Hi, i am trying to upgrade a rancher-managed cluster from k8s 1.20.6 to 1.20.11. Since 2h one of the worker nodes is in cordoned state and cluster state is "updating". Any idea where i can look what the problem might be?