The additional routes aren't getting created for the SR-IOV VRF created. They are trying to do a setup with extra routes added to the secondary (SRIOV) interface (eg. new default route for applications that bind to the 2nd interface/VRF).

To isolate the issue, adding extra routes in the "default" routing table works, it's adding the VRF setup where extra routing doesn't work.

OCP 4.14.37 & SR-IOV Network Operator v4.14.0-202412180606

Being able to use VRFs in general, and in SRIOV cases specifically, is of very high importance to us given that we'll be running Telco workloads on the platform - most of those CNF workloads will use secondary (multus) interfaces and will need to have correct routing (including a 2nd default route) added

1. Refer this documentation--> https://docs.openshift.com/container-platform/4.14/networking/hardware_networks/configuring-sriov-device.html#cnf-assigning-a-sriov-network-to-a-vrf_configuring-sriov-device 

to create pods with secondary SR-IOV network

sh-5.1# ip r s vrf vrf-1
10.0.30.8/29 dev net1 proto kernel scope link src 10.0.30.10

The route isn't in the VRF namespace (it isn't in the default routing table either)

The routes getting created for the VRF

They are using ConnectX-4 MT27710 SR-IOV nic & current CNI is Cisco ACI. 

Also, having extra routes works perfectly fine as long we don't create the extra VRF - eg. leave out the VRF metaplugin from the IPAM block and the route will be added to the global (default) routing table as expected. It's only when the VRF is added that we have the issue