Deployment Solution

I believe this issue started to occur duing the latest HF that was released...I believe it was HF2.  I am not completely sure. 

The issue occurs when trying to capture an image. We have gone through all the normal steps to try and remediate it, but it appears it keeps happening and I was wondering if anyone can assist me.

Error Message:
"Windows could not parse or process the unattend answer file for pass [specialize].  The settings specified in the answer file cannot be applied.  The error was detected while processing settings for component [Microsoft-Windows-Shell-Setup]."

Steps Taken:

• Recreated task in Altiris Console
• Cleared out all the files in both "panther" folders
• Ran sysprep manually from the Windows 10 machine
• Modified the three registry settings that was done in Windows 7 as a test

I believe there are a few other steps, but anyone seen this issue or a few other steps we can try?

Look at the log files in c:\windows\panther to start with. Then maybe c:\windows\system32\sysprep and c:\windows\setupact.log.

Post up your XML here (Clear out any company / personal information) and maybe we'll be able to spot something.

Would you be able to tell me which XML file you're referring to, so that I am not posting the wrong one? :)

Sure :) 

Unattend.xml would probably be the best to start with.

Have you validated the answer file in System Image Manager? 

Thank you for the software.  I will definitely keep this in mind when working with my unattend.xml.

I actually looked online and it appears the local unattend.xml file was missing the "key" information.  I filled that information out, captured it, and it appeared the error went away.  I am not sure why the sysprep wasn't generating a good unattend.xml file, but manually adding the key and then running sysprep seemed to have worked.  Thanks!

Yay! Glad you figured it out.

This is an HF3 issue and I would log a support ticket on this but the support site is still broken at the moment.

HF2 worked perfectly and now the unattend.xml has missing information.

It was missing information in the "specialize" pass. I had to add MAC addresses and "true" statements for DHCP in a couple of places.

This got rid of the error message but it tried to log in with the disabled local admin account.

I then noticed that the entire tail of the section which involves creating a local "Ghost" account which then runs the driver installation script was also omitted.

I would advise users to NOT upgrade to HF3 at this time until this is resolved.

EDIT: I should further clarify that the issue only seems to affect a brand new, pre-defined computer. If you instll the agent onto an existing production OS and allow it to upload its basic inventory it can be built correctly, and re-built as well.

However when you add a pre-defined computer with just computer name and MAC address and PXE boot this machine, as soon as the agent connects and is ready to accept jobs you will see the MAC address disappear and the IP address appear in its place. This does not happen to a machine that has been seen in production before.