Using Nagios to monitor high availability clusters.

[christopher baus]

Good Day,

I have been configuring nagios to monitor our redundant servers, and I am having a problem getting the results we want.  Here's a brief diagram.


         nagios monitor
           |
         router 
           |
   -----internet-----
  |                  |
  |                  |
  |                  |
main cluster   back up cluster

We have two levels of alerts, and have attempted to use dependencies and notification groups to implement the logic, but haven't been able to get exactly what we want..  I have set up host parents to represent this dependency tree.  The router is the last host both sites have in common.

Here's the logic we are looking for.

o In the case of failure of the main cluster XOR back up cluster notify the first level user (ie email).
o In the case of failure of the main cluster AND the back up cluster notify the critical user (ie cell phone).

It seems like a simple configuration, but even using all the advanced features (dependencies, escalations, etc.), I still haven't gotten it working just right.  

I am currently considering creating a virtual host who's service checks check_commands actually check the status of both servers.  But this seems like a lot of work, and considering how flexible the system is, it seems I could do this with in its basic framework.

thanks,

christopher
 



-------------------------------------------------------
This SF.net email is sponsored by: Etnus, makers of TotalView, The debugger 
for complex code. Debugging C/C++ programs can leave you feeling lost and 
disoriented. TotalView can help you find your way. Available on major UNIX 
and Linux platforms. Try it free. www.etnus.com
_______________________________________________
Nagios-users mailing list
Nagios-users at lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/nagios-users
::: Please include Nagios version, plugin version (-v) and OS when reporting any issue. 
::: Messages without supporting info will risk being sent to /dev/null