none
Cluster Client access point and generic application resources disappeared

    Întrebare

  • We have a Windows 2008 R2 cluster hosting some generic application like Apache services and some opentext applications. It is node and disk majority with 2 nodes and a Witness disk.

    Yesterday we were removing some disks which were a part of my Windows 2008 R2 cluster. While removing disks, it got hung while removing a disk which was mounted on another cluster disk(another resource in same application group). After five mins, the entire disks(46 disks) were moved to available storage and when I checked my application group, i found it was showing just two resources, one cluster disk and one Apache application- both in failed state. 

    There was an access point with name and IP address and now missing along with another two generic applications which were a part of this application cluster. Also all Cluster disks which were a part of this application cluster is now moved to available storage. I have checked cluster registry key, and can find registry values for all those missing resources on resource hive. 

    I am still able to ping the client access point with name and IP. But I cannot find anything on application cluster. I have tried cluster restarts, node restarts, but still shows same status. I have tried failover. Though it failed over between two nodes, the status is same. 


    luni, 9 iulie 2018 08:28

Răspunsuri

  • From an administrative PowerShell prompt, run Get-ClusterResource.  It will display all the resources and what group they reside in.  We have seen that when you remove a disk from a role and it has dependents, they all move to the Available Storage Group.  Available Storage in Failover Cluster Manager is only designed to display storage.  But you can view all through PowerShell.  If they are in Available Storage, you would need to use Move-ClusterResource in order to move them out and into a role.

    Thanks, John Marlin Microsoft Program Manager

    vineri, 27 iulie 2018 15:58
  • Hi all,

    Thanks for valuable inputs. With powershell command Get-ClusterResource revealed the root cause. Missing applications and Client access point were moved to Available Storage group. In MMC, available Storage Group was showing just cluster disks and I guess Applications and Client Access point were hidden(in MMC). But powershell command showed all resources and all were under Available Storage group. 

    Resolution: I moved one cluster disk from available storage group to my application group. Viola, all applications, Cleint access point and cluster disks automatically moved to application group and issue is resolved now.

    joi, 2 august 2018 17:38

Toate mesajele

  • Hi,

    Have you tried to re-add these resources? What was the result?

    Best Regards,

    Frank



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


    marți, 10 iulie 2018 05:46
  • Hi

    How did you remove the disk? directly from the node or did you follow the sequence?

    Sequence is like, 

    1. Remove the disk resource from the application,
    2. Then removing the same from cluster
    3. Then from the node

    If you had removed directly from the node, would you be able to add the same set of disks to the nodes again?
    After having added the disks back, check the cluster resources (if required reboot after confirming all the disks been added as earlier). 

    you can also go through the regkey "cluster" under hklm and check the resources are intact in the registry

    Regards
    Bala N

    miercuri, 11 iulie 2018 13:25
  • Hi,

    Just check the situation about your issue.

    Best Regards,
    Frank

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

    joi, 12 iulie 2018 02:35
  • Hi,

    Was your issue resolved? 

    If you resolved it using our solution, please "mark it as answer" to help other community members find the helpful reply quickly.
    If you resolve it using your own solution, please share your experience and solution here. It will be very beneficial for other community members who have similar questions.
    If no, please reply and tell us the current situation in order to provide further help.


    Best Regards,
    Frank

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

    marți, 17 iulie 2018 09:45
  • Were there any dependencies on the disks and were they assigned to a group? For example, say I have a file share with dependencies set as:

    File Share depends Disk1 and NetworkName

    Network Name depends on IPAddress

    If you delete Disk1 from the group, it will prompt you saying that File Share depends on Disk1 and will also be removed.  If you answer yes, it will remove any and all dependent resources.  So in this case, when deleting Disk1, then all resources would be deleted because of the dependency structures.


    Thanks, John Marlin Microsoft Program Manager

    luni, 23 iulie 2018 16:30
  • Actually the resources are already there. I was even able to ping my cluster application IP.  I can see same from Registry. After two days all resources re appeared. But lastweek when I tried to remove another disk, again same issue occurred. 
    vineri, 27 iulie 2018 15:33
  • Yes, we did same. Removed from Cluster application , then deleted from available storage and then from node.

    The issue happened when I removed a disk which was mounted(as Y:\disk1) on another disk in same cluster(Y drive)

    vineri, 27 iulie 2018 15:35
  • All file shares were made offline. It was when I removed a disk which was mounted on another cluster disk.
    vineri, 27 iulie 2018 15:36
  • From an administrative PowerShell prompt, run Get-ClusterResource.  It will display all the resources and what group they reside in.  We have seen that when you remove a disk from a role and it has dependents, they all move to the Available Storage Group.  Available Storage in Failover Cluster Manager is only designed to display storage.  But you can view all through PowerShell.  If they are in Available Storage, you would need to use Move-ClusterResource in order to move them out and into a role.

    Thanks, John Marlin Microsoft Program Manager

    vineri, 27 iulie 2018 15:58
  • Hi John, I was trying Powershell commands. But it seems that module has to be imported. I will try same tomorrow. But is there any way to make it visible on GUI. Because my application team is not much aware of Powershell commands and they need all resources in MMC.
    vineri, 27 iulie 2018 19:43
  • Hi all,

    Thanks for valuable inputs. With powershell command Get-ClusterResource revealed the root cause. Missing applications and Client access point were moved to Available Storage group. In MMC, available Storage Group was showing just cluster disks and I guess Applications and Client Access point were hidden(in MMC). But powershell command showed all resources and all were under Available Storage group. 

    Resolution: I moved one cluster disk from available storage group to my application group. Viola, all applications, Cleint access point and cluster disks automatically moved to application group and issue is resolved now.

    joi, 2 august 2018 17:38
  • To close the loop on this, great to hear that is what the issue was.  It is actually more common than you would think over the years.  Unfortunately, no, there is not a way to expose this through Failover Cluster Manager.

    Thanks, John Marlin Microsoft Program Manager

    joi, 2 august 2018 19:30