Phantom bandwidth usage through PRTG

Hi all, 

I discoverd PRTG a few weeks ago and found it played will with ePMP1000 so I've set up some monitoring...but there's something I can't figure out:

While monitoring the ethernet port of any SM, it is constantly showing ~350kbits/second of usage. When I set up a test link I found that, even when I'm not pushing data through the connection, it still shows ~350kbits/second of data - Traffic out of the ethernet port. 

The thing is, even if I have it plugged into a switch, with nothing else plugged in, it shows the same data out from the ethernet port. 

So I did some math and found that ~350kbits/second over a month is a LOT of data, and it's totally messing with my ability to see how much data my clients are actaully using. 

I used wireshark to see what was going on and it showed just a few discovery packets, nothing to justify the displayed usage. Can anyone enlighten me on what is going on and maybe how can I correct or compensate for it? 

We also use PRTG and monitor the wireless port of the AP, not the Ethernet port for our stats. We also monitor each SM indiviudually as well.  We do see constant traffic but I associated it with SNMP data, cnMaestro data, NTP, etc

I figured it out and thought I would share in case anyone else had this issue:

We have a mixed network consisting of ePMP, UBNT and Miktotik gear. The Mikrotik gear uses it's own discovery protocol to find it's relatives on the network. As soon as I blocked this protocol the average throughput on the ethernet port of my test SM dropped from ~350kbits/sec to ~50kbits/s. 

Problem solved. 

Thanks for the input!

1 Like