> Jef Poskanzer wrote:
> >
> > >I like the HTTP_VIA field sent to our WWW server from the cache.
> > >Unfortunately it doesn't seem to add a whole lot of information, at least not
> > >to someone as new to Web cacheing as me. The HTTP_VIA tells me the same name
> > >as the REMOTE_HOST field (and the address of the cache can be found in
> > >REMOTE_ADDR).
> >
> > They become different in the case of multiple hops - the Via field
> > is supposed to have an entry from each proxy along the way, like
> > a netnews Path: header.
> > ---
> > Jef
>
> Does the VIA: header also contain the address of the ultimate client?
> Like the "FORWARDED BY ... FOR ..." header in HTTP/1.0 does.
>
> We need the ultimate client address to do client authentication.
If I understand the draft http 1.1 specification correctly, the answer is no.
That's why I'd like REMOTE_HOST (or a X_REMOTE_HOST if it isn't possible to
use REMOTE_HOST) to include the client address, not the address of the last
cache.
cf. draft httpd 1.1 at http://www.w3.org/pub/WWW/Protocols/, section 14.44.
Yours,
Lars-Owe
Received on Fri Nov 29 1996 - 01:27:23 MST
This archive was generated by hypermail pre-2.1.9 : Tue Dec 09 2003 - 16:33:40 MST