Description of problem:

It seems we have a very high terminationGracePeriodSeconds on Windows VMs templates.

For example:

1. oc get template windows10-desktop-medium -n openshift -o yaml | grep -i grace
   terminationGracePeriodSeconds: 3600

1. oc get template windows2k19-highperformance-large -n openshift -o yaml | grep -i grace
   terminationGracePeriodSeconds: 3600

While RHEL has a more reasonable 180:

1. oc get template rhel8-server-medium -n openshift -o yaml | grep -i grace
   terminationGracePeriodSeconds: 180

This is annoying if the Virtual Machine fails to shutdown gracefully. The user needs to interneve manually or wait 1h.

Version-Release number of selected component (if applicable):
4.9.0

How reproducible:
100%

Steps to Reproduce:
1. Create Windows Virtual Machine using some random bytes or corrupted disk file, or any unbootable image, so it gets stuck on SeaBIOS and won't handle shutdown signal.
2. Start VM
3. Stop or Delete the VM

Actual results:

• Pod stuck in terminating for 1h

Expected results:

• More reasonable timeout, so user does not need to intervene.