The HTTP protocol handler is a significant source of bug reports and escalation. By its nature it is often difficult to diagnose the issue, involving network captures, server logs etc..
It would help a lot if there was more code instrumentation in the protocol handler. At the very least an easy way to log the HTTP traffic including headers and such.
Two things are needed:
- More logging (and formatting of logs) of HTTP events seen by the protocol handler
- A traffic capture tool that would allow a comparison with the events generated when necessary.