Problem solve Get help with specific problems with your technologies, process and projects.

DFS benefits go beyond the user, part 2

Part 2 of our 2-part series on Distributed File System for Windows Server 2003 shows how you can use DFS for fault tolerance and load balancing.

In Part One, I showed you how to create a Distributed File System (DFS) root and how to create replicas of that root. In this article, I will explain how you can use the DFS replica set you created as a way of providing fault tolerance and high availability to your users.

In the previous article, I showed you how to create a logical path to a physical location on a server. The trick to using DFS effectively is to get users to access their files through the logical path (the DFS root) rather than accessing the data directly.

For example, on my test server, the data that my users normally access is in the server's G: drive. The users normally access this data through a share named G (\ \servername\g). If after creating the DFS root users were to continue to access data through the pre-existing share, they would benefit from a sort of automatic backup because their data is being replicated to another server. However, they would completely miss out on all of the server's load balancing and fault tolerance related benefits. In order to benefit from load balancing and fault tolerance, they must access the DFS root through its dedicated share (\ \\sharename).

Assuming that the users are accessing the DFS root through the DFS share, they will automatically see increased performance due to DFS load balancing. The DFS servers will automatically balance user traffic among the various replicas so that no one replica is carrying the full user workload (assuming that the server is being accessed through the DFS share).

The real advantage to DFS, however, is that individual replicas can be taken offline (either intentionally or unintentionally) without disrupting the users. The users will continue to have full access to their files, but the access might be a little bit slower than what the users are used to if a replica is down.

If a server crashes, DFS will eventually stop sending user requests to that server, but if you realize that the server is down before the Distributed File System Service does, or if you need to take a DFS server down for maintenance, there is a very simple procedure that you can follow. To do so, begin the process by entering the MMC command at the Run prompt. When you do, Windows will load an empty Microsoft Management Console. Now, select the Add / Remove Snap-In command from the console's File menu. Windows will now display the Add / Remove Snap-In properties sheet. At this point, click the Add button that's found on the properties sheet's Standalone tab to reveal a list of all of the available snap-ins. Select the Distributed File System option from the list and click the Add button, followed by the Close and OK buttons.

Now that the DFS console is open, select your DFS root from the column on the left. The column on the right will now display the replicas that the DFS root consists of. If you want to take a replica offline, simply right click on the replica and select the Remove Target command from the resulting shortcut menu. After doing so, the target will no longer be a part of the DFS root. You are now free to do what ever you need to do to the server without disrupting the users. When you are finished, simply add the server back in as a target and enable replication.

Adding data to the DFS root
Now that you know how much the DFS root can improve the user experience and how much easier it can make your life, you may want to add additional data sets to the DFS root. To do so, you can either create additional DFS roots (see part 1 for an explanation on how to do this) or you can add the additional data to the existing root. Microsoft supports either method, but they tend to prefer adding data sets to the existing root. The reason is that doing so makes it possible to consolidate all user data under a single DFS root. This means that the users no longer need to worry about where their files are physically located. All user files are available through a single share.

To add additional data to the DFS root, right click on the DFS root you created and select the New Link command from the resulting shortcut menu. When you do, you will see a dialog box prompting you for a link name and for a path to the target.

The link name is what the users sees when they access the link. Users will see the new link as a sub directory beneath the DFS root. You must also enter the UNC path for the new link, and an optional comment. Click OK and the link will be created.

DFS improves the user performance by making file access faster and more reliable. It also improves your quality of life because you don't have to wait until late at night to do server maintenance. You can take an individual DFS server offline without disrupting your users.

Brien M. Posey, MCSE, is a Microsoft Most Valuable Professional for his work with Windows 2000 Server and IIS. Brien has served as CIO for a nationwide chain of hospitals and was once in charge of IT security for Fort Knox. As a freelance technical writer he has written for Microsoft, CNET, ZDNet, TechTarget, MSD2D, Relevant Technologies and other technology companies. You can visit Brien's personal Web site at

More information from

Dig Deeper on Windows Server storage management