Date: Thu, 23 May 2019 06:10:49 -0700
From: Udip Pant <notifications@github.com>
Reply-To: quicwg/base-drafts
 <reply+AFTOJK7VK5J6YUAUMNARFEV26PHFTEVBNHHBU6IQGY@reply.github.com>
To: quicwg/base-drafts <base-drafts@noreply.github.com>
Cc: Subscribed <subscribed@noreply.github.com>
Message-ID: <quicwg/base-drafts/issues/2699/495211676@github.com>
In-Reply-To: <quicwg/base-drafts/issues/2699@github.com>
References: <quicwg/base-drafts/issues/2699@github.com>
Subject: Re: [quicwg/base-drafts] HTTP/3 Client handling of SERVER_BUSY error
 code needs to be specified (#2699)
Mime-Version: 1.0
Content-Type: multipart/alternative;
 boundary="--==_mimepart_5ce69bd9903d5_1ef83fe221ccd95c17395b5";
 charset=UTF-8
Content-Transfer-Encoding: 7bit
Precedence: list
Archived-At: <https://mailarchive.ietf.org/arch/msg/quic-issues/0WZi_W7VMeM_vj9DTeC5K3JLCOA>


----==_mimepart_5ce69bd9903d5_1ef83fe221ccd95c17395b5
Content-Type: text/plain;
 charset=UTF-8
Content-Transfer-Encoding: 7bit

Based on the text (https://tools.ietf.org/html/draft-ietf-quic-transport-20#section-5.2.2), my interpretation is that the server can respond with SERVER_BUSY (only) when it receives Initial packet from the client (i.e. during/pre handshake). So any client with implementation of happyeyeball-like race mechanism between QUIC and TCP should fallback to TCP.

Post handshake, the server can either send GO_AWAY or stateless reset depending on the application protocol and circumstances, and client can try to re-establish QUIC connection.

Since this issue is not HTTP3 specific, I agree with @LPardue that HTTP3 is probably not the right place. (Also, what else could client do in this case anyway?). As @ianswett mentioned above, clarifying this in the quic-transport spec would be helpful.

-- 
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/2699#issuecomment-495211676
----==_mimepart_5ce69bd9903d5_1ef83fe221ccd95c17395b5
Content-Type: text/html;
 charset=UTF-8
Content-Transfer-Encoding: 7bit

<p>Based on the text (<a href="https://tools.ietf.org/html/draft-ietf-quic-transport-20#section-5.2.2" rel="nofollow">https://tools.ietf.org/html/draft-ietf-quic-transport-20#section-5.2.2</a>), my interpretation is that the server can respond with SERVER_BUSY (only) when it receives Initial packet from the client (i.e. during/pre handshake). So any client with implementation of happyeyeball-like race mechanism between QUIC and TCP should fallback to TCP.</p>
<p>Post handshake, the server can either send GO_AWAY or stateless reset depending on the application protocol and circumstances, and client can try to re-establish QUIC connection.</p>
<p>Since this issue is not HTTP3 specific, I agree with <a class="user-mention" data-hovercard-type="user" data-hovercard-url="/hovercards?user_id=6571445" data-octo-click="hovercard-link-click" data-octo-dimensions="link_type:self" href="https://github.com/LPardue">@LPardue</a> that HTTP3 is probably not the right place. (Also, what else could client do in this case anyway?). As <a class="user-mention" data-hovercard-type="user" data-hovercard-url="/hovercards?user_id=20072817" data-octo-click="hovercard-link-click" data-octo-dimensions="link_type:self" href="https://github.com/ianswett">@ianswett</a> mentioned above, clarifying this in the quic-transport spec would be helpful.</p>

<p style="font-size:small;-webkit-text-size-adjust:none;color:#666;">&mdash;<br />You are receiving this because you are subscribed to this thread.<br />Reply to this email directly, <a href="https://github.com/quicwg/base-drafts/issues/2699?email_source=notifications&amp;email_token=AFTOJK2D5N4WBMNNEHQG6JDPW2JVTA5CNFSM4HM56UX2YY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGODWCFJHA#issuecomment-495211676">view it on GitHub</a>, or <a href="https://github.com/notifications/unsubscribe-auth/AFTOJK5KFOPU4NWIJYQ3STLPW2JVTANCNFSM4HM56UXQ">mute the thread</a>.<img src="https://github.com/notifications/beacon/AFTOJK4CHRGBOLOJLC4BUZLPW2JVTA5CNFSM4HM56UX2YY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGODWCFJHA.gif" height="1" width="1" alt="" /></p>
<script type="application/ld+json">[
{
"@context": "http://schema.org",
"@type": "EmailMessage",
"potentialAction": {
"@type": "ViewAction",
"target": "https://github.com/quicwg/base-drafts/issues/2699?email_source=notifications\u0026email_token=AFTOJK2D5N4WBMNNEHQG6JDPW2JVTA5CNFSM4HM56UX2YY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGODWCFJHA#issuecomment-495211676",
"url": "https://github.com/quicwg/base-drafts/issues/2699?email_source=notifications\u0026email_token=AFTOJK2D5N4WBMNNEHQG6JDPW2JVTA5CNFSM4HM56UX2YY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGODWCFJHA#issuecomment-495211676",
"name": "View Issue"
},
"description": "View this Issue on GitHub",
"publisher": {
"@type": "Organization",
"name": "GitHub",
"url": "https://github.com"
}
}
]</script>
----==_mimepart_5ce69bd9903d5_1ef83fe221ccd95c17395b5--