draft-ietf-ngtrans-isatap-09.txt   draft-ietf-ngtrans-isatap-10.txt 
NGTRANS Working Group F. Templin NGTRANS Working Group F. Templin
Internet-Draft Nokia Internet-Draft Nokia
Expires: July 1, 2003 T. Gleeson Expires: July 4, 2002 T. Gleeson
Cisco Systems K.K. Cisco Systems K.K.
M. Talwar M. Talwar
D. Thaler D. Thaler
Microsoft Corporation Microsoft Corporation
December 31, 2002 January 03, 2002
Intra-Site Automatic Tunnel Addressing Protocol (ISATAP) Intra-Site Automatic Tunnel Addressing Protocol (ISATAP)
draft-ietf-ngtrans-isatap-09.txt draft-ietf-ngtrans-isatap-10.txt
Status of this Memo Status of this Memo
This document is an Internet-Draft and is in full conformance with This document is an Internet-Draft and is in full conformance with
all provisions of Section 10 of RFC2026. all provisions of Section 10 of RFC2026.
Internet-Drafts are working documents of the Internet Engineering Internet-Drafts are working documents of the Internet Engineering
Task Force (IETF), its areas, and its working groups. Note that Task Force (IETF), its areas, and its working groups. Note that
other groups may also distribute working documents as other groups may also distribute working documents as
Internet-Drafts. Internet-Drafts.
skipping to change at page 1, line 36 skipping to change at page 1, line 36
and may be updated, replaced, or obsoleted by other documents at any and may be updated, replaced, or obsoleted by other documents at any
time. It is inappropriate to use Internet-Drafts as reference time. It is inappropriate to use Internet-Drafts as reference
material or to cite them other than as "work in progress." material or to cite them other than as "work in progress."
The list of current Internet-Drafts can be accessed at http:// The list of current Internet-Drafts can be accessed at http://
www.ietf.org/ietf/1id-abstracts.txt. www.ietf.org/ietf/1id-abstracts.txt.
The list of Internet-Draft Shadow Directories can be accessed at The list of Internet-Draft Shadow Directories can be accessed at
http://www.ietf.org/shadow.html. http://www.ietf.org/shadow.html.
This Internet-Draft will expire on July 1, 2003. This Internet-Draft will expire on July 4, 2002.
Copyright Notice Copyright Notice
Copyright (C) The Internet Society (2002). All Rights Reserved. Copyright (C) The Internet Society (2002). All Rights Reserved.
Abstract Abstract
This document specifies an Intra-Site Automatic Tunnel Addressing This document specifies an Intra-Site Automatic Tunnel Addressing
Protocol (ISATAP) that connects IPv6 hosts and routers within IPv4 Protocol (ISATAP) that connects IPv6 hosts and routers within IPv4
sites. ISATAP treats the site's IPv4 infrastructure as a link layer sites. ISATAP treats the site's IPv4 infrastructure as a link layer
for IPv6 with no requirement for IPv4 multicast. ISATAP enables for IPv6 with no requirement for IPv4 multicast. ISATAP enables
intra-site automatic IPv6-in-IPv4 tunneling whether globally assigned intra-site automatic IPv6-in-IPv4 tunneling whether globally assigned
or private IPv4 addresses are used. or private IPv4 addresses are used.
Table of Contents Table of Contents
1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . 3 1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . 3
2. Applicability Statement . . . . . . . . . . . . . . . . . . 3 2. Applicability Statement . . . . . . . . . . . . . . . . . . 3
3. Requirements . . . . . . . . . . . . . . . . . . . . . . . . 4 3. Requirements . . . . . . . . . . . . . . . . . . . . . . . . 3
4. Terminology . . . . . . . . . . . . . . . . . . . . . . . . 4 4. Terminology . . . . . . . . . . . . . . . . . . . . . . . . 4
5. Basic IPv6 Operation on ISATAP Links . . . . . . . . . . . . 5 5. Non-Broadcast, Multiple Access (NBMA) Operation . . . . . . 4
5.1 Interface Identifiers and Address Construction . . . . . . . 5 5.1 Multicast . . . . . . . . . . . . . . . . . . . . . . . . . 5
5.2 ISATAP Link/Interface Configuration . . . . . . . . . . . . 5 5.2 Interface Identifiers and Address Construction . . . . . . . 5
5.3 Dual Stack Operation and Address Configuration . . . . . . . 6 5.3 ISATAP Link/Interface Configuration . . . . . . . . . . . . 5
5.4 Tunneling Mechanisms . . . . . . . . . . . . . . . . . . . . 6 5.4 Link Layer Address Options . . . . . . . . . . . . . . . . . 6
5.4.1 Encapsulation . . . . . . . . . . . . . . . . . . . . . . . 6 6. Automatic Tunneling . . . . . . . . . . . . . . . . . . . . 6
5.4.2 Tunnel MTU and Fragmentation . . . . . . . . . . . . . . . . 6 6.1 Dual IP Layer Operation . . . . . . . . . . . . . . . . . . 6
5.4.3 Handling IPv4 ICMP Errors . . . . . . . . . . . . . . . . . 7 6.2 Encapsulation . . . . . . . . . . . . . . . . . . . . . . . 6
5.4.4 Decapsulation . . . . . . . . . . . . . . . . . . . . . . . 7 6.3 Tunnel MTU and Fragmentation . . . . . . . . . . . . . . . . 7
5.4.5 Link-Local Addresses . . . . . . . . . . . . . . . . . . . . 7 6.4 Handling IPv4 ICMP Errors . . . . . . . . . . . . . . . . . 8
5.4.6 Ingress Filtering . . . . . . . . . . . . . . . . . . . . . 7 6.5 Local-Use IPv6 Unicast Addresses . . . . . . . . . . . . . . 8
6. Neighbor Discovery and Address Autoconfiguration . . . . . . 8 6.6 Ingress Filtering . . . . . . . . . . . . . . . . . . . . . 8
6.1 Address Resolution . . . . . . . . . . . . . . . . . . . . . 8 7. Neighbor Discovery for ISATAP Links . . . . . . . . . . . . 8
6.2 Address Autoconfiguration and Router Discovery . . . . . . . 9 7.1 Address Resolution . . . . . . . . . . . . . . . . . . . . . 9
6.2.1 Conceptual Data Structures . . . . . . . . . . . . . . . . . 9 7.2 Router and Prefix Discovery . . . . . . . . . . . . . . . . 9
6.2.2 Validity Checks for Router Advertisements . . . . . . . . . 10 7.2.1 Conceptual Data Structures . . . . . . . . . . . . . . . . . 9
6.2.3 Router Specification . . . . . . . . . . . . . . . . . . . . 10 7.2.2 Validity Checks for Router Advertisements . . . . . . . . . 10
6.2.4 Host Specification . . . . . . . . . . . . . . . . . . . . . 11 7.2.3 Router Specification . . . . . . . . . . . . . . . . . . . . 11
7. ISATAP Deployment Considerations . . . . . . . . . . . . . . 12 7.2.4 Host Specification . . . . . . . . . . . . . . . . . . . . . 11
7.1 Host And Router Deployment Considerations . . . . . . . . . 12 8. ISATAP Deployment Considerations . . . . . . . . . . . . . . 12
7.2 Site Administration Considerations . . . . . . . . . . . . . 12 8.1 Host And Router Deployment Considerations . . . . . . . . . 12
8. IANA Considerations . . . . . . . . . . . . . . . . . . . . 13 8.2 Site Administration Considerations . . . . . . . . . . . . . 12
9. Security considerations . . . . . . . . . . . . . . . . . . 13 9. IANA Considerations . . . . . . . . . . . . . . . . . . . . 13
10. Acknowledgements . . . . . . . . . . . . . . . . . . . . . . 14 10. Security considerations . . . . . . . . . . . . . . . . . . 13
11. Acknowledgements . . . . . . . . . . . . . . . . . . . . . . 14
Normative References . . . . . . . . . . . . . . . . . . . . 14 Normative References . . . . . . . . . . . . . . . . . . . . 14
Informative References . . . . . . . . . . . . . . . . . . . 15 Informative References . . . . . . . . . . . . . . . . . . . 15
Authors' Addresses . . . . . . . . . . . . . . . . . . . . . 16 Authors' Addresses . . . . . . . . . . . . . . . . . . . . . 16
A. Major Changes . . . . . . . . . . . . . . . . . . . . . . . 17 A. Major Changes . . . . . . . . . . . . . . . . . . . . . . . 17
B. Rationale for Interface Identifier Construction . . . . . . 18 B. Rationale for Interface Identifier Construction . . . . . . 18
C. Dynamic Per-neighbor MTU Discovery . . . . . . . . . . . . . 19 C. Dynamic Per-neighbor MTU Discovery . . . . . . . . . . . . . 19
Intellectual Property and Copyright Statements . . . . . . . 21 Intellectual Property and Copyright Statements . . . . . . . 21
1. Introduction 1. Introduction
This document presents a simple approach that enables incremental This document presents a simple approach that enables incremental
deployment of IPv6 [1] within IPv4-based [2] sites in a manner that deployment of IPv6 [1] within IPv4-based [2] sites. We refer to this
is compatible with inter-domain tunneling mechanisms, e.g., RFC 3056 approach as the Intra-Site Automatic Tunnel Addressing Protocol
(6to4) [18]. We refer to this approach as the Intra-Site Automatic (ISATAP). ISATAP allows dual-stack nodes that do not share a
Tunnel Addressing Protocol (ISATAP). ISATAP allows dual-stack nodes physical link with an IPv6 router to automatically tunnel packets to
that do not share a physical link with an IPv6 router to the IPv6 next-hop address through IPv4, i.e., the site's IPv4
automatically tunnel packets to the IPv6 next-hop address through infrastructure is treated as a link layer for IPv6.
IPv4, i.e., the site's IPv4 infrastructure is treated as an link
layer for IPv6.
This document specifies details for the transmission of IPv6 packets This document specifies details for the operation of IPv6 over ISATAP
over ISATAP links (i.e., automatic IPv6-in-IPv4 tunneling), including links (i.e., automatic IPv6-in-IPv4 tunneling), including an
an interface identifier format that embeds an IPv4 address. This interface identifier format that embeds an IPv4 address. This format
format supports IPv6 protocol mechanisms for address configuration as supports IPv6 protocol mechanisms for address configuration as well
well as simple link-layer address mapping. Simple validity checks as simple link-layer address mapping. Also specified in this
for received packets are given. Also specified in this document is document is the operation of IPv6 Neighbor Discovery for ISATAP. The
the operation of IPv6 Neighbor Discovery for ISATAP. The document document finally presents deployment and security considerations.
finally presents deployment and security considerations for ISATAP.
2. Applicability Statement 2. Applicability Statement
ISATAP provides the following features: ISATAP provides the following features:
o treats site's IPv4 infrastructure as link layer for IPv6 using o treats site's IPv4 infrastructure as a link layer for IPv6 using
automatic IPv6-in-IPv4 tunneling (i.e., no configured tunnel automatic IPv6-in-IPv4 tunneling (i.e., no configured tunnel
state) state)
o enables incremental deployment of IPv6 hosts within IPv4 sites o enables incremental deployment of IPv6 hosts within IPv4 sites
with no aggregation scaling issues at border gateways with no aggregation scaling issues at border gateways
o requires no special IPv4 services within the site (e.g., o requires no special IPv4 services within the site (e.g.,
multicast) multicast)
o supports both stateless address autoconfiguration and manual o supports both stateless address autoconfiguration and manual
configuration configuration
o supports networks that use non-globally unique IPv4 addresses o supports networks that use non-globally unique IPv4 addresses
(e.g., when private address allocations [3] are used), but does (e.g., when private address allocations [3] are used), but does
not allow the virtual ISATAP link to span a Network Address not allow the virtual ISATAP link to span a Network Address
Translator [4] Translator [4]
o compatible with other NGTRANS mechanisms (e.g., 6to4 [18]) o compatible with other NGTRANS mechanisms (e.g., 6to4 [19])
3. Requirements 3. Requirements
The keywords MUST, MUST NOT, REQUIRED, SHALL, SHALL NOT, SHOULD, The keywords MUST, MUST NOT, REQUIRED, SHALL, SHALL NOT, SHOULD,
SHOULD NOT, RECOMMENDED, MAY, and OPTIONAL, when they appear in this SHOULD NOT, RECOMMENDED, MAY, and OPTIONAL, when they appear in this
document, are to be interpreted as described in [5]. document, are to be interpreted as described in [5].
This document also makes use of internal conceptual variables to This document also makes use of internal conceptual variables to
describe protocol behavior and external variables that an describe protocol behavior and external variables that an
implementation must allow system administrators to change. The implementation must allow system administrators to change. The
skipping to change at page 4, line 25 skipping to change at page 4, line 19
settings influence protocol behavior are provided to demonstrate settings influence protocol behavior are provided to demonstrate
protocol behavior. An implementation is not required to have them in protocol behavior. An implementation is not required to have them in
the exact form described here, so long as its external behavior is the exact form described here, so long as its external behavior is
consistent with that described in this document. consistent with that described in this document.
4. Terminology 4. Terminology
The terminology of RFC 2460 [1] applies to this document. The The terminology of RFC 2460 [1] applies to this document. The
following additional terms are defined: following additional terms are defined:
link; on-link: link, on-link, off-link:
same definitions as ([6], section 2.1). same definitions as ([6], section 2.1).
underlying link: underlying link:
a link layer that supports IPv4 (for ISATAP), and MAY also support a link layer that supports IPv4 (for ISATAP), and MAY also support
IPv6 natively. IPv6 natively.
ISATAP link: ISATAP link:
one or more underlying links used for tunneling. The IPv4 network one or more underlying links used for tunneling. The IPv4 network
layer addresses of the underlying links are used as link-layer layer addresses of the underlying links are used as link-layer
addresses on the ISATAP link. addresses on the ISATAP link.
ISATAP interface: ISATAP interface:
a node's attachment to an ISATAP link. a node's attachment to an ISATAP link.
ISATAP address: ISATAP address:
an on-link address on an ISATAP interface and with an interface an on-link address on an ISATAP interface and with an interface
identifier constructed as specified in Section 5.1 identifier constructed as specified in Section 5.2
ISATAP router: ISATAP router:
an IPv6 node that has an ISATAP interface over which it forwards an IPv6 node that has an ISATAP interface over which it forwards
packets not explicitly addressed to itself. packets not explicitly addressed to itself.
ISATAP host: ISATAP host:
any node that has an ISATAP interface and is not an ISATAP router. any node that has an ISATAP interface and is not an ISATAP router.
5. Basic IPv6 Operation on ISATAP Links 5. Non-Broadcast, Multiple Access (NBMA) Operation
ISATAP links transmit IPv6 packets via automatic tunnels using the ISATAP links transmit IPv6 packets via automatic tunnels using the
site's IPv4 infrastructure as a link layer for IPv6, i.e., the site's site's IPv4 infrastructure as a link layer for IPv6, i.e., IPv6
IPv4 infrastructure is treated as a Non-Broadcast, Multiple Access treats the site's IPv4 infrastructure as a Non-Broadcast, Multiple
(NBMA) link layer. The following subsections outline basic Access (NBMA) link layer. RFC 2491 [7] provides a general
operational details for IPv6 on ISATAP links: architecture for IPv6 over NBMA networks that forms the basis for
companion documents such as the present. The following subsections
present NBMA considerations for IPv6 on ISATAP links:
5.1 Interface Identifiers and Address Construction 5.1 Multicast
(RFC2491 [7], section 5.1) requires companion documents to specify ISATAP links most closely meet the description for connectionless
the exact mechanism for generating interface tokens (i.e., service found in the last paragraph of ([7], section 1), i.e., ISATAP
identifiers). Interface identifiers for ISATAP are compatible with addresses provide the sender with an NBMA destination address to
the EUI-64 identifier format ([8], section 2.5.1), and are which it can transmit packets whenever it desires. Thus, multicast
constructed by appending an IPv4 address on the ISATAP link to the emulation mechanisms are not required to support host-side operation
32-bit string '00-00-5E-FE'. (Appendix B includes non-normative text of the IPv6 neighbor discovery protocol.
explaining the rationale for this construction rule.)
5.2 Interface Identifiers and Address Construction
([7], section 5.1) requires companion documents to specify the exact
mechanism for generating interface tokens (i.e., identifiers).
Interface identifiers for ISATAP are compatible with the EUI-64
identifier format ([8], section 2.5.1), and are constructed by
appending an IPv4 address on the ISATAP link to the 32-bit string
'00-00-5E-FE'. (Appendix B includes non-normative text explaining
the rationale for this construction rule.)
Global and Local-use ISATAP addresses are constructed as follows: Global and Local-use ISATAP addresses are constructed as follows:
| 64 bits | 32 bits | 32 bits | | 64 bits | 32 bits | 32 bits |
+------------------------------+---------------+----------------+ +------------------------------+---------------+----------------+
| global or local-use unicast | 0000:5EFE | IPv4 Address | | global or local-use unicast | 0000:5EFE | IPv4 Address |
| prefix | | of ISATAP link | | prefix | | of ISATAP link |
+------------------------------+---------------+----------------+ +------------------------------+---------------+----------------+
Figure 1 Figure 1
skipping to change at page 5, line 43 skipping to change at page 5, line 46
For example, the global unicast address: For example, the global unicast address:
3FFE:1A05:510:1111:0:5EFE:8CAD:8108 3FFE:1A05:510:1111:0:5EFE:8CAD:8108
has a prefix of '3FFE:1A05:510:1111::/64' and an ISATAP interface has a prefix of '3FFE:1A05:510:1111::/64' and an ISATAP interface
identifier with embedded IPv4 address: '140.173.129.8'. The address identifier with embedded IPv4 address: '140.173.129.8'. The address
is alternately written as: is alternately written as:
3FFE:1A05:510:1111:0:5EFE:140.173.129.8 3FFE:1A05:510:1111:0:5EFE:140.173.129.8
(Similar examples for local-use addresses are made obvious by the Examples for local-use addresses are obvious from the above and with
above and with reference to the IPv6 addressing architecture reference to ([8], section 2.5.6).
document.)
5.2 ISATAP Link/Interface Configuration 5.3 ISATAP Link/Interface Configuration
ISATAP Link/Interface configuration is consistent with (RFC2491 [7], ISATAP Link/Interface configuration is consistent with ([7], sections
sections 5.1.1 and 5.1.2). 5.1.1 and 5.1.2).
Using the terminology of Section 4, an ISATAP link consists of one or An ISATAP link consists of one or more underlying links that support
more underlying links that support IPv4 for tunneling within a site. IPv4 for tunneling within a site. ISATAP interfaces are configured
ISATAP interfaces are configured over ISATAP links; each IPv4 address over ISATAP links; each IPv4 address assigned to an underlying link
assigned to an underlying link is seen as a link-layer address for is seen as a link-layer address for ISATAP.
ISATAP.
5.3 Dual Stack Operation and Address Configuration 5.4 Link Layer Address Options
([7], section 5.2) requires companion documents to specify the
contents of the [NTL], [STL], [NBMA Number] and [NBMA Subaddress]
fields for link layer address options. For ISATAP links:
o the [NTL] and [STL] fields MUST be zero
o the [NBMA Number] encodes a 4-octet IPv4 address
o the [NBMA Subaddress] field is omitted
([7], section 5.2) does NOT require companion documents to specify
the value for [Length], i.e., the total length of the option in 8
octets. Senders may therefore set [Length] to any value between 1
and 255; when [Length] is greater than 1, receivers treat any bytes
that follow the [NBMA Number] as null-padding.
6. Automatic Tunneling
The common tunneling mechanisms specified in ([9], sections 2 and 3)
are used, with the following noted specific considerations for
ISATAP:
6.1 Dual IP Layer Operation
ISATAP uses the same specification found in ([9], section 2). That ISATAP uses the same specification found in ([9], section 2). That
is, ISATAP nodes implement "IPv6/IPv4" or "dual-stack" configurations is, ISATAP nodes provide complete IPv4 and IPv6 implementations and
and operate with both stacks enabled. Address configuration and DNS are able to send and receive both IPv4 and IPv6 packets. ISATAP
considerations are the same as for ([9], sections 2.1 and 2.2) nodes operate with both their IPv4 and IPv6 stacks enabled.
5.4 Tunneling Mechanisms Address configuration considerations are the same as for ([9],
section 2.1). Additionally, ISATAP nodes require that IPv4 address
configuration take place on at least one underlying link prior to
IPv6 address configuration on an ISATAP link.
The common tunneling mechanisms specified in ([9], sections 3.1 DNS considerations are the same as ([9], sections 2.2 and 2.3).
through 3.7) are used, with the following noted specific
considerations:
5.4.1 Encapsulation 6.2 Encapsulation
The specification in ([9], section 3.1) is used. Additionally, the The specification in ([9], section 3.1) is used. Additionally, the
IPv6 next-hop address for packets sent on an ISATAP link MUST be an IPv6 next-hop address for packets sent on an ISATAP link MUST be an
ISATAP address; other packets are discarded (i.e., not encapsulated) ISATAP address; other packets are discarded and an ICMPv6 destination
and an ICMPv6 destination unreachable indication with code 3 (Address unreachable indication with code 3 (Address Unreachable) [10] is
Unreachable) [10] is returned to the source. returned to the source.
5.4.2 Tunnel MTU and Fragmentation 6.3 Tunnel MTU and Fragmentation
The specification in ([9], section 3.2) is NOT used; the The specification in ([9], section 3.2) is NOT used; the
specification in this section is used instead. specification in this section is used instead.
ISATAP uses automatic tunnel interfaces that may be configured over ISATAP uses automatic tunnel interfaces that may be configured over
multiple underlying links with diverse maximum transmission units multiple underlying links with diverse maximum transmission units
(MTUs). The minimum MTU for IPv6 interfaces is 1280 bytes ([1], (MTUs). The minimum MTU for IPv6 interfaces is 1280 bytes ([1],
Section 5), but the following considerations apply when IPv4 is used Section 5), but the following considerations for the MTU of ISATAP
as a link layer for IPv6: interfaces apply:
o nearly all IPv4 nodes accept unfragmented packets up to 1500 bytes o Nearly all IPv4 nodes connect to physical links with MTUs of 1500
bytes or larger (e.g., Ethernet)
o sub-IPv4 layer encapsulations (e.g., VPN) may occur on some paths o Sub-IPv4 layer encapsulations (e.g., VPN) may occur on some paths
o commonly-deployed VPNs use an MTU of 1400 bytes o Commonly-deployed VPNs use an MTU of 1400 bytes
Thus, ISATAP interfaces SHOULD use an MTU (ISATAP_MTU) of 1380 bytes Unless a dynamic per-neighbor MTU discovery mechanism is implemented,
(1400 minus 20 bytes for IPv4 encapsulation) to maximize efficiency ISATAP interfaces MUST use an MTU (ISATAP_MTU) of no more than 1380
and minimize IPv4 fragmentation. bytes (1400 minus 20 bytes for IPv4 encapsulation) to maximize
efficiency and minimize IPv4 fragmentation for the predominant
deployment case. ISATAP_MTU MAY be set to a larger value when the
encapsulator implements a dynamic per-neighbor MTU discovery
mechanism, but this value SHOULD NOT exceed the largest MTU of all
underlying links (minus 20 bytes for IPv4 encapsulation). Appendix C
provides non-normative considerations for dynamic per-neighbor MTU
discovery.
ISATAP_MTU MAY be set to larger values when the encapsulator uses The network layer (IPv6) forwards packets of size ISATAP_MTU or
dynamic per-neighbor MTU discovery. When larger values are used, smaller to the ISATAP interface. All other packets are dropped, and
ISATAP_MTU SHOULD NOT exceed the maximum MTU of all underlying links an ICMPv6 "packet too big" message with MTU = ISATAP_MTU is returned
minus 20 bytes for link layer encapsulation. (Appendix C provides to the source [11]. The ISATAP link layer encapsulates packets of
non-normative considerations for dynamic per-neighbor MTU discovery.) size 1380 bytes or smaller with the Don't Fragment (DF) bit NOT set
in the encapsualting IPv4 header.
As with ordinary IPv6 interfaces, the network layer (IPv6) forwards Nodes that configure ISATAP interfaces MUST have IPv4 reassembly
packets of size ISATAP_MTU or smaller to the ISATAP interface. All buffers large enough to receive packets with the DF bit not set in
other packets are dropped, and an ICMPv6 "packet too big" message the encapsulating IPv4 header. RFC 1122 [12], section 3.3.2
with MTU = ISATAP_MTU is returned to the source [11]. specifies that the Effective MTU to Receive (EMTU_R) for IPv4 nodes:
ISATAP interfaces send all packets of size 1380 bytes or smaller with "...MUST be greater than or equal to 576, SHOULD be either
the Don't Fragment (DF) bit NOT set in the encapsualting IPv4 header. configurable or indefinite, and SHOULD be greater than or equal to
the MTU of the connected network(s)".
5.4.3 Handling IPv4 ICMP Errors With reference to this specification, the EMTU_R for nodes that
configure ISATAP interfaces MUST be greater than or equal to 1500
bytes (i.e., the predominant deployment case for connected IPv4
networks) and SHOULD be either configurable or indefinite.
6.4 Handling IPv4 ICMP Errors
The specification in ([9], section 3.4) MAY be used. IPv4 ICMP The specification in ([9], section 3.4) MAY be used. IPv4 ICMP
errors and ARP failures are otherwise processed as link error errors and ARP failures are otherwise processed as link error
notifications. notifications.
5.4.4 Decapsulation 6.5 Local-Use IPv6 Unicast Addresses
The specification in ([9], section 3.6) is used.
5.4.5 Link-Local Addresses
The specification in ([9], section 3.7) is NOT used. Instead, The specification in ([9], section 3.7) is NOT used. Instead, local
link-local addresses are formed by appending an interface identifier, use IPv6 unicast addresses are formed exactly as specified in ([8],
as defined in Section 5.1, to the prefix FE80::/64. section 2.5.6).
5.4.6 Ingress Filtering 6.6 Ingress Filtering
The network layer (IPv6) destination address of a packet received on The specification in ([9], section 3.9) is used on ISATAP router
an ISATAP interface is either local (i.e., matches an address interfaces. (ISATAP host interfaces silently discard any packets
configured on the local IPv6 stack) or foreign. The decapsulator received with a foreign IPv6 destination address, i.e., an address
MUST be configured with a list of IPv4 address prefixes that are not configured on the local IPv6 stack.)
acceptable, i.e., an ingress filter list (default deny all). For
packets with foreign network layer (IPv6) destination addresses, the
link layer (IPv4) source address MUST be explicitly allowed by
ingress filtering. Packets that do not satisfy this condition are
silently discarded.
Additionally, all packets (whether foreign or local) MUST satisfy at Additionally, packets received on ISATAP host and router interfaces
least one (i.e., one or both) of the following validity checks: MUST satisfy at least one (i.e., one or both) of the following
validity checks:
o the network-layer (IPv6) source address is an on-link ISATAP o the network-layer (IPv6) source address is an on-link ISATAP
address with an interface identifier that embeds the link-layer address with an interface identifier that embeds the link-layer
(IPv4) source address (IPv4) source address
o the link-layer (IPv4) source address is in the Potential Routers o the link-layer (IPv4) source address is in the Potential Routers
List (see Section 6.2.1) List (see Section 7.2.1)
Packets that do not satisfy the above conditions are silently Packets that do not satisfy the above conditions are silently
discarded. discarded.
6. Neighbor Discovery and Address Autoconfiguration 7. Neighbor Discovery for ISATAP Links
RFC 2491 [7] provides a general architecture for IPv6 over NBMA
networks, including multicast mechanisms to support host-side
operation of the IPv6 neighbor discovery protocol. ISATAP links most
closely meet the description for connectionless service found in the
last paragraph of ([7], section 1), i.e., ISATAP addresses provide
the sender with an NBMA destination address to which it can transmit
packets whenever it desires. Thus, the RFC 2491 multicast mechanisms
are not required for address resolution and not otherwise implemented
on ISATAP links due to traffic scaling considerations (i.e., ISATAP
links are unicast-only).
RFC 2461 [6] provides the following guidelines for non-broadcast RFC 2461 [6] provides the following guidelines for non-broadcast
multiple access (NBMA) link support: multiple access (NBMA) link support:
"Redirect, Neighbor Unreachability Detection and next-hop "Redirect, Neighbor Unreachability Detection and next-hop
determination should be implemented as described in this document. determination should be implemented as described in this document.
Address resolution and the mechanism for delivering Router Address resolution and the mechanism for delivering Router
Solicitations and Advertisements on NBMA links is not specified in Solicitations and Advertisements on NBMA links is not specified in
this document." this document."
ISATAP links SHOULD implement Redirect, Neighbor Unreachability ISATAP links SHOULD implement Redirect, Neighbor Unreachability
Detection, and next-hop determination exactly as specified in [6]. Detection, and next-hop determination exactly as specified in [6].
Address resolution and the mechanisms for delivering Router Address resolution and the mechanisms for delivering Router
Solicitations and Advertisements on ISATAP links use the Solicitations and Advertisements for ISATAP links are not specified
specifications found in this document. by [6]; instead, they are specified in this document. (Note that
these mechanisms MAY potentially apply to other types of NBMA links
in the future.)
6.1 Address Resolution 7.1 Address Resolution
ISATAP addresses are resolved to link-layer addresses (IPv4) by a ISATAP addresses are resolved to link-layer addresses (IPv4) by a
static computation, i.e., the last four octets are treated as an IPv4 static computation, i.e., the last four octets are treated as an IPv4
address. ([7], section 5.2) requires companion documents to specify address.
the format for link layer address options, however, link layer
address options are not needed for address resolution in ISATAP.
Thus, no format is specified and the following specification from
([9], section 3.8) applies:
"This means that a sender of Neighbor Discovery packets
* SHOULD NOT include Source Link Layer Address options or Target
Link Layer Address options on the tunnel link.
* MUST silently ignore any received SLLA or TLLA options on the Following static address resolution, ISATAP hosts SHOULD perform an
tunnel link." initial reachability confirmation by sending unicast Neighbor
Solicitations (NSs) and receiving a Neighbor Advertisement using the
mechanisms specified in ([6], sections 7.2.2-7.2.8). (Note that
implementations MAY omit the source/target link layer options in NS/
NA messages when unicast is used.)
Following static address resolution, ISATAP hosts SHOULD implement ISATAP hosts SHOULD additionally perform Neighbor Unreachability
the reachability confirmation specifications in [6], sections Detection (NUD) as specified in ([6], section 7.3). ISATAP routers
7.2.2-7.2.8 that apply when unicast Neighbor Solicitations (NS) are MAY perform the above-specified reachability detection and NUD
used. ISATAP hosts SHOULD additionally perform Neighbor procedures, but this might not scale in all environments.
Unreachability Detection (NUD) as specified in (RFC 2461 [6], section
7.3). ISATAP routers MAY perform the above-specified reachability
detection and NUD procedures, but this might not scale in all
environments.
All ISATAP nodes MUST send solicited neighbor advertisements ([6], All ISATAP nodes MUST send solicited neighbor advertisements ([6],
section 7.2.4). section 7.2.4).
ISATAP links disable Duplicate Address Detection, as permitted by 7.2 Router and Prefix Discovery
([12], section 4).
6.2 Address Autoconfiguration and Router Discovery
Since NBMA multicast emulation mechanisms are not used on ISATAP Since NBMA multicast emulation mechanisms are not used, ISATAP nodes
links, nodes will not receive unsolicited multicast Router will not receive unsolicited multicast Router Advertisements. Thus,
Advertisements. (RFC 2462 [12], section 5.5.2) requires that hosts alternate mechanisms are required and specified below:
use stateful autoconfiguration (i.e., DHCPv6 [13]) in the absence of
Router Advertisements. When statelful autoconfiguration is not
available, nodes use alternate mechanisms (described below) for
router and prefix discovery.
6.2.1 Conceptual Data Structures 7.2.1 Conceptual Data Structures
ISATAP nodes use the conceptual data structures Prefix List and ISATAP nodes use the conceptual data structures Prefix List and
Default Router List exactly as in ([6], section 5.1). ISATAP links Default Router List exactly as in ([6], section 5.1). ISATAP links
add two new conceptual data structures "Potential Router List" and add a new conceptual data structures "Potential Router List" and the
"Stateful Autoconfiguration Server List". following new configuration variable:
A Potential Router List (PRL) and Stateful Autoconfiguration Server ResolveInterval
List (SASL) is associated with every ISATAP link. The PRL provides a Time between name service resolutions. Default and suggested
trust basis for router validation (see security considerations). minimum: 1hr
Each entry in the PRL has an IPv4 address and an associated timer. A Potential Router List (PRL) is associated with every ISATAP link.
The IPv4 address represents a router's ISATAP interface (likely to be The PRL provides a trust basis for router validation (see security
an "advertising interface"), and is used to construct the ISATAP considerations). Each entry in the PRL has an IPv4 address and an
link-local address for that interface. Similarly, each entry in the associated timer. The IPv4 address represents a router's ISATAP
SASL has an IPv4 address and associated timer. The IPv4 address interface (likely to be an "advertising interface"), and is used to
represents a DHCPv6 server attached to the ISATAP link, and is used construct the ISATAP link-local address for that interface. The
to construct the ISATAP link-local address for that DHCPv6 server. following sections specify the process for initializing the PRL:
When a node enables an ISATAP link, it first discovers IPv4 addresses When a node enables an ISATAP link, it first discovers IPv4 addresses
for the PRL and SASL. The addresses MAY be established by a DHCPv4 for the PRL. The addresses SHOULD be established by a DHCPv4 [13]
[14] option for ISATAP (option code TBD), by manual configuration, or option for ISATAP (option code TBD), by manual configuration, or by
by an unspecified alternative method (e.g., DHCPv4 vendor-specific an unspecified alternate method (e.g., DHCPv4 vendor-specific
option; DNS ([19]) fully-qualified domain names). option).
When DNS fully-qualified domain names are used, IPv4 addresses for When no other mechanisms are available, a DNS fully-qualified domain
the PRL and SASL are discovered through a static host file or by name (FQDN) [20] MAY be used. In this case, the FQDN is resolved
querying an IPv4-based DNS server to resolve the domain names into into IPv4 addresses for the PRL through a static host file, a
address records (e.g., DNS 'A' resource records) containing IPv4 site-specific name service, or by querying an IPv4-based DNS server.
addresses. Unspecified alternative methods for domain name Unspecified alternate methods for domain name resolution may also be
resolution may also be used. The following notes apply when DNS used. The following notes apply:
fully-qualified domain names are used:
1. Site administrators maintain domain names and IPv4 addresses for 1. Site administrators maintain a list of IPv4 addresses
the PRL and SASL for the site's ISATAP service, e.g., as address representing ISATAP router interfaces and make them available via
records in the site's name service. Administrators may also one or more of the mechanisms described above.
advertise the domain names in a DHCPv4 option for ISATAP.
2. There are no mandatory rules for the selection of domain names, 2. There are no mandatory rules for the selection of a FQDN, but
but administrators are encouraged to use the convention administrators are encouraged to use the convention
"(list_name).isatap.domainname" (e.g., prl.isatap.example.com). "isatap.domainname" (e.g., isatap.example.com).
3. After initialization, nodes periodically re-initialize the PRL 3. After initialization, nodes periodically re-initialize the PRL
and SASL, e.g., once per hour. When DNS is used, client DNS (after ResolveInterval). When DNS is used, client DNS resolvers
resolvers use the IPv4 transport to resolve the names and follow use the IPv4 transport to resolve the names and follow the cache
the cache invalidation procedures in [19] when the DNS invalidation procedures in [20] when the DNS time-to-live
time-to-live expires. expires.
6.2.2 Validity Checks for Router Advertisements 7.2.2 Validity Checks for Router Advertisements
A node MUST silently discard any Router Advertisement messages it A node MUST silently discard any Router Advertisement messages it
receives that do not satisfy both the validity checks in ([6], receives that do not satisfy both the validity checks in ([6],
section 6.1.2) and the following additional validity check for section 6.1.2) and the following additional validity check for
ISATAP: ISATAP:
o the network-layer (IPv6) source address is an ISATAP address and o the network-layer (IPv6) source address is an ISATAP address and
embeds an IPv4 address from the PRL embeds an IPv4 address from the PRL
6.2.3 Router Specification 7.2.3 Router Specification
Advertising ISATAP interfaces of routers behave the same as Advertising ISATAP interfaces of routers behave the same as
advertising interfaces described in ([6], section 6.2). However, advertising interfaces described in ([6], section 6.2). However,
periodic unsolicited multicast Router Advertisements are not used, periodic unsolicited multicast Router Advertisements are not used,
thus the "interval timer" associated with advertising interfaces is thus the "interval timer" associated with advertising interfaces is
not used for that purpose. not used for that purpose.
When an ISATAP router receives a valid Router Solicitation on an When an ISATAP router receives a valid Router Solicitation on an
advertising ISATAP interface, it replies with a unicast Router advertising ISATAP interface, it replies with a unicast Router
Advertisement to the address of the node which sent the Router Advertisement to the address of the node which sent the Router
Solicitation. The source address of the Router Advertisement is a Solicitation. The source address of the Router Advertisement is a
link-local unicast address associated with the interface. This MAY link-local unicast address associated with the interface. This MAY
be the same as the destination address of the Router Solicitation. be the same as the destination address of the Router Solicitation.
ISATAP routers MAY engage in the solicitation process described under ISATAP routers MAY engage in the solicitation process described under
Host Specification below, e.g., if Router Advertisement consistency Host Specification below, e.g., if Router Advertisement consistency
verification ([6], section 6.2.7) is desired. verification ([6], section 6.2.7) is desired.
6.2.4 Host Specification 7.2.4 Host Specification
All entries in the PRL are assumed to represent active ISATAP routers All entries in the PRL are assumed to represent active ISATAP routers
within the site, i.e., the PRL provides trust basis only; not within the site, i.e., the PRL provides trust basis only; not
reachability detection. When stateful autoconfiguration is available reachability detection. ISATAP nodes SHOULD use stateful
(i.e., when the SASL is non-null and at least one DHCPv6 server is configuration to assign IPv6 prefixes and default router information.
reachable), hosts may send unicast messages directly to the DHCPv6 When stateful configuration is not available, hosts MAY periodically
server as specified in ([13], section 1.1). Hosts SHOULD attempt solicit information from one or more entries in the PRL ("PRL(i)") by
stateful autoconfiguration for each entry in the SASL (i.e., until an sending unicast Router Solicitation messages using the IPv4 address
attempt succeeds) before concluding that stateful autoconfiguration ("V4ADDR_PRL(i)") and associated timer in the entry. Hosts add the
is unavailable. following variable to support the solicitation process:
When stateful autoconfiguration is unavailable, hosts MAY
periodically solicit information from one or more entries in the PRL
("PRL(i)") by sending unicast Router Solicitation messages using the
IPv4 address ("V4ADDR_PRL(i)") and associated timer in the entry.
Hosts add the following variable to support the solicitation process:
MinRouterSolicitInterval MinRouterSolicitInterval
Minimum time between sending Router Solicitations to any router. Minimum time between sending Router Solicitations to any router.
Default and suggested minimum 800,000 milliseconds (15min). Default and suggested minimum: 15min.
When a PRL(i) is selected, the host sets its associated timer to When a PRL(i) is selected, the host sets its associated timer to
MinRouterSolicitInterval and initiates solicitation following a short MinRouterSolicitInterval and initiates solicitation following a short
delay as in ([6], section 6.3.7). The manner of choosing particular delay as in ([6], section 6.3.7). The manner of choosing particular
routers in the PRL for solicitation is outside the scope of this routers in the PRL for solicitation is outside the scope of this
specification. The solicitation process repeats when the associated specification. The solicitation process repeats when the associated
timer expires. timer expires.
Solicitation consists of sending Router Solicitations to the ISATAP Solicitation consists of sending Router Solicitations to the ISATAP
link-local address constructed from the entry's IPv4 address, i.e., link-local address constructed from the entry's IPv4 address, i.e.,
skipping to change at page 12, line 8 skipping to change at page 12, line 11
6.3.7). 6.3.7).
Hosts process received Router Advertisements exactly as in ([6], Hosts process received Router Advertisements exactly as in ([6],
section 6.3.4). Hosts additionally reset the timer associated with section 6.3.4). Hosts additionally reset the timer associated with
the V4ADDR_PRL(i) embedded in the network-layer source address in the V4ADDR_PRL(i) embedded in the network-layer source address in
each solicited Router Advertisement received. The timer is reset to each solicited Router Advertisement received. The timer is reset to
either 0.5 * (the minimum value in the router lifetime or valid either 0.5 * (the minimum value in the router lifetime or valid
lifetime of any on-link prefixes received in the advertisement) or lifetime of any on-link prefixes received in the advertisement) or
MinRouterSolicitInterval; whichever is longer. MinRouterSolicitInterval; whichever is longer.
7. ISATAP Deployment Considerations 8. ISATAP Deployment Considerations
7.1 Host And Router Deployment Considerations 8.1 Host And Router Deployment Considerations
For hosts, if an underlying link supports both IPv4 (over which For hosts, if an underlying link supports both IPv4 (over which
ISATAP is implemented) and also supports IPv6 natively, then ISATAP ISATAP is implemented) and also supports IPv6 natively, then ISATAP
MAY be enabled if the native IPv6 layer does not receive Router MAY be enabled if the native IPv6 layer does not receive Router
Advertisements (i.e., does not have connection with an IPv6 router). Advertisements (i.e., does not have connection with an IPv6 router).
After a non-link-local address has been configured and a default After a non-link-local address has been configured and a default
router acquired on the native link, the host SHOULD discontinue the router acquired on the native link, the host SHOULD discontinue the
router solicitation process described in the host specification and router solicitation process described in the host specification and
allow existing ISATAP address configurations to expire as specified allow existing ISATAP address configurations to expire as specified
in ([6], section 5.3) and ([12], section 5.5.4). Any ISATAP in ([6], section 5.3) and ([14], section 5.5.4). Any ISATAP
addresses added to the DNS for this host should also be removed. In addresses added to the DNS for this host should also be removed. In
this way, ISATAP use will gradually diminish as IPv6 routers are this way, ISATAP use will gradually diminish as IPv6 routers are
widely deployed throughout the site. widely deployed throughout the site.
Routers MAY configure an interface to simultaneously support both Routers MAY configure an interface to simultaneously support both
native IPv6, and also ISATAP (over IPv4). Routing will operate as native IPv6, and also ISATAP (over IPv4). Routing will operate as
usual between these two domains. Note that the prefixes used on the usual between these two domains. Note that the prefixes used on the
ISATAP and native IPv6 interfaces will be distinct. The IPv4 ISATAP and native IPv6 interfaces will be distinct. The IPv4
address(es) configured on a router's ISATAP interface(s) SHOULD be address(es) configured on a router's ISATAP interface(s) SHOULD be
added (either automatically or manually) to the site's address added (either automatically or manually) to the site's address
records for ISATAP router interfaces. records for ISATAP router interfaces.
7.2 Site Administration Considerations 8.2 Site Administration Considerations
The following considerations are noted for sites that deploy ISATAP: The following considerations are noted for sites that deploy ISATAP:
o ISATAP links are administratively defined by a set of router o ISATAP links are administratively defined by a set of router
interfaces, a set of stateful autoconfiguration servers, and set interfaces and set of nodes which discover those interface and
of nodes which discover those interface and server addresses Thus, server addresses Thus, ISATAP links are defined by administrative
ISATAP links are defined by administrative (not physical) (not physical) boundaries.
boundaries.
o ISATAP hosts and routers can be deployed in an ad-hoc and o ISATAP hosts and routers can be deployed in an ad-hoc and
independent fashion. In particular, ISATAP hosts can be deployed independent fashion. In particular, ISATAP hosts can be deployed
with little/no advanced knowledge of existing ISATAP routers, and with little/no advanced knowledge of existing ISATAP routers, and
ISATAP routers can deployed with no reconfiguration requirements ISATAP routers can deployed with no reconfiguration requirements
for hosts. for hosts.
o When stateful autoconfiguration is not available, ISATAP nodes MAY o When stateful autoconfiguration is not available, ISATAP nodes MAY
periodically send unicast Router Solicitations to and receive periodically send unicast Router Solicitations to and receive
unicast Router Advertisements from to one or more members of the unicast Router Advertisements from to one or more members of the
potential router list. A well-deployed stateful autoconfiguration potential router list. A well-deployed stateful autoconfiguration
service within the site can minimize and/or eliminate the need for service within the site can minimize and/or eliminate the need for
periodic solicitation. periodic solicitation.
o ISATAP nodes periodically refresh the entries on the PRL and SASL. o ISATAP nodes periodically refresh the entries on the PRL.
Responsible site administration can reduce the control traffic. Responsible site administration can reduce the control traffic.
At a minimum, administrators SHOULD ensure that dynamically At a minimum, administrators SHOULD ensure that dynamically
advertised information for the site's PRL and SASL are well advertised information for the site's PRL is well maintained.
maintained.
8. IANA Considerations 9. IANA Considerations
A DHCPv4 option code for ISATAP (TBD) [20] is requested in the event A DHCPv4 option code for ISATAP (TBD) [21] is requested in the event
that the IESG recommends this document for standards track. that the IESG recommends this document for standards track.
9. Security considerations 10. Security considerations
Site administrators are advised that, in addition to possible attacks Site administrators are advised that, in addition to possible attacks
against IPv6, security attacks against IPv4 MUST also be considered. against IPv6, security attacks against IPv4 MUST also be considered.
Responsible IPv4 site security management is strongly encouraged. In Responsible IPv4 site security management is strongly encouraged. In
particular, border gateways SHOULD implement filtering to detect particular, border gateways SHOULD implement filtering to detect
spoofed IPv4 source addresses at a minimum; ip-protocol-41 filtering spoofed IPv4 source addresses at a minimum; ip-protocol-41 filtering
SHOULD also be implemented. SHOULD also be implemented.
If IPv4 source address filtering is not correctly implemented, the If IPv4 source address filtering is not correctly implemented, the
skipping to change at page 13, line 49 skipping to change at page 13, line 50
Advertisements they receive from on-link routers, as indicated by a Advertisements they receive from on-link routers, as indicated by a
value of 255 in the IPv6 'hop-limit' field. Since this field is not value of 255 in the IPv6 'hop-limit' field. Since this field is not
decremented when ip-protocol-41 packets traverse multiple IPv4 hops decremented when ip-protocol-41 packets traverse multiple IPv4 hops
([9], section 3), ISATAP links require a different trust model. In ([9], section 3), ISATAP links require a different trust model. In
particular, ONLY those Router Advertisements received from a member particular, ONLY those Router Advertisements received from a member
of the Potential Routers List are trusted; all others are silently of the Potential Routers List are trusted; all others are silently
discarded. This trust model is predicated on IPv4 source address discarded. This trust model is predicated on IPv4 source address
filtering, as described above. filtering, as described above.
The ISATAP address format does not support privacy extensions for The ISATAP address format does not support privacy extensions for
stateless address autoconfiguration [21]. However, since the ISATAP stateless address autoconfiguration [22]. However, since the ISATAP
interface identifier is derived from the node's IPv4 address, ISATAP interface identifier is derived from the node's IPv4 address, ISATAP
addresses do not have the same level of privacy concerns as IPv6 addresses do not have the same level of privacy concerns as IPv6
addresses that use an interface identifier derived from the MAC addresses that use an interface identifier derived from the MAC
address. (This issue is the same for NAT'd addresses.) address. (This issue is the same for NAT'd addresses.)
10. Acknowledgements 11. Acknowledgements
Some of the ideas presented in this draft were derived from work at Some of the ideas presented in this draft were derived from work at
SRI with internal funds and contractual support. Government sponsors SRI with internal funds and contractual support. Government sponsors
who supported the work include Monica Farah-Stapleton and Russell who supported the work include Monica Farah-Stapleton and Russell
Langan from U.S. Army CECOM ASEO, and Dr. Allen Moshfegh from U.S. Langan from U.S. Army CECOM ASEO, and Dr. Allen Moshfegh from U.S.
Office of Naval Research. Within SRI, Dr. Mike Frankel, J. Peter Office of Naval Research. Within SRI, Dr. Mike Frankel, J. Peter
Marcotullio, Lou Rodriguez, and Dr. Ambatipudi Sastry supported the Marcotullio, Lou Rodriguez, and Dr. Ambatipudi Sastry supported the
work and helped foster early interest. work and helped foster early interest.
The following peer reviewers are acknowledged for taking the time to The following peer reviewers are acknowledged for taking the time to
review a pre-release of this document and provide input: Jim Bound, review a pre-release of this document and provide input: Jim Bound,
Rich Draves, Cyndi Jung, Ambatipudi Sastry, Aaron Schrader, Ole Rich Draves, Cyndi Jung, Ambatipudi Sastry, Aaron Schrader, Ole
Troan, Vlad Yasevich. Troan, Vlad Yasevich.
The authors acknowledge members of the NGTRANS community who have The authors acknowledge members of the NGTRANS community who have
made significant contributions to this effort, including Rich Draves, made significant contributions to this effort, including Rich Draves,
Alain Durand, Nathan Lutchansky, Karen Nielsen, Art Shelest, Margaret Alain Durand, Nathan Lutchansky, Karen Nielsen, Art Shelest, Margaret
Wasserman, and Brian Zill. Wasserman, and Brian Zill.
The authors also wish to acknowledge the work of Quang Nguyen [22] The authors also wish to acknowledge the work of Quang Nguyen [23]
under the guidance of Dr. Lixia Zhang that proposed very similar under the guidance of Dr. Lixia Zhang that proposed very similar
ideas to those that appear in this document. This work was first ideas to those that appear in this document. This work was first
brought to the authors' attention on September 20, 2002. brought to the authors' attention on September 20, 2002.
Normative References Normative References
[1] Deering, S. and R. Hinden, "Internet Protocol, Version 6 (IPv6) [1] Deering, S. and R. Hinden, "Internet Protocol, Version 6 (IPv6)
Specification", RFC 2460, December 1998. Specification", RFC 2460, December 1998.
[2] Postel, J., "Internet Protocol", STD 5, RFC 791, September [2] Postel, J., "Internet Protocol", STD 5, RFC 791, September
skipping to change at page 15, line 25 skipping to change at page 15, line 27
IPv6 Hosts and Routers", draft-ietf-ngtrans-mech-v2-01 (work in IPv6 Hosts and Routers", draft-ietf-ngtrans-mech-v2-01 (work in
progress), November 2002. progress), November 2002.
[10] Conta, A. and S. Deering, "Internet Control Message Protocol [10] Conta, A. and S. Deering, "Internet Control Message Protocol
(ICMPv6) for the Internet Protocol Version 6 (IPv6) (ICMPv6) for the Internet Protocol Version 6 (IPv6)
Specification", RFC 2463, December 1998. Specification", RFC 2463, December 1998.
[11] McCann, J., Deering, S. and J. Mogul, "Path MTU Discovery for [11] McCann, J., Deering, S. and J. Mogul, "Path MTU Discovery for
IP version 6", RFC 1981, August 1996. IP version 6", RFC 1981, August 1996.
[12] Thomson, S. and T. Narten, "IPv6 Stateless Address [12] Braden, R., "Requirements for Internet Hosts - Communication
Autoconfiguration", RFC 2462, December 1998. Layers", STD 3, RFC 1122, October 1989.
[13] Droms, R., "Dynamic Host Configuration Protocol for IPv6
(DHCPv6)", draft-ietf-dhc-dhcpv6-28 (work in progress),
November 2002.
[14] Droms, R., "Dynamic Host Configuration Protocol", RFC 2131, [13] Droms, R., "Dynamic Host Configuration Protocol", RFC 2131,
March 1997. March 1997.
[14] Thomson, S. and T. Narten, "IPv6 Stateless Address
Autoconfiguration", RFC 2462, December 1998.
[15] Mogul, J. and S. Deering, "Path MTU discovery", RFC 1191, [15] Mogul, J. and S. Deering, "Path MTU discovery", RFC 1191,
November 1990. November 1990.
[16] Postel, J., "Internet Control Message Protocol", STD 5, RFC [16] Postel, J., "Internet Control Message Protocol", STD 5, RFC
792, September 1981. 792, September 1981.
[17] Baker, F., "Requirements for IP Version 4 Routers", RFC 1812, [17] Baker, F., "Requirements for IP Version 4 Routers", RFC 1812,
June 1995. June 1995.
[18] Droms, R., "Dynamic Host Configuration Protocol for IPv6
(DHCPv6)", draft-ietf-dhc-dhcpv6-28 (work in progress),
November 2002.
Informative References Informative References
[18] Carpenter, B. and K. Moore, "Connection of IPv6 Domains via [19] Carpenter, B. and K. Moore, "Connection of IPv6 Domains via
IPv4 Clouds", RFC 3056, February 2001. IPv4 Clouds", RFC 3056, February 2001.
[19] Mockapetris, P., "Domain names - implementation and [20] Mockapetris, P., "Domain names - implementation and
specification", STD 13, RFC 1035, November 1987. specification", STD 13, RFC 1035, November 1987.
[20] Droms, R., "Procedures and IANA Guidelines for Definition of [21] Droms, R., "Procedures and IANA Guidelines for Definition of
New DHCP Options and Message Types", BCP 43, RFC 2939, New DHCP Options and Message Types", BCP 43, RFC 2939,
September 2000. September 2000.
[21] Narten, T. and R. Draves, "Privacy Extensions for Stateless [22] Narten, T. and R. Draves, "Privacy Extensions for Stateless
Address Autoconfiguration in IPv6", RFC 3041, January 2001. Address Autoconfiguration in IPv6", RFC 3041, January 2001.
[22] Nguyen, Q., "http://irl.cs.ucla.edu/vet/report.ps", spring [23] Nguyen, Q., "http://irl.cs.ucla.edu/vet/report.ps", spring
1998. 1998.
[23] Lahey, K., "TCP Problems with Path MTU Discovery", RFC 2923, [24] Lahey, K., "TCP Problems with Path MTU Discovery", RFC 2923,
September 2000. September 2000.
Authors' Addresses Authors' Addresses
Fred L. Templin Fred L. Templin
Nokia Nokia
313 Fairchild Drive 313 Fairchild Drive
Mountain View, CA 94110 Mountain View, CA 94110
US US
skipping to change at page 17, line 15 skipping to change at page 17, line 15
Microsoft Corporation Microsoft Corporation
One Microsoft Way One Microsoft Way
Redmond, WA 98052-6399 Redmond, WA 98052-6399
US US
Phone: +1 425 703 8835 Phone: +1 425 703 8835
EMail: dthaler@microsoft.com EMail: dthaler@microsoft.com
Appendix A. Major Changes Appendix A. Major Changes
changes from version 09 to version 10:
o Rearranged/revised sections 5, 6, 7
o updated MTU section
changes from version 08 to version 09: changes from version 08 to version 09:
o Added stateful autoconfiguration mechanism o Added stateful autoconfiguration mechanism
o Normative references to RFC 2491, RFC 2462 o Normative references to RFC 2491, RFC 2462
o Moved non-normative MTU text to appendix C o Moved non-normative MTU text to appendix C
changes from version 07 to version 08: changes from version 07 to version 08:
skipping to change at page 19, line 26 skipping to change at page 19, line 28
ISATAP_MTU is larger than 1380 bytes, the encapsulator must implement ISATAP_MTU is larger than 1380 bytes, the encapsulator must implement
a dynamic link layer mechanism to discover per-neighbor MTUs. a dynamic link layer mechanism to discover per-neighbor MTUs.
IPv4 path MTU discovery [15] relies on ICMPv4 "fragmentation needed" IPv4 path MTU discovery [15] relies on ICMPv4 "fragmentation needed"
messages, but these do not provide enough information for stateless messages, but these do not provide enough information for stateless
translation into ICMPv6 "packet too big" messages (see: RFC 792 [16] translation into ICMPv6 "packet too big" messages (see: RFC 792 [16]
and RFC 1812 [17], section 4.3.2.3). Additionally, ICMPv4 and RFC 1812 [17], section 4.3.2.3). Additionally, ICMPv4
"fragmentation needed" messages can be spoofed, filtered, or not sent "fragmentation needed" messages can be spoofed, filtered, or not sent
at all by some forwarding nodes. Thus, IPv4 Path MTU discovery used at all by some forwarding nodes. Thus, IPv4 Path MTU discovery used
alone is inadequate and can result in black holes that are difficult alone is inadequate and can result in black holes that are difficult
to diagnose [23]. to diagnose [24].
The ISATAP encapsulator may implement an alternate per-neighbor MTU The ISATAP encapsulator may implement an alternate per-neighbor MTU
discovery mechanism, e.g., periodic and/or on-demand probing of the discovery mechanism, e.g., periodic and/or on-demand probing of the
IPv4 path to the decapsulator. Probing consists of sending packets IPv4 path to the decapsulator. Probing consists of sending packets
larger than 1380 bytes with the DF bit set in the IPv4 header. larger than 1380 bytes to the neighbor and receiving positive
Neighbor Solicitation (NS) packets with padding bytes added should be confirmation of receipt. Two methods are possible:
used for this purpose, since successful delivery results in a
positive acknowledgement that the probe succeeded, i.e., in the form
of a Neighbor Advertisement (NA) from the decapsulator. (NB: Setting
the DF bit prevents decapsulators from receiving probe packets that
would overrun the receive buffer on an underlying link, thus no
maximum receive unit (MRU) is required.)
Implementations may choose to couple the probing process with In the first method, the encapsulator does NOT set the DF bit in the
neighbor cache management procedures ([6], section 7), e.g. to IPv4 header of probe packets. In this case, the encapsulator must
maintain timers, state variables and/or a queue of packets waiting have a priori knowledge of the decapsulator's reassembly buffer size
for probes to complete. Packets retained on the queue are forwarded and should have a priori knowledge of the decapsulator's link MTU.
when probes succeed, and provide state for sending ICMPv6 "packet too This method has the advantage that probe packets will be delivered
big" messages to the source when probes fail. Implementations may even if the network performs fragmentation, thus ordinary data
choose to store per-neighbor MTU information in the IPv4 path MTU packets may be used for probing resulting in greater efficiency.
discovery cache, in the ISATAP link layer's private data structures, Disadvantages for this method include:
etc.
ICMPv4 "fragmentation needed" messages may result when a link o special mechanisms required on both encapsulator and decapsulator
restriction is encountered but may also come from denial of service
attacks. Implementations should treat ICMPv4 "fragmentation needed" o extra state required on both encapsulator and decapsulator
messages as "tentative" negative acknowledgments and apply heuristics
to determine when to suspect an actual link restriction and when to o complex protocol signalling between encapsulator and decapsulator
ignore the messages. IPv6 packets lost due actual link restrictions
are perceived as lost due to congestion by the original source, but o possible extended periods of network fragmentation
robust implementations minimize instances of such packet loss without In the second (and preferred) method, the encapsulator sets the DF
bit in the IPv4 header of probe packets. Neighbor Solicitation (NS)
packets with padding bytes added should be used for this purpose,
since successful delivery results in a positive acknowledgement that
the probe succeeded, i.e., in the form of a Neighbor Advertisement
(NA) from the decapsulator. Setting the DF bit prevents the network
from fragmenting the packets and protects decapsulators from
receiving packets that might overrun the IPv4 reassembly buffer.
Additionally, special mechanisms and state are needed only on the
encapsulator, and no complex protocol signalling between the
encapsulator and decapsulator is required.
In either method, implementations may choose to couple the probing
process with neighbor cache management procedures ([6], section 7),
e.g. to maintain timers, state variables and/or a queue of packets
waiting for probes to complete. Packets retained on the queue are
forwarded when probes succeed, and provide state for sending ICMPv6
"packet too big" messages to the source when probes fail.
Implementations may choose to store per-neighbor MTU information in
the IPv4 path MTU discovery cache, in the ISATAP link layer's private
data structures, etc.
Additional notes:
1. Per-neighbor MTUs may vary dynamically due to fluctuations in the
IPv4 forwarding path and/or multipath routing (e.g., when QoS
routing is used in the IPv4 network). For such neighbors,
encapsulators should detect a "losing battle" and reduce the
per-neighbor MTU size to no more than 1380 bytes.
2. When not probing, encapsulators may send packets to a neighbor
with MTU greater than 1380 bytes either with the DF bit set or
not set. When the DF bit is set, undetected packet loss may
occur in the network if the neighbor's MTU decreases. When the
DF bit is NOT set, undetected packet loss is less likely but may
occur either in the network or at the neighbor's reassembly
buffer.
3. ICMPv4 "fragmentation needed" messages may result when a link
restriction is encountered but may also come from denial of
service attacks. Implementations should treat ICMPv4
"fragmentation needed" messages as "tentative" negative
acknowledgments and apply heuristics to determine when to suspect
an actual link restriction and when to ignore the messages. IPv6
packets lost due actual link restrictions are perceived as lost
due to congestion by the original source, but robust
implementations minimize instances of such packet loss without
ICMPv6 "packet too big" messages returned to the sender. ICMPv6 "packet too big" messages returned to the sender.
Intellectual Property Statement Intellectual Property Statement
The IETF takes no position regarding the validity or scope of any The IETF takes no position regarding the validity or scope of any
intellectual property or other rights that might be claimed to intellectual property or other rights that might be claimed to
pertain to the implementation or use of the technology described in pertain to the implementation or use of the technology described in
this document or the extent to which any license under such rights this document or the extent to which any license under such rights
might or might not be available; neither does it represent that it might or might not be available; neither does it represent that it
has made any effort to identify any such rights. Information on the has made any effort to identify any such rights. Information on the
 End of changes. 

This html diff was produced by rfcdiff 1.25, available from http://www.levkowetz.com/ietf/tools/rfcdiff/