Date: Wed, 12 Dec 2018 10:35:53 -0800
From: Igor Lubashev <notifications@github.com>
Reply-To: quicwg/base-drafts
 <reply+0166e4abc0732bfb6b94c27d0ebde09851e0814dc86cf55692cf000000011829170992a169ce1731b88c@reply.github.com>
To: quicwg/base-drafts <base-drafts@noreply.github.com>
Cc: Subscribed <subscribed@noreply.github.com>
Message-ID: <quicwg/base-drafts/pull/2109/c446695165@github.com>
In-Reply-To: <quicwg/base-drafts/pull/2109@github.com>
References: <quicwg/base-drafts/pull/2109@github.com>
Subject: Re: [quicwg/base-drafts] Endpoints MAY (not SHOULD) ignore IPv4 ICMP
 PTB messages (#2109)
Mime-Version: 1.0
Content-Type: multipart/alternative;
 boundary="--==_mimepart_5c11550933ba2_74643fd6242d45bc143748";
 charset=UTF-8
Content-Transfer-Encoding: 7bit
Precedence: list
Archived-At: <https://mailarchive.ietf.org/arch/msg/quic-issues/do4yrfVcCZiJDdp7oklbzBMLqP8>


----==_mimepart_5c11550933ba2_74643fd6242d45bc143748
Content-Type: text/plain;
 charset=UTF-8
Content-Transfer-Encoding: 7bit

DPLPMTUD is not a requirement for all QUIC implementations.  The SHOULD in question is written as a requirement for all.

Looking at DPLPMTUD, however, I am realizing that it is possible that the question is moot.  DPLPMTUD draft (#3.4) considers doing anything at all (like simply matching to an existing connection using IP addresses) sufficient for validation (it says that doing more for validation is a SHOULD, implying that just matching IPs is still acceptable).  Since doing anything less is impractical when processing PTB messages, any implementation that uses PTB messages is implicitly validating them.  If so, I'd be happy with some text making it explicit that just matching on IPs and, maybe, UDP ports is sufficient to consider PTB validated.

-- 
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/2109#issuecomment-446695165
----==_mimepart_5c11550933ba2_74643fd6242d45bc143748
Content-Type: text/html;
 charset=UTF-8
Content-Transfer-Encoding: quoted-printable

<p>DPLPMTUD is not a requirement for all QUIC implementations.  The SHOUL=
D in question is written as a requirement for all.</p>
<p>Looking at DPLPMTUD, however, I am realizing that it is possible that =
the question is moot.  DPLPMTUD draft (<a class=3D"issue-link js-issue-li=
nk" data-error-text=3D"Failed to load issue title" data-id=3D"189968216" =
data-permission-text=3D"Issue title is private" data-url=3D"https://githu=
b.com/quicwg/base-drafts/issues/3" data-hovercard-type=3D"issue" data-hov=
ercard-url=3D"/quicwg/base-drafts/issues/3/hovercard" href=3D"https://git=
hub.com/quicwg/base-drafts/issues/3">#3</a>.4) considers doing anything a=
t all (like simply matching to an existing connection using IP addresses)=
 sufficient for validation (it says that doing more for validation is a S=
HOULD, implying that just matching IPs is still acceptable).  Since doing=
 anything less is impractical when processing PTB messages, any implement=
ation that uses PTB messages is implicitly validating them.  If so, I'd b=
e happy with some text making it explicit that just matching on IPs and, =
maybe, UDP ports is sufficient to consider PTB validated.</p>

<p style=3D"font-size:small;-webkit-text-size-adjust:none;color:#666;">&m=
dash;<br />You are receiving this because you are subscribed to this thre=
ad.<br />Reply to this email directly, <a href=3D"https://github.com/quic=
wg/base-drafts/pull/2109#issuecomment-446695165">view it on GitHub</a>, o=
r <a href=3D"https://github.com/notifications/unsubscribe-auth/AWbkq1C8Ms=
Q6AivO-NjINtMf7dT9CV9Oks5u4UyJgaJpZM4ZKoYb">mute the thread</a>.<img src=3D=
"https://github.com/notifications/beacon/AWbkq0387B2PeIeRzoD1FCfhwKt-cS44=
ks5u4UyJgaJpZM4ZKoYb.gif" height=3D"1" width=3D"1" alt=3D"" /></p>
<script type=3D"application/json" data-scope=3D"inboxmarkup">{"api_versio=
n":"1.0","publisher":{"api_key":"05dde50f1d1a384dd78767c55493e4bb","name"=
:"GitHub"},"entity":{"external_key":"github/quicwg/base-drafts","title":"=
quicwg/base-drafts","subtitle":"GitHub repository","main_image_url":"http=
s://assets-cdn.github.com/images/email/message_cards/header.png","avatar_=
image_url":"https://assets-cdn.github.com/images/email/message_cards/avat=
ar.png","action":{"name":"Open in GitHub","url":"https://github.com/quicw=
g/base-drafts"}},"updates":{"snippets":[{"icon":"PERSON","message":"@igor=
lord in #2109: DPLPMTUD is not a requirement for all QUIC implementations=
.  The SHOULD in question is written as a requirement for all.\r\n\r\nLoo=
king at DPLPMTUD, however, I am realizing that it is possible that the qu=
estion is moot.  DPLPMTUD draft (#3.4) considers doing anything at all (l=
ike simply matching to an existing connection using IP addresses) suffici=
ent for validation (it says that doing more for validation is a SHOULD, i=
mplying that just matching IPs is still acceptable).  Since doing anythin=
g less is impractical when processing PTB messages, any implementation th=
at uses PTB messages is implicitly validating them.  If so, I'd be happy =
with some text making it explicit that just matching on IPs and, maybe, U=
DP ports is sufficient to consider PTB validated."}],"action":{"name":"Vi=
ew Pull Request","url":"https://github.com/quicwg/base-drafts/pull/2109#i=
ssuecomment-446695165"}}}</script>
<script type=3D"application/ld+json">[
{
"@context": "http://schema.org",
"@type": "EmailMessage",
"potentialAction": {
"@type": "ViewAction",
"target": "https://github.com/quicwg/base-drafts/pull/2109#issuecomment-4=
46695165",
"url": "https://github.com/quicwg/base-drafts/pull/2109#issuecomment-4466=
95165",
"name": "View Pull Request"
},
"description": "View this Pull Request on GitHub",
"publisher": {
"@type": "Organization",
"name": "GitHub",
"url": "https://github.com"
}
}
]</script>=

----==_mimepart_5c11550933ba2_74643fd6242d45bc143748--