**Session Date/Time:** 23 Jul 2026 07:00 # [NETMOD](../wg/netmod.html) ## Summary The NETMOD Working Group met at IETF 120. The session began with working group status updates, followed by progress reports on active WG documents including modules addressing IANA guidelines, YANG packages, schema comparison, and the roadmap for YANG Next / YANG 2.0. The group also discussed several individual proposals covering YANG configuration templates, subtree replacements, normalized forms for string-derived types, library feature comparisons, and YANG schema test coverage tools. Multiple polls were conducted during the session to gauge consensus on adopting new work. --- ## Key Discussion Points ### Working Group Status & Admin * **Slides**: [01) Session Intro & WG Status](https://datatracker.ietf.org/meeting/126/materials/slides-126-netmod-01-session-intro-wg-status-01) * Chairs Lou Berger and Kent Watsen opened the session. James Cumming acted as session secretary. * **Document Status**: * RFC 9507 (BCP 216) has been published. * System config draft (`draft-ietf-netmod-sysconfig`) is in the RFC Editor Queue. * `draft-ietf-netmod-yang-versioning-reqs` is on hold and will not be published individually; it will eventually be processed when the broader versioning work concludes. * Ahmed Elhassany provided an update on the expired Anydata validation draft, stating they plan to submit an updated document before the next IETF meeting. --- ### Guidance for Managing YANG Modules in RFCs and IANA Registries * **Slides**: [03) Guidance for Managing YANG Modules in RFCs and IANA](https://datatracker.ietf.org/meeting/126/materials/slides-126-netmod-02-iana-yang-guidance-00) * **Presenter**: Rob Wilton * **Active Draft**: `draft-ietf-netmod-iana-yang-guidance` * Rob Wilton highlighted that the document is designed to guide both IANA and the RFC Editor, and has been updated to incorporate feedback from both bodies. * Mahesh Jethanandani queried if the issue regarding identities versus enumerations raised by Amanda (IANA) was resolved. Rob Wilton clarified that recent communication indicated the RFC Editor and IANA are comfortable with the current text, requiring only minor clarifications. * **Next Steps**: A minor revision addressing Chin's and Amanda's feedback will be submitted, followed immediately by a Working Group Last Call (WGLC). --- ### YANG Packages Update * **Slides**: [02) YANG Packages Update](https://datatracker.ietf.org/meeting/126/materials/slides-126-netmod-yang-packages-update-00) * **Presenter**: Rob Wilton * **Active Draft**: `draft-ietf-netmod-yang-packages` * Rob Wilton summarized changes in the `-09` release, which has grown to 110 pages due to the inclusion of comprehensive examples. * **Key updates**: * Added a `version-description` field. * The `includes-package` list is now keyed by both `name` and `version` to manage metadata locations appropriately. * Added `depends-on` for bug fix packages (following suggestions from Joe Clarke). * Added features enablement under mount points. * Joe Clarke and Per Andersson discussed the utility of transaction IDs and absolute vs. relative paths in chat. Rob Wilton confirmed his intent to open-source a Cisco-implemented package compiler to aid adoption. * **Next Steps**: Address the remaining 6 minor issues and resolve the dependency on the `ypath` work. The draft is targeted for WGLC by IETF 127. --- ### YANG Schema Comparison * **Slides**: [04) YANG Schema Comparison](https://datatracker.ietf.org/meeting/126/materials/slides-126-netmod-draft-ietf-netmod-yang-schema-comparison-09-00) * **Presenter**: Jan Kundrát * **Active Draft**: `draft-ietf-netmod-yang-schema-comparison` * Jan Kundrát reported that the document now supports editorial change definitions and has been split into two modules, with only the smaller one containing extensions being standardized. * **Tooling**: Implementations are available in Libyang (`yanglint`) and Bijang. Ladislav Lhotka confirmed that the `yanglint` output can be evaluated directly. * **Next Steps**: Rob Wilton will perform an author-level review. Once any final issues are incorporated into a new version, the document will be ready for WGLC. --- ### YANG Next Agreement & YANG 2.0 * **Slides**: * [05) YANG Next Agreement](https://datatracker.ietf.org/meeting/126/materials/slides-126-netmod-yang-next-agreement-00) (Presenter: Rob Wilton) * [06) The YANG 2.0 Data Modeling Language](https://datatracker.ietf.org/meeting/126/materials/slides-126-netmod-the-yang-20-data-modeling-language-00) (Presenter: Kent Watsen) * **Active Drafts**: `draft-ietf-netmod-yang-next-agreement` and `draft-ietf-netmod-yang2` * Rob Wilton detailed the 150+ open issues on the YANG Next tracker and proposed a scoping approach where changes are classified into clarifications, minor/major enhancements, or deprecated elements. * Ahmed Elhassany emphasized the need to clearly define the baseline purpose of YANG 2.0 (e.g., acknowledging that its usage has evolved beyond simple device configuration). * Reshad Rahman expressed concern about naming the language "YANG 2.0" if it implies too many non-backwards-compatible changes, though Joe Clarke pointed out that the transition from YANG 1.0 to 1.1 was also technically non-backwards-compatible. * Maria Matějka asked if pressing issues could be published as separate standalone drafts to avoid delaying them for the 5-year YANG 2.0 cycle. Lou Berger confirmed that the WG is contribution-driven and can progress standalone extensions if there is interest. * Kent Watsen reported that the baseline YANG 2.0 draft (`draft-ietf-netmod-yang2`) is adopted, clean of XML/Netconf-specific sections, and open for Pull Requests (PRs). * **Tooling / Process**: Bi-weekly virtual interim meetings will be scheduled to review GitHub PRs and track consensus. Rob Wilton proposed experimenting with an LLM-assisted interactive web page to gather member opinions on specific GitHub issues. --- ### XML Encoding of Data Modeled with YANG * **Slides**: [07) XML Encoding of Data Modeled with YANG](https://datatracker.ietf.org/meeting/126/materials/slides-126-netmod-xml-encoding-of-data-modeled-with-yang-00) * **Presenter**: Kent Watsen * **Active Draft**: `draft-ietf-netmod-yang-xml` * Kent Watsen noted the draft has been adopted. Future updates will focus on alignment with the `draft-ietf-netmod-yang2` draft, including the removal of the `anyxml` node. * Maria Matějka and Alex Huang Feng raised concerns about keeping XML, JSON (RFC 7951), and CBOR encodings synchronized to avoid discrepancies. Rob Wilton noted that while minor encoding additions (such as floats or doubles) might be needed, the underlying encoding rules do not require major rewrites. --- ### YANG Configuration Templates * **Slides**: [08) YANG Configuration Templates](https://datatracker.ietf.org/meeting/126/materials/slides-126-netmod-yang-configuration-templates-00) * **Presenter**: Kent Watsen * Kent Watsen reviewed three main technical issues: 1. *Validation at definition time*: Proposed a new `schema-path` leaf (using YANG XPath/ypath) to specify where a template is rooted, allowing for structural validation. 2. *Programmatic elements (loops/conditions)*: Kent noted that these were previously objected to. Joe Clarke expressed surprise at the objection, noting that loops/conditions are highly valuable in real-world deployments. Lou Berger confirmed the design remains open for discussion. 3. *Non-NMDA Server support*: Proposed an `expand-template` flag for `` and `` RPCs. Reshad Rahman and James Cumming raised concerns about clients reading back different configurations from what was written if expansion is forced on retrieval. * Maria Matějka indicated in the chat that BIRD intends to align its config templates with this YANG template work. --- ### Enhancements to the YANG Language for Capturing Subtree Replacements * **Slides**: [09) Enhancements to the YANG Language for Capturing Subtree Replacements](https://datatracker.ietf.org/meeting/126/materials/slides-126-netmod-yang-enhancements-to-capture-subtree-replacements-03) * **Presenter**: Balaji (representing Rajesh Balaji) * Balaji presented a proposal for a `deprecation-info` extension statement in YANG 1.1 to machine-readably point deprecated nodes to their replacements. * Ladislav Lhotka suggested this work should be incorporated as a substatement of `status` within the YANG Next/YANG 2.0 effort. * Kent Watsen (speaking as a contributor) objected to adopting the document in its current state, stating it should contain structured data (like reason, date, or translation paths) rather than a simple unstructured string. Qiufang Sheng supported adopting the document immediately to solve a real, present problem rather than waiting for YANG 2.0. --- ### Extending Normalized Forms to String-Derived Types * **Slides**: [10) Extending Normalized Forms to String-Derived Types](https://datatracker.ietf.org/meeting/126/materials/slides-126-netmod-10-extending-normalized-forms-to-string-derived-types-00) * **Presenter**: Don Fedyk * Don Fedyk presented a method for normalizing divergent string-derived types (such as MAC addresses using colons versus dashes, or different casings) to avoid indexing and comparison errors. * Michael Richardson noted that serialization formats like CBOR benefit from knowing a string is a MAC address to encode it as raw bytes. * Ladislav Lhotka questioned how the normalized value interacts with `` replies. Joe Clarke suggested this behaves similarly to "display hints" in SNMP MIBs and should be generalized in YANG Next. --- ### YANG Libraries Feature Comparison * **Slides**: [11) YANG Libraries Feature Comparison](https://datatracker.ietf.org/meeting/126/materials/slides-126-netmod-yang-libraries-feature-comparison-00) * **Presenter**: Vivek * Vivek shared an analysis comparing the feature completeness of three prominent YANG parsers: Libyang, YangKit, and YANG Tools. * The analysis evaluated support for schema loading, CBOR, Anydata validation, XPath, structure extensions, and schema comparison. * Mahesh Jethanandani, Per Andersson, and Robert Wilton highlighted the value of correct, tool-validated examples in RFCs to prevent parser issues. --- ### YANG Coverage * **Slides**: [12) YANG Coverage](https://datatracker.ietf.org/meeting/126/materials/slides-126-netmod-12-yang-coverage-00) * **Presenter**: Ignas Bagdonas * Ignas Bagdonas presented a Proof of Concept (PoC) tool designed to calculate the testing/example coverage of a YANG schema based on XML/JSON instance documents. * Mahesh Jethanandani supported the utility of the tool and suggested integrating it into GitHub CI/CD workflows. * Lou Berger questioned whether updates to RFC authoring guidelines (e.g., establishing strict `` tags for examples) should be separated from the coverage tool draft itself to expedite adoption. --- ## Decisions and Action Items ### Session Polls * **Poll 1**: Are there any objections to the WG working on templates? * **Result**: yes: 0, no: 29, no_opinion: 0 (total: 51) * **Poll 2**: Is this document a suitable starting point for the WG to base its work? (*Regarding the YANG Configuration Templates draft*) * **Result**: yes: 22, no: 0, no_opinion: 10 (total: 51) * **Poll 3**: Is there interest in working on this problem in the WG? (*Regarding draft-ietf-netmod-yang-enhancements-to-capture-subtree-replacements*) * **Result**: yes: 20, no: 0, no_opinion: 6 (total: 50) * **Poll 4**: Is this document a suitable starting point for the WG to base its work? (*Regarding draft-ietf-netmod-yang-enhancements-to-capture-subtree-replacements*) * **Result**: yes: 14, no: 1, no_opinion: 10 (total: 50) ### Key Actions * **`draft-ietf-netmod-iana-yang-guidance`**: Rob Wilton to apply minor updates from Chin and Amanda (IANA). Once submitted, chairs will initiate a WGLC. * **`draft-ietf-netmod-yang-schema-comparison`**: Jan Kundrát and Rob Wilton to align on a final pre-WGLC review. WGLC expected on the next revision. * **YANG 2.0 / Next GitHub Repo**: Bi-weekly virtual interims will be scheduled to process incoming Pull Requests and review active tracker issues. --- ## Next Steps * Initiate formal adoption call on the mailing list for the **YANG Configuration Templates** draft. * Initiate formal adoption call on the mailing list for the **Subtree Replacements / Deprecation Info** draft (addressing contributor comments regarding unstructured strings vs. structured attributes). * Continue discussion on the mailing list regarding the **Normalized Forms for String-Derived Types** draft to resolve open questions on canonical vs. normalized representations and CBOR impacts.