I am experiencing something odd. It is also intermittent. I am running the Hallengren Backup solution with Delete before backup configuration. I recently started getting a failures. The failure code is 3013 which may be "The process cannot access the file because it is being used by another process." amongst other things, with no other comments being recorded to the command log table. the Previous Backup job has been deleted, when I log on to see what is happening. So I am possibly thinking that I need some space\time between the delete and the new backup because the space isn't getting reclaimed fast enough to a lot for the new back up file. When I run it manually it runs fine and all of the disk space is available.
My thought is to separate out the Delete step from the stored proc, and make it its own set in the job.
Before I did that I figured I'd as if anyone has had this issue before.
Are you sure it's not due to some other process, such as a tape backup or even AVS? Any time we've had this issue, it's always been a tape backup process. We would work with the tape backup team to schedule their process to occur so that it does not collide with the full backup delete (the one that you really need to make sure gets deleted for storage reasons).
That was the first we did was check all of the other processes. SQL 2005
We have a couple other databases that get back up too, these are smaller. when all were in the same job it failed on this DB also. this is the largest database. it is just under 400GB
There is nothing else happening at the same time. All I can think of is the space isn't all the way reclaimed before it starts.
I would bet quite a bit of money that it's another process causing it. You just haven't caught the process yet.
At my last job, we had a database over 15TB in size and used Ola's backup script. No issue with the delete, besides what I mentioned regarding the tape backup collision.