SQL Server Clustering SQL 2000/2005/2008 Dependencies between Cluster Groups

  • Can you create dependencies between Cluster Groups. The reason I am asking or am I missing something, but If you follow microsofts instructions for clustering you can end up with some resources not failing over.

    If you do as microsoft say and separate the SQL Server cluster group and it resources, disks etc on from the cluster group with the Quorum drive on, then if something fails in either group it only failsover that group. So you can end up with server1 being live, but the resoures i.e the disk for SQL server on server2.

  • Your disk definitions need to be in the same cluster group as the one you want to install your sqlserver instance on.

    Create a cluster resource group that only holds the resources, needed for the cluster. (msdtc, quorum, cluster name and cluster ip address).

    Create another cluster group for each sqlserver instance you want to host on the cluster. This group will start with the allocation of one or more drive.

    The install process of sqlserver will then define the instances ip address, name, services (dbengine, sqlagent, ...) in the targeted resource group.

    It (the install process) will also add the needed dependencies in the resource group.

    Johan

    Learn to play, play to learn !

    Dont drive faster than your guardian angel can fly ...
    but keeping both feet on the ground wont get you anywhere :w00t:

    - How to post Performance Problems
    - How to post data/code to get the best help[/url]

    - How to prevent a sore throat after hours of presenting ppt

    press F1 for solution, press shift+F1 for urgent solution 😀

    Need a bit of Powershell? How about this

    Who am I ? Sometimes this is me but most of the time this is me

  • If a resource fails on one of the sql instances what will make the server failover. The way I understand it only one server if live at any one time and this server is the one with access to the quorum drive. So if another server group failsover it will move to the failover server but the group with the quorom will stay where it is.

  • Jamie Wormald (8/3/2010)


    If a resource fails on one of the sql instances what will make the server failover. The way I understand it only one server if live at any one time and this server is the one with access to the quorum drive. So if another server group failsover it will move to the failover server but the group with the quorom will stay where it is.

    Depending off your settings for the cluster group members failover may occur.

    Each resource group can failover by itself.

    You can fail over one resource group indepentent from other resource groups.

    Both nodes of your cluster can have resource groups active. Off course a resource group can only be active at one node at a time.

    Both nodes must be up for failover to be able to use the mechanisme.

    So a passive node, is a computer that mainly functions as a heating device until the cluster service or a cluster admin moves and activates a resouce group to it.

    The resource group holding the quorum, can be active on any of the nodes.

    I always put the quorums resouce group at the node where i want to install an instance, for the rest I don't care where it is active.

    Johan

    Learn to play, play to learn !

    Dont drive faster than your guardian angel can fly ...
    but keeping both feet on the ground wont get you anywhere :w00t:

    - How to post Performance Problems
    - How to post data/code to get the best help[/url]

    - How to prevent a sore throat after hours of presenting ppt

    press F1 for solution, press shift+F1 for urgent solution 😀

    Need a bit of Powershell? How about this

    Who am I ? Sometimes this is me but most of the time this is me

  • Thank you very much.

    I think I understand the idea, as long as the resource for the application or instance are in the same cluster group then it does matter which node they are on, even if the primary cluster group with the quorom is on another node.

    I have just tested this theory by failing over our uat instance and it is still serving.

  • Jamie Wormald (8/3/2010)


    The way I understand it only one server if live at any one time and this server is the one with access to the quorum drive.

    is this a Windows 2003 or 2008 cluster?

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

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

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

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