locked
Failover Cluster Issue | IP conflict RRS feed

  • Question

  • Hi All,

    I am facing one issue in my existing Fail over clustering, as we have two node clustering.

    both the server showing the IP conflict for virtual IP (Cluster IP) and failover is not happening.

    Error :

    Cluster IP address resource 'Cluster IP Address' cannot be brought online because a duplicate IP address '192.168.17.179' was detected on the network.  Please ensure all IP addresses are unique.

    192.168.17.179 is my virtual IP.

    "Ethernet adapter Private:

       Connection-specific DNS Suffix  . :
       Description . . . . . . . . . . . : Citrix PV Network Adapter #0
       Physical Address. . . . . . . . . : 1A-54-16-4F-D2-F8
       DHCP Enabled. . . . . . . . . . . : No
       Autoconfiguration Enabled . . . . : Yes
       IPv4 Address. . . . . . . . . . . : 192.168.17.178(Preferred)
       Subnet Mask . . . . . . . . . . . : 255.255.255.0
       IPv4 Address. . . . . . . . . . . : 192.168.17.179(Preferred)
       Subnet Mask . . . . . . . . . . . : 255.255.255.0
       Default Gateway . . . . . . . . . :
       NetBIOS over Tcpip. . . . . . . . : Enabled"

    (This is my one of the private network adapter setting for heartbeat)

    here i see the 179 ip, and only this is the network adapter where i get 179.

    Please suggest.

    Thanks,


    rizwan

    Friday, January 23, 2015 11:53 AM

Answers

  • The error message pretty much means what it says - it is finding that the address you are trying to use is already on the network - maybe another server is using that address.

    Is this coming from the cluster validation wizard?  Make sure you get that run without error (and hopefully without warnings).  It does a really good job of helping to find configuration issues.


    . : | : . : | : . tim


    • Edited by Tim CerlingMVP Friday, January 23, 2015 10:57 PM
    • Proposed as answer by Alex Lv Tuesday, January 27, 2015 2:52 AM
    • Marked as answer by Alex Lv Sunday, February 8, 2015 2:32 AM
    Friday, January 23, 2015 10:56 PM
  • Hi Rizwan

    Correct me if I am wrong please..
    It seems you configured the ip 192.168.17.179 as your Virtual IP for the cluster and also you assigned that ip to one of your hearbeat. If so that's the problem

    The Virtual IP of the cluster should be unique, different from any of the ips assigned to any of your NICs.

    Best regards


    Joaquin Camarero Muñoz


    • Edited by Joaquin Camarero Muñoz Monday, January 26, 2015 9:21 PM
    • Proposed as answer by Alex Lv Tuesday, January 27, 2015 2:53 AM
    • Marked as answer by Alex Lv Sunday, February 8, 2015 2:32 AM
    Monday, January 26, 2015 9:08 PM
  • Hi jcamarero,

    Moreover, it is important in any cluster that there are no NICs on the same node that are configured to be on the same subnet.

    The related article:

    http://blogs.technet.com/b/askcore/archive/2010/02/22/windows-server-2008-failover-clusters-networking-part-2.aspx

    I’m glad to be of help to you!


    Please remember to mark the replies as answers if they help and unmark them if they provide no help. If you have feedback for TechNet Support, contact tnmff@microsoft.com

    • Proposed as answer by Alex Lv Sunday, February 8, 2015 2:32 AM
    • Marked as answer by Alex Lv Sunday, February 8, 2015 2:32 AM
    Tuesday, January 27, 2015 2:57 AM

All replies

  • Hi Riz,

    Can you make the cluster group offline manually,ensure all the resources associated with this group is offline. Then failover the group to another node and see if it works?

    Thanks,

    Umesh.S.K

    Friday, January 23, 2015 11:56 AM
  • The error message pretty much means what it says - it is finding that the address you are trying to use is already on the network - maybe another server is using that address.

    Is this coming from the cluster validation wizard?  Make sure you get that run without error (and hopefully without warnings).  It does a really good job of helping to find configuration issues.


    . : | : . : | : . tim


    • Edited by Tim CerlingMVP Friday, January 23, 2015 10:57 PM
    • Proposed as answer by Alex Lv Tuesday, January 27, 2015 2:52 AM
    • Marked as answer by Alex Lv Sunday, February 8, 2015 2:32 AM
    Friday, January 23, 2015 10:56 PM
  • Not sure if this is helpful but do you also have a NLB adapter configured also, I have a similar issue and found that it was related to either my NLB or the IP of the cluster, I think the logic is that the ip address may be assigned to more than one adapter and during some failover  instances it may report as being a duplicate address. I wish I could be more specific but I do occasionally see these and ignore them as I know I haven't duplicated addresses. They only appear once in a long while.
    Monday, January 26, 2015 8:46 PM
  • Hi Rizwan

    Correct me if I am wrong please..
    It seems you configured the ip 192.168.17.179 as your Virtual IP for the cluster and also you assigned that ip to one of your hearbeat. If so that's the problem

    The Virtual IP of the cluster should be unique, different from any of the ips assigned to any of your NICs.

    Best regards


    Joaquin Camarero Muñoz


    • Edited by Joaquin Camarero Muñoz Monday, January 26, 2015 9:21 PM
    • Proposed as answer by Alex Lv Tuesday, January 27, 2015 2:53 AM
    • Marked as answer by Alex Lv Sunday, February 8, 2015 2:32 AM
    Monday, January 26, 2015 9:08 PM
  • Hi jcamarero,

    Moreover, it is important in any cluster that there are no NICs on the same node that are configured to be on the same subnet.

    The related article:

    http://blogs.technet.com/b/askcore/archive/2010/02/22/windows-server-2008-failover-clusters-networking-part-2.aspx

    I’m glad to be of help to you!


    Please remember to mark the replies as answers if they help and unmark them if they provide no help. If you have feedback for TechNet Support, contact tnmff@microsoft.com

    • Proposed as answer by Alex Lv Sunday, February 8, 2015 2:32 AM
    • Marked as answer by Alex Lv Sunday, February 8, 2015 2:32 AM
    Tuesday, January 27, 2015 2:57 AM
  • I just had the same issue. My 2-nodes Exchange cluster went offline and complaining of an IP address conflict.  The cluster came back on-line "automagically" after a few minutes.

    After thoroughly review of the subnet (A production subnet with very restrictive IP allocation) we couldn't find the issue.

     This post pointed me in vmware direction and guess what?  An automatic virtual machine migration of one of the nodes failed. (It's called vSphere DRS)

    In this case was vmware the culprit.  A rule was created to keep all nodes in a different host and only moved them if extremely necessary.  The problem had not reoccurred again.

    Wednesday, August 26, 2015 4:38 AM