I think in this case it's the backend server having a moment of difficulty. We're investigating it right now.