Re: [quicwg/base-drafts] CID's should be compared in constant time (#2477)

Kazuho Oku <notifications@github.com> Sat, 16 February 2019 03:19 UTC

Date: Fri, 15 Feb 2019 19:19:51 -0800
From: Kazuho Oku <notifications@github.com>
Reply-To: quicwg/base-drafts <reply+0166e4ab3a334a24284bcbd8dbfa1be470991fd1d7371d6792cf00000001187f435792a169ce187e182d@reply.github.com>
To: quicwg/base-drafts <base-drafts@noreply.github.com>
Cc: Subscribed <subscribed@noreply.github.com>
Message-ID: <quicwg/base-drafts/issues/2477/464278972@github.com>
In-Reply-To: <quicwg/base-drafts/issues/2477@github.com>
References: <quicwg/base-drafts/issues/2477@github.com>
Subject: Re: [quicwg/base-drafts] CID's should be compared in constant time (#2477)
Mime-Version: 1.0
Content-Type: multipart/alternative; boundary="--==_mimepart_5c678157c76f4_192a3faeed8d45c419614"; charset="UTF-8"
Content-Transfer-Encoding: 7bit
Precedence: list
Archived-At: <https://mailarchive.ietf.org/arch/msg/quic-issues/ELRSQqsU5_ePClgyyP0ot7I3MPc>

I think @mikkelfj is technically correct. A simple attack would be like the following.

Assume there's a server using an ordered list to maintain CIDs. When it receives a packet, it scans the list in ascending order until it finds a matching CID (or if reaches the end of the list without finding one).

An attacker can bisect the server to see which CIDs is active, by observing the difference in the latency of receiving stateless resets. This is much convenient than a brute-force attack that tries every possible CID to see if one is active.

If we need to be clear about the constant time requirement is a different question.

-- 
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/2477#issuecomment-464278972

[quicwg/base-drafts] CID's should be compared in … Mike Bishop
Re: [quicwg/base-drafts] CID's should be compared… Mike Bishop
Re: [quicwg/base-drafts] CID's should be compared… MikkelFJ
Re: [quicwg/base-drafts] CID's should be compared… Kazuho Oku
Re: [quicwg/base-drafts] CID's should be compared… MikkelFJ
Re: [quicwg/base-drafts] CID's should be compared… Dmitri Tikhonov
Re: [quicwg/base-drafts] CID's should be compared… Martin Thomson
Re: [quicwg/base-drafts] CID's should be compared… ianswett
Re: [quicwg/base-drafts] CID's should be compared… Jana Iyengar
Re: [quicwg/base-drafts] CID's should be compared… MikkelFJ
Re: [quicwg/base-drafts] CID's should be compared… Martin Thomson
Re: [quicwg/base-drafts] CID's should be compared… Martin Thomson