I've been using nginx on our rails site for a few weeks now and I started seeing some crashes related to running a rails daemon.

I upgraded to nginx/0.8.6, built from source, in the hopes of stopping the crashes. Now whenever I access the root domain I receive a 403 forbidden error instead of the request being passed off to our thin servers. I can access anything relative to root w/o problems, i.e. for example mydomain.com fails with 403 but mydomain.com/home is passed off to the rails application just fine.

The only thing I changed in the config file was proxy_redirect false; to proxy_redirect off;

Any clues as to where or how to tackle this issue?

The error log shows: directory index of "/vol/www/mydomain.com/public/" is forbidden,

Obviously I don't want it to do a directory index but rather pass the request to our thin server.

I've verified permissions. nginx worker processes run as www-data which is the owner of the directory structure. nginx master runs as root.

I switched back to nginx version: nginx/0.6.31 installed via apt-get install. Things are working fine now using the same nginx.conf and vhost conf files.

the nginx/0.8.6 causing the problem was built using the following configure specification:

./configure \
--sbin-path=/usr/sbin \
--conf-path=/etc/nginx/nginx.conf \
--error-log-path=/var/log/nginx/error.log \
--pid-path=/var/run/nginx.pid \
--lock-path=/var/lock/nginx.lock \
--http-log-path=/var/log/nginx/access.log \
--http-client-body-temp-path=/var/lib/nginx/body \
--http-proxy-temp-path=/var/lib/nginx/proxy \
--http-fastcgi-temp-path=/var/lib/nginx/fastcgi \
--without-mail_pop3_module \
--without-mail_smtp_module \
--without-mail_imap_module \
--without-http_ssi_module \
--without-http_userid_module \
--without-http_auth_basic_module \
--without-http_autoindex_module \
--without-http_browser_module

I'm trying to find out how the ubuntu hardy nginx package was built to see if perhaps my configure specification was the source of the problem.

Do you have an index file explicitly stated in the config file?

index [i]your-index-filename[/i];

--
Jim Ohlstein

Hello Jim,

I have tried it with an index index.html and w/o that directive. There is no index.html in the root folder. Currently the statement is in the vhost file.

I've reinstalled nginx/0.6.31 using apt-get install and things are back to working fine w/o any need to change the nginx config or the vhost. I've attached both btw.

I rebuilt version nginx-0.8.6 using the following ./configure specification which I believed was used to build the ubuntu nginx package:
[quote="./configure --conf-path=/etc/nginx/nginx.conf \\
--error-log-path=/var/log/nginx/error.log \\
--pid-path=/var/run/nginx.pid \\
--lock-path=/var/lock/nginx.lock \\
--http-log-path=/var/log/nginx/access.log \\
--http-client-body-temp-path=/var/lib/nginx/body \\
--http-proxy-temp-path=/var/lib/nginx/proxy \\
--http-fastcgi-temp-path=/var/lib/nginx/fastcgi \\
--with-debug --with-http_stub_status_module \\
--add-module=../nginx_upload_module-2.0.9 \\
--with-http_flv_module --with-http_ssl_module --with-http_dav_module \\
--with-http_realip_module --with-http_xslt_module \\
--with-sha1=/usr/include/openssl"]

[/quote]

Things seemed to work fine for a bit with the new version then suddenly I started to received the 403 error.

The environment I am testing this under is an EC2 instance of Ubuntu hardy.

I have encountered the same error about 6 months ago and still don't know the ways how to solve this until now. I just ignored it and tried another one. It makes my head aches with this problem :(

"I can accept failure, everyone fails at something. But I can't accept not trying."
Editor @ [url=http://www.daily-reviews.com]Daily Reviews[/url]