We used to contract out our NOC functionality, which used Netcool exclusively. When we decided to move our NOC in-house, it was decided that we wanted the same functionality as Netcool but wanted to look at open source products. I think our requirements are somewhat different from most users of OpenNMS as such there were some specifics that needed to be met, OpenNMS has met or exceeded them all and in fact we really the only product we found capable of meeting them. While it seems most people use OpenNMS for systems monitoring (and we do use it for that), we are much concerned with alarms occurring in our switched network. (T1s, Trunk groups, call failure rates, site temperature, etc).
We are currently monitoring 657 nodes with 2000 interfaces, but we also have all remote sites reporting environmental alarms via SNMP. We also get events into OpenNMS via send-event.pl and syslog. If we have a device that is going to produce an event, we can always find a way to get it into OpenNMS. All of our alarms are divided into categories and shown on the OpenNMS main page as graphs (emulating Netcool). We have also been able to integrate OpenNMS into our own ticketing system.
While it’s true OpenNMS did not provide some of things we needed out-of-the-box; having the code, an active public support channel and mailing list, along with our paid support, we have been able to exceed what we had with Netcool and do it much cheaper! I am looking forward to some more enhancements to take place on the TL1 monitor so I can get rid of my custom code and implement that as well.
We also love the fact that we can open a support ticket and get to talk to the people that write and maintain the code, not just some support technician.
- Thomas Michelbrink, Excel Telecommunications, USA