On Fri, Feb 05, 2010 at 03:04:07PM -0500, JohnSchulze wrote:

> Maxim Dounin Wrote:
> -------------------------------------------------------
> > Could you please provide nginx -V output, full
> > config and debug
> > log for faulty request(s)? See here for
> > instructions how to
> > obtain debug log:
> >
> > http://nginx.org/en/docs/debugging_log.html
> >
> > Note well: using try_files and not using dynamic
> > hostname
> > resolution as suggested in previous reply is
> > anyway preffered.
> > But config as shown in your post should work
> > without problems, and
> > it would be good to investigate problem roots.
> >
> > Maxim Dounin
>
> Many thanks for your suggestions, I have changed to use the try_files method instead (is this a new directive?). Unfortunately the behaviour is exactly the same. I've run off a debug log for a page request but it's 99kb(!) so I can't really post it all here. In it I can see there are double requests for some files and for some of these I can't see the 200. Here is a part of the debug log where you can see two requests for "main.css" (there should only be one request) and it looks to me as if only one of them is finished with proper response headers:

> 2010/02/05 19:42:53 38652#171748: *4 http filename: "c:\nginx/../rails/MySite/public/main.css"
> 2010/02/05 19:42:53 38652#171748: *4 add cleanup: 007C1ECC
> 2010/02/05 19:42:55 160580#180468: master WaitForMultipleObjects: 4
> 2010/02/05 19:42:55 160580#180468: reap worker
> 2010/02/05 19:42:55 160580#180468: worker process 38652 exited with code C0000005

The worker exited abnormaly. I do not know how to troubleshoot such bugs
on Windows without coredumps. Probably you may look the DrWatson log file
for stack dump to estimate the point of failure.


--
Igor Sysoev
http://sysoev.ru/en/

_______________________________________________
nginx mailing list
nginx@nginx.org
http://nginx.org/mailman/listinfo/nginx