We have one SBS 2008 riding on a cluster that has started to fail its full backups but only at certain times.
The KB indicates that the problem is resident on the source if the error falls on a read or on the destination if on a write.
In this case our failure was on a write so we started to focus in on the destination.
For this cluster setup we have the backups stream across the wire to the standalone DC on an HP MicroServer that was also protected by ShadowProtect.
We looked into network connectivity as well as for disk I/O errors in the Event Logs with no results.
The last place to look was in the ShadowProtect setup on the DC itself.
Sure enough, the DC was set to run an incremental close to the same time the one backup on the SBS VM was failing.
We changed the standalone DC backup schedule to run one incremental at night to avoid any further conflicts with the VM backups that were streaming to it.
We now had a successful backup set on the SBS VM.
Co-Author: SBS 2008 Blueprint Book
Chef de partie in the SMBKitchen
Find out more at