I have setup Nagios and nsclient++ on a number of my Windows servers. They are all in the same subnet so no routing or firewall stuff is taking place in between the endpoints, and I have verified that the firewalls on the servers are not causing trouble. The problem is that the scheduled checks sometimes fail with "connection refused" and sometimes work! It is a frustrating problem to resolve because I do not know what to look for.
One place I did look is in the nsclient++ logs, where I am seeing this recurring error:
...\trunk\modules\CheckSystem\PDHCollector.cpp:148: Failed to query performance counters: \238...
This sounds promising, but I couldn't find much on Google about this error as it pertains to NSClient++