Hi,
I’m new to the technical side of things and looking for advice on setting up a more granular backup strategy for our self-hosted Pressbooks instance at my institution.
We already perform regular full system and database backups, but those are more suited for disaster recovery than for restoring specific book content. What we need is a book-level backup solution that would allow us to easily and selectively roll back changes when content is accidentally modified or deleted.
We host active book projects, each with multiple authors and collaborators, so the risk of unintentional changes is real.
Here are my main questions:
- Is there a programmatic way to back up all books individually on a regular schedule (e.g., twice a week/once a day)? Can this be automated with scripts or cron jobs?
- Can we regularly export all books (Pressbooks XML format (or WXR?)) using the Pressbooks API or a CLI tool? Do I understand correctly that this XML will not contain media files (images, PDFs, etc.) as binary files, but rather references their URLs? If so, how could we backup those files as well?
- From a best practices perspective:
- Should we ask authors to manually export (or clone) their own books as a safeguard before making substantial changes or edits?
- Or should all backup/versioning be handled at the system admin level?
- Are there any plugins you’d recommend for version control, automated exports, or rollback capabilities?
Any tips or examples from your own workflows would be helpful. Thanks
Marie-Hélène Vézina
Libraries, Université de Montréal