<div class="gmail_quote"><blockquote class="gmail_quote" style="margin: 0pt 0pt 0pt 0.8ex; border-left: 1px solid rgb(204, 204, 204); padding-left: 1ex;"> I was directed in the irc forum to look at <a href="http://nagios.sourceforge.net/docs/3_0/statetypes.html" target="_blank">http://nagios.sourceforge.net/docs/3_0/statetypes.html</a> and I learned that services or hosts need to have a hard state failure, in order for notifications to go out. I've been trying to reboot my servers in order to get my notifications to go out. I'd like to check that I'll actually get notifications as soon as there is a problem, for example, a host goes offline, I'd like to know after about 2 minutes if possible! </blockquote><div> Then just turn a server off and wait a bit... ;-) You could also create a fake host, one with an unpingable IP address. As for how much time it takes to notify once the host/service becomes unresponsive, these three definitions (outside of any separate escalation policies) will dictate how fast the first alert will go out: max_check_attempts normal_check_interval retry_check_interval In your case, you'd probably want to set max_check to '2' and normal and retry to '1'. In this case, assuming "interval_length=60", the host (or service) would be checked every minute. When the host becomes unreachable/down, the host would then be rechecked once, one minute after it first becomes unreachable/down. If it was still down, the state would then become 'hard' and an alert sent. Total time before alert was sent could be anywhere from 61 seconds to two minutes, depending on when the host is first checked after it goes down. You really need to read the object definitions in the Nagios documentation. </div></div>-- James Moseley