Description of problem:
pod network connectivity metrics are being publish with component
specific names. This makes is awkward to aggregate metrics across components.
For example, openshift_apiserver_endpoint_check_count and openshift_kube_apiserver_endpoint_check_count should be a single metric such as pod_network_connectivity_check_count, with a label to indicate the component. This better models the best practices already in use for other metrics which are common to multiple components.
Tried OCP 4.6.0-0.nightly-2020-09-02-210353, open the web console-> Monitoring -> Metrics, query keywords: pod_network_connectivity_check_count, pod_network_connectivity_check_tcp_connect_latency_gauge and pod_network_connectivity_check_dns_resolve_latency_gauge. the component label of queried results only displayed openshift-apiserver，not found the openshift-kube-apiserver, is this the expected result?
Since the bug only related to PR https://github.com/openshift/cluster-kube-apiserver-operator/pull/933, it works as https://bugzilla.redhat.com/show_bug.cgi?id=1874521#c3. So move the bug verified.