Nagios Agent - Server Architecture

This support forum board is for support questions relating to Nagios XI, our flagship commercial network monitoring solution.
Locked
MSPk
Posts: 317
Joined: Fri Aug 24, 2012 12:03 am

Nagios Agent - Server Architecture

Post by MSPk »

Hi,

I'm trying to understand how the parameter info from the monitored server is sent to the nagios management server. Currently we are configuring the service on nagios and as I understnd nagios creates a check for each parameter we try to monitor locally on a windows server and executes a check as per the polling interval, this way we have the server contact the agent each time the check is executed to get the info.

Do we have any other architecture in nagios where we can configure it in such a way that it contacts the agent once and collect the metrics of all the services/parameters (like CPU,memory, local drive, page file, services, processes) being monitored on the server in one go and send it to or submit the result to nagios when asked?
yancy
Posts: 523
Joined: Thu Oct 06, 2011 10:12 am

Re: Nagios Agent - Server Architecture

Post by yancy »

MSPk,

There are two dominant methods for checking.

1) Active Checks: An agent is installed on the client (such as windows server) and the agent listens for connections. Nagios performs the polling, scheduling and processing of check results.

2) Passive Checks: Some client process on the host runs the checks and submits the results back to Nagios. Nagios Simply listens for results and is passive in this transaction.

From your question, it sounds like you're looking for a passive check architecture where the client returns data to Nagios on it's own terms.

One particular solution is NRDS_Win which simply runs a windows scheduled task to return check data to Nagios via http or https
http://exchange.nagios.org/directory/Ad ... in/details

Regards,

-Yancy
Locked