This browser is no longer supported.
Upgrade to Microsoft Edge to take advantage of the latest features, security updates, and technical support.
Hello everyone We have a DAG cluster of two Exchange 2016 Std.5 Databases and 300 mailboxes in our office. Each node of the cluster has two NICs (MAPI and Replication) on the same network 10.50.1.0/24 and a shared witness folder. As part of building a fault-tolerant infrastructure, one of the nodes of the Exchange cluster will be moved to the Data Center. The Data Center has a different subnet 10.50.122.0/24, but in the same Active Directory site. There is a 1gbit/s bandwidth channel between the central office and the data center. Please tell me, is there a standard scheme for dividing two nodes of the same DAG cluster in different subnets and how to do it correctly and recommended? On one of the nodes, the addresses will change to the MAPI NIC and replication. Is it possible to use NICs for replicas for different networks?
A little clarification. The Active Directory site is bound to the subnet 10.50.0.0/16-by mask 16. But we use addressing by 24 mask. If I change the mask from 24 to 16 on the Exchange nodes, will this mean that logically then the nodes are in the same subnet and additional settings will then be required. Or is it impossible to do this, since this may affect the anti-spam clients and border servers? Thanks!
You can simply Change the IP on that existing server and it should be fine - as long as the two subnet can communicate.
"In this example, the MAPI network in each DAG member is on a separate subnet. As a result, the DAG requires two IP addresses, one for each subnet on the MAPI network. -"
This is referring to the quorum CNO object I was referring to before. If your cluster isnt using an IP now, then you dont need to do this.
OK, then all you need to do is build the new servers in the other subnet, ensure the nodes can communicate, and add them to the DAG.
You can remove the replication network as well if desired.
I meant to move one of the two DAG servers to a different subnet.
And here it is a little unclear, if you read what: In this example, the MAPI network in each DAG member is on a separate subnet. As a result, the DAG requires two IP addresses, one for each subnet on the MAPI network. - That is, in another subnet, the cluster node will simply have a different IP address (for example, it was 10.50.1.22 and after moving to the data center, I changed the address to 10.50.122.22 for this server)? Or do I need to add another IP address to the already existing server address?