Triggered alarm does not clear when the monitored value is no longer above the threshold for the alarm

Products

• VMware vCenter Server

Issue/Introduction

This article provides workarounds for an alarm that has been triggered, but has not cleared even though the monitored value is no longer above the threshold for the alarm.

Environment

VMware vCenter Server 5.0.x
VMware vCenter Server 4.0.x
VMware vCenter Server 4.1.x
VMware vCenter Server 5.1.x
VMware VirtualCenter 2.0.x

Resolution

This issue has been resolved in vCenter Server 4.0 Update 1. For more information see the VMware vCenter Server 4.0 Update 1 Release Notes.

If you can not upgrade to vCenter Server 4.0 Update 1 or to workaround this issue on other releases of vCenter Server, see the following workaround options:

Note: Before you begin please refer to KB1003490 for important information on restarting the mgmt-vmware service.

Symptoms

Alarm was triggered by high CPU usage and did not clear after CPU usage was down.

Workaround #1

To reset the Alarms Reporting Tolerance:

1. Click on the host that has the alarm.
2. Click Alarms. Record which alarm is still persistent.
3. Click Definitions.
4. Select the field the CPU Usage alarm has for "Defines In" (e.g. Datacenter) - click Datacenter
5. Right-click the alarm and click Edit Settings.
6. Click the Reporting tab and set the value 0 for Tolerance and 180 for Frequency. Lower the value if you want or set the value back to the original setting.

Workaround #2

Log in to the ESX Server service console as root, and run the following commands:

# service mgmt-vmware restart
# service vmware-vpxa restart

After you have run the commands, the alarms start to update properly.

For more information on working with Alarms, see Working with Alarms in the Basic Administration Guide.