Should the Host Metrics source scrape all network interfaces on the host?

[hickersonj]

I'm using the host_metrics sink with the collectors: network in a k3s cluster and the only metrics I see are for tunl0, eth0 and lo devices. The hosts network has several more NICs. Is there something wrong with my setup, or is it monitoring only within the node/pod?

Example metric output:

{
  "name": "network_transmit_errs_total",
  "namespace": "host",
  "tags": {
    "collector": "network",
    "device": "tunl0",
    "host": "vector-zx4pc"
  },
  "timestamp": "2023-06-08T07:35:27.245175986Z",
  "kind": "absolute",
  "counter": {
    "value": 0
  }
}
{
  "name": "network_receive_bytes_total",
  "namespace": "host",
  "tags": {
    "collector": "network",
    "device": "eth0",
    "host": "vector-zx4pc"
  },
  "timestamp": "2023-06-08T07:35:27.245175986Z",
  "kind": "absolute",
  "counter": {
    "value": 584086
  }
}