From mboxrd@z Thu Jan  1 00:00:00 1970
Return-Path: <linux-kernel-owner@vger.kernel.org>
Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand
	id S1751570Ab1HLVPA (ORCPT <rfc822;w@1wt.eu>);
	Fri, 12 Aug 2011 17:15:00 -0400
Received: from youngberry.canonical.com ([91.189.89.112]:54496 "EHLO
	youngberry.canonical.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org
	with ESMTP id S1751091Ab1HLVO6 (ORCPT
	<rfc822;linux-kernel@vger.kernel.org>);
	Fri, 12 Aug 2011 17:14:58 -0400
Date: Fri, 12 Aug 2011 16:13:53 -0500
From: Serge Hallyn <serge.hallyn@canonical.com>
To: Daniel Lezcano <daniel.lezcano@free.fr>
Cc: akpm@linux-foundation.org, containers@lists.linux-foundation.org,
        bonbons@linux-vserver.org, oleg@tv-sign.ru,
        linux-kernel@vger.kernel.org
Subject: Re: [PATCH 2/2] Notify container-init parent a 'reboot' occured
Message-ID: <20110812211353.GA19030@peqn>
References: <1313094241-3674-1-git-send-email-daniel.lezcano@free.fr>
 <1313094241-3674-3-git-send-email-daniel.lezcano@free.fr>
 <20110811210951.GA17349@peqn>
 <4E444A04.3070403@free.fr>
 <20110812162908.GA9304@peqn>
 <4E45904F.60904@free.fr>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <4E45904F.60904@free.fr>
User-Agent: Mutt/1.5.21 (2010-09-15)
Sender: linux-kernel-owner@vger.kernel.org
List-ID: <linux-kernel.vger.kernel.org>
X-Mailing-List: linux-kernel@vger.kernel.org

Quoting Daniel Lezcano (daniel.lezcano@free.fr):
> On 08/12/2011 06:29 PM, Serge Hallyn wrote:
> > Quoting Daniel Lezcano (daniel.lezcano@free.fr):
> > ...
> >>> Doesn't this mean that an unprivileged task in a container can shut
> >>> down the container?
> >> Ha ha ! Right, good catch :)
> >>
> >> Yes, rethinking about it, we can do what initially proposed Bruno by
> >> just preventing to reboot when we are not in the init_pid_ns. Actually, 
> >> the sys_reboot occurs after the services shutdown and "kill -1 SIGTERM"
> >> and "kill -1 SIGKILL", and would not make sense to do that in a child
> >> pid namespace, except if we are in a container where we don't want to
> >> reboot :)
> >>
> >> So IMO, it is safe to do:
> >>
> >> 	if (!ns_capable(current_pid_ns()->user_ns, CAP_SYS_BOOT))
> >>  		return -EPERM;
> >> 	
> >> 	if (pid_ns != &init_pid_ns)
> >> 		return pid_namespace_reboot(pid_ns, cmd, buffer);
> > So I don't know if you want to prepend
> > http://kernel.ubuntu.com/git?p=serge/linux-syslogns.git;a=commit;h=63556e9a39bcd75ec4a88333425800905013c73e
> > to your patchset, or just check nsown_capable(CAP_SYS_BOOT) for now,
> > but as soon as you resend with that I'll happily, nay,
> > ecstatically ack.
> 
> Ok, in order to not mix the functionnalities, I will send in a separate
> patch the nsown_capable change.

Actually the nsown_capable was a temp hack fix anyway, for a problem
we don't actually have.

I forgot you are not using user namespaces yet anyway.  So you can just
leave it as

	if (!capable(CAP_SYS_BOOT))
		return -EPERM;

for now.  Since you're root in the init user namespace, that'll pass
(as soon as you don't drop it from bounding set of course)

thanks,
-serge