https://prow.ci.openshift.org/view/gs/origin-ci-test/pr-logs/pull/openshift_ovn-kubernetes/1771/pull-ci-openshift-ovn-kubernetes-master-e2e-aws-ovn-local-gateway/1694101470443474944

{ 2023-08-22T22:45:51Z reason/ReusedPodIP podIP 10.131.0.9 is currently assigned to multiple pods: ns/e2e-provisioning-8830 pod/pod-subpath-test-preprovisionedpv-68cp node/ip-10-0-74-171.us-west-1.compute.internal uid/42d61041-f472-49e9-a3a9-02dd8af20de9;ns/openshift-monitoring pod/prometheus-adapter-58b4b5657d-p76s7 node/ip-10-0-74-171.us-west-1.compute.internal uid/6d32d8f6-04dd-452f-9175-f0b01900edb2}

The test failure is legitimate; the same IP is assigned to both pods. It looks like there was an ovnkube-controller restart while another previous pod had the IP but had been CNI DEL-eted. Rough timeline can be found from:

https://gcsweb-ci.apps.ci.l2s4.p1.openshiftapps.com/gcs/origin-ci-test/pr-logs/pull/openshift_ovn-kubernetes/1771/pull-ci-openshift-ovn-kubernetes-master-e2e-aws-ovn-local-gateway/1694101470443474944/artifacts/e2e-aws-ovn-local-gateway/gather-extra/artifacts/pods/openshift-ovn-kubernetes_ovnkube-node-ngzf8_ovn-controller.log

with:

grep "10.131.0.9[^0-9]|prometheus-adapter-58b4b5657d-p76s7|collect-profiles-28212360-zq4dt|pod-subpath-test-preprovisionedpv-68cp"