public inbox for linux-kernel@vger.kernel.org
 help / color / mirror / Atom feed
From: Vivek Goyal <vgoyal@redhat.com>
To: Mimi Zohar <zohar@linux.vnet.ibm.com>
Cc: "Kasatkin, Dmitry" <dmitry.kasatkin@intel.com>,
	linux-security-module@vger.kernel.org,
	linux-kernel@vger.kernel.org
Subject: Re: [PATCH 2/2] ima: Support appraise_type=imasig_optional
Date: Mon, 18 Feb 2013 13:21:14 -0500	[thread overview]
Message-ID: <20130218182114.GB25922@redhat.com> (raw)
In-Reply-To: <1360855815.3524.615.camel@falcor1.watson.ibm.com>

On Thu, Feb 14, 2013 at 10:30:15AM -0500, Mimi Zohar wrote:
> On Thu, 2013-02-14 at 10:03 -0500, Vivek Goyal wrote:
> > On Wed, Feb 13, 2013 at 05:27:01PM -0500, Mimi Zohar wrote:
> > 
> > [..]
> > > > Yep, I got that. Default policy gets overruled when a new policy is
> > > > loaded.
> > > > 
> > > > In secureboot mode, somehow above rule needs to take effect by default.
> > > > One option would be that kernel can enforce above rule.
> > > > (I guess by adding it to both default_list as well as policy list).
> > > 
> > > The default policy is empty, but can be replaced with boot command line
> > > options.  The existing options are ima_tcb and/ ima_appraise_tcb.
> > > Please feel free to define an additional policy.
> > 
> > I think just defining a new command line option is not sufficient
> > for secureboot use case.
> > 
> > - One can easily remove kernel command line option without breaking
> >   booting and easily bypass secureboot restrictions.
> 
> > - I guess this is one mandated rule by secureboot. There might still
> >   be a user policy which can co-exist with this rule.
> > 
> > So to me this is not a new policy. It is just one mandatory rule which
> > gets appended to any policy in secureboot mode. Think of it as mandatory
> > rule imposed by kernel for any policy user can define. And in secureboot
> > mode a user can not get rid of this rule. (Otherwise it breaks user
> > space signing and one can bypass secureboot and boot into unsigned
> > kernel).
> 
> Your rule allows both signed and unsigned files to be executed.  Signed
> files will just have more capabilities.  The ima_appraise_tcb option
> requires all files owned by root to be signed, otherwise access is
> denied.  The two policies simply can not co-exist.

Thinking loud. I guess we might have to extend ima policy/rules to allow
multiple appraise rules to co-exist. And access permission will finally
depend on if all the rules in same category return success.

Thanks
Vivek

  reply	other threads:[~2013-02-18 18:21 UTC|newest]

Thread overview: 47+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2013-02-11 20:11 [RFC PATCH 0/2] ima: Support a mode to appraise signed files only Vivek Goyal
2013-02-11 20:11 ` [PATCH 1/2] ima: Do not try to fix hash if file system does not support security xattr Vivek Goyal
2013-02-12 11:45   ` Mimi Zohar
2013-02-12 14:27     ` Vivek Goyal
2013-02-11 20:11 ` [PATCH 2/2] ima: Support appraise_type=imasig_optional Vivek Goyal
2013-02-11 22:10   ` Mimi Zohar
2013-02-12 14:26     ` Vivek Goyal
2013-02-12 17:14       ` Mimi Zohar
2013-02-12 18:52         ` Vivek Goyal
2013-02-12 18:57           ` Vivek Goyal
2013-02-13 12:14             ` Kasatkin, Dmitry
2013-02-13 13:29               ` Vivek Goyal
2013-02-13 13:36                 ` Kasatkin, Dmitry
2013-02-13 13:49                   ` Vivek Goyal
2013-02-13 14:03                   ` Mimi Zohar
2013-02-13 14:38                   ` Vivek Goyal
2013-02-13 15:26                     ` Kasatkin, Dmitry
2013-02-13 15:29                       ` Kasatkin, Dmitry
2013-02-13 15:39                         ` Vivek Goyal
2013-02-13 15:30                       ` Vivek Goyal
2013-02-13 22:27                         ` Mimi Zohar
2013-02-14 15:03                           ` Vivek Goyal
2013-02-14 15:30                             ` Mimi Zohar
2013-02-18 18:21                               ` Vivek Goyal [this message]
2013-02-19 21:54                                 ` Mimi Zohar
2013-02-13 15:51                     ` Mimi Zohar
2013-02-12 20:05           ` Mimi Zohar
2013-02-13 12:31   ` Kasatkin, Dmitry
2013-02-13 12:56     ` Mimi Zohar
2013-02-13 13:13       ` Kasatkin, Dmitry
2013-02-13 13:44         ` Mimi Zohar
2013-02-13 16:59           ` Vivek Goyal
2013-02-14 12:57             ` Mimi Zohar
2013-02-14 15:23               ` Vivek Goyal
2013-02-14 15:35                 ` Mimi Zohar
2013-02-14 16:17                   ` Vivek Goyal
2013-02-14 16:31                     ` Vivek Goyal
2013-02-14 19:49                     ` Mimi Zohar
2013-02-14 20:54                       ` Vivek Goyal
2013-02-14 20:57                         ` Vivek Goyal
2013-02-14 21:54                           ` Mimi Zohar
2013-02-13 17:33           ` Kasatkin, Dmitry
2013-02-13 17:51             ` Vivek Goyal
2013-02-13 18:20               ` Kasatkin, Dmitry
2013-02-13 21:45             ` Mimi Zohar
2013-02-14 14:40               ` Vivek Goyal
2013-02-14 15:48                 ` Mimi Zohar

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20130218182114.GB25922@redhat.com \
    --to=vgoyal@redhat.com \
    --cc=dmitry.kasatkin@intel.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-security-module@vger.kernel.org \
    --cc=zohar@linux.vnet.ibm.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox