Hackers Toolkit v2.0/Hackers_Toolkit_v2.0.iso/HTML/archive/Rainbow/ncsc/aquancsc.txt

Text File | 1999-11-04 | 253KB | 3,600 lines

FOREWORD

The National Computer Security Center is issuing A Guide to Understanding Security
Modeling in Trusted Systems as part of the "Rainbow Series" of documents produced by our
Technical Guidelines Program. In the Rainbow Series, we discuss, in detail, the features of the
Department of Defense Trusted Computer System Evaluation Criteria (DoD 5200.28-STD) and
provide guidance for meeting each requirement. The National Computer Security Center, through
its Trusted Product Evaluation Program, evaluates the security features and assurances of
commercially-produced computer systems. Together, these programs ensure that organizations are
capable of protecting their important data with trusted computer systems. Security modeling, in its
various forms, is an important component of the assurance of the trust of a computer system.

A Guide to Understanding Security Modeling in Trusted Systems is intended for use by
personnel responsible for developing models of the security policy of a trusted computer system.
At lower levels of trust, this model is generally the system"s philosophy of protection. At higher
trust levels, this also includes informal and formal models of the protection mechanisms within a
system. This guideline provides information on many aspects of security modeling, including the
process of developing a security policy model, security modeling techniques, and specific ways to
meet the requirements of the Department of Defense Trusted Computer System Evaluation
Criteria.

As the Director, National Computer Security Center, I invite your suggestions for revising this
document. We plan to review and revise this document as the need arises.

Patrick R. Gallagher, Jr.       October 1992

Director

National Computer Security Center

ACKNOWLEDGMENTS

Special recognition and acknowledgment for their contributions to this document are
extended to the following: Dr. James Williams, as primary author of this document; Barbara
Mayer, Capt. James Goldston, USAF, and Lynne Ambuel, for project management.

Special thanks are also given to the many evaluators, vendors, and researchers whose careful
reviews and suggestions contributed to this document, with special recognition for the advice of
David Bell, Daniel Schnackenberg, Timothy Levin, Marshall Abrams, Deborah Bodeau, Leonard
La Padula, Jonathan Millen, William L. Harkness, Thomas A. Ambrosi, Paul Pittelli, Dr. Michael
Sinutko, Jr, and COL Rayford Vaughn, USA.

TABLE OF CONTENTS

FOREWORD        i

ACKNOWLEDGMENT  ii

1.      INTRODUCTION    1

1.1 Background  1

1.2 Purpose     2

1.3 Control Objectives  3

1.3.1 The Assurance Objective   3

1.3.2 The Security Policy Objective     4

1.4 Historical Overview 5

1.5 Content and Organization    7

2.      OVERVIEW OF A SECURITY MODELING PROCESS 9

2.1 Security Models and Their Purpose   9

2.2 Security Modeling in the System Development Process 11

2.3 Identifying the Contents of a Security Model        14

2.3.1 Step 1: Identify Requirements on the External Interface   16

2.3.2 Step 2: Identify Internal Requirements    18

2.3.3 Step 3: Design Rules of Operation for Policy Enforcement  20

2.3.4 Step 4: Determine What is Already Known   21

2.3.5 Step 5: Demonstrate Consistency and Correctness   22

2.3.6 Step 6: Demonstrate Relevance     23

2.4 Presentation for Evaluation and Use 24

3       .SECURITY MODELING TECHNIQUES           27

3.1 Basic Concepts      27

3.1.1 Data Structures and Storage Objects       28

3.1.2 Processes and Subjects    29

3.1.3 Users and User Roles      31

3.1.4 I/O Devices       33

3.1.5 Security Attributes       35

3.1.6 Partially Ordered Security Attributes     37

3.1.7 Nondisclosure Levels      39

3.1.8 Unlabeled Entities and the Trusted Computing Base 40

3.2 Nondisclosure and Mandatory Access Control  41

3.2.1 External-Interface Requirements and Model 43

3.2.2 Information-Flow Model    45

3.2.3 Application to the Reference Monitor Interface    47

3.2.4 Access-Constraint Model   48

3.2.5 Tailoring the Models      51

3.3 Need-to-Know and Discretionary Access Control       53

3.3.1 DAC Requirements and Mechanisms   54

3.3.2 User Groups and User Roles        55

3.3.3 Sources of Complexity and Weakness        56

3.3.4 Tight Per-User Access Control     59

3.4 TCB Subjects-Privileges and Responsibilities        61

3.4.1 Identifying Privileges and Exemptions     62

3.4.2 Responsible Use of Exemptions     65

3.5 Integrity Modeling  66

3.5.1 Objectives, Policies, and Models  67

3.5.2 Error Detection and Recovery      68

3.5.3 Encapsulation and Level-Based Access Control      71

4. TECHNIQUES FOR SPECIFIC KINDS OF SYSTEM              5

4.1 Operating Systems   75

4.1.1Traditional Access Control Models  75

4.1.2 The Models of Bell and La Padula  77

4.2 Networks and Other Distributed Systems      78

4.2.1 Systems and Their Components      78
4.2.2 Model Structure and Content       80
4.2.3 Network Security Models   82
4.2.4 Individual Component Security Models      84
4.2.5 Underlying Models of Computation  85

4.3 Database Management Systems 86

4.3.1 System Structure  87
4.3.2 Integrity Mechanisms and Policies 89
4.3.3 Aggregation       90

4.3.4 Inference 91
4.3.5 Partially Automated Labeling      92

4.4 Systems with Extended Support for Label Accuracy    94

4.4.1 Factors Inhibiting Accuracy in Labeling   94
4.4.2 Floating Sensitivity Labels       95
4.4.3 Compartmented Mode Workstations (CMW)     95
4.4.4 The Chinese Wall Security Policy  96

5. MEETING THE REQUIREMENTS             99

5.1 Stated Requirements on the Security Model   99
5.2 Discussion of the B1 Requirements   100
5.3 Discussion of the B2 Requirements   103
5.4 Discussion of the B3 Requirements   104
5.5 Discussion of the A1 Requirements   105

APPENDIX A. SECURITY LEVELS AND PARTIALLY ORDERED SETS          107

A.l Terminology 107

A.2 Embeddings  109
A.3 Cartesian Products  109

 A.4 Duality    110

APPENDIX B. AVAILABLE SUPPORT TOOLS             113

B.1 FDM: the Formal Development Methodology     114

B.2 GVE: the Gypsy Verification Environment     115

APPENDIX C. PHILOSOPHY OF PROTECTION OUTLINE            17

APPENDIX D. SECURITY MODEL OUTLINE              21

APPENDIX E. GLOSSARY            127

REFERENCES              45

1.      INTRODUCTION

This document provides guidance on the construction, evaluation, and use of security policy
models for automated information systems (AIS) used to protect sensitive information whose
unauthorized disclosure, alteration, loss, or destruction must be prevented. In this context, sensitive
information includes classified information as well as unclassified sensitive information.

1.1     BACKGROUND

The National Computer Security Center (NCSC) was established in 1981 and acquired its
present name in 1985. Its main goal is to encourage the widespread availability of trusted AISs. In
support of this goal, the DoD Trusted Computer System Evaluation Criteria (TCSEC) was written
in 1983. It has been adopted, with minor changes, as a DoD standard for the protection of sensitive
information in DoD computing systems. [NCSC85] The TCSEC is routinely used for the
evaluation of commercial computing products prior to their accreditation for use in particular
environments. This evaluation process is discussed in Trusted Product Evaluations: A Guide for
Vendors. [NCSC90c]

The TCSEC divides AISs into four main divisions, labeled D, C, B, and A, in order of
increasing security protection and assurance. Divisions C through A are further divided into
ordered subdivisions referred to as classes. For all classes (C1, C2, B1, B2, B3 and A1), the TCSEC
requires system documentation that includes a philosophy of protection. For classes B1, B2, B3,
and A1, it also requires an informal or formal security policy model.

Although the TCSEC is oriented primarily toward operating systems, its underlying concepts
have been applied much more generally. In recognition of this, the NCSC has published a Trusted
Network Interpretation [NCSC87] and a Trusted Database Management System Interpretation.
[NCSC91] In addition, the NCSC also provides a series of guidelines addressing specific TCSEC
requirements, of which this document is an example.

1.2     PURPOSE

This guideline is intended to give vendors and evaluators of trusted systems a solid
understanding of the modeling requirements of the TCSEC and the Trusted Network Interpretation
of the TCSEC (TNI). It presents modeling and philosophy of protection requirements for each of
the classes in the TCSEC, describes possible approaches to modeling common security
requirements in various kinds of systems, and explains what kinds of mode is are useful in an
evaluation. It is intended for vendors, evaluators, and other potential builders and users of security
models.

This guideline discusses the philosophy of protection requirement, explains how it relates to
security modeling, and provides guidance on documentation relating to the system's philosophy of
protection and security policy model. It explains the distinction between informal and formal
security policy models as well as the relationships among application-independent models,
application-dependent security models, and model interpretations. It also explains which
techniques may be used to meet the modeling requirements at levels B1 through Al as well as the
advantages and disadvantages of the various modeling techniques. Finally, it discusses the specific
TCSEC modeling requirements.

This guideline also addresses human aspects of modeling. It describes how modeling captures
basic security requirements and how the security modeling effort leads to better systems and
provides a basis for increased assurance of security.

Finally, this guideline answers common questions about NCSC recommendations regarding
the construction of security models. Security policies and models are supplied by vendors in
response to customer needs and TCSEC requirements; they are not supplied by the NCSC or the
TCSEC. The TCSEC does, however, set minimum requirements in the areas of mandatory and
discretionary access control. The TCSEC does not require particular implementation techniques;
this freedom applies, by extension, to modeling techniques. Acceptability of a technique depends
on the results achieved. More specifically, a security model must provide a clear and accurate
description of the security policy requirements, as well as key ideas associated with their
enforcement. Moreover, one must be able to validate the model using assurance techniques
appropriate to the proposed evaluation class. Any vendor-supplied modeling technique which
appears to be compatible with the security modeling requirements will be seriously considered
during the course of a product evaluation.

Topics which are closely related to security modeling include formulation of security policy
objectives, design specification and verification, covert channel analysis, and implementation
correspondence analysis. These topics are addressed only to the extent necessary to establish their
influence on the security modeling process. All but the first of these topics are addressed in other
guidelines in this series. Security objectives for trusted systems traditionally include
nondisclosure, integrity, and denial of service. [cf NCSC88, AIS Security] However, the modeling
of denial of service is not addressed in this guideline, due to the lack of adequate literature on this
topic. This guideline is written with the understanding that security modeling will continue to
evolve in response to new policy objectives and modeling techniques associated with future trusted
system designs.

Reading and understanding this guideline is not sufficient to allow an inexperienced
individual to develop a model. Rather, he or she should read this document in conjunction with one
or more of the published models in the list of references. This guideline assumes a general
familiarity with the TCSEC and with computer security. A general text such as Building a Secure
Computer System [GASS87] may provide useful background reading. Additional mathematical
background needed for formal modeling can be found in general texts on mathematical structures
such as Introduction to Mathematical Structures [GALO89].

The approaches to security modeling presented in this document are not the only possible
approaches to satisfying TCSEC modeling requirements, but are merely suggested approaches.
The presentation of examples illustrating these approaches does not imply NCSC endorsement of
the products on which these examples are based. Recommendations made in this document are not
supplementary requirements to the TCSEC. The TCSEC itself (as supplemented by announced
interpretations [NCSC87, NCSC88a, NCSC88b, NCSC91]) is the only metric used by the NCSC
to evaluate trusted computing products.

1.3     CONTROL OBJECTIVES

The requirements of the TCSEC were derived from three main control objectives: assurance,
security policy, and accountability. The security modeling requirements of the TCSEC support the
assurance and security policy control objectives in systems rated B1 and above.

1.3.1                   THE ASSURANCE OBJECTIVE

The TCSEC assurance control objective states, in part,

"Systems that are used to process or handle classified or other sensitive information
must be designed to guarantee correct and accurate interpretation of the security policy and
must not distort the intent of that policy." [NCSC85, º 5.3.3]

Assurance in this sense refers primarily to technical assurance rather than social assurance. It
involves reducing the likelihood that security mechanisms will be subverted as opposed to just
improving people's confidence in the utility of the security mechanisms.

1.3.2           THE SECURITY POLICY OBJECTIVE

The TCSEC security policy control objective states, in part,

"A statement of intent with regard to control over access to and dissemination of
information, to be known as the security policy, must be precisely defined and
implemented for each system that is used to process sensitive information. The security
policy must accurately reflect the laws, regulations, and general policies from which it is
derived." [NCSC85, º 5.3.1]

 The security policy objective given in the TCSEC covers "mandatory security policy,"
"discretionary security policy," and "marking" objectives. The mandatory security policy
objective requires the formulation of a mandatory access control (MAC) policy that regulates
access by comparing an individual's clearance or authorization for information to the classification
or sensitivity designation of the information to be accessed. The discretionary access control
objective requires the formulation of a discretionary access control (DAC) policy that regulates
access on the basis of each individual's need-to-know, Finally, the marking objective gives labeling
requirements for information stored in and exported from systems designed to enforce a mandatory
security policy.

The access controls associated with security policies serve to enforce nondisclosure and
integrity. Nondisclosure controls prevent inappropriate dissemination of information. Integrity
controls prevent inappropriate modification of information.

The security policy control objective requires that the security policy accurately reflect the
laws, regulations, and general policies from which it is derived. These may include the revised
DoD Directive 5200.28, Security Requirements for Automated Information Systems (AISs).
[DOD88a] Section D of Directive 5200.28 gives policy relating to both integrity and
nondisclosure. It mentions safeguards "against sabotage, tampering, denial of service, espionage,
fraud, misappropriation, misuse, or release to unauthorized users." Enclosure 2 defines relevant
terms and, in particular, extends the concept of "user" to include processes and devices interacting
with an automated information system on behalf of users. Enclosure 3 gives general security
requirements, including data-integrity requirements. The revised 5200.28 does not use the MAC/
DAC terminology. Instead, it requires accurate marking of sensitive information and the existence
of an (undifferentiated) access control policy based partly on the identity of individual users. It
treats need-to-know as a form of least privilege.

Security policy relating to nondisclosure stems from Executive Order 12356 [REAG82] and,
in the case of DoD systems, from DoD 5200.1-R. [DOD86] Depending on the application, the
security policy may be constrained by a variety of other regulations as well. The collection,
maintenance, use, and dissemination of personal information is protected by DoD Directive
5400.11 [DOD82, º E.2] and by Public Law 100-503 [CONG88]. Classified and other sensitive
information needed for the conduct of federal programs is protected by the Computer Security Act
of 1987, Public Law 100-235 [CONG87], which requires safeguards against loss and unauthorized
modification.

1.4     HISTORICAL OVERVIEW

The starting point in modeling a MAC policy is the observation that security levels are
partially ordered (see Appendix A). This fact was first reflected in the design of the ADEPT-50,
an IBM 360-based, time-sharing system with both mandatory and discretionary controls.
[WEIS69, LAND81] The central role of partial orderings in MAC policy models was then
recognized in subsequent work. [POPE73; BELL73; BELL74a] These partial orderings on security
levels have become known as dominance relations.

Independently of this, access control matrices were used by Lampson and Graham [LAMP71;
GRAH72] to represent protection data. These works modeled discretionary access control using
matrices whose rows and columns represented subjects and objects. Subjects were active entities
such as processes and users. Objects were entities such as data structures that played a passive role,
but often included subjects as well. Each entry in an access matrix told what a given subject was
allowed to do with a given object by giving a set of allowed operations such as read, write, execute,
or change ownership. The AIS used the access matrix to mediate all accesses by subjects to objects.

An Air Force Computer Security Technology Planning Study [ANDE72], referred to as "the
Anderson Report," discussed use of a reference monitor to enforce nondisclosure policies and
advanced the idea of formal security verification- the use of rigorous mathematical proofs to give
(partial) assurance that a program design satisfies stated security requirements. This idea was
investigated by Schell, Downey and Popek. [SCHE73]

At that time, Bell and La Padula adapted access control matrices for use in modeling both
mandatory and discretionary access controls and codified the reference monitor concept using state
machines. Their machine states included access matrices and other security-relevant information
and are now often referred to as protection states. Having established the necessary framework,
they observed that a portion of DoD policy for nondisclosure of classified information could be
formalized as invariant properties of protection states [LAPA73], that is, as properties which hold
in all reachable states of the system. Their invariants constrained the allowed accesses between
subjects and objects. The most significant of these was their *-property It included a form of the
simple security property and guaranteed that the security level of every object read by an untrusted
subject was at or below the security level of every object written by that subject. [BELL74] These
ideas and their use in enforcing nondisclosure are covered in detail in Sections 3.2.4 and 4.1.

To complete their state machine, Bell and La Padula introduced a set of state transformations,
called rules of operation, that modeled basic changes in a protection state and then rigorously
proved that the rules of operation preserved the identified state invariants. [LAPA73, BELL74,
BELL76] This work contains the first widely discussed use of mathematical proof for studying
multilevel security.

In parallel with Bell and La Padula, Walter and others developed a similar approach to
verifying multilevel security. [WALT74, WALT74a] A strong point of the work by Walter et al.
is the use of modularity and data abstraction techniques that have since been accepted as
indispensable for verifying large systems.

The intent of the state invariants identified by Bell and La Padula is that information is
allowed to flow from one entity to another only if the second entity is at an equal or higher security
level than the first. Denning and others attempted to formalize this idea directly using "information
flow" models. [DENN76, DENN77, COHE77, REIT79, ANDR80]. These models differed in style
from access control models in that they referred explicitly to information flow, but, in contrast to
more recent investigations, did not actually define information flow. Denning's work did,
however, point out an interesting distinction. In the conditional assignment

if a = 0 then b := c,

information flows explicitly from c to b (when a = 0) and implicitly from a to b (when b  c).
Denning also pointed out that, in the case of the ADEPT-50, access control can easily allow
implicit information flow. [DENN77] This problem is discussed in detail in Section 3.2.

Discrepancies between information flow and access control open up the possibility of covert
channels- paths of information flow that malicious processes can use to bypass the access control
mechanism and thereby violate underlying security objectives. [cf LAMP73] Information flow
models and concern over covert channels have led to the development of techniques and tools for
covert channel analysis that are capable of detecting a variety of covert channels. [MILL76,
MILL81, FEIE77, FEIE80]

The developments just described provided the technical background for the security modeling
requirements given in the TCSEC. A variety of important developments not explicitly reflected in
the TCSEC have taken place in the last decade and will be presented later in this guideline. The
class of security policies that can be formally modeled has been greatly extended, beginning with
Biba's (pre-TCSEC) work on integrity. [BlBA77] Work has also gone into tailoring nondisclosure
security to particular applications. Finally, various external-interface models have been developed
that do not constrain internal system structure, an example of which is the noninterference model
of Goguen and Meseguer. [GOGU82] These models provide a rationale for rigorously assessing
new approaches to access control.

1.5     CONTENT AND ORGANIZATION

Section 2 presents an overview of the security modeling process, with emphasis on
correctness and utility. Section 3 presents technical detail on how to model concepts of interest,
including nondisclosure and integrity policies, mandatory and discretionary access controls, and
exemption from access control within the Trusted Computing Base (TCB).

Section 4 shows how to apply the techniques from Section 3 to various kinds of systems;
including operating systems, networks, database systems, and multilevel workstations. Finally,
Section 5 summarizes all of the TCSEC security modeling requirements for B1, B2, B3, and A1
computing systems.

Appendix A presents facts about lattices and partially ordered sets that are needed for Sections
3 and 4. Appendix B contains brief descriptions of available support tools. Appendices C and D
contain suggested outlines for a philosophy of protection and a security policy model. Finally,
Appendix E is a glossary giving definitions of technical terms. This glossary includes all terms that
are introduced in italics throughout the guideline.

When TCSEC requirements are discussed in this guideline, they are identified either by the
key words "must" or "shall" or by explicit quotations from the TCSEC. By way of contrast, when
desirable but optional actions and approaches are discussed, they are presented without exhortation
and are instead accompanied by an explanation of specific advantages or benefits. In a few cases,
possible requirements are designated by "should," because the implications of the TCSEC are not
fully understood or agreed upon.

2.       OVERVIEW OF A SECURITY MODELING PROCESS

A security model precisely describes important aspects of security and their relationship to
system behavior. The primary purpose of a security model is to provide the necessary level of
understanding for a successful implementation of key security requirements. The security policy
plays a primary role in determining the content of the security model. Therefore, the successful
development of a good security model requires a clear, well-rounded security policy. In the case
of a formal model, the development of the model also must rely on appropriate mathematical
techniques of description and analysis for its form.

Sections 2.1 and 2.2 explain what security models describe, why they are useful, and how they
are used in the design of secure systems. Section 2.3 introduces general definitions relating to
security models and explains how security models are created. Finally, Section 2.4 discusses the
presentation of a security model in a modeling document.

2.1     SECURITY MODELS AND THEIR PURPOSE

Early security models focused primarily on nondisclosure of information. More recently, the
importance of data as a basis for decisions and actions has stimulated interest in integrity models.
[DOD88a, WHIT84] For example, nondisclosure properties alone do not protect against viruses
that can cause unauthorized, malicious modification of user programs and data.

A wide variety of concepts can impact nondisclosure and integrity in particular system
designs. As a result, the content of security models is quite varied. Their primary purpose is to
provide a clear understanding of a system's security requirements. Without such an understanding,
even the most careful application of the best engineering practices is inadequate for the successful
construction of secure systems.

Inadequacies in a system can result either from a failure to understand requirements or from
flaws in their implementation. The former problem, defining what a system should do, is relatively
difficult in the case of security. The definition must be precise in order to prevent undesired results,
or subtle flaws, during the implementation of the system design.

During the entire design, coding, and review process, the modeled security requirements may
be used as precise design guidance, thereby providing increased assurance that the modeled
requirements are satisfied by the system. The precision of the model and resulting guidance can be
significantly improved by casting the model in a formal language.

Once the system has been built, the security model serves the evaluation and accreditation
processes. It contributes to the evaluators' judgement of how well the developers have understood
the security policy being implemented and whether there are inconsistencies between security
requirements and system design. Moreover, the security model provides a mechanism for tailoring
the evaluation process to the vendor's needs because it presents the security concept that is
supported by the system being evaluated. The inclusion of particular facts in the security model
proclaims to evaluators and potential customers that those facts are validated at the level of
assurance which the TCSEC associates with that system's evaluation class.

Upon successful evaluation and use, the security model provides a basis for explaining
security-relevant aspects of system functionality. Later, during maintenance, it provides a basis for
guidance in making security-relevant modifications. Finally, by suppressing inessential design
detail, security models facilitate a broader understanding of security that can be applied to
increasingly larger classes of systems. Among the many examples of such generalizations is the
adaptation of traditional reference monitor concepts referenced in the TNI to provide a basis for
understanding network security requirements. [NCSC87]

The intended purpose of a security model suggests several desirable properties. The
requirements captured by a good model pertain primarily to security, so that they do not unduly
constrain the functions of the system or its implementation. A good model accurately represents
the security policy that is actually enforced by the system. Thus, it clarifies both the strengths and
the potential limitations of the policy. (As an extreme example, if the system can simply declassify
all objects and then proceed normally, as in McLean's System Z [MCLE87], a good model would
show this.) Finally, a good model is simple and therefore easy to understand; it can be read and
fully understood in its entirety by its intended audience. This last property cannot be achieved
without significant care in choosing the contents, organization, and presentation of the security
model. For example, the desire to provide a security model with the "look and feel" of UNIX,
might need to be tempered by the need for simplicity and abstraction. [cf NCSC90b, Sec. 6.2]

2.2     SECURITY MODELING IN THE SYSTEM DEVELOPMENT PROCESS

Security requirements are best identified early in the system development process. Not
identifying security requirements in a timely fashion is likely to have devastating effects on
security assurance, security and application functionality, development schedule, and overall
development costs. For example, in the case of a development using DOD-STD-2167A, [DOD88]
this identification process would be part of the system requirements analysis. The identification of
security requirements begins with the identification of high-level security objectives (as described
in Section 1.3) and the methods by which they are to be met, including automated, procedural, and
physical protection methods. This identification of security requirements and their derivation from
identified higher-level security objectives is the initial material for a philosophy of protection
(POP). As indicated in Appendix C, the philosophy of protection may also include a broad range
of other topics such as the structure of the trusted computing base (TCB) and physical and
procedural security mechanisms.

Those requirements in the philosophy of protection which deal with automated protection
methods provide an initial definition of security for a security policy model. The model's purpose
is to precisely state these requirements and to compare them with key aspects of the security
enforcement mechanism. A security policy model in this sense contains two essential portions: a
"definition of security" portion that captures key security requirements and a "rules of operation"
portion that shows how the definition of security is enforced.

The model's definition of security can be used to avoid major system development errors. It
can be used to guide the design of basic software protection mechanisms and to influence the
design, selection and use of underlying firmware and hardware protection mechanisms. The initial
draft model, and supporting documentation, provides guidance as to system security during
reviews of the system design. However, there often are discrepancies between the design and the
model. Some of these are resolvable and can be identified and corrected during the normal design
review process. In some cases, however, discrepancies are unavoidable and can only be resolved
by making some assumptions that simplify the problem. These assumptions need to be justifiable,
based on the model. These discrepancies can also be addressed through procedural restrictions on
the use of the system. There are some portions of a security model that may require design
information that is not initially available and must, therefore, be postponed. Possible examples
include detailed rules of operation for a security kernel and security models for specific security-
critical processes. In such cases, the system designer must ensure that discrepancies are noted and
the completed system will satisfy the completed model.

To ensure that a design satisfies modeled security requirements, it is necessary to give a model
interpretation which shows how the model relates to the system. For evaluation classes B1 and B2,
this involves explaining how each concept in the model is embodied by the system design and
informally demonstrating that the modeled requirements are met. Since the model's rules of
operation must conform to the model's definition of security, the model interpretation need
demonstrate only that the rules of operation are adhered to. For classes B3 and A1, the model
interpretation is done in two steps. The design, as reflected in a top-level specification (TLS), is
shown to be consistent with the model, and the implementation is shown to be consistent with the
TLS. For Class B3, an informal descriptive top-level specification (DTLS) is used, an informal
correspondence from the DTLS to the model is performed, and the implementation is shown to be
consistent with the DTLS. At Class A1, the DTLS is supplemented with a formal top-level
specification (FTLS), a formal verification proves that the FTLS is consistent with the model, and
the implementation is shown to be consistent with the FTLS. A fuller summary of model-
interpretation requirements is given in Section 5.

The role of security modeling in relation to other aspects of system development is
summarized in Figure 2.1. Aspects that do not directly involve modeling, per se, are shaded in grey.
Requirements concerning the model are summarized at the beginning of Section 5. The broadest
document is the philosophy of protection (POP); it covers higher-level security objectives, derived
security policies constraining the design and use of the system, and the protection mechanisms
enforced by the TCB. The POP, the security policy, and the model all cover the system's definition
of security. Both the POP and the model cover key aspects of the TCB protection mechanisms. At
B2 and above, the formal model supports a rigorous proof showing that the rules of operation
enforce the definition of security, and the DTLS gives a functional description of the TCB
protection mechanisms with emphasis on the TCB interface. At A1, the FTLS formalizes a large
portion of the DTLS in order to verify that the TCB design satisfies the modeled security
requirements.



Figure 2.1. Security Documentation

The above paragraphs refer to a single security model but networks, database systems, and
other complex systems may involve several security models or submodels. When a system is made
up of several complex components or subsystems, interfaces between the components or
subsystem layers must be modeled, if they play a key role in security protection. In this case, the
best approach may be to develop separate models for each component or layer in order to show
how the various subsystems contribute to overall system security. A danger with this approach,
however, is that the combined effect of the various submodels may not be obvious. This is
discussed further in Section 4.

2.3     IDENTIFYING THE CONTENTS OF A SECURITY MODEL

The most basic strategy in identifying the contents of a security model is perhaps that of divide
and conquer. The modeling effort may be subdivided according to higher-level security objectives.
Requirements on a system can be mapped to derived requirements on subsystems. Requirements
for a particular objective can be classified according to level in a requirements hierarchy.

Security modeling provides a five-stage elaboration hierarchy for mapping a system's
security policy requirements to the behavior of the system. As a result of this hierarchy, the phrases
"security policy" and "security model" have taken on a variety of meanings. The five relevant
stages are as follows:

1.                      Higher-level policy objectives

2.              External-interface requirements

3.                      Internal security requirements

4.              Rules of operation

5.      Top-level specification

A higher-level objective specifies what is to be achieved by proper design and use of a computing
system; it constrains the relationship between the system and its environment. The TCSEC control
objectives belong to this first level of the requirements hierarchy. An external-interface
requirement applies a higher-level objective to a computing system's external interface; it explains
what can be expected from the system in support of the objective but does not unduly constrain
internal system structure. Internal security requirements constrain relationships among system
components and, in the case of a TCB-oriented design, among controlled entities. Rules of
operation explain how internal requirements are enforced by specifying access checks and related
behaviors that guarantee satisfaction of the internal requirements. A top-level specification is a
completely functional description of the TCB interface. It also specifies behavior of system
components or controlled entities.

In various contexts, the phrase security policy may refer to any or all of the first four stages
of elaboration. In many contexts, the term security policy refers to organizational security policy.
[cf NCSC85, Glossary] From a modeling perspective, organizational policies are the source of
higher-level policy objectives. In much of the literature on security modeling, a system security
policy is part of the requirements stages of development and provides the definition of security to
be modeled. Additional thoughts on the importance of distinguishing among policy objectives,
organizational policies, and system policies may be found in Sterne's article "On the Buzzword
"Security Policy." [STER91] The terms "AIS security policy" and "automated security policy" are
synonyms for "system security policy."

The security policy model is a model of the security policy enforced by the TCB. It is
simultaneously a model of the policy and of the system for which the model is given. The portions
of the system which are constrained by the model belong to the TCB by definition. The model's
definition of security may contain external-interface requirements, but more traditionally consists
of internal requirements. Its rules of operation may, in some cases, amount to a top-level
specification. There is no firm boundary between rules of operation and top-level specifications:
both explain how internal requirements are enforced by the TCB. However, more detail is required
in a TLS, including accurate descriptions of the error messages, exceptions, and effects of the TCB
interface. Security requirements occur implicitly in rules of operation and top-level specifications.
In this form they are often referred to as security mechanisms. Internal requirements are sometimes
classified as "policy" or "mechanism" according to whether or not they are derived from a specific
higher-level policy objective.

Conceptually, the security modeling process takes place in two main phases. Requirements
modeling takes place after a system security policy is fairly well understood. This is accomplished
by constraining the system's design and use based on the higher-level objectives. Rules of
operation may be deferred until after the definition of security is established and the basic
architecture of the system has been identified.

The following paragraphs contain general suggestions for the construction of security models.
Suggestions pertaining to specific kinds of security requirements and specific kinds of systems are
given in Sections 3 and 4, respectively.

These suggestions are broken down into six steps that could be carried out with respect to an
entire system and security policy or, more simply, for a given component and higher-level policy
objective. The first step is to identify externally imposed security policy requirements on how the
system (or component) must interact with its environment. The second is to identify the internal
entities of the system and to derive the requirements portion of the model in such a way as to extend
the external-interface requirements. The third is to give rules of operation in order to show how the
modeled security requirements can be enforced. After completion of step three enough information
generally is available to determine reliably whether the model is inherently new or can be
formalized by known techniques. At classes B2 and above, this is the fourth step. The first three
steps taken together result in the identification of several submodels. The fifth step is to
demonstrate consistency among these various submodels, so that they fit together in a reasonable
way to form a completed security policy model. Finally, the sixth step is to demonstrate relevance
of the model to the actual system design. Typically, the majority of the security modeling effort is
associated with these last two steps and associated revisions to the model. The total effort needed
to produce a security policy model varies considerably, but is often between six staff-months and
two staff-years.

2.3.1           STEP 1: IDENTIFY REQUIREMENTS ON THE EXTERNAL INTERFACE

The first step is to identify major security requirements and distinguish them from other kinds
of issues. These identified requirements should adequately support known higher-level policy
objectives for use of the system. An emphasis on external-interface requirements helps prevent an
unrecognized mixing of security and design issues. Such mixing could interfere with the
understanding of security and could impose unnecessary constraints on the system design.

External-interface requirements for a computer system can be described in several ways. An
elegant, but possibly difficult, approach is to limit the discussion purely to data crossing the system
interface; this is the "black-box" approach. The best known example of the black-box approach is
noninterference (see Section 3.2.1). Alternatively, one can describe the system in terms of its
interactions with other entities in its environment, such as other computing systems, users, or
processes. Finally, one can give a hypothetical description of internal structure that ensures the
desired external-interface behavior.

In general, the system's interaction with its environment is constrained by the sensitivity of
the information handled and by the authorizations of the individuals and systems accessing the
system. Identified user roles, associated privileges, and the extent to which certain roles are
security-critical also limit the system's interface to the environment. These constraints determine
security attributes that are associated with the system's inputs and outputs. Security attributes may
include classification, integrity, origin, ownership, source of authorization, and intended use,
among others. The use of such attributes in the construction of security models is discussed in
Section 3.

In the case of mandatory access control policies, information must be accurately labeled and
handled only by authorized users. This requirement places restrictions on how information is input
to the system and, implicitly, on how it will be processed. Authorized handling of information is
modeled in terms of constraints on what information may flow from one user to another:
information input to the system as classified information should be output only to authorized
recipients.

In addition to general regulations, there are often site-dependent and application-dependent
constraints that may need to be modeled. In particular, there may be site-dependent constraints on
allowed security labels.

Having identified the security requirements on the system interface, it is necessary to decide
on the requirements to be covered in the model's definition of security. It must then be decided
which of these requirements should be modeled directly or indirectly in terms of internal
requirements on system entities and the TCB interface. The set of requirements to include is
constrained by minimal TCSEC requirements, the need to adequately support relevant policy
objectives, and the need for a simple, understandable model. The inclusion of more requirements
may provide more useful information for accreditation once the system is evaluated, but it also
increases the difficulty of the vendor's assurance task. The inclusion of more requirements also
suggests a more careful structuring of the model in order to show how various aspects of security
fit together.

Several factors may influence a decision of whether to directly include external-interface
requirements in the security model. The direct inclusion of external-interface requirements can
help explain how the model supports higher-level policy objectives. In the case of an application
security model, the direct modeling of user-visible operations may be more relevant to end users,
a point of view reflected in the SMMS security model. [LAND84] In the case of a network security
model, understanding is facilitated by modeling of the network's interaction with hosts in its
environment, as will be discussed in Section 4.2.

2.3.2           STEP 2: IDENTIFY INTERNAL REQUIREMENTS

To support the identified external requirements, the system must place constraints on the
controlled entities of the system. These internal constraints traditionally form the model's
definition of security. In a model whose definition of security contains external-interface
requirements, internal constraints can provide a needed bridge between the definition of security
and the rules of operation.

The controlled entities themselves should be identified at a level of granularity that is fine
enough to allow needed security-relevant distinctions among entities. At class B2 and above, the
controlled entities should include all (active and passive) system resources that are accessible
outside of the TCB. For convenience, controlled entities may be grouped into subclasses in any
way that facilitates understanding of the system. Such groupings will depend on the system to be
modeled. For an operating system, the relevant controlled entities might include buffers, segments,
processes, and devices. In most models, it has been convenient to group entities according to
whether they play an active or a passive role in order to help show how the TCB implements the
reference monitor concept . For networks and other complex systems, identification of controlled
entities may need to be preceded by identification of subsystems and their derived security
requirements.

Constraints on controlled entities are best stated as general properties and relationships that
apply to all (or a broad class of) entities and accesses. Greater generality eliminates unnecessary
constraints on the system design, improves on's intuition about the model, and can greatly reduce
the overall effort needed to validate correctness of the identified constraints.

The identification of necessary constraints on controlled entities and their interactions is a
nontrivial task; familiarity with similar systems and security models can be of great benefit. To
define the necessary constraints, it will be necessary to label each entity with appropriate security
attributes and to identify possible kinds of interactions, or accesses, between controlled entities. As
used here, an access is any interaction that results in the flow of information from one entity to
another. [cf DOD88a]

In describing access constraints, it may be useful to know that some kinds of accesses are
highly restricted, as when a process accesses a file by executing it. The notion of causality may also
be important: a transfer of information from entity e1 to e2 might occur because e1 wrote e2,
because e2 read e1,or because a third entity directly copied e1 to e2 without actually reading e1. [cf
MCLE90, Sec. 2]

In the particular case of state machine models, constraints often take the form of state
invariants, as in the work of Bell and La Padula. Recent efforts suggest that state-transition
constraints are an attractive alternative to state invariants for certain kinds of security requirements.
[MCLE88; NCSC90b, Sec. 6.7] The simplest well-known, state-transition constraint is the
tranquility principle of Bell and La Padula, which says that the security level of a controlled entity
cannot change during a state transition. Tranquility can be artificially rephrased as a state invariant:
in any reachable state, the level of an entity coincides with its level in the initial state. Consider,
however, the DAC requirement that a subject which gains access to an object must possess the
necessary DAC permissions when the access request is made. This is harder to rephrase as a state
invariant because DAC permissions can change from one state to the next. Good tutorial examples
of state invariants and state-transition constraints may be found in Building a Secure Computer
System [GASS87, Sec. 9.5.1, 9.5.2]; more complex examples occur in the TRUSIX work
(NCSC90b, Sec. 2].

The number of internal requirements in a typical model varies considerably depending on the
desired level of assurance, the complexity of the policy and system being modeled, and the
granularity of the individual requirements. For example, the Trusted UNIX Working Group
(TRUSIX) model covers TCSEC access control requirements for a B3 Secure Portable Operating
System Interface for Computer Environments (POSIX) system and has eleven state invariants and
ten state-transition constraints. [NCSC90b]

The question of which security requirements to cover in the model's internal requirements is
answered as much by issues of good engineering practice as by the TCSEC. At a minimum, the
model must cover the control of processes outside the TCB. Such processes are potentially a
significant security risk because they may be of unknown functionality and origin. According to
current practice, those portions of the TCB that do not support user-requested computation do not
have to be modeled. For example, the security administrator interface does not have to be modeled.
However, if significant user-requested computation is performed by some portion of the TCB, it
should be modeled. A typical example would be a trusted downgrading process available to
authorized users (not necessarily the security administrator). A more extreme example would be a
multilevel database management system implemented as a TCB subject whose database consisted
of a single highly-classified file.

Finally, there is the possibility that some security requirements might be included in the
model, but only in the rules of operation. The rules of operation may allow the application of good
security practice closer to the implementation level. The rules of operation can then contain
security conditions that are not explicitly required by the security policy. Rules of operation can
not make up for an inadequate definition of security, however. The task of inferring a modeled
security policy from the rules of operation may be excessively difficult, especially if the rules are
complex. [cf NCSC90b, Sec. 6.8]

2.3.3           STEP 3: DESIGN RULES OF OPERATION FOR POLICY ENFORCEMENT

How can the modeled security requirements on system entities be enforced? The rules of
operation answer this question in broad terms by describing abstract interactions among system
entities with particular emphasis on access control and other policy-enforcement mechanisms. In
the case of an operating system kernel, the rules of operation would typically describe state
transformations and associated access checks needed to uphold the definition of security. In the
case of a formal model, this step amounts to the construction of a miniature FTLS. It is a useful
preliminary step, even if a full FTLS specification is planned. The rules of operation together with
the modeled requirements on controlled entities form the security policy model.

In giving rules of operation, it is important that system behavior be adequately represented.
However, actual interactions among controlled entities need not be directly specified when they
can be described as a composition of several rules of operation. There is no need for an
implementation to directly support an individual rule where several rules have been combined to
describe an action. An appropriate level of detail for rules of operation is illustrated by the work of
Bell and La Padula. [BELL76] Good tutorial examples may be found in Building a Secure
Computer System. [GASS87, º 9.5.1 (Step 3)] More complex examples may be found in the
TRUSIX work. [NCSC90b] The rules of operation will usually be more readable if they are not too
detailed. At class B3 and above, it is especially important to avoid details that are not security
relevant. This is because their inclusion in the model may force their inclusion in the TCB in order
to achieve a successful model interpretation, thereby violating the TCB minimization
requirements. [NCSC85, Sec.3.3.3.1.1]

In the case of a B1 informal model, the rules of operation may reasonably contain information
that, for higher evaluation classes, would be found in a DTLS. Thus, a stronger emphasis on policy
enforcement than on policy requirements may be useful. This is especially true if the TCB is large
and complex, as, for example, when security is retrofitted onto a previously unevaluated system.
[BODE88]

A formal model needs to formalize the idea that a particular state transformation is being
executed on behalf of a particular subject. Two traditional approaches are to add an extra input to
the state transformation that gives the subject id and to add a "current subject" field to the system
state. With the former approach, accompanying documentation should explain clearly how the
subject-identity parameter is passed, in order to avoid the erroneous impression that the subject is
responsible for identifying itself to the TCB. The latter approach suggests that the system being
modeled has a single central processing unit. This may be a problem for the model interpretation
if the system actually contains multiple CPUs. See the TRUSIX work for further discussion.
[NCSC90b, Sec. 6.13]

Finally, in designing rules of operation, it may be convenient to separate access decisions
from other kinds of system functionality, including the actual establishment or removal of access.
This separation facilitates exploration of new access control policies. Only the access decision
portion is affected by a change in policy because access enforcement depends only on the access
decision, not on how it was made. [LAPA90, ABRA90] Isolation of the access decision
mechanism occurs in the LOCK system design [SAYD87] and in the security policy architecture
of Secure Ware's Compartmented Mode Workstation [NCSC91b, Section 2.2.1.].

2.3.4           STEP 4: DETERMINE WHAT IS ALREADY KNOWN

Usually some, if not all, aspects of the identified security model will have been studied before.
The identification of previously used terminology allows security issues to be presented in a
manner that is more easily understood; and making the connection to previously studied issues may
provide valuable insight into their successful solution.

For classes B2 and above, the modeling effort must be based on accepted principles of
mathematical exposition and reasoning. This is because formal models and mathematical proofs
are required. The chosen mathematical formalism must allow for an accurate description of the
security model and should provide general mechanisms for its analysis. This is necessary so that
specific interactions among the entities of the model can be verified as being appropriate.

In practice, needed mathematical techniques are usually adapted from previous security
modeling efforts because security modeling, per se, is generally much easier than the development
of new mathematical techniques. Extensive use of past modeling techniques is especially feasible
in the case of fairly general and familiar policy requirements. System-dependent modeling is
generally required for policy enforcement, but established techniques are often available for
demonstrating consistency with the modeled requirements.

If new mathematical techniques are used, their credibility is best established by exposure to
critical review, in order to uncover possible errors in the mathematics and its intended use. This
review process is facilitated by the development of comparative results that give useful
relationships between new models and old ones.

2.3.5           STEPS: DEMONSTRATE CONSISTENCY AND CORRECTNESS

Since the security provided by a system is largely determined by its security model, it is
important that the model capture needed security requirements and that its rules of operation show
how to enforce these requirements.

The first crucial step of identifying security requirements on the system interface cannot be
directly validated by mathematical techniques. [cf MCLE85] Human review is needed to establish
what security requirements need to be addressed and whether these requirements are reflected in
later mathematical formalizations. For systems in class B2 and above, real-world interpretations
are assigned to key constructs of the formal model in order to provide a common semantic
framework for comparing the model and the security policy.

The appropriate technique for validating requirements on controlled entities (Step 2) depends
partly on the novelty of the approach. Informal human review is required to assure that the modeled
requirements support the original system security policy. Careful comparisons with previous
models of the same or related policies may help to show how new modeling techniques relate to
previously accepted techniques for handling particular policy concepts. (See, for example,
[MCLE87; MCLE90, Sec. 3 & Sec. 4].) An alternate technique is to give an external-interface
model and then mathematically prove that the constraints on system entities imply satisfaction of
the external-interface model. The external-interface model is easily compared with a security
policy or policy objective. The constraints on system entities are those identified in the
requirements portion of the security policy model. This technique is illustrated in Section 3.2.

If the rules of operation are given correctly (Step 3), they will comply with the constraints
given in the requirements portion of the model. A formal proof of compliance is required for
classes B2 and above. In a state-transition model, state invariants are proved by induction: one
proves that they hold for the initial state and are preserved by each state transition given in the rules
of operation. (See, for example, [CHEH81].) State-transition constraints are straightforwardly
proved by showing that each state transition satisfies the constraints.

2.3.6           STEP 6: DEMONSTRATE RELEVANCE

The rules of operation should be a correct abstraction of the implementation. A preliminary
model interpretation that is done as part of the modeling process can provide an informal check on
whether the rules of operation are sensible and relevant. This model interpretation shows how
enforcement mechanisms modeled in the rules of operation are realized by the system. A model
interpretation explains what each model entity represents in the system design and shows how each
system activity can be understood in terms of the given rules of operation. Thus, for example, a
"create file" operation in the system might be explained as involving a restricted use of the rules
for "create object," "write object," and/or "set permissions" in the model, depending on the actual
arguments to the "create file" command.

An appropriate, somewhat novel, example of a model interpretation is found in the TRUSIX
work. [NCSC90b, Sec. 4] The model interpretation is described as an informal mapping from the
TRUSIX DTLS to the TRUSIX model. This interpretation first explains how UNIX entities are
represented in the model and the DTLS. For example, messages, semaphores, and shared memory
are entities that the DTLS refers to as interprocess communication (IPC) objects but that are treated
as "files" in the model. Thirty-six UNIX system interface functions are then described by giving
pseudocode that shows how each function can be expressed in terms of the state transformations
given by the rules of operation. In other words, the transformations given in the rules of operation
form a toy security kernel that could be used to implement UNIX if efficiency were irrelevant.

The SCOMP and Multics model interpretations are examples based on Secure Computer
System: Unified Exposition and Multics Interpretation. [BELL76]. In the SCOMP interpretation,
a set of security-relevant kernel calls and trusted processes are identified as `SCOMP rules of
operation." [HONE85] For each such SCOMP rule, a brief summary of security-relevant
functionality is given and then interpreted as the combined effect of restricted use of one or more
rules from the above work by Bell. [BELL76] The specific restrictions are enumerated in an
appendix, along with a rationale for omitting some kernel calls and trusted functions from the
interpretation. The correspondence between the rules in this work [BELL76] and actual SCOMP
behavior is rather complex. The Multics interpretation, by way of contrast, is more sketchy, but the
actual correspondence appears to be simpler. [MARG85]

2.4     PRESENTATION FOR EVALUATION AND USE

The most important factors in the successful presentation of a security model are the
simplicity of the model, its appropriateness, and an understanding of its intended uses.

The presentation of the model must demonstrate that the model correctly describes the system
security policy. A clear explanation of the policy from which the model is derived should be
available for this demonstration. Sufficiency of the model may be demonstrated by presenting the
relationship of the model to the policy and by carefully explaining and justifying the modeling
decisions behind this relationship. In addition to modeled requirements, the system may support
other security requirements that are not suitable for inclusion in the model, especially if it is a
formal model. Unmodeled security requirements can be presented in an appendix so that TCB
developers have all of the security requirements in a single document.

An overview of the model interpretation is needed so that readers can understand the
relevance of the system's security model to the security policy and to the overall system-
development process. All of these topics may be legitimately covered in a philosophy of
protection. In fact, it is highly desirable that the philosophy of protection be the first document
delivered by the vendor in a system evaluation, so that it can serve as a basis for further
understanding of the security model and other system documentation.

In the case of a formal model, an informal explanation or presentation of the model prepared
for a general audience is highly desirable. This is partly for reasons mentioned in Section 2.3.5 and
partly because of the general need to acquaint system developers, implementors, and end users
with the basic security principles that are to be supported by the system. The informal presentation
can reasonably be included in the philosophy of protection.

A well-written explanation of the model's definition of security can be used by potential
buyers of a secure system to determine compatibility between the computing system and their
organization's security policies and needs. To evaluate the ability of a computing system to support
these policies, potential users may wish to construct an informal model of their security policies
and compare it to the definition of security supported by the computing system. This use of the
security model and the fact that some aspects of security modeling can only be validated by social
review suggest that portions of the security model and the philosophy of protection be made
available to a relatively wide audience, for example, treating them as publicly released,
nonproprietary documents.

For systems at the B2 level or above, the requirement of mathematical proof necessitates the
presentation of a rigorous mathematical formalization as well. For a mathematical audience,
standards of precision may preclude a style that will be appropriate for a general audience. In this
case, an informal presentation of the model is also needed in order to reach the general audience.

At the A1 level of assurance, the formal model is directly involved in the formal verification
of the FTLS. As a result, it may be appropriate to present the model in the formal specification
language of an endorsed verification system (see Appendix B). Authors and readers must be fully
aware of the nuances of the descriptive notation of the verification system used in order to avoid
errors due to discrepancies between author intent and reader expectation. Furthermore, if a formal
verification system is used to check proofs, it is necessary to achieve a correct translation of what
is to be proved into the language of the verification system. Further discussion of these points may
be found in [FARM86; NCSC90b, Sec. 6.3].

3.       SECURITY MODELING TECHNIQUES

Having introduced the major topics involved in security modeling, we now consider detailed
modeling issues with emphasis on mathematical structure as well as empirical justification. In this
section, a review of basic concepts is followed by discussions on the modeling of various kinds of
policies and objectives: nondisclosure, need-to-know, integrity, and special-purpose policies of
particular TCB subjects.

In most cases, modeling techniques are introduced, with details handled via references to the
available literature. Inclusion of these references does not imply NCSC endorsement of referenced
products incorporating the techniques discussed.

3.1     BASIC CONCEPTS

As discussed in Section 2, the identification of controlled entities plays a crucial role in the
development of a security policy model. At B2 and above, the controlled entities must include all
system resources. If the policy is multifaceted, with separate subpolicies for mandatory and
discretionary access controls, it may be acceptable to decompose the system into different sets of
controlled entities for different subpolicies. A secure computing system may decompose into data
structures, processes, information about users, I/O devices, and security attributes for controlled
entities. In general, the number of different kinds of entities depends on what security-relevant
distinctions are to be made. The following paragraphs discuss how to perform such a
decomposition for typical kinds of entities, with the goal of modeling security requirements in such
a way as to allow an accurate, useful model interpretation.

An explicitly controlled entry is one that has explicitly associated security attributes. In the
TRUSIX model, for example, the explicitly controlled entities are referred to as "elements". They
consist of subjects and three kinds of objects; files, directories, and entries (i.e., links). [NCSC90b,
Sec. 6.9-6.11] In addition to explicitly controlled entities, a system will have implicitly controlled
entities. Such an entity might be contained in an explicitly controlled entity or might be a
composite entity composed of explicitly controlled entities having potentially different security
attributes.

The discussion of security attributes in this section emphasizes security levels used for
mandatory access control because the TCSEC labeling requirements apply primarily to mandatory
access control and auditing. However, much of what is said about MAC labels applies to other
kinds of security attributes including, for example, discretionary access control lists.

It is necessary to model creation and destruction of most kinds of controlled entities. A
simpler model results if the same approach is used in all cases. In a formal model, one can model
the creation of new controlled entities by modeling changes in the set of all controlled entities. A
fixed set of possible controlled entities, together with an indication of which entities are available
for use in a particular state (e.g., which subjects are active, which objects or object-ids are
allocated) can also be used. [cf NCSC90b]

3.1.1           DATA STRUCTURES AND STORAGE OBJECTS

In this guideline, a data structure is a repository for data with an internal state, or value, that
can be written (i.e., changed) and/or read (i.e., observed) by processes (and, possibly, devices)
using well defined operations available to them. A data structure is distinct from its value at a
particular time, which is, in turn, distinct from the information conveyed by that value. The
information content depends not only on the value but also on how it is interpreted. Similarly, a
data structure is distinct from a name for that data structure.

A data structure that is explicitly associated with exactly one security level by the security
policy model is a storage object. There are no a priori restrictions on the level of abstraction at
which data structures and storage objects are modeled. In particular, objects may be abstract data
structures that are accessed using high-level operations provided by an encapsulation mechanism.
In this case, a user would have access to the high-level operations but could not access the
encapsulated data directly via more concrete operations that may have been used to define the high-
level operations. This lack of direct access would have to be enforced by the TCB.

Ordinarily, the storage objects in a security model are disjoint. This means that, in principle,
changes to one object do not force changes to other objects. If a state-machine model is used,
disjoint storage objects can be modeled as separate components of the underlying machine state.
If two objects at different security levels were not disjoint, then access to data in their intersection
would be allowed or denied according to which label is used. This ambiguity can be resolved by
explicitly associating a level with their intersection. In this case, the associated level allows the
intersection to also be regarded as a separate storage object, thereby restoring disjointness. Thus,
disjointness in storage objects simplifies one's understanding of how security labels are used to
control access. It has also been argued that disjointness simplifies covert channel analysis via
shared resource matrices.

In some systems, collections of objects are combined to form multilevel data structures that
are assigned their own security levels. Multilevel data structures called "containers" are used in the
Secure Military Message System (SMMS) to address aggregation problems. [LAND84] The
level of a container must dominate the levels of any objects or containers inside it.

When an object or other controlled entity is created, it must be assigned security attributes and
initialized in such a way as to satisfy the object reuse requirement. The modeling of object reuse
policies (e.g., objects are erased when allocated) can be useful, but object reuse is not found in
traditional access control models because the object reuse requirement deals with the content of
objects and TCB data structures. The initialization of security attributes, however, plays an
essential role in access control and can be modeled by distinguishing between "active" and
"inactive" entities, as in [NCSC90b]. The security attributes of a newly created object may be taken
from, or assigned by, the subject that created it. In many systems, creating an object is essentially
a special case of writing to it: its value changes from undefined to null. If this change is visible to
non-TCB subjects and the new object is created by a non-TCB subject, then the security level of
the new object must dominate that of the creating subject. DAC attributes, by way of contrast, are
usually given by system- or user-supplied defaults.

3.1.2           PROCESSES AND SUBJECTS

A process may create, destroy, and interact with storage objects and other processes, and it
may interact with I/O devices. It has an associated run-time environment and is distinct from the
program or code which defines it. For instance, the program would ordinarily be modeled as
information contained in a storage object. An explicitly controlled process is a subject. It normally
has a variety of associated security attributes including a security level, hardware security
attributes such as its process domain, the user and user group on whose behalf it is executing,
indications of whether it belongs to the TCB, and indications of whether it is exempt from certain
access control checks.

The issues regarding disjointness of storage objects mentioned in Section 3.1.1 apply to
subjects as well. If two subjects share memory, it is advisable to model the shared memory as an
object separate from either subject in order to achieve the disjointness needed for information-flow
analysis. Local memory need not be separately modeled, but must be properly taken into account
when modeling the subject. Registers (including the instruction pointer) are technically shared
memory but can often be treated abstractly as unshared local memory. This is especially true if
registers are saved and restored during process swaps and the information they contain is not
shared between processes.

Security levels for processes are ordinarily similar to security levels for storage objects. The
TCSEC requires that a subject have a nondisclosure level that is dominated by the clearance and
authorization of its user. A TCB subject may reasonably be assigned separate levels for reading
and writing in order to allow partial exemption from access control (see Section 3.4). Other TCB-
related entities may also need separate levels for reading and writing in some systems. One such
example is /dev/null in UNIX. [GLIG86] This object is system high in the sense that any process
can write to it; but it is system low in the sense that any process can read from it because its value
is always null.

In general, the security-relevant attributes of a subject are part of its run-time environment.
Some attributes are relatively static and are inherited from the subject's executable code or are
stored with its code. In UNIX, the property of being a set-user-id program is a statically determined
attribute, as is the effective user-id. Other attributes are dynamically assigned and are inherited
from a subject's parent or are actively assigned by the parent (assuming it has a parent process). In
the latter case, the parent must be relied upon to assign the child's security attributes appropriately.
As a result the parent usually belongs to the TCB in this latter case. A secure terminal server, for
example, might create a subject and assign it security attributes based on the determined identity
of a user logging in.

There are potential difficulties in associating user-ids with TCB subjects. A terminal server
would appear to operate on behalf of the user currently logged in or on behalf of the Identification
and Authentication mechanism if no one was logged in. A print spooler acts on behalf of the users
in its print queue. Such TCB subjects are, in reality, associated with multiple user-ids, a fact which
is relevant to the design of the audit mechanism. A TCB subject that does not act on behalf of a
given user can be handled either as having a dynamically modifiable user-id or as acting on behalf
of the system (equivalently, as acting on behalf of a system-defined pseudo-user).

When a subject creates a child subject by executing a program, the resulting child might
belong to the TCB and be exempt from certain access control checks, even if the original subject
was not in the TCB. This is possible if exemptions are associated with the program object itself.

With the advent of multitasking languages such as Ada, there is a question of when two
threads of control belong to the same subject. Pragmatically, to be the same subject, they should
have the same security attributes. To qualify as separate subjects, their separation should be
enforceable by the TCB. Finally, nothing explicitly prohibits a multithreaded process from
consisting of several subjects operating at different security levels. Of course, intraprocess
communication will be somewhat limited for such a process unless it contains TCB subjects that
are exempt from some of the MAC constraints.

3.1.3           USERS AND USER ROLES

User-related topics often turn up in security models in spite of the fact that users are not
controlled entities and thus do not need to be directly modeled. The users of a system may perform
specific user roles by executing associated role-support programs. In general, a user may engage
in a combination of several roles. As a matter of policy, a given user role may require system-
mediated authorization and may provide specific system resources needed for performance of the
role. Although the following paragraphs discuss only individual user roles, most of the basic
concepts extend to roles for other kinds of clients in the environment of a computing system
including user groups and, in the case of a network, hosts or subjects running on hosts.

The TCSEC requires support for certain trusted user roles. For systems at B2 and above, these
roles include the security administrator role and the system operator role. The administrator role
governs security attributes of users, moderates discretionary and mandatory access control, and
interprets audit data. The operator role ensures provision of service and performs accounting
activities. [NCSC90a] These roles may be subdivided so that they can be shared by multiple users.
In addition, they do not preclude the addition of other trusted roles. In general, trusted user roles
are characterized by the fact that they involve handling security-critical information in a way that
violates access restrictions placed on non-TCB subjects. As a result, processes that support trusted
roles can be misused and must have restricted access. By definition, such trusted role processes
have security properties that are unacceptable without special constraints on the behavior of their
users. These observations suggest (but do not mandate) the modeling of support for user roles,
especially trusted user roles. Trusted role processes are usually treated as TCB subjects. Additional
information on modeling them is given in Section 3.4.

Instructive examples of role definitions may be found in Secure Xenix [GLIG86] and the
SMMS security model [LAND84]. Secure Xenix restructured the Guru role into four separate
roles. The SMMS security model included a system security officer role, a downgrader role and a
releaser role. In both of these systems, the roles support separation of duty in that each role has
privileges not available to the other roles. Separation of duty has been emphasized by Clark and
Wilson, who have given a general scheme for constraining user roles by using triples of the form
(user, program, object-list) in order to control how and whether each user may access a given
collection of objects. [CLAR87] A similar role-enforcement mechanism is supported by type
enforcement. [BOEB85] The type-enforcement mechanism itself assigns "types" to both subjects
and objects; subject types are referred to as "domains." The accesses of each subject are
constrained by a "type table" based on subject domain and object type. Each user has a set of
domains associated with subjects running on his behalf. These domains are used to define user
roles. [cf THOM90]

3.1.4           I/O DEVICES

Although users are external to the computing system and need not be directly modeled, it is
still useful to model their allowed interactions with the system in order to document security-
relevant constraints that are enforced by the system. User interactions may be modeled in terms of
constraints on I/O devices if there is a convention for discussing the current user of a device.
Devices which are accessible to subjects outside the TCB should be modeled either implicitly or
explicitly as part of the interface between the non-TCB subjects and the TCB. An additional reason
for interest in I/O devices is that I/O typically accounts for a large portion of the TCB. The
following paragraphs present general information on the modeling of devices with emphasis on
device security requirements and then discuss when they should be modeled as objects, as subjects,
or as their own kind of entity.

Normally, security models discuss abstract encapsulated devices rather than actual physical
devices. An encapsulated device is an abstract representation of a device, its associated device
driver, and possibly other associated entities such as attached hardware and dedicated device
buffers. Thus, for example, one might discuss a line printer connected to the system but not the
actual RS-232 device used to achieve the connection or its associated device driver.

By definition, an input device injects information into the system in a way that the system
cannot entirely control. The next state of the system may depend on the actual input as well as on
the current state and the particular state transformation being executed. This fact can be
accommodated in several ways in an FTLS or a model that addresses object content. One can treat
the actual input as a parameter to the transformation. If there are only a few different input values,
one can associate different transformations with different inputs. Finally, one can treat the
transformation as being nondeterministic, meaning that several different states can result from
executing the transformation in a given state.

Abstract encapsulated devices are often passive entities, in contrast to their underlying
hardware. Security requirements for devices, however, differ significantly from those for either
storage objects or controlled processes:

╖       External policy on use of the system requires that devices pass information only to
authorized users.

╖       Devices may transport either unlabeled data or labeled data and are classified as single
level or multilevel devices accordingly.

╖       At B2 and above, the TCSEC requires that every device have a minimum and maximum
device level that represents constraints imposed by the physical environment in which
the device is located.

Authorized use of a device may be enforced by requiring that any piece of information output
by the device have a security level that is dominated by the clearance and authorization of the
recipient. A combination of procedural and automated methods are needed to correctly associate
the security level of the information, the user who will receive that information, and the security
level of that user. Typically, the device itself will also have a security level. The user who receives
information from a device may well be different than the user who sent that information to the
device. If a device with local memory (an intelligent terminal, for example) is allocated to different
users or different security levels at different times, it will typically need to be reset between users
in order to meet requirements on authorized transmission of information.

Both single-level and multilevel devices may handle multiple levels of information. A single-
level device can only handle a single level at a time but may have some convention for changing
levels. A multilevel device, by way of contrast, can handle different levels of data without altering
its security designation, but still might be used in such a way as to carry data at only one fixed
security level.

The TCSEC requirement for device ranges does not explain how they are to be used. The most
common option is to require that the level of any object transmitted by the device be within the
range. Another option is to require that the security clearance of any user be within the range.
Separate decisions regarding device minimum and device maximum are also possible. In the
Trusted Xenix system, a user with a secret clearance can have an unclassified session on a
terminal with a secret device minimum. The intended interpretation of the device minimum is that
the device is in a restricted area that should contain only users whose clearance is at least the device
minimum. If the device minimum is secret, then a login attempt by an uncleared user is treated as
a violation of physical security. [GLIG86]

The question of whether devices should be modeled in the same way as other kinds of
controlled entities depends on the complexity of the devices involved, observed similarities
between devices and other modeled entities, and goals relating to the simplicity and accuracy of
the model. The TRUSIX group decided to model devices in the same way as file objects, a decision
that depended heavily on both the nature of UNIX and the specific goals of the TRUSIX effort.
[NCSC90b, º 6.9] The SMMS model treats devices as containers in order to capture the fact that
the device maximum must dominate the level of all data handled by the device. [LAND84] Yet
another alternative is to model devices by modeling their device drivers as TCB subjects. In
general, more sophisticated I/O devices need greater modeling support, with the limiting case
being intelligent workstations modeled as autonomous computing systems.

3.1.5           SECURITY ATTRIBUTES

A security attribute is any piece of information that may be associated with a controlled entity
or user for the purpose of implementing a security policy. The attributes of controlled entities may
be implicit; they need not be directly implemented in data structures. Labels on a multilevel tape,
for example, can be stored separately from the objects they label, provided there is an assured
method of determining the level of each object on the tape. [cf NCSC88b, C1-C1-05-84]

Primary attributes of security policies that are usefully reflected in the security model include
locus of policy enforcement, strength and purpose of the policy, granularity of user designations,
and locus of administrative authority. These policy aspects lead to a taxonomy of policies and
security attributes.

There are several types of security attributes of any given system: informational attributes,
access control attributes, nondisclosure attributes, and integrity attributes. Informational attributes
are maintained for use outside the given computing system, whereas access control attributes limit
access to system resources and the information they contain. Purely informational attributes are
somewhat uncommon; an informative example is given in Section 4.4. Access control attributes
may be classified according to what they control. A loose access control attribute controls access
to the entity it is associated with, whereas a tight access control attribute also controls access to
information contained in that entity. Thus, access restrictions determined by tight attributes must
propagate from one object to another when (or before) information is transferred, because control
over the information must still be maintained after it leaves the original entity. Nondisclosure
attributes are used to prevent unauthorized release of information, whereas integrity attributes are
used to prevent unauthorized modification or destruction of information.

Attributes can also be classified by the granularity and authority of their control. The user
granularity of an attribute may be "coarse," controlling access on the basis of broadly defined
classes of users, or it can be "per-user," controlling access by individual users and processes acting
on their behalf. Finally, centralized authority implies that policy for use of attributes is predefined
and takes place under the control of a system security administrator, whereas distributed authority
implies that attributes are set by individual users for entities under their control. This classification
of security attributes is based partly on the work of Abrams. [ABRA90]

When applied to typical security policies for B2 systems, these distinctions among security
attributes might take the form given in Figure 3.1. MAC policies must enforce nondisclosure
constraints on information and may enforce integrity constraints as well. They must regulate access
to information on the basis of user clearance and labeled sensitivity of data. The involvement of
user clearances typically comes at the expense of per-user granularity because several users are
likely to have the same clearance. Authority to assign security attributes is usually somewhat
centralized. Users can create entities at various levels, but ability to change the levels of existing
entities is usually restricted to authorized users.

DAC policies must enforce access constraints on both reading and writing. They must provide
per-user granularity as criteria for access decisions, although they often include a group
mechanism for coarse gained access decisions as well. DAC policies are normally used to enforce
both nondisclosure and integrity, but constraints on access to named objects do not necessarily
imply corresponding constraints on access to information in those objects. Authority to change
discretionary attributes is usually distributed among users on the basis of ownership.

                MAC     DAC

        Binding Strength:       Tight   Loose

        Purpose:        Nondisclosure10 Nondisclosure & Integrity

        User Granularity:       Coarse  Per-user

        Authority       Centralized     Distributed10



Figure 3.1. Typical Classification of Access Control Policies

3.1.6           PARTIALLY ORDERED SCURITY ATTRIBUTES

A security attribute belonging to a partially ordered set may be referred to as a level. The
associated partial ordering may be referred to as the dominance relation; in symbols, L1 is
dominated by L2 if and only if L1 ú L2. The use of partially ordered levels is usually associated
with tight access control policies and constraints on information flow. Constraints on information
flow can arise for several different reasons. As a result a multifaceted policy might have a different
partial ordering for each reason. Often, the combined effect of constraints associated with several
partial orderings can be expressed in terms of a single composite ordering. In this case, facts about
Cartesian products of partially ordered sets given in Appendix A may be used to simplify the
formal model and, possibly, the system's implementation as well. The following paragraphs
discuss the connection between levels and constraints on information flow, the use of these
constraints for supporting nondisclosure and integrity objectives, and the tailoring of level-based
policies to particular applications.

A dominance relation is often viewed as an abstraction of allowed information flows:
information can flow from entity E1 to entity E2 only if the level of E1 is dominated by the level of
E2. This rather general view, which is an analogue of the original *-property of Bell and La Padula,
allows the illustration of some basic issues in the use of levels, but it is overly simple in some
respects. It does not address whether information flows properly or whether the flow is direct or
indirect. Moreover, this view does not contain any explicit assumption about why information may
be prevented from flowing from one entity to another.

Partially ordered levels and related constraints on information flow have been suggested for
use in enforcing both nondisclosure and integrity objectives. The motivation for these suggestions
may be understood from the following considerations. Suppose that access controls could enforce
the above information-flow constraints perfectly. Suppose that the two objects contain the same
information, but one is labeled at a higher level than the other. In this situation, information is less
likely to be disclosed from the higher-level object because fewer subjects have a sufficiently high
level to receive this information. Conversely, if lower-level subjects can write higher-level objects,
then inappropriate modification of the lower-level object is less likely because fewer subjects have
a sufficiently low level to write to it. This dual relationship may cause the goals of nondisclosure
and integrity to conflict, especially if the ordering on levels is strongly hierarchical. As explained
in Section 3.5, this duality between nondisclosure and integrity has some significant limitations. A
given set of security levels designed for enforcing nondisclosure may be in appropriate for
enforcing integrity, and not all integrity policies use partially ordered security attributes.

When a computing system is accredited for use at a particular site, it is assigned a host
accreditation range - a set of levels at which the host may store, process, and transmit data. A
host range can prevent the aggregation of certain classes of data. If several objects have different
levels and no level in the host range dominates all of them, then there is no legitimate way of
concatenating these objects. This use of host ranges is discussed further in Section 4.4.

A desirable constraint holds between a host range and device ranges: if the host range contains
a level that does not belong to any device range, a non-TCB subject running at that level can not
communicate bidirectionally without using covert channels. By way of contrast, a device range
may reasonably contain levels not in the host range. Suppose, for example, that A and B are
incomparable levels dominated by a level C that does not belong to the host range. A device might
reasonably use C as a maximum device level in order to allow use of the device at either level A
or level B. But, in this case, the system would need to check that each object input from the device
actually belonged to the host range.

3.1.7           NONDlSCLOSURE LEVELS

The following paragraphs discuss the structure of nondisclosure levels as it relates to their
abstract mathematical properties, to TCSEC requirements, and to their intended use. Analyses
given in the following paragraphs suggest that the structure of nondisclosure levels that are used
to enforce nondisclosure policies is often not needed for modeling purposes. If their structure plays
no significant role in a given model, their inclusion is unnecessary and may limit the applicability
of the model.

The TCSEC requires that nondisclosure levels contain a classification component and a
category component. The hierarchical "classification" component is chosen from a linearly
ordered set and the nonhierarchical "category" component must belong to a set of the form *(C),
the set of all subsets of C, for some set C of categories. [cf NCSC85, Sec. 9.0, Sec. 3.1.1.4] In some
applications, thousands of categories may be needed. In others, only the four clearance levels
"unclassified", "confidential", "secret", and "top secret" are needed, as a result of Executive Order
12356. [REAG82] These facts illustrate the importance of configuration-dependent host
accreditation ranges to remove unused security levels.

As explained in Appendix A, any partially ordered set may be fully embedded in one based
on categories. As a result, TCSEC constraints on nondisclosure levels do not restrict the class of
partially ordered sets that may be used for such levels (although these constraints do affect the use
of human-readable names). The decomposition of levels into classification and category
components can be configuration-dependent and may be given along with the host accreditation
range. This fact is of some interest for computing systems with both commercial and military
applications. [cf BELL90] In this case, the model should be general enough to embrace all intended
configurations.

Even if labels are explicitly assumed to contain classification and category components,
nothing in the TCSEC prevents the addition of vendor-supplied components because such
additions do not invalidate the mandated access checks. Thus, for example, each label could
contain a release date after which the contained information will be valueless and, therefore,
unclassified. If release dates are chronologically ordered, later dates would represent a higher
nondisclosure level. The initial specification of release dates, like that of other nondisclosure
attributes, needs to be handled by trusted path software. The regrading from classified -outdated to
unclassified would be carried out by trusted software in conformance with accepted downgrading
requirements.

A nondisclosure level is, by definition, commensurate with the level of harm that could result
from unauthorized disclosure, and it should also be commensurate with the level of assurance
against unauthorized disclosure that is to be found in the system's TCB and surrounding physical
environment. Various rules of thumb have been worked out to correlate risk with levels of
nondisclosure and assurance. [DOD88a, NCSC85a] While these relationships are not likely to
show up in the security model itself, they may affect the kinds of security attributes that are
included in the nondisclosure level.

3.1.8           UNLABELED ENTITIES AND THE TRUSTED COMPUTING BASE

In addition to controlled entities, there are system resources that either are not user accessible
or are part of the mechanism by which user-accessible resources are controlled. These latter
resources are the software, hardware, and internal data structures which make up the TCB. At
higher levels of assurance, significant effort goes into minimizing the size and complexity of the
TCB in order to reduce the overall effort needed to validate its correct operation.

The TCB typically consists of the implemented access control mechanism and other entities
that must be protected in order to maintain the overall security of the system. A TCB process which
is not part of the access control mechanism may be assigned security attributes and controlled as a
subject in order to help enforce the principle of least privilege within the TCB. Conversely, certain
subjects may need to be included within the TCB because they assign security levels to input,
support trusted user roles, transform data in such a way as to legitimately alter the data security
level, or perform some other security-critical function. A data structure may also be security-
critical, as when it contains user-authentication data, a portion of the audit trail, audit-control data,
or security attributes of controlled entities. It is not always necessary to assign security attributes
to TCB processes and security-critical data structures, but this is often done to enforce the principle
of least privilege within the TCB and to regulate access by non-TCB subjects to security critical
data structures.

If a piece of data can be accessed as a direct effect of a system call (i.e. access directly
specified in a parameter) then it must be accounted for in the interpretation of controlled entities in
such away as to satisfy MAC requirements. But some data structures may not be directly
accessible. Possible examples include security labels, the current access matrix, internal object
names that are not accessible to users of the system, and transient information related to access
control, opening of files, and so forth. A data structure which is not directly accessible does not
have to be labeled. A decision to supply a label may complicate the modeling process, whereas a
decision not to supply a label may increase the difficulty of the covert channel analysis.

While there is no explicit requirement to model the TCB, the model must capture security
requirements imposed by the TCSEC, including reference monitor requirements relating to non-
TCB subjects and the entities they manipulate. [cf NCSC87, Appendices B.3.4, B.7.1] A possible
approach to modeling these reference monitor requirements is discussed in Section 3.2.4. If
significant aspects of the system security policy are embodied in TCB subjects that are exempt
from modeled access constraints on non-TCB subjects, then exempt subject modeling is also
needed. This topic is discussed further in Section 3.4.

3.2     NONDISCLOSURE AND MANDATORY ACCESS CONTROL

How can nondisclosure requirements be accommodated in a model's definition of security?
To what extent can access control succeed in enforcing nondisclosure? What impact do
nondisclosure and access control requirements have on trusted systems design? The first of these
questions is customarily addressed by imposing access constraints on controlled entities (e.g., the
*-property). But various research efforts have contributed additional approaches that provide a
useful context in which to explain how access constraints can support nondisclosure objectives.
Both traditional and newer research-related approaches are discussed below. The ordering is top-
down, beginning with nondisclosure requirements on the external system interface, as in the
modeling paradigm described in Section 2.3.

Section 3.2.1 shows how nondisclosure requirements can be formalized in an external-
interface model. In Section 3.2.2, external-interface requirements are elaborated to obtain an
information-flow model, in order to facilitate later analysis. Section 3.2.3 introduces the reference
monitor interface and applies the information-flow requirements to individual subject instructions.
Section 3.2.4 presents an access-constraint model that ensures satisfaction of the information-flow
requirements at the reference monitor interface. Finally, Section 3.2.5 only briefly discusses rules
of operation because of their system-specific nature.

Each of the three models presented in Sections 3.2.1, 3.2.2, and 3.2.4 provides adequate
conceptual support for nondisclosure requirements found in the TCSEC mandatory security
objective and could serve as a definition of mandatory security in a security policy model.
Adequacy of the access-constraint model, in particular, is established by comparison with the
previous two models. Comments regarding the impact of these models on overall system design
are distributed among the various subsections, with the third access-constraint model providing the
most explicit basis for designing rules of operation and judging correctness of implementation.

These models are very simple and need to be adjusted to accommodate policy variations
found in particular trusted systems. These models do not address aggregation and inference. For
sake of simplicity, no accommodation is made for trusted processes or discretionary access control.
The presented access-constraint model is especially relevant to systems in which all user-initiated
computation is performed by subjects outside of the TCB. It may not be adequate for systems
whose TCB contains major trusted applications (e.g., a multilevel DBMS implemented as a TCB
subject). Finally, the entire analysis assumes that the system to be modeled is deterministic in the
sense that its behavior is completely determined by its combined input sequence and data initially
in the system.

3.2.1           EXTERNAL-INTERFACE REQUIREMENTS AND MODEL

Consider a system where each input or output is labeled with a nondisclosure level. Users
work with I/O streams containing items at a given level and provide inputs in such a way that the
level of a given input stream accurately reflects the sensitivity of the informatIon it contains. The
system creates each output item by extracting information from input items (possibly at several
different levels) and affixing an appropriate label. A combination of automated and procedural
methods is used to combine input streams into a single input sequence and to separate the resulting
combined output sequence into separate output streams at different levels.

The actual nondisclosure requirement is this: outputs labeled with a given level must contain
only information whose sensitivity is dominated by that of their label. This requirement is pictured
in Figure 3.2, where lighter shadings represent higher information security levels:  for level A
dominates  for level B, which dominates  for level C.

This requirement is difficult to model (let alone implement) because it talks about the actual
sensitivity of output information, whereas the system is only given the attributed sensitivity of its
data. These difficulties can be avoided with the following alternate requirement: a given labeled
output must not contain information derived from data whose attributed sensitivity fails to be
dominated by the level of the output's label. This alternate requirement applies to both data
supplied in the input streams, and data residing in the system itself. This alternate requirement is
slightly stronger than the original provided that information at a given level cannot be synthesized
by aggregation and inference from information at strictly lower levels. In this case, any classified
information in the system either came from the input stream or was already contained in the system
when it was installed level .

Figure 3.2 Intended Use of a Secure System

In the first external-interface model, each item in the system's input stream is taken to be a
labeled value, as is each item in the output stream. There are two "nondisclosure security"
requirements which are given relative to an arbitrary level L:

Noninterference:

Any output stream at level L remains unchanged when inputs at levels not dominated by L are
altered or removed.

Nonobservability:

Any output stream at level L remains unchanged when data in the system whose sensitivity is
not dominated by L is altered or removed.

These two terms and the requirements they name are similar; the only distinction is that
noninterference discusses independence from high-level input whereas nonobservability
discusses independence from high-level data in the system.

As with any mathematical modeling effort, there is a need to specify the physical
interpretation of the model. There are also choices to be made as to which aspects of the system's
observable behavior are actually addressed by this external-interface model. An "accurate"
interpretation of this model is a physical system whose I/O streams and data are related according
to the above two requirements. In a "complete" interpretation of this model, the input stream would
contain all external influences on the system being modeled; the output stream would contain all
observable effects of the system on its environment; and the data in the system would include all
data that can influence the value of the output stream. An accurate interpretation can be "useful"
without being complete if it includes all outputs associated with normal use as well as enough of
the inputs and system state to predict the included outputs. It is standard engineering practice to
avoid details that would make the modeling problem intractable and then consider them in a
separate covert channel analysis.

The noninterference requirement is due to Goguen and Meseguer. (GOGU82] Under useful
interpretations, this requirement rules out resource-exhaustion channels associated with those
aspects of the system that are covered by the model's interpretation. Suppose that some system
resource (such as memory) is used for processing both low-level and high-level inputs, and that the
processing of low-level inputs cannot proceed when that resource is unavailable. In this case, the
system may not emit a low-level diagnostic explaining the nature of the problem, since the
diagnostic could reveal "interference" by high-level input. The low-level diagnostic, if allowed,
would have provided a resource-exhaustion channel.

The nonobservability requirement was developed as part of the LOCK verification effort. It
covers situations in which classified data is entered during system configuration using information
paths not addressed in the security modeling process. A superficially stronger requirement is that
the system initially contains no classified information. This stronger requirement occurs implicitly
in the original noninterference models of Goguen and Meseguer. [GOGU82, GOGU84] With this
stronger requirement, useful physical interpretations would need to include any classified inputs
that occurred during the construction, installation, or booting of the system.

3.2.2           INFORMATION-FLOW MODEL

The information-flow model is discussed next because of its close relationship to
noninterference. The requirements of this model are motivated by an informal view of how
information flows through a deterministic state-machine system. The possible paths of information
flow during a state transition are depicted as arrows in Figure 3.3, where I, O,  and S abbreviate
input, output, and state, respectively

.

Figure 3.3. Information Flows in a Deterministic State Machine

Each input is assumed to induce a state transition and a (possibly empty) sequence of labeled
outputs. As indicated in the above diagram, there are just four possible flows: directly from input
to output, from an input to the next system state, from a given state to the next state, and from a
given state to an output. Correspondingly, there are four flow-security requirements for the
information-flow model that must hold for any nondisclosure level L. They refer to "portions" of
the state, meaning collections of variables (i.e., state components):

I/O Security:

An output at level L can only be induced by an input whose level is dominated by L.

I/S Security:

An input at level L can affect only those portions of the system state whose levels
dominate L.

S/O Security:

An output at level L can depend only on those portions of the system state whose levels
are dominated by L.

S/S Security:

A portion of the state which is at level L can affect only those portions of the state whose
levels dominate L.

To see the effect of these requirements, suppose, for example, that the current time is
maintained as a state component that is implicitly incriminated by every input instruction
according to the time needed for its execution. The I/S security property implies that the clock level
must dominate every level in the input stream because every input affects this particular state
component. This observation does not imply that all "system" clocks have to be system high,
however. One possibility is to partition time into several disjoint "slices." This partitioning
effectively creates a virtual clock for each time slice. Processes running in a given time slice affect
only the virtual clock associated with that time slice, so that the level of the virtual clock need only
dominate the levels of these processes. Consequently, a process that reads the virtual clock for its
time slice must have a level that dominates those of other processes running in that time slice.

The four requirements of the information-flow model imply those of the nondisclosure-
requirements model. This fact is an example of an "unwinding" theorem in that it shows how to
recast nondisclosure in terms of individual inputs and state transitions. An informal justification of
this fact can be given as follows. First, consider the nonobservability requirement: information at
a given level L in the output stream cannot come directly from portions of the state not dominated
by L (by the S/O security property), and it cannot come indirectly from a preceding state transition
(by the S/S security property). Now consider noninterference: information in the output stream at
a given level L cannot come directly from inputs not dominated by L (by the I/O security property),
and it cannot come indirectly via an intermediate system state as a result of the I/S security and
nonobservability properties.

All four of the flow-security requirements are necessary for noninterference. Moreover, there
is a partial converse: in the case of systems that contain "print" commands capable of outputting
arbitrary state components, the four flow-security requirements follow from the nondisclosure
requirements.

This information-flow model is similar to those of Feiertag, Levitt, and Robinson [FEIE77]
but it makes some minor improvements: an input that causes a state change may also induce output;
a given input may induce outputs at several different levels; and it is not required that each input
be associated with an identified user. The correctness of this model with respect to the
nondisclosure-requirements model is formally proven in [WILL91]. The result is a variant of the
unwinding theorem of Goguen and Meseguer. [GOGU84, RUSH85] An exposition of Rushby's
treatment of the unwinding theorem can be found in the article "Models of Multi level Security."
[MILL89]

3.2.3           APPLICATION TO THE REFERENCE MONITOR INTERFACE

Security policy models traditionally emphasize the reference monitor interface and cover the
processing of subject instructions but ignore issues associated with the sequencing of subject
instructions and their synchronization with external inputs. These ignored issues are handled
separately via covert channel analysis. This is due to the lack of general, well-developed modeling
techniques for dealing with time, concurrence, and synchronization. Subject-instruction
processing, by way of contrast, is readily modeled. In particular, the above external-interface and
information-flow models can be used for subject-instruction processing, just by giving them a new
physical interpretation.

Under this new interpretation, the inputs of the state machine are the instructions that subjects
execute, including system traps whose semantics are defined by the TCB's kernel-calI software.
External system inputs are also included in the case of "input instructions." Each instruction is
executed to produce a state change and zero or more outputs. The outputs include external system
outputs and, possibly, feedback to the unmodeled instruction-sequencing mechanism.

Notice that subjects do not literally execute instructions (since they are untrusted). The TCB
itself executes each subject instruction on behalf of an associated subject controlled by the TCB.
In particular, each hardware instruction available to processes outside of the TCB (i.e., each "user-
mode" instruction) is executed by the CPU, which is part of the TCB. To ensure nondisclosure
security, all subject instructions, including user-mode hardware instructions, must satisfy the
reinterpreted model requirements, either by virtue of the hardware design or by enforced
restrictions on their use. For example, security may be violated if a subject can follow a kernel call
with a "branch-to-previous C ontext" instruction that inadvertently restores all of the access
privileges used during the processing of that kernel call. Instructions implemented by the hardware
but not used by compilers should be modeled if the compilers are bypassable. In the unusual case
where non-TCB subjects can directly execute microcode instructions, these too need to be
modeled.

The actual accesses of a given subject to various objects in its environment are determined by
the instructions it executes (or attempts to execute). Therefore, it is this stream of subject
instructions that the reference validation mechanism must mediate in order to carry out the
recommendations of the Anderson Report. [ANDE72] The application of noninterference and
information flow to subject-instruction processing dates back to [FEIE77] where a form of
noninterference is referred to as "property P1." The validation of noninterference as applied to
LOCK subject-instruction processing is presented in [HAIG87a; FINE90]. Keefe and Tsai have
adapted noninterference for use in modeling DBMS schedulers. [KEEF90] It can be shown that
information flow for subject-instruction processing, together with a variant of noninterference for
subject-instruction sequencing, implies noninterference and nonobservability for the entire system.
[WlLL91]

3.2.4           ACCESS--CONSTRAINT MODEL

An access-constraint model can be obtained by expanding the information-flow model of
instruction processing to include traditional notions of access control, including subjects, objects,
a current-access matrix, and access constraints. This is not a complete access control model in the
traditional sense because it lacks rules of operation. It is a definition of mandatory security for
instruction processing; it does not show how access constraints are actually enforced.

The access-constraint model assumes that the instruction processing state is made up of
labeled state components called objects. The model does not explicitly assume that subjects are
controlled processes, but it does assume that every computation involving either access to objects
or output has an associated subject. Each subject has a nondisclosure level and is assumed to
include its local data space (including stack, program counter, and so forth). Consequently, each
subject is also considered to be an object that could passively be acted on by other subjects. The
system state, st, contains a "current-access matrix," Ast (s, o), that associates each subject-object
pair with a set of "modes." For simplicity, the possible modes are taken to be just "observe" and
"modify."

The requirements of the access-constraint model fall into three groups: traditional
requirements, constraints on the semantics of observation and modification, and I/O requirements.
These requirements are formulated for systems with file-like objects that are opened (in accordance
with simple security and the *-property), accessed for a period of time (in accordance with the
observe-semantics and modify-semantics requirements), and then closed. Each of the eight
requirements must hold in every reachable state.

Simple Security:

A subject may have observe access to an object only if its level dominates that of the
object.

*-Property:

A subject may have modify access to an object only if its level is dominated by that of
the object.

Tranquility:

The level of a given subject or object is the same in every reachable state.

Variants of simple security and the *-property are found in virtually all mandatory security
models. The tranquility requirement can be weakened without compromising the mandatory
security objective, but possibly at the expense of a more complicated model. [cf MCLE88]

The requirements which constrain the semantics of reading and writing are a major factor in
deciding what checks must appear in the rules of operation. Other major factors include the actual
system design and the degree of detail needed in the model.

Observe Semantics:

A subject that executes an instruction whose behavior depends on the value of some state
component must have observe access to that state component.

Modify Semantics:

A subject that executes an instruction which modifies a given state component must have
modify access to that state component.

The observe-semantics requirement is slightly stronger than necessary. A subject (e.g., a mail
program) that knows the existence of two objects at a higher level might reasonably cause
information to be transferred from one to the other by means of a "blind" copy instruction, but this
is directly ruled out by the observe-semantics requirement. As noted by McLean [MCLE90, Sec.
4], Haigh's analysis contains a similar restriction. [HAIG84] A very careful treatment of what
constitutes observation may be found in "A Semantics of Read." [MARC86] The use of a
semantics of reading and writing may be found also in several other security modeling efforts,
including those by Cohen, [COHE77] Popek, [POPE78] and Landwehr [LAND84].

The following requirements constrain allowed associations between subjects and I/O streams.
They assume that each input is read on behalf of an associated subject referred to as its "reader."

Reader Identification:

The decision as to which subject is the reader for a given input must be based only on
information whose level is dominated by that of the reader.

Reader Authorization:

The level of the data read must be dominated by that of the reader.

Writer Authorization:

A subject that executes an instruction which produces an output must have a level that is
dominated by that of the output.

In some implementations, the associations between inputs and subjects are relatively static, and
their validation is straightforward. In others, subjects are created dynamically to service inputs as
they arrive, and explicit modeling may be useful.

If the eight requirements of this access-constraint model are satisfied, then so are the
requirements of the information-flow model. [WILL91] This observation supports the thesis that
access constraints can provide an adequate definition of mandatory security. Related comparisons
of access control models and information-flow models may be found in the works by Gove and
Taylor. [GOVE84, TAYL84] Unfortunately, this access-constraint model also shares some
potential weaknesses of the previous models. In particular, use of the simple security and *-
properties to enforce nondisclosure rests on the following implicit assumption: a non-TCB subject
either outputs information at its own level or outputs information of an unknown lower level that
must, therefore, be given worst-case protection. This assumption ignores the possibility that a
process may be able to produce information which is more highly classified than its inputs through
some form of aggregation or inference. Security models which address this possibility in the case
of database management systems are discussed in Section 4.3.

3.2.5           TAILORING THE MODELS

The remaining tasks in modeling nondisclosure are to tailor the definition of mandatory
security to meet specific system needs and to provide rules of operation describing the kinds of
actions that will be exhibited by the system being modeled. The following paragraphs discuss
adaptations relating to lack of current access and the desirability of modeling error diagnostics,
trusted operations, and nondeterminacy.

In most systems there are some operations that access objects without being preceded by an
operation that provides access permission. For these operations, authorization must be checked on
every access, and either the model or its interpretation must treat the combined effects of the simple
security and observe-semantics properties, and of the *-property and modify-semantics properties.
If this is done in the model, the result is as follows:

Observe Security:

A subject may execute an instruction whose behavior depends on the value of an object
only if its security level dominates that of the object.

Modify Security:

A subject may execute an instruction that modifies an object only if its level dominates
that of the object.

These axioms omit reference to the traditional current-access matrix and are particularly well-
suited to systems that do not have an explicit mechanism for granting access permissions.

Although it is necessary to model unsuccessful execution resulting from attempted security
violations, it is not necessary to model resulting error diagnostics. If the model only covers normal
use of the system, it is both acceptable and traditional to omit error diagnostics, as would typically
be the case in an informal model of a B1 system. For higher evaluation classes, however, an
available option is to give detailed rules of operation that explicitly model some or all error returns.
Their inclusion in the model can provide an alternative to examining them by means of a more
traditional covert channel analysis, as well as additional information for judging correctness of the
system's design and implementation. Error diagnostics resulting from unsuccessful instruction
executions can reasonably be modeled either as output or as information written into the subject's
data space.

A variant of the above modeling strategy has been carried out for the LOCK system. The
LOCK verification is based on noninterference and nonobservability applied to subject-instruction
processing (as opposed to the entire system). The inputs consist of LOCK kernel requests and user-
mode machine instructions. LOCK uses a "conditional" form of noninterference in which certain
"trusted" inputs are explicitly exempted from the noninterference requirement. The LOCK model
was elaborated by means of an unwinding theorem and then augmented to obtain an access control
model. Technically, the LOCK noninterference verification is an extension of traditional access
control verification because the first major step in proving noninterference was to verify the
traditional Bell & La Padula properties for the LOCK access control model. This access-control
verification represents about half of the LOCK noninterference verification evidence. The LOCK
developers compared noninterference verification with a traditional covert channel analysis
technique based on shared resource matrices. [HAIG87, FINE89] They have since concluded that
the noninterference approach is preferable, especially if both nonobservability and noninterference
are verified, because of the extensive hand analysis associated with the shared resource matrix
approach.

Noninterference has been generalized to nondeterministic systems in several ways. A variety
of nonprobabilistic generalizations has been proposed for dealing with nondeterminacy, but they
do not provide a full explanation of nondisclosure because of the possibility of noisy information
channels. [WITT90] Despite this limitation, nonprobabilistic generalizations of noninterference
provide useful insight into the nature of security in distributed systems. An interesting result is that
a security model can be adequate for sequential systems, but is not adequate for a distributed
system. This is because the process of "hooking up" its sequential components introduces new
illegal information channels that are not addressed by the model. [MCCU88a] A state-machine
model that overcomes this lack of "hook-up" security has been provided by McCullough.
[MCCU88] It relies on state-transition techniques and, like the original Goguen and Meseguer
models, has a demonstrated compatibility with traditional design verification methodologies.

3.3     NEED-TO-KNOW AND DISCRETIONARY ACCESS CONTROL

Discretionary access control (DAC) mechanisms typically allow individual users to protect
objects (and other entities) from unauthorized disclosure and modification. Many different DAC
mechanisms are possible, and these mechanisms can be tailored to support a wide variety of user-
-controlled security policy objectives. Users may impose need-to-know constraints by restricting
read access and may guard the integrity of their files by restricting write access. As explained
below, the use of group names may also allow specific objects and processes to be associated with
specific user roles in support of least privilege. Discretionary security mechanisms are more varied
and tend to be more elaborate than mandatory mechanisms. The policy requirements for them are
weaker in order to allow for this variation. As a result, a well-understood discretionary security
model can play a larger role both in clarifying what is provided in a particular system and in
encouraging an elegant security design.

Traditionally, systems have been built under the assumption that security objectives related to
DAC are both user-enforced and user-supplied. A variety of well-known weaknesses are traceable
to this assumption. By way of contrast, vendor cognizance of user security objectives allows the
development of a DAC security model whose mechanisms correctly support higher-level, user-
enforced security policies. Moreover, modeling of these higher-level policies would provide a
suitable basis for validating correctly designed DAC mechanisms and for supplying guidance on
their use for policy enforcement.

DAC mechanisms and requirements are summarized in Section 3.3.1. Group mechanisms and
their use in supporting user roles are covered in Section 3.3.2. Section 3.3.3 discusses traditional
weaknesses in meeting common user-enforced security objectives and Section 3.3.4 presents
mechanisms that overcome some of these weaknesses. Further information on DAC mechanisms
may be found in A Guide to Understanding Discretionary Access Control in Trusted Systems.
[NCSC87a] The formalization of control objectives such as need-to-know and least privilege, as
well as the subsequent verification of access control mechanisms with per-user granularity, are
research topics that have yet to be adequately explored.

3.3.1           DAC REQUIREMENTS AND MECHANISMS

Separate DAC attributes for reading and writing are traditional but not required. DAC security
attributes must be able to specify users both explicitly and implicitly (i.e., by specifying a user
group whose membership might be controlled by another user or user group). For systems at B3
and above, DAC attributes must give explicit lists of individuals and groups that are allowed or
denied access. A wide variety of relationships among individual and group permissions and denials
are possible. [cf LUNT88]

The assignment of DAC attributes may be carried out by direct user interaction with the TCB,
by (non-TCB) subjects acting on behalf of the user [NCSC88b, C1-CI-01-86], or by default. This
last alternative is needed in order to ensure that an object is protected until such time as its DAC
attributes are set explicitly. [NCSC88b, C1-CI-03-86] The default attributes for an object may be
inherited (as when a new object is created in UNIX by copying an object owned by the user) or
they may be statically determined.

A user who has responsibility for assigning DAC attributes to an object may be regarded as
an owner of the object. Typically, the user who creates an object has responsibility for assigning
DAC attributes and is thus the initial owner in this sense, but this is not a requirement. [NCSC88b,
C1-CI-03-85] In UNIX, each file has a unique explicit owner, but root can also modify DAC
attributes and it is thus an implicit co-owner. Some systems provide for change of ownership and
for multiuser ownership; others do not. A variety of issues arise in the case of multiple owners.
May any owner grant and revoke access, or are some owners more privileged than others? How is
coordination among owners achieved; is agreement among owners required? The answers can
differ for read and write accesses and for granting and revoking. Analogous issues arise when
transfer of ownership occurs. A traditional approach to these issues is to let any owner grant or
revoke access or ownership; another is to adopt a principle of "least access," so that all owners must
grant a particular kind of access in order for it to become available. [cf LUNT88]

By tradition, the entities controlled by the DAC mechanism must include all user-accessible
data structures controlled by MAC. However, the explicitly controlled entities (i.e. named objects)
may be different from MAC storage objects. Examples where storage objects and named objects
differ are found in some database systems (see Section 4.3). Operating systems can also have this
feature. For example, files might be named objects which are made up of individually labeled
segments that are the storage objects.

3.3.2           USER GROUPS AND USER ROLES

Access control lists determine triples of the form [user/group, object, access-mode] and
thereby provide a limited variety of triples of the sort used for role enforcement. In fact,
generalizing the mechanism to allow arbitrary programs in place of access modes would provide a
general mechanism of the sort used for role enforcement, as was discussed at the end of Section
3.1.3. In the case of a trusted role, all associated programs would belong to the TCB.

Some systems allow a user to participate in several groups or roles. Some of these systems
require a user to set a "current group" variable for a given session. In this case, the user would run
only programs with execute access in the current group, and they would access only files associated
with that group. Again, the effect is to create a three-way constraint involving user groups,
processes, and storage objects. In order for a user to participate in several groups or roles, it must
be possible for some groups to be subgroups of others or for a user to be associated with several
different groups. The possibility of subgroups involves some interesting implementation issues.
[SAND88] In the SMMS [LAND84], groups represent user roles, each user has a unique
"individual" role, there is a notion of current role, and a given user may have several current roles.
This use of individual roles or groups is a way of tying individual accesses to group accesses so
that the two kinds of access do not have to be discussed separately: individual access is implicitly
given by the rules for group access applied to one-member groups.

The ability to define user groups may be distributed. It is usually assumed that the class
authorized to define user groups is larger than the class of system security personnel but smaller
than the entire user population. A group associated with a trusted user role (e.g., downgrader,
security administrator) would necessarily be controlled by a system administrator.

The owner of a group and the users who specify access to their objects in terms of that group
need to clearly understand both the criteria for group membership and the entitlements associated
with membership in that group. Such understandings depend partly on the mechanics of the group
mechanism, which may be clarified by including groups in the security model.

3.3.3            SOURCES OF COMPLEXITY AND WEAKNESS

In security policies and definitions of security, complexity tends to inhibit effective user
understanding. As such, it is a weakness that, in some cases, may be offset by accompanying
advantages. The following paragraphs discuss several sources of complexity and weakness in DAC
mechanisms including objects that are not disjoint, the coexistence of two or more access control
mechanisms, discrepancies between allowed and authorized accesses, and the use of "set-user-id"
to allow object encapsulation. Finally, the most common weakness found in DAC mechanisms is
discussed, namely, that they are loose; that is, they control access to objects without necessarily
controlling access to the information they contain.

It can happen that named objects overlap so that a given piece of data can be associated with
several different sets of security attributes. In this case, they can be called disjoint. This is
essentially what happens if ownership is associated with file names rather than with files. In this
case, a given piece of data can have multiple owners, each of whom can give or revoke access to
it. As with mandatory access controls, a lack of disjointness tends to interfere with one's ability to
determine allowed accesses. It usually implies that named objects are different from storage
objects, a fact which is a source of complexity that may have some advantages. Discretionary
access may be controlled to a finer or coarser level of object granularity than mandatory access.
Aggregation problems can be addressed by allowing some objects to be subobjects of others and
by imposing stricter access controls on an aggregate object than on its components. In the case of
aggregate objects, there is the additional problem that a change in the permissions of the aggregate
may logically entail a change in the permissions of its components.

When a C2 or B1 system is created by retrofitting security into a previously unevaluated
system, it may happen that the new mechanism supports access control lists in addition to the old
discretionary mechanism supported. In this case, the discretionary portion of the security model
can play a useful role in showing how the two mechanisms interact. [BODE88]

If a process has its discretionary permission to access an object revoked while the process is
using it, some implementations allow this usage to continue, thereby creating a distinction between
authorized and allowed accesses. This distinction is both an added complexity and a weakness that
needs to be modeled in order to accurately reflect the utility of the mechanism. The principal reason
for allowing this distinction is efficiency of implementation. However, in virtual memory systems,
immediate revocation can often be handled efficiently by deleting access information in an object's
page table. [KARG89]

Although discretionary access is ordinarily thought of as relating to users and objects,
processes also acquire discretionary permissions, either dynamically when they are executed, or
statically from their executable code. With dynamic allocation, the permissions may be those
associated with the user who invoked the process; the process's user id would be an example. In
command and control systems, there is often a "turnover" mechanism in which the user id of a
process can change in order to allow a smooth transfer of control when a user's shift ends. In
UNIX, the user id of a shell program changes in response to a "switch user" command.

With static allocation, security attributes might be associated with a subject on the basis of its
executable code. The UNIX "set user id" feature provides an example of this type of allocation.
The owner of a program can specify that it is a "set-id" program, meaning that it will run with the
owner's "effective user id" and thereby assume its owner's permissions when executed by other
users. The purpose of this mechanism is to allow programmers to create "encapsulated" objects.
Such an object is encapsulated by giving it owner-only access, so that it can be handled only by
programs whose effective user id is that of the owner. Other users can access the object by only
executing "encapsulating" set-user-id programs that have the same owner as the encapsulated
object. The UNIX set-user-id option is a source of added complexity, and, as explained below, is
vulnerable to a variety of misuses. Other methods of object encapsulation are thus well worth
investigating. Suggested patches to the UNIX set-user-id mechanism have been considered in the
design of Secure Xenix [GLIG86], and modeling of the set-user-id mechanism itself is illustrated
in the TRUSIX work. [NCSC90b]

Discretionary access controls are inherently loose. This can cause information to be disclosed
even though the owner has forbid it. For example, breaches of need-to-know security may occur
when user i, who owns a file f gives permission to access information in f to user j but not to user
k, and then k indirectly obtains access to f. This can happen in a variety of ways, including the
following:

╖ j copies f to a file that k has read access to.

╖ a Trojan horse acting under authority of either i or j gives k a copy of f.

╖ a Trojan horse acting under authority of i gives k read access to f.

╖ a poorly designed set-user-id program created by j is run by k (with the discretionary
permissions of j) and is misused to give k a copy of f.

Each case involves an extension of access that is analogous to downgrading in MAC. If this
extension occurs without the knowledge and permission of f owner, then the intent of the owner's
protection is violated.

The first three of the above five weaknesses are specific to policies dealing with
nondisclosure, and they need not carry over to other policies dealing with unauthorized
modification. The fourth and fifth weaknesses, by way of contrast, involve tampering with DAC
security attributes by non-TCB subjects, a feature that seriously affects any policy relying on DAC
attributes.

The first of the above weaknesses appears to be an inherent aspect of traditional DAC
mechanisms. An interesting fact about these mechanisms is that there is no general algorithm that
can decide, for an arbitrary system and system state, whether a given user can ever obtain access
to a given object. [HARR76] As explained below, such weaknesses are not forced by DAC
requirements, despite their prevalence in existing systems.

3.3.4            TIGHT PER-USER ACCESS CONTROL

Tight controls on the distribution of information within a computing system originate from
efforts to provide DAC-like mechanisms that have useful information-flow properties [MILL84]
and from efforts to provide automated support for "ORCON" and similar release markings that are
used in addition to security classifications. (ISRA87, GRAU89] Some typical release markings
are:

ORCON-dissemination and extraction of information controlled by originator

NOFORN-not releasable to foreign nationals

NATO-releasable to NATO countries only

REL -releasable to specified foreign countries only

EYES ONLY -viewable by members of specified offices only

PERSONAL FOR -releasable to specified individuals only.

Release markings are used by the originator(s) of a document for providing need-to-know
information. Some release markings such as NOFORN, NATO, and REL <>, have coarse user
granularity and, as explained in Appendix A.4, can be handled via nondisclosure categories. Others
have per-user granularity but, unlike traditional DAC mechanisms, restrict access to both the
document and the information it contains. A catalogue of release markings and an accompanying
language syntax may be found in the article "Beyond the Pale of MAC and DAC - Defining New
Forms of Access Control." [MCCO90]

Tight access control mechanisms designed to support need-to-know differ from traditional
DAC mechanisms in several crucial respects. The explicitly controlled entities (i.e., "named
objects") include processes as well as data structures. In addition, a user's ability to modify their
security attributes is highly constrained. The first such mechanism was proposed by Millen
[MILL84]; a minor variant of it follows.

Each controlled entity is associated with two sets of users, a "distribution" set and a
"contribution" set. These are obtained by evaluating the entity's "access expression" whenever the
object is involved in an access check. Access expressions are the DAC attributes of the policy; their
semantics explain the interplay among individual and group authorizations and denials. For the
sake of brevity these are not modeled. The distribution and contribution sets enforce nondisclosure
and integrity constraints, respectively. Smaller distribution sets represent a higher level of
nondisclosure, and smaller contribution sets represent a higher level of integrity. The empty set is
both the highest nondisclosure level and the highest integrity level. Information may flow from
entity f to entity g, provided the distribution set for f contains the distribution set for g and the
contribution set for f is contained in the contribution set for g.

In Millen's model, devices have a controlling influence on user behavior. A terminal, for
example, is modeled as two separate devices-a keyboard for input and a screen for output. When
a user logs in, the distribution and contribution sets for both the keyboard and the screen are set (by
a secure terminal server) to {i}. Communication from other users (or files that they own) is enabled
by extending the contribution set for the screen. Communication to other users is enabled by
extending the distribution set for the keyboard. The keyboard contribution set and the screen
distribution set must always contain {i} in order to reflect the actual presence of the person using
the terminal.

The distribution sets and contribution sets of Millen's model may be viewed as levels whose
partial ordering is directly tied to an information-flow policy. Consequently, this DAC mechanism
is tight enough to control information flow, and covert storage channel analysis can be used to
check the extent to which distribution sets control disclosure in an actual implementation. In
Millen's model, DAC permissions for an object are set (by its creator or by default) when it is
created and are never modified thereafter. As a result, DAC Trojan horses of the sort discussed in
Section 3.3.3 are impossible. The ability to dynamically change discretionary attributes without
losing tightness would require some nontrivial additions to the policy. Expansion must be done
only by trusted path software in response to appropriate owner authorization. Ownership must
propagate so that it is associated not only with controlled entities but also with the information they
contain. One option is for ownership to propagate in the same way as contribution sets, with each
owner supplying access constraints. A slightly different option is suggested in McCollum's work.
[MCCO90]

Similarities between MAC and the tight DAC of Millen's model suggest the possibility of a
single mechanism meeting both sets of requirements. Moreover, both sets of requirements stem
from the same policy objective in Security Requirements for Automated Information Systems
(AISs). [DOD88a] The consistency of the two sets can be informally justified as follows: assume
the system supports co-ownership by maintaining separate distribution and contribution sets for
each owner, taking intersections in the obvious way. Each input must be co-owned by a system
security officer (SSO), and there is a trusted path mechanism that allows a user to select the SSO's
distribution set from a collection of SSO-controlled groups. For example, the groups might be TS,
S+, and U+ where, by definition, U+ is the group of all users, S+ is the union of TS and the secret
users, and TS is the top secret-users. Users are instructed to select the SSO's distribution set
according to the sensitivity of their data, and, as a result, the main MAC requirements for labeling
and access control are satisfied.

A disadvantage of tight DAC is that there are many innocuous violations of the policy, as, for
example, when a user creates a file with owner-only access and then mails it to a colleague. Pozzo
has suggested that, if the user is authorized to extend access, a trusted path mechanism should
interrupt the program causing the violation in order to obtain the user's permission, thereby
minimizing the inconvenience associated with tight control. [POZZ86] Another strategy for
minimizing unnecessary access violations, which has been suggested by Graubartis, is to allow
DAC security attributes to float so that when a process reads a file, for example, the distribution
list for the process would be intersected with that of the file. [GRAU89] A disadvantage of
propagated ownership is that the set of owners tends to expand, and this is inconvenient if all
owners must agree on access control decisions.

3.4     TCB SUBJECTS-PRIVILEGES AND RESPONSIBILITIES

TCB processes are often exempt from some of the access constraints placed on non-TCB
subjects and are, therefore, able to access data and perform actions that are not available to non-
TCB subjects. The responsible use of such exemptions by the TCB is properly part of the system
security policy. Exemptions in this sense are not a license to violate policy. If a TCB process is
exempt from some of the constraints placed on non-TCB subjects but not others, it may be useful
to treat it as a TCB subject so that the TCB can enforce those constraints from which it has not been
exempt.

The trusted-role programs identified in Section 3.1.3 are usually exempt in this sense because
they have access to security critical data structures and, in many cases, address role-related
extensions of the basic system security policy. Device-related subjects are also likely to be exempt.
A secure terminal server needs to handle inputs at a variety of security levels and may be
implemented as a TCB subject exempt from some of the MAC constraints. A print server is also
likely to be implemented as a multilevel TCB subject because of the need to save, label, and print
files at several different security levels.

In some cases, exempt subjects conform to the same overall policy that the system enforces
on non-TCB subjects. In others, exempt subjects provide limited but significant extensions to the
basic system policy. As a result, the presence of unmodeled exempt subjects can make it difficult
to determine the actual system security policy by looking at the security policy model. [cf
LAND84] To the extent that the policy for exempt subjects differs from the policy described in the
system security model, the validity of the model as a representation of the system is compromised
as are assurances derived from an analysis of the model. The extent of the compromise tends to be
influenced by the extent to which such exempt subjects interact with nonexempt subjects.

The actual rules enforced by the system include both what may be done by non-TCB subjects
and what is done by TCB subjects. Unmodeled special cases can be avoided by directly addressing
the policies associated with exempt subjects in the model. [cf ABRA90] The following paragraphs
address the modeling of exemptions and their legitimate use by TCB processes. There is no explicit
requirement to model TCB subjects and their exemptions from access control, but there may be
implicit modeling requirements that apply to some subjects, especially those that are directly
involved in the access control mechanism. In the case of subjects exempt from mandatory access
checks, it is often appropriate to substitute covert channel analysis for explicit modeling.

3.4.1           IDENTIFYING PRIVILEGES AND EXEMPTIONS

If the principle of least privilege is followed in allocating exemptions to TCB subjects, then
the extent of a subject's exemptions are both a partial measure of the need to model its behavior
and an indicator of what should be modeled. This information indicates the extent to which a TCB
subject's privileges may be abused and, thus, the amount of assurance that must be provided
regarding the subject's correctness. Information on the extent of a subject's exemptions can be
provided by an explicit identification of exemptions, by its security attributes, and by specific
knowledge of the resources accessible to it. These identification techniques provide a variety of
techniques for modeling and implementing exemptions.

A useful first step in describing exemptions is to classify them according to various kinds of
security requirements, such as mandatory access control, discretionary access control, auditing,
and service assurance. The purpose of this classification is to guarantee that any process which fails
to have a particular kind of exemption will not interfere with the corresponding kind of security
requirement. Each named exemption or "privilege" is associated with a particular kind of
functionality allowed by that exemption. Ideally, this functionality should be available only
through possession of the named exemption. As with other security attributes, it is important to
know how a process inherits its exemptions (e.g., dynamically from its parent or statically from its
executable code). Whether this classification is an explicit part of the model will depend largely on
whether it appears in the system design. Including exemptions explicitly in the design simplifies
the analysis of exempt subjects but also complicates the access control mechanism.

In the Army Secure Operating System (ASOS), exemptions are an explicit part of the access
control mechanism. [DIVI90] The association between exemptions and kinds of security allows
the model to assert, for example, that any process (trusted or otherwise) satisfies the *-property
unless it has the "security-star-exemption." To possess an exemption, a process must receive that
exemption both statically during system generation and dynamically from its parent process. Thus,
most ASOS subjects can never have exemptions. Those subjects that can have exemptions will
only run with the exemptions when they are necessary. The TRUSIX work also illustrates the use
of exemptions for the case of a trusted login server: two exemptions and associated transformations
allow (a new invocation of) the server to change its real user id and raise its security level.
[NCSC90b]

The DAC mechanism can be extended to identify TCB subjects and security-critical objects
in the following way. The system has one or more distinguished pseudousers. Some or all security-
critical data structures are owned by these pseudousers. A named object owned by a pseudouser
has owner-only access, and DAC is designed in such a way that nothing can ever alter the access
permissions of an object owned by a pseudouser. (This implies that the system's DAC mechanism
is free from some of the traditional weaknesses mentioned in Section 3.3.3.) Only TCB subjects
are allowed to act on behalf of pseudousers and, therefore, are the only subjects capable of
accessing objects owned by their pseudousers. Thus, if the audit trail, for example, is owned by an
"auditor" pseudouser, then editing of audit files can be performed only by specially designed TCB
subjects acting on behalf of the auditor, if at all.

The MAC mechanism can also be extended to allow partial exemptions. Security-critical data
structures can be treated as objects with special security levels possessed only by TCB subjects.
More significantly, subjects can be partially exempt from the *-property. Each subject is provided
with two separate security labels, an "alter-min" label that gives the minimum level to which a
subject may write and a "view-max" label that gives the maximum level from which it may read.
A process is partially trusted to the extent that its alter-min level fails to dominate its view-max
level. [SCHE85, BELL86, DION81] A straightforward application of partially trusted processes is
found in the GEMSOS design. [SCHE85] Each process has two security labels and is classified as
"single-level" or "multilevel," according to whether the two labels are equal or distinct. Each
security label has separate nondisclosure and integrity components. The nondisclosure component
of the view-max label must dominate the nondisclosure component of the alter-min label, whereas,
the integrity component of the alter-min label dominates the integrity component of the view-max
label because of the partial duality between nondisclosure and integrity mentioned in Section 3.1.6.

Finally, as an example of the above identification techniques, consider the requirement that
only certain administrative processes can access user-authentication data. One possibility is to treat
this independently of other policy requirements. User-authentication data can be stored in
unlabeled TCB data structures that are not available to non-TCB subjects or even most TCB
subjects. This nonavailability constraint (and exemption from it) might be implemented via
hardware isolation or by an explicit software exemption mechanism. This approach could be
explicitly modeled.

A second possibility is to permanently place user-authentication data in named objects owned
by an "administrator" pseudouser and allow only certain programs to run on behalf of this
pseudouser. A (fixable) drawback of this approach is that such programs may need to run on behalf
of actual users for auditing purposes.

A third possibility is to place user-authentication data in a storage object which has a unique
security level that is incomparable with any level available to users of the system. Only certain
administrative programs are allowed to run at this level; such programs may need to be partially
trusted in order to access data at other security levels. Notice that other TCB-only security levels
may be needed to ensure that these administrative programs do not have access to other security-
critical data structures which are also being protected by the MAC mechanism.

3.4.2           RESPONSIBLE USE OF EXEMPTIONS

In modeling an exempt subject, the goal is to prohibit abuses of privilege that might result
from exemptions by placing explicit constraints on that subject's behavior. The main challenge in
formulating these constraints is to achieve an appropriate level of abstraction. In most cases, the
requisite security requirement is considerably weaker than a detailed statement of functional
correctness. The following paragraphs first discuss subjects that conform to the overall policy
enforced for non-TCB subjects and then discuss those that do not. These latter subjects include,
primarily, the trusted-role programs identified in Section 3.1.3.

An exempt subject that conforms to the basic policy illustrated by the modeling of non-TCB
subjects usually does not require separate modeling unless it supports a significant extension of the
TCB interface that is not covered by the general model. This is the case of a subject which is also
a trusted DBMS, for example. However, the modeling of a policy-conforming exempt subject can
provide additional insight into its proper design and is particularly valuable if the subject is visible
at the user interface. For example, a scheduler is user-visible, should be policy-conforming, and
could be treated as an exempt subject whose behavior is justified either by modeling
noninterference-like requirements [cf KEEF90; MAIM90] or by performing a covert channel
analysis.

A secure terminal server is another example of a user-visible, policy conforming, exempt
subject. The main security-critical requirements for a secure terminal server are that each user and
terminal have a trusted path (or paths) for exchanging security-critical information with the TCB
and that there be no "cross talk" between a given trusted path and any other I/O channel. The issue
of how logical channels are multiplexed onto a given terminal is not security-relevant, as long as
it is correct. The fact that the multiplexing includes an Identification and Authentication (I&A)
protocol is security-relevant, but modeling details of the I&A mechanism is unlikely to add much
assurance unless it is accompanied by an analysis of subvertability.

In the case of a trusted-role program, misuse and resulting breaches of security can be
prevented through a combination of software checks and procedural constraints on correct use of
the program. If the program and the trusted role it supports are designed together, then the design
analysis can identify errors of use and can determine whether they are easily detected through
automated checks. The automated checks are appropriately covered in a trusted-process security
model. Knowledge of errors that are not caught by the automated checks can be recast as informal
policies associated with the trusted role itself. These procedural policies are reasonably included
in the system's Trusted Facility Manual. A reader of the high-level system documentation should
be able to see how a given misuse of a trusted role is inhibited through an appropriate combination
of software mechanisms required by the model and procedural constraints found in the definition
of the trusted role. One should be able to see, for example, how a system administrator is inhibited
from falsifying evidence about the behavior of other users by editing the audit trail.

Security properties for trusted-role processes often involve constraints that hold over a
sequence of events, as opposed to constraints on individual state transitions. The use of locks to
ensure proper sequencing of events may be needed, at least in some cases. [cf LAND89] Modeling
of trusted-role processes is often omitted on grounds of restricted use and lack of accepted
examples, but this argument is weak because the higher level of user trust is offset by a greater
potential for abuse.

3.5     INTEGRITY MODELING

Although both integrity and nondisclosure are important for both commercial and military
computing systems, commercial systems have historically placed greater emphasis on integrity,
and military systems more emphasis on nondisclosure. Accordingly, guidelines on integrity policy
are under development by the National Institute of Science and Technology (NIST).

Although the TCSEC does not impose specific requirements for integrity policy modeling, it
does provide for vendor-supplied policies and models. As mentioned in Section 1.3.2, DoD
Directive 5200.28 includes both integrity and nondisclosure requirements. In addition, the NCSC
evaluation process accommodates any security policy that meets minimum TCSEC requirements
and is of interest to the Department of Defense.

Although the following paragraphs do not offer guidance on the formulation of system
integrity policies, they do consider relevant modeling techniques and TCSEC requirements. A brief
discussion of integrity objectives, policies, and models is given in order to provide an overall
picture of the field of security policy modeling. This is followed by a brief taxonomy of integrity-
related concepts and their relationship to security modeling. Topics covered in the taxonomy fall
into two broad areas: error handling and integrity-oriented access control. Examples relating to
TCB integrity are included with the taxonomy as indications of possible relationships between
integrity modeling and related assurance issues for the TCB. The presented taxonomy is based
largely on the one found in "A Taxonomy of Integrity Models, Implementations and Mechanisms."
[ROSK90]

3.5.1           OBJECTIVES, POLICIES, AND MODELS

The term integrity has a variety of uses in connection with computer security [cf RUTH89],
all stemming from a need for information that meets known standards of correctness or
acceptability with respect to externally supplied real-world or theoretical situations. A closely
related need is the ability to detect and recover from occasional failures to meet these standards.
These needs lead to derived objectives for user integrity, data integrity, and process integrity in
order to maintain and track the acceptability of information as it is input, stored, and processed by
a computing system.

Commercial experience as a source of integrity objectives, policies, and mechanisms is
covered in the landmark paper by Clark and Wilson. [CLAR87; KATZ89] Their paper includes a
separation-of-duty objective for promoting user integrity; application-dependent, integrity-
validation processes for ensuring data integrity; the application-dependent certification of
"transformation procedures" for establishing process integrity; and a ternary access control
mechanism for constraining associations among users, processes, and data. A thorough discussion
of their work is given in the Report on the Invitational Workshop on Integrity Policy in Computer
Information Systems (WIPCIS). [KATZ89]

Their requirement for well-formed transactions suggests that for high-integrity processes,
application-dependent software modeling may be needed as input to the certification process. As
discussed in Section 3.5.3, the access control mechanism of Clark and Wilson determines which
procedures may act on a given object and thereby provides an encapsulation mechanism similar to
those associated with data abstraction in modern programming and specification languages.

3.5.2           ERROR DETECTION AND RECOVERY

In the following paragraphs, general observations on the nature of error handling are followed
by a variety of examples that arise in connection with integrity policies.

An error can only be recognized if an unexpected value occurs or if an unexpected
relationship among two or more values occurs. The possibility of anomalous values and
relationships may be built in as part of the system design, as in the case of audit records and
message acknowledgments. It may be added by an application, be of external origin resulting from
syntactic and semantic constraints on the structure of the application data, or be external in the form
of information held by multiple users (in addition to the computing system).

The detection of anomalous values and relationships is only partially automated in most cases.
Typically, an initial error or anomalous situation is detected, perhaps automatically. This discovery
may be followed by further automated or manual investigation to find related errors and, perhaps,
the root cause of these errors. Finally, corrective action removes the errors and/or inhibits the
creation of new errors.

The automated portions of this three-part, error-handling process are more likely to be suitable
for security modeling. In the case of a partially automated error-handling mechanism, modeling
can help clarify which portions of the mechanism are automated (namely, those that are modeled).
If detection is automated and recovery is manual, there may be additional issues associated with
the design of an alarm mechanism (e.g., timeliness, avoidance of masking high-priority alarms
with low-priority alarms).

As already mentioned, the TCB audit mechanism required for systems at classes C2 and
above is a built-in, error-rejection mechanism. It is usually not modeled but could be; Bishop has
provided an example. [BISH90] Recent integrity articles have suggested that audit records should
allow for a complete reconstruction of events so that errors can be traced to their source. [SAND90,
CLAR89] This integrity requirement on audit trails is potentially amenable to formal analysis.

"Checkpoint and restore" is another built-in mechanism found in many systems. Modeling of
this mechanism may be no more interesting than for auditing. However, systems at classes B3 and
above have a trusted-recovery requirement. The state invariants that come out of the formal
modeling and verification process effectively define exceptional values of the system state that
must not exist in a newly restored state. Moreover, run-time checking of these state invariants may
be necessary after an unexplained system failure.

A variety of integrity mechanisms promote user integrity by collecting the same or related
information from several different users. Supervisory control and N-person control allow two or
more users to affirm the validity of the same information. Supervisory control involves sequential
production and review of an action, whereas N-person control refers to simultaneous or
independent agreement on taking action. A possible approach to modeling N-person control is
given by McLean. [MCLE88, Sec. 3] The subjects of the model include composite subjects made
up of N ordinary subjects acting on behalf of different users. The explicit inclusion of these N-
person subjects invites questions about their security attributes; for example, what is their security
level? In both supervisory and N-person control, lack of agreement is an error condition, and the
system performs error handling by blocking the action being controlled.

Certain access controls may be suspended by any user in an emergency, but the system may
require that a state of emergency be explicitly declared before suspending these access controls.
The explicit declaration together with the actual violation of the controls provides two different
indications of the need for a typical action. The decision to constrain a particular action via
supervisory, N-person, or emergency-override control might be made by an appropriately
authorized user, as opposed to the system's designers. In all three of these mechanisms, the actions
to be controlled need not be vendor-defined.

A related strategy for promoting user integrity is separation of duty. This involves defining
user roles in such a way that no one user can commit a significant error without detection. Instead,
there would have to be collusion among several users. For example, if no one person can order
goods, accept delivery, and provide payment, then it is difficult for a user to buy nonexistent items
from himself. A more generic example recommended by Clark and Wilson is that no person who
is authorized to use a transformation procedure has participated in its certification. Separation of
duty promotes user integrity, if roles are designed in such a way that several different users
contribute overlapping, partially redundant views of an external situation modeled in the
computing system. In most cases, design of the roles is application-dependent. Provisions for role
enforcement are part of the system design. Detection of errors can be either manual or automated
(but application-dependent). Error recovery is largely manual. In the Clark-Wilson model,
separation of duty is enforced via access control triples, but is itself not formally modeled.

According to Clark and Wilson, consistency checks on the structure of user-supplied data are
needed initially to guarantee that data has been properly entered [CLAR87] and can be run later as
a check against inappropriate modification [CLAR89]. Typically, these Integrity Validation
Procedures (IVPs) compare new information against previously computed information, as in a
program that compares actual inventory data against previously computed inventory. The use of
IVPs is explicitly modeled by Clark and Wilson, but the fact that they reject anomalous data is not.

Redundancy to improve process integrity is used in high-availability systems. Two or more
processes with different hardware and/or algorithms are run with the same expected result, and a
voting algorithm combines the results. An interesting feature of these data- and process-integrity
promoting algorithms is that they apparently increase integrity through aggregation and inference.
In this respect, integrity is similar to nondisclosure, not dual to it. Interesting formal models of
voting algorithms include the "Byzantine Generals Problem", in which a process can give
inconsistent results [LAMP82], and clock synchronization algorithms. [LAMP87, RUSH89]

In general, all error-handling mechanisms exploit redundancy of the sort discussed in
information theory and conform to the same general principles that provide the basis for error-
correcting codes [cf HAMM80] used to suppress noise. What sets integrity-related mechanisms
apart is their emphasis on user-induced errors.

3.5.3           ENCAPSULATION AND LEVEL-BASED ACCESS CONTROL

The following paragraphs discuss encapsulation mechanisms, the use of level-based access
control and integrity hierarchies, and the use of partially trusted subjects to achieve encapsulation.

Encapsulation mechanisms ensure what Clark and Wilson refer to as "internal consistency."
They provide a limited set of high-level operations that can be certified to maintain a known
structure on the data that they encapsulate. Several mechanisms suitable for performing
encapsulation have been discussed in Sections 3.1.3, 3.3.2, and 3.3.3. Another encapsulation
mechanism is message passing, as illustrated in the design of Smalltalk. [GOLD80] As discussed
below, level-based access control with partially trusted subjects also provides an encapsulation
capability similar to type enforcement.

If an encapsulation mechanism is supported, it may be used to provide tamper proofing for
the TCB and to enforce the principle of least privilege within the TCB. Type enforcement has been
used for this purpose. [BOEB85] In the LOCK system, each security-critical entity (for example,
the password file) is of a type that can be accessed only by the appropriate TCB subjects. This
typing information is preset and, in the case of TCB entities, cannot be modified, even by the
system security officer. Since the type enforcement mechanism is formally verified, this
verification provides a partial verification of TCB integrity as a special case. Type enforcement has
also been used to extend the LOCK TCB for a trusted DBMS application. [STAC90] In the
extension, new TCB subjects are straightforwardly prevented from interfering with the old part of
the TCB through static access restrictions in the type-enforcement table.

The crucial idea behind access control based on integrity levels was alluded to in Section
3.1.6. That information may flow from one entity to another only if the latter's integrity level is at
or below that of the former, thereby preventing the latter from being contaminated with low-
integrity information. Thus, if the integrity ordering is inverted for purposes of access control, then
the *-property will automatically enforce the desired property. As observed by Roskos, [ROSK90]
this duality applies not only to access control, but to higher-level nondisclosure policies as well. In
the case of noninterference, for example, if inputs from user A cannot "interfere" with outputs to
user "B," then A cannot compromise the integrity of these outputs. [PITT88] The practical utility
of these observations is undetermined.

For level-based integrity to be useful, there must be some convention for assigning integrity
levels to controlled entities. Biba [BIBA77] suggested a hierarchy dual to that given by Executive
Order 12356, which, as mentioned in Section 3.1.7, defines the terms "confidential," "secret," and
"top secret." Thus, integrity levels would be classified according to the level of harm which could
result from unauthorized or inappropriate modification of information. The dual of "secret," for
example, would be a level indicating that inappropriate modification could result in serious, but
not exceptionally grave, damage to national security. While this suggestion has not found wide
acceptance, there are clear-cut examples of integrity orderings associated with the possibility of
serious harm. The inadvertent substitution of "exercise" data for "live" data is one such example;
effectively, "exercise" is a lower integrity level than "live." What Biba's suggestion lacks (in
contrast to the work of Clark and Wilson) is a convention for controlling not only who may modify
data but also how they do it.

There are several other integrity measures not based on level of harm. [cf PORT85] User
contribution sets form an integrity ordering under the dual of the subset relation, as was noted in
Section 3.3.4. A careful analysis of user roles can also give indications of needed integrity and
nondisclosure levels. [LIPN82] The results of trusted error-detection and integrity-validation
procedures can be stored as (components of) integrity levels in order to disable inappropriate uses
of flawed data and to enable appropriate error-recovery procedures.

Integrity levels (or components) that are based on freedom from various kinds of error cannot
be preserved during processing unless the procedures involved do not introduce these kinds of
errors. This motivates the use of software assurance hierarchies. The necessary level of software
assurance is found by working backwards from a level of acceptable risk, which depends both on
the expected level of software errors and the level of harm which may result from these errors in a
particular operating environment (the latter being bounded by the relevant Biba integrity levels).
If software at several levels of assurance is needed on the same system, then an access control
mechanism is needed to be sure that software used in a given application is backed up by adequate
assurance. A technique that suggests itself is to include assurance levels as components of integrity
levels.

A well-defined hierarchy of assurance levels is given in the TCSEC for evaluating TCBs. It
has been adapted to subsystems [cf NCSC88a] and is readily adapted to individual processes; the
hierarchy itself is summarized in Appendix D of the TCSEC. An easily applied assurance hierarchy
based on credibility of origin has also been suggested by Pozzo: [POZZ86] software supplied by a
system's vendor is regarded as having higher assurance than software developed by a company's
support staff, and production software typically has higher assurance than experimental software.
Software of unknown origin has lowest assurance. Such a hierarchy presupposes an ability to track
data sources, as described in DoD 5200.28-R. [DODS8a, Enclosure2, Req. 7]

The use of level-based integrity for TCB encapsulation is suggested by Millen. [MILL84] One
or more integrity categories can be reserved for TCB entities, as a means of protecting them from
non-TCB entities. An application of this idea is found in the network described by Fellows.
[FELL87] In this network, each communication process in the TCB is a partially trusted subject
that sends messages labeled with its own unique integrity category, thereby encoding authorship
of the message in the security label itself.

Lee and Shockley have argued that level-based integrity with partially trusted subjects can
partially fulfill the requirements of Clark and Wilson. [LEE88, SHOC88] The ASOS integrity
policy may be viewed as a special case of this general partially trusted subject mechanism. Access
checking in ASOS allows any process to read below its own integrity level; the implicit assumption
is that any subject with reasonable integrity can be relied on to make its own integrity checks when
needed. In effect, each subject in ASOS is treated as if it had a separate view-max security label
obtained by setting the integrity component of its security label to integrity-low. [DIVI90]

4.      TECHNIQUES FOR SPECIFIC KINDS OF SYSTEMS

Though relevant to many systems, some security modeling techniques are best illustrated by
a particular kind of system. While traditional access control models are applicable to a wide class
of systems, they are nicely illustrated by operating systems because they both contain file-like
objects that are opened, accessed for a period of time, and then closed. The issue of selecting an
underlying model of computation occurs in any modeling effort, but models not based on state
machines have occurred most frequently in networks. Trusted application models often take the
form of models for database systems. Issues having to do with label accuracy are illustrated in
Compartmented Mode Workstations (CMWs) and the Secure Military Message System (SMMS).
The treatment of these topics is largely introductory because of the breadth (and, in some cases,
relative newness) of the material to be covered.

4.1     OPERATING SYSTEMS

In modeling operating systems, there is especially rich literature to draw on. The following
paragraphs discuss traditional access control models and the models of Bell and La Padula in
particular.

4.1.1           TRADITIONAL ACCESS CONTROL MODELS

An access control model traditionally involves a set of states, together with a set of primitive
operations on states. Each state contains a set S of "subjects," a set O of "objects," and an "access"
matrix A. For each subject s and object o, A [s, o] is a set of access rights, such as "read," "write,"
"execute," and "own." In the context of an access control model, rules of operation are axioms or
definitions describing the primitive operations.

The simplest useful access control model is perhaps the HRU model of Harrison, Ruzzo and
Ullman. [HARR76] The HRU model assumes that every subject is also an object, the rationale
being that (almost) every subject has local memory that it can access. In the HRU model, the
primitive operations are, create s, create o, destroy s, destroy o, enter r into A [s, o], and delete r
from A [s, o], where r varies over all supported access rights.

To use an access control model for security modeling, one needs to add functions that extract
security labels from subjects and objects and to explain how operations are affected by security
labels. In some cases, it may also be appropriate to model other attributes of objects, such as
directory hierarchies. These additions to the basic access control model facilitate a clear
presentation, but are not theoretically necessary for discussing security, as can be seen from
Pittelli's translation of a Bell and La Padula model into the HRU model. [PITT87] One can model
the addition of new subjects or objects by allowing s and o to change in response to the create and
delete operations, or one can use fixed sets of subjects and objects together with an indication of
which subjects are active. Finally, if an access control model is to discuss information flow, it
needs to discuss which objects are actually written as a result of a given operation. This can be done
by adding functions that extract object content. A complete information-flow analysis would also
need to address flows to and from non-object entities. Flows to and from the current-access matrix,
for example, could be handled either by treating rows of the matrix as objects or by treating these
flows in the covert channel analysis.

The assumption that subjects are objects is associated with several interesting observations
that need to be modeled whether or not subjects are regarded as objects. Typically, every subject
has read and write access to itself (that is, to its stack and local data area). This fact can be modeled
as the invariant (read, write) ┐ A [s, s]. Consequently, if a subject outside the TCB is allowed to
change its security level, the new level should dominate the old one. This is because information
at the old level in its local memory can be transferred only to entities whose level dominates the
subject's new and (hence) old security level. The assumption that subjects are objects also provides
a means of modeling interprocess communication: a process P1 may send a message or signal to
process P2 if and only if P1 has write access to P2.

Hierarchical file systems have a potential for introducing various covert storage channel
problems. In many cases, these channels can be closed through the use of additional access control
checks that depend on the structure of the directory hierarchy. It is acceptable to include such
access control checks (and hence, directory structures) in the model. One problem is that if a
directory contains a file at a lower security level, then it is difficult to delete the directory without
users at the lower security level finding out, since they can no longer access the file after it is
deleted. A widely imitated solution proposed by Walter is to require a compatibility property for
directories, namely that the level of each file dominate that of its parent directory. [WALT74]
Alternatively, the system could just remove access and leave actual deletion of files and directories
to a trusted garbage collector. A similar problem that is not ruled out by compatibility arises when
a lower-level process creates a directory at a higher level. If a higher-level process then adds a
(higher-level) file to the directory, a UNIX-like operating system would refuse a later request by
the lower-level process to delete the directory, thereby signaling the existence of the higher-level
file. [GLIG86]

The compatibility principle was regarded by the TRUSIX working group as being a generally
useful property of directories, independently of its use to discourage covert channels. They
considered four possible approaches to modeling directories and, for sake of generality, decided
that security labels would be assigned to individual directory entries as well as directories.
[NCSC90b, Sec. 6.10, 6.11]

Kernel call based implementations of operating systems invariably rely on some form of
exception mechanism. At higher assurance levels, it is likely to be a matter of policy that
exceptions not return information about higher-level entities. In this case, kernel exceptions can be
included in the security model in order to represent this policy, either directly or in summary form.
One might, for example, use the three exception classes "access-denied," "user-error," and "system
error". [BELL76]

For further information on access control models, the reader is referred to Cryptography and
Data Security [DENN82, Ch.4] and "Models of Multilevel Security" [MILL89].

4.1.2           THE MODELS OF BELL AND LA PADULA

As mentioned in Section 1.4, the work of Bell and La Padula identified the main steps in the
security modeling process and provided the first real examples of security verification by proving
that their rules of operation preserved necessary state invariants. The identified invariants codified
important mandatory access control requirements and provided guidance for their implementation.

Not surprisingly, fifteen years of close scrutiny has produced an understanding of areas where
refinement of the original approach is desirable in future efforts. The question of correspondence
to externally imposed MAC security policy was handled quite informally. As indicated in Section
3.2 (and in the LOCK verification effort), a closer correspondence is possible with the use of
external-interface models. In fact, the Bell and La Padula models contain some well-known
weaknesses. A lack of attention to local process memory is related in part to the fact that subjects
need not be objects. A subject may, in the absence of the tranquility principle [LAPA73, p. 19],
lower its security level. As a result, the "change subject current security level" and "change object
level" rules provide a variety of opportunities for untrusted downgrading of information. The
significance of these and similar rules was not well understood until much later. [MILL84,
MCLE85, MCLE87, BELL88, LAPA89]

The Bell and La Padula models have a relatively narrow focus. None of the models explicitly
mentions multilevel devices, external interfaces, or user identities; and there is no modeling of
integrity. With the exception of the *-property, exemptions from access control are not modeled,
so that there is no basis for relating exemptions to use of privilege, subject integrity, or trusted user
roles. The Multics model interpretation was necessarily incomplete because design work on Secure
Multics was still in progress in 1976. Some of the notational conventions used in the models (e.g.,
overuse of subscripts and lack of mnemonic naming conventions) are avoided in many of the more
recent security modeling efforts.

4.2     NETWORKS AND OTHER DISTRIBUTED SYSTEMS

Which network decomposition techniques lend themselves to security analysis? How should
this decomposition be reflected in the model? How should the overall network security policy be
reflected in the model? What is the role of individual components in enforcing the overall policy?
Should nondeterminacy and the distributed nature of a network be reflected in the model's
underlying model of computation; that is, should the network be treated as a state machine or as
something else? The following subsections discuss these questions as they relate to network
security objectives and the requirements of the Trusted Network Interpretation (TNI). [NCSC87]

4.2.1           SYSTEMS AND THEIR COMPONENTS

A network system is an automated information system that typically consists of a
communications subnet and its clients. Clients are entities that the subnet interacts with (e.g., host
computers, other networks, users, electronic censors), as in Figure 4.1. Some clients may play a
distinguished role in the network, performing key distribution or network administration services.
The communications subnet might consist of message switches and transmission lines.

Figure 4.1. A Physical Network Decomposition

In addition to a physical decomposition of a network, there is also the possibility of
decomposing along protocol layers, so that a network handling a layer n protocol decomposes into
abstract protocol machines and network(s) handling a more primitive layer n-1 protocol. [cf ISO84,
TANE88] In this second approach, it is possible to reassemble the protocol machines at various
layers into physical components, so that the second approach is compatible with the first. In Figure
4.2, abstract protocol machines are collected to form interface units that may either be part of or
separate from corresponding hosts. Each interface unit includes protocol handlers for all layers.

Figure 4.2. A Network Protocol Decomposition

The partitioning of a network system results in optional subsystem components, such as a
communications subnet, as well as individual components whose further subdivision is not useful
for modeling purposes. [NCSC87, Sec. I.3] In the following paragraphs, network systems and
subsystems are both referred to as networks. Individual and subsystem components are both
referred to as components. Network partitioning has a variety of security-relevant ramifications.
There is a need for both system security models and component models as well as the need to
consider TCB interfaces both in components and in the entire network system. In contrast to
operating systems, there is little point in assuming that a network is either "up" or "down." Rather,
it is necessary to model the network across a range of partially functioning states. [FELL87] The
principle of "mutual suspicion" can be applied to components in order to minimize security
violations in case an individual component is subverted. [SCHA85] In this case, each component
enforces a component policy that is strong enough not only to contribute to the overall system
policy but also to detect suspicious behavior in other components.

Networks are usually extensible, and, as a result, analysis of their security cannot depend on
configuration details that change as the network expands. A similar problem is posed by the need
for fault tolerance. Component failures and subsequent restarts should not lead to violations of
network security policy. [FELL87]

Some networks have covert channel problems related to indeterminate information in
message headers. In some cases, these "header" channels are visible in the model itself. Other
covert channels result from interconnection subtleties and from nondeterminacy caused by channel
noise and by race conditions associated with distributed computation. Similar covert-channel
problems can arise in other nondeterministic systems, and, as mentioned in Section 3.2, a
probabilistic analysis may be needed in order to rule out certain kinds of covert channels.
[WITT9O] However, nonprobabilistic analyses have also provided some useful insights.
[MCCU88, MCCU88a, JOHN88]

4.2.2           MODEL STRUCTURE AND CONTENT

A crucial modeling requirement for networks is that "the overall network policy must be
decomposed into policy elements that are allocated to appropriate components and used as the
basis for the security policy model for those components". [NCSC87, Sec. 3.1.3.2.2] One way of
meeting this requirement is to provide a security policy model consisting of several submodels. A
network security model can give an overall definition of security for the network in order to clarify
how the network supports higher-level security objectives. A structural model can explain how the
network is partitioned. This is particularly useful if the component structure is directly involved in
the network's policy enforcement strategy. Finally, each component is given a security policy
model. Correctness is shown by demonstrating that the component model properties together with
the structural model properties imply the requirements in the network security model. The given
demonstrations may rely either on external interface properties of the components [cf MOORE90,
BRIT84, FREE88] or on modeled internal properties [cf GLAS87].

For a component that is part of a larger network, external-interface requirements will often be
apparent from the network modeling effort. For a separately evaluated component, rated B2 or
higher, external-interface requirements will usually be given in an accompanying network security
architecture description. [NCSC87, Sec. 3.2.4.4, A.3] The NCSC Verification Working Group has
concluded that access control requirements imposed by the network security architecture should
be modeled. Consequently, the definition of security for the component model needs to contain or
comply with such external-interface requirements. It is easier to demonstrate conformance if these
requirements are directly included. This "external" portion of a component model is essentially
new and is distinct from internal requirements or rules of operation. In the case of MAC
requirements, this external interface portion may be regarded as a generalization of TCSEC device
labeling requirements. [NCSC87; Sec. 3.1.1.3.2, 3.2.1.3.4] A model for a network system, in
contrast to a component, may not need such external-interface requirements because the
environment of the network system need not have a well-defined security architecture and is likely
to contain only trusted entities such as users and other evaluated network systems.

Security-critical entities in the network or its containing network system are collectively
referred to as the network TCB (NTCB). A second network modeling requirement is that a
component model should cover all interfaces between security-critical network entities and other
kinds of entities (see Figure 4.3). Collectively, these interfaces contain the reference monitor
interface for the NTCB. Relevant non-NTCB entities for a network might belong either to the
network or to its environment. Whole components of a network may lie outside of the NTCB.
Individual components may also contain non-NTCB entities (as in the case of an ordinary host, for
example).

When modeling a component, it may be desirable to model not only the interface between its
TCB and non-TCB entities but also the interface between its TCB and other NTCB entities in order
to give a better description of the security services provided by that component. For an A1system,
this additional modeling information can provide correctness criteria for the FTLS, since each
component FTLS must describe the interface between that component's TCB and the TCB
portions of other components. [NCSC87, Sec. 4.1.3.2.2] These additional interfaces are indicated
by dashed lines in Figure 4.3.

Figure 4.3. Modeled Interfaces to the TCB Portion of Component A

4.2.3           NETWORK SECURITY MODELS

The security policy model for a network often contains a network security portion, although
this is not required by the TNI. Overall network security modeling is especially useful for a network
that is too large for effective system testing because, in this case, overall security assurance can be
achieved only by some form of static analysis. A network security model can give a precise
description of security services provided by the network and can help to identify the boundaries of
the system being modeled. It can identify security-relevant aspects of communication among major
components and can describe associated security requirements. The entities described in this model
need not be local to individual physical components. This model partially determines and may
share the external-interface requirements found in the various component models.

For network systems rated B1 and above, mandatory access control plays a major role in
modeling user-related communication. [SCHN85] Other useful security requirements include
correct delivery [GLASS87]; discretionary access control; label, header, or message integrity
[GLAS87, FREE88]; and encryption [BRIT84]; among others. [ISO89; NCSC87, Sec.9]

In many networks, there is a significant amount of TCB software devoted to security-critical
communications protocols. A clear understanding of the basis for these protocols is required in
order to completely assess security assurance. A network security model can provide this
understanding by providing requirements that serve as correctness criteria for security-critical
protocols found in the NTCB portion of the network. Label association protocols and label
integrity requirements are of particular relevance in the case of MAC policy modeling.

Access control modeling may be influenced by the network topology. For example, access
enforcement may be carried out by a communications subnet, while access decisions are performed
by a designated host that serves as an access decision facility. [BRIT84]

One approach to specifying mandatory access control is to regard hosts as subjects (partially
trusted subjects, in the case of multilevel secure hosts) and host "liaisons" as "communication"
objects. [BELL86, BELL88a] Host liaisons are temporary connections established between pairs
of hosts for the purpose of exchanging messages at a given security level, such as transport layer
connections. This approach is similar to that taken in several real systems. In the Boeing local area
network (LAN) model the subjects are arbitrary LAN clients. [SCHN85] In the model described
by Fellows subjects are hosts. [FELL87] In this latter modeling effort, the host liaison approach
was found to be satisfactory for expressing system security requirements with the possible
exception of what came to be known as the "entelechy" property. This property means that a host
may send or receive messages over a connection if and only if it has current access to that
connection. These models are internal-requirements models for network systems. Because these
models treat hosts very simply, they also serve as external-interface specifications of their
communication subnets.

A more explicit external-interface model was given for the Multinet Gateway System (MGS).
The clients of the MGS are "external systems" that communicate with the MGS over message
ports. [FREE88] Messages have security labels, message ports are multilevel devices, and the
expected access constraints are enforced between messages and message ports. In addition, the
model contains limited information about the structure of messages and declares that any delivered
message must be legitimately "derived from" messages received by the system at the same security
level. In other words, no spoofing or regrading is allowed. The MGS model and FTLS are, in part,
partitioned along protocol layers.

In some cases lower-level component models involve system-only messages which are not
visible at the system interface. [FELL87, FREE88] In these cases special security labels are used
to separate system-only messages from normal client messages so that privileged information in
system-only messages cannot accidentally leave the system. Lower level component specifications
had to distinguish among several kinds of messages, such as acknowledgments and other
"protocol" messages as well as between encrypted and plain text messages. Messages are split into
header and data components. In networks, some understanding of object structure is essential to
enforcing security, and this understanding necessarily turns up in derived security requirements for
network components.

In some cases, the network model may be trivial and therefore unneeded. A network system
consisting of single-level hosts connected by one-way links, for example, might have only the
obvious requirement that the level of a destination host dominate that of a receiving host. Even this
simple case may be worth modeling, if the cascading problem is considered. Considering the
cascading problem may determine when network connections increase the risk of compromise.
[see MILL88]

4.2.4           INDIVIDUAL COMPONENT SECURITY MODELS

Like other component models, individual component security models should accommodate
relevant external-interface requirements inherited from their containing networks. An individual
component which contains entities that are not security critical must contain a reference monitor,
and its model should constrain the reference monitor interface. In this case, the model will include
a structure analogous to that presented in Section 3.2, including internal requirements and rules of
operation. Correctness of the model can be proved by showing that the rules of operation imply
both internal- and external-interface requirements. For some components, this proof may involve
showing that the internal requirements already imply the external-interface requirements. For
others, the internal requirements may apply only to the internal reference monitor interface, so that
they are unrelated to the external-interface requirements and have properties that are enforced by
different software.

As with other kinds of computing systems, an individual component may contain subjects that
do not act on behalf of particular users; the TNI refers to these as internal subjects. In the case of
networks, it is common for internal subjects to exist outside the TCB, as in the case of routing
protocols which are not involved in security enforcement. In some components, all of the subjects
are internal, and the isolation of untrusted processes is so simple and straight-forward that
modeling of the reference monitor interface is unnecessary.

An individual component may lie entirely within the NTCB. In this case, its external-interface
requirements may already be sufficient as a description of component policy enforcement because
an internal-security model offers no essentially new insights. Alternatively, a simplified functional
description, in the form of an FTLS or rules of operation, may be useful in showing how the
component goes about meeting its external-interface requirements. In this case, the functional
description should provably imply the external-interface requirements.

4.2.5           UNDERLYING MODELS OF COMPUTATION

State machines have been successfully used as a basis for security modeling in several
network evaluations. However, concerns about their convenience for partitioning a network TCB
into components have been expressed by some researchers and system developers. State invariants
may not be testable in practice because they can discuss widely separated state components that
cannot be accessed in a timely fashion. Two elementary state transitions may be concurrent,
leading to a partially ordered view of event histories in which two events can occur without either
fully preceding the other. Parts of the secure state may be replicated in different locations, which
imposes consistency requirements as well as forced discrepancies between the system state and the
Cartesian product of the component states. [FELL87] Because of these perceived inconveniences,
various history mechanisms have been used as an alternative to state machines, including I/O
histories [GLASS87; FREE88], time-stamped buffer histories [BRIT84], and event histories
[MCCU88]. Event histories are essentially the "trace" semantics for Hoare's communicating
sequential processes. [HOARE85]

The differing approaches used in the above security modeling efforts underscore the fact that
a formal security model depends on an underlying model of computation. Models of computation
that could reasonably be used for networks include modal logic with communication based on
shared variables [MANN82, PNUE81], axiomatically defined process algebras [MILN83,
HOAR85], and Petri nets augmented with data flow primitives [AGER82, CHIS85]. Specification
languages based on process algebras [EIJK89] and on communicating state machines [DIAZ89]
have been developed for describing communications protocols and may be adaptable for use in
writing network security models.

In general, the underlying model of computation for a security model should be chosen to
facilitate the statement and proof of the key security properties that will be studied. In the case of
networks, several different options have been tried and many more are plausible, but no preferred
approach has yet been established.

4.3     DATABASE MANAGEMENT SYSTEMS

The field of database security is relatively new and has provided a large variety of questions
and issues, some of which are outlined in the following paragraphs. As a result, there are many
unresolved issues, and significant effort may be needed to obtain a good security design. This
section is based in part on observations drawn from works by Hubbard and Hinke [HUBB86,
HINK90], from the Trusted Database Management System Interpretation of the Trusted Computer
System Evaluation Criteria (TDI), [NCSC91] and from similar security modeling issues for other
trusted applications.

Perhaps the most obvious modeling issue is how the DBMS security model should be related
to the DBMS data model, that is, to the high-level user's model of how the DBMS manages data.
Can the security model serve as a simplified data model? If so, is it consistent with the data model
supplied to users of the system? Are all of the major entities in the data model covered in the
security model? In the case of a relational DBMS, for example, does the model and/or
accompanying documentation explain how relations, views, records, fields, and view definitions
relate to the appropriate entities (e.g.,named objects) of the security model?

Commercial database systems are commonly designed to run on a variety of different
operating systems. This suggests that a database system might be modeled separately from its
underlying operating systems, if there is a common explanation of how its security features interact
with those of the underlying operating systems. Database systems raise a variety of security policy
issues that are not covered in the TCSEC. The large quantity of user-supplied information found in
typical databases increases the potential for erroneous data (and data classifications) as well as the
potential for the unauthorized release of information due to aggregation and inference. However,
database systems can partially automate the labeling of data through the use of content-dependent
classification rules, thereby eliminating some kinds of user errors.

4.3.1           SYSTEM STRUCTURE

Many of the structural considerations regarding database systems apply to application
systems in general. [cf PAYN90] A discussion of these considerations is followed by DBMS-
specific observations about the design of objects for mandatory and discretionary access control.

A DBMS (or similar application) can be built on an existing trusted system or can be built
from scratch. A third alternative is to build on a modified system, but, in terms of modeling, this is
similar to starting from scratch. Building on a trusted operating system offers a well-defined and
well-documented basis that provides process isolation and user authentication. Depending on the
DBMS design, the operating system may also provide mandatory or discretionary access control.
However, it may be necessary to hide or "turn off" some of the original security mechanisms, if
they are inconsistent with and impact the operation of the DBMS security policy. The DBMS may
also need to hide the original TCB/user interface, if the DBMS provides a significantly different
user interface than the underlying operating system. Starting from scratch can give a more unified
design, may simplify the security modeling effort, and is particularly appropriate for database
machines with simple, built-in operating systems.

The DBMS-supplied notion of security can be presented in an external-interface model that
gives the intended user-view of security. If the DBMS is built on a trusted operating system, the
external model can be followed with an internal requirements model that identifies the role of the
underlying operating system and its security policy model. The external-interface model itself will
be more relevant to users of the application if it emphasizes concepts associated with the DBMS.
Terminology found in the OS security model, by way of contrast, may be inappropriate either
because it refers to entities hidden by the application or is very abstract in order to apply to a wide
range of product applications.

The security models associated with a trusted DBMS and its underlying operating system
illustrate several issues that arise in other large applications. To ensure that the system design is
consistent with the security policy, it is necessary to provide evidence that the model (or models)
of underlying subsystems support the application policy. The decomposition of an overall security
policy model into submodels for subsystems is particularly advantageous when subsystems are
developed by different groups over a protracted period of time. Despite efforts to keep the system
model abstract, there may be inconsistencies between it and off-the-shelf subsystems used in the
development. As a result, the system security model will need to be maintained in an interactive
process as the design and product selection proceed.

The granularity of storage objects for database systems is typically finer han that of the
underlying operating system. The distinction between active and passive entities tends to be
somewhat blurred in the case of object-oriented systems. Existing examples suggest that the data
model has a strong influence on the mandatory security policy, as can be seen by comparing
relational [e.g., DENN88], functional [THUR89], entity/relationship [GAJN88], and object-
oriented security models [e.g., JAJO90; KEEF89b; LUNT90; LARR90]. These observations
suggest that the underlying OS MAC facility may well need to be supplanted. Database entities
may be viewed as multilevel structures that are divided up and stored in single-level files. In this
case, the DBMS might contain non-TCB subjects operating at all security levels used by the
DBMS. This approach is illustrated by the LOCK Data Views [HAIG90] and SeaViews
[LUNT88b] systems. Alternatively, the DBMS itself may take responsibility for the assignment of
security labels so that entities which are single-level objects to the operating system are multilevel
data structures to the DBMS, as in the ASD [GARV90] and MITRE prototypes [DAVI88]. In
either case, it is important to understand the relationship between database entities and the storage
objects of the underlying operating system, and it is appropriate to explain and perhaps formally
model the protocol for passing objects between the DBMS and the underlying operating system.

The named objects in a DBMS may differ not only from those of the underlying operating
system but also from the storage objects of the DBMS itself. User views are traditionally used as
a means of controlling access in relational database systems [KORT86, Ch.13.2]; and it is
reasonable to use views as named objects. In this case a user may have access to several
overlapping views so that the ability to access a given piece of information would depend on the
user's view of it. Consequently, the user's ability to access that information might not be
immediately apparent. In degenerate cases, two named objects might coincide or, similarly, a given
object might have several names.

4.3.2           INTEGRITY MECHANISMS AND POLICIES

Since pragmatic approaches to data integrity are, in varying degrees, part of most database
systems, [KORT86, Ch 13.3; FERN81, Ch.8] the modeling of DBMS integrity policies may lead
to a better formal understanding of integrity that can be applied much more generally. Issues of
interest include whether integrity constraints obscure other critical security concerns, whether
there are built-in mechanisms for monitoring or controlling the propagation of inconsistencies, and
whether users are able to learn which integrity checks succeed or fail for a given piece of
information.

Simple data integrity checks such as "A record field of type `month' should be an integer
between 1 and 12" guarantee structural integrity. If such checks are enforced on input, a
corresponding integrity failure is unlikely, and therefore significant. Labeling constraints of the
sort discussed in Section 4.3.5 can be modeled in much the same way as simple integrity checks;
they enforce a form of label integrity.

A more complicated check such as "A record field of type `department' should only be
populated by department names, of which there are currently 93," however, can be invalidated in
at least two ways, by adding a misspelled department and by altering the set of official department
names. In more complicated cases, it may not be possible to guarantee data integrity because
inconsistencies do not uniquely identify an erroneous entry, so that inconsistencies in the database
must be allowed for.

A model of a data integrity policy can show how information about integrity checks is
associated with data and how it propagates during data processing. For example, indication of
whether a given record satisfies a particular integrity check is a security attribute that might be
modeled as a category in an integrity label. The model could explain how this attribute is handled
during data processing.

Finally, some system integrity issues appear to be unique to secure database systems. The
scheduling of single-level updates associated with a series of multilevel updates has to be both
correct and efficient, and it must allow secure recovery after system failure. [KEEF90]

4.3.3           AGGREGATlON

The aggregation of less classified information to form a more highly classified body of
information is illustrated by the following example: locations of individuals are unclassified, but
total troop strength in a given area is classified, suggesting that a comprehensive database of troop
locations should itself be classified. In some cases, aggregation problems can be handled by
separating individual pieces of information into several or many categories. Once this is done,
aggregate levels are automatically represented by a set of categories and thus by a higher security
level. With this approach, no explicit modeling of aggregation control is needed.

An alternate approach is to place objects in "containers" that are classified at a higher level
and allow them to be read only by subjects that are at the higher level and by TCB subjects that are
authorized to extract individual objects as a form of controlled downgrading, as in the SMMS
model. [LANDS4] Methods for simulating containers in a relational database system have been
shown by Meadows. [MEAD90]

Yet another approach to controlling aggregation is to keep track of user queries and provide
context dependent classification of query results. In this case, the first N items from an aggregate
might be automatically downgraded, but the remaining items could only be obtained by users
acting at the level of the aggregate. [HAlG90; cf LUNT89] This approach can be formalized by
modeling the access histories of containers.

A rather general approach to aggregation is to introduce an aggregate level function, g, that
associates each set of data items with a security level. One expects that if H is a subset of K, then
g (H) ú g(K). Additionally, one expects that, for any A, B, and C, if g(A) ú g (B), then g(A ╗ C) ú
g(B ╗ C), since the information in A ╗  C is obtainable from information at levels dominated by
g(B ╗  C), since g(A) ú g(B)  ú g(B ╗ C) and  g(C) úg(B ╗ C). Meadows provides further
information regarding the construction of aggregate-level functions. [MEAD90b]

4.3.4           INFERENCE

In contrast to aggregation, inference allows new information to combine with preexisting
knowledge in a given environment, obtaining results that are not contained in the original
aggregate. The inference of restricted information from authorized queries and database
information is illustrated by the following examples:

1. The total.quantity of an item is classified, but its total cost and cost per item are not; two
unclassified queries suffice to retrieve a piece of classified information. [DENN82]

2. An uncleared user query of the form, "Give me the contents of all containers which contain
the fact that Flight 127 is carrying bombs to the front," might elicit a response of "access
denied" because one or more of the requested records is classified, thereby indicating that
Flight 127 is secretly carrying bombs to the front. [WISE90]

3. A customer's bank balance is restricted information. A user of a statistical database obtains
the total of all bank balances for bank customers in Smalltown, along with a corresponding
list of all customer names (of which there is only one). [KORT86]

The first inference example can be addressed by classifying one of the facts which led to the
inference, by enforcing separation of duty so that no uncleared user is authorized to get both facts,
or by maintaining user access histories and providing context dependent classification of query
results. [HAIG90] A possible objective for the modeling effort is to provide enough information
to see which options are available in the modeled system.

The second example looks superficially like a covert channel associated with error returns but
is potentially more dangerous because it is so easy to use. It can be avoided by redefining the query
semantics so that the phrase "all containers" implicitly refers to all containers at or below the user's
level. With this change, the appropriate query response is no such records exist." A strong model
and model interpretation would simply rule out this example.

The third example underscores the fact that inference problems also arise in commercial
systems. [cf KORT86, Ch. 13.5; FERN81, Ch. 13] The need to model inference control thus
extends beyond traditional multilevel security.

An extensive quantitative analysis of the inference problem may be found in Crytography and
Data Security [DENN82] In the LOCK Data Views effort, much of the inference problem is
delegated to a database system security officer who can specify content- and context- dependent
classification rules. The classification of a field in a record may depend on its value, on other fields
in the record it is displayed with, or on previous queries made by the user. [HAIG90] An interesting
consequence of inference problems is that there may be no lowest, safe security level for a given
piece of information: an otherwise unclassified piece of information might allow users at level {a}
to infer information at level {a, b} and users at level {p} to infer information at level {p, q}. Thus,
{a, b}, {p, q}, and {a, b, p, q} would be admissible levels for this piece of information, whereas
{a} and {p} would not. These and other considerations have led to an explicit axiomatization of
inference in terms of an "infer" function [HAIG90] that could be included in a model's definition
of security in order to explicitly capture an inference control policy.

Inference constraints may be enforced during database design, updating, and query
processing. In principle, enforcement need only occur during query processing, but this approach
may also be the least efficient. [cf KEEF89, HINK88, ROWE89, FORD90] Thus, locus of policy
enforcement may be a significant factor in modeling inference control.

4.3.5           PARTIALLY AUTOMATED LABELING

As discussed in Section 3.2, the usual labeling paradigm is that users know the sensitivities of
their inputs and supply these sensitivities when the information is being entered. In the case of
databases, however, information about object classifications may not be available when a database
schema is defined. Including this classification information as part of the schema definition not
only provides a uniform approach to classification, but also improves labeling accuracy by
allowing specially authorized users to assign or constrain object sensitivity levels.

In actuality, information rather than data or containers for data is classified. Consequently,
a rule for classifying storage objects in a database may depend not only on the particular object but
also on the data it contains and on how the data is used, as is illustrated by the following examples:

1. A ship's location might be unclassified when it was in a U.S. port but classified if it were
fifty miles from Tripoli. [GRAU90]

2. Names of manufacturing companies are unclassified but most information about a
particular spy plane, including its manufacturer, is classified.

In the first example, classification is content-dependent. As noted by Graubart, content-
dependent classification is antithetical to the tranquility principle. [GRAU90] In the second
example, whether or not the name of the plane's manufacturer is classified depends on its use in a
given context. Lunt has argued that context-sensitive classifications are easily confused with
inference problems [LUNT89] and that object-oriented systems are especially appropriate for
handling context dependencies. [LUNT90]

Lack of tranquility exposes what has been called the "multi party update conflict" problem.
Two users working at different classifications attempt to store conflicting reports, leading to
questions about which report and which classification is correct, and whether or not the mistaken
user should be so informed. One possibility is that the less -classified user has been given a "cover
story" to hide publicly observable aspects of a classified situation. In this case, a possible approach
to the update conflict is to store both the cover story and the truth in the same spot in the database
with the two entries being distinguished only by their classification. This approach, known as
"polyinstantiation," is not the only approach to multiparty update conflicts [KEEF90b], and it may
well be inappropriate in situations where cover stories are not involved. [SMIT90]

A fairly wide variety of content-and/or context-dependent classification mechanisms have
been investigated, but a general approach to the problem has yet to be developed. [SMIT88] As
indicated in Section 3.2.1, traditional approaches to MAC modeling begin with the assumption that
data sensitivity is user-supplied. As a result, some aspects may need to be rethought when
modeling policies for automated data labeling.

4.4     SYSTEMS WITH EXTENDED SUPPORT FOR LABEL ACCURACY

As mentioned in Section 3.1.5, security labels are used both to control information flow and
to provide accurate security markings for data. These goals may well conflict, and this conflict may
be theoretically unavoidable in some situations. [JONE75; DENN82, º 5.2.1] Policies and coping
strategies for maintaining label accuracy have turned up in security models, whereas label accuracy
itself has not yet been explicitly modeled.

The following paragraphs discuss pragmatic factors that inhibit accuracy, some coping
strategies suggested by these factors, and the realization of these strategies in Compartmented
Mode Workstations (CMWs). As a final example, the "Chinese wall" policy for stock market
analysts is introduced and shown to be a variant of the workstation security policy.

4.4.1           FACTORS INHIBITING ACCURACY IN LABELING

In practice, several factors may interfere with accuracy in labeling, including the need to
maintain labels across different working environments, the use of complex labeling schemes, and
data fusion.

In many applications, the collection, processing, and distribution of information all take place
in different environments. It can happen that labels are assigned during collection and used to
control the ultimate distribution, but the majority of processing takes place in a system-high
environment where automated access control is not supported or in a multilevel environment where
not all security markings are used to control access by users. In these cases, it is necessary to
accurately maintain security markings even when they are not needed to enforce access control in
the information processing environment.

Some applications may involve the use of literally thousands of categories. Moreover, the
required security markings may include not only a DoD classification but also code words,
handling caveats, and/or release markings as well. Such markings usually form a lattice, but the
rules for combining markings may be complex. [WOOD87]

The fusion of data at several security levels requires a convention for labeling fused data.
Even if aggregation and inference issues are ignored, it is still necessary to find a level which
dominates that of every data source. If the level of fused data can be predicted in advance, then
aggregate data can be maintained in multilevel data structures, as in the SMMS. [LAND84]
However, finding an appropriate aggregate level may be hindered by difficulty in combining labels
or by the fact that the sources cannot be predicted in advance, either because they arrive in real time
or because the avoidance of inappropriate sources is part of the fusion task itself.

4.4.2           FLOATING SENSITIVITY LABELS

If the security levels form a semi-lattice, then some form of "high water mark" policy may be
used to maintain an upper bound on fused data. In such a policy, the TCB maintains the least upper
bound of all data levels associated with an entity in a "floating" label. While the TCSEC
requirements do not rule out floating labels, the MAC requirements do imply that the level of a
subject must never float above the clearance of its user. At B2 and above, the covert channel
requirements suggest that label data should not be exploitable as a covert channel. Unfortunately,
most high watermark policies not only allow, but actually force, the existence of covert channels
based on the use of floating labels. [DENN82, Ch.5.3]

The level of a process can be kept from floating too high by going to a dual-label mechanism
in which a fixed label contains the subject's maximum security level, which dominates that of the
floating label. In this situation, the fixed label is used only for access control and need only include
those security attributes that are actually used for access control. Factors which favor acceptability
of a covert channel include low bandwidth and the absence of application software that can exploit
covert channels.

4.4.3           COMPARTMENTED MODE WORKSTATlONS (CMW)

The CMW design described by Woodward [WOOD87] is for a dual label system. The fixed
label contains a "sensitivity" level and is the only label used for access control. The floating label
contains an "information" level that consists of a second sensitivity level and additional security
markings. The sensitivity levels are used to control nondisclosure and consist of DoD clearance
and authorization components. The security markings form a lattice; hence, so do the information
levels. The intended use of the two labels is that the fixed label enforces an upper bound on the
sensitivity of information held by an entity, while the information level describes the current
security level of information held. The covert channel problem resulting from the use of floating
labels can lead to erroneous information labels but cannot be used to violate the access control
policy enforced by the fixed labels.

A CMW security model has been developed by Bodeau and Millen. [MILL90] Its MAC
policy may briefly be summarized as follows: when a user creates a new file or process, it receives
a sensitivity level equal to that of the user's login level. By way of contrast, the information label
for a newly created file contains the lowest possible information level. In particular, its sensitivity
component is "unclassified." A sensitivity label never changes throughout the life of an entity, but
the information label is allowed to float upwards in such a way that its sensitivity component is
always dominated by the level of that entity's sensitivity label.

The rules of operation are such that, whenever an operation is invoked that involves an
(intended) information flow from an entity E1 to an entity E2, the sensitivity labels are checked to
be sure that the sensitivity of E2 dominates that of E1. The maximum of the information levels for
E1 and E2 is then computed, and it becomes the new value of the information label for E2. In
addition to the properties just described, the CMW model also discusses privilege and DAC. The
model does not address direct process-to-process communication, but does treat pipes as objects.
As is appropriate when modeling a policy that is subject to misuse, the existence of a channel based
on information labels is derivable from the model itself.

4.4.4           THE CHINESE WALL SECURITY POLICY

As presented by Brewer and Nash, the "Chinese Wall" Policy is a mandatory access control
policy for stock market analysts. [BREW89] This organizational policy is legally binding in the
United Kingdom stock exchange. According to the policy, a market analyst may do business with
any company. However, every time the analyst receives sensitive "inside" information from a new
company, the policy prevents him from doing business with any other company in the same
industry because that would involve him in a conflict of interest. In other words, collaboration with
one company places a "Chinese wall" between him and all other companies in the same industry.
Same ness of industry might be judged according to business sector headings found in listings of
the stock exchange, for example. Notice that this policy does not, by itself, prohibit conspiracies.
For example, one market analyst can give information about a company in industry I to a company
in industry J. Subsequently, another analyst can transfer this information from the company in
industry J back to some other company in industry I. Analogous conspiracies between colliding
processes, however, are explicitly ruled out in the corresponding system security policy and its
model. [cf BREW89, Axiom 6]

Brewer and Nash argue that this policy cannot be modeled using models in the style of Bell
and La Padula that obey tranquility. However, this policy can easily be modeled as a variant of the
CMW model. Information and sensitivity levels are both sets of categories, where each category
represents a different company. There is an accreditation range for information levels. A level
belongs to the accreditation range if and only if it does not contain two or more companies from
the same industry. The model must be extended slightly by adding information labels for users.
Every user and every controlled entity has a system-high sensitivity level, reflecting the fact that
an analyst may work with any particular company. Every user starts out with an empty (system-
low) information label. Each time an analyst attempts to read information in a new category (i.e.,
company), his information level floats up, unless it goes outside of the accreditation range, in
which case his attempt is rejected. Notice that, as in the CMW example, there are thousands of
categories. While the rule for combining categories is straightforward, the accreditation range is
not. Additional thoughts on the use of accreditation ranges in controlling aggregation may be found
in the paper "Extending the Brewer-Nash Model to a Multilevel Context." [MEAD90b]

5.      MEETING THE REQUIREMENTS

At evaluation class C1 and above, a "description of the manufacturer's philosophy of
protection and an explanation of how this philosophy is translated into the TCB" is required. For
evaluation classes B1 and above, this requirement is supplanted with explicit security modeling
requirements. This section contains a listing of these requirements, followed by a discussion of
how a security policy model can meet these requirements, as well as satisfy related assurance and
architectural requirements.

5.1     STATED REQUIREMENTS ON THE SECURITY MODEL

Requirements that are new at a given evaluation level are presented in bold face.

5.1.1 B1 REQUIREMENTS

1. An informal or formal model of the security policy supported by the TCB shall be
maintained over the life cycle of the ADP system. An informal or formal description
of the security policy model enforced by the TCB shall be available.

2. [The security model] shall be demonstrated to be consistent with its axioms.

3. An explanation [shall be] provided to show that it is sufficient to enforce the security
policy.

4. The specific TCB protection mechanisms shall be identified and an explanation given
to show that they satisfy the model.

5.1.2 B2 REQUIREMENTS

1. A  formal model of the security policy supported by the TCB shall be maintained over the
life cycle of the ADP system. A formal description of the security policy model enforced
by the TCB shall be available.

2. [The model] shall be proven consistent with its axioms.

3. [The model shall be] proven sufficient to enforce the security policy.

4. The specific TCB protection mechanisms shall be identified and an explanation given to
show that they satisfy the model.

5.1.3 B3 REQUIREMENTS

1. A  formal model of the security policy supported by the TCB shall be maintained over the
life cycle of the ADP system. A formal description of the security policy model enforced
by the TCB shall be available.

2. [The model] shall be proven consistent with its axioms.

3. [The model shall be] proven sufficient to enforce the security policy.

4. The specific TCB protection mechanisms shall be identified and an explanation given to
show that they satisfy the model.

5. A convincing argument shall be given that the DTLS is consistent with the model.

5.1.4 A1 REQUIREMENTS

1. A formal model of the security policy supported by the TCB shall be maintained over the
life cycle of the ADP system. A formal description of the security policy model enforced
by the TCB shall be available.

2. [The model] shall be proven consistent with its axioms.

3. [The model shall be] proven sufficient to enforce the security policy.

4. The specific TCB protection mechanisms shall be identified and an explanation given to
show that they satisfy the model.

5. A convincing argument shall be given that the DTLS is consistent with the model.

6. A combination of formal and informal techniques shall be used to show that the
FTLS is consistent with the model. This verification evidence shall be consistent with
that provided within the state-of-the-art of the particular National Computer
Security Center endorsed formal specification and verification system used.

7. During the entire life cycle, i.e., during the design, development and maintenance of the
TCB, a configuration management system shall be in place... that maintains control of
changes to the formal model....

5.2     DISCUSSION OF THE B1 REQUIREMENTS

1. Provided Model of the Security Policy Enforced by the TCB. The decision of whether to
use a formal or informal model is a matter of choice, as can be seen from the TCSEC Glossary. An
informal security policy model should be presented with enough precision that a formal
mathematical model could be constructed if needed. A formal security model must conform to
accepted standards of mathematical rigor.

The security model must present a definition of security and an enforcement policy (i.e., rules
of operation) for controlled system entities. The controlled entities may be correctly interpreted as
storage objects, devices, processes, and other controlled system resources. The model must present
the system's notion of access and explain how access checks (or constraints) succeed in preventing
unauthorized access. Thus, the model must consist of more than just a high-level description of
security requirements.

Since formal proof is not required, a firm distinction between abstract security requirements
and concrete enforcement mechanisms may be unnecessary. It is absent from the [NCSC88]
definition of a security policy model. A distinction between the reference monitor and other
portions of the TCB is also not required, but the role of TCB subjects in the enforcement of the
security policy must be taken into account. Moreover, in the case of a retrofitted system that did
not originally support trusted user roles, the interface between TCB subjects and the basic system
kernel may be too complicated for TCB-subject modeling to be valuable.

2. Demonstration of Internal Consistency. The second modeling requirement covers two
related concerns. The model must not contain inconsistencies. In other words, it must not
contradict itself. Moreover, a formal or informal demonstration must show that the modeled
enforcement policy is sufficient to achieve any modeled security requirements ("axioms"). In other
words, the rules of operation must imply the definition of security.

3. Explanation of Sufficiency to Enforce the Security Policy. An explanation must be
provided to show that the model's description of the security policy is adequate-that the model
leads to a system whose TCB enforces the advertised system security policy. For the model to be
sufficient or adequate in this sense, it must include key ideas used in the design of the security
enforcement mechanisms. The resulting rules of operation should provide a basis for
understanding how the system's main security enforcement mechanisms enforce the security
policy. In the case of networks and other complex systems, models of key subsystems are needed
in addition to a model of the entire system. In particular, "the overall network policy must be
decomposed into policy elements that are allocated to appropriate components and used as the
basis for the security policy model for those components". [NCSC87, Sec. 3.1.3.2.2]

The system security policy itself must include the minimum requirements of Section 3.1.1 of
the TCSEC, and the model should tailor Section 3.1.1 to the particular system at hand. The access
control requirements in Section 3.1.1 must always be modeled, whereas the need to model the
labeling and object reuse requirements will vary from one system to the next. [cf NCSC87, Sec.
3.1.4.4]

The security policy model must include a description of DAC. An explanation of how DAC
interacts with MAC is encouraged but not explicitly required. In particular, there is no a priori
requirement for DAC objects to coincide with MAC objects.

The security policy model must include a description of MAC. The decomposition of
sensitivity levels into clearance, nondisclosure category, or other components need not be
explicitly included unless such decomposition is essential to an understanding of the model.
Flagrant examples of illegal information channels may be regarded as MAC policy violations, even
though a covert channel analysis is not required. More specifically, a documented (or trivially
inferred) use of a system function must not result in an illegal transfer of information.

Modeling of the following security policy requirements may be useful, but is traditionally not
required:

╖ A process acting on behalf of a user must have a label that is dominated by the clearance
and authorization of that user.

╖ Information flowing across a device must have an implicitly or explicitly associated
sensitivity level, according to whether the device is classified as single-level or multi level.

╖ Object reuse and process initialization do not happen in such a way as to allow unauthorized
disclosure.

╖ In a network, the overall network security policy is enforced by the NTCB.

╖ Additional vendor supplied security requirements, whether derived from governing
regulations or customer needs, may be enforced by the TCB.

4. TCB Protection Mechanisms and Correspondence to Model. The TCSEC modeling
requirements must be met in away that is consistent with the needs of the system development
process and with actual TCB protection mechanisms. The vendor must supply a model
interpretation showing how the rules of operation in the model relate to the actions of the TCB. The
model interpretation must address the reference monitor interface and show how subject
instructions are accounted for.

The following aspects of a system do not have to be modeled although their representation in
the model may be useful:

╖ Error diagnostics,

╖ The contents of storage objects and other controlled entities,

╖ The scheduling of subjects and their synchronization with external inputs,

╖ Internal TCB structure,

╖ Portions of the TCB that do not support user-requested computation (e.g., security-
administrator functions), and

╖ An individual network component that has a very simple (or nonexistent) reference monitor
and contains no subjects which act on behalf of users.

5.3 DISCUSSION OF THE B2 REQUIREMENTS

1. Provided Model of the Security Policy Enforced by the TCB. The model must be written
in a formal mathematical notation; either mathematical English or a well-defined formal
specification language is acceptable. From the TCSEC Glossary definition of a formal security
policy model, it is clear that the model must describe both what security is and how it is enforced.
As discussed in Section 3.2, the definition of security can be formalized using either external-
interface requirements or internal requirements on controlled entities. The explanation of how
security is enforced typically takes the form of rules of operation.

2. Demonstration of Internal Consistency. A mathematical proof must be given showing
that the rules of operation ensure satisfaction of the modeled security requirements.

3. Explanation of Sufficiency to Enforce the Security Policy. The proof referred to in this
requirement is an informal, rather than a mathematical proof, because sufficiency depends on the
typically informal security policy. The intent of this documentation requirement is that stronger
evidence of sufficiency be provided at B2 than at B1. If the system has a novel approach to
mandatory or discretionary access control, it may be necessary to model both what the system does
and what Section 3.2.1 of the TCSEC requires and then to prove that the system does what is
required. Stronger evidence of sufficiency is possible, if the model is compatible with other B2
security, accountability, and assurance requirements. For this reason, explicit inclusion of the
following requirements may be useful:

╖ The TCB shall support the assignment of minimum and maximum security levels to all
attached physical devices. These security levels shall be used by the TCB to enforce
constraints imposed by the physical environments in which the devices are located.

╖ [The TCB design shall] separate those elements [of the TCB] that are protection-critical
from those that are not.

╖ The TCB modules shall be designed such that the principle of least privilege is enforced.

╖ The user interface to the TCB shall be completely defined and all elements of the TCB
identified.

╖ The TCB shall support separate operator and administrator functions.

╖ The system developer shall conduct a thorough search for covert storage channels.

╖ The TCB shall support a trusted communication path between itself and the user for initial
login and authentication. Communications via this path shall be initiated exclusively by the
user.

If separate security models are given for the security kernel and other components of the TCB,
then the pieces must fit together correctly, and arguments should be given to the effect that overall
system security is achieved and necessary checks are not inadvertently omitted. As indicated in
Section 3.2, the notion of which storage channels are covert is partially reflected in the security
model, and the inclusion of information flow requirements in the model's definition of security can
reduce the effort needed to perform covert channel analysis.

4. TCB Protection Mechanisms and Correspondence to Model. The B2 requirements for
validation of TCB protection mechanisms are basically the same as those discussed at the end of
Section 5.2, except for an implicit completeness requirement. This requirement is that "the TCB
shall enforce a mandatory access control policy over all resources (i.e., subjects, objects, and   I/O
devices) that are directly or indirectly accessible by subjects external to the TCB." This
requirement should be reflected in the model interpretation, and its satisfaction may require
extensions to the model in order to account for all accessible resources in the implementation. The
model interpretation itself may be accomplished in two steps, by first mapping the model to the
DTLS and then the DTLS to the TCB implementation.

5.4 Discussion of the B3 Requirements

The B3 requirements include the B2 requirements as well as several new requirements.

1, 2. Provided Model and Internal Consistency. The basic structure of the security model is
the same at B3 as at B2.

3. Explanation of Sufficiency to Enforce the Security Policy. There is one additional
security policy requirement at B3, namely that access controls "shall be capable of specifying, for
each named object, a list of named individuals and a list of groups of named individuals with their
respective modes of access to that object. For each named object, it shall be possible to specify a
list of named individuals and a List of groups of named individuals for which no access to the
object is to be given."

In addition, the model may usefully reflect, and must be consistent with, the following B3
architectural assurance requirement: "The TCB shall be designed and structured to use a complete,
conceptually simple protection mechanism with precisely defined semantics. This mechanism
shall play a central role in enforcing the internal structuring of the TCB and the system. Significant
system engineering shall be directed toward minimizing the complexity of the TCB and excluding
from the TCB modules that are not protection-critical."

4, 5. TCB Protection Mechanisms and DTLS Correspondence to Model. The validation of
TCB protection mechanisms is normally performed by showing that the model is an abstraction of
the DTLS and then mapping the DTLS to the actual TCB implementation. (See Section 2.3.6.)

5.5 DISCUSSION OF THE A1 REQUIREMENTS

1-5. Previously Introduced Requirements. The first five requirements are the same at A1 as
at B3 (except for the fact that the FTLS rather than the DTLS is mapped to the implementation).
The requirements for a formal covert channel analysis invite, but do not require, the use of
information flow or similar external-interface models.

6. FTLS Correspondence to Model. The main new requirement at this level is that a "formal"
mathematical proof be given showing that the security model is an abstraction of the FTLS. This
proof may be carried out either by hand or with the use of an endorsed formal verification system.
In either case, the vendor is responsible for providing an understandable, logically correct
justification of correspondence. This justification normally begins with the formulation of a
rigorous conjecture to the effect that the model is an abstraction of the FTLS. If part or all of the
proof is presented to a verification system, then the vendor must demonstrate that the presented
portion is correctly codified in the formal language of the verification system. This demonstration
may require informal argument and an appeal to the semantics of the system's formal specification
language.

There are two common ways of showing that the FTLS is consistent with the model:
comparing the FTLS with the rules of operation and comparing the FTLS directly with the model's
definition of security. In the latter case, the rules of operation and model interpretation are similar
in purpose to the FTLS and its implementation correspondence, respectively. This parallelism of
requirements does not necessarily imply duplication of effort, however, because there are no
explicit requirements which force a distinction between the FTLS and the rules of operation.

7. Configuration Management for the Model. The requirement that the model and related
documentation be maintained under configuration management is not a modeling requirement,
perse. However, because of this requirement, extra care is needed to ensure that the model is given
in a form that contributes to its maintainability.

APPENDIX A. SECURITY LEVELS AND PARTIALLY ORDERED SETS

Policies that control information flow often rely on partially ordered sets of security attributes.
This appendix introduces partial orderings as they relate to information flow and presents basic
facts about partial orderings that are relevant to the modeling and implementation of label-based
security policies. Many of these facts can also be found in the undergraduate text [ABBO69].

Most processes satisfy two basic information flow properties:

reflexivity: A process can access any information it possesses. That is, information can
always flow from a process to itself.

transitivity: If information can flow from process P1 to process P2 and can flow from P2 to
P3, then information can flow from P1 to P3.1

These two properties of information flow determine a preordering relation between processes. If
processes are labeled in such a way as to make economical use of label values, then the following
property may also hold as well:

 antisymmetry: If information can flow from a process with label L1 to a process with label
L2, and conversely, then L1 = L2.

Thus, the intended relationship between labels and information flow leads to consideration of a
reflexive, transitive, antisymmetric relation on label values, that is, a partial ordering. The intended
relationship between information flow and this partial ordering can be expressed by saying that L1
ú L2, if information is allowed to flow from controlled processes with label L1 to controlled
processes with label L2. This relation is traditionally referred to as dominance: L1, is dominated by
L2 if and only if L1, ú L2.

The following sections introduce basic terminology and show how partial orderings can be
constructed and manipulated through the use of embeddings, Cartesian products, and dual
orderings.

A.l TERMINOLOGY

Formally, ú is taken to be a partial ordering on a set  whose elements are referred to as
levels. A pair of the form┼ (, ú) is a partially ordered set. For any levels L1 and L2, L1