3.5 Configuring the Driver Health Job

The Driver Health job evaluates the conditions for a driver’s health states and assigns the driver the appropriate state. It evaluates the conditions for the health states, assigns the driver with the appropriate state, executes any actions associated with the assigned state, and stores the driver’s transaction data. The default polling interval is one minute. The job also executes any actions associated with the assigned state. For more information about setting up health monitoring for a driver, see Monitoring Driver Health in the Identity Manager 4.0.2 Common Driver Administration Guide.

  1. Make sure you’ve already added the job to the driver set. If you haven’t, see Section 2.0, Adding a Predefined Job.

  2. On the General page in iManager, configure the following options:

    Enable Job: Leave this option selected unless you don’t want the job to run.

    Delete the Job After it Runs Once: Select this option if you want the job to run one time only and then be deleted.

    Servers: Select the servers where you want the job to run. Multiple servers are available only if the driver is running on multiple servers.

    Email Server: To monitor the job, you (or others) can receive e-mail notifications whenever certain results occur for the job. You configure e-mail notification on the Results page. However, to enable this e-mail notification to work, you must first specify the e-mail server that will be used to send the notifications. Click the button to locate and select the Default Notification Collection object or any other notfTemplateCollection object that defines an SMTP mail server.

    Display Name: Displays the name assigned to the job.

    Description: Displays the description assigned to the job.

  3. On the Schedule page, specify when you want the job to run:

    The Driver Health job is a continuously running job, meaning that it does not stop unless a health state action shuts it down or it is shut down manually. The job must run continuously to be able to support transaction data collection for use in Transactions History conditions.

    If the job does stop, it is restarted based on the schedule.

    Run on a Schedule: Runs the job on a daily, weekly, monthly, or yearly schedule. You can also specify a custom schedule; use the crontab standard when specifying a custom schedule. The default schedule checks every minute to see if the job is running. If the job is not running, it is started.

    Run Manually: Runs the job only when you initiate it through the Run Now option (see Section 5.0, Starting a Job).

  4. On the Scope page, add the driver sets or individual drivers you want the job applied to.

    By default, the job applies to all drivers in the driver set. You can, however, modify the list so that the job applies only to specific drivers in the driver set. You can also add other driver sets or drivers from other driver sets.

  5. On the Parameters page, select the type of action you want the job to perform:

    Login ID: This defaults to the login ID that was used when creating the driver job. You should change this only if you want the driver to authenticate by using different credentials.

    Login password: This is the password required for the login ID that you supplied in the Login ID field.

    Subscriber Heartbeat: Controls whether the Driver Health job does a heartbeat query on a driver's Subscriber channel before performing a healthcheck on the driver.

    Polling interval: Determines how often the job evaluates the conditions for the health states, assigns the driver the appropriate state, executes any actions associated with the assigned state, and stores the driver’s transaction data. The default polling interval is one minute.

    Polling interval units: Specifies the time unit (minutes, hours, days, weeks) for the number specified in the Polling interval setting.

    Duration sampling data is kept: Specifies how long a driver’s transaction data is kept. The default, two weeks, causes a transaction to be retained for two weeks before being deleted. A longer duration provides a greater time period that can be used in Transactions History conditions, but requires more memory. For example, to use a Transactions History condition that evaluates of the number of publisher reported events for the last 10 days, you need to keep transaction data for at least 10 days.

    To store transaction data for one driver every minute (Polling interval) for two weeks (Duration sampling data is kept) requires approximately 15 MB of memory.

    For more information about setting up health monitoring for a driver, see Memory Requirements for Driver Health in the Identity Manager 4.0.2 Common Driver Administration Guide.

    Duration units: Specifies the time unit (minutes, hours, days, weeks) for the number specified in the Duration sampling data is kept setting.

  6. On the Results page, define the actions you want performed based on the results for the job.

    Each time the job runs, it generates Intermediate results and a Final result. The possible results are Success, Warning, Error, and Aborted. Intermediate results are generated at various points throughout the job. The Final result is generated when the job is finished. For each result, you can specify the action you want performed when it occurs: 1) generate an event for Novell Audit or Novell Sentinel and 2) generate an e-mail notification. For example, you might want no action to occur for an Intermediate Success result and an e-mail notification to be sent for an Intermediate Error result.

    To define an action for a result:

    1. Click the No action link next to the result to display the Result Notification dialog box.

    2. Select Audit result if you want to generate an event for Novell Audit or Novell Sentinel.

      or

      Select Send email, then fill in the recipient and e-mail template information.

    3. Click OK to save your changes.

  7. When finished configuring the job, click OK to save your changes.

    The job is added to the job list.

  8. Select the job in the list (by selecting the check box next to the job name), then click Get Status.

    The Job Status dialog box displays any configuration errors. Because the job requires rights to the driver object and those rights have not yet been granted, you see an Insufficient rights to driver object error.

  9. Click Grant rights, then click OK to confirm the action.

  10. If other errors are displayed, resolve the errors. Otherwise, click Close to close the Job Status dialog box.