ISSUE
A Unitrends backup failure job log reports the following error:
Read error 0 in file <file path>: file backup corrupt.
RESOLUTION
The solution is to configure VSS to write the shadow copies of the problem drive to a different drive which has more available space and less load during the backup. The ideal situation would be to have a separate drive dedicated to shadow copies that is not being backed up and is also not the drive containing the paging file. If this is not possible, specifying a separate drive that contains the paging file is typically still better than storing the shadow copies on the drive being backed up.
The instructions in the related articles below are operating system specific and show how to configure the shadow copy storage location of a volume to go to a different volume.
CAUSE
In short, "Read error 0 in file..." is the result of the VSS shadow being removed in the middle of the backup. To confirm this, you may get the "Start Date:" and "End Date:" from the backup. Then view the SYSTEM event logs in your Windows system. Look for volsnap events in that time span.
Possible events to be seen are:
Event ID: 25
Source: volsnap
The shadow copies of volume <volume> were deleted because the shadow copy storage could not grow in time. Consider reducing the I\O load on the system or choose a shadow copy storage volume that is not being shadow copied.
Microsoft Volume Shadow Copy Service (VSS) is part of the Windows operating system. When Unitrends is used to backup a volume or when a backup is created for a VSS aware application, Unitrends calls VSS at the beginning of the job to take a snapshot of the volume. This snapshot is also known as a shadow copy, and is supposed to persist for the duration of the backup of the volume. However, the above errors indicate that the I/O load on the drive being backed up was too high. Because of this high load on the drive, VSS encountered a failure and deleted the shadow copy snapshot from which Unitrends was backing up prior to the end of the job. The corrupt data message is shown in the Unitrends job logs as it tries to read open files through the shadow copy that was destroyed.
By default VSS is configured to write the shadow copies of a volume to a storage location on the volume itself. This can result in very high I/O during the backup of a volume since the drive is experiencing load from the backup reading files, load from VSS reading from and writing to the shadow copy, and load from any other applications on the system using that same volume.
Event ID: 36
Source: volsnap
The shadow copies of volume <volume> were aborted because the shadow copy storage could not grow due to a user imposed limit.
To increase your limit, you will need to go to the Windows Explorer "Configure Shadow Copies" screen. There you can configure the settings for each volume to increase the space allocated for VSS for that volume. Here, if necessary, you can also change the volume that will house the shadows for the volume in question.
NOTES
For additional information regarding this and other related volsnap errors, refer to https://support.microsoft.com/en-us/kb/826936.
General VSS troubleshooting
https://helpdesk.kaseya.com/hc/en-gb/articles/4407521158545
How to confirm shadowstorage is setup correctly
https://helpdesk.kaseya.com/hc/en-gb/articles/4407510344977