SQL server 2005 2 node active/passive cluster manual failover method

  • Hi,

    I will be doing a manual failover of our single instance, 2 node active/passive cluster soon. I am not very experienced when it comes to SQL clustering, and inherited this cluster when I started the job a few months ago.

    In cluster Administrator We have 2 groups, Clustergroup and SQLServergroup. I understand to initiate the manual failover from cluster administrator you right click on the group and select move group.

    What I need to know is

    1. do I need to perform this action on each group

    2. If I perform this action on the clustergroup, will it automatically move the SQLserver group

    Clustergroup has the quorum disk, cluster ip address, cluster name, and MSDTC resource

    SQLserver group has sql server, sql server agent,sql server fulltext and analysis services. Also the shared data drives, SQL ip address, SQL network name and GxClusPlugin (whateverr that is!)

    Any response will be gratefully received.

  • If all the resources required for the SQL Services are in a group and dependencies have been configured correctly, the yes, move the group and all relevant resources will failover.

    -----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
    This thing is addressing problems that dont exist. Its solution-ism at its worst. We are dumbing down machines that are inherently superior. - Gilfoyle

  • Thanks for your response Henrico. I understand that the resources belonging to each group will failover if I move the group. But as I have 2 separate groups, Cluster group and SQLserver group, do I need to fail each of these over separately? Or if I failover the Cluster group will SQLserver group follow suit. There is no obvious depency set up between groups.

    If you right click on the properties for each group you get 3 tabs, general, failover and failback. Nothing about dependencies

  • Looks like the 2 groups don't failover together in case of a manual failover, you have to manually fail over each. It is whether you need to failover the quorum (cluster group) first or the SQL services, shared disks (sqlserverGroup) first.

    Any thoughts?

    Alternatively I could just reboot the active server, then the failover will take care of itself, I would have liked to do it in a more controlled way though.

Viewing 4 posts - 1 through 3 (of 3 total)

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