The local site is preferred and used over a more expensive site. However,
for this to work properly, sites, subnets and site links need to be
correctly defined and configured. The idea of the data centre being first
to answer is out of date. That's how NT 4 worked with the NetBIOS mailslot
messages. AD doesn't do this. AD is site aware.
Ideally, all physical subnets will be correctly defined in AD and associated
with a site. Those site objects will be in different site links, where the
site link contains the remote site and the main site. Or at least, this is
the best way if you have a hub-and-spoke network topology. This can vary,
but in most cases is the best fit. If you have a different design, your AD
guys should know what they're doing. From what you've said this isn't the
case, so lets assume this is the best design for you.
With such a design in place, and the server objects for the DCs in the
correct site, and all DCs registering in DNS, your local box will always be
used for logon and DFS. If you want more advice, you need to take a peek at
the sites and services snap in. You can look at this even if you're not an
administrator. It's DSSITE.MSC in the ADMINPAK.
--
Paul Williams
Microsoft MVP - Windows Server - Directory Services
http://www.msresource.net |
http://forums.msresource.net