Four steps to easily create a Win2003 Distributed File system

When Microsoft first introduced the distributed filesystem (Distributed file system, referred to as DFS), it focused on itself the end user's desire to make things easier. The idea of this technique is that the user itself does not need to know which server resources are real. They can access the file system simply through a special share, and they can access all the data they need, whether centrally stored locally or distributed across many different servers.

While it's always good to make things simple on the client side, I think DFS is more useful than just load balancing and fault tolerance. DFS can be used to assign a user's workload to multiple servers that contain copies of files. In addition, if a replica's server fails, or if an administrator needs to perform offline maintenance on a single server, the user can continue to access the replica server continuously without interruption by using DFS. (In the second part, I'll discuss load balancing and fault-tolerant issues in more detail.) )

In the first section, I'm going to show you how to create a Distributed file system in Windows 2003 server.

First step: Create a console

To start this process, first enter the Microsoft Management Console (Admin console, MMC) command in the Run command. When you enter, Windows will load an empty admin console. Now, you will select the Add or remove snap-in (Snap-In) command from the File menu in the console. Windows then displays the Properties menu for the Add or remove snap-in. At this point, click the Add button based on the "Standalone" tab of the Properties menu to display a list of all available snap-through cells. Select the Distributed File System option from the table column and click the "Add" button, then click the "Close" and "Confirm" button.

Step two: Create a DFS root

"Root" is the topmost level of the DFS hierarchy. A root contains more than one shared folder. To create a root in a server that currently contains some of your data, you need to right-click the Distributed File System container (container) in the console and select the new Root command from the shortcut menu. Doing so enables Windows to load the new Root Installation Wizard. Click "Next" to bypass the Welcome interface of the installation wizard. The wizard will ask you whether you want to create a separate root or create a domain root. The root of a domain exists only in Windows Server 2003. They support automatic data replication. For the purposes of this article, you should select the root option for the domain, and then click Next.

The next interface to the wizard is to ask you which domain is the primary domain for the root you are creating. Select the appropriate domain, and then click Next. Now, you must enter the name of the server that will be the host for the newly created root. This server must be a member of the domain you selected in the previous step. Enter the name of the server that fully meets the requirements of the domain, and then click Next.

Step three: Name the root

The next interface of the wizard requires you to enter the name of the root you created. Windows will create a shared name that is consistent with the name of the root you entered. I also recommend that you enter some descriptive text to help you remember what this root is for. Click "Next" to continue our work.

Fourth Step: Folder selection

In this interface, the wizard will ask you to specify a folder in which Windows can assign shared files to. I recommend that you choose a folder that already contains your data. Click "Next" and then click "Finish" to end the root creation.

Your DFS share should now be activated. To confirm that the root is valid, right-click it, and then select Check Status on the shortcut menu. The correct status should be displayed as "online."

How to create a replica server

This is the preparation for the second part of this series, and I want to show you how to create a replica server.

To start this process, first go to the server you want to create as a replica, and select a volume with sufficient disk space available. After this step, create a folder on this volume to store the data that you will copy to the server.

Once you have created the folder, you need to set the folder to be shared and use the same share name that you assigned to the DFS root. I recommend that you have full control over everyone at the shared level and that you use NTFS permissions to protect your data, rather than relying on a shared-level license. After you set up the share, you need to wait 15 minutes for the network to recognize the share. As you wait, open the service console and start the Distributed File system service.

Now go back to the DFS console, right-click the DFS root you created earlier, and select the new root Target command from the shortcut menu. Now you will be prompted to enter the name of the server at the root target you want to create. In my system, I need to enter the server's NetBIOS name, not its fully compliant domain name, in order for Windows to identify the server. Here you need to make your own choices and click "Next". After a while, you will see an interface that informs you that you have completed the wizard. Click "Finish", and then continue with the work below.

This time your Dfs root should be turned on and running. Now you should initialize the replication process. You can right-click the Dfs root with the right mouse button and choose Set Copy on the shortcut menu. Windows will load the Setup Replication Wizard. Click "Next" to go over the Welcome interface of the wizard, and you will be prompted to choose to initialize the main file. Select the shared file that currently contains the data, and then click Next. In the next interface you see, the wizard will ask you to select a replicated topology. I recommend that you use a fully meshed topology. Click "Finish". You may then see a message that prompts you to set the Replication service (file Replication service) on the replica server to start automatically. If this information appears, then use the Services console on the server to change the startup mode of the File Replication Service to start automatically. Start the service, and then click Yes to retry the replication settings.

Beyond the user's DFS advantage

In the first part (four easy to create Win2003 Distributed File system), I have explained how to create a Distributed file system (Distributed file System,dfs) root and how to create a copy of this root. In the second section, I'll explain how to use the DFS replica settings you created so that it can provide a way for your users to be fault tolerant and highly effective.

In the previous article, I introduced you to how to create a logical path for a physical location on the server. The trick to effectively use DFS is to have users access their files through logical paths (that is, DFS root) instead of directly accessing the data.

For example, on my test server, the data that my users typically access is located in the G: drive. Users typically access this data by sharing a path named G (\ \ server name \g). If, after creating the Dfs root, users access data through pre-existing shares, they will benefit from an automated backup, since their data has been replicated to another server as well. However, they may completely miss the benefits of load balancing and fault tolerance on all servers. To be able to derive benefits from load balancing and fault tolerance, they must access the DFS root through a dedicated shared (\ \ domain name. com\ share name).

Assuming that users are accessing the DFS root through a DFS share, they will naturally see improved performance due to DFS load balancing. DFS servers are able to balance user traffic across multiple replicas, so none of the replicas consume all the user effort (assuming the server is accessed through a DFS share). )

However, the real advantage of Dfs is that a separate copy can be used offline, whether intentional or unintentional, without disrupting the user. Users still have full access to their files, but the speed of access may be slower than when the copy has been completed.

If the server is corrupted, DFS will eventually stop sending a user request to the server, but if the Distributed File system service discovers a problem, you find that the server is down, or if you need to maintain a DFS server for downtime, So here's a very simple method you can use. To start this process, first enter the MMC command in the Run Command dialog box. When you enter this command, Windows loads an empty Microsoft Management console. Now, select the Add or remove snap-in (Snap-In) command from the File menu in the console. Windows will now display the Add or Remove Snap-in property menu. At this point, click the "Add" button based on the "Standalone" tab of the Properties menu to display the table columns for all available snap-in units. Select the Distributed File System option in the table column and click the "Add" button, and then click the "Close" and "Confirm" buttons.

Now that the DFS console is open, select your Dfs root in the column on the left. The copy contained in the DFS root will be displayed in the column on the right. If you want to get a copy offline, simply right-click on the copy and select Move Target from the shortcut menu. After such an operation, the target will no longer be part of the DFS root. Now you can do whatever you want to do with the server, and none of this needs to interrupt the user. Once you have completed the operation of the server, simply add the server as a target and start copying.

Adding data to the DFS root

Now that you know how much the DFS can improve the user experience and make your life easier, you may also want to add additional data to the DFS root. To do this, you can create additional DFS roots (the first part describes how to create a Dfs root), or you can add additional data to a root that has already been created. Microsoft supports both methods, but they are more likely to add data to existing roots. The reason for this is that doing so would make it possible to merge all user data that is under a single DFS root. This means that users no longer need to worry about the physical location of their files. In a separate share, all user files are valid.

To add additional data to the DFS root, you need to first right-click the Dfs root you created, and select New Connection on the shortcut menu. When you do this, you will see a dialog box that prompts you to enter the name of the connection and the path to the destination.

The name of the connection is the name that the user sees when accessing the connection. The user will see this new connection as a subordinate directory under the DFS root. You must also enter a common naming convention (UNC) path for the new connection, along with an optional annotation. Click "OK" and then the connection is created.

DFS can make file access faster and more stable, thereby improving user performance. It also improves the quality of your life, because you don't have to wait until the dead of night to do server maintenance. You can take offline operations on a single DFS server without disrupting the user.