Building a SQL Server 2008 R2 Cluster: Part III


In the previous part of this post, we have configured the disks for our Windows Cluster. The disks are created from a RAID 5 LUN. As you can see, we have a LUN of 60GB however only 40GB is available for shared storage. image

As we are going to install SQL Server on a windows cluster environment, we are going to add a new disk for the MSDTC resource. Here is a list of all the available disks available to our cluster.image

Now, we have all our disks ready, its time to add it to both our nodes. To begin with, I am going to log on to our first node, BLRC100. Both the nodes will access these shared disk with the iSCSI service available with Windows Server 2008. Open the iSCSI Initiator from the administrative tools (Start Menu –> Administrative Tools –> iSCSI Initiator). If the iSCSI service is not running on this server, you will get the following screen, click on Yes to start the Microsoft iSCSI service on this server.image

If prompted, unblock the iSCSI service in the Windows Firewall so that it can communicate with the storage box.image

The iSCSI Initiator properties window will be opened. Click on the Discovery tab. In this tab, we are going to add the storage server portal. Click on Add Portal, and provide the IP address or FQDN of the storage portal and click on OK. If you get the following error message, make sure the iSCSI service is running on the storage server.image

To start the iscsi service on the storage server, click on the Setting tab from the Services (Services –> iSCSI Services) drop-down menu.image

If the iSCSI service is running on the storage server, after you will add the portal on the Discovery tab of the iSCSI Initiator properties tab, you will get the following screen.image

Now that we have added the target portal, click on the Targets tab of the iSCSI initiator properties. You will find all the disks in inactive state. Select one disk and click on the Log On button, you will get the following screen. Make sure, you select the checkbox next to Automatically restore this connection when the computer starts. Click on OK to continue. imageOnce you have added some of these disks, you will get the following screen. In the below screenshot, the disks next to the green-line are connected; however the disks next to the red-line are still inactive. image

Once all the disks are active, click on the Favorite Targets tab to list all the shared disk available to the Cluster node. Click on Ok to close the iSCSI Initiator properties window. We need to repeat all the above steps on the other node, to make these disks available. image

Once you open the Server Manager (Start Menu –> Administrative Tools –> Server Manager) and browse to Storage –> Disk Management. You will find all the disks are available offline. Right-click on the disk, and select Online from the drop-down menu.image

Once the disk is online, right-click on the disk and select Initialize Disk from the available option. image

After you initialized the disk, format the disk. Make sure you have selected NTFS file system for the disk. Once the disk is formatted, bring it offline. You need to initialized all the available disks, format it with NTFS partition and then bring it offline. After you have completed all the disks, time to log on to the other node. image

As we have completed all the above steps on the first node of our windows cluster, time to move to the second node of our cluster system, BLRC101. Shut down the cluster node BLRC100 & start the node BLRC101. On this node, we have to all the iSCSI service as we have done for node BLRC100, bring the disks online. As we have already formatted the disks earlier, we don’t have to format it once again. Once all the disks are online, you will get the following screen. image

Since all the disks are ready for cluster installation, we are going to take these disks offline and shutdown this node. Then we are going to add a network adapter on each node, which will create a HeartBeat Network (10.200.41.100/8, 10.200.41.101/8) for the cluster. Once we add the network adapter, we are going to install Dot Net Framework v 2.0 & 3.5 on both the nodes. Since DotNet framework is mandatory for SQL Server cluster installation, we are going to install it before starting the Windows Cluster installation.

As we have installed and configured all the necessary hardware and software, its time to start the windows cluster installation. In the next post, we are going to start the installation of Windows Cluster.

About these ads

About Sudeepta Ganguly

A SQL Server User... Still Learning
This entry was posted in Clustering, How To, SQL Server, SQL Server Installation and tagged , , . Bookmark the permalink.

5 Responses to Building a SQL Server 2008 R2 Cluster: Part III

  1. Pingback: Building a SQL Server 2008 R2 Cluster: Part II | SQL…. Still Learning

  2. Pingback: Building a SQL Server 2008 R2 Cluster: Part IV | SQL…. Still Learning

  3. Andrew Bruce says:

    Great stuff, Sudeepta. One question on this post: “Since all the disks are ready for cluster installation, we are going to take these disks offline and shutdown this node. Then we are going to add a network adapter on each node, which will create a HeartBeat Network (10.200.41.100/8, 10.200.41.101/8) for the cluster.”
    It is unclear to me as a n00b what that means – you are adding the network adapter using Oracle VM, yes? But you are configuring the adapter with IPv4 networking *after* restarting each node? Correct? I also assume that Node A would have 10.200.41.100 and Node B would have 10.200.41.101 for the new adapter, yes?

    Thanks again, now onto number 4. I’m tempted to take this low cost solution and use it instead of VMware for a customer…

    • Hi Andrew,
      Glad you like the post. Yes I was using Oracle VM for these posts. I had shutdown both the node to add the network adapters to each node. I could have add all the adapters while creating the VM initially, but I thought of adding it when required. The purpose of this series is to learn how to install a cluster (Windows +SQL).

      About your thought on implementing the same for your customer, allow me to inform you that, Cluster(Windows +SQL) is a costly solutions, and for a production environment, you might want to implement it using MS Hyper-V server/ VMWare ESX Server. If you consider the cost of the software licenses, you might want to use it on a supported hardware :). You may consider of using “Database Mirroring” as well.

  4. Pingback: Building a SQL Server 2008 R2 Cluster: Part VIII | SQL…. Still Learning

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s