NSClient++ Help (#1) - watching the watcher? (#617) - Message List

watching the watcher?

Hi gang,

I'm just about to rollout nsclient++ to a bunch of servers, but I wanted to understand what the common thinking is about watching the watcher.

In particular, I'd like to monitor nsclient++ to make sure it doesn't impact the systems in a significant way. I've had bad experiences with monitoring agents in the past (not this one!).

right now, i've got a few checks defined in my nagios configuration that watch:

\Process(nsclient++)\% Processor Time \Process(nsclient++)\Working Set \Process(nsclient++)\Thread Count

Will this give me the information I need?

My nscp configuration generally uses Perfmon counters (probably 20-30 or so per server) and some external scripts (about 10-15 per server).

Will these datapoints help me keep track of the impact of nsclient++? Will this capture metrics (cpu, mem, etc) for external scripts that are "shelled" out?

any other tips or recommendations?

thanks,

J

Tree View Flat View (newer first) Flat View (older first)
  • Message #1882

    Humm..

    There is a "DebugLogMetrics?" plugin you can use...

     + Added new module A_DebugLogMetrics.dll which can be used to generate debug info.
       Enable the module and a file called process_info.csv will be created under %APP_DATA%/nsclient++/process_info.csv which contains metrics.
    

    But that only gives you information on NSClient++ not any scripts you are using (and yes it is just performance counters so you can do it by hand).

    Not really sure how to get info from scripts though...

    Michael Medin

  • Message #1884

    Notice that the DebugLogMetrics? will have a negative impact on the performance of NSClient++ (as it needs to log debug information)

    Michael Medin

Tree View Flat View (newer first) Flat View (older first)

Subscriptions