Default VA1 deployment[1] when done with "OVNKubernetes" network_type do not work.

Noticed when validating the default switch[3]

The issue happens when using 192.168.122.1 as dns forwarder for DNSMasq CR[[4][5]

Related Jira thread[6] where others too hit the issue.

Issue is dnsmasq pods can't reach the dns server at 192.168.122.1 and thus not functional. And edpm nodes can't resolve any DNS address and thus fails operations like compute nodes are not registered etc.

Workaround is to see some other DNS server as forwarder which is reachable/routable from the openshift pods/ worker nodes.

[1] https://ci-framework.pages.redhat.com/docs/ci-framework/05_deploy_va.html

 [2]

cifmw_devscripts_config_overrides:

  network_type: "OVNKubernetes"

 [3][ https://github.com/openstack-k8s-operators/ci-framework/pull/918|https://github.com/openstack-k8s-operators/ci-framework/pull/918]

 [4] https://github.com/openstack-k8s-operators/architecture/blob/007ccf00d6c3d4aa16f684fc2f1104fa80f77b1f/va/hci/stage4/openstackcontrolplane.yaml#L51

 [5] https://github.com/fultonj/architecture/blob/ci-framework/validated_arch_1/stage4/openstackcontrolplane.yaml#L31

 [6] https://redhat-internal.slack.com/archives/CQXJFGMK6/p1700255782023559