With this one I have not yet tried using it in a restore. The full backup is quite large and the test environment is not ready for a change [testing is in progress on other development]
In the past with a different database I did successfully , multiple times, use a full backup and differential to restore a database from a backup that had supposedly failed.
It leads me to distrust the "verify" and the backup messages.
But I don't trust assuming that a "failed" backup actually will be good when it is really needed.
I am interested to see if there are known bugs in the statements used to verify and identify backup files.