Deploying Exchange Server in a Multi-Site Environment

Deployment of Exchange Server in a multi-site environment means having two or more servers of Exchange Server in different geo-locations. In this setup, the databases are constantly synchronized across site-to-site tunnels or some kind of express routes between the sites. The idea behind this setup is to ensure business continuity and disaster recovery. If something happens to a location, the data will be recovered and the services continued to work, even with limited resources.

In this article, we will be discussing the process involved in setting up a multi-site deployment with Exchange Server.

How to Deploy Exchange Server in Multi-Site Environment?

For multi-site deployment of Exchange Server, you need the following:

  • Two or more servers (located at different sites).
  • Extended network where the two sites can communicate.
  • Exchange Server license - be it Standard or Enterprise.
  • Setup a Database Availability Group (DAG) for data replication.
  • Network team to configure any changes needed to the network.

Once the above is achieved and the secondary site is set up with proper electricity monitoring, temperature, and security, you need to consider the following for this setup.

Infrastructure Design

The infrastructure design can differ depending on the business needs. The Active Directory on the primary site can be used as a file share witness (FSW). Here is an example (see the below image) of a design of a multi-site setup with two nodes, file share witness, and active directory.

infrastructure design

It is also important to have an Active Directory Server in both sites. This will ensure that authentication and DNS can happen even if the connection between sites is not available. In case of a failover, nothing will work since Exchange Server is dependent on the Active Directory Schema (as it holds the configuration). Also, the authentication of the users will not happen.

Network Requirements

To setup a Database Availability Group (DAG), you will need two networks:

  • A network for communication of the servers - between each other and the clients.
  • A network for replication - to be used for seeding and log shipping.

The replication network must be on different subnet from the production and can communicate with their respective match on the secondary site.

With regards to interfaces on the Exchange Servers, you will need two interfaces - one for the production and the other for the Database Availability Group (DAG). However, it is suggested to have non-teamed interfaces as this will reduce complexity and improve the stability of traffic. For the speed of the interface depending on the load, you can opt for 1Gbit interface or 10Gbit interface.

IP Addresses Requirements

Since multi-site deployment cannot be setup with Exchange Server alone, you also need to setup the Windows Cluster Service to manage the cluster of Exchange Servers. Below is the table showing the IP Addresses required for the Cluster, MAPI Network, Replication Network, and the Witness Service.

 
Component  IP Address Required
Cluster 1
Availability Group

1

MAPI Network 1 per cluster member
Replication Network 1 per cluster member
Witness 1

Exchange Server Network Design

Taking into consideration the infrastructure design, below is the diagram of all the components that are required for the cluster setup and the network needed.

Exchange Server Network Design

Let’s explore each component:

  • The Database Availability Group (DAG) role is to seed, log ship, and maintain the steady and real-time replication between the active database and the copy database.
  • The Cluster will manage the cluster of servers and their services. It will help to initiate a failover and to see the health (uptime) of each Exchange Server, along with the help of the witness to withhold the voting majority of the cluster.

Setup Requirements

When setting up the Exchange Servers to act as a Database Availability Group (DAG), it is important that both servers have the same specifications. Since the replication is done in real-time, you must have the same specifications to ensure that the data is replicated with no delay. The same disk configuration is also needed as this is important for the Exchange Servers to operate smoothly.

The last thing to take into consideration is the update level and application version. It’s important that both servers have the same operating system with the same Exchange Server installed. All the updates must match. This will ensure smooth installation and operation of the cluster.

Best Practices to Follow to Ensure Integrity of Setup

Here are some best practices you can follow to ensure integrity of the setup:

  • Both Exchange Server nodes should be monitored when it comes to uptime, replay of data between the source and destination, and the uptime of server.
  • You should also consider the patching method for the servers. Ensure that when the server is patched or any maintenance work is done, it will be outside the working hours and the servers will be put in maintenance mode. Failure to do this might result in issues on the Exchange Server and possible data corruption.
  • Periodic testing of the failover process must be done and documented. In case of failure, this will ensure that the services will automatically switch to the secondary site while having the latest copy of the data and no data will be lost. This test would also involve access of the users from VPN or other connections.
  • Backup and recovery are an essential part of the upkeep and assurance of the servers’ recovery in case of a disaster. A robust backup solution must be in place with application awareness and compatibility with Database Availability Groups (DAG). Although backups have a copy of all the data, the data can still get corrupted due to misconfiguration, malware attack, failure of hardware/software, or sudden loss of power. So, you must ensure that all the data can be recovered from the damaged databases. You can always restore from backup but this must be done as a last resort as it would involve losing the data between the last healthy backup and when the server failed.
  • To ensure that the data is always recoverable, you can keep in hand third-party tools, such as Stellar Repair for Exchange. This tool can open databases independently, in any state, of any size, and from any version of Exchange Server. It can granularly export user mailboxes, user archives, shared mailboxes, public folders, disabled mailboxes, and even deleted and purged data to PST file and other formats. In the case of a server recovery, you can granularly export the data directly to a live Exchange Server or Exchange Online with automatic mailbox matching.

To Conclude

For deploying the Exchange Server in a multi-site environment, you need careful planning and proper resources. This will ensure load balancing, high availability, and disaster recovery across varied locations. In this article, we have outlined the necessary requirements you need for configuring Exchange Server in a multi-site environment. We have also mentioned some best practices that must be followed to ensure the integrity of such setup.



Was this article helpful?
About The Author
author image
Anubhuti Sinha linkdin Icon

Anubhuti is a tech enthusiast with deep expertise in Microsoft Exchange Server.

Table of Contents

WHY STELLAR® IS GLOBAL LEADER

Why Choose Stellar?
  • 0M+

    Customers

  • 0+

    Years of Excellence

  • 0+

    R&D Engineers

  • 0+

    Countries

  • 0+

    PARTNERS

  • 0+

    Awards Received