+

Search Tips   |   Advanced Search

Enable the on demand router to work with IBM Enterprise Workload Manager

Use this task to enable IBM Enterprise Workload Manager™ to monitor Intelligent Management.

To perform this task, install IBM Enterprise Workload Manager. See the IBM Enterprise Workload Manager information center for installation instructions.

By performing this task, the on demand router (ODR) makes Application Response Measurement (ARM) calls to report response times to IBM Enterprise Workload Manager. IBM Enterprise Workload Manager can use the response time information to monitor the environment both inside and outside of the Intelligent Management domain. See Intelligent Management and IBM Enterprise Workload Manager.


Tasks

  1. Configure service and transaction classes in IBM Enterprise Workload Manager. The service and transaction classes that we configure in IBM Enterprise Workload Manager override any service classes configured in Intelligent Management. See the Enterprise Workload Manager information center for more information.

  2. Configure IBM Enterprise Workload Manager to manage Intelligent Management. After you complete this step, any service policies configured in Intelligent Management are not valid. IBM Enterprise Workload Manager has its own configured service and transaction classes.

  3. Copy the ARM libraries from IBM Enterprise Workload Manager to the Intelligent Management configuration. Include the ARM libraries from IBM Enterprise Workload Manager in install_root/bin. Remember the name of this class so that we can supply the information in the administrative console. Alternatively, we can add two custom properties to the application server or ODR Java virtual machine (JVM). First, define the java.library.path custom property and set its value to the IBM Enterprise Workload Manager library path, for example c:\IBM\VE2\EWLMMS\classes\ms. Next, create the ws.ext.dirs custom property and set its value to the folder containing the ARM libraries provided by IBM Enterprise Workload Manager, for example, c:\IBM\VE2\EWLMMS\classes\ARM.
  4. Enable ARM reporting. On the request metrics administrative console page, use the following settings:

    • Select the appropriate application component, such as servlets or Web services, that needs to report ARM statistics.
    • Define the trace level is optional.

    • Enter the name of the IBM Enterprise Workload Manager ARM factory in the ARM transaction factory implementation class name. The transaction class factory name for IBM Enterprise Workload Manager is com.ibm.wlm.arm40SDK.transaction.Arm40TransactionFactory.

    See getting performance data from request metrics.

  5. Enable ARM reporting on your other tiers and applications, such as IBM DB2, IBM HTTP Server, and also for the various components running on WebSphere Application Server. We can enable ARM instrumentation in specific WAS components by selecting the appropriate components. See getting performance data from request metrics. See the product documentation for our specific tier applications regarding more information on enabling ARM reporting. Any tiers that send requests to the Intelligent Management tier must be ARM enabled to facilitate the interaction between IBM Enterprise Workload Manager and Intelligent Management. Any tiers that receive requests from the Intelligent Management tier do not need to have ARM enabled, however, enabling ARM on these tiers is strongly encouraged.

  6. To enable WAS and Intelligent Management with IBM Enterprise Workload Manager, we must manually update the WAS server.policy files. We must update the server.policy file for all application servers and on demand routers. See the server.policy file permission.

  7. Restart all applications and servers that IBM Enterprise Workload Manager is now monitoring.

The on demand routers in the Intelligent Management infrastructure will begin to make ARM calls to report response times on requests. TheIBM Enterprise Workload Manager can monitor the Intelligent Management environment.


Subtopics


Related:

  • Intelligent Management and IBM Enterprise Workload Manager
  • Configure the autonomic request flow manager
  • Getting performance data from request metrics
  • server.policy file permissions