nginx automated installer

[xanubi]

I have a problem, the system is ignoring it. Htaccess.
I have a configuration that generates the file with php. Js when it does not exist, however the. Htaccess is being ignored and the file is not generated.
How do I get the nginx read the error before the 404. Htaccess. ?

Like i said on your PM, if you let nignx proxy the request, apache will get that and read whatever you have (htaccess, etc.).

[n00bie]

How does Nginx compare to Apache? How would I benefit of using Nginx instead of Apache?

[SunShellNET]

It is working on my personal server where I host less than 10 domains. But not on a production server.

[Solokron]

Most sites we see that require a dedicated environment where we could stick nginx in are heavy PHP websites. Any idead on when you will have nginx using spawn-fcgi from the LightTPD project to get it serving up PHP? That is what I am interested in.

*Or even PHP-FPM?

[Solokron]

So I figured I would test this little installer out.

The below thread looked interesting and I was hoping it was integrated but immediately viewing Apache status results in

"Failed to receive status information from Apache.
Unable to connect to local httpd server."

I would like to give some sugestions, based on what i see after install and production.

======================
1) createvhosts.py
======================
Add more file types, like this:
location ~* \.(gif|jpg|jpeg|png|ico|wmv|avi|asf|asx|mpg|mpeg|mp4|pls|mp3|mid|wav|swf|flv|htm|html|txt|js|css|exe|zip|rar|gz|tgz|uha|7z|doc|docx|xls|xlsx|pdf)$ {

======================
2) STATUSURL="http://localhost/whm-server-status"
======================
On files:
/usr/local/apache/bin/apachectl
/etc/rc.d/init.d/httpd

Change the line:
STATUSURL="http://localhost/whm-server-status"
To:
STATUSURL="http://localhost:8081/whm-server-status"

======================
3) Service and Chkcfg
======================
Edit file /etc/chkserv.d/chkservd.conf and add nginx:1

Add a file named nginx to /etc/chkserv.d/ and inside put this:
service[nginx]=x,x,x,service nginx restart,nginx,root


:::::::::::::::::::::::::::::::::::::::::::::::::::::

That's it. With this settings, the "apache status" option of cpanel will work again, and the nginx service will be monitored and you can choose it from cpanel option "service manager". I hope this helps.

[bmbraga]

I'm having another problem, did the installation on 2 servers, one works perfectly fine. In the second I did, the settings are 100% equal, but the "apache status of failed.
What should I look?

[ChrisRHS]

I had an issue with the Apache status as well, on a fresh server. Once I created an account though, it worked as it should.

[Solokron]

I was hoping that would be the case as well but still no go. Time for me to hit the error logs.

I have it corrected and seems to be working great. I noticed the following files are included:

/etc/nginx/fastcgi.conf
/etc/nginx/fastcgi.conf.default
/etc/nginx/fastcgi_params
/etc/nginx/fastcgi_params.default

have you made any progress with fastcgi. I could integrate it manually but my concern is cPanel integration.

*Also, do you have SMP enabled in this compile?

I had an issue with the Apache status as well, on a fresh server. Once I created an account though, it worked as it should.

[n00bie]

how does nginx compare to apache? How would i benefit of using nginx instead of apache?

bump bump!

[kernow]

This worked great on a redhat server ( thanks ) but failed on a centos 64 bit:

Installation Complete -- run /etc/init.d/nginx restart to start nginx      
[root@host] publicnginx >> /etc/init.d/nginx restart   
Restarting nginx daemon: nginxcat: /var/run/nginx.pid: No such file or directory
kill: usage: kill [-s sigspec | -n signum | -sigspec] pid | jobspec ... or kill -l [sigspec]
 not running[warn]: conflicting server name "xx.xx.xx.x" on 0.0.0.0:80, ignored                        
[emerg]: could not build the server_names_hash, you should increase either server_names_hash_max_size: 2048 or server_names_hash_bucket_size: 128                                                                                 
 already running.

Any ideas?

[nsetiono]

What I did was add a /etc/nginx/proxy.conf file with:

proxy_redirect off;
proxy_set_header Host $host;
proxy_set_header X-Real-IP $remote_addr;
proxy_set_header X-Forwarded-For $proxy_add_x_forwarded_for;
client_max_body_size 10m;
client_body_buffer_size 128k;
proxy_connect_timeout 90;
proxy_send_timeout 90;
proxy_read_timeout 90;
proxy_buffer_size 4k;
proxy_buffers 4 32k;
proxy_busy_buffers_size 64k;
proxy_temp_file_write_size 64k;
proxy_redirect http://206.214.223.169:8081 http://206.214.223.169;
proxy_pass http://206.214.223.169:8081/;

And replace the above in /scripts/createvhosts.py with:

include /etc/nginx/proxy.conf

Then just run /scripts/createvhosts.py to regenerate. Seems to be working so far.

this one can be used if you only have 1 ip address but if more then 1 i think it's will not worked or i'm wrong ?

[Solokron]

I am running it on a CentOS 64-bit server with no errors. Your error looks to be a config issue, not a binary issue.

This worked great on a redhat server ( thanks ) but failed on a centos 64 bit:

Installation Complete -- run /etc/init.d/nginx restart to start nginx      
[root@host] publicnginx >> /etc/init.d/nginx restart   
Restarting nginx daemon: nginxcat: /var/run/nginx.pid: No such file or directory
kill: usage: kill [-s sigspec | -n signum | -sigspec] pid | jobspec ... or kill -l [sigspec]
 not running[warn]: conflicting server name "xx.xx.xx.x" on 0.0.0.0:80, ignored                        
[emerg]: could not build the server_names_hash, you should increase either server_names_hash_max_size: 2048 or server_names_hash_bucket_size: 128                                                                                 
 already running.

Any ideas?

[kernow]

Its the same script that worked OK on our redhat server, so it should work on the centos. No changes have been made to the script.

[Solokron]

ok, I am going to save some of you a lot of headaches when it comes to rewrites with nginx. I spent all day trying out rewrites and trying to figure out why they were not working. Even checked with a heavy nginx user and it seemed to threw him off as well for a bit. So here you go, if you plan on doing any .htm or .html rewrites, remove the damn file types in the location ~* \. area as nginx will completely ignore any rewrite rules if they are immediately proxied in that area.

[kernow]

Just to clarify

...remove the damn file types in the location ~* \.

Are you saying to remove files types from every clients root directory if their using rewrite rules ?