Hello, I am creating a cluster in Rancher via "Cluster Management" -> Create -> Harvester. After successfully creating the cluster, I observe that one of the initial master nodes is designated as the "server" (bootstrap node) in the RKE2 configuration file, as shown below. When I scale up the master nodes and then scale down the current bootstrap node, the "server" entry in /etc/rancher/rke2/config.yaml.d/50-rancher.yaml is updated to point to a new (elected?) bootstrap node. However, if I disable the bootstrap node, the cluster becomes dysfunctional. This suggests that I have a single point of failure (SPOF), and my cluster is not highly available (HA). I understand that I can configure kube-vip to provide a single VIP address for the master nodes. My question is: where should I configure the VIP to ensure high availability? Am I missing something in my understanding? Here is the relevant configuration from /etc/rancher/rke2/config.yaml.d/50-rancher.yaml: { ... "server": "https://192.168.4.76:9345", ... }