Transient VSS errors appear to be randomly and occur many hours into the backup.
For that reason you need to suspect the errors are due to a VSS shadow area overrun which can be easily prevented.
Try using the tool VSSUIRUN.EXE from the command prompt (see screenshots here:)
and set ‘no limit’ for all drives/volumes listed. This setting is an upper limit to how much temp space VSS may use.
The amount of temp space needed is the total of all write block change operations on that volume for the entire duration of a backup. Naturally this may vary greatly
and the system default is usually quite low at around 300MB; hence, it’s no surprise to see backups failing sporadically.
You can safely apply this tweak on all servers you have as a precaution.
If you want to confirm the VSS shadow space setting is indeed the cause, have a look at the Windows Event Viewer logs (app and sys) at around the time when the backup reported the VSS error.
You may find VSS and/or VolSnap errors, ‘disk warnings’, or any other kind of error ± 10 mins from the above time. In the case of disk warnings, these are actual
disk errors and usually precede a hard disk fatality
As always, if you are looking for server backup software solutions, use BackupChain 🙂