This guide walks through deploying a containerized HTTP service on a Kubernetes cluster, covering container images, deployments, services, and ingres configuration.

Building and Pushing the Container Image

First, build the container image for your HTTP server and push it to a container registry. In this example, the image chengfengsunce/httpserver:0.0.1 is pushed to Docker Hub.

Creating the Deployment

A Deployment manages replica pods and ensures the desired number of instances remain running. Below is the YAML configuration:

apiVersion: apps/v1
kind: Deployment
metadata:
  name: httpserver
  labels:
    app: httpserver
spec:
  replicas: 3
  selector:
    matchLabels:
      app: httpserver
  template:
    metadata:
      labels:
        app: httpserver
    spec:
      containers:
      - name: httpserver
        image: chengfengsunce/httpserver:0.0.1
        imagePullPolicy: IfNotPresent
        ports:
        - containerPort: 80
        livenessProbe:
          httpGet:
            path: /healthz
            port: 8080
            scheme: HTTP
          initialDelaySeconds: 5
          periodSeconds: 10
        readinessProbe:
          httpGet:
            path: /healthz
            port: 8080
            scheme: HTTP
          initialDelaySeconds: 5
          periodSeconds: 10
        resources:
          requests:
            cpu: 20m
            memory: 20Mi
          limits:
            cpu: 200m
            memory: 100Mi

Deploy it using:

kubectl create -f deployment.yaml

Verify the Deployment status:

kubectl get deployment

Creating the Service

A Service provides network acess to the pod group. Create a Service resource for internal cluster communication:

apiVersion: v1
kind: Service
metadata:
  name: product-server
spec:
  ports:
  - name: http
    port: 80
    targetPort: 80
  - name: grpc
    port: 81
    targetPort: 81
  selector:
    app: product-server
  type: ClusterIP

Apply and verify:

kubectl create -f service.yaml
kubectl get service

Sample output:

NAME             TYPE        CLUSTER-IP   EXTERNAL-IP   PORT(S)           AGE
product-server   ClusterIP   10.107.6.57  <none>       80/TCP,81/TCP     6h10m

Tip: Deployment and Service definitions can be combined in a single YAML file separated by ---.

Configuring Ingress for External Access

To expose services externally, install and configure the ingress-nginx controller. Apply the ingress controller deployment:

kubectl create -f ingress-nginx.yaml

Check the controller pod status:

kubectl get pods -n ingress-nginx

Create a Ingress resource to route external traffic:

apiVersion: networking.k8s.io/v1beta1
kind: Ingress
metadata:
  name: nginx-web
  annotations:
    nginx.ingress.kubernetes.io/rewrite-target: /
spec:
  rules:
  - host: product.com
    http:
      paths:
      - path: /
        backend:
          serviceName: product-server
          servicePort: 80
      - path: /grpc
        backend:
          serviceName: product-server
          servicePort: 81

Deploy and verify:

kubectl create -f ingress.yaml
kubectl get ingress

Sample output:

NAME HOSTS ADDRESS PORTS AGE nginx-web product.com 80 8h

<h2>Verifying Self-Healing Behavior</h2>

<p>Kubernetes automatically replaces failed pods. List all pods in the default namespace:</p>

<code>kubectl get pods</code>

<p>Delete a specific pod to test recovery:</p>

<code>kubectl delete pods product-server-599cfd85cc-ppmhx</code>

<p>Immediately check pod status again:</p>

<code>kubectl get pods</code>

<p>The deleted pod shows terminating status while a new one spawns to maintain the replica count. The Deployment controller detects the pod loss and schedules a replacement, demonstrating the self-healing capability of Kubernetes.</p>