SQL Server 2012 AlwaysOn Groups and FCIs Part 4

  • Comments posted to this topic are about the item SQL Server 2012 AlwaysOn Groups and FCIs Part 4

    -----------------------------------------------------------------------------------------------------------

    "Ya can't make an omelette without breaking just a few eggs" 😉

  • thank you for the wonderful post. I do have few questions:

    i) I was able to set AG with partial clustering ( no quorum ,shared resources), can i still set up auto failover in AG?

    ii) How do u decide which IP to use, does it have to be outside of the range listed in GUI?

  • curious_sqldba (5/20/2014)


    thank you for the wonderful post.

    You're welcome.

    curious_sqldba (5/20/2014)


    i) I was able to set AG with partial clustering ( no quorum ,shared resources), can i still set up auto failover in AG?

    A cluster is a cluster is a cluster. If you have no disk or fileshare witness you are relying on Majority Node Set, for this you need an odd number of nodes. How many nodes are in the cluster?

    curious_sqldba (5/20/2014)


    ii) How do u decide which IP to use, does it have to be outside of the range listed in GUI?

    Not entirely sure what you mean here, could you expand a little

    -----------------------------------------------------------------------------------------------------------

    "Ya can't make an omelette without breaking just a few eggs" 😉

  • Good read mate.

    Trying to get my head around this...Potentially a stupid question but does it allow you to:

    - create a 2 node cluster, 2 Standalone instance, with no shared disk?

    If so, in the above scenario:

    - If primary fails, same as DB mirroring: force online with data-loss accepted

    - If secondary instance fails, does that still wipe out the AWO group? (on node majority).

    Carlton.

  • Carlton Leach (5/21/2014)


    does it allow you to:

    - create a 2 node cluster, 2 Standalone instance, with no shared disk?

    Yes that's pretty much a basic configuration with an AlwaysOn group across 2 replicas

    Carlton Leach (5/21/2014)


    If so, in the above scenario:

    - If primary fails, same as DB mirroring: force online with data-loss accepted

    If you don't have synchronous with auto failover then yes.

    Carlton Leach (5/21/2014)


    - If secondary instance fails, does that still wipe out the AWO group? (on node majority).

    Carlton.

    not if you configure a cluster witness appropriately

    -----------------------------------------------------------------------------------------------------------

    "Ya can't make an omelette without breaking just a few eggs" 😉

  • Thanks for the informative article! I realize that this article is almost a year old, but I have a few questions that I'm hoping you can help me with.

    I see that you have different instance names for the FCI and the stand-alone instances within the AG. Would it be possible to use the same instance names for all instances? I'm not sure if that would be allowed since all instances are part of the same WSFC.

    Would it be possible to use the same drive letters\paths for the data and log files with all instances? This would prevent having issues with things like ALTERing the DATABASE to ADD FILEs on the primary and having it fail due to the drives being different on secondary replicas.

    Would application and user connections connect to the AG listener name?

    How reliable has this configuration been? Have you run into any unanticipated issues with this setup?

    Thanks!!

Viewing 6 posts - 1 through 5 (of 5 total)

You must be logged in to reply to this topic. Login to reply