Reverse proxy

Reverse Proxy Support

Most people probably have OSPi exposed only on their internal network at something like http://xxx.xxx.xxx.xxx/ where xxx.xxx.xxx.xxx is an internal ip address issued by your home router. A reverse proxy is a web server that sits between a service on an internal network (such as OSPi) and the open Internet. It allows you to host OSPi at a sub-path of a fully qualified domain name, or FQDN, such as https://mydomain.com/some/path/to/OSPi.

Why a Reverse Proxy?

There are several simpler alternatives to a reverse proxy if you would like to expose your OSPi to the Internet. Alternatives include hosting at the root domain (e.g. https://mydomain.com ), at a subdomain (e.g. https://ospi.mydomain.com) or on another port (e.g. https://mydomain.com:8080). A reverse proxy is useful if you want meet the following 3 requirements.

• Host multiple web-based services from a single FQDN. (http://mydomain.com/ospi and https://mydomain.com/otherservice)
• Use standard web traffic ports for all services (80,443)
• Use a single SSL certificate for all services

The last point is really the key. Most of the simpler options mention above would require you to purchase separate SSL certificates from a Central Authority (CA) such Verisign, Network Solutions, etc for each service you wish to host securely. The exception is hosting on another port. You could use a single certificate for multiple ports, but you would need to configure your router to forward each new port to the correct internal IP.

You could also use self-signed certificates but you would receive browser security warnings and trying to use the mobile app would most likely not work.

Regardless of how you expose OSPi to the internet IF you do it is HIGHLY RECOMMENDED that you access OSPi over an secure SSL connection.

Example setup

There are many different web servers which can provide reverse proxy support. Two popular, and free, ones are http://nginx.com/ and http://httpd.apache.org/. Below is a sample NGINX configuration. Setting up NGINX and configuring it to forward traffic to your router is out of the scope of this documentation but can be found at http://nginx.org/en/docs/.

NGINX Example

    # This section permantly redirects all unsecure web traffic (HTTP to HTTPS)
    server {
      listen 80;
      server_name mydomain.com;
      return 301 https://mydomain.com$request_uri;
    }
    
    #This section sets up the ssl connection and the routing from the internet to OSPi
    server {
      listen			443 ssl;
      client_max_body_size    300M;
      server_name		mydomain.com;
      ssl_certificate		"/Volumes/hdd 1/docs/ssl_cert/mycert.crt"; #replace mycert with actual filename
      ssl_certificate_key	"/Volumes/hdd 1/docs/ssl_cert/mykey.key"; #replace mykey with actual filename

      #charset koi8-r;

      access_log  /usr/local/var/log/nginx/access.log upstreamlog;
   
      # sets up the sub-path to ospi (i.e. https://mydomain.com/ospi/)  by setting the location '/ospi/'
      # removing SSL (since OSPi isn't configured for it), and
      # adds information to the request that OPSi used to construct the correct links
      # in the OSPi application (i.e. https://mydomain/ospi/page.html instead of http://xxx.xxx.xxx.xxx:8080/page.html)
      location /ospi/ {
        proxy_pass http://xxx.xxx.xxx.xxx:8080/; #internal network location of your OSPi
        proxy_set_header Host $host;
        proxy_set_header X-Forwarded-For $proxy_add_x_forwarded_for;
        proxy_set_header X-Forwarded-Proto $scheme;
        proxy_set_header X-Scheme $scheme;
        proxy_set_header X-Script-Name /ospi;
      }
    }