[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <CAHC9VhRODBvRXGh+c5LBCwS2=pfxYpTfc-Wrt3DXD93wgq5yxQ@mail.gmail.com>
Date: Fri, 30 Jan 2026 16:47:08 -0500
From: Paul Moore <paul@...l-moore.com>
To: Stephen Smalley <stephen.smalley.work@...il.com>
Cc: cgzones@...glemail.com, selinux@...r.kernel.org,
Ondrej Mosnacek <omosnace@...hat.com>, linux-kernel@...r.kernel.org
Subject: Re: [RFC PATCH] selinux: prevent truncation of status map
On Fri, Jan 30, 2026 at 4:08 PM Stephen Smalley
<stephen.smalley.work@...il.com> wrote:
> On Fri, Jan 30, 2026 at 12:11 PM Christian Göttsche
> <cgoettsche@...tendoof.de> wrote:
> >
> > From: Christian Göttsche <cgzones@...glemail.com>
> >
> > Currently the SELinux status map can be truncated, given the necessary
> > permissions, leading to foreign user space processes getting a bus error
> > (SIGBUS) while concurrently making use of the status map.
> > For example systemd can be killed that way, see [1].
> >
> > Override the setattr inode handler and check for O_TRUNC in the open
> > handler to prevent truncations.
> >
> > Link [1]: https://github.com/systemd/systemd/issues/37349
> > Closes: https://github.com/SELinuxProject/selinux/issues/475
> > Signed-off-by: Christian Göttsche <cgzones@...glemail.com>
>
> This issue can't be limited to just the status node, so if we care
> about preventing this, we ought to do it for them all.
> When would we want to allow truncation of _any_ selinuxfs file?
Fair point.
--
paul-moore.com
Powered by blists - more mailing lists