Network Working Group Robert Elz
Internet Draft University of Melbourne
Expiration Date: November 1996
May 1996
Identifying IPv6 Interfaces in Link-Local Addresses
draft-ietf-ipngwg-iid-02.txt
Status of this Memo
This document is an Internet-Draft. Internet-Drafts are working
documents of the Internet Engineering Task Force (IETF), its areas,
and its working groups. Note that other groups may also distribute
working documents as Internet-Drafts.
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."
To learn the current status of any Internet-Draft, please check the
"1id-abstracts.txt" listing contained in the Internet-Drafts Shadow
Directories on ftp.is.co.za (Africa), nic.nordu.net (Europe),
munnari.oz.au (Pacific Rim), ds.internic.net (US East Coast), or
ftp.isi.edu (US West Coast).
1. Abstract
This draft proposes a change to the way that IPv6 link local
addresses are constructed, so that a node can guarantee that all of
its link local addresses are unique within the node. The current
definition of a link local address, a well known prefix, some number
of zero bits, and a link specific unique token, ensures that it will
be unique on the link, which is what is required for communications
using those addresses over the link, but does not require uniqueness
of the addresses within a node. In some cases all of a nodes
interfaces may share the same link local address. Even the
possibility of this means that link local addresses, which may in
some situations be the only addresses that exist, cannot be used for
internal definition of interfaces, or other purposes within the node.
This draft suggests a method by which nodes may overcome this
problem, by redefining the bits between the prefix and the token to
be available to be used by the node to cause the addresses to be
unique.
kre [Page 1]
Internet Draft draft-ietf-ipngwg-iid-02.txt May 1996
2. Introduction
IPv6 created link local addresses, by which all interfaces could
always be numbered, regardless of any other addressing which may, or
may not, be available. These addresses are only suitable for
communicating within that link, and are only unique on the link.
[Addrconf] defines link local addresses, the various IPv6 over foo
specs define mechanisms for generating link local addresses such that
they are highly likely to be unique, and [addrconf] defines methods
for detecting most cases in which this procedure has failed to
generate unique link local addresses.
However, nothing, so far, has defined any method for ensuring that
link local addresses are unique within a node, nor for that matter,
for justifying that any such uniqueness is useful. This draft
attempts to achieve both of those purposes.
It is intended that this draft serve the purpose of encouraging
debate and discussion on this issue, and then perhaps cause some
modifications to published RFCs, or other drafts. It is not intended
that this draft ever, itself, be more widely published. Because of
this, several terms (eg: "addrconf") are not defined here, and
assumed to be understood by the reader. Examination of IPv6 RFCs and
drafts should provide explanations.
No apology is made for failing to misspell "neighbour" in this draft.
3. Identifying Interfaces
It is usual for a node with more than one interface to need, from
time to time, a mechanism to identify a particular interface amongst
the interfaces available. Currently, with IPv4, this has been done
on an ad hoc basis, as IPv4 addresses could not be used. Not all
interfaces necessarily posses an IPv4 address.
However, with IPv6, all (IPv6) interfaces will have a link local
address. This address is intended to allow communications over the
attached links and so is defined to be usable only on that link.
With a minor modification, the link local address could also serve
the purpose of identifying interfaces within a node for IPv6. This
relies upon all interfaces having a link local address, however this
is already specified by [addrconf]. It also relies on the link local
addresses being unique within the node, which is a property they do
not currently have.
If this is adopted, implementations could, if they wanted, use link
local addresses as their standard method for interface identification
kre [Page 2]
Internet Draft draft-ietf-ipngwg-iid-02.txt May 1996
for IPv6, eliminating the various methods used for IPv4, often not
very satisfactory.
Many IPv4 implementations have used the concatenation of a variable
length interface-type string, with a number indicating which
interface of that type this is (eg: "le1", or "fta0"). This has
multiple drawbacks, for coding it means the need to deal with a
different kind of label than the addresses that are used to refer to
interfaces on all other nodes, there is usually no stable number
which can be expected to remain invariant over hardware
reconfigurations (installing a new interface may change the number
assigned to one that was otherwise untouched), and there are usually
no tools available for mapping more user friendly (or stable) names
to these identifiers.
Without this change, using link local addresses to identify
interfaces is not possible, unless an implementation can guarantee,
by some other means, that the tokens used by all interfaces will
differ, always.
4. The method.
A link local address is created by taking a well known prefix
(FE80::/10) [addrspec] and appending a link dependent unique token in
the low order bits [addrconf]. The precise method, and means of
generating the unique token is specified in the various "IP over foo"
specifications for links of type "foo" [IPv6/Ether, IPv6/FDDI, ...].
For the purposes of this draft, the current [addrspec] link local
address shall be considered to be comprised of three fields, the
prefix, the intermediate-zeroes, and the token.
This draft proposes inserting a new field between the token and the
prefix. That is, the intermediate-zeroes will be split into two
fields, which we shall call the interface identifier, and the
discretionary bits.
The interface identifier will be node defined with the sole purpose
of disambiguating interfaces within a node if the token required on
several links happens to be the same.
The discretionary bits can be used by the node for any purpose, and
are no longer required to be zero.
kre [Page 3]
Internet Draft draft-ietf-ipngwg-iid-02.txt May 1996
4.1. Placement Option 1
The new interface identifier field will be placed in towards the left
of the link local address. This is to guarantee that it can be in
the same positions for all link types, regardless of the length of
the token to be appended. This guarantees that if the interface
identifier is unique within the node for all interfaces, then the
generated link local addresses will also be unique within the node
for all interfaces, regardless of the values of the discretionary
bits or token. It also achieves maximal benefit for the "::"
notational convention by keeping as many zeroes as possible in
contiguous positions, though implementations are permitted to place
any value they desire in the discretionary bits. In this option, the
discretionary bits will be in lesser significant bit positions than
the interface identifier.
It is suggested that the low order 16 bits of the high order 32 bits
of the link local address be allocated to interface identifier. This
allows another 6 bits between the current defined prefix, and the new
field, which will be reserved for future use, potentially to define a
different format for link local addresses at some future date. Using
high order bits also has ramifications (or more precisely non-
ramifications) with respect to multicast address selection for
neighbour discovery, which will be expanded upon below. The various
"IPv6 over foo" specifications will be altered to show this field.
4.2. Placement Option 2
The new interface identifier field will be placed in towards the
right of the link local address. This places it just next to the
token, which is assumed to have a defined maximum length. The
interface identifier field will be at a fixed bit position regardless
of token length, to the left of the longest possible token. This
guarantees that if the interface identifier is unique within the node
for all interfaces, then the generated link local addresses will also
be unique within the node for all interfaces, regardless of the
values of the discretionary bits or token. It also achieves maximal
benefit for the "::" notational convention by keeping as many zeroes
as possible in contiguous positions, though implementations are
permitted to place any value they desire in the discretionary bits.
In this option, the discretionary bits will be in more significant
bit positions than the interface identifier.
It is suggested that the high order 16 bits of the low order 64 bits
of the link local address be allocated to interface identifier. This
assumes that the longest permitted token is 48 bits. These bits are
beyond the range of bits used for multicast address selection for
neighbour discovery, which will be expanded upon below. The various
kre [Page 4]
Internet Draft draft-ietf-ipngwg-iid-02.txt May 1996
"IPv6 over foo" specifications will be altered to show this field.
4.3. The New Fields
The "addrconf" specification will define the contents of the
interface identifier field. This will specify that a node may insert
any value in this field that it desires, but that it is intended that
the field be used to cause all link local addresses assigned to the
node to be unique. It will be recommended that nodes use the field
to hold some interface hardware-specific value (or software-specific,
for virtual interfaces) which is likely to remain constant over time,
even if similar interfaces are added or removed. Thus, this field
alone will be unique amongst all interfaces to the node, and so is
sufficient to identify interfaces, regardless of whether or not the
token varies from one interface to another.
The discretionary bits may be used by the implementation for any
purpose.
5. Duplicate address detection
[Addrconf] specifies that any address generated by a node must be
tested for uniqueness by being tested by the Duplicate Address
Detection (DAD) algorithm.
For link local addresses as originally defined, this amounts to a
test for uniqueness of the link specific token on the link, as all
other bits in the address are the same for all nodes.
[Addrconf] uses this feature to permit DAD to be avoided for
additional addresses created from the same token on the same link,
when created in a standard manner - such as that specified for
stateless autoconfiguration in [addrconf]. Such addresses are formed
the same way in all nodes on the link, with the token inserted -
known uniqueness of the token guarantees uniqueness within the same
scope of the other generated address. Uniqueness in wider scopes is
derived from the known properties of the prefix to which the token is
appended.
As modified here, performing DAD on the link local address does not
any longer amount to any uniqueness guarantee of the token, as two
link local addresses (from different nodes) may have the same token,
yet differ in interface identifier, or discretionary bits.
To avoid the extra burden of testing all autonomously configured
addresses, this drafts specifies than when testing a link local
address for uniqueness using DAD, the address tested shall be formed
with the interface identifier and discretionary bits, that is, the
kre [Page 5]
Internet Draft draft-ietf-ipngwg-iid-02.txt May 1996
intermediate-zeroes, set to zero. This is the same address that was
previously tested.
Whenever a node that implements this modification receives a
Neighbour Solicitation packet containing a target address with prefix
FE80::/10 it should consider only those 10 bits, and the final token
bits, for which the size will be determined by the nature of the link
over which the NS is received, when comparing the target address
requested, and its own link local address for the link, all other
bits of both the target address sought, and the local link local
address should be considered to be zero.
Nodes that have been implemented according to previous versions of
[Addrconf] and [Discovery], or which simply do not wish to make use
of the interface identifier or discretionary bits, will have zeroes
in all the intermediate-zeroes bits of their own link local
addresses. When processing a ND packet sent for DAD purposes, simple
comparison of the addresses will work, as that ND packet will also
contain zeros in the intermediate-zeroes bits. Other ND packets may
contain other values in those bits, but this node will only be
concerned with those containing its own link local address, and that
will be as this node defined it.
The Neighbour Advertisement returned in response to receiving a
Neighbour Solicitation containing a target address with prefix
FE80::/10, and sent from the unspecified address, that is, a DAD
packet, should contain the responding node's link local address,
modified to contain only zeroes in the intermediate-zeroes bits. A
node that has not been modified to implement this specification will
do this naturally, as the intermediate-zeroes bits in its link local
address will all be zero anyway. Nodes implementing these
modifications must check received NS packets for their source address
being unspecified, which they must do anyway to reply correctly, and
if so, and if a reply is to be sent, ensure that the intermediate-
zeroes bits are cleared from the link local address.
When receiving Neighbour Advertisement packets during the DAD
process, and the target address therein has a prefix of FE80::/10, a
node implementing this specification must consider only the prefix,
and token when comparing the returned target address and the
tentative link local address for the node. The intermediate-zeroes
bits must, for this purpose, and only this purpose, be treated as if
they were set to zero the tentative link local address. Those bits
will always be zero in the NA reply received.
This procedure then returns DAD of the link local address to being a
uniqueness test of the token on the link, which then allows the token
to be used to generate other unique addresses without testing those -
kre [Page 6]
Internet Draft draft-ietf-ipngwg-iid-02.txt May 1996
including the link local address as defined here.
Note that as defined here, an implementation that chooses not to make
use of the interface identifier or discretionary bits fields, and
always uses them as intermediate-zeroes as currently defined, need
not alter its implementation from that pertaining to the previous
definition of link local addresses. Received NS packets for DAD will
contain zeroes in all relevant bits, and thus will appear as if this
specification did not exist. The returned NA from the node will
contain the interface link local address, just as now, which will be
the address with the intermediate-zeroes field. Ordinary NS and NA
packets directed to and from the node, will contain the node's link
local address, as always, whatever bits it contains, as would any
other packets to or from the node using its link local address.
Nodes must thus never assume that link local addresses will always
contain zeroes in the intermediate-zeroes bits. Nodes should not
ever be tempted to make assumptions about the values of any addresses
of any other nodes.
[Addrconf] and [Discovery] will be updated, as specified below, to
include these procedures.
6. Multicast Address Generation
[Discovery] specifies the algorithm by which the solicited-node
multicast address is generated. That uses only the low bits of the
IPv6 address. By positioning the interface identifier in the upper
bits of the link local address, the same solicited-node multicast
address will be generated, whatever interface identifier is chosen.
On some media, the token might be less than 32 bits wide. In such
cases the node may choose to use some of the bits used for multicast
address generation for other purposes. This draft strongly
recommends against this practice, but does not prohibit it. However,
nodes must be prepared to receive Neighbour Solicitation packets sent
to either the node's link local address, or to the address formed
from the prefix FE80::/16 and the interface token, with zeroes
between (in the intermediate-zeroes field). This may mean accepting
two different multicast addresses where one would ordinarily be
sufficient.
kre [Page 7]
Internet Draft draft-ietf-ipngwg-iid-02.txt May 1996
7. Rationale
By guaranteeing that all interfaces have unique addresses within the
node, the node can use those unique addresses to identify the
interface, rather than having to invent some new name space for this
purpose. Using addresses also allows all the standard tools to be
used for interface identification, that are used for host
identification. Interfaces can be named by Domain Name System names,
which can be manipulated in the same way as other DNS names, and
translated by standard routines into the addresses used as interface
identifiers.
Because a link local address must exist for all interfaces [addrconf]
and must always be generated in a standard way, the address is
effectively available for internal uses instantly the interface is
made known to the rest of the system - even before DAD is performed.
This means that no other interface identification is required, the
unique link local address can be, if the implementation desires, the
only interface identification provided. If DAD fails, [addrconf]
specifies that the interface be shut down. Even then the link local
address will still be unique within the node, and can still be used
to refer to the inactive interface.
With all interfaces identified by unique intra-node addresses,
implementations can, if they desire, make use of address fields in
the regular API for interface identification. For example, an
application might make use of source routing via a local interface in
order to direct packets to be transmitted over a specific link -
which would not imply that a routing header would be present in the
packet transmitted if the local address was all it contained, or that
that address would ever be present in a transmitted routing header.
Directing packets through specific interfaces, especially if
unnumbered, has not been easy, or even possible, in many current IP
implementations.
Implementations with multiple tokens available to create link local
addresses have the opportunity to connect multiple interfaces to one
link if that will produce a useful configuration - perhaps to achieve
higher bandwidth through a switch that permits multiple parallel
conversations. An implementation with only a single token would need
to invent a token, which is certainly not to be recommended, survive
with a shared link local address, or modify some other part of the
link local address, presumably using the technique described here.
With unique link local addresses, stateful autoconfiguration could be
used to obtain other addressing, where stateless autoconfiguration
would generate the same addresses for all such interfaces. Without
that, there would be no unique token for DHCP to use to assign
different addresses,
kre [Page 8]
Internet Draft draft-ietf-ipngwg-iid-02.txt May 1996
Where the node concerned is participating in routing protocols, such
as OSPF for IPv6, it also seems that having multiple interfaces on
one link sharing the same link local address might be a problem.
(Precise details unavailable...)
8. Specific Document Changes Suggested
In order to implement this change, the documents listed in this
section must be modified as indicated, or in such a way as to lead to
the same effect. Some of the documents concerned are, at the date of
this memo, available only as Internet Drafts. Because the changes
below relate to a specific version of each document, the normal
practice of never referencing an Internet Draft other than as "work
in progress" will be suspended here, and specific drafts will be
referenced.
Note that in these changes, the reference annotations have been
changed so as to be compatible with those used in this document,
rather than those used in the original. This is for clarity here
only, clearly the references in each case should be in the format
required by the specific document.
8.1. Changes to RFC1884
In section 2.4.8, on page 11, the diagram showing the format of a
link local address must be changed to either:
| 10 | 16
| bits | 6 | bits| 96-n bits | n bits |
+----------+---+-----+---------------+----------------------------+
|1111111010| 0 | IID | discretionary | interface ID |
+----------+---+-----+---------------+----------------------------+
or:
| 10 | 16
| bits | 54 bits | bits| 48 bits |
+----------+-------------------+-----+----------------------------+
|1111111010| discretionary | IID | interface ID |
+----------+-------------------+-----+----------------------------+
Which is chosen will depend upon which placement of the interface
identifier (IID) is decided to be better. See section 4.
No other changes are required.
kre [Page 9]
Internet Draft draft-ietf-ipngwg-iid-02.txt May 1996
8.2. Changes to draft-ietf-addrconf-ipv6-auto-07.txt
8.2.1. Changes to Terminology
In section 2, somewhere on pages 5 to 8, the following definitions
should be added:
interface identifier
- An integer identifier for each interface attached to a
node that is unique within the node, and relatively
stable. That is, the identifier should not normally
change across resets of the IPv6 layer, nor reboots of
the node, and should generally be unchanged by the
addition of deletion of other interfaces.
discretionary bits
- A value that the implementation or node can set to any
value for any purpose when defining the enclosing
structure or value. Once set, the bits must not be
altered as long as the particular structure remains
defined, or the value required.
8.2.2. Changes to link local address formation
In section 5.3, in the paragraph beginning "A link-local address is
formed ..." on page 15, will be replaced by the following paragraphs.
A link local address is formed from five component parts. The
most significant bits are always the well known prefix FE80::/10
[Addrspec]. The least significant bits are the interface token,
chosen in an interface specific way. Between those are placed
fields containing zeroes (padding), an interface identifier
field, and a field containing any value that the implementation
desires to place there.
The interface identifier field is a 16 bit field that can
contain either zeroes, or a value chosed by the implementation
that is unique among all interfaces connected to the node. Such
a value should be chosen in a way that long term stability of
the value chosen is likely. That is, neither a random number,
nor a simple count of interfaces as they are configured is
appropriate. Suitable choices would relate to the particular
hardware configuration, such as the slot number used on the bus,
or the address in I/O space of the interface adaptor, or
similar. It is highly desirable that the interface identifier
not alter merely because some other interface is added or
removed.
kre [Page 10]
Internet Draft draft-ietf-ipngwg-iid-02.txt May 1996
Following those paragraphs, one of the following paragraphs will be
added, depending on which choice is made for the positioning of the
interface identifier field.
Either:
The fields of the link local address will be formed in the
order, from most to least significant bits, the well known
prefix (FE80::/10) first, then a six bit zero field, then the 16
bit interface identifier field, then an M bit discretionary
field, followed by an N bit interface token. M shall be 96 - N.
If the interface token is more than 96 bits in length,
autoconfiguration fails, and manual configuration is required.
Or:
The fields of the link local address will be formed in the
order, from most to least significant bits, the well known
prefix (FE80::/10) first, then 54 discretionary bits, then the
16 bit interface identifier field, followed by an optional
variable width field of zeroes, followed by an N bit interface
token. If the interface token is less than 48 bits, the zero
filled field is included, and is 48 - N bits wide. If the
interface token is exactly 48 bits, no field of zeroes is
included. If the interface token is more than 48 bits in
length, autoconfiguration fails, and manual configuration is
required.
8.2.3. Changes to Duplicate Address Detection
In section 5.4.2, on page 16, the paragraph which begins "To check an
address, a node sends ..." shall be replaced by:
To check an address, a node sends DupAddrDetectTransmits
Neighbor Solicitations, each separated by RetransTimer
milliseconds. If the tentative address is not a link local
address, the solicitation's Target Address is set to the address
being checked. If the tentative address is a link local
address, the solicitation's Target Address is set to the address
being checked, modified such that only the prefix and interface
token fields are non-zero, all other bits of the link local
tentative address must be zero. The IP source is set to the
unspecified address and the IP destination is set to the
solicited-node multicast address of the Target Address.
In section 5.4.3, on page 17, the following paragraph should precede
the paragraphs that currently exist:
kre [Page 11]
Internet Draft draft-ietf-ipngwg-iid-02.txt May 1996
While performing Duplicate Address Detection for a link local
address, the node must consider a received Neighbor Solicitation
message sent with a Target Address that is also a link local
address, and which has the same interface token value as the
node's tentative link local address, to be a Neighbor
Solicitation sent to that tentative link local address.
In section 5.4.4 on page 18, the following paragraph should precede
the paragraph that currently exists:
While performing Duplicate Address Detection for a link local
address, the node must consider a received Neighbor
Advertisement message sent with a Target Address that is also a
link local address, and which has the same interface token value
as the node's tentative link local address, to be a Neighbor
Advertisement sent to that tentative link local address.
8.3. Changes to draft-ietf-ipngwg-discovery-06.txt
In section 4.3, on page 24, the Target Address field of the Neighbour
Solicitation packet is changed to be:
The IP address of the target of the solicitation. It MUST NOT
be a multicast address. If Duplicate Addres Detection is in
progress [Addrconf] and the address being tested ia a link local
address, then only the prefix (FE80::/10) and the interface
token are non-zero, all other bits MUST be zero.
In section 4.4, on page 26, the paragraph describing the Target
Address field of a Neighbour Advertisment shall be augmented as
follows.
Note that when responding to a Neighbour Solicitation sent from
the unspecified address (Source Address of the Neighbour
Solicitation) seeking a link local address, the Target Address
MUST be the link local address modified by including only the
prefix (FE80::/10) and interface token fields. All other bits
MUST be zero. This should be equivalent to the address in the
Target Address field of the Neighbour Solicitation packet, but
need not be identical to the node's link local address on the
interface.
In section 7.2.4, on page 61, the paragraph that begins "If the
source of the solicitation is the unspecified address, ..." will be
altered to be:
If the source of the solicitation is the unspecified address,
the node MUST set the Solicited flag to zero and multicast the
kre [Page 12]
Internet Draft draft-ietf-ipngwg-iid-02.txt May 1996
advertisement to the all-nodes address. Further, if the Target
Address is a link local address, then only the prefix and
interface token fields may have non-zero values, all other bits
of the Target Address MUST be zero. Otherwise, the node MUST
set the Solicited flag to one and unicast the advertisement to
the Source Address of the solicitation.
8.4. Changes to draft-ietf-ipngwg-ethernet-ntwrks-02.txt
On page 2, in the section entitled "Stateless Autoconfiguration and
Link-Local Addresses", the paragraph which begins: "The IPv6 Link-
local address ..." will be changed to be:
The IPv6 Link-local address [Addrspec] for an Ethernet interface
is formed by appending the interface's IEEE 802 address to the
80-bit prefix created by appending the interface's host specific
interface identifier [Addrconf] and other host or implementation
defined bits to the well known ten bit prefix FE80::/10.
The diagram that immediately follows will be replaced by either:
+-------+-------+-------+-------+-------+-------+------+------+
| FE 80 | IID | <--- discretionary value -- |
+-------+-------+-------+-------+-------+-------+------+------+
| --- ---> | Ethernet Address |
+-------+-------+-------+-------+-------+-------+------+------+
or:
+-------+-------+-------+-------+-------+-------+------+------+
| FE 80 <--- discretionary value --- --- --- ---> |
+-------+-------+-------+-------+-------+-------+------+------+
| IID | Ethernet Address |
+-------+-------+-------+-------+-------+-------+------+------+
Which is chosen will depend upon which placement of the interface
identifier (IID) is decided to be better.
8.5. Changes to draft-ietf-ipngwg-fddi-ntwrks-03.txt
On page 4, in the section entitled "Stateless Autoconfiguration and
Link-Local Addresses", the paragraph which begins: "The IPv6 Link-
local address ..." will be changed to be:
The IPv6 Link-local address [Addrspec] for an FDDI interface is
formed by appending the interface's IEEE 802 address to the 80-
bit prefix created by appending the interface's host specific
interface identifier [Addrconf] and other host or implementation
kre [Page 13]
Internet Draft draft-ietf-ipngwg-iid-02.txt May 1996
defined bits to the well known ten bit prefix FE80::/10.
The diagram that immediately follows will be replaced by either:
+-------+-------+-------+-------+-------+-------+------+------+
| FE 80 | IID | <--- discretionary value -- |
+-------+-------+-------+-------+-------+-------+------+------+
| --- ---> | FDDI Address |
+-------+-------+-------+-------+-------+-------+------+------+
or:
+-------+-------+-------+-------+-------+-------+------+------+
| FE 80 <--- discretionary value --- --- --- ---> |
+-------+-------+-------+-------+-------+-------+------+------+
| IID | FDDI Address |
+-------+-------+-------+-------+-------+-------+------+------+
Which is chosen will depend upon which placement of the interface
identifier (IID) is decided to be better.
8.6. Changes to draft-ietf-ipngwg-pppext-ipv6cp-02.txt
In section 5, the paragraph which begins "The most significant 10
bits of the address" on page 10 will be replaced by either:
The most significant 10 bits of the address is the Link-Local
prefix FE80::. This is followed by six more bits of zero, then
the 16 interface identifier bits, then 64 discretionary bits,
followed by the Interface Token field.
or:
The most significant 10 bits of the address is the Link-Local
prefix FE80::. This is followed by 54 discretionary bits, then
the 16 interface identifier bits, then 16 bits of zero, followed
by the Interface Token field.
The choice of replacement text will depend upon which placement of
the interface identifier field is considered best. The immediately
preceding diagram in the draft will be replaced by either:
| 10 bits | 6 | 16 | 64 bits | 32 bits |
+----------+---+-----+--------------------------+-----------------+
|1111111010| 0 | IID | discretionary | Interface Token |
+----------+---+-----+--------------------------+-----------------+
or:
kre [Page 14]
Internet Draft draft-ietf-ipngwg-iid-02.txt May 1996
| 10 bits | 54 bits | 16 | 16 | 32 bits |
+----------+------------------------------------+-----------------+
|1111111010| discretionary | IID | 0 | Interface Token |
+----------+------------------------------------+-----------------+
Which is chosen will depend upon which placement of the interface
identifier (IID) is decided to be better. Note that in the second
option, the token field remains 48 bits wide though the PPP token is
defined as only being 32 bits. The remaining 16 bits are zero
padded. This ensures that the interface identifier field will remain
in the same bit positions in all link local addresses.
8.7. Changes to other link specific documents
Other documents specifying IPv6 implementation details on other link
types will be changed in ways similar to those indicated above.
9. Security Considerations
Addressing and security are unrelated concepts. Attempts to pretend
otherwise are misguided.
10. References
[IPv6] "Internet Protocol, Version 6 (IPv6) Specification",
S. Deering, R. Hinden, RFC1883, January 4, 1996.
[Addrspec] "IP Version 6 Addressing Architecture",
R. Hinden, S. Deering, RFC1884, January 4, 1996.
[IPv6/Ether] "A Method for the Transmission of IPv6 Packets over
Ethernet Networks"
Matt Crawford, Work In Progress (soon to be an RFC).
[IPv6/FDDI] "A Method for the Transmission of IPv6 Packets over
FDDI Networks"
Matt Crawford, Work In Progress (soon to be an RFC).
[Discovery] "Neighbor Discovery for IP Version 6 (IPv6)"
Thomas Narten, Erik Nordmark, W A Simpson,
Work In Progress (soon to be an RFC).
[Addrconf] "IPv6 Stateless Address Autoconfiguration"
Susan Thomson, Thomas Narten,
Work In Progress (soon to be an RFC).
kre [Page 15]
Internet Draft draft-ietf-ipngwg-iid-02.txt May 1996
[DHCPv6] "Dynamic Host Configuration Protocol for IPv6 (DHCPv6)"
J. Bound, C. Perkins,
Work In Progress (soon to be an RFC).
11. Acknowledgements
My thanks to Matt Crawford for assisting getting this draft into a
semi-presentable state, which is not to imply that he agrees with the
proposal. Also to Dennis Ferguson for pointing out additional
justification for use of a method like this. The IPNGWG working
group also devoted some valuable meeting time to discussion of this
issue, all who contributed there also contributed here.
12. Author's Address
Robert Elz
University of Melbourne
Parkville
Victoria 2052
Australia
EMail: kre@munnari.OZ.AU
kre [Page 16]
