Description of problem:

In our documentation we state how to edit the number of CPU here:

https://docs.redhat.com/en/documentation/openshift_container_platform/4.17/html/virtualization/virtual-machines#virt-edit-vms

It then states:

Enter the desired number of vCPU sockets and click Save.

If the VM is migratable, a live migration is triggered. If not, or if the changes cannot be live-updated, a RestartRequired condition is added to the VM. 

We should create a warning box here stating thatthere are limitations to how much we can increase CPU or memory before a restart is actually required (which is not really hotplug at that point).

By default a VM is created with 4x the number of CPU "available" for hotplug, and 4x the amount of memory set as "maxGuest" 

For a 1 CPU / 2 GB small VM this translates to:

<maxMemory unit='KiB'>8388608</maxMemory>
  <memory unit='KiB'>2097152</memory>
  <currentMemory unit='KiB'>2097152</currentMemory>
  <vcpu placement='static' current='1'>4</vcpu>
  <vcpus>
    <vcpu id='0' enabled='yes' hotpluggable='no' order='1'/>
    <vcpu id='1' enabled='no' hotpluggable='yes'/>
    <vcpu id='2' enabled='no' hotpluggable='yes'/>
    <vcpu id='3' enabled='no' hotpluggable='yes'/>
  </vcpus> 

You can only hot plug up to 4 total cpu and 8GB memory without requiring a restart.

This gives a false impression to customers that all hotplug operations can be done without restart. 

We have at least one customer who can't afford downtime and didn't realize this ahead of time.

There are ways to proactively increase that ratio to allow for more flexibility if needed.

I've started a KCS here to try and address alternatives https://access.redhat.com/solutions/7121992

Ideally we'd warn customers ahead of time of this scenario this in our documentation