| RFC 9172 | Bundle Protocol Security (BPSec) | January 2022 |
| Birrane, III & McKeever | Standards Track | [Page] |
This document defines a security protocol providing data integrity and confidentiality services for the Bundle Protocol (BP).¶
This is an Internet Standards Track document.¶
This document is a product of the Internet Engineering Task Force (IETF). It represents the consensus of the IETF community. It has received public review and has been approved for publication by the Internet Engineering Steering Group (IESG). Further information on Internet Standards is available in Section 2 of RFC 7841.¶
Information about the current status of this document, any errata, and how to provide feedback on it may be obtained at https://www.rfc-editor.org/info/rfc9172.¶
Copyright (c) 2022 IETF Trust and the persons identified as the document authors. All rights reserved.¶
This document is subject to BCP 78 and the IETF Trust's Legal Provisions Relating to IETF Documents (https://trustee.ietf.org/license-info) in effect on the date of publication of this document. Please review these documents carefully, as they describe your rights and restrictions with respect to this document. Code Components extracted from this document must include Revised BSD License text as described in Section 4.e of the Trust Legal Provisions and are provided without warranty as described in the Revised BSD License.¶
This document defines security features for the Bundle Protocol (BP) [RFC9171] and is intended for use in Delay-Tolerant Networking (DTN) to provide security services between a security source and a security acceptor. When the security source is the bundle source and the security acceptor is the bundle destination, the security service provides end-to-end protection.¶
The Bundle Protocol specification [RFC9171] defines DTN as referring to "a network architecture providing communications in and/or through highly stressed environments" where "BP may be viewed as sitting at the application layer of some number of constituent networks, forming a store-carry-forward overlay network". The phrase "stressed environment" refers to multiple challenging conditions including intermittent connectivity, large and/or variable delays, asymmetric data rates, and high bit error rates.¶
It should be presumed that the BP will be deployed in an untrusted network, which poses the usual security challenges related to confidentiality and integrity. However, the stressed nature of the BP operating environment imposes unique conditions where usual transport security mechanisms may not be sufficient. For example, the store-carry-forward nature of the network may require protecting data at rest, preventing unauthorized consumption of critical resources such as storage space, and operating without regular contact with a centralized security oracle (such as a certificate authority).¶
An end-to-end security service that operates in all of the environments where the BP operates is needed.¶
BPSec provides integrity and confidentiality services for BP bundles, as defined in this section.¶
Integrity services ensure that changes to target data within a bundle can be discovered. Data changes may be caused by processing errors, environmental conditions, or intentional manipulation. In the context of BPSec, integrity services apply to plaintext in the bundle.¶
Confidentiality services ensure that target data is unintelligible to nodes in DTN, except for authorized nodes possessing special information. Generally, this means producing ciphertext from plaintext and generating authentication information for that ciphertext. In this context, confidentiality applies to the contents of target data and does not extend to hiding the fact that confidentiality exists in the bundle.¶
NOTE: Hop-by-hop authentication is NOT a supported security service in this specification, for two reasons:¶
This document defines the security services provided by the BPSec. This includes the data specification for representing these services as BP extension blocks and the rules for adding, removing, and processing these blocks at various points during the bundle's traversal of a delay-tolerant network.¶
BPSec addresses only the security of data traveling over the DTN, not the underlying DTN itself. Furthermore, while the BPSec protocol can provide security-at-rest in a store-carry-forward network, it does not address threats that share computing resources with the DTN and/or BPSec software implementations. These threats may be malicious software or compromised libraries that intend to intercept data or recover cryptographic material. Here, it is the responsibility of the BPSec implementer to ensure that any cryptographic material, including shared secrets or private keys, is protected against access within both memory and storage devices.¶
Completely trusted networks are extremely uncommon. Among untrusted networks, different networking conditions and operational considerations require security mechanisms of varying strengths. Mandating a single security context, which is a set of assumptions, algorithms, configurations, and policies used to implement security services, may result in too much security for some networks and too little security in others. Default security contexts are defined in [RFC9173] to provide basic security services for interoperability testing and for operational use on the terrestrial Internet. It is expected that separate documents will define different security contexts for use in different networks.¶
This specification addresses neither the fitness of externally defined cryptographic methods nor the security of their implementation.¶
This specification does not address the implementation of security policies and does not provide a security policy for the BPSec. Similar to cipher suites, security policies are based on the nature and capabilities of individual networks and network operational concepts. This specification does provide policy considerations that can be taken into account when building a security policy.¶
With the exception of the Bundle Protocol, this specification does not address how to combine the BPSec security blocks with other protocols, other BP extension blocks, or other best practices to achieve security in any particular network implementation.¶
This document is best read and understood within the context of the following other DTN documents:¶
The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", "SHOULD", "SHOULD NOT", "RECOMMENDED", "NOT RECOMMENDED", "MAY", and "OPTIONAL" in this document are to be interpreted as described in BCP 14 [RFC2119] [RFC8174] when, and only when, they appear in all capitals, as shown here.¶
This section defines terminology that either is unique to the BPSec or is necessary for understanding the concepts defined in this specification.¶
The application of security services in DTN is a complex endeavor that must consider physical properties of the network (such as connectivity and propagation times), policies at each node, application security requirements, and current and future threat environments. This section identifies those desirable properties that guide design decisions for this specification and that are necessary for understanding the format and behavior of the BPSec protocol.¶
Security services within this specification must allow different blocks within a bundle to have different security services applied to them.¶
Blocks within a bundle represent different types of information. The primary block contains identification and routing information. The payload block carries application data. Extension blocks carry a variety of data that may augment or annotate the payload or that otherwise provide information necessary for the proper processing of a bundle along a path. Therefore, applying a single level and type of security across an entire bundle fails to recognize that blocks in a bundle represent different types of information with different security needs.¶
For example, a payload block might be encrypted to protect its contents and an extension block containing summary information related to the payload might be integrity signed but unencrypted to provide waypoints access to payload-related data without providing access to the payload.¶
A bundle can have multiple security blocks, and these blocks can have different security sources. BPSec implementations MUST NOT assume that all blocks in a bundle have the same security operations applied to them.¶
The Bundle Protocol allows extension blocks to be added to a bundle at any time during its existence in DTN. When a waypoint adds a new extension block to a bundle, that extension block MAY have security services applied to it by that waypoint. Similarly, a waypoint MAY add a security service to an existing block, consistent with its security policy.¶
When a waypoint adds a security service to the bundle, the waypoint is the security source for that service. The security block(s) that represent that service in the bundle may need to record this security source, as the bundle destination might need this information for processing.¶
For example, a bundle source may choose to apply an integrity service to its plaintext payload. Later a waypoint node, representing a gateway to another portion of the delay-tolerant network, may receive the bundle and choose to apply a confidentiality service. In this case, the integrity security source is the bundle source and the confidentiality security source is the waypoint node.¶
In cases where the security source and security acceptor are not the bundle source and bundle destination, respectively, it is possible that the bundle will reach the bundle destination prior to reaching a security acceptor. In cases where this may be a practical problem, it is recommended that solutions such as bundle encapsulation be used to ensure that a bundle be delivered to a security acceptor prior to being delivered to the bundle destination. Generally, if a bundle reaches a waypoint that has the appropriate configuration and policy to act as a security acceptor for a security service in the bundle, then the waypoint should act as that security acceptor.¶
The security policy enforced by nodes in the delay-tolerant network may differ.¶
Some waypoints will have security policies that require the waypoint to evaluate security services even if the waypoint is neither the bundle destination nor the final intended acceptor of the service. For example, a waypoint could choose to verify an integrity service even though the waypoint is not the bundle destination and the integrity service will be needed by other nodes along the bundle's path.¶
Some waypoints will determine, through policy, that they are the intended recipient of the security service and will terminate the security service in the bundle. For example, a gateway node could determine that, even though it is not the destination of the bundle, it should verify and remove a particular integrity service or attempt to decrypt a confidentiality service, before forwarding the bundle along its path.¶
Some waypoints could understand security blocks but refuse to process them unless they are the bundle destination.¶
A security context is the set of assumptions, algorithms, configurations, and policies used to implement security services. Different contexts may specify different algorithms, different polices, or different configuration values used in the implementation of their security services. BPSec provides a mechanism to define security contexts. Users may select from registered security contexts and customize those contexts through security context parameters.¶
For example, some users might prefer a SHA2 hash function for integrity, whereas other users might prefer a SHA3 hash function. Providing either separate security contexts or a single, parameterized security context allows users flexibility in applying the desired cipher suite, policy, and configuration when populating a security block.¶
Whenever a node determines that it must process more than one security block in a received bundle (either because the policy at a waypoint states that it should process security blocks or because the node is the bundle destination), the order in which security blocks are processed must be deterministic. All nodes must impose this same deterministic processing order for all security blocks. This specification provides determinism in the application and evaluation of security services, even when doing so results in a loss of flexibility.¶
This specification defines two types of security block: the Block Integrity Block (BIB) and the Block Confidentiality Block (BCB).¶
Security operations in a bundle MUST be unique; the same security service MUST NOT be applied to a security target more than once in a bundle. Since a security operation is represented by a security block, this means that multiple security blocks of the same type cannot share the same security targets. A new security block MUST NOT be added to a bundle if a preexisting security block of the same type is already defined for the security target of the new security block.¶
This uniqueness requirement ensures that there is no ambiguity related to the order in which security blocks are processed or how security policy can be specified to require certain security services be present in a bundle.¶
Using the notation OP(service, target), several examples illustrate this uniqueness requirement.¶
NOTES:¶
A single security block MAY represent multiple security operations as a way of reducing the overall number of security blocks present in a bundle. In these circumstances, reducing the number of security blocks in the bundle reduces the amount of redundant information in the bundle.¶
A set of security operations can be represented by a single security block when all of the following conditions are true.¶
When representing multiple security operations in a single security block, the information that is common across all operations is represented once in the security block; the information that is different (e.g., the security targets) is represented individually.¶
If a node processes any security operation in a security block, it is RECOMMENDED that it process all security operations in the security block. This allows security sources to assert that the set of security operations in a security block are expected to be processed by the same security acceptor. However, the determination of whether a node actually is a security acceptor or not is a matter of the policy of the node itself. In cases where a receiving node determines that it is the security acceptor of only a subset of the security operations in a security block, the node may choose to only process that subset of security operations.¶
A security target is a block in the bundle to which a security service applies. This target must be uniquely and unambiguously identifiable when processing a security block. The definition of the extension block header from [RFC9171] provides a "block number" field suitable for this purpose. Therefore, a security target in a security block MUST be represented as the block number of the target block.¶
Each security block uses the Canonical Bundle Block Format as defined in [RFC9171]. That is, each security block is comprised of the following elements:¶
Security-specific information for a security block is captured in the block-type-specific data field.¶
The structure of the security-specific portions of a security block is identical for both the BIB and BCB block types. Therefore, this section defines an Abstract Security Block (ASB) data structure and discusses its definition, its processing, and other constraints for using this structure. An ASB is never directly instantiated within a bundle, it is only a mechanism for discussing the common aspects of BIB and BCB security blocks.¶
The fields of the ASB SHALL be as follows, listed in the order in which they must appear. The encoding of these fields MUST be in accordance with the canonical forms provided in Section 4.¶
This field identifies which optional fields are present in the security block. This field SHALL be represented as a CBOR unsigned integer whose contents shall be interpreted as a bit field. Each bit in this bit field indicates the presence (bit set to 1) or absence (bit set to 0) of optional data in the security block. The association of bits to security block data is defined as follows.¶
Implementations MUST set reserved bits to 0 when writing this field and MUST ignore the values of reserved bits when reading this field. For unreserved bits, a value of 1 indicates that the associated security block field MUST be included in the security block. A value of 0 indicates that the associated security block field MUST NOT be in the security block.¶
This field captures one or more security context parameters that should be used when processing the security service described by this security block. This field SHALL be represented by a CBOR array. Each entry in this array is a single security context parameter. A single parameter SHALL also be represented as a CBOR array comprising a 2-tuple of the Id and value of the parameter, as follows.¶
The logical layout of the parameters array is illustrated in Figure 1.¶
+----------------+----------------+ +----------------+ | Parameter 1 | Parameter 2 | ... | Parameter N | +------+---------+------+---------+ +------+---------+ | Id | Value | Id | Value | | Id | Value | +------+---------+------+---------+ +------+---------+
This field captures the results of applying a security service to the security targets of the security block. This field SHALL be represented as a CBOR array of target results. Each entry in this array represents the set of security results for a specific security target. The target results MUST be ordered identically to the Security Targets field of the security block. This means that the first set of target results in this array corresponds to the first entry in the Security Targets field of the security block, and so on. There MUST be one entry in this array for each entry in the Security Targets field of the security block.¶
The set of security results for a target is also represented as a CBOR array of individual results. An individual result is represented as a CBOR array comprising a 2-tuple of a result Id and a result value, defined as follows.¶
The logical layout of the security results array is illustrated in Figure 2. In this figure, there are N security targets for this security block. The first security target contains M results and the Nth security target contains K results.¶
+--------------------------+ +---------------------------+ | Target 1 | | Target N | +----------+----+----------+ +---------------------------+ | Result 1 | | Result M | ... | Result 1 | | Result K | +----+-----+ .. +----+-----+ +---+------+ .. +----+------+ | Id |Value| | Id |Value| | Id |Value| | Id | Value| +----+-----+ +----+-----+ +----+-----+ +----+------+
A BIB is a BP extension block with the following characteristics.¶
Notes:¶
A BCB is a BP extension block with the following characteristics.¶
The block processing control flags value can be set to whatever values are required by local policy with the following exceptions:¶
The BCB modifies the contents of its security target(s). When a BCB is applied, the security target body data are encrypted "in-place". Following encryption, the security target block-type-specific data field contains ciphertext, not plaintext.¶
Notes:¶
The security block types defined in this specification are designed to be as independent as possible. However, there are some cases where security blocks may share a security target; this sharing creates processing dependencies.¶
If a BCB and a BIB share a security target, an undesirable condition occurs: a waypoint would be unable to validate the BIB because the shared security target has been encrypted by the BCB. To address this situation, the following processing rules MUST be followed:¶
These restrictions on block interactions impose a necessary ordering when applying security operations within a bundle. Specifically, for a given security target, BIBs MUST be added before BCBs. This ordering MUST be preserved in cases where the current BPA is adding all of the security blocks for the bundle or where the BPA is a waypoint adding new security blocks to a bundle that already contains security blocks.¶
In cases where a security source wishes to calculate both a plaintext integrity-protection mechanism and encrypt a security target, a BCB with a security context that generates an integrity-protection mechanism as one or more additional security results MUST be used instead of adding both a BIB and then a BCB for the security target at the security source.¶
Each security context MUST define its own context parameters and results. Each defined parameter and result is represented as the tuple of an identifier and a value. Identifiers are always represented as a CBOR unsigned integer. The CBOR encoding of values is as defined by the security context specification.¶
Identifiers MUST be unique for a given security context but do not need to be unique amongst all security contexts.¶
An example of a security context can be found in [RFC9173].¶
This section provides two examples of BPSec blocks applied to bundles. In the first example, a single node adds several security operations to a bundle. In the second example, a waypoint node received the bundle created in the first example and adds additional security operations. In both examples, the first column represents blocks within a bundle and the second column represents the block number for the block, using the terminology B1...Bn for the purpose of illustration.¶
In this example, a bundle has four non-security-related blocks: the primary block (B1), two extension blocks (B4, B5), and a payload block (B6). The bundle source wishes to provide an integrity signature of the plaintext associated with the primary block, the second extension block, and the payload. The bundle source also wishes to provide confidentiality for the first extension block. The resultant bundle is illustrated in Figure 3 and the security actions are described below.¶
Block in Bundle ID
+==========================================+====+
| Primary Block | B1 |
+------------------------------------------+----+
| BIB | B2 |
| OP(bib-integrity, targets = B1, B5, B6)| |
+------------------------------------------+----+
| BCB | B3 |
| OP(bcb-confidentiality, target = B4) | |
+------------------------------------------+----+
| Extension Block (encrypted) | B4 |
+------------------------------------------+----+
| Extension Block | B5 |
+------------------------------------------+----+
| Payload Block | B6 |
+------------------------------------------+----+
The following security actions were applied to this bundle at its time of creation.¶
Consider that the bundle as it is illustrated in Figure 3 is now received by a waypoint node that wishes to encrypt the second extension block and the bundle payload. The waypoint security policy is to allow existing BIBs for these blocks to persist, as they may be required as part of the security policy at the bundle destination.¶
The resultant bundle is illustrated in Figure 4 and the security actions are described below. Note that block IDs provided here are ordered solely for the purpose of this example and are not meant to impose an ordering for block creation. The ordering of blocks added to a bundle MUST always be in compliance with [RFC9171].¶
Block in Bundle ID
+==========================================+====+
| Primary Block | B1 |
+------------------------------------------+----+
| BIB | B2 |
| OP(bib-integrity, target = B1) | |
+------------------------------------------+----+
| BIB (encrypted) | B7 |
| OP(bib-integrity, targets = B5, B6) | |
+------------------------------------------+----+
| BCB | B8 |
|OP(bcb-confidentiality,targets = B5,B6,B7)| |
+------------------------------------------+----+
| BCB | B3 |
| OP(bcb-confidentiality, target = B4) | |
+------------------------------------------+----+
| Extension Block (encrypted) | B4 |
+------------------------------------------+----+
| Extension Block (encrypted) | B5 |
+------------------------------------------+----+
| Payload Block (encrypted) | B6 |
+------------------------------------------+----+
The following security actions were applied to this bundle prior to its forwarding from the waypoint node.¶
Security services require consistency and determinism in how information is presented to cipher suites at security sources, verifiers, and acceptors. For example, integrity services require that the same target information (e.g., the same bits in the same order) is provided to the cipher suite when generating an original signature and when validating a signature. Canonicalization algorithms transcode the contents of a security target into a canonical form.¶
Canonical forms are used to generate input to a security context for security processing at a BP node. If the values of a security target are unchanged, then the canonical form of that target will be the same even if the encoding of those values for wire transmission is different.¶
BPSec operates on data fields within bundle blocks (e.g., the block-type-specific data field). In their canonical form, these fields MUST include their own CBOR encoding and MUST NOT include any other encapsulating CBOR encoding. For example, the canonical form of the block-type-specific data field is a CBOR byte string existing within the CBOR array containing the fields of the extension block. The entire CBOR byte string is considered the canonical block-type-specific data field. The CBOR array framing is not considered part of the field.¶
The canonical form of the primary block is as specified in [RFC9171] with the following constraint.¶
All non-primary blocks share the same block structure and are canonicalized as specified in [RFC9171] with the following constraints.¶
Security contexts MAY define their own canonicalization algorithms and require the use of those algorithms over the ones provided in this specification. In the event of conflicting canonicalization algorithms, algorithms defined in a security context take precedence over this specification when constructing canonical forms for that security context.¶
This section describes the security aspects of bundle processing.¶
Security blocks must be processed in a specific order when received by a BP node. The processing order is as follows.¶
If a received bundle contains a BCB, the receiving node MUST determine whether it is the security acceptor for any of the security operations in the BCB. If so, the node MUST process those operations and remove any operation-specific information from the BCB prior to delivering data to an application at the node or forwarding the bundle. If processing a security operation fails, the target SHALL be processed according to the security policy. A bundle status report indicating the failure MAY be generated. When all security operations for a BCB have been removed from the BCB, the BCB MUST be removed from the bundle.¶
If the receiving node is the destination of the bundle, the node MUST decrypt any BCBs remaining in the bundle. If the receiving node is not the destination of the bundle, the node MUST process the BCB if directed to do so as a matter of security policy.¶
If the security policy of a node specifies that a node should have applied confidentiality to a specific security target and no such BCB is present in the bundle, then the node MUST process this security target in accordance with the security policy. It is RECOMMENDED that the node remove the security target from the bundle because the confidentiality (and possibly the integrity) of the security target cannot be guaranteed. If the removed security target is the payload block, the bundle MUST be discarded.¶
If an encrypted payload block cannot be decrypted (i.e., the ciphertext cannot be authenticated), then the bundle MUST be discarded and processed no further. If an encrypted security target other than the payload block cannot be decrypted, then the associated security target and all security blocks associated with that target MUST be discarded and processed no further. In both cases, requested status reports (see [RFC9171]) MAY be generated to reflect bundle or block deletion.¶
When a BCB is decrypted, the recovered plaintext for each security target MUST replace the ciphertext in each of the security targets' block-type-specific data fields. If the plaintext is of a different size than the ciphertext, the framing of the CBOR byte string of this field must be updated to ensure this field remains a valid CBOR byte string. The length of the recovered plaintext is known by the decrypting security context.¶
If a BCB contains multiple security operations, each operation processed by the node MUST be treated as if the security operation has been represented by a single BCB with a single security operation for the purposes of report generation and policy processing.¶
If a received bundle contains a BIB, the receiving node MUST determine whether it is the security acceptor for any of the security operations in the BIB. If so, the node MUST process those operations and remove any operation-specific information from the BIB prior to delivering data to an application at the node or forwarding the bundle. If processing a security operation fails, the target SHALL be processed according to the security policy. A bundle status report indicating the failure MAY be generated. When all security operations for a BIB have been removed from the BIB, the BIB MUST be removed from the bundle.¶
A BIB MUST NOT be processed if the security target of the BIB is also the security target of a BCB in the bundle. Given the order of operations mandated by this specification, when both a BIB and a BCB share a security target, it means that the security target must have been encrypted after it was integrity signed; therefore, the BIB cannot be verified until the security target has been decrypted by processing the BCB.¶
If the security policy of a node specifies that a node should have applied integrity to a specific security target and no such BIB is present in the bundle, then the node MUST process this security target in accordance with the security policy. It is RECOMMENDED that the node remove the security target from the bundle if the security target is not the payload or primary block. If the security target is the payload or primary block, the bundle MAY be discarded. This action can occur at any node that has the ability to verify an integrity signature, not just the bundle destination.¶
If a receiving node is not the security acceptor of a security operation in a BIB, it MAY attempt to verify the security operation anyway to prevent forwarding corrupt data. If the verification fails, the node SHALL process the security target in accordance with local security policy. If a payload integrity check fails at a waypoint, it is RECOMMENDED that it be processed in the same way as a failure of a payload integrity check at the bundle destination. If the check passes, the node MUST NOT remove the security operation from the BIB prior to forwarding.¶
If a BIB contains multiple security operations, each operation processed by the node MUST be treated as if the security operation has been represented by a single BIB with a single security operation for the purposes of report generation and policy processing.¶
If it is necessary for a node to fragment a bundle payload, and security services have been applied to that bundle, the fragmentation rules described in [RFC9171] MUST be followed. As defined there and summarized here for completeness, only the payload block can be fragmented; security blocks, like all extension blocks, can never be fragmented.¶
Due to the complexity of payload-block fragmentation, including the possibility of fragmenting payload-block fragments, integrity and confidentiality operations are not to be applied to a bundle representing a fragment. Specifically, a BCB or BIB MUST NOT be added to a bundle if the "Bundle is a fragment" flag is set in the bundle processing control flags field.¶
Security processing in the presence of payload-block fragmentation may be handled by other mechanisms outside of the BPSec protocol or by applying BPSec blocks in coordination with an encapsulation mechanism. A node should apply any confidentiality protection prior to performing any fragmentation.¶
There exists a myriad of ways to establish, communicate, and otherwise manage key information in DTN. Certain DTN deployments might follow established protocols for key management, whereas other DTN deployments might require new and novel approaches. BPSec assumes that key management is handled as a separate part of network management; this specification neither defines nor requires a specific strategy for key management.¶
When implementing BPSec, several policy decisions must be considered. This section describes key policies that affect the generation, forwarding, and receipt of bundles that are secured using this specification. No single set of policy decisions is envisioned to work for all secure DTN deployments.¶
Adding a BIB to a security target that has already been encrypted by a BCB is not allowed. If this condition is likely to be encountered, there are (at least) three possible policies that could handle this situation.¶
BPAs must process blocks and bundles in accordance with both BP policy and BPSec policy. The decision to receive, forward, deliver, or delete a bundle may be communicated to the report-to address of the bundle in the form of a status report, as a method of tracking the progress of the bundle through the network. The status report for a bundle may be augmented with a "reason code" explaining why the particular action was taken on the bundle.¶
This section describes a set of reason codes associated with the security processing of a bundle. The communication of security-related status reports might reduce the security of a network if these reports are intercepted by unintended recipients. BPSec policy SHOULD specify the conditions in which sending security reason codes are appropriate. Examples of appropriate conditions for the use of security reason codes could include the following.¶
Security reason codes are assigned in accordance with Section 11.2 and are as described below.¶
Given the nature of DTN applications, it is expected that bundles may traverse a variety of environments and devices that each pose unique security risks and requirements on the implementation of security within BPSec. For this reason, it is important to introduce key threat models and describe the roles and responsibilities of the BPSec protocol in protecting the confidentiality and integrity of the data against those threats. This section provides additional discussion on security threats that BPSec will face and describes how BPSec security mechanisms operate to mitigate these threats.¶
The threat model described here is assumed to have a set of capabilities identical to those described by the Internet Threat Model in [RFC3552], but the BPSec threat model is scoped to illustrate threats specific to BPSec operating within DTN environments; therefore, it focuses on on-path attackers (OPAs). In doing so, it is assumed that the delay-tolerant network (or significant portions of the delay-tolerant network) are completely under the control of an attacker.¶
BPSec was designed to protect against OPA threats that may have access to a bundle during transit from its source, Alice, to its destination, Bob. An OPA node, Olive, is a noncooperative node operating on the delay-tolerant network between Alice and Bob that has the ability to receive bundles, examine bundles, modify bundles, forward bundles, and generate bundles at will in order to compromise the confidentiality or integrity of data within the delay-tolerant network. There are three classes of OPA nodes that are differentiated based on their access to cryptographic material:¶
If Olive is operating as a privileged node, this is tantamount to compromise; BPSec does not provide mechanisms to detect or remove Olive from the delay-tolerant network or BPSec secure environment. It is up to the BPSec implementer or the underlying cryptographic mechanisms to provide appropriate capabilities if they are needed. It should also be noted that if the implementation of BPSec uses a single set of shared cryptographic material for all nodes, a legitimate node is equivalent to a privileged node because KM == KA == KB. For this reason, sharing cryptographic material in this way is not recommended.¶
A special case of the legitimate node is when Olive is either Alice or Bob (i.e., KM == KA or KM == KB). In this case, Olive is able to impersonate traffic as either Alice or Bob, respectively, which means that traffic to and from that node can be decrypted and encrypted, respectively. Additionally, messages may be signed as originating from one of the endpoints.¶
Once Olive has received a bundle, she is able to examine the contents of that bundle and attempt to recover any protected data or cryptographic keying material from the blocks contained within. The protection mechanism that BPSec provides against this action is the BCB, which encrypts the contents of its security target, providing confidentiality of the data. Of course, it should be assumed that Olive is able to attempt offline recovery of encrypted data, so the cryptographic mechanisms selected to protect the data should provide a suitable level of protection.¶
When evaluating the risk of eavesdropping attacks, it is important to consider the lifetime of bundles on DTN. Depending on the network, bundles may persist for days or even years. Long-lived bundles imply that the data exists in the network for a longer period of time and, thus, there may be more opportunities to capture those bundles. Additionally, the implication is that long-lived bundles store information within that remains relevant and sensitive for long enough that, once captured, there is sufficient time to crack encryption associated with the bundle. If a bundle does persist on the network for years and the cipher suite used for a BCB provides inadequate protection, Olive may be able to recover the protected data either before that bundle reaches its intended destination or before the information in the bundle is no longer considered sensitive.¶
NOTE: Olive is not limited by the bundle lifetime and may retain a given bundle indefinitely.¶
NOTE: Irrespective of whether BPSec is used, traffic analysis will be possible.¶
As a node participating in the delay-tolerant network between Alice and Bob, Olive will also be able to modify the received bundle, including non-BPSec data such as the primary block, payload blocks, or block processing control flags as defined in [RFC9171]. Olive will be able to undertake activities including modification of data within the blocks, replacement of blocks, addition of blocks, or removal of blocks. Within BPSec, both the BIB and BCB provide integrity-protection mechanisms to detect or prevent data manipulation attempts by Olive.¶
The BIB provides that protection to another block that is its security target. The cryptographic mechanisms used to generate the BIB should be strong against collision attacks, and Olive should not have access to the cryptographic material used by the originating node to generate the BIB (e.g., KA). If both of these conditions are true, Olive will be unable to modify the security target or the BIB, and thus she cannot lead Bob to validate the security target as originating from Alice.¶
Since BPSec security operations are implemented by placing blocks in a bundle, there is no in-band mechanism for detecting or correcting certain cases where Olive removes blocks from a bundle. If Olive removes a BCB, but keeps the security target, the security target remains encrypted and there is a possibility that there may no longer be sufficient information to decrypt the block at its destination. If Olive removes both a BCB (or BIB) and its security target, there is no evidence left in the bundle of the security operation. Similarly, if Olive removes the BIB, but not the security target, there is no evidence left in the bundle of the security operation. In each of these cases, the implementation of BPSec must be combined with policy configuration at endpoints in the network that describe the expected and required security operations that must be applied on transmission and that are expected to be present on receipt. This or other similar out-of-band information is required to correct for removal of security information in the bundle.¶
A limitation of the BIB may exist within the implementation of BIB validation at the destination node. If Olive is a legitimate node within the delay-tolerant network, the BIB generated by Alice with KA can be replaced with a new BIB generated with KM and forwarded to Bob. If Bob is only validating that the BIB was generated by a legitimate user, Bob will acknowledge the message as originating from Olive instead of Alice. Validating a BIB indicates only that the BIB was generated by a holder of the relevant key; it does not provide any guarantee that the bundle or block was created by the same entity. In order to provide verifiable integrity checks, the BCB should require an encryption scheme that is Indistinguishable under adaptive Chosen Ciphertext Attack (IND-CCA2) secure. Such an encryption scheme will guard against signature substitution attempts by Olive. In this case, Alice creates a BIB with the protected data block as the security target and then creates a BCB with both the BIB and protected data block as its security targets.¶
If Olive is in an OPA position within the delay-tolerant network, she is able to influence how any bundles that come to her may pass through the network. Upon receiving and processing a bundle that must be routed elsewhere in the network, Olive has three options as to how to proceed: not forward the bundle, forward the bundle as intended, or forward the bundle to one or more specific nodes within the network.¶
Attacks that involve rerouting the bundles throughout the network are essentially a special case of the modification attacks described in this section, one where the attacker is modifying fields within the primary block of the bundle. Given that BPSec cannot encrypt the contents of the primary block, alternate methods must be used to prevent this situation. These methods may include requiring BIBs for primary blocks, using encapsulation, or otherwise strategically manipulating primary block data. The details of any such mitigation technique are specific to the implementation of the deploying network and are outside of the scope of this document.¶
Furthermore, routing rules and policies may be useful in enforcing particular traffic flows to prevent topology attacks. While these rules and policies may utilize some features provided by BPSec, their definition is beyond the scope of this specification.¶
Olive is also able to generate new bundles and transmit them into the delay-tolerant network at will. These bundles may be either 1) copies or slight modifications of previously observed bundles (i.e., a replay attack) or 2) entirely new bundles generated based on the Bundle Protocol, BPSec, or other bundle-related protocols. With these attacks, Olive's objectives may vary, but may be targeting either the Bundle Protocol or application-layer protocols conveyed by the Bundle Protocol. The target could also be the storage and computing capabilities of the nodes running the bundle or application-layer protocols (e.g., a denial of service to flood on the storage of the store-and-forward mechanism or a computation that would process the bundles and perhaps prevent other activities).¶
BPSec relies on cipher suite capabilities to prevent replay or forged message attacks. A BCB used with appropriate cryptographic mechanisms may provide replay protection under certain circumstances. Alternatively, application data itself may be augmented to include mechanisms to assert data uniqueness and then be protected with a BIB, a BCB, or both along with other block data. In such a case, the receiving node would be able to validate the uniqueness of the data.¶
For example, a BIB may be used to validate the integrity of a bundle's primary block, which includes a timestamp and lifetime for the bundle. If a bundle is replayed outside of its lifetime, then the replay attack will fail as the bundle will be discarded. Similarly, additional blocks, such as the Bundle Age, may be signed and validated to identify replay attacks. Finally, security context parameters within BIBs and BCBs may include anti-replay mechanisms such as session identifiers, nonces, and dynamic passwords as supported by network characteristics.¶
Because of the diversity of networking scenarios and node capabilities that may utilize BPSec, there is a risk that a single security context mandated for every possible BPSec implementation is not feasible. For example, a security context appropriate for a resource-constrained node with limited connectivity may be inappropriate for use in a well-resourced, well-connected node.¶
This does not mean that the use of BPSec in a particular network is meant to happen without security contexts for interoperability and default behavior. Network designers must identify the minimal set of security contexts necessary for functions in their network. For example, a default set of security contexts could be created for use over the terrestrial Internet, and they could be required by any BPSec implementation communicating over the terrestrial Internet.¶
To ensure interoperability among various implementations, all BPSec implementations MUST support at least the current, mandatory security context(s) defined in IETF Standards Track RFCs. As of this writing, that BP mandatory security context is specified in [RFC9173], but the mandatory security context(s) might change over time in accordance with usual IETF processes. Such changes are likely to occur in the future if/when flaws are discovered in the applicable cryptographic algorithms, for example.¶
Additionally, BPSec implementations need to support the security contexts that are required by the BP networks in which they are deployed.¶
If a node serves as a gateway between two or more networks, the BPSec implementation at that node needs to support the union of security contexts mandated in those networks.¶
BPSec has been designed to allow for a diversity of security contexts and for new contexts to be defined over time. The use of different security contexts does not change the BPSec protocol itself, and the definition of new security contexts MUST adhere to the requirements of such contexts as presented in this section and generally in this specification.¶
Implementers should monitor the state of security context specifications to check for future updates and replacement.¶
Security blocks uniquely identify the security context to be used in the processing of their security services. The security context for a security block MUST be uniquely identifiable and MAY use parameters for customization.¶
To reduce the number of security contexts used in a network, security context designers should make security contexts customizable through the definition of security context parameters. For example, a single security context could be associated with a single cipher suite and security context parameters could be used to configure the use of this security context with different key lengths and different key management options without needing to define separate security contexts for each possible option.¶
A single security context may be used in the application of more than one security service. This means that a security context identifier MAY be used with a BIB, with a BCB, or with any other BPSec-compliant security block. The definition of a security context MUST identify which security services may be used with the security context, how security context parameters are interpreted as a function of the security operation being supported, and which security results are produced for each security service.¶
Network operators must determine the number, type, and configuration of security contexts in a system. Networks with rapidly changing configurations may define relatively few security contexts with each context customized with multiple parameters. For networks with more stability, or an increased need for confidentiality, a larger number of contexts can be defined with each context supporting few, if any, parameters.¶
| Context Type | Parameters | Definition |
|---|---|---|
| Key Exchange AES | Encrypted Key, IV | AES-GCM-256 cipher suite with provided ephemeral key encrypted with a predetermined key encryption key and cleartext initialization vector. |
| Pre-Shared Key AES | IV | AES-GCM-256 cipher suite with predetermined key and predetermined key-rotation policy. |
| Out-of-Band AES | None | AES-GCM-256 cipher suite with all info predetermined. |
Developers or implementers should consider the diverse performance and conditions of networks on which the Bundle Protocol (and, therefore, BPSec) will operate. Specifically, the delay and capacity of DTNs can vary substantially. Developers should consider these conditions to better describe the conditions in which those contexts will operate or exhibit vulnerability, and selection of these contexts for implementation should be made with consideration for this reality. There are key differences that may limit the opportunity for a security context to leverage existing cipher suites and technologies that have been developed for use in more reliable networks:¶
When developing security contexts for use with BPSec, the following information SHOULD be considered for inclusion in these specifications.¶
Security contexts MUST state whether their associated cipher suites generate ciphertext (to include any authentication information) that is of a different size than the input plaintext.¶
If a security context does not wish to alter the size of the plaintext, it should place overflow bytes and authentication tags in security result fields.¶
Other Security Blocks (OSBs) may be defined and used in addition to the security blocks identified in this specification. BIB, BCB, and any future OSBs can coexist within a bundle and can be considered in conformance with BPSec if all of the following requirements are met by any future identified security blocks.¶
Additionally, policy considerations for the management, monitoring, and configuration associated with blocks SHOULD be included in any OSB definition.¶
NOTE: The burden of showing compliance with processing rules is placed upon the specifications defining new security blocks, and the identification of such blocks shall not, alone, require maintenance of this specification.¶
This specification includes fields that require registries managed by IANA.¶
This specification allocates two block types from the existing "Bundle Block Types" registry defined in [RFC6255].¶
| Value | Description | Reference |
|---|---|---|
| 11 | Block Integrity | This document |
| 12 | Block Confidentiality | This document |
The "Bundle Block Types" registry notes whether a block type is meant for use in BP version 6, BP version 7 (BPv7), or both. The two block types defined in this specification are meant for use with BPv7.¶
This specification allocates five reason codes from the existing "Bundle Status Report Reason Codes" registry defined in [RFC6255].¶
| BP Version | Value | Description | Reference |
|---|---|---|---|
| 7 | 12 | Missing security operation | This document, Section 7.1 |
| 7 | 13 | Unknown security operation | This document, Section 7.1 |
| 7 | 14 | Unexpected security operation | This document, Section 7.1 |
| 7 | 15 | Failed security operation | This document, Section 7.1 |
| 7 | 16 | Conflicting security operation | This document, Section 7.1 |
BPSec has a Security Context Identifier field for which IANA has created a new registry named "BPSec Security Context Identifiers". Initial values for this registry are given below.¶
The registration policy for this registry is Specification Required (see [RFC8126]).¶
The value range: signed 16-bit integer.¶
| Value | Description | Reference |
|---|---|---|
| < 0 | Reserved | This document |
| 0 | Reserved | This document |
Negative security context identifiers are reserved for local/site-specific uses. The use of 0 as a security context identifier is for nonoperational testing purposes only.¶
The following participants contributed technical material, use cases, and useful thoughts on the overall approach to this security specification: Scott Burleigh of the IPNGROUP, Angela Hennessy of the Laboratory for Telecommunications Sciences, Amy Alford and Cherita Corbett of the Johns Hopkins University Applied Physics Laboratory (JHU/APL), and Angela Dalton of AMD Research.¶
Additionally, Benjamin Kaduk of Akamai Technologies provided a detailed technical review that resulted in a stronger and more precise specification.¶