none
Removing and replacing virtual network adapter on one node in an Always-on SQL cluster RRS feed

  • Question

  • We have a SQL Always-on cluster consisting of two nodes. They are both virtual machines running Windows 2012 and SQL Server 2012 on VMWare.

    I want to change the adapter type from E1000 (Windows 2012 suffers from instability when using this type of virtual NIC) to VMXNET3 on one of the servers. The server in question is usually the secondary node in the cluster, which is used in case of failover. Hopefully this makes what I want to do easier? This would be achieved by removing the network adapter and adding another of the correct type. The same network details (IP address etc) will be applied to the new virtual NIC.

    I am unsure of the best way to go about doing this and ensuring that every works as it did previously after the change. Should I remove the server from the cluster temporarily (how is this achieved?) for example?

    Any thoughts or advice would be most welcome.

    Kind regards


    Friday, February 19, 2016 11:19 AM

Answers

  • Hi,

    If your resources are located and active on the node with proper hardware configuration, you can stop the cluster service on the second node-> rightclick node, more actions, stop cluster service.

    What I did as test was (2x 2012r2std, sql 2012, 2x NIC for each vm)

    Stop service (in cluster manager)

    Disable Cluster Service in services.msc (just to make sure it wont start until node is ready)

    Uninstall NIC's in device manager.

    Shutdown VM

    Remove/add NIC in hyper-v

    Start VM

    Configure IP address's

    Set Cluster Service to Automatic

    Restart VM

    And then the node came up by it self, and I could move the sql ressource just for testing.

    /\


    Best Regards,

    Jesper Vindum, Denmark

    Systems Administrator

    Help the forum: Monitor(alert) your threads and vote helpful replies or mark them as answer, if it helps solving your problem.

    Friday, February 19, 2016 11:00 PM

All replies

  • Hi,

    If your resources are located and active on the node with proper hardware configuration, you can stop the cluster service on the second node-> rightclick node, more actions, stop cluster service.

    What I did as test was (2x 2012r2std, sql 2012, 2x NIC for each vm)

    Stop service (in cluster manager)

    Disable Cluster Service in services.msc (just to make sure it wont start until node is ready)

    Uninstall NIC's in device manager.

    Shutdown VM

    Remove/add NIC in hyper-v

    Start VM

    Configure IP address's

    Set Cluster Service to Automatic

    Restart VM

    And then the node came up by it self, and I could move the sql ressource just for testing.

    /\


    Best Regards,

    Jesper Vindum, Denmark

    Systems Administrator

    Help the forum: Monitor(alert) your threads and vote helpful replies or mark them as answer, if it helps solving your problem.

    Friday, February 19, 2016 11:00 PM
  • Hello

    Many thanks for your prompt and thorough response. This sounds like a good way to I will give your method a try shortly. I will let you know how I get on.

    Many thanks for you help once again.


    • Edited by Netmales Monday, February 22, 2016 9:38 AM
    Monday, February 22, 2016 9:38 AM