Recover VM Operations Fail with Unknown Errors

Valid on Windows operating systems.

Symptom

Recover VM jobs fail. You can submit the Recover VM job, however, the following message appears in the Activity Log:

Failed to recover virtual disks.	

In addition, VDDK reports the following error message:

Unknown Error.	

Solution 1:

To correct this problem, consider the following solutions:

• Recover VM jobs can fail when there is not enough free disk space on the original datastore. VDDK returns the message because the VDDK API (currently) does not support the capability to detect the amount of free disk space on the original datastore. (The datastore is the location where you specified to recover the virtual machine.) To correct this problem, free the amount of disk space on the original datastore that is required to complete the operation and then resubmit the job.
• Network disturbance and high network traffic can cause Recover VM jobs to fail. To correct this problem, verify that the proxy server and the ESX Server system or the vCenter Server system can communicate with each other though the network, then resubmit the job.
• Multiple concurrent connections consisting of backup or recover VM jobs to the ESX Server system or the vCenter Server system, which includes vSphere SDK connections through the VMware vSphere Client, can cause the jobs to fail. To correct this problem, close all unnecessary connections and then resubmit the job. For information about the maximum quantity of concurrent connections that are allowed, see Cannot Open VMDK File.
• Examine the Tasks and Events sections of the VMware vSphere Client log to discover internal errors for the specific virtual machine. Correct the internal errors and then resubmit the job.

  Example: Another application or operation is using the VMDK file. To correct this problem, release the file and resubmit the job.

Solution 2:

This problem can occur under the following conditions:

• VDDK did not process a snapshot properly.
• VDDK did not delete a snapshot manually or internal to the virtual machine.

To correct this problem, resubmit the job. If the job fails again, delete the recovered virtual machine and resubmit the job.