Pat Riehecky wrote:
> This is a bit of a odd duck, but....
>
> The university I work for has a bunch of library pages that can only be
> accessed from on campus as they are hosted off site and authenticated by
> IP address.  
This sounds like a perfect scenario for an acceleration setup.  You can 
dispense with having users set proxy in their browser and only require 
authentication for off-site access.
In short, the Squid box acts like the origin server (using a domain 
within your control: http://offsite.library.iwu.edu/ or some such).  
ACLs are set up such that access from within your campus network is 
allowed through the acceleration setup without authentication, access 
from outside is allowed WITH authentication, and all other access is 
denied (forcing those who are using your proxy for all internet traffic 
to repent, and helping prevent abuse of the system).  Allowed requests 
are relayed by your Squid server to the remote library site, and the 
content is ultimately served by your Squid server (as it is now).
>
> Wow, you got all the way down here... dang....
>
> I will accept vaguely half formed, partially coherent theories just to
> keep my own mental gears turning.  Anything at all you could contribute
> would be tremendously helpful (this includes, the proposed task is
> impossible proofs as well, but sadly I would need a strong argument to
> hand up the chain as they look at me funny when I say this doesn't sound
> possible).
>   
Half-formed, partially coherent, I can handle.  Fleshing this setup out 
is left as an exercise for the reader.  The FAQ sections on accelerators 
(http://wiki.squid-cache.org/SquidFaq/ReverseProxy) and ACLs 
(http://wiki.squid-cache.org/SquidFaq/SquidAcl) should help a lot... 
Questions regarding further clarification of this framework are welcome.
> Pat
>   
Chris
Received on Tue May 15 2007 - 15:56:08 MDT
This archive was generated by hypermail pre-2.1.9 : Fri Jun 01 2007 - 12:00:05 MDT