Description of problem:

Currently when running an IPI on Baremetal deployment, when the installer creates the bootstrap VM, via the command journalctl -f -u bootkube.service will show every second an error similar to:

Apr 24 18:52:05 localhost bootkube.sh[3385]: "99_openshift-machineconfig_99-master-ssh.yaml": unable to get REST mapping for "99_openshift-machineconfig_99-master-ssh.yaml": no matches for kind "MachineConfig" in version "machineconfiguration.openshift.io/v1"

This keeps polling forever until the master nodes are up and get the MCO. It is not necessarily an error but to the untrained eye it can seem like an error. Instead it be great if either these errors were shown only after a given period of time, or during some point of the install were checking for this would indeed be warranted. 


How reproducible: 100% , in versions 4.3 and above

Steps to Reproduce:
1. Install IPI on BM
2. Log into the bootstrap VM
3. Check out the bootstrap VM bootkube.service journalctl

Actual results:
Flooding of 

"99_openshift-machineconfig_99-master-ssh.yaml": unable to get REST mapping for "99_openshift-machineconfig_99-master-ssh.yaml": no matches for kind "MachineConfig" in version "machineconfiguration.openshift.io/v1"

making it harder to see if anything else is happening or of concern. 

Expected results:

Only poll periodically or trigger something in the installer of when this indeed should be checked.