auto-scale kubernetes deployments based on arbitrary datadog metric-series.
- Configure a k8s job to run at some interval, per k8s deployment, ie: 5 minutes.
- Set the job env vars for configuration
note: it currently depends on having a kubeconf locally available. I will update it eventually, to use in-cluster authentication
run deps.sh
./deps.sh
go get k8s.io/client-go/...
go get -u k8s.io/apimachinery/...
export GOPATH=$(pwd)
go build