FailedScheduling event in specific environments with two replicas

1. Proposed title of this feature request
FailedScheduling event in specific environments with two replicas

2. What is the nature and description of the request?
This is a follow-up of RHBZ #2024173 where it was found that in specific situation a FailedScheduling during the Deployment update as no suitable OpenShift Container Platform - Node is found.

This is mostly happening when using 3 infra OpenShift Container Platform - Node(s), which is recommended and documented in https://docs.openshift.com/container-platform/4.9/machine_management/creating-infrastructure-machinesets.html

While the Teams are following https://github.com/openshift/enhancements/blob/master/CONVENTIONS.md#high-availability it seems that this could be optimized to prevent FailedScheduling events from happening and make this a more convenient experience.

3. Why does the customer need this? (List the business requirements here)
Customers are concerned that a Red Hat recommendation can cause transient FailedScheduling event. Even though there is no impact, they are wondering why Red Hat would provide configuration or Deployments that are not able to update without causing any failure event.

4. List any affected packages or components.
kubernetes