Thanks @Hodge. I do suspect boost is the problem. The pump leak was found while troubleshooting the boost, and is probably unrelated like you say.
I will look into getting both sorted...