Quick "escalation" question

[Carroll, Jim P [Contractor]]

> is it necessary to define an escalation for EVERY Service of 
> every host 
> in the escalation.cfg??  if I want all pages for all problems for a 
> specific box sent to the same contact group then can I get away with 
> just a host escalation?

Yes.

> If I don't define an escalation for say http on box #1 and it 
> goes down 
> will no one get notified?? or will it just use the host escalation in 
> liew of the missing service escalation?

The latter.

> as far as Nagios is concern how does it deal with hosts vs. 
> services?? I 
> am under the impression the only "host" check being done is the 
> check-hosts-alive as defined in the hosts.cfg so the bulk of 
> what your 
> monitoring in nagios will be in the services.cfg I am monitoring much 
> more then ping...  so again, is only a host escalation sufficient to 
> trigger a notification in the event of a service failure?

Nagios only does the host check if all service checks for that host are hard down.

> I am monitoring over 300+ services on 80+ hosts so if I can get away 
> with only needing to have 80 host escalations rather then 
> 300+ service 
> escalations + 80 host escalations I would welcome the cut 
> down in time...

Try 128 hosts, 1080 active checks, and *many* service dependencies to cut down on the NRPE churn.  ;)

And yes, we're using hostgroupescalation definitions.  :)

jc

> --Chris
> 
> 
> 
> -------------------------------------------------------
> This sf.net email is sponsored by:ThinkGeek
> Welcome to geek heaven.
> http://thinkgeek.com/sf
> _______________________________________________
> Nagios-users mailing list
> Nagios-users at lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/nagios-users
> 


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