Using Nagios to monitor "service-less" hosts

[Tedman Eng]

> -----Original Message-----
> From: Andy Shellam (Mailing Lists)
> [mailto:andy.shellam-lists at mailnetwork.co.uk]
> Sent: Tuesday, November 07, 2006 8:44 AM
> To: Nagios Users mailinglist
> Subject: Re: [Nagios-users] Using Nagios to monitor 
> "service-less" hosts
> 
> 
> Personally I've not noticed any performance decrease having 
> host checks 
> set - it's on a moderately spec'd server, and if a gateway 
> does go down, 
> I know about it almost immediately and can take action.
> 
> JOOI, why is it said that a host check degrades performance?  
> All it's 
> doing is running a command like a service? 

Host checks preempt all other check operations, they are not performed in
parallel like service checks are.  In a small monitoring environment (less
than a few hundred service checks), this may have little noticable impact.
But if you watch the scheduling queue closely while performing some "host
down" experiments with Nagios 2.0 active host checks enabled, you'll see the
adverse effects firsthand.  High service check latency compounds as more and
more hosts are actively checked.

It's generally a better solution to set up a ping service, even if it's the
only service (if it's responding to a request, then it's considered a
service)

An good analogy is to think of hosts as "containers" for the services.  Host
checks are tests to see if the entire container is missing.

-------------------------------------------------------------------------
Using Tomcat but need to do more? Need to support web services, security?
Get stuff done quickly with pre-integrated technology to make your job easier
Download IBM WebSphere Application Server v.1.0.1 based on Apache Geronimo
http://sel.as-us.falkag.net/sel?cmd=lnk&kid=120709&bid=263057&dat=121642
_______________________________________________
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