ArgoCD pods fail to schedule in VKS Cluster. Warnings in "kubectl describe pod" output show these errors or similar:
Warning FailedScheduling 2m51s (x33 over 13m) default-scheduler DRS cannot find a host to power on or migrate the virtual machine.
Additionally, on the vCenter side you might see issues with deploying OVF templates due to a datastore that has been "deleted or not been completely created".
ESXi hosts are configured with swap file locations that are referencing invalid or nonexistent datastores. This can happen even if configured to be set at default same location as the virtual machine's configuration file.
Update swap file locations for the hosts to use specific datastores that are known to work.