**Session Date/Time:** 21 Jul 2025 07:30 # nmop ## Summary This IETF NMOP session covered several drafts related to network management, telemetry, anomaly detection, and service and infrastructure maps (SIMAP). Discussions included updates to existing documents, potential future work, and operator experiences with network incidents. Key topics involved message broker telemetry, anomaly detection lifecycle and semantics, CMAP concepts and modeling, and operator requirements for network management. ## Key Discussion Points * **NMOP Message Broker Telemetry:** * The draft is adopted and updates were presented, including improved use of IETF platform manifest, clearer naming conventions, and optional metadata flags. * Feedback from Rob regarding Yang Push implementation was addressed. * Discussion on future augmentation for IPFIX data collection, with a preference for keeping it separate to declare victory. * **Network Anomaly Detection (NAD):** * Three documents related to architecture, semantics, and lifecycle were discussed. * Terminology updated from "cause" to "trigger". * Considerations for seasonality and maintenance windows were added to the semantic metadata annotations. * Discussion about distinguishing between rule-based and knowledge-based anomaly detection. * Suggestion to cover other components in the architecture framework document. * A question about effectiveness on detecting anomalies with some Swisscom statistics provided. * Question about extensibility into multi-modal data formats with response focused on relevant metrics. * **SIMAP Concept:** * Update on the NMOP SIMAP concept draft, with focus on defining terms and use cases * Agreement to close the 11 open issues that were reviewed on the mailing list * Ready for working group last call with condition. * SIMAP Modeling Draft updates: Discussion about augmenting RCF8345 to support CIMAP requirements. * **SIMAP External Relationship:** * Presentation on a draft formalizing links to external information within SIMAP. * Template-based approach for querying external data sources. * Consideration of alternative solutions and a request for feedback. * **SIMAP Applicability to Transport Networks:** * Discussion on the impact of applying SIMAP to transport networks (OTN and WDM). * Use cases: service provisioning * Discussion about the scope of "transport network" and the appropriate working group for this work (NMOP vs. CCAMP). * Ensure the requirements of the model fit within the SIMAP concept. * **RFC 3535 Update:** * Presentation on updating RFC 3535 with current operator requirements. * Discussion on the appropriate venue for this work and its relation to other initiatives (e.g., ONIONS). * Goal is to identify work NMOP can undertake as well as tasks it can farm out to other groups. * Challenge on how to handle vendors using proprietary models instead of the IETF standard models. * **Operator Incident Sharing:** * MTS shared an incident involving a leaf switch swap leading to partial blackholing due to interface order. * NBN shared an incident involving generators to manage power outages, and discussed the use of knowledge graphs to improve problem resolution. * Opportunity to potentially correlate S-Flow data with BMP data for detection. * General Discussion * Suggestion to use the four month gaps to provide review comments for documents. ## Decisions and Action Items * **Action Item (Chairs):** Rashad and Buna to consider the existing issues and confirm if the CMAP concept document is ready for working group last call after the IETF meeting. * **Action Item (Luis):** Consider feedback and potential next steps for each requirement listed in the RFC 3535 bis draft to clarify direction. * **Action Item (Mahesh):** Consider where is the final home for the RFC 3535bis draft and whether NMOP should farm out the work. * **Action Item (Pierre Francois):** Prepare zero dash one doc and present to working group. * **Action Item (Pierre Francois):** Push dash o1 of the SIMAP external relationship draft and engage on mailing list. * **Action Item:** Discuss which components of the Architecture framework doc needs to be covered in the working group. * **Action Item:** Read the ZSM liaison statement and determine if any overlap/conflict needs to be communicated. * **Decision:** Decide whether a design team needs to be created for the NMOP. ## Next Steps * Continue refining the NMOP Message Broker Telemetry draft based on feedback and implementation experience. * Seek further feedback from the working group on the NAD drafts, especially regarding Yang model changes. * Push forward on the CMAP concept and modelling drafts. * Engage with other working groups within the Ops Area to gather input on the RFC 3535 update draft. * Present the results of the NMOP at the next OPS Area meeting for feedback. * Continue operator incident sharing to inform NMOP work items. --- **Session Date/Time:** 23 Jul 2025 14:00 # nmop ## Summary This NMOP session included presentations on validating configure subscription yang push publication, the open source project Anomali lifecycle and semantics, and the CEMAP hackathon results. An update was also given from the design team on knowledge graph. Open mic time focused on coordination between working groups and locations for young models. ## Key Discussion Points * **Validating Configure Subscription Yang Push Publication:** Thomas presented on hackathon activities validating Yang Push implementations, including extending Lib-Yang for any data validation and validating the message broker telemetry message schema. Key findings included gaps in Lib-Yang related to data store mapping and Yang Structures support. * **Open Source Net Anomaly Lifecycle and Semantics:** Vincenzo presented on Antagonist, an open-source project used to validate network anomaly detection drafts. The presentation covered moving towards a production-ready version, a new UI, and validation with Swisscom data. The discussion focused on data model scalability and the refinement process. There was question and answer regarding the backend. * **CEMAP Hackathon Results:** Pierre François reported on the CEMAP hackathon, focusing on modeling VPN services steered into SRV6 routes. An implementation of the Vivex draft on modeling external relationships was used. The discussion included feedback on SRV6 models and adaptation of code. * **Knowledge Graph Update:** Michael gave an update from the knowledge graph design team, focusing on unifying the work of the three related drafts. The goal is to unlock connections between isolated data sources using semantic web ideas. The team will create an initial POC fusing early CMAP bottle and examined the Noria ontology. Discussion included scalability, the formality of the IETF process, and use cases for anomaly and incident management. * **Any Data Discussion:** The chair noted the any data implementation and questioned where there would be any testing. It was revealed that work has begun and a pull request on liby is available. ## Decisions and Action Items * **Knowledge Graph Design Team:** Those interested in joining the knowledge graph design team should contact the chair. * **SRV6 Topology Coordination:** The SRV6 ops chair, Olga, and Pierre Francois will collaborate on coordinating SRV6 topology modeling and alignment between CMAP and SRV6 ops. ## Next Steps * **Yang Push Validation:** Address identified gaps in Lib-Yang related to data store mapping and Yang Structures support. * **Anomali Lifecycle and Semantics:** Finish backend implementation for the GUI and complete testing. * **CEMAP:** Seek feedback on SRV6 models and adapt the code accordingly. Publish SRV6 models * **Knowledge Graph:** Create initial POC using early CMAP model and examine the Noria ontology.