Hi Tony,

Thank you for your question.

Issue1:

A: we could do that which have no active user base in the off-site datacenter.

Issue2:

A: we could refer to the following links:

http://blogs.technet.com/b/winde76/archive/2011/03/23/step-by-step-create-a-database-availability-group-dag.aspx

http://blogs.technet.com/b/rmilne/archive/2014/07/21/vmware-issues-with-exchange-dag.aspx

https://technet.microsoft.com/en-us/library/dd298065(v=exchg.150).aspx

Notice: although some links is about Exchange 2010, but it was adapted to Exchange 2013.

Issue3:

A: we think HA/DR in a single site is a better solution.

If there are any questions regarding this issue, please be free to let me know.

Best Regard,

Jim

Hello all,

I have been tasked with migrating Exchange from 2010 (SBS 2011) to Exchange 2013 (virtualised on Server 2012).  I'm still at the planning stage.  One of the requirements is that we implement High Availability and disaster recovery as we have had a few 24 hour power outages recently - Exchange online is NOT an option (although it's my first choice).

We are a single site, forest and domain of about 75 users (although have expnsion plans to increase that to about 200+).  My plan so far is to rent a couple of 2012 Data Centre servers somewhere and use these as the failover site in case we any more power outages/disasters, with a couple of DAG's.

I have been reading

https://technet.microsoft.com/en-GB/library/dd979799(v=exchg.150).aspx (amongst others) and have a few concerns about whether my plan will work.

1. Would I be able to implement this scenario if there is no active user base in the off-site (rented) data centre?
2. Are there any step by step tutorials for implementing this kind of scenario?
3. What alternatives do I have to implement HA/DR in a single site - my boss is stronly againt Exchange online?

Thanks
Tony

Hi

You would want to look at a DAG. The passive copies of the DB's will be in the DC.

My suggestion would be to get a consultant that can assist you with this deployment, the planning has to be correct especially with the number of outages you are having.