10.5 Agent Self Monitoring

In addition to monitoring the operating system, server hardware, and application resources, most organizations find it useful to monitor the operation of the AppManager components themselves. You can choose from three basic methods for monitoring the operation of AppManager agents on your Windows agent computers:

  • Run the NT_RemoteServiceDown on one or more agent computers to remotely monitor the NetIQ Corporation AppManager Client Resource Monitor and NetIQ Corporation AppManager Client Communication Manager on other agent computers by listing the agent computer names for the Machine list parameter and netiqmc.exe,netiqccm for the Services parameter.

  • Run the AMAdmin_AgentSelfMon Knowledge Script to monitor the status of the scripting engine and other low-level components that the Client Resource Monitor uses to ensure the agent is running jobs properly.

    When you run the AgentSelfMon Knowledge Script, the Client Resource Monitor sets a timestamp in the Windows registry at each interval. At subsequent intervals, the Client Communication Manager compares the timestamp value with a threshold that specifies the maximum amount of time, in seconds, that can elapse between timestamps. If the age of the timestamp value exceeds the threshold you specify, the Client Communication Manager (netiqccm.exe) automatically restarts the Client Resource Monitor (netiqmc.exe). If the timestamp is within an acceptable range, the job simply updates the timestamp value and waits for the next iteration.

  • Run the AMAdmin_AgentHealth Knowledge Script to monitor the Windows Application log for events generated by the Client Resource Monitor and the Client Communication Manager that indicate general, communication, job, security, or upgrade problems. Both services log specific “self-monitoring” information, which the AgentHealth Knowledge Script can check for. You can further filter log entries by specifying a combination of include and exclude strings for the Description field.

  • Run the AMHealth_HeartbeatWin Knowledge Script to monitor the heartbeat of the AppManager agent computer. A heartbeat is a periodic signal generated by an Appmanager agent computer to indicate that it is still running. For more information, see the module management guide on the AppManager Modules Documentation page.