It may not be apparent based on my reaction to this particular issue, but I'm a strong advocate for DNN and its hybrid profit/nonprofit open source model, that has evolved over that last decade+. However, I generally can't bring my self to fix what some else broke at my own expense.

BUT... I'm willing to consider it if someone can provide info on whats wrong with the code. If you know definitively that it worked prior to 7.1 surely someone also has a pretty good guess what the problem is. Who was closest to the change?

 I haven't had a chance to look into this, but the issue is likely to be based on the change of password format. Prior to 7.1.0 the default format was encryption, from 7.1.0 on the default is hashing. The main difference here is that hashing is a 1 way operation, so passwords cannot be determined (meaning any GetPassword(user) logic fails)- this meant that some pieces of functionality had to be changed e.g. the old change password function didn't work as it requires the user to enter the "old" and "new" passwords, but as the system could not read the "old" password, this did not work. We worked around this with the password reset logic. I suspect Q&A may be something similar i.e. somewhere in there it may contain some GetPassword(user) logic, but have not had time to look into this. It may even be that Q&A cannot work in a hashed environment and that we have to add code to determine if hashing is used and disable this function, but until I get some time (or someone else gets time) I cant say for sure.