• Status: Solved
  • Priority: Medium
  • Security: Public
  • Views: 279
  • Last Modified:

MS Exchange 2007 High Availability

Hi All,

I am looking to implement a Microsoft Exchange 2007 mail server scenario.

Mail for our client is critical and i would like to have another mail server running at another site which is connected via a managed WAN.

In the event of a server going down we would like to switch over to the other replicating server partner and it can then become the primary reciever of all mail. But i would also like this replicating server to act as a local cache / front end server so that the mail data on the WAN is as minimal as possible.

I was thinking.

CCR with 2007 Exchange
Front End / Back End Scenario with Exchange 2007
Perhaps a secondary MX record?

Has anyone had a lot of experience with this and have any other better ideas. I have tested this scenario previously and it has worked with smaller mailbox servers. This server could potentially have multiple storage groups and 80-100GB of mail.

  • 3
  • 2
  • 2
2 Solutions
so problem 1) CCR is only for the mailbox role and cannot be shared with any other exchange 2007 roles.  So you would need to have dedicated mailboxes nodes in each datacentre

Check out the Exchange Storage Calcutator for infor on how many disks you will need and also the band width for CCR replication you will need http://msexchangeteam.com/archive/2007/01/15/432207.aspx

If you want to CCR across two locations, they need to be in the same AD site and you need to use windows 2008 as it will allow this.

As for the HUB and CAS roles, it's usually recommended that you have atleast two in each location for redundancy, but you could virtualise them using hyper-v or ESX (http://www.google.com/url?sa=U&start=2&q=http://technet.microsoft.com/en-us/library/cc794548.aspx&ei=dGXxSdN-ieiUB6XWkcoM&usg=AFQjCNFYfGgrcOr63dcHT8Xh-9Zdud7-dw)

As for the MX record, as long as you have internet connectivity in the both locations, you could load balance outbound and use MX preferences to deliver to both.
I would recommend that you keep both data centres active, as in the servers are running all this time and share the load
As far as I understand your needs, i'm usually in this scenario, implementing an Cross-SCR design :
Server A and B are MB, HT and CAS
Server A hosts two SG with one MB store each, replicated by SCR to server B
Server B hosts two SG with one MB store each, replicated by SCR to server A

For external access, I'm using ISA 2006 (Sp1 now), with a server farm.
I'm implementing the same certificate on ISA and both exchnage servers

Working very well, as long as you know that SCR is MANUAL switchin/failover.

ronmarccAuthor Commented:
So you can actually have mail delivering to both, i didnt realise this fact, i made the assumption lower priority MX records are just used when the primary MX is not avaliable.

I will most definately be virtualising the datacenter.

But i was just curious on the best approach in the event the datacenter has a total internet failure, so that the other exchange servers will take over and simply re-synchronise when the internet is backup.

The datacenter is virtualised with blades, so its fairly redundant.

We just have no real availability in the event of an internet issues.

Problems using Powershell and Active Directory?

Managing Active Directory does not always have to be complicated.  If you are spending more time trying instead of doing, then it's time to look at something else. For nearly 20 years, AD admins around the world have used one tool for day-to-day AD management: Hyena. Discover why

If internet fails, for sure you will not receive anything ;)
having two independent internet access is something you can implement, but it depends of your Budget.
ronmarccAuthor Commented:
Well I guess at the end of the day that is what i wanted the mail going to two different servers on two different internet connections at two different sites to avoid this from happening.

So CCR or SCR could work for either scenario, one is manual and one is automatic.

Secondary to that,

Is the topology still front end / back end.

I would like to have a local cache (so to speak) at some of the larger sites to reduce time to load mailboxes etc.. Is this something that can be implemented like 2003 frontend/backend scenarios?
how are things going?
ronmarccAuthor Commented:
Apologies, both you guys have answered my question and i am happy with the path i am moving on with.
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

Making Bulk Changes to Active Directory

Watch this video to see how easy it is to make mass changes to Active Directory from an external text file without using complicated scripts.

  • 3
  • 2
  • 2
Tackle projects and never again get stuck behind a technical roadblock.
Join Now