This message was deleted.

my other option is to recreate a second cluster with RKE2 for instance, manage it with rancher, and just move the workload to that. Would it be a preferred aproach?

If you want to have Rancher deployed on top of your microk8s you can do it. At the end of the day Rancher Management is just a helm chart that runs on top of any Kubernetes cluster given that the Kubernetes version is supported. When you do that, the cluster where Rancher Management is deployed becomes the "local" cluster (local is an especial name to designated the cluster where Rancher is deployed) and you can manage it. If your microk8s cluster is Highly Available, Rancher will be highly available (it uses several replicas and uses the underlying etcd as storage). May that mess up with your exiting SAs and Role Bindings? I don't think so (but I'm not 100% sure). Connecting with Google SSO is better done after Rancher is deployed, as you can leverage the existing integration.

thanks for the confirmation.