:
  name: nodeport-service
spec:
  type: NodePort
  ports:
    - port: 80
      nodePort: 31001
  selector:
    app: demo

Create and verify the service with the following commands:

kubectl create -f nodePortService.yaml
kubectl get services nodeport-service

Once created, the service will be accessible externally via the node IP and port (e.g., 172.16.2.201:31001).

However, this method has limitations:

• Each port can only be used by one service.
• Port range is limited to 30000-32767.
• Service becomes unreachable if the node IP changes.

These constraints make NodePort suitable for development and testing, but not ideal for production environments.

Accessing Cluster Applications via LoadBalancer Service

To overcome the limitations of NodePort, a LoadBalancer Service provides a more robust solution. It exposes the service externally using a cloud provider's load balancer, typically assigning a public IP address.

Below is an example YAML definition for a LoadBalancer Service:

apiVersion: v1
kind: Service
metadata:
  name: demo
  namespace: default
spec:
  clusterIP: 10.3.255.28
  loadBalancerIP: 106.52.99.55
  ports:
    - name: tcp-80-80
      nodePort: 31504
      port: 80
      protocol: TCP
      targetPort: 80
  selector:
    app: demo
    k8s-app: demo
    qcloud-app: demo
  type: LoadBalancer

After deploying this service, external access can be achieved via:

• Node IP and port
• LoadBalancer IP
• Domain name (configured with DNS pointing to the LoadBalancer IP)

Deploying Microsoft SQL Server in Kubernetes

This section demonstrates deploying Microsoft SQL Server on Linux using Kubernetes objects. The deployment goals are:

• Deploy SQL Server 2017 on Linux.
• Persist data to /var/mssql on the host.
• Set the SA password to 123456abcD.
• Expose port 1433 using NodePort 30338.

The YAML definition includes a Deployment and a Service:

apiVersion: apps/v1
kind: Deployment
metadata:
  name: mssql
  labels:
    app: mssql
spec:
  replicas: 1
  selector:
    matchLabels:
      app: mssql
  template:
    metadata:
      labels:
        app: mssql
    spec:
      containers:
        - name: mssql
          image: mcr.microsoft.com/mssql/server:2017-latest-ubuntu
          imagePullPolicy: Always
          ports:
            - containerPort: 1433
          env:
            - name: ACCEPT_EULA
              value: "Y"
            - name: SA_PASSWORD
              value: "123456abcD"
          resources:
            limits:
              cpu: "2"
              memory: 2096Mi
            requests:
              cpu: 100m
              memory: 827Mi
          volumeMounts:
            - name: data-vol
              mountPath: /var/opt/mssql/
      volumes:
        - name: data-vol
          hostPath:
            path: /var/mssql
---
apiVersion: v1
kind: Service
metadata:
  name: mssql
spec:
  ports:
    - name: tcp-1433-1433
      port: 1433
      targetPort: 1433
      nodePort: 30338
      protocol: TCP
  selector:
    app: mssql
  type: NodePort

Apply the configuration using:

kubectl apply -f mssqlserver.yaml

Check the deployment status with:

kubectl get svc -o wide -l app=mssql
kubectl get po -o wide -l app=mssql
kubectl get deployment -o wide -l app=mssql

Once the deployment is ready, you can connect to the SQL Server instance using a database management tool.