Exchange 2010 DC Failover

Hi Experts,
We are planning to do maintenance of our primary data center, and eventually, entire data center will remain unavailable for at least 10 hours, now, I need to fail-over all exchange services, and make it running from secondary data center. Could you please provide me the steps to do so.  Please find the below for server roles.

Primary Site:
 Server01: Mailbox, CAS, Hub
 Server02:        Mailbox, CAS, Hub      
Server03      Mailbox (No mailbox exist, just role installed), Client Access
Server04      ET
Secondary Site:
Server07      Mailbox, CAS, Hub
Server08      ET
i. DAG Member servers are :  Server01, Server02, Server07
ii. Outlook clients are connected via load balancer which is pointing to primary CAS array namely
 P-CASArray.domain.com and we have another CAS array in secondary site S-CASArray.domain.com.
iii. Webmail is pointing to load balancer and which is also pointing to server01 and server02.
iv. Some applications are using IMAP/POP service from the server03
v. BES service is pointing to server03
Could you please advise how I can make available the following services and steps to do data center fail-over of exchange 2010.
1.      Steps to do DAG fail-over to secondary site
2.      Outlook connectivity for clients and availability
3.      Outlook anywhere availability
4.      Webmail availability-
5.      Active Sync client connectivity
6.   BES service availability( If possible)
ipsec600Asked:
Who is Participating?
 
Adam FarageEnterprise ArchCommented:
Perform a datacenter switchover: http://technet.microsoft.com/en-us/library/dd351049%28v=exchg.141%29.aspx

Read this through, and let me know if you have questions.
0
 
ipsec600Author Commented:
Hi Adam, thank you so much for the reference link, it is most useful and clarifies most of the query, could you please clarify cas array part, in my case all outlook clients are connected to primary cas array and it is pointing to load balancer of primary site, now while I moved all DB to secondary site then what should I need to do for outlook connectivity? In secondary site we have another cas array, so should I change in DNS IP for primary cas array and point to secondary site cas array IP.
0
 
Adam FarageEnterprise ArchCommented:
in my case all outlook clients are connected to primary cas array and it is pointing to load balancer of primary site, now while I moved all DB to secondary site then what should I need to do for outlook connectivity? In secondary site we have another cas array, so should I change in DNS IP for primary cas array and point to secondary site cas array IP.

It depends on the level of DR you are partaking in, or which services or servers are being failed back and forth. Most cases, you will proxy / redirect (only services that redirect is the OWA and EAS service, the others proxy) from one site to another. This is usually a tad bit easier to deal with, as you dont have to worry about changing DNS A records and waiting for the TTL to expire.

More about proxy and redirection: http://exchangelaboratory.com/2013/04/04/exchange-proxy-and-redirection-exchange-2007-and-2010-explained/

Now with that said, lets go through a few scenarios:

Lets say you have two sites, with CAS Array 1 in Site 1 and CAS Array 2 in Site 2. Both sites have (on the CAS services) an InternalURL and ExternalURL.

If you fail over all databases within Site 1 to Site 2, but keep the CAS Arrays online when users connect into Site 1, they will redirect (or proxy, depending on the service) to Site 2 and then make there connection there. If its a proxy, it will then traverse between Site 2 to Site 1.

Here is the key to getting this done all together. If you fail back the CAS within Site 1 CAS Array, keeping one available at a time as you do patching there should be absolutely no reason you cannot keep DNS the way it is.

So no, you do not need to change DNS A records in your scenario... since you are using a load balancing solution just keep one CAS alive in each CAS Array the whole time.
0
 
ipsec600Author Commented:
Awesome!!! Thank you so much Adam for the detail insight this clarifies everything, the schedule is postponed for the time being for that I got chance to test in development environment, and will open separate question for any issue.

Once again thanks for your excellent support.
0
Question has a verified solution.

Are you are experiencing a similar issue? Get a personalized answer when you ask a related question.

Have a better answer? Share it in a comment.

All Courses

From novice to tech pro — start learning today.