Reference task: https://issues.redhat.com/browse/MULTIARCH-4421

Repo: https://github.com/openshift/openshift-tests-private/blob/master/test/extended/networking/cloud_egressip_ovn.go
Ran all the tests again after the PR https://github.com/openshift/openshift-tests-private/pull/1615 was merged.

Result: error: 3 fail, 5 pass, 33 skip (44m18s)

Failing tests:
[sig-networking] SDN NonHyperShiftHOST-ConnectedOnly-NonPreRelease-PstChkUpgrade-Author:jechen-High-56875-OVN egressIP should still be functional post upgrade. [Disruptive] [Serial]
[sig-networking] SDN OVN EgressIP NonHyperShiftHOST-Longduration-NonPreRelease-ConnectedOnly-Author:jechen-High-54647-No stale or duplicated SNAT on gateway router after egressIP failover to new egress node. [Disruptive] [Serial]
[sig-networking] SDN nmstate NonHyperShiftHOST-Author:qiowang-Medium-54350-Verify that nmstate work well with SDN egressIP [Disruptive] [Serial]

RH QE Jean Chen opened bugs for the failing test scenarios.

OCPBUGS-33071 [New] : on IBM powervs cluster, getting "unable to start container process" when trying to execute ovn-nbtrl command on ovnkube-node northd container
OCPBUGS-33072 [New] : on IBM powervs cluster, Egress IP Health Server is not started in ovnkube-node pods in ovnkube-controller containers for master nodes
OCP-61344 Jean is working on fixing it. https://privatebin.corp.redhat.com/?cff3aa02ec338a12#7B91Eu35fMRZJs5cHuLThPsvkiEBWtd1kjnoKad271v6
He will submit PR for the fix to cluster_helpers.go