Scheduled backup failing


I have a scheduled backup thats been running on the same centos7 box for a couple of years.

It backups up to ftp space provided by the data centre.

Its a full backup, and old backups are set to be deleted after 4 days. There is enough space for this.

Just in the last couple of weeks its started failing because old backups arent being deleted and allocation gets full.
The failed process doesnt complete and ends up hanging around growling.
I have to kill the process, ftp into to storage and manually delete old backups.

As far as the provider is concerned nothing has changed re the ftp space.

Does anybody have any ideas why virtualmin is failing to delete the old backups?