Web Servers General Talk - Multiple TCP/HTTP servers with only one IP : not as simple

This is Interesting: Free IT Magazines  
Home > Archive > Web Servers General Talk > April 2005 > Multiple TCP/HTTP servers with only one IP : not as simple





You are viewing an archived Text-only version of the thread. To view this thread in it's original format and/or if you want to reply to this thread please [click here]

Author Multiple TCP/HTTP servers with only one IP : not as simple
alt250@gmail.com

2005-04-23, 7:47 am

I have one Windows application which is a kind of real-time TCP and
HTTP server.
In order for the client software to be able to connect to a server on
the Internet without problems, the server must listen on ports HTTP
(80) and HTTPS (443).

I want to host multiple software servers on the same computer which as
only one public IP address.
So, for example, if i have 2 software servers on this computer they
will need to bind at the same time on ports 80 and 443 of the same
public IP.
AFAIK this is imposible.

I am currently looking for solutions in order to solve this problem.

One solution i imagine :

Note : The client software can communicate with the server both in TCP
and HTTP, so i can use only HTTP in this case.

Make the software servers listen on internal IPs or on another port on
the same computer.
For example, server1 will listen on 10.1.0.1 and server2 on 10.1.0.2.

This way it is possible to use a reverse proxy which will listen on
ports 80 and 443 of the public IP and redirect to the good internal
server depending on the client request URL.
For example, if the client send a GET http://server1.com/Index.html
this request will be interpreted by the reverse proxy which will
redirect to server1 on 10.1.10.1.

I tried with different reverse proxies (Apache, Squid, AT32 and a few
others), and this always failed.

This failed because when the reverse proxy receive a request, says for
http://server1.com/Bigfile.zip, it redirects the request to the
internal
server1, acting itself as the HTTP client. Then it waits to receive
BigFile.zip completly before transfering it back to the real client.
I suppose this mechanism is used in order to cache the web page for
next client requests.

The problem is my application use HTTP to stream data. For the reverse
proxy this is like big web pages which will never be complete. So this
didn't work.

One solution to this problem will be to use a reverse proxy which does
not wait for the internal page to be fully transfered before sending it

back to the client. But currently i didn't found one !

They are problably other solutions too, i am open to your suggestions.

Thank you in advance

Sponsored Links






Free braindumps | Software forum | Database administration forum

Copyright 2003 - 2008 webservertalk.com