Wish list: Elevate SNMP priority within SM

It would be nice if SNMP traffic could be given a higher priority than user traffic.  Pretty please.

We monitor each SM with Cacti and whenever the customer hits their bandwidth limit Cacti starts missing SMNP responses.  Resulting graphs show chunks of missing data making them look just like SM's with bad radio links dropping sessions.  Canopy SMNP support is so extensive.  Clearly it is a high priority within Cambium and that is a good thing.  And yet the data fails to be returned making it untrustworthy... only because the customer happens to be maxing out their sustained data rate limit at some level less than the SM's actual throughput capability.  The SM is choosing to ignore the SMNP request even though the 450SM has the bandwidth and CPU power to spare.  If we start reaching the 450AP throughput limit then we will ease off on the polling rate, so let us decide when too much SMNP data is too much.


Thanks for the feedback Paul.  I am investigating this with our development team, and will get back to you here!

Sorry for the delayed response.

1 Like

We are having the same issue. It is inconvenient enough that I am investigating how hard it would be to modify the poller cmd.php in cacti to try snmp again if the device has been recently connected.

Anyone else use SNMP regularly and lose chunks of data due to customer maxing out their bandwidth allocation?  Are we, and @dsoodak, the only people reliying on SNMP to monitor SMs?

@CambiumMatt: Did your development team get back to you about this?

We are seeing this exact same thing. I would like to see this for all products, PTP and ePMP.