I am using the latest pull of branch 3.3, and t_branch_timeout() doesn't work if the previous branch resulted in a reply (i.e. a negative final reply).
That is to say, t_branch_timeout() == FALSE on a genuine timeout, as long as the preceding branch provided a result (such as 404).
t_branch_timeout() == TRUE if this is a timeout, on the first branch.
In other words, it appears that if t_any_replied() == TRUE, then t_branch_timeout() == FALSE, even for this branch, and even if the branch really did time out (no response at all from destination).
That's not how the documentation says it should behave. Am I doing something wrong? If so, how do I find out if _this particular_ branch timed out, from the failure_route[]?