Re: [quicwg/base-drafts] Stream correlation in Transport (#672)

Mike Bishop <notifications@github.com> Wed, 28 June 2017 23:30 UTC

Date: Wed, 28 Jun 2017 23:30:00 +0000
From: Mike Bishop <notifications@github.com>
Reply-To: quicwg/base-drafts <reply+0166e4ab5d9b69149f202c8ae7979de2c27356951c28654592cf00000001156bfdf892a169ce0e43a87c@reply.github.com>
To: quicwg/base-drafts <base-drafts@noreply.github.com>
Cc: Subscribed <subscribed@noreply.github.com>
Message-ID: <quicwg/base-drafts/pull/672/c311820173@github.com>
In-Reply-To: <quicwg/base-drafts/pull/672@github.com>
References: <quicwg/base-drafts/pull/672@github.com>
Subject: Re: [quicwg/base-drafts] Stream correlation in Transport (#672)
Mime-Version: 1.0
Content-Type: multipart/alternative; boundary="--==_mimepart_59543bf84453d_56513ff182cbbc3c6661e"; charset="UTF-8"
Content-Transfer-Encoding: 7bit
Precedence: list
Archived-At: <https://mailarchive.ietf.org/arch/msg/quic-issues/Ls4625eX7Sk8P_wZw4b0y4dmcjE>

Good point.  Most things fall quickly into the bucket of “associating to these is not valid” – all unidirectional streams, response streams, and bidirectional streams that have seen a response are the same. But every pub/sub stream remains valid forever, as does a bidirectional stream that never gets answered.  There are some possible memory attacks there we’d need to iron out if we went this way.

From: MikkelFJ [mailto:notifications@github.com]
Sent: Wednesday, June 28, 2017 4:10 PM
To: quicwg/base-drafts <base-drafts@noreply.github.com>
Cc: Mike Bishop <Michael.Bishop@microsoft.com>; Author <author@noreply.github.com>
Subject: Re: [quicwg/base-drafts] Stream correlation in Transport (#672)

I have so far only looked at transport.md - I think the exact encoding is not too important for now - notably var length integers might replace bits in the type byte.

One thought is that it might not be desirable to check for association errors because the initial stream might be closed and gone before a response arrives. Checking validity would require state to be retained - in principle indefinitely. But the association type may still be useful to and API layer.

—
You are receiving this because you authored the thread.
Reply to this email directly, view it on GitHub<https://na01.safelinks.protection.outlook.com/?url=https%3A%2F%2Fgithub.com%2Fquicwg%2Fbase-drafts%2Fpull%2F672%23issuecomment-311817033&data=02%7C01%7CMichael.Bishop%40microsoft.com%7Cf33c31e5652d4d04e59b08d4be7ac60f%7C72f988bf86f141af91ab2d7cd011db47%7C1%7C0%7C636342881899061982&sdata=ZllkXdSDwGP7V8GLwBoLvFousQKYgRxOzb8G73zkJQo%3D&reserved=0>, or mute the thread<https://na01.safelinks.protection.outlook.com/?url=https%3A%2F%2Fgithub.com%2Fnotifications%2Funsubscribe-auth%2FAEE2haeX7CVj98vhhchOYqzJb1J7qq1tks5sIt05gaJpZM4OIm8f&data=02%7C01%7CMichael.Bishop%40microsoft.com%7Cf33c31e5652d4d04e59b08d4be7ac60f%7C72f988bf86f141af91ab2d7cd011db47%7C1%7C0%7C636342881899061982&sdata=b%2FJhB8vT5XBHIZjK6X8MXEV6X6STZ6e7cvcI40hVXcY%3D&reserved=0>.

-- 
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/672#issuecomment-311820173

[quicwg/base-drafts] Stream correlation in Transp… Mike Bishop
Re: [quicwg/base-drafts] Stream correlation in Tr… MikkelFJ
Re: [quicwg/base-drafts] Stream correlation in Tr… Mike Bishop
Re: [quicwg/base-drafts] Stream correlation in Tr… Martin Thomson
Re: [quicwg/base-drafts] Stream correlation in Tr… Mike Bishop
Re: [quicwg/base-drafts] Stream correlation in Tr… Mike Bishop