Back to work.

I found that one of the problems that I was having with the current project is that a change had been made to the serialization format of one of the objects in C++, but it hadn't been replicated in the Java test code. I sent an e-mail to the rest of the team suggesting that we need to start making these changes in both code bases in the future. :)

Otherwise, I think everything else that's going wrong is a matter of messed up configurations on the test machines. I think.