Getting hosts to not show up on the dashboard

[highness]

How can I have some hosts not show up in the alerts without making the services inactive?

We typically get a list of Windows hosts that need to be added to monitoring - we add them, but typically the firewall rules aren't in place to allow access to port 12489 until the host goes live.

So, those hosts are alerting for a few days (or if a project gets delayed, weeks). What is the best way of moving those hosts out of the monitoring dashboards without making the checks inactive?

Our initial way of thinking was to create a service group of pending and move the services to that service group, but the services are still alerting on the main dashboard.

[Box293]

You could create service dependencies. Create a service that checks the NSClient++ version. If this service is down then all of the other services will not execute / notify.

If you have XI 5 then you can use the new nagios directive called host_down_disable_service_checks.

• It’s best described as automatic service dependencies on their own hosts.
  System wide setting, this applies across the board, it is not granular.
  Define in /usr/local/nagios/etc/nagios.cfg
  host_down_disable_service_checks=1
  Can also be defined via CCM > Advanced > Nagios Core Main Config
  Restart nagios monitoring engine or Apply Configuration for setting to take effect
  Can reduce the load on the XI host as plugins will not be executed if this host is down.
  Keep in mind that if the host is down then any defined service dependencies will be ignored.

I would also like to point out that your host check intervals should be tuned so that the host goes down before the services. I have this explained in detail here:
http://sites.box293.com/nagios/guides/c ... -intervals

With all of these options, once the host is down then these services will appear under the tactical overview as known on problem hosts. Because the serivces are still checking / failing while the host max_retries are preformed, you will still some failed service objects that won't alert, but will have a critical status.

Does that info help?

[highness]

That helps!

We haven't migrated to XI v5 yet, but plan on doing so in the February time frame (this is our busy time of the year).

Thanks!

[rkennedy]

Sounds good. I will leave this thread open for now, feel free to post back in a couple months when you migrate to XI 5!