On Fri, 30 Mar 2018 06:40:57 +0100 Nick Kew wrote: > > On 29 Mar 2018, at 18:22, Eric Covener <covener@xxxxxxxxx> wrote: > > > > On Thu, Mar 29, 2018 at 11:11 AM, Andrew Beverley <andy@xxxxxxxxxxx> wrote: > >> Dear all, > >> > >> I am trying to debug a fastcgi server-side application problem, which I > >> believe was caused by the value of the CONTENT_LENGTH environment > >> variable not matching the size of the request body sent to the > >> application. > >> > >> In order to debug, I need to recreate the situation, which was caused > >> by an unknown client request. > > I haven’t verified whether this’ll do the job for you, but have you tried > using RequestHeader (mod_headers) to fake it? > > > Something like mod_deflate or mod_substitute changing the length> > > I’ve certainly encountered problems with filters invalidating headers. > And (in the distant past) fixed some of them. It’s entirely possible > there are still unfixed cases. Thanks for the replies and suggestions. After several more hours of debugging, it turns out that CONTENT_LENGTH was in fact correct. It was actually a bug in the web application's framework, which was attempting to re-read the request's body, after its file pointer was already at the end (and it therefore read no data the second time). Many thanks for the help, Andy --------------------------------------------------------------------- To unsubscribe, e-mail: users-unsubscribe@xxxxxxxxxxxxxxxx For additional commands, e-mail: users-help@xxxxxxxxxxxxxxxx
|