Nginx – How to host multiple apps on one server and point multiple domains to them

1. For email, I don't think you can do this with DNS. The only time DNS is involved with mail is for determining which server to send mail to for the requested domain. So whilst you can have the MX records for both domain1.com and domain1.net pointing to mail.domain1.com, say, you will need to configure that server to accept mail for both domains, and route it to the correct mailbox as appropriate. Exchange will do all of the above out of the box.

For web requests, (and this answers #3 too), it depends on whether you care about the host name that ends up in the address bar of the browser. If you don't then you could just CNAME www.domain1.com to www.domain1.net, and as long as you have both hostnames as vaild host headers for the site in IIS/Apache, the page will load and all will work. I wouldn't recommend this however as all it takes is one absolute href and the user is bounced to the other domain, not to mention issues with SSL and wider business branding problems, SEO, the list goes on.

To do that properly, use a redirect rule and bounce all traffic to one domain.

2. IIS redirect is the neatest way, keeps everything consistent. You could just add A records for both in DNS and add both as accepted host headers for the site in IIS though. Up to you.

3. As 1. but bear in mind that if you want to use SSL you will need a dedicated server IP for each site, so it might be simpler to redirect everything to one place and then secure that.

HTH

My http/https configuration using centos and nginx from the EPEL repo is as follows:

/etc/nginx/nginx.conf

user              nginx;
worker_processes  1;

error_log  /var/log/nginx/error.log;

pid        /var/run/nginx.pid;

events {
    worker_connections  1024;
}

http {
    include           /etc/nginx/mime.types;
    default_type      application/octet-stream;

    log_format  main  '$remote_addr - $remote_user [$time_local] "$request" '
                      '$status $body_bytes_sent "$http_referer" '
                      '"$http_user_agent" "$http_x_forwarded_for"';

    access_log         /var/log/nginx/access.log  main;
    sendfile           on;
    tcp_nopush         on;
    keepalive_timeout  65;
    gzip  on;

    server {
        listen          80;
        server_name     rproxy.yourdomain.co.uk;
        ## use when niginx listens on a nonstandard port - the port in the rewrite 
        ## rule needs to be the port nginx is actually listening on.
        #rewrite     ^(.*)   https://rproxy.yourdomain.co.uk:80$1 permanent; 
        access_log      /var/www/rproxy.yourdomain.co.uk/log/nginx.access.log;
        error_log       /var/www/rproxy.yourdomain.co.uk/log/nginx_error.log debug;

        location / {
            proxy_pass  http://192.168.xx.x:81;
            proxy_redirect          default;
        }

        error_page  404              /404.html;
        location = /404.html {
            root   /usr/share/nginx/html;
        }

        error_page   500 502 503 504  /50x.html;
        location = /50x.html {
            root   /usr/share/nginx/html;
        }

    }

    server {
        listen          443;
        server_name     rproxy.yourdomain.co.uk;
        ## use when niginx listens on a nonstandard port - the port in the rewrite 
        ## rule needs to be the port nginx is actually listening on.
        #rewrite     ^(.*)   https://rproxy.yourdomain.co.uk:443$1 permanent;
        ssl                             on;
        ssl_certificate                 /etc/pki/tls/certs/ca.crt;
        ssl_certificate_key             /etc/pki/tls/private/ca.key;
        ssl_session_timeout             5m;
        ssl_protocols                   SSLv2 SSLv3 TLSv1;
        ssl_ciphers                     ALL:!ADH:!EXPORT56:RC4+RSA:+HIGH:+MEDIUM:+LOW:+SSLv2:+EXP;
        ssl_prefer_server_ciphers       on;

        access_log      /var/www/rproxy.yourdomain.co.uk/log/nginx.access.log;
        error_log       /var/www/rproxy.yourdomain.co.uk/log/nginx_error.log debug;

        location / {
            proxy_pass  https://192.168.xx.x:444;  
            proxy_redirect          default;
        }

        error_page  404              /404.html;
        location = /404.html {
            root   /usr/share/nginx/html;
        }

        error_page   500 502 503 504  /50x.html;
        location = /50x.html {
            root   /usr/share/nginx/html;
        }

    }

    include /etc/nginx/conf.d/*.conf;
}

/etc/nginx/conf.d/proxy.conf

## $proxy_port setting string is needed when
## nginx is listening on a nonstandard port
#proxy_set_header        Host            $host:$proxy_port;
#proxy_set_header        X-Real-IP       $remote_addr:$proxy_port;
#proxy_set_header        X-Forwarded-For $proxy_add_x_forwarded_for:$proxy_port;
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;
client_header_buffer_size 64k;
proxy_connect_timeout   90;
proxy_send_timeout      90;
proxy_read_timeout      90;
proxy_buffer_size   16k;
proxy_buffers       32   16k;
proxy_busy_buffers_size 64k;

The link below gives the full details of my centos and nginx configuration:

http://www.zen212739.zen.co.uk/centos5-5/pub-centos-rproxy-nginx-install.txt

I only use one domain for webmail access, but adding another domain to this nginx config should not be too difficult.

Using nginx for an imap proxy is a little more difficult so set up. I am still looking into this myself.

Smtp traffic needs a store and forward mail server. A separate postfix setup would probably be a better option.

I set up a intel atom based centos xen server with a few vm's to essentially the same thing that you are attempting to do.