vMotion fails at 68% with the error: An error occurred restoring the virtual machine state during migration (2012207)

Symptoms

• Cannot vMotion a virtual machine.
• vMotion fails at 68%.
• You see these error:
• • A General System Error
  • Failed to receive migration.An error occurred restoring the virtual machine state during migration. 
    
• If you navigate to the Virtual Machine folder, you see three vswp files instead of two
• When using Windows 8, you observe errors in the vmware.log similar to:
  
  11-01T15:39:00.158Z| vmx| DUMPER: Item 'CtrCountX' [-1, -1] not found.
  11-01T15:39:00.158Z| vmx| VMGenCtrCheckpoint: Failing restore from old Win8 checkpoint
• In the /var/log/hostd.log file on the destination host, you see entries similar to:
  
  01-18T01:51:47.816Z [392C2B90 info 'ha-eventmgr'] Event 137 : Error message on virtual_machine_name on pfc-vsphere3 in ha-datacenter: Failed to receive migration.
  -->
  18T01:51:47.817Z [38DEEB90 verbose 'vm:/vmfs/volumes/4ec6366a-da4ccca7-5c3c-d067e5f90149/virtual_machine_name/virtual_machine_name.vmx'] VMotionLastStatusCb: Failed with error 7: Failed to resume on destination.
  01-18T01:51:47.817Z [38DEEB90 verbose 'vm:/vmfs/volumes/4ec6366a-da4ccca7-5c3c-d067e5f90149/virtual_machine_name/virtual_machine_name.vmx'] VMotionResolveCheck: Operation in progress
  01-18T01:51:47.817Z [39850B90 verbose 'vm:/vmfs/volumes/4ec6366a-da4ccca7-5c3c-d067e5f90149/virtual_machine_name/virtual_machine_name.vmx'] Retrieved current VM state from foundry 5, 2
  01-18T01:51:47.818Z [39850B90 verbose 'vm:/vmfs/volumes/4ec6366a-da4ccca7-5c3c-d067e5f90149/virtual_machine_name/virtual_machine_name.vmx'] VMotionResolveCheck: Operation in progress
  01-18T01:51:47.818Z [38B95B90 info 'vm:/vmfs/volumes/4ec6366a-da4ccca7-5c3c-d067e5f90149/virtual_machine_name/virtual_machine_name.vmx'] Disconnect check in progress.
  01-18T01:51:47.818Z [38A40B90 info 'vm:/vmfs/volumes/4ec6366a-da4ccca7-5c3c-d067e5f90149/virtual_machine_name/virtual_machine_name.vmx'] Answered question _vmx1
  01-18T01:51:47.818Z [38A40B90 verbose 'vm:/vmfs/volumes/4ec6366a-da4ccca7-5c3c-d067e5f90149/virtual_machine_name/virtual_machine_name.vmx'] Retrieved current VM state from foundry 5, 2
  01-18T01:51:47.818Z [38A40B90 verbose 'vm:/vmfs/volumes/4ec6366a-da4ccca7-5c3c-d067e5f90149/virtual_machine_name/virtual_machine_name.vmx'] VMotionResolveCheck: Operation in progress
  01-18T01:51:47.822Z [38B95B90 info 'vm:/vmfs/volumes/4ec6366a-da4ccca7-5c3c-d067e5f90149/virtual_machine_name/virtual_machine_name.vmx'] Disconnect check in progress.
  01-18T01:51:47.822Z [39850B90 verbose 'vm:/vmfs/volumes/4ec6366a-da4ccca7-5c3c-d067e5f90149/virtual_machine_name/virtual_machine_name.vmx'] Version status of tools changed to: 2
  01-18T01:51:47.823Z [38B95B90 info 'vm:/vmfs/volumes/4ec6366a-da4ccca7-5c3c-d067e5f90149/virtual_machine_name/virtual_machine_name.vmx'] Unmounting the vm.
  01-18T01:51:47.824Z [38B95B90 info 'vm:/vmfs/volumes/4ec6366a-da4ccca7-5c3c-d067e5f90149/virtual_machine_name/virtual_machine_name.vmx'] VMDB unmount initiated.
  01-18T01:51:47.824Z [38B95B90 info 'vm:/vmfs/volumes/4ec6366a-da4ccca7-5c3c-d067e5f90149/virtual_machine_name/virtual_machine_name.vmx'] Unmounting VM complete.
  01-18T01:51:47.829Z [38DEEB90 verbose 'vm:/vmfs/volumes/4ec6366a-da4ccca7-5c3c-d067e5f90149/virtual_machine_name/virtual_machine_name.vmx'] VMotionStatusCb: Completed
  01-18T01:51:47.829Z [38DEEB90 verbose 'vm:/vmfs/volumes/4ec6366a-da4ccca7-5c3c-d067e5f90149/virtual_machine_name/virtual_machine_name.vmx'] VMotionResolveCheck: Firing ResolveCb
  01-18T01:51:47.829Z [38DEEB90 info 'VMotionDst (1326851502585845)'] ResolveCb: VMX reports needsUnregister = true for migrateType MIGRATE_TYPE_VMOTION
  01-18T01:51:47.830Z [38B95B90 info 'vm:/vmfs/volumes/4ec6366a-da4ccca7-5c3c-d067e5f90149/virtual_machine_name/virtual_machine_name.vmx'] Mount state values have changed.
  01-18T01:51:47.830Z [38B95B90 info 'Libs'] VmdbDbRemoveCnx: Removing Cnx from Db for '/db/connection/#9f/'
  01-18T01:51:47.830Z [38B95B90 info 'Libs'] VmdbCnxDisconnect: Disconnect: closed pipe for sub cnx '/db/connection/#9f/' (0)
  01-18T01:51:47.830Z [38DEEB90 info 'VMotionDst (1326851502585845)'] ResolveCb: Failed with fault: (vmodl.fault.SystemError) {
  -->    dynamicType = <unset>,
  -->    faultCause = (vmodl.MethodFault) null,
  -->    faultMessage = (vmodl.LocalizableMessage) [
  -->       (vmodl.LocalizableMessage) {
  -->          dynamicType = <unset>,
  -->          key = "msg.checkpoint.mrestoregroup.failed",
  -->          message = "An error occurred restoring the virtual machine state during migration.
  --> ",
  -->       },
  -->       (vmodl.LocalizableMessage) {
  -->          dynamicType = <unset>,
  -->          key = "msg.checkpoint.migration.failedReceive",
  -->          message = "Failed to receive migration.
  --> ",
  -->       }
  -->    ],
  -->    reason = "Failed to resume on destination.",
  -->    msg = ""
  --> }
  01-18T01:51:47.831Z [38DEEB90 verbose 'VMotionDst (1326851502585845)'] Migration changed state from MIGRATING to UNREGISTERING
  01-18T01:51:47.831Z [38DEEB90 verbose 'VMotionDst (1326851502585845)'] Finish called
  

Cause

This issue occurs if the state of the virtual machine cannot be determined from the .vmx file.

Starting with ESXi 5.0, by default, each virtual machine has two vswp files when the virtual machine is powered on. This issue occurs when the machine has three vswp files, instead of two.

Resolution

To work around this issue in a View environment:

1. Log in to the VMware View Administrator page.
2. Click Inventory > Desktop.
3. Click the virtual machine for which vMotion failed.
4. Click the More commands drop down and select Enter Maintenance Mode.
5. In vCenter Server, right-click the virtual machine and click Power > Power off.
6. Right-click the datastore and browse to the virtual machine storage location.
7. Verify that the vswp files are no longer present.
8. Right-click the virtual machine and click Power > Power On. 
   
   Note: Do not use Restart or Reset.
   
9. Right-click the datastore and browse to the virtual machine storage location.
10. Verify that the virtual machine has only two vswp files.
11. Perform the vMotion again.

To work around this issue in a non-View environment:

1. Connect to vCenter Server using the vSphere Client.
2. Right-click the virtual machine and click Power > Power off.
3. Right-click the datastore and browse to the virtual machine storage location.
4. Verify that the vswp files are no longer present.
5. Right-click the virtual machine and click Power > Power On. 
   
   Note: Do not use Restart or Reset.
   
6. Right-click the datastore and browse to the virtual machine storage location.
7. Verify that the virtual machine has only two vswp files.
8. Perform the vMotion again.

If the issue persists and you still see three vswp files:

1. Connect to vCenter Server using the vSphere Client.
2. Right-click the virtual machine and click Power > Power off.
3. Right-click the datastore and browse to the virtual machine storage location.
4. Right-click the vswp file and click Move To.
5. Expand and select a temporary destination location where you want to save the the files and then click Move. You can also rename the vswp file.
6. Right-click the virtual machine and click Power > Power On.
   
   Note: Do not use Restart or Reset.
   
7. Right-click the datastore and browse to the virtual machine storage location.
8. Verify that the virtual machine has only two vswp files.
9. Perform the vMotion again.

Source KB