none
Change subnet mask for 2012R2 Hyper-V cluster RRS feed

  • Question

  • Hi there,

    Would anyone be able to please advise as to how I can change the subnet mask for a Windows Server 2012R2 Cluster.

    There are two windows server 2012R2 hosts with failove clustering and the hyper-v role installed. 

    Each host has teams of 2X1GB nics for Management (cluster comms allowed), Live Migration, Cluster (Dedicated), VM Network.

    Host1: Management: 192.168.1.1/24, Cluster : 192.168.4.1, Live Migration: 192.168.5.1.

    Host2: Management: 192.168.1.2/24, Cluster : 192.168.4.2, Live Migration: 192.168.5.2.

    The subnet mask only needs to be changed on the Management Nic team. It needs to have a subnet mask of 255.255.254.0 as the network has changed to 192.168.0.0/23 to allow for more IPs. The DNS (192.168.1.11) and default gateway(192.168.1.254) remain the same. The new host IP on host 1 would therefore be 192.168.1.1/ 255.255.254.0 and on host 2 it would be 192.168.1.2/ 255.255.254.0.

    Thanks,

    HA

    Monday, March 23, 2015 10:26 PM

Answers

  • Hi Sam,

    Thanks for the reply. If I change the subnet mask on the hosts, the cluster IP vanishes from the nic configuration and the cluster shows as failed. Additionally, the cluster manager sees the managment network as a new network. It removes the old one and creates a new one generically called 'Cluster Network 1'.

    Switching back to the old subnet and changing the name back to Managment and restarting the hosts brings the cluster online again.

    Regards,

    HA

    What you're seeing is normal because the cluster name object is bound to both an IP and a subnet. When you change the subnet, you are literally creating a new network and the CNO no longer knows where it belongs.

    Are you sure your cluster went offline? Or did it just take the CNO offline? Your nodes should stay up and so should the roles. Only the CNO should have crashed. If anything else happened, the rest of your cluster networking is not configured correctly. You need at least one other network that the cluster can communicate across while modifying this one.

    Assuming you want to retain the original CNO IP, do the following:

    1. Change the subnet mask on one node. Ensure that nothing other than the CNO crashes. If something else goes down, fix the rest of your networking so that at least one other network can carry cluster traffic and come back.
    2. Change the subnet mask on the other node.
    3. In Failover Cluster Manager, check the Networks node. The new network should appear with a Cluster Use of Cluster and Client. If not, something is wrong with one of the IPs. Fix that and come back.
    4. Still in FCM, click on the name of your cluster in the tree in the left pane. In the center pane, near the bottom, there is a "Cluster Core Resources" section. It should show your "Server Name" as being offline, and if you expand it, it should also show the CNO address as also being offline. Double-click either item, and you should be greeted with a similar window to the following:
    5. See where it says, "No network?" That's because when I originally created this CNO, it was 192.168.25.20/24. Now that the IP is on 192.168.24/23, the CNO is lost. Now click Add. Fill out the new window like this:
    6. When you return to the first window, both entries will be there. They'll look similar, except that one shows the new network. Highlight the old one without a network and click Remove. Then click OK.
    7. Back at the main Failover Cluster Manager window, right-click on your CNO under "Server Name" in the "Cluster Core Resources" and click Bring Online. After a few moments, it should come up.
    8. For clean-up purposes, go to the Networks node and rename your new network as desired.

    I apologize for not having more screenshots. You only get two per post so I picked the ones that I thought were most relevant.


    Eric Siron
    Altaro Hyper-V Blog
    I am an independent blog contributor, not an Altaro employee. I am solely responsible for the content of my posts.

    • Marked as answer by ha20 Tuesday, March 24, 2015 1:52 PM
    Tuesday, March 24, 2015 1:23 PM

All replies

  • You just change the mask, not sure what the question is..

    Sam Boutros, Senior Consultant, Software Logic, KOP, PA http://superwidgets.wordpress.com (Please take a moment to Vote as Helpful and/or Mark as Answer, where applicable) _________________________________________________________________________________ Powershell: Learn it before it's an emergency http://technet.microsoft.com/en-us/scriptcenter/powershell.aspx http://technet.microsoft.com/en-us/scriptcenter/dd793612.aspx

    Tuesday, March 24, 2015 2:46 AM
  • Hi Sam,

    Thanks for the reply. If I change the subnet mask on the hosts, the cluster IP vanishes from the nic configuration and the cluster shows as failed. Additionally, the cluster manager sees the managment network as a new network. It removes the old one and creates a new one generically called 'Cluster Network 1'.

    Switching back to the old subnet and changing the name back to Managment and restarting the hosts brings the cluster online again.

    Regards,

    HA

    Tuesday, March 24, 2015 11:36 AM
  • Hi,

    the old school is to have separate network for Cluster comm, however nowadays no need actually for this. you can direct the HB traffic through live migration network. by the way i have same setup you have now and am doing the following:

    1- One network for VM Migration

    2- the remaining card, create Hyper-V switch on to of and allow management OS to share this network

    this will end up to have 1 Nic for Migration, 1 Nic for VM Traffic and another virtual Nic for Management

    if you are not ok with this, you can use the newly introduced feature which is Network Virtualization

    https://technet.microsoft.com/en-us/library/jj945275.aspx

    http://blogs.technet.com/b/keithmayer/archive/2012/10/08/gettingstartedwithhypervnetworkvirtualization.aspx

    Thanks,


    Kareem Behery

    Tuesday, March 24, 2015 12:27 PM
  • Hi Sam,

    Thanks for the reply. If I change the subnet mask on the hosts, the cluster IP vanishes from the nic configuration and the cluster shows as failed. Additionally, the cluster manager sees the managment network as a new network. It removes the old one and creates a new one generically called 'Cluster Network 1'.

    Switching back to the old subnet and changing the name back to Managment and restarting the hosts brings the cluster online again.

    Regards,

    HA

    What you're seeing is normal because the cluster name object is bound to both an IP and a subnet. When you change the subnet, you are literally creating a new network and the CNO no longer knows where it belongs.

    Are you sure your cluster went offline? Or did it just take the CNO offline? Your nodes should stay up and so should the roles. Only the CNO should have crashed. If anything else happened, the rest of your cluster networking is not configured correctly. You need at least one other network that the cluster can communicate across while modifying this one.

    Assuming you want to retain the original CNO IP, do the following:

    1. Change the subnet mask on one node. Ensure that nothing other than the CNO crashes. If something else goes down, fix the rest of your networking so that at least one other network can carry cluster traffic and come back.
    2. Change the subnet mask on the other node.
    3. In Failover Cluster Manager, check the Networks node. The new network should appear with a Cluster Use of Cluster and Client. If not, something is wrong with one of the IPs. Fix that and come back.
    4. Still in FCM, click on the name of your cluster in the tree in the left pane. In the center pane, near the bottom, there is a "Cluster Core Resources" section. It should show your "Server Name" as being offline, and if you expand it, it should also show the CNO address as also being offline. Double-click either item, and you should be greeted with a similar window to the following:
    5. See where it says, "No network?" That's because when I originally created this CNO, it was 192.168.25.20/24. Now that the IP is on 192.168.24/23, the CNO is lost. Now click Add. Fill out the new window like this:
    6. When you return to the first window, both entries will be there. They'll look similar, except that one shows the new network. Highlight the old one without a network and click Remove. Then click OK.
    7. Back at the main Failover Cluster Manager window, right-click on your CNO under "Server Name" in the "Cluster Core Resources" and click Bring Online. After a few moments, it should come up.
    8. For clean-up purposes, go to the Networks node and rename your new network as desired.

    I apologize for not having more screenshots. You only get two per post so I picked the ones that I thought were most relevant.


    Eric Siron
    Altaro Hyper-V Blog
    I am an independent blog contributor, not an Altaro employee. I am solely responsible for the content of my posts.

    • Marked as answer by ha20 Tuesday, March 24, 2015 1:52 PM
    Tuesday, March 24, 2015 1:23 PM
  • Hi Eric,

    Thanks for the prompt and helpful reply. You are correct. My nodes stayed up and the VMs kept running as well. Live migration across nodes worked too. It was only the CNO object that crashed. I'll try this out.

    Thanks again.

    Regards,

    HA

    Tuesday, March 24, 2015 1:52 PM