Sounds like this issue can be reproduced in 4.10.0-0.nightly-2022-11-24-131934
when I scale up one worker by BareMetalHost
apiVersion: metal3.io/v1alpha1
kind: BareMetalHost
metadata:
name: sriov27
namespace: openshift-machine-api
spec:
online: true
bmc:
address: ipmi://10.xx.xx.xx
credentialsName: openshift-sriov27-bmc-secret
bootMACAddress: 04:3f:72:00:01:02
The worker marked as localhost
# oc get node
NAME STATUS ROLES AGE VERSION
localhost Ready worker 98m v1.23.12+8a6bfe4
master-0 Ready master 159m v1.23.12+8a6bfe4
master-1 Ready master 160m v1.23.12+8a6bfe4
master-2 Ready master 159m v1.23.12+8a6bfe4
worker-0 Ready worker 140m v1.23.12+8a6bfe4
worker-1 Ready worker 140m v1.23.12+8a6bfe4
Here is the must-gather logs and Networkmanager ovs-configuration logs
http://file.apac.redhat.com/~zzhao/bug2069754/