Re: [TLS] Session Lifetime

Eric Rescorla <> Mon, 29 November 2010 18:44 UTC

Return-Path: <>
Received: from localhost (localhost []) by (Postfix) with ESMTP id 977173A6C26 for <>; Mon, 29 Nov 2010 10:44:12 -0800 (PST)
X-Virus-Scanned: amavisd-new at
X-Spam-Flag: NO
X-Spam-Score: -101.472
X-Spam-Status: No, score=-101.472 tagged_above=-999 required=5 tests=[AWL=-0.096, BAYES_00=-2.599, FM_FORGED_GMAIL=0.622, HTML_MESSAGE=0.001, J_CHICKENPOX_81=0.6, USER_IN_WHITELIST=-100]
Received: from ([]) by localhost ( []) (amavisd-new, port 10024) with ESMTP id xp9iqREFfCsn for <>; Mon, 29 Nov 2010 10:44:11 -0800 (PST)
Received: from ( []) by (Postfix) with ESMTP id 0A7453A6BDB for <>; Mon, 29 Nov 2010 10:44:10 -0800 (PST)
Received: by gyb13 with SMTP id 13so2539472gyb.31 for <>; Mon, 29 Nov 2010 10:45:20 -0800 (PST)
MIME-Version: 1.0
Received: by with SMTP id d24mr9466668agj.160.1291056320669; Mon, 29 Nov 2010 10:45:20 -0800 (PST)
Received: by with HTTP; Mon, 29 Nov 2010 10:45:20 -0800 (PST)
In-Reply-To: <20101129175911.GA19368@sescenties>
References: <> <> <> <20101129175911.GA19368@sescenties>
Date: Mon, 29 Nov 2010 10:45:20 -0800
Message-ID: <>
From: Eric Rescorla <>
To: Seth David Schoen <>
Content-Type: multipart/alternative; boundary="001485f9a68c60e7280496357993"
Subject: Re: [TLS] Session Lifetime
X-Mailman-Version: 2.1.9
Precedence: list
List-Id: "This is the mailing list for the Transport Layer Security working group of the IETF." <>
List-Unsubscribe: <>, <>
List-Archive: <>
List-Post: <>
List-Help: <>
List-Subscribe: <>, <>
X-List-Received-Date: Mon, 29 Nov 2010 18:44:12 -0000

On Mon, Nov 29, 2010 at 9:59 AM, Seth David Schoen <> wrote:

> Eric Rescorla writes:
> > The upper limit is just a suggestion anyway.
> >
> > I'm unaware of any reason why weeks to months isn't acceptab;e.
> One concern is that resumed sessions work like cookies to identify
> users, but users may not have convenient ways to control them as
> they can control cookies.  There has been recent concern about
> browser functionality that is equivalent to HTTP cookies but not
> subject to user control like cookies.
> Session resumption can have this effect too.
> Of course, this concern is attenuated for sites where users have
> to log in and have only one account, but it could be meaningful on
> HTTPS sites where users have no account or have multiple accounts.

So, that's certainly true (and of course CSSC offers an even more powerful
in this respect), but I don't generally feel that lifetime is that useful a
way to control this.
Rather, the same mechanisms that blow away all other browser state (e.g.,
browsing mode,, clear cookies,etc.) should also blow away session state.


A separate performance consideration is that ideally the client
> and server should have matched expectations about how long to
> expect sessions to persist -- whether that's "not at all" or "a
> year".

It's not clear to me that that's that useful. If the client just keeps
session state
as long as is convenient, then the server can extract close-to-optimal
with no additional protocol effort.