RFC 8815

Internet Engineering Task Force (IETF)                    M. Abrahamsson
Request for Comments: 8815
BCP: 229                                                        T. Chown
Category: Best Current Practice                                     Jisc
ISSN: 2070-1721                                              L. Giuliano
                                                  Juniper Networks, Inc.
                                                               T. Eckert
                                             Futurewei Technologies Inc.
                                                             August 2020

    Deprecating Any-Source Multicast (ASM) for Interdomain Multicast


   This document recommends deprecation of the use of Any-Source
   Multicast (ASM) for interdomain multicast.  It recommends the use of
   Source-Specific Multicast (SSM) for interdomain multicast
   applications and recommends that hosts and routers in these
   deployments fully support SSM.  The recommendations in this document
   do not preclude the continued use of ASM within a single organization
   or domain and are especially easy to adopt in existing deployments of
   intradomain ASM using PIM Sparse Mode (PIM-SM).

Status of This Memo

   This memo documents an Internet Best Current Practice.

   This document is a product of the Internet Engineering Task Force
   (IETF).  It represents the consensus of the IETF community.  It has
   received public review and has been approved for publication by the
   Internet Engineering Steering Group (IESG).  Further information on
   BCPs is available in Section 2 of RFC 7841.

   Information about the current status of this document, any errata,
   and how to provide feedback on it may be obtained at

Copyright Notice

   Copyright (c) 2020 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, as they describe your rights and restrictions with respect
   to this document.  Code Components extracted from this document must
   include Simplified BSD License text as described in Section 4.e of
   the Trust Legal Provisions and are provided without warranty as
   described in the Simplified BSD License.

Table of Contents

   1.  Introduction
   2.  Background
     2.1.  Multicast Service Models
     2.2.  ASM Routing Protocols
       2.2.1.  PIM Sparse Mode (PIM-SM)
       2.2.2.  Embedded-RP
       2.2.3.  BIDIR-RP
     2.3.  SSM Routing Protocols
   3.  Discussion
     3.1.  Observations on ASM and SSM Deployments
     3.2.  Advantages of SSM for Interdomain Multicast
       3.2.1.  Reduced Network Operations Complexity
       3.2.2.  No Network-Wide IP Multicast Group-Address Management
       3.2.3.  Intrinsic Source-Control Security
   4.  Recommendations
     4.1.  Deprecating Use of ASM for Interdomain Multicast
     4.2.  Including Network Support for IGMPv3/MLDv2
     4.3.  Building Application Support for SSM
     4.4.  Developing Application Guidance: SSM, ASM, Service
     4.5.  Preferring SSM Applications Intradomain
     4.6.  Documenting an ASM/SSM Protocol Mapping Mechanism
     4.7.  Not Filtering ASM Addressing between Domains
     4.8.  Not Precluding Intradomain ASM
     4.9.  Evolving PIM Deployments for SSM
   5.  Future Interdomain ASM Work
   6.  Security Considerations
   7.  IANA Considerations
   8.  References
     8.1.  Normative References
     8.2.  Informative References

   Authors' Addresses

1.  Introduction

   IP Multicast has been deployed in various forms, within private
   networks, the wider Internet, and federated networks such as national
   or regional research networks.  While a number of service models have
   been published, and in many cases revised over time, there has been
   no strong recommendation made by the IETF on the appropriateness of
   those models to certain scenarios, even though vendors and
   federations have often made such recommendations.

   This document addresses this gap by making a BCP-level recommendation
   to deprecate the use of Any-Source Multicast (ASM) for interdomain
   multicast, leaving Source-Specific Multicast (SSM) as the recommended
   interdomain mode of multicast.  Therefore, this document recommends
   that all hosts and routers that support interdomain multicast
   applications fully support SSM.

   This document does not make any statement on the use of ASM within a
   single domain or organization and, therefore, does not preclude its
   use.  Indeed, there are application contexts for which ASM is
   currently still widely considered well suited within a single domain.

   The main issue in most cases with moving to SSM is application
   support.  Many applications are initially deployed for intradomain
   use and are later deployed interdomain.  Therefore, this document
   recommends that applications support SSM, even when they are
   initially intended for intradomain use.  As explained below, SSM
   applications are readily compatible with existing intradomain ASM
   deployments using PIM-SM, as PIM-SSM is merely a subset of PIM-SM.

2.  Background

2.1.  Multicast Service Models

   Any-Source Multicast (ASM) and Source-Specific Multicast (SSM) are
   the two multicast service models in use today.  In ASM, as originally
   described in [RFC1112], receivers express interest in joining a
   multicast group address, and routers use multicast routing protocols
   to deliver traffic from the sender(s) to the receivers.  If there are
   multiple senders for a given group, traffic from all senders will be
   delivered to the receivers.  Since receivers specify only the group
   address, the network -- and therefore the multicast routing protocols
   -- are responsible for source discovery.

   In SSM, by contrast, receivers specify both group and source when
   expressing interest in joining a multicast stream.  Source discovery
   in SSM is handled by some out-of-band mechanism (typically in the
   application layer), which drastically simplifies the network and how
   the multicast routing protocols operate.

   IANA has reserved specific ranges of IPv4 and IPv6 address space for
   multicast addressing.  Guidelines for IPv4 multicast address
   assignments can be found in [RFC5771], while guidelines for IPv6
   multicast address assignments can be found in [RFC2375] and
   [RFC3307].  The IPv6 multicast address format is described in

2.2.  ASM Routing Protocols

2.2.1.  PIM Sparse Mode (PIM-SM)

   The most commonly deployed ASM routing protocol is Protocol
   Independent Multicast - Sparse Mode (PIM-SM), as detailed in
   [RFC7761].  PIM-SM, as the name suggests, was designed to be used in
   scenarios where the subnets with receivers are sparsely distributed
   throughout the network.  Because receivers do not indicate sender
   addresses in ASM (but only group addresses), PIM-SM uses the concept
   of a Rendezvous Point (RP) as a "meeting point" for sources and
   receivers, and all routers in a PIM-SM domain are configured to use a
   specific RP(s), either explicitly or through dynamic RP-discovery

   To enable PIM-SM to work between multiple domains, an interdomain,
   inter-RP signaling protocol known as Multicast Source Discovery
   Protocol (MSDP) [RFC3618] is used to allow an RP in one domain to
   learn of the existence of a source in another domain.  Deployment
   scenarios for MSDP are given in [RFC4611].  MSDP floods information
   about all active sources for all multicast streams to all RPs in all
   the domains -- even if there is no receiver for a given application
   in a domain.  As a result of this key scalability and security issue,
   along with other deployment challenges with the protocol, MSDP was
   never extended to support IPv6 and remains an Experimental protocol.

   At the time of writing, there is no IETF interdomain solution at the
   level of Proposed Standard for IPv4 ASM multicast, because MSDP was
   the de facto mechanism for the interdomain source discovery problem,
   and it is Experimental.  Other protocol options were investigated at
   the same time but were never implemented or deployed and are now
   historic (e.g., [RFC3913]).

2.2.2.  Embedded-RP

   Due to the availability of more bits in an IPv6 address than in IPv4,
   an IPv6-specific mechanism was designed in support of interdomain
   ASM, with PIM-SM leveraging those bits.  Embedded-RP [RFC3956] allows
   routers supporting the protocol to determine the RP for the group
   without any prior configuration or discovery protocols, simply by
   observing the unicast RP address that is embedded (included) in the
   IPv6 multicast group address.  Embedded-RP allows PIM-SM operation
   across any IPv6 network in which there is an end-to-end path of
   routers supporting this mechanism, including interdomain deployment.

2.2.3.  BIDIR-RP

   BIDIR-PIM [RFC5015] is another protocol to support ASM.  There is no
   standardized option to operate BIDIR-PIM interdomain.  It is deployed
   intradomain for applications where many sources send traffic to the
   same IP multicast groups because, unlike PIM-SM, it does not create
   per-source state.  BIDIR-PIM is one of the important reasons for this
   document to not deprecate intradomain ASM.

2.3.  SSM Routing Protocols

   SSM is detailed in [RFC4607].  It mandates the use of PIM-SSM for
   routing of SSM.  PIM-SSM is merely a subset of PIM-SM [RFC7761].

   PIM-SSM expects the sender's source address(es) to be known in
   advance by receivers through some out-of-band mechanism (typically in
   the application layer); thus, the receiver's designated router can
   send a PIM Join message directly towards the source without needing
   to use an RP.

   IPv4 addresses in the 232/8 ( to range are
   designated as Source-Specific Multicast (SSM) destination addresses
   and are reserved for use by source-specific applications and
   protocols.  For IPv6, the address prefix ff3x::/32 is reserved for
   source-specific multicast use.  See [RFC4607].

3.  Discussion

3.1.  Observations on ASM and SSM Deployments

   In enterprise and campus scenarios, ASM in the form of PIM-SM is
   likely the most commonly deployed multicast protocol.  The
   configuration and management of an RP (including RP redundancy)
   within a single domain is a well-understood operational practice.
   However, if interworking with external PIM domains is needed in IPv4
   multicast deployments, interdomain MSDP is required to exchange
   information about sources between domain RPs.  Deployment experience
   has shown MSDP to be a complex and fragile protocol to manage and
   troubleshoot.  Some of these issues include complex Reverse Path
   Forwarding (RPF) rules, state attack protection, and filtering of
   undesired sources.

   PIM-SM is a general-purpose protocol that can handle all use cases.
   In particular, it was designed for cases such as videoconferencing
   where multiple sources may come and go during a multicast session.
   But for cases where a single, persistent source for a group is used,
   and receivers can be configured to know of that source, PIM-SM has
   unnecessary complexity.  Therefore, SSM removes the need for many of
   the most complex components of PIM-SM.

   As explained above, MSDP was not extended to support IPv6.  Instead,
   the proposed interdomain ASM solution for PIM-SM with IPv6 is
   Embedded-RP, which allows the RP address for a multicast group to be
   embedded in the group address, making RP discovery automatic for all
   routers on the path between a receiver and a sender.  Embedded-RP can
   support lightweight ad hoc deployments.  However, it relies on a
   single RP for an entire group that could only be made resilient
   within one domain.  While this approach solves the MSDP issues, it
   does not solve the problem of unauthorized sources sending traffic to
   ASM multicast groups; this security issue is one of biggest problems
   of interdomain multicast.

   As stated in RFC 4607, SSM is particularly well suited to either
   dissemination-style applications with one or more senders whose
   identities are known (by some out-of-band mechanism) before the
   application starts running or applications that utilize some
   signaling to indicate the source address of the multicast stream
   (e.g., an electronic programming guide in IPTV applications).
   Therefore, SSM through PIM-SSM is very well suited to applications
   such as classic linear-broadcast TV over IP.

   SSM requires applications, host operating systems, and the designated
   routers connected to receiving hosts to support Internet Group
   Management Protocol, Version 3 (IGMPv3) [RFC3376] and Multicast
   Listener Discovery, Version 2 (MLDv2) [RFC3810].  While support for
   IGMPv3 and MLDv2 has been commonplace in routing platforms for a long
   time, it has also now become widespread in common operating systems
   for several years (Windows, Mac OS, Linux/Android) and is no longer
   an impediment to SSM deployment.

3.2.  Advantages of SSM for Interdomain Multicast

   This section describes the three key benefits that SSM with PIM-SSM
   has over ASM.  These benefits also apply to intradomain deployment
   but are even more important in interdomain deployments.  See
   [RFC4607] for more details.

3.2.1.  Reduced Network Operations Complexity

   A significant benefit of SSM is the reduced complexity that comes
   through eliminating the network-based source discovery required in
   ASM with PIM-SM.  Specifically, SSM eliminates the need for RPs,
   shared trees, Shortest Path Tree (SPT) switchovers, PIM registers,
   MSDP, dynamic RP-discovery mechanisms (Bootstrap Router (BSR) /
   AutoRP), and data-driven state creation.  SSM simply utilizes a small
   subset of PIM-SM, alongside the integration with IGMPv3/MLDv2, where
   the source address signaled from the receiver is immediately used to
   create (S,G) state.  Eliminating network-based source discovery for
   interdomain multicast means the vast majority of the complexity of
   multicast goes away.

   This reduced complexity makes SSM radically simpler to manage,
   troubleshoot, and operate, particularly for backbone network
   operators.  This is the main operator motivation for the
   recommendation to deprecate the use of ASM in interdomain scenarios.

   Note that this discussion does not apply to BIDIR-PIM, and there is
   (as mentioned above) no standardized interdomain solution for BIDIR-
   PIM.  In BIDIR-PIM, traffic is forwarded to the RP instead of
   building state as in PIM-SM.  This occurs even in the absence of
   receivers.  Therefore, BIDIR-PIM offers a trade-off of state
   complexity at the cost of creating unnecessary traffic (potentially a
   large amount).

3.2.2.  No Network-Wide IP Multicast Group-Address Management

   In ASM, IP multicast group addresses need to be assigned to
   applications and instances thereof, so that two simultaneously active
   application instances will not share the same group address and
   receive IP multicast traffic from each other.

   In SSM, no such IP multicast group management is necessary.  Instead,
   the IP multicast group address simply needs to be assigned locally on
   a source like a unicast transport protocol port number: the only
   coordination required is to ensure that different applications
   running on the same host don't send to the same group address.  This
   does not require any network-operator involvement.

3.2.3.  Intrinsic Source-Control Security

   SSM is implicitly secure against off-path unauthorized/undesired
   sources.  Receivers only receive packets from the sources they
   explicitly specify in their IGMPv3/MLDv2 membership messages, as
   opposed to ASM, where any host can send traffic to a group address
   and have it transmitted to all receivers.  With PIM-SSM, traffic from
   sources not requested by any receiver will be discarded by the First-
   Hop Router (FHR) of that source, minimizing source attacks against
   shared network bandwidth and receivers.

   This benefit is particularly important in interdomain deployments
   because there are no standardized solutions for ASM control of
   sources and the most common intradomain operational practices such as
   Access Control Lists (ACLs) on the sender's FHR are not feasible for
   interdomain deployments.

   This topic is expanded upon in [RFC4609].

4.  Recommendations

   This section provides recommendations for a variety of stakeholders
   in SSM deployment, including vendors, operators, and application
   developers.  It also suggests further work that could be undertaken
   within the IETF.

4.1.  Deprecating Use of ASM for Interdomain Multicast

   This document recommends that the use of ASM be deprecated for
   interdomain multicast; thus, implicitly, it recommends that hosts and
   routers that support such interdomain applications fully support SSM
   and its associated protocols.  Best current practices for deploying
   interdomain multicast using SSM are documented in [RFC8313].

   The recommendation applies to the use of ASM between domains where
   either MSDP (IPv4) or Embedded-RP (IPv6) is used.

   An interdomain use of ASM multicast in the context of this document
   is one where PIM-SM with RPs/MSDP/Embedded-RP is run on routers
   operated by two or more separate administrative entities.

   The focus of this document is deprecation of interdomain ASM
   multicast, and while encouraging the use of SSM within domains, it
   leaves operators free to choose to use ASM within their own domains.
   A more inclusive interpretation of this recommendation is that it
   also extends to deprecating use of ASM in the case where PIM is
   operated in a single operator domain, but where user hosts or non-PIM
   network edge devices are under different operator control.  A typical
   example of this case is a service provider offering IPTV (single
   operator domain for PIM) to subscribers operating an IGMP proxy home
   gateway and IGMPv3/MLDv2 hosts (computer, tablets, set-top boxes).

4.2.  Including Network Support for IGMPv3/MLDv2

   This document recommends that all hosts, router platforms, and
   security appliances used for deploying multicast support the
   components of IGMPv3 [RFC3376] and MLDv2 [RFC3810] necessary to
   support SSM (i.e., explicitly sending source-specific reports).
   "IPv6 Node Requirements" [RFC8504] states that MLDv2 must be
   supported in all implementations.  Such support is already widespread
   in common host and router platforms.

   Further guidance on IGMPv3 and MLDv2 is given in [RFC4604].

   Multicast snooping is often used to limit the flooding of multicast
   traffic in a Layer 2 network.  With snooping, an L2 switch will
   monitor IGMP/MLD messages and only forward multicast traffic out on
   host ports that have interested receivers connected.  Such snooping
   capability should therefore support IGMPv3 and MLDv2.  There is
   further discussion in [RFC4541].

4.3.  Building Application Support for SSM

   The recommendation to use SSM for interdomain multicast means that
   applications should properly trigger the sending of IGMPv3/MLDv2
   source-specific report messages.  It should be noted, however, that
   there is a wide range of applications today that only support ASM.
   In many cases, this is due to application developers being unaware of
   the operational concerns of networks and the implications of using
   ASM versus SSM.  This document serves to provide clear direction for
   application developers who might currently only consider using ASM to
   instead support SSM, which only requires relatively minor changes for
   many applications, particularly those with single sources.

   It is often thought that ASM is required for multicast applications
   where there are multiple sources.  However, RFC 4607 also describes
   how SSM can be used instead of PIM-SM for multi-party applications:

   |  SSM can be used to build multi-source applications where all
   |  participants' identities are not known in advance, but the multi-
   |  source "rendezvous" functionality does not occur in the network
   |  layer in this case.  Just like in an application that uses unicast
   |  as the underlying transport, this functionality can be implemented
   |  by the application or by an application-layer library.

   Some useful considerations for multicast applications can be found in

4.4.  Developing Application Guidance: SSM, ASM, Service Discovery

   Applications with many-to-many communication patterns can create more
   (S,G) state than is feasible for networks to manage, whether the
   source discovery is done by ASM with PIM-SM or at the application
   level and SSM/PIM-SSM.  These applications are not best supported by
   either SSM/PIM-SSM or ASM/PIM-SM.

   Instead, these applications are better served by routing protocols
   that do not create (S,G), such as BIDIR-PIM.  Unfortunately, many
   applications today use ASM solely for service discovery.  One example
   is where clients send IP multicast packets to elicit unicast replies
   from server(s).  Deploying any form of IP multicast solely in support
   of such service discovery is, in general, not recommended.  Dedicated
   service discovery via DNS-based Service Discovery (DNS-SD) [RFC6763]
   should be used for this instead.

   This document describes best practices to explain when to use SSM in
   applications -- e.g., when ASM without (S,G) state in the network is
   better, or when dedicated service-discovery mechanisms should be
   used.  However, specifying how applications can support these
   practices is outside the scope of this document.  Further work on
   this subject may be expected within the IETF.

4.5.  Preferring SSM Applications Intradomain

   If feasible, it is recommended for applications to use SSM even if
   they are initially only meant to be used in intradomain environments
   supporting ASM.  Because PIM-SSM is a subset of PIM-SM, existing
   intradomain PIM-SM networks are automatically compatible with SSM
   applications.  Thus, SSM applications can operate alongside existing
   ASM applications.  SSM's benefits of simplified address management
   and significantly reduced operational complexity apply equally to
   intradomain use.

   However, for some applications, it may be prohibitively difficult to
   add support for source discovery, so intradomain ASM may still be

4.6.  Documenting an ASM/SSM Protocol Mapping Mechanism

   In the case of existing ASM applications that cannot readily be
   ported to SSM, it may be possible to use some form of protocol
   mapping -- i.e., to have a mechanism to translate a (*,G) join or
   leave to a (S,G) join or leave for a specific source S.  The general
   challenge in performing such mapping is determining where the
   configured source address, S, comes from.

   There are existing vendor-specific mechanisms deployed that achieve
   this function, but none are documented in IETF documents.  This may
   be a useful area for the IETF to work on as an interim transition
   mechanism.  However, these mechanisms would introduce additional
   administrative burdens, along with the need for some form of address
   management, neither of which are required in SSM.  Hence, this should
   not be considered a long-term solution.

4.7.  Not Filtering ASM Addressing between Domains

   A key benefit of SSM is that the receiver specifies the source-group
   tuple when signaling interest in a multicast stream.  Hence, the
   group address need not be globally unique, so there is no need for
   multicast address allocation as long the reserved SSM range is used.

   Despite the deprecation of interdomain ASM, it is recommended that
   operators not filter ASM group ranges at domain boundaries, as some
   form of ASM-SSM mappings may continue to be used for some time.

4.8.  Not Precluding Intradomain ASM

   The use of ASM within a single multicast domain such as a campus or
   enterprise is still relatively common today.  There are even global
   enterprise networks that have successfully been using PIM-SM for many
   years.  The operators of such networks most often use Anycast-RP
   [RFC4610] or MSDP (with IPv4) for RP resilience, at the expense of
   the extra operational complexity.  These existing practices are
   unaffected by this document.

   In the past decade, some BIDIR-PIM deployments have scaled
   interdomain ASM deployments beyond the capabilities of PIM-SM.  This,
   too, is unaffected by this document; instead, it is encouraged where
   necessary due to application requirements (see Section 4.4).

   This document also does not preclude continued use of ASM with
   multiple PIM-SM domains inside organizations, such as with IPv4 MSDP
   or IPv6 Embedded-RP.  This includes organizations that are
   federations and have appropriate, nonstandardized mechanisms to deal
   with the interdomain ASM issues explained in Section 3.2.

4.9.  Evolving PIM Deployments for SSM

   Existing PIM-SM deployments can usually be used to run SSM
   applications with few-to-no changes.  In some widely available router
   implementations of PIM-SM, PIM-SSM is simply enabled by default in
   the designated SSM address spaces whenever PIM-SM is enabled.  In
   other implementations, simple configuration options exist to enable
   it.  This allows migration of ASM applications to SSM/PIM-SSM solely
   through application-side development to handle source-signaling via
   IGMPv3/MLDv2 and using SSM addresses.  No network actions are
   required for this transition; unchanged ASM applications can continue
   to coexist without issues.

   When running PIM-SM, IGMPv3/MLDv2 (S,G) membership reports may also
   result in the desired PIM-SSM (S,G) operations and bypass any RP
   procedures.  This is not standardized but depends on implementation
   and may require additional configuration in available products.  In
   general, it is recommended to always use SSM address space for SSM
   applications.  For example, the interaction of IGMPv3/MLDv2 (S,G)
   membership reports and BIDIR-PIM is undefined and may not result in
   forwarding of any traffic.

   Note that these migration recommendations do not include
   considerations on when or how to evolve those intradomain
   applications best served by ASM/BIDIR-PIM from PIM-SM to BIDIR-PIM.
   This may also be important but is outside the scope of this document.

5.  Future Interdomain ASM Work

   Future work may attempt to overcome current limitations of ASM
   solutions, such as interdomain deployment solutions for BIDIR-PIM or
   source-access-control mechanisms for IPv6 PIM-SM with embedded-RP.
   Such work could modify or amend the recommendations of this document
   (like any future IETF Standards Track / BCP work).

   Nevertheless, it is very unlikely that any ASM solution, even with
   such future work, can ever provide the same intrinsic security and
   network- and address-management simplicity as SSM (see Section 3.2).
   Accordingly, this document recommends that future work for general-
   purpose interdomain IP multicast focus on SSM items listed in
   Section 4.

6.  Security Considerations

   This document adds no new security considerations.  It instead
   removes security issues incurred by interdomain ASM with PIM-SM/MSDP,
   such as infrastructure control-plane attacks and application and
   bandwidth/congestion attacks from unauthorized sources sending to ASM
   multicast groups.  RFC 4609 describes the additional security
   benefits of using SSM instead of ASM.

7.  IANA Considerations

   This document has no IANA actions.

8.  References

8.1.  Normative References

   [RFC1112]  Deering, S., "Host extensions for IP multicasting", STD 5,
              RFC 1112, DOI 10.17487/RFC1112, August 1989,

   [RFC3307]  Haberman, B., "Allocation Guidelines for IPv6 Multicast
              Addresses", RFC 3307, DOI 10.17487/RFC3307, August 2002,

   [RFC3376]  Cain, B., Deering, S., Kouvelas, I., Fenner, B., and A.
              Thyagarajan, "Internet Group Management Protocol, Version
              3", RFC 3376, DOI 10.17487/RFC3376, October 2002,

   [RFC3810]  Vida, R., Ed. and L. Costa, Ed., "Multicast Listener
              Discovery Version 2 (MLDv2) for IPv6", RFC 3810,
              DOI 10.17487/RFC3810, June 2004,

   [RFC3956]  Savola, P. and B. Haberman, "Embedding the Rendezvous
              Point (RP) Address in an IPv6 Multicast Address",
              RFC 3956, DOI 10.17487/RFC3956, November 2004,

   [RFC4291]  Hinden, R. and S. Deering, "IP Version 6 Addressing
              Architecture", RFC 4291, DOI 10.17487/RFC4291, February
              2006, <https://www.rfc-editor.org/info/rfc4291>.

   [RFC4607]  Holbrook, H. and B. Cain, "Source-Specific Multicast for
              IP", RFC 4607, DOI 10.17487/RFC4607, August 2006,

   [RFC5771]  Cotton, M., Vegoda, L., and D. Meyer, "IANA Guidelines for
              IPv4 Multicast Address Assignments", BCP 51, RFC 5771,
              DOI 10.17487/RFC5771, March 2010,

   [RFC7761]  Fenner, B., Handley, M., Holbrook, H., Kouvelas, I.,
              Parekh, R., Zhang, Z., and L. Zheng, "Protocol Independent
              Multicast - Sparse Mode (PIM-SM): Protocol Specification
              (Revised)", STD 83, RFC 7761, DOI 10.17487/RFC7761, March
              2016, <https://www.rfc-editor.org/info/rfc7761>.

   [RFC8313]  Tarapore, P., Ed., Sayko, R., Shepherd, G., Eckert, T.,
              Ed., and R. Krishnan, "Use of Multicast across Inter-
              domain Peering Points", BCP 213, RFC 8313,
              DOI 10.17487/RFC8313, January 2018,

8.2.  Informative References

   [RFC2375]  Hinden, R. and S. Deering, "IPv6 Multicast Address
              Assignments", RFC 2375, DOI 10.17487/RFC2375, July 1998,

   [RFC3170]  Quinn, B. and K. Almeroth, "IP Multicast Applications:
              Challenges and Solutions", RFC 3170, DOI 10.17487/RFC3170,
              September 2001, <https://www.rfc-editor.org/info/rfc3170>.

   [RFC3618]  Fenner, B., Ed. and D. Meyer, Ed., "Multicast Source
              Discovery Protocol (MSDP)", RFC 3618,
              DOI 10.17487/RFC3618, October 2003,

   [RFC3913]  Thaler, D., "Border Gateway Multicast Protocol (BGMP):
              Protocol Specification", RFC 3913, DOI 10.17487/RFC3913,
              September 2004, <https://www.rfc-editor.org/info/rfc3913>.

   [RFC4541]  Christensen, M., Kimball, K., and F. Solensky,
              "Considerations for Internet Group Management Protocol
              (IGMP) and Multicast Listener Discovery (MLD) Snooping
              Switches", RFC 4541, DOI 10.17487/RFC4541, May 2006,

   [RFC4604]  Holbrook, H., Cain, B., and B. Haberman, "Using Internet
              Group Management Protocol Version 3 (IGMPv3) and Multicast
              Listener Discovery Protocol Version 2 (MLDv2) for Source-
              Specific Multicast", RFC 4604, DOI 10.17487/RFC4604,
              August 2006, <https://www.rfc-editor.org/info/rfc4604>.

   [RFC4609]  Savola, P., Lehtonen, R., and D. Meyer, "Protocol
              Independent Multicast - Sparse Mode (PIM-SM) Multicast
              Routing Security Issues and Enhancements", RFC 4609,
              DOI 10.17487/RFC4609, October 2006,

   [RFC4610]  Farinacci, D. and Y. Cai, "Anycast-RP Using Protocol
              Independent Multicast (PIM)", RFC 4610,
              DOI 10.17487/RFC4610, August 2006,

   [RFC4611]  McBride, M., Meylor, J., and D. Meyer, "Multicast Source
              Discovery Protocol (MSDP) Deployment Scenarios", BCP 121,
              RFC 4611, DOI 10.17487/RFC4611, August 2006,

   [RFC5015]  Handley, M., Kouvelas, I., Speakman, T., and L. Vicisano,
              "Bidirectional Protocol Independent Multicast (BIDIR-
              PIM)", RFC 5015, DOI 10.17487/RFC5015, October 2007,

   [RFC6763]  Cheshire, S. and M. Krochmal, "DNS-Based Service
              Discovery", RFC 6763, DOI 10.17487/RFC6763, February 2013,

   [RFC8504]  Chown, T., Loughney, J., and T. Winters, "IPv6 Node
              Requirements", BCP 220, RFC 8504, DOI 10.17487/RFC8504,
              January 2019, <https://www.rfc-editor.org/info/rfc8504>.


   The authors would like to thank members of the IETF MBONE Deployment
   Working Group for discussions on the content of this document, with
   specific thanks to the following people for their contributions to
   the document: Hitoshi Asaeda, Dale Carder, Jake Holland, Albert
   Manfredi, Mike McBride, Per Nihlen, Greg Shepherd, James Stevens,
   Stig Venaas, Nils Warnke, and Sandy Zhang.

Authors' Addresses

   Mikael Abrahamsson

   Email: swmike@swm.pp.se

   Tim Chown
   Harwell Oxford
   Lumen House, Library Avenue
   OX11 0SG
   United Kingdom

   Email: tim.chown@jisc.ac.uk

   Lenny Giuliano
   Juniper Networks, Inc.
   2251 Corporate Park Drive
   Herndon, Virginia 20171
   United States of America

   Email: lenny@juniper.net

   Toerless Eckert
   Futurewei Technologies Inc.
   2330 Central Expy
   Santa Clara, California 95050
   United States of America

   Email: tte@cs.fau.de