children, parents, grandparents, great grandparents...

[jwgill]

I'm monitoring a remote site.

First, checking that the private router endpoint between the main site and the remote site is up.

Then for all the hosts at that site, I set that router as the parent. Below that router, I have a vm host and have set the vm guests to reference the vm host as their parent. Below that i have a further set of ... and below that ... and ....

Basically children, reference parents. Those parents reference parents,.. those ....

Do I understand this properly, that when one of the parent hosts is down, all children in the tree, should become unknown/unreachable?

[scottwilkerson]

Do I understand this properly, that when one of the parent hosts is down, all children in the tree, should become unknown/unreachable?

They do not become unknown, but if their check host check returns a non-OK state it would be unreachable because the parent was down. This will cause all of this hosts children to not send notifications, however they still perform their checks as normal and will report the results in the UI.

[jwgill]

When the parent hosts are unreachable, the individual service check will still continue to test?

[scottwilkerson]

When the parent hosts are unreachable, the individual service check will still continue to test?

Yes, but notifications are suppressed.

There is a nagios Core configuration setting that can be set in the nagios.cfg that changes this behavior if you set the following

Code: Select all

execute_service_checks=0

https://assets.nagios.com/downloads/nag ... gmain.html