Description of problem:
Openstack jobs are passing at a rate of 16%, most of the failing tests seem to be having connection issues, e.g.:

# [sig-instrumentation][Late] Alerts shouldn't report any alerts in firing state apart from Watchdog and AlertmanagerReceiversNotConfigured [Suite:openshift/conformance/parallel]
fail [github.com/openshift/origin/test/extended/util/client.go:693]: Jun  3 08:57:05.772: the server is currently unable to handle the request (get users.user.openshift.io e2e-test-prometheus-2fsc6-user)


list of occurrences:
https://search.apps.build01.ci.devcluster.openshift.com/?search=the+server+is+currently+unable+to+handle+the+request&maxAge=168h&context=1&type=junit&name=4.5.*openstack%7Copenstack.*4.5&maxMatches=5&maxBytes=20971520&groupBy=job

sample job:
https://prow.svc.ci.openshift.org/view/gcs/origin-ci-test/logs/release-openshift-ocp-installer-e2e-openstack-4.5/715


Version-Release number of selected component (if applicable):
4.5 on openstack

How reproducible:
common

Starting with api-server since the request failure seems to be trying to reach the apiserver, but certainly likely this is a network/routing/loadbalancing config issue specific to the openstack platform configuration.