<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN"> <html> <head> <meta content="text/html;charset=ISO-8859-1" http-equiv="Content-Type"> </head> <body bgcolor="#ffffff" text="#000000"> Emmett Hogan wrote: <blockquote cite="mid41E42B68.4060406@Examen.com" type="cite"> <meta content="text/html;charset=ISO-8859-1" http-equiv="Content-Type"> <title></title> Okay...now I am REALLY confused. I do not understand why you would need to make all services dependent on a check_ping service when the Nagios Documentation states: By adding the dependency on a check_ping service aren't you adding another level of complexity that is already handled by default?? </blockquote> Yes, you are, unless you're doing something else for your host check (or you've disabled them). In our case, we can't ping. We can, however, SSH from the Nagios server to (most) of the remote boxes. Our host check is, therefore, a check_ssh module check (is SSH running on remote host? If so, host is up. If not, host is down). So we wrap a bunch of modules in an SSH tunnel, and any checks that require this SSH connectivity that fail will cause a host check to be performed, which checks SSH connectivity. So we don't have to make a dependency upon ssh services. Short answer: Don't make services dependent upon network connectivity. Nagios handles that. </body> </html>