SNMP Bandwidth Monitoring

We are trying to monitor the bandwidth utilization of CPE devices using PRTG. Setup goes find and the program sees the device but we are only seeing like 1kbps or 10kbps on the graps. This is for both CPE and AP. Any suggestions?

I see the same thing, a SNMP walk doesn’t show many options.
I’m hoping someone else has solved this.

Can someone state if they are able to see bandwidth utilization via SNMP? We are still having the same issue. SNMP read sees a port but the port when selected to monitor only shows 10k or so worth of utilization not whats actually used. Its the same for all CPE’s that we tried.

Try polling the AP for specific SM bandwidth Usage. It’s a pain but it should work.

How do you do this? I haven’t been able to figure it out.

Well I can’t speak directly for the moto gear. We use the Gemtek 3.65 CPE’s (just not with Moto’s FW) and Purewave 3.65 Base Stations.

When I tried setting up SNMP monitoring for the SM’s I determined that there weren’t and OID’s for traffic in the CPE it’s self.

So I started looking at the AP and noticed that the CPE’s and their up and down traffic counters where listed. So I started digging through the SNMP of the AP by doing snmpwalks and trying to determine the SNMP OIDs for UP and DOWN traffic of a given SM.

I was able to determine the OID’s for the information, I then duplicated the Network Interface Statistics Template and modified it with the proper OID’s and generated the graph.

It took some time to figure it out but it works.

Note the graphs will be listed under the AP…not the client within Cacti

These are probably the fields you want - uplink/downlink bytes and packets for the radio link and ethernet link (NO packet counter on radio AFAICS) on the CPE:

Motorola-CPE-PRIVATE-MIB::gemtekDevCpeWimaxTotalUpLinkDataByte.0 106736342
Motorola-CPE-PRIVATE-MIB::gemtekDevCpeWimaxTotalDownLinkDataByte.0 964699610

Motorola-CPE-PRIVATE-MIB::gemtekDevCpeLanTotalDownLinkDataByte.0 73704864
Motorola-CPE-PRIVATE-MIB::gemtekDevCpeLanTotalUpLinkDataByte.0 948120150
Motorola-CPE-PRIVATE-MIB::gemtekDevCpeLanTotalDownLinkDataPackets.0 437783
Motorola-CPE-PRIVATE-MIB::gemtekDevCpeLanTotalUpLinkDataPackets.0 6108368

.1.3.6.1.4.1.10529.300.1.1.11.0 106738751
.1.3.6.1.4.1.10529.300.1.1.12.0 964705079

.1.3.6.1.4.1.10529.300.1.2.2.0 73704998
.1.3.6.1.4.1.10529.300.1.2.3.0 948126013
.1.3.6.1.4.1.10529.300.1.2.4.0 437785
.1.3.6.1.4.1.10529.300.1.2.5.0 6108462

IMPORTANT NOTE: ‘UpLink’ and ‘DownLink’ are reversed in a sense if you use the LAN fields. (‘uplink’ is ‘to customer over LAN’) I graph the second field (gemtekDevCpeLanTotalUpLinkDataByte.0) as ‘cpe downlink’ and the graphs make sense - customers provisioned for 10mbps down/1mbps up for example it’s obvious they can’t be sustaining 6mbps uplink…

BTW - I graph the LAN instead of the WiMax fields - there will almost always be a difference in bridge mode, definitely a difference in NAT mode. Up to you which you are more interested in tracking.

I graph throughput and radio stats (stack dBm and CINR the right way you have a band, top is signal bottom is noise floor) for all the PMP320CPEs on our network, polled every 10 mins.

j