hpa-operator

The hpa-operator will will create automatically Horizontal Pod Scaling base on Annotations in spec.metadata in Deployment. The annotations are

hpa.apixio.com/min -> minimum number of replicas
hpa.apixio.com/max -> maximum number of replicas 
hpa.apixio.com/template -> name of hpa template you want to use
hpa.apixio.com/behavior -> scaling behavior name, use default if not defined

How it work

Simply checking Deployment by using Deployment Informer and Lister. If deployment has the valid annotations, it will create or update HPA with the same name as Deployment Name.
When Deployment is deleted, it also deleted either.
HPA Templates were stored in ConfigMap that was mounted to hpa-operator as volume at /template

Installing

HPA Template configmap

kubectl create -f https://raw.githubusercontent.com/zduymz/hpa-operator/master/examples/hpatemplate.yaml

HPA Operator with RBAC

kubectl create -f https://raw.githubusercontent.com/zduymz/hpa-operator/master/examples/hpa-rbac.yaml

HPA Operator without RBAC

kubectl create -f  https://raw.githubusercontent.com/zduymz/hpa-operator/master/examples/hpa.yaml

Testing

Create nginx deployment

kubectl create -f https://raw.githubusercontent.com/zduymz/hpa-operator/master/examples/nginx.yaml

Send requests

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

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

Build yourself

You need to install golang >= 1.20.

# Install requirements package
go mod tidy
# build on macos
make macos
# build on linux
make linux

Note

We have some environment variables

K8S_MASTER: k8s master url (automatically look up if empty)
K8S_CONFIG: kubeconfig file (automatically look up if empty)
HPA_TEMPLATES: directory contain hpa templates (default is /template/)