On 30 December 2010 12:44, stan <[email protected]> wrote:
> I am still confused by why the service def's have a filed for hostname. As > you point out, services don't necisarily relate to hosts. They are > functions provided, or statuses of a machine. and I would think thta > servicegroups would be where the assoctation betwen the service, and > thehosts that you want to check for that service wouldbe. > > Am I missing something? > Oh, I see. No, I don't think you are missing anything. I read somewhere (probably Wolfgang Barth's excellent book) that in Nagios each service should be taken to be the combination of both the service_description *and* the host_name. Certainly in Nagios it is not possible to have a service which exists without a host. If you look at the objects.cache file (probably under /usr/local/nagios/var/) then you will see how Nagios stores each service definition. Even if you have defined services based on a hostgroup, Nagios will actually internally regard each service on each host as a separate entity - every service will have an entry for both host_name and service_description. The inbuilt service/host dependency and the host parent/child relationships are important for Nagios in determining when it should notify - in particular it won't notify for a service if the host is down and usually won't notify for a host if the parent host is down. The logic Nagios uses for determining dependencies and reachability make Nagios a very powerful tool for ensuring you only receive notifications for those hosts and services which are actually causing the problem, not all the hosts and services which are down as a result of the underlying problem. I agree with you it can be a little counter-intuitive at times. In particular if you only have one service (for example ping) associated with a host, it's not always obvious whether it would be appropriate to acknowledge the service or the host when the ping fails. ------------------------------------------------------------------------------ Learn how Oracle Real Application Clusters (RAC) One Node allows customers to consolidate database storage, standardize their database environment, and, should the need arise, upgrade to a full multi-node Oracle RAC database without downtime or disruption http://p.sf.net/sfu/oracle-sfdevnl _______________________________________________ Nagios-users mailing list [email protected] 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
