I don't even think I should try to store the security tables in DB2, agree?



So I will just have a separate SQL instance for security? I am fine with that and see pretty detailed instructions on how to support the tables in a separate database.



Just really making sure I didn't miss anything here. Does the above sound solid to you?



Also, how would this work if I decide to architect my application so that it will support a truly "disconnected mode" as in I just found out that users might need to run it at times on a laptop with no internet access, then save the data to the server at a later time. I was going to just use XML or a local SQLEXPRESS or SQL Compact database, but I'm not sure how not being able to connect to the security tables will work in that scenario.




Keith Chisarik