From mboxrd@z Thu Jan 1 00:00:00 1970 Return-path: Received: from hrndva-omtalb.mail.rr.com ([71.74.56.122]) by merlin.infradead.org with esmtp (Exim 4.80.1 #2 (Red Hat Linux)) id 1UIhXj-00079l-Db for kexec@lists.infradead.org; Thu, 21 Mar 2013 15:36:40 +0000 Date: Thu, 21 Mar 2013 10:37:25 -0500 From: "Serge E. Hallyn" Subject: Re: [PATCH 01/12] Security: Add CAP_COMPROMISE_KERNEL Message-ID: <20130321153725.GA3656@austin.hallyn.com> References: <1363797717.2580.10.camel@falcor1.watson.ibm.com> <1363798166.2553.29.camel@x230.sbx07502.somerma.wayport.net> <1363802506.2580.55.camel@falcor1.watson.ibm.com> <1363803158.2553.33.camel@x230.sbx07502.somerma.wayport.net> <1363806968.2580.86.camel@falcor1.watson.ibm.com> <1363811856.2553.37.camel@x230.sbx07502.somerma.wayport.net> <1363813877.2580.120.camel@falcor1.watson.ibm.com> <1363814289.2553.41.camel@x230.sbx07502.somerma.wayport.net> <20130321134348.GA3934@redhat.com> MIME-Version: 1.0 Content-Disposition: inline In-Reply-To: <20130321134348.GA3934@redhat.com> List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Sender: "kexec" Errors-To: kexec-bounces+dwmw2=infradead.org@lists.infradead.org To: Vivek Goyal Cc: Matthew Garrett , "linux-efi@vger.kernel.org" , "linux-pci@vger.kernel.org" , "kexec@lists.infradead.org" , James Morris , "linux-kernel@vger.kernel.org" , "linux-security-module@vger.kernel.org" , Mimi Zohar , "Serge E. Hallyn" Quoting Vivek Goyal (vgoyal@redhat.com): ... > Giving CAP_MODIFY_KERNEL to processess upon signature verification > will simplify things a bit. > > Only thing is that signature verification alone is not sufficient. We > also need to make sure after signature verification executable can > not be modified in memory in any way. So that means atleast couple of > things. Also what about context? If I construct a mounts namespace a certain way, can I trick this executable into loading an old singed bzImage that I had laying around? ISTM the only sane thing to do, if you're going down this road, is to have CAP_MODIFIY_KERNEL pulled from bounding set for everyone except a getty started by init on ttyS0. Then log in on serial to update. Or run a damon with CAP_MODIFIY_KERNEL which listens to a init_net_ns netlink socket for very basic instructions, like "find and install latest signed bzImage in /boot". Then you can at least trust that /boot for that daemon is not faked. _______________________________________________ kexec mailing list kexec@lists.infradead.org http://lists.infradead.org/mailman/listinfo/kexec From mboxrd@z Thu Jan 1 00:00:00 1970 From: "Serge E. Hallyn" Subject: Re: [PATCH 01/12] Security: Add CAP_COMPROMISE_KERNEL Date: Thu, 21 Mar 2013 10:37:25 -0500 Message-ID: <20130321153725.GA3656@austin.hallyn.com> References: <1363797717.2580.10.camel@falcor1.watson.ibm.com> <1363798166.2553.29.camel@x230.sbx07502.somerma.wayport.net> <1363802506.2580.55.camel@falcor1.watson.ibm.com> <1363803158.2553.33.camel@x230.sbx07502.somerma.wayport.net> <1363806968.2580.86.camel@falcor1.watson.ibm.com> <1363811856.2553.37.camel@x230.sbx07502.somerma.wayport.net> <1363813877.2580.120.camel@falcor1.watson.ibm.com> <1363814289.2553.41.camel@x230.sbx07502.somerma.wayport.net> <20130321134348.GA3934@redhat.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Return-path: Content-Disposition: inline In-Reply-To: <20130321134348.GA3934-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org> Sender: linux-efi-owner-u79uwXL29TY76Z2rM5mHXA@public.gmane.org To: Vivek Goyal Cc: Matthew Garrett , Mimi Zohar , James Morris , "linux-kernel-u79uwXL29TY76Z2rM5mHXA@public.gmane.org" , "linux-security-module-u79uwXL29TY76Z2rM5mHXA@public.gmane.org" , "linux-efi-u79uwXL29TY76Z2rM5mHXA@public.gmane.org" , "kexec-IAPFreCvJWM7uuMidbF8XUB+6BGkLq7r@public.gmane.org" , "linux-pci-u79uwXL29TY76Z2rM5mHXA@public.gmane.org" , "Serge E. Hallyn" List-Id: linux-efi@vger.kernel.org Quoting Vivek Goyal (vgoyal-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org): ... > Giving CAP_MODIFY_KERNEL to processess upon signature verification > will simplify things a bit. > > Only thing is that signature verification alone is not sufficient. We > also need to make sure after signature verification executable can > not be modified in memory in any way. So that means atleast couple of > things. Also what about context? If I construct a mounts namespace a certain way, can I trick this executable into loading an old singed bzImage that I had laying around? ISTM the only sane thing to do, if you're going down this road, is to have CAP_MODIFIY_KERNEL pulled from bounding set for everyone except a getty started by init on ttyS0. Then log in on serial to update. Or run a damon with CAP_MODIFIY_KERNEL which listens to a init_net_ns netlink socket for very basic instructions, like "find and install latest signed bzImage in /boot". Then you can at least trust that /boot for that daemon is not faked. From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from hrndva-omtalb.mail.rr.com ([71.74.56.122]:24052 "EHLO hrndva-omtalb.mail.rr.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1754529Ab3CUPgi (ORCPT ); Thu, 21 Mar 2013 11:36:38 -0400 Date: Thu, 21 Mar 2013 10:37:25 -0500 From: "Serge E. Hallyn" To: Vivek Goyal Cc: Matthew Garrett , Mimi Zohar , James Morris , "linux-kernel@vger.kernel.org" , "linux-security-module@vger.kernel.org" , "linux-efi@vger.kernel.org" , "kexec@lists.infradead.org" , "linux-pci@vger.kernel.org" , "Serge E. Hallyn" Subject: Re: [PATCH 01/12] Security: Add CAP_COMPROMISE_KERNEL Message-ID: <20130321153725.GA3656@austin.hallyn.com> References: <1363797717.2580.10.camel@falcor1.watson.ibm.com> <1363798166.2553.29.camel@x230.sbx07502.somerma.wayport.net> <1363802506.2580.55.camel@falcor1.watson.ibm.com> <1363803158.2553.33.camel@x230.sbx07502.somerma.wayport.net> <1363806968.2580.86.camel@falcor1.watson.ibm.com> <1363811856.2553.37.camel@x230.sbx07502.somerma.wayport.net> <1363813877.2580.120.camel@falcor1.watson.ibm.com> <1363814289.2553.41.camel@x230.sbx07502.somerma.wayport.net> <20130321134348.GA3934@redhat.com> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii In-Reply-To: <20130321134348.GA3934@redhat.com> Sender: linux-pci-owner@vger.kernel.org List-ID: Quoting Vivek Goyal (vgoyal@redhat.com): ... > Giving CAP_MODIFY_KERNEL to processess upon signature verification > will simplify things a bit. > > Only thing is that signature verification alone is not sufficient. We > also need to make sure after signature verification executable can > not be modified in memory in any way. So that means atleast couple of > things. Also what about context? If I construct a mounts namespace a certain way, can I trick this executable into loading an old singed bzImage that I had laying around? ISTM the only sane thing to do, if you're going down this road, is to have CAP_MODIFIY_KERNEL pulled from bounding set for everyone except a getty started by init on ttyS0. Then log in on serial to update. Or run a damon with CAP_MODIFIY_KERNEL which listens to a init_net_ns netlink socket for very basic instructions, like "find and install latest signed bzImage in /boot". Then you can at least trust that /boot for that daemon is not faked.