Description of problem:

When installing using proxy + externalDNS, authentication cluster operator never gets available.

Logs from authentication-operator pod:

E0619 13:51:20.270807       1 controller.go:129] {AuthenticationOperator2 AuthenticationOperator2} failed with: error checking current version: unable to check route health: failed to GET route: dial tcp 10.0.111.196:443: i/o timeout

10.0.111.196 IP address is the API floating ip:

[morenod@morenod-laptop ~]$ openstack floating ip list --long | grep 10.0.111.196
| d7bb6c14-82e0-45d6-9e66-875b6bb8d72e | 10.0.111.196        | 192.168.0.5      | 6e88657a-e418-40e5-8631-5a43ab54e70a | 316eeb47-1498-46b4-b39e-00ddf73bd2a5 | 542c6ebd48bf40fa857fc245c7572e30 | a4684936-c0d0-491b-b623-7659ba1ea501 | None   | preserve mrnd-13-46-px                                     | []   | None     | None       |
[morenod@morenod-laptop ~]$ openstack port list | grep mrnd-13-46-px | grep 192.168.0.5
| 6e88657a-e418-40e5-8631-5a43ab54e70a | mrnd-13-46-px-l6tr2-api-port       | fa:16:3e:c7:79:f7 | ip_address='192.168.0.5', subnet_id='8f3467f5-2771-4f1c-af5f-656ea1ee1657'                      | DOWN   |
[morenod@morenod-laptop ~]$ 


Compared with an installation without proxy or externalDNS, check done on controller.go:129 returns an EOF, not a timeout:

E0619 13:00:51.627750       1 controller.go:129] {AuthenticationOperator2 AuthenticationOperator2} failed with: error checking current version: unable to check route health: failed to GET route: EOF

This seems to be controller, as operator gets available even with the EOF error:

I0619 13:00:53.430840       1 event.go:278] Event(v1.ObjectReference{Kind:"Deployment", Namespace:"openshift-authentication-operator", Name:"authentication-operator", UID:"9dbd31c7-0d3b-4e8e-9360-0b4774253913", APIVersion:"apps/v1", ResourceVersion:"", FieldPath:""}): type: 'Normal' reason: 'OperatorStatusChanged' Status for clusteroperator/authentication changed: Degraded changed from True to False ("RouteHealthDegraded: failed to GET route: EOF")
I0619 13:01:01.413373       1 status_controller.go:172] clusteroperator/authentication diff {"status":{"conditions":[{"lastTransitionTime":"2020-06-19T13:00:53Z","reason":"AsExpected","status":"False","type":"Degraded"},{"lastTransitionTime":"2020-06-19T13:01:01Z","message":"Progressing: got '404 Not Found' status while trying to GET the OAuth well-known https://192.168.3.157:6443/.well-known/oauth-authorization-server endpoint data","reason":"_WellKnownNotReady","status":"True","type":"Progressing"},{"lastTransitionTime":"2020-06-19T13:01:01Z","status":"False","type":"Available"},{"lastTransitionTime":"2020-06-19T12:49:58Z","reason":"AsExpected","status":"True","type":"Upgradeable"}]}}
I0619 13:01:01.428440       1 status_controller.go:172] clusteroperator/authentication diff {"status":{"conditions":[{"lastTransitionTime":"2020-06-19T13:00:53Z","reason":"AsExpected","status":"False","type":"Degraded"},{"lastTransitionTime":"2020-06-19T13:01:01Z","message":"Progressing: got '404 Not Found' status while trying to GET the OAuth well-known https://192.168.3.157:6443/.well-known/oauth-authorization-server endpoint data","reason":"_WellKnownNotReady","status":"True","type":"Progressing"},{"lastTransitionTime":"2020-06-19T13:01:01Z","status":"False","type":"Available"},{"lastTransitionTime":"2020-06-19T12:49:58Z","reason":"AsExpected","status":"True","type":"Upgradeable"}]}}
I0619 13:01:01.428948       1 event.go:278] Event(v1.ObjectReference{Kind:"Deployment", Namespace:"openshift-authentication-operator", Name:"authentication-operator", UID:"9dbd31c7-0d3b-4e8e-9360-0b4774253913", APIVersion:"apps/v1", ResourceVersion:"", FieldPath:""}): type: 'Normal' reason: 'OperatorStatusChanged' Status for clusteroperator/authentication changed: Degraded message changed from "RouteHealthDegraded: failed to GET route: EOF" to "",Progressing changed from Unknown to True ("Progressing: got '404 Not Found' status while trying to GET the OAuth well-known https://192.168.3.157:6443/.well-known/oauth-authorization-server endpoint data"),Available changed from Unknown to False ("")

Version-Release number of the following components:
4.6.0-0.nightly-2020-06-19-051412

How reproducible:

Steps to Reproduce:
1.Install IPI on OSP using proxy and externalDNS
2.Check status of authentication cluster operator
3.

Actual results:
Cluster operators authentication and console (dependency from authentication) are not getting Available, making the installation failed

Expected results:
Authentication cluster operator captures the timeout as it does with the EOF error and continues its initialization


Additional info:
Please attach logs from ansible-playbook with the -vvv flag