locked
Hardware upgrade of Failover Cluster nodes RRS feed

 > 

  • Question

  • Question
    Sign in to vote
    0
    Sign in to vote

    Hello,
    I have planned an CPU and Memory upgrade of two Windows servers 2012 that are nodes in Failover Cluster.

    I am a bit worried about the Cluster after the replacement of the hardware. 

    If you have some knowledge about this situational can you please let me know what kind of issue I can expect from the Cluster?
    I guess that the Failover will not be working before I run the validator again and before I reactivate the windows servers.

    I'll be really grateful of some information on this topic!


    Thank you in advance!

    Best Regards,

     Kristiyan Dobrev

      
    Thursday, February 15, 2018 8:14 AM

All replies

  • Question
    Sign in to vote
    1
    Sign in to vote
    Should not have an issue.  Memory upgrade will simply be recognized.  CPU upgrade may require a BIOS upgrade, but that has nothing to do with the cluster.  If running VMs on the cluster, you may end up with a situation that prevents Live Migration of VMs unless the VMs are configured for processor compatibility.

    tim

    Thursday, February 15, 2018 1:04 PM
  • Question
    Sign in to vote
    0
    Sign in to vote
    Thank you very much!

    BIOS upgrade will be performed as well.

    Next week I'll post the result of the upgrade.
    Friday, February 16, 2018 7:38 AM
  • Question
    Sign in to vote
    0
    Sign in to vote

    Some more recommendations:

    - drain roles/

    - set maintenance mode for CSVs/

    - enable compatibility checkmark in VM settings if different models of CPUs will be used/


    Sunday, February 18, 2018 9:21 PM
  • Question
    Sign in to vote
    0
    Sign in to vote

    Hi Kristiyan Dobrev,

    Just to check have you got any progress with the issue, welcome to feedback the result. And if the above reply could be of help, you may mark useful reply as answer.

    Best Regards,

    Anne

    Please remember to mark the replies as answers if they help.
    If you have feedback for TechNet Subscriber Support, contact tnmff@microsoft.com.

    Thursday, February 22, 2018 6:54 AM
  • Question
    Sign in to vote
    0
    Sign in to vote

    Hello,

    Sorry for the late respond!


    Everything went extremely well.

    We had a window of 8 hours so we shutdown every VM on both nodes and than the nodes them self.

    After flashing the new BIOS and changing the Memory and CPU we started the machines one by one. There was no issue with anything. Cluster is working fine and there was no windows licencing issue as i suspected earlier.

    Thank you all :)

    Best Regards,

    Kristiyan Dobrev

     
    Thursday, March 15, 2018 8:04 AM
  • Question
    Sign in to vote
    0
    Sign in to vote
    You do not mention if you tested live migration to the new hosts.  If that is not successful, you may need to enable processor compatibility.

    tim

    Thursday, March 15, 2018 1:14 PM