Installing CAS role Exchange 2010

Posted on 2011-10-03
Last Modified: 2012-08-14
I'm in the process of installing Exchange 2010 into an Exchange 2003 FE/BE environment.  For budgetary purposes and to make best use of the existing new hardware I am planning to install the CAS, HT and MBX roles on both new servers.  I am also planning to setup a CAS array and a DAG, except I will not load balance the CAS array due to Microsoft limitations and no external hardware load balancer.  With that said, I need clarification on the naming convention of the DAG, CAS array and Internet facing name space.  

If I want the OWA name to be, do I use this same name for the CAS array and/or DAG name?  

When prompted for the Internet facing name during the CAS install, will I input the same name during the installation on both CAS servers or is this going to throw an error?  If I am not load balancing, I'm confused how both servers can have the same OWA address/configuration?
Question by:wpstech

    Author Comment

    i should add that I was planning to setup a DAG so that each of the (2) Ex 2010 servers would each have 2 active DB's and hold the 2 passive DB's of each other
    LVL 11

    Expert Comment

    there won't be a cas array because that would require a third server or nlb device, so you will need two different internet facing fqdn's.  One for each cas server, that is if both cas servers will be internet facing.
    LVL 6

    Accepted Solution

    i think you are confused with DAG namespace and CAS array name space.

    1. DAG name space is just for dag specific only, outlook, activesync and any other client application is not bothered with it. you can have DAG name wht ever you want, but make sure its diffrent from CAS array if you opted to have it.

    2. Cas array name space is very import now as it is used in outlook and any other client which connect to exchage server. there is an attribute named as rpcclientaccessserver which is actually on database only. if you run this command ( Get-MailboxDatabase <DB name> | fl RpcClientAccessserver) you will see you first CAS server name in it.

     x named CAS server will connect that particular databse to provide clients connectivty and in outlook in server name (where we do check name)  you will see CAS array or cas server name.

    Now important thing. if you want to keep same name space for CAS array and external URL you can have it, but in your enviorment i dont see NLB device as we have all roles running on both servers and we will have DAG too.
    reason- Windows NLB and Cluster service use few same files as both are windows components. but if you have hardware NLB then u can use CAS array.  

    You cannt use same name space for both CAS server with out CAS array.

    Author Comment

    I've read many references online to an option where DAG and CAS array can both be configured/deployed on a 2 server Exchange 2010 environment w/ all 3 roles installed on both servers, while this configuration must sacrifice load balancing it still offers HA.  

    My plan was to create a CAS array, point the DB's on both servers to the CAS array, and create a DNS record that points the CAS array to ExServer1.  I realize that this doesnt offer load balancing, which at this point is fine.  If ExServer1 ever fails all I have to do is update the DNS record to point the CAS array to ExServer2.  

    So...based on this scenario, when I begin the installation on the servers once I do the install on the 1st server should I then go ahead and create the CAS array and the DAG before installing Ex2010 on the second server, or will that even matter?  Am I going to be prompted for the "internet facing name" during the installation of CAS on the second server?  If so, is it going to let me use the same name that I used on the first installation of CAS?

    Lastly, does the DAG get a FQDN?  If so, do I need to use the same or different name for the CAS?  Does it matter?

    Here are a few links i've referenced:

    I'd really like to hear from an expert whose deployed this same 2 server scenario.  

    Author Comment

    Update:  I've also read references to creating a CAS array and pointing the CAS array name to the DAG IP in DNS, rather than pointing it directly to a physical server such as ExServer1.  

    Which of these options is the proper/correct option for deploying a CAS and DAG on a 2 server setup w/out load balancing?  

    Please advise.

    Author Comment

    LVL 6

    Assisted Solution

    1. To create CAS array you need to have 2 CAS servers. So fisrt go ahead and install 2 CAS/HUB server role if you dont want to change the RPCClientaccess attribute later. But if you have installed all roles at the same time on both server its just a matter of few commands using which you can change it.

    2. You need to specify DAG name not FQDN, its diffrent from CAS array name.
         Get-DatabaseAvailabilityGroup DAG1 -Status | fl  
    This example displays the list of servers in a DAG named DAG1. It also displays real-time status information for DAG1. In addition, several properties of the DAG, such as the file share witness directory and share configuration information are also displayed

    So DAG name will just be used as identity to find whats the status like in exchange 2007 we use

    3.Yes you will get a prompt for internet facing server when u will install 2nd exchange CAS server, you can use the same name space too.  

    Please follow these articles which can help you to deploy DAG and some info too

    Must read article and you can follow it in your deployment

    Author Comment

    the question still remains...if deploying CAS array w/ no load balancing and DAG on same boxes, do I point the CAS to the DAG IP or to the IP of one of the CAS servers?  See questions and related articles above.  I've read references to doing it both ways.
    LVL 6

    Assisted Solution

    Yes. it should go to the IP of the CAS server.

    Author Closing Comment

    still wasnt a clear answer to the question.

    Featured Post

    Maximize Your Threat Intelligence Reporting

    Reporting is one of the most important and least talked about aspects of a world-class threat intelligence program. Here’s how to do it right.

    Join & Write a Comment

    Suggested Solutions

    Title # Comments Views Activity
    Mircosoft Exchange Server 12 34
    Office365 Enrolment 2 26
    Exchange 2013 + Security Group 2 16
    MS Excel default colors in styles 2 12
    Set up iPhone and iPad email signatures to always send in high-quality HTML with this step-by step guide.
    Local Continuous Replication is a cost effective and quick way of backing up Exchange server data. The following article describes the steps required to configure Local Continuous Replication. Also, the article tells you how to restore from a backup…
    In this video we show how to create a Distribution Group in Exchange 2013. We show this process by using the Exchange Admin Center. Log into Exchange Admin Center.: First we need to log into the Exchange Admin Center. Navigate to the Recipients >>…
    how to add IIS SMTP to handle application/Scanner relays into office 365.

    728 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

    15 Experts available now in Live!

    Get 1:1 Help Now