{
    "pages": [
        {
            "page_number": 1,
            "text": " \n  \n"
        },
        {
            "page_number": 2,
            "text": " \nThe International Handbook of Computer Security\n \n \n \n \n \n \n \nJae K. Shim, Ph.D. \nAnique A. Qureshi, Ph.D., CPA, CIA \nJoel G. Siegel, Ph.D., CPA\n \n \n \n \n \n \n \n \n \n \n \n \n \n \nThis book is available at a special discount when ordered in bulk quantities. For information, contact \nSpecial Sales Department, AMACOM, a division of American Management Association, 1601 \nBroadway, New York, NY 10019.\n \n \n \n \n \n \nThis publication is designed to provide accurate and authoritative information in regard to the subject \nmatter covered. It is sold with the understanding that the publisher is not engaged in rendering legal, \naccounting, or other professional service. If legal advice or other expert assistance is required, the \nservices of a competent professional person should be sought.\n \n"
        },
        {
            "page_number": 3,
            "text": " \n \n \n \n © 2000 The Glenlake Publishing Company, Ltd. \n \n \n \n \n All rights reserved. \n \n \n \n \n Printed in the United Stated of America \n \n \n \n \n ISBN: 0-8144-0579-7 \n \n \n \n \n \nThis publication may not be reproduced, stored in a retrieval system, or transmitted in whole or in \npart, in any form or by any means, electronic, mechanical, photocopying, recording, or otherwise, \nwithout the prior written permission of the publisher.\n \n \n \n \n \n \nAMACOM \nAmerican Management Association \nNew York • Atlanta • Boston • Chicago • Kansas City •\nSan Francisco • Washington, D.C. \nBrussels • Mexico City • Tokyo • Toronto\n \n \n \n \n \n Printing number \n \n \n \n \n 10 9 8 7 6 5 4 3 2 1 \n \nDedication \n \n \n \n \n \n \nChung Shim \nDedicated Wife \n \nShaheen Qureshi \nLoving Wife \n \nAqsa Qureshi \nWonderful Daughter \n \n"
        },
        {
            "page_number": 4,
            "text": " \nRoberta Siegel \nLoving Wife, Colleague, and Partner\n \nAcknowledgements \n \n \n \n \n \nWe express our deep appreciation to Barbara Evans for her exceptional editing efforts. Special thanks \ngo to Jimmy Chang, microcomputer consultant at Rand Corporation in Santa Monica for coauthoring \nChapters 3 and 4, to Allison Shim for her word processing work, and to Roberta Siegel for \ncontributing her expertise in computer security.\n \n \n \n \n \n We acknowledge with great appreciation the advice and suggestions of Dr. John Walker, CPA, an \ninternationally recognized leading expert on computer security.\nTable of Contents \n \n \n \n \n About the Authors \n \n \n \n \n \nix \n \n \n \n \n What This Book Will Do for You \n \n \n \n \n \nxi \n \n \n \n \n Chapter 1—Organizational Policy \n \n \n \n \n \n1 \n \n \n \n \n Chapter 2—Physical Security and Data Preservation \n \n \n \n \n \n11 \n \n \n \n \n Chapter 3—Hardware Security \n \n \n \n \n \n33 \n \n \n \n \n Chapter 4—Software Security \n \n \n \n \n \n67 \n \n \n \n \n Chapter 5—Personnel Security \n \n \n \n \n \n109 \n \n"
        },
        {
            "page_number": 5,
            "text": " \n \n \n Chapter 6—Network Security \n \n \n \n \n \n117 \n \n \n \n Appendix 6.A—Commercial Firewalls \n \n \n \n \n \n145 \n \n \n \n Appendix 6.B—Firewall Resellers \n \n \n \n \n \n153 \n \n \n \n Appendix 6.C—Public Domain, Shareware, etc. \n \n \n \n \n \n163 \n \n \n \n \n Chapter 7—Security Policy \n \n \n \n \n \n165 \n \n \n \n Appendix 7.A—Sources of Information Security Policies \n \n \n \n \n \n178 \n \n \n \n Appendix 7.B—Sample Computer Policy \n \n \n \n \n \n179 \n \n \n \n \n Chapter 8—Contingency Planning \n \n \n \n \n \n191 \n \nAppendix 8.A—Business Impact Analysis Worksheet \n \n \n \n \n \n213 \n \n \n \n Appendix 8.B—Communications Assessment Questionnaire \n \n \n \n \n \n215 \n \n \n \n Appendix 8.C—Insurance Recovery Program \n \n \n \n \n \n217 \n \n \n \n Appendix 8.D—Making an Insurance Claim \n \n \n \n \n \n219 \n \n"
        },
        {
            "page_number": 6,
            "text": " \n \n \n Chapter 9—Auditing and Legal Issues \n \n \n \n \n \n221 \n \n \n \n \n Appendix—Security Software \n \n \n \n \n \n235 \n \n   \n \nAbout the Authors \n \n \n \n \n \nJae K. Shim, Ph.D., is professor of business administration at California State University, Long \nBeach. Dr. Shim received his MBA and Ph.D. degrees from the University of California at Berkeley. \nFor over 20 years a consultant on information systems development and computer applications, he is \nnow president of the National Business Review Foundation, a management and computer consulting \nfirm. Dr. Shim has more than 50 books to his credit and has published some 50 articles in \nprofessional journals, including the Journal of Systems Management, Financial Management, the \nJournal of Operational Research, Omega, Data Management, Management Accounting, Simulation \nand Games, Long Range Planning, the Journal of Business Forecasting, Decision Sciences, \nManagement Science, and Econometrica.\n \n \n \n \n \n \nIn 1982 Dr. Shim received the Credit Research Foundation Outstanding Paper Award for one of his \narticles on financial modeling. He has also received a Ford Foundation Award, a Mellon Research \nFellowship, and an Arthur Andersen Research Grant.\n \n \n \n \n \n \nAnique Qureshi, Ph.D., CPA, CIA, is associate professor of accounting and information systems at \nQueens College of the City University of New York. He is an expert in computer applications, \nespecially those related to the World Wide Web. Dr. Qureshi has written two books for Prentice-Hall \nand has contributed chapters to books published by both Prentice-Hall and McGraw-Hill. His articles \nhave appeared in Accounting Technology, the CPA Journal, Management Accounting, the National \nPublic Accountant, and Internal Auditing.\n \n \n \n \n \n \nJoel G. Siegel, Ph.D., CPA, is a consultant to businesses on computer applications and professor of \naccounting, finance, and information systems, Queens College of the City University of New York. \nHe was previously associated with Coopers and Lybrand, CPAs, and Arthur Andersen, CPAs. He has \nserved as consultant to numerous organizations including Citicorp, ITT, and the American Institute of \nCertified Public Accountants (AICPA). Dr. Siegel is the author of 60 books, published by Glenlake \nPublishing, the American Management Association, Prentice-Hall, Richard Irwin, McGraw-Hill, \nHarperCollins, John Wiley, Macmillan, Probus, International Publishing, Barron's, and AICPA. He \nhas written over 200 articles on business topics, many on computer applications to business. His \narticles have appeared in such journals as Computers in Accounting, Financial Executive, Financial \nAnalysis Journal, the CPA Journal, National Public Accountant, and Practical Accountant. In 1972, \n"
        },
        {
            "page_number": 7,
            "text": "he received the Outstanding Educator of America Award. Dr. Siegel is listed in Who's Who Among \nWriters and Who's Who in the World. He formerly chaired the National Oversight Board.\nWhat This Book Will Do for You \n \n \n \n \n \nComputers are an integral part of everyday operations. Organizations depend on them. A computer \nsystem failure will have a critical impact on the organization. Potential vulnerabilities in a computer \nsystem that could undermine operations must therefore be minimized or eliminated.\n \n \n \n \n \n \nThe International Handbook of Computer Security is written primarily to help business executives \nand information systems/computer professionals protect their computers and data from a wide variety \nof threats. It is intended to provide practical and thorough guidance on a wide range of computer \nsecurity issues, emphasizing practical guidance rather than theory. Topics discussed include company \nsecurity policies, physical security, data preservation, hardware and software security, personnel \nsecurity, network security, contingency planning, and legal and auditing issues.\n \n \n \n \n \n \nSecurity concerns have heightened in recent years. You've probably seen news stories about \ncomputer data errors, thefts, burglaries, fires, and sabotage. Moreover, the increased use of \nnetworked computers, including the Internet, Intranets, and Extranets, has had a profound effect on \ncomputer security. The greatest advantage of remote access through networks—convenience—is \nwhat makes the system more vulnerable to loss. As the number of points from which a computer can \nbe accessed increases, so does the threat of attack.\n \n \n \n \n \n \nThe major steps in managing computer security are discussed in this book. We help you as a business \nexecutive identify resources in your own organization that need to be protected. Sometimes, thinking \ninformation is not valuable to anyone else, your organization may not be willing to take security \nprecautions.\n \n \n \n \n \n \nThis is a serious mistake. Hackers often steal or destroy private or confidential data simply because \nit's there! Other hackers may delete or destroy files in an attempt to cover their illegal activity. You \nneed a comprehensive security plan in your organization; a casual attitude towards computer security \nis never justified.\nWe also analyze the costs and benefits of various security safeguards. Cost includes not only the direct \ncost of a safeguard, such as equipment and installation costs, but also the indirect costs, such as \nemployee morale and productivity losses.\n \n \n \n \n \n \nIt's important to recognize that increasing security typically results in reduced convenience. \nEmployees may resent the inconvenience that accompanies security safeguards. And indeed, too \nmuch security can be just as detrimental as too little. You'll need to find a balance.\n \n \n"
        },
        {
            "page_number": 8,
            "text": " \n \n \n \nWe cannot over-emphasize the importance of contingency planning. If security is violated, how do \nyou recover? What are the legal consequences? What will be the financial impact? In planning \ncomputer security policies and financial support, be sure to perform a risk analysis.\n \n \n \n \n \n \nComputer security risks fall into three major categories: destruction, modification, and disclosure. \nEach may be further classified into intentional, unintentional, and environmental attacks. One threat \ncomes from computer criminals and disgruntled employees who intend to defraud, sabotage, and \n''hack.\" Another comes from computer users who are careless. A final threat comes from the \nenvironment; your organization must protect itself from disasters like fire, flood, and earthquakes. An \neffective security plan must consider all these types of threats.\n \n \n \n \n \n \nWe do not neglect insurance. What is the company's risk exposure? Your insurance policies should \ncover such risks as theft, fraud, intentional destruction, and forgery, as well as business interruption \ninsurance to cover additional expenses and lost profits during downtime.\n \n \n \n \n \n \nThroughout this book, we provide extensive examples to illustrate practical applications, and answers \nto common questions. Checklists, charts, graphs, diagrams, report forms, schedules, tables, exhibits, \nillustrations, and step-by-step instructions are designed to enhance the handbook's practical use. The \ntechniques we spell out can be adopted outright or modified to suit your own needs.\nChapter 1— \nOrganizational Policy \n \n \n \n \n \nToday the cost to businesses of stolen, misused, or altered information can be high, especially if real \nor purported damages to customers can be traced back to mismanagement. That's why you must \nvalue your information resources within the context of your business goals and constraints.\n \n \n \n \n \n \nThe objective of security management is to eliminate or minimize computer vulnerability to \ndestruction, modification, or disclosure. But before we can discuss information security, we must see \nhow that security works.\n \n \n \n \n \n \nA key consideration is the physical location of the organization. Naturally, more security is needed in \nareas of high crime, although this may take the form of less expensive generic physical security \nmeasures. Who uses the information will also affect the security measures chosen. Some users need \nto alter data; others simply need to access it.\n \n \n \n \n \n If a security plan is to be effective, top management must be fully convinced of the need to take \ncounteractive steps. To assess the seriousness of a computer breakdown or loss of data, each business \n \n"
        },
        {
            "page_number": 9,
            "text": "has to evaluate threats to the company, the potential losses if the threats are realized, and the time and \ncost that will be necessary to recover from any breach in security.\n \n \n \n \n \nThe proliferation of networks scatters security issues across the globe and increases the need for \ninexpensive but effective levels of security. Physical security measures reflect the location of each \ncomponent, but procedural measures, especially in a large organization, though they may seem \nobtrusive are of equal importance.\n \n \n \n \n \n \nPersonal computers are another potential security threat. More and more people operate their PCs \nwith telecommunications services to connect to central computers and network services. To limit the \ndamage that can be done, each user must be identified and that identity authenticated. The user is then \nallowed to perform only authorized actions.\nAudits can be very valuable for detecting security violations and deterring future violations. A security \nviolation may be indicated from customer or vendor complaints that show discrepancies or errors; on \nthe other hand, variance allowances can cover up fraudulent activity.\n \n \n \n \n \n \nAudit trails used to produce exception reports are especially valuable to managers. Standard \nquestions include who accessed what data, whether the data were altered, or whether access-only \nemployees attempted alteration. Exception reports are best used daily because they are after-the-fact \nreports. You may also choose to look only at reports from areas of high vulnerability or where there \nis a history of corruption or attempted corruption.\n \n \n \n \n \n \nA good manager will know the types and forms of information generated and how the information is \nused by the business before planning how to manage it. Security measures in an information \nresource management program must be practical, flexible, and in tune with the needs of the business. \nA risk-management approach recognizes alternatives and decision choices at each step in \ninformation resources management in order to develop a program that meshes with ongoing business \npractices.\n \n \n \n \n \n \nIt is your responsibility as a manager to (1) assist with the design and implementation of security \nprocedures and controls, and (2) ensure that these remain effective by continuous internal audits. To \ndo this you must:\n \n \n \n \n \n • Identify the risks. \n \n \n \n \n • Evaluate the risks. \n \n \n"
        },
        {
            "page_number": 10,
            "text": "  • Install appropriate controls.  \n \n \n \n \n • Prepare a contingency plan. \n \n \n \n \n • Continually monitor those controls against the plan. \n \n \n \n \n \nMisuse of information is costly. Ask yourself, \"Where in the business scheme does this information \nwork?\" identifying not only the department but also the type of usage (strategic, tactical, operational, \nor historical). This will help you determine how secure that information must be. Its value must \njustify the expense of protecting business data. For instance, because encryption is relatively \nexpensive, it's usually reserved for higher business use (strategic or tactical). Operational business \nuses may use simpler controls such as passwords.\nSecurity Administration \n \n \n \n \n \nSecurity should be administered in the context of how the organization needs to control, use, and \nprotect its information. Protection needs to be appropriate and reasonable given management's risk \nposture. Three levels of security (physical, procedural, and logical) used in tandem can reduce the \nrisks.\n \n \n \n \n \n Physical Security  \n \n \n \n \n \nPhysical security, the first line of defense, is the one that usually comes to mind when you hear the \nword \"security.\" This level literally separates those who are authorized to use certain types of \ninformation from those who are not. It also creates and maintains an environment in which the \nequipment is not exposed to damaging environment hazards like extreme heat or flooding, natural \ndisasters, fire, power failure, or air conditioning failure.\n \n \n \n \n \n \nDetection devices warn of an environmental failure, and automatic systems can protect against \ndamages. Heat and smoke sensors and thermostats for temperature and humidity are standard \nequipment in computer centers. Attached to automatic shutoff devices they protect your computer \nsystem should critical limits be exceeded. Some natural disasters cannot be foreseen, especially in the \nusually windowless domain of the computer center, but disruption of service can be kept to a \nminimum by using backup centers.\n \n \n \n \n \nAt backup centers themselves, physical security takes on a heightened purpose. Your company may \nwant to join a data center insurance group. The group data center should be able to handle the total \n"
        },
        {
            "page_number": 11,
            "text": " \nworkload of each member organization; in the event of service failure, the data center assumes the \ndata processing role for that organization. During regular operations the data center may be used by a \nthird party.\n \n \n \n \n \n \nHuman control is more elusive. Traffic, especially at the beginning and end of the business day, can \noverburden card-access systems. The physical layout of the building and the routes employees use to \nreach their workplaces can also overburden checkpoints. Guards, usually low-paid, are susceptible to \nbribery and relaxation of standards. Additionally, during high traffic times there may not be enough \nguards to check employee ID badges, or register visitors.\n \n \n \n \n \n Procedural Security  \n \n \n \n \n \nDaily users of information systems gain great insight into their workings. They can identify holes in \nthe process. Employees generally know if their system is being audited (as they should, to discourage \ncorruption); if they are not being audited, the temptation to tamper with the system may be too great \nto resist. Companies with high turnover are particularly susceptible to employee modifications of the \nsystem.\nCareful hiring and processing of employees, then, is one way to instill procedural security. Threats \nfrom mentally unstable employees are obvious. However, without the proper safeguards all current and \nformer employees have access to the company's computer resources. Among the proper safeguards:\n \n \n \n \n \n • Revoke passwords as soon as an employee is terminated or if he is even suspected of infringement. \n \n \n \n \n • Use lists of authorized personnel to control entrance into the system. \n \n \n \n \n • Constantly monitor logs generated by computer systems that report access to sensitive areas. \n \n \n \n \n • All transactions processed should be reviewed and audited. \n \n \n \n \n \nThese actions constitute a fundamental level of control over business operations that lets the whole \norganization know that management is concerned with security and is devoting time and money to \nseeing that its security objectives are met.\n \n \n \n \n \n Logical Security  \n \n"
        },
        {
            "page_number": 12,
            "text": " \n \n \n Computer hardware or software should automatically control the people and programs trying to \naccess computer resources. Data encryption is an example.\n \n \n \n \n \n Generally, all three levels of security must be combined to form the right mix for a given element. \nThis is called an access control system. Its goals are to:\n \n \n \n \n \n • Prevent unauthorized physical or logical access to facilities or to information via electronic formats, \n \n \n \n \n • Track user computing and telecommunication activities, and \n \n \n \n \n • Establish a basis for, and then enforce, a set of authorizations for all persons and programs \nattempting to use electronic information resources.\nEstablishing a Security Policy \n \n \n \n \n \nEvery organization should have a security policy that defines the limits of acceptable behavior and \nhow the organization will respond to violations of such behavior. The policy assigns accountability \nand delegates authority across the organization. It will naturally differ from organization to \norganization, based on unique needs. Optional policies include:\n \n \n \n \n \n • No playing of computer games on corporate computers. \n \n \n \n \n • No visiting adult web sites using corporate Internet accounts or computers. \n \n \n \n \n • An embargo against the use of a specific protocol if it cannot be administered securely. \n \n \n \n \n • A prohibition against taking copies of certain corporate electronic documents out of the office. \n \n \n \n \n • No use of pirated software. \n \n \nQuestions you must answer include: How will violators be reprimanded or punished? Will the \n"
        },
        {
            "page_number": 13,
            "text": "  organization respond to violators inside the organization? Will it be different from the response to \nviolators outside the organization? What civil or criminal actions might be taken against violators?\n  \n \n \n \n \n Security policy should not be set piecemeal. This leads to inefficiencies, holes in the system, poor \nvaluation of information elements, and inconsistencies. And it costs more to set policy piecemeal.\n \n \n \n \n \n Publishing the policy is vital. \n \n \n \n \n \nThe owners of information can best assign information elements to a particular classification. Top \nmanagement is in the best position to evaluate consequences. About 1 percent of all business \ninformation should have the highest level (and therefore costliest) classification. Mid-range \nclassifications typically have about 40 percent of all business information.\n \n \n \n \n \n \nPolicy statements set program goals, give detailed directions for carrying out procedures, and explain \nabsolute requirements of the information security system. Policy statements should be concise and \nnot require modification for at least five years; standards or procedures usually must be modified no \nmore often than every three years.\n \n \n \n \n \n \nYour security policy should be a broad statement that guides individuals and departments as they \nwork to achieve certain goals. Specific actions needed to realize goals will be contained in supporting \nstandards rather than in the policy document.\nThe security policy should be concise and to the point, generally not exceeding 10 pages. It should be \neasy to understand. It should emphasize the roles of individuals and departments. It is not the purpose \nof the security policy to educate individuals. That objective is better achieved through training.\n \n \n \n \n \n \nThe rationale for a security policy should be stated, explaining its purpose, including why data \nintegrity must be maintained. Come down hard on the importance of maintaining the confidentiality \nand privacy of information resources. The organization must have information continuously \navailable; any interruption can have serious financial consequences.\n \n \n \n \n \n \nComputer security must be everyone's responsibility, so the computer security policy should \nencompass all locations of the company and all of its subsidiaries. Because security is only as strong \nas its weakest link, everyone in the organization must be held to the same set of standards. This \nmeans that the standards have to be flexible enough to be used in a wide variety of circumstances \nwhile remaining consistent across the organization.\n \n \n \nThe security policies apply to all data and computer facilities, including standalone computers, \n"
        },
        {
            "page_number": 14,
            "text": "  \nInternet and Intranet sites, local area networks (LANs), and wide area networks (WANs), as well as \nall forms of electronic communication, including email, fax, and data transmissions. They should \nalso encompass relevant printed material, such as documentation and technical specifications.\n  \n \n \n \n \n \nComputer security is a means to an end, not an end in itself; it is an integral component of your \norganization's overall risk management strategy. It should therefore be evaluated periodically to \nrespond to changes in technology or circumstances. Assign authority for issuing and amending the \nsecurity policy to a committee such as the Information Technology Management Committee that \nmust determine when circumstances justify departure from the policy. All exceptions must have \ncommittee approval.\n \n \n \n \n \n \nFor a security policy to proceed, all individuals and departments must participate. It is well \nestablished that individuals are more likely to accept the security policy (or any other policy!) if they \nhave had input during its creation, but the real benefit of employee participation is the knowledge \nthey bring.\n \n \n \n \n \n \nThe relationship between the computer security policy and other corporate policies should be spelled \nout. For example, the computer security policy should be used in conjunction with the firm's policies \nfor the internal control structure and contingency plans, including business interruption and \nresumption plans.\nThe policy should ensure compliance with all laws. Privacy and confidentiality issues have a serious \neffect on computer security. Increased governmental regulation is likely. The legal department should \nhelp department heads comply with the laws.\n \n \n \n \n \n The responsibilities of the Information Systems department and its security personnel should be \ndefined in the security policy document. These responsibilities might be to:\n \n \n \n \n \n • Be responsible for all computer networks and communications. \n \n \n \n \n • Provide systems development methodology for security needs. \n \n \n \n \n • Ensure that security personnel have the training and skills to perform their duties. \n \n \n \n \n • Provide computer security assistance to other departments. \n \n \n"
        },
        {
            "page_number": 15,
            "text": "  • Be responsible for all cryptographic methods and keys.  \n \n \n \n \n • Manage virus detection software for both networked and standalone computers. \n \n \n \n \n • Acquire hardware or operating systems as needed. \n \n \n \n \n • Authorize the use of networks. \n \n \n \n \n • Review, evaluate, and approve all contracts related to information systems. \n \n \n \n \n For personal computer systems, the security policy should address additional precautions; for \ninstance:\n \n \n \n \n \n • All original data should be backed up regularly. \n \n \n \n \n • Virus detection software must always be used on PCs, especially before copying data or programs \nonto the network.\n \n \n \n \n \n \n• Certain types of confidential or important data should never be stored on a local hard drive; instead \nsuch data should be stored on the network, or on floppy or compact disks or a removable hard drive, \nso that it may be stored in a secure place.\n \n \n \n \n \n • Standards should be established for remote access. \n \n \n \n \n \n• PCs should not be directly connected to the Internet, since the Internet is a source of both virus \ninfections and hackers. Internet access should be only through the company's Internet server, which \ncan protect itself.\nAdditional policy components can include the policies regarding the hiring, performance, and firing of \ninformation workers, though they should not be overly specific.\n \n \n \n \n \n Security should be continuous in all situations, and not limited to protecting against intentional \n \n"
        },
        {
            "page_number": 16,
            "text": "attacks. The board of directors should write a clear statement of security intention, including:\n \n \n \n \n • Definitions of behaviors that will be tolerated or that will result in disciplinary action or dismissal, \n \n \n \n \n • Standards of protection necessary at every company location, and \n \n \n \n \n • Allocation of responsibility to one person (ideally) or to a group, with the authority to carry out the \npolicy, set budgets, and approve objectives.\n \n \n \n \n \n The Security Administrator \n \n \n \n \n The security administrator sets policy, subject to board approval. He also investigates, monitors, \nadvises employees, counsels management, and acts as a technical specialist.\n \n \n \n \n \n \nThe security administrator establishes the minimal fixed requirements for information classification \nand the protection each classification needs in terms of physical, procedural, and logical security \nelements. He assigns responsibilities to job classifications and explains how to manage exceptions to \npolicy.\n \n \n \n \n \n \nThe security administrator advises other information security administrators and users on the \nselection and application of security measures, giving advice on how to mark (written and electronic \n\"stamps\") and handle processes, select software security packages, train security coordinators, and \nsolve problems.\n \n \n \n \n \n \nThe security administrator investigates all computer security violations, advises senior management \non matters of information resource control, consults on matters of information security, and provides \ntechnical consultation for business activities.\n \n \n \n \n \n Finally . . . \n \n \n \n \nFinding and keeping qualified employees requires a large cash outlay, especially when qualified \nindividuals are scarce. Computer security will depend partly on how well those employees are \nsupervised and motivated. One theory is that employees who know that their company values its \n"
        },
        {
            "page_number": 17,
            "text": "Security for system components should be commensurate with their value to the business. Total security \nis not possible; even attempting it would be prohibitively costly, as well as overly burdensome to users. \nTherefore, top management should be aware of the varying risks of computer information loss or \nmodification. They should be part of the design and implementation of the security policy, with the \nsecurity administrator reporting directly to senior management. \n \nsecurity, reviews its practices, alters faulty programs, and punishes wayward employees as well as \noutsiders will be less likely to commit fraud and more likely to report it. \n  \nChapter 2— \nPhysical Security and Data Preservation \n \n \n \n \n \nThe first line of defense for a computer system is to protect it physically: the plant, the equipment, \nand the personnel. Physical security protects the data, its integrity, accuracy, and privacy. An \neffective physical security system will prevent a security failure. However, should a system be \nsuccessfully attacked, it should create an audit trail for investigators.\n \n \n \n \n \n \nComputer equipment is at higher risk if it is easily accessible by the public or in a high crime area. \nAnd, of course, sometimes people authorized to be on your premises steal. The cost of theft can be \nvery significant, far higher than the replacement price of the stolen equipment, because the company \nmay also lose valuable data, especially if your work has not been properly backed up.\n \n \n \n \n \n Computer Facilities \n \n \n \n \n \nIn the past, when computing tended to be centralized, it was easier to label a structure as the \n''computer center.\" With distributed computing, that is no longer possible. All areas where computing \nis done and from where an attack may be launched are vulnerable. Unauthorized access to computer \nfacilities should be restricted through the use of surveillance equipment.\n \n \n \n \n \n \nFacilities should be designed to protect computers, taking into account environmental factors like \nheating, cooling, dehumidifying, ventilating, lighting, and power systems. For example, the ducts of \nair conditioning units should be secured against access with heavy-gauge screens.\n The following safeguards help protect computer facilities from both accidents and disasters like fire \nand floods:\n \n \n \n \n \n • Adequate emergency lighting for safe evacuation in case of fire or other disaster. \n \n"
        },
        {
            "page_number": 18,
            "text": " \n \n \n • Fireproof containers to protect media (disks, tapes, or other output). \n \n \n \n \n • User manuals for equipment and software to maintain continuity of proper operations. \n \n \n \n \n • Surge protectors to protect the computer system against power line disturbances. \n \n \n \n \n As computers become smaller, they can be housed in smaller areas and this changes the way facilities \nare designed. The layout of computer facilities is important in planning for computer security.\n \n \n \n \n \n Central computer facilities should be housed near wire distribution centers but away from junctions \nof water or steam pipes. The room should be sealed tightly to minimize smoke or dust from outside.  \n \n \n \n \n Wire management is simple with multilevel computer racking furniture, which offers space flexibility \nand which is available from several suppliers:\n \n \n \n \n \n • ACS Computer Network Racking Systems (http://ourworld.compuserve.com/JLukach/) \n \n \n \n \n • Ergonomic Workstations Ltd. (http://www.ergo-ws.com/) \n \n \n \n \n • Information Support Concepts (http://www.iscdfw.com/) \n \n \n \n \n • LANSTAR (http://lanstur.com/) \n \n \n \n \n • Page Concepts (http://www.pagec.com/) \n \n \n \n \n • PC Innovations, Inc. (http://www.pcinnov.com/) \n \n \n \n \n • Salix Group (http://www.salixgroup.com/) \n \n"
        },
        {
            "page_number": 19,
            "text": " \n \n \n • Stacking Systems, Inc. (http://www.stackingsystems.com/) \n \n \n \n \n • Systems Manufacturing Corp. (http://www.smcplus.com) \n \n \n \n \n • Workstation Environments (http://www.workenv.com/) \n \n \n \n \n Roll-out shelves may be used for quick access to servers. Security cabinets should be used for \ncontrolled access to critical hardware and server systems.\nIf wiring is a concern, cables can generally be run along the walls. Racking shelves generally contain \nmultistage openings for improved access to cables with a wide range of plugs and cable connectors.\n \n \n \n \n \n \nAluminum channels or I-beams can be used to raise components and cabinets if there is danger of \nflooding. Placing network equipment next to processing equipment can save cabling costs. Smaller \ncomponents may be stacked vertically to conserve floor space and reduce cable costs. The Salix \nGroup, for example, offers Spectro Data for networks; it is not limited by layout size and can be used \nfor a high-capacity four-level configuration.\n \n \n \n \n \n \nMultilevel units are cost-effective, and if they are ergonomically designed, productivity increases. \nThe main work surface should provide vibration-free areas for screen, keyboard, and digitizing \npalette, with additional workspace for accessing other documents and equipment.\n \n \n \n \n \n \nAmericon (Stacking Systems, Inc.), for instance, offers server cabinetry for both active monitoring \nand closet environments. Its Network Solutions cabinetry may be used when floor space is at a \npremium. Its LAN Manager consoles allow for multiple stacking of servers, monitors, keyboards, \nand mice, along with desk surfaces and storage space. The LAN Commander cabinets contain these \nsecurity features:\n \n \n \n \n \n • Lock-in suspension glide shelving \n \n \n \n \n • Seismic strapping for servers \n \n \n \n \n • 180-degree rotating doors for access to both sides of the server \n \n"
        },
        {
            "page_number": 20,
            "text": " \n \n \n • Whisper-cool exhaust fans \n \n \n \n \n • Heavy rated casters for moving from place to place \n \n \n \n \n • Movement stabilization once the cabinet has been spotted \n \n \n \n \n • Rear access through sliding doors \n \n \n \n \n Optional accessories include: \n \n \n \n \n • Remote access for consoles as far away as 250 feet \n \n \n \n \n • Pullout server shelves \n \n \n \n \n • EIA rack mounts for Ethernet equipment \n \n \n \n \n • Induction fans for cooling when not on a raised floor \n \n \n \n \n \nWorkspace Resources (http://www.workspace-resources.com) provides design and marketing services \nfor the office and contract furniture industry. It coordinates the needs of businesses with the \ncapabilities of furniture manufacturers.\nEnvironmental Considerations* \n \n \n \n \n Computer facilities are susceptible to damage from a variety of environmental factors: \n \n \n \n \n \n• Heat can cause electronic components to fail. Air conditioning is generally essential for reliable \noperation. Take simple precautions to ensure that air can circulate freely. Backup power should be \navailable to air conditioning the computer system even if the primary power fails.\n \n \n"
        },
        {
            "page_number": 21,
            "text": " \n \n \n \n• Water is an obvious enemy of computer hardware. Floods, rain, sprinkler system activity, burst \npipes, etc., can do significant damage. Check that water pipes are routed away from computer \nfacilities. Instead of a traditional sprinkler system, consider using a less potentially harmful fire-\nextinguishing agent.\n \n \n \n \n \n \n• Humidity at either extreme is harmful. High humidity can lead to condensation, which can corrode \nmetal contacts or cause electrical shorts. Low humidity may permit the buildup of static electricity. \nThe floors of computer facilities should either be bare or covered with anti-static carpeting. Monitor \nhumidity continuously to keep it at acceptable levels.\n \n \n \n \n \n \n• Dust, dirt, and other foreign particles can interfere with proper reading and writing on magnetic \nmedia, among other problems. Personnel should not be allowed to eat or drink around computers. \nThe air should be filtered and the filters replaced regularly.\n \n \n \n \n \n \n• Power failure can render all equipment useless. Brownouts and blackouts are the most visible sign \nof power failure. However, voltage spikes, which can cause serious damage, are much more \ncommon. Spikes like those produced by lightning may either damage equipment or randomly alter or \ndestroy the data. A drop in line voltage can also lead to malfunction of computer equipment. Voltage \nregulators and line conditioners should be used if electricity fluctuates. Think about installing an \nuninterruptible power supply.\n \n \n \n * Shim et al, Information Systems Management Handbook (N.J.: Prentice-Hall, 1999).\nMaintenance and Preventive Care  \n \n \n \n \n \nRegular maintenance can help prevent the unexpected downtime that can be caused by the weather \nand other environmental factors. Run diagnostic programs as part of regular maintenance and keep a \nmaintenance log. You can quickly identify recurring problems by scanning the logs. At a minimum, \nlog the following information:\n \n \n \n \n \n • Type of equipment serviced \n \n \n \n \n • Manufacturer and identification number of equipment serviced \n \n \n \n \n • Date of service \n \n"
        },
        {
            "page_number": 22,
            "text": " \n \n \n • Services performed, including the results of diagnostic tests \n \n \n \n \n • A note indicating whether the service was scheduled or not \n \n \n \n \n \nComputer areas should be kept cleaned and dusted, with no eating, drinking, or smoking allowed. Set \nup programs to train your personnel in proper handling of computer equipment, peripherals, magnetic \nmedia, and CD-ROMs, reminding them of basic things like not putting magnetic media near \ntelephones, radios, or other electric equipment, and writing labels before placing them on disks.\n \n \n \n \n \n \nSet up a regular cleaning schedule for computers and peripheral equipment, and use cleaning \nproducts recommended by the manufacturer. Never spray electrical equipment directly with cleaning \nliquids. Clean keyboard surfaces with a damp cloth and vacuum with special computer vacuums.\n \n \n \n \n \n \nPrinters need to be cleaned to remove fibers, dust particles, and lint. Magnetic media devices, \nespecially the read/write heads and transport rollers, can be cleaned with commercial products. Dust, \nsmoke, fingerprints, and grease building up on recording surfaces can lead to crashes or permanent \ndamage to the equipment and magnetic media.\n \n \n \n \n \n Simple precautions, such as using static-resistant dust covers, can protect equipment, but never use \nthem when the equipment is in use or it may overheat.\n \n \n \n \n \n Water Alert Systems  \n \n \n \n \n \nWater alert systems should be installed wherever water might damage computer equipment, generally \nin the basement or in floors above the computer systems. Water sensing systems, which are especially \nuseful in protecting electrical cables under the floor, should be installed within suspended ceilings and \ninside water-cooled computer cabinets and process cooling equipment. The water sensors should \nactivate both an alarm and a drainage pump.\nStatic Electricity  \n \n \n \n \n \nStatic electricity results from an excess or deficiency of electrons. An individual can easily become \ncharged to several thousands of volts. While the current from electrostatic discharges is too low to \nharm humans, it can do a lot of damage to electronic equipment.\n \n \n \n"
        },
        {
            "page_number": 23,
            "text": "  \nYou can protect against electrostatic discharges by grounding, shielding, filtering, and limiting \nvoltage. Vinyl flooring is generally better than carpeting to avoid static electricity buildup. Simple \nprecautions can also minimize the dangers, such as:\n  \n \n \n \n \n • Using anti-static sprays \n \n \n \n \n • Grounding computer equipment \n \n \n \n \n • Using anti-static floor and table mats \n \n \n \n \n • Maintaining a proper level of humidity \n \n \n \n \n Humidity Control  \n \n \n \n \n \nHumidity should be tightly controlled. When air is too dry, static electricity is generated. When it is \ntoo high, above 80 percent, there may be problems with electric connections and a process similar to \nelectroplating starts. Silver particles migrate from connectors onto copper circuits, thus destroying \nelectrical efficiency. A similar process affects the gold particles used to bond chips to circuit boards. \nAn optimal relative humidity level is 40 to 60 percent.\n \n \n \n \n \n Wires and Cables \n \n \n \n \n \nIn distributed computing, it's essential to protect the wiring system. Generally there are two options \nfor wires and cables, copper or optical fiber. While fiber optics offer significant performance and \nsecurity advantages, they cost more to install. However, the cost disadvantage rapidly diminishes as \nthe volume of data to be transferred increases.\n \n \n \n \n \n \nFiber optics work by sending light signals along very thin strands of glass or plastic fiber. The fiber's \ncore is surrounded by cladding. The cladding causes the reflections, which guide the light through the \nfiber.\n \n \n \n \n \n Two common types of fiber are multimode and singlemode. Multimode, which has a larger core, is \nused with LED sources for LANs.\n"
        },
        {
            "page_number": 24,
            "text": "Singlemode fiber, which has a smaller core, is used with laser sources. Plastic optical fiber has a much \nlarger core; it uses visible light.\n \n \n \n \n \n Cables and wires are fragile. A buffer coating protects the fiber from damage. Additional protection \nis provided by an outer covering, the jacket.\n \n \n \n \n \n \nIt is not possible to repair damaged wires; they must be replaced. In the process, the electrical \nproperties of cables may be affected, in turn affecting the reliability of the data. Establish alternate \npaths for cables that are critical.\n \n \n \n \n \n \nFiber optics are more secure than copper. It is relatively easy for someone to tap copper lines if they \ncan obtain access to them at any point. Such wiretaps are very difficult to detect. In contrast, it is \nmuch harder and more expensive to tap optical fibers. Moreover, normal operations are disturbed by \na fiber optics tap, which can therefore be detected more easily. Yet even with fiber optics, a skilled \nperson with proper equipment might tap the system undetected, so though fiber optics provides a \ndeterrent to crime, they are not perfectly secure. Of course, the best way to protect sensitive data is to \nuse encryption.\n \n \n \n \n \n Fiber optics are not affected by electrical or magnetic interference. Copper wires have to be shielded \nwith cabling and grounded metal conduits.\n \n \n \n \n \n \nOn the other hand, the ends of all fiber optic cables must be microscopically smooth. They have to be \nexactly aligned and positioned. This requires expensive special equipment and highly trained \npersonnel.\n \n \n \n \n \n An experienced person should certify any data wiring. The person should: \n \n \n \n \n • Perform a visual inspection. \n \n \n \n \n • Check that each cable is connected correctly. \n \n \n \n \n • Check that there are no crossed pairs. \n \n \n"
        },
        {
            "page_number": 25,
            "text": "  • Use a reflectometer to detect if there are any constrictions, bad terminations, or external \ninterference.\n  \n \n \n \n \n Purchase orders for any wiring should specify: \n \n \n \n \n • Who will certify the wiring. \n \n \n \n \n • What equipment will be used to test the wiring. \n \n \n \n \n • What standards will apply. \n \nProtecting Information \n \n \n \n \n The integrity, accuracy, and privacy of data are essential in any organization. Data lacks integrity if \nanything is:\n \n \n \n \n \n • Missing \n \n \n \n \n • Incomplete \n \n \n \n \n • Inconsistent \n \n \n \n \n • Poorly designed (in a database environment) \n \n \n \n \n Data accuracy is not the same as data integrity. Data is accurate if \n \n \n \n \n • It is reliable, and \n \n \n \n \n • The data is what it purports to be. \n \n"
        },
        {
            "page_number": 26,
            "text": " \n \n \n Data privacy requires that only authorized individuals have access to data. \n \n \n \n \n Destroying Data  \n \n \n \n \n \nData that is no longer needed must be destroyed. Information on magnetic media is typically \n\"destroyed\" by overwriting on it. While this appears to destroy the information, there are many \nsubtleties to consider. For example, if the new file is shorter than the old file, information may \nremain on magnetic media beyond the new file's end-of-file marker. Any information beyond that \ncan be easily retrieved. Overwriting the entire medium is safer but time-consuming. Instead, use \nother methods, such as degaussing. Degaussers are essentially bulk erasure devices; when used \nwithin their specifications, they provide adequate protection.\n \n \n \n \n \n \nFormatting a disk does not safely destroy all information. Magnetic media may retain a latent image \nof the preceding bit value after the writer insertion of a new bit value because it is not possible to \ncompletely saturate the magnetization. While normal read/write operations are not affected by this \nlimitation, it does pose a security threat exploitable by anyone with sophisticated equipment.\n \n \n \n \n \n \nPapers and other soft materials, such as microfiche and floppy disks, can be shredded. Some shredders \ncut in straight lines or strips; others cross-cut or produce particles. Some shredders disintegrate \nmaterial by repeatedly cutting and passing it through a fine screen. Others may grind the material and \nmake pulp out of it.\nBurning is another way to destroy sensitive data. As with shredding, burning means that the storage \nmedium can no longer be reused. Yet even with burning, you need to be careful. It's possible using \nspecial techniques, for example, to retrieve printed information from intact paper ashes, even though \nthe information may no longer be visible to the human eye.\n \n \n \n \n \n Controlling Access \n \n \n \n \n Access controls guard against improper use of equipment, data files, and software. The oldest method \nof restricting physical access is with a lock. Locks are of two types, preset and programmable.\n \n \n \n \n \n \nWith preset locks, it's not possible to change the access requirements without physically modifying \nthe locking mechanism. The combination on programmable locks, whether mechanical or electronic, \ncan be more easily changed as security needs change, but their basic problem is that the entry codes \nare often easy for an observer to obtain. To overcome this problem, some electronic locks use a touch \nscreen that randomly varies the digit locations for each user and restrict directional visibility to a \n \n"
        },
        {
            "page_number": 27,
            "text": "perpendicular angle.\n \n \n \n \n \nMake sure there's only one door for access into a secured access, and the entrance should not be \ndirectly from a public place. It should be selfclosing and it shouldn't have a hold-open feature. A \ncombination or programmable lock may be sufficient. Install an alarm system.\n \n \n \n \n \n \nOne development in access control combines security with asset management. For example, it's \npossible to link a laptop with a specific individual and detect when the asset is moved in, out, or \nwithin a facility.\n \n \n \n \n \n Security guards and guard dogs can also be used to restrict access; their physical presence serves as a \ndeterrent.\n \n \n \n \n \n Pre-employment screening and bonding are essential when hiring security guards. Certain states, \nsuch as New York, have mandatory training requirements for guards.\n \n \n \n \n \n \nThe limitations of guards, however, are well-known. They can easily become bored with routine \nwork and may not fulfill their duties as expected. It's easy for someone to forge identification to get \npast a guard. Through procedural error guards may also allow unauthorized individuals access to \nrestricted areas.\n \n \n \n \n \n \nDogs have excellent hearing and a keen sense of smell. Guard dogs can be trained to \"hold\" intruders \ntill security personnel arrive. On the other hand, security dogs mean you'll need additional liability \ninsurance and training and maintaining dogs is expensive. Finally, they generally cannot differentiate \nbetween authorized and unauthorized visitors. \nStill, security is enhanced if guards or dogs patrol the facilities often at random intervals. This \npsychological deterrence lets a potential intruder know that he might be caught. A determined attacker, \nof course, is unlikely to be bothered by psychological deterrents, so guards and dogs should always be \nbacked up through other means.\n \n \n \n \n \n \nSomething as simple as lights can greatly enhance security. Lights make it easier for security \npersonnel to carry out surveillance. Lights also make it harder for intruders to enter the facilities. \nLights may be: left on all the time, put on timer or ambient control, activated by motion detectors, or \nmanually operated.\n \n \n \n \n \n To limit access a security system must be able to discriminate between authorized and unauthorized \nindividuals. The three general discrimination methods are:\n \n"
        },
        {
            "page_number": 28,
            "text": " \n \n \n \n \n• Identification, comparing the physical characteristics of an individual with previously stored \ninformation. Access thus depends on who the person is. It may verify the individual's signature, \npersonnel number, code, voice print, palm print, fingerprint, teeth print, or other personal trait. \nSecondary authentication, such as the user's place of birth, may be required for highly sensitive \ninformation.\n \n \n \n \n \n \n• Users name plus passwords based on some combination of letters or numbers. There should be no \nlogic to the password, so it cannot be easily guessed. Access depends on what the person knows. \nPasswords should be changed regularly; inactive passwords (e.g., more than four months old) should \nbe deleted. When an employee leaves, block his password immediately. If a user changes a password, \nyou'll need controls to prevent use of the old password. Passwords should not be shared. Access \ncontrol software allows a minimum password time period in which a new password cannot be \nchanged or a new password matching an old one will be rejected.\n \n \n \n \n \n \n• Cards/keys. Access can depend on what a person possesses: Cards, keys, badges, etc. Improper \naccess may be signalled by an alarm. Evaluate any unauthorized access pattern. You might want to \nlook into smart cards, in which the user enters both an identification number and a randomly \ngenerated code that changes each time it's used or at stated times.\nComputer and terminal access controls include: \n \n \n \n \n • Automatic shut-off: The system signs off the user if the user fails to sign off after a transmission is \ncompleted.\n \n \n \n \n \n • Call-back: A phone call is made to the terminal site to verify the user's identity before access is \ngranted.\n \n \n \n \n \n • Time lock: Access is denied to the system during specified hours, such as after normal business \nhours.\n \n \n \n \n \n \nWithin the plant, areas containing sensitive data should be accessible only to authorized personnel. \nThese areas, including the computer rooms, should have only a single entry door that can be operated \nby an encoded magnetic ID card or by physical controls, such as a librarian keeping a log. A lockout \nshould result from repeated errors.\n \n \n \nYour logs should be automatic; they should list ID number, time of access, and function performed. \nData dictionary software provides an automated log of access to software and file information. Use \n"
        },
        {
            "page_number": 29,
            "text": "  intrusion detection devices like cameras and motion detectors to monitor sensitive areas for the \npresence of unauthorized individuals.\n  \n \n \n \n \n \nAre your people diligently honoring the controls you're set up over processing, maintaining records, \nand file or software modification? Each individual function (e.g., accounts receivable, payroll) may \nrequire its own password so that users have access only to limited areas. The computer can keep an \ninternal record of the date and time each file was last updated to compare against the log. The hours \nto access key files can be limited to prevent unauthorized access after normal working hours.\n \n \n \n \n \n \nFiles should be assigned different levels of confidentiality and security, such as Top Secret, \nConfidential, Internal Use Only, and Unrestricted. Confidential information should not be displayed \non computer screens.\n \n \n \n \n \n \nTo control access to sensitive data, map access requirement to system components based on job \nfunction, with an appropriate segregation of duties. Temporary employees should be restricted to a \nspecific project, activity, system, and time period. If you want to avoid possible data manipulation, \ndon't give programmers free access to the computer area or the library. Keep those important disks \nlocked up.\n Hardware Security \n \n \n \n \n While computer hardware has improved tremendously in reliability and speed, these technological \nadvances have not always been good for computer security and data integrity.\n \n \n \n \n \n \nParity checks and data redundancies are critical for error-free data processing. Extra bits included at \npredetermined locations help catch certain types of errors when data is moved back and forth \nbetween different devices or from storage to registers.\n \n \n \n \n \n \n• Vertical redundancy checks (VRC), though common, have some problems. VRC are simple and \ninexpensive to implement. First, you determine whether there should be an odd or an even number of \n''1\" bits in each character's binary code. An error is detected if the correct number is not transmitted. \nThe basic flaw with the approach is that two errors may offset each other, allowing the error to go \nunnoticed. Furthermore, there is no standardization on the use of odd or even parity.\n \n \n \n \n \n \n• Longitudinal redundancy checks (LRC) provide an additional safeguard since VRC may not detect \nall the errors. This technique involves the use of an extra character generated after some \npredetermined number of data characters. The bits in the extra character provide parity for its row. \nLRD has its limitations. It cannot correct multiple errors or errors in ambiguous position (ambiguous \nbit is correct for VRC but incorrect for LRD), or errors that do not result in both a VRC and LRC \n \n"
        },
        {
            "page_number": 30,
            "text": "indication.\n \n \n \n \n \n• Cyclical redundancy checks (CRC) are typically used when extra assurance of the accuracy of data \nis needed. A large number of redundant data bits is used, which requires longer transmission times \nand extra space in memory. The primary advantage of this technique is that any single error, whether \nin data bit or parity bit, would be detected.\n \n \n \n \n \n Hardware typically has several features to protect the data during input, output, and processing. \n \n \n \n \n • Dual-Read reads the same data twice and compares the two results. Any discrepancy indicates an \nerror.\n \n \n \n \n \n • Read-After Write reads the data immediately after it's recorded to verify after it is recorded to verify \nthe accuracy of the write function.\n• Echo Check is used to verify the reception of a signal when data is transmitted to another computer or \nto peripheral devices such as printers.\n \n \n \n \n \n \n• Replication is an important feature for critical applications. A backup computer/site is used in case \nof failure of the primary computer. Fault-tolerant or fail-safe computers contain at least two \nprocessors that operate simultaneously; if one fails, the other processors pick up the load. When a \ncritical application requires extensive communication facilities, the backup equipment should contain \nboth communication equipment and a processor. Repairs or replacement of malfunctioning \nequipment should be immediate.\n \n \n \n \n \n • Overflow may result when an arithmetic operation, such as dividing by zero, results in values \nbeyond a computer's allowable range. This function is typically built into the computer hardware.\n \n \n \n \n \n \n• Interrupts are generated when the hardware detects deviations in order to maintain the integrity of \nthe data processing system. For example, input/output (I/O) interrupts result when a previously busy \ndevice becomes available. The equipment then checks after each I/O interrupt to determine if the data \nhas been written or read without error. I/O interrupts are generated when the Escape or Enter key is \npressed. From a security perspective, interrupts can affect logs or cause the execution of unauthorized \nprograms. Other types of interrupts include program check, machine check, and external. Program \ncheck interrupts terminate the program as a result of improper instructions or data. Machine check \ninterrupts are generated by defective circuit modules, open drive doors, and parity errors. External \ninterrrupts result from pressing an Interrupt key, from signal from another computer, or from timer \naction. From a security perspective, for example, the built-in electronic clock in the processor can be \nused to generate an interrupt at a specified interval to ensure that sensitive jobs do not remain on the \n \n"
        },
        {
            "page_number": 31,
            "text": "computer long enough to be manipulated. Plan for the possibility of loss of data does not result \nbecause of interrupts.\n \n \n \n \n \nMost integrated circuit chips on hardware equipment are inscrutable to a lay person. There are \nhundreds of thousands of transistors on a small semiconductor. Still, it's possible for a bug to be \nplanted into electronic equipment, and it may be very difficult to detect. Several techniques may be \nused to seal hardware against such tampering.\nKeep records of hardware failure and computer down times Schedule regular maintenance, and record \nthe results If computer equipment needs frequent servicing, personnel might be tempted to bypass \ncontrols and take shortcuts, raising the possibility of human errors considerably Analyze your records \nfor unfavorable trends in downtime or frequently unscheduled service calls\n \n \n \n \n \n The hardware inventory logs for all computer equipment and peripherals should contain at least the \nfollowing information:\n \n \n \n \n \n • Description of the hardware \n \n \n \n \n • Manufacturer's name \n \n \n \n \n • Model number \n \n \n \n \n • Serial number \n \n \n \n \n • Company identification number \n \n \n \n \n • Date of purchase \n \n \n \n \n • Name, address, and phone number tor the source ot the item, whether store or manufacturer \n \n \n \n \n • Date warranty expires \n \n \n \n \n"
        },
        {
            "page_number": 32,
            "text": " • Department or location where the hardware equipment will be used \n \n \n \n \n • Name and title of individual responsible for the equipment \n \n \n \n \n • Signature of the responsible individual or department head \n \n \n \n \n • If the equipment is taken off premises, the date and time the equipment is checked out, and the date \nand time it's returned, along with the signature of the authorized individual\n \n \n \n \n \n \nHardware inventory logs should be stored in a secure location with a copy stored off-site All hardware \nshould be etched or engraved with the company name, address, telephone number, manufacturer's \nserial number, and company's identification number To prevent theft, locking devices should secure \ncomputer equipment and peripherals to desktops, etc.\nSoftware and Devices for Physical Security \n \n \n \n \n A wide variety of software and devices is available to prevent computer theft. Computer Security \nProducts, Inc. (http://www.computersecurity.com) provides an excellent assortment.\n \n \n \n \n \n CompuTrace Theft Recovery Software  \n \n \n \n \n \nCompuTrace Theft Recovery Software is primarily for laptop computers, but it may be used with \ndesktops. Once the software is installed, it works silently and transparently. Regularly and often, it \nuses the computer's modem to place a toll-free call to a monitoring center after checking to see if the \nmodem is attached and in use. It turns off the modem speaker when making its scheduled call. The \ncomputer's serial number and the origination telephone number are recorded with each call.\n \n \n \n \n \n \nIf the computer is stolen, you call CompuTrace's theft hot line to activate the Theft Recovery \nAssistance Procedure. The next time the stolen computer's modem dials in to the monitoring center, \nCompuTrace acquires the origination telephone number and determines its location. Local law \nenforcement authorities are then notified.\n \n \n \n \n \n CompuTrace is available for DOS and Windows-based systems. It cannot be deleted; it even survives \na hard-drive format. The only way to delete it is to use a registered copy of the uninstall disk.\n \n \n"
        },
        {
            "page_number": 33,
            "text": " \n \n \n CompuTrace, which uses less than 7K of memory is not detectable by antivirus software and does \nnot appear in any directory. It's fully automated and does not interfere with other applications.\n \n \n \n \n \n \nIt works from any phone line in North America. It works even if the phone number is unlisted. It \ndoesn't rely on Caller-ID technology. It even works from hotel and office phones that require you to \ndial a prefix to reach an outside line. If CompuTrace doesn't detect a dial tone when it first calls out, \nit will try again with various prefix combinations.\n \n \n \n \n \n \nThough CompuTrace's default calling schedule is usually 5 to 7 days, you may change it. It's also \npossible to program the computer to call in with greater frequency once it has been reported stolen. If \nthe modem is not connected or is in use at the scheduled call time, CompuTrace keeps on trying \nperiodically till the modem is available.\n \n \n \n \n \n \nAs an added benefit, CompuTrace may be used to manage computer assets in large organizations. \nThe CompuTrace Monitoring Center provides up-to-the-minute listings of all computers and their \nlocations. It's easy to determine whether the computer is in a regional office, at an employee's home, \nor on the road. Monitoring reports can be downloaded from a private Internet web site. Reports can \nbe distributed via email or fax.\n \n \n \n  \nCompuTrace is available from Computer Security Products, Inc. (800.466.7636). At the time of this \nwriting, CompuTrace was available with:\n \n \n \n \n \n • 1-year Monitoring Service for \n \n \n \n \n $89.95 \n \n \n \n \n • 2-year Monitoring Service for \n \n \n \n \n $149.95 \n \n \n \n \n • 3-year Monitoring Service for \n \n \n \n \n $199.95 \n \n \n \n \n Quantity discounts are available: \n \n"
        },
        {
            "page_number": 34,
            "text": " \n  \n \n \n \n • 26—100 units \n \n \n \n \n 15% \n \n \n \n \n • 101–250 units \n \n \n \n \n 20% \n \n \n \n \n • 251–500 units \n \n \n \n \n 25% \n \n \n \n \n • 501–1,000 \nunits\n \n \n \n \n \n 28% \n \n \n \n \n • over 1,000 \nunits\n \n \n \n \n \n 30% \n \n \n \n \n PC and Peripheral Security  \n \n \n \n \n \nMost computer equipment and peripherals can be quickly secured with steel cables, an easy and \ninexpensive theft deterrent. Special fasteners protect RAM chips and internal components. Cover \nlocks can be used to:\n \n \n \n \n \n • Lock the computer case \n \n \n \n \n • Block access to disk drive slots \n \n \n \n \n • Block access to the CD-ROM \n \n \n \n \n • Block access to the on/off switch \n \n"
        },
        {
            "page_number": 35,
            "text": " \n \n \n The base of the cover lock can be attached to most flat surfaces. The locks may be keyed alike or \ndifferently. Master keying is also possible.\n \n \n \n \n \n \nLock-down plates provide additional security. The Cavalier Security System, for example, consists of \ntwo steel plates. The base plate contains the lock and is secured to a table. The insert or top plate is \nattached to the equipment to be protected. The plates come in various sizes depending on the width \nand length of the equipment to be secured. By selecting a size slightly smaller than the equipment's \nfootprint, the lock-down plates appear less obtrusive.\nNetwork Alarm  \n \n \n \n \n \nEtherLock Alarm System (www.computersecurity.com), which plugs into the hub, allows you to use \nyour network as an alarm system. It interfaces with your existing alarm system and uses the \nnetwork's wiring to secure computers; an alarm is triggered if a network wire is unplugged. It does \nnot affect system performance or network throughput.\n \n \n \n \n \n \nLockSoft Remote Management Software for EtherLock systems \n(www.computersecurity.com/etherlock/locksoft.htm) allows for control of the EtherLock system from \nany computer on the network. A central monitoring site can be notified of the attempted theft. \nRunning LockSoft software with EtherLock lets you perform the following tasks from the central \nconsole:\n \n \n \n \n \n • Receive network-based alarm reports when computers are disconnected. \n \n \n \n \n • View the connection status of all protected devices. \n \n \n \n \n • Remove individual devices from the protection loop for maintenance or relocation. \n \n \n \n \n • Arm, disarm, and test all EtherLock systems connected to the network. \n \n \n \n \n • Allow password-protected access to secure individual computers. This feature lets administrators \ngive notebook users the flexibility to disconnect their machines.\n \n \n \nAt the time of this writing the cost of EtherLock 10T Base Unit was $1,948. The base unit can hold \n"
        },
        {
            "page_number": 36,
            "text": "  \nup to 16 Protection Modules, each costing $799 and supporting up to 12 devices. Therefore, the full \nsystem can protect up to 192 computers and peripherals on a single hub. Its modular design allows \nfor expansion as the LAN grows. The minimum configuration requires one protection module.\n  \n \n \n \n \n \nThe LockSoft software that comes with EtherLock computer security systems is available for \nWindows and DOS-based systems. Administrator software is included; it collects data on the \nEtherLock system and the devices being protected.\n \n \n \n \n \n \nTo protect laptop computers, the NoteLock security bracket ($19.95) may be used in conjunction with \nthe EtherLock security system. You can connect to or disconnect from the network using the Ethernet \ncable. The LockSoft program simply asks you to enter a personal password. Personnel can be alerted \nif an attempt is made to remove a secured laptop computer from the network. Logging off from the \nnetwork or powering down the computer does not affect the security features; only the appropriate \npassword can be used to disconnect from the network.\nThe SimmLock security bracket ($19.95) is designed to protect memory chips (SIMMs), \nmicroprocessors, hard drives, and other internal components. Security personnel are alerted if any \nattempt is made to remove the computer case or access its internal components. SimmLock brackets \ncan be affixed to monitors, external hard drives, and other peripheral equipment not directly connected \nto the network.\n \n \n \n \n \n Asset Tracking  \n \n \n \n \n \nTamper-proof asset-tracking security tags should be affixed on computers and peripherals. STOP \n(Security Tracking of Office Products) asset tags are available from Computer Security Products, Inc. \n(http://www.computersecurity.com/stop/index.html). Security plates or tags help in three ways: (1) \nthey deter theft—a thief is less likely to steal tagged equipment; (2) they help in recovering stolen \nequipment; and (3) you can use these tags for asset management.\n \n \n \n \n \n \nSTOP plates link equipment data to a worldwide tracking and retrieval service. If equipment is lost or \nstolen, law enforcement authorities can be notified to track it. The barcode on STOP tags can be used \nto track equipment day to day and can interface with the Microsoft Access database.\n \n \n \n \n \n \nThe STOP security plate, made of photo-anodized aluminum, is secured to equipment using \ncyanoacrylate adhesive. It takes about 800 pounds of pressure to remove the security plate. If the \nplate is removed, the equipment casing will be noticeably altered.\n \n \n \n \n \nBehind each plate is an indelible tattoo, \"Stolen Property,\" that is chemically etched into the \nequipment. If someone succeeds in removing the security plate using special tools, the indelible \n"
        },
        {
            "page_number": 37,
            "text": " \nmarking is exposed, as are the company identification number (optional) and a toll-free number for \nverification and anti-theft information. This tattoo cannot be removed without defacing the case. \nDefacing is recognized by police and equipment sellers as a sign that the property is stolen.\n \n \n \n \n \n \nEach security plate bears a warning that the property is monitored and traceable. It also warns that a \ntattoo has been etched into the equipment. Each plate also has a barcode to track information and a \ntoll-free telephone number to call in case lost or stolen equipment is found.\n \n \n \n \n \n Once equipment is registered, the STOP retrieval service will oversee its return. In any case of theft, \nSTOP will help register the loss with law enforcement agencies in the United States and abroad.\n \n \nSTOP'S hand-held barcode scanner, along with its asset tracking software, helps you maintain the \ninventory of valuable equipment. Inventory records are updated simply by scanning tags. The software \nwill report on missing or out-of-plate hardware. It can also report on mobile equipment by registering \nwho borrowed the equipment and when it was due.\n \n \n \n \n \n \nThe software is network-ready and customizable. It's based on the Microsoft Access database, but the \nsoftware includes a runtime module, so Microsoft Access is not required to use the software. Source \ncode is available for you to customize it.\n \n \n \n \n \n \nEach STOP security plate costs $25. Quantity discounts can significantly reduce the cost of each \nplate. For example, if 10 or more plates are purchased, the price drops to $15 each. If more than 500 \nplates are ordered, the security plates cost less than $9 each. The Tattoo Activating Gel costs $2.50 \nfor up to 10 security plates. For customized plates, the minimum order is 200 units and requires a \none-time setup charge of $250.\n \n \n \n \n \n The price of the security plate includes unlimited use for three years of STOP'S anti-theft and \nretrieval hotlines and its recovery service.\n \n \n \n \n \n \nAfter the first three years, unlimited use of these services costs $1 per year per machine, or $4 \nlifetime per machine. For large sites, a $200 flat fee per year covers an unlimited number of \nmachines.\n \n \n \n \n \n The STOP asset tracking management software costs $200 but is provided free with an order for 500 \nor more security plates.\n \n \n \n \n \n The Intermec hand-held barcode scanner package costs $2,500 and includes: \n"
        },
        {
            "page_number": 38,
            "text": " \n \n \n \n • Communications dock and cable \n \n \n \n \n • Charger \n \n \n \n \n • Light wand and cord \n \n \n \n \n • Power supply \n \n \n \n \n • Barcode creation software \n \n \n \n \n • STOP Asset Tracking Software \n \n \n \n \n The Xyloc System  \n \n \n \n \n \nXyloc access cards may be used to secure desktop computers and laptops. The card automatically \nlocks the computer and blanks the screen when the authorized user with the card leaves a pre-defined \narea. It also automatically unlocks the computer system when the authorized user returns with the \ncard. The computer's session work is preserved when the computer is locked. Background tasks \ncontinue to run even when the system is locked.\nThe pre-defined area, the \"active zone,\" can be set from one to 50 feet. It allows access by many users \nto a single computer. Communication between the access card and the lock is encrypted to prevent an \nattacker from grabbing the code to create a clone. The system can be programmed to deploy \nincrementally to individual PCs, to workgroups, or enterprisewide.\n \n \n \n \n \n \nThe Xyloc system is suitable when you need high security and restricted access to the computer \nsystem and system files. The system works either alone using the access card or, for even greater \nsecurity, in conjunction with a password. On a LAN, it's possible to remotely manage several Xyloc \nsystems from a central facility. The software maintains the audit trail and logs events. This \ninformation may be used, for example, to determine if proper security procedures are being followed.\n \n \n \n \n \n The Xyloc key contains a low-power radio transceiver with a unique user identification code. It's \npowered by a lithium battery that typically lasts six months to a year. A battery meter lets you know \n \n"
        },
        {
            "page_number": 39,
            "text": "how much power remains. The Xyloc Card Key is compatible with other ID and security badges. \nThere's an encrypted channel for all communications with the lock.\n \n \n \n \n The lock is a small device containing a low-powered transceiver. The lock simply plugs into the \nkeyboard, serial, or USE port; it's powered by the port and contains a built-in status light.\n \n \n \n \n \n \nThe Xyloc Access Card system ($189.95; see http://www.seattlecomp-sec.com) consists of two \npieces of hardware and software for the access control card. Each package includes one card key, one \nlock, and software. Additional pieces are available separately.\n \n \n \n \n \n Card Technology  \n \n \n \n \n \nMany manufacturers are combining multiple technologies, such as bar codes, magnetic strips, \nproximity, and smart cards, on a single card. Such a card may also serve as a photo ID, which in \neffect gives it an additional function. A universal reader device that can support multiple formats will \nbe required.\n \n \n \n \n \n \nCurrent access control technology typically works by keeping doors locked. It denies access to \neveryone except those who can show or do something to get through the door. Technology is now \nproceeding in a new direction. Doors are left open, closing only when an unauthorized person tries to \nenter. For example, users might carry cards with chips that would tell the door that the person is an \nauthorized user and that it's okay to stay open. The approach of anyone not carrying the appropriate \ncard would close the doors.\nThere are potential problems associated with this technology. For example, assume a group of four \npeople approaches. Only three of the four are authorized. The system should be capable of stopping the \ngroup and letting only the authorized individuals through.\n \n \n \n \n \n \nSoftware for access control systems can help in collecting and managing a wide variety of data, data \nthat could help determine, for example, the total amount of time spent on site by each cardholder. \nAccess control data may be used to determine which employees are still in the facility during an \nemergency. This may save lives by helping authorities determine who might be trapped inside.\n \n \n \n \n \n Visual Surveillance  \n \n \n \n \n \nVideo surveillance is becoming increasingly popular. Cameras are more affordable. Image quality \nhas improved tremendously. The components are getting smaller and more reliable. Cameras are \nmore functional and responsive. Features such as panning, tilting, and zooming are common.\n \n"
        },
        {
            "page_number": 40,
            "text": " \n \n \n \n \nDigital videos, digital transmission of data, and digital storage are likely to increase the use of \nsurveillance equipment. Digital storage allows security personnel to retrieve specific scenes quickly. \nImage quality tends to be much better than ordinary videotape.\n \n \n \n \n \n \nDigital technology makes it possible to record and view images at the same time. Improvements in \ntransmission media may mean that cameras at remote sites will replace more security officers. \nRemote monitoring and recording is becoming more feasible because of price decreases in \ncomponents, including chips and memory.\n \n \n \n \n \n Biometric Devices  \n \n \n \n \n \nBiometrics for access control purposes is on the horizon. It hasn't gained widespread popularity \nprimarily because of its cost and lack of accuracy. Both are likely to diminish with improvements in \ncomputer processing. A facial recognition system for door access will soon be widely available. \nCompanies are working on integrating fingerprint sensor technology into keyboards in order to \nrestrict access to a terminal or a network. Miniature cameras at computer workstations may control \naccess through facial recognition technology.\nChapter 3— \nHardware Security \n \n \n \n \n \nSoftware security depends on hardware security. If the hardware can be stolen or surreptitiously \nreplaced, secure software will not help. Before the invention of the personal computer, computer \nmainframes were so huge that they took up an entire room. To secure these machines, IT managers \nlocked the rooms. Now small and portable laptop and palmtop computers are easily stolen.\n \n \n \n \n \n \nCompanies use computers for storing sensitive information, doing online transactions, and accessing \nprivate and public networks. IT managers looking to protect their investments must consider securing \nthe perimeter and allowing only authorized users access to their computers.\n \n \n \n \n \n Some hardware problems are common: \n \n \n \n \n • Equipment and removable media can be stolen or substituted. \n \n \n \n \n • Changing hardware setup parameters can circumvent security. \n"
        },
        {
            "page_number": 41,
            "text": " \n \n \n \n • Systems can be booted by unauthorized users or unauthorized software. \n \n \n \n \n • Boot media can be rewritten by unauthorized software. \n \n \n \n \n • Unauthorized software can be executed from removable media. \n \n \n \n \n Some of the safeguards that can be taken are: \n \n \n \n \n • Locking doors and security equipment \n \n \n \n \n • Having lockable cases and keyboards and removable media drives \n \n \n \n \n • Having a key or password-protected configuration and setup \n \n \n \n \n • Requiring a password to boot \n \n \n \n \n • Requiring a password to mount removable media \n \n• Using read-only media \n \n \n \n \n • Storing removable media in secured areas \n \n \n \n \n \nOrganizations that store and transmit sensitive and valuable information over both public and private \nnetworks should be concerned with information security (see Figure 31). For example, security \nbreach can occur in hardware, software, network connections, or authentication procedures. Today, \nsecuring information is more difficult not only because PCs are portable but also because of cheap \nhigh-speed modems and the Internet Attacks are becoming more sophisticated Hackers have tools to \nautomate attacks.\n \n \n \n"
        },
        {
            "page_number": 42,
            "text": "  \n  \n \n \n \n \n \nFigure 3.1: \nCSI/FBI 1998 Computer Crime and Security Survey \nSource: Computer Security Institute\n \n \nSecurity breaches cost companies millions of dollars not only in losses but also in increased staff hours, \ndecreased productivity, loss of credibility in the marketplace, and legal liability. Of 520 companies \nsurveyed in the 1998 Computer Security Institute/FBI Computer Crime and Security Survey, 64% \nreported a security breach. The total financial loss for that 64% was estimated at over $136 million. \nThese figures are very conservative; they do not reflect corporations that did not realize they had a \nsecurity breach or those that are reluctant to report one for fear of negative press.\n \n \n \n \n \n Note particularly in Figure 3.1 that $50.6 million in financial losses were attributable to unauthorized \ninsider access.\n \n \n \n \n \n Physical Security \n \n \n \n \n \nPhysical security is almost everything that happens before a user (or an attacker) starts typing \ncommands on the keyboard. Surprisingly, many organizations are not deeply concerned about \nphysical security. It may be the form of security most likely to be forgotten because physical security \nthreats, practices, and protections are different for practically every site (Garfinkel, 1996).\n \n \n \n \n \n Yet, physical security is not that hard: \n \n"
        },
        {
            "page_number": 43,
            "text": " \n \n \n • Access should only be given to those who need specified data. \n \n \n \n \n • Computer rooms should have bars on any windows or electronic detection devices as needed. \n \n \n \n \n • Access to rooms containing computers should be restricted to specific personnel. These rooms can \nbe controlled with locked doors, posted guards, and similar restraints.\n \n \n \n \n \n • Protect computers with lockable equipment enclosures, lockable power switches, fasteners, and \nother securing devices.\n \n \n \n \n \n • Make sure other electronic devices do no interfere with computers, and that the computers do not \ninterfere with other electronic devices.\n \n \n \n \n \n • Keep computers in a comfortable environment to prolong their life and ensure the safety of stored \ndata.\n \n \n \n \n \n • Shield computers from sudden surges or drops in electrical line voltage with surge protectors or \nuninterruptible power supplies.\nData Integrity \n \n \n \n \n \nData integrity is as important as the actual hardware Data integrity refers to the validity of data If \nintegrity is compromised, there is no point in protecting your data Data integrity can be compromised \nin a number of ways:\n \n \n \n \n \n • Human errors when data is entered \n \n \n \n \n • Errors occurring when data is transmitted between computers \n \n \n \n \n • Software bugs or viruses \n \n \n"
        },
        {
            "page_number": 44,
            "text": "  • Hardware malfunctions, such as disk crashes  \n \n \n \n \n • Natural disasters, such as fires and floods \n \n \n \n \n To minimize threats to data integrity: \n \n \n \n \n • Back up data regularly. \n \n \n \n \n • Control access to data. \n \n \n \n \n • Design user interfaces that prevent the input of invalid data. \n \n \n \n \n • Use error detection and correction software when transmitting data. \n \n \n \n \n Network Backup  \n \n \n \n \n \nIn a world where intruders can come from both inside and outside, IT managers need a contingency \nplan to make sure that if an intruder sabotages an entire system, they can bring the data back. \nBacking up data to a network backup server is critical. Having a centralized backup location can save \nyou time and money searching for backup tapes. Organizations that support heterogeneous clients \nalso need this capability.\n \n \n \n \n \n \nRetrospect Remote 3.0 from Dantz Development Corp. is a full-featured backup server. With the \nRemote Pack 1.0 client software running on a dedicated Macintosh, it can back up Windows as well \nas Macintosh clients, handling all backup, restore, duplication and backup server tasks, whether client \nsources are hard drives, external or removable media, or files in subdirectories (Windows) or folders \n(Mac).\nAccess Control System \n \n \n \n \n Consider securing the perimeter of your building. By allowing only unauthorized users access to your \nfacility, you greatly reduce the risk of someone stealing proprietary information.\n \n \n"
        },
        {
            "page_number": 45,
            "text": " \n \n \n \nRAND Corporation, a respected nonprofit policy research organization, controls physical security \nwith staff access cards, alarms, surveillance cameras, and its own guard force. Surveillance cameras \nrecord all entrances to and exits from the building, including the entire parking lot. The guards \nscreen, sign in, and escort visitors to their destinations and patrol both interior and perimeter of the \nfacility. They investigate incidents and complaints and may escort staff to their cars. Only RAND \nstaff members are allowed in the facility.\n \n \n \n \n \n Memory Data Integrity Checking  \n \n \n \n \n \nEnsuring the integrity of data stored in memory is an easy way to make sure your data is secured \nbefore being saved to floppy disk or on the network. Consider buying computers that use data \nintegrity checking to prevent later errors. Especially for computers with critical roles, such as servers, \nan error correction code (ECC) capable memory controller is a good idea.\n \n \n \n \n \n \nThe two primary methods to ensure the integrity of data stored in memory are parity and ECC. Parity \nis the most common: One bit is added to every 8 bits (1 byte) of data. The limitation is that the parity \nmethod can detect an error but cannot correct it. ECC is more comprehensive: It can correct as well \nas detect 1 -bit errors, usually without the user even knowing that there has been an error. Kingston \n(www.kingston.com), an independent manufacturer, sells data integrity checking memory products \nfor use in workstations, servers, desktops, portables, and printers.\n \n \n \n \n \n Deploying a Security System \n \n \n \n \n \nWhen planning a security system, form a strategy rather than randomly deploying gizmos. Three \npossible strategies are perimeter security, interior security, and physical protection of individual \nassets. While you want to keep intruders out, you also want to limit the amount of damage they can do \nonce they are inside.\nPerimeter Security  \n \n \n \n \n \nPerimeter security means preventing entry by criminals, who choose the easiest points, usually doors \nand windows. Windows can be protected with small, easily mounted devices that sound an alarm \nwhen the glass is broken.\n \n \n \n \n \n \nAccess control systems are complex, high-tech information networks but they are essential to office \nsecurity. Authorization can take many forms, including entering a code into a keypad, sliding a card \nthrough a scanner, or pressing a button on a wireless remote control. ADT, an electronic security \ncompany, sells access control systems like a card reader (see Figure 3.2) to help companies track the \n \n"
        },
        {
            "page_number": 46,
            "text": "comings and goings of their employees.\n \n \n \n \n \n \n \n \n \n \n \nFigure 3.2: \nADT Card Reader \nSource: ADT\n \n \n \n \n \n Interior Security  \n \n \n \n \n \nInterior security is as much about deterring intruders as about catching them. Motion detectors can be \nused to set off alarms. Surveillance cameras can detect theft when they are located so as to catch the \nintruder's eye. Deltavision (www.deltavision.com), a Canadian CCTV manufacturer, sells a wall or \nceiling-mounted camera that interfaces with Deltavision recorders for 24-hour real-time surveillance \n(see Figure 3.3). If you're worried about espionage, you can buy cameras disguised as clocks, \ntelephones, or even exit signs.\n \n \n \n \n \n \nFigure 3.3: \nDeltavision designer mount \ncamera \nSource: Deltavision\n \n \n \n \n \n Physical Protection of Individual Assets  \n \n \n"
        },
        {
            "page_number": 47,
            "text": "  \nMany computing devices are small enough to be put in a briefcase and carried away. Cable kits can \nbe an inexpensive solution to this problem. Some hardware devices can be connected to computers \nand set to trigger an alarm. Software can be installed in a laptop to let police track down stolen \nlaptops.\n  \n \n \n \n \n Integrated Firewall Appliances \n \n \n \n \n \nFor IT managers looking for strong security without any impact on network performance, Cisco \nSystems, Inc. (www.cisco.com), the worldwide leader in networking for the Internet, provides the \nPIX Firewall Series. Unlike typical CPU-intensive proxy servers that perform extensive processing \non each data packet, a PIX Firewall uses a non-UNIX, secure, embedded system. Its performance of \nup to 256,00 simultaneous connections, over 6,500 connections per second, and nearly 170 megabits \nper second (Mbps) throughput is dramatically greater than that delivered by other appliance-like \nfirewalls or those based on general-purpose operating systems.\n \n \n \n \n \n \nThe PIX series is based on the adaptive security algorithm (ASA), which effectively protects access \nto the internal host network by comparing inbound and outbound packets to entries in a table. Access \nis permitted only if passage can be validated. Cut-through proxy, which enhances authentication, \nchallenges a user initially at the application later, but once the user is authenticated and policy is \nchecked, the PIX Firewall shifts session flow to a lower layer for dramatically faster performance. \nPIX Firewalls allow you to accommodate thousands of users without affecting performance.\n \n \nTotal VPN Solution \n \n \n \n \n \nIn the software security chapter, it will be shown that companies rely on virtual private networks \n(VPN) to allow telecommuters, customers, suppliers, and branch offices access their internal \ninformation. Unfortunately, VPN security software requires extensive encrypting, with decrypting at \nthe server end, which means that users experience performance degradation. As your organization \ngrows and more people use VPN, performance of the host computer will be further impeded.\n \n \n \n \n \n \nChrysalis-ITS, Inc. (www.chrysalis.com), the provider of high-performance network security \nsolutions, now offers a hardware solution, Luna VPN. Luna VPN extends security network \nconnections across Intranet, Extranet, and Internet by combining security software and hardware \nacceleration into one package. With Luna VPN hardware, the encryption and decryption processing \nis offloaded from the host computer to increase network throughput and remove bottlenecks. The \nLuna VPN combination increases performance by up to 11 times for DES operations, and up to 19 \ntimes for Triple-DES operations (Figure 3.4).\n \n \n \n"
        },
        {
            "page_number": 48,
            "text": "  \n  \n \n \n \n \n \nFigure 3.4: \nLuna VPN Network Throughput Performance \nSource: Security Assurance White Paper from Chrysalis-ITS, 1999\nProcessor overhead can also be decreased when Luna VPN is added, because it can significantly \noffload the host CPU. Adding a second processor to either Solaris or Windows NT systems does not \noffload processing from the host processor as effectively.\n \n \n \n \n \n Notebook Security \n \n \n \n \n As computer theft and network break-ins become a growing concern for notebook users and their \ncompanies, major computer vendors are offering products to safeguard their machines:\n \n \n \n \n \n • IBM offers a Smart Card Security Kit on its notebooks' hard drives. \n \n \n \n \n • HP offers smart cards on its OmniBook notebooks, Vectra desktop PCs, and Kayak workstations. \n \n \n \n \n • Dell offers a hard-drive password feature on Latitude notebooks, OptiPlex desktops, and Precision \nworkstations.\n \n \n \n \n \n • Compaq uses Fingerprint Identification Technology (biometrics instead of passwords) to log on to a \nnetwork on its Deskpro, Armada, and Professional Workstation systems.\n \n \n \n \n \n Smart-card technology offers numerous security advantages: \n \n"
        },
        {
            "page_number": 49,
            "text": " \n \n \n • Users can log in from any PC on the network, enabling ''roaming\" between PCs and easy import of \ntemporary or flex-time personnel.\n \n \n \n \n \n • A digital signature on the card provides instant proof of sender identity and message authentication. \n \n \n \n \n • Permanent encryption keys ensure that stored information is transmitted easily without \ncompromising confidentiality.\n \n \n \n \n \n • Chip circuitry for smart cards is tamper-proof, becoming void upon exposure to light. \n \n \n \n \n • Sensitive information is stored on the card itself, unlike biometrics user-authentication systems, \nwhich store data within a system database, where it might be accessed illegally.\n \n \n \n \n \n • Smart cards can display an employee's identification photograph and double as a swipe card for \nentry to company facilities.\nTokens: \nDual-Factor Security \n \n \n \n \n \nIt's not hard for a sophisticated hacker or a determined insider to gain access to a supposedly secure \nsystem if your confidential resources are protected only with a single factor, a user password. \nSecurity Dynamics Technologies, Inc. (www.securitydynamics.com), has developed a SecurID token \nfor a two-factor identification. Factor one is something secret that only the user knows, such as a PIN \nor a password. The second factor is something the user possesses, the SecurID token, a small device \nthe size of a credit card that displays a constantly changing ID code (Figure 3.5).\n \n \n \n \n \n \n \n \n \n"
        },
        {
            "page_number": 50,
            "text": "  \nFigure 3.5: \nSecurID Token \nSource: Security Dynamics\n  \n \n \n \n \n \nThe SecurID is a microprocessor-based handheld device that generates unique, one-time, \nunpredictable access codes every 60 seconds. To gain access, a user must enter a name, a personal \nidentification number, and the code currently displayed on the SecurID token. The security server \ncompares the code with the proper password for that time period. The tokens require no card readers \nor time-consuming challenge/response procedures.\n \n \n \n \n \n The tamperproof SecurID token can: \n \n \n \n \n • Prevent unauthorized access to information resources \n \n \n \n \n • Authenticate users at network, system, application, or transaction level\n• Be used from any PC, laptop, or workstation—ideal for remote access and VPNs \n \n \n \n \n • Work seamlessly with WebID for secure Web access \n \n \n \n \n • Work across access control modules (ACMs) for multiple platform enterprise security \n \n \n \n \n \nSecuring sensitive data is especially important when it involves access to clinical data. Beth Israel \nDeaconess Medical Center in Boston uses the SecurID token to authenticate access to its medical \ndatabases. When Beth Israel Hospital and Deaconess Medical Center merged in October 1996, the \nSecurID token gave them a single way with an identical look and feel to get on both systems: Beth \nIsrael stores its patient records in a custom-built Mumps-based system, while Deaconess uses a \nSybase repository (Harrison, 1999).\n \n \n \n \n \n \nThe hospitals looked at other security tools like biometrics but the hardware would have had to be \ndeployed on every PC. They were concerned with error rates caused by a greasy fingertip, which \ncould lock a medical worker out of access to urgently needed medical data. The hospitals are now \nlooking at smart cards to consolidate applications that currently include swipe cards, digital \ncertifications, VPN, and tokens.\n \n \n \n \n \n The SecurID token is interoperable with Cisco and Intel routing products and compatible with VPNs.  \n"
        },
        {
            "page_number": 51,
            "text": "Some users may send authentication passwords remotely over public networks.\n \n \n \n \n The Smart Card \n \n \n \n \n \nA smart card is a credit-card-sized device that has an embedded microprocessor, a small amount of \nmemory, and an interface that allows it to communicate with a workstation or network. According to \nthe Smart Card Industry Association, some 1.6 billion smart cards were issued in 1998. In 2000, that \nnumber is expected to be 2.8 billion. The most common application is the phone card, which stores a \nset value for use in public pay phones (see Table 3.1).\n \n \n \n \n \n With smart cards, passwords and Ids are securely encrypted on the card; when combined with \nbiometrics finger-scanning technologies, users can even skip the PIN code.\nTable 3.1: Smart Card Applications and Projected Growth Rates by Units Deployed \n \n \n \n \n Application Cards \nIssued\n \n \n \n \n \n \nIn Cards Issued\n \n \n \n \n \n In Annual Average  \n \n \n \n \n Smart Card \nApplications\n \n \n \n \n \n 1996 (millions)  \n \n \n \n \n 2000 (millions)  \n \n \n \n \n Growth Rate (%)  \n \n \n \n \n Pay phone \n \n \n \n \n \n605\n \n \n \n \n \n \n1,500\n \n \n \n \n \n \n29\n \n \n \n \n \n GSM \n \n \n \n \n \n20\n \n \n \n \n \n \n45\n \n \n \n \n \n \n25\n \n \n \n \n \n Health care \n \n \n \n \n \n70\n \n \n \n \n \n \n120\n \n \n \n \n \n \n14\n \n \n \n \n \n Banking \n \n \n \n \n \n40\n \n \n \n \n \n \n250\n \n \n \n \n \n \n105\n \n \n \n \n \n Identity/access \n \n \n \n \n20\n \n \n \n \n \n300\n \n \n \n \n \n280\n \n"
        },
        {
            "page_number": 52,
            "text": " \n  \n \n \n \n \n \n \n \n Transportation \n \n \n \n \n \n15\n \n \n \n \n \n \n200\n \n \n \n \n \n \n247\n \n \n \n \n \n Pay TV \n \n \n \n \n \n15\n \n \n \n \n \n \n75\n \n \n \n \n \n \n80\n \n \n \n \n \n Gaming \n \n \n \n \n \n5\n \n \n \n \n \n \n200\n \n \n \n \n \n \n780\n \n \n \n \n \n Metering/vending \n \n \n \n \n \n10\n \n \n \n \n \n \n80\n \n \n \n \n \n \n140\n \n \n \n \n \n Retail/loyalty \n \n \n \n \n \n5\n \n \n \n \n \n \n75\n \n \n \n \n \n \n280\n \n \n \n \n \n Source: Smart Card Industry Association \n \n \n \n \n \nA smart card can be personalized to the user. For example, a company using smart cards to provide \naccess to network services would store network privileges and preferred system settings on each \nemployee's card, which could then be used from any terminal with a card reader, customizing the \ndesktop to the employee's preference.\n \n \n \n \n \n Two characteristics make smart cards especially well-suited for security-sensitive or personal data \napplications:\n \n \n \n \n \n \n1. The onboard processor can service requests from the network and return the results without \ndivulging the sensitive data. For example, a smart card could digitally sign data without divulging the \nuser's private key.\n \n \n \n2. The user can carry data on the smart card rather than entrusting it to network storage. For example, \n"
        },
        {
            "page_number": 53,
            "text": "  a smart card could be used to carry personal information about the user, such as medical records or \ndigital certificates.\n  \n \n \n \n \n \nMicrosoft Windows NT4.0, Windows 95, and Windows 98 all support smart cards and smart card \nreaders based on specifications established by the Personal Computer Smart Card (PC/SC) \nWorkgroup of leading companies. Windows-compatible smart card solutions can be used with \nInternet Explorer to authenticate a secure connection and with Outlook Express or Outlook 98 for \nsending and receiving secure e-mail.\nThrough the use of public key cryptography and X.509 certificates, smart cards securely store private \nas well as public key certificates. Embedded private key is the digital certification representation of a \nuser's identify. Smart card allows you to digitally sign and encrypt messages, provides access to \nprotected Intranet sites, and can enable a single network sign-on (Backman, 1999).\n \n \n \n \n \n Two-Factor Authentication  \n \n \n \n \n \nSecurity Dynamics Technologies, Inc. offers two-factor authentication technology in a smart card \nversion for use on corporate Intranets. As with the SecurID token, its design incorporates a PIN \nnumber and a unique password. The smart card version, the SecurID 100, however, will generate the \nunique password automatically and transmit it to the server without any user involvement. The same \ncard could be used to gain access to the facilities or store value for use in company cafeteria and \nvending machines.\n \n \n \n \n \n \nWith the SecurID 1100 smart card, you can customize features to provide building access by \nincorporating magnetic stripes or holograms. The card can be combined with corporate employee \nbadge systems and other security applications. Other applications can be integrated with the smart \ncard to customize it to your own environment.\n \n \n \n \n \n Smart Cards in Banks  \n \n \n \n \n \nGuarding the privacy of bank customers requires the strongest possible security. One company that \nbelieves smart cards can provide a high level of protection is the Union Bank of Switzerland (UBS), \namong the largest banks in the world. As of March 1999, UBS had deployed 12, 500 smart cards \namong its employees and was adding about 1,000 per month. Employees use the smart cards to gain \naccess to the company network and files. Eventually UBS plans to deploy smart cards to all 35,000 \nof its Swiss employees.\n \n \n \nBecause the UBS network is decentralized, it did not fit well with server-based authentication \nsolutions. Smart cards, however, allow for authentication and a single sign-on (SSO). Schumann \n"
        },
        {
            "page_number": 54,
            "text": "  \nSecurity Software, Inc. (www.shumannsoftware.com), offers the Secure Single Sign-On (SAM/SSSO) \nto deliver strong authentication and confidential data transfer. Unlike other systems, SAM/SSSO does \nnot require a central authentication server. It uses DES encryption and RSA-based public key infra-\nstructure (PKI). The General Security Service (GSS) API (www.gssc.net) provides the common \ninterface between the smart card and secured applications. Where strong authentication has not been \nadded to an application, the smart card also supports conventional log-in via a script engine that \nautomatically logs on with a user ID and password stored in the smart card.\n \nThe basic components of SAM/SSSO are a smart card and a smart card reader. User log-ins and \npasswords are stored on a tamper-proof smart card and are accessible only with a single sign-on PIN \nfor off-site as well as on-site security. With SAM/SSSO hours of wasted productivity spent in changing \nor losing passwords can be avoided.\n \n \n \n \n \n \nUBS's smart card single sign-on process starts when the user enters smart card, ID, and password into \nthe reader (Figure 3.6). The user clicks on the desired application, triggering the scripting engine to \nlaunch the application script, which may be on the workstation or the network server. The script \ndelivers the ID and password from the smart card to the application log-on screen.\n \n \n \n \n \n \n \n \n \n \n \n \nFigure 3.6: \nUBS's Smart Card Single Sign-On \nSource: Information Security Magazine, 1999\n \n \n \n \n \n Smart Card Vulnerabilities  \n \n \n \n \n \nWhile the flexibility of smart cards gives them many uses for businesses—access control, e-\ncommerce, authentication, privacy protection—it can also be susceptible to attacks. Think about the \nfollowing vulnerabilities of smart cards:\n• Attacks by the cardholder \n \n \n \n — Against the terminal \n \n"
        },
        {
            "page_number": 55,
            "text": " \n \n — Against the data owner \n \n \n \n — Against the issuer \n \n \n \n — Against the software manufacturer \n \n \n \n \n • Attacks by the terminal owner against the issuer \n \n \n \n \n • Attacks by the issuer against the cardholder \n \n \n \n \n • Attacks by the manufacturer against the data owner \n \n \n \n \n \nIT managers can prevent these attacks if they use both a smart card and a portable smart card reader \nfrom one vendor. Cylink Corporation, (www.cylink.com), an ISO 9001-certified provider of \nencryption-based network security solutions to Fortune 500 companies, provides an advanced public \nkey smart card, the PrivateCard, and a smart card reader, PrivateSafe. Using them in combination, \nyou can be assured that sensitive private information never leaves the card.\n \n \n \n \n \n \nThe PrivateCard Smart Card is an isolated tamper-proof smart card with an onboard microprocessor \nand control program. The user's digital signature public and private key pair is produced on the card \nand never leaves it. Since all sensitive private key cryptographic functions are performed securely on \nthe smart card chip, it protects the private key from attacks and prevents exposure to a potentially \nhostile external environment. PrivateCard can be integrated with other security products.\n \n \n \n \n \n PrivateCard has the following features: \n \n \n \n \n • RSA key generation supports up to 1,024-bit RSA functions. \n \n \n \n \n • Random number generation is preformed inside the smart card chip itself. \n \n \n"
        },
        {
            "page_number": 56,
            "text": "  • RSA private key functions such as decrypt and digital signatures are all performed on-chip.  \n \n \n \n \n • Multiple keys and data objects may be stored on the card. \n \n \n \n \n • Authorized users can perform access control and PIN management securely. \n \n \n \n \n • Passwords can be changed after a specified period. \n \n \n \n \n • Multiple-level file system allows information to be secured by directory.\nThe PrivateSafe Smart Card Reader connects between the keyboard and the PC (see Figure 3.7); no \nadditional hardware or external power supply is needed. Bypassing keyboard input, PrivateSafe isolates \nthe user's private information on the smart card so that it never reaches the PC. It generates the public \nkey pairs right on the PrivateCard so that the user's private key is never exposed to the PC.\n \n \n \n \n \n \n \n \n \n \n \n \nFigure 3.7: \nPrivateSafe Smart Card Reader \nSource: Cylink Corporation\n \n \n \n \n \n Tokens \n \n \nA token is a small device no bigger than a credit card that displays a constantly changing ID code. \n"
        },
        {
            "page_number": 57,
            "text": "  \nOnce the user enters a password, the card displays an ID that can be used to log into a network. As an \nalternative to entering an ID code from a token or deploying a smart card reader, Aladdin Knowledge \nSystems (www.aks.com), an information security company, has developed eToken (see Figure 3.8), a \ncar-key-sized token that plugs into a computer's Universal Serial Bus (USB) port (a standard feature \non virtually all PCs and laptops manufactured since 1997). USB provides true \"plug and play\" for up \nto 127 peripherals.\n \n \n \n \n \n \n \nFigure 3.8: \nAladdin's eToken \nSource: Aladdin Knowledge Systems\n \n \n \n \n \n \nFlexible enough to hold information without an expensive reader, eToken contains its own processor \nchip to encrypt information; store private keys, passwords, digital certificates, and digital cash; and \nprovide two-factor authentication for secure access to VPNs, remote access servers, subscription-\nbased Web content, and back office applications.\n \n \n \n \n \n \nAladdin's eToken technology opens and guards doors in e-commerce, e-banking, VPNs, Extranets, \nand WANs. It can protect sensitive data and resources with file encryption and access control, and it \ncan sign or encrypt electronic messages so that they cannot be forged, changed, or intercepted.\n \n \n \n \n \n eToken has the following features: \n \n \n \n \n • Easy to use: No additional hardware or software is needed. You just insert eToken into the USB \nport on a desktop, laptop, monitor, or keyboard and type in a password.\n \n \n \n \n \n • Cost-effective: The cost is between $20 and $50 per token, depending on the size of the setup. \n \n \n \n \n • Compact and convenient: Only the size of a car key, eToken is portable. \n \n \n \n \n • Highly secure: Credentials are stored in a tamper-proof container, providing a higher degree of \n \n"
        },
        {
            "page_number": 58,
            "text": "security than software-only solutions.\n \n \n \n \n • Versatile: eToken can contain a large number of private keys for different applications.\nBiometrics \n \n \n \n \n \nBiometrics identifies people based on unique physical characteristics or behavioral traits using body \nfeatures such as fingerprints, eyes, or faces, and ways of doing things, such as speaking or writing \nsignatures. The catalyst for biometrics is the growing realization that passwords can get lost, \nmisplaced, forgotten, or written on notes and stuck on computer monitors for anyone to use. Valuable \ncorporate data is often compromised as a result.\n \n \n \n \n \n \nBiometric technology has not yet gained wide acceptance. Of the $100 billion spent on private \nsecurity, Mentis Corp., a market research firm in Durham, NC, says, the biometric market totaled \nonly $100 million in 1999. Mentis predicts, however, that the market will grow from 27% to 35% \nthrough 2000 as pattern-recognition software improves, computers become better able to handle \nbiometric applications, and prices fall.\n \n \n \n \n \n The term \"biometrics product\" has multiple definitions: \n \n \n \n \n • A component that captures a human characteristic \n \n \n \n \n • Associated hardware \n \n \n \n \n • Application software \n \n \n \n \n • Image-matching software \n \n \n \n \n • A stand-alone product \n \n \n \n \n • A complete solution \n \n \n \n \n"
        },
        {
            "page_number": 59,
            "text": " • A platform or environment that supports biometric authentication \n \n \n \n \n \nThis technology is still expensive without being as effective as other security products. Moreover, \npeople see having their eyeballs scanned as an invasion of privacy. They fear a Big Brother-like \nagency keeping tabs on fingerprints and voiceprints. Once they understand how limited the data \nneeded is, they are likely to become more comfortable with the technology.\n \n \n \n \n \n Some of the commercially available biometric identification methods are (see Table 3.2): \n \n \n \n \n • Voiceprint: Sound waves generated by an individual speaking a given word or password are \ncompared to stored patterns.\n \n \n \n \n \n • Fingerprints: Ridges on fingers are converted to a digital template that can be compared with \ndatabase records.\n• Palm prints: The ridges of the palm are measured for comparison with a database \n \n \n \n \n • Hand geometry: Measures the size and shape of the hand \n \n \n \n \n • Hand veins: The vein pattern on the back of the hand is scanned, creating a digital template that can \nbe matched against stored patterns\n \n \n \n \n \n • Handwriting acoustic emissions: Analyzes sounds generated when a person signs his or her name \n \n \n \n \n • Iris map: A video image of the colored portion of the eye is mapped by computer, creating a digital \ncode based on the individual pattern of the ins\n \n \n \n \n \n • Facial thermographs: Use an infrared camera to capture heat emission patterns, there is a unique \nsignature when heat passes through facial tissue\n \n \n \n \n \n • Facial identification: Converts a video image of the face to a digital template to be compared with a \nrecorded image.\n \n \n"
        },
        {
            "page_number": 60,
            "text": " \n \n \n TABLE 3.2: Common biometric techniques and how they rate \n \n \n \n \n \n \nUser Criteria\n \n \n \n \n \n \nSystem Criteria\n \n \n \n \n \n \n Intrusiveness  \n \n \n \n \n \nEffort\n \n \n \n \n \n Accuracy  \n \n \n \n \n \nCost\n \n \n \n \n \n Dynamic signature \n \n \n \n \n verification \n \n \n \n \n \nExcellent\n \n \n \n \n \n \nFair\n \n \n \n \n \n \nFair\n \n \n \n \n \n Excellent  \n \n \n \n \n Face geometry \n \n \n \n \n \nGood\n \n \n \n \n \n \nGood\n \n \n \n \n \n \nFair\n \n \n \n \n \n \nGood\n \n \n \n \n \n Finger scan \n \n \n \n \n \nFair\n \n \n \n \n \n \nGood\n \n \n \n \n \n \nGood\n \n \n \n \n \n \nGood\n \n \n \n \n \n Hand geometry \n \n \n \n \n \nFair\n \n \n \n \n \n \nGood\n \n \n \n \n \n \nFair\n \n \n \n \n \n \nFair\n \n \n \n \n \n Passive iris scan \n \n \n \n \n \nPoor\n \n \n \n \n \n Excellent  \n \n \n \n \n Excellent  \n \n \n \n \n \nPoor\n \n \n \n \n \n Retina scan \n \n \n \n \n \nPoor\n \n \n \n \n \n \nPoor\n \n \n \n \n \n Very good  \n \n \n \n \n \nFair\n \n \n \n \n \n Voice print \n \n \n \n \n Very good  \n \n \n \n \n \nPoor\n \n \n \n \n \n \nFair\n \n \n \n \n \n Very good  \n \n \n \n \n Source: International Biometric Group, New York \n \n"
        },
        {
            "page_number": 61,
            "text": " \n  \n \n \n \n \nGenerally, biometric authentication is a two-phase process The first is scanning personal \ncharacteristics of users, such as fingerprints irises, faces, signatures, or voiceprints, into the computer \nKey features are then converted to unique templates, which are stored in the computer as encrypted \nnumerical data In the second phase, the user presents that personal characteristic and the computer \ncompares it with the template in the database. But in the real world, matches will rarely be perfect due \nto extraneous factors and background interference.\nVoiceprint  \n \n \n \n \n \nWith VoiceGuardian from Keyware Technologies (www.biometrics.org), provider of Layered \nBiometric Verification solutions, the enrollment process consists of repeating a pass phrase three \ntimes. An example would be ''My voice is my password.\" The voiceprint may be stored on the \nsecurity server, smart card, or local PC. The second stage of verification is accomplished by speaking \nthe pass phrase one time; the live voice sample is then evaluated against the stored voiceprint. If the \nvoiceprint is accepted, the user can access the secure resources.\n \n \n \n \n \n \nAuthentication is performed through analysis by a verification engine of an individual's speech \npatterns at the phoneme level, looking for the points of inflection and articulation that form a unique \nvoice pattern. Although users do not consider voiceprint as intrusive, its accuracy currently is only \nfair.\n \n \n \n \n \n Fingerprint  \n \n \n \n \n \nYour fingerprints have unique characteristics such a whorls, arches, loops, ridge endings, and ridge \nbifurcations. Verification systems capture the flat image of a finger and perform one-to-one \nverification. Compaq Computer Corp. (www.compaq.com) uses Fingerprint Identification \nTechnology to simplify the log-in process (see Figure 3.9). Compaq claims that up to 50 percent of \ncalls to corporate help desks are related to forgotten or expiring passwords. Furthermore, passwords \ncan be duplicated, forged, or stolen. Compaq's Fingerprint Identification Technology makes two \nconnections to your PC, one between the keyboard and the PC and the other to the parallel port. It \nuses software algorithm technologies to convert the image into a unique map of minutiae points.\n \n \n \n \n \n \nMinutiae points are unique data points that describe the fingerprint. The encrypted map, rather than \nthe actual fingerprint, is stored within the network (see Figure 3.10). Since you cannot recreate the \nfingerprint from the data, your fingerprint image is never stored anywhere on the network.\n \n \n \n"
        },
        {
            "page_number": 62,
            "text": " \n \n \n \n \n \n \nFigure 3.9: \nCompaq Fingerprint \nIdentification Technology \nSource: Compaq Computer \nCorporation\n \n \n \n \n \n \n \n \n \n \n \n \nFigure 3.10: \nMinutiae Points \nSource: Compaq Computer Corporation\n \n \n \n \n \n \nAn authorized user only has to register those unique minutiae points once to the server to set up a \ndigital record. To get logged on to the network, the user only has to place the registered finger on the \nreader attached to his or her PC. Fingerprint Identification Technology is compatible with both \nWindows 95 and Windows NT Workstation 4.0.\nSince fingerprint authentications are unique, not easily copied, and relatively inexpensive, they are \nmore likely to have widespread use than other biometric solutions. Fingerprints can be used in a wide \nrange of applications, such as the Internet e-commerce, credit card, and ATM authentication. Once the \ncost of biometric systems comes down enough and standards mature, for many passwords may well \nbecome a thing of the past.\n \n \n"
        },
        {
            "page_number": 63,
            "text": " \n \n \n Iris Recognition  \n \n \n \n \n \nYour iris, the colored portion of your eye, is one of a kind. Even your own right and left iris patterns \nare completely different. Each iris contains more than 266 measurable discriminators (a fingerprint \ncontains about 35) (see Figure 3.11). Moreover, the iris remains stable, protected, and virtually \nunchanged from 18 months of age until death.\n \n \n \n \n \n \n \n \n \n \n \n \nFigure 3.11: \nIris Identification \nSource: Sensar Inc.\n \n \n \n \n \n \nAlthough the biometric identifier such as an iris offers the convenience of not having to carry a photo \nID and various cards, keys, and codes, some people worry that we are moving toward a world where \nthe price of convenience is our personal privacy. The public considers iris scan technology as \nintrusive because the camera is taking a picture of one's eye.\n \n \n \n \n \n \nThe two types of iris recognition systems are active and passive. The active system must be manually \nfocused and the user must be close to the camera. The price of the active iris scan system is several \nthousand dollars, including the camera and the software necessary to run the application on the PC.\nThe passive system is substantially easier to use: It incorporates a set of cameras to automatically \nlocate the user's face and eye, removing the need to manually focus the camera. Passive iris scan is the \nmost expensive biometric technology—it costs tens of thousands of dollars—because the technology, \navailable only for a few years, has not yet been mass-produced.\n \n \n \n \n \n \nThe passive Sensar Iris Identification System (see Figure 3.12) uses three video cameras to get a \nhigh-quality image from as far away as three feet. It maps the iris and converts it into a digital bar \ncode in less than two seconds. The system consists of two modules, the Identification Optical \nPlatform and the Identification Process Platform, connected by a cable up to 10 feet long. Sensar \n \n"
        },
        {
            "page_number": 64,
            "text": "claims that the matching probability of the Iris Identification System is greater than that of DNA \ntesting.\n \n \n \n \n \n \n \n \n \n \n \nFigure 3.12: \nIris Identification System \nSource: Sensar Inc.\n \n \n \n \n \n \nBank United, the largest bank headquartered in Texas, is the first bank in the United States to \nintroduce Iris Recognition ATMs. Thousands of consumers in Houston, Dallas, and Fort Worth can \nwithdraw cash from their accounts at the ATM just by looking at it; the camera will instantly \nphotograph the customer's iris. If the iris data matches the record stored at the time of enrollment, \naccess will be granted. Positive identification can be read through glasses, contact lenses, and most \nsunglasses. As more people rely on iris recognition and as the prices fall, we are likely to see more \nuse of this technology.\nFace Recognition  \n \n \n \n \n \nYour face is yours alone. Securing your computer network through face recognition makes it easy to \nunlock access to your company's applications and data. Miros Inc. in Wellesley, MA, \n(www.miros.com) provides TrueFace Network (see Figure 3.13), which incorporates neural network \nrecognition technology with True-Face Isolator, the first automated neural network face location.\n \n \n \n \n \n \n \n \n"
        },
        {
            "page_number": 65,
            "text": " \n \n \n \nFigure 3.13: \nTrueFace Network \nSource: Miros Inc.\n \n \n \n \n \n Selecting a Biometric System  \n \n \n \n \n \nMany applications of biometrics are in use today. Before deciding on whether to use biometrics in \nyour organization, review the following steps recommended by the Biometrics Consortium \n(www.biometrics.org):\n \n \n \n \n \n • Identify the level of security in the current application, measuring five basic parameters. \n \n \n \n 1. Total elapsed time taken for enrolling a person \n \n \n \n 2. Total elapsed time taken by an individual to successfully use the application \n \n \n \n 3. Percentage of false rejections \n \n \n \n 4. Percentage of false acceptances \n \n \n \n \n 5. Uniformity of performance of the application across the population who will use the \napplication\n \n \n \n \n \n • What improvements are required in the future application for each of the five parameters? Define \nthe target figures for the new application.\n \n \n \n \n \n • Select a biometric and a manufacturer (lists of reputable manufacturers are available from the \nBiometrics Consortium).\n \n \n \n \n \n • Perform a \"black box\" screening test run by your own employees or an independent consultant to \n \n"
        },
        {
            "page_number": 66,
            "text": "see how well the biometric device performs against your defined parameters.\n \n \n \n \n • Implement the biometric fully, with integration, implementation, planning, and testing. \n \n \n \n \n Questions to be answered: \n \n \n \n \n • Biometric type \n \n \n \n — How suitable is the biometric to the application? \n \n \n \n — Are there any persons or groups who cannot use the biometric? \n \n \n \n \n • Biometric device \n \n \n \n — Is the price of the device likely to fit the proposed budget? \n \n \n \n — Is the size of the biometric device significant? \n \n \n \n \n • Biometric manufacturer \n \n \n \n — What is the reliability, quality of service, and reputation of the manufacturer? \n \n \n \n — Who are the manufacturer's major clients? \n \n \n \n \n • Fraud potential \n \n \n \n — Can the application be cheated if the criminal learn as much about the application as the \nmanufacturer knows?\n \n"
        },
        {
            "page_number": 67,
            "text": " \n \n \n — If the biometric is used for verification, can the token be forged easily? \n \n \n \n \n Remote Access Security \n \n \n \n \n \nBanks of modems have typically provided users remote access to corporate networks by dialing \nthrough the public switched telephone network. This is expensive in terms of equipment, support \ncosts, and the cost of telecommunications service. With the Internet, companies can dramatically \nreduce their remote access costs by implementing VPN technology that fosters business-to-business \nintegration.\nExtending your network to the outside world raises the question of how to protect network security and \ndata integrity. Contivity Extranet Switch 2500 from Bay Networks (www.nortelnetworks.com) provides \nsecure Extranets for business partners and secure connectivity for remote users, yet with industrial-\nstrength remote access for corporate travelers and trading partners. It also provides the functionality, \nperformance, and security of a secured Extranet.\n \n \n \n \n \n Contivity Extranet Switch 2500 has the following benefits: \n \n \n \n \n • Cost-effective: You can leverage the cost by using the Internet as the network infrastructure to \nprovide a secure Extranet at both the central site and the branch office.\n \n \n \n \n \n • Flexible and easy to use: Designed as a \"plug and play\" device, it allows you to work with existing \nnetwork infrastructure components such as routers, firewalls, and authentication servers.\n \n \n \n \n \n • High performance: It's designed with all the hardware and software necessary to support VPN \ntunnels, encryption, compression, and filters for up to 400 simultaneous active users.\n \n \n \n \n \n • Flexible security: The security architecture is integrated but ensures that only authorized users can \naccess the network.\n \n \n \n \n \n • Standards-based technology: It supports constantly changing VPN and Extranet technology \nstandards.\n \n \n \n"
        },
        {
            "page_number": 68,
            "text": "  • High availability: It allows up to 400 simultaneous users.  \n \n \n \n \n • Flexible management: It's fully configurable from any browser using HyperText Markup Language \n(HTML) and Java configuration and monitoring.\n \n \n \n \n \n • Broad client support: It supports industry-standard PPTP clients from third parties. \n \n \n \n \n \nContivity Extranet Switch 2500 supports leading certificate authorities such as Entrust; radius \nproviders, such as Bay Secure Access Control, Funk, and Merit, and token providers like Security \nDynamics, AXENT, Secure Computing, and LeeMah Datacom, so that you can easily incorporate the \nExtranet into your existing security infrastructure. It can use either its own internal lightweight \ndirectory access protocol (LDAP) server or an external LDAP server such as Netscape Directory \nServer to authenticate and differentiate among users.\nIntrusion Detection System \n \n \n \n \n \nThough many network security products like firewalls and authentication systems provide critical \nsecurity, they offer limited visibility in the network data stream: Hackers could be using alternate \nmeans of access to your networks, such as dialing into a modem, dialing into someone's PC, or \nattacking from inside the organization. The NetRanger system from Cisco is the industry's first \nenterprise-scale, real-time network intrusion detection system to report and terminate unauthorized \nactivity throughout a network. The NetRanger system can operate in both Internet and Intranet \nenvironments.\n \n \n \n \n \n \nThe NetRanger system has two components, the sensor and the director. The sensors, which are high-\nspeed network appliances, analyze the content and context of individual packets to determine if \ntraffic is authorized. If activity appears suspicious, suggesting a SATAN attack, a ping sweep, or the \ntransmission of a secret research project code word, the sensors would then forward alarms to the \ndirector management console and remove the offender from the network. The director, a high-\nperformance software-based management system, centrally monitors the activity of multiple sensors.\n \n \n \n \n \n \nThe proactive response functionality of the sensor allows users to automatically eliminate specific \nconnections identified with unauthorized activity. The director remotely controls the configuration of \nthe sensors from one centralized location. It can feed alarm information into a database archive to \ngenerate custom graphs and reports.\n \n \n \n \n \n The NetRanger system reduces costs and ensures consistent security enforcement. In the NetRanger \nsystem:\n \n \n"
        },
        {
            "page_number": 69,
            "text": " \n \n \n • Real-time intrusion detection is transparent to legitimate traffic and network usage. \n \n \n \n \n • Real-time response to unauthorized activity blocks offenders from accessing the network or \nterminates the offending sessions.\n \n \n \n \n \n • A comprehensive attack signature list detects a wide range of attacks and can detect content and \ncontext-based attacks.\nLocking down Computers \n \n \n \n \n Steel Security Cables  \n \n \n \n \n \nIn order to stop someone from physically taking your computer, you need to lock it down. Steel \nsecurity cables like those from Computer Security Products (www.computersecurity.com) are an easy \nway to deter hardware theft (see Figure 3.14). The company's PC Security Kits come with a hex \nfastener to protect internal components like RAM from theft. The disc fasteners attach to your \ncomputer. A cable can be secured to any immovable object.\n \n \n \n \n \n \n \n \n \n \n \n \nFigure 3.14: \nPC Security Kits \nSource: Computer Security Products Inc.\n \n \n \n \n \n Universal Drive Lock  \n \n \nIn order to stop unauthorized users from transferring a virus by floppy disk, you have to secure the \n"
        },
        {
            "page_number": 70,
            "text": "  \nfloppy drive on your computer. Universal Drive Lock from Innovative Security Products, \n(www.wesecure.com) provider of security devices for computers and office equipment, will lock up \nyour external or internal drives, including nearly all floppy (see Figure 3.15), CD-ROM, Syquest, \nIomega, tape back-up, Mac floppy with flat bezels, and notebook/laptop drives. Universal Drive Lock \nlets you leave a diskette or CD in the drive while using the lock.\n \n \n \n \n \n \n \nFigure 3.15: \nUniversal Drive Lock \nSource: Innovative Security \nProducts\n \n \n \n \n \n Universal Drive Lock prevents: \n \n \n \n \n • Introduction of an external virus to your computer and network \n \n \n \n \n • Removal of sensitive files by unauthorized individuals \n \n \n \n \n • Introduction of unauthorized software to PCs and networks \n \n \n \n \n The Ultimate Security Kit  \n \n \n \n \n \nThe Ultimate Security Kit (see Figure 3.16) from Innovative Security Products will both lock down a \ncomputer and sound an alarm. The kit will protect your PC or Mac with a 100+ decibel alarm if the \nunit is tampered with or the cable is cut. Two steel lock-down plates and a strong liquid adhesive \noffer up to 1,000 lbs. of holding strength.\n \n \n \n \n \n The Ultimate Security Kit prevents: \n \n"
        },
        {
            "page_number": 71,
            "text": " \n \n \n • Theft of the monitor \n \n \n \n \n • Theft of internal components such as the CPU \n \n \n \n \n • Theft of the keyboard \n \n \n \n \n • Theft of the printer or other peripheral \n \n \n \n \n \n \n \nFigure 3.16: \nThe Ultimate Security Kit \nSource: Innovative Security Products\n \n \n \n \n \n Laptop Security \n \n \n \n \n \nComputer theft is second only to auto theft: In 1995 there were 208,000 notebook computers stolen \nin the U.S.; the losses in hardware, software, and data amounted to approximately $2 billion. Since \nlaptops are becoming lighter and smaller, thieves have a greater chance to abscond with one within a \nfew seconds. The data stored on a laptop can be more valuable than the laptop itself, especially if you \nare the target of industrial espionage.\n \n \n \n \n \n Here are some ways to help keep your laptop secure: \n \n \n \n \n • Keep good records. Serial numbers and a detailed description will help the police trace a stolen \n \n"
        },
        {
            "page_number": 72,
            "text": "laptop and help you make an insurance claim.\n \n \n \n \n • Using an etching tool, mark your name and phone number on your laptop. \n \n \n \n \n • Never leave a laptop in a hotel room unattended. If the hotel has a safe, keep it there. \n \n \n \n \n • Buy separate insurance if your business or home insurance does not cover your laptop.\n• Keep your laptop in an ambiguous carrying case that looks like a traditional briefcase or any piece of \ncarry-on luggage.\n \n \n \n \n \n Security Devices  \n \n \n \n \n \nOne preventive measure is a security cable. Master Lock and Kensington (www.kensington.com) \nhave jointly developed the Universal Notebook Security Cable (see Figure 3.17), a simple system for \nquick and easy installation. The lock fits into your computer's built-in security slot, a feature \ndeveloped by Kensington that is now an industry standard.\n \n \n \n \n \n \n \n \n \n \n \n \nFigure 3.17: \nKensington Universal Notebook \nSecurity Cable Kit \nSource: Computer Security Products \nInc.\n \n \n \n \n \n To secure your portable computer, insert the lock into the built-in slot in your computer and wrap the \ncable around an immovable object like a table or desk. If no slot is available, the Security Slot \n \n"
        },
        {
            "page_number": 73,
            "text": "Adapter Kit can be installed to secure the device.\n \n \n \n \n \nComputer Security Products, Inc., has a product, CompuTrace Theft Recovery Software, to recover \nstolen laptops. CompuTrace, which works with any PC equipped with a modem, silently and \nperiodically calls the CompuTrace Monitoring Center. The software is loaded onto the hard disk with \na leading-edge stealth technology that hides from detection even by anti-virus software. Once \nactivated, CompuTrace will make regular calls to a toll-free tracking line, logging and archiving every \ncall for analysis.\nCompuTrace is scheduled to call its toll-free tracking line every five to seven days. It will detect when \nthe modem is connected and not in use before calling. When it initiates a call, CompuTrace disables the \nmodem speaker. With each call, the computer's serial number and originating telephone number are \nlogged. If the laptop has been reported stolen, CompuTrace will use its theft recovery procedure to \nlocate it. The Theft Recovery Team coordinates the recovery of the stolen computer with local law \nenforcement.\n \n \n \n \n \n Since CompuTrace does not appear in any directory, it is virtually impossible to remove. It cannot be \nerased off the hard disk by deleting files, formatting the hard disk, or even partitioning the hard disk.  \n \n \n \n \n CompuTrace: \n \n \n \n \n • Reduces losses due to computer theft. \n \n \n \n \n • Can be incorporated into a theft deterrent program to reduce losses. \n \n \n \n \n • Makes it easy to manage your company's computer assets. \n \n \n \n \n • Is undetectable, for maximum effectiveness. \n \n \n \n \n • Is fully automated and maintenance-free. \n \n \n \n \n • Is compatible with all phone systems and popular security software products. \n \n \n \n \n Summary \n"
        },
        {
            "page_number": 74,
            "text": " \n \n \n \n \nHardware security is more than locking the door to the computer room. Companies are using the \nInternet, Intranets, and Extranets to exchange information. Their employees use laptops wherever \nthey go, carrying with them proprietary information. Hardware devices such as computer locks and \nsmart cards help companies protect this information. Securing the organization's perimeter with \nalarms, surveillance cameras, and security guards make it transparent to intruders that security is \nenforced. Consider all possible security breaches and implement a security solution that encompasses \nthem all.\n \n \n \n \n \n With any security issue, the battle is ongoing. You will constantly need newer tools and technologies \nto keep one step ahead of all intruders\nReferences \n \n \n \n Backman, Dan, ''Smartcards: The Intelligent Way to Security,\" Network Computing Online (May 15, \n1999). (www.nwc.com/909/909ws/2.html).\n \n \n \n \n Garfinkel, Simson, & Spafford, Gene, Practical UNIX & Internet Security (O'Reilly & Associates, \n1966).\n \n \n \n \n Harrison, Ann, \"Tokens: Not Just for Security Any More,\" ComputerWorld (Mar. 22, 1999), p. 64. \n \n \n \n Innovative Security Products, available on the World Wide Web at www.wesecure.com, 1998. \n \n \n \n An Introduction to Computer Security: The NIST Handbook, Special Publication 800-12, (Feb. 7, \n1996). (www.csrc.ncsl.nist.gov/nistpubs/800-12).\n \n \n \n \n McCooey, Eileen, \"Security Becomes a Priority,\" Windows Magazine (Jan. 1, 1999), p. 50. \n \n \n \n NIST Computer Security Handbook of Cryptography, available at \nhttp://security.isu.edu/isl/hk_crypt.html.\n \n \n \n \n Pasmore, Frank, \"Weaving a Security Web,\" Information Security Magazine, available at \nwww.infosecuritymag.com, April 1998.\n \n"
        },
        {
            "page_number": 75,
            "text": " \n \n \n Yasin, Rutrell, \"Better Snapshots of Enterprise Security,\" InternetWeek (Feb. 22, 1999), p. 23.\nChapter 4*— \nSoftware Security \n \n \n \n \n Vital information needs to flow freely from both inside and outside the company. But vital \ninformation is often both sensitive and confidential. It must be safeguarded.\n \n \n \n \n \n Security Breaches \n \n \n \n \n \nSecurity breaches are up in every category, says a 1998 Information Security industry study. The \nstudy surveyed 1,050 readers who worked in computer/data processing, education, finance/insurance, \ngovernment, the law, manufacturing, health care, or the military.\n \n \n \n \n \n \nThree out of four organizations experienced a virus in 1998, up from 68 percent in 1997. Employee \naccess abuses are the second most common type of breach after viruses (Figure 4.1). Disgruntled \nemployees can create havoc for companies with sensitive information. Nearly one in five companies \nexperienced a dramatic increase in leakages of proprietary information, Information Security said.\n \n \n \n \n \n \nAlthough three out of four organizations experienced a virus last year, nine out of ten have a virus \nprotection product in place (Table 4.1). With the increased growth in the Internet, Internet security is \nanother major problem your company has to deal with, yet as Table 4.1 shows, only six out often \norganizations have in place Intemet/Intranet/Web security. That is likely to change as more \norganizations have a Web presence and engage in e-commerce. Table 4.2 shows the products and \nservices organizations plan to purchase in the near future.\n \n \n \n \n*This chapter was coauthored by Robert Fonfeder, Ph.D., CPA, professor in the School of Business at Hofstra \nUniversity.\n \n \n \n"
        },
        {
            "page_number": 76,
            "text": " \n  \n \n \n \n \n \nFigure 4.1: \nIndustry Breaches; Types of Breaches Experienced \nSource: Information Security\n \n \n \n \n TABLE 4.1: Top 10 Information Security Products and Services Now In Use \n \n \n \n \n Virus Protection \n \n \n \n \n 91% \n \n \n \n \n Backup storage \n \n \n \n \n 90% \n \n \n \n \n Access controls \n \n \n \n \n 85% \n \n \n \n \n Physical security \n \n \n \n \n 80% \n \n \n \n \n Firewalls \n \n \n \n \n 74% \n \n \n \n \n \n \n \n"
        },
        {
            "page_number": 77,
            "text": " Client/server security \n \n 73% \n \n \n \n \n LAN/WAN security \n \n \n \n \n 67% \n \n \n \n \n Disaster-recovery services \n \n \n \n \n 61% \n \n \n \n \n E-mail security \n \n \n \n \n 61% \n \n \n \n \n Internet/Intranent/Web security \n \n \n \n \n 60% \n \n \n \n \n Source: Information Security \n \n \n \n TABLE 4.2: Top 10 Products and Services Organizations Plan to Buy in the \n\"Near Future\"\n \n \n \n \n \n Encryption \n \n \n \n \n 32% \n \n \n \n \n Training/education \n \n \n \n \n 28% \n \n \n \n \n Virtual private networks (VPN) \n \n \n \n \n 27% \n \n \n \n \n Internet/Intranet/Web security \n \n \n \n \n 27% \n \n \n \n \n Firewalls \n \n \n \n \n 26% \n \n"
        },
        {
            "page_number": 78,
            "text": " \n  \n \n \n \n E-mail security \n \n \n \n \n 25% \n \n \n \n \n Smart cards \n \n \n \n \n 21% \n \n \n \n \n Disaster recovery services \n \n \n \n \n 21% \n \n \n \n \n Network/communications security \n \n \n \n \n 18% \n \n \n \n \n Client/server security \n \n \n \n \n 18% \n \n \n \n \n Source: Information Security \n \n \n \n \n Malicious Codes \n \n \n \n \n IT managers need to address a variety of malicious codes. The best known are viruses, but others are \nworms, Trojan horses, droppers, and bombs.\n \n \n \n \n \n Worms  \n \n \n \n \n \nA worm is a program on either a non-networked or networked computer that replicates itself but does \nnot infect other programs. In a network setting a worm copies itself to and from floppy disks or \nacross network connections. In a non-networked computer it can copy itself to different locations on \nhard disks. In either case, worms often steal and vandalize computer data.\n \n \n \n \n \n Trojan Horses  \n"
        },
        {
            "page_number": 79,
            "text": " \n \n \n \n \nA Trojan horse is a bad program hidden inside a seemingly harmless program. When that program is \nrun, the Trojan horse launches in order to perform an unexpected task. Trojan horses do not replicate \nthemselves, but they can steal passwords, delete data, format hard drives, or cause other problems.\nDroppers  \n \n \n \n \n \nA dropper is a program designed to avoid anti-virus detection, usually by encryption that prevents \nanti-virus software from noticing it. Typical functions of droppers include transport and installation \nof viruses. They wait on the system for a specific event, at which point they infect the system with \nthe virus they contain.\n \n \n \n \n \n Bombs  \n \n \n \n \n \nA bomb is a malicious script or scheduling program that activates when a specific event occurs. \nSome bombs activate at a specific time, typically using the system clock. A bomb could be \nprogrammed, for instance, to erase all DOC files from your hard disk on New Year's Eve, or to pop \nup a message on a famous person's birthday. Another bomb might wait for the twentieth time a \nprogram is launched and then erase the program's template files.\n \n \n \n \n \n Viruses  \n \n \n \n \n A computer virus is a program that replicates and spreads by attaching itself to other programs. When \nthe infected program is run, the virus executes an event, which can be:\n \n \n \n \n \n • Benign, displaying a message on a certain date \n \n \n \n \n • Annoying, slowing performance or altering the screen display \n \n \n \n \n • Catastrophic, erasing files, destroying data, or crashing systems \n \n \n \n \n \nViruses can be spread through droppers and use the worm idea to replicate themselves. They can act \nas Trojan horses by attaching themselves to an existing program, hiding inside the program, \nlaunching when the program launches, and committing unwanted acts.\n \n"
        },
        {
            "page_number": 80,
            "text": " \n \n \n \n The different types of viruses include: \n \n \n \n \n \n• Boot sector viruses, infectors residing on a hard disk that are read and executed by the computer at \nboot time. Boot sector viruses can infect either the DOS boot sector or the master boot record, being \nloaded into memory during the boot process.\n \n \n \n \n \n • File infectors are parasitic viruses that attach themselves to executable files. These viruses wait in \nmemory for you to run another program. They replicate simply through the use of the computer.\n• Macro viruses affect programs such as Microsoft Office that ship with programs built-in. When a \ndocument or template containing the macro virus is opened, the virus does its damage. These viruses \nare also programmed to copy themselves into other documents to continually spread.\n \n \n \n \n \n • Multipartite viruses combine boot sector with file infections. \n \n \n \n \n \n• Stealth viruses mislead anti-virus software into acting as if nothing is wrong. They retain \ninformation about the files they have infected, wait in memory, and intercept anti-virus programs \nlooking for altered files, giving the anti-virus programs the old information rather than the new.\n \n \n \n \n \n • Polymorphic viruses alter themselves when they replicate so that anti-virus software looking for \nspecific patterns will not find them all. Those viruses that survive can continue replicating.\n \n \n \n \n \n • Future new viruses are likely to be numerous and unpredictable. \n \n \n \n \n How Viruses Spread \n \n \n \n \n \nViruses come from a variety of sources, and because they can attach themselves to legitimate \nsoftware, they can pass many security defenses. In a 1991 study of major U.S. and Canadian \ncomputer users by Dataquest, the market research firm for the National Computer Security \nAssociation, viruses were shown to spread from the following sources:\n \n \n \n \n \n • 87% of users blamed an infected diskette. \n \n"
        },
        {
            "page_number": 81,
            "text": " \n \n \n • 43% of the diskettes responsible for introducing a virus into a corporate computing environment \nwere brought from home.\n \n \n \n \n \n • 6% of infected diskettes were demo disks, diagnostic disks used by service technicians, or shrink-\nwrapped software disks.\n \n \n \n \n \n • 71% of infections occurred in a networked environment, making rapid spread of viruses a serious \nrisk.\n \n \n \n \n \n • 7% had come from software downloaded from an electronic bulletin board service. \n \n \n \n \n \nViruses can be downloaded from trial programs, a macro from a specific program, or an attachment to \nan e-mail message. A virus delivered as an e-mail attachment, however, does nothing until you \ndouble-click on the attachment One way to protect yourself from this virus is never to open \nattachments that are executable files or data files for programs.\nSymptoms of Viruses \n \n \n \n \n \nMost common viruses give off no symptoms of infection; anti-virus tools are necessary to identify \nand eradicate them However, the many viruses that are flawed do provide some tip-offs to their \npresence Symantec, maker of Norton anti-virus software, \n(www.Symantec.com/avcenter/security/index.html) suggests that you watch for:\n \n \n \n \n \n • Changes in the length of programs \n \n \n \n \n • Changes in the file date or time stamp \n \n \n \n \n • Longer program load times \n \n \n \n \n • Slower system operation \n \n \n \n \n • Reduced memory or disk space \n"
        },
        {
            "page_number": 82,
            "text": " \n \n \n \n • Bad sectors on your floppy \n \n \n \n \n • Unusual error messages \n \n \n \n \n • Unusual screen activity \n \n \n \n \n • Failed program execution \n \n \n \n \n • Failed system boot-ups when booting or accidentally booting from the A drive \n \n \n \n \n • Unexpected writes to a drive \n \n \n \n \n Anti-Virus Software \n \n \n \n \n Norton Anti-Virus  \n \n \n \n \n \nTo fully protect your corporate network from viruses, you will need to install anti-virus software \nNorton Anti Virus software from Symantec is a good choice for Windows workstations and servers \nFeatures in version 5.0 let you quarantine infected files and automatically protect you against not \nonly viruses but also malicious ActiveX and Java applets. Norton Anti Virus runs in the background.\n \n \n \n \n \n Some key features of Norton Anti Virus 5.0 are: \n \n• Quarantine: It isolates infected files in a safe corner of your computer until you can repair the \ndamage.\n \n \n \n \n \n • Scan and deliver: It easily sends quarantined or other suspicious files to Symantec for evaluation \nand repair.\n \n \n"
        },
        {
            "page_number": 83,
            "text": " \n \n \n • LiveUpDate: automatically retrieves new virus definitions from Symantec as often as once a week. \n \n \n \n \n • Protection against malicious codes: detects and removes dangerous forms of ActiveX code, Java \napplets, and Trojan horses.\n \n \n \n \n \n • 24-hour protection: runs constantly in the background. \n \n \n \n \n Dr. Solomon's Anti-Virus Deluxe  \n \n \n \n \n \nAnother good anti-virus software product is Dr. Solomon's Anti-Virus Deluxe from McAfee. Dr. \nSolomon's WinGuard scanner and NetGuard provide 24-hour virus protection from Internet \ndownloads, shared files, e-mail, floppies, and hard disks. The program detects viruses in compressed \nand archives files. It also includes an SOS disk that lets you boot from a clean diskette, even if your \noperating system will not load. With the SOS disk you can have a virus-free system before you install \nDr. Solomon's anti-virus software. Dr. Solomon provides free automatic update to the most current \nversion of the software and protection updates to the hundreds of new viruses that appear each \nmonth.\n \n \n \n \n \n Total Virus Defense  \n \n \n \n \n \nWith over 22,000 existing viruses and over 300 new ones being created each month, protecting a \ndiverse network from them is no easy task. Total Virus Defense 4.0 is a complete virus security \nsolution from Network Associates, a provider of enterprise anti-virus solutions, and Dr. Solomon, a \ncompany in detection and cleaning technology. Total Virus Defense keeps your network protected \nenterprise-wide, at the desktop, file and groupware servers, and the Internet.\n \n \n \n \n \n \nTotal Virus Defense ensures protection of your PC from all sources of viruses, including floppy \ndisks. Internet downloads, e-mail attachments, networks, shared files, CD-ROM, online services, and \neven most popular compressed file types. The total Virus Defense Suite includes:\n \n \n \n \n \n • VirusScan: detects and removes viruses from desktop clients. \n \n \n \n \n • NetShield: detects virus-infected files transmitted to and from the server to prevent the spread of \nviruses throughout the network.\n"
        },
        {
            "page_number": 84,
            "text": "GroupShield: scans groupware environments (Lotus Notes or Microsoft Exchange) to stop viruses \nbefore they're distributed.\n \n \n \n \n \n • WebShield: scans all inbound and outbound e-mail passing through an SMTP e-mail gateway, \nautomatically cleans or quarantines files, and sends alerts to system administrators.\n \n \n \n \n \n \nThe Total Virus Defense AutoUpdate feature works through scheduled downloads from a central \nserver. Autolmmune lets your company automatically detect, remove, and create a cure for \npreviously unknown viruses. Anti-Virus Informant allows network administrators to proactively \nmonitor virus defense.\n \n \n \n \n \n Firewalls \n \n \n \n \n \nA firewall is a system or group of systems that enforces an access control policy between two \nnetworks. It's a security enforcement point that separates a trusted network from an untrusted one, \nsuch as the Internet (see Figure 4.2). It screens all connections between networks, determining which \ntraffic should be allowed and which should not. The security administrator determines in advance \nhow the firewall should set the rules to separate networks.\n \n \n \n \n \n \n \n \n \n \n \n \nFigure 4.2: \nFirewall \nSource: Network Associates, 1999\nFirewalls commonly: \n \n \n \n \n • Secure network access with a perimeter defense. \n \n"
        },
        {
            "page_number": 85,
            "text": " \n \n \n • Control all connections into and out of the network. \n \n \n \n \n • Filter data according to defined rules. \n \n \n \n \n • Authenticate users and applications before permitting them to access internal resources. \n \n \n \n \n • Log activities for security auditing purposes. \n \n \n \n \n • Actively notify the appropriate people when suspicious events occur. \n \n \n \n \n \nFirewalls not only protect the internal corporate network from the public Internet, they can separate \nindividual departments from the rest of the network. With internal firewalls, security administrators \ncan apply different access control rules across a wide variety of working groups and network subnets, \nthus providing a layer of protection against internal breaches. For example, a firewall can be placed \nin front of the Human Resources department to deter access by other departments to sensitive \npersonnel data.\n \n \n \n \n \n When you connect to the Internet, you're putting three things at risk: \n \n \n \n \n • Your data—the information you keep on the computers \n \n \n \n \n • Your resources—the computers themselves \n \n \n \n \n • Your reputation—which can be affected by others using your identity \n \n \n \n \n \nThe Internet presents marvelous opportunities for people to exchange ideas and information. At the \nsame time, the risks are out there. A firewall is the most effective way to connect to the Internet and \nstill protect your network (Chapman, 1995), though building a firewall requires significant expense \nand effort, and the restrictions it places on insiders can be a major annoyance.\n \n \n \n"
        },
        {
            "page_number": 86,
            "text": "  Once in place, firewalls do offer significant benefits. They allow you to:  \n \n \n \n \n • Focus your security decisions. \n \n \n \n \n • Enforce security policy. \n \n \n \n \n • Log Internet activity efficiently. \n \n \n \n \n • Limit your exposure.\nFirewalls offer excellent protection against network threats, but they are not a complete security \nsolution (Chapman, 1995). Certain threats outside the control of the firewall need to be addressed by \nother means. For example, a firewall works well in protecting your network from outsiders but not \nfrom someone on your side of the firewall—58% of security threats originate from employees, \ncontractors, consultants, and other internal users (see Figure 4.3). Firewalls:\n \n \n \n \n \n • Do not protect you against malicious insiders. \n \n \n \n \n • Do not protect you against connections that do not go through it. \n \n \n \n \n • Do not protect against completely new threats. \n \n \n \n \n • Do not protect against viruses. \n \n \n \n \n \n \n \n"
        },
        {
            "page_number": 87,
            "text": " \n \n \n \nFigure 4.3: \nSecurity Breaches \nSource: Information Week/Price Waterhouse Coopers, 1998\n \n \n \n \n \n Packet Filter Firewalls  \n \n \n \n \n \nMost organizations that want a high-performance yet easy-to-configure firewall will implement \npacket filter firewalls. In packet filtering the firewall checks each incoming or outgoing IP packet \nheader against a table of access control rules. If the address and port information are acceptable, the \npacket proceeds through the firewall directly to its destination. If not, the packet is dropped at the \nfirewall. Elaborate rules can be developed to allow or disallow certain packets.\nPacket-filtering firewalls are fast because they operate at the network layer and make only cursory \nchecks into the validity of a given connection (see Figure 4.4). Unfortunately, they have low security \nand no effect on application vulnerabilities. Once a connection has been approved by the firewall, the \noutside source can connect directly through to the target destination, potentially exposing the internal \nnetwork to attack by direct connection with untrusted external sources.\n \n \n \n \n \n \n \n \n \n \n \n \nFigure 4.4: \nPacket Filter Firewall \nSource: Network Associates White Paper, 1999\n \n \n \n \n \n Application Proxy Firewall  \n \n \n \n \n \nGovernment agencies such as DARPA (Defense Advanced Research Projects Agency) and the U.S. \nDepartment of Defense started looking for a better approach to firewall security. The application \nproxy firewall that resulted provides full application-level awareness of attempted connections by \nexamining everything at the highest layer of the protocol stack (Figure 4.5).\n \n"
        },
        {
            "page_number": 88,
            "text": " \n \n \n \n \nAn application proxy firewall can easily see the granular details of each attempted connection up \nfront and implement security policies accordingly (Gauntlet, NAI, 1999). The built-in proxy function \nterminates the client connection at the firewall and initiates a new connection to the internal protected \nnetwork, thus preventing any direct contact between trusted and untrusted systems. A proxy firewall \nmakes it more difficult for hackers on the outside to exploit vulnerabilities on systems inside the \nfirewall.\n \n \n \n \n \n \n \n \n \n \nFigure 4.5: \nApplication Proxy Firewall \nSource: Network Associates White Paper, 1999\n \n \n \n \n \n Proxy Server  \n \n \n \n \n \nSpoofing is a way of using a valid internal IP address to pretend that the foreign system is inside your \nsecure walls. Packet filtering alone is not enough to stop a break-in attempt because it does not \nprevent spoofing. As a result, many companies have migrated to proxy server firewalls. The proxy \nserver resides between a client application such as a Web browser and a real server. It allows internal \nclients to access the Internet from behind a firewall, intercepting all requests from clients within the \nfirewall and forwarding them to the real server.\n \n \n \n \n \n However, there is a downside to proxy servers: The overhead required to run an application proxy is \nconsiderably higher than a packet filter, which significantly slows network traffic.\n \n \n \n \n \n Hybrid Firewalls  \n \n"
        },
        {
            "page_number": 89,
            "text": " \n \n \n \nIn 1998 a few vendors like Check Point and Network Associates addressed this problem by designing \n''hybrid\" firewalls. As the name suggests, hybrid firewalls combine the best of both worlds: the speed \nof a packet filter and the stringent security features of an application proxy (Foote, 1999). Hybrid \nfirewalls use packet filtering during peak hours when there is a lot of network traffic, but if the \nfirewall perceives a threat, the application proxies take control. Once a connection has been \nestablished through a proxy, all subsequent traffic is filtered, thereby balancing speed with security.\nA Complete Firewall Security System \n \n \n \n \n \nThe Internet can link a company with customers, remote employees, suppliers, and business partners \nat a fraction of the cost of other linkages. To remain competitive means that private networks must be \nable to extend to the outside world.\n \n \n \n \n \n \nWith the increasingly sophisticated security threats from new technologies, having stand-alone \nfirewalls to protect a network is not enough. Integrated network security that actively communicates \nwith other security tools and responds to new attacks by modifying security measures accordingly is \nneeded (see Figure 4.6). The role of the firewall has to evolve from a stand-alone system enforcing \naccess rules into a distributed system integrating components throughout the network.\n \n \n \n \n \n \n \n \n \n \n \n \nFigure 4.6: \nDistributed Firewall System \nSource: Network Associates\nA complete security system must have the following components: \n \n \n \n \n • Firewalls that guard against unauthorized entry from outsiders or individual departments \n \n \n \n \n • Vulnerability scanners that proactively scan the internal network for potential security holes or \n \n"
        },
        {
            "page_number": 90,
            "text": "weaknesses\n \n \n \n \n • Intrusion detection monitors that watch internal network traffic and servers for signs of attack \n \n \n \n \n • Virus and content scanners searching for malicious code such as viruses, Trojan horses, or hostile \nJava and ActiveX applets\n \n \n \n \n \n • A central event manager that meaningfully integrates the distributed security components (all \nmonitors, sensors, and scanners)\n \n \n \n \n \n \nThus companies cannot rely on passive firewalls that guard only the front entrance. Firewalls have to \nwork with other security components to actively respond to changing threats. Network Associates \nnow has an active firewall system called the Gauntlet Active Firewall (see Figure 4.7) that combines \nproactive vulnerability scanning, real-time intrusion detection monitoring, anti-virus scanning, and \nvirtual private networking into a single active system.\n \n \n \n \n \n \n \n \n \n \n \n \nFigure 4.7: \nThe Active Firewall \nSource: Network Associates\nThis scenario can demonstrate how the Gauntlet Active Firewall works: Every morning at 2:00 a.m. the \nvulnerability scanner runs through a series of routine vulnerability tests. During one of these, it \ndiscovers that an employee has set her PC up as an insecure FTP server that exposes data on the private \nnetwork to attack. The scanner forwards this information to the Event Orchestrator so that it can check \nthe customer's security policy to determine the appropriate course of action.\n \n \n"
        },
        {
            "page_number": 91,
            "text": " \n \n \n \nThe Event Orchestrator then instructs the firewall to shut down the FTP server until the IT staff \narrives the following morning. The Event Orchestrator can also send a trouble ticket describing the \nproblem to the help desk. When in the morning the user calls the help desk because her FTP service \nis not working, the IT staff, already aware of that, can then advise her of the security risks her actions \nhave created.\n \n \n \n \n \n Encryption \n \n \n \n \n \nEncryption is the translation of data into a secret code, a method of scrambling information so that \nonly the intended receiver can use it. It uses algorithms to mathematically combine \"keys\" with plain \ntext to form encrypted or cipher text. It's the most effective security device because you must have a \nspecific key to unlock the data.\n \n \n \n \n \n \nAny digital data can be encrypted, including e-mail messages, telephone calls, movies, pictures, and \ncomputer files. Today there are both hardware devices and software packages available for \nencrypting data.\n \n \n \n \n \n \nThere are two main types of encryption: symmetric (secret key) encryption and asymmetric (public \nkey) encryption. Secret key encryption uses one key. Public key encryption uses a pair of keys called \npublic and private keys.\n \n \n \n \n \n Secret Key Encryption  \n \n \n \n \n \nBefore the creation of the public/private key method, a technology called symmetric cryptography \nwas used. Here the sender uses the secret key (a large binary number) algorithm to mathematically \nencrypt the communication and the recipient uses the same key to decrypt it. Secret key encryption is \nrelatively fast and small. It's often generated in Web browsers each time they open a secure \ntransaction. Using secret key cryptography, it's safe to send encrypted messages without fear of \ninterception because it's not likely that the message can be deciphered (see Figure 4.8).\n \n \n \n \n"
        },
        {
            "page_number": 92,
            "text": " \n \n \n \n \nFigure 4.8: \nSymmetric Keys \nSource: Symantec Corporation\n \n \n \n \n \n \nThe problem with this method, however, is figuring out how to deliver the secret key to the recipient \neither through another secure transmission channel or in person. If the secret key is intercepted, all \ncommunications are compromised. Moreover, the infrastructure is impractical for a large number of \npeople. If you have a large organization, you may need thousands (if not millions) of secret keys. In \naddition, secret key cryptography assumes that the parties who share a key will not disclose or \nmodify the key.\n \n \n \n \n \n The method does not ensure the authenticity of the sending party. If Bob receives an encrypted \nmessage, he has no way of proving who sent it.\n \n \n \n \n \n \nThe best-known secret-key system is the Data Encryption Standard (DES) published by the National \nInstitute of Standards and Technology as Federal Information Processing Standards (FIPS) 46-2. \nNIST is an agency of the U.S. Department of Commerce's Technology Administration. FIPS is the \nofficial series of publications relating to standards and guidelines mandated by the Secretary of \nCommerce.\n \n \n \n \n \n \nSince 1977 DES has been the federal government's standard method for encrypting sensitive \ninformation. The algorithm standard has, moreover, evolved from solely a U.S. government \nalgorithm into one that is used globally by the commercial sector. It's the most widely accepted, \npublicly available cryptographic system today.\n \n \n \n \n \n Public Key Cryptography  \n \n \nA major advance occurred with the invention of public key cryptography. Public key (asymmetric) \ncryptography uses a \"public\" key and a \"private\" key that are mathematically tied together. Using this \n"
        },
        {
            "page_number": 93,
            "text": "  \nmethod a message is encrypted with a public key, and the private key is used to decrypt the same \nmessage (see Figure 4.9). The public key may be distributed to the world because it is used only for \nencrypting. Its owner must keep the private key confidential. The encrypted message can only be \ndecrypted with its matching private key. This protocol is used with Secure Sockets Layers (SST) \ntechnology, which is the standard protocol for secure Web-based communications.\n \n \n \n \n \n \n \n \n \n \nFigure 4.9: \nPublic/Private Keys Source Symantec Corporation\n \n \n \n \n \n \nPublic key technology is a breakthrough because it solves the key management problem associated \nwith symmetric key encryption in large organizations. It also reduces the possibility of key \ncompromise during delivery.\n \n \n \n \n \n \nBefore the invention of public key cryptography, it was virtually impossible to provide key \nmanagement to large-scale networks. The number of keys required to secure communications among \nthose users increased substantially as the number of users grew. For example, a network of 100 users \nwould require almost 5,000 keys if it used only secret key cryptography. Doubling the network to \n200 users increases the number of keys needed to almost 20,000.\n \n \n \nPublic key encryption, however, is bulky and slow. It doesn't easily encrypt a file once for a number \nof different people. Users who lose their keys will not be able to decrypt files encrypted with those \nkeys. It also doesn't guarantee that John's public key is not someone else pretending to be John. As a \nresult, most encryption systems use both types of encryption. \nCombination Encryption \n \n \n \n \n \nPublic/private key pairs are very long and time-consuming. To speed the process and make it more \nsecure, both methods—symmetric and public key encryption—are used together. In the scenario \nshown in Figure 4.10, Bill has a message that he wants to send to Maria. Bill sues a symmetric key \nto encrypt the document and then encrypts the symmetric key using Maria's public key.\n \n \n"
        },
        {
            "page_number": 94,
            "text": "  \n \n \n \n \n \n \n \n \n \n \nFigure 4.10: \nEncrypt with Symmetric and Public Keys \nSource: Symantec Corporation\n \n \n \n \n \n When Maria receives the encrypted message from Bill, she uses her private key to decrypt the \nsymmetrical key (see Figure 4.11). Then Maria uses the symmetric key to decrypt the message.\n \n \n \n \n \n \n \n \n \n \nFigure 4.11: \nDecrypt Using Private and Symmetric Keys \nSource: Symantec Corporation\n \n \n"
        },
        {
            "page_number": 95,
            "text": " \n \n \n Encryption in On-line Trading \n \n \n \n \n \nOne of the major uses of public/private key technology is in on-line trading where security is a major \nconcern for both investors and brokers. The top five on-line brokers—E*Trade, DLJdirect, Discover \nBrokerage, Waterhouse, and Datek—ensure the security of customer's account information by using \nencryption to safeguard against hackers intercepting a customer's user ID and password and posing as \nthe actual customer.\n \n \n \n \n \n \nThe on-line brokerage firm keeps one key—the private key (Figure 4.12). It makes the other key -- \nthe public key—available to customers. Anyone who finds the public key can use it only to send the \nfirm a private message (see Figure 4.13). It cannot be used to decrypt messages sent to the firm or to \nimitate the firm. Therefore, the firm can send the public key to customers using e-mail or it can even \npost the key on its Web site.\n \n \n \n \n \n \nFigure 4.12: \nPublic and Private Keys \nSource: VeriSign Inc.\n \n \n \n \n \n \n \n"
        },
        {
            "page_number": 96,
            "text": " \n \n \n \n \nFigure 4.13: \nEncrypt and Decrypt \nSource: VeriSign Inc.\n \nIn public/private key encryption, the customer's computer verifies the broker's server by checking the \nbroker's digital signature and comparing it against the signature on record in a digital certificate. \nDigital certificates are sent by a third-party registrar like VeriSign Inc. Then the customer's and \nbroker's computers use a session key. The session key (Figure 4.14) is used to encrypt all the data \ntransferred during the session. The key ranges in length from 40 to 128 bits, depending on the level of \nsecurity supported by the Web browsers of both parties. This complicated conversion makes an \nintercepted message virtually impossible to decode. Even if a hacker happens to access a customer's \naccount, funds can be transferred only to a bank account the customer has previously specified.\n \n \n \n \n \n \n \n \n \n \n \n \nFigure 4.14: \nSession Key Used to Place an Order \nSource: VeriSign Inc.\n"
        },
        {
            "page_number": 97,
            "text": " \n \n \n \n \n \nFigure 4.15: \nEach Transaction Uses a Session Key \nSource: VeriSign Inc.\n \n \n \n \n \n \nEach transaction uses a session key (Figure 4.15). When a customer buys, the transaction gets \nencrypted using the session key. When the firm receives the order, the firm uses its session key to \ndecrypt it. This process is not contingent on the next transaction. When a customer buys again, a new \nsession key is created. Even if a hacker decodes one transaction, the next will be entirely different.\n \n \n \n \n \n Digital Signatures \n \n \n \n \n \nAuthentication generally has two components, one to prove that the information came from a specific \nuser and the second to prove that the information has not been altered. Authentication is possible \nthrough the use of a digital signature, a technique that uses public key encryption to ''sign\" electronic \ndocuments.\n \n \n \n \n \n \nDigital signature is a code that can be attached to an electronically transmitted message that uniquely \nidentifies the sender. Like a written signature, the purpose of a digital signature is to guarantee that \nthe individual sending the message really is who he or she claims to be. Digital signatures not only \nallow for strong authentication and data integrity but also non-repudiation: When participants transact \nbusiness electronically using digital signatures, they cannot deny the transaction.\n Digital signature schemes have two steps: \n \n \n"
        },
        {
            "page_number": 98,
            "text": "  1. Generating a message digest (\"hash\") of the message  \n \n \n \n \n 2. Generating the signature by combining the message digest with the user's secret key \n \n \n \n \n \nAdding a signature to a message validates the integrity of the message. Digital signature confirms \nthat the message has not been altered since you signed it because any changes to the message after \nyou sign will invalidate the signature.\n \n \n \n \n \n \nAll the trust in the signature, however, is tied to whether or not you can trust the signer's public key. \nIf you cannot verify that the public key belongs to the specific person, corporation, or authorized \nagent, then digital signatures can be forged.\n \n \n \n \n \n Public Key Infrastructure  \n \n \n \n \n \nSince key management and digital signatures are based on public key technologies, good information \nsecurity requires a public key infrastructure (PKI). PKI is a system of digital certificates, certificate \nauthorities, and other registration authorities that authenticate the validity of each party to an Internet \ntransaction. There is no single PKI, or even a single agreed standard for setting up a PKI. \nImplementing public key encryption, digital signature, and other security services on a broad scale \nwill require many certificate authorities.\n \n \n \n \n \n \nEntrust/PKI. Entrust Technologies, Inc., provides the Entrust/PKI to allow people to encrypt, \ndigitally sign, and authenticate electronic transactions across all applications. Entrust/PKI is a \nmanaged PKI. In a managed PKI, the users in one certificate authority (CA) can communicate \nsecurely with users from another CA. A managed PKI is thus a method for maintaining trustworthy \nelectronic relationships between CAs. Entrust/PKI policy management provides security, flexibility, \nand ease of use through a wide range of network policy options.\n \n \n \n \n \n \n \nKerberos. Kerberos is an authentication system created at the Massachusetts Institute of Technology \n(MIT) as a solution to network security problems carried by insiders. Designed to enable two parties \nto exchange private information across an insecure network connection, it works by assigning a \nunique key, called a ticket, to each user that logs on to the network. The ticket is then embedded in \nmessages to identify the sender of the message. Kerberos provides the tools of authentication and \nstrong cryptography to help secure information systems across your enterprise.\nPretty Good Privacy. Pretty Good Privacy (PGP) is a technique for encrypting messages developed by \nPhilip Zimmerman. PGP is one of the most common ways to protect messages on the Internet because \nit is effective, easy to use—and free. PGP is based on the public-key method. To encrypt a message \nusing PGP, you need the PGP encryption package, which is available from a number of sources, \n \n"
        },
        {
            "page_number": 99,
            "text": "including MIT.\n \n \n \n \n \nPGP Enterprise Security Suite, with applications for Win32, Mac, and UNIX environments, is a \ncomplete e-mail, file, disk, and network security product from Network Associates. PGP users can \nsend strongly encrypted files to non-PGP users with self-decrypting files. The Suite integrates \nsecurity and authentication for e-mail, files, and disks by integrating:\n \n \n \n \n \n • PGP VPN Client, which uses either PGP or X.509-based certificates for secure VPN and remote \nInternet access to your network\n \n \n \n \n \n • PGP Desktop Security, which secures e-mail and files by encrypting and digitally signing for data \nprotection and authentication\n \n \n \n \n \n \n• Certificate Server, through which organizations can create and manage a unified public-key \ninfrastructure that enables confidential communications across a corporate Intranet or Extranet or the \npublic Internet\n \n \n \n \n \n • Policy Management Agent, which enforces corporate security policies for e-mail communications \nacross internal and external networks\n \n \n \n \n \n • PGP Software Developers Toolkit, which can incorporate trusted and peer-reviewed PGP \ncryptographic capabilities into new and existing applications\n \n \n \n \n \n \nWith PGP Enterprise Security Suite, e-mails, files, and disks can be secured and authenticated as \nquickly as doing a \"save.\" The Suite also supports current standards such as IPSec and X.509, the \nmost widely used standard for defining digital certificates.\nThe Orange Book \n \n \n \n \n \nTrusted Computer System Evaluation Criteria, better known as the Orange Book, is a U.S. \ngovernment publication that standardizes security system requirements. It defines four broad \ncategories of security for host-based environments:\n \n \n \n \n \n • Minimal security (least) \n \n"
        },
        {
            "page_number": 100,
            "text": " \n \n \n • Discretionary protection \n \n \n \n \n • Mandatory protection \n \n \n \n \n • Verified protection (most) \n \n \n \n \n The objectives of the Orange Book are to: \n \n \n \n \n 1. Provide a way of assessing the level to which you can trust a given computer system. \n \n \n \n \n 2. Provide guidelines to manufacturers as to what to build into their systems to satisfy various \nsecurity needs.\n \n \n \n \n \n 3. Serve as a basis for specifying security requirements so you can purchase a coordinated security \nsystem.\n \n \n \n \n \n In today's complex business environments, though, the Orange Book's security classifications are \nsomewhat limited (Innovative Security Products, 1998).\n \n \n \n \n \n Security in a Distributed Network \n \n \n \n \n \nIn the past IT managers planning to secure their companies' applications and systems had to use a \nvariety of tools to gather all the reports. AXENT Technologies, Inc. (www.axent.com) has alleviated \nthis by creating a product called Enterprise Security Manager 5.0 (ESM). ESM allows administrators \nto check, manage, and enforce security policies from a central console, offloading repetitive tasks \nassociated with managing security policy from human staff members to computers.\n \n \n \n \n \n \nESM is based on manager/agent architecture. Software agents residing in servers and workstations \ngather pertinent data and send them to the ESM console. A single console can collect data from and \nmanage up to 10,000 software agents. ESM console and advanced reporting features provide a \nsnapshot of security levels across the enterprise in a single graph (Yasin, 1999). Only data that is \nabsolutely necessary gets sent between agents and managers, using less networking bandwidth for \n"
        },
        {
            "page_number": 101,
            "text": "security checks. All agents can be run manually on a schedule, depending on what you need to \nproactively manage your corporate computing environment.\nESM offers the following benefits: \n \n \n \n \n • Consistency across platforms: It evaluates all systems based on a standard security policy. \n \n \n \n \n • Broad platform support: It supports over 35 operating systems, including Windows NT, UNIX, \nNetWare, and Open VMS.\n \n \n \n \n \n • Dynamic configuration: It allows you to customize policies to meet the needs of your unique \nenvironment.\n \n \n \n \n \n • Integrated reporting capabilities: It allows you to go from enterprise-wide reporting to individual \nsecurity settings with just a few clicks of the mouse.\n \n \n \n \n \n • Intuitive graphical user interface (GUI): The interface for configuring, managing, and reporting on \nsecurity policies across your enterprise is easy to use.\n \n \n \n \n \n • Fast correction of security problems: You can quickly correct faulty security settings and update \ngroups of systems to security settings mandated in the security policy.\n \n \n \n \n \n • Compatibility with existing systems: It integrates easily with existing security applications and \nprocesses.\n \n \n \n \n \n • Manager/agent architecture: ESM's manager and agents can be different systems, allowing you to \nleverage your existing security systems.\n \n \n \n \n \n • Hierarchical approach: It's scalable to your enterprise network. \n \n \n \n \n Total Network Security \n \n \n"
        },
        {
            "page_number": 102,
            "text": "  \nFor IT managers looking for a one-stop network security solution, IBM's Integrated Security \nSolutions enables companies to turn to a single vendor for all security software, hardware, and \nservices needed when moving their business to the Internet. The system has three parts:\n  \n \n \n \n \n • IBM SecureWay FirstSecure integrates security for Web-based systems with legacy-based systems\n• Tivoli Availability provides uninterrupted network services. \n \n \n \n \n • Tivoli Administration provides centralized, consistent ways to manage a secure network. \n \n \n \n \n The Internet Security Market \n \n \n \n \n \nAccording to a study released March 30, 1999, by International Data Corp. (IDC), the worldwide \nmarket for Internet security software grew from $2 billion in 1997 to an estimated $3.1 billion in \n1998. Security threats continue to increase as more companies provide access to the internal systems \nvia the Internet for things like e-commerce, prompting many companies to increase spending on \nsecurity software (Perez, 1999). Based on the IDC study of 300 companies with more than $100 \nmillion in annual revenues, the 1999 market for Internet security software is estimated at $4.2 billion \nworldwide, growing to $7.4 billion in 2002.\n \n \n \n \n \n \nIn general, the Internet software security market is expected to have a compound annual revenue \ngrowth rate of about 30% between now and 2002, IDC said. IDC reports that the top three segments \nof this market are likely to be:\n \n \n \n \n \n • Firewall, the fastest-growing segment, with an estimated 40% compound annual growth \n \n \n \n \n • Anti-virus software, likely to be the largest segment of the market in 2002, accounting for almost \nhalf of the worldwide revenues from Internet security software\n \n \n \n \n \n • Encryption software, plus software authorization, authentication, and administration \n \n \n \n \n Security Protocols \n \n \n \n \n SSL  \n"
        },
        {
            "page_number": 103,
            "text": " \n \n \n \n \n \nSecure Sockets Layer (SSL) is the Internet security protocol for point-to-point connections. With the \ngrowth of the Internet, many applications need to securely transmit data to remote applications and \ncomputers. SSL was designed to solve this problem. Many popular Web browsers like Netscape \nCommunication and Internet Explorer use SSL to protect against eavesdropping, tampering, and \nforgery. In SSL, when clients and servers make connections, they authenticate each other. Once \nauthenticated, a \"secure pipe\" is established and data can be securely exchanged (see Figure 4.16). \nSSL uses strong encryption technologies from RSA Data Security.\n \n \n \n \n \n \n \n \n \nFigure 4.16: \nSecure Socket Layer \nSource: RSA Data Security\n \n \n \n \n \n Some practical applications of SSL are: \n \n \n \n \n • Client/server systems: Securing database access \n \n \n \n \n • Financial: Remote banking programs \n \n \n \n \n • Information systems: Remote access and administration applications \n \n \n \n \n • Travel industry: Create online reservation systems and secure information transfer \n \n"
        },
        {
            "page_number": 104,
            "text": " \n \n \n IPSec  \n \n \n \n \n \nInternet Protocol Security (IPSec), a set of protocols developed by the Internet Engineering Task \nForce (IETF) for encryption and authentication of TCP/IP traffic, is the leading standard for \ncryptographically based authentication, integrity, and privacy services.\n \n \n \n \n \n \nAt the IP layer, computers on a network communicate by routing datagram packets that contain data, \ndestination addresses, source addresses, and other information. In a corporate LAN or the Internet, \nwhere packet datagrams are transmitted \"as is,\" unencrypted, a malicious attacker could hijack, forge, \nor modify them. IPSec secures the network packets to create a secure network of computers over \ninsecure channels. It enables users to communicate securely with a remote host over the Internet via \nVPNs. Where SSL authenticates and encrypts communications between clients and servers at the \napplication layer, IPSec secures the underlying network layer.\n \n \n Some practical applications of IPSec are: \n \n \n \n \n • VPN software and hardware: Creating secure networks over insecure means of transmission such as \nthe Internet\n \n \n \n \n \n • Remote access software and hardware: Securing access to network resources \n \n \n \n \n • Firewall products: Securing VPN tunneling to link business partners and members of the enterprise \n \n \n \n \n S/MIME  \n \n \n \n \n \nSecure Multipurpose Internet Mail Extensions (S/MIME) is the electronic standard that protects \nmessages from unauthorized interception and forgery. S/MIME uses public-key encryption \ntechnology to secure transmission, storage, authentication, and forwarding of secret data. Where SSL \nsecures a connection between a client and a server over an insecure network, S/MIME is used to \nsecure users, applications, and computers.\n \n \n \n \n \n Some practical applications of S/MIME are: \n \n \n \n \n • Electronic data exchange (EDI): digital signatures on contracts \n"
        },
        {
            "page_number": 105,
            "text": " \n \n \n \n • Financial messaging: store and transfer bank statements \n \n \n \n \n • Content delivery: electronic bill payment \n \n \n \n \n • Health care: secure patient records and health claims \n \n \n \n \n Like SSL and IPSec, S/MIME is based on the RSA algorithm for digital signatures and digital \nenvelopes (see Table 4.3).\n \n  \nTABLE 4.3: Security Protocols Overview \n \n \n \n \n \nProtocol\n \n \n \n \n \n \nSummary\n \n \n \n \n \n \nSSL \n(Secure Socket Layer)  \n \n \n \n \n \nAllows a \"secure pipe\" between any two \napplications for transfer of data and mutual \nauthentication\n \n \n \n \n \n \nIPSec \n(IP Security Protocol)  \n \n \n \n \n \nStandard for cryptographically-based \nauthentication, integrity, and confidentiality \nservices at the IP datagram layer\n \n \n \n \n \n \nS/MIME \n(Secure MIME)\n \n \n \n \n \n \nGuarantees the secure transmission, storage, \nauthentication, and forwarding of secret data at the \napplication level\n \n \n \n \n \n Source: RSA Data Security \n \n"
        },
        {
            "page_number": 106,
            "text": " \n \n \n VPNs \n \n \n \n \n \nVirtual private networks (VPNs) allow communication between sites using a public network such as \nthe Internet to go over an encrypted channel. Since network communication over the Internet is \nvulnerable to \"snooping\" (electronic eavesdropping), setting up a VPN guarantees private \ncommunication between two sites. VPNs also represent cost savings over true private networks such \nas leased line an frame relay because they reduce connectivity costs.\n \n \n \n \n \n A successful VPN deployment is much more than simple data encryption. A complete VPN solution \nincludes:\n \n \n \n \n \n \n• Security: granular access control, user authentication, standards-based data encryption to guarantee \nthe security of network connections, authenticity of users, and privacy and integrity of data \ncommunications\n \n \n \n \n \n • Traffic control: band-width management, quality of service, and acceleration of data \nencryption/decryption to guarantee VPN reliability and performance\n \n \n \n \n \n • Enterprise management: a comprehensive security policy that integrates enterprise-wide VPNs and \ncentralized policy-based management\n Remote Access Security Solution \n \n \n \n \n \nThe Internet has brought many corporations more flexibility in allowing their workers to \ntelecommute or work in remote locations, but the corporations need to ensure that remote users have \naccess to the resources they need without sacrificing corporate network security. They have to \nprevent unauthorized users from gaining access to the corporate network by impersonating legitimate \nusers. AXENT Technologies, Inc., has a complete remote access security solution called Security \nBriefcase for Windows 95 and Windows NT. Security Briefcase provides a complete enterprise \npolicy-based solution by protecting the local disk, preventing unauthorized users from accessing \ncorporate information, and securing the session over the Internet (see Figure 4.17).\n \n \n \n"
        },
        {
            "page_number": 107,
            "text": "  \n  \n \n \n \n \n \nFigure 4.17: \nSecurity Briefcase \nSource: AXENT Technologies Inc.\n \n \n \n \n \n Defender  \n \n \n \n \n \nPasswords may be compromised with hacker tools such as password sniffers, network sniffers, and \ndictionary attacks. Once passwords are stolen, unauthorized users can easily access your files. \nSecurity Briefcase provides two-factor user authentication consisting of a token and a PIN.\nPower VPN  \n \n \n \n \n \nVPNs encapsulate data packets, securing your Internet communications. The Power VPN feature of \nAXENT's Briefcase (see Figure 4.18), which does not require modification of your application \nsoftware or the network, reduces your telecommunications costs by providing access directly to your \nVPN through any ISP, eliminating the need for maintaining costly toll-free numbers. Users dial into a \nlocal ISP and tunnel securely to the corporate network.\n \n \n \n \n \n \n \n"
        },
        {
            "page_number": 108,
            "text": " \n \n \n \n \nFigure 4.18: \nComparison Between Firewall Only and PowerVPN \nSource: AXENT Technologies Inc.\n \n \n \n \n \n \nPowerVPN consists of a client that runs on a PC or laptop and a gateway behind the firewall (see \nFigure 4.18). The secure and authenticated connection is between the PC that uses the PowerVPN \nencryption session to connect to the Power VPN gateway at the corporate network. Unlike VPN from \nother firewall vendors, Power VPN operates independent of the firewall, making firewalls instantly \ncompatible with PowerVPN. PowerVPN increases throughput by compressing the data before \nencrypting it.\n PCShield  \n \n \n \n \n \nThe PCShield feature of the Security Briefcase protects data on your PC from unauthorized users by \nproviding automatic file encryption and decryption, and centralized key management, all in one \nproduct. PCShield provides centralized administration of PC and laptop security. A central database \ncontains all user accounts, groups, file access control rules, and audit standards, making it possible to \nknow which computers each user can access. This entire secure authentication is integrated into the \nproduct, making the process transparent to the Windows operating systems and applications.\n \n \n \n \n \n \nPCShield automatically protects new files from unauthorized access. It secures files that are stored \nlocally, on file servers, on floppy drives, or even transmitted across the network. PCShield adds an \naccess control label to each protected file so that the protection stays with the file wherever it resides. \nIt also encrypts data automatically. When a PC is left unattended or connected to the network, users \nmust re-enter their passwords to regain access to the system. Information cannot be accessed on a \nprotected stolen laptop.\n \n \n \n \n \n Doing Business over the Internet \n"
        },
        {
            "page_number": 109,
            "text": " \n \n \n \n Certificate Authorities  \n \n \n \n \n \nBusiness is built on trust. On the Internet that trust is easily compromised. To secure your data over \nthe Internet, certificate authorities (CA) are the only way to go (Bruno, 1998). A CA with software \nthat can generate and manage digital certificates can either be built in-house or outsourced to a third \nparty. To build an in-house CA server, you must:\n \n \n \n \n \n • Have a good working knowledge of public-key cryptography \n \n \n \n \n • Decide on either top-down CA or cross-certification \n \n \n \n \n • Weigh performance and scalability \n \n \n \n \n • Make sure the CA software uses a standard certificate to ensure that different packages can \ncommunicate\n \n \n \n \n \n • Make CAs comply with LDAP (lightweight directory access protocol) to make it easier to store \ncertificates\n \n \n \n \n \n • Decide on price and protection \n \n \n \n \n • Decide on whether to control your own security \n \nIn deciding whether to outsource CA, you must: \n \n \n \n \n • Decide on what people and budget it takes to support certificates \n \n \n \n \n • Analyze the speed of deployment \n \n \n"
        },
        {
            "page_number": 110,
            "text": "  • Leverage the third party's expertise and high-security facilities  \n \n \n \n \n • Determine if the burden of handling CA should be placed on a third party \n \n \n \n \n • Decide on whether to control your own security \n \n \n \n \n • Decide which third party is most trustworthy in its policies and practices \n \n \n \n \n \nWhether developed in-house or outsourced, a CA has a straightforward task: It verifies the identities \nof end-users by issuing certificates—unique, encrypted, digital Ids—that are attached to e-mail, \ntransaction records, or files sent over the Internet or Intranet (Bruno, 1998). A certificate is like a \ndigital passport that lets the recipient know that the sender is who he or she claims to be.\n \n \n \n \n \n All CAs are based on public-key cryptography; they typically have three components: \n \n \n \n \n • A database that stores public keys \n \n \n \n \n • A cryptographic engine that generates the actual certificate \n \n \n \n \n • A PKI (public key infrastructrure) engine that tracks the expiration date of issued certificates \n \n \n \n \n E-mail Security \n \n \n \n \n \nE-mail has become such an integral part of the working world that it also poses significant security \nthreats. The accessibility given by the Internet provides hackers with a direct line to personal and \ncorporate in-boxes (Paone, 1999). Malicious code can be attached to incoming messages; a simple \nmouse click by the targeted user is all it takes to activate them. Spam (junk e-mail) can flood in-\nboxes with time wasting, unnecessary, or inappropriate content (Paone, 1999). Spam attacks can clog \nLANs or strain the Internet gateway.\n \n \n \n \n \nYet many companies and individuals do not give much thought to e-mail security. Unlike other \n"
        },
        {
            "page_number": 111,
            "text": " \nsecurity initiatives, e-mail is a more ongoing challenge than installing a firewall (Paone, 1999). One \ntool that can be used in desktop and corporate settings is virus-scanning software, but there is no \nsingle methodology that can protect against all e-mail-based threats (see Table 4.4).\n \n  \n TABLE 4.4: E-mail Security Holes \n \n \n \n \n Vulnerability \n \n \n \n \n \nProblem\n \n \n \n \n \n Possible solutions \n \n \n \n \n Unencrypted e-\nmail\n \n \n \n \n \n \nCan be read by unknown and unauthorized \nthird parties while it traverses the Internet, \nor even by unauthorized employees within \nthe Internet\n \n \n \n \n \n \nE-mail encryption, \nvirtual private \nnetworking solutions\n \n \n \n \n \n \nMalicious \nattachments and \nviruses\n \n \n \n \n \n \nUnsuspecting users may open e-mail \nattachments from unfamiliar senders and \nunleash viruses or malicious active content \nonto their desktops\n \n \n \n \n \n \nVirus scanning \nproducts for both the \nclient and the \nfirewall\n \n \n \n \n \n Spam \n \n \n \n \n \nMore of a nuisance than a security \nproblem, spam (junk e-mail) can clutter the \nLAN and in-boxes with unwanted and \noften inappropriate message\n \n \n \n \n \n E-mail filters \n \n \n \n \n Source: Paone, 1999 \n \n \n \n \n Regardless of policies and products, e-mail security will remain a moving target as hackers continue \nto find their way around existing barriers. Some suggestions for securing e-mail include:\n \n \n \n \n \n • Not opening attachments from unfamiliar sources \n \n \n \n \n • Reporting instances of spam \n"
        },
        {
            "page_number": 112,
            "text": " \n \n \n \n • Using encryption when sending highly sensitive information across the Internet \n \n \n \n \n Access Control \n \n \n \n \n \nAccess control is the mechanisms and policies that restrict access to computer resources. If an \nunauthorized user has physical access to your computer, that person can usually bypass a password-\nprotected screen by rebooting the computer. Symantec Corporation has a product called Norton Your \nEyes Only for Windows 95 that gives PC users easy, secure, and integrated access control. Your Eyes \nOnly addresses the following security needs:\n• Public/private key on-the-fly encryption/decryption \n \n \n \n \n • Centralized management \n \n \n \n \n • Mobile and remote users \n \n \n \n \n • Access control for multiple users of a single machine \n \n \n \n \n \nYour Eyes Only is available in two versions, one for end users and one for administrators. It uses \npublic/private key technology to encrypt and decrypt files. A user who enters a password gains \naccess only to his or her private key. Your Eyes Only lets you select key lengths of between 360 and \n2,048 bits. It offers on-the-fly encryption and decryption that works automatically within applications \nas files are opened and closed.\n \n \n \n \n \n Your Eyes Only offers the following access control features: \n \n \n \n \n • BootLock: protects the boot process from unauthorized users, allowing Windows 95 users to lock an \nentire PC so that unauthorized users cannot start it even from a floppy disk.\n \n \n \n \n \n • ScreenLock: protects the system while the user is temporarily away. It blanks the screen and locks \naccess to the PC.\n \n \n"
        },
        {
            "page_number": 113,
            "text": " \n \n \n \n• SmartLock folders: offer on-the-fly encryption. You can specify the encryption method and the files \nto lock. Any files in SmartLock folders are automatically decrypted when opened and encrypted \nwhen closed.\n \n \n \n \n \n Remote Access via the Internet \n \n \n \n \n \nMost companies still use costly private dial-up to connect remote users. In addition, companies must \ninstall and configure encryption and tunneling software for a large number of mobile users. Sun \nMicrosystems i-Planet is designed to leverage the cost of computing and offer access to your \ncorporate Intranet from the Internet.\n \n \n \n \n \n \ni-Planet software uses Java to eliminate the need for remote users to dial into a corporate modem pool \nor use security and authentication software tied to a laptop. By using the Internet standard for remote \naccess, i-Planet enables remote users to access the private network from various technologies, such as \ncable modems, DSL, or ISDN (see Figure 4.19). With i-Planet users can use any browser enabled \nwith Java technology to read e-mail, download data, check appointment calendars, and access other \nenterprise information.\n \n \n \n \n \n \n \n \n \nFigure 4.19: \ni-Planet Topology \nSource: Sun Microsystems\n \n \n \n \n \n The i-Planet architecture is a server software-based solution that does not require any client \ninstallation, management, or configuration; users need only a Web browser and an Internet \nconnection. The client application is either pushed to the client as HTML Web pages or Java applet \n \n"
        },
        {
            "page_number": 114,
            "text": "that is downloaded on demand. Once users clear the authentication, i-Planet creates a VPN on the fly. \nAuthentication is handled over an encrypted SSL channel between the client Web browser and the i-\nPlanet authentication system.\n \n \n \n \n i-Planet software offers remote access security for: \n \n \n \n \n • Internal HTML front ends and Web sites  \n \n \n \n \n • E-mail access  \n \n \n \n \n • Calendar access, supporting the Common Desktop Environment (CDE) calendar and Sun Calendar \nServer applications\n • Xll and PC applications, allowing you to run existing missioncritical PC or Xll applications without \nany change\n \n \n \n \n \n • File access, optimizing mail of large files and attachments by delegating as many tasks to be \nperformed on the server as possible\n \n \n \n \n \n • Terminal emulation, using local Telnet client and VT100 emulator to allow encrypted Telnet \nsessions from the Internet\n \n \n \n \n \n Summary \n \n \n \n \n \nToday computer security has to do with protecting your sensitive information from both outsiders \nand insiders. You need a security policy that covers all risks. Security tools such as firewalls, anti-\nvirus software, and encryption will help your company deter access to unauthorized users.\n \n \n \n \n \n \nBelieving that your environment is secured is not enough. You have to take a proactive approach to \nsecurity, making sure that newer technologies are implemented to keep up with sophisticated hacker \ntools. A safe and secure computer environment will protect your investments for the coming years.\n References \n \n \n \n An Analysis of Security Incidents on the Internet, 1989–1995, www.cert.org, April 17, 1997. \n"
        },
        {
            "page_number": 115,
            "text": " \n \n \n Backman, Dan, ''Smartcards: The Intelligent Way to Security,\" Network Computing, \nhttp://www.nwc.com/909/909w5/2.html, May 15, 1998.\n \n \n \n \n Bhimani, Anish, \"All Eyes on PKI,\" Information Security, www.infosecurity.com, October 1998. \n \n \n \n Bruno, Lee, \"Certificate Authorities, Who Do You Trust?\" Data Communications, Mar. 21, 1998, pp. \n54-61.\n \n \n \n \n Chapman, D. Brent, & Zwicky, Elizabeth D., Building Internet Firewalls O'Relly & Associates, Inc., \n1995).\n \n \n \n \n Foote, Steven, \"19 Infosecurity Predictions for '99,\" www.inforsecuritymag.com/nov/cover.htm. 1998. \n \n \n \n Garfinkel, Simson, & Spafford, Gene, Practical UNIX & Internet Security, O'Relly & Associates, \nInc., June 27, 1999.\n \n \n \n \n \"A Gauntlet Firewall Executive White Paper,\" \nwww.nai.com/media/doc/products/gauntlet_adaptiveproxy.doc.\n \n \n \n \n Innovative Security Products, www.wesecure.com, 1998. \n \n \n \n An Introduction to Computer Security: The NIST Handbook, Special Publication 800-12, , February \n7, 1996.http: //csrc.ncsl.nist.gov/nistpubs/800-12/\n \n \n \n \n M2 Presswire, \"Hewlett-Packard: HP corporate PCs, notebooks, workstations now optimized for \nsmart cards,\" Nov. 18, 1998.\n \n \n \n \n McCooey, Eileen, \"Security Becomes Priority,\" Windows Magazine (Jan. 1, 1999), p. 50. \n \n \n"
        },
        {
            "page_number": 116,
            "text": " National Institute of Standards and Technology, Computer Security Handbook Cryptography, \nhttp://security.isu.edu/isl/hk_crypt.html.\n \n \nNetwork Associates Executive White Paper, \"The Active Firewall,\" \nhttp://www.nai.com/international/uk/asp_set/products/tns/activesecurity/acts_literature.asp, August \n18, 1999.\n \n \n \n \n Paone, Joe, \"Slaying the E-mail Demons: Making Mail Safe,\" MicroTimes (Apr. 2, 1999), pp. 117–\n27.\n \n \n \n \n Pasmore, Frank, \"Weaving a Secure Web,\" Information Security, www.infosecuritymag.com, 1999. \n \n \n \n Perez, Juan Carlos, \"IDC: Internet security market grows by $1 billion,\" InfoWorld Electric, \nhttp://www.infoworld.com/cgi-bin/displaystory.pl?990330.ensecurity.htm, Mar. 30, 1999.\n \n \n \n \n Smart Cards, www.microsoft.com, Mar. 24, 1999. \n \n \n \n Yasin, Rutrell, \"Better Snapshots of Enterprise Security,\" InternetWeek (Feb. 22, 1999), p. 23. \n \n \n \n \n Websites for Computer Security \n \n \n \n Aladdin Knowledge Systems: http://www.aks.com/  \n \n \n \n AXENT Technologies, Inc.: http://www.axent.com/ Biometric Consortium: \nhttp://www.biometrics.org/\n \n \n \n \n Chrysalis-ITS Security Assurance White Paper: http://www.chrysalis.com/read/VPNwhitepaper.htm  \n \n \n \n Cisco Systems, Inc.: http://www.cisco.com/  \n \n \n \n Computer Security Products, Inc.: http: www.computersecurity.com/  \n \n"
        },
        {
            "page_number": 117,
            "text": " \n \n Computer Viruses—an Executive Brief, http://www.symantec.com/avcenter/reference/corpst.html  \n \n \n \n Deltavision: http://www.deltavision.com/  \n \n \n \n Entrust Technologies, Inc.: http://www.entrust.com/  \n \nKerberos: The Network Authentication Protocol: http://web.mit.edu/kerberos/www/  \n \n \n \n Network Associates, Inc.: http://www.nai.com/  \n \n \n \n NIST Computer Security Resource Clearinghouse: http://csrc.nist.gov/  \n \n \n \n Security Information: http://www.zdnet.com/devhead/filters/security/  \n \n \n \n Sensar, Inc.: http://www.sensar.com/  \n \n \n \n Shumann Security Software, Inc.: http://www.schumannsoftware.com/  \n \n \n \n Sun Microsystems i-Planet: http://www.swest2.sun.com/software/cover/  \n \n \n \n Symantec Security Center: http://www.symantec.com/avcenter/security/index.html\nChapter 5— \nPersonnel Security \n \n \n \n \n \nEvery staff position that interacts with the computer system should be evaluated from a security \nperspective. Establish criteria for filling each position; spell out the process to be used to evaluate \ncandidates, screen applicants, and conduct background checks. Specify the training program for new \nemployees.\n \n \n \n"
        },
        {
            "page_number": 118,
            "text": "  \nDefine each position clearly, describing its functions. Assess the sensitivity level. Sensitivity level is \ndetermined by the degree of harm a person could cause in that position: For instance, a person might \nbe in a position to steal or disclose confidential data, interrupt critical processing functions, commit \nfraud, or steal resources. What type of computer access will be needed for the position? Employees \nshould be given only the access needed to perform their duties.\n  \n \n \n \n \n \nSegregation of duties is an important control to ensure that employees do not perform incompatible \nduties. This means you must allocate responsibilities so that a single employee cannot perpetrate an \nerror and cover it up in the normal course of his or her duties. You'll need checks and balances to \nprevent or catch irregularities, so that the duties of one person automatically check the work of \nanother.\n \n \n \n \n \n \nFunctions such as systems analysis, programming, computer operations, and data control should be \nsegregated. The electronic data processing (EDP) department should be organizationally independent \nof the operating departments. It should not have the authority to correct data entered by users in other \ndepartments.\nScreening \n \n \n \n \n \nBefore you hire anyone for a sensitive position, perform a pre-employment background check. (For \nless sensitive positions, post-employment screening may be sufficient.) For highly sensitive \npositions, the background check should provide positive evidence that the candidate can perform the \nduties required honestly.\n \n \n \n \n \n The applicant's prospective supervisor should not conduct the background check; the personnel \ndepartment or an independent agency should do it.\n \n \n \n \n \n \nA significant percentage of applicants lie either on resumes or in interviews. These individuals should \nbe rejected not only because they may not be qualified but also because they lack the integrity \nrequired for sensitive positions.\n \n \n \n \n \n \nMany individuals will not divulge a criminal past to prospective employers. If the employees \nthereafter commits a criminal act that affects others, the employer may be held liable. Employers \nhave a legal obligation to thoroughly investigate their employees' backgrounds—without violating an \nemployee's privacy. Check an applicant's previous address, professional and bank references, credit \nhistory, criminal record, and previous employers.\n \n \n \n \n \n When applicants lie about their work history, they tend to: \n \n"
        },
        {
            "page_number": 119,
            "text": " \n \n \n • Increase the amount of time they have spent at a job \n \n \n \n \n • Create fictitious employers \n \n \n \n \n • Boost their salary levels \n \n \n \n \n • Tell you that their employer has gone out of business \n \n \n \n \n • Exaggerate job duties and inflate job titles \n \n \n \n \n • Falsely claim to be self-employed or independent consultants \n \n \n \n \n \nCandidates may also fabricate educational credentials and professional experience. Having only a \nyear or two of college, they may claim that they have a degree. A person may have even assumed the \nidentity of a real graduate or an established professional. Candidates may have bought their degrees \nfrom unaccredited mail-order universities. Some fabricate honors and awards. Others may list fake \npublications on their resumes. Applicants should be asked to provide copies of their publications and \nproofs of honors and awards.\n \n \n \n \n \n \nMany employers do a poor job of checking references. They either don't bother with the references or \nask superficial questions. To avoid legal hassles, all applicants should be required to sign a release \nform. The applicants should promise to hold their references harmless for what they say. Before any \nformal offer is made, if the reference check is not acceptable, it's not necessary to give the candidate a \nspecific reason for not being chosen. Keep records of reference checks to refute allegations of \ndiscriminatory or illegal actions. A conscientious reference check may also defend your company in \nany later litigation over that employee's actions.\nYou can access hundreds of public sources to check an applicant's background. Checking from public \nsources does not require the applicant's permission. Public records often contain conviction records and \nin some jurisdictions arrest records.\n \n \n \n \n \n \nFederal agencies collect a vast amount of data on individuals. The FBI's Identification Division \ncontains the largest repository of criminal records in the United States. The Freedom of Information \nAct (FIA) allows you to access many government files, including criminal conviction records. A \nconviction in itself is generally not a sufficient reason to deny the applicant the job; the conviction \nshould relate to the job you're trying to fill.\n \n"
        },
        {
            "page_number": 120,
            "text": " \n \n \n \n \nThe financial and credit records of prospective applicants can be accessed through credit bureaus and \nprivate investigative agencies. The Fair Credit Reporting Act governs investigative consumer reports, \nwhich are frequently used by prospective employers to screen candidates.\n \n \n \n \n \n Corporate records and limited partnership records are public. These records are one way to tell if the \napplicant actually owned a business, and for how long.\n \n \n \n \n \n \nFrom court records, you can learn if the applicant was a participant in any civil or criminal litigation. \nCity and county filings may give you information about such financial factors as whether there are \nany liens against the applicant. Additional sources of financial information are real property records \nand tax rolls. Financial situation may have an effect on a person's ability to perform the job; a \nfinancially troubled candidate may have a greater incentive to commit fraud.\n \n \n \n \n \n The following companies perform pre-employment background verification: \n \n \n \n \n • Accurate Data Services: http://www.acudata.com/  \n \n \n \n \n • Alexander Information Group: http://www.alexinfogp.com/  \n \n \n \n \n • American Background: http://www.americanbackground.com/  \n \n \n \n \n • American Labor Resources: http://www.amlabor.com/  \n \n• Application Profiles, Inc.: http://www.ap-profiles.com/  \n \n \n \n \n • Background Check International, LLC: http://www.bcint.com/  \n \n \n \n \n • Barrientos & Associates: http://www.emcsat.com  \n \n \n \n \n • Confidential Research Associates: http://www.mrjcomp.com/cra/  \n"
        },
        {
            "page_number": 121,
            "text": " \n \n \n \n • Employment Research Service: http://www.employment-research.com/  \n \n \n \n \n • EMPFacts Factual Data: http://www.employmentscreen.com/  \n \n \n \n \n • Indepth Profiles: http://www.idprofiles.com/  \n \n \n \n \n • Informus Corp.: http://www.informus.com/  \n \n \n \n \n • Integri-Net: http://www.integer-net.com/  \n \n \n \n \n • Justifacts: http://www.justifacts.com/  \n \n \n \n \n • On-Line Screening Services, Inc.: http://www.onlinescreening.com/  \n \n \n \n \n • Pennell & Associates: http://www.pennellinvestigations.com/  \n \n \n \n \n • PEV: http://pev.frickco.com/  \n \n \n \n \n • Pre-Employment Screening, Inc.: http://www.ddpes.com/  \n \n \n \n \n • RCE Information Services: http://www.iglou.com/datadetective/  \n \n \n \n \n • SafeHands: http://www.safehands.com/  \n \n \n \n \n • StafTrack: http://www.staftrack.com/  \n \n"
        },
        {
            "page_number": 122,
            "text": " \n \n \n • USDatalink: http://www.usdatalink.com/  \n \n \n \n \n Legal Agreements \n \n \n \n \n Insist that new hires in sensitive jobs sign employment agreements with non-disclosure provisions. \nThe agreement should specify:\n \n \n \n \n \n • Scope of the employee's duties \n \n \n \n \n • That the employee will work solely for your organization and may not work anywhere else without \nadvance written permission\n \n \n \n \n \n • That the organization is entitled to reap the benefit of the employee's work product \n \n \n \n \n • That the employee will not reveal secrets to unauthorized individuals, within as well as outside the \norganization\n• The conditions for terminating the employee \n \n \n \n \n • That the employee will return all materials, including notes, handbooks, computer programs, \ncommercial documents, and software at your request, and always at the end of the employment term  \n \n \n \n \n • That the employee may not keep any copies of items owned by the organization \n \n \n \n \n • The limitations on the types of work the employee may do after leaving the organization \n \n \n \n \n • That the employee will not engage in any unfair business practices, including corporate spying or \nengaging in a business that competes with the employer's business\n \n \n \n \n \n • That the employee will abstain from any activity that may hurt the employer or its interests \n \n"
        },
        {
            "page_number": 123,
            "text": " \n \n \n Training New Employees \n \n \n \n \n \nTraining new employees in computer security responsibilities and duties is a highly effective way of \nenhancing security. You may want to limit a new employee's access to the computer systems till \nsecurity training is completed, but it's also essential that computer security training be ongoing. These \nare some of the issues to be addressed during training:\n \n \n \n \n \n • The organization's data back-up policies \n \n \n \n \n • The type of data that should be encrypted \n \n \n \n \n • How the data encryption keys are managed \n \n \n \n \n • What types of data may be shared with colleagues \n \n \n \n \n • What types of data are available to the public \n \n \n \n \n Performance Appraisal \n \n \n \n \n Performance and skill level of employees should be routinely documented, using formal performance \nevaluation systems. Give employees feedback about their performance regularly.\n \n \n \n \n \n \nAn effective review procedure can help prevent job frustration and stress. It can also help maintain \nemployee morale. It's important to be concerned about the threat of psychological dissatisfaction. \nDisgruntled employees may do intentional damage. Moreover, job turnover associated with \ndissatisfied employees disrupts the operation and maintenance of computer systems. Discontent can \nact as a catalyst for computer crime or sabotage. Watch for possible indicators of discontent, such as:\n• Low quality or low production output \n \n \n \n \n • Complaints \n \n"
        },
        {
            "page_number": 124,
            "text": " \n \n \n • Late arrivals \n \n \n \n \n • Excessive absenteeism \n \n \n \n \n • Putting off vacations \n \n \n \n \n • Excessive unwarranted overtime \n \n \n \n \n Quick action, like communicating with the employee on a one-to-one basis, can significantly \nminimize job discontent.\n \n \n \n \n \n Exit Procedures \n \n \n \n \n \nSpecial security issues arise when an employee leaves. An employee may leave on mutually \nagreeable terms, including retirement, promotion, accepting a better position at another company, or \ntransferring willingly to another department. An employee may also leave on unfriendly terms, as \nwhen the employee is fired, unwillingly transferred to another department, forced into retirement, or \ndemoted.\n \n \n \n \n \n \nFor a mutually agreeable termination, follow a standard exit procedure. Upon leaving the \norganization, employees should be required to return badges, keys, and company materials. Change \ntheir access codes and passwords, and even locks if necessary.\n \n \n \n \n \n \nData files, especially encrypted files and the keys to decrypt them, other documents, and all backed-\nup files should be returned. Don't let the employee keep copies of anything; you should both sign an \nagreement attesting that all copies have been returned.\n \n \n \n \n \n Employees an cause considerable damage if terminated, for example, they may: \n \n \n \n \n • Intentionally input erroneous data \n \n"
        },
        {
            "page_number": 125,
            "text": " \n \n \n • Erase data files and destroy back-ups \n \n \n \n \n • Make copies of data files for personal use or for competitors \n \n \n \n \n • Misfile or destroy important documents \n \n• Create \"random\" errors that are difficult to trace and costly to correct \n \n \n \n \n \nWhile most employees can do some harm to the computer system, systems personnel can do the \nmost. From a security perspective, termination of systems personnel requires great caution. For \nexample, systems personnel may delete or destroy data or program files. They may also place logic \nbombs to harm the system (erase data, deny access, etc.), activating the code after their departure. \nThey can set these in place long before they're even notified of termination. Protect your organization \nwith controls over modification of system files.\n \n \n \n \n \n \nWhen an employee is leaving on unfriendly terms, his or her access to the computer system should be \nrestricted as quickly as possible. These accounts should be closed before or at the same time as the \nemployee is notified of the termination.\n \n \nChapter 6— \nNetwork Security \n \n \n \n \n \nNetworks may be broadly classified as either wide area networks (WANs) or local area networks \n(LANs). Security is needed for both. Computers in a WAN may be thousands of miles apart; \ncomputers in a LAN are usually closer together, such as in the same building or plant. Data-switching \nequipment may be used in LANs but not as often in WANs.\n \n \n \n \n \n \nOn the Internet security is needed to prevent unauthorized changes to your Web site. Businesses \nselling information-related products over the Internet, such as software vendors that allow their \npaying customers to download upgrades, need a way to discriminate between paying customers and \nvisitors not paying.\n \n \n \n \n \n \nSecurity administrators face the risk that an attacker will be able to break into the network. The \nattacks may range from direct attacks by both hackers and insiders to automated attacks using \nnetwork worms. Such an attacker might:\n \n \n"
        },
        {
            "page_number": 126,
            "text": " \n \n \n • Read access: Read or copy confidential information. \n \n \n \n \n \n• Write access: Write to a network—perhaps infecting the system with a virus or plant Trojan horses \nor back doors. The attacker may also destroy confidential information by deleting it or writing over \nit.\n \n \n \n \n \n • Denial of service: Deny authorized users normal network services. An attack may consume CPU \ntime or network bandwidth or fill up memory.\n \n \n \n \n \n \nSecurity risks in using a server on the Internet include inappropriate configuration of FTP (file \ntransfer protocol) settings. If FTP access to your server is allowed, configure it to prevent \nunauthorized modifications to files.\nThere must be a secure communication link for data transmission between interconnected host \ncomputer systems of the network. A major form of network communication security is cryptography to \nsafeguard transmitted data confidentiality. Cryptographic algorithms may be either symmetric (private \nkey) or asymmetric (public key).\n \n \n \n \n \n \nThe two popular encryption methods are link-level and end-to-end security. The former safeguards \ntraffic independently on every communication link; the latter safeguards messages from source to \ndestination. Link-level enciphers the communications line at the bit level, deciphering it as it enters \nthe nodes. End-to-end enciphers information at the entry point to the network, deciphering it at the \nexit point, thus providing security over information inside the nodes.\n \n \n \n \n \n You should have a list of authorized users, general or specific: \n \n \n \n \n • Who is allowed into the facilities? \n \n \n \n \n • When may they enter? \n \n \n \n \n • For what purposes may they enter? \n \n \n \n \n A variety of tools is available to help the IT manager implement the security plan, including \nencryption tools, route and packet filtering, and firewalls.\n \n"
        },
        {
            "page_number": 127,
            "text": " \n \n \n \n \nYou must have a network security policy. Your company also needs an internal corporate security \npolicy, once you decide how critical it is to protect the integrity of the computing system and the \nsecurity of your Web site.\n \n \n \n \n \n \nThe internal security plan should be distributed to everyone who uses the facilities, with written \nguidance for employees on the proper use of passwords. Tell them the types of words that should not \nbe used as passwords, and the policy about how often the password is to be changed.\n \n \n \n \n \n \nThere must be positive authentication before a user can access a terminal, an on-line application, or \nthe network environment, and you may want date and time constraints. Your employees should have \naccess to information only on a need-to-know basis. Unauthorized use should deactivate or lock a \nterminal. Diskless workstations may give you a safer environment.\nPasswords \n \n \n \n \n \nMost LAN or communication software packages contain encryption and security features. Passwords \nare included in virtually every package. However, people generally don't select good passwords or \nchange them often enough. Hackers can easily breach security by guessing passwords.\n \n \n \n \n \n \nThe effectiveness of passwords is greatly diminished when users do not select good passwords. \nPeople tend to make certain mistakes. They share them with other individuals or write them down. If \nyou need to write down a password to remember it, the purpose is defeated. These are good \nguidelines for choosing passwords:\n \n \n \n \n \n • Don't choose a password that's a word or a name in English or any other language. Hackers often \nuse dictionaries to find out passwords.\n \n \n \n \n \n • Avoid patterns like 123456, 12468, asdf, or qwerty (from the keyboard). \n \n \n \n \n • Don't use geographical names like Vegas or Florida. \n \n \n \n \n \n• If your system requires that the password contain both numerals and letters, don't just add a number \nto a word. Hackers know that most people will choose a word and add the numeral one (e.g., CAT1 \nor 1CAT).\n \n \n"
        },
        {
            "page_number": 128,
            "text": " \n \n \n • Encourage a combination of upper and lower-case characters. Non-alphabetic characters also make \nit more difficult for hackers to guess passwords.\n \n \n \n \n \n \n• An excellent technique is to use the first letter of a phrase to create a password. For example, ''I \nWas Born In New York\" would yield the password IWBINY. Although not a word that's easy to \nguess, it's easy to remember.\n \n \n \n \n \n \n• Change passwords regularly; encourage this by programming the computer system to require new \npasswords. The system should keep a history of older passwords and check to ensure that users don't \nuse the same password again or choose one they've used recently.\n \n \n \n \n \n \nGive users security guidelines and give new users a course in security precautions and how to select a \ngood password. Users must understand why a good password is essential. The Web site of Symantec, \nmakers of security software, advises about selecting a good password, so users can evaluate the \nstrength of the ones they choose: \nhttp://www.symantec.com/avcenter/security/passwords/passwordanalysis/html\nPasswords provide good protection from casual or amateur hackers, but experienced hackers can \ntypically bypass the password system, especially in the UNIX environment. Software programs are \navailable to assist new hackers, even those with limited knowledge, to find or guess passwords.\n \n \n \n \n \n The aim of most hackers is to obtain unlimited access to the computer system, typically by: \n \n \n \n \n • Finding bugs or errors in system software \n \n \n \n \n • Taking advantage of an incorrect installation \n \n \n \n \n • Looking for human errors \n \n \n \n \n Many hackers are authorized users with limited access trying to get unlimited access. These hackers \nwill have a valid user ID and password and will be looking for weaknesses in the system.\n \n \n \n \n \n In most UNIX systems, passwords are stored in an encrypted file. Some systems use a shadow \npassword file where the original data is stored. Passwords are generally encrypted using the data \n \n"
        },
        {
            "page_number": 129,
            "text": "encryption standard (DES) algorithm.\n \n \n \n \n \nThe encryption method used is essentially irreversible. While it's easy to encrypt a password, it's \nextremely difficult, almost impossible, to decrypt it, but hackers can discover passwords through \nbrute force, especially those that consist of only six lower-case characters. Passwords for accounts \nthat are likely to attract hackers must obviously not consist only of lower-case characters.\n \n \n \n \n \n \nA serious design flaw can sometimes result in the creation of a \"universal\" password, one that \nsatisfies the requirements of the log-in program without the hacker actually knowing the correct \npassword. In one case, for example, a hacker could enter an overly long password. It could overwrite \nthe actual password, thus allowing the hacker access.\n \n \n \n \n \n Modem Connections \n \n \n \n \n Any time a user connects to the network using a modem, additional risks are introduced into the \nsystem. These can be minimized with dial-in modems.\nSimply keeping the telephone number secret is not sufficient. The many hackers who dial in all the \ntelephone numbers in an entire prefix could randomly discover yours.\n \n \n \n \n \n \nIn the past many companies used dial-back techniques to reduce modem risk. Nowadays, caller ID \naccomplishes the same objective. Essentially the network allows access only from pre-identified \ntelephone numbers. The obvious disadvantage is that the telephone numbers of authorized users must \nbe known in advance. This makes it difficult for users who travel.\n \n \n \n \n \n Another way to minimize risk from dial-in modems is to use hardware encryption devices on both \nends of the connection, but these tend to be expensive.\n \n \n \n \n \n \nA good telecommunications software program will have numerous protocol options, enabling \ncommunications with different types of equipment. Some programs error check information or \nsoftware programs received. Desirable features in telecommunications programs include menus \nproviding help, telephone directory storage, and automatic log-on and redial.\n \n \n \n \n \n Saboteur's Tools \n \n \n"
        },
        {
            "page_number": 130,
            "text": "  \nWhile in recent years ingenious procedures have been developed to preserve computer security, \nmany computer systems are still astonishingly insecure. Saboteurs have at hand a wide variety of \ntechniques to overcome security, among them:\n  \n \n \n \n \n \nTrojan Horse: The saboteur places a hidden program within the normal programs of the business. \nThe computer functions normally while the hidden program collects data, modifies programs and \nfiles, destroys data, or even causes a complete shutdown of operations. Trojan horses can be \nprogrammed to destroy all traces of their existence after execution.\n \n \n \n \n \n \nSalami Techniques: The perpetrator adapts the computer program to cause very small changes that \nare unlikely to be discovered but whose cumulative effect can be substantial. For example, the \nperpetrator may steal 10 cents from the paycheck of each individual and transfer it to his own \naccount.\n \n \n \n \n \n \nBack Door or Trap Door: In developing a computer program programmers sometimes insert a code to \nallow them to bypass standard security procedures. Once the programming is complete the code, \neither accidentally or intentionally, may remain in the program. Attackers use this extra code to \nbypass security.\nTime Bomb/Logic Bomb: A code may be inserted into a computer program that causes damage when a \npredefined condition occurs.\n \n \n \n \n \n \nMasquerade: A computer program is written to simulate a real program, perhaps the log-in screen \nand related dialogue. When a user attempts to log in, the program captures the user's ID and \npassword and displays an error message, prompting the user to log in again. The second time, the \nprogram allows the user to log in. The user may never know that the first log in was fake, to capture \nthe access code.\n \n \n \n \n \n \nScavenging: Normally when the user \"deletes\" data, that information is not actually destroyed; \ninstead, the space is made available for the computer to write on later. A scavenger may thus be able \nto steal sensitive data that the user thought had been deleted but that actually was still available.\n \n \n \n \n \n \nViruses: Viruses are like Trojan horses except that the illegal code can replicate itself. A virus can \nspread rapidly throughout the system; eradicating it can be expensive and cumbersome. To guard \nagainst viruses, take care in using programs from diskettes or in copying software from bulletin \nboards or outside the company. Use disks only from verified sources. The best precaution is to use a \ncommercial virus scanner on all downloaded files before using them.\n \n \n \n \n \nData Manipulation: The most common and easiest way of committing fraud is to add or alter data \n"
        },
        {
            "page_number": 131,
            "text": " \nbefore or during input. The best way to detect this is to use audit software to scrutinize transactions \nand review trails that indicate additions, changes, and deletions to data files. The use of batch totals, \nhash totals, and check digits can also help prevent this type of crime.\n• A batch total is a reconciliation between the total daily transactions processed by the micro and \nmanual totals determined by an individual other than the computer operator. Material deviations must \nbe investigated.\n \n \n \n \n \n • A hash total is found by adding values that would not typically be added together, such as employee \nand product numbers, so the total has no meaning other than for control purposes.\n \n \n \n \n \n \n• A check digit is used to ascertain whether an identification number (e.g., account number, employee \nnumber) has been correctly entered by adding a calculation to the ID number and comparing the \noutcome to the check digit.\n \n \n \n \n \n \nPiggybacking: Physical piggybacking to gain access to controlled areas occurs when an authorized \nemployee goes through a door using a magnetic ID card and an unauthorized employee right behind \nalso enters the premises. The unauthorized employee is then in a position to commit a crime. In \nelectronic piggybacking, an authorized employee leaves a terminal or desktop and an unauthorized \nindividual uses it to gain access.\nDesigning Secure Networks* \n \n \n \n \n \nThe architecture of a network includes hardware, software, information link controls, standards, \ntopologies, and protocols. A protocol relates to how computers communicate and transfer \ninformation. There must be security controls for each component within the architecture to assure \nreliable and correct data exchanges. Otherwise the integrity of the system may be compromised.\n \n \n \n \n \n In designing the network, it's necessary to consider three factors: \n \n \n \n \n \n1. The user should get the best response time and throughput. Minimizing response time entails \nshortening delays between transmission and receipt of data; this is especially important for interactive \nsessions between user applications. Throughput means transmitting the maximum amount of data per \nunit of time.\n \n \n \n \n \n \n2. The data should be transmitted within the network along the leastcost path, as long as other factors, \nsuch as reliability, are not compromised. The least-cost path is generally the shortest channel between \ndevices with the fewest intermediate components. Low priority data can be transmitted over \nrelatively inexpensive telephone lines; high priority data can be transmitted over expensive high-\nspeed satellite channels.\n \n"
        },
        {
            "page_number": 132,
            "text": " \n \n \n \n \n3. Reliability should be maximized to assure proper receipt of all data. Network reliability includes \nthe ability not only to deliver error-free data, but also to recover from errors or lost data. The \nnetwork's diagnostic system should be able to locate component problems and perhaps even isolate \nthe faulty component from the network.\n \n \n \n \n \n Network Media  \n \n \n \n \n When you choosing a network medium, you need to consider: \n \n \n * Shim et al. Information Systems Management Handbook, (N.J.: Prentice-Hall, 1999).\n• Technical reliability \n \n \n \n \n • The type of business you're in \n \n \n \n \n • The number of people who will need to access data simultaneously \n \n \n \n \n • Number of micros \n \n \n \n \n • The physical layout of your current equipment \n \n \n \n \n • Frequency of updating \n \n \n \n \n • Compatibility \n \n \n \n \n • Cost \n \n \n \n \n • Geographic dispersion \n \n"
        },
        {
            "page_number": 133,
            "text": " \n \n \n • Network operating software and support \n \n \n \n \n • Applications software \n \n \n \n \n • Expandability (adding workstations) \n \n \n \n \n • Restriction to PCs (or can cheaper terminals be used?) \n \n \n \n \n • Ease of access in sharing equipment and data \n \n \n \n \n • Need to access disparate equipment, like other networks and mainframes \n \n \n \n \n • Processing needs \n \n \n \n \n • Speed \n \n \n \n \n • Data storage capability \n \n \n \n \n • Maintenance \n \n \n \n \n • Noise \n \n \n \n \n • Connectivity mechanisms \n \n \n \n \n • Ability of network to conduct tasks without corrupting data \n \n \n \n \n Network Topologies  \n"
        },
        {
            "page_number": 134,
            "text": " \n \n \n \n \nA network configuration or topology is the physical shape of the network, the layout of linked \nstations. A node is a workstation. A bridge is a connection between two similar networks. Network \nprotocols are software implementations providing support for network data transmission. A server is \na micro or a peripheral performing tasks such as data storage within a LAN.\n \n \n \n \n \n \nNetwork servers are of several types. A dedicated server is a central computer used only to manage \nnetwork traffic. A computer used simultaneously as a local workstation is called a non-dedicated \nserver. In general, dedicated servers provide faster network performance because they do not take \nrequests from local users as well as network stations. In addition, these machines are not susceptible \nto crashes caused by local users' errors. Dedicated servers are expensive; they cannot be disconnected \nfrom the network and used as stand-alone computers. Non-dedicated servers have a higher price-\nperformance ratio for companies that need occasional use of the server as a local workstation.\nThe most common types of network topology are: \n \n \n \n \n \n• The hierarchical topology (also called vertical or tree structure) is attractive for several reasons. \nThe software to control the network is simple and the topology provides a concentration point for \ncontrol and error resolution. However, it also presents potential bottleneck and reliability problems. \nNetwork capabilities may be completely lost if there is a failure at a higher level.\n \n \n \n \n \n \n• The horizontal (or bus) topology popular in LANs offers simple traffic flow between devices. This \ntopology permits all devices to receive every transmission; in other words, a single station broadcasts \nto multiple stations. The biggest disadvantage is that since all computers share a single channel, a \nfailure in the channel results in loss of the network. One way to get around this problem is with \nredundant channels. Another disadvantage of this topology is that the absence of concentration points \nmakes it more difficult to isolate faults. A bus network usually needs a minimum distance between \ntaps to reduce noise. To identify a problem, each system element must be checked. A bus topology is \nsuggested for shared databases but not for single-message switching. It employs minimum topology \nto fill a geographic area while at the same time having complete connectivity.\n \n \n \n \n \n \n• The star topology is widely used for data communication systems. The software is not complex and \ncontrolling traffic is simple. All traffic emanates from the hub or the center of the star. Though \nsimilar to the hierarchical network, however, the star topology has more limited distributed \nprocessing capabilities. The hub routes data traffic to other components. It also isolates faults, which \nis relatively simple in this configuration. The potential of bottleneck at the hub may cause serious \nreliability problems, however. One way to enhance reliability is to establish a redundant backup of \nthe hub node.\n \n \n \nA star network is best when there's a need to enter and process data at many locations with day-end \ndistribution to different remote users. It's easy to identify errors in this system, since each \n"
        },
        {
            "page_number": 135,
            "text": "  \ncommunication must go through the central controller. Maintenance is easy if the central computer \nfails the network. There's a high initial cost because each node must be hooked up to the host \ncomputer in addition to the cost of the host. Expansion is easy: You only need to run a wire from the \nnew terminal to the host computer.\n \n• The ring topology sends data flows in a circular direction. Each station receives the data and then \ntransmits it to the next. One main advantage is that bottlenecks like those in the hierarchical or star \nnetworks are relatively uncommon. The structure is organized. The primary disadvantage is that the \nentire network can be lost if the channel between two nodes fails. Establishing a backup channel can \nusually alleviate this problem. Other ways to overcome it are switches to automatically route the traffic \nround the failed node, or redundant cables.\n \n \n \n \n \n \nA ring network is more reliable and less expensive when there is minimal communication between \nterminals. This type of network is best when there are several users at different locations who have to \naccess updated data on a continual basis, because multiple data transmissions can occur \nsimultaneously. The ring network permits users to create and update shared databases. With a ring, \nhowever, there is greater likelihood of error compared to a star because numerous intervening parties \nhandle data. In light of this, data in a ring system should make an entire circle before being removed \nfrom the network.\n \n \n \n \n \n \n• The mesh topology is very reliable, though complex. Its structure makes it relatively immune to \nbottlenecks and other failures. The multiplicity of paths makes it relatively easy to route traffic \naround failed components or busy nodes.\n \n \n \n \n \n LANs and WANs  \n \n \n \n \n \nWAN and LAN topologies usually take different shapes. A WAN structure tends to be more \nirregular. Since an organization generally leases lines at considerable cost, it attempts to keep them \nfully utilized. To accomplish this, data for a geographical area is often routed through one channel; \nhence the irregular shape of the WAN network.\n \n \n \n \n \n \nThe LAN topology tends to be more structured. Since channels in a LAN network are relatively \ninexpensive, the owners are generally not concerned with maximum utilization of channels. \nFurthermore, since LANs usually reside in a single building, the situation is inherently more \nstructured. LANs are flexible, fast, and compatible. They maximize equipment utilization, reduce \nprocessing cost, reduce errors, and ease information flow.\nLANs use ordinary telephone lines, coaxial cables, fiber optics, or other devices like interfaces. Fiber \noptics produce good performance and reliability but are of high cost. LAN performance depends on \nphysical design, protocols supported, and transmission band-width. Band-width is the frequency range \nof a channel; it reflects transmission speed along the network. As more devices become part of a LAN, \ntransmission speed decreases.\n \n \n"
        },
        {
            "page_number": 136,
            "text": " \n \n \n Two or more LANs may be connected. Each node becomes a cluster of stations (subnetworks). The \nLANs communicate with each other. Advantages of interfacing networks are that:\n \n \n \n \n \n • Total network costs are lower. \n \n \n \n \n • Individual subnetworks meet particular needs, increasing flexibility. \n \n \n \n \n • More reliable and higher-cost subnetworks can be used for critical activities. \n \n \n \n \n • If one LAN fails, the other still functions. \n \n \n \n \n Disadvantages are that: \n \n \n \n \n • Complexity is greater. \n \n \n \n \n • Some network functions may not be able to cross network boundaries. \n \n \n \n \n Communications Security  \n \n \n \n \n \nThe communication systems used to link data between two or more sites should be reliable, private, \nand secure. Communication systems are easily affected by environmental factors, hardware \nmalfunctions, and software problems. Attacks on computers that do not require physical access fall \nunder the domain of communications security.\n \n \n \n \n \n \nThe increased use of computer technology has also increased dependence on telecommunications. All \ntypes of data—sound, video, and traditional—are transferred over networks. Communications \nsecurity means ensuring that the physical links between the computer networks function at all times. \nThis also means that during data transmission, breakdowns, delays, and disturbances must be \nprevented. Take care to prevent unauthorized individuals from tapping, modifying, or otherwise \nintercepting data transmission.\nSix considerations in communications security are: \n \n"
        },
        {
            "page_number": 137,
            "text": " \n \n \n 1. Line security: restricting unauthorized access to the communication lines connecting the various \nparts of computer systems.\n \n \n \n \n \n 2. Transmission security: preventing unauthorized interception of communications. \n \n \n \n \n \n3. Digital signatures: authenticating the sender or message integrity to the receiver. A secure digital \nsignature process is comprised of (1) a method of signing a document that makes forgery infeasible, \nand (2) validating that the signature is the one whose it purports to be.\n \n \n \n \n \n \n4. Cryptographic security: rendering the information unintelligible if transmission is intercepted by \nunauthorized individuals. Security coding (encryption) of sensitive data is necessary. When the \ninformation is to be used, it can be decoded. A common coding method is DES. For even greater \nsecurity, double encryption may be used: Encryption is processed twice using two different keys. \n(You may also encrypt files on a hard disk to prevent an intruder from reading the data.)\n \n \n \n \n \n 5. Emission security: preventing electromagnetic radiation from electronic devices. These emissions \ncan be intercepted, without wires, by unauthorized individuals.\n \n \n \n \n \n \n6. Technical security: preventing the use of devices such as microphones, transmitters, or wiretaps to \nintercept data transmissions. Security modems allow only authorized users to access confidential \ndata. A modem may have graduated levels of security. Different users may be assigned different \nsecurity codes. There can be password and call-back features. There may be built-in audit trails, \nallowing you to monitor who is accessing private files.\n \n \n \n \n \n \nVANs offer both communication services and specialized data processing. Generally, a company has \nno direct control over a VAN'S security. However, VAN security has a direct effect on the client's \noverall security.\n \n \n \n \n \n Communication security may be in the form of: \n \n \n \n \n \n• Access control, which guards against improper use of the network. For example, Kerberos \ncommercial authentication software can be added to a security system to verify a user's existence by \nencrypting passwords transmitted around networks. Password control and other user authentication \ndevices, like the SecurID from Security Dynamics (800-SECURID) or Vasco Data Security's Access \nKey II(800-238-2726).\n"
        },
        {
            "page_number": 138,
            "text": "Do not accept a prepaid call if it's not from a network user. Hackers don't typically spend their own \nfunds. Review data communications billings and verify each host-to-host connection. Review all dial-\nup terminal users. Are the telephone numbers unlisted and changed periodically? Control specialists \nshould try to make unauthorized access to the network to test whether the security is working properly.\n \n \n \n \n \n • Identification, which identifies the origin of a communication within the network, as by digital \nsignals or notarization.\n \n \n \n \n \n • Data confidentiality, which prevents unauthorized disclosure of information within the \ncommunication process.\n \n \n \n \n \n \n• Data integrity, which guards against unauthorized changes (e.g., adding, deleting) of data at both \nthe receiving and sending points, as through cryptographic methods. Anti-virus software should be \ninstalled at both the network server and workstations to alert users when viruses are trying to enter \nthe system.\n \n \n \n \n \n \n• Authentication, which substantiates the identify of an originating or user entity within the network, \nverifying that the entity is actually the one being claimed and that the information being transmitted \nis appropriate. Examples of security controls are passwords, time stamping, synchronized checks, \nnonrepudiation, and multiple-way handshakes. Biometric authentication methods measure body \ncharacteristics. Keystroke dynamics is another possibility for identification.\n \n \n \n \n \n • Digital signature, signing messages with a private key. \n \n \n \n \n • Routing control, inhibiting data flow to network elements identified as unsecure, such as relays, \nlinks, or subnetworks.\n \n \n \n \n \n • Traffic padding, analyzing data for reasonableness. \n \n \n \n \n \n• Interference minimization, eliminating or curtailing radar/radio transmission interference. For a \nsmall network, one workstation may be used as the backup and restore for other nodes. In a large \nnetwork, several servers may perform backups, since the failure of one could have disastrous effects \non the entire system. Access to backup files must be strictly controlled.\nToken-Ring and Ethernet Networks  \n \n \n"
        },
        {
            "page_number": 139,
            "text": "  \nTraditional token-ring and Ethernet networks work on the broadcast principle, sending information in \nunits called frames. Each frame contains information about a variety of items, including the sender's \nand the receiver's addresses. The sender broadcasts a frame that every receiver can see.\n  \n \n \n \n \n \nAt any given moment only one computer in the network is broadcasting and all others are receiving. \nAnother computer may broadcast after the first is finished. While all machines on a network can see \nthe broadcasting computer's frame, under ideal conditions only the computer whose address matches \nthe receiver's address in the frame should be able to access the frame's contents.\n \n \n \n \n \n Sniffers  \n \n \n \n \n \nSnifters are programs designed to capture certain information. Network managers use snifters to \nanalyze network traffic and network statistics. Hackers, however, may use snifters to steal \ninformation, such as passwords.\n \n \n \n \n \n \nCertain actions can minimize sniffing risk. The most obvious is to limit access. If the hacker cannot \naccess the LAN, snifters cannot be used. However, because it's often possible to restrict access to \nnetworks too tightly, other alternatives should be considered.\n \n \n \n \n \n \nSwitched versions of token-ring and Ethernet networks may minimize sniffing. With a switched \nLAN, each user has his own port on the switch. A virtual connection is established with the \ndestination port for each frame sent. If the destination address in the frame doesn't mach, the risk \nassociated with sniffing is significantly reduced. Switched networks tend to be more expensive; \ncompletely switched networks are rare.\n \n \n \n \n \n \nProbably the best way to minimize sniffing risk is data encryption. It's important that the key is never \nsent over the network. Traditional information, such as the time, is used to enhance the encryption \nscheme.\n \n \n \n \n \n Data Flow  \n \n \n \n \n Data-switching equipment routes data through the network to its destinations. For instance, such \nequipment can route data around failed or busy devices or channels.\nRouters at each site are used to communicate with routers at other sites. Routers provide information \nabout the individuals and the resources available in the LAN. They're responsible for directing the flow \nof information. It's possible to configure the routers so that certain types, such as FTP or Telnet do not \nallow either incoming or outgoing access. It's also possible to enable or disable certain routers to \n \n"
        },
        {
            "page_number": 140,
            "text": "receive information only from certain network addresses.\n \n \n \n \n Route and packet filtering requires significant technical knowledge as well as time. Most routers \ndon't provide a security or audit trail, even though you need to know:\n \n \n \n \n \n • Who tried to break into the computer system \n \n \n \n \n • How often they tried \n \n \n \n \n • What means they used to attempt the break-in \n \n \n \n \n Data Transmission  \n \n \n \n \n Data transmission between computers in a network uses one of three methods: \n \n \n \n \n • Simplex transmission is in one direction only. An example is radio or television transmission. \nSimplex transmission is rare in computer networks.\n \n \n \n \n \n \n• Half-duplex transmission, found in many systems, can flow in both directions but not in both at the \nsame time. In other words, once a query is transmitted from one device, it must wait for a response to \ncome back.\n \n \n \n \n \n \n• Full-duplex transmission can transmit information in both directions simultaneously it doesn't have \nthe stop-and-wait aspects of a half-duplex system. Because of its high throughput and fast response \ntime, full-duplex transmission is common.\n \n \n \n \n \n Security Layers  \n \n \n \n \n Security should be provided in different layers. Both networking facilities and telecommunication \nelements must be secure. Make sure you have controls over both host computers and subnetworks.\n \n \n"
        },
        {
            "page_number": 141,
            "text": " \n \n \n \nNetwork traffic may be over many subnetworks, each having its own security levels, depending on \nconfidentiality and importance. Each may need different security services and controls. Note that \nsecurity aspects of each subnetwork have to be distributed to the gateways so as to incorporate \nsecurity factors into routing decisions.\nNetwork Backup  \n \n \n \n \n Backup capability is especially important in networks so that if one computer fails another can take \nover the load. This might be critical in certain industries such as the financial.\n \n \n \n \n \n Secure Sockets Layer \n \n \n \n \n \nWhen Secure Sockets Layer (SSL) is enabled (see \nhttp://developer1.netscape.com/docs/manuals/security/sslin/contents.htm), a Web browser will \ndisplay a lock or similar symbol to indicate that the data transfer is secure. Another way to tell if a \nWeb site is secure is to look at its address: It should start with ''https://\" rather than simply \"http://.\"\n \n \n \n \n \n \nMost Web-based monetary transactions are secured using SSL. Many Web severe/client products \nsupport SSL connections. To transact business on the web, you need access to such a server as well \nas a digital certificate.\n \n \n \n \n \n While using SSL for encryption greatly enhances security and confidentiality, it does slow the \ncommunication interchange: All the data has to be encrypted and then decrypted.\n \n \n \n \n \n \nThe SSL protocol was developed by Netscape. It operates by layering a security protocol on top of an \nunderlying connection transport protocol such as HTTP, Telnet, NNTP, FTP, and TCP/IP. SSL is \nbuilt into Netscape's client and server products.\n \n \n \n \n \n \nWhen building a Web site, you can enable SSL by configuring a security-enabled http (https) process \non the server. Web pages that require SSL access can be specified. Common Gateway Interface \n(CGI) routines can be written on the server side to integrate SSL into existing applications.\n \n \n \n \n \n \nSSL provides data encryption, checks for data integrity, and provides server—and, if required, \nclient—authentication for a TCP/IP connection. SSL is open and nonproprietary. Encryption, \ndecryption, and authentication are transparent for applications using SSL.\n \n \n"
        },
        {
            "page_number": 142,
            "text": " \n \n \n \nSSL is used extensively to encrypt and authenticate communications between clients and servers on \nthe Web. The transport layer security standard (TLS) of the Internet Engineering Task Force (IETF) is \nbased on SSL.\nYou can confirm and authenticate an SSL server's identity when you send sensitive information, such \nas a credit card number, to the server. The digital certificate used to prove authenticity serves as the key \nto SSL. Anyone with the correct software can become a certificate authority (CA), but usually there are \nonly certain trusted CAs that a Web browser is programmed to accept; VeriSign, Inc., is one of the \nbest-known.\n \n \n \n \n \n \nPublic-key cryptography techniques may be used to check if a server's certificate and public ID are \nvalid. Similarly, a server can check that the client's certificate and public ID are valid. Without \npublic-key cryptography, encrypted communication could take place between two or more users only \nif they shared the keys. Each user would have to maintain several keys for communicating with \nvarious users.\n \n \n \n \n \n \nPublic-key cryptography (see Chapter 4) allows parties to communicate securely without sharing \nsecret keys. Each party establishes a key pair: one private key and one public key. The public key is \navailable to all nodes on a network; it's used to encrypt messages to the node. The private key used to \ndecrypt the messages never leaves its node.\n \n \n \n \n \n \nTCP/IP (the Transmission Control Protocol/Internet Protocol) provides the rules for transporting and \nrouting data over the Internet. Protocols like the Hypertext Transport Protocol (HTTP) use TCP/IP to \ncarry out tasks like displaying Web pages. The SSL protocol runs in the middle between TCP/IP and \nhigher-level protocols like HTTP, allowing SSL-enabled clients and servers to authenticate \nthemselves and making an encrypted connection possible.\n \n \n \n \n \n \nThe \"strength\" of an SSL connection depends on the bit level: 40-bit SSL connections tend to be \nweak; a 128-bit SSL connection is extremely strong. 128 bits is approximately 340 septillion times \n(340,000,000,000,000,000,000,000,000) larger than 40 bits.\n \n \n \n \n \n \nIt's presently illegal for US companies to export internationally anything above a 56-bit encryption. \nSoftware security companies are trying to overcome these restrictions by developing encryption \ntechnology outside the United States.\n \n \n \n \n \n \nThe SSL protocol includes two subprotocols. The SSL Record Protocol defines the format that will be \nused for data transmission. The SSL Handshake Protocol determines how the record protocol will \nexchange data between server and client when the SSL connection is first established. It's used to \neither authenticate the server to the client or the client to the server. It also allows client and server to \n"
        },
        {
            "page_number": 143,
            "text": "select from cryptographic algorithms or ciphers supported by both client and server.\nPublic-key and symmetric key encryption are both used by the SSL protocol. While symmetric key \nencryption tends to be faster, public-key encryption provides better authentication. Common ciphers \nare:\n \n \n \n \n \n • Data Encryption Standard (DES). Triple DES applies DES three times and supports 168-bit \nencryption. Its key size makes it one of the strongest ciphers supported by SSL.\n \n \n \n \n \n • Digital Signature Algorithm (DSA), for authentication of digital signatures. \n \n \n \n \n • Key Exchange Algorithm (KEA), for key exchange. \n \n \n \n \n • Message Digest (MD5), used to create digital signatures. It is a commonly used cipher. \n \n \n \n \n • RSA, a company name for authentication and encryption. RSA key exchange algorithm is common \nfor SSL connections; it's the most popular cipher for commercial applications.\n \n \n \n \n \n • Secure Hash Algorithm (SHA-1), for secure data transmission. \n \n \n \n \n \n• SKIPJACK, a classified symmetric-key algorithm used in FORTEZZA-compliant hardware. The \nFORTEZZA encryption system is used by US government agencies for sensitive but not classified \ndata. FORTEZZA ciphers use KEA instead of RSA. FORTEZZA cards and DSA are used for client \nauthentication.\n \n \n \n \n \n \nPerformance can suffer when using public-key cryptography, so it's usually limited to digital \nsignatures or small amounts of data. Symmetric key encryption, such as DES, is typically used for \nbulk data.\n \n \n \n \n \n \nYour security administrator should decide which cipher suites to enable or disable considering the \nnature of the data, the need for confidentiality and security, and the speed of the cipher. The national \norigin of the parties is another consideration; certain ciphers may only be used within the US and \nCanada. Thus, if your organization disables the weaker ciphers, you automatically restrict access to \nclients within the US and Canada.\n \n \n"
        },
        {
            "page_number": 144,
            "text": " \n \n \n SSL Handshake  \n \n \n \n \n The following sequence of events is typical in an SSL connection: \n \n \n \n \n • The client provides the server with its own SSL version number, cipher settings, and other \ncommunications-related data.\n • The server provides the client with its SSL version number, cipher settings, and other \ncommunications-related data.\n \n \n \n \n \n • The server sends its certificate, requesting the client's certificate, if necessary. \n \n \n \n \n • The client authenticates the server. If the server cannot be authenticated, the client is warned that an \nencrypted and authenticated connection cannot be established.\n \n \n \n \n \n \n• The client creates a \"pre-master\" secret for the SSL connection and encrypts it with the server's \npublic key. The encrypted pre-master is then sent to the server. The client may also be asked to sign \nand send data as well as its certificate to authenticate itself.\n \n \n \n \n \n • The session will be terminated if the server cannot authenticate the client. \n \n \n \n \n • The server uses its private key to decrypt the pre-master and generate the \"master\" secret using the \npre-master.\n \n \n \n \n \n \n• Using the master secret, session keys are generated by both client and server. These symmetric \nsession keys are used to encrypt and decrypt the data. The keys ensure that the data is not tampered \nwith between the time it's sent and the time it's received.\n \n \n \n \n \n • The SSL session begins once the handshake is completed. Both client and server use the session \nkeys to encrypt and decrypt data and to verify data integrity.\n \n \n \n \n \n Authentication  \n \n"
        },
        {
            "page_number": 145,
            "text": " \n \n \n \nAuthentication of both clients and servers requires encrypting data with one key of a public-private \nkey pair and decrypting it with the other. For server authentication the client encrypts the pre-master \nsecret with the server's public key. The associated private key alone can decrypt the pre-master \nsecret. This gives the client reasonable assurance about the server's identity.\n \n \n \n \n \n \nTo authenticate itself the client encrypts random pieces of data using its own private key. In other \nwords, it creates a digital signature that can be validated, using the public key in the client's \ncertificate, only if the corresponding private key had been used. If the server cannot validate the \ndigital signature, the session will be terminated.\nSSLRef  \n \n \n \n \n \nSSLRef is an advanced software developer's tool-kit to help provide security features in TCP/IP \napplications using the SSL protocol. ANSI C source code is provided for incorporation into TCP/IP \napplications. SSLRef may be downloaded free for noncommercial use. While there are no license \nrestrictions on SSLRef, there are export restrictions.\n \n \n \n \n \n Kerberos  \n \n \n \n \n \nThe Kerberos protocol is used in a client/server environment to authenticate the client to the server \nand the server to the client. After authenticating identity, Kerberos may then be used to encrypt data. \nKerberos does not send across any data that may allow an attacker to impersonate the user.\n \n \n \n \n \n Kerberos is available free in the form of source code from the Massachusetts Institute of Technology \nand is also sold by several vendors in commercial software products.\n \n \n \n \n \n \nWhen a client accesses a network service, the client asserts to the server that it's running on behalf of \nan authorized user. Without authentication, there's virtually no security. With Kerberos \nauthentication, the client proves its identity to the server.\n \n \n \n \n \n \nIn the traditional environment, a user's identity is authenticated by checking the password during the \nlog-in process. Without Kerberos authentication the user would need to enter a password to access \neach network service. This is at minimum inconvenient, and it still doesn't provide security when \naccessing services on a remote machine. Without encryption it would be easy for anyone to intercept \nthe password during transit.\n \n \n \n"
        },
        {
            "page_number": 146,
            "text": "  \nKerberos eliminates the need to use passwords. Instead, a key is used to encrypt and decrypt short \nmessages and provide the basis for authentication. The client, to prove its identity, presents a ticket \nissued by the Kerberos authentication server. Secret information, such as a password that only an \nauthorized user would know, is contained in the ticket.\n  \n \n \n \n \n Kerberos is not effective against password-guessing attacks. A hacker who intercepts a few encrypted \nmessages may launch an attack by randomly trying passwords to see if the messages decrypt.\n \n \n \n \n \n \nKerberos assumes that workstations or machines are reasonably secure and only the network \nconnections are vulnerable. A trusted path for passwords is needed. For example, if the password is \nentered into a program containing a Trojan horse (that is, the program has been modified to capture \ncertain information), Kerberos will not provide any protection. Moreover, if transmissions between \nthe user and the authentication program can be intercepted, Kerberos will be ineffective.\n \nBoth the user and the network service must have keys registered with the Kerberos authentication \nserver. The user's key is derived from the user-selected password. The network service key is selected \nrandomly.\n \n \n \n \n \n \nMany types of software used by the international community require Kerberos. Because the US \nrestricts export of cryptography, a version of Kerberos called Bones is available for international \nusers. All the DES routines have been stripped from Bones, which is used to \"trick\" other software \ninto thinking that Kerberos is installed.\n \n \n \n \n \n \nTo use Kerberos, a Kerberos principal must be established. This principal is like a regular account on \na machine; certain information, such as the user name and password, are associated with it. The \ninformation is stored encrypted in the Kerberos database. To be effective, Kerberos has to be \nintegrated into the computer system. It protects only data from software that is configured to use it.\n \n \n \n \n \n The server, if possible, should be physically secure. Ideally, the machine should be dedicated to \nrunning the authentication server. Access to the machine should be tightly restricted.\n \n \n \n \n \n \nEach user's initial password has to be registered with the authentication server. The registration \nprocedure depends on the number of users. In-person registration provides the best control if the \nnumber of users is small. Consider other procedures, such as a log-in program on a trusted system, \nwhen the number of users is large.\n \n \n \n \n \n \nSeveral tools can enhance the security provided by Kerberos. Onetime passwords generated by a \ndevice are particularly useful. Commercial products are available that combine one-time-passwords \nwith Kerberos.\n \n"
        },
        {
            "page_number": 147,
            "text": " \n \n \n \n Sources of Information about Kerberos  \n \n \n \n \n Additional information about Kerberos is available from the following sites: \n \n \n \n \n • http://nii.isi.edu/info/kerberos/  \n \n \n \n \n • http://nii.isi.edu/publications/kerberos-neuman-tso.html  \n \n \n \n \n • http://www.cybersafe.com  \n \n \n \n \n • http://www.latticesoft.com>  \n \n \n \n \n • http://www.stonecast.net  \n \n \n \n \n • http://www.wrq.com  \n \nSecuring the DMZ \n \n \n \n \n \nThe DMZ (demilitarized zone) of a firewall deserves special attention (firewalls are explained in \ndetail in Chapter 4). The DMZ is that part of the network that doesn't belong to either the internal \nnetwork or the Internet. This is generally the area between the access router and the bastion host, a \nparticularly vulnerable system that has been fortified against attacks (bastions may be part of the \nfirewall).\n \n \n \n \n \n \nPutting access control lists on the access router creates a DMZ. An access control list contains the \nrules that define which packets are permitted to enter. The access router, which connects an internal \nnetwork with the Internet, is the first line of defense against attacks from the Internet.\n \n \n \n \n \n If your system has different security levels, one option is to divide the DMZ into several separate \nzones so that, for instance, even an attacker who gains access will be unable to attack the private \n \n"
        },
        {
            "page_number": 148,
            "text": "network because the bastion hosts are on a separate LAN.\n \n \n \n \n \nMost organizations don't secure their Web servers strictly, so that unauthenticated users might be \nable to run CGI or other executable programs on the servers. If this is done on a bastion host, \nhowever, the entire security mechanism might be compromised.\n \n \n \n \n \n Services should be split up not only by host but also by network, limiting the level of trust between \nhosts on the networks.\n \n \n \n \n \n \nUse redundant components to achieve maximum security. A single failure, such as a software bug, \nshould not compromise the entire system. Risk related to software bugs, for instance, can easily be \nreduced by:\n \n \n \n \n \n • Keeping up to date on software fix patches \n \n \n \n \n • Using well-known products that have been around a while \n \n \n \n \n • Running only necessary services \n \n \n \n \n \nIf your emphasis is on maintaining security rather than allowing access, you may decide to block \neverything and allow access on a limited or case-by-case basis. In deciding which services to permit, \nask yourself:\n \n \n \n \n \n • What effect will allowing this service have on security? \n \n \n \n \n • How does permitting the service affect firewall architecture? \n \n \n \n \n • Will an attacker be able to exploit an inherent weakness? \n \n• How well known is the service? \n \n \n \n \n"
        },
        {
            "page_number": 149,
            "text": " • Is the protocol for this service published? \n \n \n \n \n Sources of Firewall Information  \n \n \n \n \n You can find more information about firewalls at: \n \n \n \n \n • http://sunsite.unc.edu/LDP/HOWTO/Firewall-HOWTO.html  \n \n \n \n \n • ftp://tp.tis.com/pub/firewalls/  \n \n \n \n \n • http://www.dark.net/pub/mjr/pubs/index.shtml  \n \n \n \n \n • ftp://ftp.research.att.com/dist/internet_security/  \n \n \n \n \n • http://www.net.tamu.edu/tfp/security/TAMU/  \n \n \n \n \n • http://www.cs.purdue.edu/coast/firewalls/  \n \n \n \n \n Pretty Good Privacy (PGP) \n \n \n \n \n \nPretty Good Privacy (PGP) can be used to encrypt e-mail messages or digitally sign messages. If the \nencrypted e-mail is intercepted, it will appear to be garbage. Digital signatures can also be used when \nthe content of the message isn't secret but the sender wants to confirm that he wrote it. PGP has \nproven very effective.\n \n \n \n \n \n \nPGP relies on a public-key encryption scheme, but there's no need for a secure channel over which to \nsecure the key. The public key used to encrypt the data is typically available in a public database. The \ndecryption key is private.\n \n \n \nPGP is available free for non-commercial use. ViaCrypt PGP, available for commercial use primarily \n"
        },
        {
            "page_number": 150,
            "text": "  in the US and Canada, costs under $100. For the international community, commercial PGP versions \nare available from Ascom Systec AG (IDEA@ascom.ch).\n  \n \n \n \n \n To use PGP, you need two key rings, one public and one private. The public ring holds your public \nkey and the public keys of people known to you. The private ring holds your secret key or keys.\n \n \n \n \n \n \nPublic-key encryption is much slower than convention. PGP combines two algorithms, RSA and \nIDEA, to encrypt plain text. PGP's IDEA encryption scheme currently requires a 128-bit key. Using \nthe fastest current technology, it would still take several trillion years to break the encryption scheme.\n \n \n \n \n \n \nTo launch a successful attack against such a scheme, an attacker would have to understand the \nmathematical transformation that takes place between plain and cipher text. The complexity of the \ntransformation makes it extremely difficult to solve the mathematical problem.\nAt present PGP gives you three choices for key size: 512, 768, or 1,024 bits. It's also possible to specify \nthe number of bits for your key. Increased key size means increased security, but key size also affects \nthe program's running time during generation.\n \n \n \n \nOnly the RSA portion is affected by key size during encryption; the main body of the message is not \naffected. Therefore it's best to use the 1,024-bit key size. Note, though, that individuals running certain \nPGP versions may be unable to handle very large keys.\n \n \n \n \n \n Each time PGP is run, a different session key is generated. This enhances security. \n \n \n \n \nUsing PGP, it's possible to apply a digital signature to a message. If a trusted copy of the sender's \npublic key is available, it can be used to check the signature. It's impossible of anyone else to create \nthe signature without the secret key. This will also detect if someone has tampered with the message.\n \n \n \n \n \nIf the contents of a message are not secret but it's essential to allow others to verify the identity of the \nsender, you may use clear signing digital signatures. Clear signing works only on text files, not on \nbinary files.\n \n \n \n \n \nYou may sign someone's key if you wish to inform others that you believe the key belongs to that \nperson. Other people may rely on your signature to decide whether or not that person's key is valid. A \nkey-signing party is used to get together various users of PGP and extend the degree of trust on the \nWeb.\n \n \n"
        },
        {
            "page_number": 151,
            "text": " \n \n \nIf the secret key ring is stolen or lost, the key should be revoked. Using a backup copy of the secret \nkey ring, a key revocation certificate may be uploaded to one of the public key servers. Before \nuploading the revocation certificate, it's a good idea to add a new ID to the old key to inform others \nwhat your new key ID will be. Without a backup copy of the secret key ring, it's impossible to create a \nrevocation certificate.\n \n \n \n \n Although there are several public key servers, it's only necessary to send your key to one of them. That \nserver will send your key to the others. Keys may be submitted or extracted at: http://pgp5.ai.mit.edu/.  \n \n \n \n \n Sources of PGP Information  \n \n \n \n \n You can get further information about PGP at: \n \n \n \n • \nhttp://dir.yahoo.corn/Computers_and_Internet/Security_and_Encryption/PGP_Pretty_Good_Privacy/  \n \n \n \n \n • http://www.pgpi.com/links/  \n \n• http://home.epix.net/~alf/Security/  \n \n \n \n \n • http://web.mit.edu/network/pgp.html  \n \n \n \n \n • http://thegate.gamers.org/~tony/pgp.html  \n \n \n \n \n • http://www.nai.corn/default_pgp.asp  \n \n \n \n \n Security Analysis Tool for Auditing Networks \n \n \n \n \n The Security Analysis Tool for Auditing Networks (SATAN) was written by Dan Farmer and Wietse \nVenema because computer systems are becoming more vulnerable to attacks via networks.\n \n \n"
        },
        {
            "page_number": 152,
            "text": " \n \n \n \nSATAN gathers information about a remote host or set of hosts by probing various services such as \nFinger or FTP provided by the host. It identifies potential security flaws and bugs, such as \nincorrectly configured network services or known system bugs.\n \n \n \n \n \n \nSATAN consists of several programs, each testing for a specific weakness. Additional programs may \nbe added by putting their executable file into SATAN'S main directory. The entire SATAN package, \nincluding source code and documentation, is freely available via FTP from various sites, including:\n \n \n \n \n \n SATAN Demo \n \n \n \n \n • http://www.fish/com/satan/  \n \n \n \n \n North America \n \n \n \n \n • ftp://ftp.mcs.anl/gov/pub/security  \n \n \n \n \n • ftp://coast.cs.purdue.edu/pub/tools/unix/satan  \n \n \n \n \n • ftp://vixen.cso.uiuc.edu/security/satan-1.1.1.tar.Z  \n \n \n \n \n • ftp://ftp.acsu.buffalo.edu/pub/security/satan-1.1.1,tar.Z  \n \n \n \n \n • ftp://ftp.acsu.buffalo.edu/pub/security/satan-1.1.1,tar.gz  \n \n \n \n \n • ftp://ftp.net.ohio-state.edu/pub/security/satan/satan-1.1.1.tar.Z  \n \n \n \n \n • ftp://ftp.cerf.net/pub/software/unix/security/  \n \n \n \n \n • ftp://ftp.tisl.ukans.edu/pub/security/satan-1.1.1.tar.Z  \n"
        },
        {
            "page_number": 153,
            "text": " \n \n \n \n • ftp://ftp.tcst.com/pub/security/satan-1.1.1.tar.Z  \n \n \n \n \n • ftp://ftp.orst.edu/pub/packages/satan/satan-1.1.1.tar.Z  \n \n \n \n \n • ftp://ciac.llnl.gov/pub/ciac/sectools/unix/satan/satan.tar.Z  \n \nAustralia \n \n \n \n \n • ftp://ftp.dstc.edu.au:/pub/security/ satan/satan-1.1.1.tar.Z  \n \n \n \n \n • ftp://coombs.anu.edu.au/pub/security/satan/  \n \n \n \n \n • ftp://ftp.auscert.org.au/pub/mirrors/ftp.win.tue.nl/satan-1.1.1.tar.Z  \n \n \n \n \n Europe \n \n \n \n \n • ftp://ftp.denet.dk/pub/security/tools/satan/ satan-1.1.1. tar.Z  \n \n \n \n \n • http://ftp.luth.se/pub/unix/security/satan-I.I.I.tar.Z  \n \n \n \n \n • ftp://luth.se/pub/unix/security/ satan-1.1.1.tar.Z  \n \n \n \n \n • ftp://ftp.wi.leidenuniv.nl/pub/security/  \n \n \n \n \n • ftp://ftp.cs.ruu.nl/pub/SECURITY/ satan-1.1.1.tar.Z  \n \n \n \n \n • ftp://ftp.cert.dfn.de/pub/tools/net/satan/ satan-1.1.1.tar.Z  \n"
        },
        {
            "page_number": 154,
            "text": " \n \n \n \n • ftp://ftp.csi.forth.gr/pub/security/ satan-1.1.1.tar.Z  \n \n \n \n \n • ftp://informal!k.uni-kiel.de/pub/sources/security/MIRROR.ftp. win.tue.nl  \n \n \n \n \n • ftp://ftp.kulnet.kuleuven.ac.be/pub/mirror/ftp,win.tue.nl/security/  \n \n \n \n \n • ftp://ftp.ox.ac.uk/pub/comp/security/software/satan/satan-1.1.1.tar.Z  \n \n \n \n \n • ftp://ftp.nvg.unit.no/pub/security/satan-1.1.1.tar.Z  \n \n \n \n \n • ftp://cnit.nsk.su/pub/unix/security/satan  \n \n \n \n \n Courtney  \n \n \n \n \n \nCourtney, available free, may be thought of as the anti-SATAN. It monitors attacks from SATAN. \nThus, if SATAN is probing your system, Courtney notifies you and gives you a chance to trace the \nprobe. Courtney can be downloaded from:\n \n \n \n \n \n http://ciac.llnl.gov.ciac/ToolsUnixNetMon.html#Courtney  \n \n \n \n \n Automated Vulnerability Testing \n \n \n \n \n \nGenerally, it's not that controls are lacking or weak but that they aren't appropriately configured. Most \ncomputer software, for example, comes with default settings. If these are left unchanged, security may \nbe compromised.\nThe existence of controls gives management and users the false impression that their data is secure. \nMany hackers exploit well-known security weaknesses. They rely on common errors made by system \nadministrators, such as not protecting the system with the latest security patches.\n \n \n"
        },
        {
            "page_number": 155,
            "text": " \n \n \n \nWhile incompetence or lack of time or other resources certainly contributes to this problem, many \nadministrators make mistakes because of the nature of computer systems. Most systems support a \nwide variety of services, and the security mechanism must be flexible enough to meet the needs of a \nvariety of users. This flexibility can often lead to security weaknesses.\n \n \n \n \n \n \nMany automated tools audit computer systems and report potential security weaknesses. They can \nexamine thousands of files on a multi-user computer system and identify vulnerabilities resulting \nfrom improper use of controls or mismanagement. Automated tools may test for adequate virus \nprotection or for the ability to plant Trojan horses or worms.\n \n \n \n \n \n \nAutomated tools for testing security vulnerability are different from automated tools that monitor \nactivity or detect intrusion. Monitors and intrusion detection systems analyze activity as it occurs. \nVulnerability testing tools search for potential weaknesses.\n \n \n \n \n \n Stand-alone Systems  \n \n \n \n \n \nTo identify vulnerabilities in a stand-alone system, automated testing tools analyze shared executable \nfiles. Vulnerability testing tools are used to analyze files whose modification or disclosure would \nallow the hacker to circumvent system controls and gain unauthorized access.\n \n \n \n \n \n \nThe testing tools can check, for instance, if the passwords are long enough. Short passwords can be \nguessed easily using brute force. These tools can also check to see if passwords are changed often \nenough.\n \n \n \n \n \n \nTo prevent the planting of a Trojan horse program, user start-up files should be protected against \nmodification, because the start-up files are always executed. Write-access in a user's personal file \nspace should similarly be limited.\n \n \n \n \n \n \nThe audit trail should always be maintained. Automated tools can check to see if modification \nprivileges for system binaries are restricted to systems staff. System binary files should be reviewed \nfor unexpected changes, and secure defaults should be specified.\n \n \n \n \n \n \nIn a networked environment, computer systems generally share data and other resources. Security \ncontrols for access paths in networks can be tested. A network host will have the potential \nvulnerabilities not only of a stand-alone system but also of the networked system.\nTesting Techniques  \n"
        },
        {
            "page_number": 156,
            "text": " \n \n \n \n Vulnerability tests may audit the system or launch a mock attack. Testing thus may be passive or \nactive, broad or narrow.\n \n \n \n \n \n \nActive tests are intrusive; they identify vulnerabilities by exploiting them. Passive tests examine the \nsystem to infer the existence of vulnerabilities. An active test may launch a dictionary attack or \nrandomly try common or short passwords. If successful, it would log the results for review by \nsecurity personnel. A passive test might check the protection of a password file. It may copy the file, \nencrypt it, and compare encrypted strings. Both types of tests provide useful information. However, \nactive tests are riskier. Individual circumstances and professional judgment affect the choice of tests.\n \n \n \n \n \n \nVulnerability testing programs may be narrow, examining only a single vulnerability, or broad, \nexamining the entire system. While single vulnerability tests tend to be simple, they may not reveal \nthe joint effect of lack of testing. Weaknesses in multiple controls may compound the effect of \nvulnerabilities. System testing provides better information than a series of single tests, making it \neasier to determine the total risk.\n \n \n \n \n \n Conclusion \n \n \n \n \n \nComputer networks play a dominant role in transmitting information within and between firms. A \nnetwork is simply a set of computers (or terminals) connected by transmission paths. These paths are \nusually telephone lines but may be other media, such as wireless or infrared transmission, radio \nwaves, or satellites. The network serves one purpose: exchange of data.\n \n \n \n \n \n \nEncryption must be used whenever sensitive or confidential information is transmitted. The Internet \nnetwork is so open that anything can be read or snatched at many locations between the originating \nsite and the destination site. Encryption should be used when putting any secret or sensitive \ninformation on the Internet server.\n \n \n \n \n \n In a networked environment, the more connections, the more complex the system—and the greater the \nlikelihood of security being compromised.\nAppendix 6.A— \nCommercial Firewalls* \n \n \n \n \n \n \n \n \nOS/ \n \n"
        },
        {
            "page_number": 157,
            "text": "  Product\nVendor   \n \n E-mail \n \n Hardware \nPlatforms\n \n \n  Contact \nInformation  \n \n \n \n \n Actane Controller \n \n \n \n \n info@actane.com \n \n \n \n \n SNMP\nPC\n \n \n \n \n \nLe California Bat D2 \n2, Rue Jean Andreani\n13084 Aix-En-Provence\nCEDEX 2 \nFrance\n \n \n \n \n AFS 2000 Internet Device\nInternet Devices\n \n \n \n \n \n Sales@internetdevices.com \n \n \n \n \n - \n \n \n \n \n (408)541-1400 \n \n \n \n \n Alta Vista Firewall\nAlta Vista\n \n \n \n \n \n - \n \n \n \n \n WNT\nPC\n \n \n \n \n \n - \n \n \n \n \n Ascend Secure Access \nFirewall\n \n \n \n \n \n info@ascend.com \n \n \n \n \n PC \n \n \n \n \nOne Ascend Plaza \n1701 Harbor Bay Parkway\nAlameda, CA 94502 USA\n(510) 769-6001\n \n \n \n \n aVirt Gateway \naVirt Gateway Solutions \n \n \n \n \n sales@aVirt.com \n \n \n \n \n PC \n \n \n \n \n51 West Centre. Suite 604\nOrem, UT 84043 USA\n(800) 41-AVIRT or \n(801) 802-7450\n \n \n \n \n BIGfire \nBiodata and AB Systems \n \n \n \n \n Sales@biodata.com.sg \n \n \n \n \n - \n \n \n \n \nBurg Lichtenfels \nD-35104 Lichtenfels, \nGermany \nTel:++49-6454-912070\n \n \n \n \n BorderWare BorderWare\nTechnologies\n \n \n \n \n \n info@borderware.com \n \n \n \n \n POS, PC \n \n \n \n \n6695 Millcreek Drive, Unit 1\nMississauga, Ontario L5N 5R8\nEurope:++44 181 893 6066\nNorth America: 1 905 814 \n7900\n"
        },
        {
            "page_number": 158,
            "text": " \n \n \n \n Brimstone \nSOS Corporation \n \n \n \n \n sales@soscorp.com \n \n \n \n \n any PC \n \n \n \n \n40 Broad street, Suite 2175\nNew York, NY 10004 USA\n(800) SOS-UNIX or \n(212) 809-5900\n \n \n \n \n Bull S.A.:Firewall and\nNetwall\n \n \n \n \n \n - \n \n \n \n \n UNIX/WS \n \n \n \n \n - \n \n \n \n \n Centri\nCisco  \n \n \n \n \n info@cisco.com \n \n \n \n \n WNT\nPC\n \n \n \n \n \n170 W. Tasman Dr. \nSan Jose, CA 95134 USA\n(408) 526-4000\nProduct\nVendor  \n \n \n \n \n E-mail \n \n \n \n \n \nOS/ \nHardware \nPlatforms\n \n \n \n \n \n Contact \nInformation \n \n \n \n \n Cisco IOS Firewall\nCisco\n \n \n \n \n \n info@cisco.com \n \n \n \n \n Router \n \n \n \n \n170 W. Tasman Dr.\nSan Jose, CA 95134 USA\n(408) 526-4000\n \n \n \n \n Cisco PIX\nCisco\n \n \n \n \n \n info@cisco.com \n \n \n \n \n POS \n \n \n \n \n170 W. Tasman Dr.\nSan Jose, CA 95134 USA\n(408) 526-4000\n \n \n \n \n Citadel \nCitadel Data Security \n \n \n \n \n info@cdsec.com \n \n \n \n \n UNIX\nPC\n \n \n \n \n \nUnit 3, 46 Orange Street\nCape Town, 8001 South \nAfrica \nTel:+27 21 23-6065\n \n \n \n \n Conclave \nInternet Dynamics \n \n \n \n \n sales@sterling.com \n \n \n \n \n SUN\nWS  \n \n \n \n \n4600 Lakehurst Court\nDublin, OH 43016\nUSA \n(800) 700-5579 or\n"
        },
        {
            "page_number": 159,
            "text": "(33)1 4417-6400\n \n \n \n \n ConSeal PC \nSignal 9 Solutions \n \n \n \n \n firewall@signal9.com \n \n \n \n \n W95\nPC  \n \n \n \n \n (613) 599-9010 \n \n \n \n \n COOL-FIRE\nSymbolic\n \n \n \n \n \n mt@symbolic.it \n \n \n \n \n - \n \n \n \n \nViale Mentana \n29 1-43100 Parma, Italy\nTel:+39 521 776180\n \n \n \n \n \nCryptoSystem\nPyroWall \nRADGUARD\n \n \n \n \n \n info@radguard.com \n \n \n \n \n - \n \n \n \n \n24 Raoul Wallenberg street\nTel Aviv 69719 Israel\nTel:972 3 645 5444\n \n \n \n \n CSM Proxy Plus\nCSM-USA\n \n \n \n \n \n Var_sales@csm-usa.com \n \n \n \n \n Many \nPC WS \n \n \n \n \n360 So. Ft.Lane Suite #1B\nLayton, UT 84041 USA\n(801) 547-0914\n \n \n \n \n CyberGuard \nCyberGuard Corporation \n \n \n \n \n info@cybg.com \n \n \n \n \n UNIX \nWS PC \n \n \n \n \n2000 W. Commercial Blvd.\nFt.Lauderdale, FL 33309 \nUSA \n(954) 958-3900\n \n \n \n \n Cybershield \nData General \n \n \n \n \n - \n \n \n \n \n UNIX\nWS\n \n \n \n \n \n (508) 898-5000 \n \n \n \n \n Product\nVendor  \n \n \n \n \n E-mail \n \n \n \n \n \nOS/ \nHardware \nPlatforms\n \n \n \n \n \n Contact \nInformation \n \n \n \n \n CYCON Labyrinth \nCYCON Technologies \n \n \n \n \n labyrinth@cycon.com \n \n \n \n \n UNIX \n \n \n \n 11240 Waples Mill Rd.,\nSuite 403\n \n \n \n"
        },
        {
            "page_number": 160,
            "text": " \n \n \n PC \n \n \n \n Fairfax, VA 22030 USA\n(703) 383-0247\n \n \n \n \n Digital Firewall Service \nDigital Equipment Corp. \n \n \n \n \n email directory \n \n \n \n \n UNIX\nWNT  \n \n \n \n \nUnited States Contact:\nDick Calandrella\n(508) 496-8626\n \n \n \n \n \nDEC SecurityGate for \nOpen VMS[*] \nDigital Equipment Corp.\n \n \n \n \n \n email directory \n \n \n \n \n VMS \n \n \n \n \n contacts or \n(508) 568-6868 \n \n \n \n \n Elron Firewall \nElron Software \n \n \n \n \n info@elronsoftware.com \n \n \n \n \n POS \n \n \n \n \nOne Cambridge Centre\nCambridge, MA 02142 \nUSA \n(617) 692-3000\n \n \n \n \n \nenter works \nNetSeer LIght \nenter Works.com\n \n \n \n \n \n Michael Lazar \n \n \n \n \n - \n \n \n \n \n19886 Ashburn Road\nAshburn, VA 20147 USA\n(800) 505-5144 or\n(703) 724-3800\n \n \n \n \n ExFilter V1.1.2 \n \n \n \n \n exfilter@exnet.com or \nexfiler@exnet.co.uk\n \n \n \n \n \n SUN\nWS  \n \n \n \n \n - \n \n \n \n \n FLUX EF Enhanced Firewall \nINS Inter Networking Systems \n \n \n \n \n flux-info@ins.de \n \n \n \n \n FluxOS \n \n \n \n \nINS GmbH P.O Box \n101312 \nD-44543 Castrop\nGermany \nTel: +49 2305 101 0\n \n \n \n \n Firewall/plus\nNetwork-1\n \n \n \n \n \n sales@network-1.com Alyssa \nEarnhart\n \n \n \n \n \n \nWNT\nDOS \n \nPC\n \n \n \n \n(972) 606-8200 for Texas\n(800) 638-9751 \nWorld Headquarters\n \n"
        },
        {
            "page_number": 161,
            "text": " \n \n \n \n Fort Knox Firewall Device\nInternet Devices\n \n \n \n \n \n sales@internetdevices.com \n \n \n \n \n - \nPC \n \n  \n (408) 541-1400 x314\n \n \n \n \n Freegate \nFreeGate Corporation \n \n \n \n \n info@freegate.com \n \n \n \n \n - \n \n \n \n \n1208 E. Arques \nSunnyvale, CA 94086 \nUSA \n(408) 617-1000\n \nProduct\nVendor  \n \n \n \n \n E-mail \n \n \n \n \n \nOS/ \nHardware \nPlatforms\n \n \n \n \n \n Contact \nInformation \n \n \n \n \n \nGauntlet \nTrusted Information Systems\nNetworks Associates\n \n \n \n \n \n tis@tis.com(us) \ninfo@eu.tis.com(non-us) \n \n \n \n \n - \n \n \n \n \n - \n \n \n \n \n \nGEMINI Trusted Security\nFirewall \nGemini Computers\n \n \n \n \n \n tft@geminisecure.com \n \n \n \n \n - \n \n \n \n \n (408) 373-8500 \n \n \n \n \n \nGFX Internet Firewall System \nGNAT Box \nGlobal Technology Associates\n \n \n \n \n \n gfx-sales@gta.com\ngb-sales@gta.com  \n \n \n \n \n \nUNIX/POS\n \nPC\n \n \n \n \n \n3504 Lake Lynda Drive\nSuite 160 \nOrlando, FL 32817 USA\n(800) 775-4GTA or\n(407) 380-0220\n \n \n \n \n \nGlobeServer \nData Quest Information\nSystems\n \n \n \n \n \n info@dqisystems.com \n \n \n \n \n LINUX\nPC\n \n \n \n \n \n7509 Kingston Pike Suite \n313 \nKnoxville, TN 37919 USA\n(423) 588-4757\n \n \n \n \n HSC GateKeeper \n \n \n \n \n info@hsc.fr \n \n \n142, rue de Rivoli\n"
        },
        {
            "page_number": 162,
            "text": "Herve Schauer Consultants\n \n \n  75039 Paris Cedex 01\nTel:+33 (1) 46.38.89.90\n \n \n \n \n IBM Firewall for AIX\nIBM\n \n \n \n \n \n peter_crotty@vnet.ibm.com \n \n \n \n \n AIX\nWS  \n \n \n \n \nIBM Internet Firewall\nPO Box 12195 \nMail Drop B44A/B501\nRTP, NC 27709 US\nTel:+1 919-254-5074\n \n \n \n \n ICE.Block \nJ. River, Inc. \n \n \n \n \n info@jriver.com \n \n \n \n \n UNIX\nPC\n \n \n \n \n \n124 North First Street\nMinneapolis, MN 55401 \nUSA \n(612) 339-2521\n \n \n \n \n Instant Internet \nDeerfield Communications \n \n \n \n \n info@deerfield.com \n \n \n \n \n WNT, W95\nPC\n \n \n \n \n \n (517)732-8856 \n \n \n \n \n Interceptor \nTechnologic \n \n \n \n \n info@tlogic.com \n \n \n \n \n - \nPC \n \n \n \n \n4170 Ashford Dunwoody \nRd. \nSte. 465 \nAtlanta, GA 30319 USA\n(404) 843-9111\n \n \n \n \n \nInterLock \nMCI Worldcom Advanced\nNetworks\n \n \n \n \n \n info@ans.net \n \n \n \n \n SUN\nWS  \n \n \n \n \n1875 Campus Commons \nDr. \nReston, VA 22091 USA\n(800) 456-8267 or\n(703) 758-7700\n \n \n \n \n Inter-Ceptor \nNetwork Security International \n \n \n \n \n - \n \n \n \n \n - \n \n \n \n \n John Shepard at\n(516) 674-0338\nProduct\nVendor  \n \n \n \n E-mail \n \n \n \n \n \nOS/ \nHardware \nPlatforms\n  \n \n \n Contact \nInformation \n"
        },
        {
            "page_number": 163,
            "text": " \n \n \n \n \n \n IPAD 1200 \nNetmatrix Internet Co. \n \n \n \n \n sales@ipad-canada.com \n \n \n \n \n POS \n \n \n \n \nNetmatrix Corporation\n#36001, 6449 Crowchild Tr.\nCalgary. Alberta Canada\nT3E 7C6 \n(403) 686-1169\n \n \n \n \n IRX Firewall \nLivingston Enterprises \n \n \n \n \n info@livingston.com \nsales@livingston.com \n \n \n \n \n Router \n \n \n \n \n4464 Wiilow Road \nPleasanton, CA 94588 USA\n(510) 737-2100\n \n \n \n \n Iware \nInternetware \n \n \n \n \n Paul Singh \n \n \n \n \n Novell \n \n \n \n \n505 W. Olive Ave., Suite 420\nSunnyvale, CA 94086 USA\n(408) 244-6141\n \n \n \n \n iWay-One \nBateTech Software \n \n \n \n \n www.workgroup.co.za\nsales@batetech.com\n \n \n \n \n \n WNT \n \n \n \n \n7550 W. Yale Ave., B130\nDenver, CO 80227 USA\n(303) 763-8333\n \n \n \n \n Juniper \nObtuse Systems \n \n \n \n \n info@obtuse.com \n \n \n \n \n - \n \n \n \n \n Alberta, Canada \n \n \n \n \n KarlBridge/KarlBrouter\nKarlNet Inc.\n \n \n \n \n \n sales@KarlNet.com \n \n \n \n \n Bridge/\nRouter  \n \n \n \n \n Columbus, OH USA\n(614) 263-KARL\n \n \n \n \n \n Lucent Managed Firewall\nLucent Technologies\n \n \n \n \n \n firewall@lucent.com \n \n \n \n \n \nWNT \nUNIX\nPOS\n \n \n \n \n \n480 Red Hill Road \nMiddletown, NJ 07748 USA\n(800) 288-9785\n \n \n \n \n MIMEsweeper\nIntergralis\n \n \n \n \n \n info@us.integralis.com \n \n \n \n \n WNT\nPC\n \n \n \n \n UK:+44(0) 1734 306060\nUS: (206) 889-5841\n \n"
        },
        {
            "page_number": 164,
            "text": " \n \n \n M>Wall \nMATRAnet \n \n \n \n \n delplanque@matranet.com \n \n \n \n \n \nUNIX \nWNT \nWS PC\n \n \n \n \n \n \n18 rue Grange Dame Rose BP \n262 - 78147 Velizy Cedex\nFrance \nTel: +33 (0) 1 34 58 44 58\n \n \n \n \n \nNetCS \nNetCS Informationstechnik\nGmbH\n \n \n \n \n \n Oliver Korfmacher \n \n \n \n \n Router \n \n \n \n \nKatharinenstrasse 18 \nD-10711 Berlin, Germany\nTel: +49.30/89660-0\n \n \n \n \n NetGate \nSmall Works \n \n \n \n \n info@smallworks.com \n \n \n \n \n SUN\nWS  \n \n \n \n \n (512) 338-0619 \n \n \n \n \n \nNetGuard Control Center\n(was Guardian) \nLanOptics\n \n \n \n \n \n sales@lanoptics.com \n \n \n \n \n \nWNT\nW95 \nPC\n \n \n \n \n \n2445 Midway Rd. \nCarrollton, TX 75006 USA\nTel: (972) 738-6900\nProduct\nVendor  \n \n \n \n \n E-mail \n \n \n \n \n \nOS/ \nHardware \nPlatforms\n \n \n \n \n Contact \nInformation\n \n \n \n \n NetRoad/FireWARE/FireWALL\nUkiah Software\n \n \n \n \n \n - \n \n \n \n \n WNT \nNWR PC \n \n \n \n \n - \n \n \n \n \n NetSafe \nSiemens Nixdorf \n \n \n \n \n info@swn.sni.be \n \n \n \n \n UNIX\nWS\n \n \n \n \n \nSNS RD21 Rue de \nNiverlie 11 \nB-5020 Namur Belgium\nTel: +32 (0) 81/55.47.00\n \n \n \n \n Netscreen-100 \nNetscreen Technologies \n \n \n \n \n info@netscreen.com \n \n \n \n \n POS \n \n \n \n \n4699 Old Ironside Drive\nSte. 300 \nSanta Clara, CA 95054 \nUSA \nTel: (408) 970\n \n"
        },
        {
            "page_number": 165,
            "text": " \n \n \n Net Security Master\nSOLsoft SA\n \n \n \n \n \n info@solsoft.com \n \n \n \n \n \nUNIX \nWNT \nWS PC\n \n \n \n \n \n4 bis, rue de la Gare\n92300 Levallois\nFrance \nTel: +33 147155 500\n \n \n \n \n Netra Server \nSun MicroSystems \n \n \n \n \n - \n \n \n \n \n SUN\nWS  \n \n \n \n \nSunSoft, Inc.\n2550 Garcia Ave.\nMountain View, CA \n94043 \n(800) SUN-SOFT\noutside US: 1\n3200\n \n \n \n \n \nNetwork Systems ATM \nFirewall \nBorderGuard \nNetwork Control Facility \nThe Security Router \nStorageTek Network Systems \nGroup\n \n \n \n \n \n webmaster@network.com \n \n \n \n \n - \n \n  \n Tel: (800) NET INFO\n \n \n \n \n Nokia IP & VPN Series \nNokia Telecommunications \n \n \n \n \n info@iprg.nokia.com \n \n \n \n \n Router \n \n \n \n \n232 Java Drive\nSunnyvale, CA 94089 \nUSA \nTel: (408) 990\n \n \n \n \n Norman Firewall \nNorman Data Defense Systems \n \n \n \n \n norman@norman.com \n \n \n \n \n UNIX \nWS PS \n \n \n \n \n3040 Williams Dr. 6th \nFloor \nFairfax, VA 22031 USA\n(703) 573-8802\n \n \n \n \n Novix \nFireFox \n \n \n \n \n - \n \n \n \n \n Novell \n \n  \n (800) 230-6090\n \n \n \n \n Orion \n \n \n \n \n info@zebu.com \n \n \n \n - \n  \n Samantha Agee\n"
        },
        {
            "page_number": 166,
            "text": "Zebu Systems\n \n \n \n(206) 781-9566\n \nProduct\nVendor  \n \n \n \n \n E-mail \n \n \n \n \n \nOS/ \nHardware \nPlatforms\n \n \n \n \n Contact \nInformation\n \n \n \n \n Phoenix Adaptive Firewall\nProgressive Systems\n \n \n \n \n \n info@progressivesystems.com \n \n \n \n \n - \n \n \n \n \n2000 W. Henderson \nRd. \nSte. 400 \nColumbus, OH \n43220 USA\n(614) 326-4600\n \n \n \n \n PORTUS & PORTUS-ES \nLivermore Software Laboratories \n \n \n \n \n portusinfo@Isli.com \n \n \n \n \n ALL\nALL \n \n \n \n \n1830 S. Kirkwood, \nSuite 205 \nHouston, TX 77077 \nUSA \n(281) 759-3274\n \n \n \n \n PrivateNet \nNEC Technologies \n \n \n \n \n info@privatenet.nec.com \n \n \n \n \n BSD \nUNIX \n \n \n \n \n(800) 668-4869\nDepartment Code: \nYCB\n \n \n \n \n Pyramid Firewall\nDataTec\n \n \n \n \n \n antoniob@datatec.co.uk \n \n \n \n \n UNIX\nPOS  \n \n \n \n \n650 Wharfedale Rd.\nWinnersh \nWokingham\nBerkshire RG41 5TP \nUK \nTel: 0118 925 6213\n \n \n \n \n Quiotix \n \n \n \n \n jbs@Quiotix.com \n \n \n \n \n - \n \n \n \n \n - \n \n \n \n \n Raptor \nAxent Technologies \n \n \n \n \n info@axent.com \n \n \n \n \n \nUNIX, \nWNT \nWS, PC\n \n \n \n 2400 Research Blvd.\nRockville, MD \n20850 \n"
        },
        {
            "page_number": 167,
            "text": " \n(888) 44-AXENT\n \n \n \n \n \nSecureLan Secure Access\nSecureFrame \nCylink\n \n \n \n \n \n - \n \n \n \n \n - \n \n \n \n \n8229 Boone Blvd., \nSuite 650 \nVienna, VA 22182 \nUSA \n(800) 449-1162\n \n \n \n \n SecurIT Firewall \nMilkyway Networks \n \n \n \n \n Info@milkyway.com \n \n \n \n \n \nUNIX \nWNT \nWS PC\n \n \n \n \n \n150-2650 \nQueensview Drive\nOttawa, ON, Canada \nK2B 8H6 \n(800) 206-0922\n \n \n \n \n \nSidewinder, SecureZone, and \nSecure Computing Firewall for \nNT \nSecure Computing\n \n \n \n \n \n sales@securecomputing.com \n \n \n \n \n PC \n \n \n \n \nContact Info\nWorldwide\nTel: +1 (408) 487\n1900\n \n \n \n \n Site Patrol \nBBN Planet Corp. \n \n \n \n \n Gregg Lebovitz \n \n \n \n \n - \n \n \n \n \n - \n \n \n \n \n SmartWall\nV-ONE\n \n \n \n \n \n sales@v-one.com \n \n \n \n \n - \n \n \n \n \n20250 Century Blvd.\nGermantown, MD \n20874 USA\n(301) 515-5200\n \n \n \n \n SonicWall \nSonic Systems \n \n \n \n \n info@sonicsys.com \n \n \n \n \n \nAny \nPOS\n \nOS/\n \n \n \n \n \n575 N. Pastoria Ave.\nSunnyvale, CA \n94086 USA\n(408) 736-1900\n \nProduct\nVendor  \n \n \n \n E-mail \n \n \n \n \n \nOS/ \nHardware \nPlatforms\n  \n \n \n Contact \nInformation \n"
        },
        {
            "page_number": 168,
            "text": " \n \n \n \n \n \n SPF-100/SPF-200 \nSun MicroSystems \n \n \n \n \n sunscreen@incog.com \n \n \n \n \n POS \nWS, PC \n \n \n \n \nMountain View, CA 94043\nUSA \n(415) 960-3200\n \n \n \n \n Sygate \nSyberGen \n \n \n \n \n - \n \n \n \n \n W95, WNT\nPC\n \n \n \n \n \n - \n \n \n \n \n \nTurnstyle Firewall System \n(TFS) \nTurnstyle Internet Module \n(TIM) \nAtlanta Systems Group\n \n \n \n \n \n US Sales and Marketing\nCanada\n \n \n \n \n \n UNIX\n-\n \n \n \n \n US: (516) 737-6435\nCAN: (506) 453-3505\n \n \n \n \n VCS Firewall \nThe Knowledge Group \n \n \n \n \n sales@ktgroup.co.uk \n \n \n \n \n UNIX\nPC\n \n \n \n \n \nConcorde Road \nPatchway, Bristol UK\nTel: ++44 (0) 117 900 \n7500\n \n \n \n \n VPCom \nAshley Laurent \n \n \n \n \n Jeffrey Goodwin \n \n \n \n \n UNIX\nNT\n \n \n \n \n \n707 West Avenue, Suite \n201 \nAustin, TX 78701 USA\n(512) 322-0676\n \n \n \n \n Watchguard \nWatchguard Technologies \n \n \n \n \n sales@watchguard.com \n \n \n \n \n All\nPC \n \n \n \n \nContacts: \n(888) 682-1855 or\n(206) 521-8340\n \n \n \n \n WebSENSE \nNetPartners Internet Solutions \n \n \n \n \n sales@netpart.com \n \n \n \n \n - \n \n \n \n \n9210 Sky Park Court\nSan Diego, CA 92123 USA\n(800) 723-116 or \n(619) 505-3020\n \n \n \n \n \n"
        },
        {
            "page_number": 169,
            "text": "  WinGate \nDeerfield Communications  \n \n  info@deerfield.com  \n \n  WNT, W95\nPC\n  \n \n  (517) 732-8856  \n \n \n \n \n ZapNet! \nIPRoute/Secure \n \n \n \n \n info@iproute.com \n \n \n \n \n WNT\nPC\n \n \n \n \n \nSuite 400-120, \n10945 State Bridge Rd.\nAlpharetta, GA 30202 \nUSA \n(770) 772-4567\n \n \n \n \n UNIX, LINUX, WIN= Windows, W95= Windows95, WNT= WindowsNT \n \n \n \n \n POS= Proprietary or embedded operating system \n \n \n \n \n WS= RISC-based workstation \n \n \n \n \n PC= Intel-based personal computer \n \nAppendix 6.B— \nFirewall Resellers \n \n \n \n \n Company \n \n \n \n \n Products/\nServices  \n \n \n \n \n E-mail \n \n \n \n Contact\nInformation\n \n \n \n \n ANR \n \n \n \n \n NetGuard Control Center and \nSecurity consulting\n \n \n \n \n \n info@anr.co.il \n \n \n \n \n20 Admonit St.\nNetanya 42204 Isra\nTel: 972\n \n \n \n \n Astra Network \n \n \n \n \n Security and network consulting, \nFirewall installation & admin.\n \n \n \n \n \n infosec@man.net \n \n \n \n \n2633 Portage Ave.\nWinnipeg, MB Can\nR3J 0P7\n(204) 987\n \n"
        },
        {
            "page_number": 170,
            "text": " \n \n \n Atlantic Computing\nTechnology\n \n \n \n \n \n BorderWare \n \n \n \n \n info@atlantic.com \n \n \n \n \n84 Round Hill Roa\nWethersfield, CT 0\n(203) 257\n \n \n \n \n \nARTICON \nInformation Systems\nGmbH\n \n \n \n \n \n BorderWare \n \n \n \n \n - \n \n \n \n \n - \n \n \n \n \n Bell Atlantic Network\nIntegration\n \n \n \n \n \n Network and firewall design \n \n \n \n \n sales.lead@bani.com \n \n \n \n \n52 East Swedesford\nFrazer, PA 19355 U\n(800) 742\n \n \n \n \n BRAK Systems \n \n \n \n \n Firewall-1, WebSENSE and \nothers\n \n \n \n \n \n tony@brak.com \n \n \n \n \n1 City Centre, Driv\nSuite 801\nMississauga, Ontar\nCanada L6S 4T2\n(905) 272\n \n \n \n \n Cadre Computer\nResources\n \n \n \n \n \n Firewall-1, Secured ISP, Internet \ndevelopment\n \n \n \n \n \n info@ccr.com \n \n \n \n \n3000 Chemed Cent\n255 East Fifth Stree\nCincinnati, OH 452\n(513) 762\n \n \n \n \n C-CURE \n \n \n \n \n Information Security Architects \n \n \n \n \n luc.dooms@c-cure.be \n \n \n \n \nK. Rogierstraat 27\nB-2000 Antwerpen\nBelgium\nTel +32 (0) 3 216.5\n \n \n \n \n \nCentaur \nCommucication\nGmbH\n \n \n \n \n \n Firewall-1 and others, Security \ncosulting\n \n \n \n \n \n info@centaur.de \n \n \n \n \nUrbanstrasse 68\n74074 Heilbronn G\nTel +49 7131 799 0\n \n \n \n \nCitadel Security \n \n \n \nGauntlet, many others, and \n \n \n \ninfocit@citadel.\n \n726 High Street\n"
        },
        {
            "page_number": 171,
            "text": " Management Systems \n \n Internet consulting\n \n \n com.au\n \n \n  Armadale, VIC AU\nTel 03 9500 2990\n \nCompany \n \n \n \n \n Products/\nServices  \n \n \n \n \n E-mail \n \n \n \n Contact\nInformation\n \n \n \n \n CleverMinds \n \n \n \n \n Gauntlet, others, and security \nconsulting\n \n \n \n \n \n Jack Boyle \n \n \n \n Bedford, MA U\nTel: (781) 275\n \n \n \n \n Cohesive Systems \n \n \n \n \n Centri Firewall Information \nservices\n \n \n \n \n \n info@gi.net \n \n \n \n \n755 Page Mill R\nSuite A\nPalo Alto, CA 9\nUSA\n(800)682\n \n \n \n \n Comark \n \n \n \n \n Firewall-1, and others, Security \nconsulting\n \n \n \n \n \n webteam@comark\nnet\n \n \n \n \n \n444 Scott Drive\nBloomingdale, I\n60108\nUSA\nTel: (630) 924\n \n \n \n \n Comnet Solutions \n \n \n \n \n Shareware firewall: Midiator \nOne\n \n \n \n \n \n firewall@comnet\ncom.au\n \n \n \n \n \nComNet Solutio\nPty Ltd Unit 4/1\nOld Castle Hill R\nCastle Hill, NSW\nAustralia\nTel +61 2 899 5\n \n \n \n \n Computer Software\nManufaktur (CSM) \n \n \n \n \n CSM Proxy & Proxy Plus Internet \nGateway with fw features\n \n \n \n \n \n sales@csm-usa com \n \n \n \n \nP.O. Box 1105\nLayton, UT 840\nUSA\n(801) 547\n \n \n \n \n Com Tech \nCommunications \n \n \n \n Firewall-1 and others Online \nSecurity Services\n \n \n \n \n rof@comtech.com.au \n \n \n \n Australia\nTel: 0412 16337\n"
        },
        {
            "page_number": 172,
            "text": " \n \n \n \n \n \n Connect GmbH \n \n \n \n \n SmartWall BorderGuard \nNetStalker and Virus Wall\n \n \n \n \n \n Armin Bolenius \n \n \n \n \nPichlmayrstr. 26\nD-83024 Rosenh\nGermany\nTel: +49\n \n \n \n \n Collage \nCommunications \n \n \n \n \n Many firewalls and network \nservices\n \n \n \n \n \n cyberguard@CollCom \nCOM\n \n \n \n \n \n12 Tulip Lane\nPalo Alto, CA 9\nUSA\n \n \n \n \n Comcad GmbH \n \n \n \n \n Firewall-1 others, and security \nand network services\n \n \n \n \n \n frank.recktenwald@comcad.de \n \n \n \n \nIndustriestr.23\n51399 Burscheid\nGermany\nTel +49\n \n \n \n \n Conjungi \n \n \n \n \n Gauntlet \n \n \n \n \n simon@conjungi.com \n \n  \n Seattle, WA US\n \n \n \n \n CREDO NET \n \n \n \n \n Raptor and security consulting \n \n \n \n \n info@credo.net \n \n \n \n \n22941 Triton W\nSuite 241\nLaguna Hills, C\n(888) 88\nCompany \n \n \n \n \n Products/\nServices  \n \n \n \n \n E-mail \n \n \n \n Contact\nInformation\n \n \n \n \n CyberCorp \n \n \n \n \n NetGuard Control Center \n \n \n \n \n amuse@cyberservices.com \n \n \n \n \n2934 West Roy\nSuite 3136\nIrving, TX 7506\nUSA\n(972) 738\n \n \n \n \n Cypress Systems \n \n \n \n \n Raptor \n \n \n \n \n rmck \n \n \n \n P.O. Box 9070\nMcLean, VA 22\nUSA\n"
        },
        {
            "page_number": 173,
            "text": "(703) 273\n \n \n \n \n Data General \n \n \n \n \n Cybershield and Raptor \n \n \n \n \n sense@dg.com \n \n \n \n \n4400 Computer\nWestboro, MA \nUSA\n(800) 4DG\n \n \n \n \n \nDecision-Science\nApplications \nApplications\n \n \n \n \n \n BorderWare FireWall-1\nSidewinder and more\n \n \n \n \n \n infosec@dsava.com \n \n \n \n \n1110 N. Glebe \nSuite 400\nArlington, VA \nUSA\n(703) 875\n243-\n \n \n \n \n Deerfield \nCommunications\n \n \n \n \n \n WinGate Instant Internet \n \n \n \n \n info@deerfield.com \n \n  \n (517) 732\n \n \n \n \n DFC International Ltd. \n \n \n \n \n IBM Firewall \n \n \n \n \n sales@dfc.com \n \n \n \n \n52 Mowatt Cou\nThornhill, Onta\nL3T 6V5\nTel: (905) 731\n \n \n \n \n Digital Pathways UK \n \n \n \n \n Sidewinder \n \n \n \n \n - \n \n \n \n \n - \n \n \n \n \n Dimension Data \nSecurity\n \n \n \n \n \n Firewall-1 and security products \n \n \n \n \n mouritz@ddsecurity.co.za \n \n \n \n \n158 Jan Smuts A\nRosebank P.O. \n3234\nParklands 2121\nSouth Africa\nTel: +27 (0)11\n5116\n \n \n \n \n DNS Telecom \n \n \n \n \n SecureNet and SecurSite, \nWatchguard and others\n \n \n \n \n \n preynes@dnstele com.fr \n \n \nImmeuble La F\n2 \nplace des vosge\n"
        },
        {
            "page_number": 174,
            "text": "  \n92051\nParis la Difense\nFrance\nTel: +33 (0)1 4\n17\n \n \n \n \n Dynavar Networking \n \n \n \n \n Ascend, Cisco PIX and many \nothers\n \n \n \n \n \n sales@dynavar.com \n \n \n \n \n#300, 1550 \nSW \nCalgary, Albert\nCanada T2R 1K\nTel: (403) 571\nCompany \n \n \n \n \n Products/\nServices  \n \n \n \n \n E-mail \n \n \n \n Contact\nInformation\n \n \n \n \n East Coast Software \n \n \n \n \n NetGuard Control Center and \nsecurity consulting\n \n \n \n \n \n infotec@eastcoastsw.com.au \n \n \n \n \nPO Box 6494\nSt Kilda Road\nCentral, Melb\nVictoria 3004\nTel: 61\n \n \n \n \n Electric Mail \n \n \n \n \n Firewall-1 and Internet products \n \n \n \n \n info@elmail.co.uk \n \n \n \n \nMerlin Place, \nRoad\nCambridge\nUnited Kingd\nTel: +44 (0) 1\n501 333\n \n \n \n \n EMJ America \n \n \n \n \n AFS 2000, BorderWare and \nothers\n \n \n \n \n \n mrkusa@emji.net \n \n \n \n \n1434 Farringt\nRoad\nApex, NC 275\nUSA\n(800) 548\n \n \n \n \n Enstar Networking \n \n \n \n \n Firewall-1 and security \nconsulting\n \n \n \n \n \n baustin@enstar.com \n \n \n \n \n8304 Esters B\nSuite 840\nDallas, TX 75\nUSA\n"
        },
        {
            "page_number": 175,
            "text": "(800) 367\n \n \n \n \n Enterprise System \nSolutions\n \n \n \n \n \n BorderWare \n \n \n \n \n - \n \n \n \n \n - \n \n \n \n \n FishNet Consulting \nServices\n \n \n \n \n \n Firewall-1, others, and security \nconsulting\n \n \n \n \n \n info@kcfishnet.com \n \n \n \n \n7007 College \nSuite 450\nOverland Park\n66211\nUSA\n(913) 498\n \n \n \n \n Garrison Technologies \n \n \n \n \n Security consulting, firewalls, \naudits, etc\n \n \n \n \n \n sales@garrison.com \n \n \n \n \n100 Congress\nSuite 2100\nAustin, TX 78\nUSA\n(512) 302\n \n \n \n \n GearSource \n \n \n \n \n NetGuard Control Center PIX \n \n \n \n \n sales@GearSource.com \n \n \n \n \n5015 Victor S\nDallas, TX 75\nUSA\n(214) 821\n \n \n \n \n Global Data Systems \n \n \n \n \n Firewalls and security \n \n \n \n \n glasane@gdsconnect.com \n \n  \n (781) 740\n \n \n \n \n Global Technology \nAssociates\n \n \n \n \n \n GTA Firewalls and security \nservices\n \n \n \n \n \n david@globaltech.co.uk \n \n \n \n \n71 Portland R\nWorthing\nWest Sussex B\n1QG\nEngland\nTel: 44(0) 190\n51 51\n \nCompany \n \n \n \n Products/\nServices  \n \n \n \n E-mail \n \n \n Contact\nInformation\n"
        },
        {
            "page_number": 176,
            "text": " \n \n \n \n \n \n \n Graphics Computer \nSystems\n \n \n \n \n \n Firewall-1 and security services \n \n \n \n \n sales@gcs.com.au \n \n \n \n \n97 Highbury\nBurwood, V\nAustralia 31\nTel: +61 3 9\n8522\n \n \n \n \n \nHaystack Labs now \nowned by Network \nAssociates\n \n \n \n \n \n Stalker \nIntrusion detection system \n \n \n \n \n info@haystack.com \n \n \n \n \n10713 RR62\nSuite 521\nAustin, TX 7\nUSA\n(512) 918\n \n \n \n \n Herve Schauer \nConsultants\n \n \n \n \n \n HSC GateKeeper \n \n \n \n \n info@hsc.fr \n \n \n \n \n142, rue de R\n75039 Paris \n0!\nFrance\nTel: +33 (1)\n46.38.89.90\n \n \n \n \n HomeCom Internet \nSecurity Services\n \n \n \n \n \n Firewall and security sales and \nconsulting\n \n \n \n \n \n security@homecom.com \n \n \n \n \n1900 Gallow\nVienna, VA\nUSA\n(703) 847\n \n \n \n \n IConNet \n \n \n \n \n Internet in a Rack (IR) Network \nhardware and software\n \n \n \n \n \n info@iconnet.net \n \n  \n -\n \n \n \n \n Inflo Communications \nLimited\n \n \n \n \n \n Raptor firewall and others \n \n \n \n \n sales@inflo.co.uk \n \n \n \n \nMountcharle\nDonegal Tow\nDonegal Irel\nTel: +353 73\n \n \n \n \n Ingress Consulting \nGroup\n \n \n \n \n BorderWare and others \n \n \n \n \n sales@nohackers.com \n \n \n \n 60 Guild Str\nNorwood, M\n"
        },
        {
            "page_number": 177,
            "text": " \n02062 USA\n(888) INGR\n \n \n \n \n Integralis UK \n \n \n \n \n MIMEsweeper \n \n \n \n \n msw.support@integralis.co.uk \n \n \n \n U.K. +44 (0\n306060\n \n \n \n \n Integralis USA \n \n \n \n \n \n info@us.integralis.com \n \n  \n US (206) 88\n \n \n \n \n Intercede Ltd. \n \n \n \n \n MilkyWay SecureIT \n \n \n \n \n sales@mtercede.co.uk \n \n \n \n \n1 Castle Stre\nHinckley,\nLeicestershir\nLE10 IDA\nUK\nTel: +44 (0)\n250 266\n \n \n \n \n INTERNET GmbH \n \n \n \n \n BorderWare consulting \n \n \n \n \n Ingmar Schraub \n \n \n \n \nAm Burgack\nD\nGermany\nTel: +49\n3999\n \n \n \n \n Jerboa \n \n \n \n \n Independent security and firewall \nconsulting\n \n \n \n \n \n info@jerboa.com \n \n \n \n \nBox 382648\nCambridge, \n02238\n(617) 492\n Company \n \n \n \n \n Products/\nServices  \n \n \n \n \n E-mail \n \n \n \n Contact\nInformation\n \n \n \n \n Kerna Commuications \nLtd.\n \n \n \n \n \n Security consultants, firewall \ninstallation and support\n \n \n \n \n \n sales@kerna.ie \n \n \n \n \n3 Arbourfield\nTerrace,\nDundrum, Du\nIreland\nTel: +353\n2964396\n"
        },
        {
            "page_number": 178,
            "text": " \n \n \n \n The Knowledge Group \n \n \n \n \n VCS Firewall and Security \nproducts\n \n \n \n \n \n sales@ktgroup.co.uk \n \n \n \n \nConcorde Ro\nPatchway, Br\nUK\nTel: +44 (0) 1\n7500\n \n \n \n \n LANhouse \nCommunications\n \n \n \n \n \n Ascend, Centri Firewall, Gauntlet \nand security consulting\n \n \n \n \n \n sales@lanhouse.com \n \n \n \n \n510 King Stre\nSuite 202\nToronto, Onta\nCanada\nM5A 1M1\nTel: (416) 36\n \n \n \n \n LURHQ Corporation \n \n \n \n \n Security consulting, firewalls, \nWeb server security\n \n \n \n \n \n info@lurhq.com \n \n \n \n \nP.O. Box 286\nConway, SC \nUSA\n(843) 347\n \n \n \n \n \nMadison Technology\n \nGroup\n \n \n \n \n \n \nSecurity/FW design &\n \nimplementation\n \n \n \n \n \n Steveng@microlan.com \n \n \n \n \n331 Madison \n6th Floor\nNYC, NY 10\nUSA\n(212) 883\n \n \n \n \n Master Software \nTechnology\n \n \n \n \n \n Alta Vista and CS consulting \n \n \n \n \n sales@masteredge.com \n \n \n \n \n92 Montvale \nStoneham, M\n02180\nUSA\n(617) 438\n \n \n \n \n media communications \neur ab\n \n \n \n \n \n Gauntlet security \n \n \n \n \n neil@medcom.se \n \n \n \n \nBox 1144\n111 81 Stock\nSweden\nTel: \n+46.708.4322\n(GSM)\n"
        },
        {
            "page_number": 179,
            "text": " \n \n \n \n Mergent International \n \n \n \n \n Gauntlet security consulting \n \n \n \n \n info@mergent.com \n \n  \n (800) 688\n \n \n \n \n Midwest Systems \n \n \n \n \n Gauntlet Sidewinder VAR \n \n \n \n \n lschwanke@midwestsys.com \n \n \n \n \n2800 Southcr\nWest\nBurnsville, M\n55306\nUSA\n(612) 894\n \n \n \n \n MPS Ltd. \n \n \n \n \n M>Wall and security consulting \n \n \n \n \n johnt@mpsuk.com \n \n \n \n \nThe Manor S\nGreat Somerf\nNr. Chippenh\nWiltshire SN\nEngland\nTel: +44 (0) 1\n721414\nCompany \n \n \n \n \n Products/\nServices  \n \n \n \n \n E-mail \n \n \n \n Contact\nInformation\n \n \n \n \n NetGuard \n \n \n \n \n NetGuard Control Center \n \n \n \n \n solutions@ntguard.com \n \n \n \n \n11350 Random Hil\nSuite 750\nFairfax, VA 22030\n(703) 359\n \n \n \n \n NetPartners Internet \nSolutions\n \n \n \n \n \n WebSENSE, Firewall-1, Raptor, \nand others\n \n \n \n \n \n sales@netpart.com \n \n \n \n \n9665 Chesapeake D\nSuite 350\nSan Diego, CA 921\n(800) 723\n(619) 505\n \n \n \n \n Netrex \n \n \n \n \n Secure Internet Solutions \nFirewall-1\n \n \n \n \n \n info@netrex.com \n \n \n \n \n3000 Town Center\nSuite 1100\nSouthfield, MI 480\n(800) 3 NETREX\n"
        },
        {
            "page_number": 180,
            "text": " \n \n \n \n Network Associates \n \n \n \n \n Gauntlet and many security \nproducts\n \n \n \n \n \n contacts \n \n \n \n \n3965 Freedom Circ\nSanta Clara, CA 95\nUSA \n(408) 988\n \n \n \n \n Network Security \n \n \n \n \n Firewall-1, NetScreen and others \n \n \n \n \n info@nsec.net \n \n \n \n \n369 River Road\nNorth Tonawanda,\nNY 14120 USA\n(716) 692\n \n \n \n \n Obtuse Systems \n \n \n \n \n Juniper \n \n \n \n \n info@obtuse.com \n \n  \n Alberta, Canada\n \n \n \n \n Breakwater Security \nAssociates\n \n \n \n \n \n Many firewalls and security \nconsulting\n \n \n \n \n \n info@breakwater.net \n \n  \n (206) 770\n \n \n \n \n Orbis Internet \n \n \n \n \n Sidewinder and security \nconsulting\n \n \n \n \n \n dan@orbis.net \n \n \n \n \n475 Cleveland Ave\nSuite 222\nSt. Paul, MN 55104\n(612) 603\n \n \n \n \n Qualix \n \n \n \n \n Firewall-1 and secunty products \n \n \n \n \n qdirect@qualix.com \n \n \n \n \n177 Bovet Road, 2\nSan Mateo, CA 944\nUSA \n(650) 572\n \n \n \n \n Racal Airtech Ltd \n \n \n \n \n Raptor Security services \n \n \n \n \n Sohbat Ali \n \n \n \n \nMeadow View Hou\nLong Crendon, Ay\nBuckinghamshire, \nHP 18 9EQ\nTel: 01844 201800\nCompany \n \n \n \n Products/\nServices  \n \n \n \n E-mail \n \n \n Contact\nInformation\n"
        },
        {
            "page_number": 181,
            "text": " \n \n \n \n \n \n \n Racal Gaurdata \n \n \n \n \n \nRaptor \n \nSecurity services\n \n \n \n \n \n Sohbat Ali \n \n \n \n \n480 Spring Pa\nPlace\nSuite 900\nHerndon, VA\nUSA\n(703) 471\n \n \n \n \n Reese Web \n \n \n \n \n Raptor \n \n \n \n \n pp001261@interramp.com \n \n \n \n \nRocky Point \nHarbour\n3309 Diamon\nCircle\nTampa, FL 33\nUSA\n(813) 286\n \n \n \n \n Sandman Security of \nSmoke N' Mirrors Inc\n \n \n \n \n \n Firewall-1 Raptor and others \n \n \n \n \n Ben Taylor \n \n \n \n \n1165 Herndon\nSuite 200\nHerndon, Va \nUSA\n(703) 318\n \n \n \n \n Sea Change Pacific \nRegion\n \n \n \n \n \n Border Ware \n \n \n \n \n jalsop@seachange.com\nmichael@seawest\n \n \n \n \n \n5159 Beckton\nVictoria, Briti\nColumbia\nV8Y 2C2\n(604) 658\n \n \n \n \n Sea Change Europe Ltd \n \n \n \n \n Border Ware \n \n \n \n \n jalsop@seachange.com \npeter@sea-europe.co.uk \n \n \n \n \n470 London R\nSlough, Berks\n8QY\nUK\nTel: 44\n \n \n \n \n Secure Network Systms \n \n \n \n \n Many firewalls and security and \nnetwork consulting\n \n \n \n \n \n info@blanket.com \n \n \n \n Lawrence, Ka\nUSA\n \n"
        },
        {
            "page_number": 182,
            "text": " \n \n \n SecureXpert Labs FSC \nInternet Corp\n \n \n \n \n \n \nSecurity consulting and \ninformation \nFireWall-1 and others\n \n \n \n \n \n security@securexpert.com \n \n \n \n \n - \n \n \n \n \n Serverware Group plc \n \n \n \n \n iWay-One \n \n \n \n \n sale@serverwr.de.mon.co.uk \n \n \n \n Tel: (44) 1732\n464624\n \n \n \n \n Sherwood Data Systems \nLtd\n \n \n \n \n \n Karl Bridge/Karl Brouter \n \n \n \n \n sales@gbnet.com \n \n \n \n \nHigh Wycomb\nTel: +44 (0) 1\n464264\n \n \n \n \n Silicon Graphics \n \n \n \n \n Gauntlet and Security \n \n \n \n \n Sales \n \n \n \n \n39001 West 1\nRd.\nFarmington, M\n48331\nUSA\n(800) 800\n \n \n \n \n Siemens Nixdorf \n \n \n \n \n TrustedWeb \n \n \n \n \n info@trustedweb.com \n \n \n \n \nFitxwilliam C\nLeeson Close\nDublin 2, Irel\nCompany \n \n \n \n \n Products/\nServices  \n \n \n \n \n E-mail \n \n \n \n Contact\nInformation\n \n \n \n \n SkyNet Czech Republic \n \n \n \n \n Gauntlet \n \n \n \n \n Roman.Paklik@SkyNet.CZ \n \n \n \n \nKabatnikova 5\n602 00 Brno\nCzech Republi\nTel: +420 5 41\n79\n \n \n \n \n SMC Electronic \nCommerce Ltd\n \n \n \n \n \n Gauntlet, many other FWs, and \nsecurity consulting\n \n \n \n \n \n info@smcgroup.co.za \n \n \n \n 52 Wierda Rd W\nWierda Vally, \nSandton\n"
        },
        {
            "page_number": 183,
            "text": "South Africa\nTel: +27 11 26\n \n \n \n \n Softway Pty Ltd \n \n \n \n \n Gauntlet and security consulting \n \n \n \n \n enquiries@softway.com.au \n \n \n \n \nP.O. Box 305\nStrawberry Hil\nNSW 2012 Au\nTel: +61 2 969\n \n \n \n \n Stallion Ltd. \n \n \n \n \n Firewall-1 and security services \n \n \n \n \n stallion@stallion.ee \n \n \n \n \nMustamae tee 5\nTallinn 10621\nEstonia\nTel: +372 6 56\n \n \n \n \n Stonesoft Corporation \n \n \n \n \n Stonebeat and Firewall-1 \n \n \n \n \n info@stone.fi \n \n \n \n \nTaivalm=E4ki \nFIN\nEspoo, Finland\nTel: +358 9 47\nTel: +358 9 47\n1282\n \n \n \n \n StorageTek Network \n \n \n \n \n ATM, NCF, BorderGuard \n \n \n \n \n webmaster@network.com \n \n  \n Tel: (800) NET\n \n \n \n \n Sun Tzu Security \n \n \n \n \n Firewalls and security consulting \n \n \n \n \n info@suntzu.net \n \n  \n Tel: (414) 289\n \n \n \n \n Symbolic \n \n \n \n \n COOL-FIRE and security \nservices\n \n \n \n \n \n mt@symbolic.it \n \n \n \n \nViale Mentana\n29 I\nItaly\nTel: +39 521 7\n \n \n \n \n Technology Management \nSystems\n \n \n \n \n \n BlackHole \n \n \n \n \n tmsinc@erols.com \n \n \n \n \nVienna, Va 22\nUSA\nTel: (703) 768\n \n \n \n \n \n"
        },
        {
            "page_number": 184,
            "text": "  Technology Transition \nServices\n  \n \n  Fort Knox Firewall and security \nconsulting\n  \n \n  sales@techtranserv.com  \n \n  \n100 Blue Run R\nIndianola, PA \nUSA\n \n \n \n \n Trident Data Systems \n \n \n \n \n SunScreen Firewall-1 and security \nconsulting\n \n \n \n \n \n \nAnthony Dinga (east)\n \nBob Hermann (west) \nCharlie Johnson \n(midwest)\n \n \n \n \n \n5933 W. Centr\nBlvd.\nSuite 700\nLos Angles, CA\nUSA\nTel: (310) 645\nCompany \n \n \n \n \n Products/\nServices  \n \n \n \n \n E-mail \n \n \n \n Contact\nInformation\n \n \n \n \n Tripcom Systems \n \n \n \n \n Firewall-1 and Internet \nconsulting\n \n \n \n \n \n Adam Horwitz \n \n \n \n Naperville, IL\nTel: (708) 778\n \n \n \n \n Trusted Information \nSytems\n \n \n \n \n \n Gauntlet \n \n \n \n \n tis@tis.com (US) \ninfo@eu.tis.com (non-US)\n \n \n \n \n \nTel: (301 854\nor \n(410) 442\n \n \n \n \n Trusted Network \nSolutions\n \n \n \n \n \n Gauntlet and security consulting \n \n \n \n \n \n-\n \n \n \n \n Johannesburg,\nSouth Africa\n \n \n \n \n UNIXPAC Australia \n \n \n \n \n Raptor and network consulting \n \n \n \n \n info@unixpac.com.au \n \n \n \n \nCremorne, Au\nTel: (02) 9953\nor \n1 800 022 137\n \n \n \n \n Uunet \n \n \n \n \n Raptor and others security \nconsulting\n \n \n \n \n \n info@uu,net \n \n \n \n \n3060 William\nFairfax, VA 2\nUSA\n(800) 488\n(703) 206\n \n"
        },
        {
            "page_number": 185,
            "text": " \n \n \n Vanstar \n \n \n \n \n All major security products and \nconsulting\n \n \n \n \n \n Jrecor@vanstar.com \n \n \n \n \n30800 Telegra\nSuite 1850\nBingham Farm\n48312\nUSA\n(810) 540\n \n \n \n \n We Connect People Inc. \n \n \n \n \n Security FWs, Internet consulting \nand more\n \n \n \n \n \n sales@wcpinc.com \n \n \n \n California, US\n(408) 421\n \n \n \n \n WheelGroup \n \n \n \n \n \nNetRanger/NetSonar, intrusion \ndetection and other security \nproducts\n \n \n \n \n \n sales@wheelgroup.com \n \n \n \n \n13750 San Ped\nSuite 670\nSan Antonio, T\n78232\nUSA\n(210) 494\n \n \n \n \n X + Open Systems Pty \nLtd\n \n \n \n \n \n Security FWs and network \nservices\n \n \n \n \n \n info@xplus.com.au \n \n \n \n \nP.O. Box 6456\nShoppingworl\nNorth Sydney\n2059\nAU\nTel: +61 2 995\n \n \n \n \n Zeuros Limited \n \n \n \n \n Raptor and network services \n \n \n \n \n les@zeuros.co.uk \n \n \n \n \nTudor Barn, F\nLane\nRotherwick, \nHampshire,\nRG27 9BE, U\nTel: 44(0) 125\n760081\n \n \n \n \n ZONEOFTRUST.COM \n \n \n \n \n Security products, FWs and \nservices\n \n \n \n \n \n info@zoneoftrust.com \n \n \n \n \n22941 Triton W\n2nd Flr.\nLaguna Hills, \n92653\nUSA\n(714) 859\n"
        },
        {
            "page_number": 186,
            "text": "Appendix 6.C— \nPublic Domain, Shareware, Etc. \n \n \n \n \n • Drawbridge \n \n \n \n \n Available at net.tamu.edu  \n \n \n \n \n • Freestone by SOS Corporation \n \n \n \n \n Freestone is an application gateway firewall package, a genetic derivative of Brimstone produced by \nSOS Corporation. Freestone can be retrieved from the Columbia, SOS, and COAST FTP sites\n \n \n \n \n \n • fwtk firewall toolkit \n \n \n \n \n Available from ftp.tis.com Look in /pub/firewalls and /pub/firewalls/toolkit for documentation and \ntoolkit.\n \n \n \n \n \n • ISS \n \n \n \n \n \nInternet Security Scanner is an auditing package that checks domains and nodes searching for well-\nknown vulnerabilities and generating a log for the administrator to take corrective measures. For \ninformation, history, and the commercial versions by the originator, visit: www.iss.net. For the \npublicly available version, I have no current ftp site, but you might try various search engines.\n \n \n \n \n \n • Mediator One \n \n \n \n \n Non-commercial shareware firewall. \n \n \n \n \n • router-config tool \n \n \n"
        },
        {
            "page_number": 187,
            "text": "  \nA set of KomShell scripts (known to run cleanly on UnixWare and Solaris, and with very minor \ntweeking on SGI's but doesn't run with pd-ksh) to build the very complex router configurations \nneeded for high-quality packet-filtering firewalls (only generates configs for Cisco, though there are \nfacilities for adding other router types). Software is available from the Freebird Archive.\n  \n \n \n \n \n • SOCKS \n \n \n \n \n The SOCKS package, developed by David Koplas and Ying Da Lee. Available by ftp from \nftp.nec.com. Also see: www.socks.nec.com\n \n \n \n \n* The source of these appendices is www.waterw.com/~manowar/vendor.html. Please check this web site for the \nlatest information\nChapter 7— \nSecurity Policy \n \n \n \n \n \nSecurity concerns have heightened in recent years. News stories about viruses and computer fraud \ndominate. Information technology (IT) managers have to decide how to protect information and \ncomputer technology.\n \n \n \n \n \n \nIn developing a security policy, you must consider not only actual security threats but also the \nsecurity perceptions of the public. For example, although it is in fact much safer to provide a credit \ncard number over the Internet to a legitimate company than to give it to an unscrupulous employee in \na face-to-face transaction, many people express concern about providing credit card numbers over the \nInternet. Not responding to the public's concerns may result in significant financial loss.\n \n \n \n \n \n \nAs a manager you face several trade-offs concerning security management. While some may be \nquantified, others—such as determining the organization's strategic direction—cannot. That's why \nboth the IT and the human resources departments should be considered in formulating a security \nplan.\n \n \n \n \n \n Without an adequate security policy, your organization is vulnerable to many threats, including: \n \n \n \n \n • Theft of both electronic and physical resources, including data \n \n \n \n \n • Unauthorized modification of data \n"
        },
        {
            "page_number": 188,
            "text": " \n \n \n \n • Fraud and other illegal activities \n \n \n \n \n • Disclosure of confidential and proprietary information \n \n \n \n \n • Unintentional errors caused by carelessness \n \n \n \n \n • Intentional sabotage caused by current or former disgruntled employees\n• Spying or sabotage by competitors \n \n \n \n \n • Inability to continue business after an emergency or disaster \n \n \n \n \n Managing Computer Security \n \n \n \n \n In formulating a policy you must first ask yourself some questions: \n \n \n \n \n 1. What resources need to be protected? \n \n \n \n \n 2. Against whom must we protect our system? \n \n \n \n \n 3. How much can we spend to protect the system? \n \n \n \n \n 4. What benefits will we derive from the expenditure? Is the benefit worth the cost? \n \n \n \n \n 5. What happens if security is compromised? How will we respond? \n \n \n \n \n 6. What are our contingency plans? \n \n"
        },
        {
            "page_number": 189,
            "text": " \n \n \n \nPrivate, confidential, proprietary data is typically one important resource you want to protect. \nOtherwise, your financial assets might be compromised. Other valuable resources include CPU \nprocessing cycles or computer time. While attackers are typically interested in obtaining access to \nconfidential data, some may simply want to deny legitimate users access to computer facilities.\n \n \n \n \n \n \nEven information not likely to be valuable to anyone else should be protected. Hackers often steal or \ndestroy data simply because it's there. They may also delete or destroy files in an attempt to cover up \nillegal activity.\n \n \n \n \n \n \nWhen you're calculating the cost of security precautions, consider not only the direct but also the \nindirect costs. Direct costs may be for equipment, installation, and training. Indirect costs include the \neffects on morale and productivity.\n \n \n \n \n \n \nIt's prudent to recognize that increasing security decreases convenience. Employees and others, such \nas customers and suppliers, may resent the inconvenience. Too much security may be just as \ndetrimental as too little; the goal is an optimal equilibrium.\n \n \n \n \n \n \nIn spite of precautions, security will eventually be compromised. The steps you take to recover from \nsuch a breach may mean the difference between success and failure in your business. Think about \nsuch questions as the following:\n• What will be the financial impact of a security breach? \n \n \n \n \n • Do you have enough insurance? \n \n \n \n \n • What may be the legal consequences of a breach? \n \n \n \n \n • How will lost data, information, or assets be recovered? \n \n \n \n \n • How will the breach affect employees, suppliers, or customers? \n \n \n \n \n • How can similar events be prevented in the future? \n \n"
        },
        {
            "page_number": 190,
            "text": " \n \n \n Creating the Policy and the Plan \n \n \n \n \n The purpose of a security plan is to assign accountability. The security policy should define what is \nand is not acceptable.\n \n \n \n \n \n \nThe plan should clearly specify the penalties for unacceptable behavior. How will you reprimand \nviolators inside the organization? How will you deal with violators outside? What type of civil or \ncriminal action might you take?\n \n \n \n \n \n \nThe security policy should be integrated with your company's other policies and plans. For example, \nthink about the internal control structure. Plan for contingencies. Make sure all your plans comply \nwith the laws.\n \n \n \n \n \n \nThe role of the information systems department should be specified in the policy document. The \ndepartment should be responsible for, among other things, ensuring that security personnel are \nadequately trained and properly qualified. IS security personnel should be able to assist other \ndepartments with their security needs. Other responsibilities of security personnel might include:\n \n \n \n \n \n • Assisting in acquiring hardware or operating systems \n \n \n \n \n • Managing security of the communications networks \n \n \n \n \n • Establishing standards for remote access \n \n \n \n \n • Installing and maintaining virus detection software \n \n \n \n \n • Selecting cryptographic techniques and keys \n \n \n \n \n • Backing up critical data \n \n \n \n \n • Evaluating and approving IS-related contracts \n"
        },
        {
            "page_number": 191,
            "text": " \n \n \n \n The Security Policy  \n \n \n \n \n \nThe security policy should be a broad statement that guides personnel and departments in achieving \ncertain goals. It should be concise and easy to read. It should not specify actions. The purpose of the \nsecurity policy is not to educate or train individuals. That should be provided for in manuals and \nseminars.\nThe security policy, then, is written at a broad organizational level. The standards, guidelines, and \nprocedures go into supporting documents. Some questions the policy should answer are:\n \n \n \n \n \n • Why is it important to have a security policy? \n \n \n \n \n • What is the meaning of data integrity? \n \n \n \n \n • Why must data integrity be maintained? \n \n \n \n \n • Why should data be kept confidential? \n \n \n \n \n • What are the consequences to the organization if data is unavailable or compromised? \n \n \n \n \n Standards, Guidelines, and Procedures  \n \n \n \n \n \nStandards, guidelines, and procedures provide the guidance the members of your organization need to \nrealize the goals defined in the policy document. They give people clear instructions on how to meet \norganizational goals. Standards and guidelines specify the technologies and methodologies that may \nbe used; procedures offer more detailed guidance to achieve particular security objectives.\n \n \n \n \n \n \nAll three should be published in handbooks, regulations, or manuals, both physical and electronic, as \non the corporate Intranet or on CD-ROM. Providing the information in an electronic format has \nseveral advantages, among them easy access. It also makes it easier for you to keep the information \ncurrent.\n \n \n"
        },
        {
            "page_number": 192,
            "text": " \n \n \n The purpose of standards is to specify a uniform set of technologies or procedures. Standards are \ntypically mandatory; users may not exercise their own discretion in the areas they cover.\n \n \n \n \n \n \nGuidelines, on the other hand, are provided because it's not always possible, appropriate, or cost-\neffective to impose standards. Guidelines give users some latitude in meeting goals. Guidelines are \nused to ensure, for example, that specific security measures are not overlooked. Guidelines inherently \nrecognize that security measures may be correctly implemented in more than one way.\n \n \n \n \n \n Procedures offer step-by-step guidance in adhering to standards and guidelines.\nBuy-in  \n \n \n \n \n \nIf you want your security policy to be effective, solicit participation from across the organization, \nboth individuals and departments. Senior management's support is essential. You must provide the \nresources, financial and otherwise, to implement the policy adequately.\n \n \n \n \n \n \nAt the same time, users are far more likely to accept security policy, guidelines, and procedures if \nthey had input in creating them. Their participation will help you create a better policy plan. The \nspecialized knowledge each brings will result in a superior document.\n \n \n \n \n \n \nSet up a senior management committee (the Information Security Management Committee) with \nauthority to issue and amend the security policy. Make sure that people get from committee approval \nfor any exceptions to the security policy.\n \n \n \n \n \n Scope  \n \n \n \n \n \nYour computer security policy should apply to all facilities and locations of the company. The same \nset of standards should be enforced throughout the company. While it's essential that security \nstandards be applied consistently, they should be flexible enough to be used in a variety of situations.\n \n \n \n \n \n \nYour security policies should encompass all types of computer systems, including stand-alone PCs, \nLANs, WANs, the Internet, and the Intranet. They should cover all types of data transmission, \nincluding e-mail, FTP, and fax.\n \n \n \n \n \n Risk Analysis and Management \n"
        },
        {
            "page_number": 193,
            "text": " \n \n \n \n \nComputer security planning is an integral part of your organization's overall risk management \nstrategy. Each individual's and each department's responsibility should be clearly identified in the \nsecurity plan, which should be updated regularly to meet changes in technology or circumstances.\n \n \n \n \n \n You must therefore do a thorough risk analysis. Computer security risks fall into five major \ncategories:\n \n \n \n \n \n 1. Destruction of data or equipment \n \n \n \n \n 2. Theft of data or equipment \n \n \n \n \n 3. Malfunction of equipment or bugs in software \n \n4. Modification of data \n \n \n \n \n 5. Disclosure of data \n \n \n \n \n The cause of risk may be: \n \n \n \n \n • Intentional attack \n \n \n \n \n • Unintentional or accidental loss \n \n \n \n \n • Environmental threat \n \n \n \n \n \nIntentional threat comes from computer criminals and disgruntled employees setting out to defraud, \nsabotage, alter data, or steal equipment or data. Unintentional loss may result from computer users \nwho are careless. Accidental loss may be due to equipment malfunction. Environmental threats \ninclude fires, floods, earthquakes, lightning, and power outages. An effective security plan must \ncover all these threats.\n \n"
        },
        {
            "page_number": 194,
            "text": " \n \n \n \n \nTake out insurance policies to cover such risks as theft, fraud, intentional destruction, and forgery. \nDon't forget business interruption insurance, which covers lost profits and additional expenses during \ndowntime.\n \n \n \n \n \n \nYour risk analysis should take into account not only the security but also the reliability of your \nsystem; this can be compromised by errors, failures, and faults. An error is a deviation from \nexpectations. Some errors are acceptable because they can be overcome; others are simply \nunacceptable. An unacceptable error is a failure. If the failure can have serious consequences, it's \nconsidered a critical failure. A fault is a condition that results in a failure.\n \n \n \n \n \n \nIt's important to note, however, that system reliability is conceptually distinct from system security. \nThe purpose of computer system security is to protect against intentional misuse. System security \ndoesn't really consider malfunctions or bugs unless they will allow a perpetrator to breach security. \nStill, improving one factor will typically enhance the effects of the other. That's why both security \nand reliability should be considered in managing risk.\n \n \n \n \n \n The Security Administrator \n \n \n \n \n \nThe security administrator is responsible for customizing security policies and standards to the \norganization and for the planning, execution, and maintenance of the computer security system. The \nadministrator should regularly interact with other departments to learn of their changing needs. Both \ntechnical computer knowledge and management skills are therefore necessary attributes for the \nsecurity administrator, as well as a thorough understanding of the organization's internal control \nstructure.\nSpecific aspects of an organization, such as its size and inherent risks, have to be taken into account in \nsetting up the security administration department. The department must ensure the information systems \ndata is reliable and accurate. Members of the department should keep abreast of organizational \nrequirements in dynamic environments to keep the security system efficient while monitoring staff to \nensure compliance with policies. It is particularly important for you to have in place specific \nprocedures for hiring and recruiting staff for this department (see Chapter 5).\n \n \n \n \n \n The Human Factor  \n \n \n \n \n \nTypically, security systems depend more on people and their attitudes toward security than on the \nlatest technology. Typically, too, the greatest security threat in an organization comes not from \noutsiders but from insiders. Personnel incompetence, indifference, and negligence are likely to cause \nmore harm than sabotage or intentional acts by unauthorized hackers.\n \n"
        },
        {
            "page_number": 195,
            "text": "  \n \n \n \n \n \nSecurity policies and procedures often conflict with people's ideas of good manners. Trusting others \nand sharing things with them is viewed positively. Security policies, on the other hand, require that \nyou distrust others and not share information. For example, it's normally considered polite to hold a \ndoor open for someone behind you. However, in a restricted area, such politeness will result in a \nsecurity breach. Similarly, sharing user IDs or passwords may allow an unauthorized individual to \naccess sensitive information.\n \n \n \n \n \n Psychological Factors1 \n \n \n \n \n To make sure the people in your company accept the security policy, M.E. Kabay, director of \neducation for the National Computer Security Association, recommends:\n \n \n \n \n \n • Before attempting to implement policies and procedures, build up a consistent view of information \nsecurity among your colleagues.\n \n \n \n \n \n • Introduce security policies over time, don't rush them into place. \n \n \n \n1M.E. Kabay, ''Psycho-Social Factors in the Implementation of Information Security Policy,\" National Computer \nSecurity Association (see http://nsi.org/).\n \n \n• Present case studies to help get people ready to accept security requirements. \n \n \n \n \n • Give your people many realistic examples of security requirements and breaches. \n \n \n \n \n • Inspire a commitment to security rather than merely describing it. \n \n \n \n \n • Emphasize improvement effects rather than failure reduction. \n \n \n \n \n • Explore the current beliefs of employees and managers. \n \n"
        },
        {
            "page_number": 196,
            "text": " \n \n \n • Don't portray computer crime with any positive images or words. \n \n \n \n \n • Praise comments that are critical of computer crime or that support the security policies. \n \n \n \n \n • Challenge rather than ignore employees who dismiss security concerns or flout the regulations. \n \n \n \n \n • Identify the senior executives most likely to set a positive tone for security training. \n \n \n \n \n • Immediately couple frightening consequences with effective and achievable security measures. \n \n \n \n \n • Present objections to a proposal and offer counter-arguments rather than giving a one-sided diatribe. \n \n \n \n \n • Make sure repeated novel reminders of security issues are part of your security awareness program. \n \n \n \n \n • Include small gifts in your security awareness program. \n \n \n \n \n • Find a charismatic leader to help generate enthusiasm for better security. \n \n \n \n \n • Encourage specific employees to take on public responsibility for information security within their \nwork groups.\n \n \n \n \n \n • Rotate the security role periodically. \n \n \n \n \n • Incorporate into your security training information on how to tell when someone may be engaging \nin computer crime.\n \n \n \n \n \n • Build a corporate culture that rewards responsible behavior such as reporting security violations. \n \n"
        },
        {
            "page_number": 197,
            "text": " \n \n \n • Develop clearly written security policies and procedures. \n \n \n \n \n • Be sure your security procedures make it easy to act in accordance with security policy.\n• Emphasize the seriousness of failing to act in accordance with security policies and procedures. \n \n \n \n \n • Enforce standards of security so that employees will later follow the standards more rigorously. \n \n \n \n \n • Create a working environment in which employees are respected; this is more conducive to good \nsecurity than one that devalues and abuses them.\n \n \n \n \n \n • Have security supervisors get to know the staff. \n \n \n \n \n • Encourage social activities in the office. \n \n \n \n \n • Pay special attention to social \"outliers\" during training programs. \n \n \n \n \n • Monitor compliance closely to security requirements. \n \n \n \n \n • Work with the outliers to resist the herd's anti-security bias. \n \n \n \n \n • Before discussing security at a meeting, have one-on-one discussions with participants. \n \n \n \n \n • Remain impartial; encourage open debate in security meetings. \n \n \n \n \n • Bring in experts from the outside when faced with groupthink. \n \n \n \n \n • Meet again after a consensus has been built and play devil's advocate. \n \n"
        },
        {
            "page_number": 198,
            "text": " \n \n \n Seminars  \n \n \n \n \n \nThe importance of computer security must be instilled in all employees. Direct communication is \ntypically more effective at persuading individuals than mass media like videos or books. Personalized \nmessages stimulate thought and are likely to be more persuasive. An excellent way to both \nindoctrinate new employees and update the skills of current employees is to periodically schedule \nsecurity seminars. Security professionals can communicate your company's rules and procedures at \nthese seminars, as well as answer questions and address the security needs of employees.\n \n \n \n \n \n \nAssign senior executives who are liked and respected to lead the seminars. A speaker's attractiveness \nand social status have an immediate effect on the audience. For a few days, the speaker's personal \ncharacteristics will continue to influence the audience, though the effect typically declines with the \npassage of time till only the message remains with the audience. The speaker should understand \nsecurity issues and honestly believe in the policies he or she is advocating.\n \nIn trying to persuade employees, it's useful to present a balanced view, especially when trying to \nconvince those who initially disagree with your policy. By presenting both sides of the argument you \nshow your audience that you understand their perspective and have sound reasons for your own.\n \n \n \n \n \n Other Communication Channels  \n \n \n \n \n \nLecturing employees for a few hours per year, however, is unlikely to lead to improved security by \nitself. People need time to accept and acclimate to change; that's why new security policies should \ntypically be phased in over time.\n \n \n \n \n \n \nRepetition of the security message is helpful in building support for security policies. Security \nawareness may be enhanced with mugs, posters, and newsletters. Ongoing activities will yield better \nresults than occasional training seminars alone, necessary though these are.\n \n \n \n \n \n \nVideos and case studies containing examples of security breaches can have a beneficial effect. \nExposing individuals to different security scenarios helps increase their awareness of security issues. \nHumans are not good intuitive statisticians. Intuitive human judgment is often prone to bias. \nJudgment is easily distorted when individuals tend to rely on small samples, easily available data, \nand personal anecdotal experience.\n \n \n \n \n \n \nThe use of questionnaires, focus groups, and interviews, may serve several purposes. They can help \nyou obtain useful information about employees. They may also help to modify employee's beliefs, \nleading to a greater commitment to security. If you publicly support enhanced security, employees \n \n"
        },
        {
            "page_number": 199,
            "text": "will change their perceptions and be more committed to security. When they take responsibility \nopenly, their commitment to the task increases.\n \n \n \n \n Behavior Modification  \n \n \n \n \n \nYou can modify employee behavior by rewarding employees who support and punishing employees \nwho violate security policies. The reward may be as simple as verbal praise. The punishment may be \na simple verbal warning, disciplinary action, or even removal from the job. Employees who \nflagrantly violate security policies should certainly hear about it.\n \n \n \n \n \n \nThe use of fear to change the attitudes of employees works only in certain situations. Too much fear \nabout catastrophic consequences is likely to result in employees rejecting the message. When you tell \nemployees about catastrophic consequences, also show them how to counter threats.\nAccount Administration \n \n \n \n \n \nNew users are continually being added to your system while old users must be deleted. Establish a \nwritten procedure for requesting, creating, maintaining, and closing user accounts. The purpose of \naccount administration is to ensure that:\n \n \n \n \n \n • The user is authorized. \n \n \n \n \n • The user has access privileges appropriate to the job. \n \n \n \n \n • The user is not engaged in unauthorized activities. \n \n \n \n \n • Information about the user is current. \n \n \n \n \n \nThe user's supervisor should initiate the account creation process by requesting an account from the \nsystems manager. The request should specify access level and the applications to which the user \nshould have access. The approval of an applications manager may be required before the systems \nmanager grants access to a particular application.\n \n \n \nThe user access level should be part of the account profile. Specific applications may have built-in \n"
        },
        {
            "page_number": 200,
            "text": "  access controls or may rely on third-party software for access control. The systems manager must \nensure that a user's access is consistent with the request from the supervisor.\n  \n \n \n \n \n \nOn being issued the user account and password, employees should undergo security training. At the \nvery least, users should be provided with the written rules and guidelines and be required to sign an \n\"account assignment\" document indicating their understanding of those rules and guidelines. The \ndocument may be used to discipline or even prosecute users who violate them.\n \n \n \n \n \n \nThere are two techniques for creating user account IDs. The ID could be for a specific job title \n(SALESREP4) or for a specific employee (JACK_BLACK). From an auditing perspective, job title \nIDs simplify the process. However, if the account ID is for a specific job title, you'll need controls to \nensure that the password is changed as soon as the employee changes jobs or leaves.\n \n \n \n \n \n \nThe user's supervisor should notify the systems and applications managers when a user is reassigned \nor the account is no longer required. The personnel department should also be required to notify \nthose managers when there's a change in personnel or duties.\n \n \n \n \n \n \nAccess level privileges change. The change might be temporary or permanent. An employee may be \ntemporarily performing the duties of another employee who is sick or on vacation. An employee may \nalso be permanently assigned a different function or transferred to another department.\nWhen an employee takes on additional duties during the absence of another employee, take care to \nensure that he or she is not performing incompatible duties. From a control perspective, no one should \nbe in a position to perpetrate an irregularity and cover it up in the normal course of the day. Temporary \naccess privileges should also be removed as soon as they are no longer needed.\n \n \n \n \n \n \nUser accounts should be reviewed regularly to help detect unauthorized or illegal activities. The \nreview may be of a sample of user accounts or the entire system. The level of access of each user \nshould also be reviewed. Make sure that all new accounts have supervisory approval, and that the \nlevel of access granted is warranted by the job responsibilities.\n \n \n \n \n \n Check that the accounts of all personnel who left the organization or were reassigned were properly \nclosed, comparing data from the personnel department and the system manager's records.\n \n \n \n \n \n Examine account records to assure that all users signed a statement acknowledging their \nunderstanding of rules and guidelines and that they have taken security awareness training.\n \n \n \nFor certain functions, periodic screening checks of personnel may be warranted. For example, an \n"
        },
        {
            "page_number": 201,
            "text": "  \nindividual living a lifestyle considerably in excess of income is a red flag; it may be the fruit of \nfraudulent activity. The individual may be stealing corporate assets or giving competitors proprietary \ninformation.\n  \n \n \n \n \n Review the controls over account management during an audit and check compliance with controls. \nJust because a control exists doesn't mean users are actually following it.\n \n \n \n \n \n \nOne important control we have already mentioned is segregation of duties so that the employee \ncannot commit and conceal an illegal activity in the normal course of duties. Mandating that \nemployees take vacations is also important from a security perspective. Some fraudulent activities \nrequire the perpetrators to take certain actions on a regular basis to prevent the fraud from being \ndiscovered.\n \n \n \n \n \n \nIn addition to mandatory vacations, you may want to rotate job assignments periodically. This serves \nthe same purpose: It prevents a perpetrator from covering up illegal activities. Rotating job \nassignments has other benefits: When several individuals are trained to perform a single function, the \norganization is not excessively dependent on any single employee.\nConclusion \n \n \n \n \n \nThe primary benefit of the computer security policy is to prevent or minimize the loss of assets or \nresources due to a security breach. The document also provides a decision-making framework for \npurchasing software and hardware. It gives guidelines for steps to take after a security breach to \nprevent further breaches or losses.\n \n \n \n \n \n A formal risk assessment should be part of the document. Assets to be protected, threats to those \nassets, and safeguards for those assets should be analyzed.\n \n \n \n \n \n \nThe rights and obligations of the users should be specified, along with rules for account use. \nConditions that apply to personal users as well as public access accounts should be detailed. The \ndocument should also include criteria of acceptability of user software along with data access and use \npolicies.\n \n \n \n \n \n \nUsers should be specifically warned against disclosing their passwords or other information that \ncould potentially threaten the computer system. Privacy policies, including those applying to \ndisclosure of confidential information to third parties, should be prominent, with special attention to \nuser privacy, including conditions when the company may access a user's files.\n \n \n \n"
        },
        {
            "page_number": 202,
            "text": "  The security policy should state what will happen if there is a security breach. It should answer \nquestions like:\n  \n \n \n \n \n • If a security breach is in progress, who should be notified, and how? \n \n \n \n \n • What audit trail must be maintained? What log files should be kept? \n \n \n \n \n • How will the affected computer system or network be identified and isolated? \n \n \n \n \n • What are the legal ramifications of entrapment? When must security officers identify themselves? \n \n \n \n \n • How will violators be punished? \n \n \n \n \n • When will law enforcement authorities be notified? \n \n \n \n \n • How will the organization recover from a security breach? \n \nAppendix 7.A— \nSources of Information Security Policies \n \n \n \n \n \nYou can find prewritten information security policies at the Web site for Baseline Software, Inc. \n(http://www.baselinesoft.com/). You may want to use Baseline as a starting point, customizing as \nneeded.\n \n \n \n \n \n \nInformation Security Policies Made Easy Version 7 (7th edition) by Charles Cresson Wood, October \n31, 1999, sets out some 840 policies, each with an explanation. The book comes with a CD-ROM \ncontaining files that may be edited in various word processing programs. It may be accessed at \nwww.baselinesoft.com. Topics covered include:\n \n \n \n \n \n • Computer emergency response teams \n \n \n \n \n"
        },
        {
            "page_number": 203,
            "text": " • Computer viruses \n \n \n \n \n • Contingency planning \n \n \n \n \n • Data classification \n \n \n \n \n • Digital signatures \n \n \n \n \n • Electronic commerce \n \n \n \n \n • Electronic mail \n \n \n \n \n • Employee surveillance \n \n \n \n \n • Encryption \n \n \n \n \n • Firewalls \n \n \n \n \n • Internet \n \n \n \n \n • Intranets \n \n \n \n \n • Local area networks \n \n \n \n \n • Logging controls \n \n \n \n \n • Microcomputers \n \n"
        },
        {
            "page_number": 204,
            "text": " \n \n \n • Outsourcing security functions \n \n \n \n \n • Password selection \n \n \n \n \n • Portable computers \n \n \n \n \n • Privacy issues \n \n \n \n \n • Telecommuting \n \n \n \n \n • Telephone systems \n \n \n \n \n • User training \n \n \n \n \n • Web pages \n \nAppendix 7.B: \nSample Computer Policy \n \n \n \n \n Georgia Institute of Technology's Computer and Network Usage Policy (Available at \nwww.business.gatech.edu/depts/andits/internal.htm)\n \n \n \n \n \n Preface \n \n \n \n \n \nRespect for intellectual labor and creativity is vital to academic discourse and enterprise. This \nprinciple applies to works of all authors and publishers in all media. It encompasses respect for the \nright to acknowledgment, the right to privacy, and the right to determine the form, manner, and terms \nof publication and distribution. Because electronic information is volatile and easily reproduced, \nrespect for the work and personal expression of others is especially critical in computer \nenvironments. Violations of authorial integrity, including plagiarism, invasion of privacy, \nunauthorized access, and trade secret and copyright violations, may be grounds for sanctions against \n \n"
        },
        {
            "page_number": 205,
            "text": "members of the academic community. The EDUCOM Code.\n \n \n \n \n 1— \nBackground and Purpose  \n \n \n \n \n \nThis document constitutes an Institute-wide policy intended to allow for the proper use of all Georgia \nTech computing and network resources, effective protection of individual users, equitable access, and \nproper management of those resources. This should be taken in the broadest possible sense. This \npolicy applies to Georgia Tech network usage even in situations where it would not apply to the \ncomputer(s) in use. These guidelines are intended to supplement, not replace, all existing laws, \nregulations, agreements, and contracts which currently apply to these services.\n \n \n \n \n \n \nCampus units that operate their own computers or networks may add, with the approval of the unit \nhead, individual guidelines which supplement, but do not relax, this policy. In such cases, the unit \nshould inform their users and the Information Resources Security Coordinator in OIT prior to \nimplementation.\n \n \n \n \n \n \nAccess to networks and computer systems owned or operated by Georgia Tech imposes certain \nresponsibilities and obligations and is granted subject to Institute policies and local, state, and federal \nlaws. Appropriate use should always be legal, ethical, reflect academic honesty, reflect community \nstandards, and show restraint in the consumption of shared resources. It should demonstrate respect \nfor intellectual property; ownership of data; system security mechanisms; and individuals rights to \nprivacy and to freedom from intimidation, harassment, and unwarranted annoyance. Appropriate use \nof computing and networking resources includes instruction; independent study; authorized research; \nindependent research; communications; and official work of the offices, units, recognized student and \ncampus organizations, and agencies of the Institute.\n \n \n2— \nDefinitions  \n \n \n \n \n 2.1— \nAuthorized Use \n \n \n \n \n \nAuthorized use of Georgia Tech-owned or operated computing and network resources is use \nconsistent with the education, research, and service mission of the Institute, and consistent with this \npolicy.\n \n \n \n \n \n 2.2— \n \n"
        },
        {
            "page_number": 206,
            "text": "Authorized Users\n \n \n \n \n \nAuthorized users are: (1) current faculty, staff, and students of the Institute; (2) anyone connecting to \na public information service (see section 6.5); (3) others whose access furthers the mission of the \nInstitute and whose usage does not interfere with other users' access to resources. The policy Access \nby External Entities to Institute Information Technology Resources (OIT, 11/3/93, and any \nsubsequent revisions) may apply. In addition, a user must be specifically authorized to use a \nparticular computing or network resource by the campus unit responsible for operating the resource.\n \n \n \n \n \n 3— \nIndividual Privileges  \n \n \n \n \n \nIt is the following individual privileges, all of which are currently existent at Georgia Tech, that \nempower each of us to be productive members of the campus community. It must be understood that \nprivileges are conditioned upon acceptance of the accompanying responsibilities.\n \n \n \n \n \n 3.1— \nPrivacy \n \n \n \n \n \nTo the greatest extent possible in a public setting we want to preserve the individual's privacy. \nElectronic and other technological methods must not be used to infringe upon privacy. However, \nusers must recognize that Georgia Tech computer systems and networks are public and subject to the \nGeorgia Open Records Act. Users, thus, utilize such systems at their own risk.\n \n \n \n \n \n 3.2— \nFreedom of Expression \n \n \n \n \n The constitutional right to freedom of speech applies to all members of the campus no matter the \nmedium used.\n \n \n \n \n \n 3.3— \nOwnership of Intellectual Works \n \n \n \n \n People creating intellectual works using Georgia Tech computers or networks, including but not \nlimited to software, should consult Determination of Rights and Equities in Intellectual Property \n \n"
        },
        {
            "page_number": 207,
            "text": "(Board of Regents Policy Manual, section 603.03, 2/2/94 and any subsequent revisions), and related \nGeorgia Tech policies.\n \n \n \n \n 3.4— \nFreedom from Harassment and Undesired Information\nAll members of the campus have the right not to be harassed by computer or network usage by others. \n(See 4.1.3.)\n \n \n \n \n \n 4— \nIndividual Responsibilities  \n \n \n \n \n \nJust as certain privileges are given to each member of the campus community, each of us is held \naccountable for our actions as a condition of continued membership in the community. The interplay \nof privileges and responsibilities within each individual situation and across campus engenders the \ntrust and intellectual freedom that form the heart of our community. This trust and freedom are \ngrounded on each person's developing the skills necessary to be an active and contributing member \nof the community. These skills include an awareness and knowledge about information and the \ntechnology used to process, store, and transmit it.\n \n \n \n \n \n 4.1— \nCommon Courtesy and Respect for Rights of Others \n \n \n \n \n \nYou are responsible to all other members of the campus community in many ways, including to \nrespect and value the rights of privacy for all, to recognize and respect the diversity of the population \nand opinion in the community, to behave ethically, and to comply with all legal restrictions regarding \nthe use of information that is the property of others.\n \n \n \n \n \n 4.1.1— \nPrivacy of Information  \n \n \n \n \n \nFiles of personal information, including programs, no matter on what medium they are stored or \ntransmitted, may be subject to the Georgia Open Records Act if stored on Georgia Tech's computers. \nThat fact notwithstanding, no one should look at, copy, alter, or destroy anyone elses personal files \nwithout explicit permission (unless authorized or required to do so by law or regulation). Simply \nbeing able to access a file or other information does not imply permission to do so.\n \n \n \nSimilarly, no one should connect to a host on the network without advance permission in some form. \n"
        },
        {
            "page_number": 208,
            "text": "  People and organizations link computers to the network for numerous different reasons, and many \nconsider unwelcome connects to be attempts to invade their privacy or compromise their security.\n  \n \n \n \n \n 4.1.2— \nIntellectual Property  \n \n \n \n \n You are responsible for recognizing (attributing) and honoring the intellectual property rights of \nothers.\n \n \n \n \n \n 4.1.3— \nHarassment  \n \n \n \n \n No member of the community may, under any circumstances, use Georgia Tech's computers or \nnetworks to libel, slander, or harass any other person.\n \n \n \n \n \n \nThe following shall constitute Computer Harassment: (1) Intentionally using the computer to annoy, \nharass, terrify, intimidate, threaten, offend or bother another person by conveying obscene language, \npictures, or other materials or threats of bodily harm to the recipient or the recipient's immediate \nfamily; (2) Intentionally using the computer to contact another person repeatedly with the intent to \nannoy, harass, or bother, whether or not any actual message is communicated, and/or where no \npurpose of legitimate communication exists, and where the recipient has expressed a desire for the \ncommunication to cease; (3) Intentionally using the computer to contact another person repeatedly \nregarding a matter for which one does not have a legal right to communicate, once the recipient has \nprovided reasonable notice that he or she desires such communication to cease (such as debt \ncollection); (4) Intentionally using the computer to disrupt or damage the academic, research, \nadministrative, or related pursuits of another; (5) Intentionally using the computer to invade the \nprivacy, academic or otherwise, of another or the threatened invasion of the privacy of another.\n4.2— \nResponsible Use of Resources \n \n \n \n \n \nYou are responsible for knowing what information resources (including networks) are available, \nremembering that the members of the community share them, and refraining from all acts that waste \nor prevent others from using these resources or from using them in whatever ways have been \nproscribed by the Institute and the laws of the State and Federal governments. Details regarding \navailable resources are available in many ways, including consulting your Computing Support \nRepresentative (CSR) (see section 6.4), conferring with other users, examining on-line and printed \nreferences maintained by OIT and others, visiting the OIT Information Center, and contacting the \nOIT Helpdesk.\n \n \n"
        },
        {
            "page_number": 209,
            "text": " \n \n \n 4.3— \nGame Playing \n \n \n \n \n \nLimited recreational game playing, that is not part of an authorized and assigned research or \ninstructional activity, is tolerated (within the parameters of each department's rules). Institute \ncomputing and network services are not to be used for extensive or competitive recreational game \nplaying. Recreational game players occupying a seat in a public computing facility must give up that \nseat when others who need to use the facility for academic or research purposes are waiting.\n \n \n \n \n \n 4.4— \nInformation Integrity \n \n \n \n \n \n \nIt is your responsibility to be aware of the potential for and possible effects of manipulating \ninformation, especially in electronic form, to understand the changeable nature of electronically stored \ninformation, and to verify the integrity and completeness of information that you compile or use. Do \nnot depend on information or communications to be correct when they appear contrary to your \nexpectations; verify it with the person who you believe originated the message or data.\n4.5— \nUse of Desktop Systems \n \n \n \n \n \nYou are responsible in coordination with your CSR for the security and integrity of Institute \ninformation stored on your personal desktop system. This responsibility includes making regular disk \nbackups, controlling physical and network access to the machine, and installing and using virus \nprotection software. Avoid storing passwords or other information that can be used to gain access to \nother campus computing resources.\n \n \n \n \n \n 4.6— \nAccess to Facilities and Information \n \n \n \n \n 4.6.1— \nSharing of Access  \n \n \n \n \n Computer accounts, passwords, and other types of authorization are assigned to individual users and \nmust not be shared with others. You are responsible for any use of your account.\n \n \n \n"
        },
        {
            "page_number": 210,
            "text": "  4.6.2— \nPermitting Unauthorized Access   \n \n \n \n \n You may not run or otherwise configure software or hardware to intentionally allow access by \nunauthorized users. (See section 2.2.)\n \n \n \n \n \n 4.6.3— \nUse of Privileged Access  \n \n \n \n \n Special access to information or other special computing privileges are to be used in performance of \nofficial duties only. Information that you obtain through special privileges is to be treated as private.  \n \n \n \n \n 4.6.4— \nTermination of Access  \n \n \n \n \n \nWhen you cease being a member of the campus community (graduate or terminate employment), or \nif you are assigned a new position and/or responsibilities within the Institute, your access \nauthorization must be reviewed. You must not use facilities, accounts, access codes, privileges, or \ninformation for which you are not authorized in your new circumstances.\n \n \n \n \n \n 4.7— \nAttempts to Circumvent Security \n \n \n \n \n Users are prohibited from attempting to circumvent or subvert any system's security measures. This \nsection does not prohibit use of security tools by system administration personnel.\n \n \n \n \n \n 4.7.1— \nDecoding Access Control Information  \n \n \n \n \n You are prohibited from using any computer program or device to intercept or decode passwords or \nsimilar access control information.\n \n \n \n \n \n 4.7.2— \n \n"
        },
        {
            "page_number": 211,
            "text": "Denial of Service\n \n \n \n \n \nDeliberate attempts to degrade the performance of a computer system or network or to deprive \nauthorized personnel of resources or access to any Institute computer system or network are \nprohibited.\n \n \n \n \n \n 4.7.3— \nHarmful Activities  \n \nThe following harmful activities are prohibited: creating or propagating viruses; disrupting services; \ndamaging files; intentional destruction of or damage to equipment, software, or data belonging to \nGeorgia Tech or other users; and the like.\n \n \n \n \n \n 4.7.4— \nUnauthorized Access  \n \n \n \n \n \nYou may not: \ndamage computer systems \nobtain extra resources not authorized to you \ndeprive another user of authorized resources \ngain unauthorized access to systems \nby using knowledge of: \na special password \nloopholes in computer security systems \nanother user's password \naccess abilities you used during a previous position at the Institute\n \n \n \n \n \n 4.7.5— \nUnauthorized Monitoring  \n \n \n \n \n You may not use computing resources for unauthorized monitoring of electronic communications. \n \n \n \n \n 4.8— \nAcademic Dishonesty \n \n \n \n \n You should always use computing resources in accordance with the high ethical standards of the \nInstitute community. Academic dishonesty (plagiarism, cheating) is a violation of those standards.\n \n"
        },
        {
            "page_number": 212,
            "text": " \n \n \n \n 4.9— \nUse of Copyrighted Information and Materials \n \n \n \n \n You are prohibited from using, inspecting, copying, and storing copyrighted computer programs and \nother material, in violation of copyright.\n \n \n \n \n \n 4.10— \nUse of Licensed Software \n \n \n \n \n \nNo software may be installed, copied, or used on Institute resources except as permitted by the owner \nof the software. Software subject to licensing must be properly licensed and all license provisions \n(installation, use, copying, number of simultaneous users, term of license, etc.) must be strictly \nadhered to.\n \n \n \n \n \n 4.11— \nPolitical Campaigning; Commercial Advertising \n \n \n \n \n \nBoard of Regents policy (section 914.01) states ''The use of System materials, supplies, equipment, \nmachinery, or vehicles in political campaigns is forbidden.\" The Georgia Tech Faculty Handbook \n(section 6.15.3.8(b)) states \"Political campaign and commercial advertisement shall not be displayed \non the campus.\" The use of Institute computers and networks shall conform to these policies.\n \n \n \n \n \n 4.12— \nPersonal Business \n \n \n \n \n \nComputing facilities, services, and networks may not be used in connection with compensated outside \nwork nor for the benefit of organizations not related to Georgia Tech, except: in connection with \nscholarly pursuits (such as faculty publishing activities); in accordance with the Institute Consulting \nPolicy or the policy Access by External Entities to Institute Information Technology Resources (OIT, \n11/3/93, and any subsequent revisions); or in a purely incidental way. This and any other incidental \nuse (such as electronic communications or storing data on singleuser machines) must not interfere \nwith other users' access to resources (computer cycles, network bandwidth, disk space, printers, etc.) \nand must not be excessive. State law restricts the use of State facilities for personal gain or benefit.\n5— \nGeorgia Tech Privileges  \n \n"
        },
        {
            "page_number": 213,
            "text": " \n \n \n \nOur society depends on institutions like Georgia Tech to educate our citizens and advance the \ndevelopment of knowledge. However, in order to survive, Georgia Tech must attract and responsibly \nmanage financial and human resources. Therefore, Tech has been granted by the State, and the \nvarious other institutions with which it deals, certain privileges regarding the information necessary \nto accomplish its goals and to the equipment and physical assets used in its mission.\n \n \n \n \n \n 5.1— \nAllocation of Resources \n \n \n \n \n Georgia Tech may allocate resources in differential ways in order to achieve its overall mission. \n \n \n \n \n 5.2— \nControl of Access to Information \n \n \n \n \n \nGeorgia Tech may control access to its information and the devices on which it is stored, \nmanipulated, and transmitted, in accordance with the laws of Georgia and the United States and the \npolicies of the Institute and the Board of Regents.\n \n \n \n \n \n 5.3— \nImposition of Sanctions \n \n \n \n \n Georgia Tech may impose sanctions and punishments on anyone who violates the policies of the \nInstitute regarding computer and network usage.\n \n \n \n \n \n 5.4— \nSystem Administration Access \n \n \n \n \n \nA System Administrator (i.e., the person responsible for the technical operations of a particular \nmachine) may access others files for the maintenance of networks and computer and storage systems, \nsuch as to create backup copies of media. However, in all cases, all individuals' privileges and rights \nof privacy are to be preserved to the greatest extent possible.\n \n \n \n \n \n 5.5— \n \n"
        },
        {
            "page_number": 214,
            "text": "Monitoring of Usage, Inspection of Files\n \n \n \n \n \nUnits of Georgia Tech operating computers and networks may routinely monitor and log usage data, \nsuch as network session connection times and end-points, CPU and disk utilization for each user, \nsecurity audit trails, network loading, etc. These units may review this data for evidence of violation \nof law or policy, and other purposes.\nWhen necessary, these units may monitor all the activities of and inspect the files of specific users on \ntheir computers and networks. Any person who believes such monitoring or inspecting is necessary \nmust obtain the concurrence of the unit head and the campus Legal Division. In all cases all \nindividuals' privileges and right of privacy are to be preserved to the greatest extent possible.\n \n \n \n \n \n 5.6— \nSuspension of Individual Privileges \n \n \n \n \n \nUnits of Georgia Tech operating computers and networks may suspend computer and network \nprivileges of an individual for reasons relating to his/her physical or emotional safety and well being, \nor for reasons relating to the safety and well-being of other members of the campus community, or \nInstitute property. Access will be promptly restored when safety and well-being can be reasonably \nassured, unless access is to remain suspended as a result of formal disciplinary action imposed by the \nOffice of the Vice President for Student Services (for students) or the employee's department in \nconsultation with the Office of Human Resources (for employees).\n \n \n \n \n \n 6— \nGeorgia Tech Responsibilities  \n \n \n \n \n 6.1— \nSecurity Procedures \n \n \n \n \n \nGeorgia Tech has the responsibility to develop, implement, maintain, and enforce appropriate \nsecurity procedures to ensure the integrity of individual and institutional information, however \nstored, and to impose appropriate penalties when privacy is purposefully abridged.\n \n \n \n \n \n 6.2— \nAnti-harassment Procedures \n \n \n \n \n Georgia Tech has the responsibility to develop, implement, maintain, and enforce appropriate \nprocedures to discourage harassment by use of its computers or networks and to impose appropriate  \n"
        },
        {
            "page_number": 215,
            "text": "penalties when such harassment takes place.\n \n \n \n \n 6.3— \nUpholding of Copyrights and License Provisions \n \n \n \n \n \nGeorgia Tech has the responsibility to uphold all copyrights, laws governing access and use of \ninformation, and rules of organizations supplying information resources to members of the \ncommunity (e.g., acceptable use policies for use of Internet).\n \n \n \n \n \n 6.4— \nIndividual Unit Responsibilities \n \n \n \n \n \nEach unit has the responsibility of: \nenforcing this policy \nproviding for security in their areas \nproviding individuals equipped with Institute-owned desktop systems with resources for regular disk \nbackups (software, hardware, media, and training) and for virus protection\n \n \nIf warranted by the importance and sensitivity of information stored and processed in their facility, a \nunit must also:\n \n \n \n \n \n provide system administration personnel \n \n \n \n \n perform and verify integrity of regular media backups \n \n \n \n \n employ appropriate security-related software and procedures \n \n \n \n \n guard confidentiality of private information, including user files and system access codes \n \n \n \n \n control physical access to equipment \n \n \n \n \n provide proper physical environment for equipment \n \n"
        },
        {
            "page_number": 216,
            "text": " \n \n \n provide safeguards against fire, flood, theft, etc. \n \n \n \n \n provide proper access administration; e.g., prompt and appropriate adjustment of access permissions \nupon a user's termination or transfer\n \n \n \n \n \n control and record system software and configuration changes \n \n \n \n \n monitor system logs for access control violation attempts \n \n \n \n \n \nUnits are to designate a person employed by the unit as their Computing Support Representative \n(CSR); the Director of Client Services, Office of Information Technology is to be notified of CSR \nappointments. CSRs should be knowledgeable about their unit's computing environment and about \ncentral resources and services. The CSR serves:\n \n \n \n \n \n as the first point of contact for unit personnel seeking problem resolution, information, and other \nassistance regarding computing and networking\n \n \n \n \n \n to facilitate interaction between the unit and the Office of Information Technology \n \n \n \n \n 6.5— \nPublic Information Services \n \n \n \n \n \nUnits and individuals may, with the permission of the appropriate unit head, configure computing \nsystems to provide information retrieval services to the public at large. (Current examples include \n\"anonymous ftp\" and \"gopher.\") However, in so doing, particular attention must be paid to the \nfollowing sections of this policy: 2.1 (authorized use [must be consistent with Institute mission]), 3.3 \n(ownership of intellectual works), 4.2 (responsible use of resources), 4.9 (use of copyrighted \ninformation and materials), 4.10 (use of licensed software), and 6.4 (individual unit responsibilities). \nUsage of public services must not cause computer or network loading that impairs other services.\n \n \n \n \n \n 7— \nProcedures and Sanctions  \n \n"
        },
        {
            "page_number": 217,
            "text": " \n \n \n 7.1— \nInvestigative Contact \n \nIf you are contacted by a representative from an external organization (District Attorney's Office, FBI, \nGBI, Southern Bell Security Services, etc.) who is conducting an investigation of an alleged violation \ninvolving Georgia Tech computing and networking resources, inform the office of the Executive \nDirector for Information Technology (EDIT) and the Chief Legal Advisor immediately. Refer the \nrequesting agency to the EDIT office; that office will provide guidance regarding the appropriate \nactions to be taken.\n \n \n \n \n \n 7.2— \nResponding to Security and Abuse Incidents \n \n \n \n \n \nAll users and units have the responsibility to report any discovered unauthorized access attempts or \nother improper usage of Georgia Tech computers, networks, or other information processing \nequipment. If you observe, or have reported to you (other than as in 7.1 above), a security or abuse \nproblem with any Institute computer or network facilities, including violations of this policy:\n \n \n \n \n \n \nTake immediate steps as necessary to ensure the safety and wellbeing of information resources. For \nexample, if warranted, a system administrator should be contacted to temporarily disable any \noffending or apparently compromised computer accounts, or to temporarily disconnect or block \noffending computers from the network (see section 5.6).\n \n \n \n \n \n \nEnsure that the following people are notified: (1) your Computing Support Representative, (2) your \nunit head, (3) the Information Resources Security Coordinator (IRSC), who is located within the \nOffice of Information Technology.\n \n \n \n \n \n \nThe IRSC will coordinate the technical and administrative response to such incidents. Reports of all \nincidents will be forwarded to Student Services (for apparent policy violations by students) or the \nunit head (for employees), and to the Executive Director for Information Technology and the Chief \nInformation Officer.\n \n \n \n \n \n 7.3— \nFirst and Minor Incident \n \n \nIf a person appears to have violated this policy, and (1) the violation is deemed minor by OIT, and (2) \nthe person has not been implicated in prior incidents, then the incident may be dealt with at the OIT \n"
        },
        {
            "page_number": 218,
            "text": "  or unit level. The alleged offender will be furnished a copy of the Institute Computer and Network \nUsage Policy (this document), and will sign a form agreeing to conform to the policy.\n  \n \n \n \n \n 7.4— \nSubsequent and/or Major Violations \n \n \n \n \n \nReports of subsequent or major violations will be forwarded to Student Services (for students) or the \nunit head (for employees) for the determination of sanctions to be imposed. Units should consult the \nOffice of Human Resources regarding appropriate action.\n \n7.5— \nRange of Disciplinary Sanctions \n \n \n \n \n \nPersons in violation of this policy are subject to the full range of sanctions, including the loss of \ncomputer or network access privileges, disciplinary action, dismissal from the Institute, and legal \naction. Some violations may constitute criminal offenses, as outlined in the Georgia Computer \nSystems Protection Act and other local, state, and federal laws;the Institute will carry out its \nresponsibility to report such violations to the appropriate authorities.\n \n \n \n \n \n 7.6— \nAppeals \n \n \n \n \n Appeals should be directed through the already-existing procedures established for employees and \nstudents.\nChapter 8— \nContingency Planning \n \n \n \n \n Objective \n \n \n \n \n \nThe purpose of computer security is to protect the information services of the organization as a \nwhole. Information should not be lost, damaged, or modified. It should be readily available to \nauthorized users. It should not be possible to accidentally or intentionally disable the computer \nsystem.\n \n \n \n \n \n \nContingency planning is a strategy to minimize the effect of disturbances and to allow for timely \nresumption of activities. The aim of contingency planning is to minimize the effects of a disruption \non your organization's operations. A disruption is any security violation, man-made or natural, \n \n"
        },
        {
            "page_number": 219,
            "text": "intentional or accidental, that affects normal operations. Disruptions in computer processing can be \nclassified into three categories:\n \n \n \n \n • Malfunctions: Minor disruptions that affect hardware, software, or data files. They're usually quite \nnarrow in scope, and it's usually possible to recover from them quickly.\n \n \n \n \n \n \n• Disasters: Disruptions to the entire facility. They typically require the use of alternate off-site \nprocessing facilities to recover operations. Entire facilities may be disrupted for a significant period \nof time.\n \n \n \n \n \n \n• Catastrophes: The most serious type of disruption. In a catastrophe, the facilities may have been \ndestroyed. Alternate facilities are always needed to process data. It may be necessary to rebuild or \nestablish new or permanent facilities.\nRarely will a company face either a disaster or a catastrophe. Malfunctions or other minor failures are \nlikely to be the norm. For minor malfunctions, it's generally more convenient to use onsite backup \nfacilities.\n \n \n \n \n \n \nYour contingency plan should focus on the continuity of your business. Its primary purpose is to \nreduce the risk of financial loss and enhance your organization's ability to recover from a disruption \npromptly, at least cost. It should apply to all facets of your organization: staff, computer programs, \ndata, workspace, production, and vital records. Contingency planning for your information systems \nshould look at all critical areas including LANs and WANs, client server systems, distributed \ndatabases, and PCs.\n \n \n \n \n \n \nA common mistake in contingency planning is an excessive focus on computer recovery. You really \nneed a business recovery plan. Undue emphasis on the technology, rather than the business, is \ncounter-productive. Quick recovery of computer technology is useless if your organization cannot \nrecover its business. Excessive focus on computer technology results in committing too many \nresources to redundant processing facilities.\n \n \n \n \n \n \nA contingency is an event that may or may not occur. The focus of computer security contingency \nplanning is to provide options in case disruption strikes. Recovery from loss of key personnel is \nusually accomplished through succession planning and backup training. Computer facilities are \ntypically covered by insurance policies and businesses can generally recover their investment in \ncomputers and equipment.\n \n \n \n \n \n But losses in a disaster or catastrophe typically exceed what's recoverable through insurance policies. \nSome types of losses are uninsurable.\n \n"
        },
        {
            "page_number": 220,
            "text": " \n \n \n \n \nThe primary focus of computer security should always be preventive rather than corrective action, \nthough it's impossible to prevent every security breach. It's virtually impossible to anticipate every \nproblem; even if a problem can be anticipated, the cost/benefit ratio may not justify taking preventive \nmeasures. Sometimes precautionary measures may prove ineffective cause of human or other error. \nProductivity and efficiency may also be sacrificed if precautionary measures are taken too far.\n \n \n \n \n \n \nYou'll need emergency procedures for each type of potential disaster. Think about how the disaster \nmight affect data processing and business operations. How long would the service be interrupted? At \nwhat level would the company be able to operate?\n \n \n \n \n \n \n \nOrganizations are sometimes hesitant about using resources to develop a disaster or catastrophe \nrecovery plan. The probability of a disaster or catastrophe is generally low, and the high costs \nassociated with developing a detailed contingency plan may be a deterrent. Many organizations may \nfeel that the costs exceed the potential benefits.\nYet while the probability of a disaster or catastrophe may be low, the cost of being unprepared is high. \nMost businesses are heavily dependent on computer technology; even a minor interruption could have \nserious financial consequences.\n \n \n \n \n \n \nContingency planning should provide an organized way to make decisions if there's a disruption. Its \npurpose is to reduce confusion and enhance the ability of the staff to deal with the crisis. When a \ndisruption occurs, a company doesn't have the time to deliberate, plan, and organize its recovery. The \norganization needs to recover quickly. A well-tested, comprehensive recovery plan can save critical \ntime (and therefore money).\n \n \n \n \n \n The Role of Senior Management \n \n \n \n \n \nSenior management has ultimate responsibility for establishing, regulating, and monitoring \ncontingency plans. Senior management support is essential. Management must be willing to commit \nadequate resources, both tangible and intangible. The people at the top should appoint a team to \nmanage the contingency planning process.\n \n \n \n \n \n \nTop management has a fiduciary responsibility to protect the organization's assets. If after a \ndisruption there are losses that could have been prevented or minimized by planning, shareholders \nand creditors may hold senior managers as well as the board of directors personally liable.\n \n \n \n"
        },
        {
            "page_number": 221,
            "text": "  \nGovernment regulation, such as the Foreign Corrupt Practices Act of 1977 (FCPA), may impose \nadditional civil and criminal penalties. The FCPA requires all publicly held organizations to maintain \nadequate controls over their information systems. Organizations may take reasonable steps to ensure \nthe integrity of their records and the internal control structure. An organization that fails to protect its \ninformation records can be held in violation of the FCPA; penalties range from fines to \nimprisonment.\n  \n \n \n \n \n The Contingency Planning Committee \n \n \n \n \n \nA committee should formulate, test, and implement your contingency plans; the information systems \nmanager should be a key member, along with members from functional areas throughout the \norganization. The committee defines the scope of the plan, which should deal with how to:\n• Prevent disruptions \n \n \n \n \n • Minimize loss if a disruption cannot be prevented \n \n \n \n \n • Recover from a disruption in an organized and expedient manner \n \n \n \n \n \nThe planning committee should consult with all major departments and specialists within and outside \nthe organization before drafting a plan. Then each department should review the draft to suggest \nimprovements and modifications. There's plenty of expertise to call on:\n \n \n \n \n \n • Internal auditors will play a key role in evaluating the internal control structure and conducting \noperational audits. They are also familiar with the needs of external auditors.\n \n \n \n \n \n • The lawyers should be consulted with respect to the legal consequences of a disruption, including \ncompliance with government regulations, such as the FCPA.\n \n \n \n \n \n • The accounting and finance departments, which are heavily dependent on information technology, \nare likely to suffer considerably from a disruption.\n \n \n \n \n \n • The security department is responsible for coordinating the recovery efforts if there is a security \nbreach, fire, earthquake, flood, or bomb threat.\n \n \n"
        },
        {
            "page_number": 222,
            "text": " \n \n \n \n• Medical specialists should be consulted for ways to protect human life in a disaster. They should \nknow the effects of fire extinguishers, such as Halon or carbon dioxide, and other chemicals. They \ncan advise on the type of first aid equipment that should be kept available and how employees should \nbe trained.\n \n \n \n \n \n • The public relations department should be responsible for all communications to the press and \nothers outside the organization.\n \n \n \n \n \n Areas to Cover \n \n \n \n \n Human Safety  \n \n \n \n \n The primary concern in any type of planning should be the health and safety of the people who work \nfor and with you. You need a plan for:\n \n \n \n \n \n • Emergency evaluation \n \n • Alerting the fire department and other emergency response authorities \n \n \n \n \n • Health and safety concerns unique to the business (such as in a chemical manufacturing plant) \n \n \n \n \n Business Impact Analysis  \n \n \n \n \n \nIn formulating your disaster recovery plan, conduct a business impact analysis, determining the likely \ncost of each risk taken to a worst-case scenario. A business impact analysis considers how various \nthreats and vulnerabilities might affect the continuity of your business. Incorporate into your plan \nrecovery strategies for specific disasters, emphasizing your backup strategy, including the role of any \noff-site facilities that will be needed. Appendix 8.A contains an Impact Analysis Worksheet to help \nyou with this.\n \n \n \n \n \n Legal Liability  \n \n \n"
        },
        {
            "page_number": 223,
            "text": "  \nThe planning committee may not realize the existence of certain liabilities during its risk analysis. \nThe plan itself may violate certain legal requirements, exposing the organization to unnecessary \nliabilities.\n  \n \n \n \n \n \nTo avoid that, the committee will need to review corporate documents, including the Articles of \nIncorporation and the by-laws, and then consider the impact of federal, state, and local laws on your \ncontingency planning. Violation of laws may prevent recovery from insurance policies.\n \n \n \n \n \n \nFor example, the Worker Adjustment and Retraining Notification Act limits the right of an employer \nto lay off personnel or close a plant. Employers must provide at least 60 days notification to state and \nlocal officials as well as to employees. While natural disasters such as earthquakes are excluded from \nthe notification requirement, other disasters—such as fires—are not. Without adequate insurance \ncoverage, your organization may be unable to lay off personnel and must continue paying salaries \nand benefits to the remainder.\n \n \n \n \n \n Flexibility  \n \n \n \n \n \nThe recovery plan has to be flexible enough to cover a wide variety of disasters and catastrophes. \nWhen planning for resources, consider the effects on your business of a single disaster, such as a fire, \nversus the effects of a community-wide disaster, such as an earthquake or flood. In a community-wide \ndisaster, outside resources to help your organization may be severely strained. In your contingency \nplanning, identify the resources you will need in both types of disaster.\nNotification Procedure  \n \n \n \n \n The notification procedure in a disaster should be clearly specified. Ask yourself: \n \n \n \n \n • Who should be notified? \n \n \n \n \n • How are they to be notified? What if phone lines and e-mail aren't working? \n \n \n \n \n • What mobile communications equipment might be needed to provide notification? \n \n \n \n \n • Who will be responsible for notification? \n \n"
        },
        {
            "page_number": 224,
            "text": " \n \n \n • Where will the primary notification list be kept? Where will the backup list be kept? \n \n \n \n \n • How often should the notification list be updated? \n \n \n \n \n A Communications Assessment Questionnaire is presented in Appendix 8.B. \n \n \n \n \n Access to Facilities  \n \n \n \n \n \nThe recovery plan has to consider the effect of delayed access to facilities after a disaster. For \nexample, even if a fire has been put out, the authorities may not allow anyone inside the building till \nthey've assessed the amount of structural damage. Law enforcement authorities may be conducting a \ncriminal investigation (e.g., arson), so the building may be considered a crime scene. In a \nmanufacturing plant there may be danger of toxic chemical contamination. In a community-wide \ndisaster like a tornado or a flood you may not be allowed into the geographical area where your \nfacility is located for anywhere from a couple of hours to several weeks.\n \n \n \n \n \n \nIdentify early those areas of the facility that require priority access. This will expedite damage \nassessment. If urgent access to certain areas will be required, check with local authorities before \nanything happens to ascertain the proper procedure. Under special circumstances, they may grant \naccess to qualified individuals.\nEmergency Acquisitions  \n \n \n \n \n \nIf something goes wrong, what supplies will you need? The contingency plan should include the \nprocedure for acquisitions during an emergency, with a list of pre-authorized emergency supplies. \nSpecify by job title who will be responsible for emergency acquisitions, with dollar limits, and \nprocedures to authorize expenditures beyond the limits. Any special accounting requirements should \nbe discussed with the accounting department and incorporated into the contingency plan.\n \n \n \n \n \n Vital Records  \n \n \n \n \n Vital records need to be recovered quickly from off-site backup locations. The contingency plan \nshould specify:\n \n \n \n \n \n • Documents and records likely to be needed first \n"
        },
        {
            "page_number": 225,
            "text": " \n \n \n \n • Where vital records are stored \n \n \n \n \n • Equipment and other resources that might be needed for recovery \n \n \n \n \n • Where the records will be stored once recovered \n \n \n \n \n Backup Requirements  \n \n \n \n \n \nIt's essential to have a backup strategy for hardware, software, data, and documentation. Backup \nstrategy should incorporate functions that are critical for the survival of your organization. For \nexample, data files must be backed up regularly and often. All vital records, whether or not \ncomputerized, should be protected.\n \n \n \n \n \n \nSpecify short-term as well as long-term needs. Contingency planning goes beyond simply keeping a \nbackup of records for short-term recovery. In fact, it should be more focused on the long term. In the \npast, when computer centers tended to be centralized, traditional contingency plans focused on their \nrecovery. With the shift to decentralized information systems, however, contingency planning must \nnow focus on the entire organization, not simply the information processing centers or data centers.\n \n \n \n \n \n \nA written backup policy will clarify procedures and prevent operational errors. The policy should \nspecify the backup schedule for each type of data file and how long each generation of data files must \nbe kept.\n \n \n \n \n \n Recovery in a disaster can be expedited if your organization does not use unique hardware or \nsoftware. Standard technology and up-to-date equipment and software makes replacement simple.\nBackup Facilities  \n \n \n \n \n \nYou need both on-site and off-site backup of hardware, software, data, and documentation. On-site \nbackup is convenient and readily accessible; a simple fire-resistant safe may be sufficient. Store in it \nthe most current copy of the backup files.\n \n \n \nHowever, in a disaster, on-site backup tends to be inadequate. Whatever disaster—fire, earthquake, \n"
        },
        {
            "page_number": 226,
            "text": "  flood—affects the primary hardware, software, or data files will also affect the backup. That's why \noffsite backup storage is necessary.\n  \n \n \n \n \n \nAn off-site facility may be nearby of convenience or far away for enhanced protection. A nearby \nfacility may not be affected by a fire but may be susceptible to damage from natural disasters like \nearthquakes or hurricanes. For a local off-site facility a fire-resistant safe in a building within a mile \nor two may provide adequate protection.\n \n \n \n \n \n \nA remote off-site facility provides still greater protection. A fire-resistant storage area several miles \naway may be accessed weekly and used to house data files for at least several weeks. A very remote \noff-site facility—preferably at least 100 miles away—provides the greatest protection and may be \nused for archival storage.\n \n \n \n \n \n \nLarger organizations will typically keep several generations of backups in different facilities. The \nlogistics of moving files to different facilities can become complex; you'll need procedures to ensure \nthat data files are correctly backed up and transported to the right facilities.\n \n \n \n \n \n Backup sites may be categorized into three types: \n \n \n \n \n • Hot sites: Sites that can become fully operational on short notice, typically a few hours. All \nhardware and software must be compatible with the original site.\n \n \n \n \n \n \n• Warm sites: Sites that can become fully operational in a matter of days or weeks. Less expensive \nequipment and software may already be in place. More expensive equipment or software is purchased \nor leased if there is a serious disruption. A partial degradation in processing output is considered \nacceptable.\n \n \n \n \n \n \n• Cold sites: Sites that have only a skeleton structure in place. It may take anywhere from several \ndays to weeks or even months to get the site operational. Hardware and software are not in place, but \nthere is adequate electrical wiring and air conditioning.\n \n \n \n \n \n \nWhile hot sites offer considerably faster recovery time, their cost is high. Warm sites are less \nexpensive; cold sites are typically useful for longer-term computing needs. An organization may wish \nto use all three types of sites in its recovery plan. For example, a hot site could be established for \nshort-term processing needs, giving the company enough time to prepare a warm or cold site for the \nlong run.\nFor critical applications where a delay in computer processing may cause significant financial loss, you \n"
        },
        {
            "page_number": 227,
            "text": "can subscribe to a commercial hot site. Though the cost is high, these sites will guarantee that you'll \nhave properly configured hardware and software to run your data when you really need it.\n \n \n \n \n \n \nThe costs of commercial hot sites vary considerably. Most vendors charge a monthly subscription fee \nplus an activation fee and an hourly rate when you actually need to use the site. The activation fee \nand the hourly rates may be covered by insurance.\n \n \n \n \n \n \nSome vendors charge a very high activation fee to discourage use in non-emergency situations. \nOthers encourage their subscribers to use the facilities for non-emergency situations such as overload \nprocessing.\n \n \n \n \n \n \nCommercial hot sites should not be relied on for an extended period of time. For longer-term needs, \nthere are also commercial warm and cold sites. A warm site subscription typically provides for basic \ncomputer facilities with some hardware. The cold site typically provides the infrastructure for a \ncomputer facility, such as air conditioning, heating, and humidity controls; wiring for voice and data; \nelectrical wiring; and flood and fire protection. It usually does not contain hardware or software.\n \n \n \n \n \n \nIn choosing a commercial hot site, consider the number and concentration of subscribers. Too many \nmeans that the facility may not be available when needed; too few raises concerns about the financial \nsurvival of the business. Look out for a geographic concentration of subscribers: If there's a natural \ndisaster, the site may be unable to meet the needs of all its subscribers.\n \n \n \n \n \n When you're selecting a hot site, take into account: \n \n \n \n \n • Activation fees \n \n \n \n \n • Monthly fees \n \n \n \n \n • Usage fees \n \n \n \n \n • Contract period and penalty for early termination \n \n \n \n \n • Number of subscribers \n"
        },
        {
            "page_number": 228,
            "text": " \n \n \n \n • Geographical concentration of subscribers \n \n \n \n \n • Networking capacity \n \n \n \n \n • Customer service and technical assistance \n \n• Hardware and software provided \n \n \n \n \n • Ability to upgrade hardware and software if necessary \n \n \n \n \n • Vendor's expertise and experience in your industry \n \n \n \n \n • Vendor's financial stability \n \n \n \n \n • Vendor's experience with actual disasters \n \n \n \n \n • References \n \n \n \n \n • Other services provided \n \n \n \n \n \nMobile backup sites are available from several vendors. These are trailers equipped with computer \nhardware and software that can be taken to any location you desire. This setup is especially helpful \nwhere it may be difficult for the personnel to commute to a remote backup facility.\n \n \n \n \n \n To identify the best recovery strategy you'll need to conduct a thorough technical as well as cost-\nbenefit analysis. Don't forget how much it costs for wiring, air conditioning, and fire prevention.\n \n \n \n \n \n Mutual Aid Agreements  \n \n"
        },
        {
            "page_number": 229,
            "text": " \n \n \n \nMutual aid agreements are a low-cost alternative for emergency processing: Two or more \norganizations with similar equipment and applications promise to help each other in an emergency. \nThese agreements may not be legally enforceable so they shouldn't be relied on exclusively, but they \nmake sense especially for companies with special requirements that may be unable to use commercial \nhot sites.\n \n \n \n \n \n If you're thinking about a mutual aid agreement, ask yourself: \n \n \n \n \n • What are the differences in equipment between my company and the potential partner? \n \n \n \n \n • What are the differences in applications? \n \n \n \n \n • How will we maintain compatibility in equipment and software with the partner over time? \n \n \n \n \n • How often will we test our systems for compatibility? \n \n \n \n \n • In an emergency, how much processing time will they have available? If there are multiple partners, \nwhich one will we use?\n \n \n \n \n \n • What equipment, software, staff assistance, and facilities will the partner provide?\n• How long will it take before we can use the partner's facilities? \n \n \n \n \n • How long can we use the partner's facilities? \n \n \n \n \n Relocating Facilities  \n \n \n \n \n \nThe severity of the damage to your own facilities will dictate whether your organization relocates \ntemporarily or permanently. The contingency plan should cover both possibilities. Recovering from a \ndisaster may take a long time; what happens in the interim? For example, if only a portion of your \nfacilities can be used:\n \n \n"
        },
        {
            "page_number": 230,
            "text": " \n \n \n • Which departments or functions will go back in first? \n \n \n \n \n • What type of support will they require? \n \n \n \n \n • Can large departments be split up, operating out of two or more locations? \n \n \n \n \n Factors to think about with regard to relocating include: \n \n \n \n \n • Square footage needed \n \n \n \n \n • Which departments need to be in proximity \n \n \n \n \n • Communications facilities needed \n \n \n \n \n • Security concerns \n \n \n \n \n • Storage area needed \n \n \n \n \n • Access to public transportation \n \n \n \n \n • Parking needed \n \n \n \n \n • Employee requirements, such as housing, schools, day care, etc. (if relocating at a distance) \n \n \n \n \n Hardware Backup  \n \n \n"
        },
        {
            "page_number": 231,
            "text": "  \nMany organizations think that hardware backup is all that's needed for contingency planning, but \nhardware backup alone is not sufficient for most companies. While hardware is an essential element \nof contingency planning, there's a lot more to it.\n  \n \n \n \n \n \nIt is important to have backup hardware available both on and offsite. For smaller disruptions, only \none or at most a few pieces of equipment are likely to be affected. On-site backup is usually sufficient \nto resolve such problems. To save on costs you might consider using older or slower systems in \nemergencies. Sometimes partial processing capability is acceptable.\nFor more serious disruptions, which tend to be longer, off-site hardware is of considerable importance. \nWhile hardware vendors are naturally the best source for replacement equipment, the vendor may be \nunable to supply the equipment you need quickly enough for you. If that's so, used hardware dealers \nmay be able to supply critical components on short notice.\n \n \n \n \n \n Software and Data Backup  \n \n \n \n \n \nSoftware and data are intangible assets that must be protected. Software includes the operating \nsystem, utilities, and application programs. Keep backup copies not only of your primary software \nbut also of any upgrades or patches to fix bugs, and of user manuals (online or hard copy). It's \nunnecessary to keep backup copies of older generation software.\n \n \n \n \n \n \nYou need backup copies of the software and user manuals at both primary and off-site facilities, \nwhere you also need copies of the software configuration files or other special settings. Don't forget \nto upgrade software at both facilities at the same time, and set in place a procedure to ensure that \ncopies of maintenance patches to the software are also kept off-site. Check the legal requirements for \nmaking backups: Some vendors require an additional licensing fee if the software will be used in an \nalternate facility.\n \n \n \n \n \n \nData includes all source documents, electronic data files and databases, and output documents. \nUnlike software, data by its very nature changes continuously. The contingency plan should prepare \nfor rapid recovery of data.\n \n \n \n \n \n \nThere are several different techniques for backing up data. For small files the process may be as \nsimple as making a duplicate on the backup medium. When it's either not practical or not cost-\neffective to backup the entire file, try something else.\n \n \n \n \n \n \nIn a batch processing system, a new master file is created using the old master and updating it with \ntransactional changes. Keeping copies of the old master file for two or three generations, along with \ncopies of the transaction file updates, provides for continuous backup of the data.\n \n"
        },
        {
            "page_number": 232,
            "text": " \n \n \n \n \nA real-time online processing system requires different procedures. Duplicate logging of transactions \nmay be necessary. Copies of individual records may have to be kept before updating. Other \ntechniques, such as before-and-after images of master records, may be useful. Specialized backup \nprocedures are typically an integral part of database management systems.\nBefore recovering data from backup files, it's advisable always to duplicate the backup files to prevent \naccidents. Data integrity during the reconstruction process can then be ensured with special supervision \nand controls.\n \n \n \n \n \n Documentation Backup  \n \n \n \n \n \nAlong with hardware, software, and data, backup user manuals and other documentation should be \nstored in an off-site fire-resistant location. When changes are made to the primary manuals and \ndocumentation, make sure that the backup documentation is updated as well.\n \n \n \n \n \n Systems and program documentation that should be backed up include: \n \n \n \n \n • Source code for programs \n \n \n \n \n • Flowcharts \n \n \n \n \n • Program logic descriptions \n \n \n \n \n • Error conditions \n \n \n \n \n Vital Records and Source Documents  \n \n \n \n \n \nWhile most organizations store backup copies of vital records off-site, there are still numerous \ndocuments that may only be in the primary facilities because they have not yet been backup up. \nMany of these are source documents like invoices or purchase orders that originate outside the \norganization. If a source document is critical and the information in it cannot be easily reconstructed \nif the document is destroyed, you need a procedure to copy the source document.\n \n \n"
        },
        {
            "page_number": 233,
            "text": " \n \n \n Remote Backup  \n \n \n \n \n \nSeveral companies offer remote backup using the Internet. Periodically, usually once a day, your data \nfiles can be automatically backed up to an off-site computer facility where the data is stored \nencrypted. Companies that offer remote backup include:\n \n \n \n \n \n • At Backup, Inc.: http://www.atbackup.com/  \n \n \n \n \n • Atrieva: http://www.atrieva.com/  \n \n \n \n \n • BACKUP Data Protection Agency: http://members.aol.com/backupdpa/index.htm  \n \n \n \n \n • Connected Corporation: http://www.connected.com/  \n \n• CYBeRGEM Remote Backup System: http://www.cybrgem.com/backup/  \n \n \n \n \n • DataLock Remote Data Services: http://datalock.com/  \n \n \n \n \n • Filetron: http://www.back-up.com/  \n \n \n \n \n • Guardian Computer Service: http://www.guardiancomputer.com/  \n \n \n \n \n • Netsafe: http://www.evault.com/  \n \n \n \n \n • Offsite Data Management Remote Backup Service: http://odms.com/  \n \n \n \n \n • Saf-T-Net: http://www.trgcomm.com/  \n \n \n"
        },
        {
            "page_number": 234,
            "text": "  • SafeGuard Interactive: http://www.sgii.com/   \n \n \n \n \n • TeleBackup Systems, Inc.: http://www.telebackup.com/  \n \n \n \n \n \nA more costly alternative is to transmit a copy of each transaction to a remote facility. In mirror \nprocessing the data is not only transmitted but also updated on a duplicate database. With mirror \nprocessing the backup database is always current. This doesn't necessarily have to be done at a \nremote location; you may want to set it up on-site.\n \n \n \n \n \n Fire Safety \n \n \n \n \n \nFire is the most common cause of damage to computer centers. Combustible material should not be \nallowed in any computer room. Install fire-detectors in vulnerable locations. Use fire-retardant \nmaterial for office furniture, draperies, and any floor coverings. Put waste receptacles outside the \ncomputer room: Computer paper in waste receptacles can accelerate a fire.\n \n \n \n \n \n Detection Devices  \n \n \n \n \n \nFire detectors sense either thermal combustion and its byproducts or changes in temperature. They \nmay be actuated by smoke, heat, or flame. All detectors must meet government standards. Make sure \nthe devices are connected to an automatic fire alarm system.\n \n \n \n \n \n \nSmoke-actuated devices provide early warning of fires developing slowly. They should be installed \nin all air conditioning and ventilating systems. Smoke detectors typically rely on either photoelectric \nor radioactive devices.\n \n \n In the photoelectric cell, variations in the intensity of light cause changes in electric current. These \ndetectors are generally of three types:\n \n \n \n \n \n • Area-sampling devices draw in air from the area to be protected: If smoke is present in the sampled \nair, the light reflections on the photoelectric cell will trigger the alarm.\n \n \n \n \n \n • Beam devices focus a beam of light onto a photoelectric cell from across the protected area. Smoke \ncausing an obstruction in the light activates the alarm.\n \n \n"
        },
        {
            "page_number": 235,
            "text": " \n \n \n • Spot devices, unlike beam devices, contain the light source and the receiver in one unit. Smoke \nentering the detector causes the light to reflect onto the photoelectric cell, activating the alarm.\n \n \n \n \n \n \nRadioactive smoke detectors contain a minute amount of radioactive material in a special housing \n(the danger from radiation from such devices is minimal). Smoke interacting with the radioactive \nmaterial changes its ionization, which activates the alarm. Radioactive detectors are mostly \ncommonly spot types.\n \n \n \n \n \n \nThe response time for radioactive smoke detectors is affected by several variables, including the \nstratification of air currents and the nature of the combustion products. Generally, the heavier the \nparticles resulting from combustion, the longer it takes for them to reach the ceiling, where the smoke \ndetectors are usually attached, and the longer the response time of the unit.\n \n \n \n \n \n \nHeat-actuated detectors can be of two types. The first will activate the alarm when the temperature \nreaches a predetermined value. The second senses the amount of change in temperature. Typically, \nwhen the rise in temperature exceeds 15° to 20° F. the alarm is activated.\n \n \n \n \n \n \nIn highly combustible areas, the rate-of-rise temperature detectors are recommended because their \nresponse time is faster. However, fixed temperature detectors are not as prone to false alarms. Some \nheat-actuated detectors contain both types of sensors.\n \n \n \n \n \n Heat detectors are available in line or spot styles. Line-type detectors usually rely on heat-sensitive \ncables or a pneumatic tube. Spot-type detectors are placed at fixed intervals in each zone.\n \n \n \n \n \n \nFlame-actuated detectors are of two types, both expensive: flameradiation-frequency and flame-\nenergy. Flame-radiation-frequency detectors sense the flame-related flicker caused by combustion. \nFlame-energy detectors sense the infrared energy of the flame. They tend to be best-suited to \nprotecting expensive equipment; their principal advantage is that they are super-fast. These detectors \ncan also produce enough voltage to trigger the release of an extinguishing agent.\nExtinguishing Agents  \n \n \n \n \n Different types of fires require different types of extinguishing agents. Using the wrong extinguishing \nagent can do more harm than good.\n \n \n \n \n \n• Fires fed by ordinary combustible materials, such as wood, paper, plastics, and fabric, can be safety \n"
        },
        {
            "page_number": 236,
            "text": " extinguished with water or triclass (ABC) dry chemical.\n \n \n \n \n \n • Fires fed by flammable liquids and gases, such as oil, grease, gasoline, or paint, can generally be \nsafely extinguished with tri-class (ABC) dry chemical, halon, FM-200, and carbon dioxide.\n \n \n \n \n \n • Fires involving live electrical equipment should be extinguished only with a non-conducting agent \nsuch as tri-class (ABC), regular dry chemical, halon, or carbon dioxide.\n \n \n \n \n \n Hand-held fire extinguishers should be mounted on the wall, as should self-contained breathing \napparatus, because carbon dioxide discharge can suffocate humans.\n \n \n \n \n \n Electrical Fires  \n \n \n \n \n \nMost computer room fires will be electrical, caused by overheating of wire insulation or other \ncomponents. Because smoke from an electrical fire may be toxic, it should be avoided even in small \nquantities. Generally electrical fires cannot be extinguished until the heat source is eliminated.\n \n \n \n \n \n \nA power panel with circuit breakers for the major pieces of equipment should be easily accessible, \npreferably inside the computer room. The circuits should be clearly labeled so that equipment can be \nshut down quickly in an emergency. There should be separate circuits for redundant devices, and an \nemergency switch to shut down everything if there is a fire.\n \n \n \n \n \n \nIn a major fire or explosion, the only concern should be the safety of human life. Computer \nequipment and wiring is likely to be destroyed by the intense heat. That's why backup copies of disks \nand data should always be kept off-site. Not only will this help in recovering from a fire, it can also \nhelp during the fire since the staff will not be tempted to risk their lives saving data.\n \n \n \n \n \n \nCarbon dioxide, a colorless, odorless, and electrically non-conductive inert gas, is generally stored \nunder pressure as a liquid. Carbon dioxide extinguishes fire by reducing the amount of oxygen \navailable to it. It may not be effective on fires fed by materials such as metal hydrides, reactive metals \nlike sodium, potassium, magnesium, titanium, and zirconium, and chemicals containing oxygen \navailable for combustion (for more information, see \nhttp://www.afaefire.co.uk/htlm/carbort_dioxide.html).\nHalon has the potential of depleting the ozone layer. While halon is still in use, by international \nagreement it has not been manufactured since January 1, 1994. FM-200, a substitute, is not very \neffective against electrical fires. In an electrical fire, it's essential that the power be shut off because till \npower is stopped a fire extinguishing system will only suppress, not extinguish, the fire.\n \n"
        },
        {
            "page_number": 237,
            "text": " \n \n \n \n The suppression agent in FM-200 is a halogenated alkane, heptafluoropropane. Under compression \nFM-200, a colorless and odorless gas, becomes a liquid, which is stored in steel cylinders.\n \n \n \n \n \n \nOnce discharged, FM-200 returns to the gaseous state. It suppresses fire by cooling it and reducing \nthe amount of oxygen available. It's typically discharged rapidly, in no more than a few seconds. FM-\n200 helps prevent re-ignition. It doesn't leave any residue, it doesn't require cleanup after discharge, \nand it doesn't harm humans.\n \n \n \n \n \n Water Sprinklers  \n \n \n \n \n \nWater sprinkler systems are a simple, relatively inexpensive protection against fire. Most new \nbuildings are required by code to have sprinkler systems, though their accidental activation can cause \nsubstantial damage and it may take a long time before normal operations can be resumed.\n \n \n \n \n \n \nIn an electrical fire, water may even intensify the fire, causing greater damage. Sensors should \ntherefore be installed to cut off electrical power before sprinklers are turned on. It should also be \npossible to activate sprinkler heads individually to prevent damage to a wide area. A shut-off valve \ninside the computer room can help you shut off water when it's no longer needed, minimizing \ndamage if the system is accidentally activated.\n \n \n \n \n \n \nCarbon dioxide, halon, and FM-200 extinguishers don't require any cleanup after discharge. Foam or \ndry chemicals can be hard to remove. Quick removal of smoke should be a priority. The smoke or \nfire alarm should automatically activate special fans and blowers.\n \n \n \n \n \n Extinguishing Micro-computer Fires  \n \n \n \n \n \nIf computer equipment starts smoking, first cut off the equipment's electrical power. This is often \nsufficient to extinguish the fire by itself. If there are visible signs of fire, or if you can feel the heat, \nuse a fire extinguisher.\nCarbon dioxide extinguishers are often recommended for microcomputer-related fires. When using a \ncarbon dioxide extinguisher, don't spray the agent directly onto the surface of the CRT because the \nsudden drop in temperature it causes will shatter the glass.\n \n \n \n \n \n Training  \n"
        },
        {
            "page_number": 238,
            "text": " \n \n \n \n \nPersonnel should be trained for a fire emergency. Company policy should state exactly what action \nshould be taken if a fire starts or a smoke alarm goes off. Personnel should be strictly prohibited from \nrisking injury or loss of life to protect data or equipment.\n \n \n \n \n \n Preventing Damage  \n \n \n \n \n The following steps can reduce the damage caused by fire, and in the process reduce your insurance \npremiums:\n \n \n \n \n \n • Safes for storage of documents should have a minimum four-hour fire rating. \n \n \n \n \n • Walls, floors, and ceilings of computer facilities should have a minimum two-hour fire rating. \n \n \n \n \n • The fire alarm should ring simultaneously at the computer facility and the nearest fire station. In \naddition, fire alarm signals should be located to assure prompt response.\n \n \n \n \n \n • Vaults storing backup tapes and records should be in a separate building at a sufficient distance. \n \n \n \n \n • Smoke and ionization detection systems should be installed on the ceiling of the computer facilities. \nWater detection systems should be installed under the floors.\n \n \n \n \n \n • Halon, FM-22, or a similar chemical extinguishing system should be installed throughout the \nfacilities. Automatic sprinkler systems can be used in supply and support areas.\n \n \n \n \n \n • Building code and fire marshal regulations must be adhered to strictly. \n \n \n \n \n Fire Tracer and Win Tracer  \n \n \nFire Tracer products (http://www.emss.net) give the exact location of the source of a developing fire \n"
        },
        {
            "page_number": 239,
            "text": "  \nin a computer cabinet, control equipment set, or room. They can also be used for air handling units, \nreturn air grills, ducts, and large open areas, supported by air sampling techniques. With these \nproducts, air is pumped into a series of micro-bore tubes positioned in a cabinet, room, or \ncompartment being protected. The sample of air is then analyzed for the presence of either smoke \nparticles or gas. If the smoke level reaches a preset trace level, then the Fire Tracer switches from a \nsampling mode to a search mode to identify the exact location of the smoke source.\n \nThree separate alarms are triggered by the system. The first alarm level is triggered when the trace \ncycle is initiated. A pre-alarm is triggered to warn that a potential fire exists in the identified area. The \nfinal alarm indicates that a full incipient fire condition is present. Each of the three alarms may be \nconnected to a fire panel for control purposes.\n \n \n \n \n \n Win Tracer control software may be used to monitor and control Fire Tracers. Win Tracer provides: \n \n \n \n \n • Warning when smoke levels exceed preset limits \n \n \n \n \n • Location of the alarm \n \n \n \n \n • Action sequences and hazard indications \n \n \n \n \n • Remote control [reset, off/online & changing functions, such as alarm levels, etc.] \n \n \n \n \n • Display of site status [tracers which are on/offline, etc.] \n \n \n \n \n • Historical trends \n \n \n \n \n • Data logging and recording of significant events \n \n \n \n \n \nThe basic display is simple enough to be monitored by an unskilled operator. Bar graphs display \ncurrent levels of smoke or gas. Color-coding indicates when current levels exceed the alert levels. If \nthere is an alarm, there is a text description of the alarm source, supplemented by a site map \npinpointing the location, along with actions to be taken for different alarm types and locations (e.g., \nnotes to tell operator to phone fire station, clear room, whatever) and a list of hazards near the alarm \nlocations.\n \n \n"
        },
        {
            "page_number": 240,
            "text": " \n \n \n \nWin Tracer can be customized and the data analyzed. The administrator has remote programming \naccess to change text descriptions, action sequences, and hazard lists. Alarms may be acknowledged \nor cancelled. Trend graphs can be plotted to monitor compliance.\nInsurance \n \n \n \n \n \nYour insurance, with the right endorsements, must be adequate to protect against disruptions in \nbusiness. Your contingency plan should address insurance concerns both before and after a disaster. \nEspecially be sure you're covered if access to the facilities is delayed.\n \n \n \n \n \n \nStandard insurance policies treat computer equipment like industrial equipment, covering them for \nthe same threats. Standard policies don't cover computer equipment for power outages or electrostatic \ndischarges that may delete or destroy electronically stored information.\n \n \n \n \n \n \nA policy specifically geared toward electronic data processing (EDP) typically covers risks to \ncomputers, including replacement or repair. For all high-technology assets, the policy should contain \na replacement cost endorsement. Unlike assets like industrial machinery with a long life and low \ndepreciation, computer equipment tends to depreciate rapidly, losing its value. Without the \nreplacement cost endorsement, you may not be able to afford to replace computer equipment after a \ndisaster.\n \n \n \n \n \n \nEDP policies, however, typically don't cover damages resulting from loss or inability to access \ncomputer equipment. That's why you need business interruption and out-of-pocket expenses \ninsurance. Business interruption insurance covers not only losses caused directly by the disaster, but \nalso future business losses if operations can't resume quickly. Out-of-pocket expense coverage \nreimburses you for use of alternate computer facilities. Make sure your operations at the outsourcing \nvendor's facilities are covered.\n \n \n \n \n \n \nKeep your insurance policies up to date. Many organizations just keep on renewing a policy they \nbought years ago, though the business environment is continuously changing. A policy that provided \nadequate coverage even three years ago may fall short for current needs.\n \n \n \n \n \n \nMaintain a running list of organizational assets, along with their appraised values, updating it every \ntime an asset is added or retired. Certainly review it at least once or twice a year. And, of course, \nstore a copy of the list off-site.\n \n \n \n \n \n Many insurance policies require policyholders to insure their property to 80%, 90%, or 100% of its \nvalue, penalizing them if they fail to maintain the proper level. To reduce your chances of being \n \n"
        },
        {
            "page_number": 241,
            "text": "penalized, have your assets appraised periodically and insured to the value required by the policy.\n \n \n \n \n Check that your valuable papers and records coverage includes the cost of recreating documents, \nsuch as re-entering data and restoring damaged items.\nSoftware products are often excluded from insurance policies unless there is a specific endorsement, \nbut you need to have them covered in case there's a breakdown in equipment, such as a system crash.\n \n \n \n \n \n Appendix 8.C discusses how to implement an effective insurance recovery program. Appendix 8.D \ndiscusses how to save time and money if you need to file an insurance claim.\n \n \n \n \n \n Conclusion \n \n \n \n \n \nMany organizations have no workable disaster recovery plan. A common mistake is to put too much \nemphasis on electronic data processing and computer recovery and too little on keeping the business \nrunning. Many business functions may be able to survive with manual procedures; prepare \nprocedures to support the essential functions of your business until computer processing is re-\nestablished.\n \n \n \n \n \n Both user participation and the support of senior management are critical in formulating an effective \ncontingency plan.\n \n \n \n \n \n \nMany contingency plans provide unnecessary details and lack flexibility. Too much information \nmakes it difficult to update and revise the plan. Anything that can be dealt with at the time of the \ndisaster should be left out, as should anything that cannot be determined until the effect of a disaster \nhas been evaluated.\n \n \n \n \n \n Contingency plans should be reviewed, tested, and updated at least annually. At each review ask \nyourself:\n \n \n \n \n \n • What does the organization need now and in the future to survive after a disaster? \n \n \n \n \n • How are these needs likely to be fulfilled by the strategies selected in the plan? \n \n \n"
        },
        {
            "page_number": 242,
            "text": "Senior managers are ultimately responsible for contingency planning. While they may delegate many \ndetails to middle management and staff, they must play an integral part in the creation of the plan. \n  • How can the strategies be made more effective?  \n \n \n \n \n Those who were responsible for developing the contingency plan should not review, test, or evaluate \nit. Independent individuals will have greater objectivity and insight.\n \n \n \n \n \n \nDisasters or catastrophes are extremely rare. It's therefore imperative that your contingency plan be \ncost-effective, though the cost-benefit analysis is admittedly very difficult for rare events. \nFurthermore, your company's risk preference will affect how much risk you can tolerate.\nAppendix 8.A— \nBusiness Impact Analysis Worksheet1 \n \n \n \n \n \nWhat would be the effect of a worst-case scenario if any kind of disaster or calamity occurred at your \nwork facility? Calculating the ''impact value\" of a crisis will tell you how damaging a critical incident \nwill be if there is no crisis intervention.\n \n \n \n \n \n Amount of Scrutiny the Crisis Might Generate \n \n \n \n \n • What kind of news coverage might ensure? \n \n \n \n \n • What kind of agencies might get involved? OSHA? EPA? \n \n \n \n \n • How would stockholders react? \n \n \n \n \n • What about senior management's reaction and handling of event? \n \n \n \n \n On a scale of 1 to 10, how much scrutiny would your crisis incur? \n \n \n \n \n Company's Public Image \n \n"
        },
        {
            "page_number": 243,
            "text": " \n \n \n • How will the public view the crisis and your handling of it? \n \n \n \n \n • Will the company be seen as culpable? Liable? \n \n \n \n \n • What about the reactions of customers? \n \n \n \n \n • Community reaction? \n \n \n \n \n • Reaction of employees' families? \n \n \n \n \n On a scale of 1 to 10, estimate the potential damage to company image and reputation in the \naftermath of the crisis.\n \n \n \n \n \n Effect on Employee Morale and Productivity \n \n \n \n \n • How would the crisis affect employee productivity? \n \n \n \n \n • Absenteeism? \n \n \n \n \n • How much time might be spent dealing with the crisis at the expense of other work functions? \n \n \n \n \n • What about employee morale during this time? Stress? Outrage? \n \n \n \n \n On a scale of 1 to 10, estimate the amount of business interference the crisis might generate as a \nresult of its effect on employees.\n \n \n \n \n1This appendix, taken from Dan Paulk, \"Estimating the Impact Value of a Crisis\" (available through http:// \nwww.disaster-resource.com/), is based on the research of Steven Fink, Crisis Management: Planning for the \nInevitable (New York: American Management Association, 1986).\n"
        },
        {
            "page_number": 244,
            "text": "Damage to the Company's Bottom Line \n \n \n \n \n • What might be the cost in \"hard\" dollars to the company? \n \n \n \n \n • How much would be covered by insurance? \n \n \n \n \n • In terms of \"soft\" dollars, what will employee unrest, stress, absenteeism, lowered morale, and \nlowered productivity cost the company?\n \n \n \n \n \n • How would a crisis affect the company's ability to get products to market on time? Meet payroll? \nPay creditors?\n \n \n \n \n \n • What about the costs of increased workers' compensation claims? \n \n \n \n \n • Unemployment insurance claims? \n \n \n \n \n Combining hard and soft-dollar costs, what is your opinion (1 to 10) of the cost of the crisis? \n \n \n \n \n Legal Liability/Vulnerability \n \n \n \n \n • Was this incident foreseeable? \n \n \n \n \n • If so, did we take enough reasonable precautions to prevent the calamity? \n \n \n \n \n • Did we have any specific controls or policies in place to prevent the crisis? \n \n \n \n \n • Is there any \"discoverable\" information regarding this incident that could be interpreted in a \ndamning way by outsiders?\n \n \n"
        },
        {
            "page_number": 245,
            "text": " \n \n \n • Would an aggressive plaintiff's attorney smell an opportunity here? \n \n \n \n \n Considering the potential degree of legal liability, what is your estimate (1 to 10) of legal \nvulnerability regarding this incident.\n \n \n \n \n \n Impact Value of the Crisis = Total of 5 scores divided by 5:____________ \n \n \n \n \n The effect of a crisis on business functions increases as the impact value score increases. The score \ntells you how damaging a \"critical incident\" will be if there is no crisis intervention.\nAppendix 8.B— \nCommunications Assessment Questionnaire1 \n \n \n \n \n 1. Who is responsible for ensuring that backup communications are established? \n \n \n \n \n 2. What are your existing communications capabilities? \n \n \n \n A. Voice:____________________________________________________ \n \n \n \n B. Data:_____________________________________________________ \n \n \n \n C. Image transmission:_________________________________________ \n \n \n \n \n 3. Inventory your existing backup communications. \n \n \n \n A. Radio:_____________________________________________________ \n \n \n \n B. Cellular:___________________________________________________ \n \n \n \n C. Amateur radio:______________________________________________ \n"
        },
        {
            "page_number": 246,
            "text": " \n \n \n D. Data/fax:__________________________________________________ \n \n \n \n E. Public telephones:___________________________________________ \n \n \n \n F. Essential service:____________________________________________ \n \n \n \n G. Other:____________________________________________________ \n \n \n \n \n 4. Who plans to use these backup communications during a disaster? (Identify specific work groups \nor people—chances are several people think they're going to use the same backups.)\n \n \n \n \n \n 5. What are your critical circuits? What alternate communications capability do you have for them? \n \n \n \n \n 6. What are your critical systems? What alternate communications or backup protection do you have \nfor them?\n \n \n \n \n \n 7. Which systems have backup power? How long will it last? What are your plans to have backup \ngenerators available if needed?\n \n \n \n \n \n 8. What priorities have been established of communications restoration by: \n \n \n \n A. Location \n \n \n \n B. Systems/switches \n \n \n \n C. Facilities \n \n \n"
        },
        {
            "page_number": 247,
            "text": "  9. What are your alternative routing patterns? Are they activated automatically, or can you redirect \nyour traffic routes at the time of a disaster? What are your plans to redirect traffic, if you can?\n  \n \n \n \n1 This appendix is taken from Judy Bell, \"Communications Assessment Questionnaire\" (Disaster Survival \nPlanning, Inc , http://www.disaster-survival.com).\n \n \n10. What emergency restoration procedures have you established with your communications and \nequipment vendors?\n \n \n \n \n \n 11. Where do you store your communications information? How is it preserved as a vital record? \n \n \n \n \n 12. Who in your organization will need backup communications at the time of a disaster? \n \n \n \n \n 13. Who do they primarily need to talk to? \n \n \n \n \n 14. What will they use as their alternate communications if they aren't at work when the disaster \nstrikes?\n \n \n \n \n \n 15. Based on all the information examined, what do you recommend for alternate communications for \neach department and critical member of your organization?\nAppendix 8.C— \nInsurance Recovery Program1 \n \n \n \n \n The following ten steps will help your organization implement an effective insurance recovery \nprogram:\n \n \n \n \n \n \n1. Determine accurate corporate asset values based on either actual cash or replacement values. Cash \nvalue is the original cash value of the asset minus the time depreciation of the asset. Replacement \nvalue is what it would cost to replace the asset.\n \n \n \n \n \n \n2. Estimate recovery expenses by conducting several what-if scenarios to determine how much \ninsurance you will need to cover out-of pocket expenses related to the recovery of the information \ntechnology and corporate assets. This would include hot site declaration, daily usage fees, overtime, \nand travel.\n \n \n"
        },
        {
            "page_number": 248,
            "text": " \n \n \n \n3. Determine the business income coverage amount by completing a Business Impact Analysis (BIA; \nsee Appendix 8.A). If the BIA shows that your company would not have the cash reserves to meet \nforecasted operating expenses, consider business income insurance. This is expensive, so run your \nfinancial models carefully.\n \n \n \n \n \n \n4. Determine the coinsurance amount by accurately assessing how much insurance you need to carry. \nIf you don't carry insurance on a reasonable value of your assets, you will only collect a percentage \nof the actual loss rather than the full amount. The industry standard of coinsurance—a strategy \ndesigned to lower insurance premiums—is 80 percent, so if you have assets of $1 million, you need \nto be insured for at least $800,000. That's why you need an up-to-date inventory—it's very easy to \nfall below the coinsurance amount, especially during times of inflation.\n \n \n \n \n \n \n5. Request a premium reduction review. Many insurance companies offer a discount to companies \nthat have a formalized loss prevention program. They also have risk managers who can advise you on \nhow to make the loss prevention program more effective. This can save you 20 percent or more on \npremiums, depending on the sophistication of your program.\n \n \n \n \n1 This appendix is based on Tan Shreider's, \"Insurance: The Disaster Recovery Plan of Last Resort,\" Contingency \nPlanning and Management, June 1996.\n6. Properly insure proprietary software. Media coverage pays for the cost of replacement or \nreproduction only. On the other hand, commercial software is easily replaced for a nominal fee.\n \n \n \n \n \n \n7. Know your policy exclusions. Many companies have made insurance claims only to find that their \npolicy specifically excludes coverage for their most common risks. Look for the fine print regarding \nutility failures, media restoration, earthquakes, and other exclusions. Review exclusions for computer \nequipment used during travel or at employees' homes, since many policies specially exclude coverage \nfor equipment at locations not listed on the policy. Policy exclusions can be altered with a rider.\n \n \n \n \n \n \n8. See if you're over-insured in some areas. Today, when many hardware manufacturers provide \nrecovery and restoration insurance in lease agreements, you may be doubly insured. Worse, \nmanufacturers may be considered the primary carrier if the insurance they provide is more \ncomprehensive than a policy written elsewhere.\n \n \n \n \n \n \n9. Put the policy out for competitive bid. More than 50 major companies underwrite business \nrecovery policies. Placing your company's insurance needs out for competitive bid could save you as \nmuch as 40 percent on your premiums.\n \n \n \n"
        },
        {
            "page_number": 249,
            "text": "  \n10. Document your insurance response plan. If a disaster strikes your company, this will be one of the \nmost vital documents you have. It specifies the roles and responsibilities of your company and the \ninsurance company after a disaster. Just following the procedures set forth by the insurance carrier \ncould allow them to adjust the claim, which could waive your company's rights to dispute it.\n \nAppendix 8.D— \nMaking an Insurance Claim1 \n \n \n \n \n If you need to make a claim to obtain reimbursement from your insurance company for damages, the \nfollowing will help you save time and money.\n \n \n \n \n \n Hire a Reputable Claims Adjuster: A specialist experienced in dealing with insurance company \nclaims adjusters can substantially increase your settlement.\n \n \n \n \n \n \nDocument the Loss: Insurance companies require extensive documentation of losses before paying \nany claims. It's not unusual for a final settlement check to be issued years after the disaster. Proper \ndocumentation will shorten the time.\n \n \n \n \n \n \nArrange for a Special Loss-Payable Clause: Insurance policies often cover several parties under \none policy. Placing and negotiating a claim under this circumstance could require 10 or more \nendorsements before a settlement check is issued. Get a power of attorney or a waiver from other \nparties.\n \n \n \n \n \n \nReduce the Risk of Additional Loss: Insurance companies will pay for physical damage to the \nproperty for the peril insured against, but if additional damage occurs after the event and your \ncompany could have avoided or reduced that loss, payment may be forfeited. Board up those \nwindows, get a fence up, do whatever can be done to secure what's left of the property against new \nrisks.\n \n \n \n \n1 This appendix is based on Tan Shreider, \"Insurance: The Disaster Recovery Plan of Last Resort,\" Contingency \nPlanning and Management, June 1996.\nChapter 9— \nAuditing and Legal Issues \n \n \n \n \n Introduction \n \n \nMost organizations are facing a host of technologies ranging from e-commerce, the Internet, the \n"
        },
        {
            "page_number": 250,
            "text": "  \nIntranet, data mining, data analysis, data warehousing, and telecommunications to enterprise-wide \napplications, artificial intelligence expert systems and neural networks, and client-server computing. \nAll business processes today depend on effective and efficient information processing.\n  \n \n \n \n \n \nLaws and public policies to regulate the use of these technologies have not kept pace with their \ndevelopment. For example, legal boundaries to protect confidentiality and integrity of data are of \nconcern to security professionals.\n \n \n \n \n \n Security Auditing \n \n \n \n \n The field of security auditing may be broadly classified into two types, internal and external. Both \nrely on the independent appraisal function. However, their scope is different.\n \n \n \n \n \n \nInternal auditors typically work for a given organization. External auditors do not. They are typically \nCertified Public Accountants (CPAs) or Chartered Accountants (CAs) hired to perform an \nindependent audit, usually of a company's financial statements. Their primary concern is to evaluate \nthe fairness of the statements.\n \n \n \n \n \n \nThe scope of internal auditing is typically broader. Internal auditors are concerned not only with \nsafeguarding organizational assets but also with promoting operational efficiency. Their concern is \nthat the company have adequate controls and that the procedures used are cost-efficient as well as \neffective. Internal auditors typically report to top management or to the audit committee of the board \nof directors.\nThe information technology (IT) function can be audited by both internal and external auditors. \nWhoever does it must have expertise in both financial auditing and computer technology. IT auditors \ncan help your company assess the risks related to the use of computer technology.\n \n \n \n \n \n \nCoordinating the activities of IT auditors and financial auditors can enhance audit efficiency. IT \nauditors might train and guide non-IT auditors in IT procedures and methods. Or IT auditors might \nmanage the computer system during financial audit processing.\n \n \n \n \n \n \nIT auditors recommend appropriate controls, which tend to be more complex than controls in manual \nsystems. Specialized computer audit techniques must be used in highly automated environments. IT \nauditing:\n \n \n \n \n \n • Uses technological tools and expertise \n"
        },
        {
            "page_number": 251,
            "text": " \n \n \n \n • Evaluates the adequacy and effectiveness of the control systems \n \n \n \n \n • Assesses technology-related risks \n \n \n \n \n \nIT auditors review systems to ensure that they meet quality criteria, assessing their compliance with \nthe organization's systems development methodology. IT auditors may review proposed \nenhancements to computer systems to evaluate whether the system contains adequate controls. Their \nparticipation in the development process avoids the need to modify systems after they've been \nimplemented, a costly and difficult process. It may in fact sometimes be virtually impossible to \ncorrectly modify a system.\n \n \n \n \n \n While IT auditors may evaluate systems in process, the IT auditor must not assume any operational \nresponsibility. The auditors must remain independent and objective.\n \n \n \n \n \n Data centers serve the information needs of an organization. IT auditors typically review the \nfollowing aspects of data centers:\n \n \n \n \n \n • Systems development standards \n \n \n \n \n • Efficiency and effectiveness of operating and administrative procedures \n \n \n \n \n • Library control procedures \n \n \n \n \n • Structure of the data center \n \n \n \n \n • Network system \n \n \n \n \n • Backup controls \n \n • Contingency planning and disaster recovery \n"
        },
        {
            "page_number": 252,
            "text": " \n \n \n \n • Personnel practices \n \n \n \n \n • Security \n \n \n \n \n To perform their duties IT auditors must: \n \n \n \n \n • Keep current with state-of-the-art technologies \n \n \n \n \n • Understand how to use the technology to support business functions \n \n \n \n \n • Use audit tools specific to the technology needs of the organization \n \n \n \n \n IT auditors often review applications systems, with special attention to programmed control \nprocedures such as edit checks and exception reporting. Among their duties, IT auditors:\n \n \n \n \n \n • Evaluate the risks and controls associated with technology \n \n \n \n \n • Support other auditing functions during financial,operational, or compliance audits \n \n \n \n \n • Evaluate corporate computer policy and security standards \n \n \n \n \n The IT auditor is also responsible for determining whether controls are adequate and whether: \n \n \n \n \n • Transactions are processed accurately and completely \n \n \n \n \n • Transactions are properly authorized \n \n"
        },
        {
            "page_number": 253,
            "text": " \n \n \n • Errors and omissions are prevented or detected \n \n \n \n \n • Duties are segregated \n \n \n \n \n • Jobs are completed in a timely fashion \n \n \n \n \n IT auditors often assist external and operational internal auditors in: \n \n \n \n \n • Collecting, extracting, and analyzing data \n \n \n \n \n • Reviewing and testing internal controls \n \n \n \n \n • Investigating exceptions \n \nTransaction Trail \n \n \n \n \n \nThe transaction trail or audit trail allows the auditor to trace a transaction back to its origins. All \nattempts to gain access to the system should be logged chronologically. Unusual activity and \nvariations from established procedures should be identified and investigated.\n \n \n \n \n \n \nMany significant transactions occur inside the computer, so they aren't visible or directly observable. \nThe transaction trail provides information about additions, deletions, or modifications to data within \nthe system. An effective audit trail allows the data to be retrieved and certified. Audit trails will give \ninformation such as:\n \n \n \n \n \n • Date and time of the transaction \n \n \n \n \n • Who processed the transaction \n \n \n \n \n • At which terminal the transaction was processed \n"
        },
        {
            "page_number": 254,
            "text": " \n \n \n \n \nMaintaining audit trails is more difficult in an electronic environment. In a paper-based system, for \ninstance, a physical purchase order is prepared, typically in triplicate. At each state, paper work is \ndone and a physical trail established. Such a trail is normally absent in electronic transactions.\n \n \n \n \n \n Your computer software should be designed to provide an audit trail. Most commercial software \npackages have at least some audit trail capability.\n \n \n \n \n \n \nComputer security risks affect an organization's internal control structure, which in turn affects the \nability to audit the entity. Computer processing reduces human involvement, centralizes data, and \nmay eliminate segregation of duties. Centralizing data makes it possible to introduce higher quality \ncontrols over operations. However, due to reduced human involvement and less segregation of duties, \nauditors must use great care in evaluating the electronic data processing department.\n \n \n \n \n \n \nAdequate segregation of duties is crucial. No person should be in a position both to perpetrate and to \nconceal errors in the normal course of business. For example, there should be a division of duties \nbetween:\n \n \n \n \n \n • Programmers, librarians, and operators: Different individuals should develop computer \napplications, have custody of programs and data files, and operate applications.\n \n \n \n \n \n • Data processing personnel, users, and control personnel  \n \n• Individuals authorizing changes in program logic or data and those coding the changes  \n \n \n \n \n EDI and Electronic Contracting \n \n \n \n \n \nElectronic data interchange (EDI) systems are on-line systems where computers automatically \nperform transactions such as order processing and generating invoices. EDI allows trading partners to \nexchange electronic data faster, cheaper, and more accurately. The messages are structured in a \nprearranged format to facilitate automatic computer processing. The electronic messages generally \nresult in a legally binding contract.\n \n \n \nAlthough EDI can reduce costs, it can adversely affect an auditor's ability to do her job. EDI \ntransactions go through several systems. Electronic records and audit trail must be maintained \n"
        },
        {
            "page_number": 255,
            "text": "  \nthroughout. Any data used for EDI needs to be translated into a standardized format; the translation \nsoftware must maintain the audit trail. Any communication sent over the network must be accounted \nfor by communication software. Data translated into internal format by the recipient's translation \nsoftware must be tracked. Finally, the data is used by the recipient's application software.\n  \n \n \n \n \n \nIn an EDI environment, a weakness in any system can create problems not just for that entity but also \nfor its trading partners. Therefore, each function at each stage must be reviewed and appropriate \ncontrols incorporated.\n \n \n \n \n \n \nThe American Institute of Certified Public Accountants (AICPA) has issued control techniques to \nensure the integrity of an EDI system. The AICPA recommends that controls over accuracy and \ncompleteness at the application level include:\n \n \n \n \n \n • Checks on performance to determine compliance with industry standards \n \n \n \n \n • Checks on sequence numbering for transactions \n \n \n \n \n • Prompt reporting of irregularities \n \n \n \n \n • Verification of adequacy of audit trails \n \n \n \n \n • Checks of embedded headers and trailers at interchange, functional group, and transaction set level \nControl techniques at the environmental level include:: \n• Quality assurance review of vendor software \n \n \n \n \n • Segregation of duties \n \n \n \n \n • Ensuring that software is virus-free \n \n \n \n \n • Procuring an audit report from the vendor's auditors \n \n \n \n \n"
        },
        {
            "page_number": 256,
            "text": " • Evidence of testing \n \n \n \n \n To ensure that all EDI transactions are authorized, the AICPA suggests these controls: \n \n \n \n \n • Operator identification code \n \n \n \n \n • Operator profile \n \n \n \n \n • Trading partner identifier \n \n \n \n \n • Maintenance of user access variables \n \n \n \n \n • Regular changing of passwords \n \n \n \n \n \nNot all electronic messages result in electronic contracting. For example, messages with purely \ninformational content don't. Nor do intra-firm messages; the law generally distinguishes between \nintra-firm and inter-firm communications.\n \n \n \n \n \n Yet electronic contracting occurs routinely. Examples of electronic offer and acceptance modes are: \n \n \n \n \n • Purchase orders \n \n \n \n \n • Invoices \n \n \n \n \n • Payments \n \n \n \n \n • Solicitation and submission of bids \n \n \n \n \n"
        },
        {
            "page_number": 257,
            "text": " • Filing documents with the government \n \n \n \n \n • Advertising goods and services \n \n \n \n \n Trading partner agreements or EDI agreements are essential to electronic contracting. These \nagreements:\n \n \n \n \n \n • Clarify each party's rights and obligations \n \n• Specify the risk and liability of each party \n \n \n \n \n • Help avoid misunderstandings \n \n \n \n \n \nA trading partner agreement gives the parties the legal right to enforce it. These agreements affect \npartners only; they don't cover third parties, such as VANs. The legal and EDI communities (e.g., the \nAmerican Bar Association and the EDI Council of Canada) have drafted several model trading \npartner agreements. There are also model agreements for specific industries and countries. Though \nthe models provide a fair and balanced contract, most businesses will want to customize them to their \nown needs.\n \n \n \n \n \n \nA trading partner agreement should state the intent of the parties to transact business electronically. It \nshould specify whether all trade or only a specified portion of the trade between the two parties is \ncovered by the agreement. It should clearly specify which transaction sets will constitute a legally \nenforceable offer and acceptance and how electronic payments will be made. The parties should \nacknowledge that they will not repudiate the validity, integrity, or reliability of EDI transactions and \nwill consider them the equivalent of paper-based transactions.\n \n \n \n \n \n The parties must agree on the time and place of receipt of EDI communications. There are several \npossibilities. Receipt may take place when:\n \n \n \n \n \n • A message is sent by the sender's computer system \n \n \n \n \n • A message is received by the receiver's computer system \n \n"
        },
        {
            "page_number": 258,
            "text": " \n \n \n • A message is received at the receiver's mailbox on a VAN or other third party computer system \n \n \n \n \n • An acknowledgement of receipt is sent by the recipient \n \n \n \n \n • An acknowledgement of receipt is received by the sender. \n \n \n \n \n \nAcknowledgements, typically used to verify communications, provide proof of a transaction's \nintegrity and authority. Use cryptographic methods whenever possible, especially when the \nauthenticity of a transaction is crucial. Sometimes electronic signatures are used to verify the integrity \nof a message; typically these signatures are created cryptographically. However, a signature doesn't \nhave to be encrypted; any symbol or party's name may be considered sufficient as a signature for \npurposes of offer and acceptance as long as the EDI system is trustworthy. The location of the \nsignature in messages should be agreed on in advance and be as uniform as possible.\nSecurity considerations need special attention when you're drafting a trading partners agreement. \nSecurity provides confidence that transactions are authentic. It's needed to ensure that the transactions \nremain confidential. The EDI system's security is essential in determining whether electronic \ncontracting is legally enforceable.\n \n \n \n \n \n \nFrom a legal perspective trading partner agreements generally require commercially reasonable \nsecurity—but the definition of commercially reasonable is vague, differing from industry to industry. \nFor example, banking will require a much higher level of security than on-line retailing of software.\n \n \n \n \n \n \nThe trading partner agreement should discuss the security responsibilities of each party. For instance, \nto what extent is one party responsible for ensuring the security of its trading partner? What actions \nwill be taken if security is breached? Basic EDI security risks include:\n \n \n \n \n \n • Access violations \n \n \n \n \n • Message modifications \n \n \n \n \n • Interruptions or delays \n \n \n \n \n • Message rerouting \n \n"
        },
        {
            "page_number": 259,
            "text": " \n \n \n • Message repudiation \n \n \n \n \n \nWithout access controls, an unauthorized individual could initiate a transaction by pretending to be \nan authorized trading partner. Fictitious purchase orders may be sent or fictitious payments made. \nThe reliability and integrity of the EDI system break down without appropriate access controls. \nGreater security is achieved by combining several access control techniques. Most common are \ntechniques based on:\n \n \n \n \n \n • Something a person knows, such as a password \n \n \n \n \n • Something a person possesses, such as a magnetic card or a token \n \n \n \n \n • Some unique attribute of a person, such as a fingerprint, a voice print, or a retinal pattern \n \n \n \n \n \nUnauthorized individuals may intentionally modify electronic messages. Messages may also be \nmodified unintentionally through hardware, software, or transmission error. Authentication of \nmessages is a major concern, especially with respect to repudiation of a transaction. Irrevocable proof, \nsuch as a digital signature, minimizes the risk of repudiation.\nAuditing Contingency Plans \n \n \n \n \n \nData processing serves the information needs of most organizations and its survival in a disaster is \noften critical. Auditors are therefore especially concerned about the viability of the disaster recovery \nplan.\n \n \n \n \n \n \nYour contingency plan is a valuable document that needs to be audited like any other asset. An \nauditor is responsible for investigating, evaluating, and verifying controls, which may reduce the \nrisks associated with various types of disasters.\n \n \n \n \n \n \nAvoiding disaster is always preferable to recovering from disaster. An effective contingency plan, \naudited regularly, can sometimes help prevent a disaster. For example, sabotage can be prevented, \nbut when it isn't, saboteurs can often hide their activities. Auditing may help detect the crime.\n \n \n \n \n \n Controls \n"
        },
        {
            "page_number": 260,
            "text": " \n \n \n \n \nManagement is responsible for installing and maintaining controls, which are used to reduce the \nprobability of attack on computer security. The auditor is responsible for determining whether the \ncontrols are adequate and whether they are being complied with. As more controls are incorporated, \nthe operating costs tend to increase. Some types of controls are discussed below.\n \n \n \n \n \n \nDeterrent controls are used to encourage compliance with controls. Deterrents are relatively \ninexpensive to implement. Since their purpose is to deter crime, however, it's often difficult to \nmeasure their effectiveness. Deterrent controls are meant to complement other controls; they're not \nsufficient by themselves.\n \n \n \n \n \n \nPreventive controls are the first line of defense. Their purpose is to thwart perpetrators trying to gain \naccess to your system. They also help prevent unintentional errors from affecting the system and the \ndata. For example, pre-numbered documents ensure that there isn't a failure in recording a \ntransaction. Data validation and review procedures prevent the recording of an incorrect or \nincomplete transaction, or duplicates of a transaction.\n \n \n \n \n \n \nDetective controls help detect an error once a system has been violated. These controls prevent the \nerror from harming the system. Their purpose is to focus attention on the problem. For example, a bait \nfile will identify unauthorized use: A dummy (non-existent) record is processed. Or there may be a \ncomparison between standard run time and actual run time to spot possible misuse.\nCorrective controls reduce the impact of the threat after a loss has occurred. They aid in recovering \nfrom damage or in reducing the effect of damage. Corrective controls may provide data for recovery \nprocedures. For instance, lost information on floppies may be restored with utility programs.\n \n \n \n \n \n \nApplication controls are built into software to deter or detect irregularities and minimize errors. \nApplication controls typically include input, processing, change, testing, output, and procedural \ncontrols.\n \n \n \n \n \n • Input controls are used to ensure that transactions are authorized, processed correctly, and \nprocessed only once. Input controls may reject, correct, or resubmit data.\n \n \n \n \n \n • Processing controls ensure that transactions entered into the system are valid and accurate, that \nexternal data is not lost or altered, and that invalid transactions are reprocessed correctly.\n \n \n \n \n \n • Change controls safeguard the integrity of the system. Standards are established for making \nmodifications. All changes must be documented.\n \n"
        },
        {
            "page_number": 261,
            "text": " \n \n \n \n • Test controls ensure that a system is reliable before it becomes operational. An example would be \nthe processing of limited test data when using the new system.\n \n \n \n \n \n • Output controls authenticate other controls. They verify that authorized transactions are processed \ncorrectly. Random comparisons can be made of output to input to verify correct processing.\n \n \n \n \n \n • Procedural controls help reduce the probability of processing mistakes and assure continued \nfunctioning if a failure does occur.\n \n \n \n \n \n Audit Software \n \n \n \n \n \nComputer-assisted audit techniques are used extensively; large quantities of electronically stored data \ncan be tested quickly and accurately using audit software. Query languages can create ad hoc reports \nand perform a variety of audit procedures. Audit software functions typically:\n \n \n \n \n \n • Appraise reasonableness (e.g., accuracy of sales discounts) and trends, including aging analysis \n \n \n \n \n • Check for duplicate invoices or payments \n \n \n \n \n • Compare financial data on different files for consistency \n \n• Analyze and report data \n \n \n \n \n • Extract data from computer files \n \n \n \n \n • Provide exception reporting (e.g., excessive inventory balances or an unusual employee salary) \n \n \n \n \n • Do field comparisons to find errors or inconsistencies \n \n"
        },
        {
            "page_number": 262,
            "text": " \n \n \n • Detect fraud \n \n \n \n \n • Recalculate balances \n \n \n \n \n • Do statistical sampling and analysis \n \n \n \n \n Generalized audit software can identify errors keyed into accounting software. It provides cost \nsavings over custom software because most audits involve similar activities, such as:\n \n \n \n \n \n • Analyzing data for unusual or erroneous values \n \n \n \n \n • Analyzing or comparing data stored in two or more separate but logically related files \n \n \n \n \n • Generating and formatting reports \n \n \n \n \n • Recalculating balances \n \n \n \n \n • Selecting a sample \n \n \n \n \n • Stratifying data \n \n \n \n \n • Testing transactions \n \n \n \n \n Examples of popular generalized audit software packages are ACL Software (888.669.4225 or \nwww.acl.com) and Interactive Data Extraction and Analysis (IDEA) (888.641.2800 or www.cica.ca).  \n \n \n \n \n \nYou can customize audit software if the generalized variety doesn't fulfill the needs of your \norganization. Specialized audit software is in any case available for specific industries, such as \nbanking, health care, entertainment, or insurance.\n \n"
        },
        {
            "page_number": 263,
            "text": " \n \n \n \n Legal Liability in Security Management \n \n \n \n \n \nComputer security law is a relatively new field, and the legal establishment has yet to reach a \nconsensus on a host of important issues. Nonetheless, you can incur substantial legal liability by not \nmaintaining adequate security. Management may be held personally liable in certain instances. Be \nparticularly careful to protect privacy and other personal rights, which are easily violated due to a lack \nof computer security.\nLegislation  \n \n \n \n \n \nThe Financial Privacy Act of 1978 was passed to protect private information. The 1987 Computer \nSecurity Act further protects privacy and increases government computer security requirements. This \nact states that ''improving the security and privacy of sensitive information in the federal computer \nsystems is in the public interest.\" The private sector also has to ensure that confidential information is \nkept private.\n \n \n \n \n \n \nThe 1987 Computer Security Act gave the National Institute of Standards and Technology (NIST) \nresponsibility to develop cost-effective standards to protect confidential information in federal \ndatabases. Private companies can use NIST's work as a model for their own standards.\n \n \n \n \n \n \nOnce information is determined to be sensitive, it should be verified for accuracy before being put \ninto a database and given whatever protection is necessary to keep it confidential. Ask yourself the \nfollowing questions:\n \n \n \n \n \n • How should this information be classified? \n \n \n \n \n • How can we ensure the accuracy of the information? \n \n \n \n \n • How can we protect sensitive or confidential information? \n \n \n \n \n The Computer Fraud and Abuse Act of 1986 makes any unauthorized use (copying, damaging, \nobtaining database information, etc.) of computer hardware or software across state lines a crime.\n \n \n \n"
        },
        {
            "page_number": 264,
            "text": "  \nThe Foreign Corrupt Practices Act of 1977 applies to companies whose securities are registered or \nfiled under the Securities Exchange Act of 1934. It requires these companies to keep accurate \naccounting records and to maintain a system of internal control.\n  \n \n \n \n \n The Counterfeit Access Device and Computer Fraud and Abuse Act of 1984 covers unauthorized \nretrieval of data from the computer files of a financial institution or a credit reporting agency.\n \n \n \n \n \n The Electronic Communications Privacy Act of 1986 prohibits anyone from intercepting information \nbeing transmitted electronically.\n \n \n \n \n \n Negligence and Due Care  \n \n \n \n \n \nYou can incur liability for security violations in a variety of situations, ranging from programming \nerrors to violations of civil or criminal law. The standard to avoid liability is due care. For instance, \nyou may have properly designed and coded a computer program but because security is inadequate, a \nsaboteur places a logic bomb that causes the program to crash. The organization and its senior \nmanagers may be held personally liable for any damages arising from the crash if negligence in \nsecuring the program can be proved. Such damages may be significant if, for example, they cause a \nloss in market price of the stock or, worse, if human life is affected, as in the crash of a medical \ndiagnosis system.\nNIST has published several national standards for computer security. They cover: \n \n \n \n \n • Automated password generators \n \n \n \n \n • Contingency planning \n \n \n \n \n • Data encryption \n \n \n \n \n • Digital signatures \n \n \n \n \n • Electrical power for computer facilities \n \n \n \n \n • Key management \n"
        },
        {
            "page_number": 265,
            "text": " \n \n \n \n • Password usage \n \n \n \n \n • Physical security and risk management \n \n \n \n \n • User authentication techniques \n \n \n \n \n \nThe Department of Defense (DOD) publishes the Rainbow Series of booklets to help developers, \nevaluators, and users of trusted systems. They include information on networks, databases, and other \nproblems with distributed computer systems. The governments of Britain, the Netherlands, France, \nand Germany have themselves jointly issued detailed Information Technology Security Evaluation \nCriteria (ITSEC).\n \n \n \n \n \n \nConsider using these standards in managing your own computer security. If you don't, in a lawsuit \nalleging breach of security, the plaintiff may use your failure to follow recognized standards to prove \nthat you've been negligent, even if your organization wasn't required to follow them.\nAppendix— \nSecurity Software1 \n \n \n \n \n The following software may be used to improve the security of computers running on the Unix \noperatfing system. Several of these programs also have non-Unix versions.\n \n \n \n \n \n A \n \n \n \n Abacus Sentry  \ndetects the use of a port scanner in real time. Availability: www.psionic.com/abacus/portsentry/  \n \n \n \n Anonftpd  \nis a read-only anonymous FTP server. Availability: anonymous ftp at koobera.math.uic.edu  \n \n \n \n \nArgus  \nis a powerful tool for monitoring networks. It provides tools for analysis of network activity that can \nbe used to verify the enforcement of network security policies, network performance analysis, and \n \n"
        },
        {
            "page_number": 266,
            "text": "more. Availability: anonymous ftp at ftp.sei.cmu.edu\n \n \n \n \nArpwatch  \nis an Ethernet monitor program that keeps track of Ethernet/IP address pairings. Availability: \nanonymous ftp at ftp.ee.lbl.gov\n \n \n \n \n \n C \n \n \n \n \nCOPS (Computer Oracle and Password System)  \nidentifies security risks. It checks for empty passwords in /etc/passwrd, world-writable files, \nmisconfigure anonymous ftp, and many others. Availability: anonymous ftp at ftp.cert.org\n \n \n \n \n \nCourtney  \nidentifies the use of SATAN on a subnet. The program tcpdump (see below) will also be needed to \nrun Courtney. Availability: anonymous ftp at ciac.llnl.gov\n \n \n \n \n Crack  \nis a password cracker. Availability: www at www.users.dircon.co.uk  \n \n \n 1 The material in this appendix was taken from http://www.alw.nih.gov/Security/prog-full.html.\n \n \nCrack lib  \nchecks plain text words against those generated by Crack. Availability: anonymous ftp at \ncoast.cs.purdue.edu\n \n \n \n \n \n D \n \n \n \n \nDeslogin  \nprovides a more secure method for remote log-in than Telnet or rlogin in untrusted networks. \nDeslogin encrypts the connection using DES. Availability: anonymous ftp at ftp.uu.net\n \n \n \n \n \nDig  \nis a network utility that queries domain name servers; similar to nslookup but more flexible. \nAvailability: anonymous ftp at venera.isi.edu\n \n"
        },
        {
            "page_number": 267,
            "text": " \n \n \n Drawbridge  \nis a powerful bridging filter package. Availability: anonymous ftp at net.tamu.edu  \n \n \n \n \n F \n \n \n \n \nFping  \nis an efficient way to test whether a large number of hosts are up. Availability: anonymous ftp at \nslapshot.stanford.edu\n \n \n \n \n \n I \n \n \n \n \nIcmpinfo  \ndisplays unusual ICMP messages received by a host; it can be used to detect suspicious network \nactivity. Availability: anonymous ftp at hplyot.obspm.fr\n \n \n \n \n \nISS  \nchecks hosts within a specified range of IP addresses for security vulnerabilities in send mail, \nanonymous FTP setup, NFS, and many more. Availability: anonymous ftp at info.cert.org\n \n \n \n \n \nIPACL  \nfilters incoming and outgoing TCP and UDP in a SVR4/386 kernel. Availability: anonymous ftp at \nftp.win.tue.nl\n \n \n \n \n \n K \n \n \n \n \nKerberos  \nis an authentication system used to protect unsecure networks (export restricted). Availability: WWW \nat web.mit.edu\n \n \n \n \n \nKlaxon  \nis a daemon used to identify the use of port scanners like ISS and SATAN. Availability: anonymous \nftp at ftp.eng.auburn.edu\n \n"
        },
        {
            "page_number": 268,
            "text": " \n \n \n \n L \n \n \n \n \nL6  \nprovides a flexible and intelligent interface for periodic integrity checks of data using Perl. \nAvailability: www.pgci.ca/16.html\n \n \n \n \n \nLogdaemon  \nis a replacement for system ftp, rlogin, rexec, rshdaemons, and login program that has added security \nfeatures, such as login in failures and S/Key one-time password support. Availability: anonymous ftp \nat ftp.win.tue.nl\n \n \n \n \n \nLogsurfer  \nanalyzes any text-based log files on the fly, using contexts, and executes a corresponding action. \nAvailability: anonymous ftp at ftp.cert.dfn.de\n \n \n \n \n Lsof  \ndisplays all open files on a Unix system. Availability: anonymous ftp at vic.cc.purdue.edu\nM \n \n \n \n \nMangle  \nis a utility that either checks existing passwords for weakness or forces users to choose good \npasswords. Availability: anonymous ftp at ftp.informatik.uni-erlangen.de\n \n \n \n \n \nMerlin  \nis an interface to five popular security packages to make it easier to analyze and manage the data. \nAvailability: anonymous ftp at ciac.llnl.gov\n \n \n \n \n MD5  \nis a hash function using the authenticity of the file. Availability: anonymous ftp at rsa.com  \n \n \n \n \nMIME Object Security Services (MOSS)  \nis an extension of MultiPurpose Internet Mail Extensions (MIME) that provides authentication, \nintegrity, and confidentiality to an e-mail message (export restricted). Availability: anonymous ftp at \nftp.tis.com\n \n"
        },
        {
            "page_number": 269,
            "text": " \n \n \n \n N \n \n \n \n \nNetlog  \nis network logging and monitoring of all TCP and UDP connections on a subnet; it includes tools to \nanalyze the output. Availability: anonymous ftp at net.tamu.edu\n \n \n \n \n \nNetwork Security Scanner (NSS)  \nis a Perl that scans one host on subnet or an entire subnet for simple security problems. Availability: \nanonymous ftp at jhunix.hcf.jhu.edu\n \n \n \n \n \nNFSWatch  \nmonitors NFS request and measures response time for each RPC. Availability: anonymous ftp at \ncoast.cs.purdue.edu\n \n \n \n \n \nNmap (Network Mapper)  \nis a utility for stealth port scanning of large networks (see Syn for tracking these types of scans). \nAvailability: www.insecure.org/nmap\n \n \n \n \n \nNpasswd  \nis a replacement for the system passwd command that does not accept poor passwords. Availability: \nanonymous ftp at ftp.cc.utexas.edu\n \n \n \n \n \n O \n \n \n \n \nOPIE  \nprovides the ability to generate and use one-time passwords. (Related tools are also available for \nWindows, DOS, and Mac.) Availability: anonymous ftp at ftp.nrl.navy.mil\n \n \n \n \n \nOsh  \nis a restricted C shell that allows the administrator to control access to files and directories and to \nprovide logging. Availability: anonymous ftp at ftp.c3.lanl.gov\n \n \n \n"
        },
        {
            "page_number": 270,
            "text": "  P  \n \n \n \n \nPasswd+  \nis a proactive password checker that replaces the system passwd command and enforces the selection \nof good passwords. Availability: anonymous ftp at ftp.dartmouth.edu\n \n \n \n \n \nPGP (Pretty Good Privacy)  \nprotects documents such as e-mail from unauthorized reading using public-key encryption (some \nversions are export-restricted). Availability: US and Canada: anonymous ftp at www.eff.org or via web \nform; international: anonymous ftp at ftp.ifi.uio.no\nPinglogger  \ndetects and logs ICMP echo requests. Availability: WWW at www.students.uiuc.edu  \n \n \n \n Portmapper  \nreduces vulnerabilities and disallows proxy access. Availability: anonymous ftp at ftp.win.tue.nl  \n \n \n \n \nRIPEM (Riordan's Internet Privacy Enhanced Mail)  \nimproves the security of e-mail by verifying the authenticity of the message sender, among others \nthings (export restricted). Availability: anonymous ftp at ripem.msu.edu\n \n \n \n \n \n R \n \n \n \n \nRpcbind  \nprevents intruders from bypassing NFS export restrictions. Availability: anonymous ftp at \nftp.win.tue.nl\n \n \n \n \n \nRscan  \nis an extensible network scanner that checks for common network problems and SGI-specific \nvulnerabilities. Availability: anonymous ftp at ftp.vis.colostate.edu\n \n \n \n \n \n S \n \n \n \n SATAN  \nis a program that gathers information about the network, such as the type of machines, the services \n \n"
        },
        {
            "page_number": 271,
            "text": "available on these machines, and potential security flaws. Availability: anonymous ftp at \nftp.win.tue.nl; see also wzv.win.tue.nl for a list of mirror sites.\n \n \n \n \nScan-Detector  \ndetermines when an automated scan of UDP/TCP ports is being done on a host running this program. \nLogs to either syslog or strerr. Availability: anonymous ftp at coast.cs.purdue.edu\n \n \n \n \n \nSendmail  \na replacement for the system sendmail, this version includes all the latest patches. Availability: \nanonymous ftp at ftp.cs.berkeley.edu\n \n \n \n \n \nSendmail Wrapper  \nprovides limited protection against local sendmail attacks. Availability: anonymous ftp at \nftp.auscert.org.au\n \n \n \n \n \nShadow  \nincludes everything necessary to use the shadow password file. Availability: anonymous ftp at \nftp.cs.widener.edu\n \n \n \n \n \nSimple Socksd  \nis another implementation of Version 4 SOCKS protocol that is fast, easy to compile, and simple to \nconfigure. Availability: http at Simple SOCKS Daemon\n \n \n \n \n \nSKey  \ngenerates one-time passwords to gain authenticated access to computer hosts. Availability: anonymous \nftp at thumper.bellcore.com or coast.cs.purdue.edu\nSKIP (Simple Key-Management for Internet Protocols)  \nadds privacy and authentication at the network level. Availability: US and Canada: via web form; \ninternational: anonymous ftp at ftp.elvis.ru\n \n \n \n \n \nSmrsh  \nis a restricted shell for sendmail to limit the number of programs that can be executed by sendmail. \nAvailability: anonymous ftp at ftp.nec.com\n \n \n \n \n Socks is a package that allows various Internet services, such as gopher, ftp, and Telnet, to be used \nthrough a firewall. Availability: anonymous ftp at ftp.nec.com\n \n \n"
        },
        {
            "page_number": 272,
            "text": " \n \n \nSSH (Secure Shell)  \nis an advanced version of rlogin, rsh, and rcp that provides RSA authentication and encryption of \ncommunications as well as many other security improvements (export restricted). Availability: \nanonymous ftp at ftp.cs.hut.fi\n \n \n \n \n \nSTEL  \na system replacement for Telnet, provides strong mutual authentication and encryption. Availability: \nanonymous ftp at idea.sec.dsi.unimi.it\n \n \n \n \n \nStrobe  \ndisplays all active listening TCP ports on remote hosts. It uses an algorithm that efficiently uses \nnetwork band-width. Availability: anonymous ftp at suburbia.apana.org or minnie.cs.adfa.oz.au\n \n \n \n \n \nSudo  \nallows a system administrator to give limited root privileges to users and logs their activities. This \nversion is also known as CU-sudo. Availability: anonymous ftp at ftp.cs.colorado.edu\n \n \n \n \n \nSwatch  \nmonitors and filters log files and executes a specified action, depending on the pattern in the log. \nAvailability: anonymous ftp at ee.stanford.edu\n \n \n \n \n \nSyn  \nis a Perl utility for tracking steal port scanning. Availability: anonymous ftp at Syn \nwww.bigfat.net/unix.html\n \n \n \n \n \n T \n \n \n \n \nTCP Wrapper  \nallows a Unix system administrator to control access to network services; it also provides logging \ninformation of wrapped networks services that may be used to prevent or monitor network attacks. \nAvailability: anonymous ftp at ftp.win.tue.nl\n \n \n \n \n \nTcpdump  \ncaptures and dumps protocol packets to monitor or debug a network. Availability: anonymous ftp at \nftp.ee.lbl.gov\n \n"
        },
        {
            "page_number": 273,
            "text": " \n \n \n \nTcpr  \nis a set of Perl scripts that forwards ftp and Telnet commands across a firewall. Availability: \nanonymous ftp at ftp.alantec.com\nTiger  \nchecks for known security vulnerabilities of Unix workstations; it's similar to COPS with many \nextensions. Availability: anonymous ftp at net.tamu.edu\n \n \n \n \n \nTIS Firewall Toolkit  \nis a software package to build and maintain a system to protect a network from unwanted activities. \nAvailability: anonymous ftp at ftp.tis.com\n \n \n \n \n \nTraceroute  \ntraces the route IP packets take from the current system to a destination system. Availability: \nanonymous ftp at ftp.psc.edu\n \n \n \n \n Tripwire  \nmonitors for changes in system binaries. Availability: anonymous ftp at coast.cs.purdue.edu  \n \n \n \n \nTTY-Watcher  \nmonitors, logs, and interacts with all of the TTY on a system. Availability: anonymous ftp at \ncoast.cs.purdue.edu\n \n \n \n \n \n W \n \n \n \n \nWu-ftpd  \nis a replacement FTP server for Unix systems whose features include extensive logging and a way of \nlimiting the number of FTP users. Availability: anonymous ftp at wuarchive.wustl.edu\n \n \n \n \n \n X \n \n \n \n \nXinetd  \nis a replacement for inetd that has extensive logging and access control capabilities for both TCP and \nUDP services. Availability: anonymous ftp at qiclab.scn.rain.com\n \n"
        },
        {
            "page_number": 274,
            "text": "  \n \n \n \n \nXp-BETA  \nis an application gateway for the XII protocol that uses Socks or CERN WWW Proxy. Availability: \nanonymous ftp at ftp.mri.co.jp\n \n \n \n \n \n Y \n \n \n \n \nYPX  \nis a utility used to retrieve an NIS map from a host running the NIS daemon. Availability: anonymous \nftp at ftp.uu.net or WWW server at mls.saic.com\n"
        }
    ]
}