Submit a ticketCall us

Get a crash course on Network Monitoring delivered right to your inbox
This free 7-day email course provides a primer to the philosophy, theory, and fundamental concepts involved in IT monitoring. Lessons will explain not only how to perform various monitoring tasks, but why and when you should use them. Sign up now.

Home > Success Center > Network Performance Monitor (NPM) > Single alert with multiple conditions on the same object

Single alert with multiple conditions on the same object

Table of contents

Updated April 3, 2017

Overview

A single alert is desired when multiple conditions occur on a single object.  For example, one might want a single alert if both volume E and volume C on the same computer are over 70 percent full.  This same logic can be used to alert if the same application, present on two nodes of a manual HA system, are down.

Environment

All Orion Platform versions

Steps

  1. Set the following conditions to trigger the alert when both volume E and volume C on node dev-aus-xxxx-xx-swdev.local are over 70% full.  Note the advanced condition that the alert is triggered if 2 or more conditions are met:
  2. Add the advanced condition that the alert can be triggered if more or equal to 2 conditions are met. Without the advanced option, the alert will occur if the C drive is over 70% full OR the E drive is over 70% full.  However, with the advanced option, a single alert will be triggered when both conditions are true.

 

The 2 objects that must be fulfilled for the alert to trigger are:

  • If node name = dev-aus-xxxx-xx.swdev.local AND caption contains E:\ AND percent used is greater than 70
  • If node name = dev-aus-xxxx-xx.swdev.local AND caption contains C:\ AND percent used is greater than 70

 

One cannot use all ANDs in the condition, because the caption name will never be both C:\ and E:\ at the same time.

 

In the next example, the same SAM application is assigned to 2 nodes. The alert will be triggered when that application is not up on either of 2 nodes (manual HA solution).

 

In this case, the two nodes, sup-aus-xxxx-xx and sup-aus-yyy-yy, are not set up officially for failover.  The same application, annette wmi process monitor, is assigned to both nodes with a set of processes. Only one occurrence of this application is ever running at a time.  Because the application contains a set of processes, all processes in the application must be UP, for the application to be UP.  An alert is desired when the SAM application is not UP on either node.  This alert will also be triggered when just one of the processes of the application is not running on the active computer, and the application is not UP on the standby computer

 

The OR logic must be used since a single node cannot be named both sup-aus-xxxx-xx and sup-aus-yyy-yy. Also, the advanced option is used to trigger a single alert when neither machine has the application (which includes all processes in the application) UP.

 

The two conditions that must be fulfilled for the alert to trigger are:

  • If node name = sup-aus-xxxx-xx AND application name = annette wmi process monitor AND application status is not UP.
  • If node name = sup-aus-yyy-yy AND application name = annette wmi process monitor AND application status is not UP.

 

If you have HA officially configured for failover, it is best to write the alert against the virtual node in the HA.

 

 

Last modified
19:47, 4 Apr 2017

Tags

Classifications

Public