Nagios Configuration Parent/Child hosts

Aaron M. Segura aaron.segura at cabelas.com
Mon Jan 7 23:42:13 CET 2008


On Mon, 2008-01-07 at 14:13 -0800, Jeff Shumard - DefenseWeb
Technologies wrote:
> What you are saying is to use service dependencies instead of
> parent/client hosts.  I could do this it's just a lot of configuration
> and I was trying to see if we could do it with the parent/client host
> configuration.
> 
> Is that what you are saying to setup is to depend all the checks to the
> one NRPE_NT service?
> 
> Thank you,
> Jeff
> 
> -----Original Message-----
> From: Aaron M. Segura [mailto:aaron.segura at cabelas.com] 
> Sent: Monday, January 07, 2008 1:46 PM
> To: Jeff Shumard - DefenseWeb Technologies
> Cc: nagios-users at lists.sourceforge.net
> Subject: Re: [Nagios-users] Nagios Configuration Parent/Child hosts
> 
> On Mon, 2008-01-07 at 12:45 -0800, Jeff Shumard - DefenseWeb
> Technologies wrote: 
> > We have Nagios running its checks just fine.  What have a mix network
> of
> > Windows and UNIX Servers.  We have a windows server running NRPE_NT
> > which we run all our WMI Checks from for each of the windows hosts.
> We
> > send our checks to this host it submits a WMI Query to the correct
> host
> > then sends back the response.  This is working just find with out any
> > problems.  We had a problem last week where the NRPE_NT service
> stopped
> > accepting checks and started a ping flood.  That is the background and
> > here is my question.  Can we resolve this making this Server running
> > NRPE_NT the parent of all our windows service checks?  What we would
> > like to do is if a service goes down it then runs the check on the
> > parent to see if it is up.  If the parent is not up then it suppresses
> > all the alerts for all childs to the parent server.  Is this possible,
> I
> > have tried a couple of configurations but have not been able to get
> this
> > to work.
> > 
> > Thank you,
> > Jeff
> 
> <sorry if this is a duplicate.  Evolution crashed while sending the
> first time>
> 
> That is entirely possible.  It's what we do for all of our NRPE-based
> checks.  If the NRPE daemon is unresponsive on any given server we just
> get the one alarm for "NRPE_Alive" instead of multiple alarms for all
> the checks that depend on NRPE.
> 
> Instead of making the *server* the parent, make the NRPE_NT *service* on
> that server the parent of all NRPE_NT-based service checks that happen
> through it.
> 

Yes, that's correct.  The only way parent/client host configuration will
work is if you configure nagios to evaluate the *host* state of your
NRPE_NT server based solely on the status of the NRPE_NT service running
on it.  Of course, that also would affect services on the child servers
that don't use NRPE_NT at all.

In other words, your host-check for your NRPE_NT server would have to
solely evaluate whether the NRPE_NT service is running on that server in
order to accomplish what you want.  However, that would also cause, say,
an ICMP check of a child server to show as being "down" due to the
NRPE_NT parent host being "down", even though it has nothing to do with
the ability of your nagios server to make an ICMP request and get a
reply...which may not be what you want.

The best solution AFAIK is service-based.

I think that's right.  I'm sure some helpful soul will chime in and
correct me if not :)


-------------------------------------------------------------------------
Check out the new SourceForge.net Marketplace.
It's the best place to buy or sell services for
just about anything Open Source.
http://ad.doubleclick.net/clk;164216239;13503038;w?http://sf.net/marketplace
_______________________________________________
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





More information about the Users mailing list