monitoring unreachable hosts

[Frost, Mark {PBG}]

This topic of unreachable hosts that's come up recently has got me
thinking about an issue we have.  We have a few hosts that are behind
proxies and as such are impossible to ping.  They have a single service
which we can check through the proxy successfully.

I'm a little stuck on what to do with the host checks.  I can supress
the alerts, but they still always show as critical (kind of permanently
critical).  Maybe I'm too anal, but I don't like that.  It's
meaningless.  So I was thinking of making the host check command
identical to the service check command.  The end result being that if
the service is up, the host will also show up, if the service is down,
the host will also show down.  Still weird, but at least kind of
conceptually correct.

What concerns me about is the my understanding of how host and service
checks would work in this case.  If the service check fails, it's going
to want to fallback and try the host check which will fail also because
it's running the exact same check as the service.  However,  what if
I've disabled notfications on the host itself?  That is, I really only
ever want service alerts -- no host alerts.  In the case of a normal set
of host/service checks, Nagios will notice that the host is down,
suppress alerts about all the services being down and send out only a
"host is down" alert.

In the scenario where the host and service checks are the same and the
host notifications are disabled, wouldn't a service alert be suppressed
and I'd never hear about an issue on this host?  Again, conceptually,
the service result is all that's meaningful.  Maybe Nagios is smarter
than that and would decide that since the host notifications were
disabled that it would go ahead and send the service alerts anyway?  I
guess this could be an issue with any host that has notifications
disabled now that I think about it.

I tried earlier to disable all checking of the host in this case and
that gives me a host unreachable alert when the service goes down.  That
makes sense since the host can't be checked, but I don't want it going
into that bogus state.

What would probably be ideal is if Nagios had some way of marking a host
as being irrelevanet.  That is, the service would sort of stand alone
without any host dependency in terms of monitoring.  Kind of strange,
though.

Thanks

Mark

------------------------------------------------------------------------------
Enter the BlackBerry Developer Challenge  
This is your chance to win up to $100,000 in prizes! For a limited time, 
vendors submitting new applications to BlackBerry App World(TM) will have
the opportunity to enter the BlackBerry Developer Challenge. See full prize  
details at: http://p.sf.net/sfu/Challenge
_______________________________________________
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