Okay, that makes sense.
What about this idea then... (some sort of flag for "execute this
service test on the standard nagios retest policy for service
failures").
define host {
host_name linux-server
address 10.0.0.1
parents linux-server2
hostgroups web-servers
}
define service {
service ping-linux-servers
check_command_line ping ...
execute_on_failure true
}
define service {
service smtp-linux-servers
check_command_line
execute_on_failure false
}
define host_service {
host_name linux-server
service smtp-linux-servers
contact_group linux-admins
}
Thanks
Craig
On Fri, 2005-06-17 at 17:36 -0400, Subhendu Ghosh wrote:
>
> The model for hosts and services exist separately such that if one of a
> number of services fails on a host, the host check command is used to
> verify if the host is up/down before checking all the other services.
>
> Similarly if a host is down, parent hosts are checked to see if the path
> to host is down or just the target host. This is also done using the host
> check_command.
>
> So while it might look redundant from a data model perespective - it
> actually is quite useful.
>
> -sg
This post was automatically imported from historical nagios-devel mailing list archives
Original poster: [email protected]