linux-ext4.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: "Serge E. Hallyn" <serge@hallyn.com>
To: "Eric W. Biederman" <ebiederm@xmission.com>
Cc: "Aleksa Sarai" <cyphar@cyphar.com>,
	"Christian Brauner" <christian.brauner@ubuntu.com>,
	"Alexander Viro" <viro@zeniv.linux.org.uk>,
	"Christoph Hellwig" <hch@infradead.org>,
	linux-fsdevel@vger.kernel.org,
	"John Johansen" <john.johansen@canonical.com>,
	"James Morris" <jmorris@namei.org>,
	"Mimi Zohar" <zohar@linux.ibm.com>,
	"Dmitry Kasatkin" <dmitry.kasatkin@gmail.com>,
	"Stephen Smalley" <stephen.smalley.work@gmail.com>,
	"Casey Schaufler" <casey@schaufler-ca.com>,
	"Arnd Bergmann" <arnd@arndb.de>,
	"Andreas Dilger" <adilger.kernel@dilger.ca>,
	"OGAWA Hirofumi" <hirofumi@mail.parknet.co.jp>,
	"Geoffrey Thomas" <geofft@ldpreload.com>,
	"Mrunal Patel" <mpatel@redhat.com>,
	"Josh Triplett" <josh@joshtriplett.org>,
	"Andy Lutomirski" <luto@kernel.org>,
	"Amir Goldstein" <amir73il@gmail.com>,
	"Miklos Szeredi" <miklos@szeredi.hu>,
	"Theodore Tso" <tytso@mit.edu>, "Alban Crequy" <alban@kinvolk.io>,
	"Tycho Andersen" <tycho@tycho.ws>,
	"David Howells" <dhowells@redhat.com>,
	"James Bottomley" <James.Bottomley@hansenpartnership.com>,
	"Jann Horn" <jannh@google.com>,
	"Seth Forshee" <seth.forshee@canonical.com>,
	"Stéphane Graber" <stgraber@ubuntu.com>,
	"Lennart Poettering" <lennart@poettering.net>,
	smbarber@chromium.org, "Phil Estes" <estesp@gmail.com>,
	"Serge Hallyn" <serge@hallyn.com>,
	"Kees Cook" <keescook@chromium.org>,
	"Todd Kjos" <tkjos@google.com>,
	"Jonathan Corbet" <corbet@lwn.net>,
	containers@lists.linux-foundation.org,
	linux-security-module@vger.kernel.org, linux-api@vger.kernel.org,
	linux-ext4@vger.kernel.org, linux-unionfs@vger.kernel.org,
	linux-audit@redhat.com, linux-integrity@vger.kernel.org,
	selinux@vger.kernel.org
Subject: Re: [PATCH 00/34] fs: idmapped mounts
Date: Thu, 29 Oct 2020 21:18:05 -0500	[thread overview]
Message-ID: <20201030021805.GA20489@mail.hallyn.com> (raw)
In-Reply-To: <87361xdm4c.fsf@x220.int.ebiederm.org>

On Thu, Oct 29, 2020 at 11:37:23AM -0500, Eric W. Biederman wrote:
> Aleksa Sarai <cyphar@cyphar.com> writes:
> 
> > On 2020-10-29, Eric W. Biederman <ebiederm@xmission.com> wrote:
> >> Christian Brauner <christian.brauner@ubuntu.com> writes:
> >> 
> >> > Hey everyone,
> >> >
> >> > I vanished for a little while to focus on this work here so sorry for
> >> > not being available by mail for a while.
> >> >
> >> > Since quite a long time we have issues with sharing mounts between
> >> > multiple unprivileged containers with different id mappings, sharing a
> >> > rootfs between multiple containers with different id mappings, and also
> >> > sharing regular directories and filesystems between users with different
> >> > uids and gids. The latter use-cases have become even more important with
> >> > the availability and adoption of systemd-homed (cf. [1]) to implement
> >> > portable home directories.
> >> 
> >> Can you walk us through the motivating use case?
> >> 
> >> As of this year's LPC I had the distinct impression that the primary use
> >> case for such a feature was due to the RLIMIT_NPROC problem where two
> >> containers with the same users still wanted different uid mappings to
> >> the disk because the users were conflicting with each other because of
> >> the per user rlimits.
> >> 
> >> Fixing rlimits is straight forward to implement, and easier to manage
> >> for implementations and administrators.
> >
> > This is separate to the question of "isolated user namespaces" and
> > managing different mappings between containers. This patchset is solving
> > the same problem that shiftfs solved -- sharing a single directory tree
> > between containers that have different ID mappings. rlimits (nor any of
> > the other proposals we discussed at LPC) will help with this problem.
> 
> First and foremost: A uid shift on write to a filesystem is a security
> bug waiting to happen.  This is especially in the context of facilities
> like iouring, that play very agressive games with how process context
> makes it to  system calls.
> 
> The only reason containers were not immediately exploitable when iouring
> was introduced is because the mechanisms are built so that even if
> something escapes containment the security properties still apply.
> Changes to the uid when writing to the filesystem does not have that
> property.  The tiniest slip in containment will be a security issue.
> 
> This is not even the least bit theoretical.  I have seem reports of how
> shitfs+overlayfs created a situation where anyone could read
> /etc/shadow.
> 
> If you are going to write using the same uid to disk from different
> containers the question becomes why can't those containers configure
> those users to use the same kuid?

Because if user 'myapp' in two otherwise isolated containers both have
the same kuid, so that they can write to a shared directory, then root
in container 1 has privilege over all files owned by 'myapp' in
container 2.

Whereas if they can each have distinct kuids, but when writing to the
shared fs have a shared uid not otherwise belonging to either container,
their rootfs's can remain completely off limits to each other.

> What fixing rlimits does is it fixes one of the reasons that different
> containers could not share the same kuid for users that want to write to
> disk with the same uid.
> 
> 
> I humbly suggest that it will be more secure, and easier to maintain for
> both developers and users if we fix the reasons people want different
> containers to have the same user running with different kuids.
> 
> If not what are the reasons we fundamentally need the same on-disk user
> using multiple kuids in the kernel?
> 
> Eric

  reply	other threads:[~2020-10-30  2:18 UTC|newest]

Thread overview: 66+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2020-10-29  0:32 [PATCH 00/34] fs: idmapped mounts Christian Brauner
2020-10-29  0:32 ` [PATCH 01/34] namespace: take lock_mount_hash() directly when changing flags Christian Brauner
2020-11-01 14:41   ` Christoph Hellwig
2020-11-02 13:33     ` Christian Brauner
2020-10-29  0:32 ` [PATCH 02/34] namespace: only take read lock in do_reconfigure_mnt() Christian Brauner
2020-10-29  0:32 ` [PATCH 03/34] fs: add mount_setattr() Christian Brauner
2020-11-01 14:42   ` Christoph Hellwig
2020-11-02 13:34     ` Christian Brauner
2020-10-29  0:32 ` [PATCH 04/34] tests: add mount_setattr() selftests Christian Brauner
2020-10-29  0:32 ` [PATCH 05/34] fs: introduce MOUNT_ATTR_IDMAP Christian Brauner
2020-11-01 14:45   ` Christoph Hellwig
2020-11-02 13:29     ` Christian Brauner
2020-10-29  0:32 ` [PATCH 06/34] fs: add id translation helpers Christian Brauner
2020-11-01 14:46   ` Christoph Hellwig
2020-11-02 13:25     ` Christian Brauner
2020-10-29  0:32 ` [PATCH 07/34] capability: handle idmapped mounts Christian Brauner
2020-11-01 14:48   ` Christoph Hellwig
2020-11-02 13:23     ` Christian Brauner
2020-10-29  0:32 ` [PATCH 08/34] namei: add idmapped mount aware permission helpers Christian Brauner
2020-10-29  0:32 ` [PATCH 09/34] inode: add idmapped mount aware init and " Christian Brauner
2020-10-29  0:32 ` [PATCH 10/34] attr: handle idmapped mounts Christian Brauner
2020-10-29  0:32 ` [PATCH 11/34] acl: " Christian Brauner
2020-10-29  0:32 ` [PATCH 12/34] xattr: " Christian Brauner
2020-10-29  0:32 ` [PATCH 13/34] selftests: add idmapped mounts xattr selftest Christian Brauner
2020-10-29  0:32 ` [PATCH 14/34] commoncap: handle idmapped mounts Christian Brauner
2020-10-29  0:32 ` [PATCH 15/34] stat: add mapped_generic_fillattr() Christian Brauner
2020-10-29  0:32 ` [PATCH 16/34] namei: handle idmapped mounts in may_*() helpers Christian Brauner
2020-10-29  0:32 ` [PATCH 17/34] namei: introduce struct renamedata Christian Brauner
2020-10-29  0:32 ` [PATCH 18/34] namei: prepare for idmapped mounts Christian Brauner
2020-10-29  0:32 ` [PATCH 19/34] namei: add lookup helpers with idmapped mounts aware permission checking Christian Brauner
2020-10-29  0:32 ` [PATCH 20/34] open: handle idmapped mounts in do_truncate() Christian Brauner
2020-10-29  0:32 ` [PATCH 21/34] open: handle idmapped mounts Christian Brauner
2020-10-29  0:32 ` [PATCH 22/34] af_unix: " Christian Brauner
2020-10-29  0:32 ` [PATCH 23/34] utimes: " Christian Brauner
2020-10-29  0:32 ` [PATCH 24/34] would_dump: " Christian Brauner
2020-10-29  0:32 ` [PATCH 25/34] exec: " Christian Brauner
2020-10-29  0:32 ` [PATCH 26/34] fs: add helpers for idmap mounts Christian Brauner
2020-10-29  0:32 ` [PATCH 27/34] apparmor: handle idmapped mounts Christian Brauner
2020-10-29  0:32 ` [PATCH 28/34] audit: " Christian Brauner
2020-10-29  0:32 ` [PATCH 29/34] ima: " Christian Brauner
2020-10-29  0:32 ` [PATCH 30/34] ext4: support " Christian Brauner
2020-10-29  0:32 ` [PATCH 31/34] expfs: handle " Christian Brauner
2020-10-29  0:32 ` [PATCH 32/34] overlayfs: handle idmapped lower directories Christian Brauner
2020-10-29  0:32 ` [PATCH 33/34] overlayfs: handle idmapped merged mounts Christian Brauner
2020-10-30  9:57   ` Amir Goldstein
2020-10-29  0:32 ` [PATCH 34/34] fat: handle idmapped mounts Christian Brauner
2020-10-29  2:27 ` [PATCH 00/34] fs: " Dave Chinner
2020-10-29 16:19   ` Christian Brauner
2020-10-29  7:20 ` Sargun Dhillon
2020-10-29 15:47 ` Eric W. Biederman
2020-10-29 15:51   ` Aleksa Sarai
2020-10-29 16:37     ` Eric W. Biederman
2020-10-30  2:18       ` Serge E. Hallyn [this message]
2020-10-30 15:07       ` Seth Forshee
2020-10-30 16:03         ` Serge E. Hallyn
2020-11-03 14:10       ` Alban Crequy
2020-10-29 16:05   ` Lennart Poettering
2020-10-29 16:36     ` Sargun Dhillon
2020-10-29 16:54     ` Eric W. Biederman
     [not found]   ` <20201029161231.GA108315@cisco>
2020-10-29 16:23     ` Serge E. Hallyn
2020-10-29 16:44     ` Eric W. Biederman
2020-10-29 18:04       ` Stéphane Graber
2020-10-29 21:58 ` Andy Lutomirski
2020-10-30 12:01   ` Christian Brauner
2020-10-30 16:17     ` Serge E. Hallyn
2020-10-31 17:43     ` Andy Lutomirski

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=20201030021805.GA20489@mail.hallyn.com \
    --to=serge@hallyn.com \
    --cc=James.Bottomley@hansenpartnership.com \
    --cc=adilger.kernel@dilger.ca \
    --cc=alban@kinvolk.io \
    --cc=amir73il@gmail.com \
    --cc=arnd@arndb.de \
    --cc=casey@schaufler-ca.com \
    --cc=christian.brauner@ubuntu.com \
    --cc=containers@lists.linux-foundation.org \
    --cc=corbet@lwn.net \
    --cc=cyphar@cyphar.com \
    --cc=dhowells@redhat.com \
    --cc=dmitry.kasatkin@gmail.com \
    --cc=ebiederm@xmission.com \
    --cc=estesp@gmail.com \
    --cc=geofft@ldpreload.com \
    --cc=hch@infradead.org \
    --cc=hirofumi@mail.parknet.co.jp \
    --cc=jannh@google.com \
    --cc=jmorris@namei.org \
    --cc=john.johansen@canonical.com \
    --cc=josh@joshtriplett.org \
    --cc=keescook@chromium.org \
    --cc=lennart@poettering.net \
    --cc=linux-api@vger.kernel.org \
    --cc=linux-audit@redhat.com \
    --cc=linux-ext4@vger.kernel.org \
    --cc=linux-fsdevel@vger.kernel.org \
    --cc=linux-integrity@vger.kernel.org \
    --cc=linux-security-module@vger.kernel.org \
    --cc=linux-unionfs@vger.kernel.org \
    --cc=luto@kernel.org \
    --cc=miklos@szeredi.hu \
    --cc=mpatel@redhat.com \
    --cc=selinux@vger.kernel.org \
    --cc=seth.forshee@canonical.com \
    --cc=smbarber@chromium.org \
    --cc=stephen.smalley.work@gmail.com \
    --cc=stgraber@ubuntu.com \
    --cc=tkjos@google.com \
    --cc=tycho@tycho.ws \
    --cc=tytso@mit.edu \
    --cc=viro@zeniv.linux.org.uk \
    --cc=zohar@linux.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;
as well as URLs for NNTP newsgroup(s).