Re: [quicwg/base-drafts] Simplify TLP and RTO into Probe Timeout (#2114)

ianswett <> Wed, 12 December 2018 10:35 UTC

Return-Path: <>
Received: from localhost (localhost []) by (Postfix) with ESMTP id 4F04312F1AC for <>; Wed, 12 Dec 2018 02:35:39 -0800 (PST)
X-Virus-Scanned: amavisd-new at
X-Spam-Flag: NO
X-Spam-Score: -3.056
X-Spam-Status: No, score=-3.056 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIMWL_WL_HIGH=-1.46, 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_NONE=-0.0001, SPF_PASS=-0.001] autolearn=ham autolearn_force=no
Authentication-Results: (amavisd-new); dkim=pass (1024-bit key)
Received: from ([]) by localhost ( []) (amavisd-new, port 10024) with ESMTP id VwJr3xLkSzXt for <>; Wed, 12 Dec 2018 02:35:37 -0800 (PST)
Received: from ( []) (using TLSv1.2 with cipher ECDHE-RSA-AES128-GCM-SHA256 (128/128 bits)) (No client certificate requested) by (Postfix) with ESMTPS id 1FA2A12F1A6 for <>; Wed, 12 Dec 2018 02:35:37 -0800 (PST)
DKIM-Signature: v=1; a=rsa-sha1; c=relaxed/relaxed;; h=from:reply-to:to:cc:in-reply-to:references:subject:mime-version:content-type:content-transfer-encoding:list-id:list-archive:list-post:list-unsubscribe; s=s20150108; bh=ubXIstxS07RNIzYS8vCf76kVv0Y=; b=BWSEAtchzxjbShSR qZgrAuQTYfBRlxpy4zKm7cuIPftH60+hL4JU1ti9aKBRgIvC/AHG2ERJHcv8IdxU Vj5CEyq/jnegdd91WTC4srE0Z8q8oRLshYYWQ6VhAp+ohEFmpAO19BnyGwIhQCW5 zDV/uCteLhDzwBwnd3id6hYLG1Y=
Received: by with SMTP id filter0652p1las1-27362-5C10E478-1 2018-12-12 10:35:36.07896467 +0000 UTC m=+24434.793960267
Received: from (unknown []) by (SG) with ESMTP id QBGSwOVkQgiAU5xqTvydGw for <>; Wed, 12 Dec 2018 10:35:35.974 +0000 (UTC)
Received: from (localhost []) by (Postfix) with ESMTP id ED0A12411EF for <>; Wed, 12 Dec 2018 02:35:35 -0800 (PST)
Date: Wed, 12 Dec 2018 10:35:36 +0000
From: ianswett <>
Reply-To: quicwg/base-drafts <>
To: quicwg/base-drafts <>
Cc: Subscribed <>
Message-ID: <quicwg/base-drafts/pull/2114/review/>
In-Reply-To: <quicwg/base-drafts/pull/>
References: <quicwg/base-drafts/pull/>
Subject: Re: [quicwg/base-drafts] Simplify TLP and RTO into Probe Timeout (#2114)
Mime-Version: 1.0
Content-Type: multipart/alternative; boundary="--==_mimepart_5c10e477eba18_484b3f92096d45b828556d"; 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-SG-EID: l64QuQ2uJCcEyUykJbxN122A6QRmEpucztpreh3Pak3I1/ADK52qG0/ONbRRoRPe/toQm4rnRbT8W1 yb3kOYKQ/Fp96f++2lPUh14MT/kXyDPUZLjQICu8voBx8lv11G5g5NR7zTE4CKIY2XBVyndgI7CCZz jrTNofKYbJTSxaOYn03N+ezAYU2ZSgTZ3iuCfF3EHETvk/adIaVEZq2mJg==
Archived-At: <>
X-Mailman-Version: 2.1.29
List-Id: Notification list for GitHub issues related to the QUIC WG <>
List-Unsubscribe: <>, <>
List-Archive: <>
List-Post: <>
List-Help: <>
List-Subscribe: <>, <>
X-List-Received-Date: Wed, 12 Dec 2018 10:35:39 -0000

ianswett commented on this pull request.

-## Retransmission Timeout
+A PTO expiration is classified as spurious or valid when an ACK frame is
+received that newly acknowledges packets in flight, see {{pto-loss}}.  On a
+valid PTO, the congestion window MUST be reduced to the minimum congestion
+window and slow start is re-entered.

If you have a lot of tails(ie: HTTP traffic), then it doesn't take too many packet losses to end up hitting this.  TLP was targeted partially at cases like that, and I think you'll find changing it will produce measurable performance regressions in web traffic.  We can do something else in a follow-up, but as I said before, I'd prefer text saying SHOULD consider an acknowledgement of the first PTO as standard packet loss, not extreme congestion, and treat it as such.

You are receiving this because you are subscribed to this thread.
Reply to this email directly or view it on GitHub: