RFC 9375

A YANG Data Model for Network and VPN Service Performance Monitoring

📅 published April 2023, from ietf

RFC 9375 — A YANG Data Model for Network and VPN Service Performance Monitoring 🦋✨

okay bestie, come sit with me, because this one is a whole vibe 💅🎀 RFC 9375 is a Standards Track document from the IETF, and it was published in April 2023 ✨. The authors are B. Wu, Ed. and Q. Wu, Ed. (both from Huawei) 🌸, M. Boucadair, Ed. (Orange) 🧚, O. Gonzalez de Dios (Telefonica) 💜, and B. Wen (Comcast) 🍓. Its ISSN is 2070-1721 🔥.

The title block doesn't say it obsoletes or updates anything, so this is a brand-new standard, no drama 🙈 ✨. Its whole job is to give you a YANG data model for performance monitoring of both the underlay networks and the overlay VPN services that ride on top of them 🦋💖.

Abstract ✨

okay so the abstract is short and sweet 🥰. The data model for network topologies in RFC 8345 has vertical layers between networks, and that layering can be extended to cover network and service topologies 🌈. This document defines a YANG module for performance monitoring (PM) of both underlay networks and overlay VPN services, so you can monitor and manage performance across the topology of both layers 💅🎀.

Status of This Memo ✨

This is an Internet Standards Track document 👑.

It's a product of the Internet Engineering Task Force (IETF) 🫶. It represents the consensus of the IETF community, got public review, and was approved for publication by the Internet Engineering Steering Group (IESG) 🔥. If you want more info on Internet Standards, check Section 2 of RFC 7841 🧚.

For the current status, any errata, and how to give feedback, go to https://www.rfc-editor.org/info/rfc9375 💜.

Copyright (c) 2023 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) as they were on the publication date 📜. Please actually read those, bestie, because they explain your rights and restrictions 🙈. Code Components pulled out of this document have to include the Revised BSD License text described in Section 4.e of the Trust Legal Provisions, and they come without warranty as the Revised BSD License describes 💖.

Table of Contents 🎀

1.  Introduction
2.  Terminology
  2.1.  Acronyms
3.  Network and VPN Service Performance Monitoring Model Usage
  3.1.  Collecting Data via the Pub/Sub Mechanism
  3.2.  Collecting Data On Demand
4.  Description of the YANG Data Model
  4.1.  Layering Relationship between Multiple Layers of Topology
  4.2.  Network-Level Performance Monitoring Augmentation
  4.3.  Node-Level Performance Monitoring Augmentation
  4.4.  Performance Monitoring Augmentation at Link and Termination
        Point Level
5.  Network and VPN Service Performance Monitoring YANG Module
6.  Security Considerations
7.  IANA Considerations
8.  References
  8.1.  Normative References
  8.2.  Informative References
Appendix A.  Illustrative Examples
  A.1.  Example of VPN Performance Subscription
  A.2.  Example of VPN Performance Snapshot
  A.3.  Example of Percentile Monitoring
Acknowledgements
Contributors
Authors' Addresses

1. Introduction 🦋

okay so [RFC8969] is the framework for automating service and network management with YANG [RFC7950] data models 💅. It says the performance measurement telemetry model should be tied to services (like a Layer 3 VPN or Layer 2 VPN) or to network models, so you can monitor overall network performance and the Service Level Agreements (SLAs) 🔥.

Here's the thing 🙈: VPN performance is tied to changes in the underlay networks that carry the VPN. Like, link delay between Provider Edge (PE) and Provider (P) devices, plus packet loss on Layer 2 and Layer 3 interfaces connecting PEs and Customer Edge (CE) devices, directly hits VPN performance 😭. Putting Layer 2 / Layer 3 VPN performance together with network performance data also lets the orchestrator monitor consistently 🌈. So this document defines a YANG module for both network and VPN service performance monitoring (PM) 🫶. You can use it to monitor and manage network performance at the topology level, or on the service topology between VPN sites 💖.

The base model in Section 5 can be extended with technology-specific details, e.g., adding Explicit Congestion Notification (ECN) statistics for Layer 3 networks or VPN services to support performance-sensitive applications ✨🎀.

This document does not introduce new metrics or new mechanisms for measuring network performance. It just uses the existing mechanisms and statistics to monitor network and service performance 💅.

The YANG module here is an augmentation to the network topology data model in [RFC8345], and it draws on YANG types from [RFC6991], [RFC8345], [RFC8532], and [RFC9181] 🧚.

Appendix A has examples showing how to use the module 🦋💜.

2. Terminology ✨

These terms are defined in [RFC7950] and used in this spec 🌸:

  • augment
  • data model
  • data node

The YANG terminology is found in [RFC7950] 💖. The tree diagrams in this document follow the notation in [RFC8340] 🎀.

2.1. Acronyms 💅

These acronyms show up in the document 🫶:

CE        Customer Edge, as defined in [RFC4026]

L2VPN     Layer 2 Virtual Private Network, as defined in [RFC4026]

L3VPN     Layer 3 Virtual Private Network, as defined in [RFC4026]

L2NM      L2VPN Network Model

L3NM      L3VPN Network Model

MPLS      Multiprotocol Label Switching

OAM       Operations, Administration, and Maintenance

OSPF      Open Shortest Path First

OWAMP     One-Way Active Measurement Protocol, as defined in
          [RFC4656]

P         Provider router, as defined in [RFC4026]

PE        Provider Edge, as defined in [RFC4026]

PM        Performance Monitoring

SLA       Service Level Agreement

TP        Termination Point, as defined in [RFC8345], Section 4.2

TWAMP     Two-Way Active Measurement Protocol, as defined in
          [RFC5357]

VPLS      Virtual Private LAN Service, as defined in [RFC4026]

VPN       Virtual Private Network

3. Network and VPN Service Performance Monitoring Model Usage 🧚

Models are key for automating network management operations (Section 3 of [RFC8969]) 💅. Together with service and network models, performance measurement telemetry models are needed to monitor network performance against specific service requirements, which are usually captured in an SLA 🔥.

Here's the architecture picture, bestie 🦋:

                          +---------------+
                          |   Customer    |
                          +-------+-------+
                                  |
          Customer Service Models |
                                  |
                          +-------+---------+
                          |    Service      |
                          |  Orchestrator   |
                          +------+-+--------+
                                 | |
          Network Service Models | | Network and VPN Service PM Models
                                 | |
                          +------+-+--------+
                          |     Network     |
                          |   Controller    |
                          +-------+---------+
                                  |
          +-----------------------+------------------------+
                                Network

     Figure 1: An Example Architecture with a Service Orchestrator

The network and VPN service PM model can expose operational performance info to the layer above, like an orchestrator or a Business Support System (BSS) / Operational Support System (OSS) client app, through standard network management APIs 💜. Figure 1 shows an example in a layered model architecture described in [RFC8309] ✨.

Before using the model, the controller has to establish topology visibility of the network and VPN 🌸. For example, the controller can use network info from [RFC8345] and [YANG-SAP], or VPN info from the L3VPN Network Model (L3NM) [RFC9182] and the L2VPN Network Model (L2NM) [RFC9291] 🫶. Then the controller derives network or VPN performance data by aggregating (and filtering) lower-level data collected through monitoring counters on the devices involved 💖.

The network or VPN performance data can come from different sources 🌈. For example, per-link performance data in the underlay can be collected with a network performance measurement method such as One-Way Active Measurement Protocol (OWAMP) [RFC4656], Two-Way Active Measurement Protocol (TWAMP) [RFC5357], Simple Two-way Active Measurement Protocol (STAMP) [RFC8762], Multiprotocol Label Switching (MPLS) Loss and Delay Measurement [RFC6374], or In situ OAM (IOAM) [RFC9197] 🦋. Performance monitoring info that reflects the quality of the network or VPN service (like network performance between a source node and a destination node, or between VPN sites) can be computed and aggregated, for example, using info from the Traffic Engineering Database (TED) [RFC7471] [RFC8570] [RFC8571] or the Large-Scale Measurement Platform (LMAP) [RFC8194] 🎀.

The measurement and report intervals tied to this performance data usually depend on the config of the specific measurement or collection method, or combos of them 💅. This document defines network-wide measurement intervals so that measurement requirements for networks or VPN services line up 👑.

3.1. Collecting Data via the Pub/Sub Mechanism 🔥

Some apps, like service-assurance apps that have to keep a continuous view of operational data and state, can use the subscription model in [RFC8641] to subscribe, at the data source, to the specific network or VPN service performance data they care about 💖. For example, network or VPN topology updates can be obtained through on-change notifications [RFC8641] ✨. For dynamic PM data (like VPN Routing and Forwarding (VRF) routes or Media Access Control (MAC) entries, link metrics, and interface metrics), various notifications can be specified to get more complete data 🌈. A periodic notification [RFC8641] can be specified to get real-time performance data 🫶. For devices or controllers that keep historical performance data for a period of time, a replay notification (see [RFC5277] or [RFC8639]) can be used to get that historical data 🧚. And alarm notifications [RFC8632] can be specified to get alarms for metrics that go above or below the performance threshold 🚨💅.

The data source can then use the network and VPN service performance monitoring model defined in this document, plus the YANG-Push data model [RFC8641], to distribute specific telemetry data to target recipients 🦋✨.

To be continued in part 2 🎀💜 (don't go anywhere, bestie 🫶)

3.2. Collecting Data On Demand 🫶

okay so sometimes a service-assurance app just wants a quick snapshot of perf data from a network topology or a VPN service topology, no waiting around for updates 💅 it can grab that info through a Network Configuration Protocol (NETCONF) [RFC6241] or RESTCONF [RFC8040] interface, using the network and VPN service PM model. for example, you can use a specified link-id of a VPN as a filter in a RESTCONF GET request to pull the per-link VPN PM data 🦋 main character energy for sure ✨

4. Description of the YANG Data Model 🎀

okay so this document defines the ietf-network-vpn-pm YANG module 🧚 it's an augmentation to the ietf-network and ietf-network-topology YANG modules, so it adds new stuff on top of what those already do 💜

4.1. Layering Relationship between Multiple Layers of Topology 🌈

so [RFC8345] is the one that defines a YANG data model for network and service topologies and inventories 💖 the service topology in [RFC8345] is basically an abstract topology sitting above Layer 1 (L1), Layer 2 (L2), and Layer 3 (L3) underlay topologies. it has the generic topology elements of node, link, and termination point, and it's giving layers on layers 🍰

one typical example, described in Figure 3 of [RFC8345], is two VPN service topologies running over a common L3 topology. each VPN service topology gets mapped onto a subset of the nodes from that L3 topology, which is so cute 🥹

here's how the topology hierarchy maps the VPN service topology onto an underlying Layer 3 network topology:

                    VPN 1                       VPN 2
         +------------------------+   +------------------------+
        /                        /   /                        /
       / S1C_[VN3]..........    /   /                        /
      /         \          :   /   / S2A_[VN1]____[VN3]_S2B /
     /           \         :  /   /        *        *      /
    /             \        :............ * ....     *     /
   / S1B_[VN2]____[VN1]_S1A /   /       *     :     *    /
  +---------:-------:------+   +-------*------:-----*---+
            :        :      * * *  * *        :     *
            :         :   *                   :     *
  Site-1A   :  +-------:-*--------------------:-----*-----+ Site-1C
    [CE1]___:_/_______[N1]___________________[N2]___*____/__[CE3]
            :/       / / \             _____//      *   /
  [CE5]_____:_______/ /    \     _____/     /     *    /
Site-2A    /:        /       \  /          /    *     /
          / :                [N5]         /   *      /
         /   :     /       __/ \__       /  *       /
        /     :   /    ___/       \__   / *        /
Site-1B /       : / ___/              \ /*         /  Site-2B
[CE2]__/________[N4]__________________[N3]________/____[CE4]
     /                                          /
    +------------------------------------------+
                                  L3 Topology

  Legend:
     N:   Node
     VN:  VPN Node
     S:   Site
     CE:  Customer Edge
     __   Link within a network layer
     :    Mapping between VPN 1 service topology and L3 topology
     *    Mapping between VPN 2 service topology and L3 topology

     Figure 2: Example of Topology Mapping between VPN Service
                 Topology and an Underlying Network

so as Figure 2 shows, two VPN service topologies are built on top of one underlying Layer 3 network 🧩

VPN 1: this service topology supports Hub-and-Spoke communications for "customer 1" 👑 it connects the customer's access at three sites: Site-1A, Site-1B, and Site-1C. those sites hook up to nodes that get mapped to node 1 (N1), node 2 (N2), and node 4 (N4) in the underlying Layer 3 network. Site-1A is the Hub, and Site-1B and Site-1C are set up as Spokes 🍓

VPN 2: this service topology supports any-to-any communications for "customer 2" 🌸 it connects the customer's access at two sites: Site-2A and Site-2B. those sites connect to nodes mapped to node 1 (N1) and node 3 (N3) in the underlying Layer 3 network. Site-2A and Site-2B both have an "any-to-any" role ✨

so based on how VPN service topologies line up with underlying network topologies, the Network and VPN Service PM YANG module extends the performance status of both the underlay networks and the VPN services 💅 for example, it can give you link PM statistics and port statistics for an underlay network, like Layer 1, Layer 2, Layer 3, and OSPF networks. it can also give you VPN PM statistics, which can be split further into PM for the VPN tunnel and PM at the VPN PE access node. here's the diagram 👇

      +-----------------------------------------------------+
      |                                                     |
      |                      VPN2 Link                      |
      |              |<-------------------->|               |
      |              |                      |               |
      |      VPN2+---+---+              +---+---+VPN2       |
      |       TP1| VN1   |  Tunnel PM   |  VN3  |TP2        |
      |       ---+ PE A  |==============|  PE B +----       |
      |vpn-access+-------+              +-------+ vpn-access|
      |-interface|                              | -interface|
      |          |##############################|           |
      |          |inter-vpn-access-interface PM |           |
      |                                                     |
      +-----------------------------------------------------+
      |                                                     |
      |                                                     |
+----+ |        TP+-----+ Link  +---+ Link  +-----+TP        | +----+
| CE4+-+----------+ N1  +-------+-N2+-------+  N3 +----------+-+CE5 |
+----+ |       1-1+-----+1-2 2-1+---+2-2 3-1+-----+3-2       | +----+
      |                                                     |
      |                                                     |
      +-----------------------------------------------------+

      Legend:
        N:  node
        VN: VPN Node
        TP: Termination Point
        -:  Link

                   Figure 3: An Example of VPN PM

so Figure 3 shows an example of VPN PM and two VPN PM measurement methods: the VPN tunnel PM and the inter-VPN-access interface PM 🦋 VPN PM can also cover statistics on VPN access interfaces, the number of current VRF routes, or the L2VPN MAC entries of a VPN node. bestie, that's a lot of data 😭🫶

4.2. Network-Level Performance Monitoring Augmentation 💜

so this module can be used for performance monitoring of both the underlay networks and the VPN services. each of those shows up as its own entry in the network list [RFC8345] 🌈 the difference comes down to this:

  • when the "service" presence container is absent, it means you're monitoring the network itself 🍰
  • when the "service" presence container is present, it means you're monitoring the VPN service specified by the "service-type" leaf, like L3VPN or Virtual Private LAN Service (VPLS). those values come from [RFC9181]. when a network topology instance has L3VPN or other L2VPN network types, it represents a VPN instance that can do performance monitoring 💖

the YANG tree in Figure 4 is part of the ietf-network-vpn-pm tree, and it defines this set of network-level attributes 🧚

"vpn-id": this refers to an identifier of a VPN service defined in [RFC9181]. it's how the performance status gets correlated with the network service configuration 🔗

"vpn-service-topology": this indicates the type of VPN service topology. the model supports "any-to-any", "hub-spoke" (where Hubs can exchange traffic), and "hub-spoke-disjoint" (where Hubs cannot exchange traffic). those are taken from [RFC9181] too. these topology types describe how VPN sites talk to each other 💅

module: ietf-network-vpn-pm
  augment /nw:networks/nw:network/nw:network-types:
    +--rw service!
       +--rw service-type            identityref
       +--rw vpn-id?                 vpn-common:vpn-id
       +--rw vpn-service-topology?   identityref

                  Figure 4: Network-Level YANG Tree

4.3. Node-Level Performance Monitoring Augmentation 🎀

the YANG tree in Figure 5 is the node part of the ietf-network-vpn- pm tree 🌸

for network performance monitoring, the module defines these attributes 💖

"node-type": this indicates the device type of the PE, P device, or Autonomous System Border Router (ASBR), as defined in [RFC4026] and [RFC4364]. that way, the performance metric between any two nodes that each have a specific node type can be reported 🛸

"entry-summary": this lists a set of IPv4 statistics, IPv6 statistics, and MAC statistics. the detailed statistics are specified separately 📊

for VPN service topology, the module defines one attribute 👑

"role": this defines the role in a particular VPN service topology. the roles are taken from [RFC9181] (e.g., "any-to-any-role", "spoke-role", and "hub-role") 🫶

  augment /nw:networks/nw:network/nw:node:
    +--rw node-type?       identityref
    +--ro entry-summary
       +--ro ipv4-num
       |  +--ro maximum-routes?        uint32
       |  +--ro total-active-routes?   uint32
       +--ro ipv6-num
       |  +--ro maximum-routes?        uint32
       |  +--ro total-active-routes?   uint32
       +--ro mac-num
          +--ro maximum-mac-entries?        uint32
          +--ro total-active-mac-entries?   uint32
  augment /nw:networks/nw:network/nw:node:
    +--rw role?   identityref

                    Figure 5: Node-Level YANG Tree

okay bestie, this part is all about the link and termination point (TP) level 🦋 The YANG tree in Figure 6 is the link and TP part of the "ietf-network-vpn-pm" tree.

The "links" come in two flavors 💅: topology links (defined in [RFC8345]) and abstract links of a VPN between PEs (defined in this module).

The performance data of a link is a collection of counters and gauges that report how it's doing 📊 All of these metrics are defined as unidirectional metrics, so each direction gets its own numbers.

augment /nw:networks/nw:network/nt:link:
  +--rw perf-mon
     +--rw low-percentile?            percentile
     +--rw intermediate-percentile?   percentile
     +--rw high-percentile?           percentile
     +--rw measurement-interval?      uint32
     +--ro pm* [pm-type]
     |  +--ro pm-type          identityref
     |  +--ro pm-attributes
     |     +--ro start-time?                     yang:date-and-time
     |     +--ro end-time?                       yang:date-and-time
     |     +--ro pm-source?                      identityref
     |     +--ro one-way-pm-statistics
     |     |  +--ro loss-statistics
     |     |  |  +--ro packet-loss-count?   yang:counter64
     |     |  |  +--ro loss-ratio?          percentage
     |     |  +--ro delay-statistics
     |     |  |  +--ro unit-value?                     identityref
     |     |  |  +--ro min-delay-value?                yang:gauge64
     |     |  |  +--ro max-delay-value?                yang:gauge64
     |     |  |  +--ro low-delay-percentile?           yang:gauge64
     |     |  |  +--ro intermediate-delay-percentile?  yang:gauge64
     |     |  |  +--ro high-delay-percentile?          yang:gauge64
     |     |  +--ro jitter-statistics
     |     |     +--ro unit-value?                     identityref
     |     |     +--ro min-jitter-value?               yang:gauge64
     |     |     +--ro max-jitter-value?               yang:gauge64
     |     |     +--ro low-jitter-percentile?          yang:gauge64
     |     |     +--ro intermediate-jitter-percentile? yang:gauge64
     |     |     +--ro high-jitter-percentile?         yang:gauge64
     |     +--ro one-way-pm-statistics-per-class* [class-id]
     |        +--ro class-id             string
     |        +--ro loss-statistics
     |        |  +--ro packet-loss-count?   yang:counter64
     |        |  +--ro loss-ratio?          percentage
     |        +--ro delay-statistics
     |        |  +--ro unit-value?                     identityref
     |        |  +--ro min-delay-value?                yang:gauge64
     |        |  +--ro max-delay-value?                yang:gauge64
     |        |  +--ro low-delay-percentile?           yang:gauge64
     |        |  +--ro intermediate-delay-percentile?  yang:gauge64
     |        |  +--ro high-delay-percentile?          yang:gauge64
     |        +--ro jitter-statistics
     |           +--ro unit-value?                     identityref
     |           +--ro min-jitter-value?               yang:gauge64
     |           +--ro max-jitter-value?               yang:gauge64
     |           +--ro low-jitter-percentile?          yang:gauge64
     |           +--ro intermediate-jitter-percentile? yang:gauge64
     |           +--ro high-jitter-percentile?         yang:gauge64
     +--rw vpn-pm-type
        +--rw inter-vpn-access-interface
        |  +--rw inter-vpn-access-interface?   empty
        +--rw vpn-tunnel!
           +--ro vpn-tunnel-type?   identityref
augment /nw:networks/nw:network/nw:node/nt:termination-point:
  +--ro pm-statistics
     +--ro last-updated?               yang:date-and-time
     +--ro inbound-octets?             yang:counter64
     +--ro inbound-unicast?            yang:counter64
     +--ro inbound-broadcast?          yang:counter64
     +--ro inbound-multicast?          yang:counter64
     +--ro inbound-discards?           yang:counter64
     +--ro inbound-errors?             yang:counter64
     +--ro inbound-unknown-protocol?   yang:counter64
     +--ro outbound-octets?            yang:counter64
     +--ro outbound-unicast?           yang:counter64
     +--ro outbound-broadcast?         yang:counter64
     +--ro outbound-multicast?         yang:counter64
     +--ro outbound-discards?          yang:counter64
     +--ro outbound-errors?            yang:counter64
     +--ro vpn-network-access* [network-access-id]
        +--ro network-access-id           vpn-common:vpn-id
        +--ro last-updated?               yang:date-and-time
        +--ro inbound-octets?             yang:counter64
        +--ro inbound-unicast?            yang:counter64
        +--ro inbound-broadcast?          yang:counter64
        +--ro inbound-multicast?          yang:counter64
        +--ro inbound-discards?           yang:counter64
        +--ro inbound-errors?             yang:counter64
        +--ro inbound-unknown-protocol?   yang:counter64
        +--ro outbound-octets?            yang:counter64
        +--ro outbound-unicast?           yang:counter64
        +--ro outbound-broadcast?         yang:counter64
        +--ro outbound-multicast?         yang:counter64
        +--ro outbound-discards?          yang:counter64
        +--ro outbound-errors?            yang:counter64

Figure 6: Link and Termination Point YANG Subtree

For the data nodes of "link" in Figure 6, the YANG module defines this minimal set of link-level performance attributes 🎀

  • Percentile parameters: The module supports reporting delay and jitter metrics with percentile values. There are three percentile values for configuring the different reporting levels 💖 By default, the low percentile (10th percentile), intermediate percentile (50th percentile), and high percentile (90th percentile) are used. Configuring a percentile to 0.000 means the client isn't interested in receiving that particular percentile. If all the percentile nodes are configured to 0.000, then no percentile-related nodes will be reported for a given performance metric (e.g., one-way delay and one-way delay variation), and only peak/min values will be reported. For example, a client can tell the server it only wants high percentiles 🌸 Then, for a given link at a given "start-time", "end-time", and "measurement-interval", the "high-delay-percentile" and "high-jitter-percentile" will be reported. There's an example of percentiles in action in Appendix A.3 🧚
  • Measurement interval ("measurement-interval"): Specifies the performance measurement interval, in seconds ⏱️
  • Start time ("start-time"): Indicates the start time of the performance measurement for link statistics 🌅
  • End time ("end-time"): Indicates the end time of the performance measurement for link statistics 🌇
  • PM source ("pm-source"): Indicates the performance monitoring source. The data for a topology link can be based on, e.g., BGP - Link State (BGP-LS) [RFC8571]. The statistics of VPN abstract links can be collected using VPN OAM mechanisms, e.g., the OAM mechanisms referenced in [RFC9182], or Ethernet service OAM [ITU-T-Y-1731] referenced in [RFC9291]. Alternatively, the data can be based on the underlay technology OAM mechanisms, e.g., Generic Routing Encapsulation (GRE) tunnel OAM 🔍
  • Loss statistics: A set of one-way loss statistics attributes used to measure end-to-end loss between VPN sites or between any two network nodes. The exact loss value or the loss percentage can be reported 📉
  • Delay statistics: A set of one-way delay statistics attributes used to measure end-to-end latency between VPN sites or between any two network nodes. The peak/min values or percentile values can be reported ⚡
  • Jitter statistics: A set of one-way IP Packet Delay Variation [RFC3393] statistics attributes used to measure end-to-end jitter between VPN sites or between any two network nodes. The peak/min values or percentile values can be reported 🫨
  • PM statistics per class: "one-way-pm-statistics-per-class" lists performance measurement statistics for the topology link, or for the abstract link between VPN PEs, with given "class-id" names. This list is defined separately from "one-way-pm-statistics", which is used to collect generic metrics for unspecified "class-id" names 🏷️
  • VPN PM type ("vpn-pm-type"): Indicates the VPN performance type, which can be "inter-vpn-access-interface" PM or "vpn-tunnel" PM. These two are common VPN measurement methods. "inter-VPN-access-interface" PM is used to monitor the performance of logical point-to-point VPN connections between source and destination VPN access interfaces, while "vpn-tunnel" PM is used to monitor the performance of VPN tunnels. The "inter-VPN-access-interface" PM includes PE-PE monitoring, so usually only one of the two methods is used. "inter-VPN-access-interface" PM is defined as an empty leaf, which is not bound to a specific VPN access interface. The source or destination VPN access interface of the measurement can be augmented as needed 🦋
  • VPN tunnel type ("vpn-tunnel-type"): Indicates the abstract link protocol type of a VPN, such as GRE or IP-in-IP. The leaf refers to an identifier of the "underlay-transport" defined in [RFC9181], which describes the transport technology carrying the VPN service's traffic. If there are multiple types of tunnels between a single pair of VPN nodes, a separate link for each type of tunnel can be created 🚇

For the data nodes of "termination-point" in Figure 6, the module defines this minimal set of statistics 🌈

  • Last updated time ("last-updated"): Indicates the date and time when the counters were last updated 🕰️
  • Inbound statistics: A set of inbound statistics attributes used to measure the inbound statistics of the termination point, such as received packets, received packets with errors, and so on 📥
  • Outbound statistics: A set of outbound statistics attributes used to measure the outbound statistics of the termination point, such as sent packets, and packets that could not be sent due to errors 📤
  • VPN network access ("vpn-network-access"): Lists counters of the VPN network access defined in the L3NM [RFC9182] or the L2NM [RFC9291]. When multiple VPN network accesses are created using the same physical port, finer-grained metrics can be monitored. If a TP is associated with only a single VPN, this list is not required 💜

5. Network and VPN Service Performance Monitoring YANG Module 👑

The "ietf-network-vpn-pm" YANG module uses types defined in [RFC6991], [RFC8345], [RFC8532], and [RFC9181] 🫶

<CODE BEGINS> file "ietf-network-vpn-pm@2023-03-20.yang"
module ietf-network-vpn-pm {
  yang-version 1.1;
  namespace "urn:ietf:params:xml:ns:yang:ietf-network-vpn-pm";
  prefix nvp;

okay bestie, we're deep in the YANG module now and it is SO good 🧚💖 here's the next chunk of the module, with all the imports, identities, typedefs and groupings exactly as they are 🎀

import ietf-yang-types {
  prefix yang;
  reference
    "RFC 6991: Common YANG Data Types";
}
import ietf-vpn-common {
  prefix vpn-common;
  reference
    "RFC 9181: A Common YANG Data Model for Layer 2 and
         Layer 3 VPNs";
}
import ietf-network {
  prefix nw;
  reference
    "RFC 8345: A YANG Data Model for Network
         Topologies, Section 6.1";
}
import ietf-network-topology {
  prefix nt;
  reference
    "RFC 8345: A YANG Data Model for Network
         Topologies, Section 6.2";
}
import ietf-lime-time-types {
  prefix lime;
  reference
    "RFC 8532: Generic YANG Data Model for the Management of
         Operations, Administration, and Maintenance (OAM)
         Protocols That Use Connectionless Communications";
}

organization
  "IETF OPSAWG (Operations and Management Area Working Group)";
contact
  "WG Web:   <https://datatracker.ietf.org/wg/opsawg/>
   WG List:  <mailto:opsawg@ietf.org>

   Editor: Bo Wu
        <lana.wubo@huawei.com>

   Editor: Mohamed Boucadair
        <mohamed.boucadair@orange.com>

   Editor: Qin Wu
        <bill.wu@huawei.com>

   Author: Oscar Gonzalez de Dios
        <oscar.gonzalezdedios@telefonica.com>

   Author: Bin Wen
        <bin_wen@comcast.com>";
description
  "This YANG module defines a model for network and VPN service
   performance monitoring (PM).

   Copyright (c) 2023 IETF Trust and the persons identified as
   authors of the code.  All rights reserved.

   Redistribution and use in source and binary forms, with or
   without modification, is permitted pursuant to, and subject
   to the license terms contained in, the Revised BSD License
   set forth in Section 4.c of the IETF Trust's Legal Provisions
   Relating to IETF Documents
   (https://trustee.ietf.org/license-info).

   This version of this YANG module is part of RFC 9375
   (https://www.rfc-editor.org/info/rfc9375); see the RFC itself
   for full legal notices.";

revision 2023-03-20 {
  description
    "Initial revision.";
  reference
    "RFC 9375: A YANG Data Model for Network and VPN Service
         Performance Monitoring";
}

identity node-type {
  description
    "Base identity for node type";
}

identity pe {
  base node-type;
  description
    "Provider Edge (PE) node type.  A PE is the device or set
     of devices at the edge of the provider network with the
     functionality that is needed to interface with the
     customer.";
}

identity p {
  base node-type;
  description
    "Provider router node type.  That is, a router
     in the core network that does not have interfaces
     directly toward a customer.";
}

identity asbr {
  base node-type;
  description
    "Autonomous System Border Router (ASBR) node type.";
  reference
    "RFC 4364: BGP/MPLS IP Virtual Private Networks (VPNs)";
}

identity pm-source-type {
  description
    "Base identity from which specific performance monitoring
     mechanism types are derived.";
}

identity pm-source-bgpls {
  base pm-source-type;
  description
    "Indicates BGP-LS as the performance monitoring metric
     source.";
  reference
    "RFC 8571: BGP - Link State (BGP-LS) Advertisement of
         IGP Traffic Engineering Performance Metric
         Extensions";
}

identity pm-source-owamp {
  base pm-source-type;
  description
    "Indicates the One-Way Active Measurement Protocol (OWAMP)
     as the performance monitoring metric source.";
  reference
    "RFC 4656: A One-way Active Measurement Protocol (OWAMP)";
}

identity pm-source-twamp {
  base pm-source-type;
  description
    "Indicates the Two-Way Active Measurement Protocol (TWAMP)
     as the performance monitoring metric source.";
  reference
    "RFC 5357: A Two-Way Active Measurement Protocol (TWAMP)";
}

identity pm-source-stamp {
  base pm-source-type;
  description
    "Indicates the Simple Two-way Active Measurement Protocol
     (STAMP) as the performance monitoring metric source.";
  reference
    "RFC 8762: Simple Two-Way Active Measurement Protocol";
}

identity pm-source-y-1731 {
  base pm-source-type;
  description
    "Indicates Ethernet OAM Y.1731 as the performance monitoring
     metric source.";
  reference
    "ITU-T Y.1731: Operations, administration and
           maintenance (OAM) functions and mechanisms
           for Ethernet-based networks";
}

identity pm-source-ioam {
  base pm-source-type;
  description
    "Indicates In Situ Operations, Administration, and Maintenance
     (IOAM) as the performance monitoring metric source.";
  reference
    "RFC 9197: Data Fields for In Situ Operations, Administration,
         and Maintenance (IOAM)";
}

identity pm-type {
  description
    "Base identity for the PM type.";
}

identity pm-type-network-link {
  base pm-type;
  description
    "Indicates that the PM type is for the link in
     the network topology.";
}

identity pm-type-vpn-inter-access {
  base pm-type;
  description
    "Indicates that the PM type is for logical point-to-point VPN
     connections between source and destination VPN access
     interfaces.";
}

identity pm-type-vpn-tunnel {
  base pm-type;
  description
    "Indicates that the PM type is for VPN tunnels.";
}

typedef percentage {
  type decimal64 {
    fraction-digits 5;
    range "0..100";
  }
  description
    "Percentage to 5 decimal places.";
}

typedef percentile {
  type decimal64 {
    fraction-digits 3;
    range "0..100";
  }
  description
    "The percentile is a value between 0 and 100 to 3
     decimal places, e.g., 10.000, 99.900, and 99.990.
     For example, for a given one-way delay measurement,
     if the percentile is set to 95.000 and the 95th percentile
     one-way delay is 2 milliseconds, then the 95 percent of
     the sample value is less than or equal to 2 milliseconds.";
}

grouping entry-summary {
  description
    "Entry summary grouping used for network topology
     augmentation.";
  container entry-summary {
    config false;
    description
      "Container for VPN or network entry summary.";
    container ipv4-num {
      leaf maximum-routes {
        type uint32;
        description
          "Indicates the maximum number of IPv4 routes
           for the VPN or network.";
      }
      leaf total-active-routes {
        type uint32;
        description
          "Indicates total active IPv4 routes
           for the VPN or network.";
      }
      description
        "IPv4-specific parameters.";
    }
    container ipv6-num {
      leaf maximum-routes {
        type uint32;
        description
          "Indicates the maximum number of IPv6 routes
           for the VPN or network.";
      }
      leaf total-active-routes {
        type uint32;
        description
          "Indicates total active IPv6 routes
           for the VPN or network.";
      }
      description
        "IPv6-specific parameters.";
    }
    container mac-num {
      leaf maximum-mac-entries {
        type uint32;
        description
          "Indicates the maximum number of MAC entries
           for the VPN or network.";
      }
      leaf total-active-mac-entries {
        type uint32;
        description
          "Indicates the total active MAC entries
           for the VPN or network.";
      }
      description
        "MAC statistics.";
    }
  }
}

grouping link-loss-statistics {
  description
    "Grouping for per-link error statistics.";
  container loss-statistics {
    description
      "One-way link loss summarized information.";
    reference
      "RFC 4656: A One-way Active Measurement Protocol (OWAMP)
       ITU-T Y.1731: Operations, administration and
             maintenance (OAM) functions and mechanisms
             for Ethernet-based networks";
    leaf packet-loss-count {
      type yang:counter64;
      description
        "Total number of lost packets.";
    }
    leaf loss-ratio {
      type percentage;
      description
        "Loss ratio of the packets.  Expressed as percentage
         of packets lost with respect to packets sent.";
    }
  }
}

grouping link-delay-statistics {
  description
    "Grouping for per-link delay statistics.";
  container delay-statistics {
    description
      "One-way link delay summarized information.";
    reference
      "RFC 4656: A One-way Active Measurement Protocol (OWAMP)
       ITU-T Y.1731: Operations, administration and
             maintenance (OAM) functions and mechanisms
             for Ethernet-based networks";
    leaf unit-value {
      type identityref {
        base lime:time-unit-type;
      }
      default "lime:milliseconds";
      description
        "Time units, where the options are hours, minutes, seconds,
         milliseconds, microseconds, and nanoseconds.";
    }
    leaf min-delay-value {
      type yang:gauge64;
      description
        "Minimum observed one-way delay.";
    }
    leaf max-delay-value {
      type yang:gauge64;
      description
        "Maximum observed one-way delay.";
    }
    leaf low-delay-percentile {
      type yang:gauge64;
      description
        "Low percentile of observed one-way delay with
         specific measurement method.";
    }
    leaf intermediate-delay-percentile {
      type yang:gauge64;
      description
        "Intermediate percentile of observed one-way delay with
         specific measurement method.";
    }
    leaf high-delay-percentile {
      type yang:gauge64;
      description
        "High percentile of observed one-way delay with
         specific measurement method.";
    }
  }
}

that's the whole chunk, and the module keeps rolling from here 💅✨

okay bestie, the YANG glow-up keeps going 💅✨ here's the next batch of groupings and augments, with every single line kept exact

     grouping link-jitter-statistics {
       description
         "Grouping for per-link jitter statistics.";
       container jitter-statistics {
         description
           "One-way link jitter summarized information.";
         reference
           "RFC 3393: IP Packet Delay Variation Metric
                for IP Performance Metrics (IPPM)
            RFC 4656: A One-way Active Measurement Protocol (OWAMP)
            ITU-T Y.1731: Operations, administration and
                  maintenance (OAM) functions and mechanisms
                  for Ethernet-based networks";
         leaf unit-value {
           type identityref {
             base lime:time-unit-type;
           }
           default "lime:milliseconds";
           description
             "Time units, where the options are hours, minutes, seconds,
              milliseconds, microseconds, and nanoseconds.";
         }
         leaf min-jitter-value {
           type yang:gauge64;
           description
             "Minimum observed one-way jitter.";
         }
         leaf max-jitter-value {
           type yang:gauge64;
           description
             "Maximum observed one-way jitter.";
         }
         leaf low-jitter-percentile {
           type yang:gauge64;
           description
             "Low percentile of observed one-way jitter.";
         }
         leaf intermediate-jitter-percentile {
           type yang:gauge64;
           description
             "Intermediate percentile of observed one-way jitter.";
         }
         leaf high-jitter-percentile {
           type yang:gauge64;
           description
             "High percentile of observed one-way jitter.";
         }
       }
     }

     grouping tp-svc-telemetry {
       leaf last-updated {
         type yang:date-and-time;
         config false;
         description
           "Indicates the date and time when the counters were
            last updated.";
       }
       leaf inbound-octets {
         type yang:counter64;
         description
           "The total number of octets received on the
            interface, including framing characters.";
       }
       leaf inbound-unicast {
         type yang:counter64;
         description
           "The total number of inbound unicast packets.";
       }
       leaf inbound-broadcast {
         type yang:counter64;
         description
           "The total number of inbound broadcast packets.";
       }
       leaf inbound-multicast {
         type yang:counter64;
         description
           "The total number of inbound multicast packets.";
       }
       leaf inbound-discards {
         type yang:counter64;
         description
           "The number of inbound packets that were discarded
            even though no errors had been detected.  Possible
            reasons for discarding such a packet could be to
            free up buffer space, not enough buffer for too
            much data, etc.";
       }
       leaf inbound-errors {
         type yang:counter64;
         description
           "The number of inbound packets that contained errors.";
       }
       leaf inbound-unknown-protocol {
         type yang:counter64;
         description
           "The number of packets received via the interface
            that were discarded because of an unknown or
            unsupported protocol.";
       }
       leaf outbound-octets {
         type yang:counter64;
         description
           "The total number of octets transmitted out of the
            interface, including framing characters.";
       }
       leaf outbound-unicast {
         type yang:counter64;
         description
           "The total number of outbound unicast packets.";
       }
       leaf outbound-broadcast {
         type yang:counter64;
         description
           "The total number of outbound broadcast packets.";
       }
       leaf outbound-multicast {
         type yang:counter64;
         description
           "The total number of outbound multicast packets.";
       }
       leaf outbound-discards {
         type yang:counter64;
         description
           "The number of outbound packets that were discarded
            even though no errors had been detected to
            prevent their transmission.  Possible reasons
            for discarding such a packet could be to free
            up buffer space, not enough buffer for too
            much data, etc.";
       }
       leaf outbound-errors {
         type yang:counter64;
         description
           "The number of outbound packets that contained errors.";
       }
       description
         "Grouping for interface service telemetry.";
     }

     augment "/nw:networks/nw:network/nw:network-types" {
       description
         "Defines the service topologies types.";
       container service {
         presence "Presence of the container indicates performance
                   monitoring of the VPN service, and absence of
                   the container indicates performance monitoring
                   of the network itself.";
         description
           "Container for VPN service.";
         leaf service-type {
           type identityref {
             base vpn-common:service-type;
           }
           mandatory true;
           description
             "This indicates the network service type,
              e.g., L3VPN and VPLS.";
         }
         leaf vpn-id {
           type vpn-common:vpn-id;
           description
             "VPN identifier.";
         }
         leaf vpn-service-topology {
           type identityref {
             base vpn-common:vpn-topology;
           }
           description
             "VPN service topology, e.g., hub-spoke, any-to-any,
              and hub-spoke-disjoint.";
         }
       }
     }

     augment "/nw:networks/nw:network/nw:node" {
       description
         "Augments the network node with other general attributes.";
       leaf node-type {
         type identityref {
           base node-type;
         }
         description
           "Node type, e.g., PE, P, and ASBR.";
       }
       uses entry-summary;
     }

     augment "/nw:networks/nw:network/nw:node" {
       when '../nw:network-types/nvp:service' {
         description
           "Augments for VPN service PM.";
       }
       description
         "Augments the network node with VPN service attributes.";
       leaf role {
         type identityref {
           base vpn-common:role;
         }
         default "vpn-common:any-to-any-role";
         description
           "Role of the node in the VPN service topology.";
       }
     }

     augment "/nw:networks/nw:network/nt:link" {
       description
         "Augments the network topology link with performance
          monitoring attributes.";
       container perf-mon {
         description
           "Container for PM attributes.";
         leaf low-percentile {
           type percentile;
           default "10.000";
           description
             "Low percentile to report.  Setting low-percentile
              to 0.000 indicates the client is not interested
              in receiving low percentile.";
         }
         leaf intermediate-percentile {
           type percentile;
           default "50.000";
           description
             "Intermediate percentile to report.  Setting
              intermediate-percentile to 0.000 indicates the client
              is not interested in receiving intermediate percentile.";
         }
         leaf high-percentile {
           type percentile;
           default "95.000";
           description
             "High percentile to report.  Setting high-percentile
              to 0.000 indicates the client is not interested in
              receiving high percentile.";
         }
         leaf measurement-interval {
           type uint32 {
             range "1..max";
           }
           units "seconds";
           default "60";
           description
             "Indicates the time interval to perform PM
              measurement over.";
         }
         list pm {
           key "pm-type";
           config false;
           description
             "The list of PM based on PM type.";
           leaf pm-type {
             type identityref {
               base pm-type;
             }
             config false;
             description
               "The PM type of the measured PM attributes.";
           }
           container pm-attributes {
             description
               "Container for PM attributes.";
             leaf start-time {
               type yang:date-and-time;
               config false;
               description
                 "The date and time the measurement last started.";
             }
             leaf end-time {
               type yang:date-and-time;
               config false;
               description
                 "The date and time the measurement last ended.";
             }
             leaf pm-source {
               type identityref {
                 base pm-source-type;
               }
               config false;
               description
                 "The OAM tool used to collect the PM data.";
             }
             container one-way-pm-statistics {
               config false;
               description
                 "Container for link telemetry attributes.";
               uses link-loss-statistics;
               uses link-delay-statistics;
               uses link-jitter-statistics;
             }
             list one-way-pm-statistics-per-class {
               key "class-id";
               config false;
               description
                 "The list of PM data based on class of service.";
               leaf class-id {
                 type string;
                 description
                   "The class-id is used to identify the class
                    of service.  This identifier is internal
                    to the administration.";
               }
               uses link-loss-statistics;
               uses link-delay-statistics;
               uses link-jitter-statistics;
             }
           }
         }
       }
     }

the jitter grouping is basically the "how wiggly is the delay" stats 🦋 min, max, and three percentiles, all in gauge64, with the time unit defaulting to milliseconds 🌸

the telemetry grouping is a whole counter squad 🎀 inbound and outbound octets, unicast, broadcast, multicast, discards, errors, and unknown-protocol drops, all counter64 🔥 only last-updated is read-only, since it's config false 🙈

the augments hang the new stuff onto the network, node, and link objects 💜 the service container shows up only when the thing being monitored is a VPN service 👑 the perf-mon container on each link holds the percentile knobs, a 60-second default measurement interval, and the per-PM-type and per-class stats lists 🧚

     augment "/nw:networks/nw:network/nt:link/perf-mon" {
       when '../../nw:network-types/nvp:service' {
         description
           "Augments for VPN service PM.";
       }
       description
         "Augments the network topology link with VPN service
          performance monitoring attributes.";
       container vpn-pm-type {
         description
           "The VPN PM type of this logical point-to-point
            unidirectional VPN link.";
         container inter-vpn-access-interface {
           description
             "Indicates inter-vpn-access-interface PM, which is used
              to monitor the performance of logical point-to-point
              VPN connections between source and destination VPN
              access interfaces.";
           leaf inter-vpn-access-interface {
             type empty;
             description
               "This is a placeholder for inter-vpn-access-interface PM,
                which is not bound to a specific VPN access interface.
                The source or destination VPN access interface
                of the measurement can be augmented as needed.";
           }
         }
         container vpn-tunnel {
           presence "Enables VPN tunnel PM";
           description
             "Indicates VPN tunnel PM, which is used to monitor
              the performance of VPN tunnels.";
           leaf vpn-tunnel-type {
             type identityref {
               base vpn-common:protocol-type;
             }
             config false;
             description
               "The leaf indicates the VPN tunnel type, e.g.,
                Generic Routing Encapsulation (GRE) and Generic
                Network Virtualization Encapsulation (Geneve).";
           }
         }
       }
     }

     augment "/nw:networks/nw:network/nw:node/nt:termination-point" {
       description
         "Augments the network topology termination point with
          performance monitoring attributes.";
       container pm-statistics {
         config false;
         description
           "Container for termination point PM attributes.";
         uses tp-svc-telemetry;
       }
     }

     augment "/nw:networks/nw:network/nw:node"
           + "/nt:termination-point/pm-statistics" {
       when '../../../nw:network-types/nvp:service' {
         description
           "Augments for VPN service PM.";
       }
       description
         "Augments the network topology termination-point with
          VPN service performance monitoring attributes.";
       list vpn-network-access {
         key "network-access-id";
         description
           "The list of PM based on VPN network accesses.";
         leaf network-access-id {
           type vpn-common:vpn-id;
           description
             "The reference to an identifier for the VPN network
              access.";
         }
         uses tp-svc-telemetry;
       }
     }
   }
   <CODE ENDS>

6. Security Considerations 🔒✨

okay bestie, this part is serious, but we're gonna get through it together 💅. The YANG module in this document is a schema for data that's meant to be reached through network management protocols like NETCONF [RFC6241] or RESTCONF [RFC8040]. The lowest NETCONF layer is the secure transport layer, and the mandatory-to-implement secure transport is Secure Shell (SSH) [RFC6242]. The lowest RESTCONF layer is HTTPS, and the mandatory-to-implement secure transport is TLS [RFC8446]. 🔐

The Network Configuration Access Control Model (NACM) [RFC8341] is the thing that lets you restrict access for particular NETCONF or RESTCONF users. You can limit them to a preconfigured subset of all the available NETCONF or RESTCONF protocol operations and content. It's giving bouncer at the club 🚪💖

Now the spicy part. A bunch of data nodes in this YANG module are writable, creatable, and deletable (that's config true, which is the default). These nodes may be considered sensitive or vulnerable in some network environments 😭. Write operations (for example, edit-config) to these nodes without proper protection can have a negative effect on network operations. Those writes can lead to inaccurate or incomplete network measurements, and that can mess with the visibility and the decisions this data is supposed to inform. Unauthorized write access to the subtrees below could have the impacts shown in Table 1.

    +============+======================+============================+
    | Access     | Node                 | Potential Impact           |
    +============+======================+============================+
    | /nw:networks/nw:network/nw:network-types                       |
    +============+======================+============================+
    | write      | service type         | disable VPN PM             |
    +------------+----------------------+----------------------------+
    | write      | VPN identifier       | disable VPN PM             |
    +------------+----------------------+----------------------------+
    | write      | VPN service topology | render data unusable       |
    +============+======================+============================+
    | /nw:networks/nw:network/nw:node                                |
    +============+======================+============================+
    | write      | node type            | render data unusable       |
    +------------+----------------------+----------------------------+
    | write      | VPN topology role    | render data unusable       |
    +============+======================+============================+
    | /nw:networks/nw:network/nw:link/nvp:perf-mon                   |
    +============+======================+============================+
    | write      | percentile           | impact reporting cadence   |
    +------------+----------------------+----------------------------+
    | write      | measurement interval | impact monitoring fidelity |
    +------------+----------------------+----------------------------+
    | write      | vpn-pm-type          | impact monitoring fidelity |
    +------------+----------------------+----------------------------+

               Table 1: Write Operation Sensitivity Impact

Some of the readable data nodes in this YANG module can be sensitive or vulnerable too, so it's really important to control read access (for example, via get, get-config, or notification) to them 👀. When you're using this stuff, you need to weigh confidentiality against proper performance monitoring, because there's a trade-off there ⚖️. Unauthorized access to these subtrees could have these impacts:

  • /nw:networks/nw:network/nw:node: unauthorized read access to this subtree can disclose the operational state information of underlay network instances or VPN instances.
  • /nw:networks/nw:network/nt:link/nvp:perf-mon/nvp:one-way-pm-statistics: unauthorized read access to this subtree can disclose the operational state information of underlay network links or VPN abstract links.
  • /nw:networks/nw:network/nw:node/nt:termination-point/nvp:pm-statistics: unauthorized read access to this subtree can disclose the operational state information of underlay network termination points or VPN network accesses.

This YANG module does not define any Remote Procedure Call (RPC) operations and actions, so there's no RPC drama here 🙌.

7. IANA Considerations 🦋

IANA has registered the following URI in the "ns" subregistry within the "IETF XML Registry" [RFC3688] 📝:

  • URI: urn:ietf:params:xml:ns:yang:ietf-network-vpn-pm
  • Registrant Contact: The IESG.
  • XML: N/A; the requested URI is an XML namespace.

IANA has also registered the following YANG module in the "YANG Module Names" subregistry [RFC6020] within the "YANG Parameters" registry 🌸:

  • Name: ietf-network-vpn-pm
  • Namespace: urn:ietf:params:xml:ns:yang:ietf-network-vpn-pm
  • Maintained by IANA: N
  • Prefix: nvp
  • Reference: RFC 9375

8. References ✨

8.1. Normative References 📚

These are the docs this spec leans on hard, so you basically need them to get anything going 🤓

8.2. Informative References 🔍

These ones are good to peek at, but they aren't required reading, lowkey 🫶

Appendix A. Illustrative Examples 🧚

A.1. Example of VPN Performance Subscription 💜

Okay so Figure 7 shows a client subscribing to performance monitoring info between nodes ("node-id") A and B in the L3 network topology 🗺️ The parameter the client is obsessed with? End-to-end loss 📉

   ============== NOTE: '\' line wrapping per RFC 8792 ===============

   POST /restconf/operations/ietf-subscribed-notifications:establish-\
                                      subscription
   Host: example.com
   Content-Type: application/yang-data+json

   {
     "ietf-subscribed-notifications:input": {
       "stream-subtree-filter": {
         "ietf-network:networks": {
           "network": {
             "network-id": "example:VPN1",
             "ietf-network-vpn-pm:service": {
               "service-type": "ietf-vpn-common:l3vpn"
             },
             "node": [
               {
                 "node-id": "example:A",
                 "ietf-network-vpn-pm:node-type": "pe",
                 "termination-point": [
                   {
                     "tp-id": "example:1-0-1"
                   }
                 ]
               },
               {
                 "node-id": "example:B",
                 "ietf-network-vpn-pm:node-type": "pe",
                 "termination-point": [
                   {
                     "tp-id": "example:2-0-1"
                   }
                 ]
               }
             ],
             "ietf-network-topology:link": [
               {
                 "link-id": "example:A-B",
                 "source": {
                   "source-node": "example:A"
                 },
                 "destination": {
                   "dest-node": "example:B"
                 },
                 "ietf-network-vpn-pm:perf-mon": {
                   "pm": [
                     {
                       "pm-type": "pm-type-vpn-tunnel",
                       "pm-attributes": {
                         "one-way-pm-statistics": {
                           "loss-statistics": {
                             "packet-loss-count": {}
                           }
                         }
                       }
                     }
                   ],
                   "vpn-pm-type": {
                     "vpn-tunnel": {
                       "vpn-tunnel-type": "ietf-vpn-common:gre"
                     }
                   }
                 }
               }
             ]
           }
         },
         "ietf-yang-push:periodic": {
           "period": "500"
         }
       }
     }
   }

                   Figure 7: Example of Pub/Sub Retrieval

A.2. Example of VPN Performance Snapshot ✨

Okay bestie, this one is a snapshot 📸. Figure 8 shows the message body of a RESTCONF request that fetches the performance data for the link and termination points (TPs) belonging to "VPN1". Literally just a GET for the stats 💅

{
  "ietf-network:networks": {
    "network": {
      "network-id": "example:VPN1",
      "node": [
        {
          "node-id": "example:A",
          "ietf-network-vpn-pm:node-type": "pe",
          "termination-point": [
            {
              "tp-id": "example:1-0-1",
              "ietf-network-vpn-pm:pm-statistics": {
                "inbound-octets": "100",
                "outbound-octets": "150"
              }
            }
          ]
        },
        {
          "node-id": "example:B",
          "ietf-network-vpn-pm:node-type": "pe",
          "termination-point": [
            {
              "tp-id": "example:2-0-1",
              "ietf-network-vpn-pm:pm-statistics": {
                "inbound-octets": "150",
                "outbound-octets": "100"
              }
            }
          ]
        }
      ],
      "ietf-network-topology:link": [
        {
          "link-id": "example:A-B",
          "source": {
            "source-node": "example:A"
          },
          "destination": {
            "dest-node": "example:B"
          },
          "ietf-network-pm:perf-mon": {
            "pm": [
              {
                "pm-type": "pm-type-vpn-tunnel",
                "pm-attributes": {
                  "one-way-pm-statistics": {
                    "loss-statistics": {
                      "packet-loss-count": "120"
                    }
                  }
                }
              }
            ],
            "vpn-pm-type": {
              "vpn-tunnel": {
                "vpn-tunnel-type": "ietf-vpn-common:gre"
              }
            }
          }
        }
      ]
    }
  }
}

Figure 8: Example of VPN PM

Look at that node A and node B TP stats mirroring each other: 100 out and 150 in on one side, 150 in and 100 out on the other 🪞 And the link carries a packet-loss-count of 120 on a GRE tunnel. Iconic data, honestly 🦋

A.3. Example of Percentile Monitoring 📊

This one shows percentile measurement data that could come back for the link "example:A-B" between "example:A" and "example:B". Percentiles are giving "here's the whole distribution, not just the average" energy 🌈

{
  "ietf-network-topology:link": [
    {
      "link-id": "example:A-B",
      "source": {
        "source-node": "example:A"
      },
      "destination": {
        "dest-node": "example:B"
      },
      "ietf-network-vpn-pm:perf-mon": {
        "low-percentile": "20.000",
        "intermediate-percentile": "50.000",
        "high-percentile": "90.000",
        "pm": [
          {
            "pm-type": "pm-type-vpn-inter-access",
            "pm-attributes": {
              "one-way-pm-statistics": {
                "delay-statistics": {
                  "unit-value": "ietf-lime-time-types:milliseconds",
                  "min-delay-value": "43",
                  "max-delay-value": "99",
                  "low-delay-percentile": "64",
                  "intermediate-delay-percentile": "77",
                  "high-delay-percentile": "98"
                }
              }
            }
          }
        ],
        "vpn-pm-type": {
          "inter-vpn-access-interface": {
            "inter-vpn-access-interface": [null]
          }
        }
      }
    }
  ]
}

Figure 9: Example of VPN PM with Percentile Value

Okay the delay numbers are so telling 😭 Min is 43 ms, max is 99 ms, and the 90th percentile lives up at the high end. Percentiles tell the real story, bestie 💜

Acknowledgements 🫶

Huge thanks to Joe Clarke, Adrian Farrel, Tom Petch, Greg Mirsky, Roque Gagliano, Erez Segev, and Dhruv Dhody for reviewing this doc and dropping important input. You're all absolute legends 👑

This work is partially supported by the European Commission under Horizon 2020 Secured autonomic traffic management for a Tera of SDN flows (Teraflow) project (grant agreement number 101015857). Funding that made this happen, so cute 🌸

Contributors 🎀

The following authors contributed significantly to this document:

Michale Wang Huawei Email: wangzitao@huawei.com

Roni Even Huawei Email: ron.even.tlv@gmail.com

Change Liu China Unicom Email: liuc131@chinaunicom.cn

Honglei Xu China Telecom Email: xuhl6@chinatelecom.cn

Authors' Addresses 💌

Bo Wu (editor)

Huawei
Yuhua District
101 Software Avenue
Nanjing
Jiangsu, 210012
China
Email: lana.wubo@huawei.com

Qin Wu (editor)

Huawei
Yuhua District
101 Software Avenue
Nanjing
Jiangsu, 210012
China
Email: bill.wu@huawei.com

Mohamed Boucadair (editor)

Orange
Rennes 35000
France
Email: mohamed.boucadair@orange.com

Oscar Gonzalez de Dios

Telefonica
Madrid
Spain
Email: oscar.gonzalezdedios@telefonica.com

Bin Wen

Comcast
Email: bin_wen@comcast.com
📦 where this page comes from