Backups fail with "can't find parent uuid" or STATUS_ACCESS_VIOLATION messages. (4046670)

Backups of specific VMs consistently fail with one or the other of these messages. A sample is below:

VM Name: VM_Name_1
Result: Failed
Start Time: 1/12/2012 12:01:27 AM
End Time: 1/12/2012 12:02:24 AM
Duration (minutes): 1
Archive Size: 0 MB
Message: An internal error occurred during execution, please contact Quest support if the error persists. Error Message: 2568 - can't find parent uuid c540a959-433a-4c64-b7f9-7fbd8957f2eb [at var_open_write_base:340]

VM Name: VM_Name_2
Result: Failed
Start Time</strong>: 1/12/2012 12:05:46 AM
End Time</strong>: 1/12/2012 5:10:12 AM
Duration (minutes)</strong>: 305
Archive Size</strong>: 0 MB
Message</strong>: An internal error occurred during execution, please contact Quest support if the error persists. Error
Message: 0 [unknown (0x2694)] backup 9824 exception::handle:
Exception: STATUS_ACCESS_VIOLATION

VM Name: VM_Name_3
Result: Failed
Start Time: 1/11/2012 8:47:20 PM
End Time: 1/12/2012 4:44:31 AM
Duration (minutes): 477
Archive Size: 0 MB
Message: An internal error occurred during execution, please contact Quest support if the error persists. Error
Message: 0 [unknown (0x24D4)] backup 2992 exception::handle:
Exception: STATUS_ACCESS_VIOLATION

 

Both of these errors are tracable to the same cause; a backup is looking for a directory in the repository that it can't find.

First, verify that the repository credentials are working. Select the correct repository in the My Repositories view of vRanger. Right-click the repository and select Properties. When the window opens, verify that the Free Space area does not display "Unknown" as this is an indicator that vRanger cannot access the repository. If Free Space indicates Unknown, then there is a problem with the repository being unavailable, or the path or credentials being incorrect. This issue can also be caused by some anti-virus products. Be sure to correct this before continuing.

Then, using a file browser such as Windows Explorer, locate the savepoints in the repository for the failing VM. In the root of the repository, you will see a directory (folder) for each job's backup for a VM. Locate the directory for the failing VM and open that directory. Within that directory, you will see directories for each savepoint set. These consist of a full backup and all of it's dependent space-saving backups.

In the vRanger GUI, open the My Repository view and locate the entries for the failing VM

If the My Repository view shows that there are savepoints for the failing VM, then:
1. If you wish to keep the most recent savepoint set for a possible future restore, use your file browser to move the directory containing that set to another location. If you do not wish to keep it, simply delete the folder containing that savepoint set. Do not delete the master folder for that VM.
2. In the My Repository view, delete the savepoint set that you have just moved.

If there are no savepoints displayed for the failing VM:
1. If you wish to keep the savepoint sets for a possible future restore, use your file browser to move the directories containing the sets to another location. If you do not wish to keep them, simply delete the folders containing the savepoint sets. Do not delete the master folder for that VM.
 

Please consider the use of Andrew Gasyucheno's Repository Sync script to manage your repository cleanup tasks. This is located at http://communities.quest.com/message/35601#35601