Sippy indicates that kube-apiserver is degraded 100% of the time after attempting minor metal-ipi upgrades since roughly the time the Kube 1.30 rebase landed.

Failing job runs can be found here, but this failure appears to match the signature of most all of them:

https://prow.ci.openshift.org/view/gs/test-platform-results/logs/periodic-ci-openshift-release-master-nightly-4.17-upgrade-from-stable-4.16-e2e-metal-ipi-upgrade-ovn-ipv6/1796332443565821952

The key test is:

upgrade: Operator upgrade kube-apiserver

{Failed to upgrade kube-apiserver, operator was degraded (GuardController_SyncError): GuardControllerDegraded: Missing PodIP in operand kube-apiserver-master-2.ostest.test.metalkube.org on node master-2.ostest.test.metalkube.org  }