Nagios Support Forum

I've been searching through the nagios site for the proper term that describes the problem I'm looking to solve.
There's a document called: 'Monitoring_Architecture_Solutions_For_MSPs.pdf' on the nagios site that I think nails it. (I've pasted the url for this pdf at the bottom to save space)


The pdf calls it 'Federated Monitoring'. Here's the excerpt that describes what I'm looking to do:

"In this model, remote networks and their elements are monitored by dedicated Nagios servers. Each remote Nagios server may be managed by central NOC staff or by the client. Notifications, reports, and configuration is generally handled by each remote Nagios server.
Remote Nagios servers can be configured to transfer check results (status information) back to a central Nagios XI server at the NOC. This allows NOC staff to have a birds- eye view of the entire network, and provides them with centralized reporting and optional notifications."

I think the most key part was, "Remote Nagios servers can be configured to transfer check results (status information) back to a central Nagios XI server at the NOC". This document does not elaborate on the ways that this may be implemented.
How might one configure remote nagios core servers to transfer check results back to a central NagiosXI server? And if this is possible, could this data from the remote servers be utilized by the dashboards/reporting/performance graphing/visualization features of XI that I think look promising?

I'm very interested in maybe purchasing an XI license -- but I will have to be able to set up a proof of concept using my existing remote core servers and an XI evaluation server.

Any ideas on the best way to implement this?

https://encrypted.google.com/url?sa=t&r ... KA&cad=rja

Yeah, for sure, we actually have another doc on that.

http://assets.nagios.com/downloads/nagi ... ith_XI.pdf
http://assets.nagios.com/downloads/nagi ... ith_XI.pdf

Both of these deal with sending checks from on Nagios XI server to another, and its surprisingly easy.

So a high overview would be using either NSCA or NRDP for the transfer -- and configuring the NagiosCore servers to use outbound checks that feed the centralized NagiosXI instance that must be configured to accept inbound checks from all the NagiosCore servers at different data centers?

Yeah, you would have to configure the Nagios Core servers to send their check output to the central Nagios XI server.

You also might consider a different approach such as using DNX:

http://dnx.sourceforge.net/

hi mr. scott,

for this federated monitoring, what is the latency (if there is any) of synching data from the remote nagios instances to the central nagios? say, the instances will be installed in parts of Asia, USA and Europe and central nagios server would be in Asia?

thanks in advance.
rose