Use of BGP-4 Multiprotocol Extensions for IPv6 Inter-Domain Routing (RFC2545)
Original Publication Date: 1999-Mar-01
Included in the Prior Art Database: 2019-Feb-11
Internet Society Requests For Comment (RFCs)
P. Marques: AUTHOR [+1]
BGP-4 Multiprotocol Extensions (BGP-MP) defines the format of two BGP attributes (MP_REACH_NLRI and MP_UNREACH_NLRI) that can be used to announce and withdraw the announcement of reachability information. This document defines how compliant systems should make use of those attributes for the purpose of conveying IPv6 routing information. [STANDARDS-TRACK]
Network Working Group P. Marques Request for Comments: 2545 cisco Systems, Inc. Category: Standards Track F. Dupont Inria March 1999
Use of BGP-4 Multiprotocol Extensions for IPv6 Inter-Domain Routing
Status of this Memo
This document specifies an Internet standards track protocol for the Internet community, and requests discussion and suggestions for improvements. Please refer to the current edition of the "Internet Official Protocol Standards" (STD 1) for the standardization state and status of this protocol. Distribution of this memo is unlimited.
Copyright (C) The Internet Society (1999). All Rights Reserved.
BGP-4 Multiprotocol Extensions [BGP-MP] defines the format of two BGP attributes (MP_REACH_NLRI and MP_UNREACH_NLRI) that can be used to announce and withdraw the announcement of reachability information. This document defines how compliant systems should make use of those attributes for the purpose of conveying IPv6 routing information.
The BGP-4 protocol [BGP-4] in particular, and path vector routing protocols in general, are mostly independent of the particular Address Family for which the protocol is being used.
IPv6 falls under the generic category of protocols for which BGP-4 is suitable and, unless stated otherwise in this document, the BGP-4 procedures to apply when using BGP-4 to carry IPv6 reachability information are those defined in [BGP-4] and in subsequent documents that extend or update the BGP-4 specification.
In terms of routing information, the most significant difference between IPv6 and IPv4 (for which BGP was originally designed) is the fact that IPv6 introduces scoped unicast addresses and defines particular situations when a particular address scope must be used. This document concerns itself essentially with the necessary rules to accommodate IPv6 address scope requirements.
Marques & Dupont Standards Track [Page 1]
RFC 2545 BGP-4 Multiprotocol Extensions for IPv6 IDR March 1999
2. IPv6 Address Scopes
IPv6 defines 3 unicast address scopes [ADDR-ARCH]: global, site-local and link-local. Site-local addresses are non-link-local address which are valid within the scope of a "site" and cannot be exported beyond it. As this document makes no assumption on the characteristics of a particular routing realm where BGP-4 is used, it makes no distinction between global and site-local addresses and refers to both as "global" or "non-link-local". Network administrators must however respect address scope restrictions and should be aware that the concepts of a BGP-4 routing domain and "site" are orthogonal notions and that they may or may not coincide in a given situation.
Companion IPv6 specifications further define that only link-local address can be used when generating ICMP Redirect Messages [ND] and as next hop addresses in some routing protocols (eg. RIPng [RIP]).
This restrictions does imply that an IPv6 router must have a link- local next hop address for all directly connected...