More sophistic notification

Sheeri Kritzer sheeri.kritzer at tufts.edu
Fri Jan 7 17:09:48 CET 2005


Dear Peter,

Nagios is very simple, and that's one of its greatest assets.  What you propose 
CAN be done.

Nagios checks are scripts that are run, some with parameters, that output one of 
four exit states (CRITICAL, WARNING, UNKNOWN and OK), and maybe other things.  
If you can write a script that outputs one of those 4 exit states, you can do 
anything in Nagios.  

In your example, you want to write a script that calls the checks for CPU/load, 
memory, and processes.  It should remember the exit value for each, and after 
the third check it should output one exit code for all three -- for instance, if 
2 or 3 checks return critical then your overall check should be critical.  There 
are many permutations, too -- when do you go WARNING -- when 2 or 3 are warning? 
 or when 1 is critical?  etc.

Anyway, I'd suggest writing the ubercheck to run all 3 checks, and then 
configure nagios to only use the ubercheck (that one will call all 3 checks).

This is all in the documentation.  I don't know if it's a FAQ, but it probably 
should be.  (ie, "Is there a check that. . . ?)

-Sheeri Kritzer
Systems Administrator
University Systems Group
Tufts University
617-627-3925
sheeri.kritzer at tufts.edu


Quoting "Jevos, Peter" <Peter.Jevos at Oriflame-SW.Com>:

> 
> Hello 
> I've one question. Is Nagios capable to give more sophisticated
> notifications ? What I mean ?
>  Let's example. I'm monitoring on one machine few  values ( cpu, memory ,
> proccesses etc) There are plugins in Nagios which can monitor these values
> and send notifications if values are out of normal (warning, critical
> state). But I'd like to get only one notification if group of measure values
> are in critical state. To my example, one group of my measures values on one
> machine are: CPUload, MEM, Proccesses. So I'd like to get notification when
> all of these values are in critical state, not only two. The first purpose
> of this is that 2 values on one machine in critical state doesn't' have to
> be problem, but 4 or five is problem. The second purpose is to avoid to get
> false alarms cause CPU in critical state could be a normal in some
> environment and when you'll get hundreds notifications per day you will
> probably ignore it.
> So is this possible to define some group of alarms and on the basis of theme
> to send notification that host is in critical state cause all of measure
> values are in critical state?
> Thanx a lot for an answers and ideas
> peter
> 



-------------------------------------------------------
The SF.Net email is sponsored by: Beat the post-holiday blues
Get a FREE limited edition SourceForge.net t-shirt from ThinkGeek.
It's fun and FREE -- well, almost....http://www.thinkgeek.com/sfshirt
_______________________________________________
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