From: Murray McAllister <mmcallis@redhat.com>
To: James Morris <jmorris@namei.org>
Cc: SE Linux <selinux@tycho.nsa.gov>
Subject: Re: user guide draft: "Confined and Unconfined User Domains" review
Date: Thu, 11 Sep 2008 20:40:50 +1000 [thread overview]
Message-ID: <48C8F5B2.6000809@redhat.com> (raw)
In-Reply-To: <alpine.LRH.1.10.0809092001230.21555@tundra.namei.org>
James Morris wrote:
> On Tue, 9 Sep 2008, Murray McAllister wrote:
>
>> Each Linux user account is mapped to an SELinux user identity when a user
>> login session is created, and the mapped SELinux user identity is used in the
>> security context for processes in that session.
>
> This is a long sentence which I suspect general users would not easily
> understand. Perhaps break it into two sentences, with the second as:
>
> "The SELinux user identity is indicated in the user's process security
> context for that session."
Would the following be enough:
Each Linux user account is mapped to an SELinux user identity via
SELinux policy. By default, on Fedora 10, Linux users are mapped to the
SELinux unconfined_u user. This is seen by running the
/usr/sbin/semanage login -l command:
>
> Do you have a diagram breaking down the security context? You could refer
> to it here.
No. I will try to organize one. Is there anything specific that should
be on it?
>
>> By default, on Fedora 10,
>> Linux users are mapped to the SELinux unconfined_u user. This is seen by
>> running the id -Z and /usr/sbin/semanage login -l commands:
>>
>> # id -Z
>> unconfined_u:unconfined_r:unconfined_t:s0-s0:c0.c1023
>
> This command will have different outputs depending on how the user is
> logged in, and there are seemingly (to the reader) two different ways to
> see the SELinux user mapping (a new concept to them at this point).
>
> I suggest breaking it up so you first show the mapping via semanage, then
> show the output of 'id -Z' for one of the Unix logins, also perhaps
> explaining the flow:
>
> - Linux users are mapped to SELinux users via policy
>
> - user commences login
> - pam_selinux maps the user and sets up the resulting security context
> - user shell is launched in that context
Would an example of adding a user (useradd newuser), logging in as
newuser, then running "id -Z" help?
>
>> # /usr/sbin/semanage login -l
>>
>> Login Name SELinux User MLS/MCS Range
>>
>> __default__ unconfined_u s0-s0:c0.c1023
>> root unconfined_u s0-s0:c0.c1023
>> system_u system_u s0-s0:c0.c1023
>>
>> The first row, __default__, defines that any new Linux users created that are
>> not specifically mapped to an SELinux user, are mapped to the SELinux
>> unconfined_u user. For a description of each column, refer to Chapter 3,
>> SELinux Contexts.
>
> I think you need to refer to a concrete example with the current text.
An example of what a user sees (see above, adding newuser), or
explaining what each field is?
>
>> Unconfined Linux users are subject to executable and
>> writeable memory checks, and are also restricted by MCS (and MLS, if the MLS
>> policy is used). If they execute an object that the SELinux policy defines can
>
> Why introduce unfamiliar terminology like "execute an object" ? People
> execute applications.
I have removed these and replaced "subjects" with "processes", "execute
object" with "execute application", and so on.
>
>> transition from the unconfined_t domain to its own confined domain, the
>> unconfined Linux users are still subject to the restrictions of that confined
>> domain.
>
> Perhaps important to (re)state the security benefit of this, in that
> an unconfined user cannot override the security policy for a confined
> application just because they themselves are unconfined.
Sounds good. Thank you.
--
This message was distributed to subscribers of the selinux mailing list.
If you no longer wish to subscribe, send mail to majordomo@tycho.nsa.gov with
the words "unsubscribe selinux" without quotes as the message.
next prev parent reply other threads:[~2008-09-11 10:41 UTC|newest]
Thread overview: 12+ messages / expand[flat|nested] mbox.gz Atom feed top
2008-09-09 7:19 user guide draft: "Confined and Unconfined User Domains" review Murray McAllister
2008-09-09 10:29 ` James Morris
2008-09-11 10:40 ` Murray McAllister [this message]
2008-09-09 13:44 ` Daniel J Walsh
2008-09-15 1:27 ` Murray McAllister
2008-09-15 2:12 ` Murray McAllister
2008-09-15 11:17 ` Dominick Grift
2008-09-15 13:07 ` Daniel J Walsh
2008-09-15 13:03 ` Daniel J Walsh
2008-09-17 5:53 ` Murray McAllister
2008-09-15 11:10 ` Dominick Grift
2008-09-17 5:43 ` Murray McAllister
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=48C8F5B2.6000809@redhat.com \
--to=mmcallis@redhat.com \
--cc=jmorris@namei.org \
--cc=selinux@tycho.nsa.gov \
/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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.