Nicholas Whiting wrote:
> Hi Everyone,
>
> I would like to overide the default manner in which Nagios
> populates the event loop, objects, etc from the .cfg files on
> startup. Rather than read from local .cfg files, the Nagios daemon
> would poll a centralized server which would return an xml encoded
> map of the necessary elements that would normally be contained
> in the .cfg files. This xml would be consulted in the future if
> the central server was inaccessible.
>
> The reason I favor this approach is that I manage 3000 servers and
> would like to a have central database store the nagios configs
> for every distributed nagios client without having to extract the
> related subset of configs and deliver them to the respective nagios
> instance.
>
Have each poller check a hostgroup of its own and then spend two hours
writing a script to pull just the things needed for those hosts out of
the config. It should be around 400 lines of perl or php, I think.
> Finally, my question!! Is this possible to achieve through a NEB module?
The server-side part, yes, but it's imo the wrong way to go since it
actually has nothing to do with Nagios operations, and only with its
initialization.
For the client-side part I think you'd have to jump through hoops to get
it to work.
> I would rather write this as a discrete module then have to monkey with
> nagios source.
>
Given the lack of NEB documentation you'd still have to fiddle with (at
least read thoroughly) the Nagios source.
--
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]