It was an issue at our backend. It was fixed at 13:50 UTC today.