Re: 2.4.x/Slackware Init script vulnerability

From: Keith Owens (kaosat_private)
Date: Tue Jul 17 2001 - 16:59:43 PDT

  • Next message: Red Wolf: "RE: W2k: Unkillable Applications"

    -----BEGIN PGP SIGNED MESSAGE-----
    Hash: SHA1
    
    Content-Type: text/plain; charset=us-ascii
    
    On Tue, 17 Jul 2001 16:32:07 -0400, 
    Derek Martin <ddmat_private> wrote:
    >I also did the same thing on a Red Hat 7.1 system, with modutils 2.4.2
    >(as shipped by Red Hat), and linux 2.4.5 (pristine), and the modules.*
    >files were recreated with permissions 0644 upon reboot, so it seems
    >not to be limited to just Slackware, but also not a universal problem.
    >Since it did not happen on RH 7.1 with modutils 2.4.2, it may be that
    >the problem is actually in modutils 2.4.3 (and later, probably), and
    >not in earlier modutils.  I think this is probably not really a kernel
    >issue, per se.
    
    None of the above.  A change to the kernel in 2.4.3-pre5 or -pre6
    caused all kernel thread programs to run with umask 0, including init.
    Newer Redhat rc.sysinit sets the umask instead of trusting the kernel
    value, older Redhat and current Slackware trust the kernel.  modutils
    trust umask.
    
    >I would think that modutils should set the creation mode to 0644 when
    >creating these files.  I would also think that as a security measure,
    >modutils should verify that these files (or at least modules.dep) are
    >not world-writable (and probably also not group writable) BEFORE
    >loading modules as a result of listed dependencies...
    
    When programs should force security settings and when they should trust
    the umask is a policy question.  Users on development systems
    deliberately create modules.dep as 666 and allow modules to be owned as
    other than root so modutils allows this.  In this case I decided that
    the policy setting should come from the user via umask, instead of
    being forced by the programs.  If root's umask is 000 then lots of
    programs are insecure, should all of those programs be changed to
    ignore umask?
    
    >I'm not really
    >sure that the kernel itself should automatically set a restrictive
    >umask, as I would think it should be up to user-space programs to
    >decide that; but it probably doesn't matter much either way.
    
    The kernel normally mimics the default umask for shells, 022.  The
    change from 022 to 000 was incorrect and will be backed out in the next
    kernel release.
    
    Keith Owens, modutils maintainer.
    
    -----BEGIN PGP SIGNATURE-----
    Version: GnuPG v1.0.3 (GNU/Linux)
    Comment: Exmh version 2.1.1 10/15/1999
    
    iD8DBQE7VNFui4UHNye0ZOoRAubSAJ9NnJnZ9QoyFKs3cXDS4ys/di5QCACg5KEX
    3NjShKKQqwm3mMRlACkUFes=
    =q86Y
    -----END PGP SIGNATURE-----
    



    This archive was generated by hypermail 2b30 : Wed Jul 18 2001 - 08:29:19 PDT