Message-ID: <20231012215122.GA8245@openwall.com>
Date:   Thu, 12 Oct 2023 23:51:22 +0200
From:   Solar Designer <solar@...nwall.com>
To:     Willy Tarreau <w@....eu>
Cc:     Vegard Nossum <vegard.nossum@...cle.com>,
        linux-doc@...r.kernel.org, linux-kernel@...r.kernel.org,
        security@...nel.org, corbet@....net, workflows@...r.kernel.org,
        Greg Kroah-Hartman <gregkh@...uxfoundation.org>,
        Kees Cook <keescook@...omium.org>
Subject: Re: [RFC PATCH] Documentation: security-bugs.rst: linux-distros relaxed their rules

Hi all,

Thank you (especially Willy) for your effort on this.

Out of the 3 paragraphs, the first one looks good to me as-is, but for
the last two I propose the slightly edited versions below.

On Sat, Oct 07, 2023 at 04:04:54PM +0200, Willy Tarreau wrote:
> +Please note that the respective policies and rules are different since
> +the 3 lists pursue different goals.  Coordinating between the kernel
> +security team and other teams is difficult since occasional embargoes
> +start from the availability of a fix for the kernel security team, while
> +for other lists they generally start from the initial post to the list,
> +regardless of the availability of a fix.

---
Please note that the respective policies and rules are different since
the 3 lists pursue different goals.  Coordinating between the kernel
security team and other teams is difficult since for the kernel security
team occasional embargoes (as subject to a maximum allowed number of
days) start from the availability of a fix, while for "linux-distros"
they start from the initial post to the list regardless of the
availability of a fix.
---

I added the part in braces to explain why the difference in when
embargoes start matters.  I also moved part of that sentence for
consistency.  Finally, I replaced "other lists" with specific reference
to "linux-distros" because this paragraph talks only about 3 specific
lists and on "oss-security" there are no embargoes.

On Sat, Oct 07, 2023 at 06:39:36PM +0200, Willy Tarreau wrote:
> On Sat, Oct 07, 2023 at 06:30:11PM +0200, Vegard Nossum wrote:
> > On 07/10/2023 16:04, Willy Tarreau wrote:
> > > +As such, the kernel security team strongly recommends that reporters of
> > > +potential security issues DO NOT contact the "linux-distros" mailing
> > > +list BEFORE a fix is accepted by the affected code's maintainers and you
> > 
> > is s/BEFORE/UNTIL/ clearer?
> 
> Probably, yes.

I agree.  Also, the sentence jumps from "reporters" to "you" implying
that "you" is a reporter, but maybe it's better to make that explicit.

> > > +have read the linux-distros wiki page above and you fully understand the
> > > +requirements that doing so will impose on you and the kernel community.
> > > +This also means that in general it doesn't make sense to Cc: both lists
> > > +at once, except for coordination if a fix remains under embargo. And in
> > > +general, please do not Cc: the kernel security list about fixes that
> > > +have already been merged.

This implies that in general a fix does not remain under embargo.
However, contacting "linux-distros" only makes sense when a fix does
remain under embargo (either not yet pushed to a public list/repo, or
under the Linux kernel exception for a public not-too-revealing fix) -
otherwise, the issue should be brought to "oss-security" right away.

Edited:

---
As such, the kernel security team strongly recommends that as a reporter
of a potential security issue you DO NOT contact the "linux-distros"
mailing list UNTIL a fix is accepted by the affected code's maintainers
and you have read the distros wiki page above and you fully understand
the requirements that contacting "linux-distros" will impose on you and
the kernel community.  This also means that in general it doesn't make
sense to Cc: both lists at once, except maybe for coordination if and
while an accepted fix has not yet been merged.  In other words, until a
fix is accepted do not Cc: "linux-distros", and after it's merged do not
Cc: the kernel security team.
---

This allows possible Cc'ing of both lists in the time window between
"fix is accepted by the affected code's maintainers" and "merged".
Makes sense?  I worry this distinction between accepted and merged may
be overly complicated for some, but I don't have better wording.

> > I was thinking about this Cc: thing and would it make sense to:
> > 
> > 1) have LKML and other public vger lists reject messages that include
> > s@k.o or (linux-)distros@ on Cc? The idea being that this is probably a
> > mistake -- I believe it has happened a few times recently by mistake.
> > 
> > 2) have (linux-)distros@ reject NEW threads (i.e. no In-Reply-To:) that
> > also include s@k.o on Cc? We could include a nice message explaining why
> > and to please resend when a patch has been developed and/or a disclosure
> > is planned in the next 7 days.
> 
> I don't know, maybe it would add extra config burden, but on the other
> hand it could avoid the mistake from newcomers who have not read the
> docs first (which happened a few times already), but if l-d becomes a
> bit more flexible and tolerant to reporters' mistakes, as now documented,
> it should also be less of a problem.
> 
> > I guess the problem with this would be if
> > somebody on s@k.o does a reply-all which would add distros right back in
> > the loop -OR- a patch has already been developed and included.
> 
> Then this would be deliberate, there would an in-reply-to so that would
> not be a problem. I really doubt anyone from s@k.o would Cc linux-distros
> anyway since it would imply disclosing some details from a reporter, and
> we do not do that, it's up to the reporter to do it if they want.

I think we don't want to complicate the setup, which we'd then have to
explain somewhere.  With my concern/edit above, also the logic isn't
that simple.

Alexander

Powered by blists - more mailing lists