• Status: Solved
  • Priority: Medium
  • Security: Public
  • Views: 1601
  • Last Modified:

Help needed with Exchange 2010 namespace

In the "Namespace Planning" section of this Technet article (http://technet.microsoft.com/en-us/library/dd638104.aspx), Microsoft summarise "The key requirement is that, regardless of the naming convention you use, each datacenter should have its own unique namespace.".

I am confused as to how this should be achieved ?

Is each site a subdomain of a root forest ?  In which case, how does DAG work, because I thought DAG had to be in the same domain, not just the same forest ? (I believe this is a Windows Failover Cluster restriction that it cannot cross domains, hence it's also a DAG restriction)
0
TwentyFourSeven
Asked:
TwentyFourSeven
  • 4
  • 2
  • 2
1 Solution
 
BusbarSolutions ArchitectCommented:
because internally you cannot have the same namespace in 2 data center, or you will have to update DNS which is manual step.
0
 
TwentyFourSevenAuthor Commented:
Still confused.
0
 
BusbarSolutions ArchitectCommented:
how come you will handle mail.domain.com in the 2 data centers internally ?!
0
Creating Active Directory Users from a Text File

If your organization has a need to mass-create AD user accounts, watch this video to see how its done without the need for scripting or other unnecessary complexities.

 
TwentyFourSevenAuthor Commented:
All I want to do is achieve the recommended configuration in the Technet article I linked to in my first post. But the Technet article is not clear how this should be done.
0
 
tigermattCommented:

I hear your confusion. I can see how the mention of "namespaces" in the article could be confusing. It isn't talking about Active Directory Domains and Forests because as you said, all members of the DAG must be in the same domain. What's it's talking about is the external DNS records you use to expose the client access services in each datacenter to the Internet.

You have to plan for both datacenters to be actively providing client access services at any one time -- this could happen if you had a partial failover, such as some software issue affecting some, but not all, databases in your main datacenter.

The easiest way to discuss this might be by way of example. Let's say we have two datacenters, our primary one in London and a redundant one in Paris. The DAG is stretched across both facilities.

The London datacenter is the primary endpoint for OWA, Autodiscover, Outlook Anywhere, ECP and Exchange ActiveSync. You would create records in your external DNS namespace for owa.domain.com and autodiscover.domain.com, and point those at the Virtual IP (VIP) of the load balancer sat in front of your CAS array in London. It could be a Windows NLB array, dedicated hardware load balancer or Forefront TMG load balancing array.

If you don't expose the Paris site to the Internet, your London site has the ability to proxy OA/OWA/ECP etc to the CAS servers in the Paris site. If there were a partial failover but Client Access was still functioning in London, it would happily proxy requests to the non-Internet-facing Paris client access servers. However this is slow and cumbersome, as it relies on potentially slow WAN links between the two sites.

If the users whose mailboxes are active in Paris can log in to the Paris CAS array directly (by exposing it to the Internet), then proxying is not necessary. To expose Paris to the Internet means you need a second namespace - a second set of DNS records - dedicated to the services in that site. So you could use owa.paris.domain.com as the article suggests, or owaFR.domain.com, or anything you choose, as long as it's different to what's used in the London site. Those URLs are configured on the ExternalURL property of Paris CAS virtual directories so that London can redirect requests to the appropriate URL if a user's mailbox is activated on a database in Paris.

So if you are in a partial failover scenario of some but not all databases, both sites are exposed to the Internet. All users still access resources through owa.domain.com but London will automatically redirect requests to owa.paris.domain.com if it was necessary.

This doesn't make a difference in a full failover scenario. In a full failover where you lose Client Access services in the London site (loss of power or all routes to the Internet), you must update the main owa and autodiscover.domain.com records to point to the Paris CAS array external IP address so that users can still log on. This should be documented as part of your failover procedure.

In short: the additional records are simply there so that the Paris site is exposed to the Internet and can have requests redirected in the event of a partial failover.

-Matt
0
 
TwentyFourSevenAuthor Commented:
Awarding points to tigermatt:......
0
 
TwentyFourSevenAuthor Commented:
Now that's what I call an answer to a question !  Other "experts" on EE should take note !

Well deserved 500 points tigermatt !

Thank you !
0
 
tigermattCommented:
Glad to be of assistance. :)
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.

Join & Write a Comment

Featured Post

Get your problem seen by more experts

Be seen. Boost your question’s priority for more expert views and faster solutions

  • 4
  • 2
  • 2
Tackle projects and never again get stuck behind a technical roadblock.
Join Now