Configure Kubernetes Horizontal Pod Autoscaler

Enable metrics plugin on Minikube

minikube addons enable metrics-server

Start a deployment and expose it as a service

kubectl apply -f https://k8s.io/examples/application/php-apache.yaml

Create Horizontal Pod Autoscaler

kubectl autoscale deployment php-apache --cpu-percent=50 --min=1 --max=10 

This command creates an HPA with a minimum of one pod and maximum of ten pods running. It will try to keep the averga usage of the pods at 50% and if there is more usage, it will automatically create more pods to keep the usage at 50%.

To test it, run the following on a different terminal

  kubectl run -it --rm load-generator --image=busybox /bin/sh

Then, to create the trafic neede for the HPA start working.

**while true; do wget -q -O- http://php-apache; done**

Back on the first terminal, monitor the HPA with

kubectl get hpa -w

To have more control over the HPA we can also create it with a mainfest file, the equivalent to the command kubectl autoscale deployment php-apache --cpu-percent=50 --min=1 --max=10 would be:

apiVersion: autoscaling/v2beta2
kind: HorizontalPodAutoscaler
metadata:
  name: php-apache
spec:
  scaleTargetRef:
    apiVersion: apps/v1
    kind: Deployment
    name: php-apache
  minReplicas: 1
  maxReplicas: 10
  metrics:
  - type: Resource
    resource:
      name: cpu
      target:
        type: Utilization

HPA can also be configured depending on the cpu usage or requests recieved, it is a good idea save it as a file to keeep record of the configuration for the HPA.

Reference Links

• https://kubernetes.io/docs/tasks/run-application/horizontal-pod-autoscale/