Utilizing high availability scalability high performance using kubernetes Ingress Nginx And Hpa

the steps:

#i use docker desktop that acts as a cluster: to act as a handler for public domain of the app)
load balancer (cluster) -> ingress <-> hpa <-> the application in pods
# hpa: horizontal pod autoscaling

1. Install the app you want to scale
2. Make the Dockerfile
3. Build it, for example:
   docker build -t shohiebsense/network-traffic-monitor-8085:latest .
4. Push it to harbor/registry or whatever, you can build private harbor/registry tho but i use docker hub https://hub.docker.com/
   docker push shohiebsense/network-traffic-monitor-8085:latest
5. Install the metrics, tools to monitor the hpa.
   kubectl apply -f https://github.com/kubernetes-sigs/metrics-server/releases/latest/download/components.yaml
6. Check if metrics installed
   kubectl get deployment metrics-server -n kube-system
   6a. log if it’s running
   kubectl logs -n kube-system deployment/metrics-server
7. Apply the deployment, load_balancer, and ingress (.yaml file).
   kubectl apply -f deployment.yaml
kubectl apply -f ingress.yaml
kubectl apply -f load_balancer.yaml
8. You can use bash script or yaml file to do autoscaling kubectl autoscale deployment network-traffic-monitor-deployment-8085 --cpu-percent=3 --min=2 --max=5
9. using .yaml

   apiVersion: autoscaling/v2
   kind: HorizontalPodAutoscaler
   metadata:
     name: network-traffic-monitor-hpa-8085
   spec:
     scaleTargetRef:
   apiVersion: apps/v1
   kind: Deployment
   name: network-traffic-monitor-deployment-8085
     minReplicas: 2
     maxReplicas: 7
     metrics:
     - type: Resource
   resource:
     name: cpu
     target:
       type: Utilization
       averageUtilization: 3  # Lower the CPU target to 3%, it's small i know, to see the scaling by generating new pods in action
   

10. Test your app, I use locust
11. Monitor your pods, can use watch or k9s

    watch kubectl get pods -o wide
    

12. https://k9scli.io/

Repo:
https://github.com/shohiebsense/kubernetes_8085-autoscaler-hpa-ingress-nginx