@millions-afternoon-34677 I'm new to Harvester but a tip: check if the node your VM got scheduled to has storage to host a copy of your VM data in its Longhorn. It could be that Longhorn can not host a replica of the Kubernetes Volume providing root disk to the VM, and needs to pull all data through the network. If you have like 1G network and a busy VM it can explain the long startup. As an alternative check in Longhorn which nodes have replicas and migrate the vm one of those nodes. If the warning disappears and startup is fast then it confirms the above.