I have am Monitoring a Windows Host on an MPLS

[slansing]

Nagios would absolutely have to have a way to communicate with the network/address at the end point, or the end point would have to send check results passively back up to Nagios. Traps can be a bit more advanced, but they are certainly one way to do this, you could also send up NSCA check results, or NRDP results.

[billperrotta]

Please recommend which way you think is the easiest way to do this?

I am not familiar with any, and tried to set up snmp traps according to nagios 11 document.
However I do not have a trap wizard with nagios core.


How can the endpoint communicate with Nagios, when the point to point network goes down and switches to dial backup?

My boss would like the grid to turn yellow instead of red and still record the server stats when the main mpls is down.

[slansing]

How can the endpoint communicate with Nagios, when the point to point network goes down and switches to dial backup?

My boss would like the grid to turn yellow instead of red and still record the server stats when the main mpls is down.

This is totally up to you, and your network team. You need to set up proper networking routes so you can validly monitor those systems. A quick heads up, yellow indicates a warning in nagios, red is a critical, green is an okay, and orange is an unknown. If you only have access to those endpoint systems when the dial up connection comes up you can set whatever networking equipment that is in between you and the endpoint as a parent host, that would change the endpoint systems to an unknown state while they are unreachable.

http://nagios.sourceforge.net/docs/3_0/ ... ility.html