Hello, I have Rancher and a downstream RKE2 cluster. Rancher is installed on 6 nodes (a k3s cluster), where 3 control planes are recorded in Cloudflare as the ingress-nginx-controller. The issue occurs when I run kubectl commands on the downstream RKE2 cluster. I performed a PostgreSQL restore in the downstream RKE2 cluster using kubectl, where the kubeconfig context endpoint is pointing to my Rancher website. The restore process got stuck, and Several Rancher nodes became unresponsive / experienced downtime, and I was unable to access the Rancher web UI. Does anyone have suggestions on how to handle this?