Voice Infrastructure Monitoring

When it comes to proactive monitoring, many of us who support larger organizations are all familiar with software suites that provide a “dashboard” view of their infrastructure. HP OpenView, CA Unicenter, Nagios, and Hyperic all come to mind. Most, if not all, of these platforms include SNMP support for handling traps as well as importing MIBs. They tie together monitoring and even management of directory services, database and application servers, and core network infrastructure. It seems as though there is somewhat of a lack for PBX / media gateway / convergence server unified monitoring. Sure, we can ping the device and collect SNMP statistics on port traffic, etc, but what about the statistics that really matter? There is a void of such a proactive monitoring environment to analyze call traffic patterns, voice circuit functionality, busy ports, etc, in the voice realm. So, my question is this: for those of you managing such an environment, what are YOU doing to anticipate problems before they actually occur? Or, are we still chasing after user reports of dropped calls or “all ports busy” messages?