Create HPA-Controller.md

[udaybambal]

Continuing with respect to PR 20 and using same Apache based webpage.

Kubernetes HPA Controller (Horizontal Pod Autoscaler) on Minikube Cluster

In this demo, we will see how to deploy HPA controller. HPA will automatically scale the number of pods based on CPU utilization.

Pre-requisites to implement this project:

• Create 1 virtual machine on AWS with 2 CPU, 4GB of RAM (t2.medium)
• Setup minikube on it Minikube setup.
• Ensure you have the Metrics Server installed in your Minikube cluster to enable HPA. If not already installed, you can install it using:

minikube addons enable metrics-server

• Check minikube cluster status and nodes :

minikube status
kubectl get nodes

What we are going to implement:

• In this demo, we will create an deployment & service files for Apache and with the help of HPA, we will automatically scale the number of pods based on CPU utilization.

Steps to implement HPA:

Update the Deployments:

We'll modify the existing Apache deployment YAML files to include resource requests and limits. This is required for HPA to monitor CPU usage.

#apache-deployment.yaml
apiVersion: apps/v1
kind: Deployment
metadata:
  name: apache-deployment
spec:
  replicas: 1
  selector:
    matchLabels:
      app: apache
  template:
    metadata:
      labels:
        app: apache
    spec:
      containers:
      - name: apache
        image: httpd:2.4
        ports:
        - containerPort: 80
        resources:
          requests:
            cpu: 100m
          limits:
            cpu: 200m
---
apiVersion: v1
kind: Service
metadata:
  name: apache-service
spec:
  selector:
    app: apache
  ports:
    - protocol: TCP
      port: 80
      targetPort: 80
  type: ClusterIP

Apply the updated deployments:

kubectl apply -f apache-deployment.yaml

Create HPA Resources

We will create HPA resources for both Apache and NGINX deployments. The HPA will scale the number of pods based on CPU utilization.

#apache-hpa.yaml
apiVersion: autoscaling/v2beta2
kind: HorizontalPodAutoscaler
metadata:
  name: apache-hpa
spec:
  scaleTargetRef:
    apiVersion: apps/v1
    kind: Deployment
    name: apache-deployment
  minReplicas: 1
  maxReplicas: 5
  metrics:
  - type: Resource
    resource:
      name: cpu
      target:
        type: Utilization
        averageUtilization: 20

Apply the HPA resources:

kubectl apply -f apache-hpa.yaml

• port forward to access the Apache service on browser.

  kubectl port-forward svc/apache-service 8081:80 --address 0.0.0.0 &

Verify HPA

You can check the status of the HPA using the following command:

kubectl get hpa

This will show you the current state of the HPA, including the current and desired number of replicas.

Stress Testing

To see HPA in action, you can perform a stress test on your deployments. Here is an example of how to generate load on the Apache deployment using 'BusyBox':

kubectl run -i --tty load-generator --image=busybox /bin/sh

Inside the container, use 'wget' to generate load:

while true; do wget -q -O- http://apache-service.default.svc.cluster.local; done

This will generate continuous load on the Apache service, causing the HPA to scale up the number of pods.