Architecture and Framework for IPv6 over Non-Broadcast Access
draft-thubert-6man-ipv6-over-wireless-13
The information below is for an old version of the document.
| Document | Type |
This is an older version of an Internet-Draft whose latest revision state is "Replaced".
|
|
|---|---|---|---|
| Authors | Pascal Thubert , Michael Richardson | ||
| Last updated | 2023-03-03 | ||
| Replaced by | draft-ietf-6man-ipv6-over-wireless | ||
| RFC stream | Internet Engineering Task Force (IETF) | ||
| Formats | |||
| Additional resources | Mailing list discussion | ||
| Stream | WG state | Call For Adoption By WG Issued | |
| Document shepherd | (None) | ||
| IESG | IESG state | I-D Exists | |
| Consensus boilerplate | Unknown | ||
| Telechat date | (None) | ||
| Responsible AD | (None) | ||
| Send notices to | (None) |
draft-thubert-6man-ipv6-over-wireless-13
6MAN P. Thubert, Ed.
Internet-Draft Cisco Systems
Updates: RFC4291 (if approved) M. Richardson
Intended status: Informational Sandelman
Expires: 4 September 2023 3 March 2023
Architecture and Framework for IPv6 over Non-Broadcast Access
draft-thubert-6man-ipv6-over-wireless-13
Abstract
This document extends RFC 4291 for IPv6 over modern Non-Broadcast /
Non-Transitive networks that decouples the physical concepts of
interface, link and broadcast domain from the IP concepts of
Interface, Link and Subnet, with the primary goal to apply on
networks where broadcast is undesirable. A study of the issues with
classical ND over wireless media is presented, and a framework to
solve those issues within the new architecture is proposed.
Status of This Memo
This Internet-Draft is submitted in full conformance with the
provisions of BCP 78 and BCP 79.
Internet-Drafts are working documents of the Internet Engineering
Task Force (IETF). Note that other groups may also distribute
working documents as Internet-Drafts. The list of current Internet-
Drafts is at https://datatracker.ietf.org/drafts/current/.
Internet-Drafts are draft documents valid for a maximum of six months
and may be updated, replaced, or obsoleted by other documents at any
time. It is inappropriate to use Internet-Drafts as reference
material or to cite them other than as "work in progress."
This Internet-Draft will expire on 4 September 2023.
Copyright Notice
Copyright (c) 2023 IETF Trust and the persons identified as the
document authors. All rights reserved.
Thubert & Richardson Expires 4 September 2023 [Page 1]
Internet-Draft IPv6 over NBMA March 2023
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 Revised BSD License text as
described in Section 4.e of the Trust Legal Provisions and are
provided without warranty as described in the Revised BSD License.
Table of Contents
1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . 3
2. Issues with IPv6 ND-Based Access . . . . . . . . . . . . . . 3
2.1. ND-Classic and ND-Proxies . . . . . . . . . . . . . . . . 3
2.2. The case of Wireless . . . . . . . . . . . . . . . . . . 5
2.3. The case of Overlays . . . . . . . . . . . . . . . . . . 6
2.4. Power and Sustainability . . . . . . . . . . . . . . . . 8
2.5. Security and Privacy . . . . . . . . . . . . . . . . . . 8
2.6. More Middleboxes . . . . . . . . . . . . . . . . . . . . 9
2.7. Summary of Issues . . . . . . . . . . . . . . . . . . . . 11
3. An Architecture for IPv6 over Non-Broadcast Networks . . . . 11
3.1. Basic Concepts . . . . . . . . . . . . . . . . . . . . . 12
3.2. Terminology . . . . . . . . . . . . . . . . . . . . . . . 14
3.2.1. IP Links . . . . . . . . . . . . . . . . . . . . . . 14
3.2.2. IP Interfaces . . . . . . . . . . . . . . . . . . . . 16
3.2.3. IP Subnets . . . . . . . . . . . . . . . . . . . . . 17
3.2.4. ND Proxies . . . . . . . . . . . . . . . . . . . . . 19
3.2.5. Subnet Gateway Protocols . . . . . . . . . . . . . . 19
3.2.6. Acronyms . . . . . . . . . . . . . . . . . . . . . . 20
3.3. IP Models . . . . . . . . . . . . . . . . . . . . . . . . 21
3.3.1. Physical Broadcast Domain . . . . . . . . . . . . . . 21
3.3.2. link-layer Broadcast Emulations . . . . . . . . . . . 22
3.3.3. Mapping the IPv6 link Abstraction . . . . . . . . . . 24
3.3.4. Mapping the IPv6 subnet Abstraction . . . . . . . . . 25
3.4. Stateful Address Autoconfiguration and Subnet Routing . . 26
4. A Framework for Stateful Address Autoconfiguration and Subnet
Routing . . . . . . . . . . . . . . . . . . . . . . . . . 26
4.1. Implementing Stateful Address Autoconfiguration . . . . . 26
4.2. links and Link-Local Addresses . . . . . . . . . . . . . 27
4.3. Subnets and Global Addresses . . . . . . . . . . . . . . 28
4.4. Anycast and Multicast Addresses . . . . . . . . . . . . . 29
5. WiND Applicability . . . . . . . . . . . . . . . . . . . . . 29
5.1. Case of LPWANs . . . . . . . . . . . . . . . . . . . . . 30
5.2. Case of Infrastructure BSS and ESS . . . . . . . . . . . 30
5.3. Case of Mesh Under Technologies . . . . . . . . . . . . . 32
5.4. Case of DMB radios . . . . . . . . . . . . . . . . . . . 32
5.4.1. Using ND-Classic only . . . . . . . . . . . . . . . . 32
5.4.2. Using Wireless ND . . . . . . . . . . . . . . . . . . 32
Thubert & Richardson Expires 4 September 2023 [Page 2]
Internet-Draft IPv6 over NBMA March 2023
6. IANA Considerations . . . . . . . . . . . . . . . . . . . . . 36
7. Coexistence with IPv6 ND . . . . . . . . . . . . . . . . . . 36
8. Privacy Considerations . . . . . . . . . . . . . . . . . . . 39
9. Security Considerations . . . . . . . . . . . . . . . . . . . 40
10. Acknowledgments . . . . . . . . . . . . . . . . . . . . . . . 40
11. Normative References . . . . . . . . . . . . . . . . . . . . 41
12. Informative References . . . . . . . . . . . . . . . . . . . 42
Authors' Addresses . . . . . . . . . . . . . . . . . . . . . . . 46
1. Introduction
IEEE Std. 802.1 [IEEE Std. 802.1] Ethernet Bridging provides an
efficient and reliable broadcast service for wired networks;
applications and protocols have been built that heavily depend on
that feature for their core operation. Unfortunately, Low-Power
Lossy Networks (LLNs) and Wireless Local Area Networks (WLANs)
generally do not benefit from the same reliable and cheap broadcast
capabilities as Ethernet links.
Similarly, the use of broadcast is discouraged on large scale virtual
networks that are overlaid over distributed physical locations as
meshes of point-to-point (P2P) links, as it represents massive
amounts of replication. All in all, as IPv6 Links migrate from a
physical wire to virtual or intangible media, modern networks exhibit
the common requirement to decouple physical properties such as
broadcast capabilities and transitivity that are perceived and
handled at the lower layers from the abstractions that are
manipulated at the IP layer.
This document presents an architecture for IPv6 over modern Non-
Broadcast / Non-Transitive networks that decouples the lower-layer
concepts of links and broadcast domains from the IP concepts of Links
and Subnets. A study of the issues with classical ND over wireless
media is presented, and a framework to solve those issues within the
new architecture is proposed.
2. Issues with IPv6 ND-Based Access
2.1. ND-Classic and ND-Proxies
Though ND-Classic was the state of the art when designed for an
Ethernet wire at the end of the twentieth century, it must be
reevaluated for the new technologies, such as wireless and overlays,
that evolved since then. In particular, relying on broadcast
operation may be inefficient (too many replications over constrained
links), unreliable (broadcast may be lost in transmission),
detrimental to network operations (broadcast storms), prone to
multiplication attacks (a unicast from the outside may cause a
Thubert & Richardson Expires 4 September 2023 [Page 3]
Internet-Draft IPv6 over NBMA March 2023
broadcast inside), and detrimental to privacy (an observer inside the
network may listen to broadcasts and learn more than it needs for its
own operation).
The IPv6 ND Neighbor Solicitation (NS) [RFC4861] message is used as a
multicast IP packet for Address Resolution (AR) and Duplicate Address
Detection (DAD) [RFC4862]. In those cases, the NS message is sent at
the Network Layer to a Solicited-Node Multicast Address (SNMA)
[RFC4291] and should in theory only reach a very small group of
nodes. It is intended for one Target, that may or may not be present
in the network, but it is often turned into a MAC-Layer broadcast and
effectively reaches most of the nodes that are attached to the layer
2 link.
DAD was designed for the efficient broadcast operation of Ethernet.
Experiments show that DAD often fails to discover the duplication of
IPv6 addresses in large wireless access networks [DAD ISSUES]. In
practice, IPv6 addresses very rarely conflict, not because the
address duplications are detected and resolved by the DAD operation,
but thanks to the entropy of the 64-bit Interface IDs (IIDs) that
makes a collision quasi-impossible for randomized IIDs.
Arguably, IPv6 ND brought the support if multicast that should have
reduced the broadcast traffic for DAD and AR, but in practice IPv6 ND
multicast messages are mapped to link-layer broadcast on Ethernet and
Wireless networks, and replication on overlays. The multicast model
to Solicited Node Multicast Addresses (SNMA) probably does not make
much economical sense either, since it would mean close to one state
per address is every switch (there's usually only one address with
the same SNMA, though the birthday paradox applies) when a complete
ND cache would require only one state in every router, and there are
typically far less routers than switches in the subnet.
Multicast NS transmissions may occur when a node joins the network,
moves, or wakes up and reconnects to the network. Over a very large
fabric, this can generate hundreds of broadcasts per second. If the
broadcasts were blindly copied over Wi-Fi, the MAC-layer broadcast
traffic associated to ND IP-layer multicast could consume enough
bandwidth to cause a substantial degradation to the unicast service
[MCAST EFFICIENCY]. To protect their bandwidth, some networks
throttle ND-related broadcasts, which reduces the capability for the
ND protocol to operate as expected.
The IPv6 ND Neighbor Advertisement (NA) [RFC4861] message can also be
sent multicast, as a gratuitous information that can be used to
override the neighbor cache entries (NCEs) in all nodes with the Link
Layer Address (LLA) of the sender.
Thubert & Richardson Expires 4 September 2023 [Page 4]
Internet-Draft IPv6 over NBMA March 2023
This problem can be alleviated by reducing the size of the broadcast
domain that encompasses wireless access links. This has been done in
the art of IP subnetting by partitioning the subnets and by routing
between them, at the extreme by assigning a /64 prefix to each
wireless node (see [RFC8273]).
Another way to split the broadcast domain within a subnet is to proxy
at the boundary of the wired and wireless domains the Network Layer
protocols that rely on link-layer broadcast operations. [IEEE Std.
802.11] recommends deploying proxies for the IPv4 Address Resolution
Protocol (ARP) and IPv6 ND at the APs. This requires the exhaustive
list of the IP addresses for which proxying is provided. Forming and
maintaining that knowledge is a hard problem in the general case of
radio connectivity, which keeps changing with movements and
variations in the environment that alter the range of transmissions.
[SAVI] suggests discovering the addresses by snooping the ND-Classic
protocol, but that can also be unreliable. An IPv6 address may not
be discovered immediately due to a packet loss. It may never be
discovered in the case of a "silent" node that is not currently using
one of its addresses, e.g., a printer that waits in wake-on-lan
state. A change of anchor, e.g. due to a movement, may be missed or
misordered, leading to unreliable connectivity and an incomplete list
of addresses.
2.2. The case of Wireless
As opposed to unicast transmissions, the broadcast transmissions over
wireless links are not subject to automatic retries (ARQ) and can be
very unreliable. Reducing the speed at the physical (PHY) layer for
broadcast transmissions can increase the reliability, at the expense
of a higher relative cost of broadcast on the overall available
bandwidth. As a result, protocols designed for bridged networks that
rely on broadcast transmissions often exhibit disappointing behaviors
when employed unmodified on a local wireless medium (see [MCAST
PROBLEMS]).
Like Transparent Bridging, the IPv6 [RFC8200] Neighbor Discovery
[RFC4861] [RFC4862] Protocol (ND-Classic) is reactive, and relies on
reactive Network Layer multicast for AR to locate an on-link
correspondent and DAD to ensure the uniqueness of an IPv6 address in
the case of Stateless Address Autoconfiguration (SLAAC). On Ethernet
LANs and most WLANs and Low-Power Personal Area Networks (LoWPANs),
the Network Layer multicast operation is typically implemented as a
link-layer broadcast for the lack of an adapted and scalable link-
layer multicast operation.
Thubert & Richardson Expires 4 September 2023 [Page 5]
Internet-Draft IPv6 over NBMA March 2023
It results that on wireless, an ND-Classic multicast message is
typically broadcasted. So even though there are very few nodes
subscribed to the Network Layer multicast group, and there is at most
one intended Target, the broadcast is received by many wireless nodes
over the whole subnet (e.g., the ESS fabric). And yet, the broadcast
transmission being unreliable, the intended Target may effectively
have missed the packet.
On paper, a Wi-Fi station must keep its radio turned on to listen to
the periodic series of broadcast frames, which for the most part will
be dropped when they reach Network Layer. In order to avoid this
waste of energy and increase its battery life, a typical battery-
operated device such as an IoT sensor or a smartphone will blindly
ignore a ratio of the broadcasts, making ND-Classic operations even
less reliable.
Wi-Fi [IEEE Std. 802.11] Access Points (APs) deployed in an Extended
Service Set (ESS) act as [IEEE Std. 802.1] bridges between the
wireless stations (STA) and the wired backbone. As opposed to the
classical Transparent (aka Learning) Bridge operation that installs
the forwarding state reactively to traffic, the bridging state in the
AP is established proactively, at the time of association. This
protects the wireless medium against broadcast-intensive Transparent
Bridging lookups. The association process registers the link-layer
(MAC) Address (LLA) of the STA to the AP proactively, i.e., before it
is needed. The AP maintains the list of the associated addresses and
blocks the lookups for destinations that are not registered. This
solves the broadcast issue for the link-layer lookups, but the
Network Layer problem remains.
2.3. The case of Overlays
Layer-2 Overlays (VLANs) reduce the broadcast domain from the
physical one, whereas Layer-3 overlays (e.g., VxLAN) can extend the
Subnet beyond the limits of the physical network, enabling to deploy
a Subnet over large physical domains. Layer-3 overlays are a clear
indication of the need to decouple the Subnet from the limits of
physical network.
Thubert & Richardson Expires 4 September 2023 [Page 6]
Internet-Draft IPv6 over NBMA March 2023
A Layer-3 overlay is typically a partial or full mesh of point to
point tunnel connections between routers. In case of a full mesh, a
BUM (Broadcast, Unknown, and Multicast) forwarding over the overly
can be implemented as ingress replication, which becomes detrimental
to the operation of the ingress router as the overlay grows in
number. This can be optimized in multisite operations so only one
copy flies across sites in Data Center Interconnect (DCI)
environments. Still, BUM processing is costly as it spreads across
all sites that the overlay spans, and its use should be limited to
operations where a real broadcast operation is desired, e.g., every
node is interested in receiving the packet.
As discussed in Section 2.2, multicast IPv6 ND messages are in fact
broadcasted over the network, meaning that in the case of an overlay,
they contribute to the BUM traffic as full broadcast. While a
broadcast IPv6 ND RA message may be of interest within a site or a
subsite to all local nodes, it is probably of little interest to
other sites that are served by other routers, even when the overlay
spans across both sites. The same goes for a local printer, the
broadcast mDNS lookup should reach local printers but not faraway
ones. This is another indication of the need to decouple the span of
the Subnet from the lower layer broadcast domain.
Broadcast IPv6 ND NS messages used for lookup and DAD intend to reach
at most one node. When those messages is used over the fabric and
the target is a silent node, which location is unknown to the fabric,
either the NS message is broadcasted, or the packet is lost, meaning
that the silent node will not be reachable again until it provides
its own sign of life. Using a BUM transmission that reaches every
nodes to communicate with at most one is a misuse of the overlay
resources and should be replaced by unicast-based methods.
In data centers, overlays are typically combined with multihoming at
the edge. An advanced Network Interface Card (NIC) is equipped with
more than one physical adapter (typically Ethernet) for redundancy,
and the physical adapters connect to different leaf switches (aka
ToR) to the same cloud network. The server (say, using Kubernetes)
needs a single address and would rather use that address on both
ports to the same Subnet, and use either adapter for its own reasons,
e.g., load balancing, independently of IP addressing considerations.
This means that the Interface abstraction that the server needs is a
logical construct, decoupled from the physical adapters, and capable
to encompass more than one.
Thubert & Richardson Expires 4 September 2023 [Page 7]
Internet-Draft IPv6 over NBMA March 2023
2.4. Power and Sustainability
Misusing broadcast entails extraneous power consumption in any
switched or wireless environment. In the wireless case, broadcasts
are sent at the slowest speed available in order to maximize the
chances that all nodes in the BSS will receive the frame; not only
does this consume more precious bandwidth than anycast transmission
and generate interferes which may cascade in losses and retries in
adjacent networks, this also means that power is consumed for a
longer time (can be hundred time more) than for unicast. Power is
also wasted when replicating a packet to span an overlay. In all
cases, to make the internet greener, we must reconsider the use of
broadcast over large access networks. To maintain the capability to
build the Subnets we want, this means that we must decouple the
Subnet from the broadcast domain, make the broadcast domain small and
local, and refocus the use of broadcast to the cases where all nodes
are interested.
Constrained IoT devices conserve their power by placing themselves in
deep sleep for most of the time. While a device is sleeping, it
cannot answer IPv6 ND messages for AR and DAD. This makes IPv6 ND
unsuitable to IoT devices. The 6LoWPAN WG has determined that the
most appropriate model for a constrained network is a pull model
where the device wakes up, negotiates with its router(s) for
addresses and connectivity for some amount of time in the future, and
goes back to sleep. The router can then perform a role of sleep
proxy that defends the address(es) and holds traffic for the device
till the device wakes up and pulls it from the router. Additionally,
when the constrained network grows into a mesh, broadcast operations
become rapidly inacceptable in terms of power and bandwidth, and the
not-only model whereby for the most part hosts do not lookup one
another, is mandatory.
2.5. Security and Privacy
Broadcasting NS and NA messages exposes to the source to the whole
network, including nodes that do not need to know that the source is
present on the network. A passive listener my learn a lot on the
presence of others and the source has no control over that. A more
private approach has each node see and connect to only a subset of
the routers using the not-onlink model, leaving the source the
capability to shortcut to the destination based on redirect messages
from the router, provided that the source wishes to do so.
Once it has discovered the presence of neighbor addresses, it becomes
easy for the onlink attacker to impersonate any host in the network,
either by sourcing packets with a stolen address, or by overriding
the neighbor caches with a NA that indicates the attacker's LLA. It
Thubert & Richardson Expires 4 September 2023 [Page 8]
Internet-Draft IPv6 over NBMA March 2023
is thus desirable to avoid exposing the host IPv6 address in
broadcast ND messages. Additionally, it is desirable that only the
address owner can source packets with that address, and that another
party may not claim and obtain that address.
The reactive NS lookup method can also be leveraged from the outside
of the network to perform DOS attacks on constrained resources. The
attacker just needs to forge any random address from the subnet
prefix and send the packet. The Subnet ingress router will have to
store that packet for the duration of the lookup time, and broadcast
an NS to lookup the fake address. This both locks memory in the
router and consumes bandwidth in the Subnet.
To avoid the lookup delays and associated attacks, it makes sense to
use a proactive method whereby the router knows all the address
mappings in advance, so that if the destination address is not
present in the router tables, then the destination does not exist in
the network and the packet can safely be dropped by the forwarding
engine.
2.6. More Middleboxes
The above problems have been observed for at least since ealy 2000s.
A number of actions were taken. For instance, IEEE std 802.11 [IEEE
Std. 802.11] mandates the support of a middlebox operation called
"ARP proxy" for IPv4 and IPv6 in the Access Point (AP) that cancels
broadcasts over a BSS when the IP target of the ARP/ND message is not
associated to the AP. With IPv4, the expectation is that the STA
owns one address, and that the address is obtained via DHCP right
after the association, so it is pretty easy to snoop the address in
the DHCP exchange (as long as it remains in the clear) and cancel
undesirable ARP lookups.
In contrast to IPv4, IPv6 enables a node to form multiple addresses,
some of them temporary and with a particular attention paid to
privacy. Addresses may be formed and deprecated asynchronously to
the association. Even if the knowledge of IPv6 addresses used by a
STA can be obtained by snooping protocols such as IPv6 ND and DHCPv6,
or by observing data traffic sourced at the STA, such methods provide
only an imperfect knowledge of the state of the STA at the AP. This
may result in a loss of connectivity for some IPv6 addresses, in
particular for addresses rarely used and in a situation of mobility.
This may also result in undesirable state persistence in the AP when
a STA ceases to use an IPv6 address. It follows that snooping
protocols is not a recommended technique and that it should only be
used as last resort.
Thubert & Richardson Expires 4 September 2023 [Page 9]
Internet-Draft IPv6 over NBMA March 2023
Because IPv6 ND is so easy to attack, some vendors have deployed
undocumented proprietary counter measures as middlebox operation in
the switches and routers. Those middleboxes snoop the ND protocol
messages, filter them or modify them, for instance to change their
scope from multicast to unicast. Based on the snooped information,
the middlebox may for instance drop an RA message that appears to be
coming from a host (e.g., as inferred because it arrives on a
wireless interface), or an NA message that appears to be coming from
a device that does not appear to be the owner. But, for the lack of
an explicit contract between the host and the middlebox, the
middlebox cannot determine who the earl owner is, and it may reject
the rightful owner.
The middlebox may also block SLAAC from a host above a fixed number
of addresses, to protect the network resources (addresses are an
expensive and thus limited resource in a managed network) unbeknownst
to the host. When that happens, the host believes that it can use
the address but fails to connect with it. This might happen even if
the host has ceased to use other addresses and is now within the
quota, because ND lacks a method for the host to know how many
addresses it can own, and for the router to know which addresses the
host uses at any point of time. The infrastructure needs a
deterministic knowledge of the addresses in use, and for that a
contract must be passed between the host and the network to ensure
that the addresses are usable.
Maybe the most insidious side effect of those middleboxes is that as
opposed to NAT, they are obfuscated and proprietary. From one vendor
to the next, and one product generation to the next, their behavior
may evolve and affect the ND protocol in different fashions. In the
short term, this may only impact some specific deployments, which may
have to work around the issues. But worse than that, this may affect
the future and the capability we have to evolve the protocols, like
firewalls impact our capability to develop new transports in parallel
to TCP and UDP. We must either standardize (e.g., for ND proxy) or
eliminate those middlebox activities, and for that, the IPv6 ND
protocol must evolve to a model where the middleboxes are not needed
anymore. This means that we need a model that minimizes the use of
broadcasts, and where a contract provides mutual guarantees for the
host that need IPv6 addresses and the routers that provide
reachability and protection for these addresses.
Thubert & Richardson Expires 4 September 2023 [Page 10]
Internet-Draft IPv6 over NBMA March 2023
2.7. Summary of Issues
IPv6 ND inherited 2 majors design points from IPv4, a strong coupling
of logical and physical concepts, which creates unacceptable
constraints on modern deployments with virtual and intangible links,
and a reactive operation for AR and DAD that requires an extensive
use of broadcast. While IPv4 and IPv6 behaviors are similar for
addresses obtained via DHCP, the cost of AR and DAD makes IPv6
significantly more expensive with Stateless Address Autoconfiguration
(SLAAC) is used.
And while IPv4 supported NBMA and P2MP models (e.g., on Frame Relay
leveraging OSPFv2), the IPv6 promise to support NBMA (for ATM)
remained unmet to this day, and only P2P and Transit links are
properly supported by IPv6 ND. For those reasons, IPv6 with SLAAC
can be significantly less attractive than IPv4 to some network
administrators.
IPv6 ND exposes all addresses to all nodes in the network, which is
unfit for privacy. It is prone to DOS attacks from outside the
network and impersonation attacks from the inside, with no method to
prove the address ownership and perform Source Address Validation
(SAVI) later on the data traffic. To protect against such threats,
the vendors had to introduce middleboxes that interfere with the
protocol operation and affect the capability to evolve the protocol
in the future.
IPv6 ND lack a support for mobility (which typically entails a
sequence counter maintained by the host and the deprecation of state
that is based on older sequences) and for anycast. This makes it
very hard for the network to defend the addresses on behalf of the
owner, e.g., when the owner is temporarily disconnected. It results
that the operation of the middleboxes is unsatisfactory and may cause
discontinuities in connectivity.
The broadcast operation is not only detrimental to bandwidth, it is
also an issue for energy conservation and sustainability. Devices
must be always attached and always powered on to answer lookups and
DADs, which makes IPv6 ND unapplicable to constrained IoT devices
that sleep for the vast majority of their time. The key design
points in this architecture derive from the original observations
made at the 6LoWPAN WG, and focus on avoiding waste of constrained
resources such as spectrum and energy, by using broadcasts only when
broadcast is really needed, and decoupling the IP abstraction of a
Subnet from the broadcast domains.
3. An Architecture for IPv6 over Non-Broadcast Networks
Thubert & Richardson Expires 4 September 2023 [Page 11]
Internet-Draft IPv6 over NBMA March 2023
3.1. Basic Concepts
This document introduces a new architecture to IPv6 Neighbor
Discovery that is designed to apply to the WLANs and LoWPANs types of
networks, as well as other Non-Broadcast Multi-Access (NBMA) networks
such as Data-Center overlays and more generally as a replacement to
the classical IPv6 ND in any network where the issues discussed in
Section 2 are detrimental to the network operation. The architecture
leverages the not-onlink model and routing inside the subnet, which
enables to form potentially large MLSNs without creating a large
broadcast domain at the link-layer.
To support the deployment agility that virtual (e.g., VxLAN overlays
and pseudowires) and intangible (e.g., wireless, laser, and quantum)
links enable, the IP abstractions of Interface, Link, and Subnet are
decoupled from their classical physical counterparts of interface,
link, and broadcast domains. The Subnet is defined by a /64 prefix,
as the longest aggregation that can be advertised in the IGP with the
outside. Host routes and longer prefixes are advertised inside the
Subnet only, using a separate Subnet Gateway protocol.
Any device that owns an address within the Subnet prefix belongs to
the subnet, this is now decoupled from the physical connectivity and
broadcast domain. Instead, the IPv6 routers that serve a Subnet must
form a connected dominating set such that every host in the Subnet is
connected to at least one router and the routers are connected to one
another directly (classical NBMA, aka full mesh) or indirectly via
other routers (Point to MultiPoint, P2MP, aka partial mesh). The
not-onlink model is used throughout, so hosts do not look each other
up, saving all the associated broadcast. Instead, they rely on the
routers to forward the packets inside and outside the Subnet. This
way, the Subnet can have any structure needed for the deployment,
where hosts can move from router to router in the Subnet, or anywhere
in the Internet provided they can lay a mobility tunnel to one of the
routers for use as IP Link to the Subnet.
All IP links are abstracted as Point-to-Point, though a lower-layer
broadcast service may be used by the router to send RAs to a subset
of local hosts in the subnet, or by the host to send an RS message to
a subset of the routers. An IP Interface bundles one or more
subInterfaces, one per Subnet that can be reached through that
Interface. A Global IPv6 address is installed on the subInterface
that connects to the Subnet from which the address derives. The IP
Interface connects to one or more IP Links (to different neighbors)
over the same or over different physical interfaces (they are
decoupled). A Link-Local Address is associated to the IP Link
directly. Each SubInterface connects to the subset of those IP Links
that reach other nodes in the Subnet.
Thubert & Richardson Expires 4 September 2023 [Page 12]
Internet-Draft IPv6 over NBMA March 2023
In a fashion similar to a Wi-Fi Association, IPv6 Hosts register
their addresses to their serving router(s), which may reject the
registration, e.g., in case of a duplication. To support distributed
routers in the Subnet, an abstract registrar service maintains the
state of all active registrations in the Subnet and answers queries
to lookup mappings, validate ownership, and avoid duplications. The
registration is abstract to the routing service and the registrar
service, and it can be protected to prevent impersonation attacks.
The registration enables the network to know deterministically all
the IPv6 addresses and LLA mapping currently in use, and eliminates
the need for lookups and DAD, and for the associated broadcasts.
With the registration, the routers have a complete knowledge of the
hosts they serve and in return, hosts obtain routing services for
their registered addresses. The abstract routing service allows an
ingress router to find a path to the destination address within the
subnet. It can be a simple reflection in a Hub-and-Spoke subnet that
emulates an IEEE Std. 802.11 Infrastructure BSS at the Network Layer.
It can also be a full-fledge routing protocol, e.g., RPL, which is
designed to adapt to various LLNs such as WLAN and WPAN radio meshes,
see [RFC9010] for more, or BGP/EVPN, for application in data centers,
see [RFC8365] for more. It can be based on overlay tunnels between
ingress router and egress router leveraging a resolver service such
as LISP, see [RFC7834] for more. Finally, the routing service can
also be an ND proxy that emulates an IEEE Std. 802.11 Infrastructure
ESS at the Network Layer, as specified in the IPv6 Backbone Router
[RFC8929].
The abstract registrar service maintains the mapping between the
registered node link-layer address and the registered IPv6 address.
It contains meta data that enables to ascertain that the second
registration for the same address is performed for the same
registered node, so it also binds the registered node with the
registered IPv6 address. The registrar service provides APIs to look
up a link-layer address for an IPv6 address as well as validate IPv6
address ownership. The registrar can be implemented as a mapping
server ala LISP, a distributed state ala ND proxy, or a synchronized
state ala EVPN. In the former case, this enables the reactive
lookups to be performed as unicast requests to the map resolver. In
the latter, the address mapping is synchronized by the routing
protocol and known to all the routers for all nodes in the IP Subnet,
so there is never a need for a reactive lookup.
On the one hand, the Architecture proposed in this document avoids
the use of broadcast operation for DAD and AR, and on the other hand,
it supports use cases where subnet and link-layer domains are not
congruent, which is common in wireless networks unless a specific
link-layer emulation is provided.
Thubert & Richardson Expires 4 September 2023 [Page 13]
Internet-Draft IPv6 over NBMA March 2023
The address registration provides a contractual interface between the
nodes and the routers where nodes can ask for addresses which will be
guaranteed to be operational for a contractual lifetime, and the
network may accept or refuse granting additional addresses based on
state (e.g., duplicate address) as well as policy (e.g., quota).
This ways hosts and routers agree deterministically on which
addresses will be served to which nodes in the Subnet. That
interface is agnostic to the router to router and router to
registrar. The latter interfaces can be implemented in various
fashions that can blend in existing technologies such as legacy IPv6
ND network through ND proxy, EVPN- and LISP-based overlays.
3.2. Terminology
3.2.1. IP Links
For a long time, the term link has been used to refer to the layer 2
communication medium that can be leveraged at layer 3 to instantiate
one IP hop. In this document we conserve that term but differentiate
it from an IP link, which is a layer 3 abstraction that represents
the layer 2 link but is not the layer 2 link, like the map is not the
country.
With IPv6, IP has moved to layer 3 abstractions for its operations,
e.g., with the use of link local address (LLA), and that of IP
multicast for link-scoped operations. At the same time, the concept
of an IP link emerged as an abstraction that represents how IP layer
considers the layer 2 link:
* An IP link connects an IP node to one or more other IP nodes using
a lower layer subnetwork. The lower layer subnetwork may comprise
multiple lower layer links, e.g., in the case of a switched fabric
or a mesh-under LLN.
* an IP link defines the scope of an LLA, and defines the domain in
which the LLA must be unique
* An IP Link is attached to a physical interface, and one Link Local
Address is associated to the IP Link.
* an IP link provides a subset of the connectivity that is offered
by the physical link at the lower layer; if the IP link is
narrower than the layer 2 reachable domain, then layer 3 filters
must restrict the link-scoped communication to remain between
peers on a same IP link. More than one IP link may be installed
on the same physical interface to connect to different peers.
Thubert & Richardson Expires 4 September 2023 [Page 14]
Internet-Draft IPv6 over NBMA March 2023
* an IP link can be Point to Point (P2P), Point to Point (P2MP,
forming a partial mesh), NBMA (non-broadcast multi-access, fully
meshed), or transit (broadcast-capable and any-to-any).
It is a network design decision to use one IP link model or another
over a given lower layer network, e.g., to map a Frame Relay network
as a P2MP IP link, or as a collection of P2P IP links. As another
example, an Ethernet fabric may be bridged, in which case the nodes
that interconnect the layer 2 links are L2 switches, and the fabric
can be abstracted as a single transit IP link; or the fabric can be
routed, in which case the P2P IP links are congruent with the layer 2
links, and the nodes that interconnect the links are routers.
This architecture only uses P2P Link abstractions as shown in
Figure 1, where an IP link is identified by a pair of local and
remote Link Layer Address. A physical interface may enable to reach
to more than one peer at layer 2; in that case, this architecture
maps each peer relationship as a different IP Link. A Link Local
Address only needs to be unique within that peer to peer
relationship.
+--------------------
|
| IP Link 1 ---------------------------> Node 1
Link Layer Address 1 Link Layer Address N1
| Link Local Address LLA1 Link Local Address LLAN1
|
| IP Link 2 ---------------------------> Node 2
Link Layer Address 2 Link Layer Address N2
| Link Local Address LLA2 Link Local Address LLAN2
| ...
|
| (LLA 1 may be the same as LLA 2)
+--------------------
Physical interface
Figure 1: P2P Link Abstraction
Thubert & Richardson Expires 4 September 2023 [Page 15]
Internet-Draft IPv6 over NBMA March 2023
If only the physical interface but not the Link-Layer address of the
peer is visible from the IP layer when processing a message, then the
IP layer cannot discriminate the IP Link of packets arriving on the
same interface, and for that reason, it will reject a second
registration for the same Link Local Address by a second peer,
meaning that a Link Local Address will have to be unique on a
physical interface across IP Links. In that case, the Link Local
Address of the peer is used to identify the IP Link, and all the
addresses registered to this node with the same peer Link Local
Address as source will be associated to the same IP Link to that
peer.
3.2.2. IP Interfaces
As is the case for links, the term interface has been historically
confused between the physical object that provides physical
connectivity and the IP layer logical object that connects the host
with the IP Link:
* an IP Interface is an abstraction that connects the host with a
collection of IP Links (for the purpose of link local
communication) and bundles the interfaces for each IP Subnet in
subInterfaces. The host installs at least one link-local address
on an IP Interface for each IP Link that is connected through that
interface, and one subInterface per Subnet. The same Link-Local
address may be reused over different IP Links as long as it is not
a collision for the peer on that IP Link. Similarly, the host
installs at least one global (or unique local as appropriate)
address on an IP subInterface for the associated Subnet, and the
address is advertised over each IP Link in the Subnet.
* an IP Interface can be P2P, in which case it connects to a single
IP Link, or P2MP, in which case it aggregates multiple IP Links.
In a multihomed host, a single IP Interface can be installed to
connect to the IP Links associated to different physical
interfaces, in which case the same IPv6 address may be advertised
on more than one physical interface. Conversely, when more than
one Subnet is reachable over a physical interface, more than one
IP Interface may leverage that physical interface for
transmission.
Thubert & Richardson Expires 4 September 2023 [Page 16]
Internet-Draft IPv6 over NBMA March 2023
+--------------------
|
| IP SubInterface a ------------------> IP Link A
| IP Subnet a::/64 IP Link B
| IP Addresses a::1 .. a::n ...
| IP Link N
|
| IP SubInterface b ------------------> IP Link A
| IP Subnet b::/64 IP Link D
| IP Addresses b::1 .. b::n ...
| IP Link P
| ...
|
| (Link A and B may be attached to different physical interfaces)
| (Link A may belong to both subInterfaces a and b)
|
+--------------------
IP Interface
Figure 2: Interface Abstraction
3.2.3. IP Subnets
IPv6 builds another abstraction, the IP subnet, over one shared IP
link or over a collection IP links, forming a MLSN in the latter
case. An MLSN is formed over IP links (e.g., P2P or P2MP) that are
interconnected by routers that either inject hosts routes in an IGP,
in which case the topology can be anything, or perform ND proxy
operations, in which case the structure of links must be strictly
hierarchical to avoid loops.
Thubert & Richardson Expires 4 September 2023 [Page 17]
Internet-Draft IPv6 over NBMA March 2023
+--------------------
| router 4
| ^ ^
| +----------------------------------+ / |
| | L2 broadcast domain IP Link IP Link
| | / | |
| | v | v
| | router1 <--IP Link--> router 2 <--IP Link--> router 3
| | ^ ^ ^ ^ | ^
| | | \ / | | |
| | IP Link IP Link IP Link IP Link | IP Link
| | | \ / | | |
| | v v v v | v
| | host 1 host 2 host 3 | host 4
| | |
| +----------------------------------+
|
| (Different IP Links may be sustained by different media)
|
+--------------------
IP Subnet
Figure 3: Subnet Abstraction
It is a network design decision to use one IP subnet model or another
over a given lower layer network. A switched fabric can host one or
more IP subnets, in which case the IP links can reach all and beyond
one subnet. On the other hand, a subnet can encompass a collection
of links; in that case, the scope of the link local addresses, which
is the IP Link, is narrower than the span of the subnet.
A subnet prefix is associated with the IP subnet, and a node is a
member of an IP subnet when it has an IP address that derives from
that prefix. The IP address is either a Unique Local (ULA) or a
Global Unicast Address (GUA), and as opposed to the case of LLAs, the
scope of the address is not limited to the IP subnet.
The switched and routed fabric above could be the exact same network
of physical links and boxes, what changes is the way the networking
abstractions are mapped onto the system, and the implication of such
decision include the capability to reach another node at layer-2, and
the size of the broadcast domain and related broadcast storms.
Thubert & Richardson Expires 4 September 2023 [Page 18]
Internet-Draft IPv6 over NBMA March 2023
3.2.4. ND Proxies
[RFC8929] defines bridging and routing IPv6 ND proxies for
registering nodes / registered addresses. Both forms of ND proxies
interconnect IP links and enable to isolate the layer 2 broadcast
domains. But in the case of a bridging proxy, the layer 2 unicast
communication can still exist between the layer 2 domains that are
covered by the layer 3 links, whereas in the base of a routing proxy,
they are isolated, and packets must be routed back and forth.
Bridging proxies are possible between compatible technologies and
translational bridges (e.g., Wi-Fi to Ethernet), whereas routing
proxies are required between non-bridgeable technologies and
desirable to avoid exposing the layer 2 addresses across, e.g., for
reasons of stability and scalability.
ND proxies can also serve IPv6 nodes that still rely on classical
IPv6 ND in a coexistence scenario. The ND proxy intercepts (snoops)
the multicast NS messages from the nodes and, in case or AR or DAD,
polls the registrar to lookup whether an active mapping exists for
the Target. When that is the case, the ND proxy may forward the NS
message as a Layer-2 unicast to the node that owns the binding, else
it may either drop the multicast or broadcast it at Layer-2. Once
the node formed an address, the ND proxy fills the registrar to
associate the IPv6 address with the node. The method is brittle,
since there is no contract with the node to guarantee the ownership,
no "contract", as discussed in Section 2.6, so for those addresses,
the registrar may be inaccurate.
3.2.5. Subnet Gateway Protocols
The /64 boundary creates a wall between the traditional Interior
Gateway Protocols (IGP) that operate between Subnets and manipulate
shorter than 64 prefixes, and Subnet Gateway Protocols (SGP) that
operate inside a Subnet and manipulate shorter than 64 prefixes,
typically /128 host routes, and possibly more specific data like
Link-Layer Address mappings and Address Proof of Ownership.
Thubert & Richardson Expires 4 September 2023 [Page 19]
Internet-Draft IPv6 over NBMA March 2023
As opposed to classical IGPs, an SGP must support rapid mobility of
addresses to cope with wireless devices and virtual machines
mobility. In that regard, an SGP operates mores as a MANET protocol
than as a classical IGP. Ideally, there should be no stale route,
and no microloop. A classical method in MANETs to achieve this is to
sequence the movements and advertise the sequence in the routing
protocol, so only routes with the most recent sequence can be
followed, and once a packet starts following a route with a certain
sequence, it must be discarded rather to have to follow a path with
an older sequence. To support this approach, the node that registers
an address must be the owner of a mobility sequence number and update
that sequence when it moves.
Multihoming being a classical requirement in DC environments, the SGP
must be able to differentiate not only address duplication from
movement, but also from anycast addresses, which can be advertised
from multiple places in a coordinated (same mobility sequence) or
uncoordinated fashion. For unicast addresses, an token that
identifies the address owner can be used for address duplication
avoidance, and if that token is cryptographic, it can be used as
registration ownership verifier as well.
3.2.6. Acronyms
This document uses the following abbreviations:
6BBR: 6LoWPAN Backbone Router
6LN: 6LoWPAN Node
6LR: 6LoWPAN Router
ARO: Address Registration Option
DAC: Duplicate Address Confirmation
DAD: Duplicate Address Detection
DAR: Duplicate Address Request
EDAC: Extended Duplicate Address Confirmation
EDAR: Extended Duplicate Address Request
MLSN: Multi-link subnet
LLN: Low-Power and Lossy Network
LoWPAN: Low-Power Wireless Personal Area Network
NA: Neighbor Advertisement
NBMA: Non-Broadcast Multi-Access
NCE: Neighbor Cache Entry
ND: Neighbor Discovery
NDP: Neighbor Discovery Protocol
NS: Neighbor Solicitation
RPL: IPv6 Routing Protocol for LLNs
RA: Router Advertisement
RS: Router Solicitation
SGP: Subnet Gateway Protocol
Thubert & Richardson Expires 4 September 2023 [Page 20]
Internet-Draft IPv6 over NBMA March 2023
VLAN: Virtual Local Area Network
WiND: Wireless Neighbor Discovery
WLAN: Wireless Local Area Network
WPAN: Wireless Personal Area Network
3.3. IP Models
3.3.1. Physical Broadcast Domain
At the physical (PHY) Layer, a broadcast domain is the set of nodes
that may receive a transmission that one sends over an interface, in
other words the set of nodes in range of the radio transmission.
This set can comprise a single peer on a serial cable used as point-
to-point link. It may also comprise multiple peer nodes on a
broadcast radio or a shared physical resource such as the Ethernet
wires and hubs for which ND-Classic was initially designed.
On WLAN and LoWPAN radios, the physical broadcast domain is defined
relative to a particular transmitter, as the set of nodes that can
receive what this transmitter is sending. Literally every frame
defines its own broadcast domain since the chances of reception of a
given frame are statistical. In average and in stable conditions,
the broadcast domain of a particular node can be still be seen as
mostly constant and can be used to define a closure of nodes on which
an upper Layer abstraction can be built.
A PHY Layer communication can be established between two nodes if the
physical broadcast domains of their unicast transmissions overlap.
On WLAN and LoWPAN radios, that relation is usually not reflexive,
since nodes disable the reception when they transmit; still they may
retain a copy of the transmitted frame, so it can be seen as
reflexive at the MAC Layer. It is often symmetric, meaning that if B
can receive a frame from A, then A can receive a frame from B. But
there can be asymmetries due to power levels, interferers near one of
the receivers, or differences in the quality of the hardware (e.g.,
crystals, PAs and antennas) that may affect the balance to the point
that the connectivity becomes mostly uni-directional, e.g., A to B
but practically not B to A.
It takes a particular effort to place a set of devices in a fashion
that all their physical broadcast domains fully overlap, and that
specific situation can not be assumed in the general case. In other
words, the relation of radio connectivity is generally not
transitive, meaning that A in range with B and B in range with C does
not necessarily imply that A is in range with C.
Thubert & Richardson Expires 4 September 2023 [Page 21]
Internet-Draft IPv6 over NBMA March 2023
3.3.2. link-layer Broadcast Emulations
We call Direct MAC Broadcast (DMB) the transmission mode where the
broadcast domain that is usable at the MAC layer is directly the
physical broadcast domain. IEEE Std. 802.15.4 [IEEE Std. 802.15.4]
and IEEE Std. 802.11 [IEEE Std. 802.11] OCB (for Out of the Context
of a BSS) are examples of DMB radios. DMB networks provide mostly
symmetric and non-transitive transmission. This contrasts with a
number of link-layer Broadcast Emulation (LLBE) schemes that are
described in this section.
In the case of Ethernet, while a physical broadcast domain is
constrained to a single shared wire, the IEEE Std. 802.1 [IEEE Std.
802.1] bridging function emulates the broadcast properties of that
wire over a whole physical mesh of Ethernet links. For the upper
layer, the qualities of the shared wire are essentially conserved,
with a reliable and cheap broadcast operation over a transitive
closure of nodes defined by their connectivity to the emulated wire.
In large switched fabrics, overlay techniques enable a limited
connectivity between nodes that are known to a Map Resolver. The
emulated broadcast domain is configured to the system, e.g., with a
VXLAN network identifier (VNID). Broadcast operations on the overlay
can be emulated but can become very expensive, and it makes sense to
proactively install the relevant state in the mapping server as
opposed to rely on reactive broadcast lookups to do so.
An IEEE Std. 802.11 [IEEE Std. 802.11] Infrastructure Basic Service
Set (BSS) also provides a transitive closure of nodes as defined by
the broadcast domain of a central AP. The AP relays both unicast and
broadcast packets and provides the symmetric and transitive emulation
of a shared wire between the associated nodes, with the capability to
signal link-up/link-down to the upper layer. Within a BSS, the
physical broadcast domain of the AP serves as emulated broadcast
domain for all the nodes that are associated to the AP. Broadcast
packets are relayed by the AP and are not acknowledged. To increase
the chances that all nodes in the BSS receive the broadcast
transmission, AP transmits at the slowest PHY speed. This translates
into maximum co-channel interferences for others and the longest
occupancy of the medium, for a duration that can be a hundred times
that of the unicast transmission of a frame of the same size.
For that reason, upper layer protocols should tend to avoid the use
of broadcast when operating over Wi-Fi. To cope with these problems,
APs may implement strategies such as turn a broadcast into a series
of unicast transmissions, or drop the message altogether, which may
impact the upper layer protocols. For instance, some APs may not
copy Router Solicitation (RS) messages under the assumption that
Thubert & Richardson Expires 4 September 2023 [Page 22]
Internet-Draft IPv6 over NBMA March 2023
there is no router across the wireless interface. This assumption
may be correct at some point of time and may become incorrect in the
future. Another strategy used in Wi-Fi APS is to proxy protocols
that heavily rely on broadcast, such as the Address Resolution in ARP
and ND-Classic, and either respond on behalf or preferably forward
the broadcast frame as a unicast to the intended Target.
In an IEEE Std. 802.11 [IEEE Std. 802.11] Infrastructure Extended
Service Set (ESS), infrastructure BSSes are interconnected by a
bridged network, typically running Transparent Bridging and the
Spanning tree Protocol or a more advanced Layer 2 Routing (L2R)
scheme. In the original model of learning bridges, the forwarding
state is set by observing the source MAC address of the frames. When
a state is missing for a destination MAC address, the frame is
broadcasted with the expectation that the response will populate the
state on the reverse path. This is a reactive operation, meaning
that the state is populated reactively to the need to reach a
destination. It is also possible in the original model to broadcast
a gratuitous frame to advertise self throughout the bridged network,
and that is also a broadcast.
The process of the Wi-Fi association prepares a bridging state
proactively at the AP, which avoids the need for a reactive broadcast
lookup over the wireless access. In an ESS, the AP may also generate
a gratuitous broadcast sourced at the MAC address of the STA to
prepare or update the state in the learning bridges so they point
towards the AP for the MAC address of the STA. WiND emulates that
proactive method at the Network Layer for the operations of AR, DAD
and ND proxy.
In some instances of WLANs and LoWPANs, a Mesh-Under technology
(e.g., a IEEE Std. 802.11s or IEEE Std. 802.15.10) provides meshing
services that are similar to bridging, and the broadcast domain is
well-defined by the membership of the mesh. Mesh-Under emulates a
broadcast domain by flooding the broadcast packets at the link-layer.
When operating on a single frequency, this operation is known to
interfere with itself, and requires inter-frame gaps to dampen the
collisions, which reduces further the amount of available bandwidth.
As the cost of broadcast transmissions becomes increasingly
expensive, there is a push to rethink the upper Layer protocols to
reduce the dependency on broadcast operations.
Thubert & Richardson Expires 4 September 2023 [Page 23]
Internet-Draft IPv6 over NBMA March 2023
3.3.3. Mapping the IPv6 link Abstraction
As introduced in Section 3.2.1, IPv6 defines a concept of link, link
scope and Link-Local Addresses (LLA), an LLA being unique and usable
only within the Scope of a Link. The ND-Classic [RFC4861] DAD
[RFC4862] process uses a multicast transmission to detect a duplicate
address, which requires that the owner of the address is connected to
the link-layer broadcast domain of the sender.
On a wired medium, the IP link is often confused with the physical
broadcast domain because both are determined by the serial cable or
the Ethernet shared wire. Ethernet Bridging reinforces that illusion
with a link-layer broadcast domain that emulates a physical broadcast
domain over the mesh of wires. But the difference shows on legacy
P2MP and NBMA networks such as ATM and Frame-Relay, on shared links
and on newer types of NBMA networks such as radio and composite
radio-wires networks. It also shows when private VLANs or link-layer
cryptography restrict the capability to read a frame to a subset of
the connected nodes.
In Mesh-Under and Infrastructure BSS, the IP link extends beyond the
physical broadcast domain to the emulated link-layer broadcast
domain. Relying on Multicast for the ND operation remains feasible
but becomes highly detrimental to the unicast traffic, and becomes
less and less energy-efficient and reliable as the network grows.
On DMB radios, IP links between peers come and go as the individual
physical broadcast domains of the transmitters meet and overlap. The
DAD operation cannot provide once and for all guarantees over the
broadcast domain defined by one radio transmitter if that transmitter
keeps meeting new peers on the go.
The scope on which the uniqueness of an LLA must be checked is each
new pair of nodes for the duration of their conversation. As long as
there's no conflict, a node may use the same LLA with multiple peers
but it has to perform DAD again with each new peer. A node may need
to form a new LLA to talk to a new peer, and multiple LLAs may be
present in the same radio interface to talk to different peers. In
practice, each pair of nodes defines a temporary P2P link, which can
be modeled as a sub-interface of the radio interface.
The DAD and AR procedures in ND-Classic expect that a node in a
subnet is reachable within the broadcast domain of any other node in
the subnet when that other node attempts to form an address that
would be a duplicate or attempts to resolve the MAC address of this
node. This is why ND is applicable for P2P and transit links, but
requires extensions for more complex topologies.
Thubert & Richardson Expires 4 September 2023 [Page 24]
Internet-Draft IPv6 over NBMA March 2023
3.3.4. Mapping the IPv6 subnet Abstraction
As introduced in Section 3.2.3, IPv6 also defines the concept of a
subnet for Global and Unique Local Addresses (GLA and ULA). All the
addresses in a subnet share the same prefix, and by extension, a node
belongs to a subnet if it has an address that derives from the prefix
of the subnet. That address must be topologically correct, meaning
that it must be installed on an interface that is connected to the
subnet.
Unless intently replicated in different locations for very specific
purposes, a subnet prefix is unique within a routing system; for
ULAs, the routing system is typically a limited domain, whereas for
GLAs, it is the whole Internet.
For that reason, it is sufficient to validate that an address that is
formed from a subnet prefix is unique within the scope of that subnet
to guarantee that it is globally unique within the whole routing
system. Note that a subnet may become partitioned due to the loss of
a wired or wireless link, so even that operation is not necessarily
obvious, more in [DAD APPROACHES].
The IPv6 aggregation model relies on the property that a packet from
the outside of a subnet can be routed to any router that belongs to
the subnet, and that this router will be able to either resolve the
destination link-layer address and deliver the packet, or, in the
case of an MLSN, route the packet to the destination within the
subnet.
If the subnet is known as on-link, then any node may also resolve the
destination link-layer address and deliver the packet, but if the
subnet is not on-link, then a host in the subnet that does not have a
Neighbor Cache Entry (NCE) for the destination will also need to pass
the packet to a router, more in [RFC5942].
On Ethernet, an IP subnet is often congruent with an IP link because
both are determined by the physical attachment to a shared wire or an
IEEE Std. 802.1 bridged domain. In that case, the connectivity over
the IP link is both symmetric and transitive, the subnet can appear
as on-link, and any node can resolve a destination MAC address of any
other node directly using ND-Classic.
But an IP link and an IP subnet are not always congruent. In the
case of a Shared Link, individual subnets may each encompass only a
subset of the nodes connected to the link. Conversely, in Route-Over
Multi-link subnets (MLSN) [RFC4903], routers federate the links
between nodes that belong to the subnet, the subnet is not on-link
and it extends beyond any of the federated links.
Thubert & Richardson Expires 4 September 2023 [Page 25]
Internet-Draft IPv6 over NBMA March 2023
3.4. Stateful Address Autoconfiguration and Subnet Routing
This Architecture defines a new operation for ND that is based on 2
major paradigm changes, a proactive address registration by hosts to
their attachment routers and routing to host routes (/128) within the
subnet. This allows ND to avoid the expectations of transit links
and subnet-wide broadcast domains.
The proactive address registration, called Stateful Address
Autoconfiguration (SFAAC) by opposition to SLAAC, is agnostic to the
method used for Address Assignment, e.g., Manual, Semantically Opaque
Autoconfiguration [RFC7217], randomized [RFC8981], or DHCPv6
[RFC8415]. It does not change the IPv6 addressing [RFC4291] or the
current practices of assigning prefixes, typically a /64, to a
subnet. But the DAD operation is performed as a unicast exchange
with the abstract egistrar service.
This Architecture combines SFAAC with the not-onlink model on the IP
Interfaces. Hosts do not expect the IP Subnet to be reachable over
the L2 broadcast domain and rely on their routers to forward the
packets inside and outside the Subnet. In turn, the router expose to
each other all the IPv6 addresses that are either owned or registered
to it as host routes over a Subnet Gateway Protocol, an IGP that is
specialized in routing inside the subnet and can be decoupled with
the routing protocol used between subnets.
4. A Framework for Stateful Address Autoconfiguration and Subnet
Routing
4.1. Implementing Stateful Address Autoconfiguration
SFAAC was initially standardized for IoT and wireless links as
[RFC6775], [RFC8505], and [RFC8928]. A new option in NS/NA messages,
the Extended Address Registration Option (EARO) signals that the
Target Address is being registered and provides the registration
parameters [RFC8505]. This method allows to prepare and maintain the
host routes in the routers and avoids the reactive Address Resolution
in ND-Classic and the associated link-layer broadcasts transmissions.
The EARO provides information to the router that is independent to
the routing protocol and routing can take multiple forms, from a
traditional IGP to a collapsed Hub-and-Spoke model where only one
router owns and advertises the prefix. [RFC8505] is already
referenced as the registration interface to "RIFT: Routing in Fat
Trees" [I-D.ietf-rift-rift] and "RPL: IPv6 Routing Protocol for
Low-Power and Lossy Networks" [RFC6550] with [RFC9010].
Thubert & Richardson Expires 4 September 2023 [Page 26]
Internet-Draft IPv6 over NBMA March 2023
Wireless ND (WiND) is an example instantiation of the Architecture
presented in Section 3; it combines SFAAC with a Backbone Router
(6BBR) ND proxy function (more in [RFC8929]) operating as a Layer-3
Access Point. Multiple 6BBRs placed along the wireless edge of a
Backbone link handle IPv6 Neighbor Discovery and forward packets over
the backbone on behalf of the registered nodes on the wireless edge.
This enables to span a subnet over an MLSN that federates edge
wireless links with a high-speed, typically Ethernet, backbone (as a
Layer-3 ESS). The ND proxy maintains the reachability for Global
Unicast and Link-Local Addresses within the federated MLSN, either as
a routing proxy where it replies with its own MAC address or as a
bridging proxy that typically forwards the multicast ND messages as
unicast Layer-2 frames to their target. The wireless nodes can form
any address they want and move freely from a wireless edge link to
another, without renumbering. In that case, the registrar is
distributed between the 6BBR, each 6BBR maintaining only a state for
the subset of the addresses that were registered to it and for which
it is authoritative. When the 6BBR is not currently authoritative
for a new address being registered to it, it relies on Classical ND
that is used reactively over the backbone to obtain an existing
registration state in
This framework allows other implementations of the abstract concept
of the registrar. For instance, [EVPN-SFAAC] allows to distribute
the registrar in every router, and leverages EVPN as the method to
synchronize the registrar state between routers. In that case, BGP
acts both as the SGP to announce the reachability of the addresses
and as the synchronization protocol between the distributed
registrar. All the routers know proactively the mapping for all the
addresses, and there is no need for a reactive lookup as is the case
for WiND. As another example, a Locator/ID Separation Protocol
(LISP) Map-Resolver [RFC6830] could support the EDAR/EDAC exchange
either directly or via a proxy, and serve as registrar.
The framework allows for mixed environments with registrations and
classical ND, using [RFC8929] to perform ND proxy operations on
behalf of registered address and respond to DAD and lookups from
legacy nodes, and prevent registering nodes from autoconfiguring
addresses that exist in legacy nodes by performing DAD on behalf of
the registering nodes, more in Section 7.
4.2. links and Link-Local Addresses
For Link-Local Addresses, DAD is typically performed between
communicating pairs of nodes and an NCE can be populated with a
single unicast exchange. In the case of a bridging proxies, though,
the Link-Local traffic is bridged over the backbone and the DAD must
proxied there as well.
Thubert & Richardson Expires 4 September 2023 [Page 27]
Internet-Draft IPv6 over NBMA March 2023
For instance, in the case of Bluetooth Low Energy (BLE)
[RFC7668][IEEEstd802151], the uniqueness of Link-Local Addresses
needs only to be verified between the pair of communicating nodes,
the central router and the peripheral host. In that example, 2
peripheral hosts connected to the same central router can not have
the same Link-Local Address because the addresses would collision at
the central router which could not talk to both over the same
interface. The DAD operation from SFAAC is appropriate for that use
case, but the one from ND is not, because the peripheral hosts are
not on the same broadcast domain.
On the other hand, the uniqueness of Global and Unique-Local
Addresses is validated at the subnet Level, using a logical registrar
that is global to the subnet.
4.3. Subnets and Global Addresses
SFAAC extends ND-Classic for Hub-and-Spoke (e.g., BLE) and Route-Over
(e.g., RPL) Multi-link subnets (MLSNs).
In the Hub-and-Spoke case, each Hub-Spoke pair is a distinct IP Link,
and a subnet can be mapped on a collection of links that are
connected to the Hub. The subnet prefix is associated to the Hub.
Acting as a router, the Hub advertises the prefix as not-on-link to
the spokes in RA messages Prefix Information Options (PIO). Acting
as hosts, the Spokes autoconfigure addresses from that prefix and
register them to the Hub with a corresponding lifetime.
Acting as a registrar, the Hub maintains a binding table of all the
registered IP addresses and rejects duplicate registrations, thus
ensuring a DAD protection for a registered address even if the
registering node is sleeping.
The Hub also maintains an NCE for the registered addresses and can
deliver a packet to any of them during their respective lifetimes.
It can be observed that this design builds a form of Network Layer
Infrastructure BSS.
A Route-Over MLSN is considered as a collection of Hub-and-Spoke
where the Hubs form a connected dominating set of the member nodes of
the subnet, and IPv6 routing takes place between the Hubs within the
subnet. A single logical registrar is deployed to serve the whole
mesh.
The registration in [RFC8505] is abstract to the routing protocol and
provides enough information to feed a routing protocol such as RPL as
specified in [RFC9010]. In a degraded mode, all the Hubs are
Thubert & Richardson Expires 4 September 2023 [Page 28]
Internet-Draft IPv6 over NBMA March 2023
connected to a same high speed backbone such as an Ethernet bridging
domain where ND-Classic is operated. In that case, it is possible to
federate the Hub, Spoke and Backbone nodes as a single subnet,
operating ND proxy operations [RFC8929] at the Hubs, acting as 6BBRs.
It can be observed that this latter design builds a form of Network
Layer Infrastructure ESS.
4.4. Anycast and Multicast Addresses
While IPv6 ND is defined for unicast addresses only,
[I-D.ietf-6lo-multicast-registration] extends [RFC8505] for anycast
and multicast IPv6 addresses.
[I-D.ietf-6lo-multicast-registration] can be used as a replacement
for MLDv2 [RFC3810] for use cases where broadcast are not desirable,
and when a device push model such as SFAAC is preferred over a
network pull such as MDv2 and classical ND. With [RFC8505], the host
does not need to define SNMAs for its unicast addresses and does not
perform the associated MLDv2 operation. With
[I-D.ietf-6lo-multicast-registration], MLDv2 and its extensive use of
broadcast can be totally eliminated.
In the case of anycast, the signal enables the 6BBRs to accept more
than one registration for the same address, and collectively elect
the registering host receives a packet for a given anycast address.
5. WiND Applicability
WiND applies equally to P2P links, P2MP Hub-and-Spoke, link-layer
Broadcast Domain Emulation such as Mesh-Under and Wi-Fi BSS, and
Route-Over meshes.
There is an intersection where The IP link and the IP subnet are
congruent and where both ND and WiND could apply. These includes
P2P, the MAC emulation of a PHY broadcast domain, and the particular
case of always on, fully overlapping physical radio broadcast domain.
But even in those cases where both are possible, WiND is preferable
vs. ND because it reduces the need of broadcast.
This is discussed in more details in the introduction of [RFC8929].
There are also a number of practical use cases in the wireless world
where links and subnets are not congruent:
Thubert & Richardson Expires 4 September 2023 [Page 29]
Internet-Draft IPv6 over NBMA March 2023
* The IEEE Std. 802.11 infrastructure BSS enables one subnet per AP,
and emulates a broadcast domain at the link-layer. The
Infrastructure ESS extends that model over a backbone and
recommends the use of an ND proxy [IEEE Std. 802.11] to
interoperate with Ethernet-connected nodes. WiND incorporates an
ND proxy to serve that need, which was missing so far.
* Bluetooth is Hub-and-Spoke at the link-layer. It would make
little sense to configure a different subnet between the central
and each individual peripheral node (e.g., sensor). Rather,
[RFC7668] allocates a prefix to the central node acting as router,
and each peripheral host (acting as a host) forms one or more
address(es) from that same prefix and registers it.
* A typical SmartGrid networks puts together Route-Over MLSNs that
comprise thousands of IPv6 nodes. The 6TiSCH architecture
[RFC9030] presents the Route-Over model over an IEEE Std. 802.15.4
Time-Slotted Channel-Hopping (TSCH) [IEEEstd802154] mesh, and
generalizes it for multiple other applications.
Each node in a SmartGrid network may have tens to a hundred others
nodes in range. A key problem for the routing protocol is which
other node(s) should this node peer with, because most of the
possible peers do not provide added routing value. When both
energy and bandwidth are constrained, talking to them is a waste
of resources and most of the possible P2P links are not even used.
Peerings that are actually used come and go with the dynamics of
radio signal propagation. It results that allocating prefixes to
all the possible P2P links and maintain as many addresses in all
nodes is not even considered.
5.1. Case of LPWANs
LPWANs are by nature so constrained that the addresses and subnets
are fully pre-configured and operate as P2P or Hub-and-Spoke. This
saves the steps of neighbor Discovery and enables a very efficient
stateful compression of the IPv6 header.
5.2. Case of Infrastructure BSS and ESS
In contrast to IPv4, IPv6 enables a node to form multiple addresses,
some of them temporary to elusive, and with a particular attention
paid to privacy. Addresses may be formed and deprecated
asynchronously to the association.
Snooping protocols such as ND-Classic and DHCPv6 and observing data
traffic sourced at the STA provides an imperfect knowledge of the
state of the STA at the AP. Missing a state or a transition may
Thubert & Richardson Expires 4 September 2023 [Page 30]
Internet-Draft IPv6 over NBMA March 2023
result in the loss of connectivity for some of the addresses, in
particular for an address that is rarely used, belongs to a sleeping
node, or one in a situation of mobility. This may also result in
undesirable remanent state in the AP when the STA ceases to use an
IPv6 address while remaining associated. It results that snooping
protocols is not a recommended technique and that it should only be
used as last resort, when the WiND registration is not available to
populate the state.
The recommended alternative method is to use the WiND Registration
for IPv6 Addresses. This way, the AP exposes its capability to proxy
ND to the STA in Router Advertisement messages. In turn, the STA may
request proxy ND services from the AP for all of its IPv6 addresses,
using the Extended Address Registration Option, which provides the
following elements:
* The registration state has a lifetime that limits unwanted state
remanence in the network.
* The registration is optionally secured using [RFC8928] to prevent
address theft and impersonation.
* The registration carries a sequence number, which enables to
figure the order of events in a fast mobility scenario without
loss of connectivity.
The ESS mode requires a proxy ND operation at the AP. The proxy ND
operation must cover Duplicate Address Detection, Neighbor
Unreachability Detection, Address Resolution and Address Mobility to
transfer a role of ND proxy to the AP where a STA is associated
following the mobility of the STA.
The WiND proxy ND specification that associated to the Address
Registration is [RFC8929]. With that specification, the AP
participates to the protocol as a Backbone Router, typically
operating as a bridging proxy though the routing proxy operation is
also possible. As a bridging proxy, the backbone router either
replies to NS lookups with the MAC address of the STA, or preferably
forwards the lookups to the STA as link-layer unicast frames to let
the STA answer. For the data plane, the backbone router acts as a
normal AP and bridges the packets to the STA as usual. As a routing
proxy, the backbone router replies with its own MAC address and then
routes to the STA at the IP layer. The routing proxy reduces the
need to expose the MAC address of the STA on the wired side, for a
better stability and scalability of the bridged fabric.
Thubert & Richardson Expires 4 September 2023 [Page 31]
Internet-Draft IPv6 over NBMA March 2023
5.3. Case of Mesh Under Technologies
The Mesh-Under provides a broadcast domain emulation with symmetric
and Transitive properties and defines a transit link for IPv6
operations. It results that the model for IPv6 operation is similar
to that of a BSS, with the root of the mesh operating as an Access
Point does in a BSS/ESS.
While it is still possible to operate ND-Classic, the inefficiencies
of the flooding operation make the associated operations even less
desirable than in a BSS, and the use of WiND is highly recommended.
5.4. Case of DMB radios
IPv6 over DMB radios uses P2P links that can be formed and maintained
when a pair of DMB radios transmitters are in range from one another.
5.4.1. Using ND-Classic only
DMB radios do not provide MAC level broadcast emulation. An example
of that is IEEE Std. 802.11 OCB which uses IEEE Std. 802.11 MAC/PHYs
but does not provide the BSS functions.
It is possible to form P2P IP links between each individual pairs of
nodes and operate ND-Classic over those links with Link-Local
addresses. DAD must be performed for all addresses on all P2P IP
links.
If special deployment care is taken so that the physical broadcast
domains of a collection of the nodes fully overlap, then it is also
possible to build an IP subnet within that collection of nodes and
operate ND-Classic.
If an external mechanism avoids duplicate addresses and if the
deployment ensures the connectivity between peers, a non-transit Hub-
and-Spoke deployment is also possible where the Hub is the only
router in the subnet and the Prefix is advertised as not on-link.
5.4.2. Using Wireless ND
Though this can be achieved with ND-Classic, WiND is the recommended
approach since it uses unicast communications which are more reliable
and less impacting for other users of the medium.
Thubert & Richardson Expires 4 September 2023 [Page 32]
Internet-Draft IPv6 over NBMA March 2023
The routers send RAs with a SLLAO at a regular period. The period
can be indicated in the RA-Interval Option [RFC6275]. If available,
the message can be transported in a compressed form in a beacon,
e.g., in OCB Basic Safety Messages (BSM) that are nominally sent
every 100ms.
An active beaconing mode is possible whereby the Host sends broadcast
RS messages to which a router can answer with a unicast RA.
A router that has Internet connectivity and is willing to serve as an
Internet Access may advertise itself as a default router [RFC4191] in
its RA messages. The RA is sent over an unspecified IP link where it
does not conflict to anyone, so DAD is not necessary at that stage.
The host instantiates an IP link where the router's address is not a
duplicate. To achieve this, it forms a Link-Local Address that does
not conflict with that of the router and registers to the router
using [RFC8505]. If the router sent an RA(PIO), the host can also
autoconfigure an address from the advertised prefix and register it.
(host) (router)
| |
| DMB link |
| |
| IPv6 ND RS |
|-------------->|
|-----------> |
|------------------>
| IPv6 ND RA |
|<--------------|
| |
| NS(EARO) |
|-------------->|
| |
| NA(EARO) |
|<--------------|
| |
...
| |
| NS(EARO) |
|-------------->|
| |
| NA(EARO) |
|<--------------|
| |
Figure 4: RFC 8505 Registration Flow for a Link-Local Address
Thubert & Richardson Expires 4 September 2023 [Page 33]
Internet-Draft IPv6 over NBMA March 2023
The lifetime in the registration should start with a small value
(X=RMin, TBD), and exponentially grow with each re-registration to a
larger value (X=Rmax, TBD). The IP link is considered down when
(X=NbBeacons, TDB) expected messages are not received in a row. It
must be noted that the physical link flapping does not affect the
state of the registration and when a physical link comes back up, the
active registrations (i.e., registrations for which lifetime is not
elapsed) are still usable. Packets should be held or destroyed when
the IP link is down.
P2P links may be federated in Hub-and-Spoke by edge routers, and the
Subnet may comprise multiple edge routers, in which case each
advertises its registered addresses over the SGP as illustrated in
Figure 5. Note that the Extended DAR/DAC exchange can be omitted if
it can be replaced with the information that is distributed in the
SGP, see for instance [RFC9010] which applies to IoT environments,
which needs only the first EDAR/EDAC exchange, and [EVPN-SFAAC], for
EVPN-based wireless deployments in enterprise and campus, which does
not use EDAR/EDAC at all.
Thubert & Richardson Expires 4 September 2023 [Page 34]
Internet-Draft IPv6 over NBMA March 2023
6LoWPAN Node 6LR 6LBR
(host) (router) (registrar)
| | |
| | |
| IPv6 ND RS | |
|-------------->| |
|-----------> | |
|------------------> |
| IPv6 ND RA | |
|<--------------| |
| | |
| NS(EARO) | |
|-------------->| |
| | Extended DAR |
| |-------------->|
| | |
| | Extended DAC |
| |<--------------|
| NA(EARO) |
|<--------------|<inject in SGP> ->
...
| | |
| NS(EARO) | |
|-------------->| |
| | Extended DAR |
| |-------------->|
| | |
| | Extended DAC |
| |<--------------|
| NA(EARO) | |
|<--------------|<maintain in SGP> ->
| | |
...
Figure 5: RFC 8505 Registration Flow for a Global Address
An example Hub-and-Spoke is an OCB Road-Side Unit (RSU) that owns a
prefix, provides Internet connectivity using that prefix to On-Board
Units (OBUs) within its physical broadcast domain. An example of
Route-Over MLSN is a collection of cars in a parking lot operating
RPL to extend the connectivity provided by the RSU beyond its
physical broadcast domain. Cars may then operate NEMO [RFC3963] for
their own prefix using their address derived from the prefix of the
RSU as CareOf Address.
Thubert & Richardson Expires 4 September 2023 [Page 35]
Internet-Draft IPv6 over NBMA March 2023
6. IANA Considerations
This specification does not require IANA action.
7. Coexistence with IPv6 ND
The framework allows for a mixed environment with both models,
classical ND and SFAAC, coexist. With [RFC8929], an ethernet
backbone link operating Classical ND federates a MultiLink Subnet
(MLSN) of wireless links and/or meshes, and routers called Backbone
Routers (6BBR) operate as ND proxies.
In a wireless deployments, the Backbone Routers are placed along the
wireless edge of a backbone (e.g., in Access Points) and federate
multiple wireless links to form a single MLSN, echoing the Wi-Fi ESS
structure but at Layer-3, as shown in Figure 6. In that example,
Optimistic Duplicate Address Detection (ODAD) [RFC4429] allows the
IPv6 address to be used before completion of DAD, so the whole flow
below can happen in the milliseconds that follow the Wi-Fi
association.
Thubert & Richardson Expires 4 September 2023 [Page 36]
Internet-Draft IPv6 over NBMA March 2023
6LN(STA) 6BBR(AP) 6LBR default GW
| | | |
| Wi-Fi Access BSS | IPv6 Backbone (e.g., Ethernet) |
| | | |
| RS(multicast) | | |
|----------------->| | |
| RA(PIO, Unicast) | | |
|<-----------------| | |
| NS(EARO) | | |
|----------------->| | |
| | Extended DAR | |
| |--------------->| |
| | Extended DAC | |
| |<---------------| |
| | |
| | NS-DAD(EARO, multicast) |
| |--------> |
| |-------------------> |
| |----------------------------------->|
| | |
| | RS(no SLLAO, for ODAD) |
| |----------------------------------->|
| | if (no fresher Binding) NS(Lookup) |
| | <----------------|
| | <-------------------------|
| |<-----------------------------------|
| | NA(SLLAO, not(O), EARO) |
| |----------------------------------->|
| | RA(unicast) |
| |<-----------------------------------|
| | |
| IPv6 Packets in Optimistic Mode |
|<----------------------------------------------------->|
| | |
| |
| NA(EARO) | <DAD timeout>
|<--------- -------|
| |
Figure 6: Initial Registration Flow to a 6BBR Acting as a ND Proxy
In use cases such as overlays, a Map Resolver acting as 6LBR may be
deployed on the Backbone Link to serve the whole subnet, and EDAR/
EDAC messages (or equivalent alternates, e.g., using LISP) can be
used in combination with DAD to enable coexistence with IPv6 ND over
the backbone. The 6LBR proactive operations will then coexist on the
Backbone with the reactive classical IPv6 ND operation. Nodes that
support [UNICAST AR] may query the mappings they look up with the
Thubert & Richardson Expires 4 September 2023 [Page 37]
Internet-Draft IPv6 over NBMA March 2023
6LBR before attempting the reactive operation, which may be avoided
if the 6LBR is conclusive, either detecting a duplication or
returning a mapping. This model also enables a snooping switch
acting as ND proxy to intercept Ar and DAD NS messages and perform
unicast lookups to the resolver and only broadcast the original NS
messgae when the unicast lookup fails.
Note that the RS sent initially by the 6LN (e.g., a Wi-Fi STA) is
transmitted as a multicast, but since it is intercepted by the 6BBR,
it is never effectively broadcast at layer-2. The multiple arrows in
Figure 6 associated to the ND messages on the backbone denote a real
Layer-2 broadcast.
It is not necessary to isolate the registering nodes in separate
physical links, but it is preferred with wireless links as it enables
to isolate the broadcast domain on the ethernet link from the
wireless links at the Access Points. In other words, the 6BBRs
collectively form a global registrar for the Subnet that aggregates
the information in each local registrar in the 6LBR. The global
registrar is distributed between the 6BBRs, which leverage classical
IPv6 ND (AR and DAD) to lookup information that they do not have
locally from the other 6BBRs and from nodes that are connected to the
backbone.
In the case of wireless meshes, RPL may be used as local SGP in each
mesh as shown in Figure 7. More details on the operation of WiND and
RPL over the MLSN can be found in section 3.1, 3.2, 4.1 and 4.2.2 of
[RFC9030].
Thubert & Richardson Expires 4 September 2023 [Page 38]
Internet-Draft IPv6 over NBMA March 2023
6LoWPAN Node 6LR 6LBR 6BBR
(RPL leaf) (router) (root)
| | | |
| 6LoWPAN ND |6LoWPAN ND+RPL | 6LoWPAN ND | ND-Classic
| LLN link |Route-Over mesh|Ethernet/serial| Backbone
| | | |
| IPv6 ND RS | | |
|-------------->| | |
|-----------> | | |
|------------------> | |
| IPv6 ND RA | | |
|<--------------| | |
| | <once> | |
| NS(EARO) | | |
|-------------->| | |
| 6LoWPAN ND | Extended DAR | |
| |-------------->| |
| | | NS(EARO) |
| | |-------------->|
| | | proxy registration
| | | |
| | | NS-DAD (EARO)
| | | |------>
| | | ND proxy
| | | |
| | | NA(EARO) |<timeout>
| | |<--------------|
| | Extended DAC | |
| |<--------------| |
| NA(EARO) | | |
|<--------------| | |
| | | |
Figure 7: Initial Registration Flow with 6BBR ND-Proxy
8. Privacy Considerations
Classical ND exposes all addresses to all nodes in the Subnet, which
is a privacy issue and makes impersonation attacks easier. In
contrast, in switched and wireless networks, a host is not on-path of
the unicast packets for registration and for data for other hosts, so
it cannot snoop the other addresses in the network. A rogue host can
only discover the existence of an addresses by trying and failing to
register that address, but for that it would need to fathom which
address to try and that can be very hard in a /64 address space taht
is used wisely. For that reason, this framework limits that
knowledge to on-path snooping switches, to the routers and to the
abstract registrar, which are typically more controlled / harder to
Thubert & Richardson Expires 4 September 2023 [Page 39]
Internet-Draft IPv6 over NBMA March 2023
hack than the common host. When classical ND and SFAAC coexist
within the same Subnet, all addresses in the Subnet, including
registered addresses, can be snooped in the broadcast domain where
classical IPv6 ND is operated. It makes sense to reduce that domain
to the maximum and control which device connect to it.
The exposure of addresses can be further reduced if the exchanges
with the registrar (e.g., EDAR and EDAC) are encrypted, e.g., using a
public key associated with the registrar. The registration and
routing exchanges could also be encrypted to avoid leaking the
addresses to snopping switches, but this is typically not done inside
a physical site where the networking gear is tightly controlled. In
a DCI environment, the inter-side (SD-WAN) links are typically
encrypted, to the exchanges are obfuscated from an on-path listener.
9. Security Considerations
The registration model [RFC8505] implemented by this framework allows
for a model where the ingress routers have a full knowledge of all
the addresses in the Subnet. The ingress router can thus discard any
packet which destination appears to be in the Subnet from its prefix,
but is not known, meaning that it does not exist. This mostly
defeats the traditional DoS scanning attacks against ND whereby the
remote attacker sends volumes of packets to as many non-existent
addresses to saturate the Neighbor Cache and clog the Subnet internal
bandwidth in broadcasts.
When the ownership verifier is cryptographic, this framework enables
a zerotrust model whereby only the address owner can advertise an
address in ND and as source of data packets, more in [RFC8928]. This
defeats the classical impersonation attacks against IPv6 ND and
allows to disable the proprietary middlebox software aimed at
protecting the address ownership against onlink rogues.
10. Acknowledgments
Many thanks to the participants of the 6lo WG where a lot of the work
discussed here happened, following work at ROLL, 6TiSCH, and mainly
6LoWPAN.
Special thanks to Brian Carpenter and Eric Levy-Abegnoli who provided
support and useful comments throughout the development of this
architecture, and to Erik Nordmark and Zach Shelby with whom this
work really started during IETF 72 in Dublin.
Also many thanks to Eduard Vasilenko, XiPeng Xiao, for their
contributions and support to this work at 6MAN, v6Ops, and ETSI IPE.
Thubert & Richardson Expires 4 September 2023 [Page 40]
Internet-Draft IPv6 over NBMA March 2023
11. Normative References
[RFC3963] Devarapalli, V., Wakikawa, R., Petrescu, A., and P.
Thubert, "Network Mobility (NEMO) Basic Support Protocol",
RFC 3963, DOI 10.17487/RFC3963, January 2005,
<https://www.rfc-editor.org/info/rfc3963>.
[RFC4191] Draves, R. and D. Thaler, "Default Router Preferences and
More-Specific Routes", RFC 4191, DOI 10.17487/RFC4191,
November 2005, <https://www.rfc-editor.org/info/rfc4191>.
[RFC4861] Narten, T., Nordmark, E., Simpson, W., and H. Soliman,
"Neighbor Discovery for IP version 6 (IPv6)", RFC 4861,
DOI 10.17487/RFC4861, September 2007,
<https://www.rfc-editor.org/info/rfc4861>.
[RFC4862] Thomson, S., Narten, T., and T. Jinmei, "IPv6 Stateless
Address Autoconfiguration", RFC 4862,
DOI 10.17487/RFC4862, September 2007,
<https://www.rfc-editor.org/info/rfc4862>.
[RFC5942] Singh, H., Beebee, W., and E. Nordmark, "IPv6 Subnet
Model: The Relationship between Links and Subnet
Prefixes", RFC 5942, DOI 10.17487/RFC5942, July 2010,
<https://www.rfc-editor.org/info/rfc5942>.
[RFC6275] Perkins, C., Ed., Johnson, D., and J. Arkko, "Mobility
Support in IPv6", RFC 6275, DOI 10.17487/RFC6275, July
2011, <https://www.rfc-editor.org/info/rfc6275>.
[RFC6830] Farinacci, D., Fuller, V., Meyer, D., and D. Lewis, "The
Locator/ID Separation Protocol (LISP)", RFC 6830,
DOI 10.17487/RFC6830, January 2013,
<https://www.rfc-editor.org/info/rfc6830>.
[RFC8200] Deering, S. and R. Hinden, "Internet Protocol, Version 6
(IPv6) Specification", STD 86, RFC 8200,
DOI 10.17487/RFC8200, July 2017,
<https://www.rfc-editor.org/info/rfc8200>.
[RFC8505] Thubert, P., Ed., Nordmark, E., Chakrabarti, S., and C.
Perkins, "Registration Extensions for IPv6 over Low-Power
Wireless Personal Area Network (6LoWPAN) Neighbor
Discovery", RFC 8505, DOI 10.17487/RFC8505, November 2018,
<https://www.rfc-editor.org/info/rfc8505>.
Thubert & Richardson Expires 4 September 2023 [Page 41]
Internet-Draft IPv6 over NBMA March 2023
[RFC8928] Thubert, P., Ed., Sarikaya, B., Sethi, M., and R. Struik,
"Address-Protected Neighbor Discovery for Low-Power and
Lossy Networks", RFC 8928, DOI 10.17487/RFC8928, November
2020, <https://www.rfc-editor.org/info/rfc8928>.
[RFC8929] Thubert, P., Ed., Perkins, C.E., and E. Levy-Abegnoli,
"IPv6 Backbone Router", RFC 8929, DOI 10.17487/RFC8929,
November 2020, <https://www.rfc-editor.org/info/rfc8929>.
12. Informative References
[RFC3810] Vida, R., Ed. and L. Costa, Ed., "Multicast Listener
Discovery Version 2 (MLDv2) for IPv6", RFC 3810,
DOI 10.17487/RFC3810, June 2004,
<https://www.rfc-editor.org/info/rfc3810>.
[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>.
[RFC4429] Moore, N., "Optimistic Duplicate Address Detection (DAD)
for IPv6", RFC 4429, DOI 10.17487/RFC4429, April 2006,
<https://www.rfc-editor.org/info/rfc4429>.
[RFC4903] Thaler, D., "Multi-Link Subnet Issues", RFC 4903,
DOI 10.17487/RFC4903, June 2007,
<https://www.rfc-editor.org/info/rfc4903>.
[RFC6550] Winter, T., Ed., Thubert, P., Ed., Brandt, A., Hui, J.,
Kelsey, R., Levis, P., Pister, K., Struik, R., Vasseur,
JP., and R. Alexander, "RPL: IPv6 Routing Protocol for
Low-Power and Lossy Networks", RFC 6550,
DOI 10.17487/RFC6550, March 2012,
<https://www.rfc-editor.org/info/rfc6550>.
[RFC6775] Shelby, Z., Ed., Chakrabarti, S., Nordmark, E., and C.
Bormann, "Neighbor Discovery Optimization for IPv6 over
Low-Power Wireless Personal Area Networks (6LoWPANs)",
RFC 6775, DOI 10.17487/RFC6775, November 2012,
<https://www.rfc-editor.org/info/rfc6775>.
[RFC7668] Nieminen, J., Savolainen, T., Isomaki, M., Patil, B.,
Shelby, Z., and C. Gomez, "IPv6 over BLUETOOTH(R) Low
Energy", RFC 7668, DOI 10.17487/RFC7668, October 2015,
<https://www.rfc-editor.org/info/rfc7668>.
Thubert & Richardson Expires 4 September 2023 [Page 42]
Internet-Draft IPv6 over NBMA March 2023
[RFC7217] Gont, F., "A Method for Generating Semantically Opaque
Interface Identifiers with IPv6 Stateless Address
Autoconfiguration (SLAAC)", RFC 7217,
DOI 10.17487/RFC7217, April 2014,
<https://www.rfc-editor.org/info/rfc7217>.
[RFC7834] Saucez, D., Iannone, L., Cabellos, A., and F. Coras,
"Locator/ID Separation Protocol (LISP) Impact", RFC 7834,
DOI 10.17487/RFC7834, April 2016,
<https://www.rfc-editor.org/info/rfc7834>.
[RFC8273] Brzozowski, J. and G. Van de Velde, "Unique IPv6 Prefix
per Host", RFC 8273, DOI 10.17487/RFC8273, December 2017,
<https://www.rfc-editor.org/info/rfc8273>.
[RFC8365] Sajassi, A., Ed., Drake, J., Ed., Bitar, N., Shekhar, R.,
Uttaro, J., and W. Henderickx, "A Network Virtualization
Overlay Solution Using Ethernet VPN (EVPN)", RFC 8365,
DOI 10.17487/RFC8365, March 2018,
<https://www.rfc-editor.org/info/rfc8365>.
[RFC8415] Mrugalski, T., Siodelski, M., Volz, B., Yourtchenko, A.,
Richardson, M., Jiang, S., Lemon, T., and T. Winters,
"Dynamic Host Configuration Protocol for IPv6 (DHCPv6)",
RFC 8415, DOI 10.17487/RFC8415, November 2018,
<https://www.rfc-editor.org/info/rfc8415>.
[RFC8981] Gont, F., Krishnan, S., Narten, T., and R. Draves,
"Temporary Address Extensions for Stateless Address
Autoconfiguration in IPv6", RFC 8981,
DOI 10.17487/RFC8981, February 2021,
<https://www.rfc-editor.org/info/rfc8981>.
[I-D.ietf-rift-rift]
Przygienda, T., Sharma, A., Thubert, P., Rijsman, B.,
Afanasiev, D., and J. Head, "RIFT: Routing in Fat Trees",
Work in Progress, Internet-Draft, draft-ietf-rift-rift-16,
12 September 2022, <https://datatracker.ietf.org/doc/html/
draft-ietf-rift-rift-16>.
[RFC9010] Thubert, P., Ed. and M. Richardson, "Routing for RPL
(Routing Protocol for Low-Power and Lossy Networks)
Leaves", RFC 9010, DOI 10.17487/RFC9010, April 2021,
<https://www.rfc-editor.org/info/rfc9010>.
[DAD ISSUES]
Yourtchenko, A. and E. Nordmark, "A survey of issues
related to IPv6 Duplicate Address Detection", Work in
Thubert & Richardson Expires 4 September 2023 [Page 43]
Internet-Draft IPv6 over NBMA March 2023
Progress, Internet-Draft, draft-yourtchenko-6man-dad-
issues-01, 3 March 2015,
<https://datatracker.ietf.org/doc/html/draft-yourtchenko-
6man-dad-issues-01>.
[MCAST EFFICIENCY]
Vyncke, E., Thubert, P., Levy-Abegnoli, E., and A.
Yourtchenko, "Why Network-Layer Multicast is Not Always
Efficient At Datalink Layer", Work in Progress, Internet-
Draft, draft-vyncke-6man-mcast-not-efficient-01, 14
February 2014, <https://datatracker.ietf.org/doc/html/
draft-vyncke-6man-mcast-not-efficient-01>.
[RFC9030] Thubert, P., Ed., "An Architecture for IPv6 over the Time-
Slotted Channel Hopping Mode of IEEE 802.15.4 (6TiSCH)",
RFC 9030, DOI 10.17487/RFC9030, May 2021,
<https://www.rfc-editor.org/info/rfc9030>.
[MCAST PROBLEMS]
Perkins, C. E., McBride, M., Stanley, D., Kumari, W. A.,
and J. C. Zúñiga, "Multicast Considerations over IEEE 802
Wireless Media", Work in Progress, Internet-Draft, draft-
ietf-mboned-ieee802-mcast-problems-15, 28 July 2021,
<https://datatracker.ietf.org/doc/html/draft-ietf-mboned-
ieee802-mcast-problems-15>.
[SAVI] Bi, J., Wu, J., Lin, T., Wang, Y., and L. He, "A SAVI
Solution for WLAN", Work in Progress, Internet-Draft,
draft-bi-savi-wlan-24, 13 November 2022,
<https://datatracker.ietf.org/doc/html/draft-bi-savi-wlan-
24>.
[UNICAST AR]
Thubert, P. and E. Levy-Abegnoli, "IPv6 Neighbor Discovery
Unicast Lookup", Work in Progress, Internet-Draft, draft-
thubert-6lo-unicast-lookup-02, 8 November 2021,
<https://datatracker.ietf.org/doc/html/draft-thubert-6lo-
unicast-lookup-02>.
[DAD APPROACHES]
Nordmark, E., "Possible approaches to make DAD more robust
and/or efficient", Work in Progress, Internet-Draft,
draft-nordmark-6man-dad-approaches-02, 19 October 2015,
<https://datatracker.ietf.org/doc/html/draft-nordmark-
6man-dad-approaches-02>.
Thubert & Richardson Expires 4 September 2023 [Page 44]
Internet-Draft IPv6 over NBMA March 2023
[EVPN-SFAAC]
Thubert, P., Przygienda, T., and J. Tantsura, "Secure EVPN
MAC Signaling", Work in Progress, Internet-Draft, draft-
thubert-bess-secure-evpn-mac-signaling-03, 31 January
2022, <https://datatracker.ietf.org/doc/html/draft-
thubert-bess-secure-evpn-mac-signaling-03>.
[I-D.ietf-6lo-multicast-registration]
Thubert, P., "IPv6 Neighbor Discovery Multicast and
Anycast Address Listener Subscription", Work in Progress,
Internet-Draft, draft-ietf-6lo-multicast-registration-12,
22 November 2022, <https://datatracker.ietf.org/doc/html/
draft-ietf-6lo-multicast-registration-12>.
[IEEE Std. 802.15.4]
IEEE standard for Information Technology, "IEEE Std.
802.15.4, Part. 15.4: Wireless Medium Access Control (MAC)
and Physical Layer (PHY) Specifications for Low-Rate
Wireless Personal Area Networks".
[IEEE Std. 802.11]
IEEE standard for Information Technology, "IEEE Standard
for Information technology -- Telecommunications and
information exchange between systems Local and
metropolitan area networks-- Specific requirements Part
11: Wireless LAN Medium Access Control (MAC) and Physical
Layer (PHY) Specifications".
[IEEEstd802151]
IEEE standard for Information Technology, "IEEE Standard
for Information Technology - Telecommunications and
Information Exchange Between Systems - Local and
Metropolitan Area Networks - Specific Requirements. - Part
15.1: Wireless Medium Access Control (MAC) and Physical
Layer (PHY) Specifications for Wireless Personal Area
Networks (WPANs)".
[IEEEstd802154]
IEEE standard for Information Technology, "IEEE Standard
for Local and metropolitan area networks -- Part 15.4:
Low-Rate Wireless Personal Area Networks (LR-WPANs)".
[IEEE Std. 802.1]
IEEE standard for Information Technology, "IEEE Standard
for Information technology -- Telecommunications and
information exchange between systems Local and
metropolitan area networks Part 1: Bridging and
Architecture".
Thubert & Richardson Expires 4 September 2023 [Page 45]
Internet-Draft IPv6 over NBMA March 2023
Authors' Addresses
Pascal Thubert (editor)
Cisco Systems, Inc
Building D
45 Allee des Ormes - BP1200
06254 Mougins - Sophia Antipolis
France
Phone: +33 497 23 26 34
Email: pthubert@cisco.com
Michael C. Richardson
Sandelman Software Works
Email: mcr+ietf@sandelman.ca
URI: http://www.sandelman.ca/
Thubert & Richardson Expires 4 September 2023 [Page 46]