In Exchange Server, high availability, also called Database Availability Group (DAG), is a setup that ensure uninterrupted service or connectivity in case anything goes wrong. In this setup, if you have a cluster of 3 nodes (which are 3 Exchange Servers), all the databases will be replicated in real-time between the servers. If one server fails, all the services will operate from another node. All this without any interruption to services for the client - user devices or Outlook. This means that users can keep sending and receiving emails without interruption.
Benefits of High Availability
Here are some benefits of high availability setup:
- No interruption to the services.
- Easy maintenance with bi-weekly patching and putting servers in maintenance mode.
- Failsafe operations where the changeover is automatic.
- Piece of mind with geolocated servers.
Components of High Availability
In a high availability setup, there are a number of components that you should know about.
1. Database Availability Groups (DAGs)
This is one of the fundamental blocks of having a high availability system. When you have a number of Exchange Mailbox Servers, the databases are constantly replicated from the main copy to all the servers so that each and every server will have a real-time copy of the main database. This will ensure that if the services are switched to another server, the database remains available and everything will continue as normal.
The data is replicated between the active copy and the passive copies in all the other nodes in the cluster. The replication is done in two parts:
- Block mode is the part where each update in the database is written in the active database log buffer. It is then shipped to a log buffer on each passive mailbox copy in block mode.
- File mode is the part where the closed transaction log files are copied from the active database to the passive copies.
2. Active Directory and DNS
Active Directory is very important as most of the configuration of Exchange Server and high availability is stored in the schema. Anything that happens in the cluster of Exchange Servers (creations, modifications, or queries) is done through the Active Directory Domain Services (AD DS).
Since the system is name-based, there is a common reference for the clients to access the email server. The name also depends on the DNS service. This is an important aspect of the availability group. If the DNS has any issues or there are delays during a failover or failback, the clients will not be able to query the active server. This makes it invisible to the users to whom the server is servicing. So, this minimizes the manual work or changes to the users. The servers, which are a part of the Database Availability Group are situated in different locations, can be accessed via a common name.
3. Active Manager
This is the core of a high availability system. This acts like a mediator in the Database Availability Group (DAG) and is responsible for failure monitoring of the active Exchange Server. It sits on the Mailbox Servers and manages all the switchovers, maintenance modes, and failovers of the cluster to ensure smooth data flow between the servers and their clients.
Things to Consider while Setting Up High Availability System
Although you can have a high availability system, you need to also take into consideration the inevitable and unforeseen issues. For this reason, you must have a robust business continuity plan, when things go wrong. You should always ask the below questions:
- What would happen if the system goes down?
- What if there is a sudden power cut?
- What to do in case of a malware or virus attack?
- Where is the backup stored and what is the frequency?
- What if the databases are not synchronizing due to an error?
- Has the failover ever been tested and documented?
These are some of the things to consider while building a high availability system.
To Conclude
Above, we have explained the concept and components of a high availability setup in Exchange Server. In a high availability system, in case something happens to the main server, the services and database switch to another server. But what if the database gets corrupted? In such a case, you can consider the right tools that can have your back when you have database issues. You need the right tool to recover the data as quickly as possible with minimal effort and downtime.
Tools such as Stellar Repair for Exchange give you a peace of mind as it can open the databases from any version of Exchange Server, in any state, any size, and without a working Exchange Server. After a quick scan, you can easily and granularly export user mailboxes, user archives, shared mailboxes, disabled mailboxes, public folders, and deleted and purged items. Data can be exported to PST or directly to a live Exchange Server database of any version or to a Microsoft 365 tenant.
4 min read
