Video Screencast Help
Symantec Secure Login will be live on Connect starting February 25. Get the details here.

How to Create Software Delivery in Altiris

Created: 14 Jan 2009 • Updated: 14 Jan 2009 | 1 comment
Language Translations
Raman's picture
0 0 Votes
Login to vote

Using Software Delivery Solution we can delivery any MSI/Wise Package to any number of clients at a time. Before creating a delivery task, we have to create the packages. Common package creation is explained below:

To create a new package in Software Delivery, go to Resources > Software Management > Software Delivery Package > Windows right click on Windows, click on New > Software Delivery Package

Package Tab

  • Name: This field is how this “package” will be named across the Altiris Console, including any Wizards or other selection dialogs. Give a name to this Package.
  • Description: Here we can give a description of this package, give any thing which describes about the package we are creating.
  • Publisher: This label will only be used by the Software Portal.
  • Version: By Default it will be 1.0, and we can change this manually if required.
  • Package Source: This sets the method the Notification Server will obtain details about the Package. Each option is described below. Note that this does not necessarily provide how the Altiris Agents will access this package.
    • Package does not contain source files—No package source means no download will occur for this package. This is typically used for actions against items that already exist on the destination systems (that is, run a start command for antivirus services).
    • Access package from a local directory on the NS computer—As described, this option uses the local computer on the Notification Server for the package source.
    • Access package from an existing UNC—In the format of \\servername\share, a UNC can be used as the package source.
    • Note: This UNC will be used by all Altiris Agents in obtaining the package.
  • Access package from a URL—Note the screenshot below. Two fields appear when using this option. Package Location should be the actual URL the package resides in. The Package directory requires a UNC to the same location for snapshot creation purposes. This is required to manage the snapshot of the package.
  • Package Location: Defines the actual location of the package. Refer to the definition of a package for how that location will be used.
  • Installers found: This option is only used with PCT Solution (no longer being developed), and used by Software Migration tasks. In common uses this field will not affect anything about the Software Delivery Package.
  • Package files will be deleted from the client computer if unused for: This drop-down controls how long unused packages will remain on the client computers before deletion.

    Programs tab

    The following is a detailed explanation of Programs configuration, including an understanding of what underlies each option on the Programs screen. Multiple Programs can be created for a single “Package.” This view shows the first part of a standard Software Delivery Solution Package configuration page:

    • Name: This is the name of the action item, or Program. This name will be shown in the Software Portal configuration page and all other selectors dealing with the setting up of Package/Program items.
    • Description: For personal use. Use as desired.
    • Applications and Installer: These items are only used with PCT Solution, and limitedly with Migration Tasks. For most use cases these are not used.
    • Command line: The command line runs against the working directory. Please see Best Practices for information on how to configure this.
      Note: The command line controls exactly what occurs on the client. It is important to understand how the Altiris Agent treats this, and some of the caveats of this.
      • TEST IT: Always test your command line to ensure it works correctly. It should never:
        • Have a screen that requires user interaction that is user prompts for install configuration.
        • Require any user interaction of any kind that is Informational boxes or warnings that require an OK button to be clicked.
      • Command-line switches must contain appropriate input data to avoid user-input prompts. A quiet or unattended switch should commonly be used.
      • Most Executions without command-lines will require user prompts.
      • If your package contains multiple folders, using the Working Directory field is essential to put the execution in the right place. By default the execution occurs in the root directory of the package.

      Reboots should not be conducted by the MSI or EXE you are executing. The After running actions should be used for reboots

    • Working Directory: The two main uses of this field are:
      • Executing a file within a subfolder in the Package.
      • On a “No Source Files” package and the execution occurs from a local location on the target computer, use this to specify the directory on the local client the execution should occur.
    • Success Codes: Place all codes you wish to trigger a “Successful” status. For example, error code 3010 means the installation was successful, but a required reboot did not occur.
    • Failure Codes: This field is unnecessary. All codes not found in the Success codes field will automatically be labeled a “Failure.”

    Second Section of the Programs Configuration Page

    • Estimated Disk Space: This field does not change functionality in any way but can be used for size-tracking purposes. Edits to this field are always manual.
    • Estimated Run Time: This field does not affect the execution, but can be used for tracking purposes. All edits to this field are manual.
    • Terminate After: If left blank, or if set to 0, the terminate time will be 360 minutes (6 hours). This is a default Altiris Agent setting added to avoid multiple processes hanging for reoccurring tasks. The value always represents minutes. The Altiris Agent will kill the Main task and any child tasks spawned from the parent when the signified time runs out.
    • After Running: This configuration item only occurs after a successful completion of the Task associated with the Program. See the descriptions of these options below:
      • No action required: Self explanatory.
      • Restart Computer: After the Task receives a return code, the computer will initiate a reboot.
        Note: This will occur whether the code was a success or failure. If the execution fails, it will still reboot the system. Note that the deferral option (in minutes, shown in the following screenshot) configuration ties to the Warn option in a Task.
      • Log off user: The configuration looks like the one above, but merely logs the user off instead of rebooting. The deferral options are identical.

    Third Section of the Programs Configuration Page

    • Starting Window: This option only comes into play when the option User Input Required is checked. If unchecked, the execution will not be visible on the logged-on user’s desktop.
    • Run with rights: The execution of the task will load a limited user stack for the user specified. System Account works well for most executions, and a specified user can be used with Domain user accounts for specific admin tasks if the System Account has been locked down. The logged on user option uses the user who is logged on, but still loads its own stack for the execution.
    • Program can run: Select the option that best fits the action required. If the install requires a user to be logged on, choose that option, and so on.
    • Note: Tasks that run when no user is logged on will not show no matter how the execution is configured.
    • User input required: This option is more than what it seems. This option will pipe the execution to the user’s desktop, making it visible. It also loads a more robust user stack, so certain installations will benefit from having this option checked if the installs have user-related install items; that is, an MSI install might install to the CURRENT USER registry hive, so it only installs some of the components for the user whose context is being used for execution. By adding the ALLUSERS=1 or similar input command, the MSI can install for all users. However without the User input required option checked, the MSI fails due to the incomplete user stack loaded during execution.
    • Run once for: The two options, each logged on user, and Specified users or groups will run an install for multiple users. This is only available when the option Run with rights is set to ‘Logged on user’.

    Advanced tab

    This tab controls Package Server configuration and provides additional download options.

    • Agent display name and Agent display description: These two options change how this package shows up in the Altiris Agent UI. If you use the default, the Package name will be used for the Name.
    • Enable verbose reporting of package status events: By disabling this option, this stops the event type AeX SWD Package from being sent to the server. If the server load is heavy on the NS, disabling this type of events can benefit, though all knowledge of the download procedure for this package will not exist at the NS.
    • Use Alternate Download Destination on Client: The default location for downloading SWD Packages is C:\Program Files\Altiris\Altiris Agent\Software Delivery\{<GUID OF PACKAGE>}\cache. This location will always be used for Snapshot management. Note the following:
      • The Altiris Agent will duplicate the package in both the default cache directory and the alternate location.
      • The package is copied over once full received to the cache directory.
      • The alternate location will not be updated until directly before the execution of the Task. This means updates to the package will be received to the cache location, and will not be updated until directly before a scheduled execution by a recopy.
      • Package Destination Location on package servers (leave blank for default): This works differently than the Alternate Destination for the Altiris Agent. The “Managed” package, including share creation and snapshot management, will be the Alternate location and no where else.
      • Assign Package to: These options control how the Client downloads the package. Package Servers will not be addressed in this document.
      • Apply, Cancel, and Update Distribution Point
      • When all changes have been made to the Package dialog desired, click Apply to save the changes or Cancel to discard changes

      Comments 1 CommentJump to latest comment

      DeborahKahmke's picture

      I would like more info about choosing the correct Package Source and examples of when you would choose each one to better understand the benefits of each. Also, more information on the success codes would be great. As in how to set different success codes for installed but not rebooted, installed and rebooted successfully, not installed but rebooted, etc.


      Deborah Kahmke

      Login to vote