Notification Escalation - [Suppression Workaroun Posted]

[uaslam]

Currently we have critical notifications going out to a Primary on call pager and a Backup on call pager.
We need the first 2 notifications to only go out to Primary for ALL services... What is an easy way we could set this up?

The XI GUI only allows this for individual services.

[scottwilkerson]

I would set the primary to be the contact on all services, then create an escalation using the escalation wizard that sends the escalated messages to both the primary and secondary (you can choose more than one service at a time when running the escalation wizard.

Then you can remove the backup contact from all services using the Bulk modification tool in the CCM.

[uaslam]

Is it possible to setup service escalation for a service_group?

[slansing]

I believe you can only escalate hosts, hostgroups, and service groups at this time.

[uaslam]

I believe you can only escalate hosts, hostgroups, and service groups at this time.

So thats a yes for Service Groups?

[lmiltchev]

Actually, slansing meant to say: " I believe you can only escalate hosts, hostgroups, and services at this time." Sorry about that. You cannot setup escalations for service groups.

[uaslam]

I would set the primary to be the contact on all services, then create an escalation using the escalation wizard that sends the escalated messages to both the primary and secondary (you can choose more than one service at a time when running the escalation wizard.

What's an easy way to identify all service checks which have "primary" listed as a contact? (for the purpose of setting up an escalation)

[sreinhardt]

The easiest way that I would suggest is to grep through the service configs. Something like:

Code: Select all

grep -in 'primary' /usr/local/nagios/etc/services/*

The service config on the left of each line will show the service config name that has that contact. However that exact command will not show what specific service it is listed for. It will however add the line that primary is listed as a contact so at least you can go directly there.

Otherwise you can masquerade as that user, and provided they are not an admin most anything that is shown for hosts and services they will receive notifications for. The one exception would be if they are a contact for the host and not services, it will still show both, due to how permissions are handled.

[uaslam]

So my problem is this.. for example I have "/ Disk Usage" service check for multiple hosts. Some of which are in a staging environment (no notifications to on call, primary or backup). But production hosts with the same check have on call notifications. These service checks have primary listed as default contact in the service definition.

Now I want to setup escalation for only production host services (in other words where primary is listed as a contact).
But when I try to add a new service escalation policy, it gives me only one option for "/ Disk Usage" and Im guessing that this would apply to all services with that name and would therefore escalate notifications for staging hosts as well ?

Also, The staging hosts do still send out notifications to appropriate contacts but no on call notification or escalation is required.

[sreinhardt]

I believe you are correct. This should definitely apply to everything. The best option I can give you would be to redefine that service group and check. One for production and one for staging. While its not ideal to have to recreate these things, we would generally suggest to keep intermingling of production and testing to a minimum.