Disabling Service check when host is down.

[[email protected]]

Enhancement request;

I have posted messages whto the support forumj http://support.nagios.com/forum/viewtopic.php?f=7&t=9532 which have suggested the parent/child approach. Im not in favour of this approach as it create a substantial body of worek, plus it would create an additional admin overhead as machines were moved or configurations changes occured within the network.

One suggestion was very informative, showing in detail how Nagios operates and how thyis could be achieved using external scripts. http://monitoringtt.blogspot.ie/2011/05/nagios-service-checks-based-on-host.html. I will attempt to adopt this approach as a test. However, I am concerned that external scripts run over several hundred machines will incur a significant performance hit on our nagios boxes.

A better approach would be to have this as an option within the host/hostgroup configs ...

Suggestion:
Can a switch be added to the configuration of hosts and / or groups to disable service checks if the host is down and respond with Unknown, and of course to re-enable when the host is reachable again.

Regards
Liam

[slansing]

This should be able to be handled through event handlers as Abrist had suggested, this is something that I believe we are looking at as an addition and you can request this feature by submitting a feature request at:

http://tracker.nagios.com/

I can assure you we go over this list quite often.

[Mitchell]

I would love to see this config option. I opened a feature request a while ago...
http://tracker.nagios.com/view.php?id=297

Regards
Ashish

[slansing]

If it is a feature that has enough interest or value rest assured that one of the dev's will comment on your submission.

[ChristopherW]

Has there been any development of this? It'd be a remarkably useful feature (it'd be brilliant if it was included in core as well as XI).

[slansing]

I've just opened up an internal feature request, we shall see if it gains traction there!

[bmonaghancertica]

I too would like this feature. The issue is that if many hosts go down, it may cause CPU usage to run away on the Nagios server due to the constant checking of services on hosts that are down.

[tmcdonald]

I have already added this to Core: https://github.com/NagiosEnterprises/na ... 724930a3e7

It's an "all-or-nothing" approach, so it can't be enabled on a per-group basis.

[asardouk]

I have already added this to Core: https://github.com/NagiosEnterprises/na ... 724930a3e7

It's an "all-or-nothing" approach, so it can't be enabled on a per-group basis.

How can we use these changes ? does host services take the same status of the host ?

i'm using router wizard to monitor router interfaces. When a router goes down the bandwidth service stay with "OK" status. How can i proceed to make it take host status when it's down ?

[tmcdonald]

The services will remain in whatever state they were in before the host went down. I was considering adding some logic to change that state to a user-defined value, but the original intent of the feature was to cut down on CPU usage during outages so I wanted to keep it lean and minimal.

I might still do this if I get enough requests for it and have the time, but for now this is how it works.