Anyone can help me about this?
-------- Messaggio Originale --------
Oggetto: Possible bug with set-cookie
Data: Thu, 04 Oct 2007 15:24:57 +0200
Da: Matteo Niccoli <matteo.niccoli(a)staff.dada.net>
I have this script:
If I execute it, with:
If I switch from 1799 to 1800, nginx stop serving this page.
It appens only with nginx. Apache is works good maybe by deleting cookie or
simple overwriting another.
How can solve this?
Testing a generated list of maps I noticed that nginx handles the
keys in a case insensitiv way, reporting conflicts for
Is this intended?
Janko Hauser email: jhauser(a)zscout.de
mobile: +49 1721 641552
-----BEGIN PGP SIGNED MESSAGE-----
Is it possible to send logs to some program instead of file?
(http://wiki.codemongers.com/NginxHttpLogModule#access_log) show only
one way (file).
Now we use 'access_log /dev/stdout' in configuration and run nginx via
'nginx | program' - but it is not elegant.
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.7 (GNU/Linux)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org
-----END PGP SIGNATURE-----
Is it possible to have a different 404 or 500 error page be shown based
on the request path? For example, any failed request that started with
"/admin" would get an error page that matched the layout of the admin
layout while any other failed request would get the standard 500 error
Posted via http://www.ruby-forum.com/.
Changes with nginx 0.6.14 15 Oct 2007
*) Change: now by default the "echo" SSI command uses entity encoding.
*) Feature: the "encoding" parameter in the "echo" SSI command.
*) Feature: the "access_log" directive may be used inside the
*) Bugfix: if all upstream servers were failed, then all servers had
got weight the was equal one until servers became alive; bug
appeared in 0.6.6.
*) Bugfix: a segmentation fault occurred in worker process if
$date_local and $date_gmt were used outside the
*) Bugfix: a segmentation fault might occur in worker process if debug
log was enabled.
Thanks to Andrei Nigmatulin.
*) Bugfix: ngx_http_memcached_module did not set
Thanks to Maxim Dounin.
*) Bugfix: a worker process may got caught in an endless loop, if the
memcached was used.
*) Bugfix: nginx supported low case only "close" and "keep-alive"
values in the "Connection" request header line; bug appeared in
*) Bugfix: sub_filter did not work with empty substitution.
*) Bugfix: in sub_filter parsing.
I hope everyone is having a nice weekend.
Can anyone out there tell me whether it would be possible to access
the remote user name and password ($user and $password in
http://$user:$firstname.lastname@example.org/) so that I can roll my own basic
It seems to me that remote_user is only available when the Auth Basic
module is enabled, but am I missing something?
In my mod_wsgi implementation (function wsgi_build_environment) I need
to copy each of the HTTP header and wsgi params in Python objects.
In the current implementation I just allocate a buffer, copy the content
in it, and then create a Python String object.
I'm assuming that using an allocator pool, allocating small buffers is
cheap, but I would like a confirmation.
An alternative implementation is to allocate only one buffer, using the
max lenght required, and reusing it for every Python object.
Thanks Manlio Perillo
I'm pleased to annunce the release 0.0.3 of mod_wsgi.
mod_wsgi is an implementation of the WSGI PEP
for the Nginx HTTP server.
The nginx mod_wsgi Mercurial repository is available at:
The tagged releases can be found at:
A snapshot of the current tip can be downloaded at:
I will try to commit only stable code, so you can "safely" use the tip.
In this release I have added full support to the WSGI 2.0 draft (that
is: WSGI 1.0 without the start_response and the write callable).
With this release WSGI applications can be executed concurrently by
Nginx, when they return a generator (see the nginx-file in the examples
In the next release I will add support for WSGI 1.0 (unfortunately the
write callable cannot be easily implemented).
Instructions for building the wsgi module can be found in the README file.
Regards Manlio Perillo
I'm using nginx to proxy for both ruby on rails and php applications. Some of my users are reporting intermittant connection reset errors in their browsers. The behavior they describe consists of the browser fully loading the web page then immediately forwarding them to a connection reset page. I have seen it first hand a couple of times but have not been able to consistently reproduce it. I checked the access log for the request and it never gets logged by nginx. Has anyone else seen this kind of behavior?
My guess is its some kind of bug with 302 handling.