Volker Maibaum wrote:
>>Brad O'Hara wrote:
>>
>>>
>>>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those options will easily become outdated, but also because of the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.
>
For this purpose hostgroups would be much better. The technical staff
can watch the hostgroup summary page and easily see if they need to
bother with calling in the expertise. The naming-convention suggestion
is very good too.
--
Andreas Ericsson [email protected]
OP5 AB www.op5.se
Tel: +46 8-230225 Fax: +46 8-230231
This post was automatically imported from historical nagios-devel mailing list archives
Original poster: [email protected]