max_check_attempts vs first_notification_delay

[matt.uebel]

I am trying to understand the host configuration directives as described here: http://nagios.sourceforge.net/docs/3_0/ ... tions.html

Suppose a host template configured as such:

Code: Select all

define host {
       name                          		linux-server
       hostgroups                    		+Production Linux
       check_command                 		check-host-alive!!!!!!!!
       use                           		generic-host
       max_check_attempts            		10
       check_interval                		3
       retry_interval                		1
       check_period                  		24x7
       contact_groups                		+admins,Unix-Linux
       notification_interval         		60
       first_notification_delay      		0
       notification_options          		d,u,r,
       register                    		0

}	

Now, my understanding is that, at the point an active check fails, Nagios does not send an alert because of the max_check_attempts being set to greater than 1. In this case, it will try again every 1 minute until it reaches 10 attempts. At that point, it will immediately send a notification because the first_notification_delay is set to 0.

Is this right?

Given that it is correct, if we wanted a host to immediately notify at active host check failure, we should set max_check_attempts to 1, and first_notification_delay to 0.

[sreinhardt]

Your logic seems sound, and yes this particular definition will go soft warn\critical, then check every 1 min for 10 times, you are correct.