So I’m working on a Web service for a client. I have a suite of unit tests that work properly, but when the client tries to access the Web service, the request fails. Before I made some changes to the service (including upgrading some libraries that shouldn’t matter), both my tests and the customer’s client worked as expected. So I’m left with the problem of figuring out how my tests differ in their operation from the customer’s client application. The best way I know of to figure this out is to log the full incoming requests and see how they differ, but I don’t know how to do that. Do you have any ideas? (The web service runs on Tomcat 5 and Apache 2.)