I am using USMT 4.0 to migrate users from local accounts to a new domain using /md*:newdomain. This works fine but on some restores loadstate seems to be stuck before closing itself down at around 9x%, other machines restore fine using the same scripts. I tracked it down in the loadstate.log file to one line after all the user profiles and the "System" profile has finished restoring. If the line reads CRegShortNameFixMemTable::Read, table has 0 entries then USMT cleans up and closes a millisecond after. If the line reads CRegShortNameFixMemTable::Read, table has xxx entries (can range from 100 to 400 odd) then USMT starts to loop multiple times for each restored account doing "something", and the progress % in loadstateprogress.log stays the same. This can just keep going for an hour or more while it keeps progressing through the accounts. 4 accounts can take shorter than 2 accounts, there doesn't seem to be any way to predict the time based on number of accounts or something special with those accounts. I cannot find any information on what that function is doing nor why it does it, and why only on certain machines. If anyone can help it would be appreciated! Many thanks

if you enable verbose logging and use MIG_ENABLE_DIAG do you get more info to help with troubleshooting this ? I describe how you can set this up here Step by Step ConfigMgr 2007 Guides | Step by Step ConfigMgr 2012 Guides | I'm on Twitter > ncbrady

if you enable verbose logging and use MIG_ENABLE_DIAG do you get more info to help with troubleshooting this ? I describe how you can set this up here Step by Step ConfigMgr 2007 Guides | Step by Step ConfigMgr 2012 Guides | I'm on Twitter > ncbrady

Thanks Niall, I have verbose logging enabled so 'll set MIG_ENABLE_DIAG on top of that and see what I get.