To generate alerts related to CPU usage on your accounting virtual machines, you add a second symptom to your vRealize Operations alert definition after you add the first symptom. The second symptom is related to host memory usage for the hosts on which the accounting virtual machines operate.

Prerequisites

Add the virtual machine CPU usage symptom. See Add a Virtual Machine CPU Usage Symptom to the Alert Definition.

Procedure

  1. In the Alert Definition Workspace window, after you configure the Name and Description, Base Object Type, and Alert Impact, click Next.
  2. Configure the symptom related to host systems for the virtual machines.
    1. From the Select Symptom drop-down menu, select Metric / Property.
    2. From the Defined On drop-down menu, select Self.
    3. Click Create New to add new symptom.
  3. Configure the host system symptom in the Add Symptom Definition workspace window.
    1. From the Base Object Type drop-down menu, expand vCenter Adapters and select Host System.
    2. In the metrics list, expand Memory and drag Usage (%) to the workspace on the left.
    3. From the threshold drop-down menu, select Dynamic Threshold.
      Dynamic thresholds use vRealize Operations analytics to identify the trend metric values for objects.
    4. In the Symptom Definition Name text box, enter a name similar to Host memory usage above trend.
    5. From the criticality drop-down menu, select Warning.
    6. From the threshold drop-down menu, select Above Threshold.
    7. Leave the Wait Cycle and Cancel Cycle at the default values of 3.
      This Wait Cycle setting requires the symptom condition to be true for three collection cycles before the symptom is triggered. This wait avoids triggering the symptom when a short spike occurs in host memory usage.
    8. Click Save.
    The dynamic symptom identifies when the hosts on which the accounting virtual machines run are operating above the tracked trend for memory usage.
    The dynamic symptom is added to the symptom list.
  4. In the Alert Definition Workspace window, drag Host memory usage above trend from the symptoms list to the symptom workspace on the left.
    The Self-Host System symptom set is added to the symptom workspace.
  5. On the Self-Host System symptom set, from the value type drop-down menu for This Symptom set is true when, select Any.
    With this configuration, when any of the hosts running accounting virtual machines exhibit memory usage that is above the analyzed trend, the symptom condition is true.
  6. At the top of the symptom set list, from the Match {operator} of the following symptoms drop-down menu, select Any.
    With this configuration, if either of the two symptom sets, virtual machine CPU usage or the host memory, are triggered, an alert is generated for the host.

Results

You defined the second symptom set for the alert definition and configured how the two symptom sets are evaluated to determine when the alert is generated.

What to do next

Add recommendations to your alert definition so that you and your engineers know how to resolve the alert when it is generated. See Add Recommendations to the Alert Definition.