How do I set up an Exchange 2010 failover cluster?

Question

Clustering

Hello!
At the moment there is a powerful physical server, there are 20 virtual machines with different roles on it.
Two of these servers are organized into an Exchange 2010 cluster via the Failover DAG.
I managed to find out that the cluster has a separate IP address, Quorum Configuration - Node Majority.
Please help me understand the possible scenarios for organizing an Exchange 2010 failover cluster, what is the difference between them in terms of speed, the ability to automatically switch from one server to another in case of failure, and the necessary recovery actions.
Is the current configuration optimal or are there better options?
There are about 20 mailboxes, but a large volume of letters.

Answer 1

Your current scheme is a bit unclear.
1. If you have 1 physical. the server - that about what fault tolerance/HA there is a speech?
2. Quorum Configuration - Node Majority assumes the operation of the cluster with a successful quorum of more than half of the nodes. In your case (with two nodes), the cluster will completely fall apart if one of them fails. Maybe there is Node and File Share Majority ?
3. There are detailed official docks and a lot of books on Exchange HA.

Answer 2

Please help me understand the possible scenarios for organizing an Exchange 2010 failover cluster, what is the difference between them in terms of speed, the ability to automatically switch from one server to another in case of failure, and the necessary recovery actions.

DAGs provide protection at the database level, the Exchange 2010 server. Switching time between Exchange databases is about 30 seconds, and given the fact that clients communicate with databases through Client Access Servers, they will not notice this transition at all .
But ensuring database fault tolerance is only half the battle; you also need to ensure that users themselves have continuous access to these databases. CAS (Client Access Service) is responsible for working with email clients in Exchange 2010. To organize the fault tolerance of this service, there is a Client Access Array, which allows you to combine several CAS servers into an array, and then configure access to the FQDN of this array using a standard Windows NLB (Network Load Balancing) cluster or a hardware balancer.
In addition to the Mailbox Server and Client Access Server roles, you need to ensure the failover of the Hub Transport Server role, but you don’t need to do anything with it, just install this role on two or more Exchange 2010 servers and the failover will be provided automatically, i. to. Exchange Hub Transport servers are fault tolerant by default.
But in your scheme there is a bottleneck about which you described, this is one physical server. Here it is better to take care of fault tolerance at the virtualization level with several physical. servers using a cluster. Here the separation is quite large in the choice of types of clusters, everything will depend on your needs.

Answer 3

Thanks for the info!
I checked that NLB is configured, all three roles are configured on both nodes.
I conditionally called it a "powerful server", the hoster has a whole cluster there, with uninterrupted operation and other necessary characteristics.