How To - Set the frequency and sensitivity of alerts

Because the potential always exists for random CPU utilization peaks, network throughput peaks, and other issues which may cause a device not to respond on the first attempt, IPSentry provides a way to limit the potential for false alerts through the IPSentry Alert Schedule on each device.
 
Also, you may not wish to be notified each and every time that IPSentry cycles on a failed device.  Sometimes the first page is enough, but second and third notifications should be delayed and serve only as reminders.  Again, IPSentry provides a method to resolve this issue on the IPSentry Alert Schedule.
 
One final function of the Alert Schedule is the ability to limit the total number of alerts generated by a failed device.  It may just simply be unacceptable to send a pager alert, every minute, until the problem is resolved - especially when the technician responsible must drive an hour or two to correct the problem.
 
In the example below, we will configure an alert to be triggered only after the 2nd failure.  That alert will then only be triggered every 5 cycles until a maximum of 10 alerts have been sent.  This schedule would provide us with adequate coverage of alerts without killing the batteries on our pager due to our 2 hour drive in order to correct the problem.

1. 
   Select to modify an entry.
    
   
2. 
   Select the alert tab for the alert you wish to tune.
    
   
3. 
   At the bottom of the alert configuration, you will see a section titled Alert Schedule
    
   
4. 
   Set the First After value to 2
    
   
5. 
   Set the Every value to 5
    
   
6. 
   Set the Max value to 10
   

The alert will only triggered "First, After 2 failures" and will be triggered "Every 5 Failures" thereafter until a "Max of 10" alerts have been generated.  Once the monitored item is successful, the internal alert counters are reset and the process is reset.
 
Note: It is possible that if you have a device which fails often but recovers immediately, the Max Alerts will have no net effect in that if the device fails twice, then recovers, then fails twice, then recovers, then fails twice, and so on... the alert schedule counters will keep being reset and the alert will continue to be triggered.  Each time a successful monitoring takes place, the internal counters for the above actions are reset to zero.