**Session Date/Time:** 23 Jul 2026 12:00 # [CORE](../wg/core.html) ## Summary The CORE working group met at IETF 126. The session covered status updates on working group documents, three post-Working Group Last Call (WGLC) reviews, several active working group drafts, and four new individual submissions. Key advancements included progressing URI path abbreviations and conditional observe query parameters toward shepherding, successful interoperability testing for KUDOS (Key Update for OSCORE), and discussions on non-traditional response forms, CORECONF datastores, OSCORE-capable proxies, quantum-resistant Group OSCORE, DNS-based SID files, asynchronous task resources, and diagnostic notation. --- ## Key Discussion Points ### 1. WG Status and Administration * **Presenter:** Marco Tiloca * **Slides:** [Chairs' slides](https://datatracker.ietf.org/meeting/126/materials/slides-126-core-chairs-slides-02) * **Status Updates:** Three new RFCs have been published since the last meeting. Three documents are currently approved for publication at the RFC Editor. The CoAP PubSub Architecture was successfully submitted to the IESG. --- ### 2. URI Path Abbreviation in CoAP * **Presenter:** Christian Amsüss * **Draft:** [draft-ietf-core-uri-path-abbrev](https://datatracker.ietf.org/meeting/126/materials/slides-126-core-uri-path-abbrev) * **Slides:** [URI Path abbreviation in CoAP](https://datatracker.ietf.org/meeting/126/materials/slides-126-core-uri-path-abbreviation-in-coap-02) * **Discussion:** * Christian Amsüss noted that recent updates incorporated editorial feedback and shifted the terminology from "reject" to "refuse" to avoid clashing with formal CoAP/UDP definitions. * In response to early IANA reviews, registry text was reorganized to exclude designated experts and move relevant passages directly into the main text body. * Carsten Bormann volunteered to sheperd the document. --- ### 3. Conditional Query Parameters for CoAP Observe * **Presenter:** Bill Silverajan * **Draft:** [draft-ietf-core-conditional-attributes](https://datatracker.ietf.org/meeting/126/materials/slides-126-core-conditional-attributes) * **Slides:** [Conditional Query Parameters for CoAP Observe](https://datatracker.ietf.org/meeting/126/materials/slides-126-core-conditional-query-parameters-for-coap-observe-00) * **Discussion:** * Bill Silverajan presented version 13, highlighting the resolution of GitHub issue #59. Discovery is now tentatively handled using the interface (`if`) attribute `core.conditional`. * The group confirmed that if a server receives an unsupported conditional observe parameter, it should fall back to treating the request as a standard observe (no-op on the unsupported condition) rather than rejecting the entire request. * Marco Tiloca (acting as shepherd) and Carsten Bormann confirmed that the draft is ready to proceed to the shepherding stage despite the slightly mismatched filename. --- ### 4. Key Update for OSCORE (KUDOS) * **Presenter:** Rikard Höglund * **Draft:** [draft-ietf-core-oscore-key-update](https://datatracker.ietf.org/meeting/126/materials/slides-126-core-oscore-key-update) * **Slides:** [Slides KUDOS CoRE IETF 126](https://datatracker.ietf.org/meeting/126/materials/slides-126-core-slides-kudos-core-ietf-126-00) * **Discussion:** * Rikard Höglund outlined updates in version 14, including the removal of a redundant YANG model for SCHC and the addition of an implementation status section. * Successful hackathon interoperability tests were performed in forward secrecy mode between Java (Californium) and C (uOSCORE/uEDHOC) implementations. * Consecutive execution tests revealed a minor bug where the replay window failed to reset on context re-derivation, which the authors plan to resolve. * Lessons learned include clarifying the sorting mechanism to use "short lex order" and prefixing vector data types with their lengths per the TLS RFC. * Christian Amsüss noted via chat that he has editorial suggestions regarding renaming the "x byte." --- ### 5. Observe Notifications as CoAP Multicast Responses * **Presenter:** Marco Tiloca * **Draft:** [draft-ietf-core-observe-multicast-notifications](https://datatracker.ietf.org/meeting/126/materials/slides-126-core-observe-multicast-notifications-00) * **Slides:** [Observe Multicast Notifications](https://datatracker.ietf.org/meeting/126/materials/slides-126-core-observe-multicast-notifications-00) * **Discussion:** * Marco Tiloca detailed updates in versions 14 and 15, which include the optimization of phantom requests using [draft-ietf-core-cacheable-oscore](https://datatracker.ietf.org/meeting/126/materials/slides-126-core-cacheable-oscore) and the generalization of the client-counting option into a generic "nudge" option. * Clarifications were added regarding the age calculation of multicast notifications inside the error response payload (`last-notice` field) and making transport details (`tp-info`) optional for specialized deployments. * A successful Java-based implementation in Eclipse Californium was completed. * Carsten Bormann suggested recruiting more implementers (e.g., Christian Amsüss) to test interoperability before issuing WGLC. Alexander Pelov expressed interest in drafting a Python implementation. --- ### 6. CoAP: Non-traditional Response Forms * **Presenter:** Carsten Bormann * **Draft:** [draft-ietf-core-responses](https://datatracker.ietf.org/meeting/126/materials/slides-126-core-responses) * **Slides:** [CoAP Non-traditional Response Forms](https://datatracker.ietf.org/meeting/126/materials/slides-126-core-coap-non-traditional-response-forms-00) * **Discussion:** * Carsten Bormann explored response types deviating from the classic 1-to-1 paradigm, covering "configured requests" and "embedded requests" (e.g., using a `Response-For` option). * The `Leisure-For-Responses` option was proposed to specify response thresholds. * The group agreed to place the complex `Respond-To` option on the backburner (documenting its design considerations in an appendix) while progressing with the safer `Response-For` and `Leisure-For-Responses` options. * Esko Dijk and Marco Tiloca discussed naming and functional overlaps with the "multicast timeout" options defined in [draft-ietf-core-groupcomm-proxy](https://datatracker.ietf.org/meeting/126/materials/slides-126-core-groupcomm-proxy). --- ### 7. CoAP Management Interface (CORECONF 'unified' Datastore) * **Presenter:** Wojciech Kozaczynski * **Draft:** [draft-ietf-core-comi](https://datatracker.ietf.org/meeting/126/materials/slides-126-core-comi) * **Slides:** [CORECONF 'unified' Datastore](https://datatracker.ietf.org/meeting/126/materials/slides-126-core-coreconf-unified-datastore-00) * **Discussion:** * Wojciech Kozaczynski highlighted that the unified datastore specification is currently incomplete and that implementing full RESTCONF-style transactions is too resource-heavy for constrained nodes. * He proposed moving away from atomic transactions to a simpler "one-by-one" item processing sequence or "make-before-break" group boundaries encoded via CBOR maps. * Alexander Pelov supported stripping transaction complexity out of the baseline draft to allow shipping COMI sooner, leaving transaction-level atomicity to future extensions. * Esko Dijk inquired about client feedback structure during partial failures. Carsten Bormann suggested leveraging CoAP problem details and stripping out COMI-specific variants. --- ### 8. OSCORE-capable Proxies * **Presenter:** Rikard Höglund * **Draft:** [draft-ietf-core-oscore-capable-proxies](https://datatracker.ietf.org/meeting/126/materials/slides-126-core-oscore-capable-proxies) * **Slides:** [OSCORE-capable proxies slides CoRE IETF 126](https://datatracker.ietf.org/meeting/126/materials/slides-126-core-oscore-capable-proxies-slides-core-ietf-126-00) * **Discussion:** * Rikard Höglund presented version 7, which restructures the algorithm for processing incoming requests at an OSCORE-speaking proxy without altering its core logical state machine. * A new guidance section (5.1) was added to address how a proxy decides to establish an OSCORE leg with an origin server. * The authors asked whether the working group should define and register two new `SvcParamKeys` for discovering nested OSCORE support and OSCORE-to-proxy legs via DNS Service Records (SVC). --- ### 9. KEM for Group OSCORE * **Presenter:** Marco Tiloca * **Slides:** [KEM for Group OSCORE](https://datatracker.ietf.org/meeting/126/materials/slides-126-core-kem-for-group-oscore-00) * **Discussion:** * Marco Tiloca introduced a new individual proposal to secure the pairwise mode of Group OSCORE with post-quantum algorithms using Key Encapsulation Mechanisms (KEMs), such as ML-KEM. * The mechanism uses the Group Manager to coordinate the exchange of public KEM keys and ciphertexts among nodes. * Christian Amsüss pointed out that in the event of an ECDSA breach, source authentication within the group would fail, but the confidentiality of the pairwise mode would downgrade gracefully back to the group level. * François Navas volunteered to review the draft. --- ### 10. SID in DNS * **Presenter:** Laurent Toutain * **Slides:** [SID in DNS](https://datatracker.ietf.org/meeting/126/materials/slides-126-core-sid-in-dns-00) * **Discussion:** * Laurent Toutain presented a mechanism to resolve YANG SID allocations by utilizing reverse DNS structures (e.g., mapping a SID to a pointer referencing the source SID/YANG file). * Maria Matějka raised concerns regarding corporate lifetimes, indicating that if a manufacturer goes bankrupt, DNS records may disappear. She argued that sold devices must remain self-contained by carrying their own SID/YANG files. * Laurent and Alexander Pelov clarified that the draft simply standardizes a DNS interface to existing SDO/manufacturer SID registries, not the business model of registries. * Sandoche Balakrichenan (AFNIC) noted that registries like AFNIC see economic viability in hosting such identifiers. Esko Dijk questioned how private/enterprise SIDs would be handled if companies choose not to publish them. --- ### 11. CoAP Extensions for Asynchronous Task Resources * **Presenter:** Linzhou * **Slides:** [CoAP Extensions for Asynchronous Task Resources](https://datatracker.ietf.org/meeting/126/materials/slides-126-core-coap-extensions-for-asynchronous-task-resources-01) * **Discussion:** * Linzhou introduced a framework for modeling, monitoring, and controlling long-running operations (such as firmware updates) as addressable task resources. * Christian Amsüss asked at what abstraction level these tasks operate (e.g., service-level vs operating system threads). Linzhou confirmed it targets the service level. * Carsten Bormann welcomed the draft, noting that standardized asynchronous task handling has been lacking in CoAP. * Marco Tiloca suggested investigating the Access Identifier Format (AIF) in the ACE working group to handle separate access controls for the execution resource and the task-tracking resource. * Esko Dijk pointed out via chat some conceptual similarities to the topic lifecycle in CoAP PubSub. --- ### 12. CoAP Diagnostic Message Notation * **Presenter:** Carsten Bormann * **Slides:** [CoAP Diagnostic Message Notation](https://datatracker.ietf.org/meeting/126/materials/slides-126-core-coap-diagnostic-message-notation-00) * **Discussion:** * Carsten Bormann presented a new individual submission aimed at establishing a uniform diagnostic notation for documenting CoAP transactions in specifications, tool outputs, and Continuous Integration (CI) verification. * Rikard Höglund asked how multi-response flows (like observe notifications) would be represented. Carsten suggested utilizing distinct method and response codes. * Christian Amsüss supported adopting this format for command-line tool outputs (e.g., aiocoap-client). * Martine Lenders asked if the notation supports showing payloads in multiple parallel formats (e.g., raw hex and parsed CBOR). Carsten replied that CDDL toolchains are evolving to support this. --- ## Decisions and Action Items * **`draft-ietf-core-uri-path-abbrev`:** Christian Amsüss to apply parallel updates matching IANA early-review feedback. Carsten Bormann to submit the shepherding write-up post-vacation. * **`draft-ietf-core-conditional-attributes`:** Proceed to shepherding review under Marco Tiloca. * **`draft-ietf-core-oscore-key-update` (KUDOS):** Working Group Last Call (WGLC) requested. * **`draft-ietf-core-responses`:** The WG decided to place the `Respond-To` option on the backburner (moving it to the appendix as design guidance) and focus on progressing `Response-For` and `Leisure-For-Responses`. --- ## Next Steps * **KUDOS Interoperability:** Organize a virtual interop session in September to test consecutive executions and verify context-reset behavior. * **Multicast Notifications Interoperability:** Recruit additional developers to implement [draft-ietf-core-observe-multicast-notifications](https://datatracker.ietf.org/meeting/126/materials/slides-126-core-observe-multicast-notifications) before initiating a WGLC. * **CORECONF / COMI:** Revise the datastore transaction model in [draft-ietf-core-comi](https://datatracker.ietf.org/meeting/126/materials/slides-126-core-comi) to replace complex atomic rollbacks with simple, sequential item processing. * **Review Volunteers:** François Navas to review the post-quantum KEM Group OSCORE draft before IETF 127. Working group participants to review and provide feedback on the Asynchronous Task Resources and CoAP Diagnostic Message Notation drafts on the mailing list.