TODO VPC Peering

TODO Helm Chart

    Download helm chart
    remove other edge&firefox&node's from helm chart

    check this repo for scaling pods {
        https://github.com/SeleniumHQ/docker-selenium/issues/1688
        https://github.com/prashanth-volvocars/docker-selenium/blob/auto-scaling/charts/selenium-grid/values.yaml
        }

TODO kube ok

TODO

Keda

selenium autoscaler

    install https://keda.sh/ plugin  in kube
    helm repo add kedacore https://kedacore.github.io/charts
    kubectl create namespace keda
    helm install keda kedacore/keda --namespace keda
    helm install -f values.yaml docker-selenium

NOTES

    try graphql  or /status

Autoscaling

    pods are scale-in and scale-out by keda,
    nodes are scaled out and scaled-in by autoscaling group by aws
    By how individual grid isolated components scale's by default it have only 1 replica enabled in helm
    Increase resource size for pod's