Thanks @rpfernandezjr . This actually is a feature that was added to the PyLTI13 project . We plan on updating to this version in the next major release and could probably make the it combine both the LTI_CONFIG and database settings.

Ahh thanks, I remember this was a conversation in the past, with you guys and Etienne. I didn't see an issue for it so wasn't sure if it fell through the cracks.

I was looking at a way to configure this in the database #1318 but got into a little blocker figuring out how to make it compatible both ways easily.

Have you tried to disable the deployment id checks? There's a configuration for that now.

"LTI_CONFIG_DISABLE_DEPLOYMENT_ID_VALIDATION": true,

@jennlove-um Follow up with Unizin to see if they have set the configuration per Matt's question above. If this is not a current issue for them, then remove from this release. Explore other options to address this issue.

We had a PR here to use the Django Admin UI from the library, but the problem with that is you need a way to launch the tool in the first place. So it either needs a way to populate the database directly or to read and use the config values to be used.

One idea we've had as a middle ground between disabled checks and full checks that we've seen some tools do is to have you define a single deployment id, and once the tool is launched from that it's marked as being allowed from any deployment id on that configuration. This might be a good idea to implement in our local code, or even better use when/if we switch to the https://github.com/academic-innovation/django-lti library.