Description of problem: OpenShift liveness/readiness probes are failing after reboot of an OpenStack compute node (and therefore all OpenShift worker/master nodes running on that compute node). After the OpenShift nodes were up again (on the same OpenStack compute node) all liveness and readiness probes failed until the Pods got deleted/recreated manually. Tcpdump (before deleting the Pods) shows that packets are leaving the OpenShift node towards the OpenStack Router but nothing is sent to the Pods. Following log entries can be found in the kuryer-controller Pod after the reboot: WARNING kuryr_kubernetes.controller.drivers.utils [-] Port <Port ID> is in DOWN status but still associated to a trunk. This should not happen. Trying to delete it from the trunk.: openstack.exceptions.ConflictException: ConflictException: 409: Client Error for url: https://<openstack api>:13696/v2.0/ports/<Port ID>, Port <Port ID> is currently a subport for trunk <Trunk ID>.ESC[00m Version-Release number of selected component (if applicable): OpenStack v16.1.7 OpenShift v4.8.27
The needinfo request[s] on this closed bug have been removed as they have been unresolved for 120 days