Surfaces in tests:
[sig-node] node-lifecycle detects unexpected not ready node
[sig-node] node-lifecycle detects unreachable state on node

https://prow.ci.openshift.org/view/gs/test-platform-results/logs/periodic-ci-openshift-release-master-ci-4.18-e2e-azure-ovn-techpreview-serial/1858341879754526720

In this job, we see:

node/ci-op-jxk5kmn5-d8a6e-dgknq-master-2 - reason/UnexpectedUnreachable unexpected node unreachable at from: 2024-11-18 05:39:24.934825905 +0000 UTC m=+4531.474291883 - to: 2024-11-18 05:39:24.934825905 +0000 UTC m=+4531.474291883}
 
15 seconds later we also see an event about api unreachable.

We see similar problems in https://prow.ci.openshift.org/view/gs/test-platform-results/logs/periodic-ci-openshift-release-master-ci-4.18-e2e-azure-ovn-upgrade/1859160753831940096

In this job, we see:

node/ci-op-c53845j5-431b2-9596s-worker-centralus2-2k8sd - reason/UnexpectedNotReady unexpected node not ready at from: 2024-11-20 10:50:50.757173918 +0000 UTC m=+601.172900086 - to: 2024-11-20 10:50:50.757173918 +0000 UTC m=+601.172900086}

And at 10:50, we see kubelet lease failures and apiserver unavailability. We would like to narrow down the issue between kubelet, apiserver and the load-balancer.