Client Management Suite

I am trying to create a package that will upgrade Crystal Reports. After many attempts the best approach was to create a policy that runs a package delivery task to move over the contents into the drivers folder and then run a .bat file to execute the setup.exe. The Policy runs as follows: Pre-download detection check for CR - Not Detected Downloading package for CR - Success Package Delivery Task - Failed Execute install command for CR - Not Run The problem is the task actually works. I can see all of the files in the drivers folder. after the package delivery task completes. The odd thing is then if I go back into AeXAgentActivate and launch the Application Task manually it then returns a success code and moves onto running the .bat script. Is there a reason I am getting a failure code even though the task completes and is there a way for the execute install command to run despite the task returning a failure code?

Are you able to show us the contents of your .bat file that's producing this error code?

Sure I can attach a screenshot of the .bat file. The .bat file runs just fine once it gets to that point. It is the package delivery task that runs right before the .bat file that is the hang-up though.

Do you see log files for the installation in %TEMP% for the user this is being executed as (if SYSTEM, that will be C:\WINDOWS\TEMP). I'm wondering what the install log says when you get this error code.

I checked the %TEMP% folder, but I guess I am not sure what I am looking for. There was no log file or folder and nothing else had a modified date with todays date. I did check the Altiris agent logs and found this.

The command line is basically a call to CMD with your command line. CMD returns 1003 back to the agent, hence the failure - again, this is what CMD returned to us, so what it means is not necessarily what we show. Any non-zero return value is reported as failed. When you create or edit a software package, you can specificy success and failure codes - put 1003 as a success if that is what it means. Otherwise, troubleshoot at the command line and determine what is going on, and if needed, run a bat file that returns a 0 if it does what you intend it to do. Another option is to create a software package with your files and batch file, import it into a SW package, and run the policy as a standard Managed Software Delivery policy. You will also have the option to add success/failure codes for the package (found in the command line edit dialog). The main point is that when we interface with the operating system, we report success or failure on the return value from the OS, not what actually happened.