Exchange Server 2010 : Deploying a Database Availability Group (part 1) - Creating the File Share Witness

Deployment of a Database Availability Group (DAG) is a fairly straightforward process but it does have several steps that must occur in the correct order. By becoming familiar with the requirements and the process, the implementation should be fairly uneventful.

Because a DAG doesn’t require a Windows 2008 Cluster to be prebuilt, the instructions will assume that the administrator has already built a basic Exchange Server 2010 mailbox server

Requirements for DAG

You will need two or more servers that are capable of supporting the Exchange Server 2010 Mailbox role. You don’t need shared storage with a DAG because the transactions are shipped to the replicas and applied locally. This allows up to 16 independent databases and sets of log files on up to 16 different servers with independent media. You will want to follow the same standards as you would with a normal mailbox server in terms of database sizes and hardware specifications to support your anticipated user load.

To set up a DAG, you need the following:

• Two or more servers running Windows Server 2008, Enterprise Edition

• Two network interfaces per server

• Exchange Server 2010 Enterprise Edition

• 1 File share witness per DAG

• No longer limited to a single AD site

You need to pre-create the file share that will be used as the file share witness by the cluster. The permissions on the share will be configured for you by Exchange Server 2010. The file share witness cannot be hosted on a member of the DAG.

Creating the File Share Witness

While a Database Availability Group isn’t a cluster in the traditional sense, it does utilize some of the functions of Windows 2008 clustering in order to track system availability. One of the components required for this function is a file share witness. Administrators familiar with Exchange Server 2007 may recall that a file share witness acts like a voting node in a cluster, specifically a Majority Node Set cluster. This allows a cluster to have as few as 2 traditional nodes with the FSW acting like a 3^rd node. This allows the cluster to achieve a majority if a node were to fail. Were there only 2 voting nodes total, the majority of two is two, so only one being up and running wouldn’t constitute a majority and it wouldn’t be able to make cluster related decisions. Each Database Availability Group will need its own file share witness. To create this share, follow these steps:

1.	Create a directory on a Hub Transport server or a File server.
2.	Right-click the directory, and choose Share.
3.	Click Share.
4.	Click Done.

The configuration of the Database Availability Group will configure the necessary permissions on the share for you.

Creating the DAG via GUI

Once the Exchange Server mailbox servers are prepared, they can be joined into a new Database Availability Group by performing the following steps:

1.	Launch the Exchange Management Console.
2.	Expand Organization Configuration.
3.	Click Mailbox.
4.	In the middle pane, click the Database Availability Group tab (see Figure 1). Figure 1. Creating the DAG.
5.	In the right pane, click New Database Availability Group.
6.	When prompted, enter a unique name for the Database Availability Group along with the file share witness path and directory that were created earlier (see Figure 2). Click New. Figure 2. Entering the information for a new DAG.
7.	When the wizard has completed, click Finish.

At this point, the DAG has been created, however it has no members. Add member mailbox servers to the DAG with the following steps:

1.	Launch the Exchange Management Console.
2.	Expand Organization Configuration.
3.	Click Mailbox.
4.	In the middle pane, click the Database Availability Group tab.
5.	Right click the DAG created in the previous steps and choose Manage Database Availability Group Membership.
6.	When the wizard appears, click Add and choose the mailbox servers from the list that you want to join to the DAG (see Figure 3). Click Manage. Figure 3. Adding members to the DAG.
7.	The wizard might take several minutes to complete. When it has added all the necessary nodes, click Finish (see Figure 4). Figure 4. Completing the new DAG member wizard.

When this process has been completed on both nodes, the systems are ready for the rest of the configuration process to continue.

1.	Return to Exchange Management Console and expand Organization Configuration.
2.	Click Mailbox. In the middle pane, click the Database Management tab (see Figure 5). Figure 5. The newly created DAG.
3.	In the lower pane, right-click the database you wish to replicate within the DAG.
4.	Choose Add Mailbox Database Copy.
5.	When the wizard launches, browse for the server in the DAG to which you want to replicate the mailbox database. Pick a Replay Lag Time and a Truncation Lag Time (see Figure 6). Figure 6. Adding a DAG replica.
6.	Enter a unique preferred list sequence number and click Add.
7.	When the wizard completes, click Finish.

When the Database Availability Group is created, a computer object is created in Active Directory to represent the Failover cluster virtual network name account. If a DAG is going to be recreated with the same name, it is necessary to disable or delete this computer account or the process will error out and fail.