Inconsistently Long Restore Times for Cloned Environments RRS feed

  • Question

  • I have automated tests running throughout the day.  I can restore the same cloned environment several times during the same day.  Sometimes the process will take two minutes.  Sometimes it will take over two hours.  I've also seen this behavior with different clones.  I've tried moving the environment to different Hyper-V servers and saw the same behavior.  Utilization on the servers is low.

    Windows Server 2008R2 (core on Hyper-V hosts), MTM 2010 RTM, SCVMM 2008 R2

    • Moved by Lily_WuModerator Thursday, November 1, 2012 6:02 AM relate to lab management (From:Testing with Visual Studio Test Manager (MTM))
    Wednesday, October 31, 2012 6:08 PM

All replies

  • Hi TI,

    Thank you for your post.

    I am trying to involve someone familiar with this topic to further look at this issue. There might be some time delay. Appreciate your patience.

    Thank you for your understanding and support.

    Best Regards,

    Lily Wu [MSFT]
    MSDN Community Support | Feedback to us

    Thursday, November 1, 2012 6:02 AM
  • Hi TI Dave,

    Sorry for the late response. And could you please offer me more information about in which scenario the process takes 2 minutes and in which scenario it takes 2 hours? Also take a look at the event logs in the Event Viewer to see if there is anything useful, if so, post here.


    Vicky Song [MSFT]
    MSDN Community Support | Feedback to us

    Monday, November 5, 2012 9:14 AM
  • We have a similary problem. the restore to snapshot operation takes longer and longer time. You can see this in the tfs build log.

    It helps to reboot the hyper-v host. The MS support team was unable to find a reason, so now we are rebooting all hosts every week and planing to do an upgrade to 2012 as soon as it fully supports 2012 on all servers (waiting for SCVMM 2012 sp1)

    Monday, November 5, 2012 10:03 AM
  • Sorry for taking so long to reply.  I did not realize that you had responded. The exact same environment may take 2 minutes or 4 hours at any given time.  I found the following errors on the SCVMM (but nothing at the time of the slowness): 

    Volume Shadow Copy Service error: Unexpected error VSS_E_WRITER_STATUS_NOT_AVAILABLE. An older active writer session state is being overwritten by a newer session. The most common cause is that the number of parallel backups has exceeded the maximum supported limit.  hr = 0x80042409. 

    The Collect Procedure for the "DTSPipeline100" service in DLL "C:\Windows\system32\DTSPipelinePerf100.dll" generated an exception or returned an invalid status. The performance data returned by the counter DLL will not be returned in the Perf Data Block. The first four bytes (DWORD) of the Data section contains the exception code or status code.

    • Edited by TI Dave Monday, November 19, 2012 8:55 PM
    Monday, November 19, 2012 8:35 PM