Fix a race in the hybrid overlay CARRY patch causing pods to claim the management port IP address.
@ross, could you help verified this bug? thanks
Verified other-config:exclude_ips is set 4.4.0-0.nightly-2020-02-03-110037 $ oc get pod -o wide -n test1 NAME READY STATUS RESTARTS AGE IP NODE NOMINATED NODE READINESS GATES test-rc-gwcc6 1/1 Running 0 22m 10.128.2.7 v9579 <none> <none> sh-4.2# ovn-nbctl list Logical_Switch v9579 _uuid : 14fc1d83-6e1f-4897-95aa-9a1d5b9b44b9 acls : [1ae8eb74-9abb-4804-b112-a1f7290fd68d] dns_records : [] external_ids : {gateway_ip="10.128.2.1/23"} load_balancer : [465a25ec-15b0-4679-a3d0-0d5a26e47935, 5c270e5b-f270-43a2-ade2-04304d25c153] name : "v9579" other_config : {exclude_ips="10.128.2.3", subnet="10.128.2.0/23"} other_config=exclude_ips set to 10.128.2.3, pod IP set to 10.128.2.7
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-2020:0581