OpenNMS is event driven. Events can be received, modified, assigned a severity, reduced (in the case of duplicate events) and can cause notification actions, such as sending an e-mail or SMS, or running a corrective script, to be performed. Numerous event sources can be used, from SNMP traps to Syslog messages to TL/1 logs, and of course it is extremely easy to add custom events. OpenNMS can serve as the central repository for all information available from devices on the network.

We have clients with over 70,000 devices being managed by OpenNMS. While often overlooked, the ability to add, change and remove devices from the management platform becomes incredibly important on a network of this size. OpenNMS can automatically discover the network, gather information from external sources (such as a company database) or a combination of the two. This allows for a fine grain control over device management with results in fewer false positives.

OpenNMS has the ability to collect time-series data from a number of protocols: HTTP, SNMP, JMX and even WMI. This data can be stored, graphed, as well as measured for thresholds. Not only does it provide the basic high/low thresholding, OpenNMS can also trigger alerts on complex thresholds (involving more than one value) as well as relative and absolute changes in collected values. Whether measuring CPU utilization or the price of an eBay auction, OpenNMS can do it.

If it is reachable over the network, OpenNMS can monitor its performance. From simple “pings” and port checks up to complex monitors for web sites, mail services and mobile telephony communication, OpenNMS comes with a number of monitors to insure that those services are both active and responsive. In an age where a slow web page can result in a quick loss of revenue, a solid service monitoring solution can pay for itself in no time.