Wyoming Public Radio’s network infrastructure has expanded at an amazing rate in the past three years.  When I started here, the only internal networking was for the automation system.  Now, however, virtually every piece of equipment we bring into the plant has an Ethernet port.  We have at least five different “virtual” networks running internally now (along with equipment around the state connected via microwave).  It totals over 70 different ‘nodes’ in our network.

I decided it was long past time to implement some form of network monitoring for the WPR plant.  After a bit of research, I found something called ‘OpenNMS’.  This package is an open source piece of software that allows monitoring of networked devices using a variety of methods.

The ‘Dashboard’ view of OpenNMS.  This gives me an ‘at a glance’ view of network status.  I can group various types of devices together for easier monitoring.  The software can E-Mail me to alert me to potential problems.

Monitoring the number of listeners on one of our web streams…I gather this data via HTTP from the status page of the stream server.  Configuring this data source was a bit tricky but once the configuration was complete, OpenNMS started to collect, record and graph the data.

Another useful thing to monitor is Eb/No from our satellite receivers.  This gives us a good indication of the performance of our satellite links.  The satellite receivers provide this data via SNMP.  OpenNMS provides a significant amount of SNMP data gathering without any additional configuration.  In this case, however, I had to dig a bit to find out exactly where to gather the data from the receivers.  Once I found the data and configured OpenNMS to gather it, viola!  Very useful graphs.

One final graph…Bandwidth on our 1 gigabit fiber backbone link to the UW campus.  We consume almost 80 megabits of bandwidth during peak periods of web stream listening.  This data was gathered from the network switch.  No additional configuration in OpenNMS was required to generate this particular graph.  This is one example of how powerful this software can be even without additional customization.

If you would be interested in implementing something similar in your broadcast plant, leave a comment and I would be happy to share my experiences in configuring the software.