this is a dummy bug to backport https://issues.redhat.com/browse/SDN-4173

1. Let's say you have a 3 node cluster
2. create test namespace with 1 pod on every node
3. create egress firewall 
kind: EgressFirewall
apiVersion: k8s.ovn.org/v1
metadata:
  name: default
spec:
  egress:
  - type: Allow
    to:
      cidrSelector: 1.2.3.0/24 
4. count number of ovs flows on any node (every node should have 1 pod from that namespace)
use https://github.com/openshift/network-tools
`network-tools ovn-count-flows -l 10 <ovnkube pod on a chosen node name>`
find output like
Number of OVS flows=1 for ACL with name and ids = "EF:default:0" {"k8s.ovn.org/id"="default-network-controller:EgressFirewall:default:0", "k8s.ovn.org/name"=default, "k8s.ovn.org/owner-controller"=default-network-controller, "k8s.ovn.org/owner-type"=EgressFirewall, rule-index="0"}

the number of flows in that case is 1
5. The difference between before/after the fix should equal <number of nodes-1>, so before the fix the same rule should result in 3 OVS flows