I've noticed that if you run out of storage your current backupjob will be corrupt and the backupjob fails with:
Rolling back changes made by incomplete job run
Error: Root element is missing.
This has happened to me twice since upgrading to 9.5 and the same error both times. Before upgrading this wasn't a problem, if you run out of space of course the job will fail but run the next time if there where space available.
I've opened a case with the support. Got the expected answer - Create a new backup set.
I tried to explain that I see this as bug introduced in 9.5, but got no understanding in return.
Is this know? Will it be fixed or this is how it will work now?
The strangest in this story is when I today was going to copy the current backupfiles to an other storage and create a new backup set. Now I notice that the job has run during the night, and all old backup files are completely gone?! Only one full backup file left in the folder!
This last thing I haven't been able to recreate.