Policy Framework Working Group B. Moore
INTERNET-DRAFT E. Ellesson
Category: Standards Track IBM
J. Strassner
A. Westerinen
Cisco Systems
July, 2000
Policy Core Information Model -- Version 1 Specification
<draft-ietf-policy-core-info-model-07.txt>
Friday, July 14, 2000, 1:33 PM
Status of this Memo
This document is an Internet-Draft and is in full conformance with all
provisions of Section 10 of RFC2026.
Internet-Drafts are working documents of the Internet Engineering Task
Force (IETF), its areas, and its working groups. Note that other
groups may also distribute working documents as Internet-Drafts.
Internet-Drafts are draft documents valid for a maximum of six months
and may be updated, replaced, or obsoleted by other documents at any
time. It is inappropriate to use Internet-Drafts as reference
material or to cite them other than as "work in progress."
The list of current Internet-Drafts can be accessed at
http://www.ietf.org/ietf/1id-abstracts.txt
The list of Internet-Draft Shadow Directories can be accessed at
http://www.ietf.org/shadow.html
Copyright Notice
Copyright (C) The Internet Society (2000). All Rights Reserved.
Abstract
This document presents the object-oriented information model for
representing policy information developed jointly in the IETF Policy
Framework WG and as extensions to the Common Information Model (CIM)
activity in the Distributed Management Task Force (DMTF). This model
defines two hierarchies of object classes: structural classes
representing policy information and control of policies, and
association classes that indicate how instances of the structural
classes are related to each other. Subsequent documents will define
mappings of this information model to various concrete
implementations, for example, to a directory that uses LDAPv3 as its
access protocol. The components of the CIM schema are available via
the following URL: http://www.dmtf.org/spec/cims.html [1].
Moore, et al. Expires: Jul 2000 + 6 months [Page 1]
Internet Draft Policy Core Information Model July 2000
Table of Contents
1. Introduction......................................................4
2. Modeling Policies.................................................5
2.1. Policy Scope.................................................7
2.2. Declarative versus Procedural Model..........................8
3. Overview of the Policy Core Information Model.....................9
4. Inheritance Hierarchies for the Policy Core Information Model....12
5. Details of the Model.............................................13
5.1. Reusable versus Rule-Specific Conditions and Actions........13
5.2. Roles.......................................................15
5.2.1. Roles and Role Combinations...............................15
5.2.2. The PolicyRoles Property..................................18
5.3. Local Time and UTC Time in PolicyTimePeriodConditions.......18
5.4. CIM Data Types..............................................20
6. Class Definitions................................................20
6.1. The Abstract Class "Policy".................................21
6.1.1. The Property "CommonName (CN)"............................22
6.1.2. The Multi-valued Property "PolicyKeywords"................22
6.1.3. The Property "Caption" (Inherited from ManagedElement)....23
6.1.4. The Property "Description" (Inherited from ManagedElement)23
6.2. The Class "PolicyGroup".....................................23
6.3. The Class "PolicyRule"......................................24
6.3.1. The Property "Enabled"....................................26
6.3.2. The Property "ConditionListType"..........................26
6.3.3. The Property "RuleUsage"..................................27
6.3.4. The Property "Priority"...................................27
6.3.5. The Property "Mandatory"..................................27
6.3.6. The Property "SequencedActions"...........................28
6.3.7. The Multi-valued Property "PolicyRoles"...................28
6.4. The Class "PolicyCondition".................................29
6.5. The Class "PolicyTimePeriodCondition".......................31
6.5.1. The Property "TimePeriod".................................32
6.5.2. The Property "MonthOfYearMask"............................33
6.5.3. The Property "DayOfMonthMask".............................34
6.5.4. The Property "DayOfWeekMask"..............................35
6.5.5. The Property "TimeOfDayMask"..............................35
6.5.6. The Property "LocalOrUtcTime".............................36
6.6. The Class "VendorPolicyCondition"...........................36
6.6.1. The Multi-valued Property "Constraint"....................37
6.6.2. The Property "ConstraintEncoding".........................38
6.7. The Class "PolicyAction"....................................38
6.8. The Class "VendorPolicyAction"..............................39
6.8.1. The Multi-valued Property "ActionData"....................39
6.8.2. The Property "ActionEncoding".............................40
6.9. The Class "PolicyRepository"................................40
7. Association and Aggregation Definitions..........................41
7.1. Associations................................................41
7.2. Aggregations................................................41
7.3. The Abstract Aggregation "PolicyComponent...................41
7.4. The Aggregation "PolicyGroupInPolicyGroup"..................41
7.4.1. The Reference "GroupComponent"............................42
7.4.2. The Reference "PartComponent".............................42
Moore, et al. Expires: Jul 2000 + 6 months [Page 2]
Internet Draft Policy Core Information Model July 2000
7.5. The Aggregation "PolicyRuleInPolicyGroup"...................42
7.5.1. The Reference "GroupComponent"............................43
7.5.2. The Reference "PartComponent".............................43
7.6. The Aggregation "PolicyConditionInPolicyRule"...............43
7.6.1. The Reference "GroupComponent"............................44
7.6.2. The Reference "PartComponent".............................44
7.6.3. The Property "GroupNumber"................................44
7.6.4. The Property "ConditionNegated"...........................44
7.7. The Aggregation "PolicyRuleValidityPeriod"..................45
7.7.1. The Reference "GroupComponent"............................45
7.7.2. The Reference "PartComponent".............................45
7.8. The Aggregation "PolicyActionInPolicyRule"..................46
7.8.1. The Reference "GroupComponent"............................46
7.8.2. The Reference "PartComponent".............................46
7.8.3. The Property "ActionOrder"................................47
7.9. The Abstract Association "PolicyInSystem"...................48
7.10. The Weak Association "PolicyGroupInSystem".................48
7.10.1. The Reference "Antecedent"...............................48
7.10.2. The Reference "Dependent"................................49
7.11. The Weak Association "PolicyRuleInSystem"..................49
7.11.1. The Reference "Antecedent"...............................49
7.11.2. The Reference "Dependent"................................49
7.12. The Association "PolicyConditionInPolicyRepository"........49
7.12.1. The Reference "Antecedent"...............................50
7.12.2. The Reference "Dependent"................................50
7.13. The Association "PolicyActionInPolicyRepository"...........50
7.13.1. The Reference "Antecedent"...............................50
7.13.2. The Reference "Dependent"................................51
7.14. The Aggregation "PolicyRepositoryInPolicyRepository".......51
7.14.1. The Reference "GroupComponent"...........................51
7.14.2. The Reference "PartComponent"............................51
8. Intellectual Property............................................52
9. Acknowledgements.................................................52
10. Security Considerations.........................................52
11. References......................................................54
12. Authors' Addresses..............................................55
13. Full Copyright Statement........................................56
14. Appendix A - Class Identification in a Native CIM Implementation57
14.1. Naming Instances of PolicyGroup and PolicyRule.............57
14.1.1. PolicyGroup's CIM Keys...................................57
14.1.2. PolicyRule's CIM Keys....................................58
14.2. Naming Instances of PolicyCondition and Its Subclasses.....59
14.2.1. PolicyCondition's CIM Keys...............................61
14.3. Naming Instances of PolicyAction and Its Subclasses........62
14.4. Naming Instances of PolicyRepository.......................64
14.5. Role of the CreationClassName Property in Naming...........64
14.6. Object References..........................................65
Moore, et al. Expires: Jul 2000 + 6 months [Page 3]
Internet Draft Policy Core Information Model July 2000
1. Introduction
This document presents the object-oriented information model for
representing policy information currently under joint development in
the IETF Policy Framework WG and as extensions to the Common
Information Model (CIM) activity in the Distributed Management Task
Force (DMTF). This model defines two hierarchies of object classes:
structural classes representing policy information and control of
policies, and association classes that indicate how instances of the
structural classes are related to each other. Subsequent documents
will define mappings of this information model to various concrete
implementations, for example, to a directory that uses LDAPv3 as its
access protocol.
The policy classes and associations defined in this model are
sufficiently generic to allow them to represent policies related to
anything. However, it is expected that their initial application in
the IETF will be for representing policies related to QoS (DiffServ
and IntServ) and to IPSec. Policy models for application-specific
areas such as these may extend the Core Model in several ways. The
preferred way is to use the PolicyGroup, PolicyRule, and
PolicyTimePeriodCondition classes directly, as a foundation for
representing and communicating policy information. Then, specific
subclasses derived from PolicyCondition and PolicyAction can capture
application-specific definitions of conditions and actions of
policies.
Two subclasses, VendorPolicyCondition and VendorPolicyAction, are also
included in this document, to provide a standard extension mechanism
for vendor-specific extensions to the Policy Core Information Model.
This document fits into the overall framework for representing,
deploying, and managing policies being developed by the Policy
Framework Working Group. It traces its origins to work that was
originally done for the Directory-enabled Networks (DEN)
specification, reference [5]. Work on the DEN specification by the
DEN Ad-Hoc Working Group itself has been completed. Further work to
standardize the models contained in it will be the responsibility of
selected working groups of the CIM effort in the Distributed
Management Task Force (DMTF). DMTF standardization of the core policy
model is the responsibility of the SLA Policy working group in the
DMTF.
This document is organized in the following manner:
o Section 2 provides a general overview of policies and how they are
modeled.
o Section 3 presents a high-level overview of the classes and
associations comprising the Policy Core Information Model.
o The remainder of the document presents the detailed specifications
for each of the classes and associations.
Moore, et al. Expires: Jul 2000 + 6 months [Page 4]
Internet Draft Policy Core Information Model July 2000
o Appendix A overviews naming for native CIM implementations. Other
mappings, such as LDAPv3, will have their own naming mechanisms.
The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT",
"SHOULD", "SHOULD NOT", "RECOMMENDED", "MAY", and "OPTIONAL" in this
document are to be interpreted as described in RFC 2119, reference
[3].
2. Modeling Policies
The classes comprising the Policy Core Information Model are intended
to serve as an extensible class hierarchy (through specialization) for
defining policy objects that enable application developers, network
administrators, and policy administrators to represent policies of
different types.
One way to think of a policy-controlled network is to first model the
network as a state machine and then use policy to control which state
a policy-controlled device should be in or is allowed to be in at any
given time. Given this approach, policy is applied using a set of
policy rules. Each policy rule consists of a set of conditions and a
set of actions. Policy rules may be aggregated into policy groups.
These groups may be nested, to represent a hierarchy of policies.
The set of conditions associated with a policy rule specifies when the
policy rule is applicable. The set of conditions can be expressed as
either an ORed set of ANDed sets of condition statements or an ANDed
set of ORed sets of statements. Individual condition statements can
also be negated. These combinations are termed, respectively,
Disjunctive Normal Form (DNF) and Conjunctive Normal Form (CNF) for
the conditions.
If the set of conditions associated with a policy rule evaluates to
TRUE, then a set of actions that either maintain the current state of
the object or transition the object to a new state may be executed.
For the set of actions associated with a policy rule, it is possible
to specify an order of execution, as well as an indication of whether
the order is required or merely recommended. It is also possible to
indicate that the order in which the actions are executed does not
matter.
Policy rules themselves can be prioritized. One common reason for
doing this is to express an overall policy that has a general case
with a few specific exceptions.
For example, a general QoS policy rule might specify that traffic
originating from members of the engineering group is to get Bronze
Service. A second policy rule might express an exception: traffic
originating from John, a specific member of the engineering group, is
to get Gold Service. Since traffic originating from John satisfies
the conditions of both policy rules, and since the actions associated
with the two rules are incompatible, a priority needs to be
Moore, et al. Expires: Jul 2000 + 6 months [Page 5]
Internet Draft Policy Core Information Model July 2000
established. By giving the second rule (the exception) a higher
priority than the first rule (the general case), a policy
administrator can get the desired effect: traffic originating from
John gets Gold Service, and traffic originating from all the other
members of the engineering group gets Bronze Service.
Policies can either be used in a stand-alone fashion or aggregated
into policy groups to perform more elaborate functions. Stand-alone
policies are called policy rules. Policy groups are aggregations of
policy rules, or aggregations of policy groups, but not both. Policy
groups can model intricate interactions between objects that have
complex interdependencies. Examples of this include a sophisticated
user logon policy that sets up application access, security, and
reconfigures network connections based on a combination of user
identity, network location, logon method and time of day. A policy
group represents a unit of reusability and manageability in that its
management is handled by an identifiable group of administrators and
its policy rules would be consistently applied
Stand-alone policies are those that can be expressed in a simple
statement. They can be represented effectively in schemata or MIBs.
Examples of this are VLAN assignments, simple YES/NO QoS requests, and
IP address allocations. A specific design goal of this model is to
support both stand-alone and aggregated policies.
Policy groups and rules can be classified by their purpose and intent.
This classification is useful in querying or grouping policy rules.
It indicates whether the policy is used to motivate when or how an
action occurs, or to characterize services (that can then be used, for
example, to bind clients to network services). Describing each of
these concepts in more detail,
o Motivational Policies are solely targeted at whether or how a
policy's goal is accomplished. Configuration and Usage Policies
are specific kinds of Motivational Policies. Another example is
the scheduling of file backup based on disk write activity from 8am
to 3pm, M-F.
o Configuration Policies define the default (or generic) setup of a
managed entity (for example, a network service). Examples of
Configuration Policies are the setup of a network forwarding
service or a network-hosted print queue.
o Installation Policies define what can and cannot be put on a system
or component, as well as the configuration of the mechanisms that
perform the install. Installation policies typically represent
specific administrative permissions, and can also represent
dependencies between different components (e.g., to complete the
installation of component A, components B and C must be previously
successfully installed or uninstalled).
Moore, et al. Expires: Jul 2000 + 6 months [Page 6]
Internet Draft Policy Core Information Model July 2000
o Error and Event Policies. For example, if a device fails between
8am and 9pm, call the system administrator, otherwise call the Help
Desk.
o Usage Policies control the selection and configuration of entities
based on specific "usage" data. Configuration Policies can be
modified or simply re-applied by Usage Policies. Examples of Usage
Policies include upgrading network forwarding services after a user
is verified to be a member of a "gold" service group, or
reconfiguring a printer to be able to handle the next job in its
queue.
o Security Policies deal with verifying that the client is actually
who the client purports to be, permitting or denying access to
resources, selecting and applying appropriate authentication
mechanisms, and performing accounting and auditing of resources.
o Service Policies characterize network and other services (not use
them). For example, all wide-area backbone interfaces shall use a
specific type of queuing.
Service policies describe services available in the network. Usage
policies describe the particular binding of a client of the network
to services available in the network.
These categories are represented in the Policy Core Information Model
by special values defined for the PolicyKeywords property of the
abstract class Policy.
2.1. Policy Scope
Policies represent business goals and objectives. A translation must
be made between these goals and objectives and their realization in
the network. An example of this could be a Service Level Agreement
(SLA), and its objectives and metrics (Service Level Objectives, or
SLOs), that are used to specify services that the network will provide
for a given client. The SLA will usually be written in high-level
business terminology. SLOs address more specific metrics in support of
the SLA. These high-level descriptions of network services and metrics
must be translated into lower-level, but also vendor- and device-
independent specifications. The Policy Core Information Model classes
are intended to serve as the foundation for these lower-level, vendor-
and device-independent specifications.
It is envisioned that the definition of the Policy Core Informational
Model in this draft is generic in nature and is applicable to Quality
of Service (QoS), to non-QoS networking applications (e.g., DHCP and
IPSec), and to non-networking applications (e.g., backup policies,
auditing access, etc.).
Moore, et al. Expires: Jul 2000 + 6 months [Page 7]
Internet Draft Policy Core Information Model July 2000
2.2. Declarative versus Procedural Model
The design of the Policy Core Information Model is influenced by a
declarative, not procedural, approach. More formally, a declarative
language is used to describe relational and functional languages.
Declarative languages describe relationships between variables in
terms of functions or inference rules, to which the interpreter or
compiler can apply a fixed algorithm in order to produce a result. An
imperative (or procedural) language specifies an explicit sequence of
steps to follow in order to produce a result.
It is important to note that this information model does not rule out
the use of procedural languages. Rather, it recognizes that both
declarative as well as procedural languages can be used to implement
policy. This information model is better viewed as being declarative
because the sequence of steps for doing the processing of declarative
statements tends to be left to the implementer. However, we have
provided the option of expressing the desired order of action
execution in this policy information model, and for expressing whether
the order is mandatory or not. In addition, rather than trying to
define algorithms or sets of instructions or steps that must be
followed by a policy rule, we instead define a set of modular building
blocks and relationships that can be used in a declarative or
procedural fashion to define policies.
Compare this to a strictly procedural model. Taking such an approach
would require that we specify the condition testing sequence, and the
action execution sequence, in the policy repository itself. This
would, indeed, constrain the implementer. This is why the policy model
is characterized as a declarative one. That is, the information model
defines a set of attributes, and a set of entities that contain these
attributes. However, it does NOT define either the algorithm to
produce a result using the attributes or an explicit sequence of steps
to produce a result.
There are several design considerations and trade-offs to make in this
respect.
1. On the one hand, we would like a policy definition language to be
reasonably human-friendly for ease of definitions and diagnostics.
On the other hand, given the diversity of devices (in terms of
their processing capabilities) which could act as policy decision
points, we would like to keep the language somewhat machine-
friendly. That is, it should be relatively simple to automate the
parsing and processing of the language in network elements. The
approach taken is to provide a set of classes and attributes that
can be combined in either a declarative or procedural approach to
express policies that manage network elements and services. The key
point is to avoid trying to standardize rules or sets of steps to
be followed in defining a policy. These must be left up to an
implementation. Interoperability is achieved by standardizing the
building blocks that are used to represent policy data and
information.
Moore, et al. Expires: Jul 2000 + 6 months [Page 8]
Internet Draft Policy Core Information Model July 2000
2. An important decision to make is the semantic style of the
representation of the information.
The declarative approach that we are describing falls short of
being a "true" declarative model. Such a model would also specify
the algorithms used to combine the information and policy rules to
achieve particular behavior. We avoid specifying algorithms for the
same reason that we avoid specifying sets of steps to be followed
in a policy rule. However, the design of the information model more
closely follows that of a declarative language, and may be easier
to understand if such a conceptual model is used. This leads to our
third point, acknowledging a lack of "completeness" and instead
relying on presenting information that the policy processing entity
will work with.
3. It is important to control the complexity of the specification,
trading off richness of expression of data in the core information
model for ease of implementation and use. It is important to
acknowledge the collective lack of experience in the field
regarding policies to control and manage network services and
hence avoid the temptation of aiming for "completeness". We should
instead strive to facilitate definition of a set of common policies
that customers require today (e.g., VPN and QoS) and allow
migration paths towards supporting complex policies as customer
needs and our understanding of these policies evolve with
experience. Specifically, in the context of the declarative style
language discussed above, it is important to avoid having full
blown predicate calculus as the language, as it would render many
important problems such as consistency checking and policy decision
point algorithms intractable. It is useful to consider a reasonably
constrained language from these perspectives.
The Policy Core Information Model strikes a balance between complexity
and lack of power by using the well understood logical concepts of
Disjunctive Normal Form and Conjunctive Normal Form for combining
simple policy conditions into more complex ones.
3. Overview of the Policy Core Information Model
The following diagram provides an overview of the five central classes
comprising the Policy Core Information Model, their associations to
each other, and their associations to other classes in the overall CIM
model. Note that the abstract class Policy and the two extension
classes VendorPolicyCondition and VendorPolicyAction are not shown.
NOTE: For cardinalities, "*" is an abbreviation for "0..n".
Moore, et al. Expires: Jul 2000 + 6 months [Page 9]
Internet Draft Policy Core Information Model July 2000
+-----------+
| System |
..... +--^-----^--+ .....
. . 1. 1. . .
*.(a).* .(b) .(c) *.(d).*
+--v---v---------+ . . +-v---v------------+
| PolicyGroup <........ . | PolicyRepository |
| | w * . | |
+------^---------+ . +-----^---------^--+
*. . 0..1 . 0..1 .
.(e) . .(f) .(g)
*. . . .
+------v------+ w * . . .
| <................. . .
| PolicyRule | . .
| | . .
| | . .
| <........................ . .
| |* (h) . . .
| | . . .
| | . . .
| | . . .
| | . . .
| | . . .
| | . . .
| | .* .* .
| | +---------v-------v--+ .
| | | PolicyCondition | .
| | *+--------------------+ .
| | (i) ^ .
| <.............. I .
| |* . I .
| | .* ^ .
| | +----v----------------------+ .
| | | PolicyTimePeriodCondition | .
| | +---------------------------+ .
| | (j) .
| <......................... .
| |* . .
| | .* .
| | +----------v---------+* .
| | | PolicyAction <.......
+-------------+ +--------------------+
Figure 1. Overview of the Core Policy Classes and Relationships
In this figure the boxes represent the classes, and the dotted arrows
represent the associations. The following associations appear:
(a) PolicyGroupInPolicyGroup
(b) PolicyGroupInSystem
Moore, et al. Expires: Jul 2000 + 6 months [Page 10]
Internet Draft Policy Core Information Model July 2000
(c) PolicyRuleInSystem
(d) PolicyRepositoryInPolicyRepository
(e) PolicyRuleInPolicyGroup
(f) PolicyConditionInPolicyRepository
(g) PolicyActionInPolicyRepository
(h) PolicyConditionInPolicyRule
(i) PolicyRuleValidityPeriod
(j) PolicyActionInPolicyRule
An association always connects two classes. The "two" classes may,
however, be the same class, as is the case with the
PolicyGroupInPolicyGroup association, which represents the recursive
containment of PolicyGroups in other PolicyGroups. The
PolicyRepositoryInPolicyRepository association is recursive in the
same way.
An association includes cardinalities for each of the related classes.
These cardinalities indicate how many instances of each class may be
related to an instance of the other class. For example, the
PolicyRuleInPolicyGroup association has the cardinality range "*'
(that is, "0..n") for both the PolicyGroup and PolicyRule classes.
These ranges are interpreted as follows:
o The "*" written next to PolicyGroup indicates that a PolicyRule may
be related to no PolicyGroups, to one PolicyGroup, or to more than
one PolicyGroup via the PolicyGroupInPolicyGroup association. In
other words, a PolicyRule may be contained in no PolicyGroups, in
one PolicyGroups, or in more than one PolicyGroup.
o The "*" written next to PolicyRule indicates that a PolicyGroup may
be related to no PolicyRules, to one PolicyRule, or to more than
one PolicyRule via the PolicyGroupInPolicyGroup association. In
other words, a PolicyGroup may contain no PolicyRules, one
PolicyRule, or more than one PolicyRule.
The "w" written next to the PolicyGroupInSystem and PolicyRuleInSystem
indicates that these are what CIM terms "aggregations with weak
references", or more briefly, "weak aggregations." A weak
aggregation is simply an indication of a naming scope. Thus these two
aggregations indicate that an instance of a PolicyGroup or PolicyRule
is named within the scope of a System object. A weak aggregation
implicitly has the cardinality 1..1 at the end opposite the 'w'.
The associations shown in Figure 1 are discussed in more detail in
Section 7.
Moore, et al. Expires: Jul 2000 + 6 months [Page 11]
Internet Draft Policy Core Information Model July 2000
4. Inheritance Hierarchies for the Policy Core Information Model
The following diagram illustrates the inheritance hierarchy for the
core policy classes:
ManagedElement (abstract)
|
+--Policy (abstract)
| |
| +---PolicyGroup
| |
| +---PolicyRule
| |
| +---PolicyCondition
| | |
| | +---PolicyTimePeriodCondition
| | |
| | +---VendorPolicyCondition
| |
| +---PolicyAction
| |
| +---VendorPolicyAction
|
+--ManagedSystemElement (abstract)
|
+--LogicalElement (abstract)
|
+--System (abstract)
|
+--AdminDomain (abstract)
|
+---PolicyRepository
Figure 2. Inheritance Hierarchy for the Core Policy Classes
ManagedElement, as well as the four additional CIM classes from which
PolicyRepository is derived, are defined in the CIM schema [1]. These
classes are not discussed in detail in this document.
In CIM, associations are also modeled as classes. For the Policy Core
Information Model, the inheritance hierarchy for the associations is
as follows:
Moore, et al. Expires: Jul 2000 + 6 months [Page 12]
Internet Draft Policy Core Information Model July 2000
[unrooted]
|
+---PolicyComponent (abstract)
| |
| +---PolicyGroupInPolicyGroup
| |
| +---PolicyRuleInPolicyGroup
| |
| +---PolicyConditionInPolicyRule
| |
| +---PolicyRuleValidityPeriod
| |
| +---PolicyActionInPolicyRule
|
+---Dependency (abstract)
| |
| +---PolicyInSystem (abstract)
| |
| +---PolicyGroupInSystem
| |
| +---PolicyRuleInSystem
| |
| +---PolicyConditionInPolicyRepository
| |
| +---PolicyActionInPolicyRepository
|
+---Component (abstract)
|
+---SystemComponent
|
+---PolicyRepositoryInPolicyRepository
Figure 3. Inheritance Hierarchy for the Core Policy Associations
5. Details of the Model
The following subsections discuss several specific issues related to
the Policy Core Information Model.
5.1. Reusable versus Rule-Specific Conditions and Actions
Policy conditions and policy actions can be partitioned into two
groups: ones associated with a single policy rule, and ones that are
reusable, in the sense that they may be associated with more than one
policy rule. Conditions and actions in the first group are termed
"rule-specific" conditions and actions; those in the second group are
characterized as "reusable".
It is important to understand that the difference between a rule-
specific condition or action and a reusable one is based on the intent
of the policy administrator for the condition or action, rather than
Moore, et al. Expires: Jul 2000 + 6 months [Page 13]
Internet Draft Policy Core Information Model July 2000
on the current associations in which the condition or action
participates. Thus a reusable condition or action (that is, one that
a policy administrator has created to be reusable) may at some point
in time be associated with exactly one policy rule, without thereby
becoming rule-specific.
There is no inherent difference between a rule-specific condition or
action and a reusable one. There are, however, differences in how
they are treated in a policy repository. For example, it's natural to
make the access permissions for a rule-specific condition or action
identical to those for the rule itself. It's also natural for a rule-
specific condition or action to be removed from the policy repository
at the same time the rule is. With reusable conditions and actions,
on the other hand, access permissions and existence criteria must be
expressible without reference to a policy rule.
The preceding paragraph does not contain an exhaustive list of the
ways in which reusable and rule-specific conditions should be treated
differently. Its purpose is merely to justify making a semantic
distinction between rule-specific and reusable, and then reflecting
this distinction in the policy model itself.
An issue is highlighted by reusable and rule-specific policy
conditions and policy actions: the lack of a programmatic capability
for expressing complex constraints involving multiple associations.
Taking PolicyCondition as an example, there are two aggregations to
look at. PolicyConditionInPolicyRule has the cardinality * at both
ends, and PolicyConditionInPolicyRepository has the cardinality * at
the PolicyCondition end, and [0..1] at the PolicyRepository end.
Globally, these cardinalities are correct. However, there's more to
the story, which only becomes clear if we examine the cardinalities
separately for the two cases of a rule-specific PolicyCondition and a
reusable one.
For a rule-specific PolicyCondition, the cardinality of
PolicyConditionInPolicyRule at the PolicyRule end is [1..1], rather
than [0..n] (recall that * is an abbreviation for [0..n]), since the
condition is unique to one policy rule. And the cardinality of
PolicyConditionInPolicyRepository at the PolicyRepository end is
[0..0], since the condition is not in the "re-usable" repository.
This is OK, since these are both subsets of the specified
cardinalities.
For a reusable PolicyCondition, however, the cardinality of
PolicyConditionInPolicyRepository at the PolicyRepository end is
[1..1], since the condition must be in the repository. And, the
cardinality of PolicyConditionInPolicyRule at the PolicyRule end is
[0..n]. This last point is important: a reusable PolicyCondition may
be associated with 0, 1, or more than 1 PolicyRules, via exactly the
same association PolicyConditionInPolicyRule that binds a rule-
specific condition to its PolicyRule.
Moore, et al. Expires: Jul 2000 + 6 months [Page 14]
Internet Draft Policy Core Information Model July 2000
Currently the only way to document constraints of this type is
textually. More formal methods for documenting complex constraints
are needed.
5.2. Roles
5.2.1. Roles and Role Combinations
The concept of role is central to the design of the entire Policy
Framework. The idea behind roles is a simple one. Rather than
configuring, and then later having to update the configuration of,
hundreds or thousands (or more) of resources in a network, a policy
administrator assigns each resource to one or more roles, and then
specifies the policies for each of these roles. The Policy Framework
is then responsible for configuring each of the resources associated
with a role in such a way that it behaves according to the policies
specified for that role. When network behavior must be changed, the
policy administrator can perform a single update to the policy for a
role, and the Policy Framework will ensure that the necessary
configuration updates are performed on all the resources playing that
role.
A more formal definition of a role is as follows:
A role is a type of attribute that is used to select one or more
policies for a set of entities and/or components from among a much
larger set of available policies.
Roles can be combined together. Here is a formal definition of a
"role-combination":
A role-combination is a set of attributes that are used to select
one or more policies for a set of entities and/or components from
among a much larger set of available policies. As the examples
below illustrate, the selection process for a role combination
chooses policies associated with the combination itself, policies
associated with each of its sub-combinations, and policies
associated with each of the individual roles in the role-
combination.
It is important to note that a role is more than an attribute. A role
defines a particular function of an entity or component that can be
used to identify particular behavior associated with that entity or
component. This difference is critical, and is most easily understood
by thinking of a role as a selector. When used in this manner, one
role (or role-combination) selects a different set of policies than a
different role (or role-combination) does.
Roles and role-combinations are especially useful in selecting which
policies are applicable to a particular set of entities or components
when the policy repository can store thousands or hundreds of
thousands of policies. This use emphasizes the ability of the role (or
Moore, et al. Expires: Jul 2000 + 6 months [Page 15]
Internet Draft Policy Core Information Model July 2000
role-combination) to select the small subset of policies that are
applicable from a huge set of policies that are available.
An example will illustrate how role-combinations actually work.
Suppose an installation has three roles defined for interfaces:
"Ethernet", "Campus", and "WAN". In the Policy Repository, some
policy rules could be associated with the role "Ethernet"; these rules
would apply to all Ethernet interfaces, regardless of whether they
were on the campus side or the WAN side. Other rules could be
associated with the role-combination "Campus"+"Ethernet"; these rules
would apply to the campus-side Ethernet interfaces, but not to those
on the WAN side. Finally, a third set of rules could be associated
with the role-combination "Ethernet"+"WAN"; these rules would apply to
the WAN-side Ethernet interfaces, but not to those on the campus side.
(The roles in a role-combination appear in alphabetical order in these
examples, because that is how they appear in the information model.)
If we have a specific interface A that's associated with the role-
combination "Ethernet"+"WAN", we see that it should have three
categories of policy rules applied to it: those for the "Ethernet"
role, those for the "WAN" role, and those for the role-combination
"Ethernet"+"WAN". Going one step further, if interface B is
associated with the role-combination "branch-office"+"Ethernet"+"WAN",
then B should have seven categories of policy rules applied to it -
those associated with the following role-combinations:
o "branch-office"
o "Ethernet"
o "WAN"
o "branch-office"+"Ethernet"
o "branch-office"+"WAN"
o "Ethernet"+"WAN"
o "branch-office"+"Ethernet"+"WAN".
In order to get all of the right policy rules for a resource like
interface B, a PDP must expand the single role-combination it receives
for B into this list of seven role-combinations, and then retrieve
from the Policy Repository the corresponding seven sets of policy
rules. Of course this example is unusually complicated: the normal
case will involve expanding a two-role combination into three values
identifying three sets of policy rules.
Role-combinations also help to simplify somewhat the problem of
identifying conflicts between policy rules. With role-combinations,
it is possible for a policy administrator to specify one set of policy
rules for campus-side Ethernet interfaces, and a second set of policy
rules for WAN-side Ethernet interfaces, without having to worry about
conflicts between the two sets of rules. The policy administrator
simply "turns off" conflict detection for these two sets of rules, by
telling the policy management system that the roles "Campus" and "WAN"
are incompatible with each other. This indicates that the role
combination will never occur, and therefore conflicts will never
occur. In some cases the technology itself might identify
Moore, et al. Expires: Jul 2000 + 6 months [Page 16]
Internet Draft Policy Core Information Model July 2000
incompatible roles: "Ethernet" and "FrameRelay", for example. But
for less precise terms like "Campus" and "WAN", the policy
administrator must say whether they identify incompatible roles.
When the policy administrator does this, there are three effects:
1. If an interface has assigned to it a role-combination involving
both "Campus" and "WAN", then the policy management system can flag
it as an error.
2. If a policy rule is associated with a role-combination involving
both "Campus" and "WAN", then the policy management system can flag
it as an error.
3. If the policy management system sees two policy rules, where one is
tied to the role "Campus" (or to a role-combination that includes
the role "Campus") and the other is tied to the role "WAN" (or to a
role-combination that includes the role "WAN"), then the system
does not need to look for conflicts between the two policy rules:
because of the incompatible roles, the two rules cannot possibly
conflict.
+-------------------+
| Policy Repository |
+-------------------+
V
V retrieval of policy
V
+---------+
| PDP/PEP |
+---------+
v
v application of policy
v
+----------------+
| Network Entity |
+----------------+
Figure 4. Retrieval and Application of a Policy
Figure 4, which is introduced only as an example of how the Policy
Framework might be implemented by a collection of network components,
illustrates how roles operate within the Policy Framework. Because
the distinction between them is not important to this discussion, the
PDP and the PEP are combined in one box. The points illustrated here
apply equally well, though, to an environment where the PDP and the
PEP are implemented separately.
A role represents a functional characteristic or capability of a
resource to which policies are applied. Examples of roles include
Backbone interface, Frame Relay interface, BGP-capable router, web
Moore, et al. Expires: Jul 2000 + 6 months [Page 17]
Internet Draft Policy Core Information Model July 2000
server, firewall, etc. The multiple roles assigned to a single
resource are combined to form that resource's role combination. Role
combinations are represented in the PCIM by values of the PolicyRoles
property in the PolicyRule class. A PDP uses policy roles as follows
to identify the policies it needs to be aware of:
1. The PDP learns in some way the list of roles that its PEPs play.
This information might be configured at the PDP, the PEPs might
supply it to the PDP, or the PDP might retrieve it from a
repository.
2. Using repository-specific means, the PDP determines where to look
for policy rules that might apply to it.
3. Using the roles and role-combinations it received from its PEPs
as indicated in the examples above, the PDP is able to locate and
retrieve the policy rules that are relevant to it.
5.2.2. The PolicyRoles Property
As indicated earlier, PolicyRoles is a property associated with a
policy rule. It is an array holding "role combinations" for the policy
rule, and correlates with the roles defined for a network resource.
Using the PolicyRoles property, it is possible to mark a policy rule
as applying, for example, to a Frame Relay interface or to a backbone
ATM interface. The PolicyRoles property take strings of the form:
<RoleName>[&&<RoleName>]*
Each value of this property represents a role combination, including
the special case of a "combination" containing only one role. As the
format indicates, the role names in a role combination are ANDed
together to form a single selector. The multiple values of the
PolicyRoles property are logically ORed, to make it possible for a
policy rule to have multiple selectors.
The individual role names in a role combination must appear in
alphabetical order (according to the collating sequence for UCS-2
characters), to make the string matches work correctly. The role
names used in an environment are specified by the policy
administrator.
5.3. Local Time and UTC Time in PolicyTimePeriodConditions
An instance of PolicyTimePeriodCondition has up to five properties
that represent times: TimePeriod, MonthOfYearMask, DayOfMonthMask,
DayOfWeekMask, and TimeOfDayMask. All of the time-related properties
in an instance of PolicyTimePeriodCondition represent one of two types
of times: local time at the place where a policy rule is applied, or
UTC time. The property LocalOrUtcTime indicates which time
representation applies to an instance of PolicyTimePeriodCondition.
Moore, et al. Expires: Jul 2000 + 6 months [Page 18]
Internet Draft Policy Core Information Model July 2000
Since the PCIM provides only for local time and UTC time, a Policy
Management Tool that provides for other time representations (for
example, a fixed time at a particular location) will need to map from
these other representations to either local time or UTC time. An
example will illustrate the nature of this mapping.
Suppose a policy rule is tied to the hours of operation for a Help
Desk: 0800 to 2000 Monday through Friday [US] Eastern Time. In order
to express these times in PolicyTimePeriodCondition, a management tool
must convert them to UTC times. (They are not local times, because
they refer to a single time interval worldwide, not to intervals tied
to the local clocks at the locations where the PolicyRule is being
applied.) As reference [10] points out, mapping from [US] Eastern
Time to UTC time is not simply a matter of applying an offset: the
offset between [US] Eastern Time and UTC time switches between -0500
and -0400 depending on whether Daylight Savings Time is in effect in
the US.
Suppose the policy administrator's goal is to have a policy rule be
valid from 0800 until 1200 [US] Eastern Time on every Monday, within
the overall time period from the beginning of 2000 until the end of
2001. The Policy Management Tool could either be configured with the
definition of what [US] Eastern Time means, or it could be configured
with knowledge of where to go to get this information. Reference [10]
contains further discussion of time zone definitions and where they
might reside.
Armed with knowledge about [US] Eastern Time, the Policy Management
Tool would create however many instances of PolicyTimePeriodCondition
it needed to represent the desired intervals. Note that while there
is an increased number of PolicyTimePeriodCondition instances, there
is still just one PolicyRule, which is tied to all the
PolicyTimePeriodCondition instances via the aggregation
PolicyRuleValidityPeriod. Here are the first two of these instances:
1. TimePeriod: 20000101T050000/20000402T070000
DayOfWeekMask: { Monday }
TimeOfDayMask: T130000/T170000
LocalOrUtcTime: UTC
2. TimePeriod: 20000402T070000/20001029T070000
DayOfWeekMask: { Monday }
TimeOfDayMask: T120000/T160000
LocalOrUtcTime: UTC
There would be three more similar instances, for winter 2000-2001,
summer 2001, and winter 2001 up through December 31.
Had the example been chosen differently, there could have been even
more instances of PolicyTimePeriodCondition. If, for example, the
Moore, et al. Expires: Jul 2000 + 6 months [Page 19]
Internet Draft Policy Core Information Model July 2000
time interval had been from 0800 - 2200 [US] Eastern Time on Mondays,
instance 1 above would have split into two instances: one with a UTC
time interval of T130000/T240000 on Mondays, and another with a UTC
time interval of T000000/T030000 on Tuesdays. So the end result would
have been ten instances of PolicyTimePeriodCondition, not five.
By restricting PolicyTimePeriodCondition to local time and UTC time,
the PCIM places the difficult and expensive task of mapping from
"human" time representations to machine-friendly ones in the Policy
Management Tool. Another approach would have been to place in
PolicyTimePeriodCondition a means of representing a named time zone,
such as [US] Eastern Time. This, however, would have passed the
difficult mapping responsibility down to the PDPs and PEPs. It is
better to have a mapping such as the one described above done once in
a Policy Management Tool, rather than having it done over and over in
each of the PDPs (and possibly PEPs) that need to apply a PolicyRule.
5.4. CIM Data Types
Since PCIM extends the CIM Schema, a correspondence between data types
used in both CIM and PCIM is needed. The following CIM data types are
used in the class definitions that follow in Sections 6 and 7:
o uint8 unsigned 8-bit integer
o uint16 unsigned 16-bit integer
o boolean Boolean
o string UCS-2 string.
Strings in CIM are stored as UCS-2 characters, where each character is
encoded in two octets. Thus string values may need to be converted
when moving between a CIM environment and one that uses a different
string encoding. For example, in an LDAP-accessible directory,
attributes of type DirectoryString are stored in UTF-8 format. RFC
2279 [7] explains how to convert between these two formats.
When it is applied to a CIM string, a MaxLen value refers to the
maximum number of characters in the string, rather than to the maximum
number of octets.
In addition to the CIM data types listed above, the association
classes in Section 7 use the following type:
o <classname> ref strongly typed reference.
6. Class Definitions
There are a significant number of differences between CIM and LDAP
class specifications. The ones that are relevant to the abbreviated
class specifications in this document are the following:
Moore, et al. Expires: Jul 2000 + 6 months [Page 20]
Internet Draft Policy Core Information Model July 2000
o Instead of LDAP's three class types (abstract, auxiliary,
structural), CIM has only two: abstract and instantiable. The
type of a CIM class is indicated by the Boolean qualifier ABSTRACT.
o CIM uses the term "property" for what LDAP terms an "attribute".
o CIM uses the array notation "[ ]" to indicate that a property is
multi-valued. CIM defines three types of arrays: bags (contents
are unordered, duplicates allowed), ordered bags (contents are
ordered but duplicates are allowed) and indexed arrays (contents
are ordered and no duplicates are allowed).
o CIM classes and properties are identified by name, not by OID.
o CIM classes use a different naming scheme for native
implementations, than LDAP. The CIM naming scheme is documented in
Appendix A since it is not critical to understanding the
information model, and only applies when communicating with a
native CIM implementation.
o In LDAP, attribute definitions are global, and the same attribute
may appear in multiple classes. In CIM, a property is defined
within the scope of a single class definition. The property may be
inherited into subclasses of the class in which it is defined, but
otherwise it cannot appear in other classes. One side effect of
this difference is that CIM property names tend to be much shorter
than LDAP attribute names, since they are implicitly scoped by the
name of the class in which they are defined.
There is also a notational convention that this document follows, to
improve readability. In CIM, all class and property names are
prefixed with the characters "CIM_". These prefixes have been omitted
throughout this document, with one exception regarding naming,
documented in Appendix A.
For the complete definition of the CIM specification language, see
reference [2].
6.1. The Abstract Class "Policy"
The abstract class Policy collects several properties that may be
included in instances of any of the Core Policy classes (or their
subclasses). For convenience, the two properties that Policy inherits
from ManagedElement in the CIM schema are shown here as well.
The class definition is as follows:
NAME Policy
DESCRIPTION An abstract class with four properties for
describing a policy-related instance.
DERIVED FROM ManagedElement
ABSTRACT TRUE
PROPERTIES CommonName (CN)
Moore, et al. Expires: Jul 2000 + 6 months [Page 21]
Internet Draft Policy Core Information Model July 2000
PolicyKeywords[ ]
// Caption (inherited)
// Description (inherited)
6.1.1. The Property "CommonName (CN)"
The CN, or CommonName, property corresponds to the X.500 attribute
commonName (cn). In X.500 this property specifies one or more user-
friendly names (typically only one name) by which an object is
commonly known, names that conform to the naming conventions of the
country or culture with which the object is associated. In the CIM
model, however, the CommonName property is single-valued.
NAME CN
DESCRIPTION A user-friendly name of a policy-related object.
SYNTAX string
6.1.2. The Multi-valued Property "PolicyKeywords"
This property provides a set of one or more keywords that a policy
administrator may use to assist in characterizing or categorizing a
policy object. Keywords are of one of two types:
o Keywords defined in this document, or in documents that define
subclasses of the classes defined in this document. These keywords
provide a vendor-independent, installation-independent way of
characterizing policy objects.
o Installation-dependent keywords for characterizing policy objects.
Examples include "Engineering", "Billing", and "Review in December
2000".
This document defines the following keywords: "UNKNOWN",
"CONFIGURATION", "USAGE", "SECURITY", "SERVICE", "MOTIVATIONAL",
"INSTALLATION", and "EVENT". These concepts were defined earlier in
Section 2.
One additional keyword is defined: "POLICY". The role of this
keyword is to identify policy-related instances that would not
otherwise be identifiable as being related to policy. It may be
needed in some repository implementations.
Documents that define subclasses of the Policy Core Information Model
classes SHOULD define additional keywords to characterize instances of
these subclasses. By convention, keywords defined in conjunction with
class definitions are in uppercase. Installation-defined keywords can
be in any case.
The property definition is as follows:
NAME PolicyKeywords
DESCRIPTION A set of keywords for characterizing /categorizing
policy objects.
Moore, et al. Expires: Jul 2000 + 6 months [Page 22]
Internet Draft Policy Core Information Model July 2000
SYNTAX string
6.1.3. The Property "Caption" (Inherited from ManagedElement)
This property provides a one-line description of a policy-related
object.
NAME Caption
DESCRIPTION A one-line description of this policy-related
object.
SYNTAX string
6.1.4. The Property "Description" (Inherited from ManagedElement)
This property provides a longer description than that provided by the
caption property.
NAME Description
DESCRIPTION A long description of this policy-related object.
SYNTAX string
6.2. The Class "PolicyGroup"
This class is a generalized aggregation container. It enables either
PolicyRules or PolicyGroups, but not both, to be aggregated in a
single container. Loops, including the degenerate case of a
PolicyGroup that contains itself, are not allowed when PolicyGroups
contain other PolicyGroups.
PolicyGroups and their nesting capabilities are shown in Figure 5
below. Note that a PolicyGroup can nest other PolicyGroups, and there
is no restriction on the depth of the nesting in sibling PolicyGroups.
+---------------------------------------------------+
| PolicyGroup |
| |
| +--------------------+ +-----------------+ |
| | PolicyGroup A | | PolicyGroup X | |
| | | | | |
| | +----------------+ | ooo | | |
| | | PolicyGroup A1 | | | | |
| | +----------------+ | | | |
| +--------------------+ +-----------------+ |
+---------------------------------------------------+
Figure 5. Overview of the PolicyGroup class
As a simple example, think of the highest level PolicyGroup shown in
Figure 5 above as a logon policy for US employees of a company. This
PolicyGroup may be called USEmployeeLogonPolicy, and may aggregate
several PolicyGroups that provide specialized rules per location.
Hence, PolicyGroup A in Figure 5 above may define logon rules for
Moore, et al. Expires: Jul 2000 + 6 months [Page 23]
Internet Draft Policy Core Information Model July 2000
employees on the West Coast, while another PolicyGroup might define
logon rules for the Midwest (e.g., PolicyGroup X), and so forth.
Note also that the depth of each PolicyGroup does not need to be the
same. Thus, the WestCoast PolicyGroup might have several additional
layers of PolicyGroups defined for any of several reasons (different
locales, number of subnets, etc.). The PolicyRules are therefore
contained at n levels from the USEmployeeLogonPolicyGroup. Compare
this to the Midwest PolicyGroup (PolicyGroup X), which might directly
contain PolicyRules.
The class definition for PolicyGroup is as follows:
NAME PolicyGroup
DESCRIPTION A container for either a set of related PolicyRules
or a set of related PolicyGroups.
DERIVED FROM Policy
ABSTRACT FALSE
PROPERTIES NONE
No properties are defined for this class since it inherits all its
properties from Policy. The class exists to aggregate PolicyRules or
other PolicyGroups. It is directly instantiable. In an
implementation, various key/identification properties MUST be defined.
The keys for a native CIM implementation are defined in Appendix A,
Section 14.1.1. Keys for an LDAP implementation are defined in the
LDAP mapping of this information model [11].
6.3. The Class "PolicyRule"
This class represents the "If Condition then Action" semantics
associated with a policy. A PolicyRule condition, in the most general
sense, is represented as either an ORed set of ANDed conditions
(Disjunctive Normal Form, or DNF) or an ANDed set of ORed conditions
(Conjunctive Normal Form, or CNF). Individual conditions may either be
negated (NOT C) or unnegated (C). The actions specified by a
PolicyRule are to be performed if and only if the PolicyRule condition
(whether it is represented in DNF or CNF) evaluates to TRUE.
The conditions and actions associated with a policy rule are modeled,
respectively, with subclasses of the classes PolicyCondition and
PolicyAction. These condition and action objects are tied to
instances of PolicyRule by the PolicyConditionInPolicyRule and
PolicyActionInPolicyRule aggregations.
As illustrated above in Section 3, a policy rule may also be
associated with one or more policy time periods, indicating the
schedule according to which the policy rule is active and inactive.
In this case it is the PolicyRuleValidityPeriod aggregation that
provides the linkage.
A policy rule is illustrated conceptually in Figure 6. below.
Moore, et al. Expires: Jul 2000 + 6 months [Page 24]
Internet Draft Policy Core Information Model July 2000
+------------------------------------------------+
| PolicyRule |
| |
| +--------------------+ +-----------------+ |
| | PolicyCondition(s) | | PolicyAction(s) | |
| +--------------------+ +-----------------+ |
| |
| +------------------------------+ |
| | PolicyTimePeriodCondition(s) | |
| +------------------------------+ |
+------------------------------------------------+
Figure 6. Overview of the PolicyRule Class
The PolicyRule class uses the property ConditionListType, to indicate
whether the conditions for the rule are in DNF or CNF. The
PolicyConditionInPolicyRule aggregation contains two additional
properties to complete the representation of the rule's conditional
expression. The first of these properties is an integer to partition
the referenced conditions into one or more groups, and the second is a
Boolean to indicate whether a referenced condition is negated. An
example shows how ConditionListType and these two additional
properties provide a unique representation of a set of conditions in
either DNF or CNF.
Suppose we have a PolicyRule that aggregates five PolicyConditions C1
through C5, with the following values in the properties of the five
PolicyConditionInPolicyRule associations:
C1: GroupNumber = 1, ConditionNegated = FALSE
C2: GroupNumber = 1, ConditionNegated = TRUE
C3: GroupNumber = 1, ConditionNegated = FALSE
C4: GroupNumber = 2, ConditionNegated = FALSE
C5: GroupNumber = 2, ConditionNegated = FALSE
If ConditionListType = DNF, then the overall condition for the
PolicyRule is:
(C1 AND (NOT C2) AND C3) OR (C4 AND C5)
On the other hand, if ConditionListType = CNF, then the overall
condition for the PolicyRule is:
(C1 OR (NOT C2) OR C3) AND (C4 OR C5)
In both cases, there is an unambiguous specification of the overall
condition that is tested to determine whether to perform the actions
associated with the PolicyRule.
The class definition is as follows:
NAME PolicyRule
Moore, et al. Expires: Jul 2000 + 6 months [Page 25]
Internet Draft Policy Core Information Model July 2000
DESCRIPTION The central class for representing the "If
Condition then Action" semantics associated with a
policy rule.
DERIVED FROM Policy
ABSTRACT FALSE
PROPERTIES Enabled
ConditionListType
RuleUsage
Priority
Mandatory
SequencedActions
PolicyRoles
The PolicyRule class is directly instantiable. In an implementation,
various key/identification properties MUST be defined. The keys for a
native CIM implementation are defined in Appendix A, Section 14.1.2.
Keys for an LDAP implementation are defined in the LDAP mapping of
this information model [11].
6.3.1. The Property "Enabled"
This property indicates whether a policy rule is currently enabled,
from an administrative point of view. Its purpose is to allow a
policy administrator to enable or disable a policy rule without having
to add it to, or remove it from, the policy repository.
The property also supports the value 'enabledForDebug'. When the
property has this value, the entity evaluating the policy condition(s)
is being told to evaluate the conditions for the policy rule, but not
to perform the actions if the conditions evaluate to TRUE. This value
serves as a debug vehicle when attempting to determine what policies
would execute in a particular scenario, without taking any actions to
change state during the debugging.
The property definition is as follows:
NAME Enabled
DESCRIPTION An enumeration indicating whether a policy rule is
administratively enabled, administratively
disabled, or enabled for debug mode.
SYNTAX uint16
VALUES enabled(1), disabled(2), enabledForDebug(3)
DEFAULT VALUE enabled(1)
6.3.2. The Property "ConditionListType"
This property is used to specify whether the list of policy conditions
associated with this policy rule is in disjunctive normal form (DNF)
or conjunctive normal form (CNF). If this property is not present,
the list type defaults to DNF. The property definition is as follows:
NAME ConditionListType
Moore, et al. Expires: Jul 2000 + 6 months [Page 26]
Internet Draft Policy Core Information Model July 2000
DESCRIPTION Indicates whether the list of policy conditions
associated with this policy rule is in disjunctive
normal form (DNF) or conjunctive normal form (CNF).
SYNTAX uint16
VALUES DNF(1), CNF(2)
DEFAULT VALUE DNF(1)
6.3.3. The Property "RuleUsage"
This property is a free-form string that recommends how this policy
should be used. The property definition is as follows:
NAME RuleUsage
DESCRIPTION This property is used to provide guidelines on how
this policy should be used.
SYNTAX string
6.3.4. The Property "Priority"
This property provides a non-negative integer for prioritizing policy
rules relative to each other. For policy rules that have this
property, larger integer values indicate higher priority. Since one
purpose of this property is to allow specific, ad hoc policy rules to
temporarily override established policy rules, an instance that has
this property set has a higher priority than all instances that lack
it.
Prioritization among policy rules provides a basic mechanism for
resolving policy conflicts.
The property definition is as follows:
NAME Priority
DESCRIPTION A non-negative integer for prioritizing this
PolicyRule relative to other PolicyRules. A larger
value indicates a higher priority.
SYNTAX uint16
DEFAULT VALUE 0
6.3.5. The Property "Mandatory"
This property indicates whether evaluation (and possibly action
execution) of a PolicyRule is mandatory or not. Its concept is
similar to the ability to mark packets for delivery or possible
discard, based on network traffic and device load.
The evaluation of a PolicyRule MUST be attempted if the Mandatory
property value is TRUE. If the Mandatory property value of a
PolicyRule is FALSE, then the evaluation of the rule is "best effort"
and MAY be ignored.
The property definition is as follows:
Moore, et al. Expires: Jul 2000 + 6 months [Page 27]
Internet Draft Policy Core Information Model July 2000
NAME Mandatory
DESCRIPTION A flag indicating that the evaluation of the
PolicyConditions and execution of PolicyActions (if
the condition list evaluates to TRUE) is required.
SYNTAX boolean
DEFAULT VALUE TRUE
6.3.6. The Property "SequencedActions"
This property gives a policy administrator a way of specifying how the
ordering of the policy actions associated with this PolicyRule is to
be interpreted. Three values are supported:
o mandatory(1): Do the actions in the indicated order, or don't do
them at all.
o recommended(2): Do the actions in the indicated order if you can,
but if you can't do them in this order, do them in another order if
you can.
o dontCare(3): Do them -- I don't care about the order.
When error / event reporting is addressed for the Policy Framework,
suitable codes will be defined for reporting that a set of actions
could not be performed in an order specified as mandatory (and thus
were not performed at all), that a set of actions could not be
performed in a recommended order (and moreover could not be performed
in any order), or that a set of actions could not be performed in a
recommended order (but were performed in a different order). The
property definition is as follows:
NAME SequencedActions
DESCRIPTION An enumeration indicating how to interpret the
action ordering indicated via the
PolicyActionInPolicyRule aggregation.
SYNTAX uint16
VALUES mandatory(1), recommended(2), dontCare(3)
DEFAULT VALUE dontCare(3)
6.3.7. The Multi-valued Property "PolicyRoles"
This property represents the roles and role combinations associated
with a policy rule. Each value represents one role combination.
Since this is a multi-valued property, more than one role combination
can be associated with a single policy rule. Each value is a string
of the form
<RoleName>[&&<RoleName>]*
where the individual role names appear in alphabetical order
(according to the collating sequence for UCS-2). The property
definition is as follows:
Moore, et al. Expires: Jul 2000 + 6 months [Page 28]
Internet Draft Policy Core Information Model July 2000
NAME PolicyRoles
DESCRIPTION A set of strings representing the roles and role
combinations associated with a policy rule. Each
value represents one role combination.
SYNTAX string
6.4. The Class "PolicyCondition"
The purpose of a policy condition is to determine whether or not the
set of actions (aggregated in the PolicyRule that the condition
applies to) should be executed or not. For the purposes of the Policy
Core Information Model, all that matters about an individual
PolicyCondition is that it evaluates to TRUE or FALSE. (The
individual PolicyConditions associated with a PolicyRule are combined
to form a compound expression in either DNF or CNF, but this is
accomplished via the ConditionListType property, discussed above, and
by the properties of the PolicyConditionInPolicyRule aggregation,
introduced above and discussed further in Section 7.6 below.) A
logical structure within an individual PolicyCondition may also be
introduced, but this would have to be done in a subclass of
PolicyCondition.
Because it is general, the PolicyCondition class does not itself
contain any "real" conditions. These will be represented by
properties of the domain-specific subclasses of PolicyCondition.
+---------------------------------------------------------------+
| Policy Conditions in DNF |
| +-------------------------+ +-----------------------+ |
| | AND list | | AND list | |
| | +-------------------+ | | +-----------------+ | |
| | | PolicyCondition | | | | PolicyCondition | | |
| | +-------------------+ | | +-----------------+ | |
| | +-------------------+ | | +-----------------+ | |
| | | PolicyCondition | | ... | | PolicyCondition | | |
| | +-------------------+ | ORed | +-----------------+ | |
| | ... | | ... | |
| | ANDed | | ANDed | |
| | +-------------------+ | | +-----------------+ | |
| | | PolicyCondition | | | | PolicyCondition | | |
| | +-------------------+ | | +-----------------+ | |
| +-------------------------+ +-----------------------+ |
+---------------------------------------------------------------+
Figure 7. Overview of Policy Conditions in DNF
This figure illustrates that when policy conditions are in DNF, there
are one or more sets of conditions that are ANDed together to form AND
lists. An AND list evaluates to TRUE if and only if all of its
constituent conditions evaluate to TRUE. The overall condition then
evaluates to TRUE if and only if at least one of its constituent AND
lists evaluates to TRUE.
Moore, et al. Expires: Jul 2000 + 6 months [Page 29]
Internet Draft Policy Core Information Model July 2000
+---------------------------------------------------------------+
| Policy Conditions in CNF |
| +-------------------------+ +-----------------------+ |
| | OR list | | OR list | |
| | +-------------------+ | | +-----------------+ | |
| | | PolicyCondition | | | | PolicyCondition | | |
| | +-------------------+ | | +-----------------+ | |
| | +-------------------+ | | +-----------------+ | |
| | | PolicyCondition | | ... | | PolicyCondition | | |
| | +-------------------+ | ANDed | +-----------------+ | |
| | ... | | ... | |
| | ORed | | ORed | |
| | +-------------------+ | | +-----------------+ | |
| | | PolicyCondition | | | | PolicyCondition | | |
| | +-------------------+ | | +-----------------+ | |
| +-------------------------+ +-----------------------+ |
+---------------------------------------------------------------+
Figure 8. Overview of Policy Conditions in CNF
In this figure, the policy conditions are in CNF. Consequently, there
are one or more OR lists, each of which evaluates to TRUE if and only
if at least one of its constituent conditions evaluates to TRUE. The
overall condition then evaluates to TRUE if and only if ALL of its
constituent OR lists evaluate to TRUE.
The class definition of PolicyCondition is as follows:
NAME PolicyCondition
DESCRIPTION A class representing a rule-specific or reusable
policy condition to be evaluated in conjunction
with a policy rule.
DERIVED FROM Policy
ABSTRACT FALSE
PROPERTIES NONE
No properties are defined for this class since it inherits all its
properties from Policy. The class exists as an abstract superclass
for domain-specific policy conditions, defined in subclasses. In an
implementation, various key/identification properties MUST be defined
for the class or its instantiable subclasses. The keys for a native
CIM implementation are defined in Appendix A, Section 14.2. Keys for
an LDAP implementation are defined in the LDAP mapping of this
information model [11].
When identifying and using the PolicyCondition class, it is necessary
to remember that a condition can be rule-specific or reusable. This
was discussed above in Section 5.1. The distinction between the two
types of policy conditions lies in the associations in which an
instance can participate, and in how the different instances are
named. Conceptually, a reusable policy condition resides in a policy
repository, and is named within the scope of that repository. On the
Moore, et al. Expires: Jul 2000 + 6 months [Page 30]
Internet Draft Policy Core Information Model July 2000
other hand, a rule-specific policy condition is, as the name suggests,
named within the scope of the single policy rule to which it is
related.
The distinction between rule-specific and reusable PolicyConditions
affects the CIM naming, defined in Appendix A, and the LDAP mapping
[11].
6.5. The Class "PolicyTimePeriodCondition"
This class provides a means of representing the time periods during
which a policy rule is valid, i.e., active. At all times that fall
outside these time periods, the policy rule has no effect. A policy
rule is treated as valid at all times if it does not specify a
PolicyTimePeriodCondition.
In some cases a PDP may need to perform certain setup / cleanup
actions when a policy rule becomes active / inactive. For example,
sessions that were established while a policy rule was active might
need to be taken down when the rule becomes inactive. In other cases,
however, such sessions might be left up: in this case, the effect of
deactivating the policy rule would just be to prevent the
establishment of new sessions. Setup / cleanup behaviors on validity
period transitions are not currently addressed by the PCIM, and must
be specified in 'guideline' documents, or via subclasses of
PolicyRule, PolicyTimePeriodCondition or other concrete subclasses of
Policy. If such behaviors need to be under the control of the policy
administrator, then a mechanism to allow this control must also be
specified in the subclass.
PolicyTimePeriodCondition is defined as a subclass of PolicyCondition.
This is to allow the inclusion of time-based criteria in the AND/OR
condition definitions for a PolicyRule.
Instances of this class may have up to five properties identifying
time periods at different levels. The values of all the properties
present in an instance are ANDed together to determine the validity
period(s) for the instance. For example, an instance with an overall
validity range of January 1, 2000 through December 31, 2000; a month
mask that selects March and April; a day-of-the-week mask that selects
Fridays; and a time of day range of 0800 through 1600 would represent
the following time periods:
Moore, et al. Expires: Jul 2000 + 6 months [Page 31]
Internet Draft Policy Core Information Model July 2000
Friday, March 5, 2000, from 0800 through 1600;
Friday, March 12, 2000, from 0800 through 1600;
Friday, March 19, 2000, from 0800 through 1600;
Friday, March 26, 2000, from 0800 through 1600;
Friday, April 2, 2000, from 0800 through 1600;
Friday, April 9, 2000, from 0800 through 1600;
Friday, April 16, 2000, from 0800 through 1600;
Friday, April 23, 2000, from 0800 through 1600;
Friday, April 30, 2000, from 0800 through 1600.
Properties not present in an instance of PolicyTimePeriodCondition are
implicitly treated as having their value "always enabled". Thus, in
the example above, the day-of-the-month mask is not present, and so
the validity period for the instance implicitly includes a day-of-the-
month mask that selects all days of the month. If we apply this
"missing property" rule to its fullest, we see that there is a second
way to indicate that a policy rule is always enabled: have it point to
an instance of PolicyTimePeriodCondition whose only properties are its
naming properties.
The property LocalOrUtcTime indicates whether the times represented in
the other five time-related properties of an instance of
PolicyTimePeriodCondition are to be interpreted as local times for the
location where a policy rule is being applied, or as UTC times.
The class definition is as follows.
NAME PolicyTimePeriodCondition
DESCRIPTION A class that provides the capability of enabling /
disabling a policy rule according to a pre-
determined schedule.
DERIVED FROM PolicyCondition
ABSTRACT FALSE
PROPERTIES TimePeriod
MonthOfYearMask
DayOfMonthMask
DayOfWeekMask
TimeOfDayMask
LocalOrUtcTime
6.5.1. The Property "TimePeriod"
This property identifies an overall range of calendar dates and times
over which a policy rule is valid. It reuses the format for an
explicit time period defined in RFC 2445 (reference [10]): a string
representing a starting date and time, in which the character 'T'
indicates the beginning of the time portion, followed by the solidus
character '/', followed by a similar string representing an end date
and time. The first date indicates the beginning of the range, while
the second date indicates the end. Thus, the second date and time
must be later than the first. Date/times are expressed as substrings
of the form "yyyymmddThhmmss". For example:
Moore, et al. Expires: Jul 2000 + 6 months [Page 32]
Internet Draft Policy Core Information Model July 2000
20000101T080000/20000131T120000
January 1, 2000, 0800 through January 31, 2000, noon
There are also two special cases in which one of the date/time strings
is replaced with a special string defined in RFC 2445.
o If the first date/time is replaced with the string "THISANDPRIOR",
then the property indicates that a policy rule is valid [from now]
until the date/time that appears after the '/'.
o If the second date/time is replaced with the string
"THISANDFUTURE", then the property indicates that a policy rule
becomes valid on the date/time that appears before the '/', and
remains valid from that point on.
Note that RFC 2445 does not use these two strings in connection with
explicit time periods. Thus the PCIM is combining two elements from
RFC 2445 that are not combined in the RFC itself.
The property definition is as follows:
NAME TimePeriod
DESCRIPTION The range of calendar dates on which a policy rule
is valid.
SYNTAX string
FORMAT yyyymmddThhmmss/yyyymmddThhmmss, where the first
date/time may be replaced with the string
"THISANDPRIOR" or the second date/time may be
replaced with the string "THISANDFUTURE"
6.5.2. The Property "MonthOfYearMask"
The purpose of this property is to refine the definition of the valid
time period that is defined by the TimePeriod property, by explicitly
specifying which months the policy is valid for. These properties
work together, with the TimePeriod used to specify the overall time
period that the policy is valid for, and the MonthOfYearMask used to
pick out which months of that time period the policy is valid for.
This property is formatted as an octet string, structured as follows:
o a 4-octet length field, indicating the length of the entire octet
string; this field is always set to 0x00000006 for this property;
o a 2-octet field consisting of 12 bits identifying the 12 months of
the year, beginning with January and ending with December, followed
by 4 bits that are always set to '0'. For each month, the value
'1' indicates that the policy is valid for that month, and the
value '0' indicates that it is not valid.
The value 0x000000060830, for example, indicates that a policy rule is
valid only in the months May, November, and December .
Moore, et al. Expires: Jul 2000 + 6 months [Page 33]
Internet Draft Policy Core Information Model July 2000
If this property is omitted, then the policy rule is treated as valid
for all twelve months. The property definition is as follows:
NAME MonthOfYearMask
DESCRIPTION A mask identifying the months of the year in which
a policy rule is valid.
SYNTAX octet string
FORMAT 0x00000006XXX0
6.5.3. The Property "DayOfMonthMask"
The purpose of this property is to refine the definition of the valid
time period that is defined by the TimePeriod property, by explicitly
specifying which days of the month the policy is valid for. These
properties work together, with the TimePeriod used to specify the
overall time period that the policy is valid for, and the
DayOfMonthMask used to pick out which days of the month in that time
period the policy is valid for.
This property is formatted as an octet string, structured as follows:
o a 4-octet length field, indicating the length of the entire octet
string; this field is always set to 0x0000000C for this property;
o an 8-octet field consisting of 31 bits identifying the days of the
month counting from the beginning, followed by 31 more bits
identifying the days of the month counting from the end, followed
by 2 bits that are always set to '0'. For each day, the value '1'
indicates that the policy is valid for that day, and the value '0'
indicates that it is not valid.
The value 0x0000000C8000000100000000, for example, indicates that a
policy rule is valid on the first and last days of the month.
For months with fewer than 31 days, the digits corresponding to days
that the months do not have (counting in both directions) are ignored.
The encoding of the 62 significant bits in the octet string matches
that used for the schedDay object in the DISMAN-SCHEDULE-MIB. See
reference [8] for more details on this object.
The property definition is as follows:
NAME DayOfMonthMask
DESCRIPTION A mask identifying the days of the month on which a
policy rule is valid.
SYNTAX octet string
FORMAT 0x0000000CXXXXXXXXXXXXXXXX
Moore, et al. Expires: Jul 2000 + 6 months [Page 34]
Internet Draft Policy Core Information Model July 2000
6.5.4. The Property "DayOfWeekMask"
The purpose of this property is to refine the definition of the valid
time period that is defined by the TimePeriod property by explicitly
specifying which days of the week the policy is valid for. These
properties work together, with the TimePeriod used to specify the
overall time period that the policy is valid for, and the
DayOfWeekMask used to pick out which days of the week in that time
period the policy is valid for.
This property is formatted as an octet string, structured as follows:
o a 4-octet length field, indicating the length of the entire octet
string; this field is always set to 0x00000005 for this property;
o a 1-octet field consisting of 7 bits identifying the 7 days of the
week, beginning with Sunday and ending with Saturday, followed by 1
bit that is always set to '0'. For each day of the week, the value
'1' indicates that the policy is valid for that day, and the value
'0' indicates that it is not valid.
The value 0x000000057C, for example, indicates that a policy rule is
valid Monday through Friday.
The property definition is as follows:
NAME DayOfWeekMask
DESCRIPTION A mask identifying the days of the week on which a
policy rule is valid.
SYNTAX octet string
FORMAT 0x00000005XX
6.5.5. The Property "TimeOfDayMask"
The purpose of this property is to refine the definition of the valid
time period that is defined by the TimePeriod property by explicitly
specifying a range of times in a day the policy is valid for. These
properties work together, with the TimePeriod used to specify the
overall time period that the policy is valid for, and the
TimeOfDayMask used to pick out which range of time periods in a given
day of that time period the policy is valid for.
This property is formatted in the style of RFC 2445 [10]: a time
string beginning with the character 'T', followed by the solidus
character '/', followed by a second time string. The first time
indicates the beginning of the range, while the second time indicates
the end. Times are expressed as substrings of the form "Thhmmss".
The second substring always identifies a later time than the first
substring. To allow for ranges that span midnight, however, the value
of the second string may be smaller than the value of the first
substring. Thus, "T080000/T210000" identifies the range from 0800
Moore, et al. Expires: Jul 2000 + 6 months [Page 35]
Internet Draft Policy Core Information Model July 2000
until 2100, while "T210000/T080000" identifies the range from 2100
until 0800 of the following day.
When a range spans midnight, it by definition includes parts of two
successive days. When one of these days is also selected by either
the MonthOfYearMask, DayOfMonthMask, and/or DayOfWeekMask, but the
other day is not, then the policy is active only during the portion of
the range that falls on the selected day. For example, if the range
extends from 2100 until 0800, and the day of week mask selects Monday
and Tuesday, then the policy is active during the following three
intervals:
From midnight Sunday until 0800 Monday;
From 2100 Monday until 0800 Tuesday;
From 2100 Tuesday until 23:59:59 Tuesday.
The property definition is as follows:
NAME TimeOfDayMask
DESCRIPTION The range of times at which a policy rule is valid.
If the second time is earlier than the first, then
the interval spans midnight.
SYNTAX string
FORMAT Thhmmss/Thhmmss
6.5.6. The Property "LocalOrUtcTime"
This property indicates whether the times represented in the
TimePeriod property and in the various Mask properties represent local
times or UTC times. There is no provision for mixing of local times
and UTC times: the value of this property applies to all of the other
time-related properties.
The property definition is as follows:
NAME LocalOrUtcTime
DESCRIPTION An indication of whether the other times in this
instance represent local times or UTC times.
SYNTAX uint16
VALUES localTime(1), utcTime(2)
DEFAULT VALUE utcTime(2)
6.6. The Class "VendorPolicyCondition"
The purpose of this class is to provide a general extension mechanism
for representing policy conditions that have not been modeled with
specific properties. Instead, the two properties Constraint and
ConstraintEncoding are used to define the content and format of the
condition, as explained below.
Moore, et al. Expires: Jul 2000 + 6 months [Page 36]
Internet Draft Policy Core Information Model July 2000
As its name suggests, this class is intended for vendor-specific
extensions to the Policy Core Information Model. Standardized
extensions are not expected to use this class.
The class definition is as follows:
NAME VendorPolicyCondition
DESCRIPTION A class that defines a registered means to describe
a policy condition.
DERIVED FROM PolicyCondition
ABSTRACT FALSE
PROPERTIES Constraint[ ]
ConstraintEncoding
6.6.1. The Multi-valued Property "Constraint"
This property provides a general extension mechanism for representing
policy conditions that have not been modeled with specific properties.
The format of the octet strings in the array is left unspecified in
this definition. It is determined by the OID value stored in the
property ConstraintEncoding. Since ConstraintEncoding is single-
valued, all the values of Constraint share the same format and
semantics.
NOTE: CIM has a particular way of representing an array of octet
strings. (A single octet string can be represented as an ordered
array of uint8's, but this does not work for multi-valued properties
where each value is an octet string.) There is a qualifier
"Octetstring" that can be applied to a multi-valued string property.
This qualifier functions exactly like an SMIv2 (SNMP) Textual
Convention, refining the syntax and semantics of the existing CIM data
types "string". Strings with this qualifier consist of a 4-octet
length field, followed by an even number of the characters A-F and 0-
9. The length is encoded as an 8-digit hexadecimal value, which
includes the 4 octets of the length field itself. For example, the
octet string 0x4a is encoded as 0x000000063441.
A policy decision point can readily determine whether it supports the
values stored in an instance of Constraint by checking the OID value
from ConstraintEncoding against the set of OIDs it recognizes. The
action for the policy decision point to take in case it does not
recognize the format of this data could itself be modeled as a policy
rule, governing the behavior of the policy decision point.
The property is defined as follows:
NAME Constraint
DESCRIPTION Extension mechanism for representing constraints
that have not been modeled as specific properties.
The format of the values is identified by the OID
stored in the property ConstraintEncoding.
SYNTAX string
QUALIFIER Octetstring
Moore, et al. Expires: Jul 2000 + 6 months [Page 37]
Internet Draft Policy Core Information Model July 2000
6.6.2. The Property "ConstraintEncoding"
This property identifies the encoding and semantics of the Constraint
property values in this instance. The value of this property is a
single string, representing a single OID.
The property is defined as follows:
NAME ConstraintEncoding
DESCRIPTION An OID encoded as a string, identifying the format
and semantics for this instance's Constraint
property. The value is a dotted sequence of
decimal digits (for example, "1.2.100.200")
representing the arcs of the OID. The characters
in the string are the UCS-2 characters
corresponding to the US ASCII encodings of the
numeric characters and the period.
SYNTAX string
6.7. The Class "PolicyAction"
The purpose of a policy action is to execute one or more operations
that will affect network traffic and/or systems, devices, etc. in
order to achieve a desired state. This (new) state provides one or
more (new) behaviors. A policy action ordinarily changes the
configuration of one or more elements.
A PolicyRule contains one or more policy actions. A policy
administrator can assign an order to the actions associated with a
PolicyRule, complete with an indication of whether the indicated order
is mandatory, recommended, or of no significance. Ordering of the
actions associated with a PolicyRule is accomplished via a property in
the PolicyActionInPolicyRule aggregation.
The actions associated with a PolicyRule are executed if and only if
the overall condition(s) of the PolicyRule evaluates to TRUE.
The class definition of PolicyAction is as follows:
NAME PolicyAction
DESCRIPTION A class representing a rule-specific or reusable
policy action to be performed if the condition for
a policy rule evaluates to TRUE.
DERIVED FROM Policy
ABSTRACT FALSE
PROPERTIES NONE
No properties are defined for this class since it inherits all its
properties from Policy. The class exists as an abstract superclass
for domain-specific policy actions, defined in subclasses. In an
Moore, et al. Expires: Jul 2000 + 6 months [Page 38]
Internet Draft Policy Core Information Model July 2000
implementation, various key/identification properties MUST be defined
for the class or its instantiable subclasses. The keys for a native
CIM implementation are defined in Appendix A, Section 14.3. Keys for
an LDAP implementation are defined in the LDAP mapping of this
information model [11].
When identifying and using the PolicyAction class, it is necessary to
remember that an action can be rule-specific or reusable. This was
discussed above in Section 5.1. The distinction between the two types
of policy actions lies in the associations in which an instance can
participate, and in how the different instances are named.
Conceptually, a reusable policy action resides in a policy repository,
and is named within the scope of that repository. On the other hand,
a rule-specific policy action is named within the scope of the single
policy rule to which it is related.
The distinction between rule-specific and reusable PolicyActions
affects the CIM naming, defined in Appendix A, and the LDAP mapping
[11].
6.8. The Class "VendorPolicyAction"
The purpose of this class is to provide a general extension mechanism
for representing policy actions that have not been modeled with
specific properties. Instead, the two properties ActionData and
ActionEncoding are used to define the content and format of the
action, as explained below.
As its name suggests, this class is intended for vendor-specific
extensions to the Policy Core Information Model. Standardized
extensions are not expected to use this class.
The class definition is as follows:
NAME VendorPolicyAction
DESCRIPTION A class that defines a registered means to describe
a policy action.
DERIVED FROM PolicyAction
ABSTRACT FALSE
PROPERTIES ActionData[ ]
ActionEncoding
6.8.1. The Multi-valued Property "ActionData"
This property provides a general extension mechanism for representing
policy actions that have not been modeled with specific properties.
The format of the octet strings in the array is left unspecified in
this definition. It is determined by the OID value stored in the
property ActionEncoding. Since ActionEncoding is single-valued, all
the values of ActionData share the same format and semantics. See
Section 6.6.1 for a discussion of how CIM encodes an array of octet
strings.
Moore, et al. Expires: Jul 2000 + 6 months [Page 39]
Internet Draft Policy Core Information Model July 2000
A policy decision point can readily determine whether it supports the
values stored in an instance of ActionData by checking the OID value
from ActionEncoding against the set of OIDs it recognizes. The action
for the policy decision point to take in case it does not recognize
the format of this data could itself be modeled as a policy rule,
governing the behavior of the policy decision point.
The property is defined as follows:
NAME ActionData
DESCRIPTION Extension mechanism for representing actions that
have not been modeled as specific properties. The
format of the values is identified by the OID
stored in the property ActionEncoding.
SYNTAX string
QUALIFIER Octetstring
6.8.2. The Property "ActionEncoding"
This property identifies the encoding and semantics of the ActionData
property values in this instance. The value of this property is a
single string, representing a single OID.
The property is defined as follows:
NAME ActionEncoding
DESCRIPTION An OID encoded as a string, identifying the format
and semantics for this instance's ActionData
property. The value is a dotted sequence of
decimal digits (for example, "1.2.100.200")
representing the arcs of the OID. The characters
in the string are the UCS-2 characters
corresponding to the US ASCII encodings of the
numeric characters and the period.
SYNTAX string
6.9. The Class "PolicyRepository"
The class definition of PolicyRepository is as follows:
NAME PolicyRepository
DESCRIPTION A class representing an administratively defined
container for reusable policy-related information.
This class does not introduce any additional
properties beyond those in its superclass
AdminDomain. It does, however, participate in a
number of unique associations.
DERIVED FROM AdminDomain
ABSTRACT FALSE
Moore, et al. Expires: Jul 2000 + 6 months [Page 40]
Internet Draft Policy Core Information Model July 2000
7. Association and Aggregation Definitions
The first two subsections of this section introduce associations and
aggregations as they are used in CIM. The remaining subsections
present the class definitions for the associations and aggregations
that are part of the Policy Core Information Model.
7.1. Associations
An association is a CIM construct representing a relationship between
two (or theoretically more) objects. It is modeled as a class
containing typically two object references. Associations can be
defined between classes without affecting any of the related classes.
That is, addition of an association does not affect the interface of
the related classes.
7.2. Aggregations
An aggregation is a strong form of an association, which usually
represents a "whole-part" or a "collection" relationship. For
example, CIM uses an aggregation to represent the containment
relationship between a system and the components that make up the
system. Aggregation as a "whole-part" relationship often implies, but
does not require, that the aggregated objects have mutual
dependencies.
7.3. The Abstract Aggregation "PolicyComponent
This abstract aggregation defines two object references that will be
overridden in each of five subclasses, to become references to the
concrete policy classes PolicyGroup, PolicyRule, PolicyCondition,
PolicyAction, and PolicyTimePeriodCondition. The value of the
abstract superclass is to convey that all five subclasses have the
same "whole-part" semantics, and for ease of query to locate all
"components" of a PolicyGroup or PolicyRule.
The class definition for the aggregation is as follows:
NAME PolicyComponent
DESCRIPTION A generic aggregation used to establish 'part of'
relationships between the subclasses of Policy. For
example, the PolicyConditionInPolicyRule
aggregation defines that PolicyConditions are part
of a PolicyRule.
ABSTRACT TRUE
PROPERTIES GroupComponent[ref Policy[0..n]]
PartComponent[ref Policy[0..n]]
7.4. The Aggregation "PolicyGroupInPolicyGroup"
The PolicyGroupInPolicyGroup aggregation enables policy groups to be
nested. This is critical for scalability and manageability, as it
enables complex policies to be constructed from multiple simpler
Moore, et al. Expires: Jul 2000 + 6 months [Page 41]
Internet Draft Policy Core Information Model July 2000
policies for administrative convenience. For example, a policy group
representing policies for the US might have nested within it policy
groups for the Eastern and Western US.
A PolicyGroup may aggregate other PolicyGroups via this aggregation,
or it may aggregate PolicyRules via the PolicyRuleInPolicyGroup
aggregation. But a single PolicyGroup SHALL NOT do both.
The class definition for the aggregation is as follows:
NAME PolicyGroupInPolicyGroup
DESCRIPTION A class representing the aggregation of
PolicyGroups by a higher-level PolicyGroup.
DERIVED FROM PolicyComponent
ABSTRACT FALSE
PROPERTIES GroupComponent[ref PolicyGroup[0..n]]
PartComponent[ref PolicyGroup[0..n]]
7.4.1. The Reference "GroupComponent"
This property is inherited from PolicyComponent, and overridden to
become an object reference to a PolicyGroup that contains one or more
other PolicyGroups. Note that for any single instance of the
aggregation class PolicyGroupInPolicyGroup, this property (like all
Reference properties) is single-valued. The [0..n] cardinality
indicates that there may be 0, 1, or more than one PolicyGroups that
contain any given PolicyGroup.
7.4.2. The Reference "PartComponent"
This property is inherited from PolicyComponent, and overridden to
become an object reference to a PolicyGroup contained by one or more
other PolicyGroups. Note that for any single instance of the
aggregation class PolicyGroupInPolicyGroup, this property (like all
Reference properties) is single-valued. The [0..n] cardinality
indicates that a given PolicyGroup may contain 0, 1, or more than one
other PolicyGroups.
7.5. The Aggregation "PolicyRuleInPolicyGroup"
A policy group may aggregate one or more policy rules, via the
PolicyRuleInPolicyGroup aggregation. Grouping of policy rules into a
policy group is again for administrative convenience; a policy rule
may also be used by itself, without belonging to a policy group.
A PolicyGroup may aggregate PolicyRules via this aggregation, or it
may aggregate other PolicyGroups via the PolicyGroupInPolicyGroup
aggregation. But a single PolicyGroup SHALL NOT do both.
The class definition for the aggregation is as follows:
NAME PolicyRuleInPolicyGroup
Moore, et al. Expires: Jul 2000 + 6 months [Page 42]
Internet Draft Policy Core Information Model July 2000
DESCRIPTION A class representing the aggregation of PolicyRules
by a PolicyGroup.
DERIVED FROM PolicyComponent
ABSTRACT FALSE
PROPERTIES GroupComponent[ref PolicyGroup[0..n]]
PartComponent[ref PolicyRule[0..n]]
7.5.1. The Reference "GroupComponent"
This property is inherited from PolicyComponent, and overridden to
become an object reference to a PolicyGroup that contains one or more
PolicyRules. Note that for any single instance of the aggregation
class PolicyRuleInPolicyGroup, this property (like all Reference
properties) is single-valued. The [0..n] cardinality indicates that
there may be 0, 1, or more than one PolicyGroups that contain any
given PolicyRule.
7.5.2. The Reference "PartComponent"
This property is inherited from PolicyComponent, and overridden to
become an object reference to a PolicyRule contained by one or more
PolicyGroups. Note that for any single instance of the aggregation
class PolicyRuleInPolicyGroup, this property (like all Reference
properties) is single-valued. The [0..n] cardinality indicates that a
given PolicyGroup may contain 0, 1, or more than one PolicyRules.
7.6. The Aggregation "PolicyConditionInPolicyRule"
A policy rule aggregates zero or more instances of the PolicyCondition
class, via the PolicyConditionInPolicyRule association. A policy rule
that aggregates zero policy conditions is not a valid rule -- it may,
for example, be in the process of being entered into the policy
repository. A policy rule has no effect until it is valid. The
conditions aggregated by a policy rule are grouped into two levels of
lists: either an ORed set of ANDed sets of conditions (DNF, the
default) or an ANDed set of ORed sets of conditions (CNF). Individual
conditions in these lists may be negated. The property
ConditionListType (in PolicyRule) specifies which of these two
grouping schemes applies to a particular PolicyRule. The conditions
are used to determine whether to perform the actions associated with
the PolicyRule.
One or more policy time periods may be among the conditions associated
with a policy rule via the PolicyConditionInPolicyRule association.
In this case, the time periods are simply additional conditions to be
evaluated along with any other conditions specified for the rule.
The class definition for the aggregation is as follows:
NAME PolicyConditionInPolicyRule
DESCRIPTION A class representing the aggregation of
PolicyConditions by a PolicyRule.
DERIVED FROM PolicyComponent
Moore, et al. Expires: Jul 2000 + 6 months [Page 43]
Internet Draft Policy Core Information Model July 2000
ABSTRACT FALSE
PROPERTIES GroupComponent[ref PolicyRule[0..n]]
PartComponent[ref PolicyCondition[0..n]]
GroupNumber
ConditionNegated
7.6.1. The Reference "GroupComponent"
This property is inherited from PolicyComponent, and overridden to
become an object reference to a PolicyRule that contains one or more
PolicyConditions. Note that for any single instance of the
aggregation class PolicyConditionInPolicyRule, this property (like all
Reference properties) is single-valued. The [0..n] cardinality
indicates that there may be 0, 1, or more than one PolicyRules that
contain any given PolicyCondition.
7.6.2. The Reference "PartComponent"
This property is inherited from PolicyComponent, and overridden to
become an object reference to a PolicyCondition contained by one or
more PolicyRules. Note that for any single instance of the
aggregation class PolicyConditionInPolicyRule, this property (like all
Reference properties) is single-valued. The [0..n] cardinality
indicates that a given PolicyRule may contain 0, 1, or more than one
PolicyConditions.
7.6.3. The Property "GroupNumber"
This property contains an integer identifying the group to which the
condition referenced by the PartComponent property is assigned in
forming the overall conditional expression for the policy rule
identified by the GroupComponent reference.
The property is defined as follows:
NAME GroupNumber
DESCRIPTION Unsigned integer indicating the group to which the
condition identified by the PartComponent property
is to be assigned.
SYNTAX uint16
7.6.4. The Property "ConditionNegated"
This property is a boolean, indicating whether the condition
referenced by the PartComponent property is negated in forming the
overall conditional expression for the policy rule identified by the
GroupComponent reference.
The property is defined as follows:
NAME ConditionNegated
Moore, et al. Expires: Jul 2000 + 6 months [Page 44]
Internet Draft Policy Core Information Model July 2000
DESCRIPTION Indication of whether the condition identified by
the PartComponent property is negated. (TRUE
indicates that the condition is negated, FALSE
indicates that it is not negated.)
SYNTAX boolean
7.7. The Aggregation "PolicyRuleValidityPeriod"
A different relationship between a policy rule and a policy time
period (than PolicyConditionInPolicyRule) is represented by the
PolicyRuleValidityPeriod aggregation. The latter describes scheduled
activation and deactivation of the policy rule.
If a policy rule is associated with multiple policy time periods via
this association, then the rule is active if at least one of the time
periods indicates that it is active. (In other words, the time
periods are ORed to determine whether the rule is active.) A policy
time period may be aggregated by multiple policy rules. A rule that
does not point to a policy time period via this aggregation is, from
the point of view of scheduling, always active. It may, however, be
inactive for other reasons.
Time periods are a general concept that can be used in other
applications. However, they are mentioned explicitly here in this
specification since they are frequently used in policy applications.
The class definition for the aggregation is as follows:
NAME PolicyRuleValidityPeriod
DESCRIPTION A class representing the aggregation of
PolicyTimePeriodConditions by a PolicyRule.
DERIVED FROM PolicyComponent
ABSTRACT FALSE
PROPERTIES GroupComponent[ref PolicyRule[0..n]]
PartComponent[ref PolicyTimePeriodCondition[0..n]]
7.7.1. The Reference "GroupComponent"
This property is inherited from PolicyComponent, and overridden to
become an object reference to a PolicyRule that contains one or more
PolicyTimePeriodConditions. Note that for any single instance of the
aggregation class PolicyRuleValidityPeriod, this property (like all
Reference properties) is single-valued. The [0..n] cardinality
indicates that there may be 0, 1, or more than one PolicyRules that
contain any given PolicyTimePeriodCondition.
7.7.2. The Reference "PartComponent"
This property is inherited from PolicyComponent, and overridden to
become an object reference to a PolicyTimePeriodCondition contained by
one or more PolicyRules. Note that for any single instance of the
aggregation class PolicyRuleValidityPeriod, this property (like all
Moore, et al. Expires: Jul 2000 + 6 months [Page 45]
Internet Draft Policy Core Information Model July 2000
Reference properties) is single-valued. The [0..n] cardinality
indicates that a given PolicyRule may contain 0, 1, or more than one
PolicyTimePeriodConditions.
7.8. The Aggregation "PolicyActionInPolicyRule"
A policy rule may aggregate zero or more policy actions. A policy
rule that aggregates zero policy actions is not a valid rule -- it
may, for example, be in the process of being entered into the policy
repository. A policy rule has no effect until it is valid. The
actions associated with a PolicyRule may be given a required order, a
recommended order, or no order at all. For actions represented as
separate objects, the PolicyActionInPolicyRule aggregation can be used
to express an order.
This aggregation does not indicate whether a specified action order is
required, recommended, or of no significance; the property
SequencedActions in the aggregating instance of PolicyRule provides
this indication.
The class definition for the aggregation is as follows:
NAME PolicyActionInPolicyRule
DESCRIPTION A class representing the aggregation of
PolicyActions by a PolicyCondition.
DERIVED FROM PolicyComponent
ABSTRACT FALSE
PROPERTIES GroupComponent[ref PolicyRule[0..n]]
PartComponent[ref PolicyAction[0..n]]
ActionOrder
7.8.1. The Reference "GroupComponent"
This property is inherited from PolicyComponent, and overridden to
become an object reference to a PolicyRule that contains one or more
PolicyActions. Note that for any single instance of the aggregation
class PolicyActionInPolicyRule, this property (like all Reference
properties) is single-valued. The [0..n] cardinality indicates that
there may be 0, 1, or more than one PolicyRules that contain any given
PolicyAction.
7.8.2. The Reference "PartComponent"
This property is inherited from PolicyComponent, and overridden to
become an object reference to a PolicyAction contained by one or more
PolicyRules. Note that for any single instance of the aggregation
class PolicyActionInPolicyRule, this property (like all Reference
properties) is single-valued. The [0..n] cardinality indicates that a
given PolicyRule may contain 0, 1, or more than one PolicyActions.
Moore, et al. Expires: Jul 2000 + 6 months [Page 46]
Internet Draft Policy Core Information Model July 2000
7.8.3. The Property "ActionOrder"
This property provides an unsigned integer 'n' that indicates the
relative position of an action in the sequence of actions associated
with a policy rule. When 'n' is a positive integer, it indicates a
place in the sequence of actions to be performed, with smaller
integers indicating earlier positions in the sequence. The special
value '0' indicates "don't care". If two or more actions have the
same non-zero sequence number, they may be performed in any order, but
they must all be performed at the appropriate place in the overall
action sequence.
A series of examples will make ordering of actions clearer:
o If all actions have the same sequence number, regardless of whether
it is '0' or non-zero, any order is acceptable.
o The values
1:ACTION A
2:ACTION B
1:ACTION C
3:ACTION D
indicate two acceptable orders: A,C,B,D or C,A,B,D, since A and C
can be performed in either order, but only at the '1' position.
o The values
0:ACTION A
2:ACTION B
3:ACTION C
3:ACTION D
require that B,C, and D occur either as B,C,D or as B,D,C. Action
A may appear at any point relative to B,C, and D. Thus the
complete set of acceptable orders is: A,B,C,D; B,A,C,D; B,C,A,D;
B,C,D,A; A,B,D,C; B,A,D,C; B,D,A,C; B,D,C,A.
Note that the non-zero sequence numbers need not start with '1', and
they need not be consecutive. All that matters is their relative
magnitude.
The property is defined as follows:
NAME ActionOrder
DESCRIPTION Unsigned integer indicating the relative position
of an action in the sequence of actions aggregated
by a policy rule.
SYNTAX uint16
Moore, et al. Expires: Jul 2000 + 6 months [Page 47]
Internet Draft Policy Core Information Model July 2000
7.9. The Abstract Association "PolicyInSystem"
This abstract association inherits two object references from a
higher-level CIM association class, Dependency. It overrides these
object references to make them references to instances of the classes
System and Policy. Subclasses of PolicyInSystem then override these
object references again, to make them references to concrete policy
classes.
The value of the abstract superclass is to convey that all subclasses
have the same "dependency" semantics, and for ease of query to locate
all policy "dependencies" on a System. These dependencies are related
to scoping or hosting of the Policy.
The class definition for the association is as follows:
NAME PolicyInSystem
DESCRIPTION A generic association used to establish dependency
relationships between Policies and the Systems that
host them.
DERIVED FROM Dependency
ABSTRACT TRUE
PROPERTIES Antecedent[ref System[0..1]]
Dependent[ref Policy[0..n]]
7.10. The Weak Association "PolicyGroupInSystem"
This association links a PolicyGroup to the System in whose scope the
PolicyGroup is defined.
The class definition for the association is as follows:
NAME PolicyGroupInSystem
DESCRIPTION A class representing the fact that a PolicyGroup is
defined within the scope of a System.
DERIVED FROM PolicyInSystem
ABSTRACT FALSE
PROPERTIES Antecedent[ref System[1..1]]
Dependent[ref PolicyGroup[weak]]
7.10.1. The Reference "Antecedent"
This property is inherited from PolicyInSystem, and overridden to
restrict its cardinality to [1..1]. It serves as an object reference
to a System that provides a scope for one or more PolicyGroups. Since
this is a weak association, the cardinality for this object reference
is always 1, that is, a PolicyGroup is always defined within the scope
of exactly one System.
Moore, et al. Expires: Jul 2000 + 6 months [Page 48]
Internet Draft Policy Core Information Model July 2000
7.10.2. The Reference "Dependent"
This property is inherited from PolicyInSystem, and overridden to
become an object reference to a PolicyGroup defined within the scope
of a System. Note that for any single instance of the association
class PolicyGroupInSystem, this property (like all Reference
properties) is single-valued. The [0..n] cardinality indicates that a
given System may have 0, 1, or more than one PolicyGroups defined
within its scope.
7.11. The Weak Association "PolicyRuleInSystem"
Regardless of whether it belongs to a PolicyGroup (or to multiple
PolicyGroups), a PolicyRule is itself defined within the scope of a
System. This association links a PolicyRule to the System in whose
scope the PolicyRule is defined.
The class definition for the association is as follows:
NAME PolicyRuleInSystem
DESCRIPTION A class representing the fact that a PolicyRule is
defined within the scope of a System.
DERIVED FROM PolicyInSystem
ABSTRACT FALSE
PROPERTIES Antecedent[ref System[1..1]]
Dependent[ref PolicyRule[weak]]
7.11.1. The Reference "Antecedent"
This property is inherited from PolicyInSystem, and overridden to
restrict its cardinality to [1..1]. It serves as an object reference
to a System that provides a scope for one or more PolicyRules. Since
this is a weak association, the cardinality for this object reference
is always 1, that is, a PolicyRule is always defined within the scope
of exactly one System.
7.11.2. The Reference "Dependent"
This property is inherited from PolicyInSystem, and overridden to
become an object reference to a PolicyRule defined within the scope of
a System. Note that for any single instance of the association class
PolicyRuleInSystem, this property (like all Reference properties) is
single-valued. The [0..n] cardinality indicates that a given System
may have 0, 1, or more than one PolicyRules defined within its scope.
7.12. The Association "PolicyConditionInPolicyRepository"
A reusable policy condition is always related to a single
PolicyRepository, via the PolicyConditionInPolicyRepository
association. This is not true for all PolicyConditions, however. An
instance of PolicyCondition that represents a rule-specific condition
is not related to any policy repository via this association.
Moore, et al. Expires: Jul 2000 + 6 months [Page 49]
Internet Draft Policy Core Information Model July 2000
The class definition for the association is as follows:
NAME PolicyConditionInPolicyRepository
DESCRIPTION A class representing the inclusion of a reusable
PolicyCondition in a PolicyRepository.
DERIVED FROM PolicyInSystem
ABSTRACT FALSE
PROPERTIES Antecedent[ref PolicyRepository[0..1]]
Dependent[ref PolicyCondition[0..n]]
7.12.1. The Reference "Antecedent"
This property is inherited from PolicyInSystem, and overridden to
become an object reference to a PolicyRepository containing one or
more PolicyConditions. A reusable PolicyCondition is always related
to exactly one PolicyRepository via the
PolicyConditionInPolicyRepository association. The [0..1] cardinality
for this property covers the two types of PolicyConditions: 0 for a
rule-specific PolicyCondition, 1 for a reusable one.
7.12.2. The Reference "Dependent"
This property is inherited from PolicyInSystem, and overridden to
become an object reference to a PolicyCondition included in a
PolicyRepository. Note that for any single instance of the
association class PolicyConditionInPolicyRepository, this property
(like all Reference properties) is single-valued. The [0..n]
cardinality indicates that a given PolicyRepository may contain 0, 1,
or more than one PolicyConditions.
7.13. The Association "PolicyActionInPolicyRepository"
A reusable policy action is always related to a single
PolicyRepository, via the PolicyActionInPolicyRepository association.
This is not true for all PolicyActions, however. An instance of
PolicyAction that represents a rule-specific action is not related to
any policy repository via this association.
The class definition for the association is as follows:
NAME PolicyActionInPolicyRepository
DESCRIPTION A class representing the inclusion of a reusable
PolicyAction in a PolicyRepository.
DERIVED FROM PolicyInSystem
ABSTRACT FALSE
PROPERTIES Antecedent[ref PolicyRepository[0..1]]
Dependent[ref PolicyAction[0..n]]
7.13.1. The Reference "Antecedent"
This property is inherited from PolicyInSystem, and overridden to
become an object reference to a PolicyRepository containing one or
more PolicyActions. A reusable PolicyAction is always related to
Moore, et al. Expires: Jul 2000 + 6 months [Page 50]
Internet Draft Policy Core Information Model July 2000
exactly one PolicyRepository via the PolicyActionInPolicyRepository
association. The [0..1] cardinality for this property covers the two
types of PolicyActions: 0 for a rule-specific PolicyAction, 1 for a
reusable one.
7.13.2. The Reference "Dependent"
This property is inherited from PolicyInSystem, and overridden to
become an object reference to a PolicyAction included in a
PolicyRepository. Note that for any single instance of the
association class PolicyActionInPolicyRepository, this property (like
all Reference properties) is single-valued. The [0..n] cardinality
indicates that a given PolicyRepository may contain 0, 1, or more than
one PolicyActions.
7.14. The Aggregation "PolicyRepositoryInPolicyRepository"
The PolicyRepositoryInPolicyRepository aggregation enables policy
repositories to be nested. This derives from the higher level CIM
association, CIM_SystemComponent, describing that Systems contain
other ManagedSystemElements. This superclass could not be used for
the other Policy aggregations, since Policies are not
ManagedSystemElements, but ManagedElements.
The class definition for the aggregation is as follows:
NAME PolicyRepositoryInPolicyRepository
DESCRIPTION A class representing the aggregation of
PolicyRepositories by a higher-level
PolicyRepository.
DERIVED FROM SystemComponent
ABSTRACT FALSE
PROPERTIES GroupComponent[ref PolicyRepository[0..n]]
PartComponent[ref PolicyRepository[0..n]]
7.14.1. The Reference "GroupComponent"
This property is inherited from the CIM class SystemComponent, and
overridden to become an object reference to a PolicyRepository that
contains one or more other PolicyRepositories. Note that for any
single instance of the aggregation class
PolicyRepositoryInPolicyRepository, this property (like all Reference
properties) is single-valued. The [0..n] cardinality indicates that
there may be 0, 1, or more than one PolicyRepositories that contain
any given PolicyRepository.
7.14.2. The Reference "PartComponent"
This property is inherited from the CIM class SystemComponent, and
overridden to become an object reference to a PolicyRepository
contained by one or more other PolicyRepositories. Note that for any
single instance of the aggregation class
PolicyRepositoryInPolicyRepository, this property (like all Reference
Moore, et al. Expires: Jul 2000 + 6 months [Page 51]
Internet Draft Policy Core Information Model July 2000
properties) is single-valued. The [0..n] cardinality indicates that a
given PolicyRepository may contain 0, 1, or more than one other
PolicyRepositories.
8. Intellectual Property
The IETF takes no position regarding the validity or scope of any
intellectual property or other rights that might be claimed to pertain
to the implementation or use of the technology described in this
document or the extent to which any license under such rights might or
might not be available; neither does it represent that it has made any
effort to identify any such rights. Information on the IETF's
procedures with respect to rights in standards-track and standards-
related documentation can be found in BCP-11.
Copies of claims of rights made available for publication and any
assurances of licenses to be made available, or the result of an
attempt made to obtain a general license or permission for the use of
such proprietary rights by implementers or users of this specification
can be obtained from the IETF Secretariat.
The IETF invites any interested party to bring to its attention any
copyrights, patents or patent applications, or other proprietary
rights which may cover technology that may be required to practice
this standard. Please address the information to the IETF Executive
Director.
9. Acknowledgements
The Policy Core Information Model in this document is closely based on
the work of the DMTF's Service Level Agreements working group, so
thanks are due to the members of that working group. Several of the
policy classes in this model first appeared in early drafts on IPSec
policy and QoS policy. The authors of these drafts were Partha
Bhattacharya, Rob Adams, William Dixon, Roy Pereira, Raju Rajan, Jean-
Christophe Martin, Sanjay Kamat, Michael See, Rajiv Chaudhury, Dinesh
Verma, George Powers, and Raj Yavatkar.
10. Security Considerations
o This document itself does not introduce any new security issues for
the Internet, similar to the circumstances which existed when SMI
was introduced. However, it is necessary to document the
requirements for a secure policy system, in order to show that the
overall policy framework is viable. Our model for documenting
these requirements is based on prior work in the IETF on DNSSEC and
SNMPv3. One of our objectives in the policy work in the IETF is to
not break the known existing security mechanisms, or to make them
less effective, regardless of whether or not these security
mechanisms affect what flows on the wire.
Moore, et al. Expires: Jul 2000 + 6 months [Page 52]
Internet Draft Policy Core Information Model July 2000
o Users: The first step in identifying security requirements for
policy, is to identify the users of policy. The users fall into
three categories:
o Administrators of Schema: This group requires the most stringent
authorization and associated security controls. An improper or
mal-formed change in the design of the policy schema carries with
it the danger of rendering the repository inoperable while the
repository is being repaired or re-built. During this time, the
policy enforcement entities would need to continue to enforce
policies according to their prior configuration. The good news is
that it is expected that large network operators will change schema
design infrequently, and, when they do, the schema creation changes
will be tested on an off-line copy of the directory before the
operational directory is updated. Typically, a small group of
schema administrators will be authorized to make these changes in a
service provider or enterprise environment. The ability to
maintain an audit trail is also required here.
o Administrators of Schema Content: This group requires authorization
to load values (entries) into a policy repository) schema
(read/write access). An audit trail mechanism is also required
here. The effect of entering improperly formatted or maliciously-
intended data into a policy repository, could potentially result in
re-configuring mass numbers of network elements in a way that
renders them to be inoperable, or of rendering network resources
inaccessible for an extended period of time.
o Applications and PDPs: These entities must be authorized for read-
only access to the policy repository, so that they may acquire
policy for the purposes of passing it to their respective
enforcement entities.
o Security Disciplines:
o Audit Trail (Non-repudiation): We document the need for this
function in systems which maintain and distribute policy. The
dependency for support of this function is on the implementers
of these systems, and not on any specific standards for
implementation. The requirement for a policy system is that a
minimum level of auditing via an auditing facility must be
provided. Logging should be enabled. This working group will
not specify what this minimal auditing function consists of.
o Access Control/Authorization: Access Control List (ACL)
functionality must be provided. The two administrative sets of
users documented above will form the basis for two
administrative use cases which require support.
o Authentication: Authentication support on the order of that
available with TLS and Kerberos are acceptable for
authentication. We advise against using weaker mechanisms,
Moore, et al. Expires: Jul 2000 + 6 months [Page 53]
Internet Draft Policy Core Information Model July 2000
such as clear text and HTTP Digest. Mutual authentication is
recommended.
o Integrity/Privacy: Integrity/privacy support on the order of
TLS or IPSec is acceptable for encryption and data integrity
on the wire. If physical or virtual access to the policy
repository is in question, it may also be necessary to encrypt
the policy data as it is stored on the file system; however,
specification of mechanisms for this purpose are outside the
scope of this working group. In any case, we recommend that
the physical server be located in a physically secure
environment.
In the case of PDP-to-PEP communications, the use of IPSec is
recommended for providing confidentiality, data origin
authentication, integrity and replay prevention. See reference
[9].
o Denial of Service: We recommend the use of multiple policy
repositories, such that a denial of service attack on any one
repository will not make all policy data inaccessible to legitimate
users. However, this still leaves a denial of service attack
exposure. Our belief is that the use of a policy schema, in a
centrally administered but physically distributed policy
repository, does not increase the risk of denial of service
attacks; however, such attacks are still possible. If executed
successfully, such an attack could prevent PDPs from accessing a
policy repository, and thus prevent them from acquiring new policy.
In such a case, the PDPs, and associated PEPs would continue
operating under the policies in force before the denial of service
attack was launched. Note that exposure of policy systems to
denial of service attacks is not any greater than the exposure of
DNS with DNSSEC in place.
11. References
[1] Distributed Management Task Force, Inc., "DMTF Technologies: CIM
Standards", available via links on the following DMTF web page:
http://www.dmtf.org/spec/cims.html.
[2] Distributed Management Task Force, Inc., "Common Information Model
(CIM) Specification, version 2.2, June 14, 1999. This document is
available on the following DMTF web page:
http://www.dmtf.org/spec/cims.html.
[3] Bradner, S., "Key words for use in RFCs to Indicate Requirement
Levels", BCP 14, RFC 2119, March 1997.
[4] Hovey, R., and S. Bradner, "The Organizations Involved in the IETF
Standards Process", BCP 11, RFC 2028, October 1996.
Moore, et al. Expires: Jul 2000 + 6 months [Page 54]
Internet Draft Policy Core Information Model July 2000
[5] J. Strassner and S. Judd, "Directory-Enabled Networks", version
3.0c5 (August 1998). A PDF file is available at
http://www.murchiso.com/den/#denspec.
[6] J. Strassner, policy architecture BOF presentation, 42nd IETF
Meeting, Chicago, Illinois, October, 1998. Minutes of this BOF are
available at the following location:
http://www.ietf.org/proceedings/98aug/index.html.
[7] Yergeau, F., "UTF-8, a transformation format of ISO 10646", RFC
2279, January 1998.
[8] Levi, D., and J. Schoenwaelder, "Definitions of Managed Objects for
Scheduling Management Operations", RFC 2591, May 1999.
[9] R. Yavatkar and D. Pendarakis, R. Guerin, "A Framework for Policy-
based Admission Control", RFC 2753, January 2000.
[10] Dawson, F., and D. Stenerson, "Internet Calendaring and Scheduling
Core Object Specification (iCalendar)", RFC 2445, November 1998.
[11] Strassner, J., and E. Ellesson, B. Moore, R. Moats, "Policy Core
LDAP Schema", draft-ietf-policy-core-schema-07.txt, July 2000.
12. Authors' Addresses
Ed Ellesson
Tivoli Systems
Building 10, Office R2D39
3901 Miami Blvd.
Durham, NC 27703
Phone: +1 919-224-2111
Fax: +1 919-224-2540
E-mail: ed_ellesson@tivoli.com
Bob Moore
IBM Corporation, BRQA/502
4205 S. Miami Blvd.
Research Triangle Park, NC 27709
Phone: +1 919-254-4436
Fax: +1 919-254-6243
E-mail: remoore@us.ibm.com
John Strassner
Cisco Systems, Bldg 15
170 West Tasman Drive
San Jose, CA 95134
Phone: +1 408-527-1069
Fax: +1 408-527-6351
E-mail: johns@cisco.com
Andrea Westerinen
Moore, et al. Expires: Jul 2000 + 6 months [Page 55]
Internet Draft Policy Core Information Model July 2000
Cisco Systems
170 West Tasman Drive
San Jose, CA 95134
Phone: +1 408-853-8294
Fax: +1 408-527-6351
E-mail: andreaw@cisco.com
13. Full Copyright Statement
Copyright (C) The Internet Society (2000). All Rights Reserved.
This document and translations of it may be copied and furnished to
others, and derivative works that comment on or otherwise explain it
or assist in its implementation may be prepared, copied, published and
distributed, in whole or in part, without restriction of any kind,
provided that the above copyright notice and this paragraph are
included on all such copies and derivative works. However, this
document itself may not be modified in any way, such as by removing
the copyright notice or references to the Internet Society or other
Internet organizations, except as needed for the purpose of developing
Internet standards in which case the procedures for copyrights defined
in the Internet Standards process must be followed, or as required to
translate it into languages other than English.
The limited permissions granted above are perpetual and will not be
revoked by the Internet Society or its successors or assigns.
This document and the information contained herein is provided on an
"AS IS" basis and THE INTERNET SOCIETY AND THE INTERNET ENGINEERING
TASK FORCE DISCLAIMS ALL WARRANTIES, EXPRESS OR IMPLIED, INCLUDING BUT
NOT LIMITED TO ANY WARRANTY THAT THE USE OF THE INFORMATION HEREIN
WILL NOT INFRINGE ANY RIGHTS OR ANY IMPLIED WARRANTIES OF
MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE.
Moore, et al. Expires: Jul 2000 + 6 months [Page 56]
Internet Draft Policy Core Information Model July 2000
14. Appendix A - Class Identification in a Native CIM Implementation
While the CommonName property is present in the abstract superclass
Policy, and is thus available in all of its instantiable subclasses,
CIM does not use this property for naming instances. The following
subsections discuss how naming is handled in a native CIM
implementation for each of the instantiable classes in the Policy Core
Information Model.
Two things should be noted regarding CIM naming:
o When a CIM association is specified as "weak", this is a statement
about naming scopes: an instance of the class at the weak end of
the association is named within the scope of an instance of the
class at the other end of the association. This is accomplished by
propagation of keys from the instance of the scoping class to the
instance of the weak class. Thus the weak class has, via key
propagation, all the keys from the scoping class, and it also has
one or more additional keys for distinguishing instances of the
weak class, within the context of the scoping class.
o All class names in CIM are limited to alphabetic and numeric
characters plus the underscore, with the restriction that the first
character cannot be numeric. Refer to Appendix F "Unicode Usage"
in reference [2] for an exact specification of how CIM class names
are encoded in CIM strings.
14.1. Naming Instances of PolicyGroup and PolicyRule
A policy group always exists in the context of a system. In the
Policy Core Information Model, this is captured by the weak
aggregation PolicyGroupInSystem between a PolicyGroup and a System.
Note that System serves as the base class for describing network
devices and administrative domains.
A policy rule also exists in the context of a system. In the Policy
Core Information Model, this is captured by the weak association
PolicyRuleInSystem between a PolicyRule and a System.
The following sections define the CIM keys for PolicyGroup and
PolicyRule.
14.1.1. PolicyGroup's CIM Keys
The CIM keys of the PolicyGroup class are:
o SystemCreationClassName (A CIM_System key, propagated due to the
weak association, PolicyGroupInSystem)
o SystemName (A CIM_System key, propagated due to the weak
association, PolicyGroupInSystem)
o CreationClassName
o PolicyGroupName
Moore, et al. Expires: Jul 2000 + 6 months [Page 57]
Internet Draft Policy Core Information Model July 2000
They are defined in Reference [1] as follows:
NAME SystemCreationClassName
DESCRIPTION SystemCreationClassName represents the class name
of the CIM System object providing the naming scope
for the instance of PolicyGroup.
SYNTAX string [MaxLen 256]
QUALIFIER key
NAME SystemName
DESCRIPTION SystemName represent the individual name of the
particular System object, providing the naming
scope for the instance of PolicyGroup.
SYNTAX string [MaxLen 256]
QUALIFIER key
NAME CreationClassName
DESCRIPTION This property is set to "CIM_PolicyGroup", if the
PolicyGroup object is directly instantiated. Or,
it is equal to the class name of the PolicyGroup
subclass that is instantiated.
SYNTAX string [MaxLen 256]
QUALIFIER key
NAME PolicyGroupName
DESCRIPTION The identifying name of this policy group.
SYNTAX string [MaxLen 256]
QUALIFIER key
14.1.2. PolicyRule's CIM Keys
The CIM keys of the PolicyRule class are:
o SystemCreationClassName (A CIM_System key, propagated due to the
weak association PolicyRuleInSystem)
o SystemName (A CIM_System key, propagated due to the weak
association PolicyRuleInSystem)
o CreationClassName
o PolicyRuleName
SystemCreationClassName and SystemName work the same as defined for
the class PolicyGroup. See Section 14.1.1 for details.
The other two properties are defined in Reference [1] as follows:
NAME CreationClassName
DESCRIPTION This property is set to "CIM_PolicyRule", if the
PolicyRule object is directly instantiated. Or, it
is equal to the class name of the PolicyRule
subclass that is instantiated.
SYNTAX string [MaxLen 256]
QUALIFIER key
Moore, et al. Expires: Jul 2000 + 6 months [Page 58]
Internet Draft Policy Core Information Model July 2000
NAME PolicyRuleName
DESCRIPTION The identifying name of this policy rule.
SYNTAX string [MaxLen 256]
QUALIFIER key
14.2. Naming Instances of PolicyCondition and Its Subclasses
The CIM keys of the PolicyCondition class are:
o SystemCreationClassName
o SystemName
o PolicyRuleCreationClassName
o PolicyRuleName
o CreationClassName
o PolicyConditionName
Note that none of the keys are defined as propagated, although they
appear to fit this convention. The reason for this difference is
because (as indicated in Sections 5.1 and 6.4) the PolicyCondition
class is used to represent both reusable and rule-specific conditions.
This, in turn, affects what associations are valid for an instance of
PolicyCondition, and how that instance is named.
In an ideal world, an instance of the PolicyCondition class would be
scoped either by its PolicyRepository (for a reusable condition) or by
its PolicyRule (for a rule-specific condition). However, CIM has the
restriction that a given class can only be "weak" to one other class
(i.e., defined by one weak association).
To work within the restrictions of CIM naming, it is necessary to
"simulate" weak associations between PolicyCondition and PolicyRule,
and between PolicyCondition and PolicyRepository, through a technique
we'll call manual key propagation. Strictly speaking, manual key
propagation isn't key propagation at all. But it has the same effect
as (true) key propagation, so the name fits.
Figure 9 illustrates how manual propagation works in the case of
PolicyCondition. (Note that only the key properties are shown for each
of the classes.) In the figure, the line composed of 'I's indicates
class inheritance, the one composed of 'P's indicates (true) key
propagation via the weak aggregation PolicyRuleInSystem, and the ones
composed of 'M's indicate manual key propagation.
Moore, et al. Expires: Jul 2000 + 6 months [Page 59]
Internet Draft Policy Core Information Model July 2000
+------------------+
| System |
+------------------+
|CreationClassName |
|Name |
+------------------+
^ P
I PPPPPPPPPPPPPPPPPPPPPPPPPPPP
I P
+------------------+ +---------------v--------------+
| AdminDomain | | PolicyRule |
+------------------+ +------------------------------+
|CreationClassName | | System.CreationClassName |
|Name | | System.Name |
+------------------+ | CreationClassName |
^ | PolicyRuleName |
I +------------------------------+
I M
I M
+------------------+ M
| PolicyRepository | M
+------------------+ M
|CreationClassName | M
|Name | M
+------------------+ M
M M
M M
M M
+----v-------------------v----+
| PolicyCondition |
+-----------------------------+
| SystemCreationClassName |
| SystemName |
| PolicyRuleCreationClassName |
| PolicyRuleName |
| CreationClassName |
| PolicyConditionName |
+-----------------------------+
Figure 9. Manual Key Propagation for Naming PolicyConditions
Looking at Figure 9, we see that two key properties, CreationClassName
and Name, are defined in the System class, and inherited by its
subclasses AdminDomain and PolicyRepository. Since PolicyRule is weak
to System, these two keys are propagated to it; it also has its own
keys CreationClassName and PolicyRuleName.
A similar approach, though not automatic, is used in "manual key
propagation." Here is the approach for rule-specific and reusable
PolicyConditions:
Moore, et al. Expires: Jul 2000 + 6 months [Page 60]
Internet Draft Policy Core Information Model July 2000
o The manual propagation of keys from PolicyRule to PolicyCondition
involves copying the values of PolicyRule's four key properties
into four similarly named key properties in PolicyCondition. From
the point of view of the CIM specification language, the property
SystemName in PolicyCondition is a completely new key property.
However, the relationship to the Name property in System is defined
in the description of SystemName.
o The manual propagation of keys from PolicyRepository to
PolicyCondition works in exactly the same way for the first two key
properties. However, since PolicyRepository doesn't include
PolicyRule properties, the PolicyRuleCreationClassName and
PolicyRuleName have no values. A special value, "No Rule", is
assigned to both of these properties in this case, indicating that
this instance of PolicyCondition is not named within the scope of
any particular policy rule.
The following section defines the specific CIM keys for
PolicyCondition.
14.2.1. PolicyCondition's CIM Keys
PolicyCondition's key properties are defined in Reference [1] as
follows:
NAME SystemCreationClassName
DESCRIPTION SystemCreationClassName represents the class name
of the CIM System object providing the naming scope
for the instance of PolicyCondition. For a rule-
specific policy condition, this is the type of
system (e.g., the name of the class that created
this instance) in whose context the policy rule is
defined. For a reusable policy condition, this is
set to "CIM_PolicyRepository", if the
PolicyRepository object is directly instantiated.
Or, it is equal to the class name of the
PolicyRepository subclass that is instantiated.
SYNTAX string [MaxLen 256]
QUALIFIER key
NAME SystemName
DESCRIPTION The name of the System object in whose scope this
policy condition is defined. This property
completes the identification of the System object.
For a rule-specific policy condition, this is the
name of the instance of the system in whose context
the policy rule is defined. For a reusable policy
condition, this is name of the instance of
PolicyRepository that holds the policy condition.
SYNTAX string [MaxLen 256]
QUALIFIER key
NAME PolicyRuleCreationClassName
Moore, et al. Expires: Jul 2000 + 6 months [Page 61]
Internet Draft Policy Core Information Model July 2000
DESCRIPTION For a rule-specific policy condition, this property
identifies the class name of the policy rule
instance, in whose scope this instance of
PolicyCondition exists. For a reusable policy
condition, this property is set to a special value,
"No Rule", indicating that this instance of
PolicyCondition is not unique to one policy rule.
SYNTAX string [MaxLen 256]
QUALIFIER key
NAME PolicyRuleName
DESCRIPTION For a rule-specific policy condition,
PolicyRuleName completes the identification of the
PolicyRule object with which this condition is
associated. For a reusable policy condition, a
special value, "No Rule", is used to indicate that
this condition is reusable.
SYNTAX string [MaxLen 256]
QUALIFIER key
NAME CreationClassName
DESCRIPTION The class name of the PolicyCondition subclass that
is instantiated.
SYNTAX string [MaxLen 256]
QUALIFIER key
NAME PolicyConditionName
DESCRIPTION The identifying name of this policy condition.
SYNTAX string [MaxLen 256]
QUALIFIER key
14.3. Naming Instances of PolicyAction and Its Subclasses
From the point of view of naming, the PolicyAction class and its
subclasses work exactly like the PolicyCondition class and its
subclasses. See Section 14.2 and 14.2.1 for details.
Specifically, the CIM keys of PolicyAction are:
o SystemCreationClassName
o SystemName
o PolicyRuleCreationClassName
o PolicyRuleName
o CreationClassName
o PolicyActionName
They are defined in Reference [1] as follows:
NAME SystemCreationClassName
DESCRIPTION SystemCreationClassName represents the class name
of the CIM System object providing the naming scope
for the instance of PolicyAction. For a rule-
specific policy action, this is the type of system
Moore, et al. Expires: Jul 2000 + 6 months [Page 62]
Internet Draft Policy Core Information Model July 2000
(e.g., the name of the class that created this
instance) in whose context the policy rule is
defined. For a reusable policy action, this is set
to "CIM_PolicyRepository", if the PolicyRepository
object is directly instantiated. Or, it is equal
to the class name of the PolicyRepository subclass
that is instantiated.
SYNTAX string [MaxLen 256]
QUALIFIER key
NAME SystemName
DESCRIPTION The name of the System object in whose scope this
policy action is defined. This property completes
the identification of the System object. For a
rule-specific policy action, this is the name of
the instance of the system in whose context the
policy rule is defined. For a reusable policy
action, this is name of the instance of
PolicyRepository that holds the policy action.
SYNTAX string [MaxLen 256]
QUALIFIER key
NAME PolicyRuleCreationClassName
DESCRIPTION For a rule-specific policy action, this property
identifies the class name of the policy rule
instance, in whose scope this instance of
PolicyAction exists. For a reusable policy action,
this property is set to a special value, "No Rule",
indicating that this instance of PolicyAction is
not unique to one policy rule.
SYNTAX string [MaxLen 256]
QUALIFIER key
NAME PolicyRuleName
DESCRIPTION For a rule-specific policy action, PolicyRuleName
completes the identification of the PolicyRule
object with which this action is associated. For a
reusable policy action, a special value, "No Rule",
is used to indicate that this action is reusable.
SYNTAX string [MaxLen 256]
QUALIFIER key
NAME CreationClassName
DESCRIPTION The class name of the PolicyAction subclass that is
instantiated.
SYNTAX string [MaxLen 256]
QUALIFIER key
NAME PolicyActionName
DESCRIPTION The identifying name of this policy action.
SYNTAX string [MaxLen 256]
QUALIFIER key
Moore, et al. Expires: Jul 2000 + 6 months [Page 63]
Internet Draft Policy Core Information Model July 2000
14.4. Naming Instances of PolicyRepository
An instance of PolicyRepository is named by the two key properties
CreationClassName and Name that it inherits from its superclass
AdminDomain. These properties are actually defined in AdminDomain's
superclass, System, and then inherited by AdminDomain.
For instances of PolicyRepository itself, the value of
CreationClassName must be "CIM_PolicyRepository". (Recall that for
readability the prefix "CIM_" has been omitted from all class names in
this document). If a subclass of PolicyRepository (perhaps
QosPolicyRepository) is defined and instantiated, then the class name
"CIM_QosPolicyRepository" is used in CreationClassName.
The Name property simply completes the identification of the instance
of PolicyRepository.
14.5. Role of the CreationClassName Property in Naming
To provide for more flexibility in instance naming, CIM makes use of a
property called CreationClassName. The idea of CreationClassName is
to provide another dimension that can be used to avoid naming
collisions, in the specific case of instances belonging to two
different subclasses of a common superclass. An example will
illustrate how CreationClassName works.
Suppose we have instances of two different subclasses of
PolicyCondition, FrameRelayPolicyCondition and BgpPolicyCondition, and
that these instances apply to the same context. If we had only the
single key property PolicyConditionName available for distinguishing
the two instances, then a collision would result from naming both of
the instances with the key value PCName = "PC-1". Thus policy
administrators from widely different disciplines would have to
coordinate their naming of PolicyConditions for this context.
With CreationClassName, collisions of this type can be eliminated,
without requiring coordination among the policy administrators. The
two instances can be distinguished by giving their CreationClassNames
different values. One instance is now identified with the two keys
CreationClassName = "FrameRelayPolicyCondition" + PCName = "PC-1",
while the other is identified with
CreationClassName = "BgpPolicyCondition" + PCName = "PC-1".
Each of the instantiable classes in the Core Model includes the
CreationClassName property as a key in addition to its own class-
specific key property.
Moore, et al. Expires: Jul 2000 + 6 months [Page 64]
Internet Draft Policy Core Information Model July 2000
14.6. Object References
Today, all CIM associations involve two object references. CIM
decomposes an object reference into two parts: a high-order part that
identifies an object manager and namespace, and a model path that
identifies an object instance within a namespace. The model path, in
turn, can be decomposed into an object class identifier and a set of
key values needed to identify an instance of that class.
Because the object class identifier is part of the model path, a CIM
object reference is strongly typed. The GroupComponent object
reference in the PolicyGroupInPolicyGroup association, for example,
can only point to an instance of PolicyGroup, or to an instance of a
subclass of PolicyGroup. Contrast this with LDAP, where a DN pointer
is completely untyped: it identifies (by DN) an entry, but places no
restriction on that entry's object class(es).
An important difference between CIM property definitions and LDAP
attribute type definitions was identified earlier in Section 6: while
an LDAP attribute type definition has global scope, a CIM property
definition applies only to the class in which it is defined. Thus
properties having the same name in two different classes are free to
have different data types. CIM takes advantage of this flexibility by
allowing the data type of an object reference to be overridden in a
subclass of the association class in which it was initially defined.
For example, the object reference GroupComponent is defined in the
abstract aggregation class PolicyComponent to be a reference to an
instance of the class Policy. This data type for GroupComponent is
then overridden in subclasses of PolicyComponent. In
PolicyGroupInPolicyGroup, for example, GroupComponent becomes a
reference to an instance of PolicyGroup. But in
PolicyConditionInPolicyRule it becomes a reference to an instance of
PolicyRule. Of course there is not total freedom in this overriding
of object references. In order to remain consistent with its abstract
superclass, a subclass of PolicyComponent can only override
GroupComponent to be a reference to a subclass of Policy. A Policy
class is the generic context for the GroupComponent reference in
PolicyComponent.
Moore, et al. Expires: Jul 2000 + 6 months [Page 65]