After installing patch 1 for Veeam Backup & Replication 8, jobs may start to fail with the following error message:
Failed to create processing task for VM Error: Disk '<GUID>' not found.
The error is caused by the wrong disk IDs that were added to the database when the Veeam Backup & Replication 8 was running. This issue has been solved in patch 1, but wrong IDs still remain in the database and, in some cases of processing VMs in Direct SAN or Hot-Add mode, can affect the jobs.
To solve the issue, the wrong disk ID should be removed from the used Veeam database. We recommend backing up your database prior to making any changes. To apply the query, follow the next steps:
1. From SQL Management Studio, connect to the instance containing the Veeam configuration database (default names are VEEAM, VEEAMSQL2008R2 or VEEAMSQL2012);
2. Expand “Databases”;
3. Right-click “VeeamBackup” > New query;
4. Paste the appropriate query (below) and press “execute” or F5
DELETE FROM Physhost_disk WHERE disk_id = ‘GUID’
You will need to paste the GUID from the error message in the job to the query. For example, if the error message looks like:
Failed to create processing task for VM Error: Disk '3f8688c2-d5ae-4ac8-be11-2257097f8eb9' not found.
The query should look like:
DELETE FROM Physhost_disk WHERE disk_id = '3f8688c2-d5ae-4ac8-be11-2257097f8eb9'
5. Close the Veeam GUI and run Veeam again to force the GUI to update.
NOTE: This is an operation that Veeam Support is happy to walk through with you and this KB is provided mostly as a convenience for customers wishing to perform the above operations themselves. If you would like to walk through this process with a representative of Veeam Customer Support, please do not hesitate to open a case with us via the Customer Portal.
In some circumstances there may be multiple orphaned disk references in the database, the error shown in the GUI will only display the first problematic GUID the software comes across. With this in mind it is advised to run a test job after running the above query, if the test jobs fails with another GUID error run the query again for that GUID. Repeat the test process till no errors occur.