Saturday, March 18, 2017

Exchange 2010 Database Availability Group Setup

The Database Availability Group (DAG) is quietly most important feature in Exchange. In Exchange 2003, 2007 we used to have SCR, CCR, LCR clusters to maintain the high availability. In Exchange 2010 Microsoft combined CCR + SCR = DAG cluster functionalities together to improve the Exchange Database High Availability.

By now, I’m sure that most Exchange professionals have read up on the features of the DAG from the official Exchange product documentation, and the numerous articles that exist on the Internet. However, if you haven’t, you can find out what all the fuss is about within the Exchange 2010 product documentation.

In this article, we will cover mostly what is required and how to setup the Database Availability Group.

First and foremost important to understand is how we are going to setup the environment highly available all the times irrespective of Server OS level outage, VMware/Hyper-V Host Level, Site Level.

In this article, I am using the two-node DAG can be 

considered to be modeling a high availability solution 

within a single data center. However, this DAG 

configuration clearly cannot survive the loss of the 

data center itself, as no copies of the mailbox 
databases exist outside of the single site.

Before we setup DAG we need to understand what are the DAG component and how to setup properly,

DAG Components: -

DAG Name – We can have maximum 16 mailbox servers in a single DAG, Each DAG you create in exchange environment is a unique name, which ultimately used in the failover cluster.

DAG IP address – A DAG must have one or more Static IP address, In my environment, we have single AD site so I am using Single IP address but in your case, if it’s multiple AD sites, you can assign multiple IP addresses.

Witness Directory – If the witness server name has been specified then we must specify the witness directory, witness directory will keep the cluster configuration information.

Witness Server - If a DAG has an even number of mailbox servers contained within it, as is my environment DAG, a witness server will be used to act as an additional vote in the cluster to maintain cluster quorum. A CAS/Hub Transport server is typically chosen as the witness server, as long as that server role is not running on a member of the DAG.

Network Configuration: -
Network configuration playing the major role in Exchange DAG setup, if any misconfiguration on the NIC card setup leads poor behaviors. In exchange DAG setup either we can use dedicated MAPI & Replication Network as long as it is connecting to the backend physical NIC card delicately. Also, you can use single NIC card for MAPI + Replication this is fully supported.

MAPI Network: - This network used by client communication such as outlook and server to server communication.

Replication Network: -This network purely used for replication exchange transaction logs and reseeding the databases.

MAPI Network Setting
Replication Network Setting
Client for Microsoft Networks
QoS Packet Scheduler
Optionally Enabled
Optionally Enabled
File and Printer Sharing for Microsoft Networks
Optionally Enabled
Optionally Enabled
Link-Layer Topology Discovery Mapper I/O Driver
Link-Layer Topology Discovery Responder

DAG Setup Creation: -

Before we begin creating Exchange Database Availability Group, we need some of the prerequisites ready such as,
  • Add CAS/HUB server into Exchange Trusted Subsystem security group
  • Add Exchange Trusted Subsystem into server local administrator groups.
  • Create DAG Name computer object in AD.

Open Exchange Management Console Navigate to Organization Configuration and select Mailbox.
Select Database Availability Group tab, select “New Database Availability Group” option from Action pane.

As we discussed above in the article, specify the DAG Name, Witness Server and Witness Directory, Click Next.

If Witness server were already assigned to Exchange Trusted Subsystem and Local administrator group, you DAG creation will finish successfully.

Now, Right click the DAG Name that you created now, select properties

Select IP Addresses Tab and Update one or more IP address according to your environment, Click OK and Apply the changes.

Right Click DAG Name, Select “Manage Database Availability Group Membership”

Click “Add” and Select Mailbox servers that need to be part of selected DAG Member. This option helps you to add or remove servers from the DAG, when the server added to the DAG it will configure for automatic database recovery from database failures.

Click Manage

Now, we successfully completed DAG setup. Right Click the DAG you created and Select properties validate DAG Name, Witness server, Witness Directory, DAG IP address and Operation servers which are part of the DAG.

Select Database Availability DAG Network settings and ensure Replication and MAPI are enabled and all NIC card is UP.

Now, we validated all configuration settings and it’s time for us to configure the database copies.

Navigate to Organization Configuration, Select Database Management Tab
Select the database that you want to configure for High Availability

Select “Add Mailbox Database Copy

Click Browse and select Mailbox server, which needs to hold the Passive Database Copy. This creates a copy of the mailbox database in specified mailbox server and its enable continues replication from active database copy.

Click Add

Click Finish, Once database copy is created.

Repeat the steps for rest of your mailbox databases to configure HA.

Once you configured the passive database copy, Mailbox Replication service kicks to start replicating the Active Database EDB and Transaction Logs to Passive Database Server.

Once the replication is completed, it’s fully available for HA.

Thanks for Visiting. Keep watch for the further updates!