Re: "Reverse proxy" (was: Referrals)

["Chris G. Sellers" <chris.sellers@nitle.org>]

Just call it a proxy.   Reverse or forward, it's really doing the same  
thing, even if it's server to server proxying.  Does it really matter  
if the server sits between the browser and an internet gateway on a  
specific subnet or on the other side of the internet gateway?  To the  
browser, not much changes (if anything)

But - I digress from the real problem question here.  I've just seen  
too many people get confused by the 'reverse' terminology and trying  
to understand it as reverse (e.g. the server proxies back to the  
browser ?? ).



On Mar 17, 2008, at 1:26 PM, Michael Ströder wrote:
> Howard Chu wrote:
> > Michael Ströder wrote:
> > > Then use back-ldap as a LDAP reverse proxy or slapo-chain for  
> > > chasing the
> > > referrals automatically.
> > You keep using that phrase "reverse proxy" - it's just a proxy.
>
> Well, I don't want to nitpick here but this wording is commonly used  
> when dealing with web servers:
>
> 1. A forwarding proxy is one your client might have to use to reach  
> out to external servers (by some definition of "external") and
>
> 2. a reverse proxy is the connection end point for the client which  
> passes requests on to a hidden server.
>
> For 1. the client has to know about the proxy and for 2. the client  
> does not have to know (or even must not know).
>
> See also:
> http://en.wikipedia.org/wiki/Reverse_proxy
>
> Looking at this it seems appropriate to me to use the term "reverse  
> proxy" when talking about back-ldap.
>
> Ciao, Michael.
>
> P.S.: This reminds me of a feature request I wanted to submit to ITS  
> for letting libldap use a forwarding proxy via HTTP connect. Or at  
> least pass a file descriptor to ldap_initialize()...

______________________________________________
Chris G. Sellers			|	NITLE  - Technology Team
734.661.2318			|	chris.sellers@nitle.org
AIM: imthewherd			|	GoogleTalk: cgseller@gmail.com