Re: [quicwg/base-drafts] Compatible version upgrade (#1901)

Martin Thomson <notifications@github.com> Thu, 25 October 2018 01:27 UTC

Return-Path: <noreply@github.com>
X-Original-To: quic-issues@ietfa.amsl.com
Delivered-To: quic-issues@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 3674C130934 for <quic-issues@ietfa.amsl.com>; Wed, 24 Oct 2018 18:27:51 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -8.47
X-Spam-Level:
X-Spam-Status: No, score=-8.47 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIMWL_WL_HIGH=-0.47, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, HTML_MESSAGE=0.001, MAILING_LIST_MULTI=-1, RCVD_IN_DNSWL_HI=-5, SPF_PASS=-0.001] autolearn=ham autolearn_force=no
Authentication-Results: ietfa.amsl.com (amavisd-new); dkim=pass (1024-bit key) header.d=github.com
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id v4IdbIsCtwT9 for <quic-issues@ietfa.amsl.com>; Wed, 24 Oct 2018 18:27:49 -0700 (PDT)
Received: from out-5.smtp.github.com (out-5.smtp.github.com [192.30.252.196]) (using TLSv1.2 with cipher AECDH-AES256-SHA (256/256 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 4A5011292AD for <quic-issues@ietf.org>; Wed, 24 Oct 2018 18:27:49 -0700 (PDT)
Date: Wed, 24 Oct 2018 18:27:48 -0700
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=github.com; s=pf2014; t=1540430868; bh=gddsW8eeG1mZFOQz9OlN55PpDCr4VqGu4gxu4Il6rf0=; h=Date:From:Reply-To:To:Cc:In-Reply-To:References:Subject:List-ID: List-Archive:List-Post:List-Unsubscribe:From; b=zE2RygwXHuAuTdEPI0ZpqnMTMw2klaJrto2iIU11KDj9a/YmZ+1s8X0NyA1Eog1Jr Pogwmh21c5pwp+UUFAYTjAkiZhHO0TbD9lT9fLPhNIZgswYzxkRTqR2TbAtrjzfC9F nkft/0p/Hwo6eUz3mvE1V8ESciNd1hQvXUZmDbgI=
From: Martin Thomson <notifications@github.com>
Reply-To: quicwg/base-drafts <reply+0166e4ab8c08ffffb3b97ef344537e80ac3389a9460d229792cf0000000117e8de1492a169ce1640b1a8@reply.github.com>
To: quicwg/base-drafts <base-drafts@noreply.github.com>
Cc: Subscribed <subscribed@noreply.github.com>
Message-ID: <quicwg/base-drafts/pull/1901/review/168152869@github.com>
In-Reply-To: <quicwg/base-drafts/pull/1901@github.com>
References: <quicwg/base-drafts/pull/1901@github.com>
Subject: Re: [quicwg/base-drafts] Compatible version upgrade (#1901)
Mime-Version: 1.0
Content-Type: multipart/alternative; boundary="--==_mimepart_5bd11c147c8a6_5b9b3ff8782d45b4256422"; charset="UTF-8"
Content-Transfer-Encoding: 7bit
Precedence: list
X-GitHub-Sender: martinthomson
X-GitHub-Recipient: quic-issues
X-GitHub-Reason: subscribed
X-Auto-Response-Suppress: All
X-GitHub-Recipient-Address: quic-issues@ietf.org
Archived-At: <https://mailarchive.ietf.org/arch/msg/quic-issues/sjeov3H9a-npG5qMRk1e2LM_SOM>
X-BeenThere: quic-issues@ietf.org
X-Mailman-Version: 2.1.29
List-Id: Notification list for GitHub issues related to the QUIC WG <quic-issues.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/quic-issues>, <mailto:quic-issues-request@ietf.org?subject=unsubscribe>
List-Archive: <https://mailarchive.ietf.org/arch/browse/quic-issues/>
List-Post: <mailto:quic-issues@ietf.org>
List-Help: <mailto:quic-issues-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/quic-issues>, <mailto:quic-issues-request@ietf.org?subject=subscribe>
X-List-Received-Date: Thu, 25 Oct 2018 01:27:51 -0000

martinthomson commented on this pull request.



> +list of QUIC versions and select an alternative version from the list of
+supported versions.  A server MAY choose a compatible version from that list and
+continue the handshake with that version.  The server sends its first packet as
+though it was continuing with the version it selects.
+
+A QUIC version is compatible with this version if the cryptographic handshake
+message sent in the first packet can be used in both versions.  A compatible
+version is also able to identifying and acknowledge the first packet sent by the
+client in some fashion.  Other QUIC versions might have different constraints in
+determining what is compatible.  In order to facilitate this process, new QUIC
+versions could define a process for transforming the first packet from other
+compatible versions into the equivalent packet in the new version.
+
+Upgrading in this manner allows a server to upgrade without incurring the round
+trip imposed by sending a Version Negotiation packet.  It also allows clients to
+send their first packet using a widely deployed version, without the risk of

Yes, the incentive remains the same - the client offers what it believes to be most widely deployed.  This design only allows for a seamless upgrade to a compatible version.  If the client were to advertise the new, less-widely-deployed version, it risks getting a Version Negotiation packet.

> +A QUIC version is compatible with this version if the cryptographic handshake
+message sent in the first packet can be used in both versions.  A compatible
+version is also able to identifying and acknowledge the first packet sent by the
+client in some fashion.  Other QUIC versions might have different constraints in
+determining what is compatible.  In order to facilitate this process, new QUIC
+versions could define a process for transforming the first packet from other
+compatible versions into the equivalent packet in the new version.
+
+Upgrading in this manner allows a server to upgrade without incurring the round
+trip imposed by sending a Version Negotiation packet.  It also allows clients to
+send their first packet using a widely deployed version, without the risk of
+having to use that version with servers that supports a more-preferred version.
+
+A server MUST NOT send a Version Negotiation packet if it prefers a version that
+is not compatible with the version the client initially chose; a server has to
+allow the client to choose between versions that are not compatible.

No.  Because the server might not support every compatible version.  If the client offers A, where B is compatible, the server can still send Version Negotiation in response with B because it doesn't understand A.



> +A QUIC version is compatible with this version if the cryptographic handshake
+message sent in the first packet can be used in both versions.  A compatible
+version is also able to identifying and acknowledge the first packet sent by the
+client in some fashion.  Other QUIC versions might have different constraints in
+determining what is compatible.  In order to facilitate this process, new QUIC
+versions could define a process for transforming the first packet from other
+compatible versions into the equivalent packet in the new version.
+
+Upgrading in this manner allows a server to upgrade without incurring the round
+trip imposed by sending a Version Negotiation packet.  It also allows clients to
+send their first packet using a widely deployed version, without the risk of
+having to use that version with servers that supports a more-preferred version.
+
+A server MUST NOT send a Version Negotiation packet if it prefers a version that
+is not compatible with the version the client initially chose; a server has to
+allow the client to choose between versions that are not compatible.
 

If you were to implement each version in a different process, then this would not work, yes.  Unless you wanted to complicate the hand-off from a simple `exec quic-vY --disable-vn` to something more like `exec quic-vY --handoff=<translated packet+necessary keys>`.  The consequence is that you probably want to have compatible versions handled by the same instance.

>  
-The first client packet of the cryptographic handshake protocol MUST fit within
-a 1232 octet QUIC packet payload.  This includes overheads that reduce the space
-available to the cryptographic handshake protocol.
+2. The first message MUST include the transport parameters chosen by the client.
+   This enables upgrade to a compatible version (see {{version-upgrade}}).
+
+3. The first message MUST fit within a 1232 octet QUIC packet payload.  This

Yeah, and I think that we agreed that the 1232 thing was wrong.  I'll cite 4.3 of the -tls doc instead.

-- 
You are receiving this because you are subscribed to this thread.
Reply to this email directly or view it on GitHub:
https://github.com/quicwg/base-drafts/pull/1901#discussion_r227986859