Partager via

Exchange 2010 Site Resilience with two active Datacenters

  • Article

If you want an active/active user distribution you need to determine the failure models and scenarios you are willing to tolerate.

For example, if you have a WAN failure between the two datacenters and you deploy a single DAG, then one of the datacenters is going to have an outage event for the mailboxes because quorum cannot be maintained. If this is unacceptable for you, then you will need to deploy two DAGs to maintain user locality for failure events like that. This will increase the number of mailbox servers as you will need at least a mailbox server from each DAG in the other datacenter. The below diagram clarifies this:

The way to deal with this in the Mailbox Role Calculator:

  1. Launch two versions of the calculator.
  2. Populate the first version for the first DAG in your design. This DAG (DAG01) will utilize Datacenter 1 as its primary location (and thus its user population is based out of Datacenter 1). It has site resiliency by having servers and database copies located in Datacenter 2 that can be activated in the event Datacenter 1 is lost.
  3. Populate the second version for the second DAG in your design. This DAG (DAG02) will utilize Datacenter 2 as its primary location (and thus its user population is based out of Datacenter 2). It has site resiliency by having servers and database copies located in Datacenter 1 that can be activated in the event Datacenter 2 is lost.

By implementing the architecture in this way, you can ensure that for the majority of scenarios except loss of datacenter, the users remain operational in their primary datacenter location.

Comments

  • Anonymous
    January 01, 2003
    The comment has been removed

  • Anonymous
    July 03, 2010
    I did this design last week for a client and just did want to put it out there that each calculator will only show the amount of storage and server specs you would need in 1 DAG.  I still bring up a 3rd calculator that contains all users that would be a part of both calculators to see what the calculator recommends for storage as well as server specs.

  • Anonymous
    September 03, 2010
    Can i do the design with a single E2K10 server in the main datacenter and replicate to the dr site?

  • Anonymous
    January 06, 2011
    Hi, I think the statement "you can ensure that for the majority of scenarios except loss of datacenter" is not true. It really depends on your environment and what you do with namespaces

  • Anonymous
    April 14, 2011
    Hi, I am looking at deploying  16 active mailbox servers across 2 dags in an active/active solution.  Can this be configured so that if a WAN link fails or if a site fails then a full automatic failover will occur? Thanks ianc@tesl.com

  • Anonymous
    February 09, 2012
    The comment has been removed

  • Anonymous
    May 04, 2012
    Hi, I want to know strategy for Hub Transport Message Queue Site Resiliency options in the event of loss of a whole data center, how I can protect message which are still in HT queue and they are not yet send over to MBX server. We need to achieve Zero Message loss. We already got 2 Active data centers with Exchange 2010 servers except HT layer all other layers are redundant and losing a messages at other layer is already covered. I would really appreciate good suggestions. Thanks, JS