To create an alert to monitor the CPUs for the accounting department virtual machines and monitor host memory for the hosts on which they operate, you begin by describing the alert.

When you name the alert definition and define alert impact information, you specify how the information about the alert appears in vRealize Operations Cloud. The base object is the object around which the alert definition is created. The symptoms can be for the base object and for related objects.


  1. From the left menu, click Configure > Alerts, and then in the right pane, click Alert Definitions.
  2. Click Add to add a definition.
  3. Type a name and description.
    In this scenario, type Acct VM CPU early warning as the alert name, which is a quick overview of the problem. The description, which is a detailed overview. should provide information that is as useful as possible. When the alert is generated, this name and description appears in the alert list and in the notification.
  4. From the Base Object Type drop-down menu, expand vCenter Adapter and select Host System.
    This alert is based on host systems because you want an alert that acts as an early warning to possible CPU stress on the virtual machines used in the accounting department. By using host systems as the based object type, you can respond to the alert symptom for the virtual machines with bulk actions rather than responding to an alert for each virtual machine.
  5. Click Advanced Settings and configure the metadata for this alert definition.
    1. From the Impact drop-down menu, select Risk.
      This alert indicates a potential problem and requires attention in the near future.
    2. From the Criticality drop-down menu, select Immediate.
      As a Risk alert, which is indicative of a future problem, you still want to give it a high criticality so that it is ranked for correct processing. Because it is designed as an early warning, this configuration provides a built-in buffer that makes it an immediate risk rather than a critical risk.
    3. From the Alert Type and Subtype drop-down menu, select Performance under Virtualization/Hypervisor.
    4. To ensure that the alert is generated during the first collection cycle after the symptoms become true, set the Wait Cycle to 1.
    5. To ensure that the an alert is removed as soon as the symptoms are no longer triggered, set the Cancel Cycle to 1.
      The alert is canceled in the next collection cycle if the symptoms are no long true.
    These alert impact options help you identify and prioritize alerts as they are generated.


You started an alert definition where you provided the name and description, selected host system as the base object type, and defined the data that appears when the alert generated.

What to do next

Continue in the workspace, adding symptoms to your alert definition. See Add a Virtual Machine CPU Usage Symptom to the Alert Definition.