From: Eryu Guan <eguan@redhat.com>
To: Eric Biggers <ebiggers3@gmail.com>
Cc: fstests@vger.kernel.org, Theodore Ts'o <tytso@mit.edu>,
Gwendal Grignou <gwendal@chromium.org>,
Eric Biggers <ebiggers@google.com>
Subject: Re: [PATCH 2/2] common/config: don't hard-code SELinux context
Date: Tue, 14 Mar 2017 21:06:19 +0800 [thread overview]
Message-ID: <20170314130619.GD14226@eguan.usersys.redhat.com> (raw)
In-Reply-To: <20170313175935.GA108079@gmail.com>
On Mon, Mar 13, 2017 at 10:59:35AM -0700, Eric Biggers wrote:
> On Mon, Mar 13, 2017 at 12:02:26PM +0800, Eryu Guan wrote:
> > On Fri, Mar 10, 2017 at 04:50:48PM -0800, Eric Biggers wrote:
> > > From: Eric Biggers <ebiggers@google.com>
> > >
> > > If SELinux is enabled, xfstests mounts its filesystems with
> > > "-o context=system_u:object_r:nfs_t:s0" so that no SELinux xattrs get
> > > created and interfere with tests. However, this particular context is
> > > not guaranteed to be available because the context names are a detail of
> > > the SELinux policy. The SELinux policy on Android systems, for example,
> > > does not have a context with this name.
> > >
> > > To fix this, just grab the SELinux context of the root directory. This
> > > is arbitrary, but it should always provide a valid context. And any
> > > valid context *should* be okay (i.e. we don't necessarily need a
> > > "liberal" one), since one would likely encounter many other problems if
> > > they were to run xfstests in a confined context with SELinux in
> > > enforcing mode.
> > >
> > > Signed-off-by: Eric Biggers <ebiggers@google.com>
> >
> > SELINUX_MOUNT_OPTIONS has just been updated to be configurable, you can
> > set your own SELINUX_MOUNT_OPTIONS to override the default one, does
> > this work for you?
> >
> > d8b1dc1 common/config: make SELinux protection conditional
> >
> > Thanks,
> > Eryu
>
> Oh, I didn't notice this. It looks like Gwendal ran into the same problem, but
> on ChromeOS instead of Android.
>
> The problem can indeed be solved by overriding SELINUX_MOUNT_OPTIONS. But I
> think auto-detecting a valid context is better because then xfstests will just
> work without having to override SELINUX_MOUNT_OPTIONS.
>
> An exception would be that if for some reason someone actually wants to run
> xfstests in some particular SELinux context (maybe one they've set up
> specifically for xfstests), then they'd likely need to specify a particular
> context when mounting.
>
> How about just doing it both ways: use SELINUX_MOUNT_OPTIONS in the environment
> if set, otherwise mount with an auto-detected valid context?
This looks reasonable to me, and I tested ext4 ext3 and xfs with auto
group tests with selinux mount option set to `stat -c %C /`, and didn't
see any new failures.
Thanks,
Eryu
prev parent reply other threads:[~2017-03-14 13:06 UTC|newest]
Thread overview: 5+ messages / expand[flat|nested] mbox.gz Atom feed top
2017-03-11 0:50 [PATCH 1/2] common/quota: remove redundant SELinux detection code Eric Biggers
2017-03-11 0:50 ` [PATCH 2/2] common/config: don't hard-code SELinux context Eric Biggers
2017-03-13 4:02 ` Eryu Guan
2017-03-13 17:59 ` Eric Biggers
2017-03-14 13:06 ` Eryu Guan [this message]
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=20170314130619.GD14226@eguan.usersys.redhat.com \
--to=eguan@redhat.com \
--cc=ebiggers3@gmail.com \
--cc=ebiggers@google.com \
--cc=fstests@vger.kernel.org \
--cc=gwendal@chromium.org \
--cc=tytso@mit.edu \
/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