User Story

As a user I want the ability to specify the prefix for the cluster autoscaler min/max annotations so that I can utilize the autoscaler in OpenShift or Cluster-API environments.

Background

We carry a commit[0] in our OpenShift branch of the cluster autoscaler which hard codes the values for the minimum and maximum machineset annotations. To help enable the development of Cluster-API on OpenShift it would be better if we could set the prefix for these annotations dynamically at launch through an environment variable. We have done something similar with the resource group for the machinesets, see the `CAPI_GROUP` variable. This variable should be used for the min/max annotations as well.

 [0] https://github.com/openshift/kubernetes-autoscaler/pull/209/commits/20e2e2648595e529e6c9ff5637d3f05a860234cf

Steps

• Modify the min/max annotations to use the `CAPI_GROUP` environment variable to override the prefix when specified

Stakeholders

• hypershift team
• cloud team

Definition of Done

• min/max annotation prefixes can be overridden with CAPI_GROUP environment variable

• Docs

• update the README (in the cloudprovider/clusterapi directory) to indicate the additional behavior. There is a section talking about the CAPI_GROUP variable and its usage.

• Testing

• the unit tests should be updated to test the new behavior