Request for Comments: 4659 Alcatel
Category: Standards Track D. Ooms
OneSparrow
M. Carugi
Nortel Networks
F. Le Faucheur
Cisco Systems
September 2006
BGP-MPLS IP Virtual Private Network (VPN) Extension for IPv6 VPN
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 Notice
Copyright (C) The Internet Society (2006).
Abstract
This document describes a method by which a Service Provider may use
its packet-switched backbone to provide Virtual Private Network (VPN)
services for its IPv6 customers. This method reuses, and extends
where necessary, the "BGP/MPLS IP VPN" method for support of IPv6.
In BGP/MPLS IP VPN, "Multiprotocol BGP" is used for distributing IPv4
VPN routes over the service provider backbone, and MPLS is used to
forward IPv4 VPN packets over the backbone. This document defines an
IPv6 VPN address family and describes the corresponding IPv6 VPN
route distribution in "Multiprotocol BGP".
This document defines support of the IPv6 VPN service over both an
IPv4 and an IPv6 backbone, and for using various tunneling techniques
over the core, including MPLS, IP-in-IP, Generic Routing
Encapsulation (GRE) and IPsec protected tunnels. The inter-working
between an IPv4 site and an IPv6 site is outside the scope of this
document.
Table of Contents
1. Introduction ....................................................2
2. The VPN-IPv6 Address Family .....................................4
3. VPN-IPv6 Route Distribution .....................................5
3.1. Route Distribution Among PEs by BGP ........................5
3.2. VPN IPv6 NLRI Encoding .....................................6
3.2.1. BGP Next Hop encoding ...............................6
3.2.1.1. BGP Speaker Requesting IPv6 Transport ......7
3.2.1.2. BGP Speaker Requesting IPv4 Transport ......8
3.3. Route Target ...............................................8
3.4. BGP Capability Negotiation .................................8
4. Encapsulation ...................................................8
5. Address Types ..................................................10
6. Multicast ......................................................11
7. Carriers’ Carriers .............................................11
8. Multi-AS Backbones .............................................11
9. Accessing the Internet from a VPN ..............................13
10. Management VPN ................................................14
11. Security Considerations .......................................14
12. Quality of Service ............................................15
13. Scalability ...................................................15
14. IANA Considerations ...........................................15
15. Acknowledgements ..............................................15
16. References ....................................................16
16.1. Normative References .....................................16
16.2. Informative References ...................................16
1. Introduction
This document describes a method by which a Service Provider may use
its packet-switched backbone to provide Virtual Private Network
services for its IPv6 customers.
This method reuses, and extends where necessary, the "BGP/MPLS IP
VPN" method [BGP/MPLS-VPN] for support of IPv6. In particular, this
method uses the same "peer model" as [BGP/MPLS-VPN], in which the
customers’ edge routers ("CE routers") send their IPv6 routes to the
Service Provider’s edge routers ("PE routers"). BGP ("Border Gateway
Protocol", [BGP, BGP-MP]) is then used by the Service Provider to
exchange the routes of a particular IPv6 VPN among the PE routers
that are attached to that IPv6 VPN. Eventually, the PE routers
distribute, to the CE routers in a particular VPN, the IPv6 routes
from other CE routers in that VPN. As with IPv4 VPNs, a key
characteristic of this "peer model" is that the (IPv6) CE routers
within an (IPv6) VPN do not peer with each other; there is no
"overlay" visible to the (IPv6) VPN’s routing algorithm.
This document adopts the definitions, acronyms, and mechanisms
described in [BGP/MPLS-VPN]. Unless it is stated otherwise, the
mechanisms of [BGP/MPLS-VPN] apply and will not be re-described here.
A VPN is said to be an IPv6 VPN when each site of this VPN is IPv6
capable and is natively connected over an IPv6 interface or sub-
interface to the Service Provider (SP) backbone via a Provider Edge
device (PE).
A site may be both IPv4 capable and IPv6 capable. The logical
interface on which packets arrive at the PE may determine the IP
version. Alternatively, the same logical interface may be used for
both IPv4 and IPv6, in which case a per-packet lookup at the Version
field of the IP packet header determines the IP version.
This document only concerns itself with handling of IPv6
communication between IPv6 hosts located on IPv6-capable sites.
Handling of IPv4 communication between IPv4 hosts located on IPv4-
capable sites is outside the scope of this document and is covered in
[BGP/MPLS-VPN]. Communication between an IPv4 host located in an
IPv4- capable site and an IPv6 host located in an IPv6-capable site
is outside the scope of this document.
In a similar manner to how IPv4 VPN routes are distributed in
[BGP/MPLS-VPN], BGP and its extensions are used to distribute routes
from an IPv6 VPN site to all the other PE routers connected to a site
of the same IPv6 VPN. PEs use "VPN Routing and Forwarding tables"
(VRFs) to maintain the reachability information and forwarding
information of each IPv6 VPN separately.
As is done for IPv4 VPNs [BGP/MPLS-VPN], we allow each IPv6 VPN to
have its own IPv6 address space, which means that a given address may
denote different systems in different VPNs. This is achieved via a
new address family, the VPN-IPv6 Address Family, in a fashion similar
to that of the VPN-IPv4 address family, defined in [BGP/MPLS-VPN],
which prepends a Route Distinguisher to the IP address.
In addition to its operation over MPLS Label Switched Paths (LSPs),
the IPv4 BGP/MPLS VPN solution has been extended to allow operation
over other tunneling techniques, including GRE tunnels, IP-in-IP
tunnels [2547-GRE/IP], L2TPv3 tunnels [MPLS-in-L2TPv3], and IPsec
protected tunnels [2547-IPsec]. In a similar manner, this document
allows support of an IPv6 VPN service over MPLS LSPs, as well as over
other tunneling techniques.
This document allows support for an IPv6 VPN service over an IPv4
backbone, as well as over an IPv6 backbone. The IPv6 VPN service
supported is identical in both cases.
The IPv6 VPN solution defined in this document offers the following
benefits:
o From both the Service Provider perspective and the customer
perspective, the VPN service that can be supported for IPv6
sites is identical to the one that can be supported for IPv4
sites.
o From the Service Provider perspective, operations of the IPv6
VPN service require the exact same skills, procedures, and
mechanisms as those for the IPv4 VPN service.
o Where both IPv4 VPNs and IPv6 VPN services are supported over an
IPv4 core, the same single set of MP-BGP peering relationships
and the same single PE-PE tunnel mesh MAY be used for both.
o The IPv6 VPN service is independent of whether the core runs
IPv4 or IPv6. This is so that the IPv6 VPN service supported
before and after a migration of the core from IPv4 to IPv6 is
undistinguishable to the VPN customer.
Note that supporting IPv4 VPN services over an IPv6 core is not
covered by this document.
2. The VPN-IPv6 Address Family
The BGP Multiprotocol Extensions [BGP-MP] allow BGP to carry routes
from multiple "address families". We introduce the notion of the
"VPN-IPv6 address family", which is similar to the VPN-IPv4 address
family introduced in [BGP/MPLS-VPN].
A VPN-IPv6 address is a 24-octet quantity, beginning with an 8-octet
"Route Distinguisher" (RD) and ending with a 16-octet IPv6 address.
The purpose of the RD is solely to allow one to create distinct
routes to a common IPv6 address prefix, which is similar to the
purpose of the RD defined in [BGP/MPLS-VPN]. In the same way as it
is possible per [BGP/MPLS-VPN], the RD can be used to create multiple
different routes to the very same system. This can be achieved by
creating two different VPN-IPv6 routes that have the same IPv6 part
but different RDs. This allows the provider’s BGP to install
multiple different routes to the same system and allows policy to be
used to decide which packets use which route.
Also, if two VPNs were to use the same IPv6 address prefix
(effectively denoting different physical systems), the PEs would
translate these into unique VPN-IPv6 address prefixes using different
RDs. This ensures that if the same address is ever used in two
different VPNs, it is possible to install two completely different
routes to that address, one for each VPN.
Since VPN-IPv6 addresses and IPv6 addresses belong to different
address families, BGP never treats them as comparable addresses.
A VRF may have multiple equal-cost VPN-IPv6 routes for a single IPv6
address prefix. When a packet’s destination address is matched in a
VRF against a VPN-IPv6 route, only the IPv6 part is actually matched.
The Route Distinguisher format and encoding is as specified in
[BGP/MPLS-VPN].
When a site is IPv4 capable and IPv6 capable, the same RD MAY be used
for the advertisement of IPv6 addresses and IPv4 addresses.
Alternatively, a different RD MAY be used for the advertisement of
the IPv4 addresses and of the IPv6 addresses. Note, however, that in
the scope of this specification, IPv4 addresses and IPv6 addresses
will always be handled in separate contexts, and that no IPv4-IPv6
interworking issues and techniques will be discussed.
3. VPN-IPv6 Route Distribution
3.1. Route Distribution Among PEs by BGP
As described in [BGP/MPLS-VPN], if two sites of a VPN attach to PEs
that are in the same Autonomous System, the PEs can distribute VPN
routes to each other by means of an (IPv4) internal Border Gateway
Protocol (iBGP) connection between them. Alternatively, each PE can
have iBGP connections to route reflectors. Similarly, for IPv6 VPN
route distribution, PEs can use iBGP connections between them or use
iBGP connections to route reflectors. For IPv6 VPN, the iBGP
connections MAY be over IPv4 or over IPv6.
The PE routers exchange, via MP-BGP [BGP-MP], reachability
information for the IPv6 prefixes in the IPv6 VPNs and thereby
announce themselves as the BGP Next Hop.
The rules for encoding the reachability information and the BGP Next
Hop address are specified in the following sections.
3.2. VPN IPv6 NLRI Encoding
When distributing IPv6 VPN routes, the advertising PE router MUST
assign and distribute MPLS labels with the IPv6 VPN routes.
Essentially, PE routers do not distribute IPv6 VPN routes, but
Labeled IPv6 VPN routes [MPLS-BGP]. When the advertising PE then
receives a packet that has this particular advertised label, the PE
will pop this label from the MPLS stack and process the packet
appropriately (i.e., forward it directly according to the label or
perform a lookup in the corresponding IPv6-VPN context).
The BGP Multiprotocol Extensions [BGP-MP] are used to advertise the
IPv6 VPN routes in the MP_REACH Network Layer Reachability
Information (NLRI). The Address Family Identifier (AFI) and
Subsequent Address Family Identifier (SAFI) fields MUST be set as
follows:
- AFI: 2; for IPv6
- SAFI: 128; for MPLS labeled VPN-IPv6
The NLRI field itself is encoded as specified in [MPLS-BGP]. In the
context of this extension, the prefix belongs to the VPN-IPv6 Address
Family and thus consists of an 8-octet Route Distinguisher followed
by an IPv6 prefix as specified in Section 2, above.
3.2.1. BGP Next Hop encoding
The encoding of the BGP Next Hop depends on whether the policy of the
BGP speaker is to request that IPv6 VPN traffic be transported to
that BGP Next Hop using IPv6 tunneling ("BGP speaker requesting IPv6
transport") or using IPv4 tunneling ("BGP speaker requesting IPv4
transport").
Definition of this policy (to request transport over IPv4 tunneling
or IPv6 tunneling) is the responsibility of the network operator and
is beyond the scope of this document. Note that it is possible for
that policy to request transport over IPv4 (resp. IPv6) tunneling
while the BGP speakers exchange IPv6 VPN reachability information
over IPv6 (resp. IPv4). However, in that case, a number of
operational implications are worth considering. In particular, an
undetected fault affecting the IPv4 (resp. IPv6) tunneling data path
and not affecting the IPv6 (resp. IPv4) data path could remain
undetected by BGP, which in turn may result in black-holing of
traffic.
Control of this policy is beyond the scope of this document and may
be based on user configuration.
3.2.1.1. BGP Speaker Requesting IPv6 Transport
When the IPv6 VPN traffic is to be transported to the BGP speaker
using IPv6 tunneling (e.g., IPv6 MPLS LSPs, IPsec-protected IPv6
tunnels), the BGP speaker SHALL advertise a Next Hop Network Address
field containing a VPN-IPv6 address
- whose 8-octet RD is set to zero, and
- whose 16-octet IPv6 address is set to the global IPv6 address of
the advertising BGP speaker.
This is potentially followed by another VPN-IPv6 address
- whose 8-octet RD is set to zero, and
- whose 16-octet IPv6 address is set to the link-local IPv6
address of the advertising BGP speaker.
The value of the Length of the Next Hop Network Address field in the
MP_REACH_NLRI attribute shall be set to 24 when only a global address
is present, and to 48 if a link-local address is also included in the
Next Hop field.
If the BGP speakers peer using only their link-local IPv6 address
(for example, in the case where an IPv6 CE peers with an IPv6 PE,
where the CE does not have any IPv6 global address, and where eBGP
peering is achieved over the link-local addresses), the "unspecified
address" ([V6ADDR]) is used by the advertising BGP speaker to
indicate the absence of the global IPv6 address in the Next Hop
Network Address field.
The link-local address shall be included in the Next Hop field if and
only if the advertising BGP speaker shares a common subnet with the
peer the route is being advertised to [BGP-IPv6].
In all other cases, a BGP speaker shall advertise to its peer in the
Next Hop Network Address field only the global IPv6 address of the
next hop.
As a consequence, a BGP speaker that advertises a route to an
internal peer may modify the Network Address of Next Hop field by
removing the link-local IPv6 address of the next hop.
An example scenario where both the global IPv6 address and the link-
local IPv6 address shall be included in the BGP Next Hop address
field is that where the IPv6 VPN service is supported over a multi-
Autonomous System (AS) backbone with redistribution of labeled VPN-
IPv6 routes between Autonomous System Border Routers (ASBR) of
different ASes sharing a common IPv6 subnet: in that case, both the
global IPv6 address and the link-local IPv6 address shall be
advertised by the ASBRs.
3.2.1.2. BGP Speaker Requesting IPv4 Transport
When the IPv6 VPN traffic is to be transported to the BGP speaker
using IPv4 tunneling (e.g., IPv4 MPLS LSPs, IPsec-protected IPv4
tunnels), the BGP speaker SHALL advertise to its peer a Next Hop
Network Address field containing a VPN-IPv6 address:
- whose 8-octet RD is set to zero, and
- whose 16-octet IPv6 address is encoded as an IPv4-mapped IPv6
address [V6ADDR] containing the IPv4 address of the advertising
BGP speaker. This IPv4 address must be routable by the other
BGP Speaker.
3.3. Route Target
The use of route target is specified in [BGP/MPLS-VPN] and applies to
IPv6 VPNs. Encoding of the extended community attribute is defined
in [BGP-EXTCOM].
3.4. BGP Capability Negotiation
In order for two PEs to exchange labeled IPv6 VPN NLRIs, they MUST
use BGP Capabilities Negotiation to ensure that they both are capable
of properly processing such NLRIs. This is done as specified in
[BGP-MP] and [BGP-CAP], by using capability code 1 (multiprotocol
BGP), with AFI and SAFI values as specified above, in Section 3.2.
4. Encapsulation
The ingress PE Router MUST tunnel IPv6 VPN data over the backbone
towards the Egress PE router identified as the BGP Next Hop for the
corresponding destination IPv6 VPN prefix.
When the 16-octet IPv6 address contained in the BGP Next Hop field is
encoded as an IPv4-mapped IPv6 address (see Section 3.2.1.2), the
ingress PE MUST use IPv4 tunneling unless explicitly configured to do
otherwise. The ingress PE MAY optionally allow, through explicit
configuration, the use of IPv6 tunneling when the 16-octet IPv6
address contained in the BGP Next Hop field is encoded as an IPv4-
mapped IPv6 address. This would allow support of particular
deployment environments where IPv6 tunneling is desired but where
IPv4-mapped IPv6 addresses happen to be used for IPv6 reachability of
the PEs instead of Global IPv6 addresses.
When the 16-octet IPv6 address contained in the BGP Next Hop field is
not encoded as an IPv4-mapped address (see Section 3.2.1.1), the
ingress PE MUST use IPv6 tunneling.
When a PE receives a packet from an attached CE, it looks up the
packet’s IPv6 destination address in the VRF corresponding to that
CE. This enables it to find a VPN-IPv6 route. The VPN-IPv6 route
will have an associated MPLS label and an associated BGP Next Hop.
First, this MPLS label is pushed on the packet as the bottom label.
Then, this labeled packet is encapsulated into the tunnel for
transport to the egress PE identified by the BGP Next Hop. Details
of this encapsulation depend on the actual tunneling technique, as
follows:
As with MPLS/BGP for IPv4 VPNs [2547-GRE/IP], when tunneling is done
using IPv4 tunnels or IPv6 tunnels (resp. IPv4 GRE tunnels or IPv6
GRE tunnels), encapsulation of the labeled IPv6 VPN packet results in
an MPLS-in-IP (resp. MPLS-in-GRE) encapsulated packet as specified in
[MPLS-in-IP/GRE]. When tunneling is done using L2TPv3, encapsulation
of the labeled IPv6 VPN packet results in an MPLS-in-L2TPv3-
encapsulated packet, as specified in [MPLS-in-L2TPv3].
As with MPLS/BGP for IPv4 VPNs, when tunneling is done using an IPsec
secured tunnel [2547-IPsec], encapsulation of the labeled IPv6 VPN
packet results in an MPLS-in-IP- or MPLS-in-GRE-encapsulated packet
[MPLS-in-IP/GRE]. The IPsec Transport Mode is used to secure this
IPv4 or GRE tunnel from ingress PE to egress PE.
When tunneling is done using IPv4 tunnels (whether IPsec secured or
not), the Ingress PE Router MUST use the IPv4 address that is encoded
in the IPv4-mapped IPv6 address field of the BGP next hop field as
the destination address of the prepended IPv4 tunneling header. It
uses one of its IPv4 addresses as the source address of the prepended
IPv4 tunneling header.
When tunneling is done using IPv6 tunnels (whether IPsec secured or
not), the Ingress PE Router MUST use the IPv6 address that is
contained in the IPv6 address field of the BGP next hop field as the
destination address of the prepended IPv6 tunneling header. It uses
one of its IPv6 addresses as the source address of the prepended IPv6
tunneling header.
When tunneling is done using MPLS LSPs, the LSPs can be established
using any label distribution technique (LDP [LDP], RSVP-TE [RSVP-TE],
etc.).
When tunneling is done using MPLS LSPs, the ingress PE Router MUST
directly push the LSP tunnel label on the label stack of the labeled
IPv6 VPN packet (i.e., without prepending any IPv4 or IPv6 header).
This pushed label corresponds to the LSP starting on the ingress PE
Router and ending on the egress PE Router. The BGP Next Hop field is
used to identify the egress PE router and in turn the label to be
pushed on the stack. When the IPv6 address in the BGP Next Hop field
is an IPv4-mapped IPv6 address, the embedded IPv4 address will
determine the tunnel label to push on the label stack. In any other
case, the IPv6 address in the BGP Next Hop field will determine the
tunnel label to push on the label stack.
To ensure interoperability among systems that implement this VPN
architecture, all such systems MUST support tunneling using MPLS LSPs
established by LDP [LDP].
5. Address Types
Since Link-local unicast addresses are defined for use on a single
link only, those may be used on the PE-CE link, but they are not
supported for reachability across IPv6 VPN Sites and are never
advertised via MultiProtocol-Border Gateway Protocol (MP-BGP) to
remote PEs.
Global unicast addresses are defined as uniquely identifying
interfaces anywhere in the IPv6 Internet. Global addresses are
expected to be commonly used within and across IPv6 VPN Sites. They
are obviously supported by this IPv6 VPN solution for reachability
across IPv6 VPN Sites and advertised via MP-BGP to remote PEs and are
processed without any specific considerations to their global scope.
Quoting from [UNIQUE-LOCAL]: "This document defines an IPv6 unicast
address format that is globally unique and is intended for local
communications [IPv6]. These addresses are called Unique Local IPv6
Unicast Addresses and are abbreviated in this document as Local IPv6
addresses. They are not expected to be routable on the global
Internet. They are routable inside of a more limited area such as a
site. They may also be routed between a limited set of sites."
[UNIQUE-LOCAL] also says in its Section 4.7: "Local IPv6 addresses
can be used for inter-site Virtual Private Networks (VPN) if
appropriate routes are set up. Because the addresses are unique
these VPNs will work reliably and without the need for translation.
They have the additional property that they will continue to work if
the individual sites are renumbered or merged."
In accordance with this, Unique Local IPv6 Unicast Addresses are
supported by the IPv6 VPN solution specified in this document for
reachability across IPv6 VPN Sites. Hence, reachability to such
Unique Local IPv6 Addresses may be advertised via MP-BGP to remote
PEs and processed by PEs in the same way as Global Unicast addresses.
Recommendations and considerations for which of these supported