Hi, and by huge I mean Gigabytes! We have a very large site with tens of thousands of users and thousand+ courses. Our question database contains around 2-3000 unique questions and most are at system level as they are shared. All this works fine.
We use backup and restore to create a duplicate course for different schools for example.
However, when we run a backup (and select the option to backup questions) things recently started to run very slowly and finally we realized the questions table in the database was enormous.
What seems to happen is that all backups contain the entire database of questions and when restored these are duplicated so the number of questions effectively increases exponentially! We have a massive AWS server so have managed so far!!! However, we have recently had to use SQL to delete unused questions but this soon doubles up again.
Has anybody any ideas or come across this or anything similar before? We are on the fully patched 3.5 version.