Steps to create a policy action for a TEM Task
The article discusses the differences between creating and deploying a policy action for a Task versus a Fixlet.
Resolving the problem
For most practical purposes, TEM Tasks and Fixlet messages work exactly the same, but this is one case where their behavior differs. Even if the 'Automated Application' option under the 'Execution' tab of the Take Action Dialog is selected, a Task will not re-apply by default. In order to get a Task to re-apply, you need to go to the 'Relevance' tab of the Take Action Dialog, and select the first radio button, which is 'Use the relevance clause from the original Task or Fixlet message'. By default, it will be set to the second radio button, which is 'Run this action on all relevant computers, and consider the action successful once all lines of the action have been executed'. Note that TEM Console Users who do not have custom content ability will not see the 'Relevance' tab, and will therefore not be able to create policy Tasks. However, it is EXTREMELY IMPORTANT that you do not run a policy action from a task that has relevance that does not falsify. This will cause the client to continually execute the action in an unending loop.
Since the concept of the 'Relevance' tab is a bit non-intuitive, it warrants a bit more explanation. First of all, TEM Relevance is an internal language that TEM uses to determine which Fixlets and Tasks apply to specific computers. A Task or Fixlet's action can only be run on computers that evaluate its relevance clause to 'TRUE'.
The 'Relevance' tab tells the TEM Clients how to determine whether their action statuses are 'Fixed' or 'Failed'. For Fixlets, the first radio button (Use the original Relevance clause...) is selected by default. When this option is selected, a computer will only report 'Fixed' for an action if it has executed the action script and its Relevance subsequently evaluates to 'False'. Tasks, however have the second radio button (...action successful once all lines have been executed) selected by default. This instructs computers to report 'Fixed' once they have completed their action script, regardless of their relevance. The system is designed this way because many tasks are meant to always be relevant. These tasks should never be run as policy actions. However, some tasks do have relevance that will correctly falsify. How can you determine if a Task has relevance that will correctly falsify and can therefore be used as a policy? First of all, tasks that have relevance that won't falisify are generally tasks you wouldn't want to re-apply anyway. However, if you'd like further confirmation and are unfamiliar with the Relevance Language, here's a simple test.
- Take an action from the Task in question and target it to a test computer.
- At the 'Relevance' tab, select the first radio button (Use the original Relevance clause...)
- Press 'OK' to execute the action.
More support for:
IBM BigFix family
Software version: All Versions
Operating system(s): Platform Independent
Software edition: All Editions
Reference #: 1505967
Modified date: 27 October 2015