You respond to alerts for objects so that you can bring the affected objects back to the required level of configuration or performance. Based on the information in the alert and using other information provided in vRealize Operations , you evaluate the alert, identify the most likely solution, and resolve the problem.

As a virtual infrastructure administrator or operations manager, you troubleshoot problems with objects. Reviewing and responding to the generated alerts for objects is part of any troubleshooting process. In this example, you want to resolve workload problems for a virtual machine. As part of that process, you review the Alerts tab to determine what alerts might indicate or contribute to the identified problem.

The problem virtual machine is db-01-kyoto, which you use as a database server.

The following method of responding to alerts is provided as an example for using vRealize Operations and is not definitive. Your troubleshooting skills and your knowledge of the particulars of your environment determine which methods work for you.

Prerequisites

  • Verify that the vCenter Adapter has been configured for the actions in each vCenter Server instance.
  • Verify that you understand how to use the power-off-allowed option if you are running Set CPU Count, Set Memory, and Set CPU Count and Memory actions. See Working with PowerOff section in vRealize Operations Configuration Guide. .
  • Verify that the vCenter Adapter has been configured for the actions in each vCenter Server instance.
  • Verify that you understand how to use the power-off-allowed option if you are running Set CPU Count, Set Memory, and Set CPU Count and Memory actions. See the section on Working With Actions That Use Power Off Allowed in the vRealize Operations Information Center.

Procedure

  1. Enter the name of the object, db-01-kyoto, in the Search text box and select the virtual machine in the list.
    The object Summary tab appears. The Top Alerts panes display important active alerts for the object.
  2. Click the All Metrics tab.
    The All Metrics > Badge > Workload % generates a graph in the right pane that shows the workload is heavy.
  3. Click the Alerts tab.
    In this example, the alert list includes the follow alerts that might be related to the problem you are investigating.
    • Virtual machine has unexpected high CPU workload.
    • Virtual machine has unexpected high memory workload.
  4. In the upper left pane, select the vSphere Hosts and Clusters related hierarchy and select ancestor or descendant alerts to add to the list.
    You want to check for possible alerts on ancestor or descendant objects in the context of the selected hierarchy.
    1. On the toolbar, click Show Ancestor Alerts and select the Host System and Resource Pool check boxes.
      Any alerts for the host system or resource pool related to this virtual machine are added to the list.
    2. Click Show Descendant Alerts and select Datastore.
      Any alerts for the datastore are added to the list.
    In this example, there are no additional alerts for the host, resource pool, or datastore, so you begin addressing the virtual machine alerts.
  5. Click the Virtual machine has unexpected high CPU workload alert name.
    The Alert Details Summary tab appears.
  6. Review the recommendations to determine if one or more suggested recommendations can fix the problem.
    This example includes the following common recommendations:
    • Check the guest applications to determine whether high CPU workload is expected behavior.
    • Add more CPU capacity for this virtual machine.
  7. To follow the Check the guest applications to determine whether high CPU workload is expected behavior recommendation, click Actions on the title bar and select Open Virtual Machine in vSphere Client.
    The vSphere Web Client Summary tab appears so that you can open the virtual machine in the console and check which applications are contributing to the reported high CPU workload.
  8. To follow the Add more CPU Capacity for this virtual machine recommendation, click Set CPU Count for VM .
    1. Enter a value in the New CPU text box.
      The default value that appears before you provide a value is a suggested value based on analytics.
    2. To allow the action to power off the virtual machine before running the action if Hot Add for CPU is not activated, select the Power Off Allowed check box.
    3. To create a snapshot before changing the virtual machine CPU configuration, select the Snapshot check box.
    4. Click OK.
    5. Click the Task ID link and verify that the task ran successfully.
    The specified number of CPUs are added to the virtual machine.

What to do next

After a few collection cycles, return to the object Alerts tab. If the alert no longer appears, then your actions resolved the alert. If the problem is not resolved, see User Scenario: Investigate the Root Cause of a Problem by Using the Troubleshooting Tab Options for an example troubleshooting workflow.