W3C home > Mailing lists > Public > ietf-http-wg-old@w3.org > September to December 1996

Re: Server response version number

From: Anselm Baird_Smith <abaird@www43.inria.fr>
Date: Mon, 21 Oct 1996 18:29:54 +0200 (MET DST)
Message-Id: <199610211629.SAA21030@www43.inria.fr>
To: "John C. Mallery" <jcma@ai.mit.edu>
Cc: Henrik Frystyk Nielsen <frystyk@w3.org>, http-wg%cuckoo.hpl.hp.com@hplb.hpl.hp.com, w3c-http@w3.org
John C. Mallery writes:
 > What happens when a 1.1 client tries to do a 1.1 put via a 1.0
 > proxy to a 1.1 server?

(Speaking again of my own implementation which conforms to Henrik
description). The 1.1 client detect that the proxy is 1.0 either:

a) Because it has used it in the past, and that 1.0 proxy has replied
   with a 1.0 reply (to the first 1.1 request) - most probable case.
b) Because at the time it issued the PUT, it will wait (for some
   timeout value), for a 100 status code that will never come.

So at least the client <-> proxy link is ok. Now, the 1.0 proxy speaks
to the 1.1 server, through a 1.0 request, so the server can deal with
it properly.

I don't have a deep analyzises to propose here, but my intuition is
that Henrik's suggestion (of replying with the version given by the
client), is the only one that will not hurt.

Anselm.
Received on Monday, 21 October 1996 09:38:15 EDT

This archive was generated by hypermail pre-2.1.9 : Wednesday, 24 September 2003 06:32:16 EDT