Exchange 2010 DAG Disaster recovery drill

Posted on 2014-04-22
Last Modified: 2014-05-10

I have three nodes in my DAG, 2 at primary and one at DR site. Everything is working fine and I have tested the failover by shutting down the servers at the primary site and ensuring we can still send/receive emails when the databases are mounted at the DR site.
The problem I'm facing is with the DR process at the bank. The DR team wants to test DR using the same process used by other application. During this drill, all replication is cut off between the primary and secondary site. The goal is to have people working at the primary site without any interruptions, while the DR team utilizes the servers at the DR site to perform their job functions, including email.
I am not sure how I would bring the replicated copies of the exchange databases at the DR site while the primary servers are still up and databases are mounted. I cannot remove the DR server from the DAG group during this process, because this requires removing the copy of the database first. Is there a way to isolate the exchange server at the DR site without removing it from the DAG, and mount the database so users can perform DR activities. Also if it makes any difference, these are Virtual machines running on VMware.

Thanks for your comments and suggestions.
Question by:Antonio02
LVL 37

Expert Comment

by:Jamie McKillop
ID: 40017317

Using a single DAG, you cannot do what you are being asked to do. If communication between the sites is severed, the databases can only be mounted in one site or the other. To do what you want, you need to have two separate DAGs. You could then have DAG "A" active in site "A" and DAG "B" active in site "B". This would require you to double the number of Exchange servers you have.

LVL 41

Expert Comment

ID: 40017935
There are lot of myths with DAG, Remember DAG is only a HA for Mailbox role, so it provides HA for Mailbox role only. The solution you are looking practically will not work. Secondly, you need HA for all other roles, Load balancers, DNS etc. In any case you can only have one site active and require lot of manual changes.
LVL 42

Expert Comment

ID: 40018311
I think that you would need to pretty much clone your environment in DR including all Exchange servers and at least one domain controller, and then bring it all up on an isolated network, and then test on the isolated network. Otherwise you will get conflicts between the live environment and the DR environment copies of the Exchange databases.

I don't know why your DR team doesn't just fail over Exchange to DR and foo their testing then. From what you have said the main business processes already are confirmed to work, so I would think that you can fail over worth no impact to your main users and then do further testing.
Complete VMware vSphere® ESX(i) & Hyper-V Backup

Capture your entire system, including the host, with patented disk imaging integrated with VMware VADP / Microsoft VSS and RCT. RTOs is as low as 15 seconds with Acronis Active Restore™. You can enjoy unlimited P2V/V2V migrations from any source (even from a different hypervisor)


Author Comment

ID: 40023956
Thanks so much for your comments. In conclusion, what I am hearing is that even though there is no network connectivity between the primary site and the DR site, the DR server holding the copy of the active database cannot be mounted. Our goal was to take a snapshot of the DR server, mount the ex database, login and launch outlook. Perform our DR site, unmount the database, revert back to the snapshot, establish network connectivity with the DR server, and resume database copy.

Just so you know, the DR site is on a completely isolated network, and we have the same AD structure duplicated on both the primary and secondary sites.

Thanks again
LVL 37

Accepted Solution

Jamie McKillop earned 500 total points
ID: 40027217
If you take snapshots of your Exchange server and all your domain controllers in the DR site, then revert back to those snapshots before you restore connectivity between the two sites, you could get the databases mounted.

Just follow the instruction here to do the failover to the DR site -

You would skip the failback process as you would be reverting back to the snapshots that were taken before you initiated the failover process.


Author Comment

ID: 40043069
Thank you all so much for your replies. Truly appreciated.

Author Comment

ID: 40056688
Thanks again, we have a DR drill in two weeks. I will let you know how it went.

Featured Post

Too many email signature changes to deal with?

Are you constantly being asked to update your organization's email signatures? Do they take up too much of your time? Wouldn't you love to be able to manage all signatures from one central location, easily design them and deploy them quickly to users. Well, you can!

Join & Write a Comment

Suggested Solutions

Scenario:  You do full backups to a internal hard drive in either product (SBS or Server 2008).  All goes well for a very long time.  One day, backups begin to fail with a message that the disk is full.  Your disk contains many, many more backups th…
Restoring deleted objects in Active Directory has been a standard feature in Active Directory for many years, yet some admins may not know what is available.
In this video, we discuss why the need for additional vertical screen space has become more important in recent years, namely, due to the transition in the marketplace of 4x3 computer screens to 16x9 and 16x10 screens (so-called widescreen format). …
Windows 8 came with a dramatically different user interface known as Metro. Notably missing from that interface was a Start button and Start Menu. Microsoft responded to negative user feedback of the Metro interface, bringing back the Start button a…

708 members asked questions and received personalized solutions in the past 7 days.

Join the community of 500,000 technology professionals and ask your questions.

Join & Ask a Question

Need Help in Real-Time?

Connect with top rated Experts

17 Experts available now in Live!

Get 1:1 Help Now