VmwareMonitor

This monitor is part of the VMWare integration that provisiond provides. It tests the power state of a virtual machine or host system. If the power state is poweredOn, the service is considered to be up; if it is poweredOff, the service is marked as down. By default, a power state of standBy will also mark the service as down.

If you use Distributed Power Management (automatic PDF download), you can change the standBy state (see the ignoreStandBy parameter).

Optionally, a service down event can be triggered if there are pending alarms in vCenter that have not been acknowledged or cleared.

Import hosts

The VmwareMonitor requires that hosts be imported using provisiond and the VMware import function. Meridian requires network access to the VMWare vCenter server and each of the monitored hosts.

The credentials for accessing the status of a VMware entity are collected from the appropriate vCenter instance using settings from ${OPENNMS_HOME}/etc/vmware-config.xml. This file also sets specific asset fields to be assigned to imported virtual machines and host systems.

The VMware integration populates the following asset fields via provisiond:

  • VMware Management Server

  • VMware Managed Entity Type

  • foreignId (which contains an internal VMware vCenter Identifier)

For more information, see VMWare Handler.

VmwareMonitor class

VmwareMonitor class

Class Name

org.opennms.netmgt.poller.monitors.VmwareMonitor

Configuration and use

Optional monitor-specific parameters
Parameter Description Default Value

retry

Number of retries before the service is marked as down.

0

ignoreStandBy

Treat the standBy power state as an indicator that a service is up.

false

reportAlarms

Check for unacknowledged vSphere alarms for a comma-separated list of severities (red, yellow, green, gray).

Blank

This monitor implements the common configuration parameters.

Example

The following is an example of a poller-configuration.xml VmwareMonitor configuration. With these settings, the monitor will go down if any unacknowledged vSphere alarms with either red or yellow severities exist for the managed entity:

Note that you must include the monitor section for each service in your definition.

<service name="VMware-ManagedEntity" interval="300000" user-defined="false" status="on">
  <parameter key="retry" value="2"/>
  <parameter key="timeout" value="3000"/>
  <parameter key="reportAlarms" value="red, yellow"/>
</service>

<monitor service="VMware-ManagedEntity" class-name="org.opennms.netmgt.poller.monitors.VmwareMonitor"/>