VERY new to using USMT.
I am doing one-at-a-time migrations and leveraging a tool called User Profile Central (UPC) which is a GUI interface for those not doing mass migrations and using USMT/SCCM.
The interface lets me choose a computer on the network. See and Select the profiles I am interested in capturing. And referencing the .XML files available to facilitate the migration. So all the rules of how USMT utilizes .XML files apply.
SCOPE:
I only want to migrate a few key settings. (14 in all) Some Windows settings like Mapped Drives, a few office settings, etc. The owner of the UPC software included creating a custom .XML file with a list of the few Profile Folders (for like sticky notes, one note, etc from each profile) and a few Registry settings ( like mouse settings, KB, Accessibility...etc) 14 settings in all that I am looking to migrate during what we call a Life Cycle Replacement (LCR)
BASE UNDERSTANDINGS (Or lack of therein)
I like rendering things down to their most simple components. Forgive me if the following is very basic.
So reading the USMT and .XML documentation if I am correct, the MigApp and MigUser are always used for the collection to give a roadmap of everything currently on the computer.
You use a custom.xml to remove anything from the MigAppps and MigUser you do not want to be collected AND to also add to the collection anything outside of the scope of the MigApp and MigUser reference.
Put in another way everything is looked to be migrated that is found using the two .XML files as a reference. (Inclusionary for those who want a full on data and app settings migration) Then once that is established, you figure out what to remove (Exclusionary to trim away whatever not needed)
QUESTIONS:
Goal. Excluding EVERYTHING except a few basic settings that are on the custom.xml
- Do you always include MigApps and MigUser on all collections? (IE: Needed as the reference)
- Can you edit the MigApps or MigUser to include/not include anything in the scan or or are they kind of sacred and you leave them alone and use the custom.xml to become the gatekeeper of what is ignored or added to the collection? My mindset for asking is, if you do not look for a thing you do not need, the scan would go a lot faster. And if you include what you ARE looking for, no need to customize a separate .xml. I really need this explained in crayon. I am a solid technician when I understand what is happening and none of the documentation is granular enough to figure that out or explained in a way that my mind can grasp. Also is there maybe a flow chart that shows exactly what happens, what files are referenced and in which order?
As I have read it. Scanstate takes the two reference .XML files. Looks at the computer with that template in hand and makes ready to collect ALL of it. Somehow the custom .XML is referenced and removes anything referenced for exclusion from what was found AND adds what is referenced but was not found using the default .XML templates.
Am I understanding it? Thanks in advance for any crayons of knowledge you can spare. :)