If a pod add fails in ovnkube-master, and there is no subsequent update event for the pod we never try to add it again and CNI will just continuously fail. A prime example of this is if a pod is added for a node, who has not yet registered with ovnkube-master by setting its L3 gateway config.
I see retry notices in the logs after cluster bring up, so it looks like the retry is happening 2021-01-05T15:06:19.032219891Z I0105 15:06:19.032198 1 ovn.go:574] [3313f8eb-228c-4462-bc3a-e53a831fdc20/openshift-apiserver/apiserver-6888ff4744-cl7v5] retry pod setup 2021-01-05T15:06:19.032219891Z I0105 15:06:19.032205 1 ovn.go:580] [3313f8eb-228c-4462-bc3a-e53a831fdc20/openshift-apiserver/apiserver-6888ff4744-cl7v5] setup retry failed; will try again later 2021-01-05T15:10:19.033712862Z I0105 15:10:19.033649 1 ovn.go:574] [69d3db07-c2c1-4cfa-a919-9119fb355531/openshift-apiserver/apiserver-6888ff4744-824zf] retry pod setup 2021-01-05T15:10:19.033712862Z I0105 15:10:19.033685 1 ovn.go:580] [69d3db07-c2c1-4cfa-a919-9119fb355531/openshift-apiserver/apiserver-6888ff4744-824zf] setup retry failed; will try again later 2021-01-05T15:10:19.033712862Z I0105 15:10:19.033697 1 ovn.go:574] [cdb4f7b4-f3b8-4040-b9cd-7d65d013da5b/openshift-etcd/etcd-quorum-guard-54d9fcbb68-267pd] retry pod setup 2021-01-05T15:10:19.033712862Z I0105 15:10:19.033702 1 ovn.go:580] [cdb4f7b4-f3b8-4040-b9cd-7d65d013da5b/openshift-etcd/etcd-quorum-guard-54d9fcbb68-267pd] setup retry failed; will try again later Verified on 4.7.0-0.nightly-2021-01-05-055003
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 (Moderate: OpenShift Container Platform 4.7.0 security, bug fix, and enhancement update), 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/RHSA-2020:5633