For windows VMs there is a special PVC attached to the VM, prefixed by "persistent-state-for-", which contains the UEFI/TPM data for the VM.
If the virt-launcher pod is unable to mount and attach this volume (for whatever reason), this error is not reflected in the VM's status.conditions, but only in the virt-launcher pod's events.
Since this crucial info is not visible in the diagnostics tab (which is taken from the VM's conditions), the user sees the VM is "Starting" forever without any clue what is wrong, since all they can see is:

  conditions:
    - lastProbeTime: '2025-12-30T10:45:29Z'
      lastTransitionTime: '2025-12-30T10:45:29Z'
      message: Guest VM is not reported as running
      reason: GuestNotRunning
      status: 'False'
      type: Ready
    - lastProbeTime: null
      lastTransitionTime: null
      message: All of the VMI's DVs are bound and ready
      reason: AllDVsReady
      status: 'True'
      type: DataVolumesReady

all versions

Always

1. Create a windows virtual machine
2. cause a storage issue in the persistent-state-for-* PVC that will prevent it to be mounted on the virt-launcher pod.
3. observe the VM's conditions / diagnostics tab in the UI

No visible errors shown, although the VM is stuck at "Starting" forever

A clear error should be shown, suggesting the PVC/volume that can't be mounted on the pod, therefore the pod can't start (stuck at Init:0/1).
There should be a condition that is telling the user that there is a volume configured on the VM that can't be attached/mounted.

A reproducer is available at:
https://console-openshift-console.apps.bleeding.cnv.engineering.redhat.com
namespace: rsdeor
vm name: windows-2k22-virtio-maroon-porpoise-52