Need to get the following stats:

• list network interfaces
• collect following metrics (on any interface)
  • if_dropped (in) -  The total number of received dropped packets.
    • => node_network_receive_drop_total
  • if_errors (in) - The total number of received error packets
    • => node_network_receive_errs_total
  • if_octets (in) - The total number of received bytes
    • => node_network_receive_bytes_total
  • if_packets (in) - The total number of received packets
    • => node_network_receive_packets_total
  • if_dropped (out) - The total number of transmit packets dropped.
    • => node_network_transmit_drop_total
  • if_errors (out) - The total number of transmit error packets
    • => node_network_transmit_errs_total
  • if_octets (out) - The total number of bytes transmitted
    • => node_network_transmit_bytes_total
  • if_packets (out) - The total number of received packets transmitted·
    • => node_network_transmit_packets_total
  • [not MVP but required] if_packet_mcast (in) - Total number of received multicast packets
    • => node_network_receive_multicast_total

Check https://issues.redhat.com/browse/HATSTRAT-206 for complete description.

=>
Currently those metrics are available via the cluster monitoring operator (node exporter / netdev), however there are some hardcoded excluded devices.
The exclusion list is as such:

--collector.netdev.device-exclude=^(veth.*|[a-f0-9]{15}|enP.*|ovn-k8s-mp[0-9]*|br-ex|br-int|br-ext|br[0-9]*|tun[0-9]*|cali[a-f0-9]*)$

which typically results in just a few monitored interfaces.
In order to address the requirement, we should make that exclusion list configurable.

PS: listing all interfaces can also be achieved but also requires to remove or amend the exclusion filter. E.g: "group(node_network_iface_id) by (device)"