dont check service when host down

[Marc Powell]

> -----Original Message-----
> From: Hari Sekhon [mailto:hpsekhon at googlemail.com]
> Sent: Wednesday, September 19, 2007 3:13 AM
> To: Marc Powell
> Cc: Nagios-users at lists.sourceforge.net
> Subject: Re: [Nagios-users] dont check service when host down
> 
> Marc Powell wrote:
> > Hosts are only checked when a service
> > on that host changes state. Service goes critical, host is checked.
> > Service recovers, host is checked. At no time in between is the host
> > checked.
> 
> Oh, ok, this seems a little contrary to logic but perhaps I just think
> in a strange way (but then we knew that!).
> 
> I suppose this method puts the emphasis on services rather than hosts
> and seems to give much higher priority to
> services, which is fair enough actually.

That's exactly right, nagios is a service monitor. Nagios-2 and prior
really only uses host checks to determine if there is a wider host or
network issue causing a problem with a service in order to handle alerts
more appropriately.

> Perhaps the state change was from pending to OK and that triggered the
> initial host check...
> In this scenario it actually worked better to check the service
anyway...

If this is the only service on the host a pending->OK transition gives
the host a 'host assumed to be up' designation. The host isn't actually
checked. Given the limited information I'd have to say that the service
did flap at some point _or_ you have aggressive host checking enabled.

--
Marc

-------------------------------------------------------------------------
This SF.net email is sponsored by: Microsoft
Defy all challenges. Microsoft(R) Visual Studio 2005.
http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/
_______________________________________________
Nagios-users mailing list
Nagios-users at lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/nagios-users
::: Please include Nagios version, plugin version (-v) and OS when reporting any issue. 
::: Messages without supporting info will risk being sent to /dev/null