WorkerPool.status.selector

Selector The selector used by KEDA for autoscaling. The name of the label in the selector is just here to act as a mock, it is currently not used. It looks like, when using external metrics with an HPA, the HPA controller will first check if the label selector exists. Then it will check to make sure no other HPA has the same selector. Finally, as long as those two conditions are true, it will scale based on the external metric and won't use the selector for actually selecting pods (unless you tell it too, which with KEDA we don't). The selector is in the format workers.spacelift.io/not-used={WorkerPoolID} so that if a customer has multiple worker pools the HPA won't fail based on the second check (it will make sure no other HPA has the same selector). We don't actually need to add this selector to any pods because the HPA will not use it in our case with an external metric. See more details in this thread https://github.com/spacelift-io/kube-workerpool-controller/pull/122