OSEC

Neohapsis is currently accepting applications for employment. For more information, please visit our website www.neohapsis.com or email hr@neohapsis.com
 
From: Hal Flynn (flynnsecurityfocus.com)
Date: Mon May 27 2002 - 10:33:33 CDT

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

    I tried to get a copy of the trojaned source, but was unsuccessful.

    From what I can gather, there's two likely scenarios involving this
    problem.

    Scenario #1:
    The trojaned code was placed in a section of the source which was only
    executed by the user during the initial ./configure ; make ; make
    install sequence.

    This may not yield root privileges (if the user was intelligent enough to
    execute these commands from a standard user account), though this is going
    to yield minimally an unprivileged shell from the users account. From
    there, it's strictly a matter of time. If an attacker is able to gain
    access to a system with an unprivileged account, a local vulnerability
    will eventually be uncovered that gives elevated privileges.

    Scenario #2:
    The trojaned code was placed in the configure that is executed during the
    make install sequence. This would likely result in execution by root, as
    the default goes to /usr/local. Obviously, this requires administrative
    access for successful installation.

    If this is the case, obviously, the results are unpredictable. The
    connection initiation likely just gave the remote system an administrative
    shell to bind to. I'd like to think the attacker was this
    unsophisticated, but that would likely be underestimating him/her.

    Hal Flynn
    UNIX Focus Area Manager
    SecurityFocus

    "Semper Fidelis"