Group: Forum Members
If scheduled Diffs work fine, then it's not a problem with backups that occur under the scheduled user context as opposed to backups that occur under your user context, the latter of which is used when you run a job manually from the Backup Definition Files tab. There wouldn't be a configuration difference that would allow scheduled Diffs to succeed but scheduled Fulls to fail like that. However, a single blip in network connectivity can cause a Reflect job to fail (although usually with a more descriptive error), and powerline networking certainly isn't bulletproof -- and statistically you're more likely to encounter a blip during a Full than a Diff simply because a Full is a larger job. As a test, try moving either your WD network drive or your PC so that both devices are connected solely via Ethernet switch gear rather than powerline networking and see if the scheduled Full succeeds. You can even trigger it manually by going to the Scheduled Backups tab, right-clicking the Full job, and selecting Run Now. If it works, then powerline networking is your culprit. I've got a Wish List thread requesting more graceful handling of temporary network disruptions, but to date it has not been implemented.
Also, since you mention that the drives are "mapped", does that mean your destination path uses a mapped drive letter rather than a UNC path? If you want backups to run on a schedule, you should really use UNC paths. It doesn't seem to be a requirement in your case since your scheduled Diffs apparently work, but getting network drive mappings that you create in your user account to work in a scheduled backup job typically involves some extra work that can be avoided by using UNC paths instead.