some feature requests

[Volker Maibaum]

> Brad O'Hara wrote:
> > In line.
> > 
> > Andreas Ericsson wrote:
> > 
> >> Volker Maibaum wrote:
> >>  
> >>
> >>> Hi,
> >>> I have some feature requests for nagios:
> >>>
> >>> - set contactgroups on host basis for services. So that if no
contact is
> >>> set for a service check the host contact group is used. That would
make
> >>> configuration a lot easier.
> >>>
> >>>     
> >>
> >>
> >> This I like.
> >>
> >>  
> >>
> >>> - It would be nice if it would be possible to categorise the
criticality
> >>> of services and hosts. E.g. critical / uncritical / unimportant.
So that
> >>> the operating could easily decide if they have to call somebody
during
> >>> midnight or if the problem can wait till the next morning.
> >>> I know this could also be done by host and servicegroups, but it
would
> >>> be nicer to have this as a parameter. Depending on the category
the
> >>> host/service could be highlighted with different colours in the
> >>> web-frontend.
> >>>     
> >>
> >>
> >> This I don't. It's very, very simple to make the right contacts
the 
> >> contacts of critical hosts and services and have any script you
want 
> >> do the actual contacting, so adding complexity to the core to give 
> >> another way of doing the same thing doesn't sound very useful to
me.
> >>
> >>   
> > 
> > True enough.  Our problem is that we have operations staff monitor 
> > Nagios and notify people off hours based on how critical the
"resource" 
> > is not the severity of the check.
> > 
> 
> Ok. Then you're not using the first alternative, which would be to 
> create a notification script for the contacts supposed to be on call
for 
> the night. Nagios is fully capable of dialing your technical staff
and 
> telling them that "this and this unit isn't functioning". I think
you're 
> also aware that you can use several different notification-commands 
> (they don't have to notify, they can do anything you can make a
computer 
> do, really), and this in combination with the possibility of setting 
> several or different contactgroups for each host and service is, imo,
a 
> better solution than to implement configuration options. Mostly since 
> those options will easily become outdated, but also because of the
need 
> to add more bloat to the core of Nagios.
> 

The problem is not only about notification. I could implement two email
scripts, one that writes in the subject "critical" and one that writes
"not so important". But if the staff looks at the web-frontend of nagios
they don't see any difference. A "kernel panic" appears in the same way
as "the soda machine is out of water". The operating staff that looks at
nagios does not now that the soda machine is not critical. But they have
to decide if we have a critical situation and if they have to call an
expert.
 
regards, 

Volker	

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