Re: [quicwg/base-drafts] Ignoring ACK delay can result in wrong RTT calculations (#3350)

ianswett <notifications@github.com> Wed, 15 January 2020 20:14 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 74A551208EA for <quic-issues@ietfa.amsl.com>; Wed, 15 Jan 2020 12:14:41 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -6.596
X-Spam-Level:
X-Spam-Status: No, score=-6.596 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIMWL_WL_HIGH=-0.001, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, HTML_IMAGE_ONLY_28=1.404, HTML_MESSAGE=0.001, MAILING_LIST_MULTI=-1, RCVD_IN_DNSWL_HI=-5, SPF_HELO_NONE=0.001, 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 t7_cQ2s58gSm for <quic-issues@ietfa.amsl.com>; Wed, 15 Jan 2020 12:14:37 -0800 (PST)
Received: from out-20.smtp.github.com (out-20.smtp.github.com [192.30.252.203]) (using TLSv1.2 with cipher AECDH-AES256-SHA (256/256 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 97FA8120972 for <quic-issues@ietf.org>; Wed, 15 Jan 2020 12:14:37 -0800 (PST)
Received: from github-lowworker-fb56993.ac4-iad.github.net (github-lowworker-fb56993.ac4-iad.github.net [10.52.19.31]) by smtp.github.com (Postfix) with ESMTP id D31AA8C06CE for <quic-issues@ietf.org>; Wed, 15 Jan 2020 12:14:35 -0800 (PST)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=github.com; s=pf2014; t=1579119275; bh=5Tu1rY+8kTgiLG+ljssy85JmFq/TFghQw4Me9a8Uetw=; h=Date:From:Reply-To:To:Cc:In-Reply-To:References:Subject:List-ID: List-Archive:List-Post:List-Unsubscribe:From; b=lN7ffGxKZ3UpX65zowDRPhr24krPKIVZ9MEltaSlGxN0fqUHtTRhV2aweK2YyTwNS oVRLnSLxyIQIFzAMoMo2Bk7o3ykmUSuJdRHr9cB7ksMDQDC29PgErQvANZQW3PsTpv aGU7Egfh4Vugy5iINpg/FAuTs136hhEDQzFvoK5w=
Date: Wed, 15 Jan 2020 12:14:35 -0800
From: ianswett <notifications@github.com>
Reply-To: quicwg/base-drafts <reply+AFTOJK4CSK3CF47WNNUB7G54FSSSXEVBNHHCBTTN5I@reply.github.com>
To: quicwg/base-drafts <base-drafts@noreply.github.com>
Cc: Subscribed <subscribed@noreply.github.com>
Message-ID: <quicwg/base-drafts/issues/3350/574836935@github.com>
In-Reply-To: <quicwg/base-drafts/issues/3350@github.com>
References: <quicwg/base-drafts/issues/3350@github.com>
Subject: Re: [quicwg/base-drafts] Ignoring ACK delay can result in wrong RTT calculations (#3350)
Mime-Version: 1.0
Content-Type: multipart/alternative; boundary="--==_mimepart_5e1f72abc1c48_5c8f3f823fecd96015212a"; charset=UTF-8
Content-Transfer-Encoding: 7bit
Precedence: list
X-GitHub-Sender: ianswett
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/5Iht-ubPAA0fc-YU7txFMtMGDLQ>
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: Wed, 15 Jan 2020 20:14:41 -0000

I'll note that persistent delayed acks are bad for performance for a variety of reasons, but let's put that aside for now.

I believe the problematic text is:

> MUST NOT apply the adjustment if the resulting RTT sample is smaller than the min_rtt. This limits the underestimation that a misreporting peer can cause to the smoothed_rtt.

The alternative would be to allow the RTT sample to be as small as min_rtt, but no smaller, rather than ignoring the ack_delay entirely when it created a sample less than min_rtt.  The negative of this is it allows a buggy/malicious peer to always specify an overly large ack_delay, which drives smoothed to min_rtt and results in an RTTVar of 0.  This would make PTO quite aggressive, though given we're also adding max_ack_delay onto the PTO, possibly it's acceptable.

In real-world scenarios, I suspect it doesn't matter that much, which is why we went with the more conservative option.

-- 
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/issues/3350#issuecomment-574836935