none
Manually Merged Checkpoints and Starting a VM afterwards RRS feed

  • Question

  • Good morning,

    I recently needed to manually merge a checkpoint into the parent for about 5 VHDXs. The merge completed successfully. There one checkpoint file for each of the 5 virtual drives.

    I backed up the original data (avhdxs and vhdxs to another location before proceeding with the merge).

    After the successful merge, I updated the virtual machine by pointing it away from the avhdx files and to the vhdx files and attempted to apply the changes.

    When trying to apply the changes, I receive the following message:

    

    What should I do at this point?

    Respectfully,


    Tuesday, January 26, 2016 4:50 PM

Answers

  • All that should have been necessary was to delete the checkpoint that was represented by the AVHDX.  the system would have taken care of the merge.

    After that, you are getting standard errors because the system thought there was a checkpoint.  And you manually went around it.

    Since you manually went around it, be sure your copies are intact as you might need them.  Once oyu know they are good, push on and see what you get.


    Brian Ehlert
    http://ITProctology.blogspot.com
    Learn. Apply. Repeat.

    Tuesday, January 26, 2016 5:22 PM
    Moderator

All replies

  • All that should have been necessary was to delete the checkpoint that was represented by the AVHDX.  the system would have taken care of the merge.

    After that, you are getting standard errors because the system thought there was a checkpoint.  And you manually went around it.

    Since you manually went around it, be sure your copies are intact as you might need them.  Once oyu know they are good, push on and see what you get.


    Brian Ehlert
    http://ITProctology.blogspot.com
    Learn. Apply. Repeat.

    Tuesday, January 26, 2016 5:22 PM
    Moderator
  • Hi Brian,

    Just to explain why I used the manual approach: The checkpoint was created a long time ago and never deleted. Once I deleted the checkpoint via Hyper-V Manager, the pending merge would have taken forever to complete as it was stuck on 6% for hours.

    This might have been the wrong thing to do by based on my research, it was something that was possible and helped others with similar problems.

    From what you are saying in your response, are you suggesting that I refer to the backups that I made and wait for the system to merge the backups? Or is there anything that I can do to use the merged files?

    Thanks for your response!

    Tuesday, January 26, 2016 5:30 PM
  • You stated that you already manually merged the checkpoint disks.

    So, the system is out of sync. it thinks a checkpoint exists that is not there.

    - if I followed everything correctly.

    Therefore, if your copy of the VM is good (make sure it is) and on a different Hyper-V Server (this is key to ensure your copy is not linked to the original - unless you used export to make a copy, or backup software to make the copy).

    Knowing that your backup is good do one of the following:

    • change the settings of the VM to point to the merged virtual disk
    • Delete the VM and Import, and fix the virtual disk paths (manually copy the settings in case you need to recreate the VM).
    • Create a new VM and link to the merged virtual disks. Delete the original.

    After any of those, test that all your data is there before moving forward.  If data is missing, then recover a copy of the VM and try over again.

    I suggested some of the above, as it gets rid of the now orphaned checkpoint(s).

    The merge being stuck at 6% for hours tells me that something was happening, such as a full disk, slow disk, or something else.

    Also, I am assuming that you are running Hyper-V 2012 R2, which should have live merged the checkpoints without having to turn off the VM....


    Brian Ehlert
    http://ITProctology.blogspot.com
    Learn. Apply. Repeat.

    Tuesday, January 26, 2016 5:41 PM
    Moderator