COMMERCIAL SOLUTIONS for CLASSIFIED

(CSfC)

Data-at-Rest
Capability Package V5.0

Version 5.0
18 November 2020
 CHANGE HISTORY
Title Version Date Change Summary
Commercial Solutions 0.8 July 2014 Initial draft of CSfC Data-at-Rest (DAR)
for Classified (CSfC) requirements
Data-at-Rest (DAR)
Capability Package
Commercial Solutions 1.0 September 2014 Official release of CSfC DAR requirements
for Classified (CSfC)  Introduced SWFDE/FE (SF) Solution
Data-at-Rest (DAR) Design
Capability Package  Aligned with SW FDE Protection
Profile (PP) 1.0 & FE Extended
Package (EP) 1.0
Commercial Solutions 1.8 October 2014 Initial draft of CSfC DAR Version 2
for Classified (CSfC) requirements
Data-at-Rest (DAR)
Capability Package
Commercial Solutions 2.0 December 2014 Official release of CSfC DAR Version 2
for Classified (CSfC) requirements
Data-at-Rest (DAR)  Added PE/FE (PF) Solution Design
Capability Package  Aligned with MDF PP 3.0
Commercial Solutions 2.8 May 2015 Initial draft of CSfC DAR Version 3
for Classified (CSfC) requirements
Data-at-Rest (DAR)
Capability Package
Commercial Solutions 3.0 March 2016 Official release of CSfC DAR Version 3
for Classified (CSfC) requirements.
Data-at-Rest (DAR)  Added HWFDE/FE and HWFDE/SW
Capability Package FDE (HF and HS) Solution Design
 Updated requirements to reflect new
FDE Collaborative Protection Profile
(cPP) 2.0
 Discussed the associated
Independent Software Vendor (ISV)
technology which aligns with the FDE
cPP 2.0
 Added Lost and Found (LF) use case
Commercial Solutions 3.8 January 2017 Initial draft of CSfC DAR Version 4
for Classified (CSfC) requirements
Data-at-Rest (DAR)
Capability Package
Commercial Solutions 4.0 January 2018 Official release of CSfC DAR Version 4
for Classified (CSfC) requirements
Data-at-Rest (DAR)  Added Removable Media (RM)
Capability Package Solution Component and Solution

i
 Title Version Date Change Summary
Design
 Added continuous physical control
(previously positive control) guidance
 Added random password generation
 Added secure file deletion guidance
 Added optional two-factor
authentication
 Relocated Threat Section to a
separate document available on the
CSfC webpage
 Removed the Testing Section to a
separate DAR Testing Annex
document
 Changed DAR-PE-5 from minimum of
4 characters to minimum of 6
characters
Commercial Solutions 4.8 September 2019 Initial Draft of CSfC DAR Version 5
for Classified (CSfC) Requirements.
Data-at-Rest (DAR)
Capability Package
Commercial Solutions 5.0 18 November 2020  Added Enterprise Management
for Classified (CSfC) (EM) Use Case.
Data-at-Rest (DAR)  Added Unattended Operations
Capability Package (UO) Use Case.
 Added Hardware FDE/Hardware
FDE (H/H) Solution Design.
 Added “optional” DAR Location
Based Services capability.
 Added guidance for
Implementing CSfC in a High
Assurance GOTS Environment.
 Updated glossary and acronym
list
 Removed Table 17: Lost and
Found requirements table;
alternatively, dispersed the
requirements into existing tables,
now identifiable as “LF” in the
“Use Case” column.
 Changed RM from a Solution
Design to a Use Case.

ii
 Table of Contents
1 Introduction .......................................................................................................................................... 1
2 Purpose and Use ................................................................................................................................... 2
3 Legal Disclaimer .................................................................................................................................... 2
4 Data-at-Rest Protection Overview ........................................................................................................ 2
4.1 Implementing CSfC in a High Assurance GOTS Environment ....................................................... 3
4.2 Rationale for Layered Encryption ................................................................................................. 3
4.3 Solution States .............................................................................................................................. 3
4.3.1 EUD Solution States .............................................................................................................. 3
4.3.2 Enterprise Management (EM) Server & Mission Control Element (MCE) Solution States ... 4
4.4 DAR CNSA Suite ............................................................................................................................. 5
4.5 Authentication .............................................................................................................................. 5
4.6 Continuous Physical Control ......................................................................................................... 7
4.7 Red, Gray, and Black Data ............................................................................................................. 8
4.8 Cryptographic Erase (CE)............................................................................................................... 8
4.9 Provisioning ................................................................................................................................... 8
4.10 Secure File Deletion ...................................................................................................................... 8
4.11 DAR Location Based Services ...................................................................................................... 10
5 Solution Components.......................................................................................................................... 11
5.1 Software Full Disk Encryption (SWFDE) ...................................................................................... 11
5.2 File Encryption (FE) ..................................................................................................................... 11
5.3 Platform Encryption (PE) ............................................................................................................. 13
5.4 Hardware Full Disk Encryption (HWFDE) .................................................................................... 14
5.5 End User Device (EUD) ................................................................................................................ 15
5.6 DAR Enterprise Server (ES) and Mission Control Elements ........................................................ 16
6 Solution Designs .................................................................................................................................. 16
6.1 SWFDE/FE (SF) Solution Design................................................................................................... 17
6.2 PE/FE (PF) Solution Design .......................................................................................................... 17
6.3 HWFDE/FE (HF) Solution Design ................................................................................................. 18
6.4 HWFDE/SWFDE (HS) Solution Design ......................................................................................... 18
6.5 HWFDE/HWFDE (HH) Solution Design ........................................................................................ 18

iii
 7 DAR Use Cases..................................................................................................................................... 19
7.1 Lost and Found (LF) Use Case ..................................................................................................... 20
7.2 Removable Media (RM) Use Case ............................................................................................... 20
7.3 Enterprise Management (EM) Use Case ..................................................................................... 21
7.3.1 Enterprise Management via MA CP, MSC CP, or Campus WLAN CP .................................. 23
7.3.2 Enterprise Management via High Assurance GOTS solution .............................................. 24
7.3.3 Enterprise Management Key Recovery............................................................................... 24
7.4 Unattended Operations (UO) Use Case ...................................................................................... 25
8 Configuration Requirements............................................................................................................... 26
9 Requirements for Selecting Components ........................................................................................... 28
10 Configuration ...................................................................................................................................... 29
10.1 Overall Solution Requirements ................................................................................................... 30
10.2 Configuration Requirements for All DAR Components............................................................... 30
10.3 SWFDE Component Requirements ............................................................................................. 33
10.4 FE Component Requirements ..................................................................................................... 34
10.5 PE Component Requirements ..................................................................................................... 34
10.6 HWFDE Component Requirements............................................................................................. 35
10.7 End User Devices Requirements ................................................................................................. 36
10.8 Configuration Change Detection Requirements ......................................................................... 40
10.9 Device Management Requirements ........................................................................................... 41
10.10 Auditing Requirements ............................................................................................................... 42
10.11 Key Management Requirements ................................................................................................ 43
10.12 Supply Chain Risk Management Requirements .......................................................................... 44
11 Solution Operation, Maintenance, & Handling Requirements ........................................................... 45
11.1 Use and Handling of Solution Requirements .............................................................................. 45
11.2 Incident Reporting Requirements ............................................................................................... 48
12 Role-Based Personnel Requirements.................................................................................................. 49
13 Information to Support the AO ........................................................................................................... 51
13.1 Solution Testing .......................................................................................................................... 51
13.2 Risk Assessment .......................................................................................................................... 52
13.3 Registration of Solutions ............................................................................................................. 52

iv
 14 Testing Requirements ......................................................................................................................... 53
Appendix A: Glossary of Terms ................................................................................................................... 54
Appendix B: Acronyms ................................................................................................................................ 58
Appendix C: CSfC Incident Reporting Template.......................................................................................... 62
Appendix D: Password/Passphrase Strength Parameters .......................................................................... 64
Appendix E: Configuration Guidance .......................................................................................................... 67
Appendix F: Continuous Physical Control ................................................................................................... 73
Appendix G: References .............................................................................................................................. 75

Table of Figures
Figure 1: Software Full Disk Encryption ...................................................................................................... 11
Figure 2: Software File Encryption .............................................................................................................. 12
Figure 3: Platform Encryption ..................................................................................................................... 14
Figure 4: Hardware Full Disk Encryption ..................................................................................................... 15
Figure 5: Removable Media Use Case ......................................................................................................... 21
Figure 6: Enterprise Management Use Case............................................................................................... 23
Figure 7: Unattended Operations Use Case................................................................................................ 26

List of Tables
Table 1: Approved Commercial National Security Algorithm (CNSA) Suite for DAR .................................... 5
Table 2: Solution Design Summary ............................................................................................................. 16
Table 3: Use Case Summary ........................................................................................................................ 19
Table 4: Requirement Digraphs .................................................................................................................. 27
Table 5: Product Selection Requirements................................................................................................... 28
Table 6: Overall Solution Requirements ..................................................................................................... 30
Table 7: Configuration Requirements for All DAR Components ................................................................. 30
Table 8: SWFDE Component Requirements ............................................................................................... 33
Table 9: FE Component Requirements ....................................................................................................... 34
Table 10: PE Component Requirements ..................................................................................................... 34
Table 11: HWFDE Component Requirements ............................................................................................. 35

v
 Table 12: End User Device Requirements ................................................................................................... 36
Table 13: Configuration Change Detection Requirements ......................................................................... 40
Table 14: Device Management Requirements............................................................................................ 41
Table 15: Auditing Requirements ............................................................................................................... 42
Table 16: Key Management Requirements for All DAR Components ........................................................ 43
Table 17: Supply Chain Risk Management Requirements .......................................................................... 44
Table 18: Use and Handling of Solutions Requirements............................................................................. 45
Table 19: Incident Reporting Requirements ............................................................................................... 48
Table 20: Test Requirements ...................................................................................................................... 52
Table 21: Randomly Generated Minimum Password Length ..................................................................... 65
Table 22: Randomly Generated Minimum Passphrase Length................................................................... 66

vi
 1 INTRODUCTION
The Commercial Solutions for Classified (CSfC) Program within the National Security Agency (NSA) Cyber
Security Directorate (CSD) publishes Capability Packages (CP) to provide architectures and configuration
requirements that empower IA customers to implement secure solutions using independent, layered
Commercial Off-the-Shelf (COTS) products. The CPs are product-neutral and describe system-level
solution frameworks, documenting security and configuration requirements for customers and/or
Integrators. It is recommended that CSfC Trusted Integrators be employed to architect, design,
integrate, test, document, field, and support the solution. The list of CSfC Trusted Integrators can be
found at: https://www.nsa.gov/resources/commercial-solutions-for-classified-program/trusted-
integrators.

This CSfC Data-at-Rest (DAR) CP meets the demand for DAR solutions using the Commercial National
Security Algorithm (CNSA) Suite. These algorithms are used to protect up to top secret data using layers
of COTS products. As defined in Section 4.3.1, the DAR CP version 5.0 enables customers to implement
two independent layers of encryption for the purpose of providing protection for stored information on
the End User Device (EUD) or DAR protected system, while in a powered off or unauthenticated state.
This CP takes lessons learned from proof-of-concept demonstrations that have implemented the CNSA
Suite, modes of operation, standards, and protocols. These demonstrations included a layered use of
COTS products for the protection of classified information.

The DAR CP focuses on the implementation of cryptography to mitigate the risk of unauthenticated
access to classified data when the device is powered off or unauthenticated. This CP does not protect
against malicious code exploits and potential vulnerabilities from updates, operating system (OS)
misconfigurations, or the persistence of remnants of key or plaintext material in volatile memory on the
EUD when powered on and authenticated, as these conditions are outside of the scope for this version
of the CP.

While CSfC encourages industry innovation, trustworthiness of the components is paramount.

Customers and their Integrators are advised that modifying a National Information Assurance
Partnership (NIAP)-validated component in a CSfC solution may invalidate its certification and require a
revalidation process. To avoid delays, customers and Integrators who feel it is necessary to modify a
component should engage the component vendor and consult NIAP through their Assurance Continuity
Process (https://www.niap-ccevs.org/Documents_and_Guidance/ccevs/scheme-pub-6.pdf) to
determine whether such a modification will affect the component’s certification.

In case of a modification to a component, NSA’s CSfC Program Management Office (PMO) requires the
component to successfully complete the NIAP Assurance Maintenance Continuity process.
Modifications that trigger the revalidation process include, but are not limited to: configuring the
component in a manner different from its NIAP-validated configuration, and modifying the Original
Equipment Manufacturers’ code (to include digitally signing the code).

1
 2 PURPOSE AND USE
This CP provides high-level reference designs and corresponding configuration requirements that allow
customers to select COTS products from the CSfC Components List available on the CSfC web page
(https://www.nsa.gov/resources/commercial-solutions-for-classified-program/components-list), for
their DAR solution and then to properly configure those products to achieve a level of assurance
sufficient for protecting classified data while at rest. As described in Section 9, customers must ensure
that the components selected from the CSfC Components List will provide the necessary functionality
for the selected capabilities. To successfully implement a solution based on this CP, all Threshold
Requirements, or the corresponding Objective Requirements applicable to the selected capabilities,
must be implemented, as described in Sections 8 - 12.

Please provide comments on usability, applicability, and/or shortcomings to your NSA/CS Client
Advocate and the DAR Capability Package maintenance team at [email protected]. DAR CP
solutions must also comply with the Committee on National Security Systems (CNSS) policies and
instructions. Any conflicts between CNSS or local policy and this CP should be provided to the DAR CP
Maintenance team.

Additional information about the CSfC process is available on the CSfC web page
(https://www.nsa.gov/resources/commercial-solutions-for-classified-program).

3 LEGAL DISCLAIMER
This CP is provided “as is.” Any express or implied warranties, including but not limited to, the implied
warranties of merchantability and fitness for a particular purpose are disclaimed. In no event must the
United States (U.S.) Government be liable for any direct, indirect, incidental, special, exemplary or
consequential damages (including, but not limited to, procurement of substitute goods or services, loss
of use, data, profits, or business interruption) however caused and on any theory of liability, whether in
contract, strict liability, or tort (including negligence or otherwise) arising in any way out of the use of
this CP, even if advised of the possibility of such damage.

The User of this CP agrees to hold harmless and indemnify the U.S. Government, its agents and
employees from every claim or liability (whether in tort or in contract), including attorney’s fees, court
costs, and expenses, arising in direct consequence of recipient’s use of the item, including, but not
limited to, claims or liabilities made for injury to or death of personnel of user or third parties, damage
to or destruction of property of user or third parties, and infringement or other violations of intellectual
property or technical data rights.

Nothing in this CP is intended to constitute an endorsement, explicit or implied, by the U.S. Government
of any particular manufacturer’s product or service.

4 DATA-AT-REST PROTECTION OVERVIEW

The goal of the DAR solution is to protect classified data when the EUD is powered off or
unauthenticated. Unauthenticated is defined as the EUD state prior to a user presenting and having
their credentials (i.e., password, tokens, etc.) validated by both layers of the DAR solution. The DAR

2
 solution is composed dual encryption layers, an outer and inner layer. The outer layer is considered the
layer that is authenticated to first, while the inner layer is authenticated to secondly. The data owner
determines specific data that must be protected.

In this CP, when the term “EUD” is used, it is referring to any solution that contains two layers of CSfC
DAR protection. Based on the context and wording of some of the CP language, other terms such as
system, DAR solution, or device may be used. These terms can be used interchangeably depending on
context of the customer use case, and the data that the customer is protecting. This definition is further
explained in detail in Section 5.5.

4.1 IMPLEMENTING CSFC IN A HIGH ASSURANCE GOTS ENVIRONMENT

An option available to users of CSfC, is the concept of a blended solution that will use a CSfC solution, in
combination with a High Assurance GOTS solution. For these blended solutions, a High Assurance GOTS
solution can be utilized to replace the entire function of a Capability Package as a whole, but not the
individual layers of a solution or security functions provided by one of the layers. While CSfC utilizes two
layers of encryption, this is not required with High Assurance GOTS, where a single layer of encryption is
sufficient. For example, if desired, a CSfC DAR solution can be employed in an infrastructure where
network High Assurance Internet Protocol Encryptors (HAIPEs) are also being used. The DAR solution is
segmented, and its protection is provided by CSfC, while the protection of the network that the
information transits is provided by a High Assurance GOTS solution. For additional details or questions
about this process, please contact the CSfC PMO office at [email protected].

4.2 RATIONALE FOR LAYERED ENCRYPTION

A single layer of CNSA encryption, properly implemented, is sufficient to protect classified DAR.
However, a CSfC DAR solution uses two layers of CNSA encryption; not because of a deficiency in the
cryptographic algorithms, but to mitigate the risk of a failure in one of the cryptographic components
due to accidental misconfiguration, operator error, or malicious exploitation of an implementation
vulnerability, which results in the exposure of classified information. The use of multiple layers,
implemented with components meeting the CSfC vendor diversity requirements, reduces the likelihood
that a single vulnerability can be exploited to reveal protected information.

If one of the encryption layers is compromised or fails in some way, the second layer still provides the
needed encryption to safeguard the classified data. If both layers are compromised or fail
simultaneously, it is possible the classified data will become readable to a threat actor. The goal of the
DAR solution is to provide redundant protection that either minimizes the possibility of both layers
failing at the same time or requires an adversary to defeat both mechanisms.

4.3 SOLUTION STATES

The DAR solution states are identified and described in further detail in this section. Once a device is
considered classified (i.e., Powered-On with Outer Layer Authenticated State) it will not be considered
unclassified (must still be handled in accordance with the implementing organizations’ Authorizing
Official (AO) policies) until the device is in the powered-off state.

4.3.1 EUD SOLUTION STATES

Powered-Off State:

3
 In a powered-off state, the device is completely off and not in any power saving state. The EUD is
considered unclassified, but must still be handled in accordance with the implementing organizations’
AO policies. This applies to all removable media when unplugged from the host system. If the RMs have
their own power states, the product documentation must be consulted to determine how to
independently switch the product into a powered-off state.

Powered-On and Unauthenticated State:

In a powered-on and unauthenticated state, the EUD is completely on, but the user has not initially
logged into either layer. The EUD is considered unclassified, but must be handled in accordance with
the implementing organizations’ AO policies. This state cannot be entered by logging off after initial
logon. This applies to all removable media when plugged into the host system.

Powered-On with Outer Layer Authenticated State:

In a powered-on state with the outer layer authenticated, the EUD is operational where the user has
authenticated to the outer layer of encryption. The device in this state is considered classified and
should be handled accordingly. This applies to all removable media when plugged into the host system.

Powered-On with Outer and Inner-Layer Authenticated State:

In a powered-on state with the outer and inner-layer authenticated, the EUD is operational when the
user has authenticated to two layers of DAR encryption. The device in this state is considered classified
and should be handled accordingly. This applies to all removable media when plugged into the host
system.

Locked or Logged Out State:

In a locked or logged out state, the device is powered-on but most of the functionality is unavailable for
use. User authentication is required to access functionality. This functions as an access control and may
provide one layer of DAR protection. The device in this state is considered classified and should be
handled accordingly. This applies to all removable media when plugged into the host system.

4.3.2 ENTERPRISE MANAGEMENT (EM) SERVER & MISSION CONTROL ELEMENT (MCE)
SOLUTION STATES
Always on State:

The “always on state” in this section applies to the server that is acting as part of a remote access
architecture or a client-server architecture, controlling the DAR enterprise managed solution. This state
does not apply to the server that is acting as a DAR EUD with two layers of protection. The “EUD
Solution States,” described above in Section 4.3.1, is only applicable to a server, if that server is acting as
the DAR EUD being provisioned with two encryption layers to protect the server’s storage. In this CP, it is
assumed that the EM server, base station, or MCE will be protected within a secured facility, as
prescribed by the AO (i.e., Sensitive Compartmented Information Facility (SCIF), secured room, etc.).
This CP does not provide sufficient mechanisms to protect classified data on the EM server, unless that
server is also treated as a DAR EUD and is protected with two layers of DAR. In an always on state, the
DAR enterprise management server, the base station, or the Mission Control Element (MCE) is always

4
 powered on to keep processes up and running. Additional details about these components and
solutions, can be found in Sections 5.6, 7.3, and 7.4.

4.4 DAR CNSA SUITE

As the portability of EUDs increase, the requirements for when and how classified data is protected also
increases. EUDs can be used in both physically protected and physically unprotected environments.
Solutions using commercial products must protect classified data on the EUD by using two layers of
encryption with the approved CNSA Suite, referenced in Table 1. The solutions presented in this CP
have specific requirements for configuration, product selection, components, provisioning,
authentication, key management, operations, administration, roles, and use and handling.

Table 1: Approved Commercial National Security Algorithm (CNSA) Suite for DAR

CNSA Suite
Security Service Specifications
Standards
Confidentiality AES-256 FIPS PUB 197
(Encryption)
Authentication (Digital Elliptic Curve Digital FIPS PUB 186-4
Signature) Signature Algorithm
over the curve P-384
with SHA-384
RSA 3072 FIPS PUB 186-4
(Minimum)
Integrity (Hashing) SHA-384 FIPS PUB 180-4
Can protect Up to Top Secret ---------

NSA will initiate a transition to quantum resistant algorithms in the not too distant future. NSA
customers using layered commercial solutions to protect classified national security information with a
long intelligence life should begin implementing a layer of quantum resistant protection. Such
protection may be implemented today through the use of large symmetric keys coupled with specific
secure protocol standards. For more information please go to
https://apps.nsa.gov/iaarchive/programs/iad-initiatives/cnsa-suite.cfm.

4.5 AUTHENTICATION
In this CP, each layer is required to have a “known secret” (e.g., PIN, password, or passphrase),
smartcard, or Universal Serial Bus (USB) token to authenticate to each of the two encryption layers. The
permitted factors may differ based on the layer. DAR encryption products must meet requirements for
each of these factors during evaluation against the applicable protection profile (PP). These are
considered primary (validated) authentication factors for that component.

Many products offer alternate authentication mechanisms. When implementing the DAR solution,
these alternate mechanisms may be used only as a secondary (non-validated) authentication factor and

5
 must be paired with a primary authentication factor. Secondary factors may act as an additional access
control or may contribute to the product’s key chain; the product’s protection profile evaluation
guarantees there is no loss in strength when combining keys with potentially weaker sources. A layer
may use any number of authentication factors as long as one is a primary factor listed in that
component’s specific authentication requirement. As an example, layer one may use a known secret
(primary factor) along with a biometric (secondary factor), and layer two may use a smartcard. It is
important to consider the requirements, benefits, and drawbacks associated with different
authentication factors. Some considerations of popular factors are discussed below.

Known secrets are memorized values that can provide a strong authentication value if well chosen (see
Appendix A) and are typically supported by almost all products. They are at risk of being forgotten, as
well as being seen while being keyed into the device. The majority of the time, known secrets will be
weaker than tokens and are at risk of being very weak if not properly chosen.

Smartcard tokens are small integrated circuit devices that can store authentication keys. As long as they
are handled and stored properly, they provide a very strong form of authentication. They provide a
flexible option for authenticating a user to many devices and providing additional security through the
use of a PIN to use the card. Aside from the benefits, cards are susceptible to loss and damage. In
addition, they may also require a separate system for provisioning and recovery.

USB tokens are a simple form of token that provides a very strong form of authentication as long as they
are handled and stored properly. Although very easy to provision, they generally have no additional
security features, unless the USB device itself provides those features. Unfortunately, they are not
permitted in many places.

Biometric technology functions by taking a measurement of an element of the user’s body. Common
examples are fingerprints, iris scans, and facial recognition. This measurement is compared against a
template that is created during provisioning; if the measurement matches the template, the user is
authenticated. The vendor may use this authentication as an access control or may release a key to
contribute to decryption. If a key is used, it will be important to ask how that key is protected and what
authorizes the key’s release, as there are currently no methods being used to derive a biometric
measurement into a key. When using biometrics there may be instances when unauthorized users will
be authenticated to the biometric when they should not; this is called a false acceptance, and is a
condition with which all biometrics have to contend. Customers should obtain vendors’ False
Acceptance Rates (FAR) and determine how comprehensive their testing was to determine that rate.
The other rate to address is the False Rejection Rate (FRR), which is when an authorized user’s
measurements fail to authenticate. This is a usability concern and should also be discussed with the
vendor. The biometric template used to compare measurements is intended to be constructed so that
the user’s measurements are not reversible. If an adversary was able to obtain the template, they
would be unable to reconstruct the user’s fingerprint. However, this is not always the case; templates
are not well standardized and there have been cases of reconstruction. This may be a risk to the privacy
of users. One of the major risks of biometric is spoofing. This involves using other technology to
recreate the user’s measurement. Examples of spoofing include taking photos of the user’s face or
lifting fingerprints. The vendor should explain how they mitigate spoofing, and users should protect the
area being used to authenticate. Many biometrics need a fallback mechanism in case the area being

6
 used to authenticate to the biometric system becomes damaged, such as a finger being cut.
Consideration should be given to what the fallback mechanism is or the consequences if there is none.

Near Field Communication (NFC) is a short range signal. Generally, the devices are placed adjacently or
in contact to exchange information in this method. This signal can vary in how it is used during the
authentication process. There may or may not be an exchange of key material. The details of what is
exchanged needs to be discussed with the vendor. Regardless of what is exchanged, the devices should
be kept apart and treated like a Smartcard or USB token. NFC may not be permitted if the solution must
also comply with other CPs that don’t permit it.

Behavior based authentication covers a wide variety of features. The goal is to determine if an
authorized user has the device, based on whether the device is being used and handled the way the
authorized user normally uses the device. Based on this information the device may release a key,
provide an access control, or allow for a longer time before locking the device. Factors that are taken
into account include, location, connected networks, gyroscope measurements, user interaction, and
other internal sensors.

4.6 CONTINUOUS PHYSICAL CONTROL

Although the DAR solution can protect the confidentiality of data and render the EUD unclassified, it
does not protect the integrity of an EUD outside the control of approved users. It is difficult to examine
and determine whether or not a device has been tampered with; therefore, the EUD must remain in
continuous physical control at all times. The NSA requires that implementing organizations define the
circumstances in which an EUD that is part of the solution is considered outside of the continuous
physical control of authorized users (i.e., "lost"). The AO will define “continuous physical control”, and
this definition should align with the intended mission and threat environment for which the solution will
be deployed. Each organization must also define the circumstances in which an EUD that is a part of its
solution is to be considered recovered back into the continuous physical control of authorized users (i.e.,
"found").

This concept includes mechanisms for the Unattended Operations Use Case (described in Section 7.4).
AO’s should have a variety of mechanisms to ensure control of the EUD is maintained via cameras,
sensors, and other similar means. The exact means are out of scope of the DAR CP, but there should be
a high degree of confidence that the EUD is not susceptible to unauthorized physical access.

This CP requires any lost device, once found, to be rigorously investigated and/or destroyed in order to
mitigate threats to the integrity of the EUD and any connected systems, because upon being found, the
device is considered not secure unless the device meets Lost and Found (LF) requirements that are
indicated as “LF” in the Use Case column in the requirements table. AOs should consult the DAR CP Risk
Assessment (RA) to help make an informed risk decision.

See Appendix F for additional requirements information and some examples of continuous physical
control.

7
 4.7 RED, GRAY, AND BLACK DATA
This CP uses the following terminology to describe the data types that compose a DAR solution. The
terms Red, Gray, and Black identify the number of encryption layers applied to classified data for a
specific EUD state.

Red data is unencrypted classified data being processed by the EUD. After a user successfully
authenticates to the outer and inner layers of DAR encryption, the EUD is in a state of processing Red
data.

Gray data contains classified information that has been encrypted once. After a user successfully
authenticates to the outer layer of DAR encryption, but has not yet authenticated to the inner layer of
encryption, the EUD is in a state of processing Gray data.

Black data contains classified information that has been encrypted twice. An EUD is considered black
when the device is powered off and/or unauthenticated and the stored data has been encrypted with
both the outer and inner layers of DAR encryption.

4.8 CRYPTOGRAPHIC ERASE (CE)

Cryptographic Erase (CE), is a method of sanitization in which an encryption key for the encrypted data
is sanitized, making recovery of the decrypted data infeasible. In this CP, it is used to ensure clean re-
provisioning, as an additional protection triggered by failed authentication, or as an emergency method
of sanitizing the media, in the event proper destruction methods cannot be met (see DAR-EU-2 in Table
12).

4.9 PROVISIONING
Provisioning is the process through which EUDs are initialized before first use. During the provisioning
process, the Security Administrator (SA) loads and configures the DAR components for the EUD.
Provisioning is inherently an out-of-band process requiring physical access to the EUD. The DAR solution
cannot be applied to an EUD that already has data stored on it.

EUD re-provisioning or reuse of DAR components is allowed as long as it is performed in accordance

with this CP. If re-provisioning, the EUD must be at the same or higher classification level of the
previous unencrypted data stored on the approved DAR solution. Prior to re-provisioning an EUD, old
data should be removed via cryptographic erase or media zeroization, media zeroization is the full
overwrite of the disk. Re-provisioning EUD components from any non-CSfC solution is prohibited.

4.10 SECURE FILE DELETION

When deleting files via normal means (i.e., deleting followed by emptying the recycle bin, shift + delete,
etc.) from the computer, there is a possibility for residual data to remain on the underlying storage
media for extended periods of time, recoverable by forensic techniques. While the DAR CP requires
multiple layers of encryption and tries to mitigate user error, it is still possible for the device to be
compromised; in that event, securely deleting files reduces the information available to the adversary.
For these reasons, it is recommended to use applications to securely delete files.

Secure file deletion tools make use of more direct methods to mitigate the risk of data being
recoverable. Since there is not currently a method of validation for providing secure file deletion, here

8
 are some recommendations for features to include when acquiring a secure file deletion product. When
looking for a product to fulfill this purpose, the type of storage media must be considered. There are
currently two primary storage drives used today, hard disk drives (HDD) and solid state drives (SSD).
Flash USB drives fall into the same area as SSDs.

Normally when a file is deleted from a HDD, the reference to that file’s content is removed. The
majority of the data continues to reside on the disk, being treated as free space for new data to use.
This makes the role of a third party product straightforward. It should claim to directly overwrite the file
reference and file data with any value, as long as that value cannot contain sensitive data, such as the
contents of random access memory (RAM). Products may provide options for performing multiple
passes but this is not necessary, as a single pass provides sufficient security. However, if only multiple
passes are supported, they will not cause any harm.

In order to understand the residual risk, it is important to understand the basics of the complications
involved in erasing memory from a SSD. When a user deletes a file, the drive marks that area as free
space, but will not actually overwrite the data. This is for performance reasons, the memory used by the
SSD must be cleared before being written to again, which takes time. The drive works in conjunction
with the operating system to perform this task in the background when the drive does not have more
important tasks. Because of this, it is not possible for a third party file deletion tool to directly overwrite
data. There is also an upside, as regular deletion can eventually result in a direct overwrite, unlike HDDs
where the data can remain for long periods of time. Because of this, third party tools are not necessary,
and files may be deleted via normal means. However, there are other factors that affect when the SSD
drive’s background overwrite process can take place. The restrictions below detail configurations which
a user can exert control. There are other situations where a user cannot exert control, which has the
potential to result in data residing on the SSD for an extended undefined period of time. This is
acceptable since any residual data should be encrypted. If any of the restrictions below apply, some
third party products may be able to overcome them. Otherwise, the product should issue commands
that enable the SSD to clear memory as soon as possible.

 TRIM, the command issued to the SSD to clear space, may not be supported by the operating
system. Most modern operating systems do support this command; check operating system
documentation to ensure support for TRIM.
 The TRIM command may only be supported by the OS if certain file systems are being used.
Check vendor documentation to ensure a compatible file system is used.
 The way this is checked varies between operating systems. Check operating system
documentation on how to verify TRIM is enabled.
 Older SSDs may not support the TRIM command. The majority of modern drives do have
support; check vendor documentation to ensure the device supports this command.
 The operating system may not support TRIM for external drives, USB flash drives, or other
devices connected over USB, PCI E, M.2 and other interfaces. This is a common area where a
third party product may provide additional benefit.
 The operating system may not support TRIM when a Redundant Array of Independent Disks
(RAID) configuration is used.

9
 DAR products that support encrypted volumes may interfere with the TRIM command for data within
the volumes. Some products do enable TRIM to function within the encrypted volumes, check vendor
documentation for verification.

4.11 DAR LOCATION BASED SERVICES

Data-at-Rest protection can include capabilities that restrict access and authorization to a device based
on the EUD’s location, through the use of location aware technology. In this CP, a customer has the
option of employing DAR location aware technology as an added feature to a DAR solution. This does
not replace either of the two mandatory encryption layers required in a CSfC DAR solution, but is only
used as a supplemental means of protection, for defense in depth. Currently, there are no PPs to
validate this type of technology. If customers choose to implement DAR location services, the customer
should consult with the vendor and ask about specific details concerning the mechanisms and
methodologies used in their product.

DAR location services provides for precise geolocation of DAR devices using methods such as, Wi-Fi
Positioning System (WPS), Global Positioning System (GPS), and Radio Frequency Identification (RFID).
WPS achieves precision geolocation using a layered approach and may provide more accurate and
reliable geolocation for DAR systems that are operated within a building, whereas GPS may be effective
in outdoor locations. The precision geolocation of a DAR device configured for a DAR location service is
a precursor to successfully managing a variety of unattended secure DAR storage operations through
the location aware storage domain. These storage operations include:

 Storage function isolation to minimize possibility of interference

 Reliable location determination within wireless environments (i.e., IEEE 802.11ax standards)

 Location driven DAR operation that can only be operated within fixed boundaries

 Secure domain within a building driven by security policy

DAR RFID based mobility services provides an added layer of assurance to DAR mobility by enabling
administrators to track and control the location and movement of DAR devices within the confines of a
building or enclave. This may be achieved by creating policies for individual devices depending on one
or more of the following mobility related parameters:

 Physical location – Maintain continued device location status and confirm that uninterrupted
physical control (as defined by the AO) is in effect.

 Network connectivity – Provide validation of device presence within approved spaces for pre-
boot and decryption.

 Time-based operation –Revoke keys and initiate full-disk erase if the device remains outside of
approved spaces beyond an authorized time period.

 Encryption enforcement –Initiate a power down or a key revocation to enforce DAR encryption
when a device is removed from approved spaces or considered compromised.

10
 When a device exceeds or violates any of the mobility related parameters, the RFID based service
executes appropriate measures to ensure the integrity and security of data stored in the device are
maintained. These measures can include key revocation, crypto erase, device power down, etc.

5 SOLUTION COMPONENTS
This section describes the capabilities of each component. Section 6 describes the possible functional
implementations of each component within the possible Solution Designs and summarizes them in Table
2.

5.1 SOFTWARE FULL DISK ENCRYPTION (SWFDE)

Software Full Disk Encryption (SWFDE), shown in Figure 1, is used to provide one layer (either the inner
or outer layer depending on the solution implemented) of DAR protection. The National Institute of
Standards and Technology (NIST) Special Publication 800-111, Guide to Storage Encryption Technologies
for End User Devices, defines full disk encryption as follows: “Full Disk Encryption (FDE), also known as
whole disk encryption, is the process of encrypting all the data on the drive used to boot a computer,
including the computer’s OS, and permitting access to the data only after successful authentication to
the FDE product.” A user must log into the Pre-Boot Environment (PBE) with valid credentials. Once the
user is authenticated to the PBE, the SWFDE decrypts and boots the OS.

Figure 1: Software Full Disk Encryption

5.2 FILE ENCRYPTION (FE)

File Encryption (FE), shown in Figure 2, is approved to provide the inner layer of DAR protection. FE is
the process of encrypting individual files or sets of files on an EUD and permitting access to the
encrypted data only after proper authentication is provided.

11
 Figure 2: Software File Encryption
FE products currently on the market have a wide range of implementations. It is important for the user
to understand how a specific FE product operates to ensure all classified data on the EUD is encrypted.
There are many events and applications that may write data to the disk. Users should be made aware of
these through user training, unless the FE product can encrypt the data without their intervention.
Some examples of such events include:

1. Applications permitted to run on the EUD should be carefully considered. Applications may
create files (e.g., temporary files) in unprotected locations leaving classified data at risk. If an
application (e.g., file viewer) will be interacting with sensitive data and is not protected by an FE
component, that application must be evaluated against the Application Software Protection
Profile (ASPP) and meet the selection “not store any sensitive data” in FDP_DAR_EXT.1.1.
2. Paging files (e.g., swap files) are created when the system runs out of or becomes low on unused
volatile memory, also known as RAM. When this occurs, the system may write to the non-
volatile memory (e.g., hard disk) for storage. If the product cannot automatically protect this
data, the solution should disable system page files.
3. Systems restore, and other features that allow data to be restored to a previous point in time
create copies of the data. If this is enabled, it may allow an encrypted file to be restored to a
state before it was encrypted. Unless the product accounts for these types of scenarios, these
features should be disabled.
4. Memory dump files may be created when an error occurs. Memory dump files may include
classified data that existed in volatile memory when the crash occurred. Since these files are
created during a system crash, it is likely the product will not be able to properly encrypt them.
Therefore, it is recommended this feature be used with care by individuals who understand
what data will be contained within the file, or the feature should be disabled.

12
 5. Printer spool files are created when a document is sent to print. These are used to hold
documents while they are in queue for printing. If the solution is going to print any classified
information, these files should be protected.
6. Moving or deleting files: users should be informed that moving (cut/paste) a classified file into a
protected area is not sufficient for protecting it. Moving or deleting a file while it is unencrypted
may leave file contents on the disk until it is overwritten by the file system. This should apply to
all file movement for good practice, even though it would not apply in all cases. All files should
be encrypted before being deleted or moved.
FE protects the confidentiality of individual files, folders, or volumes, and may be accomplished in
several ways. The encryption may be performed by an application, platform, or the host OS. Each
encrypted file, folder, or volume will be protected by a File Encryption Key (FEK). The FEK is protected
by the user’s authentication factor, either directly or through one or more Key Encryption Keys (KEKs).

Proper user authentication is required to decrypt the FEK. The FE product will then decrypt files or
folders on an individual basis as they are requested by the user via specific applications. To ensure that
no classified data is left unprotected, the AO must be responsible for providing and enforcing a policy
that mandates automation and user compliance to encrypt all classified data.

5.3 PLATFORM ENCRYPTION (PE)

Platform Encryption (PE), shown in Figure 3, is approved to provide the outer layer of DAR protection.
PE is provided by the OS for platform-wide data encryption, transparently encrypting sensitive user data.
The PE layer requires hardware-backed secure key storage, with the goal of reducing the need for long
and complex passwords. With the exception of the hardware-specific requirements and which layer
they can be used for (PE protects the outer layer while FE protects the inner layer), there is little
distinction between PE and FE implementations. In all other respects, the two component
implementations are virtually identical; they both provide volume and FE capabilities.

13
 Figure 3: Platform Encryption
The PE component relies on the EUD to implement the requirements specified in the Mobile Device
Fundamentals (MDF) PP, along with the CSfC selected requirements. Items that meet the NIST
requirements for PE solutions are located in the CSfC Components List under “End User Device/Mobile
Platform.”

5.4 HARDWARE FULL DISK ENCRYPTION (HWFDE)

Hardware Full Disk Encryption (HWFDE), shown in Figure 4, can be used to provide the inner or outer
layer of DAR protection. HWFDE is commonly implemented via a Self-Encrypting Drive (SED). The SED
can be a standard hard drive or a solid state drive.

14
 Figure 4: Hardware Full Disk Encryption
A SED contains hardware built into the drive controller chip that automatically encrypts all data written
to the drive and decrypts all data read from the drive. The encryption and decryption is done
transparently to the user.

In some cases, the HWFDE solution will require multiple components to create an FDE solution. Some
SEDs require a product from an Independent Software Vendor (ISV) to function; this ISV commonly fills
the role of collecting initial authentication and passing it to the SED. It is essential that both parts of the
solution are chosen from the CSfC Components List.

The Authentication Key (AK) used in HWFDEs to encrypt or decrypt data is called the Data Encryption
Key (DEK), which is protected by a chain of keys originating from the authentication factor.

A user must log into the PBE, provided by the SED or an ISV, with valid credentials. Once the user is
authenticated to the PBE, the HWFDE decrypts and boots the operating system.

When discussing the use of ISVs and SEDs, the relevant information is sometimes referred to as FDE
Authorization Acquisition (AA) & Encryption Engine (EE) breakout information.

5.5 END USER DEVICE (EUD)

The EUD is either: a personal computer (e.g., desktop, laptop); consumer device (e.g., smart phone,
tablet); removable media (e.g., USB, CD); or a server (e.g., storage area network, network attached
storage, shared drives, external storage). It is important to keep the security of different power states in
mind when using these devices, referenced in Section 4.3.1. An EUD may operate within a secure

15
 physical environment, outside of a secure physical environment, or both inside and outside of a secure
physical environment as approved by the AO.

The drives that make up a Storage Area Network (SAN) or a Network Attached Storage (NAS) can be
protected via the solutions presented in this CP, but that protection is provided only when the system is
powered off. For powered on scenarios, consult the Mobile Access or Campus Wireless Local Area
Network (WLAN) CPs on the CSfC web site.

5.6 DAR ENTERPRISE SERVER (ES) AND MISSION CONTROL ELEMENTS

The DAR Enterprise Server (ES) and Mission Control Elements are assumed to be in protected
environments unless they are being treated as an EUD. Reference to the DAR ES should be distinct and
separate from the DAR EUD server. In this CP, the DAR Enterprise Server may be referred to as the DAR
ES or DAR EM Server, different from the DAR “EUD”. The DAR ES provides functions such as account
recovery, remote erase, network required authentication, and other similar functions for many endpoint
EUDs. The server may integrate with services provided by the operating system to manage accounts.

The MCE acts as a system or set of systems that manage or access remote unattended EUDs, such as
drones or unattended sites. In this situation, it is important to have mechanisms in place to ensure
continuous physical control of the EUD is maintained, as described in Section 4.6. This solution does not
need an ES, just remote access over a secure channel as described in the requirement.

6 SOLUTION DESIGNS
The CP provides the multiple solution designs listed in Figure 2. The designs describe solutions meeting
a wide variety of requirements to protect classified DAR.

The “SF” design consists of SWFDE and FE. The SF architecture is typically intended for EUDs such as
servers, desktops, laptops, and tablets.

The “PF” design consists of PE and FE. The PF architecture is typically intended for EUDs such as laptops,
tablets, and smart phones.

The “HF” design consists of HWFDE and FE. The HF architecture is typically intended for EUDs such as
servers, desktops, laptops, and tablets.

The “HS” design consists of HWFDE and SWFDE. The HS architecture is typically intended for EUDs such
as servers, desktops, laptops, and tablets.

The “HH” solution design consists of two independent HWFDE layers. The HH architecture is typically
intended for EUDs such as servers, desktops, laptops, and tablets.

Table 2: Solution Design Summary

Solution Design Designator Description
SWFDE/FE SF DAR solution design that uses FE as the inner layer
and SWFDE as the outer layer, as described in
Section 6.1.
PE/FE PF DAR solution design that uses FE as the inner layer

16
 and PE as the outer layer, as described in Section
6.2
HWFDE/FE HF DAR solution design that uses FE as the inner layer
and HWFDE as the outer layer, as described in
Section 6.3
HWFDE/SWFDE HS DAR solution design that uses SWFDE as the inner
layer and HWFDE as the outer layer, as described in
Section 6.4
HWFDE/HWFDE HH DAR solution design that uses HWFDE as the inner
layer and HWFDE as the outer layer, as described in
Section 6.5

Solution owners are encouraged to implement the Objective version of a requirement, but in cases
where this is not feasible, solution owners must implement the Threshold version of the requirement
instead.

6.1 SWFDE/FE (SF) SOLUTION DESIGN

The SWFDE/FE (SF) solution design requires SWFDE and file/folder/volume encryption. In the SF
solution design, SWFDE will be used to provide DAR protection for the outer layer, and FE will be used to
provide DAR protection for the inner layer. The SF DAR solution uses a password, passphrase,
smartcard, or USB token to provide access to classified data. Once a user inputs the correct password,
passphrase, smartcard token, or USB token, the system boots the operating system. Next, the user
authenticates to the FE, which in turn decrypts the user’s classified files.

Each layer of encryption in the SF DAR solution may use similar authentication mechanism types (e.g.,
passwords, passphrases, or tokens) but requires a unique authentication credential for each layer. For
the first layer of encryption, the user will authorize to the PBE provided by the SWFDE. For the second
layer, the user will use their OS login credentials, application credentials, or file-specific credentials to
authenticate to the FE.

6.2 PE/FE (PF) SOLUTION DESIGN

The PE/FE (PF) solution design permits platform encryption, which allows for a device to perform DAR
encryption via various implementations, all of which encrypt all sensitive data transparently. In the PF
solution design, PE will be used to provide DAR protection for the outer layer, and FE will be used to
provide DAR protection for the inner layer. The PF solution uses passwords to provide access to
classified data. Once a user inputs the correct password, the platform is decrypted, which then provides
access to user data. Next, the user authenticates to the FE, which in turn decrypts the user’s classified
files.

Each layer of encryption in the PF DAR solution may use similar authentication mechanism types (e.g.,
passwords) but requires a unique authorization credential for each layer. For the first layer of
encryption, the user will authorize to the device’s encryption. For the second layer, the user will use
their application credentials or file-specific credentials to authenticate to the FE.

17
 6.3 HWFDE/FE (HF) SOLUTION DESIGN
The HWFDE/FE (HF) solution design requires hardware full disk encryption and file/folder/volume
encryption. In the HF solution design, HWFDE will be used to provide DAR protection for the outer
layer, and FE will be used to provide DAR protection for the inner layer. The HF DAR solution uses a
password, passphrase, smartcard, or USB token to provide access to classified data. Once a user inputs
the correct password, passphrase, smartcard, or USB token, the system boots the operating system.
Next, the user authenticates to the FE, which in turn decrypts the user’s classified file.

Each layer of encryption in the HF DAR solution may use similar authentication mechanism types (e.g.,
passwords, passphrases, smartcard, or USB token) but requires a unique authentication credential for
each layer. For the first layer of encryption the user will authenticate to the PBE provided by the
HWFDE. For the second layer the user will use their OS login credentials, application credentials, or file-
specific credentials to authenticate to the FE.

6.4 HWFDE/SWFDE (HS) SOLUTION DESIGN

The HWFDE/SWFDE (HS) solution design approach requires hardware full disk encryption and software
full disk encryption. In the HS solution design, HWFDE will be used to provide DAR protection for the
outer layer, and SWFDE will be used to provide DAR protection for the inner layer. The HS DAR solution
uses a password, passphrase, smartcard or USB token to provide access to classified data. Once a user
inputs the correct password, passphrase, smartcard or USB token value to the outer layer HWFDE, the
inner layer SWFDE prompts the user to enter a password, passphrase, smartcard, or USB token in the
pre-boot environment. Once the user authenticates to the SWFDE, the OS is loaded, and the user has
access to data on the drive.

Each layer of encryption in the HS DAR solution may use similar authentication mechanism types (e.g.,
passwords, passphrases, or tokens) but requires a unique authentication credential for each layer. For
each layer of encryption the user will authenticate to a PBE provided by the HWFDE and SWFDE,
respectively.

6.5 HWFDE/HWFDE (HH) SOLUTION DESIGN

The HWFDE/HWFDE (HH) solution design requires hardware full disk encryption. In the HH solution
design, HWFDE will be used to provide DAR protection for both the inner and outer layers. The HWFDEs
used to provide DAR protection for both the inner and outer layers must meet DAR-PS-3 vendor
diversity requirements. The HWFDE DAR solution uses a password, passphrase, smartcard, or USB token
to provide access to classified data. Once a user inputs the correct password, passphrase, smartcard, or
USB token value of the outer layer HWFDE, and then the inner layer HWFDE, the operating system is
loaded and the user has access to the data on the drive. This solution design is not common, and is
specifically built for the purpose of providing dual hardware layers. An example of this solution design
could be a self-encrypting drive paired with an inline encryptor.

Each layer of encryption in the HH DAR solution design may use similar authentication mechanism types
(e.g. password, passphrases, or tokens) but requires a unique authentication credential for each layer.
For each layer of encryption, the user will authenticate to a PBE, provided by each HWFDE layer.

18
 7 DAR USE CASES
This CP provides multiple use cases that can be leveraged using a combination of the five solution
designs. When a specific use case is followed, the customer must implement all Threshold
requirements, where the applicable use case is listed in the “Use Case” column of requirements tables,
as well as the applicable Solution Design, listed in the “Solution Design” column. For multiple use cases,
separate registrations must be submitted with applicable requirements for each use case. These use
cases are listed and described in Table 3.

The “LF” Use Case provides extra protections to permit occasional brief events, where continuous
physical control of the solution is absent due to the EUD being considered lost, thereby, requiring
specific Lost and Found requirements. This use case allows a device to be used when the lost EUD is
found.

The “RM” Use Case is employed on removable media such as USB drives, microSD cards, and removable
drives for the purpose of secondary storage, or to physically move data to and from systems.

The “UO” Use Case is employed when DAR systems and devices are managed remotely, such as a one-
to-one relationship via an NSA approved Data-In-Transit (DIT) communication channel. These systems or
devices may be unmanned and/or unattended, but enforce protections that are considered to be in
continuous physical control, as defined by the AO.

The “EM” Use Case is employed in an enterprise environment for managing multiple devices from one
centralized management server, then pushed down to individual client devices. Solution components
are managed by the DAR ES, through a client-server architecture.

The "GA" Use Case is a generally applicable use case that can be largely applied to a requirement. When
listed as “GA”, the requirement is applied in a standard standalone use case. GA use case is for users not
implementing one of the other specific use cases.

Table 3: Use Case Summary

Use Case Designator Description
Lost and Found LF DAR use case that implements HS, HF, HH, and PF when the
device or system is out of continuous physical control, as defined
by the AO. Described in Section 7.1
Removable RM DAR use case that implements the SF, HF, HH, or HS solution
Media designs as described in Section 7.2.
Enterprise EM DAR use case that implements enterprise managed solutions to
Management manage multiple clients, implemented through the SF, HF, HH,
and HS solution designs, as described in Section 7.3.
Unattended UO DAR use case for managing unattended or remote managed DAR
Operations solutions and systems that implements HS, HF, HH, or SF, as
described in Section 7.4.
Generally GA DAR use case that is generally applicable to a standalone use case
Applicable and corresponding solution design.

19
 7.1 LOST AND FOUND (LF) USE CASE
The Lost and Found (LF) Use Case is when a user, intentionally or unintentionally, temporarily loses
control of a device (as defined by the AO) and plans to continue using it after it is recovered. This use
case adjusts the continuous physical control requirements from Section 4.6 and permits the device to be
used after it is found; however if the device is suspected to have been tampered with, it must be
rigorously investigated and/or destroyed. Note that as of DAR v 5.0, this use case no longer has its own
section of requirements, instead it leverages the use case column in the requirements tables.

This use case is intended to cover situations including but not limited to: devices left in vehicles or
devices forgotten in hotels for short periods of time, going through customs when traveling, and similar
events. These requirements lower the risk of using devices that have been in such conditions, but they
do not eliminate the risk. With this in mind, AOs should consider additional local policy to reduce the
situations where devices may be vulnerable to tampering.

This use case also contains a requirement to personalize the EUD. The intent of the personalization
requirement is to ensure that if an adversary removed the EUD and replaced it with another EUD of the
same make and model, it would be noticed by the end user. Personalization includes: adding stickers,
changing the screen’s background, etc. The administrator may also change settings to personalize the
devices for subsets of users, such as a login screen wallpaper. None of these changes should undermine
any security features of the device or other relevant security policy (i.e., requiring the device to be
rooted).

All of the requirements indicating “LF” in the “Use Case” column of the requirements table must be met
in order to implement the Lost and Found use case. This is a high risk use case and requires a number of
additional requirements to lower the risk. Note that the Lost and Found use case is optional. If it is not
implemented, then the device cannot be reused if it is lost. The SF solution is not allowed for the Lost
and Found use case. As explained in Section 7.2, the LF use case is also prohibited when using
removable media for DAR protection.

7.2 REMOVABLE MEDIA (RM) USE CASE

The Removable Media (RM) use case shown in Figure 5, depicts two layers of encryption employed on
the removable media device/form factor. This use case allows customers to use an external storage
device between different systems to protect DAR and has different password requirements. In the RM
use case, DAR protection is required for the outer layer and the inner layer, provided through the SF, HF,
HH, or HS solution designs. When using the RM use case, choose from the SF, HF, HH, or HS solution
designs. Requirements of the SF, HF, HH, or HS solution design should be followed with this use case.
For example, if the HF solution design is chosen, both HF and RM designated requirements are
applicable.

In this CP, removable media is defined as device(s) which have the primary purpose of providing external
storage of data protected by DAR through implementing two layers of encryption. Removable media
can include: a USB drive, a CD, a microSD card, or a removable drive. Removable media does not include
other portable computing devices such as smartphones and tablets. This use case allows customers to
transfer data using an external storage device between different systems or expand the storage of a
single system. For example, this use case can be used to transport data via a removable media device

20
 between secured facilities, using a DAR CP compliant solution with appropriate CP components to
enable decryption of the RM. This requires using two approved layers of encryption on the RM device
that is provisioned within a secured facility, then transporting the RM under continuous physical control
to access data on another secure workstation or device. If a solution includes both DAR host machines
and a separate DAR RM device, the customer must submit separate registrations.

Figure 5: Removable Media Use Case

The PF solution design cannot be employed on removable media because there are several
incompatibilities between their requirements. The PE layer is used only as the outer layer and requires
hardware-backed secure key storage, with the goal of reducing the need for long and complex
passwords. Each layer of encryption in the PF DAR solution may use similar authentication mechanism
types (e.g., passwords), but requires a unique authorization credential for each layer.

The RM use case only protects endpoints as stated in this CP or in a secured facility. The LF use case is
prohibited when using removable media for DAR protection. If the removable media is lost, and out of
continuous physical control, users must report it to their Information Systems Security Officer (ISSO) or
chain of command, as defined by the AO. The removable media is considered compromised once lost
and cannot be re-used if later found. Lost and Found requirements do not guarantee or protect the
integrity of the removable media once lost and out of continuous physical control.

7.3 ENTERPRISE MANAGEMENT (EM) USE CASE

The Enterprise Management (EM) Use Case shown in Figure 6, depicts a client-server architecture for
managing a DAR enterprise solution. The figure shows the DAR EM use case using a CSfC approved DIT
CP as the secure communication channel. The figure assumes that the DAR client has already been
initially provisioned by an administrator. Figure 6 is one example of how the EM use case could be

21
 depicted; however, there may be other cases where the illustration would be different, such as the
order of operations specific to when the user logs on versus when the DIT tunnels are established, or
several other factors that are specific to the customer implementation, further explained below.

Authentication for this use case may occur in various ways. Connecting to the network to complete the
authentication may or may not be required. One or both layers may have a DAR ES managing them.
The DIT solution may be CSfC or an approved High Assurance GOTS solution. If both DAR layers are
enterprise managed, both servers may either exist on the Red Network or one will exist on the Gray/Red
boundary while the other exists in the Red. If network access is not required for DAR authentication, or
only for the inner DAR layer, allowing the OS to boot before requiring authentication to network, then
connection and access to the EM server can be established as normal, as described in the relevant DIT
CPs. If an enterprise managed layer requires authentication with network access before the OS can
boot, then the tunnel must be established first by a pre-boot capability of the product, a network
device, an approved High Assurance GOTS solution, or endpoint virtualization.

In order for the client to communicate with the server, and the server communicate with the client,
there must be a secure communication link to and from the front and back end. This must be done
through use of one of the DIT CPs, or an approved High Assurance GOTS link. Each implementation is
required to use and comply with the latest version of the Mobile Access (MA) CP, Multi-Site Connectivity
(MSC) CP, or Campus Wireless Local Area Network CP for applicable network and configuration
requirements for establishing and setting up a secure connection that will allow the client and server to
communicate. When implementing the EM use case, customers must have an existing approved DIT CP
registration with the CSfC PMO, or submit a new DIT CP registration for approval.

The DAR EM server can perform, but is not limited to: sending keys, pushing updates to the EUD,
pushing policy and configuration changes to the EUD, and so on. The EM use case allows customers to
transfer the administrative overhead to one centralized management server, enforcing policies and
configuration changes that are pushed to individual DAR client devices.

For specific details on key management of the EM solution, such as how keys will be transmitted,
received, revoked, etc., refer to the requirements and details specified in the products’ PP. Vendors are
required to meet and comply with key management requirements found in the Protection Profiles. For
additional details, please refer to the “Protection Profile Module for File Encryption Enterprise
Management” and the “collaborative Protection Profile Module for Full Drive Encryption – Enterprise
Management.” There may be additional standards leveraged for the management of keys beyond the
security requirements defined by the PPs, such as Key Management Interoperability Protocol (KMIP);
this is expected and per component guidance, should followed for proper setup.

22
 Figure 6: Enterprise Management Use Case

7.3.1 ENTERPRISE MANAGEMENT VIA MA CP, MSC CP, OR CAMPUS WLAN CP

For implementation of a DAR EM solution via a CSfC DIT CP, the user must reference the latest DIT CPs
to access the appropriate network diagrams, and requirements for setting up a secure channel for the
DAR EM link from the client to server. As mentioned, the solution must be registered and approved by
the CSfC PMO in order to use, in combination with the DAR EM use case.

The customer has multiple options for setting up the DAR EM solution on the Red Network. In cases
where the EM servers for both layers are at the solution boundary, they may be contained in the same
physical box and virtually separated. The solution boundaries are described in the corresponding CPs;
the terms “after” and “before” are in reference to moving towards the Red Network.

For integration with the MA CP, there are three implementations:

MA Option 1: The location of both the EM servers is after the solution boundary of the inner layer.
Integration with the MA CP will only provide limited support for EM functionality, as MA trusted
channels must be connected before additional EM functionality is provided, which requires a network
connection. This may render certain DAR EM products incompatible.

MA Option 2: The location of one of the EM servers is after the solution boundary of the inner layer. The
location of the second EM server is before the solution boundary of the inner layer, with only the outer
layer established. This EM server is its own Red enclave, with the second layer of protection provided by
the trusted channel established by the ES itself. This setup will also only provide limited support for EM
functionality, as one of the ES’s will require both MA trusted channels to be connected before additional

23
 EM functionality is provided, which requires a network connection. This may render certain DAR EM
products incompatible.

MA Option 3: consists of enterprise managing one layer, either: after the Inner Solution boundary or
before it in a separate red enclave, and the second layer managed locally.

For integration with the MA CP where one of the servers is before the Inner Solution boundary, the
server must follow guidance found in the Mobile Access CP, referencing the protection of Inner TLS-
Protected Servers and Clients. With this option, the DAR ES(s) must be placed between the Gray
Firewall and Inner Firewall. As referenced in MA CP, Inner TLS-Protected servers must be managed from
the Red administration workstation. DAR ES(s) products listed on the CSfC Component’s List provide
communication channels through secure protocols (e.g., TLS). Product specific functionality and
assurance requirements for the DAR Enterprise Management server can be found in the applicable PP,
which CSfC component vendors are required to meet to be on the list.

For integration with the WLAN CP, the DAR ES must be placed after the Inner Solution boundary.
Integration with the WLAN CP will only provide limited support for EM functionality, as WLAN trusted
channels must be connected before additional EM functionality is provided, which requires a network
connection. This may render certain DAR EM products incompatible.

For integration with the MSC CP, the DAR ES must be placed after the Inner Solution boundary. As
network devices establish the trusted channels and not the client, full EM functionality is available.

For further details as it pertains to configurations, placement, and requirements for protection of DAR
ES servers on the Red Network or on the Inner Solution boundary, within a CP architecture, please
reference the applicable CP sections and requirements.

7.3.2 ENTERPRISE MANAGEMENT VIA HIGH ASSURANCE GOTS SOLUTION

For implementation of a DAR EM solution via a High Assurance GOTS solution, the High Assurance GOTS
link will serve as the approved secure channel, therefore replacing the DIT two tunnel requirement.
Reference Section 4.1 for Implementing a CSfC in a High Assurance GOTS Environment. The AO will be
responsible to ensure all CSfC transmitted data is appropriately protected by the High Assurance GOTS
link. As with the MSC CP, full EM functionality should be available.

7.3.3 ENTERPRISE MANAGEMENT KEY RECOVERY

EM products may provide support for recovery of credentials; these features may only be used if
included in the product’s evaluations per DAR-CR-10. Two general methods may be supported by DAR
EM products, they are challenge response and PIN recovery.

Challenge response operates by providing some known information to be verified by the EM server, at
which point the server returns a value to allow decryption. If the value is generated by the product at
provisioning, it must be stored securely. The product may prompt the user to provide the initial value; if
so, it must be generated according to the password rules and then stored securely. It may be generated
on the server for non-electronic distribution of recovery; if this is the case, a method of verifying the
user must be established. Any delivery of recovery credentials must be performed over a secure
channel.

24
 The second method is PIN recovery. In this case, the recovery PIN will be populated on the server for
each endpoint. This method will require a means of verifying the user, and a method for the delivery of
recovery credentials, which must be performed over a secure channel.

7.4 UNATTENDED OPERATIONS (UO) USE CASE

The Unattended Operations (UO) Use Case shown in Figure 7 is intended for customers operating DAR
solutions that may be unattended and remotely managed, mostly represented as a one-to-one
relationship. This use case differs from the EM use case, in that EM is intended for managing many
devices from a central management server, represented in more of a corporate enterprise environment.
Figure 7 depicts an MCE managing a dual CSfC DAR solution over a validated High Assurance GOTS link
or through an approved CSfC DIT solution. This use case allows customers to operate DAR solutions that
in nature, are more uncommon and considered unique scenarios.

In the UO use case, continuous physical control is defined by the AO, as an acceptable definition that
ensures adequate protection of the device(s) and/or system(s), and data residing there. Methods should
be defined to ensure the device is protected from unauthorized access and ensuring mechanisms that
will put the solution into a secure power state if such unauthorized access is detected.

The UO use case can be defined, but not limited to, an EUD in an unattended environment managed by
a MCE or other secure operational capability. This use case is managed over a remote connection, but
may be accessed locally to perform various functions. If remote power up or power down is required, a
High Assurance GOTS link may be required. The authentication process must follow the methods
defined in the requirements and should be performed via the MCE. If direct authentication of the EUD
is required, it must be within a secure location and physical token must be removed afterwards. UO
examples include, but are not limited to protection of data centers, overrun scenarios, and unmanned
vehicles or systems (e.g., UAV, UUV). The dual DAR solution is managed by an MCE, base station, or
similar solution.

When using the UO use case, some form of anti-tamper and detection capability (passive or active) is
required that enables the detection of possible adversarial compromise when the solution is remotely
managed without direct physical presence. These methods can include measures for monitoring and
detecting, such as cameras, sensors, etc.

25
 Figure 7: Unattended Operations Use Case

8 CONFIGURATION REQUIREMENTS
Sections 8 through 12 specify requirements for implementations of the five solutions, and five use cases
compliant with this CP. Only one use case and one design may be selected, with the exception that EM
may be paired with LF. The tables of requirements in the following sections have a column that specifies
which solution designs and use cases the requirement applies to, and uses the following nomenclature:

 SF design: DAR solution components include SWFDE and FE.

 PF design: DAR solution components include PE and FE.

 HF design: DAR solution components include HWFDE and FE.

 HS design: DAR solution components include HWFDE and SWFDE.

 HH design: DAR solution components include HWFDE and HWFDE.

 LF use case: DAR solution designs include PF, HF, HH, or HS.

 RM use case: DAR solution designs include SF, HF, HH, or HS.

 UO use case: DAR solution designs include SF, HF, HH or, HS.

 EM use case: DAR solution designs include SF, HF, HH, or HS.

26
  GA use case: DAR solution design include SF, PF, HF, HH, and HS.

The CP includes two categories of requirements:

 An Objective (O) requirement specifies a feature or function that is desired or expected but may
not currently be available. Organizations should implement objective requirements in lieu of
corresponding Threshold requirements where feasible.

 A Threshold (T) requirement specifies a minimum acceptable feature or function that still
provides the mandated capabilities if the corresponding objective requirement cannot
reasonably be met (i.e., due to system maturity). A solution implementation must satisfy all
applicable Threshold requirements, or their corresponding Objective requirements, in order to
comply with this CP.

In many cases, the Threshold requirement also serves as the Objective requirement (T=O). In some
cases, multiple versions of a requirement may exist in this CP. Such alternative versions of a
requirement are designated as being either a Threshold requirement or an Objective requirement.
Where both a Threshold requirement and a related Objective requirement exist, the Objective
requirement improves upon the Threshold requirement and may replace the Threshold requirement in
future versions of this CP. Objective requirements without corresponding Threshold requirements are
marked as “Optional” in the “Alternative” column, but improve upon the overall security of the solution
and should be implemented where feasible.

In order to comply with this CP, a solution must, at minimum, implement all Threshold requirements
associated with each of the solution designs and use cases it supports and should implement the
Objective requirements associated with those solution designs and use cases where feasible. For
example, a DAR solution utilizing a SWFDE and FE must implement only those Threshold requirements
applicable to the SF design. Additionally, the customer must implement Threshold requirements
applicable to the chosen DAR solution use case (e.g., RM, UO, EM, LF, or GA).

The customer may treat the device as classified; however, if they do so, they must adhere to the policies
and requirements for classified devices (note that those requirements exceed the requirements
contained within the DAR CP).

Each requirement defined in this CP has a unique identifier digraph that groups related requirements
together (e.g., KM), and a sequence number (e.g., 2). Table 4 lists the digraphs used to group together
related requirements, and identifies where they can be found in the following sections.

Table 4: Requirement Digraphs

Digraph Description Section(s) Table(s)
PS Product Selection Requirements Section 9 Table 5
SR Overall Solution Requirements Section 10.1 Table 6
CR Configuration Requirements for All DAR Section 10.2 Table 7
Components

27
 Digraph Description Section(s) Table(s)
SW SWFDE Component Requirements Section 10.3 Table 8
FE FE Component Requirements Section 10.4 Table 9
PE PE Component Requirements Section 10.5 Table 10
HW HWFDE Component Requirements Section 10.6 Table 11
EU EUD Requirements Section 10.7 Table 12
CM Configuration Change Detection Requirements Section 10.8 Table 13
DM Device Management Requirements Section 10.9 Table 14
AU Auditing Requirements Section 10.10 Table 15
KM Key Management Requirements for All DAR Section 10.11 Table 16
Components
SC Supply Chain Risk Management Requirements Section 10.12 Table 17
GD Use and Handling of Solution Requirements Section 11.1 Table 18
RP Incident Reporting Requirements Section 11.2 Table 19
TR Testing Requirements Section 13.1 Table 20

9 REQUIREMENTS FOR SELECTING COMPONENTS

In this section, a series of requirements are provided for maximizing the independence of components
within the solution. This will increase the level of effort required to compromise this solution.

Table 5: Product Selection Requirements

Solution Use Threshold
Req # Requirement Description Alternative
Designs Case /Objective
DAR-PS-1 The products used for the FE layer HF, SF, EM, T=O
must be chosen from the list of FE PF GA, LF,
products on the CSfC Components RM, UO
List.
DAR-PS-2 The products used for the SWFDE HS, SF EM, T=O
layer must be chosen from the list of GA, LF,
SWFDEs on the CSfC Components List. RM, UO

28
 Solution Use Threshold
Req # Requirement Description Alternative
Designs Case /Objective
DAR-PS-3 The Inner and Outer DAR layers must HF, HS, EM, T=O
either: SF, PF, GA, LF,
 Come from different HH RM, UO
manufacturers, where neither
manufacturer is a subsidiary
of the other; or
 Be different products from
the same manufacturer,
where NSA has determined
that the products meet the
CSfC Program’s criteria for
implementation
independence.
DAR-PS-4 (Moved to DAR-SC-2)
DAR-PS-5 The cryptographic libraries used by HF, HS, EM, O Optional
the Inner and Outer DAR layers must SF, PF, GA, LF,
be independently developed and HH RM, UO
implemented.
DAR-PS-6 The products used for the PE layer PF GA, LF T=O
must be chosen from the list of PE
products on the CSfC Components List
under the Mobile Platform section.
DAR-PS-7 The products used for the HWFDE HF, HS, EM, T=O
layer must be chosen from the list of HH GA, LF,
HWFDEs on the CSfC Components List. RM, UO
DAR-PS-8 The Operating System used must be HF, HS, EM, O Optional
approved by the General Purpose OS SF, HH GA, LF,
Protection Profile (OS PP). UO
DAR-PS-9 The products used for the Enterprise HF, HS, EM T=O
Management Server must be chosen HH, SF
from the list of DAR Enterprise
Management Servers on the CSfC
Components List.

10 CONFIGURATION
Once the products for the solution are selected, the next step is setting up the components and
configuring them in a secure manner. This section consists of generic guidance for how to configure the
components for a DAR solution.

29
 10.1 OVERALL SOLUTION REQUIREMENTS
Table 6: Overall Solution Requirements
Solution Use Case Threshold/
Req # Requirement Description Alternative
Designs Objective
DAR-SR-1 Default accounts, passwords, SF, PF, HF, EM, GA, LF, T=O
community strings, and other default HS, HH RM, UO
access control mechanisms for all
components must be changed or
removed.
DAR-SR-2 The DAR solution must be properly SF, PF, HF, EM, GA, LF, T=O
configured according to local policy and HS, HH RM, UO
U.S. Government guidance (e.g., NSA
guidelines). In the event of conflict
between the requirements in this CP
and local policy, the CSfC PMO must be
contacted.
DAR-SR-3 Each DAR component must have a SF, PF, HF, EM, GA, LF, O Optional
unique account for each user. HS, HH RM, UO
DAR-SR-4 All EUDs must remain in continuous HF, HS, SF, EM, GA, T=O
physical control at all times, as defined PF, HH RM, UO
by the AO.
DAR-SR-5 The AO must provide guidance when CE HF, HS, PF, EM, GA, LF, O Optional
should be implemented. SF, HH RM, UO
DAR-SR-6 The AO must provide procedures for HF, HS, PF, EM, GA, LF, O Optional
performing CE. SF, HH RM, UO
DAR-SR-7 At least one layer must use a trusted HF, HS, SF, EM, GA, UO O Optional
platform module for cryptographic key HH
storage.
DAR-SR-8 (Withdrawn)
DAR-SR-9 At least one layer must use a trusted HF, HS, SF, LF T=O
platform module for cryptographic key HH
storage.

10.2 CONFIGURATION REQUIREMENTS FOR ALL DAR COMPONENTS

Table 7: Configuration Requirements for All DAR Components
Solution Use Case Threshold/
Req # Requirement Description Alternative
Designs Objective
DAR-CR-1 Default encryption keys must be SF, PF, HF, EM, GA, LF, T=O
changed. HS, HH RM, UO

30
 Solution Use Case Threshold/
Req # Requirement Description Alternative
Designs Objective
DAR-CR-2 Primary user authentication credential SF, PF, HF, EM, GA, LF, T=O
values for each DAR layer mechanism HS, HH RM, UO
type must be unique (e.g., the
password for the 1st layer will not be
the same as the password for the 2nd
layer).
DAR-CR-3 DAR components must use algorithms SF, PF, HF, EM, GA, LF, T=O
for encryption selected from Table 1. HS, HH RM, UO
DAR-CR-4 Each DAR component must prevent SF, PF, HF, EM, GA, LF, O Optional
further authentication attempts after a HS, HH RM, UO
number of failed attempts defined by
the AO.
DAR-CR-5 Each DAR layer must perform a CE after SF, PF, HF, EM, GA, LF, O Optional
a number of consecutive failed logon HS, HH RM, UO
attempts as defined by the AO.
DAR-CR-6 Each DAR component must generate its SF, PF, HF, EM, GA, LF, T=O
own symmetric encryption keys on the HS, HH RM, UO
EUD or receive keys generated by the
Enterprise Management server.
DAR-CR-7 Each DAR component must permit only SF, HF, HS, GA, LF, RM, O Optional
an administrator to disable or alter its PF, HH UO
security functions.
DAR-CR-8 All EUDs must have DAR protections SF, PF, HF, EM, GA, LF, T=O
enabled at all times after provisioning. HS, HH RM, UO
DAR-CR-9 All EUDs must encrypt all classified SF, PF, HF, EM, GA, LF, T=O
data. (Refer to Section 5.2 for HS, HH RM, UO
additional information on FE.)
DAR-CR-10 All components must be implemented SF, PF, HF, EM, GA, LF, T=O
(configured) using only their NIAP- HS, HH RM, UO
approved configuration settings. Users
may change settings that are not part
of NIAP evaluation.
DAR-CR-11 Users must be restricted to designated SF, HF EM, GA, LF, T=O
user folders. RM, UO
DAR-CR-12 For use in high threat environments (as HF, HS, SF, EM, GA, UO T=O
defined by the AO) the two layers of HH
DAR must use different primary
authentication factors (i.e., Both layers
cannot use passwords. One layer may
use a password but the second layer
would then need to use a token or

31
 Solution Use Case Threshold/
Req # Requirement Description Alternative
Designs Objective
other factor).
DAR-CR-13 For use in routine threat environments HF, HS, SF, EM, GA, O Optional
(as defined by the AO) the two layers of HH RM, UO
DAR must use different primary
authentication factors (i.e., Both layers
cannot use passwords. One layer may
use a password but the second layer
would then need to use a token or
other factor).
DAR-CR-14 At least one DAR layer must use multi- HF, HS, SF, EM, GA, LF, O Optional
factor authentication. HH RM, UO
DAR-CR-15 The removable media must not be HF, HS, SF, RM T=O
bootable. HH
DAR-CR-16 The DAR Enterprise Server, must only HF, HS, EM T=O
manage one component/layer, and HH, SF
shall not manage the other
component/layer of the DAR solution.
DAR-CR-17 All administrators must use unique HF, HS, EM, GA, LF, T=O
identifiable accounts. HH, SF, PF RM, UO
DAR-CR-18 A baseline configuration that complies HF, HS, EM T=O
with this CP must be enforced on all HH, SF
registered endpoints.
DAR-CR-19 Enterprise management servers that HF, HS, EM T=O
leverage a SQL platform account HH, SF
management, must be configured
according to the guidance of the
platform and any additional
configuration guidance provided by the
component vendor.
DAR-CR-20 The two layers of DAR must use HF, HS, HH LF T=O
different primary authentication factors
(i.e., Both layers cannot use passwords.
One layer may use a password but the
second layer would then need to use a
token or other factor).
DAR-CR-21 Each DAR component must permit only HF, HS, EM T=O
an administrator to disable or alter its HH, SF
security functions.
DAR-CR-22 The administrator must configure HF, HH, EM T=O
remediation options (account lockout, HS, SF
key revocation, etc.) for failed

32
 Solution Use Case Threshold/
Req # Requirement Description Alternative
Designs Objective
authorization attempts by the user, as
determined by the AO.
DAR-CR-23 The administrator must configure HF, HH, GA, LF, RM, O Optional
remediation options (account lockout, HS, PF, SF UO
key revocation, etc.) for failed
authorization attempts by the user, as
determined by the AO.
DAR-CR-24 EUDs must require network access to HF, HH, EM O Optional
complete the authentication process HS, SF
for decryption.
DAR-CR-25 EUDs that are lost or compromised HF, HH, EM T=O
must be revoked and issue zeroize HS, SF
commands.

10.3 SWFDE COMPONENT REQUIREMENTS

Table 8: SWFDE Component Requirements
Solution Use Case Threshold/
Req # Requirement Description Alternative
Designs Objective
DAR-SW-1 The SWFDE must use Cipher Block SF, HS EM, GA, LF, T DAR-SW-2
Chaining (CBC) for data encryption. RM, UO
DAR-SW-2 The SWFDE must use XEX-based SF, HS EM, GA, LF, O DAR-SW-1
tweaked-codebook mode with cipher RM, UO
text stealing (XTS) or Galois/Counter
Mode (GCM) for data encryption.
DAR-SW-3 The SWFDE must be configured to use SF, HS EM, GA, LF, T=O
one of the following primary RM, UO
authentication options:
 A randomly generated
passphrase or password that
meets the minimum strength
set in Appendix D, or
 A randomly-generated bit
string equivalent to the
cryptographic strength of the
DEK contained on an external
USB token, or
 An external smartcard or
software capability containing
a software certificate with RSA
or Elliptic Curve Cryptography

33
 Solution Use Case Threshold/
Req # Requirement Description Alternative
Designs Objective
(ECC) key pairs per Table 1, or
 Any combination of the above.

10.4 FE COMPONENT REQUIREMENTS

Table 9: FE Component Requirements
Solution Use Case Threshold/
Req # Requirement Description Alternative
Designs Objective
DAR-FE-1 The FE product must use CBC for data SF, PF, HF EM, GA, LF, T DAR-FE-2
encryption. RM, UO
DAR-FE-2 The FE product must use XTS or GCM SF, PF, HF EM, GA, LF, O DAR-FE-1
for data encryption. RM, UO
DAR-FE-3 The FE product must use one of the SF, PF, HF EM, GA, LF, T=O
following primary authentication RM, UO
options:
 A randomly generated
passphrase or password that
meets the minimum strength
set in Appendix D, or
 A randomly-generated bit
string equivalent to the
cryptographic strength of the
DEK contained on an external
USB token, or
 An external smartcard or
software capability containing
a software certificate with RSA
or Elliptic Curve Cryptography
(ECC) key pairs per Table 1, or
 Any combination of the above.

10.5 PE COMPONENT REQUIREMENTS

Table 10: PE Component Requirements
Solution Use Case Threshold/
Req # Requirement Description Alternative
Designs Objective
DAR-PE-1 The PE must enable the “wipe sensitive PF GA, LF T=O
data” management function for
imported or self-generated
keys/secrets and/or other classified
data.

34
 Solution Use Case Threshold/
Req # Requirement Description Alternative
Designs Objective
DAR-PE-2 The PE must use CBC for data PF GA, LF T DAR-PE-3
encryption.
DAR-PE-3 The PE must use XTS or GCM for data PF GA, LF O DAR-PE-2
encryption.
DAR-PE-4 The AO must provide policy to the user PF GA, LF T=O
determining when data or keys must
be wiped.
DAR-PE-5 The PE product must use one of the PF GA, LF T=O
following primary authentication
options:
A minimum of a randomly generated
six-character, case-sensitive
alphanumeric password with the
length defined by the AO, or a
randomly generated passphrase with
the length defined by the AO.

10.6 HWFDE COMPONENT REQUIREMENTS

Table 11: HWFDE Component Requirements
Solution Use Case Threshold/
Req # Requirement Description Alternative
Designs Objective
DAR-HW-1 The HWFDE must use CBC for data HF, HS, HH EM, GA, LF, T DAR-HW-2
encryption. RM, UO
DAR-HW-2 The HWFDE must use GCM or XTS for HF, HS, HH EM, GA, LF, O DAR-HW-1
data encryption. RM, UO
DAR-HW-3 The HWFDE must be configured to use HF, HS, HH EM, GA, LF, T=O
one of the following primary RM, UO
authentication options:
 A randomly generated
passphrase or password that
meets the minimum strength
set in Appendix D, or
 A randomly-generated bit
string equivalent to the
cryptographic strength of the
DEK contained on an external
USB token, or
 An external smartcard or
software capability containing
a software certificate with RSA

35
 Solution Use Case Threshold/
Req # Requirement Description Alternative
Designs Objective
or ECC key pairs per Table 1, or
 Any combination of the above.

10.7 END USER DEVICES REQUIREMENTS

Table 12: End User Device Requirements
Solution Use Case Threshold/
Req # Requirement Description Alternative
Designs Objective
DAR-EU-1 All EUD provisioning must be SF, PF, HF, EM, GA, LF, T=O
performed through direct physical HS, HH RM, UO
access or through an enterprise
management server.
DAR-EU-2 If found after being lost, the EUD’s SF, PF, HF, EM, GA, T=O
non-volatile storage media must be HS, HH RM, UO
destroyed per NSA/CSS Storage Device
Sanitization (NSA/CSS Policy Manual 9-
12). (This does not preclude having
the device forensically analyzed by the
appropriate authority.)
DAR-EU-3 EUDs must implement the Basic SF, PF, HF, EM, GA, LF, O Optional
Input/Output System (BIOS) security HS, HH UO
guidelines specified in NIST SP 800-
147.
DAR-EU-4 All users must sign an organization- SF, PF, HF, EM, GA, LF, T=O
defined user agreement before being HS, HH RM, UO
authorized to use an EUD.
DAR-EU-5 All users must receive an organization- SF, PF, HF, EM, GA, LF, T=O
developed training course for HS, HH RM, UO
operating an EUD prior to use.

36
 Solution Use Case Threshold/
Req # Requirement Description Alternative
Designs Objective
DAR-EU-6 At a minimum, the organization- SF, PF, HF, EM, GA, LF, T=O
defined user agreement must include HS, HH RM, UO
each of the following:
 Consent to monitoring
 Operational Security (OPSEC)
guidance
 Required physical protections
to employ when operating
and storing the EUD
 Restrictions for when, where,
and under what conditions the
EUD may be used
 Responsibility for reporting
security incidents
 Verification of IA training
 Verification of appropriate
clearance
 Justification for Access
 Requester information and
organization
 Account Expiration Date
 User Responsibilities
 An overview of what
constitutes continuous
physical control and the risks
associated with using the EUD
after it is lost
DAR-EU-7 External USB tokens and smartcards, SF, PF, HF, EM, GA, LF, T=O
when used for authentication, must HS, HH UO, RM
be removed from the EUD upon or
before shut down in accordance with
AO policy.
DAR-EU-8 AO must provide guidance on storing SF, PF, HF, EM, GA, LF, T=O
and/or securing authentication HS, HH RM, UO
factors.
DAR-EU-9 The SA must disable system power SF, HF, HS, EM, GA, LF, T=O
saving states on EUDs (i.e., sleep and HH UO
hibernate).
DAR-EU-10 The EUD must power off after a period SF, HF, HS, EM, GA, LF, T=O
of inactivity defined by the AO, unless HH RM, UO
this is not supported by the device.

37
 Solution Use Case Threshold/
Req # Requirement Description Alternative
Designs Objective
DAR-EU-11 The EUDs must be provisioned within SF, PF, HF, EM, GA, LF, T=O
a physical environment certified to HS, HH RM, UO
protect the highest classification level
of the data stored on the device.
DAR-EU-12 The EUD must only be re-provisioned SF, PF, HF, EM, GA, LF, T=O
to the same or higher classification HS, HH RM, UO
level of the classified data per an AO
approved process.
DAR-EU-13 The EUD must be reported as “lost” SF, PF, HF, EM, GA, T=O
when out of continuous physical HS, HH RM, UO
control as specified by the AO.
DAR-EU-14 System folders must have user write SF, HF EM, GA, LF, T=O
permissions disabled unless UO
authorized by an administrator.
DAR-EU-15 The EUD must be protected with anti- SF, PF, HF, EM, GA, LF, O Optional
tamper or detection capabilities. HS, HH RM
DAR-EU-16 The device must be powered down HF, HS, PF, EM, GA, LF, T=O
before being handled by an SF, HH RM, UO
unauthorized party (e.g., customs) and
inspected afterwards. If the
unauthorized party required the
device to be powered on again for
inspection, the device must be
rebooted again before use.
DAR-EU-17 The absence of any expected HF, HS, PF, EM, GA, LF, T=O
authentication prompt(s) must be SF, HH RM, UO
reported as possible tampering to the
AO.
DAR-EU-18 When data is no longer needed, it HF, HS, PF, EM, GA, LF, O Optional
must be overwritten or erased by SF, HH RM, UO
secure erase tool per AO guidance.
(See Section 4.10)
DAR-EU-19 The EUD, when not in use outside of a HF, HS, PF, EM, GA, LF, O Optional
secured facility, must be kept in an SF, HH RM, UO
AO-approved locked container.
DAR-EU-20 The BIOS/Unified Extensible Firmware HF, HS, SF, EM, GA, LF, O Optional
Interface (UEFI) must be configured to HH UO
require a password before continuing
the boot process.

38
 Solution Use Case Threshold/
Req # Requirement Description Alternative
Designs Objective
DAR-EU-21 All DAR FDE components must be HF, HS, SF, EM, GA, LF, T=O
cryptographically erased before being HH RM, UO
provisioned again.
DAR-EU-22 All DAR components must be PF GA, LF O Optional
cryptographically erased before being
provisioned again.
DAR-EU-23 System folders must have user write PF GA, LF O Optional
permissions disabled, unless
authorized by an administrator.
DAR-EU-24 If supported, the EUD must have the HF, HS, SF, EM, GA, T=O
BIOS/UEFI password enabled. HH, PF UO, LF
DAR-EU-25 If the user suspects the EUD has been HF, HS, PF, EM, GA, LF, T=O
compromised, the EUD user must SF, HH RM, UO
obtain authorization from their AO
prior to use.
DAR-EU-26 Each EUD must be personalized by the HF, HS, PF, EM, GA, O Optional
end user. (This should not violate any SF, HH RM
other security features.)
DAR-EU-27 The EUD must not be used as a smart HF, HS, SF, RM T=O
card/USB Authentication Token, if it is HH
also storing encrypted user data.
DAR-EU-28 The EUD must be removed from a HF, HS, SF, RM T=O
host system before being handled by HH
an unauthorized party (e.g., customs).
DAR-EU-29 Administrators and endpoint users HF, HS, EM, GA, LF, T=O
must be restricted from making HH, PF, SF RM, UO
configuration changes based on what
the product supports, using a model of
least privilege.
DAR-EU-30 The EUD must be reported as HH, HF, LF T=O
“compromised” when tampering is HS, PF
suspected, as defined by AO policy.
DAR-EU-31 The EUD and/or non-volatile storage HH, HF, LF T=O
media, if compromised, must be HS, PF
destroyed per NSA/CSS Storage Device
Sanitization (NSA/CSS Policy Manual 9-
12). (This does not preclude having
the device forensically analyzed by the
appropriate authority.)

39
 Solution Use Case Threshold/
Req # Requirement Description Alternative
Designs Objective
DAR-EU-32 Prior to reuse, the EUD must undergo HH, HF, LF T=O
tamper detection inspection as HS, PF
established by the AO to determine if
the device has been tampered with or
substituted.
DAR-EU-33 The EUD, when outside of a secured HH, HF, LF T=O
facility and not in use, must be kept HS, PF
concealed from potential adversaries.
DAR-EU-34 If an unauthorized party takes the EUD HH, HF, LF T=O
out of sight or performs unknown HS, PF
operations, the device must be
considered compromised.
DAR-EU-35 When using commercial modes of HH, HF, LF T=O
travel (e.g., non-secure), the EUD must HS, PF
stay with the traveler and not be
placed in checked baggage.
DAR-EU-36 Each EUD must be personalized by the HH, HF, LF T=O
end user. (This should not violate any HS, PF
other security features.)
DAR-EU-37 EUDs must use boot integrity SF, HH, HF, EM, GA, T=O
verification. (see Appendix A) HS UO, LF
DAR-EU-38 EUDS must implement “DAR Location SF, HF, HS, EM, GA, LF, O Optional
based Services” features and restrict PF, HH UO
decryption of data to only approved
locations.
DAR-EU-39 The EUD must be protected with anti- SF, PF, HF, UO T=O
tamper or detection capabilities. HS, HH

10.8 CONFIGURATION CHANGE DETECTION REQUIREMENTS

Table 13: Configuration Change Detection Requirements
Solution Use Case Threshold/
Req # Requirement Description Alternative
Designs Objective
DAR-CM-1 A history of baseline configuration for SF, PF, HF, EM, GA, LF, T=O
all components must be maintained by HS, HH RM, UO
the SA.
DAR-CM-2 An automated process must ensure SF, PF, HF, EM, GA, LF, O Optional
configuration changes are logged. HS, HH RM, UO

40
 Solution Use Case Threshold/
Req # Requirement Description Alternative
Designs Objective
DAR-CM-3 Log messages generated for SF, PF, HF, EM, GA, LF, O Optional
configuration changes must include HS, HH RM, UO
the specific changes made to the
configuration.
DAR-CM-4 A history of baseline configuration for SF, PF, HF, EM, GA, LF, T=O
all components must be available to HS, HH RM, UO
the auditor.
DAR-CM-5 Configuration change logs must be SF, PF, HF, EM, GA, LF, T=O
kept for an AO defined period of time. HS, HH RM, UO

10.9 DEVICE MANAGEMENT REQUIREMENTS

Table 14: Device Management Requirements
Solution Use Case Threshold/
Req # Requirement Description Alternative
Designs Objective
DAR-DM-1 EUDs must be physically administered. SF, PF, HF, EM, GA, LF, T DAR-DM-2
HS, HH RM, UO
DAR-DM-2 EUDs must be remotely administered SF, PF, HF, EM, GA, LF, O DAR-DM-1
using an NSA-approved Data-In-Transit HS, HH RM, UO
(DIT) protection solution (e.g., NSA
Certified Product or CSfC approved
solution).
DAR-DM-3 Administration workstations must be SF, PF, HF, EM, GA, LF, T=O
dedicated for the purposes given in HS, HH RM, UO
the CP.
DAR-DM-4 Administration workstations must SF, PF, HF, EM, GA, LF, T=O
physically reside within a protected HS, HH RM, UO
facility where CSfC solution(s) are
managed.
DAR-DM-5 Administration workstations must be SF, PF, HF, EM, GA, LF, T=O
physically separated from HS, HH RM, UO
workstations used to manage non-
CSfC solutions.
DAR-DM-6 Only authorized SAs (See Section 12) SF, PF, HF, EM, GA, LF, T=O
must be allowed to administer the HS, HH RM, UO
DAR Components.

DAR-DM-7 EUDs must be remotely administered, SF, HF, HS, EM, UO T=O
but local administration may still be HH
performed.

41
 10.10 AUDITING REQUIREMENTS
Table 15: Auditing Requirements
Solution Use Case Threshold/
Req # Requirement Description Alternative
Designs Objective
DAR-AU-1 EUDs must be inspected for malicious SF, PF, HF, EM, GA, LF, T=O
physical changes in accordance with HS, HH RM, UO
AO defined policy.
DAR-AU-2 The EUDs must be configured to SF, PF, HF, EM, GA, LF, O Optional
generate an audit record of the HS, HH RM, UO
following events:
 Start-up and shutdown of any
platform audit functions.
 All administrative actions
affecting the DAR encryption
components.
 User authentication attempts
and success/failure of the
attempts.
 Software updates to the DAR
encryption components.
DAR-AU-3 Auditors must review audit logs for a SF, PF, HF, EM, GA, LF, T=O
time period as defined by the AO. HS, HH RM, UO
DAR-AU-4 Auditors must physically account for SF, PF, HF, EM, GA, LF, T=O
the EUDs after an AO-defined time HS, HH RM, UO
period.
DAR-AU-5 Administrators must periodically SF, PF, HF, EM, GA, LF, O Optional
compare solution component HS, HH RM, UO
configurations to a trusted baseline
configuration after an AO-defined
time period.

42
 Solution Use Case Threshold/
Req # Requirement Description Alternative
Designs Objective
DAR-AU-6 For DAR EM products that support HF, HS, EM T=O
auditing functions, audit records must HH, SF
be generated and recorded for:
 Encryption status of endpoints
 Recovery attempts and
success/failure of the
attempts
 Out of date endpoint versions
 Platform changes
 Registration of new endpoints
 Revocations of endpoints
 Key escrow from endpoints
 Cryptographic erase of
endpoints
 Changes to administrator
account
 Changes to policies pushed to
endpoints

10.11 KEY MANAGEMENT REQUIREMENTS

Table 16: Key Management Requirements for All DAR Components
Solution Use Case Threshold/
Req # Requirement Description Alternative
Designs Objective
DAR-KM-1 The key sizes used for each layer must SF, PF, HF, EM, GA, LF, T=O
be as specified in Table 1. HS, HH RM, UO
DAR-KM-2 DAR solution products must be initially SF, PF, HF, EM, GA, LF, T=O
keyed within a physical environment HS, HH RM, UO
certified to protect the highest
classification level of the DAR solution.
DAR-KM-3 The DAR solution must disable all key SF, PF, HF, GA, LF, RM, T=O
recovery mechanisms. HS, HH UO
DAR-KM-4 The algorithms used for each layer SF, PF, HF, EM, GA, LF, T=O
must be as specified in Table 1. HS, HH RM, UO
DAR-KM-5 If a physical recovery output is utilized, SF, HF, HS, EM T=O
it must be secured as classified HH
information, equivalent to the level of
data it is protecting.
DAR-KM-6 If recovery information is distributed SF, HF, HS, EM T=O
over a non-CSfC channel (i.e., HH

43
 Solution Use Case Threshold/
Req # Requirement Description Alternative
Designs Objective
physically, voice channel, etc.), it must
be secured as classified information,
equivalent to the level of data it is
protecting.
DAR-KM-7 The AO must determine a SF, HF, HS, EM T=O
methodology for verification of end HH
users requesting recover material,
whether recovery information is
distributed over a channel not
provided by the CSfC solution (i.e.,
physically, voice channel, etc.) or
distribution by a CSfC solution
component which is expected to
provide verification itself.

10.12 SUPPLY CHAIN RISK MANAGEMENT REQUIREMENTS

Table 17: Supply Chain Risk Management Requirements
Solution Use Case Threshold/
Req # Requirement Description Alternative
Designs Objective
DAR-SC-1 CSfC Trusted Integrators must be SF, PF, HF, EM, GA, LF, O Optional
employed to architect, design, HS, HH RM, UO
procure, integrate, test, document,
field, and support the solution.
DAR-SC-2 Each component selected from the HF, HS, SF, EM, GA, LF, T=O
CSfC Components List must go through PF, HH RM, UO
a Product Supply Chain Risk
Management (SCRM) Assessment to
determine the appropriate mitigations
for the intended application of the
component per the organization’s AO-
approved Product SCRM process. (See
Committee on National Security
System Directive (CNSSD) 505 SCRM
for additional guidance.)

44
 11 SOLUTION OPERATION, MAINTENANCE, & HANDLING
REQUIREMENTS
11.1 USE AND HANDLING OF SOLUTION REQUIREMENTS
The following requirements must be followed regarding the use and handling of the solution.

Table 18: Use and Handling of Solutions Requirements

Solution Use Case Threshold/
Req # Requirement Description Alternative
Designs Objective
DAR-GD-1 Acquisition and procurement SF, PF, HF, EM, GA, LF, T=O
documentation must not include HS, HH RM, UO
information about how the
equipment will be used, including
that it will be used to protect
classified information.
DAR-GD-2 The solution owner must allow, and SF, PF, HF, EM, GA, LF, T=O
fully cooperate with, NSA or its HS, HH RM, UO
authorized agent to perform an IA
compliance audit (including, but not
limited to, inspection, testing,
observation, interviewing) of the
solution implementation to ensure
that it meets the latest version of the
CP.
DAR-GD-3 The AO must ensure that a SF, PF, HF, EM, GA, LF, T=O
compliance audit is conducted every HS, HH RM, UO
year against the latest version of the
DAR CP.
DAR-GD-4 Results of the compliance audit must SF, PF, HF, EM, GA, LF, T=O
be provided to and reviewed by the HS, HH RM, UO
AO.
DAR-GD-5 When a new, approved version of the SF, PF, HF, EM, GA, LF, T=O
DAR CP is published, the AO must HS, HH RM, UO
ensure compliance against this new
CP within 6 months.
DAR-GD-6 Solution implementation SF, PF, HF, EM, GA, LF, T=O
information, which was provided to HS, HH RM, UO
NSA during solution registration,
must be updated every 12 (or fewer)
months (see Section 13.3).

45
 Solution Use Case Threshold/
Req # Requirement Description Alternative
Designs Objective
DAR-GD-7 The SA, auditor, user, and all SF, PF, HF, EM, GA, LF, T=O
Integrators must be cleared to the HS, HH RM, UO
highest level of data protected by the
DAR solution.
DAR-GD-8 The SA and auditor roles must be SF, PF, HF, EM, GA, LF, T=O
performed by different people. HS, HH RM, UO
DAR-GD-9 All SAs, users, and auditors must SF, PF, HF, EM, GA, LF, T=O
meet local information assurance HS, HH RM, UO
training requirements.
DAR-GD-10 Users must report lost or stolen EUDs SF, PF, HF, EM, GA, LF, T=O
to their ISSO or chain of command as HS, HH RM, UO
defined by the AO.
DAR-GD-11 Only SAs or CSfC Trusted Integrators SF, PF, HF, EM, GA, LF, T=O
must perform the installation and HS, HH RM, UO
policy configuration.
DAR-GD-12 Security critical patches (such as SF, PF, HF, EM, GA, LF, T=O
Information Assurance Vulnerability HS, HH RM, UO
Alerts (IAVAs)) must be tested and
subsequently applied to all
components in the solution in
accordance with local policy and this
CP.
DAR-GD-13 Local policy must dictate how the SA SF, PF, HF, EM, GA, LF, T=O
installs patches to solution HS, HH RM, UO
components.
DAR-GD-14 All DAR components must be SF, PF, HF, EM, GA, LF, T=O
updated using digitally signed HS, HH RM, UO
updates provided by the vendor.
DAR-GD-15 All authorized users must have the SF, PF, HF, EM, GA, LF, O Optional
ability to CE keys for both layers. HS, HH RM, UO
DAR-GD-16 When using an FE Product, the user SF, PF, HF EM, GA, LF, T=O
must ensure that no classified data RM, UO
shall be put into the file’s metadata
(e.g., filename).
DAR-GD-17 Withdrawn
DAR-GD-18 Withdrawn

46
 Solution Use Case Threshold/
Req # Requirement Description Alternative
Designs Objective
DAR-GD-19 AO must define loss of continuous SF, PF, HF, EM, GA, LF, T=O
physical control for each use case. HS, HH RM, UO
This definition must cover the
following topics:

 User handling
 EUD Transportation
 EUD Storage
 Anti-tamper mechanisms and
related policies, if any are
used.
 Device integrity measures
and related policies, if any
are used.
DAR-GD-20 Organizational-developed training HH, HF, LF T=O
must include guidance on tamper HS, PF
awareness and detection.

DAR-GD-21 Organizational-developed training SF, HF, HS, EM T=O

must include the following topics if HH
they are included in the solution for
both administrators and users:

 Checking the encryption

status of endpoints
 Using the recovery
mechanisms supported in the
NIAP evaluated configuration
 Checking for out of date
endpoint versions
 Detecting platform changes
 The registration process for
endpoints
 The revocation process for
endpoints
 The key escrow process for
endpoints
 The cryptographic erase
process for endpoints
 The process for pushing
policy changes to endpoints

47
 11.2 INCIDENT REPORTING REQUIREMENTS
Table 19 lists requirements to report security incidents to NSA regarding incidents affecting the solution.
These reporting requirements are intended to augment, not replace, any incident reporting procedures
already in use within the solution owner’s organization. It is critical that SAs and auditors are familiar
with maintaining the solution in accordance with this CP. Based on familiarity with the known-good
configuration of the solution, personnel responsible for Operations and Maintenance (O&M) will be
better equipped to identify reportable incidents.

For the purposes of incident reporting, “malicious” activity includes not only events that have been
attributed to activity by an adversary, but also any events that are unexplained. In other words, an
activity is assumed to be malicious unless it has been determined to be the result of known non-
malicious activity.

Table 19 only provides requirements directly related to the incident reporting process. See Section
10.10 for requirements supporting detection of events that may reveal that a reportable incident has
occurred.

Table 19: Incident Reporting Requirements

Solution Use Case Threshold/
Req # Requirement Description Alternative
Designs Objective
DAR-RP-1 Report a security failure in any of the SF, PF, HF, EM, GA, LF, T=O
CSfC DAR solution components. HS, HH RM, UO
DAR-RP-2 Report any malicious configuration SF, PF, HF, EM, GA, LF, T=O
changes to the DAR components. HS, HH UO, RM
DAR-RP-3 Report any evidence of a compromise SF, PF, HF, EM, GA, LF, T=O
of classified data caused by a failure of HS, HH RM, UO
the CSfC DAR solution. Compromise,
in this context, includes reporting real
or perceived access to classified data
(e.g., user or administrator access that
occurs without proper authentication
or through the use of incorrect
credentials).

DAR-RP-4 Report any evidence of malicious SF, PF, HF, EM, GA, LF, T=O
physical tampering (i.e., missing or HS, HH RM, UO
mis-installed parts) with solution
components.

48
 Solution Use Case Threshold/
Req # Requirement Description Alternative
Designs Objective
DAR-RP-5 Confirmed incidents meeting the SF, PF, HF, EM, GA, LF, T=O
criteria in DAR-RP-1 through DAR-RP-4 HS, HH RM, UO
must be reported within 24 hours of
detection via Joint Incident
Management System (JIMS) or
contacting NSA as specified in the CSfC
Registration Letter.
DAR-RP-6 At a minimum, the organization must SF, PF, HF, GA, EM, LF, T=O
provide the following information HS, HH UO, RM
when reporting security incidents:
 CSfC Registration Number
 Point of Contact (POC) name,
phone, email
 Alternate POC name, phone,
email
 Classification level of affected
solution
 Affected component(s)
manufacturer/vendor
 Affected component(s) model
number
 Affected component(s) version
number
 Date and time of incident
 Description of incident
 Description of remediation
activities
Is Technical Support from NSA
requested? (Yes/No)

12 ROLE-BASED PERSONNEL REQUIREMENTS

The roles required to administer and maintain the solution are detailed below, along with doctrinal
requirements for these roles.

End User – An end user may operate an EUD from physical locations not owned, operated, or controlled
by the Government. The end user must be responsible for operating the EUD in accordance with this CP
and an organization-defined user agreement. End user duties include, but are not limited to the
following:

1. Ensuring that the EUD is only operated in physical spaces that comply with the end user
agreement.

49
 2. Alerting the Security Administrator immediately upon an EUD being lost, stolen, or suspected of
being tampered with.

Security Administrator – The SA must be responsible for maintaining, monitoring, and controlling all
security functions for the entire suite of products composing the DAR solution. Security Administrator
duties include, but are not limited to the following:

1. Ensuring that the latest security critical software patches and updates (such as IAVAs) are
applied to each product in a timely fashion.

2. Documenting and reporting security-related incidents to the appropriate authorities.

3. Coordinating and supporting product logistic support activities including integration and
maintenance. Ensuring that the implemented DAR solution remains compliant with the latest
version of the CP.

4. Provisioning and maintaining EUDs in accordance with this CP.

Auditor - The auditor must be responsible for reviewing the actions performed by the SA and events
recorded in the audit logs to ensure that no action or event represents a compromise of the DAR
solution. The role of auditor and SA must not be performed by the same individual. Auditor duties
include but are not limited to the following:

1. Reviewing, managing, controlling, and maintaining security audit log data.

2. Documenting and reporting security-related incidents to the appropriate authorities.

3. The auditor will be given authority to access all audit records.

Integrator – Integrator duties may include but are not limited to the following:

1. Acquiring the products that compose the solution.

2. Configuring the DAR solution in accordance with the CP.

3. Testing the DAR solution.

4. Documenting the solution and its compliance to the CP.

5. Troubleshooting the solution.

In certain cases, an external integrator may be used to implement a DAR solution based on the CP. A
CSfC Trusted Integrator is one such entity. Although not required, the use of CSfC Trusted Integrators is
highly recommended. A CSfC Trusted Integrator is defined as a selected organization that has
demonstrated competency in:

1. System integration.

2. The technologies to be integrated.

50
 3. Formal testing processes.

4. Generating evidence for system authorization.

Chosen CSfC Trusted Integrator applicants are required to sign a Memorandum of Agreement (MoA)
with NSA.

13 INFORMATION TO SUPPORT THE AO

This section details items that will likely be necessary for the customer to obtain approval from the
system AO. The customer and AO have obligations to perform the following:

 The customer, possibly with support from an Integrator, instantiates a solution implementation
that follows the NSA-approved CP.
 The customer has a testing team develop a test plan and perform testing of the DAR solution,
see Section 13.1.
 The customer has system assessment and authorization performed using the RA information
referenced in Section 13.2.
 The customer provides the results from testing and from system assessment and authorization
to the AO for use in making an approval decision. The AO is ultimately responsible for ensuring
that all requirements from the CP have been properly implemented. NSA publishes compliance
matrixes requiring a short description of how requirements are met. NSA recommends that the
AO require the compliance matrix as part of their body of evidence.
 The customer registers the solution with NSA and re-registers yearly to validate its continued
use as detailed in Section 13.3. NSA publishes registration forms at:
http://www.nsa.gov/resources/commercial-solutions-for-classified-program/solution-
registration.
 Customers who want to use a variant of the solution detailed in this CP will contact NSA early in
their design phase to determine ways to obtain NSA approval.
 The AO will ensure that a compliance audit must be conducted every year against the latest
version of the DAR CP, and the results must be provided to the AO.

13.1 SOLUTION TESTING

This section provides a framework for a Test and Evaluation (T&E) plan and procedures to validate the
implementation of a DAR solution. This T&E will be a critical part of the approval process for the AO,
providing a robust body of evidence that shows compliance with this CP.

The security features and operational capabilities associated with the use of the solution must be tested.
The following is a general, high-level methodology for developing the test plan and procedures and for
the execution of those procedures to validate the implementation and functionality of the DAR solution.
The entire solution, to include each component described in Section 5, is addressed by this test plan.

1. Set up the baseline network design and configure all components.

51
 2. Document the baseline network design configuration. Include product model and serial
numbers, and software version numbers as a minimum.

3. Develop a test plan for the specific implementation using the test objectives from the DAR CP
Testing Annex. Any additional requirements imposed by the local AO should also be tested, and
the test plan must include tests to ensure that these requirements do not interfere with the
security of this solution as described in this CP.

4. Perform testing using the test plan derived in Step 3. System testing will consist of both black
box testing and gray box testing. A two-person testing approach should be used to administer
the tests. During test execution, security and non-security related discrepancies with the
solution must be documented.

5. Compile findings, including comments and vulnerability details as well as possible

countermeasure information, into a final test report to be delivered to the AO for their approval
of the solution.

6. The testing requirement in Table 20 has been developed to ensure that the DAR solution
functions properly and meets the configuration requirements from Section 8. Testing of these
requirements should be used as a minimum framework for the development of the detailed test
plan and procedures.

Table 20: Test Requirements

Solution Use Case Threshold/
Req # Requirement Description Alternative
Designs Objective
DAR-TR-1 The organization implementing the CP HF, HS, PF, EM, GA, LF, T=O
must perform all tests listed in the DAR SF, HH RM, UO
CP Testing Annex.

13.2 RISK ASSESSMENT

The Risk Assessment (RA) of the DAR solution presented in this CP focuses on the types of attacks that
are feasible against this solution and the mitigations that can be employed. Customers should contact
their NSA/CSS Customer Advocate to request the RA, or visit the Secret Internet Protocol Router
Network (SIPRNet) CSfC site for information. The process for obtaining the RA is available on the
SIPRNet CSfC website. The AO must be provided a copy of the NSA RA for their consideration in
approving the use of the solution.

13.3 REGISTRATION OF SOLUTIONS

All customers using CSfC solutions to protect information on National Security Systems (NSS) must
register their solution with NSA prior to operational use. Customers will provide their compliance
checklists and registration forms to NSA. This registration will allow NSA to track where DAR CP
solutions are instantiated and to provide AOs at those sites with appropriate information, including all
significant vulnerabilities that may be discovered in components or high-level designs approved for
these solutions. The CSfC solution registration process, as well as the compliance matrices and

52
 registration forms, are available at http://www.nsa.gov/resources/commercial-solutions-for-classified-
program/solution-registration.

Solution registrations are valid for one year, at which time customers are required to re-register their
solution in order to continue using it. Approved CPs will be reviewed twice a year, or as events warrant.
Registered users of this CP will be notified when an updated version is published. When a new version
of this CP that has been approved by the D/NM is published, customers will have six months to bring
their solutions into compliance with the new version and re-register them (see requirement DAR-GD-5).
Customers are also required to update their registrations whenever the information provided on the
registration form changes.

14 TESTING REQUIREMENTS
The testing requirements for the DAR solution are located in a separate DAR CP Testing Annex. This
document contains the specific tests that allow the Security Administrator or Integrator to ensure they
have properly configured the solution. Contact the CSfC PMO to obtain the DAR CP Testing Annex.

53
 APPENDIX A: GLOSSARY OF TERMS
Administration Workstation - This device is commonly used for logging, configuration review, and
management of the EUD.

Anti-Tamper Measures - These measures serve to deter or delay modification of an EUD. Passive anti-
tamper measures aid in detecting attempts to modify the EUD or inject a substitute device. Active anti-
tamper measures serve the same purpose as passive measures, while actively detecting and securing
the EUD. Examples include personalization options such as stickers, screen savers, wall papers, or other
personalization methods which do not interfere with the configuration of the device.

Assessment - The technical evaluation of a system’s security features performed as part of, and in
support of, the approval/accreditation process that establishes the extent to which a particular
computer system’s design and implementation meet a set of specified security requirements.

Assessment and Authorization - A comprehensive assessment of the management, operational, and

technical security controls in an information system, made in support of security accreditation, to
determine the extent to which the controls are implemented correctly, operating as intended, and
producing the desired outcome with respect to meeting the security requirements for the system. In
conjunction with the official management decision given by a senior agency official to authorize
operation of an information system and to explicitly accept the risk to agency operations (including
mission, functions, image, or reputation), agency assets, or individuals, based on the implementation of
an agreed-upon set of security controls. (NIST 800-37)

Assurance - A measure of confidence that the security features, practices, procedures, and architecture
of an information system accurately mediates and enforces the security policy.

Audit - The activity of monitoring the operation of a product from within the product. It includes
monitoring of a product for a set of pre-determined events. Each audit event may indicate rogue
behavior, or a condition that is detrimental to security, or provide necessary forensics to identify the
source of rogue behavior.

Authentication - The process of confirming the identity of a user.

Authorizing Official (AO) – A senior (Federal) official or executive with the authority to formally assume
responsibility for operating an information system at an acceptable level of risk to organizational
operations (including mission, functions, image, or reputation), organizational assets, individuals, other
organizations, and the Nation.

Authorization - The official management decision given by a senior agency official to authorize
operation of an information system and to explicitly accept the risk to agency operations (including
mission, functions, image, or reputation), agency assets, or individuals, based on the implementation of
an agreed-upon set of security controls (NIST 800-37). It can also be the decision to allow or deny a
subject access to an object. For example, after a user has been authenticated, authorization determines
if the user has the rights to perform specific actions on the device.

Boot Integrity Verification - These features ensure no code is executed during the boot process that has
not first been verified for its integrity and authenticity. Each step in the boot process should verify the
integrity of the next piece of code to execute before handing execution over to it. In current PC

54
 technology, this operates in two stages. First, the integrity and authenticity of the firmware is verified
using a platform/vendor specific technology. Second, UEFI secure boot verifies the option ROMs, and
the OS loader before execution is handed over to the operating system.

Capability Package (CP) - The set of guidance provided by NSA that describes recommended approaches
to provide architectures and configuration requirements that empower CS customers to implement
secure solutions using independent, layered COTS components to protect classified information. This
package will point to potential products that can be used as part of this solution. The CPs are product-
neutral and describe system-level solution frameworks documenting security and configuration
requirements for customers and Integrators.

Commercial National Security Algorithm (CNSA) - Set of commercial algorithms capable of protecting
data through Top Secret level (previously known as Suite B).

Committee on National Security Systems Policy No. 15 (CNSSP-15) - Policy specifies which public
standards may be used for cryptographic protocol and algorithm interoperability to protect National
Security Systems.

Compromise - Any computing resource whose confidentiality, integrity, or availability has been
adversely impacted, either intentionally or unintentionally.

Continuous Physical Control - The AO defines what is considered “Continuous Physical Control.”
Previously called “positive control.”

Cryptographic Erase - A method of sanitization in which an encryption key for the encrypted data is
sanitized, making recovery of the decrypted data infeasible.

DAR Component - Consists of a component that is part of the DAR solution (e.g., HWFDE, SWFDE, PE).

DAR Solution - A DAR Solution consists of two layered components (e.g., HWFDE and SWFDE).

Enterprise Management (EM) - DAR use case that employs a client-server architecture to provide
management of DAR components at an enterprise level.

End User Device (EUD) - Any computing or storage device that can store data when it is powered off (in
the context of this DAR document).

False Acceptance – When a different user will pass the biometric when they should not. Measured by
false acceptance rate (FAR).

False Rejection – When an authorized user’s measurements fail to authenticate. Measured by false
rejection rate (FRR).

Federal Information Processing Standards (FIPS) - A set of standards that describes the handling and
processing of information within governmental agencies.

File Encryption (FE) - File encryption is the process of encrypting individual files or sets of files on an EUD
and permitting access to the encrypted data only after proper authentication is provided.

Found Device - A lost device that has been recovered. (See Lost Device definition.)

55
 Full Disk Encryption (FDE) - Also known as whole disk encryption, is the process of encrypting all the
data on the drive used to boot a computer, including the computer’s OS, and permitting access to the
data only after successful authentication to the FDE product.

GA- DAR use case that doesn’t use a specific use case such as RM, UO, EM or LF. This is generally
applicable to requirements and used in a standalone implementation.

HF - DAR solution design that uses HWFDE as the outer layer, and FE as the inner layer.

HH- DAR solution design that uses HWFDE as the outer and inner layers.

High Assurance GOTS solution - cryptographic equipment, assembly, or component that is classified or
certified by the NSA for encrypting and decrypting classified or sensitive national security information
when appropriately keyed. (Previously referred to as Type 1)

HS - DAR solution design that uses HWFDE as the outer layer, and SWFDE as the inner layer.

IA Product - Product whose primary purpose is to provide security services (e.g. confidentiality,
authentication, integrity, access control, non-repudiation of data); correct known vulnerabilities; and/or
provide layered defense against various categories of non-authorized or malicious penetrations of
information systems or networks.

ISV - An independent software vendor is a separate vendor that provides a product for managing a self-
encrypting drive and provides a user interface to the drive. This definition is unique to this CP.

Known Secret - PIN, password, or passphrase.

Layer - Every DAR solution protects classified data with two layers (e.g., HWFDE, SWFDE, FE, and PE).

Lost Device - A device that is removed from the control of the physical security procedures defined by
the AO.

Mission Control Element (MCE) - The location and system from which a connection occurs to a remote
EUD using the UO use case.

Network Attached Storage (NAS) - A file-level computer data storage server connected to a computer
network providing data access to a group of clients. A NAS is a specialized computer built for storing
and serving files.

PF - DAR solution architecture that features a PE layer under the FE layer.

Platform Encryption (PE) - A device that has met the requirements (and high assurance use case) of the
MDF PP.

Pre-Boot Environment (PBE) - The initial software that is executed on start-up of the EUD that requires
a user to authenticate successfully before decrypting and booting an operating system. This is the layer
of authentication for the SWFDE or HWFDE product.

Protection Profile (PP) - A document used as part of the certification process according to the Common
Criteria. As the generic form of a security target, it is typically created by a user or user community and
provides an implementation independent specification of information assurance security requirements.

56
 Removable Media (RM) – A device which has the primary purpose of providing external storage of data,
protected by DAR via two layers of encryption.

Radio Frequency Identification (RFID) – Technology that uses electromagnetic fields to automatically
identify and track tags attached to objects. A mechanism that can be used with DAR location services.

Rooted - The process of modifying a device such that it allows users to attain administrative privileges
(i.e., root access).

Salt - A salt is random data that is added to a one-way function which hashes a password or passphrase
in order to defeat dictionary attacks and pre-computed rainbow tables.

Secure Erase - The process of removing specified data from a device via overwrite of that data.

SF - DAR solution architecture that features an SWFDE layer under the FE layer.

Software Full Disk Encryption (SWFDE) - A software product that provides Full Disk Encryption.

Storage Area Network (SAN) - A dedicated network that provides access to consolidated, block level
data storage. SANs devices appear like locally attached devices to the client operating system.

Supply Chain Risk Management (SCRM) - A program to establish processes and procedures to minimize
acquisition-related risks to critical acquisitions including hardware components and software solutions
from supply chain threats due to reliance on global sources of supply.

Unauthenticated State - The state an EUD is in when the identity of a user, user device, or other entity
has not been verified.

Unattended Operations (UO) - DAR use case that operates using a remote managed architecture to
manage an unattended DAR solution. Continuous physical control is defined by the AO.

Volume - A collection of separate units of logically divided media (partition) acting as a single entity that
has been formatted with a file system.

57
 APPENDIX B: ACRONYMS
Acronym Meaning
AA Authorization Acquisition
AES Advanced Encryption Standard
AK Authentication Key
AO Authorizing Official
ASPP Application Software Protection Profile
AU Auditing Requirements
BIOS Basic Input/Output System
CBC Cipher Block Chaining
CE Cryptographic Erase
CM Configuration Change Detection Requirements
CNSA Commercial National Security Algorithm
CNSS Committee on National Security Systems
CNSSD Committee on National Security Systems Directive
CNSSI Committee on National Security Systems Instruction
CNSSP Committee on National Security Systems Policy
COMSEC Communications Security
COTS Commercial Off-the-Shelf
CP Capability Package
cPP Collaborative Protection Profile
CR Configuration Requirement
CSfC Commercial Solutions for Classified
CSS Central Security Service
DAR Data-at-Rest
DEK Data Encryption Key
DIT Data-In-Transit
DM Device Management Requirements
D/NM Deputy National Manager
DSS Digital Signature Standard
ECC Elliptic Curve Cryptography
EE Encryption Engine
EM Enterprise Management
EU EUD Requirements

58
 Acronym Meaning
EUD End User Device
EP Extended Package
FAR False-Acceptance-Rate
FRR False-Rejection-Rate
FE File Encryption
FE EP File Encryption Extended Package
FEK File Encryption Key
FDE Full Disk Encryption
FIPS Federal Information Processing Standards
GA Generally Applicable
GCM Galois/Counter Mode
GD Requirements of Use and Handling of Solutions
GPS Global Positioning System
GOTS Government-off-the-Shelf
HAIPE High Assurance Internet Protocol Encryptor
HDD Hard Disk Drive
HF HWFDE and FE
HH HWFDE and HWFDE
HS HWFDE and SWFDE
HW Requirements for HWFDE Components
HWFDE Hardware Full Disk Encryption
IA Information Assurance
IAVA Information Assurance Vulnerability Alert
IAW In Accordance With
ISSO Information System Security Officer
ISV Independent Software Vendor
IT Information Technology
JIMS Joint Incident Management System
KEK Key Encryption Key
KM Key Management Requirements
KMIP Key Management Interoperability Protocol
LAN Local Area Network
LF Lost and Found

59
 Acronym Meaning
MA Mobile Access
MCE Mission Control Element
MDF Mobile Device Fundamentals
MoA Memorandum of Agreement
MSC Multi-Site Connectivity
NAS Network Attached Storage
NIAP National Information Assurance Partnership
NIST National Institute of Standards and Technology
NSA National Security Agency
NSS National Security Systems
O&M Operations and Maintenance
OCONUS Outside the Continental United States
OEM Original Equipment Manufacturer
OPSEC Operational Security
OS Operating System
PBE Pre-Boot Environment
PE Platform Encryption
PF PE and EE
PIN Personal Identification Number
PIV Personal Identity Verification
PMO Program Management Office
POC Point of Contact
PP Protection Profile
PS Product Selection
PUB Publication
RA Risk Assessment
RAID Redundant Array of Independent Disks
RAM Random Access Memory
RFID Radio Frequency Identification
RM Removable Media
RP Requirements for Incident Reporting
RPG Random Password Generation
RSA Rivest Shamir Adelman

60
 Acronym Meaning
SA Security Administrator
SAN Storage Attached Network
SCIF Sensitive Compartmented Information Facility
SCRM Supply Chain Risk Management
SED Self-Encrypting Drive
SF SWFDE and FE
SHA Secure Hash Algorithm
SHS Secure Hash Standard
SIM Subscriber Identity Module
SIPRNet Secret Internet Protocol Router Network
SR Solution Requirements
SSD Solid State Drive
SW Requirements for SWFDE
SWFDE Software Full Disk Encryption
T&E Test and Evaluation
TR Test Requirements
UAV Unmanned Aerial Vehicle
UUV Unmanned Underwater Vehicle
UEFI Unified Extensible Firmware Interface
UO Unattended Operations
U.S. United States
USB Universal Serial Bus
WLAN Wireless Local Area Network
WPS Wi-Fi Positioning System
XEX XOR Encrypt XOR
XOR Exclusive OR
XTS XEX-based tweaked-codebook mode with cipher text stealing

61
 APPENDIX C: CSFC INCIDENT REPORTING TEMPLATE
CSfC Incident Reporting Template

Point of Contact (POC) name, phone, email:

Alternate POC name, phone, email:

CSfC Registration Number:

Classification level of affected system:

Name of affected network(s):

Affected component(s) manufacturer/vendor:

Affected component(s) model number:

Affected component(s) version number:

Date and time of incident:

Description of incident:

Description of remediation activities:

62
 CSfC Incident Reporting Template

Is Technical Support from NSA Requested?

(Yes/No)

63
 APPENDIX D: PASSWORD/PASSPHRASE STRENGTH PARAMETERS
This appendix provides password and passphrase parameters for use in DAR products to address attacks
directly based on the strength of the password or passphrase. The information below, describes the
factors that provide strength to passwords and passphrases, and sets a minimum standard for use.

Strength

Entropy is used as a measure of strength for passwords and passphrases. According to NIST SP 800-63-
2, Electronic Authentication Guideline, entropy is a measure of the amount of uncertainty that an
attacker faces to determine the value of the secret. Entropy is usually stated in bits; for example, an
unpredictable password with 10 bits of entropy would have 210 or 1,024 possible combinations. The
greater the number of possible combinations, the greater the amount of time on average it will take an
attacker to find the correct password or passphrase.

Random vs. User Generated

Passwords and passphrases are required to be randomly generated as of DAR CP version 5.0. A
randomly generated value has the benefit that it will provide an objective amount of entropy, but can
be difficult for a user to remember. A user generated value may be easier to remember, but may be
predictable, therefore, lowering the entropy calculation reducing the strength of the password or
passphrase. If random generation is not a workable solution for the mission use case, then a deviation
from the DAR CP is required. There are many suggested methods for the user generation of passwords;
more information on these can be found in NIST SP 800-118, Guide to Enterprise Password
Management. These methods attempt to reduce the predictability while maintaining length and
memorability, but because they are user chosen, they are all still at risk of being predicable. If the
password or passphrase is predicable, an attacker could try a much shorter list of common or personal
values, reducing the average time to find the correct password or passphrase. The most effective way
to ensure the password or passphrase has an appropriate amount of entropy is by applying random
generation. The remainder of this appendix addresses random generation.

Randomly Generated Passwords

The strength of a password is determined by the character set and the length. The character set
describes the group of unique characters that may be chosen to create the password, such as numbers,
lower case letters, upper case letters, special characters, etc. The length simply describes the number of
characters chosen.

Randomly Generated Passphrases

The strength of a passphrase is determined by the number of words in the passphrase and the number
of words in the word list, the pool of unique words that can be chosen for the passphrase. The word list
can be adjusted by the properties of the words it includes, such as minimum word length, maximum
word length, and complexity (includes factors such as the difficulty of the word, capitalization, character
substitutions, etc.) per word. Each property has a tradeoff between strength and usability. A minimum
word length of four is recommended to maintain the effectiveness of the passphrase. This is based on
entropy per word from a word list ranging from 10,000 to 450,000, and entropy per character from a

64
 character set of 26. This ensures the entropy per set of characters of a given word is greater than the
entropy provided from selecting a word from the word list.

Multi-Factor Authorization

If a password/passphrase is being used as part of a multi-factor authorization solution per DAR-CR-14

and another factor is being used as a primary factor for that component, then the password or
passphrase does not need to comply with these rules. It is still recommended to comply with these
rules. If the other factor is not a primary factor and used as secondary, these rules still apply.

Assumptions

The product is assumed to meet one of the DAR protection profiles. All password and passphrase
conditioning assumes salting is performed, making pre-computed attacks infeasible. A salt is a random
value that is used in a cryptographic process to ensure that the results of the computations for one
instance cannot be reused by an attacker. The product is assumed to be kept up to date and the
protection mechanisms used in calculations cannot be bypassed.

Minimum Strength Calculations

CSfC provides a tool for random generation, which is available on GitHub at

https://github.com/nsacyber/RandPassGenerator. This tool must be used to generate random
passwords and passphrases. When using this tool to generate passwords and passphrases, it should be
ran on a network capable of protecting the classification of the data that is being protected. The tool
should be sent to the appropriate classified network through a Data Transfer Agent (DTA) for further
use. During registration instructions on how to download, verify, and use the tool will be provided.
Alternatively, contact the CSfC PMO at [email protected] for further instructions. The provided tool
is set to a default strength of 160 bits, this may be set lower, but must not be set below 112 bits. If
using custom word lists or character sets and not using the provided tool, Tables 21 and 22 show the
required minimum length of a password and passphrase given a set of characters or words. The
provided tool is capable of utilizing custom word lists. The user must define the size of the character set
or word list they will use. To use the tables, find the value that is less than or equal to your character set
(or word list) size in the Character Set Size (or Word List Size) column and the corresponding value in the
Minimum Password Length (or Minimum Passphrase Length) column for that row reflects the minimum
password (or passphrase) length that must be used.

Table 21: Randomly Generated Minimum Password Length

Randomly Generated Passwords
Character Set Size Minimum Password Length
75 16
58 17
47 18
38 19
32 20
27 21
23 22

65
 Randomly Generated Passwords
Character Set Size Minimum Password Length
21 23
18 24
16 25
15 26
13 27
12 28
11 29
10 30

Table 22: Randomly Generated Minimum Passphrase Length

Randomly Generated Passphrases
Minimum Passphrase
Word List Size Length
1000000 5
100000 6
20000 7
6000 8
2200 9
1000 10

66
 APPENDIX E: CONFIGURATION GUIDANCE
A number of the DAR requirements listed in the main body of this CP might require additional
configuration information in order to be fully understood by customers who are using them as the basis
for preparing Registration Packages for CSfC solutions. The list that follows, includes the additional
information on requirements that may provide further guidance in order for customers to prepare and
complete the Registration Packages. Please reference the requirements tables to see which solution
design and use case these requirements apply. If there are questions about requirements that are not
discussed in this list, please submit questions to the DAR CP maintenance team and a response will be
provided for considerations in future updates of the CP.

Requirement Clarification
DAR-SR-1: Default accounts, passwords, Not all products will have defaults. If the product does not have a default,
community strings, and other default access no action is needed for compliance with this requirement. If defaults do
control mechanisms for all components exist, vendors are required to provide guidance on how to change their
must be changed or removed. authentication factors during their protection profile validation. Please
refer to the NIAP Product Compliant List, select your component, and view
the Administrative Guidance document.

DAR-SR-6: The AO must provide procedures Vendors are required to provide guidance on how to perform a
for performing CE. cryptographic erase of the encrypted data, this may also be referred to as
changing the DEK or TSF. Wipe during their protection profile validation.
Please refer to the NIAP Product Compliant List, select your component,
and view the Administrative Guidance document.

DAR-CR-1: Default encryption keys must be Not all products have default encryption keys. If the keys are generated
changed. upon provisioning, no action is needed for compliance with this
requirement. If default encryption keys do exist, please follow the
guidance in DAR-SR-6 to perform a cryptographic erase along with any
other vendor guidance provided.

DAR-CR-3: DAR components must use Not all products allow for changing of algorithms or key sizes used. If
algorithms for encryption selected from more than one is supported, the vendor is required to provide guidance
Table 1. for selecting those options. Please refer to the NIAP Product Compliant
List, select your component, and view the Administrative Guidance
document. If no options are listed, you can confirm vendor algorithm and
key size selection in the Security Target document, which is posted on the
NIAP page.

67
 Requirement Clarification
DAR-CR-4: Each DAR component must Vendors may not include this functionality in their product, however, they
prevent further authentication attempts may include other non-configurable mitigations.
after a number of failed attempts defined by
the AO. For full disk encryptors, vendors are required to provide one of the
following options: Cryptographic erase, forced delay between attempts, or
institute a block after a number of consecutive attempts. If your FDE
consists of two products, these settings are required for the EE and
optional for the AA. The vendor is required to provide guidance for any
configurable limits. Please refer to the NIAP Product Compliant List, select
your component, and view the Administrative Guidance document. For
information on the selection the vendor made please refer to the Security
Target document.

For file encryptors, vendors are not required to provide this functionality.
Refer to any guidance provided by the vendor.

For platform encryption products, vendors are required to implement

throttling between authentication attempts. There is no configuration
needed for this. The vendor is also required to provide for a cryptographic
erase of all protected data upon a configurable number of failed
authentication attempts. The vendor is required to provide guidance on
how to configure the number of attempts. Please refer to the NIAP
Product Compliant List, select your component, and view the
Administrative Guidance document.

DAR-CR-5: Each DAR layer must perform a Please refer to the guidance given in DAR-CR-4.
CE after a number of consecutive failed
logon attempts as defined by the AO.
DAR-CR-10: All CSfC components must be Please refer to the NIAP Product Compliant List, select your component,
implemented (configured) using only their and view the Administrative Guidance document for assistance in
NIAP-approved configuration settings. configuring the product into a compliant state.

68
 Requirement Clarification
DAR-CR-11: Users must be restricted to There are multiple ways to accomplish this requirement depending on the
designated user folders. OS and software used; any method is acceptable. Here are common ways
to accomplish this on the most used Operating Systems. For all restricted
directories do the following:

Linux: Run the Nautilus file browser, right click the folder, and select
Properties. In the Permissions tab, change the Access drop down to Read
Only for all end users, then select Change.

Mac: Select the folder, select File, click the arrow next to the gear icon to
display further options, select Get Info, and click the drop down for sharing
and permissions (may have to scroll all the way to the bottom down to see
this). For all end users and groups on the device: Select the user or group
and choose Read Only. Then select the gear icon and apply all changes.
For additional information please see this page:
https://support.apple.com/kb/PH18894?locale=en_US&viewlocale=en_US

Windows: Right click on the folder, select Properties, and select the
Security tab. For all end users and groups on the device: Select the user or
group, check the Deny box for the write permission and then click Apply.
For additional information please see this page:
https://technet.microsoft.com/en-us/library/bb456977.aspx

DAR-SW-3:The SWFDE must be configured Reference random password/passphrase generation tool, reference multi-
to use one of the following primary authentication section.
authentication options:
 A randomly generated passphrase or
password that meets the minimum
strength set in Appendix D, or
 A randomly-generated bit string
equivalent to the cryptographic
strength of the DEK contained on an
external USB token or
 An external smartcard or software
capability containing a software
certificate with RSA or Elliptic Curve
Cryptography (ECC) key pairs per
Table 1.
Any combination of the above.

69
 Requirement Clarification
DAR-FE-3: The FE product must use one of Reference random password/passphrase generation tool, reference multi-
the following primary authentication authentication section.
options:
 A randomly generated passphrase or
password that meets the minimum
strength set in Appendix D, or
 A randomly-generated bit string
equivalent to the cryptographic
strength of the DEK contained on an
external USB token or
 An external smartcard or software
capability containing a software
certificate with RSA or Elliptic Curve
Cryptography (ECC) key pairs per
Table 1.
Any combination of the above.

DAR-PE-1: The PE must enable the “wipe Vendors are required to provide guidance on how to perform the “wipe
sensitive data” management function for sensitive data” management function; this may also be referred to as TSF
imported or self-generated keys/secrets Wipe, during their protection profile validation. Please refer to the NIAP
and/or other classified data. Product Compliant List, select your component, and view the
Administrative Guidance document.

DAR-PE-5: The PE must use the following for Reference random password/passphrase generation tool, reference multi-
authentication: authentication section.

 A minimum of a six-character, case

sensitive alphanumeric password
with the length and complexity as
defined by the AO, or
 A passphrase with the length and
complexity as defined by the AO.
DAR-HW-3: The HWFDE must be configured Reference random password/passphrase generation tool, reference multi-
to use one of the following primary authentication section.
authentication options:
 A randomly generated passphrase or
password that meets the minimum
strength set in Appendix D, or
 A randomly-generated bit string
equivalent to the cryptographic
strength of the DEK contained on an
external USB token or
 An external smartcard or software
capability containing a software
certificate with RSA or ECC key pairs

70
 Requirement Clarification
per Table 1.
Any combination of the above.

DAR-EU-9: The Security Administrator (SA) There are multiple ways to accomplish this requirement depending on the
must disable system power saving states on OS and software used; any method is acceptable. Here are common ways
EUDs (e.g., sleep and hibernate). to accomplish this on the most used Operating Systems:

Linux: Please refer to vendor guidance on your specific distribution.

Mac: Open Apple menu, select system preference, and select Energy
Saver. For all power plans: Drag to slider for computer sleep to Never for
sleep. Display sleep does not need to be changed.

Windows: Open control panel. Select small icons in the top right, then
select power options. For all power plans: Select Change Plan Settings,
select Change Advanced Power Settings, expand the sleep list and set
Sleep and Hibernate to Disabled.

DAR-EU-10: The EUD must power off after a There are multiple ways to accomplish this requirement depending on the
period of inactivity defined by the AO. OS and software used, any method is acceptable. Here are common ways
to accomplish this on the most used Operating Systems:

Linux: Please refer to vendor guidance on your specific distribution.

Mac: This function must be provided by third party software or running a

script.

Windows: Open task scheduler, select create task. Under the general tab:
Fill in a name. Select run whether user is logged on or not. Make sure run
with highest privileges is checked. Under the triggers tab: Click new, select
daily, then select ok. Under the actions tab: click new and enter shutdown
in the program/script box. Enter /l and /f under the add arguments
(optionally) box. Under the conditions tab: Check the box for start the task
only if the computer is idle for and Under the settings tab: Check the box if
the task fails restart every time and select an increment shorter than the
AO defined period. Uncheck the box start the task only if the computer is
on AC power. Under the attempt to restart up to box enter 999.

71
 Requirement Clarification
DAR-EU-14: System folders must have user Please refer to guidance on DAR-CR-11 and ensure end users are restricted
write permissions disabled unless authorized from writing to system folders.
by an administrator.

DAR-EU-20: The BIOS must be configured to This is a password to continue the boot process, creating a password
require a password before continuing the prompt before the FDE and/or OS login. Not all motherboards support
boot process. this feature. Generally the first screen will indicate which button is used to
change BIOS/UEFI settings; if not please refer to product documentation.
Once in the settings, browse for an option to require a password to
continue the boot process and enable it. The password does not need to
be strong, the absence of the password would indicate tampering.

DAR-EU-21: All DAR FDE components must Please refer to DAR-SR-6 guidance on how to perform a cryptographic
be cryptographically erased before being erase.
provisioned again.

DAR-EU-22: All DAR components must be Please refer to DAR-SR-6 guidance on how to perform a cryptographic
cryptographically erased before being erase.
provisioned again.

DAR-EU-24: If supported, the EUD must have This is a password that is required before allowing access to change
the BIOS/ UEFI password enabled. BIOS/UEFI settings. Not all motherboards support this feature. Generally
the first screen will indicate which button is used to change BIOS/UEFI
settings; if not please refer to product documentation. Once in the
settings, browse for an option to set a BIOS/UEFI password.

DAR-EU-26: Each EUD must be personalized Personalization means making device changes specific to each end user
by the end user. (This should not violate any that would be noticed before both layers are authenticated. This can
other security features.) include stickers, markings, wallpapers, etc.

DAR-EU-36: Each EUD must be personalized Please refer to DAR-EU-26 to personalize devices.
by the end user. (This should not violate any
other security features.) (previously DAR-LF-
12)

DAR-EU-37: EUDs must use boot integrity This requirement is based on device acquisition. Not all devices support
verification technology. (previously DAR-LF- these features. The specific features will have to be discussed with the
5) device vendor and then configured according to that vendor’s
specifications.

DAR-KM-3: The DAR solution must disable If the product supports key recovery mechanisms they are required to
all key recovery mechanisms. state how to disable those mechanisms in their documentation. Please
refer to the NIAP Product Compliant List, select your component, and view
the Administrative Guidance document. Does not apply to the EM use
case.

72
 APPENDIX F: CONTINUOUS PHYSICAL CONTROL
Since the NSA requires that implementing organizations define the circumstances in which an EUD that
is part of the solution is considered outside of the continuous physical control of authorized users (i.e.,
"lost"), Authorizing Officials will define “continuous physical control”, and that definition should align
with the intended mission and threat environment for which the solution will be deployed.
Organizations must also define the circumstances in which an EUD that is a part of that organization's
solution is to be considered recovered back into the continuous physical control of authorized users (i.e.,
"found").

In order to provide some guidance to clients who may not have experience with handling continuous
physical control issues, we have consulted several experienced organizations that have provided
examples of the criteria they use to define “continuous physical control”. The intent of this is to cite a
number of potential generic measures that can be taken as additional Continuous Physical Control
guidance without attribution to the source of these measures.

DAR customers have provided several ideas of measures they are considering to deal with particular
circumstances. Listed are some of the ideas being considered for handling EUDs:
 Package using clear one-use bags.
 Use tamper evident stickers with recorded unique serial numbers on critical screws.
 Use commercial backpacks with-pick-resistant locks.
 Lock in automobile glove box and lock the car.

Some users currently handle the issue by simply not authorizing the removal of any EUDs from the
secure location where they are housed.

Continuous Physical Control Examples:

(U) To assist in the development of well thought out definitions of continuous physical control,
segments of good definitions used by previous registrations have been provided. These examples
should be reviewed to ensure that the definition given for a registration follows the intent of the
requirement.

Traveling with EUDs. Commands will create local policy to address specifics on traveling with EUDs, to
include outside the continental U.S. (OCONUS) locations, in accordance with (IAW) local security
procedures.

The following general actions apply while traveling with EUDs:

a. Prior to travel:
(1) Do not take your device if you can do without it.
(2) Do not take information you do not need, including sensitive contact information.
(3) Ensure that the latest, most current, up-to-date antivirus protection, spyware
protection, OS security patches, and a personal firewall have been pushed and enabled
by the responsible Information Technology (IT) support.
(4) Disable infrared ports and features you do not need.

b. During travel:

73
 (1) Keep the EUD under physical control at all times when traveling.
(2) Never place the EUD in checked luggage.
(3) Never store the EUD in an airport, train station, bus station, or any public locker.
(4) If leaving the EUD in a vehicle that an AO determines is sufficient to keep the EUD
safe, then the EUD should be kept out of sight.
(5) Do not leave EUDs unattended unless required activities demand so. In the event
that they must be unattended, stow them securely and out of sight after removing the
battery and SIM card. Keep the battery and SIM card under control at all times to
maintain the protection of its information.
(6) Avoid leaving the EUD in a hotel room.
(7) Be prepared for airport security checks. Have the EUD’s batteries charged or a
power cord handy to demonstrate if necessary that it is functional.
(8) Heighten vigilance at any security or luggage-scanning checkpoint. Place EUD on the
conveyer belt only after the belongings of the person ahead of you have cleared the
scanner. If delayed, keep the EUD in view.
(9) Exercise diligence when traveling in foreign countries because criminals or local
intelligence may target the EUD for the information it contains.
(10) Do not display any sensitive information on the EUD screen when in any public
place (such as an airport terminal, train or bus station, airplane, train, bus, or taxi).
(11) Terminate connections when not using them.
(12) If the device or information is stolen, report it immediately to your home
organization and the local U.S. embassy or consulate.

c. Return from travel:

(1) Change the password.
(2) Have your command or unit examine the device for the presence of malicious
software.

74
 APPENDIX G: REFERENCES
Application Protection Profile for Application Software Version 1.2. (File Encryption April 2016
Software PP component). www.niap-ccevs.org/Profile/Info.cfm?id=394
Campus WLAN Campus WLAN CP. Available on the CSfC web page
Latest
CP https://www.nsa.gov/resources/commercial-solutions-for-classified-
version
program/capability-packages

CNSSD 505 CNSS Directive (CNSSD) Number 505, Supply Chain Risk Management March 2012
(SCRM)

CNSSI 1253 CNSS Instruction No. 1253, Security Categorization and Control Selection March 2014
for National Security Systems

CNSSI 4004 Committee on National Security Systems Instruction (CNSSI) No. 4004 January
Destruction and Emergency Protection Procedures for COMSEC and 2008
Classified Material

CNSSI 4009 CNSSI 4009, Committee on National Security Systems (CNSS) Glossary April 2015
www.cnss.gov/Assets/pdf/cnssi_4009.pdf

CNSSP 15 CNSS Policy (CNSSP) Number 15, National Information Assurance Policy on October
the Use of Public Standards for the Secure Sharing of Information Among 2016
National Security Systems Committee for National Security Systems

CSfC CSfC Components List. Available on the CSfC web page Current
Components List https://www.nsa.gov/resources/commercial-solutions-for-classified- update
program/components-list

FDE cPP AA Collaborative Protection Profile for Full Disk Encryption- Authorization September
Acquisition Version 2.0. (Software or Hardware FDE component) 2016
www.niap-ccevs.org/Profile/Info.cfm?id=406

FDE cPP EE Collaborative Protection Profile for Full Disk Encryption- Encryption Engine September
Version 2.0. (Software or Hardware FDE component) www.niap- 2016
ccevs.org/Profile/Info.cfm?id=407

FE Module PP-Module for File Encryption. (File Encryption component) www.niap- July 2019
ccevs.org/Profile/Info.cfm?id=415

FE EM Module PP-Module for File Encryption Enterprise Management. (File Encryption July 2019
Enterprise Management component) www.niap-
ccevs.org/Profile/Info.cfm?id=427

75
 FIPS 140-2 Federal Information Processing Standard 140-2, Security Requirements for May 2001
Cryptographic Modules

FIPS 180-4 Federal Information Processing Standard 180-4, Secure Hash Standard August 2015
(SHS)

FIPS 186-4 Federal Information Processing Standard 186-3, Digital Signature July 2013
Standard (DSS), (Revision of FIPS 186-2, June 2000)

FIPS 197 Federal Information Processing Standard 197, Advanced Encryption November
Standard (AES) 2001

FIPS 201-2 Federal Information Processing Standard 201, Personal Identity August 2013
Verification (PIV) of Federal Employees and Contractors National Institute
for Standards and Technology FIPS Publication
http://csrc.nist.gov/publications/fips/fips201-1/FIPS-201-1-chng1.pdf
MA CP Mobile Access CP. Available on the CSfC web page Latest
https://www.nsa.gov/resources/commercial-solutions-for-classified- version
program/capability-packages

MDF PP Protection Profile for Mobile Device Fundamentals. (Platform Encryption June 2017
component)- www.niap-ccevs.org/Profile/Info.cfm?id=417

MSC CP Multi-site Connectivity CP. Available on the CSfC web page Latest
https://www.nsa.gov/resources/commercial-solutions-for-classified- version
program/capability-packages

NIAP Product NIAP Product Compliant List. www.niap-ccevs.org/products/ Current

Compliant List update

NIST SP 800-111 NIST Special Publication 800-111, Guide to Storage Encryption November
Technologies for End User Devices 2007

NIST SP 800- NIST Special Publication 800-131A, Recommendation for Transitioning of November
131A (Rev 1) Cryptographic Algorithms and Key Lengths. E. Barker. 2015

NIST SP 800-132 Recommendation for Password-Based Key Derivation December

2010

NIST SP 800-147 NIST Special Publication 800-147, BIOS Protection Guidelines. D. Cooper, April 2011
et. al.

NIST SP 800-56A NIST Special Publication 800-56A, Recommendation for Pair-Wise Key April 2018
(Rev 3) Establishment Schemes Using Discrete Logarithm Cryptography. E. Barker,
D. Johnson, and M. Smid

76
 NIST SP 800-56B NIST Special Publication 800-56B, Recommendation for Pair-Wise Key March 2019
(Rev 2) Establishment Schemes Using Integer Factorization Cryptography. E.
Barker, et. al.

NIST SP 800-56C NIST Special Publication 800-56C, Recommendation for Key Derivation August 2020
(Rev 2) through Extraction-then-Expansion. L. Chen.

NIST SP 800-63-2 NIST Special Publication 800-63-2, Electronic Authentication Guideline August 2013

NSA CNSA NSA Guidance on CNSA Cryptography No Date

https://www.iad.gov/iad/programs/iad-initiatives/cnsa-suite.cfm Specified

NSA/CSS Policy NSA/CSS Storage Device Declassification

Manual 9-12 https://www.nsa.gov/ia/_files/government/MDG/NSA_CSS_Storage_Devi
December
Storage Device ce_Declassification_Manual.pdf
2014
Sanitization

OS PP Protection Profile for General Purpose Operating Systems. March 2016

https://www.niap-ccves.org/Profile/Info.cfm?id=400

77