Description of problem: openshift-dns fails to start with: Sep 17 17:39:43.716 W ns/openshift-dns pod/dns-default-tvnxp network is not ready: runtime network not ready: NetworkReady=false reason:NetworkPluginNotReady message:Network plugin returns error: Missing CNI default network (11 times) Version-Release number of selected component (if applicable): How reproducible: Appears in 4.2 nightly e2e runs Additional info: you can see it show up in this run https://prow.svc.ci.openshift.org/view/gcs/origin-ci-test/logs/release-openshift-ocp-installer-e2e-aws-serial-4.2/541 in the event logs (click monitor cluster while tests execute and then click open stdout)
Did this actually fail? DNS has tolerations, so it's scheduled on notready pods. But it should start up. Doug, did you see a degraded operator? Otherwise this is NOTABUG since it's a transient state.
Yeah, the DNS pod has a NotReady toleration. So this error is expected. Once the network is up, it starts just fine. The "fix" is either to accept this error as ignorable or remove the toleration.
Kicking over to Network Edge to decide whether or not they want to remove the toleration.
Thanks Casey. I think we should talk about removing the toleration to smooth things out unless there's some reason for the current behavior. Moving to 4.3 because we aren't going to block the release unless the transient state has some serious downstream effect.
verified with 4.3.0-0.nightly-2019-11-06-230519 and didn't find the issue.
Since the problem described in this bug report should be resolved in a recent advisory, it has been closed with a resolution of ERRATA. For information on the advisory, and where to find the updated files, follow the link below. If the solution does not work for you, open a new bug report. https://access.redhat.com/errata/RHBA-2020:0062