locked
Remants of an old recovery database RRS feed

  • Question

  • Recently in our Exchange 2007 CCR setup, I tried to failover the active node to the passive node for maintenance.  Upon doing so, I was presented with this error:

    ---

      Summary: 1 item(s). 0 succeeded, 1 failed.
      Elapsed time: 00:00:00


      Move mail02
      Failed

      Error:
      The storage group, 'f42b9342-60c9-4a66-a2b7-a1d45a5b1ef6' for the managed database, 'Recovery/16 (mail02)', could not be found in Active Directory.

      Exchange Management Shell command attempted:
      move-ClusteredMailboxServer -Identity 'mail02' -MoveComment 'moving' -TargetMachine 'pobox02'

      Elapsed Time: 00:00:00

    ---

     

    The applicaiton log shows MSExchangeRepl event ID 114 with the info of:

      The Move-ClusteredMailboxServer operation for 'issommail02' failed. Verify that the clustered mailbox server is online.

     

    The Database Recovery Management tool does not show a recovery database present, and I don't see a database directory or file for that old recovery group.

    Where is Exchange seeing this recovery group?  How can I eliminate it so I can failover my server?


    • Edited by SOMUM Thursday, December 8, 2011 6:30 PM
    Thursday, December 8, 2011 12:31 PM

Answers

  • Failover Cluster Manager does indeed show an offline resource of "Recovery/16 (mail02)."   

    So, would removing the offline resource in Cluster Manager clean everything up?

    Hi SOMUM,

    Yes, please remove the offline resource and try to failover the active node again.

     


    Frank Wang

    TechNet Community Support

    • Marked as answer by SOMUM Monday, December 12, 2011 1:28 PM
    Monday, December 12, 2011 8:04 AM

All replies

  • Hi SOMUM,

    Please run the cmdlet Get-MailboxDatabase to check whether you can find the DB.

    Please also run the TEST-ReplicationHealth & Get-StorageGroupCopyStatus and post the results here.

    Please go to Cluster Administrator to check whether there is any failed resource.

    At last, please run Exbpa in the Toolbox to do a Health Check.

    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 Subscriber Support, contact tnmff@microsoft.com


    Frank Wang

    TechNet Community Support

    Friday, December 9, 2011 6:20 AM
  • Ran the commands on both nodes.  Get-MaiboxDatabase did not show the old Recovery group.  Test-ReplicationHealth had these results:

    ---

    Server          Check                      Result     Error
    ------          -----                      ------     -----
    POBOX02R   ClusterNetwork             Passed
    POBOX02R   QuorumGroup                Passed
    POBOX02R   FileShareQuorum            Passed
    POBOX02R   CmsGroup                   *FAILED*   Clustered Mailbox server 'mail02' currently hosted on node 'POBOX02' is not online. Server 'MAIL02' may not be accessible by clients when IP Address, Network Name, and other critical resources are offline. The following resources are offline:
                                                                  Recovery/16 (mail02): Offline
    POBOX02R   NodePaused                 Passed
    POBOX02R   DnsRegistrationStatus      Passed
    POBOX02R   ReplayService              Passed
    POBOX02R   DBMountedFailover          Passed
    POBOX02R   SGCopySuspended            Passed
    POBOX02R   SGCopyFailed               Passed
    POBOX02R   SGInitializing             Passed
    POBOX02R   SGCopyQueueLength          Passed
    POBOX02R   SGReplayQueueLength        Passed

    ---

     

    Get-StorageGroupCopyStatus came back without any trace of the old group.  Failover Cluster Manager does indeed show an offline resource of "Recovery/16 (mail02)."  EXBPA's health check made no mention of this problem.

     

    So, would removing the offline resource in Cluster Manager clean everything up?

    Friday, December 9, 2011 1:00 PM
  • Failover Cluster Manager does indeed show an offline resource of "Recovery/16 (mail02)."   

    So, would removing the offline resource in Cluster Manager clean everything up?

    Hi SOMUM,

    Yes, please remove the offline resource and try to failover the active node again.

     


    Frank Wang

    TechNet Community Support

    • Marked as answer by SOMUM Monday, December 12, 2011 1:28 PM
    Monday, December 12, 2011 8:04 AM