| Internet-Draft | Implementation Considerations for EDHOC | October 2024 |
| Tiloca | Expires 24 April 2025 | [Page] |
This document provides considerations for guiding the implementation of the authenticated key exchange protocol Ephemeral Diffie-Hellman Over COSE (EDHOC).¶
This note is to be removed before publishing as an RFC.¶
Discussion of this document takes place on the Lightweight Authenticated Key Exchange Working Group mailing list (lake@ietf.org), which is archived at https://mailarchive.ietf.org/arch/browse/lake/.¶
Source for this draft and an issue tracker can be found at https://github.com/lake-wg/edhoc-impl-cons.¶
This Internet-Draft is submitted in full conformance with the provisions of BCP 78 and BCP 79.¶
Internet-Drafts are working documents of the Internet Engineering Task Force (IETF). Note that other groups may also distribute working documents as Internet-Drafts. The list of current Internet-Drafts is at https://datatracker.ietf.org/drafts/current/.¶
Internet-Drafts are draft documents valid for a maximum of six months and may be updated, replaced, or obsoleted by other documents at any time. It is inappropriate to use Internet-Drafts as reference material or to cite them other than as "work in progress."¶
This Internet-Draft will expire on 24 April 2025.¶
Copyright (c) 2024 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.¶
Ephemeral Diffie-Hellman Over COSE (EDHOC) [RFC9528] is a lightweight authenticated key exchange protocol, especially intended for use in constrained scenarios.¶
During the development of EDHOC, a number of side topics were raised and discussed, as emerging from reviews of the protocol latest design and from implementation activities. These topics were identified as strongly pertaining to the implementation of EDHOC rather than to the protocol in itself. Hence, they are not discussed in [RFC9528], which rightly focuses on specifying the actual protocol.¶
At the same time, implementors of an application using the EDHOC protocol or of an "EDHOC library" enabling its use cannot simply ignore such topics, and will have to take them into account throughout their implementation work.¶
In order to prevent multiple, independent re-discoveries and assessments of those topics, as well as to facilitate and guide implementation activities, this document collects such topics and discusses them through considerations about the implementation of EDHOC. At a high-level, the topics in question are summarized below.¶
Handling of completed EDHOC sessions when they become invalid, and of application keys derived from an EDHOC session when those become invalid. This topic is discussed in Section 2.¶
Enforcing of different trust models, with respect to learning new authentication credentials during an execution of EDHOC. This topic is discussed in Section 3.¶
Branched-off, side processing of incoming EDHOC messages, with particular reference to: i) fetching and validation of authentication credentials; and ii) processing of External Authorization Data (EAD) items, which in turn might play a role in the fetching and validation of authentication credentials. This topic is discussed in Section 4.¶
Effectively using EDHOC over the Constrained Application Protocol (CoAP) [RFC7252] in combination with Block-wise transfers for CoAP [RFC7959], possibly together with the optimized EDHOC execution workflow defined in [I-D.ietf-core-oscore-edhoc]. This topic is discussed in Section 5.¶
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.¶
The reader is expected to be familiar with terms and concepts related to the EDHOC protocol [RFC9528], the Constrained Application Protocol (CoAP) [RFC7252], and Block-wise transfers for CoAP [RFC7959].¶
This section considers the most common situation where, given a certain peer, only the application at that peer has visibility and control of both:¶
The EDHOC sessions at that peer; and¶
The application keys for that application at that peer, including the knowledge of whether they have been derived from an EDHOC session, i.e., by means of the EDHOC_Exporter interface after the successful completion of an execution of EDHOC (see Section 4.2 of [RFC9528]).¶
Building on the above, the following expands on three relevant cases concerning the handling of EDHOC sessions and application keys, in the event that any of those becomes invalid.¶
As a case in point to provide more concrete guidance, the following also considers the specific case where "applications keys" stands for the keying material and parameters that compose an OSCORE Security Context [RFC8613] and that are derived from an EDHOC session (see Appendix A.1 of [RFC9528]).¶
Nevertheless, the same considerations are applicable in case EDHOC is used to derive other application keys, e.g., to key different security protocols than OSCORE or to provide the application with secure values bound to an EDHOC session.¶
The application at a peer P may have learned that a completed EDHOC session S has to be invalidated. When S is marked as invalid, the application at P purges S and deletes each set of application keys (e.g., the OSCORE Security Context) that was generated from S.¶
Then, the application runs a new execution of the EDHOC protocol with the other peer. Upon successfully completing the EDHOC execution, the two peers derive and install a new set of application keys from this latest EDHOC session.¶
The flowchart in Figure 1 shows the handling of an EDHOC session that has become invalid.¶
An EDHOC session may have become invalid, for example, because an authentication credential CRED_X may have expired, or because the peer P may have learned from a trusted source that CRED_X has been revoked. This effectively invalidates CRED_X, and therefore also invalidates any EDHOC session where CRED_X was used as authentication credential of either peer in the session (i.e., P itself or the other peer). In such a case, the application at P has to additionally delete CRED_X and any stored, corresponding credential identifier.¶
The application at a peer P may have learned that a set of application keys is not safe to use anymore. When such a set is specifically an OSCORE Security Context, the application may have learned that from the used OSCORE library or from an OSCORE layer that takes part to the communication stack.¶
A current set SET of application keys shared with another peer can become unsafe to use, for example, due to the following reasons.¶
SET has reached a pre-determined expiration time; or¶
SET has been established to use for a now elapsed amount of time, according to enforced application policies; or¶
Some elements of SET have been used enough times to approach cryptographic limits that should not be passed, e.g., according to the properties of the specifically used security algorithms. With particular reference to an OSCORE Security Context, such limits are discussed in [I-D.ietf-core-oscore-key-limits].¶
When this happens, the application at the peer P proceeds as follows.¶
If the following conditions both hold, then the application moves to step 2. Otherwise, it moves to step 3.¶
Let us define S as the EDHOC session from which the peer P has derived SET or the eldest SET's ancestor set of application keys. Then, since the completion of S with the other peer, the application at P has received from the other peer at least one message protected with any set of application keys derived from S. That is, P has persisted S (see Section 5.4.2 of [RFC9528]).¶
The peer P supports a key update protocol, as an alternative to performing a new execution of EDHOC with the other peer. When SET is specifically an OSCORE Security Context, this means that the peer P supports the key update protocol KUDOS defined in [I-D.ietf-core-oscore-key-update].¶
The application at P runs the key update protocol mentioned at step 1 with the other peer, in order to update SET. When SET is specifically an OSCORE Security Context, this means that the application at P runs KUDOS with the other peer.¶
If the key update protocol terminates successfully, the updated application keys are installed and no further actions are taken. Otherwise, the application at P moves to step 3.¶
The application at the peer P performs the following actions.¶
It deletes SET.¶
It deletes the EDHOC session from which SET was generated, or from which the eldest SET's ancestor set of application keys was generated before any key update occurred (e.g., by means of the EDHOC_KeyUpdate interface defined in Appendix H of [RFC9528] or other key update methods).¶
It runs a new execution of the EDHOC protocol with the other peer. Upon successfully completing the EDHOC execution, the two peers derive and install a new set of application keys from this latest EDHOC session.¶
The flowchart in Figure 2 shows the handling of a set of application keys that has become invalid.¶
The following considers two peers that use the ACE framework for authentication and authorization in constrained environments [RFC9200], and specifically the EDHOC and OSCORE profile of ACE defined in [I-D.ietf-ace-edhoc-oscore-profile].¶
When doing so, one of the two peers acts as ACE resource server (RS) hosting protected resources. The other peer acts as ACE client and requests from an ACE authorization server (AS) an access token, which specifies access rights for accessing protected resources at the RS as well as the public authentication credential of the client, namely AUTH_CRED_C.¶
After that, C uploads the access token to the RS as part of the ACE workflow. This can occur before running EDHOC with the RS, or by means of an EAD item conveyed within an EDHOC message during the EDHOC execution. Alternatively, the AS can upload the access token to the RS on behalf of the client, as per the alternative workflow defined in [I-D.ietf-ace-workflow-and-params].¶
Consistent with the used EDHOC and OSCORE profile of ACE, the two peers run EDHOC in order to specifically derive an OSCORE Security Context as their shared set of application keys (see Appendix A.1 of [RFC9528]). At the RS, the access token is bound to the successfully completed EDHOC session and the established OSCORE Security Context.¶
After that, the peer acting as ACE client can access the protected resources hosted at the other peer, according to the access rights specified in the access token. The communications between the two peers are protected by means of the established OSCORE Security Context.¶
Later on, the application at one of the two peers P may have learned that the established OSCORE Security Context CTX is not safe to use anymore, e.g., from the used OSCORE library or from an OSCORE layer that takes part to the communication stack. The reasons that make CTX not safe to use anymore are the same ones discussed in Section 2.2 when considering a set of application keys in general, plus the event where the access token bound to CTX becomes invalid (e.g., it has expired or it has been revoked).¶
When this happens, the application at the peer P proceeds as follows.¶
If the following conditions both hold, then the application moves to step 2. Otherwise, it moves to step 3.¶
The access token is still valid. That is, it has not expired yet and the peer P is not aware that it has been revoked.¶
Let us define S as the EDHOC session from which the peer P has derived CTX or the eldest CTX's ancestor OSCORE Security Context. Then, since the completion of S with the other peer, the application at P has received from the other peer at least one message protected with any set of application keys derived from S. That is, P has persisted S (see Section 5.4.2 of [RFC9528]).¶
If the peer P supports the key update protocol KUDOS [I-D.ietf-core-oscore-key-update], then P runs KUDOS with the other peer, in order to update CTX. If the execution of KUDOS terminates successfully, the updated OSCORE Security Context is installed and no further actions are taken.¶
If the execution of KUDOS does not terminate successfully or if the peer P does not support KUDOS altogether, then the application at P moves to step 3.¶
The application at the peer P performs the following actions.¶
If the access token is not valid anymore, the peer P deletes all the EDHOC sessions associated with the access token, as well as the OSCORE Security Context derived from each of those sessions.¶
Note that, when specifically considering the EDHOC and OSCORE profile of ACE, an access token is associated with at most one EDHOC session (see Section 4.4 of [I-D.ietf-ace-edhoc-oscore-profile]).¶
If the peer P acted as ACE client, then P obtains from the ACE AS a new access token, which is uploaded to the other peer acting as ACE RS.¶
Finally, the application at P moves to step 4.¶
If the access token is valid while the OSCORE Security Context CTX is not, then the peer P deletes CTX.¶
After that, the peer P deletes the EDHOC session from which CTX was generated, or from which the eldest CTX's ancestor OSCORE Security Context was generated before any key update occurred (e.g., by means of KUDOS or other key update methods).¶
Finally, the application at P moves to step 4.¶
The peer P runs a new execution of the EDHOC protocol with the other peer. Upon successfully completing the EDHOC execution, the two peers derive and install a new OSCORE Security Context from this latest EDHOC session.¶
At the RS, the access token is bound to this latest EDHOC session and the newly established OSCORE Security Context.¶
The flowchart in Figure 3 shows the handling of an access token or of a set of application keys that have become invalid.¶
A peer P relies on authentication credentials of other peers, in order to authenticate those peers when running EDHOC with them.¶
There are different ways for P to acquire an authentication credential CRED of another peer. For example, CRED can be supplied to P out-of-band by a trusted provider.¶
Alternatively, CRED can be specified by the other peer during the EDHOC execution with P. This can rely on EDHOC message_2 or message_3, whose respective ID_CRED_R and ID_CRED_I can specify CRED by value, or instead a URI or other external reference where CRED can be retrieved from (see Section 3.5.3 of [RFC9528]).¶
Also during the EDHOC execution, an External Authorization Data (EAD) field might include an EAD item that specifies CRED by value or reference. This is the case, e.g., for the EAD items defined by the EDHOC and OSCORE profile of the ACE framework [I-D.ietf-ace-edhoc-oscore-profile]. In particular, they can be used for transporting (a reference to) an access token, which in turn specifies by value or by reference the public authentication credential of the EDHOC peer acting as ACE client.¶
When obtaining a new credential CRED, the peer P has to validate it before storing it. The validation steps to perform depend on the specific type of CRED (e.g., a public key certificate [RFC5280][I-D.ietf-cose-cbor-encoded-cert]), and can rely on (the authentication credential of) a trusted third party acting as a trust anchor.¶
Upon retrieving a new CRED through the processing of a received EDHOC message and following the successful validation of CRED, the peer P stores CRED only if it assesses CRED to be also trusted, and must not store CRED otherwise.¶
An exception applies for the two unauthenticated operations described in Appendix D.5 of [RFC9528], where a trust relationship with an unknown or not-yet-trusted endpoint is established later. That is, CRED is verified out-of-band at a later stage, or an EDHOC session key is bound to an identity out-of-band at a later stage.¶
If P stores CRED, then P will consider CRED as valid and trusted until it possibly becomes invalid, e.g., because it expires or because P gains knowledge that it has been revoked.¶
When storing CRED, the peer P should generate the authentication credential identifier(s) corresponding to CRED and store them as associated with CRED. For example, if CRED is a public key certificate, an identifier of CRED can be the hash of the certificate. In general, P should generate and associate with CRED one corresponding identifier for each type of authentication credential identifier that P supports and that is compatible with CRED.¶
In future executions of EDHOC with the other peer associated with CRED, this allows such other peer to specify CRED by reference, e.g., by indicating its credential identifier as ID_CRED_R/ID_CRED_I in the EDHOC message_2 or message_3 addressed to the peer P. In turn, this allows P to retrieve CRED from its local storage.¶
When processing a received EDHOC message M that specifies an authentication credential CRED, the peer P can enforce one of the following trust policies in order to determine whether to trust CRED.¶
NO-LEARNING: according to this policy, the peer P trusts CRED only if P is already storing CRED at message reception time, unless in cases where situation-specific exceptions apply and are deliberately enforced (see below).¶
That is, upon receiving M, the peer P should continue the execution of EDHOC only if both the following conditions hold.¶
P currently stores CRED, as specified by reference or by value in ID_CRED_I/ID_CRED_R or in the value of an EAD item; and¶
CRED is still valid, i.e., P believes CRED to not be expired or revoked.¶
Exceptions may apply and be actually enforced in cases where, during an EDHOC execution, P obtains additional information that allows it to trust and validate CRED, even though CRED is not already stored upon receiving M. Such exceptions typically rely on a trusted party that vouches for CRED, such as in the following cases:¶
In the procedure defined in [I-D.ietf-lake-authz], the EDHOC Initiator U receives an EDHOC message_2 where ID_CRED_R specifies the authentication credential of the EDHOC Responder V by value. In the same EDHOC message_2, an EAD item specifies a voucher issued by a trusted enrollment server W, which conveys authorization information about V and V's authentication credential CRED. Through a successful verification of the voucher, U is able to trust CRED (if found valid), even though it did not already store CRED upon receiving EDHOC message_2.¶
In the EDHOC and OSCORE profile of the ACE framework defined in [I-D.ietf-ace-edhoc-oscore-profile], the EDHOC peer acting as ACE client C can include an EAD item in an EDHOC message sent to the other EDHOC peer acting as ACE resource server (RS). The EAD item transports (a reference to) an access token issued by a trusted ACE authorization server (AS). In turn, the access token specifies CRED as the authentication credential of C, by value or by reference. Through a successful verification of the access token, the RS is able to trust CRED (if found valid), even if not already storing it upon receiving the EDHOC message with the EAD item.¶
If the peer P admits such an exception and actually enforces it on an authentication credential CRED, then P effectively handles CRED according to the trust policy "LEARNING" specified below.¶
LEARNING: according to this policy, the peer P trusts CRED even if P is not already storing CRED at message reception time.¶
That is, upon receiving M, the peer P performs the following steps.¶
P retrieves CRED, as specified by reference or by value in ID_CRED_I/ID_CRED_R or in the value of an EAD item.¶
P checks whether CRED is already being stored and if it is still valid. In such a case, P trusts CRED and can continue the EDHOC execution. Otherwise, P moves to step 3.¶
P attempts to validate CRED. If the validation process is not successful, P aborts the EDHOC session with the other peer. Otherwise, P trusts and stores CRED, and can continue the EDHOC execution.¶
Irrespective of the adopted trust policy, P actually uses CRED only if it is determined to be fine to use in the context of the ongoing EDHOC session, also depending on the specific identity of the other peer (see Sections 3.5 and D.2 of [RFC9528]). If this is not the case, P aborts the EDHOC session with the other peer.¶
As discussed in Section 2.3, two EDHOC peers can be using the ACE framework [RFC9200] and specifically the EDHOC and OSCORE profile of ACE defined in [I-D.ietf-ace-edhoc-oscore-profile].¶
In this case, one of the two EDHOC peers, namely PEER_RS, acts as ACE resource server (RS). Instead, the other EDHOC peer, namely PEER_C, acts as ACE client, and obtains from the ACE authorization server (AS) an access token for accessing protected resources at PEER_RS.¶
Together with other information, the access token specifies (by value or by reference) the public authentication credential AUTH_CRED_C that PEER_C is going to use when running EDHOC with PEER_RS. Note that AUTH_CRED_C will be used as either CRED_I or CRED_R, depending on whether the two peers use the EDHOC forward message flow (i.e., PEER_C is the EDHOC Initiator) or the EDHOC reverse message flow (i.e., PEER_C is the EDHOC Responder), respectively (see Appendix A.2 of [RFC9528]).¶
When the AS issues the first access token that specifies AUTH_CRED_C and is intended to be uploaded to PEER_RS, it is expected that the access token specifies AUTH_CRED_C by value, and that PEER_RS is not currently storing AUTH_CRED_C, but instead will obtain it upon receiving the access token.¶
While such an access token can be uploaded to PEER_RS before running EDHOC, it is also possible for PEER_C to upload it through a dedicated EAD item, while running EDHOC with PEER_RS. In such a case, PEER_RS has to learn AUTH_CRED_C as a new public authentication credential during an EDHOC session.¶
At least for its EDHOC resource used for exchanging the EDHOC messages of the EDHOC session in question, this requires PEER_RS to:¶
This section describes an approach that EDHOC peers can use upon receiving EDHOC messages, in order to fetch/validate authentication credentials and to process External Authorization Data (EAD) items.¶
As per Section 9.1 of [RFC9528], the EDHOC protocol provides a transport mechanism for conveying EAD items, but specifications defining those items have to set the ground for "agreeing on the surrounding context and the meaning of the information passed to and from the application".¶
The approach described in this section aims to help implementors navigate the surrounding context mentioned above, irrespective of the specific EAD items conveyed in the EDHOC messages. In particular, the described approach takes into account the following points.¶
The fetching and validation of the other peer's authentication credential relies on ID_CRED_I in EDHOC message_2, or on ID_CRED_R in EDHOC message_3, or on the value of an EAD item. When this occurs upon receiving EDHOC message_2 or message_3, the decryption of the EDHOC message has to be completed first.¶
The validation of the other peer's authentication credential might depend on using the value of an EAD item, which in turn has to be validated first. For instance, an EAD item within the EAD_2 field may contain a voucher to be used for validating the other peer's authentication credential (see [I-D.ietf-lake-authz]).¶
Some EAD items may be processed only after having successfully verified the EDHOC message, i.e., after a successful verification of the Signature_or_MAC field.¶
For instance, an EAD item within the EAD_3 field may contain a Certificate Signing Request (CSR) [RFC2986]. Hence, such an EAD item can be processed only once the recipient peer has attained proof of the other peer possessing its private key.¶
In order to conveniently handle such processing, the application can prepare in advance one "side-processor object" (SPO), which takes care of the operations above during the EDHOC execution.¶
In particular, the application provides EDHOC with the SPO before starting an EDHOC execution, during which EDHOC will temporarily transfer control to the SPO at the right point in time, in order to perform the required side-processing of an incoming EDHOC message.¶
Furthermore, the application has to instruct the SPO about how to prepare any EAD item such that: it has to be included in an outgoing EDHOC message; and it is independent of the processing of other EAD items included in incoming EDHOC messages. This includes, for instance, the preparation of padding EAD items.¶
At the right point in time during the processing of an incoming EDHOC message M at the peer P, EDHOC invokes the SPO and provides it with the following input:¶
When M is EDHOC message_2 or message_3, an indication of whether this invocation is happening before or after the message verification (i.e., before or after having verified the Signature_or_MAC field).¶
The full set of information related to the current EDHOC session. This especially includes the selected cipher suite and the ephemeral Diffie-Hellman public keys G_X and G_Y that the two peers have exchanged in the EDHOC session.¶
The other peers' authentication credentials that the peer P stores.¶
All the decrypted information elements retrieved from M.¶
The EAD items included in M.¶
Note that EDHOC might do some preliminary work on M before invoking the SPO, in order to provide the SPO only with actually relevant EAD items. This requires the application to additionally provide EDHOC with the ead_labels of the EAD items that the peer P recognizes (see Section 3.8 of [RFC9528]).¶
With such information available, EDHOC can early abort the current session in case M includes any EAD item which is both critical and not recognized by the peer P.¶
If no such EAD items are found, EDHOC can remove any padding EAD item (see Section 3.8.1 of [RFC9528]), as well as any EAD item which is neither critical nor recognized (since the SPO is going to ignore it anyway). As a result, EDHOC is able to provide the SPO only with EAD items that will be recognized and that require actual processing.¶
Note that, after having processed the EAD items, the SPO might actually need to store them throughout the whole EDHOC execution, e.g., in order to refer to them also when processing later EDHOC messages in the current EDHOC session.¶
The SPO performs the following tasks on an incoming EDHOC message M.¶
The SPO fetches and/or validates the other peer's authentication credential CRED, based on a dedicated EAD item of M or on the ID_CRED field of M (for EDHOC message_2 or message_3).¶
The SPO processes the EAD items conveyed in the EAD field of M.¶
The SPO stores the results of the performed operations, and makes such results available to the application.¶
When the SPO has completed its side processing and transfers control back to EDHOC, the SPO provides EDHOC with the produced EAD items to include in the EAD field of the next outgoing EDHOC message. The production of such EAD items can be triggered, e.g., by:¶
In the following, Section 4.1 to Section 4.3 describe more in detail the actions performed by the SPO on the different, incoming EDHOC messages.¶
Then, Section 4.4 describes further, special handling of incoming EDHOC messages in particular situations.¶
During the processing of an incoming EDHOC message_1, EDHOC invokes the SPO only once, after the Responder peer has successfully decoded the message and accepted the selected cipher suite.¶
If the EAD_1 field is present, the SPO processes the EAD items included therein.¶
Once all such EAD items have been processed the SPO transfers control back to EDHOC. When doing so, the SPO also provides EDHOC with any produced EAD items to include in the EAD field of the next outgoing EDHOC message.¶
Then, EDHOC resumes its execution and advances its protocol state.¶
During the processing of an incoming EDHOC message_4, EDHOC invokes the SPO only once, after the Initiator peer has successfully decrypted the message.¶
If the EAD_4 field is present, the SPO processes the EAD items included therein.¶
Once all such EAD items have been processed, the SPO transfers control back to EDHOC, which resumes its execution and advances its protocol state.¶
The following refers to "message_X" as an incoming EDHOC message_2 or message_3, and to "message verification" as the verification of Signature_or_MAC_X in message_X.¶
During the processing of a message_X, EDHOC invokes the SPO two times:¶
Right after message_X has been decrypted and before its verification starts. Following this invocation, the SPO performs the actions described in Section 4.3.1.¶
Right after message_X has been successfully verified. Following this invocation, the SPO performs the actions described in Section 4.3.2.¶
The flowcharts in Section 4.3.3 show the high-level interaction between the core EDHOC processing and the SPO, as well as the different steps taken for processing an incoming message_X.¶
The pre-verification side processing occurs in two sequential phases, namely PHASE_1 and PHASE_2.¶
PHASE_1 - During PHASE_1, the SPO at the recipient peer P determines CRED, i.e., the other peer's authentication credential to use in the ongoing EDHOC session. In particular, the SPO performs the following steps.¶
The SPO determines CRED based on ID_CRED_X or on an EAD item in message_X.¶
Those may specify CRED by value or by reference, including a URI or other external reference where CRED can be retrieved from.¶
If CRED is already installed, the SPO moves to step 2. Otherwise, the SPO moves to step 3.¶
The SPO determines if the stored CRED is currently valid, e.g., by asserting that CRED has not expired and has not been revoked.¶
Performing such a validation may require the SPO to first process an EAD item included in message_X. For example, it can be an EAD item in EDHOC message_2, which confirms or revokes the validity of CRED_R specified by ID_CRED_R, as the result of an OCSP process [RFC6960].¶
In case CRED is determined to be valid, the SPO moves to step 9. Otherwise, the SPO moves to step 11.¶
The SPO attempts to retrieve CRED, and then moves to step 4.¶
If the retrieval of CRED has succeeded, the SPO moves to step 5. Otherwise, the SPO moves to step 11.¶
If the enforced trust policy for new authentication credentials is "NO-LEARNING" and P does not admit any exceptions that are acceptable to enforce for message_X (see Section 3), the SPO moves to step 11. Otherwise, the SPO moves to step 6.¶
If this step has been reached, the peer P is not already storing the retrieved CRED and, at the same time, it enforces either the trust policy "LEARNING" or the trust policy "NO-LEARNING" while also enforcing an exception acceptable for message_X (see Section 3).¶
Consistently, the SPO determines if CRED is currently valid, e.g., by asserting that CRED has not expired and has not been revoked. Then, the SPO moves to step 7.¶
Validating CRED may require the SPO to first process an EAD item included in message_X. For example, it can be an EAD item in EDHOC message_2 that: i) specifies a voucher for validating CRED_R as a CWT Claims Set (CCS) [RFC8392] transported by value in ID_CRED_R (see [I-D.ietf-lake-authz]); or instead ii) an OCSP response [RFC6960] for validating CRED_R as a certificate transported by value or reference in ID_CRED_R.¶
If CRED has been determined valid, the SPO moves to step 8. Otherwise, the SPO moves to step 11.¶
The SPO stores CRED as a valid and trusted authentication credential associated with the other peer, together with corresponding authentication credential identifiers (see Section 3). Then, the SPO moves to step 9.¶
The SPO checks if CRED is fine to use in the context of the ongoing EDHOC session, also depending on the specific identity of the other peer (see Sections 3.5 and D.2 of [RFC9528]).¶
If this is the case, the SPO moves to step 10. Otherwise, the SPO moves to step 11.¶
P uses CRED as authentication credential of the other peer in the ongoing EDHOC session.¶
Then, PHASE_1 ends, and the pre-verification side processing moves to the next PHASE_2 (see below).¶
The SPO has not found a valid authentication credential associated with the other peer that can be used in the ongoing EDHOC session. Therefore, the EDHOC session with the other peer is aborted.¶
PHASE_2 - During PHASE_2, the SPO processes any EAD item included in message_X such that both the following conditions hold.¶
The EAD item has not been already processed during PHASE_1.¶
The EAD item can be processed before performing the verification of message_X.¶
Once all such EAD items have been processed, the SPO transfers control back to EDHOC, which either aborts the ongoing EDHOC session or continues the processing of message_X with its corresponding message verification.¶
During the post-verification side processing, the SPO processes any EAD item included in message_X such that the processing of that EAD item had to wait for completing the successful message verification.¶
The late processing of such EAD items is typically due to the fact that a pre-requirement has to be fulfilled first. For example, the recipient peer P has to have first asserted that the other peer does possess the private key corresponding to the public key of the other peer's authentication credential CRED determined during the pre-verification side processing (see Section 4.3.1). This requirement is fulfilled after a successful message verification of message_X.¶
Once all such EAD items have been processed, the SPO transfers control back to EDHOC. When doing so, the SPO also provides EDHOC with any produced EAD items to include in the EAD field of the next outgoing EDHOC message.¶
Then, EDHOC resumes its execution and advances its protocol state.¶
The flowchart in Figure 4 shows the high-level interaction between the core EDHOC processing and the SPO, with particular reference to an incoming EDHOC message_2 or message_3.¶
The flowchart in Figure 5 shows the different steps taken for processing an incoming EDHOC message_2 and message_3.¶
This section describes methods to perform special handling of incoming EDHOC messages in particular situations.¶
Section 3.1 discusses the case where two EDHOC peers use the ACE framework [RFC9200] and specifically the EDHOC and OSCORE profile of ACE defined in [I-D.ietf-ace-edhoc-oscore-profile].¶
In particular, Section 3.1 considers a peer PEER_C that, when running EDHOC with a peer PEER_RS, uses a dedicated EAD item for uploading an access token at PEER_RS. In turn, the access token specifies AUTH_CRED_C as the public authentication credential of PEER_C.¶
As also discussed in Section 3.1, this practically requires PEER_RS to enforce either the trust policy "LEARNING" or the trust policy "NO-LEARNING" with situation-specific, overriding exceptions (see Section 3), at least for the EDHOC resource used for exchanging the EDHOC messages of the EDHOC session in question.¶
However, PEER_RS might have reasons to enforce the trust policy "NO_LEARNING" with no exceptions. In such a case, unless PEER_RS already stores AUTH_CRED_C, the upload of the access token by means of the EAD item has to consistently fail, and PEER_RS has to consequently abort the EDHOC session.¶
This requires PEER_RS to perform an early check of the access token conveyed in the EAD item, in order to retrieve AUTH_CRED_C and determine whether it is already storing AUTH_CRED_C. The SPO can perform such a task, as part of steps 3 and 4 of PHASE_1 of the pre-verification side processing (see Section 4.3.1).¶
In order to be reliable, this task requires to perform a cryptographic validation of the access token. Moreover, the access token can be encrypted, which makes the actual retrieval of AUTH_CRED_C not straightforward. In practice, the SPO has a number of viable options.¶
The SPO can directly perform the necessary cryptographic processing of the access token, including decryption if necessary.¶
This requires to provide the SPO with the same security context, parameters, and cryptographic material that are available at a token validator component of PEER_RS, and that are shared with the ACE authorization server that has issued the access token.¶
If the SPO fails the validation and cryptographic processing of the access token, or if it does not already store the AUTH_CRED_C specified by the access token, then the SPO aborts the EDHOC session.¶
A downside of this approach is that the token validator component of PEER_RS will later repeat the cryptographic processing of the access token, when eventually provided by the SPO with the access token like if the latter was uploaded to PEER_RS as normally expected in the ACE framework.¶
The SPO can engage in an internal, synchronous exchange with a token validator component of PEER_RS that is normally responsible for the validation and cryptographic processing of access tokens. For example, the SPO can proceed as follows.¶
The SPO creates a copy CREDS_BEFORE of the stored authentication credentials of its other EDHOC peers.¶
The SPO provides the token validator component of PEER_RS with the access token to undergo cryptographic processing and validation. If such process succeeds, the token validator component stores the access token, and adds AUTH_CRED_C to the current set of authentication credentials of other EDHOC peers.¶
After learning of the successful validation of the access token from the token validator component, the SPO checks whether the set of currently stored authentication credentials of other EDHOC peers includes an element that is not included in CREDS_BEFORE.¶
In case of a positive match, then the found element is AUTH_CRED_C, which PEER_RS was not supposed to learn as per the enforced "NO-LEARNING" policy.¶
Consequently, the SPO requests the token validator element to purge the access token, deletes CREDS_BEFORE, deletes AUTH_CRED_C from the set of currently stored authentication credentials of other EDHOC peers, and aborts the EDHOC session.¶
Editor's note: TODO¶
The following are placeholder notes that specifically consider the EDHOC and OSCORE profile of ACE [I-D.ietf-ace-edhoc-oscore-profile].¶
While that profile is a relevant case in point to consider, the content of this section should discuss the topic in general. With respect to ID_CRED_R/ID_CRED_I, an access token is just an example of additional means to specify an authentication credential.¶
When it comes specifically to the EDHOC and OSCORE profile of ACE, some of this content might better fit in [I-D.ietf-ace-edhoc-oscore-profile], while this document can keep only implementation-specific guidelines from a general point of view.¶
AUTH_CRED_C is always specified (by value or by reference) in ID_CRED_R (ID_CRED_I) of EDHOC message_2 (EDHOC message_3).¶
AUTH_CRED_C can also be specified (by value or by reference) within an access token, which can be conveyed by an EAD item in an EDHOC message that PEER_C sends to PEER_RS. The details also depend on the two EDHOC peers using either the EDHOC forward message flow or the EDHOC reverse message flow (see Appendix A.2 of [RFC9528]).¶
When such an access token is uploaded by means of an EAD item, PEER_RS has to perform consistency checks between the AUTH_CRED_C specified in ID_CRED_R or ID_CRED_I on one hand, and the AUTH_CRED_C specified within the access token on the other hand.¶
This needs to explain in general terms when PEER_RS becomes able to perform the consistency check in different cases, which differ as to the use of the EDHOC forward message flow or of the EDHOC reverse message flow, and as to the specific EDHOC message including the EAD item that conveys the access token including AUTH_CRED_C.¶
Appendix A.2 of [RFC9528] specifies how to transfer EDHOC over CoAP [RFC7252]. In such a case, the EDHOC messages (possibly prepended by an EDHOC connection identifier) are transported in the payload of CoAP requests and responses, according to the EDHOC forward message flow or the EDHOC reverse message flow. Furthermore, Appendix A.1 of [RFC9528] specifies how to derive an OSCORE Security Context [RFC8613] from an EDHOC session.¶
Building on that, [I-D.ietf-core-oscore-edhoc] further details the use of EDHOC with CoAP and OSCORE, and specifies an optimization approach for the EDHOC forward message flow that combines the EDHOC execution with the first subsequent OSCORE transaction. This is achieved by means of an "EDHOC + OSCORE request", i.e., a single CoAP request message that conveys both EDHOC message_3 of the ongoing EDHOC session and the OSCORE-protected application data, where the latter is protected with the OSCORE Security Context derived from that EDHOC session.¶
This section provides guidelines and recommendations for CoAP endpoints supporting Block-wise transfers for CoAP [RFC7959] and specifically for CoAP clients supporting the EDHOC + OSCORE request defined in [I-D.ietf-core-oscore-edhoc]. The use of Block-wise transfers can be desirable, e.g., for EDHOC messages that include a large ID_CRED_I or ID_CRED_R, or that include a large EAD field.¶
The following especially considers a CoAP endpoint that may perform only "inner" Block-wise, but not "outer" Block-wise operations (see Section 4.1.3.4 of [RFC8613]). That is, the considered CoAP endpoint does not (further) split an OSCORE-protected message like an intermediary (e.g., a proxy) might do. This is the typical case for CoAP endpoints using OSCORE (see Section 4.1.3.4 of [RFC8613]).¶
The rest of this section refers to the following notation.¶
SIZE_BODY: the size in bytes of the data to be transmitted with CoAP. When Block-wise is used, such data is referred to as the "body" to be fragmented into blocks, each of which to be transmitted in one CoAP message.¶
With the exception of EDHOC message_3, the considered body can also be an EDHOC message, possibly prepended by an EDHOC connection identifier encoded as per Section 3.3 of [RFC9528].¶
When specifically using the EDHOC + OSCORE request, the considered body is the application data to be protected with OSCORE, (whose first block is) to be sent together with EDHOC message_3 as part of the EDHOC + OSCORE request.¶
SIZE_EDHOC_M3: the size in bytes of EDHOC message_3, if this is sent as part of the EDHOC + OSCORE request. Otherwise, the size in bytes of EDHOC message_3, plus, if included, the size in bytes of a prepended EDHOC connection identifier encoded as per Section 3.3 of [RFC9528].¶
SIZE_MTU: the maximum amount of transmittable bytes before having to use Block-wise. This is, for example, 64 KiB as maximum datagram size when using UDP, or 1280 bytes as the maximum size for an IPv6 MTU.¶
SIZE_OH: the size in bytes of the overall overhead due to all the communication layers underlying the application. This takes into account also the overhead introduced by the OSCORE processing.¶
LIMIT = (SIZE_MTU - SIZE_OH): the practical maximum size in bytes to be considered by the application before using Block-wise.¶
SIZE_BLOCK: the size in bytes of inner blocks.¶
ceil(): the ceiling function.¶
Before sending an EDHOC + OSCORE request, a CoAP client has to perform the following checks. Note that, while the CoAP client is able to fragment the plain application data before any OSCORE processing, it cannot fragment the EDHOC + OSCORE request or the EDHOC message_3 added therein.¶
If inner Block-wise is not used, hence SIZE_BODY <= LIMIT, the CoAP client must verify whether all the following conditions hold:¶
If inner Block-wise is used, the CoAP client must verify whether all the following conditions hold:¶
In either case, if not all the corresponding conditions hold, the CoAP client should not send the EDHOC + OSCORE request. Instead, the CoAP client can continue by switching to the purely sequential, original EDHOC workflow (see Appendix A.2.1 of [RFC9528]). That is, the CoAP client first sends EDHOC message_3 prepended by the EDHOC Connection Identifier C_R encoded as per Section 3.3 of [RFC9528], and then sends the OSCORE-protected CoAP request once the EDHOC execution is completed.¶
In order to avoid further fragmentation at lower layers when sending an EDHOC + OSCORE request, the CoAP client has to use inner Block-wise if any of the following conditions holds:¶
In particular, consistently with Section 5.2, the used SIZE_BLOCK has to be such that the following condition also holds:¶
COND7: (SIZE_BLOCK + SIZE_EDHOC_M3) <= LIMIT¶
Note that the CoAP client might still use Block-wise due to reasons different from exceeding the size indicated by LIMIT.¶
The following shows the number of round trips for completing both the EDHOC execution and the first OSCORE-protected exchange, under the assumption that the exchange of EDHOC message_1 and EDHOC message_2 do not result in using Block-wise.¶
If both the conditions COND5 and COND6 hold, the use of Block-wise results in the following number of round trips experienced by the CoAP client.¶
If the original EDHOC execution workflow is used (see Appendix A.2.1 of [RFC9528]), the number of round trips RT_ORIG is equal to 1 + ceil(SIZE_EDHOC_M3 / SIZE_BLOCK) + ceil(SIZE_BODY / SIZE_BLOCK).¶
If the optimized EDHOC execution workflow is used (see Section 3 of [I-D.ietf-core-oscore-edhoc]), the number of round trips RT_COMB is equal to 1 + ceil(SIZE_BODY / SIZE_BLOCK).¶
It follows that RT_COMB < RT_ORIG, i.e., the optimized EDHOC execution workflow always yields a lower number of round trips.¶
Instead, the convenience of using the optimized EDHOC execution workflow becomes questionable if both the following conditions hold:¶
That is, since SIZE_BODY <= LIMIT, using Block-wise would not be required when using the original EDHOC execution workflow, provided that SIZE_EDHOC_M3 <= LIMIT still holds.¶
At the same time, using the combined workflow is in itself what actually triggers the use of Block-wise, since (SIZE_BODY + SIZE_EDHOC_M3) > LIMIT.¶
Therefore, the following round trips are experienced by the CoAP client.¶
The original EDHOC execution workflow run without using Block-wise results in a number of round trips RT_ORIG equal to 3.¶
The optimized EDHOC execution workflow run using Block-wise results in a number of round trips RT_COMB equal to 1 + ceil(SIZE_BODY / SIZE_BLOCK).¶
It follows that RT_COMB >= RT_ORIG, i.e., the optimized EDHOC execution workflow might still be not worse than the original EDHOC execution workflow in terms of round trips. This is the case only if the used SIZE_BLOCK is such that ceil(SIZE_BODY / SIZE_BLOCK) is equal to 2, i.e., the plain application data is fragmented into only 2 inner blocks, and thus the EDHOC + OSCORE request is followed by only one more request message transporting the last block of the body.¶
However, even in such a case, there would be no advantage in terms or round trips compared to the original workflow, while still requiring the CoAP client and the CoAP server to perform the processing due to using the EDHOC + OSCORE request and Block-wise transferring.¶
Therefore, if both the conditions COND8 and COND9 hold, the CoAP client should not send the EDHOC + OSCORE request. Instead, the CoAP client should continue by switching to the original EDHOC execution workflow. That is, the CoAP client first sends EDHOC message_3 prepended by the EDHOC Connection Identifier C_R encoded as per Section 3.3 of [RFC9528], and then sends the OSCORE-protected CoAP request once the EDHOC execution is completed.¶
This document has no actions for IANA.¶
This section is to be removed before publishing as an RFC.¶
The author sincerely thanks Christian Amsüss, Geovane Fedrecheski, Rikard Höglund, John Preuß Mattsson, Göran Selander, and Mališa Vučinić for their comments and feedback.¶
The work on this document has been partly supported by the Sweden's Innovation Agency VINNOVA and the Celtic-Next project CYPRESS.¶