U.S. patent application number 10/961850 was filed with the patent office on 2005-04-14 for dynamically configurable distributed security system.
This patent application is currently assigned to BEA Systems, Inc.. Invention is credited to Byrne, David, Falco, Mark A., Howes, Jason, Patrick, Paul, Riendeau, Richard J., Xu, Mingde, Yagen, Kenneth D..
Application Number | 20050081055 10/961850 |
Document ID | / |
Family ID | 34426050 |
Filed Date | 2005-04-14 |
United States Patent
Application |
20050081055 |
Kind Code |
A1 |
Patrick, Paul ; et
al. |
April 14, 2005 |
Dynamically configurable distributed security system
Abstract
A system and method for a dynamically configurable security
system, comprising, a security service module capable of
dynamically instantiating one or more plugin security provider
modules, the one or more security provider modules are coupled to
the security service module wherein the one or more security
provider modules are capable of responding dynamically to changes
in configuration information, wherein the security service module
is capable of receiving one or more security information updates,
and wherein the security service module is capable of controlling
access to one or more resources based on the one or more security
information updates.
Inventors: |
Patrick, Paul; (Manchester,
NH) ; Byrne, David; (Woburn, MA) ; Yagen,
Kenneth D.; (San Francisco, CA) ; Xu, Mingde;
(San Jose, CA) ; Howes, Jason; (Cambridge, MA)
; Falco, Mark A.; (Lexington, MA) ; Riendeau,
Richard J.; (Burlington, MA) |
Correspondence
Address: |
FLIESLER MEYER, LLP
FOUR EMBARCADERO CENTER
SUITE 400
SAN FRANCISCO
CA
94111
US
|
Assignee: |
BEA Systems, Inc.
San Jose
CA
95131
|
Family ID: |
34426050 |
Appl. No.: |
10/961850 |
Filed: |
October 8, 2004 |
Related U.S. Patent Documents
|
|
|
|
|
|
Application
Number |
Filing Date |
Patent Number |
|
|
60508427 |
Oct 10, 2003 |
|
|
|
Current U.S.
Class: |
726/26 |
Current CPC
Class: |
G06F 21/6218 20130101;
H04L 63/20 20130101; H04L 63/105 20130101 |
Class at
Publication: |
713/200 |
International
Class: |
G06F 011/30 |
Claims
What is claimed is:
1. A dynamically configurable security system, comprising: a
security service module capable of dynamically instantiating one or
more plugin security provider modules; the one or more security
provider modules are coupled to the security service module wherein
the one or more security provider modules are capable of responding
dynamically to changes in configuration information; wherein the
security service module is capable of receiving one or more
security information updates; and wherein the security service
module is capable of controlling access to one or more resources
based on the one or more security information updates.
2. The system of claim 1, further comprising: a first process
capable of modifying the configuration information.
3. The system of claim 1, further comprising: a distributed
provisioning system capable of delivering the one or more security
information updates.
4. The system of claim 1 wherein: a plugin security provider module
implements an interface that is compatible with the service
provider interface.
5. The system of claim 1 wherein: a plugin security provider module
is capable of providing at least one of the following services:
authentication, authorization, auditing, role mapping and
credential mapping.
6. The system of claim 1 wherein: the one or more security
information updates include at least one of: a policy and
configuration information.
7. The system of claim 6 wherein: a policy is an association
between a resource and at least one of: an action and a role.
8. The system of claim 1 wherein: the configuration information
includes a set of parameters; and wherein a parameter includes a
name and a type.
9. A method for a dynamically configurable security system,
comprising: dynamically instantiating one or more plugin security
provider modules wherein the one or more plugin security provider
modules is capable of responding dynamically to changes in
configuration information; coupling the one or more plugin security
provider modules to a security service module; receiving one or
more security information updates; and controlling access to one or
more resources based on the one or more security information
updates.
10. The method of claim 9, further comprising: modifying the
configuration information.
11. The method of claim 9, further comprising: delivering a
security information update.
12. The method of claim 9 wherein: a plugin security provider
module implements an interface that is compatible with the service
provider interface.
13. The method of claim 9 wherein: a plugin security provider
module is capable of providing at least one of the following
services: authentication, authorization, auditing, role mapping and
credential mapping.
14. The method of claim 9 wherein: the one or more security
information updates include at least one of: a policy and
configuration information.
15. The method of claim 14 wherein: a policy is an association
between a resource and at least one of: an action and a role.
16. The method of claim 9 wherein: the configuration information
includes a set of parameters; and wherein a parameter includes a
name and a type.
17. A machine readable medium having instructions stored thereon to
cause a system to: instantiate one or more plugin security provider
modules wherein the one or more plugin security provider modules is
capable of responding dynamically to changes in configuration
information; couple the one or more plugin security provider
modules to a security service module; receive one or more security
information updates; and control access to one or more resources
based on the one or more security information updates.
18. The machine readable medium of claim 17, further comprising
instructions stored thereon to cause the system to: modify the
configuration information.
19. The machine readable medium of claim 17, further comprising
instructions stored thereon to cause the system to: deliver a
security information update.
20. The machine readable medium of claim 17 wherein: a plugin
security provider module implements an interface that is compatible
with the service provider interface.
21. The machine readable medium of claim 17 wherein: a plugin
security provider module is capable of providing at least one of
the following services: authentication, authorization, auditing,
role mapping and credential mapping.
22. The machine readable medium of claim 17 wherein: the one or
more security information updates include at least one of: a policy
and configuration information.
23. The machine readable medium of claim 22 wherein: a policy is an
association between a resource and at least one of: an action and a
role.
24. The machine readable medium of claim 17 wherein: the
configuration information includes a set of parameters; and wherein
a parameter includes a name and a type.
25. A computer signal embodied in a transmission medium,
comprising: a code segment including instructions for dynamically
instantiating one or more plugin security provider modules wherein
the one or more plugin security provider modules is capable of
responding dynamically to changes in configuration information; a
code segment including instructions for coupling the one or more
plugin security provider modules to a security service module; a
code segment including instructions for receiving one or more
security information updates; and a code segment including
instructions for controlling access to one or more resources based
on the security information.
Description
CLAIM OF PRIORITY
[0001] This application claims priority from the following
application, which is hereby incorporated by reference in its
entirety:
[0002] U.S. Provisional Patent Application No. 60/508,427 entitled
SYSTEMS AND METHODS FOR ENTERPRISE APPLICATION SECURITY, by Paul
Patrick et al., filed Oct. 10, 2003 (Attorney Docket No.
BEAS-1253US0).
CROSS-REFERENCE TO RELATED APPLICATIONS
[0003] This application is related to the following co-pending
applications which are each hereby incorporated by reference in
their entirety:
[0004] U.S. patent application Ser. No. 10/______ entitled
DISTRIBUTED ENTERPRISE SECURITY SYSTEM, by Paul Patrick et al.,
filed ______, 2004 (Attorney Docket No. BEAS-1253US1);
[0005] U.S. patent application Ser. No. 10/______ entitled
DELAGATION IN A DISTRIBUTED SECURITY SYSTEM, by Paul Patrick et
al., filed ______, 2004 (Attorney Docket No. BEAS-1491US0);
[0006] U.S. patent application Ser. No. 10/______ entitled
DISTRIBUTED SECURITY SYSTEM WITH DYNAMIC ROLES, by Paul Patrick et
al., filed ______, 2004 (Attorney Docket No. BEAS-1492US0);
[0007] U.S. patent application Ser. No. 10/______ entitled
DISTRIBUTED SECURITY SYSTEM WITH SECURITY SERVICE PROVIDERS, by
Paul Patrick et al., filed ______, 2004 (Attorney Docket No.
BEAS-1493US0);
[0008] U.S. patent application Ser. No. 10/______ entitled AN
EMBEDDABLE SECURITY SERVICE MODULE, by Paul Patrick et al., filed
______, 2004 (Attorney Docket No. BEAS-1494US0);
[0009] U.S. patent application Ser. No. 10/______ entitled
CONFIGURATION OF A DISTRIBUTED SECURITY SYSTEM, by Paul Patrick et
al., filed ______, 2004 (Attorney Docket No. BEAS-1495US0);
[0010] U.S. patent application Ser. No. 10/______ entitled POLICY
ANALYSIS TOOL, by Paul Patrick et al., filed ______, 2004 (Attorney
Docket No. BEAS-1496US0);
[0011] U.S. patent application Ser. No. 10/______ entitled
DISTRIBUTED SECURITY SYSTEM POLICIES, by Paul Patrick et al., filed
______, 2004 (Attorney Docket No. BEAS-1498US0);
[0012] U.S. patent application Ser. No. 10/______ entitled POLICY
INHERITENCE THROUGH NESTED GROUPS, by Paul Patrick et al., filed
______, 2004 (Attorney Docket No. BEAS-1499US0);
[0013] U.S. patent application Ser. No. 10/______ entitled SECURITY
PROVIDER DEVELOPMENT MODEL, by Paul Patrick et al., filed ______,
2004 (Attorney Docket No. BEAS-1500US0);
[0014] U.S. patent application Ser. No. 10/______ entitled SECURITY
CONTROL MODULE, by Paul Patrick et al., filed ______, 2004
(Attorney Docket No. BEAS-1501US0);
[0015] U.S. patent application Ser. No. 10/______ entitled
DELEGATED ADMINISTRATION FOR A DISTRIBUTED SECURITY SYSTEM, by Paul
Patrick et al., filed ______, 2004 (Attorney Docket No.
BEAS-1502US0);
[0016] U.S. patent application Ser. No. 10/______ entitled
DYNAMICALLY CONFIGURABLE DISTRIBUTED SECURITY SYSTEM, by Paul
Patrick et al., filed ______, 2004 (Attorney Docket No.
BEAS-1504US0);
[0017] U.S. patent application Ser. No. 10/______ entitled
DISTRIBUTED ENTERPRISE SECURITY SYSTEM, by Paul Patrick et al.,
filed ______, 2004 (Attorney Docket No. BEAS-1505US0); and
[0018] U.S. patent application Ser. No. 10/______ entitled
DISTRIBUTED ENTERPRISE SECURITY SYSTEM FOR A RESOURCE HIERARCHY, by
Paul Patrick et al., filed ______, 2004 (Attorney Docket No.
BEAS-1506US0).
COPYRIGHT NOTICE
[0019] A portion of the disclosure of this patent document contains
material which is subject to copyright protection. The copyright
owner has no objection to the facsimile reproduction by anyone of
the patent document or the patent disclosure, as it appears in the
Patent and Trademark Office patent file or records, but otherwise
reserves all copyright rights whatsoever.
INCORPORATION BY REFERENCE
[0020] The following U.S. patent and U.S. patent Publications are
hereby incorporated by reference in their entirety:
[0021] U.S. Pat. No. 6,158,010 entitled SYSTEM AND METHOD FOR
MAINTAINING SECURITY IN A DISTRIBUTED COMPUTER NETWORK, by Mark
Moriconi et al,. filed Dec. 5, 2000.
[0022] U.S. patent application Ser. No. 09/878,536, entitled SYSTEM
AND METHOD FOR SERVER SECURITY AND ENTITLEMENT PROCESSING, by Paul
Patrick, filed Jun. 11, 2001, now U.S. Patent Publication No.
US2002-0188869, published on Dec. 12, 2002 (Attorney Docket No.
BEAS-1084US0);
FIELD OF THE DISCLOSURE
[0023] The present disclosure relates generally to systems and
methods for a distributed security system that can be deployed to
protect enterprise applications in a heterogeneous computing
environment.
BACKGROUND
[0024] Companies are increasingly deploying network-based services
(e.g., web services) to create enterprise-wide applications in
heterogeneous computing environments. However, ensuring the
security of such services is seen as a barrier to their wide-spread
adoption.
[0025] Services are typically assembled from one or more software
applications that may each implement their own proprietary security
model and may each require their own form of administration. Thus,
the security information and security procedures from one portion
of an enterprise application may not be compatible with other
portions. Furthermore, some security models may not take into
account that users can have different policies depending on when
and how they are interacting with a service. The result is a
collection of services having incompatible silos of security.
[0026] Another reason why security remains a barrier to deploying
web services is cost. Software developers can require extensive
training before they are able to implement and deploy such systems.
Developing a security system can also be seen as a distraction from
a programmer's primary duty of developing application software.
Even if this were not so, legacy applications not designed for
dynamic changes in policy can require that security measures be
intricately tied to the application itself. This leads to a
security systems that are brittle and costly to maintain.
BRIEF DESCRIPTION OF THE DRAWINGS
[0027] FIG. 1 is a flow diagram illustration of resource request
processing in accordance to various embodiments of the
invention.
[0028] FIG. 2 is an illustration of an exemplary security service
module in accordance to various embodiments of the invention.
[0029] FIG. 3 is illustrates exemplary security service module
deployments in accordance to various embodiments of the
invention.
[0030] FIG. 4 is an illustration of information provisioning in
accordance to various embodiments of the invention.
[0031] FIG. 5 is an illustration of an exemplary authentication
provider in accordance to various embodiments of the invention.
[0032] FIG. 6 is an illustration of an exemplary security control
module service framework in accordance to various embodiments of
the invention.
[0033] FIG. 7 is an illustration of an exemplary provisioning
service provider in accordance to various embodiments of the
invention.
[0034] FIG. 8 is an illustration of exemplary security service
module and security control module interaction possibilities in
accordance to various embodiments of the invention.
[0035] FIG. 9 is an illustration of an exemplary administration
console and server in accordance to various embodiments of the
invention.
[0036] FIG. 10 is an exemplary illustration of system components in
accordance to various embodiments of the invention.
[0037] FIG. 11 is an illustration of exemplary policy analysis
functionality of the administration console in accordance to
various embodiments
[0038] FIG. 12 is an illustration of an exemplary user group
hierarchy in accordance to various embodiments.
DETAILED DESCRIPTION
[0039] The invention is illustrated by way of example and not by
way of limitation in the figures of the accompanying drawings in
which like references indicate similar elements. References to
embodiments in this disclosure are not necessarily to the same
embodiment, and such references mean at least one.
[0040] In the following description, numerous specific details are
set forth to provide a thorough description of the invention.
However, it will be apparent to one skilled in the art that the
invention may be practiced without these specific details. In other
instances, well-known features have not been described in detail so
as not to obscure the invention.
[0041] In various embodiments, systems and methods disclosed herein
provide an approach to security for enterprise-wide applications
while allowing for simple integration with third party security
products. Security can be administered and controlled through
flexible policies. The system can perform fine-grained, business
transaction-based authorization at each point of access of
application and other resources. In various embodiments, a resource
can represent a component of an application, information, a
process, a service, a function, a device and/or any other suitable
entity which can be interacted with programmatically. By way of
illustration, resources can include (but are not limited to) the
following:
[0042] Administrative resources such as an administration console
and the application resources that represent enterprise
applications. This can include individual Enterprise Application
Archive (EAR) files and individual components, such as Enterprise
Java.RTM. Bean (EJB) Java.RTM. Archive (JAR) files contained within
an EAR file.
[0043] Component Object Model (COM) resources that are designed as
program component objects.
[0044] Resources that are designed as connectors, which allow the
integration of Java.RTM. applications with existing enterprise
information systems.
[0045] Enterprise JavaBean (EJB) resources including EJB JAR files,
individual EJBs within an EJB JAR, and individual methods on an
EJB.
[0046] Java.RTM. DataBase Connectivity (JDBC) resources including
groups of connection pools, individual connection pools, and
multipools.
[0047] Java.RTM. Naming and Directory Interface (JNDI)
resources.
[0048] Java.RTM. Messaging Service (JMS) resources.
[0049] Server resources related to servers. This type of resource
includes operations that start, shut down, lock or unlock
servers.
[0050] Uniform Resource Locator (URL) resources related to web
applications. This type of resource can be a Web Application
archive (WAR) file or individual components of a Web application
(such as servlets and JSPs).
[0051] Web services resources related to services that can be
shared by and used as components of distributed, Web-based
applications. This type of resource can be an entire Web service or
individual components of a Web service (such as a stateless session
EJB, particular methods in that EJB, the Web application that
contains the web-services.xml file, and so on).
[0052] Specific business transactions, such as a money transfer or
security trade.
[0053] Application graphical user interface elements, such as web
pages, windows, dialog boxes, buttons, text fields and menu
selections.
[0054] Database, directory server or other information
structures.
[0055] Products or services in a portal.
[0056] The present disclosure is not limited to or dependent on any
type of resource. Accordingly, it will be apparent to those of
skill in the art that any resource type presently known or yet to
be developed is fully within the scope and spirit of the present
disclosure.
[0057] Resources may contain attributes that provide additional
information concerning the resource and how it may be accessed. For
example, a filetype resource attribute could indicate whether a
resource has a corresponding HTML, image, JSP or PDF
representation. By way of illustration, permission to view a
resource could be granted if the resource has a filetype equal to
PDF. Such permissions are termed policies. Policies embody the
security needs and reflect the business processes of an
organization. Policies can be analogized to declarative rules, such
as:
[0058] Only employees in the sales team may view sales
documents
[0059] Or
[0060] Only employees in the sales team who are regional managers
can alter discounts.
[0061] Policies dynamically identify resources that need to be
protected, privileges or roles allowed each resource, and
constraints that apply to each user or group of users that attempt
to access the resource. In various embodiments, roles can be
defined to dynamically associate users and/or groups of users based
on some criteria. For example, a system administrator role might
include all users having a certain skill level and only during
certain times of day (e.g., after 5:00 pm). In various embodiments,
the granting of roles can be expressed as policies (i.e., role
policies). An authorization policy can be based on role(s) such
that it determines which role(s) are permitted to access a resource
under certain conditions.
[0062] In various embodiments, the system can use several
categories of policies:
[0063] Authentication Policy--controls whether authentication is
required and the constraints and/or rules under which it is
allowed;
[0064] Role Policy--determines which users and/or groups of users
are in a role based on rules and/or constraints;
[0065] Authorization Policy--dynamically determines whether a role,
user or group of users are permitted to perform an action or access
a resource based on rules and/or constraints; and
[0066] Auditing Policy--defines auditing information to be captured
and stored in a designated place.
[0067] Once policies are deployed, security provider modules (see
below) can apply them to the resources that they are managing.
[0068] In various embodiments, each component of the system may
have a unique or shared configuration. In aspects of these
embodiments, configuration can be managed through an administration
console. Configuration also can be used to configure security
provider modules (see below) for supported service types (e.g.,
Authentication, Authorization, Auditing, Role Mapping and
Credential Mapping). For example, a security provider module
installed on an intermediate network between an untrusted external
network and a trusted internal network might be used to
authenticate incoming user requests against a customer database,
control which parts of the application user interface the user can
access based on their role in the organization, and pass their
identity to an application behind the firewall. An authentication
security provider could be configured to provide this service. An
audit security provider could also be configured so that all
transactions are audited.
[0069] In various embodiments, authentication policy is used to
determine the constraints under which a user, known to trusted
authentication services and can prove who they claim to be, is
permitted to be authenticated. These services may be embodied in an
operating system, a Lightweight Directory Access Protocol (LDAP)
directory, or a legacy application, for example. By way of
illustration, the method of authentication can depend on the type
of user directory and authentication method desired, or on the form
of identity accepted from a trusted external source (e.g., a web
application might require a simple username and password, whereas a
financial application might require more stringent authentication).
In username and password authentication, a user provides an
identification and a matching password, which a configured provider
(e.g., LDAP, Windows NT, or database) can then check against the
configured directory to validate the user's identity.
[0070] By way of illustration, two other types of authentication
are certificate-based and single sign-on. Certificate-based
authentication is typically used when an TLS/SSL or HTTPS client
requests access and presents its digital certificate to the server.
The server passes a certificate to an X.509 Identity Asserter
provider, which verifies the digital certificate and creates a
local identity. The digital certificate is issued by a trusted
certificate authority and is accepted as a form of user
authentication. A Single Sign-On form of authentication is typical
in a setting where the users are using multiple applications, each
of which maintains its own security and requires user identity to
control access. These may be within an organization, through an
employee portal or externally on a completely different domain. For
example, if the user requires access to separate applications, one
in New York with a local LDAP and the other in London that uses a
proprietary user database, a SAML identity assertion credential may
be passed between these environments and either consumed by the
remote security system directly or converted to a local identity
through credential mapping.
[0071] In various embodiments, authorization policy can be based on
the roles a user or group of users may hold at the time a resource
access request is made. Unlike groups, which are relatively static
and can persist for a session, roles are dynamic and are assigned
to users or groups of users by evaluating role policies. For
example, a group of bank tellers might have access to the same set
of applications to perform specific banking tasks. Thus, there
might be a role called Teller that encompasses the bank tellers.
Authorization and other policies could then be enacted to restrict
access to the applications only to users in the role of Teller.
Role policies can significantly reduce the overall number of
policies required and makes features like delegation easier to
manage. Role policies can be combined with authorization requests
or may be directly evaluated to support uses like application or
portal personalization.
[0072] In various embodiments, authorization policies are
associated with a resource. In aspects of these embodiments,
resources can be organized in a hierarchy. The hierarchical
children of a given resource in the hierarchy inherit the
authorization policies associated with their parent resource(s). In
aspects of these embodiments, authorization policies assigned to
individual resources take precedence or override inherited
authorization policies. By way of illustration, assume resource C
is a child of resource B, and resource B is a child of resource A.
If policy P1 is associated with A, B and C will inherit P1. If
policy P2 is associated with B (which is for the same control of
access as P1), B and C will not inherit P1, but C will inherit P2.
Finally, if C is associated with a policy P3 (which is for the same
control of access as P2), then C will not inherit P2. If P1, P2 and
P3 were all for different rules for controlling access, then B
would inherit P1 and C would inherit P1 and P2.
[0073] Generally speaking, an authorization policy can be specified
as follows (wherein items in square brackets indicate alternatives;
italic font indicates optional items):
[0074] [GRANT, DENY, DELEGATE] ([action, role], resource, subject,
delegator) IF (constraint condition).sub.1 ... IF (constraint
condition).sub.N;
[0075] Where:
[0076] GRANT permits a specified action. DENY revokes it. DELEGATE
assigns an action from one subject to another;
[0077] Action is the name of a resource or resource attribute to
grant, deny or delegate. Role is the name of a role to grant, deny
or delegate;
[0078] Resource is the fully qualified or short-hand name of the
resource that this policy will be associated with;
[0079] Subject is the fully qualified name of one or more users,
groups and/or roles that are granted/denied/delegated the
authorization policy or the role. A special subject called any
denotes that any user, group and role is potentially a subject;
[0080] Delegator is optionally specified for delegation and is the
subject delegating the role or action to the subject; and
[0081] IF (constraint condition) is one or more optional conditions
placed on the action. Conditions can include one or more arithmetic
and logical functions and expressions involving attributes of
resources or other entities in the system, such as user attributes,
group membership, dynamic attributes (e.g., time, date, location),
delegation attributes and other suitable information.
[0082] In various embodiments, and by way of illustration, user and
user group information can be stored and managed in external
directories and databases and can be gathered and/or calculated at
authentication time. For example, the source of user information
might be an employee database, which would include the fact that
the user is in the Sales department group, in the London office
group, and in the star-salesmen group. Typically, it is desirable
to establish a policy that applies to a large community of people,
as establishing policies for each and every user can be time
consuming, tedious, and hard to manage. Organizational structure,
in the form of groups, maps well to typical business policies and
is very useful for determining authorization and role policy. Thus,
it is desirable to create groups of users (or groups of groups)
whose tasks are related. For example, the following policy grants
the ability to open accounts via the resource TellerApp to all
users and groups belonging to the Tellers group from 9:00 AM to
5:00 PM, Monday through Friday:
[0083] GRANT (OpenAccount, //app/policy/TellerApp, //sgrp /Tellers)
if time24 in [900..1700] AND if dayofweek in [Monday...
Friday];
[0084] An alternative to defining a group of users is to specify a
role which defines the conditions under which a user or group of
users is provided a capability. Roles are desirable when group
membership tends not to be static (e.g., based on prevailing
conditions). Roles are used to abstract the capabilities mechanism
and represent a business task that a user or group of users are
entitled to perform. In the above example, if instead a role called
Tellers was used, the policies could be specified as follows:
[0085] GRANT (//role/Tellers, //app/policy/TellerApp, any) if
(is_employee = TRUE AND (employee.position = Teller OR
employee.position = manager));
[0086] GRANT (OpenAccount, //app/policy/TellerApp, //role /Tellers)
if (time24 in [900..1700] AND if dayofweek in [Monday...
Friday]);
[0087] The first policy defines a role named Tellers which includes
anybody that is an employee and who happens to have a position of
teller or manager. The second policy permits the OpenAccount action
to be performed by a user that holds the Tellers role, subject to
the constraints of time and day. By defining Tellers as a role, its
definition can change dynamically as employee accounts are created
and modified whereas a group definition would have to be changed
along with employee accounts to remain current.
[0088] In various embodiments, role and authorization policies can
also be based on contextual data available at run-time (e.g., a
transaction amount passed in as a parameter). In some cases, a
policy constraint can utilize information that is not locally
available and which could require the security provider module (see
below) to retrieve it from another source or compute it from
another piece of data. For example, assume that an account balance
is an attribute of a customer. To allow customers with an account
balance over $100,000 access to the premier banking features of an
application, the following policy could be defined:
[0089] GRANT (//role/premierbanking, //app/policy/bankapp,
//sgrp/bankusers/customers/) if (accountbalance > 100000);
[0090] In various embodiments, delegation is the transfer of
capabilities from one user or group to another, and optionally
under certain conditions. For example, a manager may delegate the
ability to approve expense reports to his secretary. Delegation is
often used by administrative users to control policy for other
administrative users or for application users to allow someone else
to perform an action for them or in their absence. In various
embodiments, there can be restrictions on what rights can be
further delegated by the delegate.
[0091] A policy can be used to delegate a role or privilege from
one subject (i.e., the delegator) to another (i.e., the delegate).
By way of illustration, delegation policies can be described with
the following syntax:
[0092] DELEGATE ([action, role], resource, delegate, delegator) IF
(constraint condition).sub.1 ... IF (constraint
condition).sub.N;
[0093] In various embodiments, the delegate gets the ability to
perform an action if the constraints evaluate to true and if the
delegator would also get the ability to perform an action under the
same conditions. If the delegator later loses the rights he has
delegated, the delegate also looses them. In aspects of these
embodiments, a delegator's identity information can be retrieved
dynamically in order to determine if, at time of evaluation, the
delegator has the capabilities (role) being delegated. By way of
illustration, a delegator can put additional constraints on what is
being shared. Constraints can include support for time-based
restrictions such as (but not limited to) the following:
[0094] Range of time (e.g., Jun. 5, 2003-Jun. 10, 2003);
[0095] After specific start time (e.g., after Jan. 1, 2004
17:00:00);
[0096] Until expiration time (e.g., until Jul. 1, 2003
09:00:00);
[0097] On a repeating time (e.g., Fridays, Weekends,
7:00:00-9:00:00 every day);
[0098] Related to action being delegated (e.g., Joe can approve my
trades for me if they are less than $5,000); and
[0099] Based on an attribute of the delegate (e.g., Joe can approve
my trades for me if they are less than his trading limit.).
[0100] In various embodiments, a delegate can re-delegate
capabilities/rights to another subject. The ability to re-delegate
can be protected such that re-delegation could be prevented when
the original delegation is made. In various embodiments, a
delegator can indicate if the delegate can re-delegate the rights
they are delegated.
[0101] By way of illustration, suppose George delegates the ability
to author a product requirements document (PRD) to Andy, and also
gives Andy the permission to delegate this ability further (not
shown). Assuming the following delegated roles:
[0102] DELEGATE (//role/author, //app/PRD, Andy, George)
[0103] DELEGATE (//role/editor, //app/ PRD, //group/Managers,
George);
[0104] DELEGATE (//role/reader, //app/ PRD, //group/Engineers,
George);
[0105] GRANT (//priv/write, //app/ PRD, //role/author);
[0106] The author role is delegated to Andy from George. Thus, Andy
can be in the author role whenever George could be. George also
delegates the roles of editor and reader to the groups Managers and
Engineers, respectively. As with the author role, Managers can be
in the editor role and Engineers can be in the reader role whenever
George can be in those roles.
[0107] By way of further illustration, Andy then decides to make
Ken an author on an "authorization" section of the PRD. This is
possible because George did not restrict Andy from re-delegating
this role:
[0108] DELEGATE (//role/author, //app/ PRD, Ken, Andy) if (section
= "authorization")
[0109] Assuming that Ken is a Manager and an Engineer he can get
the editor and reader roles on the PRD. Accordingly, Ken can act as
an author if he's working on the "authorization" section and if
Andy (the delegator) can act as an author on the PRD when the
section is "authorization." The first policy says that Andy is
always an author on the PRD so Andy's policy gets a grant, which
makes the delegation policy return true and Ken picks up the
"author" role assuming the given section is the "authorization"
section. If Ken were trying to "write" in the PRD, he would only be
able to if he had picked up the "author" role.
[0110] In various embodiments, delegation policies can be used to
create administrative policy. In aspects of these embodiments, the
rights of an administrator can be delegated by an administrator
with greater rights than their own. Default administrative policies
can be provided in this regard.
[0111] In various embodiments, auditing of administrative policy
actions can be performed. This can include recording whether
permission to perform an action was granted and whether the action
was actually performed.
[0112] In various embodiments, delegation rules can be evaluated
for authorization of a resource request only if a decision has not
been reached with authorization policy rules. Such an algorithm can
be used to require delegation evaluation in all cases.
[0113] FIG. 1 is a flow diagram illustration of resource request
processing in accordance to various embodiments of the invention.
Although this figure depicts functional steps in a particular order
for purposes of illustration, the process is not necessarily
limited to any particular order or arrangement of steps. One
skilled in the art will appreciate that the various steps portrayed
in this figure can be omitted, rearranged, performed in parallel,
combined and/or adapted in various ways.
[0114] In various embodiments, a client request to perform an
operation on a resource can be intercepted and provided to a
security service module (SSM). In various embodiments, an SSM can
be integrated with a server, application server, web server, a
process or any suitable device such as network firewalls, routers,
relays, appliances, etc. In aspects of these embodiments, a
container (e.g., an EJB or servlet container) can intercept the
request and present it to a SSM. By way of illustration, the SSM
can be provided with a subject that holds an authenticated identity
(e.g., an authenticated user), a resource identifier, user group
information, and/or a context handler object through which the SSM
can obtain additional information about the context of the
request.
[0115] Referring to FIG. 1, a process for handling intercepted
resource requests is illustrated. In step 100, a role mapping
process maps the user that initiated the request to role(s) based
on evaluation of one or more role policies. In various embodiments,
role mapping can occur just prior to when an access decision is
rendered for a protected resource. In step 102, zero or more
applicable authorization policies are determined based on the
subject, mapped roles and the resource (which is the target) of the
request and requested action. Based on an evaluation of the
authorization policies (if any) in step 104, permission to access
the resource is either granted in step 108 or denied in step
106.
[0116] In various embodiments, resource request processing is
performed by an SSM. FIG. 2 is an illustration of an exemplary SSM
in accordance to various embodiments of the invention. Although
this diagram depicts components as logically separate, such
depiction is merely for illustrative purposes. It will be apparent
to those skilled in the art that the components portrayed in this
figure can be combined or divided into separate software, firmware
and/or hardware components. Furthermore, it will also be apparent
to those skilled in the art that such components, regardless of how
they are combined or divided, can execute on the same computing
device or can be distributed among different computing devices
connected by one or more networks or other suitable communication
means.
[0117] In aspects of these embodiments, the SSM 220 includes
several functional layers that can reside in one or more processes.
An adaptation layer 200 ties a framework layer 201 and a services
layer 202 to a run-time environment. By way of example, a runtime
environment can include an application/web server, an application
program, a network firewall, and other suitable environments that
require enterprise security. The adaptation layer includes an
Application Program Interface (API) 214 to allow exploitation of
the framework layer in the run-time environment. In aspects of
these embodiments, an API can comprise a programmatic interface
(e.g., class, method and/or function definitions), a communication
interface such as a web service or other suitable mechanism for
exchanging messages, and/or any other suitable protocol for
invoking functionality and exchanging information. The present
disclosure is not limited to or dependent on any API implementation
presently known or yet to be developed, as will be appreciated by
those of skill in the art.
[0118] The adaptation layer can invoke services of the framework
layer via a framework programming interface (FPI) 218. Likewise,
the framework layer can invoke services of the services layer via a
service provider interface (SPI) 216. As with the adaptation layer
API, the FPI and SPI can comprise a programmatic interface (e.g.,
class, method and/or function definitions), a communication
interface such as a web service or other suitable mechanism for
exchanging messages, and/or any other suitable protocol for
invoking functionality and exchanging information. The present
disclosure is not limited to or dependent on any FPI/SPI
implementation presently known or yet to be developed, as will be
appreciated by those of skill in the art.
[0119] In aspects of these embodiments, the FPI can translate API
invocations into one or more SPI invocations. The FPI also hides
the SPI from the adaptation layer, thus preventing any dependency
on the SPI from forming. The SPI provides the FPI access to a set
of dynamically configurable security services represented as "plug
in" security provider modules ("providers"). In various
embodiments, each module is compatible with the SPI. The services
layer can be provided with a default set of providers. In another
embodiment, the adaptation layer can communicate directly with the
services layer, without an intervening framework layer (e.g., the
API could invoke the SPI directly).
[0120] By way of illustration, authentication provider(s) 204 can
authenticate, verify, and map security tokens to an internal format
and support, for example, a single sign-on capability.
Authorization provider(s) 206 can enforce authorization policies
for resources that take authorization policy into consideration.
Audit provider(s) 208 can audit some or all security actions taken
by the framework layer. Role mapping provider(s) 210 can map roles
to users/groups based on role definitions. Finally, credential
mapping provider(s) 212 can map authentication credentials for a
user to legacy application for single sign-on. The services layer
can accommodate more than one provider of the same type.
[0121] FIG. 3 illustrates exemplary SSM deployments in accordance
to various embodiments of the invention. Although this diagram
depicts components as logically separate, such depiction is merely
for illustrative purposes. It will be apparent to those skilled in
the art that the components portrayed in this figure can be
combined or divided into separate software, firmware and/or
hardware components. Furthermore, it will also be apparent to those
skilled in the art that such components, regardless of how they are
combined or divided, can execute on the same computing device or
can be distributed among different computing devices connected by
one or more networks or other suitable communication means.
[0122] In various embodiments and by way of illustration, SSMs can
be deployed in applications, network hardware components,
application servers, web servers, proxy servers, and any other
suitable systems requiring security. For example, a legacy 304 or
custom application program 302 can be modified to utilize the SSM
through an API or other suitable means. An application/web server
can also be configured to use the SSM via a J2EE servlet filter
310, security plugin provider 308, or delegated 312, as are
commonly known in the art. The SSM can also be deployed in a proxy
server 300. Finally, the SSM can be utilized in networking hardware
314 such as (but not limited to), a switch, router, relay, bridge,
repeater or a hub.
[0123] In various embodiments, an administration console and server
306 allows an administrator or other user to create, modify and
delete user, group and policy definitions. In addition, the
administrator has the capability to delegate administrative
privileges to other users. In various embodiments, the
administration console allows a user to pose "what if" policy
questions to determine what users/groups have access to a resource
under a given set of conditions. The administration console can
communicate with an administrative server which can dispense
provisioning information (e.g., policy and/or configuration
information) itself, or can utilize a distribution point process.
Provisioning information can be dispensed to SSMs directly, or via
one or more intermediate processes (e.g., SCMs).
[0124] FIG. 4 is an exemplary illustration of information
provisioning in accordance to various embodiments of the invention.
Although this diagram depicts objects/processes as logically
separate, such depiction is merely for illustrative purposes. It
will be apparent to those skilled in the art that the
objects/processes portrayed in this figure can be combined or
divided into separate software and firmware components.
Furthermore, it will also be apparent to those skilled in the art
that such objects/processes, regardless of how they are combined or
divided, can execute on the same computing device or can be
distributed among different computing devices connected by one or
more networks or other suitable communication means.
[0125] In various embodiments, policy and configuration information
is distributed to SSMs 402 via Security Control Mangers (SCMs) 400.
This is referred to as "provisioning". SCMs can reside anywhere on
a computer network. In one embodiment, an SCM resides locally on
systems that have one or more SSMs 402. An administration server
can provision the policy and configuration information to an SCM as
a provisioning request. In one embodiment, the information
provisioned to an SCM is only relevant to SSMs deployed on the same
system as the SCM. In aspects of these embodiments, only changes
("deltas") to this information are propagated to SCMs. This is
desirable since it can reduce the amount of information that needs
to be transmitted between the administration server and the SCMs.
By way of illustration, this can be accomplished by associating a
version number with provisioning information or subsets
thereof.
[0126] SCMs can cache provisioned information in a local store 404
and further provision it to one or more SSMs. In one embodiment, an
SCM provisions information to SSMs that reside on the same system
as the SCM. In various embodiments, provisioned configuration
information can specify security providers, locations of directory
servers, databases, and other suitable information. By way of
illustration, an SSM can dynamically load security providers based
on configuration information supplied to it by an SCM.
[0127] The SCM has many architectural benefits for the system.
Firstly, the SCM can serve as the sole external management
interface for all SSM components. This can eliminate redundant
management infrastructure in system components, thus allowing all
system components to take advantage of improvements in future
versions of the SCM management interface. Secondly, having a single
management interface per computing device has security benefits. An
SCM-enabled host can expose a single management communication
channel rather than one per SSM. This eliminates the need for a
listen port in each SSM process, drastically reducing the number of
open ports that are required to be secured and monitored. Finally,
the use of the SCM can greatly simplify SSM configuration. Rather
than relying on instance specific configuration files, an SSM can
retrieve its entire configuration from the SCM via a well-known
communication port.
[0128] A typical deployment of the system can consist of one or
more SSMs distributed throughout an enterprise--embedded in
applications, application servers, appliances, routers, web servers
and other suitable systems. Configuring a potentially large number
of distributed SSMs is a non-trivial task. Traditional
configuration mechanisms, such as the use of local property files,
break down as the number of configured entities increases. Rather
than forcing customers to edit and maintain a property file for
each SSM instance, the system includes a provisioning
infrastructure that allows customers to manage the configuration of
all SSMs through an administration console.
[0129] An SCM can supply provisioned information to SSMs as needed
(e.g., in response to requests by SSMs). An SCM can also convey the
information automatically without being requested to do so. In
various embodiments, an SCM only provides provisioning information
to an SSM is that relevant to that SSM. In aspects of these
embodiments, the SCM provides only deltas to SSMs. In various
embodiments, communication between system components can be
accomplished with secure protocols. By way of illustration,
mutually authenticated Transport Layer Security (TSL) connections
can be utilized between components. In addition, the SCM and SSM
can exchange Public-Key Infrastructure (X.509) certificates to
establish identity and trust.
[0130] In various embodiments, support for Security Assertion
Markup Language (SAML) is provided. SAML is a protocol for
propagating identity that can be used from one security domain to
the next and between components in the system. By way of example, a
dynamic webpage can accept a user name and a password. These can be
provided to a SAML authentication security provider via the SSM
API. The authentication provider can perform authentication and
return an artifact which can be passed back to the webpage and/or
exchanged with other applications that require SAML assertions.
Likewise, the SAML authentication provider can accept an artifact
and return an indication of whether it is valid or not.
[0131] With reference to FIGS. 1-2, role policies can be evaluated
as part of role mapping. In various embodiments and before a
decision as to whether to allow access to a resource is rendered,
role mapping can be performed to determine whether a user can be
dynamically granted a role. The computed role can access
information in the context of the resource request, including the
identity of the target resource (if available), desired action to
be performed and the parameter values (if any) of the request. The
context information can be used as parameters in an expression that
is evaluated by the SSM.
[0132] With reference to FIG. 2, a security provider module exposes
a programmatic interface(s) for integrating itself into the SSM and
for retrieving management and configuration information. In various
embodiments, a provider does not need to support management
operations, just configuration. If management operations are
supported, they can be exposed in the administration console or
other system/application. In various embodiments, a security
provider can make available a configuration schema. By way of
illustration, this feature is useful if the administration console
is to display information about the configuration properties for a
provider. An interface can be included in the SPI for retrieving
(and setting) this information. In various embodiments, the SSM can
use configuration information to enumerate and create provider
instances for a particular SSM instance and can pass configuration
information to each newly instantiated security provider during its
initialization. By way of illustration, configuration schemas can
be declared as Extensible Markup Language (XML) documents that
conform to a security provider configuration XML schema. The XML
schema can define how configuration parameters are expressed, and
the definition can include parameter name, type, and value
constraints.
[0133] FIG. 5 is an exemplary illustration of an authentication
provider in accordance to various embodiments of the invention.
Although this diagram depicts components as logically separate,
such depiction is merely for illustrative purposes. It will be
apparent to those skilled in the art that the components portrayed
in this figure can be combined or divided into separate software,
firmware and/or hardware components. Furthermore, it will also be
apparent to those skilled in the art that such components,
regardless of how they are combined or divided, can execute on the
same computing device or can be distributed among different
computing devices connected by one or more networks or other
suitable communication means.
[0134] A Java.RTM. Authentication and Authorization Service (JAAS)
authentication provider 512 as part of an SSM services layer 201
can authenticate user credentials against various sources of data.
The provider implements an SPI for authentication providers in
order to integrate itself into the services layer. In one
embodiment, configuration information can be provided to the
authentication provider as an Authenticator MBean. An MBean is a
Java.RTM. object that implements the Java.RTM. Management
Extensions (JMX) specification.
[0135] By way illustration, the provider can utilize databases 510
such as relational databases (RDBMs), Java.RTM. Database
Connectivity (JDBC) data sources, and other sources of credential
information. (To maximize performance database connection pooling
502 can be implemented in the provider.) A given user name can be
used to lookup a user's record in a database. If the lookup locates
a record in the database for the user, the password supplied can be
compared against the password in the database for the user. If
required, the record can also be checked to determined if it is
disabled/locked out. Through configuration, the provider can be
equipped with sufficient information to query a database source
through JDBC and query or retrieve the necessary information. It
can verify that the user exists in a store, validate a user
password algorithm and retrieve user group and attribute
information.
[0136] In various embodiments, providers can obtain user attribute
information in at least three ways: 1) information passed to the
provider as part of an authorization request; 2) information
obtained from another service as needed; and 3) information
obtained by evaluating policies. In various embodiments, user
attributes, group membership, security credentials and other
suitable information can be queried from repositories and made
accessible to authentication and authorization providers within the
SSM. In addition, such information can be cached to improve
performance. By way of illustration, a Lightweight Directory Access
Protocol (LDAP) interface for authorization and role mapping
providers to a separate attribute service can be provided to obtain
user-related information. Alternatively, a Simple Object Access
Protocol (SOAP) interface to the attribute service can be provided.
In yet another illustration, a database interface to a user
database can be provided. This approach could be used to embed the
attribute service in the provider. In various embodiments, the
attribute service could be a plugin to the SCM. In other
embodiments, the attribute service could be incorporated into the
provisioning information distributed by SCMs. In yet another
embodiment, the attribute service could be deployed inside an
administration server.
[0137] The attribute service can be designed to scale to provide
acceptable performance while servicing many providers
simultaneously. Caching can be used within providers and possibly
within the attribute service to improve performance. However, any
caching in the attribute service can be expired or updated upon
changes to the data to guarantee that only valid data is returned
to the authorization and role mapping providers. Also, connection
pooling by providers may be necessary to leverage multiple
instances of the attribute service.
[0138] The SCM 400 can be used for retrieving and setting
configuration information in authentication providers. In aspects
of these embodiments, security providers--themselves plugins into
the SSM--can also support their own dynamic plugins. By way of
illustration, the authentication provider can include support for
custom "provider" extension points. A provider extension handler
class 504 can manage provider extensions 508 by loading them on
initialization and unloading them on shutdown. References to
provider extension class methods can be stored in a map and
executed when needed. The configuration for a provider can contain
a class name for each provider extension. That class can implement
the method(s) for the type of provider extension it is configured
for as well as initialize( ) and shutdown( ) methods. In this way,
new functionality can be added to security providers without
requiring the provider itself to be modified.
[0139] FIG. 6 is an exemplary illustration of an SCM service
framework in accordance to various embodiments of the invention.
Although this diagram depicts objects/processes as logically
separate, such depiction is merely for illustrative purposes. It
will be apparent to those skilled in the art that the
objects/processes portrayed in this figure can be combined or
divided into separate software, firmware and/or hardware
components. Furthermore, it will also be apparent to those skilled
in the art that such objects/processes, regardless of how they are
combined or divided, can execute on the same computing device or
can be distributed among different computing devices connected by
one or more networks or other suitable communication means.
[0140] In various embodiments, an SCM 400 can include an extensible
service framework 602 that provides services to other processes
through inclusion of one or more service providers. In this
illustration, a provisioning service provider 604 and a process
management service provider 606 are shown. Many other service types
are possible and can, by way of an example, encompass performance
monitoring, event monitoring, configuration and resource discovery.
In various embodiments and by way of illustration, the SCM service
is a component framework that can facilitate the development of SCM
agents from a set of independent but cooperating components or
services. By way of illustration, service providers can be exposed
via a SOAP interface.
[0141] In various embodiments, the service framework can define the
lifecycle of service providers (e.g., initialization, commencement,
termination, etc.) as well as how they can locate and interact with
each other. In addition, the service framework can support the
static and dynamic configuration of service providers. New service
providers 608 can be added as needed. By way of illustration, the
service framework can expose an SPI that service providers
implement in order to integrate themselves into the service
framework. By way of illustration, service framework and service
provider configurations can be stored in an XML file that can be
parameterized at installation.
[0142] In various embodiments, a service manager 610 can be used to
intermediate between the service framework and services. The
service manager instantiates each service provider and equips each
with a service context that allows the provider to interact with
the service SSM. The service SSM can also contain an in-process
registry 600 in which services can be registered. By way of
illustration, the SCM can be managed by a platform-specific process
such as an embeddor (a system agent on Microsoft Windows.RTM.
systems) or a daemon (on Unix.RTM. systems). Such a process can
monitor the SCM and ensure the high availability of the SCM and its
services. The SCM can provide other core functionality to services,
such as configuration and logging.
[0143] The process management service provider (PMSP) 606 can start
and stop processes and monitor one process on behalf of another,
restarting it in the event of a crash, or stopping it when the
requesting parent process exits. This capability greatly improves
the manageability and reliability of the system. In various
embodiments, SCMs can utilize the PMSP to manage one or more
out-of-process SSMs. Security providers can also use the PMSP to
start, stop, and monitor authorization and role mapping providers
instances.
[0144] The provisioning service provider (PSP) 604 is responsible
for maintaining provisioning information for all SSMs associated
with the SCM. In various embodiments, the PSP exposes two services,
one for processing provisioning requests from distributors and
another for processing queries from SSMs. In various embodiments,
the PSP can accept provisioning information for the set of all SSMs
within a single administration domain from the distributor that, in
one embodiment, are collocated on the same system as the SCM. In
various embodiments, the PSP can associate a provisioning
information version number with each SSM it is aware of. By way of
illustration, on startup the PSP can contact a distributor, passing
the version numbers of SSMs it is aware of. If necessary, the
distributor can then send updates required to bring each SSM's
provisioned information up-to-date. Thereafter, the PSP can listen
for updates from the distributor and update a local persistent
store accordingly. Each SSM can retrieve its current provisioning
information at any time by querying the PSP via a information
client query.
[0145] In various embodiments, SCM service providers can optionally
establish trust relationships with other processes such as the
distributor. By way of illustration, each trust relationship can be
established by the creation of a public/private key pair that is
created and registered when the SCM is installed. The SCM can
retain the following information in order to establish trust
relationship:
[0146] Private key and X.509 digital certificate for the SCM in a
local keystore; and
[0147] X.509 certificates to any process with which the SCM has
established a trust relationship. The X.509 certificate can be
stored in a local keystore under an alias that is the unique
identifier associated with the trusted process.
[0148] FIG. 7 is an exemplary illustration of a provisioning
service provider in accordance to various embodiments of the
invention. Although this diagram depicts objects/processes as
logically separate, such depiction is merely for illustrative
purposes. It will be apparent to those skilled in the art that the
objects/processes portrayed in this figure can be combined or
divided into separate software, firmware and/or hardware
components. Furthermore, it will also be apparent to those skilled
in the art that such objects/processes, regardless of how they are
combined or divided, can execute on the same computing device or
can be distributed among different computing devices connected by
one or more networks or other suitable communication means.
[0149] With reference to FIG. 7 and in various embodiments, the PSP
604 exposes remote 704 and local 706 interfaces. In various
embodiments, the remote PSP interface can be used to exchange
information with one or more distributors 700. By way of
illustration, an interface can include one or more of: a
programmatic interface (e.g., class, method and/or function
definitions), a communication interface such as a web service
(e.g., SOAP over HTTP(S)), other suitable mechanism(s) for
exchanging messages, and a protocol for invoking functionality
and/or exchanging information. By way of illustration, an interface
can be implemented as XML over SOAP and can be optionally secured
using transport layer security protocol (TLS). However, the present
disclosure is not limited to or dependent on any interface
implementation presently known or yet to be developed, as will be
appreciated by those of skill in the art.
[0150] In various embodiments, the local PSP interface 706 can be
used to exchange information with one or more information query
clients 702. An information query client encapsulates interaction
with the local PSP interface from other processes and, by
introducing a level of indirection, makes other processes less
dependent on the implementation details of the local PSP interface.
By way of illustration, an information query client can be
implemented as an object that exposes one or more services through
which other processes can communicate with a provisioning
service.
[0151] The PSP is capable of receiving updates to provisioning
information through the remote PSP interface. The updates can be
full or incremental and can be propagated by distributor(s) (or
other suitable processes). In various embodiments, the SCM can
maintain provisioned information for all SSM instances for which it
is responsible. The SCM can receive and store provisioned
information updates in a local persistent store 708, and make it
available to SSMs via one or more clients. In various embodiments,
the provisioning mechanism ensures that only the provisioned
information required by an SCM is provisioned to that SCM.
Likewise, the SCM can ensure that only the provisioned information
required by an SSM is made available to that SSM. In various
embodiments, an atomic set of provisioned information updates can
be transactional across all SCMs; that is, an update can take
effect only if all SCMs are able to successfully process the
update. This ensures that all provisioning within an administration
domain is consistent across all running SCMs within that
domain.
[0152] In various embodiments, Jakarta Phoenix can serve as the SCM
service framework. Phoenix is a micro-kernel designed and
implemented on top of the Jakarta Avalon. It provides a number of
facilities to manage the environment of server applications. Such
facilities include log management, class loading, and thread
management. The Jakarta API defines a standard method of piecing
together server components to create a server. Furthermore, since
they are built on top of Jakarta Avalon, Phoenix applications
automatically take advantage of the standard code design patters,
rules and guidelines dictated by Avalon that increase the
modularity and reuse of software components. In various
embodiments, an SCM process can be managed by a Java.RTM. Service
Wrapper (JSW) process. The JSW monitors a JVM process and
automatically restarts it if that JVM crashes or hangs. This makes
it possible to install the SCM as a Windows.RTM. NT Service or as a
UNIX.RTM. daemon, allowing it to be automatically started upon
system reboot.
[0153] FIG. 8 is an exemplary illustration of SSM and SCM
interaction possibilities in accordance to various embodiments of
the invention. Although this diagram depicts objects/processes as
logically separate, such depiction is merely for illustrative
purposes. It will be apparent to those skilled in the art that the
objects/processes portrayed in this figure can be combined or
divided into separate software, firmware and/or hardware
components. Furthermore, it will also be apparent to those skilled
in the art that such objects/processes, regardless of how they are
combined or divided, can execute on the same computing device or
can be distributed among different computing devices connected by
one or more networks or other suitable communication means.
[0154] In various embodiments, an SSM 402 includes a services layer
202 having associated security providers (not shown), a
configuration service 804, and several clients (802, 702) for
communication with an SCM 400. Security providers and other
processes can obtain configuration information via the
configuration service, which exposes services to allow a security
provider to configure itself on start-up and dynamically
thereafter. The configuration service encapsulates an
implementation capable of retrieving this information--thus
insulating security providers from implementation details. By way
of illustration, a configuration service can be implemented as an
object that exposes one or more simple methods through which other
processes can communicate with a provisioning service provider. In
aspects of these embodiments and by way of illustration, the
services layer 202 can instantiate a new configuration service
during initialization and pass it to one or more security providers
through an initialization method. Both the service layer and its
providers can query the configuration service at any time for
configuration information.
[0155] In various embodiments, the configuration service 804 allows
security providers to retrieve their current provisioning
information from provisioning service provider 604 via information
query client 702. In aspects of these embodiments and by way of
illustration, provisioning information can be represented as scoped
name-value pairs of text strings (e.g., an XML document). In
various embodiments, additional implementations could be developed
that retrieve configuration and policy data from alternative
sources (e.g., a JMX MBean server). In various embodiments, the
configuration service can cache provisioning information in a
persistent store 808. Caching prevents unnecessary calls to the SCM
at startup and reduces the amount of information transferred
between the SCM and SSMs.
[0156] In various embodiments, security providers can use the PMSP
606 in order to start, stop, and monitor other processes. By way of
illustration, on startup an SCM can use PMSP to start SSMs. In one
embodiment, a process management client (not shown) can encapsulate
interaction with the PMSP thus making security providers less
dependent on the implementation details of the protocol used to
communicate with the PMSP. By way of illustration, a processes
management client can be implemented as an object that exposes one
or more services through which other processes can communicate with
a PMSP.
[0157] In various embodiments, the SCM client 802 allows the SSM to
interact with the SCM by instantiating clients. By way of
illustration, an SSM can first create an instance of the SCM
client, and then invoke functionality on the SCM client to retrieve
clients to other SCM services (e.g., provisioning). The SCM can
maintain a client registry service provider (CSRP) 814 that manages
information regarding the SSMs that are currently using the
SCM.
[0158] The SCM client encapsulates interaction with a CSRP from
other processes and, by introducing a level of indirection, makes
other processes less dependent on the implementation details of the
local PSP interface. By way of illustration, a SCM client can be
implemented as an object that exposes one or more services through
which other processes can communicate with a CSRP. In aspects of
these embodiments, the CSRP can maintain a map of SSM status
information. The status information can include, but is not limited
to, an SSM's start time and total operation time. In further
aspects, the client registry service can expose functionality for
enumerating running SSMs and retrieving SSM status. For example,
the service can allow other components/processes to register for
SSM status update notifications. The service can also expose an
interface that can receive heartbeat messages from SSM clients.
[0159] In various embodiments, the SCM service framework can
include a data store service provider 810 which can be used to
implement a persistent cache of provisioned information. This
service can expose basic create, read, update and delete (CRUD)
functionality. By way of illustration, the service can persist
information in a file system, a database, and any other suitable
storage means. In aspects of these embodiments, the data store
service provider implements an SPI compatible with the SCM service
SSM. By way of illustration, the data store service provider can be
used both by the SCM and by SSMs or other processes.
[0160] The SSM allows processes to leverage services offered
through security provider modules. By the same token, the SSM's
flexible infrastructure also allows security vendors, integrators,
and users to provide their own custom security provider modules.
Secure provider modules can be mixed and matched to create unique
security solutions, allowing organizations to take advantage of new
technology advances in some areas while retaining proven methods in
others. Thanks to the use of SCMs to distribute provisioning
information, a security solution embodied in a particular SSM
configuration can be painlessly scaled to suit large or small
deployments.
[0161] FIG. 9 is an exemplary illustration of an administration
server in accordance to various embodiments of the invention.
Although this diagram depicts objects/processes as logically
separate, such depiction is merely for illustrative purposes. It
will be apparent to those skilled in the art that the
objects/processes portrayed in this figure can be combined or
divided into separate software, firmware and/or hardware
components. Furthermore, it will also be apparent to those skilled
in the art that such objects/processes, regardless of how they are
combined or divided, can execute on the same computing device or
can be distributed among different computing devices connected by
one or more networks or other suitable communication means.
[0162] In various embodiments, the administration server includes
an administration coordinator 910 to harmonize activities of the
various administration server components. In aspects of these
embodiments, the administration server 900 includes an
administration console 904 ("console") that allows administrative
users to easily configure and manage various aspects of the system
through a unified management user interface. Configuration
information is cached in storage 912 for use by the distributor
914. Configuration settings for security provider modules can be
viewed and modified in the console wherein modifications to
configuration information can be automatically propagated back to
the effected security provider module(s). Through the console, SCMs
and SSMs can be deployed on specific machines. The console can also
be used to associate an SSM with an SCM and specify the security
provider modules that a given SSM will utilize. In various
embodiments, if an SSM has not been associated with an SCM, the SSM
can either dynamically associate with an SCM through a discovery
procedure or instantiate a new SCM process to associate with.
[0163] In aspects of these embodiments, the administration console
includes a user interface (not shown). By way of a non-limiting
example, a user interface can include one or more of the following:
1) a graphical user interface (GUI) (e.g., rendered with Hypertext
Markup Language) rendered on a display device or projected onto a
user's retina; 2) an ability to respond to sounds and/or voice
commands; 3) an ability to respond to input from a remote control
device (e.g., a cellular telephone, a PDA, or other suitable remote
control); 4) an ability to respond to gestures (e.g., facial and
otherwise); 5) an ability to respond to commands from a process on
the same or another computing device; and 6) an ability to respond
to input from a computer mouse and/or keyboard. This disclosure is
not limited to any particular UI. Those of skill in the art will
recognize that many other user interfaces are possible and fully
within the scope and spirit of this disclosure.
[0164] In various embodiments, a loader 906 imports policy
definitions 902 from one or more files, databases, other systems,
or from any other suitable source(s), into storage 912. The loader
can extract the definitions itself or enlist the help of one or
more additional processes (not shown). Security policies frequently
need more information than simple identity data to resolve
sophisticated policy decisions. In various embodiments, an optional
meta-directory or virtual directory 908 is provided to access and
synchronize user attributes from a variety of sources to support
these policies. Since these attributes may not be stored in any
single place, this directory serves to integrate attributes for
each user and cache them in storage 912 for use by SSMs. The
meta-directory or vitual directory can integrate information from a
variety of sources, including (but not limited to), application
databases, Lightweight Directory Access Protocol (LDAP) directory
servers, network databases, and others suitable sources. The
distributor 914 utilizes information in the storage 912 to provide
SCMs with policy and configuration information relevant to their
SSMs.
[0165] FIG. 10 is an exemplary illustration of system components in
accordance to various embodiments of the invention. Although this
diagram depicts objects/processes as logically separate, such
depiction is merely for illustrative purposes. It will be apparent
to those skilled in the art that the objects/processes portrayed in
this figure can be combined or divided into separate software,
firmware and/or hardware components. Furthermore, it will also be
apparent to those skilled in the art that such objects/processes,
regardless of how they are combined or divided, can execute on the
same computing device or can be distributed among different
computing devices connected by one or more networks or other
suitable communication means.
[0166] The system can employ a fully-distributed security
enforcement architecture consisting of SSMs 402 embedded in
applications, application servers, web servers, and hardware,
throughout an enterprise. To facilitate the management of a
potentially large number of distributed SSMs, the administration
server 900 uses a remote administration mechanism to distribute
configuration data to each SSM. In various embodiments, the SCM 400
is a component of this remote administration mechanism. Each SCM is
responsible for storing 404 and maintaining policy and
configuration information for all SSMs that it is associated with.
In one embodiment, an SCM is associated with the SSMs on its local
machine. When a change to an SSM's configuration or policy is made
and distributed from the administration console, an SCM receives
the change and updates its cached copy of the configuration. The
change is then propagated to the SSM which can adapt to the
configuration change dynamically or at a later time. In addition to
facilitating management, the SCM enables SSMs to operate in the
absence of the administration server. Since SCMs maintain a
persistent copy of each configuration, new SSMs can be started and
existing SSMs can continue to function, even if the Administration
server goes down.
[0167] FIG. 11 is an exemplary illustration of policy analysis
functionality of the administration console in accordance to
various embodiments. The administration console has a policy tool
user interface 1100 that allows users to create, modify, delete and
query policies in the system. The policy tool has four main areas:
a query region 1114 for providing search criteria, a results list
1116 for viewing results of a search, an action area 1132 including
buttons 1122-1128 for operating on policies, and a policy type
filter area 1130 for optionally filtering out results from the
results list. In aspects of these embodiments, results in the
result list can be sorted by columns.
[0168] The query region allows a user to find policies by
specifying zero or more policy components as search criteria. In
various embodiments, policy components can be specified as text
strings. All policies having matching components to those specified
by the user are included in the results list. User-specified
components can include more than one value, and can include wild
cards (e.g., "*" matches everything) to allow for pattern matching.
The user can specify the following policy components: zero or more
resources 1102, zero or more policy types 1104 (e.g., GRANT, DENY,
DELEGATE), zero or more privilege or role name(s) 1106, zero or
more subject name(s) 1108, zero or more delegator names 1110, and
zero or more constraints 1112. In aspects of these embodiments,
constraints can be specified in the same manner they are specified
in policy definitions. The resource 1102 can be specified as one or
more text strings or can be interactively chosen from a graphical
(tree) representation of a resource hierarchy 1118.
[0169] By way of illustration, if user named Joe wanted to find all
policies that effect him. Joe would specify the root of the
resource hierarchy in 1102 or 1118, or alternatively he would
specify a wild card "*" in text field 1102. Alternatively, if Joe
wanted to see what policies where applicable to him for specific
resources, he could specify those resources in 1102 or 1118. In the
subject field, Joe specifies his user name and, optionally, any
roles and/or groups he belongs to. By hitting the return key or by
selecting the Find button 1122 or other suitable action, a search
commences and all applicable policies are displayed in the results
list. Alternatively, the results list could be dynamically updated
as search parameters are entered. Joe could refine the results
using policy type filters 1130. For example, by selecting "All"
(the default) all policies are displayed. But results can be
limited to any combination of policy type, including but not
limited to: configuration, authentication, role, authorization and
audit.
[0170] Since the subject component for policies can include roles,
the policy tool first discovers all of the applicable role policies
for users and groups specified in the subject field 1108. Then, the
policy tool can perform the search using the user-specified
criteria but with the addition of any discovered roles to the
subject component. In addition, policy inheritance on the selected
resource(s) is taken into account. In aspects of these embodiments,
the policy tool can utilize the SSM API to map users and groups to
roles, and to select policies that satisfy the user specified
criteria.
[0171] In various embodiments, the rows in the results list
correspond to policies wherein one or more policies can be
selected. Selected policies can be modified by, for example,
further selecting the Edit button 1126. In aspects of these
embodiments, selection of the edit button can cause an a policy
editor window to appear in which selected policies can be edited.
In further aspects of these embodiments, a policy can be edited in
place in the results list. Or the selected policies can be deleted
by further selecting the Delete button 1128. New policies can be
added by selection of the New button 1124, which can cause a policy
editor window to appear.
[0172] By way of further illustration, the policy tool can be used
to find all policies delegated to a particular subject. By way of
example, this can be accomplished by putting the user's name in the
subject field 1108, and a wildcard "*" in the delegator field 1110.
This will find all policies where a delegation has been made to the
user (or a role the user is in). Likewise, the policy tool can be
used to find all policies that have been delegated by a particular
user. For example, the user's name can be provided in the delegator
field and a wildcard can be placed in the subject field.
[0173] FIG. 12 is an exemplary illustration of a user group
hierarchy in accordance to various embodiments. A group is a
collection of users that can share some common property, such as a
department, a job function, or a job title. By way of illustration,
a group named Accounting might contain users in the accounting
department. It is important to realize that this does not directly
reflect what access rights they have. That depends on the policies
defined for the group for a particular resource. A group can
contain users or other groups; users who are assigned to a group
are called group members. Nested memberships of groups within a
group form a hierarchy 1200. A policy whose subject component
includes a group is the same as defining the policy for each user
who is a member of the group. By using groups, there is no need to
define policy for each and every user; instead, each user in a
group inherits the policies applied to the group; this rule also
applies to nested groups. In aspects of these embodiments, group
policy inheritance is accomplished by automatically assigning
parent group policies to its nested group(s).
[0174] Referring to FIG. 12, users or groups inherit the policies
of any group to which they belong, either directly or indirectly
through the group to which they are assigned. When group inherits
policies from other groups, usually the parent group has fewer
access privileges than its nested group(s). By way of illustration,
the Employees group 1202 has three nested groups: Contractors 1204,
Part-Time 1206 and Middle Management 1208. The Middle Management
group has one nested group, Senior Management 1210. The Employees
group is the subject of Policy A (indicated by the tag attached to
the group name). The Middle Management group is the subject of
Policy B, but also becomes a subject of Policy A through
inheritance from the Employees group. Likewise, the Senior
Management group is the subject of Policy C and, through
inheritance, Policies A and B.
[0175] In various embodiments, the administration console provides
one or more user interfaces for creating, reading, updating and
deleting users and groups, and for nesting groups within groups.
Furthermore, attributes associated with users and groups can also
be created, read, updated and deleted from the console.
[0176] Various embodiments can be implemented using a conventional
general purpose or a specialized digital computer or
microprocessor(s) programmed according to the teachings of the
present disclosure, as will be apparent to those skilled in the
computer art. Appropriate software coding can readily be prepared
by skilled programmers based on the teachings of the present
disclosure, as will be apparent to those skilled in the software
art. The invention can also be implemented by the preparation of
integrated circuits or by interconnecting an appropriate network of
conventional component circuits, as can be readily apparent to
those skilled in the art.
[0177] Various embodiments include a computer program product which
is a storage medium (media) having instructions stored thereon/in
which can be used to program a computer to perform any of the
embodiments presented herein. The storage medium can include, but
is not limited to, any type of disk including floppy disks, optical
discs, DVD, CD-ROMs, microdrive, and magneto-optical disks, ROMs,
RAMs, EPROMs, EEPROMs, DRAMs, VRAMs, flash memory devices, magnetic
or optical cards, nanosystems (including molecular memory ICs), or
any type of media or device suitable for storing instructions
and/or data.
[0178] Stored on any one of the computer readable medium (media),
the present invention includes software for controlling both the
hardware of the general purpose/specialized computer or
microprocessor, and for enabling the computer or microprocessor to
interact with a human user or other mechanism utilizing the results
of the present invention. Such software can include, but is not
limited to, device drivers, operating systems, execution
environments/containers, and applications.
[0179] The foregoing description of the preferred embodiments of
the present invention has been provided for the purposes of
illustration and description. It is not intended to be exhaustive
or to limit the invention to the precise forms disclosed. Many
modifications and variations will be apparent to the practitioner
skilled in the art. Embodiments were chosen and described in order
to best describe the principles of the invention and its practical
application, thereby enabling others skilled in the art to
understand the invention, the various embodiments and with various
modifications that are suited to the particular use contemplated.
It is intended that the scope of the invention be defined by the
following claims and their equivalents.
* * * * *