Exchange Server 2016 – Configuring DAG Network Adapters

How to configure the network adapter to support a DAG implementation using Exchange Server 2016. The DAG can be used with a single adapter however, using two network adapters is a more elegant and efficient solution (if you have 10GB network available for the servers, then a single adapter is totally fine) for the vast majority of cases.


For this Tutorial we are going to use the scenario depicted in the diagram below where we will have two network adapters (MAPI and Replication) and we will configure the details on each one of them.

Here are some key points that need to be considered before creating the network configuration for your DAG when using Exchange Server 2016:

  • Same number of adapters among all members of the DAG
  • No more than 500ms network latency among members of the same DAG
  • When using more than a network adapter, they should use different network
  • Both networks should be independent, we don’t want routing between MAPI and Replication networks


General settings…

The first step is to label your network adapters, and for this Tutorial we are going to use MAPI and Replication as follows.


The binding order of those adapters is important, make sure that the MAPI is at the top.


MAPI Network…

That it is the easy to configure because it does not require anything special, keep default settings and configure the static IP (you can use DHCP if you want) and make sure that you have a gateway, DNS pointing out to your local Domain Controllers and we you are golden!

Replication Network…

This is the network that requires extra attention from the administrator. The first thing is to uncheck Client/and Services, the final result for the Replication network should be similar to the image below.


In the same adapter, make sure that Register this connection’s address in DNS is unchecked.