[sig-node] Managed cluster should report ready nodes the entire duration of the test run [Late] [Suite:openshift/conformance/parallel] is failing frequently in CI, see: https://sippy.ci.openshift.org/sippy-ng/tests/4.7/analysis?test=%5Bsig-node%5D%20Managed%20cluster%20should%20report%20ready%20nodes%20the%20entire%20duration%20of%20the%20test%20run%20%5BLate%5D%20%5BSuite%3Aopenshift%2Fconformance%2Fparallel%5D This is affecting the 4.7 release quite a lot, here are some recent failures: * https://prow.ci.openshift.org/view/gs/origin-ci-test/logs/periodic-ci-openshift-release-master-ci-4.7-e2e-aws-serial/1471691294148399104 * https://prow.ci.openshift.org/view/gs/origin-ci-test/logs/periodic-ci-openshift-release-master-ci-4.7-e2e-aws-serial/1471547869029732352 * https://prow.ci.openshift.org/view/gs/origin-ci-test/logs/periodic-ci-openshift-release-master-ci-4.7-e2e-aws-serial/1471462336106598400 * https://prow.ci.openshift.org/view/gs/origin-ci-test/logs/periodic-ci-openshift-release-master-ci-4.7-e2e-aws-serial/1471394344035422208 According to Vadim, this is caused by "its apiserver racing and doesn't apply RBs for SCCs fast enough". Here's a slack thread for more discussion: https://coreos.slack.com/archives/CJARLA942/p1639737113130300
Even if we aren't going to address this on the product side, we don't want 4.7 CI release blocking jobs failing consistently, so we need to address this somehow. David found [1] improving the PromQL in 4.8, so we'll backport that to 4.7's origin suite. [1]: https://github.com/openshift/origin/pull/25934