Looks like we are facing a bug when trying to spin up a hosted control plane cluster while using proxy settings to connect to the internet. For example, on our worker node, the static pod kube-apiserver-proxy.yaml doesn't contain the noProxy settings which seem to cause the failure of deploying the hosted cluster.

~~~
[root@ocpugbo2cogswo03 manifests]# cat kube-apiserver-proxy.yaml_
apiVersion: v1
kind: Pod
metadata:
  creationTimestamp: null
  labels:
    k8s-app: kube-apiserver-proxy
  name: kube-apiserver-proxy
  namespace: kube-system
spec:
  containers:
  - command:
    - control-plane-operator
    - kubernetes-default-proxy
    - --listen-addr=<IP-Addr>:6443
    - --proxy-addr=<Proxy-Addr>:<Proxy-port>
    - --apiserver-addr=<API-IP-Addr>:6443
    image: quay.io/openshift-release-dev/ocp-v4.0-art-dev@sha256:7ca95b9a71e41157c70378896758618b993ad90e6d80a23c46170da5c11f441f
    name: kubernetes-default-proxy
    resources:
      requests:
        cpu: 13m
        memory: 16Mi
    securityContext:
      runAsUser: 1001
  hostNetwork: true
  priorityClassName: system-node-critical
status: {}
~~~

Can you please check this issue.

    1. Install a cluster with ACM and HCP
    2. Try to create a hosted cluster using proxy configuration
    3. kube-apiserver-proxy is using proxy to reach API.

    The kube-apiserver-proxy is using proxy to reach API. Worker nodes are unable to reach a Hosted Control Plane's API when a cluster-wide http proxy is configured.

    kube-apiserver-proxy should not use proxy to reach API