Sign in
 
HomeOnline20132010Other VersionsLibraryForumsGalleryEHLO Blog
TechCenter >
> 
> 
Exchange High Availability

Odpovědět Exchange High Availability

  • Friday, January 25, 2013 6:24 AM
     
     
    Sign In to Vote
    0

    Hi Everyone,

    I would like some one to advise me. I have a 15Mbps point to point link from the data center(main site) to a another site, which technology would you recommend that has worked for you without issues to make exchange highly available. I've tried DAGs on a test environment and its like it has issues coming up when one server goes down. Am imagining of a setup where if exchange goes down on the main site i will get my users up and running ASAP.

    I would also appreciate if someone can also throw in a tip on how they've made other services that rely exchange highly available i.e; external mail gateway(Mail Marshall), Blackberry Server and TMG which publishes Active Sync,Owa & Outlook anywhere.

    Thanks guys in advance

    Meshack

     

All Replies

  • Friday, January 25, 2013 5:47 PM
     
     
    Sign In to Vote
    0

    With issues do you got?

    With 2Mbps a DAG for 4000 active users works really fast when a fail in the main datacenter goes down.

    Remember that the users not only need a DAG... the also need a Client Access to connect, a HUB transport to route mail and if you got 2 servers in the DAG... a Witness!

    Esteban

     
  • Friday, January 25, 2013 10:41 PM
     
     Answered
    Sign In to Vote
    1

    In configuring DAGs, you want to keep in mind that it works in quorum fashion.  If you have 2 nodes in a DAG, you will need a File Share Witness.  If you shut down one node and the File ShareWitness, the databases will dismount because of only one healthy node.  If you are to stretch a DAG between 2 AD sites, the minimum I would have is 2 nodes in each site with a File ShareWitness in the primary site.  That way if Site 2 goes dark, you can still operate.

    As for making other services highly available:

    1. External Mail Gateway - Since this is external, you could point one path to site 1 and a second path to site 2 to deliver email.  If you can set costs on those links then make sure your primary site is receiving most/all of the mail traffic

    2. BES - you would need to setup an additional BES server in site 2 with connectivity to the same DB.  I think there is guidance on this from RIM but that is essentially what you would look at setting up.

    3. TMG - This can be setup in an array but I would have one array in each site.  That way you can make one external DNS change to point to TMG and A/S, OA, OWA would all be cutover.

    There are also some namespace planning gotchas that you need to look into for your SAN certificate. I copied these from Here since they are all written out so give credit to Ross for putting the list together :)

    • Primary datacenter Internet protocol namespace
    • Secondary datacenter Internet protocol namespace
    • Primary datacenter Outlook Web App failback namespace
    • Secondary datacenter Outlook Web App failback namespace
    • Primary datacenter RPC Client Access namespace
    • Secondary datacenter RPC Client Access namespace
    • Autodiscover namespace
    • Legacy namespace
    • Transport namespace (if doing ad-hoc encryption or partner-to-partner encryption)

    Jason Apt, Microsoft Certified Master | Exchange 2010 My Blog