This is exactly how I interview, and it is the most effective way I know to separate the best from the rest. To pass it the candidate must be able to recall in great detail a past project's problem, investigation, and solution. That's hard to fake from memory if you don't have direct involvement and understanding.

I'll also bounce around and go into depth on particular technologies or protocols, etc which may be relevant as well as other projects' problems and solutions. I also sprinkle in my current interests whatever they may be and have them make assessments, not to see if they agree but to see how they go about analyzing and weighing different factors.