RFC 9468

Unsolicited Bidirectional Forwarding Detection (BFD) for Sessionless Applications

📅 published August 2023, from ietf

RFC 9468: Unsolicited Bidirectional Forwarding Detection (BFD) for Sessionless Applications 💖

okay bestie, this is the August 2023 doc, and it's a Standards Track one 🎀 (ISSN 2070-1721). It was written by E. Chen from Palo Alto Networks, N. Shen from Zededa, R. Raszuk from Arrcus, and R. Rahman from Equinix 🫶. It doesn't list anything in an "Obsoletes" or "Updates" slot. Instead, it augments RFC 9314, which is how it links up with the existing BFD YANG work ✨.

Abstract

okay so the whole point here is making "sessionless" applications that use Bidirectional Forwarding Detection (BFD) way less of a chore to run 🦋. The doc lays out procedures for "unsolicited BFD," which lets a BFD session get started by only one side. The other side doesn't need explicit per-session configuration or registration to take part, but it's still subject to certain per-interface or global policies 🔥.

It also brings in a new YANG module to configure and manage "unsolicited BFD" 💅. The module is based on YANG 1.1, as defined in RFC 7950, and conforms to the Network Management Datastore Architecture (NMDA), as described in RFC 8342. And it augments RFC 9314 🌸.

Status of This Memo

this is an Internet Standards Track document 🌈.

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

For the current status of this document, any errata, and how to give feedback, go to https://www.rfc-editor.org/info/rfc9468 🧚.

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) in effect on the date of publication of this document. Please review these documents carefully, because they describe your rights and restrictions with respect to this document 📜. Code Components extracted from this document must include Revised BSD License text as described in Section 4.e of the Trust Legal Provisions, and they're provided without warranty as described in the Revised BSD License 🫶.

Table of Contents

1.  Introduction
  1.1.  Requirements Language
2.  Procedures for Unsolicited BFD
3.  State Variables
4.  YANG Data Model
  4.1.  Unsolicited BFD Hierarchy
  4.2.  Unsolicited BFD Module
  4.3.  Data Model Example
5.  IANA Considerations
6.  Security Considerations
  6.1.  BFD Protocol Security Considerations
  6.2.  BFD Protocol Authentication Considerations
  6.3.  YANG Module Security Considerations
7.  References
  7.1.  Normative References
  7.2.  Informative References
Acknowledgments
Authors' Addresses

1. Introduction 🌷

okay so the current implementation and deployment practice for BFD ([RFC5880] and [RFC5881]) usually means BFD sessions have to be explicitly configured or registered on both sides 😩. That's totally fine when something like BGP [RFC4271] already has a "session" involving both sides to get established 💅. But it gets annoying when that prerequisite "session" doesn't naturally exist between two endpoints in an application. Simultaneous configuration and coordination may be needed on both sides before BFD can take effect. For example:

  • When BFD is used to keep track of the "liveness" of the next hop of static routes. Although only one side may need the BFD functionality, both sides currently need to be involved in specific configuration and coordination. In some cases, static routes are created unnecessarily just for BFD 💀.
  • When BFD is used to keep track of the "liveness" of the third-party next hop of BGP routes received from the Route Server [RFC7947] at an Internet Exchange Point (IXP). The third-party next hop is different from the peering address of the Route Server, so for BFD to work, two routers peering with the Route Server currently need to have routes and next hops from each other (although indirectly, via the Route Server) 🙈.

Clearly, it's beneficial and desirable to reduce or eliminate unnecessary configurations and coordination in these "sessionless" applications using BFD 🍓.

In this document, we present procedures for "unsolicited BFD." These let a BFD session be initiated by only one side and established without explicit per-session configuration or registration by the other side, subject to certain per-interface or global policies ✨.

Unsolicited BFD only impacts the initiation of BFD sessions. Nothing else in [RFC5880] changes, including (but not limited to) the Echo function and Demand mode 🦋.

With "unsolicited BFD," there's a potential risk of excessive resource usage by BFD from "unexpected" remote systems 😭. To mitigate those risks, several mechanisms are recommended in the Security Considerations section 🔒.

The procedure in this document could be applied to BFD for multihop paths [RFC5883]. However, because of security risks, this document applies only to BFD for single IP hops [RFC5881] 🚫.

Compared to "Seamless BFD" [RFC7880], this proposal involves only minor procedural enhancements to the widely deployed BFD itself. So we believe this proposal is inherently simpler, both in the protocol itself and in deployment 💖. For example, it doesn't require exchanging BFD discriminators over an out-of-band channel before BFD session bring-up 🧚.

When BGP ADD-PATH [RFC7911] is deployed at an IXP using a Route Server, multiple BGP paths (when they exist) can be made available to the clients of the Route Server, as described in [RFC7947] 🌈. Unsolicited BFD can be used by BGP route selection's route resolvability condition (Section 9.1.2.1 of [RFC4271]) to exclude routes where the NEXT_HOP isn't reachable using the procedures in this document 👑.

1.1. Requirements Language

The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", "SHOULD", "SHOULD NOT", "RECOMMENDED", "NOT RECOMMENDED", "MAY", and "OPTIONAL" in this document are to be interpreted as described in BCP 14 [RFC2119] [RFC8174] when, and only when, they appear in all capitals, as shown here 📖.

2. Procedures for Unsolicited BFD 🎀

okay so with "unsolicited BFD," one side takes the "Active role" and the other side takes the "Passive role," as described in [RFC5880], Section 6.1 💅.

Passive unsolicited BFD support MUST be disabled by default, and it MUST require explicit configuration to be enabled 🔥. On the passive side, the following BFD parameters, from [RFC5880], Section 6.8.1, SHOULD be configurable 🌸:

  • bfd.DesiredMinTxInterval
  • bfd.RequiredMinRxInterval
  • bfd.DetectMult

The passive side MAY also choose to use the values of the parameters listed above that the active side uses in its BFD Control packets 🦋. However, the bfd.LocalDiscr value MUST be selected by the passive side to allow multiple unsolicited BFD sessions ✨.

The active side starts sending BFD Control packets, as specified in [RFC5880]. The passive side doesn't send BFD Control packets at first. It only starts sending them after it has received BFD Control packets from the active side 🧚.

When the passive side receives a BFD Control packet from the active side with 0 as Your Discriminator and doesn't find an existing BFD session, the passive side SHOULD create a matching BFD session toward the active side, unless local configuration or policy doesn't permit it 🫶.

When the passive side receives an incoming BFD Control packet on a numbered interface, the source address of that packet MUST belong to the subnet of the interface on which the BFD packet is received. Otherwise, the BFD Control packet MUST NOT be processed 🚫.

The passive side MUST then start sending BFD Control packets and perform the necessary procedure for bringing up, maintaining, and tearing down the BFD session 💖. If the BFD session fails to get established within a certain amount of time (which is implementation specific but has to be at least equal to the local failure detection time), or if an established BFD session goes down, the passive side MUST stop sending BFD Control packets. It SHOULD also delete the BFD session it created, until the active side initiates BFD Control packets again 😭.

When an unsolicited BFD session goes down, an implementation may retain the session state for a period of time. Retaining this state can be useful for operational purposes 💡.

3. State Variables 🌙

this document defines a new state variable called Role 🧁:

bfd.Role

This is the role of the local system during BFD session initialization, as per [RFC5880], Section 6.1. The possible values are Active or Passive ✨.

4. YANG Data Model 💜

this section extends the YANG data model for BFD [RFC9314] to cover unsolicited BFD 🌈. The new module imports the YANG modules described in [RFC8349], since the "bfd" container in [RFC9314] is under "control-plane-protocol." The YANG module in this document conforms to the Network Management Datastore Architecture (NMDA) [RFC8342] 👑.

4.1. Unsolicited BFD Hierarchy

okay bestie 🎀 so unsolicited BFD parameters for IP single-hop sessions can be configured at 2 levels 💅:

  • globally, i.e., for all interfaces 🌍
  • for specific interfaces (this requires support for the unsolicited-params-per-interface feature) 🦋

If there's configuration at both levels, per-interface configuration takes precedence over global configuration 👑. This is lowkey the most important line in this whole section 🫶.

For operational data, there's a brand new role leaf node for BFD IP single-hop sessions ✨. Iconic. The tree diagram below uses the graphical representation of data models, as defined in [RFC8340] 🌸.

module: ietf-bfd-unsolicited

  augment /rt:routing/rt:control-plane-protocols
          /rt:control-plane-protocol/bfd:bfd/bfd-ip-sh:ip-sh:
    +--rw unsolicited?
       +--rw local-multiplier?                 multiplier
       +--rw (interval-config-type)?
          +--:(tx-rx-intervals)
          |  +--rw desired-min-tx-interval?    uint32
          |  +--rw required-min-rx-interval?   uint32
          +--:(single-interval) {single-minimum-interval}?
             +--rw min-interval?               uint32
  augment /rt:routing/rt:control-plane-protocols
          /rt:control-plane-protocol/bfd:bfd/bfd-ip-sh:ip-sh
          /bfd-ip-sh:interfaces:
    +--rw unsolicited
       +--rw enabled?                          boolean
       +--rw local-multiplier?
            bfd-types:multiplier
             {bfd-unsol:unsolicited-params-per-interface}?
       +--rw (interval-config-type)?
               {bfd-unsol:unsolicited-params-per-interface}?
          +--:(tx-rx-intervals)
          |  +--rw desired-min-tx-interval?    uint32
          |  +--rw required-min-rx-interval?   uint32
          +--:(single-interval) {bfd-types:single-minimum-interval}?
             +--rw min-interval?               uint32
  augment /rt:routing/rt:control-plane-protocols
          /rt:control-plane-protocol/bfd:bfd/bfd-ip-sh:ip-sh
          /bfd-ip-sh:sessions/bfd-ip-sh:session:
    +--ro role?   bfd-unsol:role

4.2. Unsolicited BFD Module

Here's the main character of this whole document, the full YANG module 🧚. Implementers rely on every single line, so copy it exactly 💖.

<CODE BEGINS> file "ietf-bfd-unsolicited@2023-08-31.yang"
module ietf-bfd-unsolicited {

  yang-version 1.1;

  namespace "urn:ietf:params:xml:ns:yang:ietf-bfd-unsolicited";

  prefix bfd-unsol;

  import ietf-bfd-types {
    prefix bfd-types;
    reference
      "RFC 9314: YANG Data Model for Bidirectional Forwarding
       Detection (BFD)";
  }

  import ietf-bfd {
    prefix bfd;
    reference
      "RFC 9314: YANG Data Model for Bidirectional Forwarding
       Detection (BFD)";
  }

  import ietf-bfd-ip-sh {
    prefix bfd-ip-sh;
    reference
      "RFC 9314: YANG Data Model for Bidirectional Forwarding
       Detection (BFD)";
  }

  import ietf-routing {
    prefix rt;
    reference
      "RFC 8349: A YANG Data Model for Routing Management
       (NMDA Version)";
  }

  organization
    "IETF BFD Working Group";

  contact
    "WG Web:   <https://datatracker.ietf.org/wg/bfd/>
     WG List:  <rtg-bfd@ietf.org>

     Editors:  Enke Chen (enchen@paloaltonetworks.com),
               Naiming Shen (naiming@zededa.com),
               Robert Raszuk (robert@raszuk.net),
               Reshad Rahman (reshad@yahoo.com)";

  description
    "This module contains the YANG definition for unsolicited BFD,
     as per RFC 9468.

     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
     (http://trustee.ietf.org/license-info).

     This version of this YANG module is part of RFC 9468; see
     the RFC itself for full legal notices.";

  reference
    "RFC 9468: Unsolicited Bidirectional Forwarding Detection
     (BFD) for Sessionless Applications";

  revision 2023-08-31 {
    description
      "Initial revision.";
    reference
      "RFC 9468: Unsolicited Bidirectional Forwarding Detection (BFD)
       for Sessionless Applications";
  }

  /*
   * Feature definitions
   */
  feature unsolicited-params-per-interface {
    description
      "This feature indicates that the server supports per-interface
       parameters for unsolicited sessions.";
    reference
      "RFC 9468: Unsolicited Bidirectional Forwarding Detection (BFD)
       for Sessionless Applications";
  }

  /*
   * Type Definitions
   */

  identity role {
    description
      "Base identity from which all roles are derived.
       Role of local system during BFD session initialization.";
  }

  identity active {
    base bfd-unsol:role;
    description
      "Active role.";
    reference
      "RFC 5880: Bidirectional Forwarding Detection (BFD),
       Section 6.1";
  }

  identity passive {
    base bfd-unsol:role;
    description
      "Passive role.";
    reference
      "RFC 5880: Bidirectional Forwarding Detection (BFD),
       Section 6.1";
  }

  /*
   * Augments
   */

  augment "/rt:routing/rt:control-plane-protocols/"
        + "rt:control-plane-protocol/bfd:bfd/bfd-ip-sh:ip-sh" {
    description
      "Augmentation for unsolicited BFD parameters.";
    container unsolicited {
      description
        "BFD IP single-hop unsolicited top-level container.";
      uses bfd-types:base-cfg-parms;
    }
  }

  augment "/rt:routing/rt:control-plane-protocols/"
        + "rt:control-plane-protocol/bfd:bfd/bfd-ip-sh:ip-sh/"
        + "bfd-ip-sh:interfaces" {
    description
      "Augmentation for unsolicited BFD on IP single-hop
       interface.";
    container unsolicited {
      description
        "BFD IP single-hop interface unsolicited top-level
         container.";
      leaf enabled {
        type boolean;
        default "false";
        description
          "Unsolicited BFD is enabled on this interface.";
      }
      /*
       * The following is the same as bfd-types:base-cfg-parms, but
       * without default values (for inheritance)
       */
      leaf local-multiplier {
        if-feature "bfd-unsol:unsolicited-params-per-interface";
        type bfd-types:multiplier;
        description
          "Multiplier transmitted by the local system.  Defaults to
           ../../unsolicited/local-multiplier.
           A multiplier configured under an interface takes
           precedence over the multiplier configured at the global
           level.";
      }
      choice interval-config-type {
        if-feature "bfd-unsol:unsolicited-params-per-interface";
        description
          "Two interval values or one value used for both transmit
           and receive.  Defaults to
           ../../unsolicited/interval-config-type.  An interval
           configured under an interface takes precedence over any
           interval configured at the global level.";
        case tx-rx-intervals {
          leaf desired-min-tx-interval {
            type uint32;
            units "microseconds";
            description
              "Desired minimum transmit interval of control
               packets.";
          }
          leaf required-min-rx-interval {
            type uint32;
            units "microseconds";
            description
              "Required minimum receive interval of control
               packets.";
          }
        }
        case single-interval {
          if-feature "bfd-types:single-minimum-interval";
          leaf min-interval {
            type uint32;
            units "microseconds";
            description
              "Desired minimum transmit interval and required
               minimum receive interval of control packets.";
          }
        }
      }
    }
  }

  augment "/rt:routing/rt:control-plane-protocols/"
        + "rt:control-plane-protocol/bfd:bfd/bfd-ip-sh:ip-sh/"
        + "bfd-ip-sh:sessions/bfd-ip-sh:session" {
    description
      "Augmentation for unsolicited BFD on IP single-hop session.";
    leaf role {
      type identityref {
        base bfd-unsol:role;
      }
      config false;
      description
        "Role.";
    }
  }
}
<CODE ENDS>

4.3. Data Model Example

This section shows an example of how to configure the passive end of unsolicited BFD 🍓. Here's the setup:

  • we have global BFD IP single-hop unsolicited configuration with a local-multiplier of 2 and min-interval at 50 ms 🌸
  • BFD IP single-hop unsolicited is enabled on interface eth0 with a local-multiplier of 3 and min-interval at 250 ms 🦋
  • BFD IP single-hop unsolicited is enabled on interface eth1. Since there is no parameter configuration for eth1, it inherits from the global configuration 💅
<?xml version="1.0" encoding="UTF-8"?>
<config xmlns="urn:ietf:params:xml:ns:netconf:base:1.0">
<interfaces xmlns="urn:ietf:params:xml:ns:yang:ietf-interfaces">
  <interface>
    <name>eth0</name>
    <type
        xmlns:ianaift="urn:ietf:params:xml:ns:yang:iana-if-type">
         ianaift:ethernetCsmacd</type>
  </interface>
  <interface>
    <name>eth1</name>
    <type
        xmlns:ianaift="urn:ietf:params:xml:ns:yang:iana-if-type">
         ianaift:ethernetCsmacd</type>
  </interface>
</interfaces>
<routing xmlns="urn:ietf:params:xml:ns:yang:ietf-routing">
  <control-plane-protocols>
    <control-plane-protocol>
      <type xmlns:bfd-types=
        "urn:ietf:params:xml:ns:yang:ietf-bfd-types">
         bfd-types:bfdv1</type>
      <name>name:BFD</name>
      <bfd xmlns="urn:ietf:params:xml:ns:yang:ietf-bfd">
        <ip-sh xmlns="urn:ietf:params:xml:ns:yang:ietf-bfd-ip-sh">
          <unsolicited>
            <local-multiplier>2</local-multiplier>
            <min-interval>50000</min-interval>
          </unsolicited>
          <interfaces>
              <interface>eth0</interface>
              <unsolicited>
                <enabled>true</enabled>
                <local-multiplier>3</local-multiplier>
                <min-interval>250000</min-interval>
              </unsolicited>
          </interfaces>
          <interfaces>
              <interface>eth1</interface>
              <unsolicited>
                <enabled>true</enabled>
              </unsolicited>
          </interfaces>
        </ip-sh>
      </bfd>
    </control-plane-protocol>
  </control-plane-protocols>
</routing>
</config>

5. IANA Considerations 🌈

IANA went ahead and registered this namespace URI in the "ns" subregistry of the "IETF XML Registry" [RFC3688] 🌸:

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

IANA also registered this YANG module in the "YANG Module Names" registry [RFC6020] 📜:

Name:  ietf-bfd-unsolicited
Maintained by IANA:  N
Namespace:  urn:ietf:params:xml:ns:yang:ietf-bfd-unsolicited
Prefix:  bfd-unsol
Reference:  RFC 9468

6. Security Considerations 💀

6.1. BFD Protocol Security Considerations 🔐

okay bestie, the same security considerations and protection measures from [RFC5880] and [RFC5881] apply to this document too 💅. BUT with "unsolicited BFD" there's a potential risk for excessive resource usage by BFD from "unexpected" remote systems 😭. so to mitigate that, implementations of unsolicited BFD MUST:

  • Limit the feature to specific interfaces and to single-hop BFD sessions using the procedures from [RFC5082]. See Section 5 of [RFC5881] for the details of these procedures. 🧱
  • Apply policy to process BFD packets only from certain subnets or hosts. 🚧
  • Deploy the feature only in an environment that does not offer anonymous participation. Examples include an IXP, where the IXP operator will have a business relationship with all IXP participants, or between a provider and its customers. 🤝

6.2. BFD Protocol Authentication Considerations 🔑

ngl this one is lowkey important ✨ Implementations of unsolicited BFD are RECOMMENDED to use BFD authentication; see [RFC5880]. If BFD authentication is used, the strongest BFD authentication mechanism that is supported MUST be used. 🔒

in some environments, like IXPs, BFD authentication cannot be used because there's a lack of coordination for the operation of the two endpoints of the BFD session 🙈.

in other environments, like when BFD is used to track the next hop of static routes, it IS possible to use BFD authentication 🎀. the tradeoff is the extra cost of configuring matching key chains between the two endpoints. it's a lot of setup, but it's worth it 🫶.

6.3. YANG Module Security Considerations 🛡️

the YANG module in this document defines a schema for data that's designed to be accessed via network management protocols such as 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 Mode (NACM) [RFC8341] gives you the tools to restrict access for particular NETCONF or RESTCONF users to a preconfigured subset of all available NETCONF or RESTCONF protocol operations and content 👑.

there are a bunch of data nodes in this YANG module that are writable/creatable/deletable (i.e., config true, which is the default). these data nodes may be considered sensitive or vulnerable in some network environments 🙈. write operations (e.g., edit-config) to these data nodes without proper protection can have a negative effect on network operations 💀. here are the subtrees and data nodes, plus their sensitivity/vulnerability:

/routing/control-plane-protocols/control-plane-protocol/bfd/ip-sh
/unsolicited:
  • Data node enabled enables creation of unsolicited BFD IP single-hop sessions globally, i.e., on all interfaces. See Section 6.1. 🌍
  • Data nodes local-multiplier, desired-min-tx-interval, required-min-rx-interval, and min-interval all impact the parameters of the unsolicited BFD IP single-hop sessions. Write operations to these nodes change the rates of BFD packet generation and detection time of the failures of a BFD session. ⏱️
/routing/control-plane-protocols/control-plane-protocol/bfd/ip-sh
/interfaces/interface/unsolicited:
  • Data node enabled enables the creation of unsolicited BFD IP single-hop sessions on a specific interface. See Section 6.1. 🔌
  • Data nodes local-multiplier, desired-min-tx-interval, required-min-rx-interval, and min-interval all impact the parameters of the unsolicited BFD IP single-hop sessions on the interface. 🎛️

some of the readable data nodes in this YANG module may also be considered sensitive or vulnerable in some network environments 👀. so it's important to control read access (e.g., via get, get-config, or notification) to these data nodes. here are the subtrees and data nodes, plus their sensitivity/vulnerability:

/routing/control-plane-protocols/control-plane-protocol/bfd/ip-sh
/sessions/session/role:
  • Access to this information discloses the role of the local system in the creation of the unsolicited BFD session. 🕵️‍♀️

7. References 📚

7.1. Normative References

7.2. Informative References

Acknowledgments 🫶

the authors would like to thank Acee Lindem, Alvaro Retana, Dan Romascanu, Derek Atkins, Greg Mirsky, Gyan Mishra, Henning Rogge, Jeffrey Haas, John Scudder, Lars Eggert, Magnus Westerlund, Mahesh Jethanandani, Murray Kucherawy, Raj Chetan, Robert Wilton, Roman Danyliw, Tom Petch, and Zaheduzzaman Sarker for their reviews and valuable input 💖. the squad really showed up for this one 👑.

Authors' Addresses 📮

Enke Chen
Palo Alto Networks
3000 Tannery Way
Santa Clara, CA 95054
United States of America
Email: enchen@paloaltonetworks.com

Naiming Shen
Zededa
160 W Santa Clara Street
San Jose, CA 95113
United States of America
Email: naiming@zededa.com

Robert Raszuk
Arrcus
2077 Gateway Place
San Jose, CA 95110
United States of America
Email: robert@raszuk.net

Reshad Rahman
Equinix
Canada
Email: reshad@yahoo.com
📦 where this page comes from