Configuring TIBCO Hawk

The configuration shown here is used both for configuring machine level metrics and for remote engine startup.

Procedure

  1. Install TIBCO Hawk and TIBCO Rendezvous software on the MM server machine, and on all client machines in the TIBCO BusinessEvents cluster whose machine-level metrics you want to monitor.
  2. Configure a Hawk domain. A Hawk domain specifies a group of TIBCO Hawk agents that acts as a monitoring set. Each machine has a Hawk agent and various micro agents (HMA) that provide useful machine-level metrics to the enterprise monitor.

    Use the same Hawk domain name and Rendezvous transport for all the monitored engines and for the emonitor application.

  3. Import the BE_HOME/MM/project/emonitor project into your workspace and edit the MM.cdd. If you copy files into the workspace, remember to copy the MM.cdd file to the above location. In the mm-class agent properties list, add the following property to specify the Hawk domain:
       tibco.clientVar.Domain=TIBCO Hawk Domain
  4. If you use non-default values for the Hawk transport properties, specify them in the mm-class agent properties list in the MM.cdd also. The properties are as follows:
       tibco.clientVar.TIBHawkDaemon=Rendezvous daemon used by Hawk
       tibco.clientVar.TIBHawkNetWork=Rendezvous network used by Hawk
       tibco.clientVar.TIBHawkService=Rendezvous service used by Hawk 

    To use the default client socket, omit the daemon argument. Default service is 7474, and daemon is tcp:7474. See Hawk documentation for additional information.

  5. Add the same properties you added to the MM.cdd file to the monitored project’s CDD file, in the Cluster tab properties sheet.
  6. In the BE_HOME/mm/bin/be-mm.tra file, set the tibco.env.HAWK_HOME property and the tibco.env.RV_HOME to point to the TIBCO Hawk and TIBCO Rendezvous installation root directories.
  7. In the TRA files of all monitored cluster engines, set the properties tibco.env.HAWK_HOME and tibco.env.RV_HOME.