IPv6 addresses in DNS: Representing Internet Protocol version 6 (IPv6) Addresses in the Domain Name System (DNS)
The IETF settles the AAAA versus A6 argument: AAAA for production, A6 and bit labels demoted to experiments.
Bush, R., Durand, A., Fink, B., Gudmundsson, O., and T. Hain, "Representing Internet Protocol version 6 (IPv6) Addresses in the Domain Name System (DNS)", RFC 3363, DOI 10.17487/RFC3363, August 2002, <https://www.rfc-editor.org/info/rfc3363>.
@misc{rfc3363,
series = {Request for Comments},
number = 3363,
howpublished = {RFC 3363},
publisher = {RFC Editor},
doi = {10.17487/RFC3363},
url = {https://www.rfc-editor.org/info/rfc3363},
author = {R. Bush and A. Durand and B. Fink and O. Gudmundsson and T. Hain},
title = {{Representing Internet Protocol version 6 (IPv6) Addresses in the Domain Name System (DNS)}},
pagetotal = 6,
year = 2002,
month = aug,
} RFC text
Contents
Network Working Group R. Bush
Request for Comments: 3363 A. Durand
Updates: 2673, 2874 B. Fink
Category: Informational O. Gudmundsson
T. Hain
Editors
August 2002
Representing Internet Protocol version 6 (IPv6)
Addresses in the Domain Name System (DNS)
Status of this Memo
This memo provides information for the Internet community. It does
not specify an Internet standard of any kind. Distribution of this
memo is unlimited.
Copyright Notice
Copyright (C) The Internet Society (2002). All Rights Reserved.
Abstract
This document clarifies and updates the standards status of RFCs that
define direct and reverse map of IPv6 addresses in DNS. This
document moves the A6 and Bit label specifications to experimental
status.
1. Introduction
The IETF had begun the process of standardizing two different address
formats for IPv6 addresses AAAA [RFC1886] and A6 [RFC2874] and both
are at proposed standard. This had led to confusion and conflicts on
which one to deploy. It is important for deployment that any
confusion in this area be cleared up, as there is a feeling in the
community that having more than one choice will lead to delays in the
deployment of IPv6. The goal of this document is to clarify the
situation.
This document also discusses issues relating to the usage of Binary
Labels [RFC 2673] to support the reverse mapping of IPv6 addresses.
This document is based on extensive technical discussion on various
relevant working groups mailing lists and a joint DNSEXT and NGTRANS
meeting at the 51st IETF in August 2001. This document attempts to
capture the sense of the discussions and reflect them in this
document to represent the consensus of the community.
The main arguments and the issues are covered in a separate document
[RFC3364] that reflects the current understanding of the issues.
This document summarizes the outcome of these discussions.
The issue of the root of reverse IPv6 address map is outside the
scope of this document and is covered in a different document
[RFC3152].
1.1 Standards Action Taken
This document changes the status of RFCs 2673 and 2874 from Proposed
Standard to Experimental.
2. IPv6 Addresses: AAAA RR vs A6 RR
Working group consensus as perceived by the chairs of the DNSEXT and
NGTRANS working groups is that:
a) AAAA records are preferable at the moment for production
deployment of IPv6, and
b) that A6 records have interesting properties that need to be better
understood before deployment.
c) It is not known if the benefits of A6 outweigh the costs and
risks.
2.1 Rationale
There are several potential issues with A6 RRs that stem directly
from the feature that makes them different from AAAA RRs: the ability
to build up addresses via chaining.
Resolving a chain of A6 RRs involves resolving a series of what are
nearly-independent queries. Each of these sub-queries takes some
non-zero amount of time, unless the answer happens to be in the
resolver's local cache already. Other things being equal, we expect
that the time it takes to resolve an N-link chain of A6 RRs will be
roughly proportional to N. What data we have suggests that users are
already impatient with the length of time it takes to resolve A RRs
in the IPv4 Internet, which suggests that users are not likely to be
patient with significantly longer delays in the IPv6 Internet, but
terminating queries prematurely is both a waste of resources and
another source of user frustration. Thus, we are forced to conclude
that indiscriminate use of long A6 chains is likely to lead to
increased user frustration.
The probability of failure during the process of resolving an N-link
A6 chain also appears to be roughly proportional to N, since each of
the queries involved in resolving an A6 chain has roughly the same
probability of failure as a single AAAA query.
Last, several of the most interesting potential applications for A6
RRs involve situations where the prefix name field in the A6 RR
points to a target that is not only outside the DNS zone containing
the A6 RR, but is administered by a different organization entirely.
While pointers out of zone are not a problem per se, experience both
with glue RRs and with PTR RRs in the IN-ADDR.ARPA tree suggests that
pointers to other organizations are often not maintained properly,
perhaps because they're less susceptible to automation than pointers
within a single organization would be.
2.2 Recommended Standard Action
Based on the perceived consensus, this document recommends that RFC
1886 stay on standards track and be advanced, while moving RFC 2874
to Experimental status.
3. Bitlabels in the Reverse DNS Tree
RFC 2673 defines a new DNS label type. This was the first new type
defined since RFC 1035 [RFC1035]. Since the development of 2673 it
has been learned that deployment of a new type is difficult since DNS
servers that do not support bitlabels reject queries containing bit
labels as being malformed. The community has also indicated that
this new label type is not needed for mapping reverse addresses.
3.1 Rationale
The hexadecimal text representation of IPv6 addresses appears to be
capable of expressing all of the delegation schemes that we expect to
be used in the DNS reverse tree.
3.2 Recommended Standard Action
RFC 2673 standard status is to be changed from Proposed to
Experimental. Future standardization of these documents is to be
done by the DNSEXT working group or its successor.
4. DNAME in IPv6 Reverse Tree
The issues for DNAME in the reverse mapping tree appears to be
closely tied to the need to use fragmented A6 in the main tree: if
one is necessary, so is the other, and if one isn't necessary, the
other isn't either. Therefore, in moving RFC 2874 to experimental,
the intent of this document is that use of DNAME RRs in the reverse
tree be deprecated.
5. Acknowledgments
This document is based on input from many members of the various IETF
working groups involved in this issues. Special thanks go to the
people that prepared reading material for the joint DNSEXT and
NGTRANS working group meeting at the 51st IETF in London, Rob
Austein, Dan Bernstein, Matt Crawford, Jun-ichiro itojun Hagino,
Christian Huitema. Number of other people have made number of
comments on mailing lists about this issue including Andrew W.
Barclay, Robert Elz, Johan Ihren, Edward Lewis, Bill Manning, Pekka
Savola, Paul Vixie.
6. Security Considerations
As this document specifies a course of action, there are no direct
security considerations. There is an indirect security impact of the
choice, in that the relationship between A6 and DNSSEC is not well
understood throughout the community, while the choice of AAAA does
leads to a model for use of DNSSEC in IPv6 networks which parallels
current IPv4 practice.
7. IANA Considerations
None.
Normative References
[RFC1035] Mockapetris, P., "Domain Names - Implementation and
Specification", STD 13, RFC 1035, November 1987.
[RFC1886] Thompson, S. and C. Huitema, "DNS Extensions to support IP
version 6", RFC 1886, December 1995.
[RFC2673] Crawford, M., "Binary Labels in the Domain Name System",
RFC 2673, August 1999.
[RFC2874] Crawford, M. and C. Huitema, "DNS Extensions to Support
IPv6 Address Aggregation and Renumbering", RFC 2874, July
2000.
[RFC3152] Bush, R., "Delegation of IP6.ARPA", BCP 49, RFC 3152
August 2001.
Informative References
[RFC3364] Austein, R., "Tradeoffs in Domain Name System (DNS)
Support for Internet Protocol version 6 (IPv6)", RFC 3364,
August 2002.
Editors' Addresses
Randy Bush
EMail: randy@psg.com
Alain Durand
EMail: alain.durand@sun.com
Bob Fink
EMail: fink@es.net
Olafur Gudmundsson
EMail: ogud@ogud.com
Tony Hain
EMail: hain@tndh.net
Full Copyright Statement
Copyright (C) The Internet Society (2002). All Rights Reserved.
This document and translations of it may be copied and furnished to
others, and derivative works that comment on or otherwise explain it
or assist in its implementation may be prepared, copied, published
and distributed, in whole or in part, without restriction of any
kind, provided that the above copyright notice and this paragraph are
included on all such copies and derivative works. However, this
document itself may not be modified in any way, such as by removing
the copyright notice or references to the Internet Society or other
Internet organizations, except as needed for the purpose of
developing Internet standards in which case the procedures for
copyrights defined in the Internet Standards process must be
followed, or as required to translate it into languages other than
English.
The limited permissions granted above are perpetual and will not be
revoked by the Internet Society or its successors or assigns.
This document and the information contained herein is provided on an
"AS IS" basis and THE INTERNET SOCIETY AND THE INTERNET ENGINEERING
TASK FORCE DISCLAIMS ALL WARRANTIES, EXPRESS OR IMPLIED, INCLUDING
BUT NOT LIMITED TO ANY WARRANTY THAT THE USE OF THE INFORMATION
HEREIN WILL NOT INFRINGE ANY RIGHTS OR ANY IMPLIED WARRANTIES OF
MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE.
Acknowledgement
Funding for the RFC Editor function is currently provided by the
Internet Society.
AI ELI5
On this page
§1Why a Decision Was Needed
By 2002 the IETF had begun standardizing two different ways to put an IPv6 address in the DNS: the AAAA record (RFC 1886) and the A6 record (RFC 2874). Both were Proposed Standards. The memo says this led to confusion and conflicts over which one to deploy, and that having more than one choice could delay the deployment of IPv6.
The goal is simply to clarify the situation. The memo is based on discussion on several working group mailing lists and on a joint DNSEXT and NGTRANS meeting at the 51st IETF in August 2001. The detailed arguments are in a separate document, RFC 3364; this memo summarizes the outcome. Who runs the root of the reverse IPv6 map is out of scope and is covered by RFC 3152.
In the standards action section, the memo states that it changes the status of RFCs 2673 and 2874 from Proposed Standard to Experimental.
Two groups invented two different ways to write long internet addresses in the internet's phone book. People got confused about which to use, so this memo steps in to sort it out.
§2AAAA Versus A6
The perceived consensus of the DNSEXT and NGTRANS chairs was that AAAA records are preferable for production deployment of IPv6 at the moment, that A6 records have interesting properties that need to be better understood before deployment, and that it is not known whether the benefits of A6 outweigh the costs and risks.
The rationale centers on the feature that makes A6 different: chaining, where an address is built up from several records. Resolving an N-link chain means resolving a series of nearly independent queries, so the time is expected to be roughly proportional to N, and the chance of failure also appears to be roughly proportional to N. The memo worries that users, already impatient with DNS lookups for IPv4, would not tolerate much longer delays.
A further concern is that the most interesting uses of A6 point at targets in other zones run by other organizations. The memo notes that experience with glue records and with PTR records in the IN-ADDR.ARPA tree suggests that pointers to other organizations are often not maintained properly.
The recommendation: RFC 1886 stays on the standards track and is advanced, while RFC 2874 moves to Experimental.
The fancy way builds an address by asking many other places for pieces, like a scavenger hunt. Every extra stop takes time and can go wrong, so the simple way, where the whole address sits in one place, wins for now.
| Specification | What it covers | Action recommended |
|---|---|---|
| RFC 1886 | AAAA records for IPv6 addresses | Stay on the standards track and be advanced |
| RFC 2874 | A6 records for IPv6 addresses | Move to Experimental |
| RFC 2673 | Binary (bit) labels | Move from Proposed to Experimental |
§3Bit Labels in the Reverse Tree
RFC 2673 defined a new kind of DNS label, the binary (bit) label, the first new label type since RFC 1035. The memo reports that deploying a new label type turned out to be difficult, because DNS servers that do not support bit labels reject queries containing them as malformed. The community also indicated that this label type is not needed for mapping reverse addresses.
The reasoning is that the hexadecimal text representation of IPv6 addresses appears able to express all of the delegation schemes expected to be used in the reverse DNS tree.
The recommended action is to change the status of RFC 2673 from Proposed to Experimental. Any future standardization is left to the DNSEXT working group or its successor.
Someone invented a brand new kind of label for the phone book, but older phone book machines choked on it. Since the normal way of writing the numbers already works, the new label is put on the shelf as an experiment.
§4DNAME in the IPv6 Reverse Tree
The memo says the question of DNAME records in the reverse mapping tree is closely tied to the need for fragmented A6 in the main tree: if one is necessary, so is the other, and if one is not, the other is not either.
Therefore, in moving RFC 2874 to Experimental, the intent of the memo is that the use of DNAME RRs in the reverse tree be deprecated.
A DNAME is a shortcut sign in the phone book. Because the shortcut only mattered for the fancy address method, it goes away from the reverse phone book too.
§5Security Considerations
Because the document specifies a course of action, it says there are no direct security considerations. It does note an indirect impact of the choice: the relationship between A6 and DNSSEC is not well understood throughout the community, while choosing AAAA leads to a model for using DNSSEC in IPv6 networks that parallels current IPv4 practice. The memo has no IANA actions.
Nothing new is being built here, so there is nothing new to break into. Sticking with the simpler method just keeps the lock-and-key rules the same as before.