One of the most anticipated new features in the upcoming version of Windows Server is storage replication. The...
feature allows for block-level data replication and is based on SMB 3.0. However, in the Technical Preview, storage replication is not available by default. Let's look at how to enable the feature.
If you want to use storage replication, you will need to install the Windows Volume Replication feature (Figure 1). This feature must be installed on both the source and the destination server. You will also need to install the Failover Clustering feature and the File Server role on both servers.
In order to make storage replication work, you are going to need four separate cluster disks. Two of these disks will reside at the source -- a local cluster node, and two will reside at the destination -- a remote cluster node. You will be replicating data from the source disks to the destination disks.
All four disks must be exposed as cluster storage. Since the disks work in pairs, you will have a source side data disk, a source side log disk, a destination side data disk and a destination side log disk. The source and destination data disks must be the same size and the same goes for the source and destination log disks. Microsoft recommends that your log source disk be SSD-based for performance reasons.
Once you have created a failover cluster, and attached the cluster disks, you will need to create a cluster role. In this case, we will need to use the File Server role (Figure 2).
Most of the High Availability Wizard’s prompts are self-explanatory. However, when you arrive at the wizard’s Select Storage screen, you must select your data disk, but not your log disk.
After completing the wizard, navigate through the Failover Cluster Manager to Storage Disks. Right click on your data disk and choose the Replication Enable commands from the shortcut menus (Figure 3). This will prompt Windows to launch the Configure Storage Replication wizard.
Click Next to bypass the wizard’s Welcome screen. You should then see a screen asking you to specify your source disk. As long as you haven’t included the log source disk in the File Server Role, you should see the disk listed within the wizard (Figure 4).
Select the log source disk and click Next. Now you will need to select a destination data volume. There is a glitch that sometimes causes the list of available storage to be empty. This happens because the remote cluster node does not currently own the disks. You can fix the problem by going back to the Failover Cluster Manager’s list of disks, selecting the remote disk and then using the Move Available Storage option.
Click Next again and you will be asked to specify the destination log disk. Now you will see a prompt asking you if the destination disk has been seeded. In other words, Windows wants to know if the destination disk already contains a partial copy of your data. In most cases you will probably choose the This is not a seeded disk option (Figure 5).
Now all you have to do is to click Next a couple more times and the replication process will begin. According to Microsoft, you will eventually be able to perform synchronous replication between servers either for disaster recovery or for high availability (as a part of a stretched cluster). As it stands right now however, storage replication only seems to be exposed through the Failover Cluster Manager. The next version of Windows Server is still in the technical preview phase, so there are some things that will likely change before it is released.
Uncover the ins-and-outs of storage replication