Release : ComLog 1.0, a WIN32 command prompt logger

From: Floydman (floydian_99at_private)
Date: Wed Aug 14 2002 - 15:42:02 PDT

  • Next message: Seth Arnold: "Re: Handling, possibly, encrypted data"

    Sorry for the repost, but one of the list moderator asked for only a 
    resume,so I thought the other lists might be the same.  You can get the 
    full paper at one of my websites.
    
    ComLog.pl, a WIN32 command prompt logger
    by Floydman, floydian_99at_private
    August 13th 2002
    
    This paper is available online at www.geocities.com/floydian_99 and 
    http://securit.iquebec.com
    
    This paper can be freely distributed and reproduced, as long as correct 
    credentials are maintained, and that no modifications are made to this 
    file.  For corrections, suggestions or comments, please send me an e-mail.
    
    Abstract
    
    The goal of this paper is to present a new Perl tool that I made to monitor 
    DOS sessions on Windows NT/2K (should also work on XP, but cannot try 
    it).  This tool can be used by administrator to keep a history of commands 
    typed in the DOS command prompt and the associated output, for example on 
    an IIS server.  This can help admins to figure out what the attacker has 
    done after compromising the machine via one of the numerous vulnerabilities 
    this software present.  Of course, like any keylogger out there, it can 
    also be use for malicious purpose, but I think other programs out there 
    make better spying tools.
    
    
    Preface
    
    I started writing about my own experiences in the security field after 
    reading Lance Spitzner's whitepapers, way back before the HoneyNet Project 
    grew into the multi-disciplinary team that has now joined him and make the 
    project as we know it today.  I also wanted to inspire myself of what was 
    being done on the UNIX platform (making your own tools when they don't 
    exists) to transpose it in the Windows world, where I am more 
    comfortable.  This paper follows these traditions, as the idea for it came 
    to me while reading the HoneyNet Project recent book, Know your ennemy.  In 
    the book, one passage relate to how the HoneyNet is designed to sniff all 
    the traffic that passes into it since all of this traffic is suspicious by 
    nature.  This worked great for capturing what the attackers were doing on 
    the systems, up to the day when someone was using cryptcat to hide his 
    session.  To circumvent this problem, the team made a modification in the 
    source code of bash to log the session and recompiled it for the next 
    attack.  However, the book says, since we can not do the same with 
    Microsoft code, they didn't bother to make an equivalent solution for 
    Windows.  ComLog fills that gap.
    
    Special thanks to Lance Spitzner for his comments and ideas during the 
    improvement of the pre-release.
    
    Targeted audience
    
    This document is presented to anyone who has interests in computer 
    security, honeynets, trojan horse, keyloggers, network administration and 
    computing in general.
    
    1. Introduction
    2. Purpose of the program
    3. How it works
    4. Things you need to know
    5. To Install
    6. Conclusion
    Appendix A. Source code
    Appendix B. Sample session history (console)
    Appendix C. Sample session history (history.txt)
    
    1. Introduction
    
    One of the most common ways that Windows machines are hacked is through the 
    command prompt.  This is not surprising, since this is the shell 
    environment on the Win32 platform, it permits the execution of commands 
    without the need of a GUI, and is easily accessed by executing one single 
    file, either command.com on Win 9x/Me, or cmd.exe on Win NT/2K.  While this 
    file in itself is not a problem, the thing is that it can be accessed by 
    many ways by anyone who can guess where is located this file, and most of 
    the time, it is always in the same places, because the installations were 
    made by default.  This is how most of IIS exploits work, they rely on a 
    Unicode encoding or a special format string to fool the server into 
    accessing files outside of the web directory and make a renamed copy of 
    cmd.exe (because IIS filters commands sent directly to cmd.exe, thanks 
    Microsoft for the great security), and then simply issues command to this 
    copy via a URL in their web browser.  Code Red and Nimda were doing that 
    automatically.
    
    Also, admins may want to keep an eye on what's going on in the DOS prompts 
    on their users workstations, as the inside threat remains the greatest 
    concern for computer security.  I mean, what could secretaries, accountants 
    and other office employees be doing in a DOS prompt that is related to 
    their job?  I wouldn't go as far as to remove their access to the command 
    prompt, besides it may be needed from time to time by support people.  But 
    right now, how can you tell if one of the employees on your network is not 
    doing a ping scan or is netcating his way to the restricted file server and 
    making password dumps?
    
    
    -----------------------------------------------------------------
    This list is provided by the SecurityFocus ARIS analyzer service.
    For more information on this free incident handling, management 
    and tracking system please see: http://aris.securityfocus.com
    



    This archive was generated by hypermail 2b30 : Thu Aug 15 2002 - 05:01:47 PDT