From: <bjo...@em...> - 2004-05-24 09:36:56
|
> I am still trying to figure out what to do about passive host checks,=20 > as there are a number of scenarios where they will not function as=20 > people would expect. This mainly affects distributed monitoring,=20 > where the host topology on master and slave servers does not look the=20 > same. For this reason, passive host checks would probably have to be=20 > tweaked so that DOWN states get translated into UNREACHABLE states=20 > (and vice versa) where appropriate. I'm still not sure how or if I=20 > should implement that internally in Nagios. When you implement this translation, it might be a good idea if=20 it could be turned off, eg. "use_distributed_translation=3D0" in=20 nagios.cfg. This is to be able to use an all-passive nagios top node=20 (i.e central nagios host) as a dummy that reflects the nagios leaf nodes' (i.e distributed nagios hosts') world view. :) Bj=F8rnar |