Description of problem:
openshift-apiserver pod logs have numerous TLS handshake error.*EOF logs from 10.128.0.1, 10.129.0.1, 10.130.0.1

Version-Release number of selected component (if applicable):
4.6.0-0.nightly-2020-07-14-035247

How reproducible:
Always

Steps to Reproduce:
1. Launched two envs:
One is ipi on aws with fips & etcd encryption, the other is ipi on gcp with http proxy
2. Check the openshift-apiserver pod logs

Actual results:
2. Both envs found numerous TLS handshake error from.*EOF logs, all are from 10.128.0.1, 10.129.0.1, 10.130.0.1.

$ grep "TLS handshake error from.*EOF" apiserver-6f4fdfcc87-{ggxt9,k26lt,tdm2d}.log | wc -l
185160
$ grep "TLS handshake error from.*EOF" apiserver-6f4fdfcc87-{ggxt9,k26lt,tdm2d}.log | tail -n 3
apiserver-6f4fdfcc87-tdm2d.log:I0715 11:21:31.739521       1 log.go:181] http: TLS handshake error from 10.128.0.1:35350: EOF
apiserver-6f4fdfcc87-tdm2d.log:I0715 11:21:31.941917       1 log.go:181] http: TLS handshake error from 10.130.0.1:40358: EOF
apiserver-6f4fdfcc87-tdm2d.log:I0715 11:21:32.122470       1 log.go:181] http: TLS handshake error from 10.129.0.1:57434: EOF
$ grep "TLS handshake error from.*EOF" apiserver-6f4fdfcc87-{ggxt9,k26lt,tdm2d}.log | grep -v -e 10.128.0.1 -e 10.129.0.1 -e 10.130.0.1
# none

Expected results:
2. Should have no so many TLS handshake error from.*EOF logs.

Additional info:
kube-apiserver pods don't have these logs.

# below shows 10.128.0.0/14, not sure if related to 10.128.0.1, 10.129.0.1, 10.130.0.1
oc get cm/cluster-config-v1 -n kube-system -o yaml
...
    networking:
      clusterNetwork:
      - cidr: 10.128.0.0/14
        hostPrefix: 23
...