Doing QA on the Python 3 registry instance, I found that there is a failure when attempting to convert a Short Client Token to an Adobe ID. Specifically, when using a revamped version of the adobe-id-for-short-client-token script in the circulation repo with a valid drm:clientToken value, the QA registry responded with:

<error xmlns="http://ns.adobe.com/adept" data="E_NYPL_AUTH 'bytes' object has no attribute 'xpath'"/>

This looks very much like a py2 to py3 issue with how bytestrings/unicode handling happens. I don't think it will be too hard to trace, but obviously we need some additional test cases on the registry for the Adobe endpoints.