Nagios Support Forum

I have observer that the alert which are coming in soft state are not reflected on the alarm screen but are observed in host status trend report.

Can you suggest if we can have the alert in hard state without any soft state.So that it appear as on alarm window on the first go itself.

I need that any node reboot alarm need for persist on window for some time so that we can capture it properly without any miss.

When you say alarm, do you mean notification? If you want to be notified every time you need to set your max_check_attempts to 1. Please let me know if I am misunderstanding your question.

dwhitfield wrote:When you say alarm, do you mean notification? If you want to be notified every time you need to set your max_check_attempts to 1. Please let me know if I am misunderstanding your question.

Is it possible to make the ping check alert to make it persist for 5 minutes on the frontend to that it can be captured in case of node reboot as well.
Below is my service definition which i am using :
define service{
use generic-service
hostgroup_name Bharti_SOUTH,Bharti_EAST,Bharti_WEST,Bharti_NORTH,Bharti_IN,Bharti_NORTH_CENTRAL,Bharti_SOUTH_CENTRAL
service_description Service ping
check_command check_ping!3000.0,80%!5000.0,100%
max_check_attempts 1
check_interval 2
retry_interval 2
check_period 24x7
notification_interval 2
notification_period 24x7
retain_nonstatus_information 1
notification_options w,c,r
contact_groups admins
}

That's the entire point of the max_check_attempts. The best thing to do for a reboot is to set scheduled downtime. Unless you schedule downtime, XI has no way of knowing whether a device is truly down, or if it's a reboot. You could set up some sort of logic to send in a passive check whenever there is a reboot, but setting something up like that would require the reboot slowing down long enough to send out the passive check, and adding a way to send passive checks to all of your hosts (if that doesn't already exist).

If these are production servers, they probably aren't going down that often. If these are not production servers, you don't need to be checking them every two minutes.