Exchange 2010 final draft

Oh man! Ad-blocking software has been detected! :'(

This website is run by the community, for the community... and it needs advertisements in order to keep running. Blocking our ads means your killing our stats!
Please disable your ad-block, or become a premium member to hide all advertisements and this notice.

Hey guys,

You've probably seen my earlier posts regarding Exchange 2010 and DAGs.. I've pretty much got it all down, just going over final details. Here's what I am thinking:

1. Two sites (Active/Active)
2. Two DAGs with an FSW in each site for each DAG
3. Total of 6 mailbox servers, 3 in each site... with one mailbox server serving as a backup for each site
4. for the time being we're going to use 2 active databases and passive/lagged DB's as well
5. CAS array and Hub /WNLB
6. No Edges servers, we have our own SMTP relay

I've never done an Active/Active Exchange 2010 deployment so it should be fun... lots to learn to!

 

Oh man! Ad-blocking software has been detected! :'(

This website is run by the community, for the community... and it needs advertisements in order to keep running. Blocking our ads means your killing our stats!
Please disable your ad-block, or become a premium member to hide all advertisements and this notice.

How many users are we talking hear mate?
Will these be physcial or virtual servers?

When we implement DAG's at a single site we generally have two mailbox servers as follows:

Server 1 - Active Database General Users
Server 2 - Passive Database General Users
Server 1 - Passive Database Directors
Server 2 - Active Database Directors

Server 3 - FSW

Assume you have less than 8 servers in your CAS Array with NLB, where is the incoming traffic entering for the authorative domain? Site A or B?
What happens if the primary site goes down? Your Anti Spam provider will have a secondary route, but how are you planning on failing over the CAS Servers autodiscover URL's for Outlook etc? What time frames for failover/failback are you hoping to achieve?

Are you going to have an Active copy of each database per site?

Have you got seperate subnets in place for the DAG Heartbeat traffic, both locally and across the WAN if you are doing site failover? Is the site latency <250ms (should be LOL)

If you are going to be doing site failover, which site are you placing the FSW in, the active or passive site?

Sorry, just seen this and would be good to know your thought process mate.

 

1. We're looking at 1200 users... spread out across two sites with half of the users in one active site the other half in another active site. For sure both sites will have to be scaled to support 1200 users at the minimum in case of failovers.

2. The databases are going to be setup similar to what you have outlined above.

3. Because we're using 2 DAGs there will be one FSW in each site for each DAG (2DAGs in this case)

4. There will be two servers in the CAS array at each site and 2 hub servers at each site.

5. This will be virtualized in VMware vSphere 5 environment

6. For auto redirection in case of failover, I am thinking using MAPI connection for fast connections and RPC over HTTPS (Outlook Anywhere) for slow connection which would point to the CAS in other site. Still looking at other ways

7. Regarding Authoritative traffic, that I still have to work out... Gotta talk to my network guys as well and see what can be worked out

8. Time frames for failovers I am hoping for 5 minutes at longest... but still kind of looking into this...

9. Yes each site is going to have an active copy for the local users and the passive/lagged copies should be replicating to the other active site

10. Yea we already have a dedicated vlan for the heartbeat in Exchange 2007 CCR and same would apply for the DAG heartbeat in Exchange 2010. The latency is about 90ms, so should be alright.

11. Because we're looking at two DAGs, there will be two FSW's, one in each site as mentioned above

For now its an "overview" nothing is set in stone for the nitty gritty technical specs although we at that stage right now.