W3C home > Mailing lists > Public > ietf-http-wg-old@w3.org > September to December 1995

Re: keepalives and proxies: a request and a proposal

From: Ari Luotonen <luotonen@netscape.com>
Date: Fri, 17 Nov 1995 12:16:59 -0800 (PST)
Message-Id: <199511172016.MAA19628@netscape.com>
To: Lorenzo Vicisano <vicisano@iet.unipi.it>
Cc: http-wg mailing list <http-wg%cuckoo.hpl.hp.com@hplb.hpl.hp.com>

I'm all for addition of Proxy-Connection: keep-alive to the spec; more
specifically:

 - if there are no proxies involved, everything's normal
 - if the client talks to a proxy, it *never* sends a Connection: keep-alive
 - if the client talks to a proxy and wants a persistent connection,
   it will send Proxy-Connection: keep-alive; the proxy can still
   respond with Connection: keep-alive
 - a proxy can now independently support Connection: keep-alive
   between itself and the servers, and itself and the client
 - the proxy will intercept a Connection: keep-alive received from a
   remote, as it is a reaction to the Connection: keep-alive sent by
   itself, and should not be forwarded to the client
 - a proxy will generate a new Connection: keep-alive to the client as
   a response to Proxy-Connection: keep-alive

This way, there is never a case where a client talking to an old proxy
server would fool it to forward the Connection: keep-alive to the
remote, and then the remote would respond with it, but the proxy would
close the connection in between.

Cheers,
--
Ari Luotonen				ari@netscape.com
Netscape Communications Corp.		http://home.netscape.com/people/ari/
501 East Middlefield Road
Mountain View, CA 94043, USA		Netscape Server Development Team
Received on Friday, 17 November 1995 12:20:03 EST

This archive was generated by hypermail pre-2.1.9 : Wednesday, 24 September 2003 06:31:36 EDT