Improve Wireless collection and graphing code

@wriedel Sure, but I don’t really have an idea what you are talking about :smiley:

Have a look at the code, I tried to make it easy to extend/understand.

Reposting from IRC:

  • Is $access_point_id currently used? Also why is it sometimes $access_point_ip?
  • Also $entPhysicalIndex
  • Why does /wireless/metric=clients/ have mini graphs but /wireless/metric=noise-floor not?
  • I noticed some of the wireless frequency sensors are collecting in GHz not MHz.

I have LibreNMS running on my home network which has 2 Cisco ap1042n devices on it. These are multiple radio devices (1 gigabit Ethernet, 1 2.4GHz radio and 1 5GHz radio).

Happy to test anything out if you want to verify support for autonomous Cisco APs, although I doubt there are many people using them!

OK the noise floor graphs showed up after a while, maybe there wasn’t enough data to start with.

@TotalGriffLock if you are missing data, please create an issue on GitHub supplying the requested data and what you would like to see.

When I added Aruba IAP 325 virtual controller. I’m seeing these errors in the discovery process “ERROR: A sensor already exists at this index frequency-arubaos-iap-0” and its reporting over and over. not sure what is going on. Any help would be appreciated.

I think it’s trying to add the other AP’s radios but it can’t cause the radios are all named alike?
Discovery output: https://gist.github.com/kkrumm1/d21d2f15c579c69d9ef90db9d89e40e9#file-aruba-iap-discovery

@Kevin_Krumm

Try this patch: https://p.libren.ms/view/raw/eca0ae24

1 Like

adding info for later time - https://github.com/librenms/librenms/issues/6989

Adding https://github.com/librenms/librenms/issues/4810 for wireless traffic support