few questions about service/host setup

[Evan Mann]

I've been playing with Nagios for a few weeks now and started to read some
of the lists archives and came across a few posts that say that if a service
is monited as up, then the host is not checked, and this had me rethinking
how I've got Nagios setup.

I'm using Nagios exclusively to ping routers and firewalls on their external
and internal IP addresses.  If I get no ping response after XX times or XX
intervals, send a notifcation.

I set this up by adding a host for every one of my routers/firewalls and
adding a * service group and then disabling the service groups notifcations,
only using the hosts notifcations.

If Nagios relies on a service to fail to then check the host, for the type
of monitoring I am doing, is there any reason to use Host notifcation?
Should I do everything at the service level for the most immediate response
times?  Is there something that the host level provides me that I'd want to
utilize?  

Also, is there a way I can set the service.cfg normal_check_interval using a
's' designation?  I left the multiplier in main config as 60 so everything
is based on minutes, which means if I set normal_check_interval to 1 it
checks every 3 minutes.  I'd like it to run a check every 15 seconds.  Can I
just specify 15s in normal_check_interval or do I need to change my default
multiplier down to 1 and adjust all my check values accordingly?


-------------------------------------------------------
This sf.net email is sponsored by:ThinkGeek
Welcome to geek heaven.
http://thinkgeek.com/sf