CryptoComply for Java 140-3

Certificate

Approved Algorithms (83)

Derived Review-Risk Graph (review prompts, not findings)

flowchart LR
  %% Deterministic review-risk graph for CryptoComply for Java 140-3
  %% Review prompts and evidence gaps, NOT vulnerability findings.
  subgraph CMVP["CMVP-disclosed clues"]
    C2["[low] Firmware update / recovery<br/>/ rollback (referenced in<br/>text)<br/><i>recovery</i>"]
    C3["[low] Self-test / status surface<br/>(referenced in text)<br/><i>Status Output<br/>Show Status<br/>self-test</i>"]
    C5["[low] Protocol / secure-channel<br/>references (may be KDF<br/>names, not a live channel)<br/><i>TLS<br/>SSH<br/>IKEV</i>"]
    C6["[low] Operating system / runtime<br/>referenced (boundary<br/>membership not asserted)<br/><i>operating system<br/>linux<br/>application</i>"]
  end
  subgraph Inference["Derived inference"]
    I2["Possible only, trusted<br/>code is reachable through<br/>update and recovery paths."]
    I3["Possible only, some<br/>services may process input<br/>before, or without,<br/>operator authentication."]
    I5["Possible only, a protocol<br/>is referenced, but whether<br/>it is a live channel or<br/>only a KDF/algorithm name<br/>is unconfirmed."]
    I6["Possible only, a<br/>runtime/OS is referenced,<br/>but its membership in the<br/>cryptographic boundary is<br/>not established."]
  end
  subgraph Risk["Reviewer question"]
    R2["Are update images<br/>authenticated before<br/>parsing, and are<br/>downgrade/rollback paths<br/>constrained?"]
    R3["Can unauthenticated<br/>services leak state,<br/>consume resources, or<br/>transition security state?"]
    R5["If a live TLS/SSH/IKE<br/>channel exists, could<br/>library CVEs apply, or is<br/>this only a<br/>KDF/documentation name?"]
    R6["If the OS/runtime is<br/>in-boundary, could its<br/>CVEs be hidden by<br/>firmware-only versioning?"]
  end
  subgraph Evidence["Evidence needed to close"]
    E2["confirm the disclosure<br/>itself (keyword hit,<br/>context unverified) ·<br/>update image format ·<br/>signature-before-parse<br/>proof · anti-rollback /<br/>downgrade policy"]
    E3["confirm the disclosure<br/>itself (keyword hit,<br/>context unverified) ·<br/>pre-auth reachability<br/>matrix · rate limits and<br/>output redaction ·<br/>abuse-case tests"]
    E5["confirm the disclosure<br/>itself (keyword hit,<br/>context unverified) ·<br/>library identity and<br/>version ·<br/>certificate-validation<br/>behaviour · protocol-CVE<br/>disposition"]
    E6["confirm the disclosure<br/>itself (keyword hit,<br/>context unverified) ·<br/>runtime identity and<br/>config · kernel/runtime<br/>hardening profile ·<br/>patch/backport manifest"]
  end
  C2 --> I2 --> R2 --> E2
  C3 --> I3 --> R3 --> E3
  C5 --> I5 --> R5 --> E5
  C6 --> I6 --> R6 --> E6
  classDef clue fill:#eef3f9,stroke:#6f7f91,color:#1f3a5f;
  classDef infer fill:#fff7e6,stroke:#b98500,color:#6b4e00;
  classDef risk fill:#fbe9e9,stroke:#b02a2a,color:#7a1f1f;
  classDef evidence fill:#e6f4ea,stroke:#1e7d34,color:#14532d;
  class C2,C3,C5,C6 clue;
  class I2,I3,I5,I6 infer;
  class R2,R3,R5,R6 risk;
  class E2,E3,E5,E6 evidence;

flowchart LR
  %% Deterministic clue tier for CryptoComply for Java 140-3
  %% confidence: high = structured record field; medium = structured but soft; low (dashed) = bare keyword hit, context unverified
  subgraph CMVP["CMVP-disclosed clues (deterministic)"]
    C2["[low] Firmware update / recovery / rollback (referenced in text)<br/><i>recovery</i><br/>src: text:keyword"]
    C3["[low] Self-test / status surface (referenced in text)<br/><i>Status Output<br/>Show Status<br/>self-test</i><br/>src: text:keyword"]
    C5["[low] Protocol / secure-channel references (may be KDF names, not a live channel)<br/><i>TLS<br/>SSH<br/>IKEV</i><br/>src: text:keyword"]
    C6["[low] Operating system / runtime referenced (boundary membership not asserted)<br/><i>operating system<br/>linux<br/>application</i><br/>src: text:keyword"]
  end
  classDef clueHigh fill:#eef3f9,stroke:#2f6fb0,stroke-width:2px,color:#1f3a5f;
  classDef clueMedium fill:#eef3f9,stroke:#6f7f91,color:#1f3a5f;
  classDef clueLow fill:#f7f7f7,stroke:#999,stroke-dasharray:4 4,color:#444;
  class C2,C3,C5,C6 clueLow;

Security Policy, page by page

SafeLogic Inc. CryptoComply for Java 140-3 Software Version 4.0.0 Document Version 1.1 January 16, 2025 SafeLogic Inc.

530 Lytton Ave, Suite 200

Palo Alto, CA 94301 www.safelogic.com

This document provides a non-proprietary FIPS 140-3 Security Policy for CryptoComply for Java 140-3. SafeLogic Inc.'s CryptoComply for Java 140-3 is designed to provide FIPS 140-3 validated cryptographic functionality and is available for licensing. For more information, visit www.safelogic.com/cryptocomply.

1.1.1 About FIPS 140

Federal Information Processing Standards Publication 140-3, Security Requirements for Cryptographic Modules, (FIPS 140-3) specifies the latest requirements for cryptographic modules utilized to protect sensitive but unclassified information. The National Institute of Standards and Technology (NIST) and Canadian Centre for Cyber Security (CCCS) collaborate to run the Cryptographic Module Validation Program (CMVP), which assesses conformance to FIPS 140. NIST (through NVLAP) accredits independent testing labs to perform FIPS 140 testing. The CMVP reviews and validates modules tested against FIPS

140 criteria. Validated is the term given to a module that has successfully gone through this FIPS 140

validation process. Validated modules receive a validation certificate that is posted on the CMVP’s website. More information is available on the CMVP website at: https://csrc.nist.gov/projects/cryptographic-module-validation-program.

This non-proprietary cryptographic module Security Policy for CryptoComply for Java 140-3 from SafeLogic Inc. (SafeLogic) provides an overview of the product and a high-level description of how it meets the security requirements of FIPS 140-3. This document includes details on the module’s cryptographic capabilities, services, sensitive security parameters, and self-tests. This Security Policy also includes guidance on operating the module while maintaining compliance with FIPS 140-3. CryptoComply for Java 140-3 may also be referred to as “the module” in this document.

The SafeLogic website (www.safelogic.com) contains information on SafeLogic services and products. The CMVP website maintains all FIPS 140 certificates for SafeLogic’s FIPS 140 validations. These certificates also include SafeLogic contact information.

This document may be freely reproduced and distributed, but only in its entirety and without modification.

Table 1 lists the module’s level of validation for each area in FIPS 140-3. Table 1 - Security Levels

Purpose and Use: CryptoComply for Java 140-3 is a standards-based “Drop-in Compliance™” cryptographic module for Java environments. The module delivers cryptographic services to host applications through a Java language Application Programming Interface (API). Module Type: Software Module Embodiment: Multi-Chip Stand Alone Cryptographic Boundary: The cryptographic boundary is the Java Archive (JAR) file, ccj-4.0.0.jar. The module is the only component within the cryptographic boundary and the only component that carries out cryptographic functions covered by FIPS 140-3. The module classes are executed on the Java Virtual Machine (JVM) using the classes of the Java Runtime Environment (JRE). The JVM is the interface to the computer’s Operating System (OS), which is the interface to the various physical components of the general purpose computer (GPC). As a software cryptographic module, the module operates within the Tested Operational Environment’s Physical Perimeter (TOEPP). The TOEPP physical perimeter is the physical perimeter of the GPC that the module operates on. The TOEPP includes the JVM/JRE, OS, and the GPC. The TOEPP includes the Operational Environment (OE) that the module operates in, the module itself, and all other applications that operate within the OE, including the host application for the module. The external entropy source used by the module is also within the TOEPP. The module’s block diagram is provided in Figure 1, which shows the cryptographic boundary and the logical relationship of the cryptographic module to the other software and hardware components of the TOEPP. The module’s logical interfaces are defined by its API.

Figure 1 - Module Block Diagram

Tested Module Identification – Software, Firmware, Hybrid (Executable Code Sets): Table 2 - Executable Code Sets

Confirming the Module Checksum, Functionality, and Versioning The module checksum, functionality, and versioning can be confirmed by executing the command: java -cp ccj-4.0.0.jar com.safelogic.cryptocomply.util.DumpInfo which should display: Version Info: CryptoComply® for Java version v4.0.0 FIPS Ready Status: READY Module SHA-256 HMAC: c5f6e9c3593f67ea87f08b91590c7531c53ac2540685b921807c9e82581911ee This display indicates that the JAR represents the software release ccj-4.0.0, that it has successfully passed all its startup tests, and that the software release is confirmed to have the HMAC listed above. Tested Operational Environments - Software, Firmware, Hybrid: The module operates in a modifiable operational environment under the FIPS 140-3 definitions. The cryptographic module was tested on the following operational environments on the GPC platforms detailed in Table 3. Table 3 - Tested Operational Environments

• Software/Firmware/Hybrid Vendor-Affirmed Operational Environments - Software, Firmware, Hybrid: Porting guidance is defined in the FIPS 140-3 CMVP Management Manual Section 7.9. The cryptographic module will remain compliant with the FIPS 140-3 validation when operating on any GPC provided that:
• No source code modifications were made

• The module operates on any general-purpose platform/processor that supports the specified operating system as listed on the validation entry. Or the module uses another compatible platform, such as one of the Java SE Runtime Environments (or equivalent OpenJDK Runtime Environments) listed in the table below (Table 4). The CMVP makes no statement as to the correct operation of the module or the security strengths of the generated keys when so ported if the specific operational environment is not listed on the validation certificate. Table 4 - Vendor Affirmed Operational Environments
• Software/Firmware/Hybrid

Not applicable. Modes List and Description: Table 5 - Modes of Operation

Mode Change Instructions and Status: In default operation the module will start with all algorithms and services enabled. If the module detects that the system property com.safelogic.cryptocomply.fips.approved_only is set to true the module will start in approved mode and non-approved mode functionality will not be available. The module optionally uses the Java SecurityManager. If the underlying JVM is running with a Java SecurityManager installed the module starts in approved mode by default with secret and private key export disabled. When the module is not used within the context of the Java SecurityManager, it will start by default in the non-approved mode. Refer to Security Policy Section 11.3 for additional information about the Java SecurityManager. Refer to Security Policy Section 11.4.1 for additional information on the module’s mode of operation rules.

The module implements the algorithms specified in the tables below. The module supports both an Approved mode and a Non-approved mode of operation. Please see Security Policy Section 2.4 for additional details on the modes of operation and the configuration of the Approved mode of operation. Please see Security Policy Section 11.1 for Initialization steps.

The module implements the following approved algorithms that have been tested by the Cryptographic Algorithm Validation Program (CAVP). There are algorithms, modes, and keys that have been CAVP tested but not used by the module. Only the algorithms, modes/methods, and key lengths/curves/moduli shown in this table are used by the module. Table 6 - Approved Algorithms, CAVP Tested

GCM encryption with an internally generated IV, see Security Policy Section 2.6.1 concerning external IVs. IV generation is compliant with IG C.H. Keys are not established directly into the module using key agreement or key transport algorithms.

DSA signature generation with SHA-1 is only for use with protocols. Key size only used for Signature Verification Legacy testing for signatures not specified under FIPS 186-5.

s: Keys are not established directly into the module using key agreement or key transport algorithms.

Note: CAVP testing is not provided for use of the PRFs SHA-512/224 and SHA-512/256. These must not be used in approved mode.

No parts of the protocols (TLS, SNMPv3, SSHv2, X9.63, IKEv2, SRTP), other than the approved cryptographic algorithms and the KDFs, have been reviewed or tested by the CAVP and CMVP

Legacy testing for signatures not specified under FIPS 186-5 (all moduli for ANSI X9.31, and testing with 1024 or

Table 7 - Vendor Affirmed Algorithms

Non-Approved, Allowed Algorithms: Not applicable.

Non-Approved, Allowed Algorithms with No Security Claimed. These algorithms are Allowed in Approved mode. Table 8 - Non-Approved, Allowed Algorithms with No Security Claimed

Table 9 - Non-Approved, Not Allowed Algorithms Support for additional modes of operation. Support for additional key sizes and the establishment of keys of less than 112 bits of security strength. Deterministic signature calculation, support for additional digests, and key sizes. Deterministic signature calculation, support for additional digests, and key sizes.

Keys are not directly established into the module using key agreement or transport techniques. Support for additional digests and signature formats, PKCS#1 1.5 key wrapping, support for additional key sizes. Support for additional key sizes and the establishment of keys of less than 112 bits of security strength.

IVs for GCM can be generated randomly, or via a FipsNonceGenerator. IV generation is compliant with IG C.H. Where an IV is not generated within the module the module supports the importing of GCM IVs. In approved mode, importing a GCM IV for encryption that originates from outside the module is nonconformant. In approved mode, when a GCM IV is generated randomly, the module enforces the use of an approved DRBG in line with Section 8.2.2 of SP 800-38D. In approved mode, when a GCM IV is generated using the FipsNonceGenerator, a counter is used as the basis for the nonce and the IV is generated in accordance with TLS protocol. Rollover of the counter in the FipsNonceGenerator will result in an IllegalStateException indicating the FipsNonceGenerator is exhausted and (as per IG C.H) where used for TLS 1.2, rollover will terminate any TLS session in process using the current key and the exception can only be recovered from by using a new handshake and creating a new FipsNonceGenerator. A service indicator for IV usage is provided in the module through Java logging. Setting the logging level to Level.FINE for the named logger com.safelogic.cryptocomply.jcajce.provider.BaseCipher will produce a log message when an IV which may have been produced outside the module and/or not from a compliant source is detected. The log message will be of the standard form including the detail: FINE: Passed in GCM nonce detected: <IV value> where <IV value> is a HEX representation of the IV in use. Setting the logging level to Level.FINER will produce an additional log message for any GCM IV which is used if the previous Level.FINE message is not activated. Log messages in this case will show the detail as:

FINER: GCM nonce detected: <IV value> where <IV value> is a HEX representation of the IV in use. Per IG C.H, this Security Policy also states that in the event module power is lost and restored the consuming application must ensure that any of its AES GCM keys used for encryption or decryption are re-distributed. The AES GCM mode falls under:

• IG C.H scenario 2: GCM IV is generated randomly, and the module uses an Approved DRBG that is internal to the module’s boundary. The IV length is 96 bits.
• IG C.H scenario 1 for TLS v1.2 protocol: The module is compatible with the TLS v1.2 protocol and supports acceptable AES GCM ciphersuites from Section 3.3.1 of the SP 800-52r2.

2.6.2 Enforcement and Guidance for Use of the Approved PBKDF (IG D.N conformance)

The PBKDF aligns with Option 1a in Section 5.4 of SP 800-132. In line with the requirements for SP 800-132, keys generated using the approved PBKDF must only be used for storage applications. Any other use of the approved PBKDF is non-conformant. In approved mode the module enforces that any password used must encode to at least 14 bytes (112 bits) and that the salt is at least 16 bytes (128 bits) long. The iteration count associated with the PBKDF should be as large as practical. As the module is a general purpose software module, it is not possible to anticipate all the levels of use for the PBKDF, however a user of the module should also note that a password should at least contain enough entropy to be unguessable and also contain enough entropy to reflect the security strength required for the key being generated. In the event a password encoding is simply based on ASCII, a 14byte password is unlikely to contain sufficient entropy for most purposes. The standard set of printable characters only allows for as much as 6 bits of entropy per byte. For a 14-byte password, this yields a key that has been generated using 14 * 6 bits of entropy, giving only 84 bits of security, which is well below what is required for a key with the same level of hardness as a 112-bit one. Users are referred to Appendix A (Security Considerations) of SP 800-132 for further information on password, salt, and iteration count selection. The iteration count value is provided by the user and should be appropriate to the way the algorithm is being used. (The memory hard augmentation of PBKDF provided by SCRYPT uses an iteration count of 1). For straight PBKDF with no memory hard support, the iteration count provided by the user should be at point of maximum cost bearable by the user carrying out the key derivation in the normal course of usage. To ensure sufficient whitening of the password in both cases, the module enforces a salt size of

For users interested in introducing memory hardness as a layer on top of the PBKDF the SCrypt augmentation to PBDKF based on HMAC-SHA-256 (as described in RFC 7914) is also available in nonapproved mode.

To customize the output of the cSHAKE function, the cSHAKE algorithm permits the operator to input strings for the Function-Name input (N) and the Customization String (S). The Function-Name input (N) is reserved for values specified by NIST and should only be set to the appropriate NIST specified value. Any other use of N is non-conformant. The Customization String (S) is available to allow users to customize the cSHAKE function as they wish. The length of S is limited to the available size of a byte array in the JVM running the module.

The module supports both FF1 and, in non-approved mode, FF3-1 format preserving encryption. Both are modes of AES. Table 10 shows the parameter constraints applicable to the module's implementation, as required by IG C.J. Table 10 - SP 800-38G Format-Preserving Encryption Constraints An attempt to use the FF1 or FF3-1 without meeting the radixminlen constraint or by exceeding maxlen will result in an IllegalArgumentException. Note: only FF1 should be used in approved mode.

As indicated under CAVP certificate A6047, the module supports TLS 1.2 KDF per RFC 5246, i.e. without using the extended master secret.

Approved HMAC algorithms can produce truncated versions of the specified HMAC. The right-most bits are truncated as per the NIST SP 800-107r1 (see also IG C.L and IG C.D).

The module does not include an entropy source. The module's use of an external Random Number Generator (RNG) is determined by the settings described in the subsections below. Table 11 – Non-Deterministic Random Number Generation Specification

The module makes use of the JVM's configured SecureRandom entropy source to provide entropy when required. The module will request entropy as appropriate to the security strength and seeding configuration for the DRBG that is using it and for the default DRBG will request a minimum of 256 bits of entropy. In approved mode the minimum amount of entropy that can be requested by a DRBG is 112 bits. The module will wait until the SecureRandom.generateSeed() returns the requested amount of entropy, blocking if necessary. The JVM’s entropy source can be configured through setting the security property securerandom.strongAlgorithms in the JVM's java.security file.

2.7.2 Guidance for the Use of DRBGs and Configuring the JVM's Entropy Source

A user can instantiate the default Approved DRBG for the module explicitly by using SecureRandom.getInstance("DEFAULT", "CCJ"), or by using a CryptoComplyFipsProvider object instead of the provider name as appropriate. This will seed the Approved DRBG from the live entropy source of the JVM with a number of bits of entropy appropriate to the security level of the default Approved The JVM's entropy source is checked according to SP 800-90B, Section 4.4 using the suggested C values for the Repetition Count Test (Section 4.4.1) and the Adaptive Proportion Test (Section 4.4.2) by default. These values can also be configured using the security property com.safelogic.cryptocomply.entropy.factors. This property takes a comma separated list of C values: one for 4.4.1, one for 4.4.2, and a value of H. For the default, the property would be set as: com.safelogic.cryptocomply.entropy.factors: 4, 13, 8.0

in the java.security property file. An additional option is available using the Approved Hash DRBG and the process outlined in SP 800-90A, Section 8.6.5. This can be turned on by following the instructions in Section 2.3 of the User Guide. The two DRBGs are instantiated in a chain as a "Source DRBG" to seed the "Target DRBG" in accordance with Section 7 of Draft NIST SP 800-90C, where the Target DRBG is the default Approved DRBG used by the module. The initial seed and the subsequent reseeds for the DRBG chain come from the live entropy source configured for the JVM. The DRBG chain will reseed automatically by pausing for 20 requests (which will usually equate to 5120 bytes). An entropy gathering thread reseeds the DRBG chain when it has gathered sufficient entropy (currently 256 bits) from the live entropy source. Once reseeded, the request counter is reset and the reseed process begins again. The “Source DRBG” in the chain is internal to the module and inaccessible to the user to ensure it is only used for generating seeds for the default Approved DRBG of the module. The user shall ensure that the entropy source is configured per Section 2.7.1 of this Security Policy and will block, or fail, if it is unable to provide the amount of entropy requested.

The module performs Cryptographic Key Generation in conformance to FIPS 140-3 IG D.H. The CKG for symmetric keys and seeds used for generating asymmetric keys is performed as per Section 4 of the SP 800-133r2 (using the output of a random bit generator) and is compliant with FIPS 186-5 and SP 80090Ar1 for DRBG. The seed used in asymmetric key generation is the direct output of SP 800-90Ar1 DRBG. Refer to Section 9.1 of the Security Policy for SSP generation details.

The module does not perform automatic SSP establishment, it only provides the components to the calling application, which can be used in SSP establishment.

The module implements KDFs from SP 800-135r1 (Recommendation for Existing Application-Specific Key Derivation Functions). These KDFs have been validated by the CAVP and received CVL certificates (A6047). No parts of these protocols, other than the CAVP tested components, have been reviewed or tested by the CAVP and CMVP.

As a software cryptographic module, the module supports logical interfaces only and not physical ports. All access to the module is through the module’s API. The API provides and defines the module’s logical interfaces. is also not applicable. The mapping of the FIPS 140-3 logical interfaces to the module is described in Table 12. Table 12 – Ports and Interfaces

All interfaces are logically separated by the module’s API. When the module performs self-tests, is in an error state, is generating keys, or performing zeroization, the module prevents all output on the logical data output interface as only the thread performing the operation has access to the data. The module is single-threaded, and in an error state, the module does not return any output data, only an error value.

Not applicable. The module does not support authentication.

The module supports two distinct operator roles, which are the User and Cryptographic Officer (CO). The cryptographic module implicitly maps the two roles to the services. An operator is considered the owner of the thread that instantiates the module and, therefore, only one concurrent operator is allowed. The module does not support a maintenance role and/or bypass capability. Table 13 lists all operator roles supported by the module. Table 13 - Roles

Table 14 lists the module services and corresponding details. The modes of SSP access shown in the table are defined as:

• G = Generate: The module generates or derives the SSP.
• R = Read: The SSP is read from the module (e.g. the SSP is output).
• E = Execute: The module uses the SSP in performing a cryptographic operation.
• W = Write: The SSP is updated, imported or written to the module.
• Z = Zeroize: The module zeroizes the SSP. Note: The module services are the same in the approved and non-approved modes of operation. The only difference is the function(s) used (i.e. approved/allowed or non-approved/non-allowed). Services in the module are accessed via the public APIs of the JAR file. The ability of a thread to invoke non-approved services depends on whether it has been registered with the module as approved mode only. In approved mode, no non-approved services are accessible. Refer also to Section 6.1 and 11.2 of this Security Policy for guidance. Table 14
• Approved Services Flag is accessed by calling the method CryptoServicesRegistrar.isInApprovedOnlyMode() - this method will return true if the thread is running in approved-only mode, false otherwise. Refer also to Section 2.4 of this Security Policy.

DRBG (SP 800-90Ar1) output

Used to generate random numbers, IVs and keys

Flag

N/A

Data

Counter DRBG Hash DRBG HMAC DRBG

AES Encryption Key, AES Decryption Key, AES Authentication Key, AES Wrapping Key, DH Agreement Private Key, DH Agreement Public Key, DRBG Seed, Internal State V and C value, and DRBG Key, DSA Signing Key, EC Agreement Private Key, EC Agreement Public Key, EC Signing Key, HMAC Authentication Key, KMAC Authentication Key, RSA Signing Key, RSA Key Transport Private Key, RSA Key Transport Public Key

CO / User CO / User

G E

SSP Export Operation

Returns a CSP as data that can be used for later output

Flag

SSP

Data

N/A

AES Encryption Key, AES Decryption Key, AES Authentication Key, AES Wrapping Key, DH Agreement Private Key, DH Agreement Public Key, DSA Signing Key, DSA Verification Key, EC Agreement Private Key, EC Agreement Public Key, EC Signing Key, EC Verification Key, HMAC Authentication Key, KMAC Authentication Key, RSA Signing Key, RSA Key Transport Private Key, RSA Key Transport Public Key

CO / User

R

Table 15 - Non-Approved Services Flag is accessed by calling the method CryptoServicesRegistrar.isInApprovedOnlyMode() - this method will return true if the thread is running in approved-only mode, false otherwise. Refer also to Section 2.4 of this Security Policy.

The integrity technique used by the module is HMAC-SHA-256. The integrity technique has received CAVP certificate A6047. The integrity technique is implemented by the module itself. The HMAC of the module JAR file, excluding directories and metadata, is calculated and compared to the expected value embedded within the module’s properties. If the calculated value does not match the expected value, the module raises an error and fails to load. The integrity test can be performed on demand by power cycling the host platform.

Each time the module is powered up, it runs the pre-operational tests to ensure that the integrity of the module has been maintained. Self-tests are available on demand by power cycling the module.

The module operates in a modifiable operational environment under the FIPS 140-3 definitions. The module runs on a GPC running one of the operating systems specified in the approved operational environment list (refer to Section 2.2 of this Security Policy). Each approved operating system manages processes and threads in a logically separated manner. The module’s operator is considered the owner of the calling application that instantiates the module within the process space of the Java Virtual Machine.

The module must be installed as described in Security Policy Section 11.1. No specific configuration options are required for the operational environments. No security rules, settings, or restrictions to the configuration of the operational environment are needed for the module to function in a FIPS-conformant manner.

The requirements of this section are not applicable to the module. The module is a software module and does not implement any physical security mechanisms.

The requirements of this section are not applicable to the module.

All Sensitive Security Parameters (SSPs) used by the module are described in this section in Table 16. All usage of these SSPs by the module (including all SSP lifecycle states) is described in the services detailed in Section 4.3 - Approved Services. Please note that the module does not perform automatic SSP establishment, it only provides the components to the calling application, which can be used in SSP establishment.

Table 16 - Sensitive Security Parameters (SSPs) Key Table The module does not provide persistent storage Key generator used in conjunction with an approved DRBG Import done via key constructor and/or factory (Electronic Entry) Export done via key recovery using getEncoded() method and followed by separate step to export key details as either plaintext or encrypted (Electronic Entry)

The AES GCM key and IV is generated randomly per IG C.H, and the Initialization Vector (IV) is a minimum of 96 bits. In the event module power is lost and restored, the consuming application must ensure that any of its AES GCM keys used for encryption or decryption are re-distributed. Refer to Section 2.6.1 of the Security Policy.

RSA key transport using PKCS#1 1.5 padding is deprecated through 2023 and disallowed after 2023.

Cryptographic Algorithm Self-Tests (CASTs) are performed prior to the first use of services related to the test target. CASTs also run periodically on service invocation. Pairwise Consistency Tests (PCTs) are performed on the corresponding key pairs.

Each time the module is powered up, it performs the pre-operational self-tests to confirm that sensitive data has not been damaged. The pre-operational tests include the software integrity test, which verifies the module using HMACSHA-256. Pre-operational tests also include the HMAC and SHS CASTs that are run prior to the software integrity test to ensure the correctness of the HMAC used. Pre-operational self-tests are available on demand by power cycling the module.

The module performs conditional self-tests when the conditions specified for cryptographic algorithm self-test and pair-wise consistency tests occur. The self-tests implemented are specified below. Table 17 – Conditional Algorithm Self-Tests

Table 18 – Pairwise Consistency Tests

If any of the above-mentioned self-tests fail, the module enters an error state called “Hard Error” state. Upon entering the error state, the module outputs status by way of an exception. An example exception for AES Encryption failure is: “Failed self-test on encryption: AES” The module can be recovered by power cycling, which results in execution of pre-operational self-tests and conditional cryptographic algorithm self-tests. If the tests pass, then the module will be available for use.

Each time the module is powered up, it runs the pre-operational tests to ensure that the integrity of the module has been maintained. Pre-operational self-tests are available on demand by power cycling the module. Initial CAST self-tests are available on demand by power cycling the module and then invoking the service related to the test target.

The module exists as part of the running JVM, and as such:

• Secure installation of the module requires the use of the unchanged jar to be loaded into a JVM via either the class-path or the module-path as appropriate to the JVM and its usage.
• Initialization of the module will occur on startup of the module by the JVM. The user can trigger initialization by attempting to invoke any service in the module or simply calling FipsStatus.isReady() which will only return true if the module has been successfully initialized.
• Once the JVM has loaded the module and the module has been initialized, the startup phase is over, and the module is able to provide services.
• Operation of the module consists of calling the various APIs providing services. The module code will make use of the current thread for performing any required CASTs and health tests and then provide a service object to the user, capable of performing the requested service. A User Guide is provided to operators of the module.

The JAR file representing the module needs to be installed in a JVM's class path in a manner appropriate to its use in applications running on the JVM. Functionality in the module is provided in two ways. At the lowest level there are distinct classes that provide access to the approved and non-approved services provided by the module. A more abstract level of access can also be gained by using strings providing operation names passed into the module's Java cryptography provider through the APIs described in the Java Cryptography Architecture (JCA) and the Java Cryptography Extension (JCE). When the module is used in approved mode, classes providing implementations of algorithms that are not approved or allowed are explicitly disabled. SSPs such as private and secret keys implement the Destroyable interface. Where appropriate these SSPs can be zeroized on demand by invoking the destroy() method. The return of the destroy() method indicates that the zeroization is complete.

If the underlying JVM is running with a Java SecurityManager installed, the module will be running in approved mode with secret and private key export disabled.

In the presence of a Java SecurityManager approved mode services specific to a context, such as DSA and ECDSA for use in TLS, require specific policy permissions to be configured in the JVM configuration by the Cryptographic Officer or User. The SecurityManager can also be used to restrict the ability of particular code bases to examine CSPs. In the absence of a Java SecurityManager specific services related to protocols such as TLS are available, however must only be used in relation to those protocols.

Use of the module with a Java SecurityManager requires the setting of some basic permissions to allow the module HMAC-SHA-256 software integrity test to take place as well as to allow the module itself to examine secret and private keys. The basic permissions required for the module to operate correctly with a Java SecurityManager are indicated by the Required column of Table 19. Table 19 - Available Java Permissions for SecurityManager

The module design corresponds to the module security rules. This section documents the security rules enforced by the cryptographic module to implement the security requirements of this FIPS 140-3 Level 1 module.

1. The module provides two distinct operator roles: User and Cryptographic Officer.
2. The module does not provide authentication.
3. The operator may command the module to perform the self-tests by cycling power or resetting the module.
4. Self-tests do not require any operator action.
5. Data output is inhibited during self-tests, zeroization, and error states. Output related to keys and their use is inhibited until the key concerned has been fully generated.
6. Status information does not contain CSPs or sensitive data that if misused could lead to a compromise of the module.
7. There are no restrictions on which keys or CSPs are zeroized by the zeroization service.
8. The module does not support concurrent operators.
9. The module does not have any external input/output devices used for entry/output of data.
10. The module does not enter or output plaintext CSPs from the module’s physical boundary.
11. The module does not output intermediate key values.

When the module is used within the context of Java Security Manager or the system/security property com.safelogic.cryptocomply.fips.approved_only is set to true, the module will start in approved mode and non-approved services are not accessible in this mode. When the module is not used within the context of Java Security Manager, the module will start in non-approved mode by default. Refer to Security Policy Section 2.4 for additional details.

The transition from non-approved mode to approved mode is a combination of granted permission (a) and request to change mode (b): a) com.safelogic.cryptocomply.crypto.CryptoServicesPermission “changeToApprovedModeEnabled” b) CryptoServicesRegistrar.setApprovedMode(true) The CSPs made available in non-approved mode will not be accessible once the thread transitions into approved mode. The CSPs generated using the non-approved mode cannot be passed or shared with algorithms operating in approved mode, and vice-versa. This is done by an indicator within the class (object) instantiating the key that the key was created in an approved mode or non-approved mode. Any attempt by a thread within the module to use the key in an opposite mode will result in an exception being generated by the module. For example, if an RSA private key has been created in either approved or non-approved mode, then any request to access that key will first need to confirm if the thread making the request is in the same mode.

The module cannot transition from approved mode to non-approved mode. To initiate the module in non-approved mode, either it should not be used in the context of Java Security Manager, or the module should have the permission com.safelogic.cryptocomply.crypto.CryptoServicesPermission unapprovedModeEnabled granted by the Java Security Manager.

Vulnerabilities found in the module will be reported on the National Vulnerability Database, located at the following link: https://nvd.nist.gov/ Researchers and users are encouraged to report any security related concerns to support@safelogic.com.

The module implements basic protections to mitigate against timing-based attacks against its internal implementations. There are two countermeasures used. The first countermeasure is Constant Time Comparisons, which protect the digest and integrity algorithms by strictly avoiding “fast fail” comparison of MACs, signatures, and digests so the time taken to compare a MAC, signature, or digest is constant regardless of whether the comparison passes or fails. The second countermeasure is made up of Numeric Blinding and decryption/signing verification which both protect the RSA algorithm. Numeric Blinding prevents timing attacks against RSA decryption and signing by providing a random input into the operation which is subsequently eliminated when the result is produced. The random input makes it impossible for a third party observing the private key operation to attempt a timing attack on the operation as they do not have knowledge of the random input and consequently the time taken for the operation tells them nothing about the private value of the RSA key. Decryption/signing verification is carried out by calculating a primitive encryption or signature verification operation after a corresponding decryption or signing operation before the result of the decryption or signing operation is returned. The purpose of this is to protect against Lenstra's CRT attack by verifying the correctness of the private key calculations involved. Lenstra's CRT attack takes advantage of undetected errors in the use of RSA private keys with CRT values and, if exploitable, can be used to discover the private value of the RSA key.

Appendix: References and Acronyms The following standards are referred to in this Security Policy. Table 20 - References

The following acronyms are used in this Security Policy. Table 21 - Acronyms