CU is Doing AWS UPI.

Check Logs bundle, As per this API comes up.

~~~
Jan 27 07:08:26 ip-172-38-50-24 bootkube.sh[4690]: https://localhost:2379 is healthy: successfully committed proposal: took = 11.804375ms
Jan 27 07:08:26 ip-172-38-50-24 bootkube.sh[2365]: Starting cluster-bootstrap...
Jan 27 07:08:28 ip-172-38-50-24 bootkube.sh[4759]: Starting temporary bootstrap control plane...
Jan 27 07:08:28 ip-172-38-50-24 bootkube.sh[4759]: Waiting up to 20m0s for the Kubernetes API
Jan 27 07:08:29 ip-172-38-50-24 bootkube.sh[4759]: Still waiting for the Kubernetes API: Get "https://localhost:6443/readyz": dial tcp [::1]:6443: connect: connection refused
Jan 27 07:08:57 ip-172-38-50-24 bootkube.sh[4759]: API is up

~~~

Later on, this installation failed due to some issues with multiple cluster operators failed.   

While performing AWS UPI 4.12 ( so we must-have manually created all control plane nodes) the Control Plane Machine Set operator is not supported on clusters with manually provisioned machines it clearly identifies this documentation.

"The Operator requires the Machine API to be operational and is therefore not supported on clusters with manually provisioned machines"

Refer Doc> https://docs.openshift.com/container-platform/4.12/machine_management/control_plane_machine_management/cpmso-about.html#cpmso-limitations_cpmso-about

So need to make an inactive control plane machine set during the UPI installation.