Tweaking connectivity and CPU alerts

I have the following rules setup to alert me if the server goes down for 15 minutes or if CPU load stays higher than the threshold for 45 minutes . If I reboot a server " the server fully reboots and come back up in 5 minutes" , I get alerted for conn and cpu .. I want not to get alerted if the server rebooted and hobbit looses connectivity for 5 minutes or so and the same for CPU.. How to fix that ?

The rules are listed below..

MAIL email at domain.com TIME=*:0000:2359 SERVICE=conn DURATION>15 REPEAT=45 RECOVERED

MAIL email at domain.com TIME=*:0000:2359 SERVICE=cpu DURATION>45 REPEAT=60 RECOVERED