So far in the previous articles we have see how to install and setup NAS4Free with iSCSI targets. A very easy and straight forward process. The aim of this article instead is to provide the steps on how to configure the iSCSI target on a Windows Server and in particular with the purpose of serving a SQL Failover Cluster
So as per previous article we are using the built-in iSCSI Client to list and connect to available targets. As soon as we do so all we have have to do is to review and accept the default settings as per screenshot below where you can see the iSCSI Server IQN unique name and the name of the resource. For now we can leave the option for the Multi-Path or Microsoft MP-IO disabled as we are connecting to these resources with a single connection. I will elaborate more on this topic in a separate article
Next step is to launch the Disk Manager applet and we’ll notice a new Disk is available. Let’s start bu setting this disk from Offline to Online
At this point we can now use Windows to initialise and format the Disk
As per screenshot
Let’s create a simple volume..
As per screenshot
Let’s use the maximum size
Let’s assign a drive letter
It is now time to format this drive with NTFS in orderfor the SQL Failvocer cluster to use it. Let’s also provide a volume label and run a quick format
A quick summary before committing changes and in a matter of few seconds the new volume will be mounted and ready for Windows to be used
When repeating the same steps for the other volumes we’ll get a configuration similar to this one:
- Disk for Quorum
- Disk for SQL Backups
- Disk for SQL Data
- Disk for SQL Log
- Disk for SQL TempDB
All these disk drives are now ready to be used from the first node of the Cluster
On the other member nodes of the cluster all we have to do is to repeat the same steps. the major difference is that once disks are “Online” onto the new cluster nodes they do not need to be formatted again. Another recommendation of better a finishing touch would be to make sure the same drives are mounted exactly with the same drive letters across different node members of the Cluster as this is also one of the requirements to setup a SQL Failover Cluster
At this point we can now take a quick break, maybe a quick espresso! Everything is ready to start the deployment of a SQL Failover Cluster Instance.
Add Comment