WDIFF

draft-ietf-dnsop-ipv6-dns-issues-10.txt --> draft-ietf-dnsop-ipv6-dns-issues-11.txt

DNS Operations WG A. Durand
Internet-Draft SUN Microsystems, Inc.
Expires: April 24, 2005 January 17, 2006 J. Ihren
Autonomica
P. Savola
CSC/FUNET
October 24, 2004
July 16, 2005

Operational Considerations and Issues with IPv6 DNS
draft-ietf-dnsop-ipv6-dns-issues-10.txt
draft-ietf-dnsop-ipv6-dns-issues-11.txt

Status of this Memo

This document is an Internet-Draft and is subject to all provisions
of section 3 of RFC 3667.

By submitting this Internet-Draft, each author represents that any
applicable patent or other IPR claims of which he or she is aware
have been or will be disclosed, and any of which he or she become becomes
aware will be disclosed, in accordance with
RFC 3668. Section 6 of BCP 79.

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.

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."

The list of current Internet-Drafts can be accessed at
http://www.ietf.org/ietf/1id-abstracts.txt.

The list of Internet-Draft Shadow Directories can be accessed at
http://www.ietf.org/shadow.html.

This Internet-Draft will expire on April 24, 2005. January 17, 2006.

Abstract

This memo presents operational considerations and issues with IPv6
Domain Name System (DNS), including a summary of special IPv6
addresses, documentation of known DNS implementation misbehaviour,
recommendations and considerations on how to perform DNS naming for
service provisioning and for DNS resolver IPv6 support,

Durand, et al. Expires April 24, 2005 January 17, 2006 [Page 1]

Internet-Draft Considerations and Issues with IPv6 DNS October 2004

service provisioning and for DNS resolver IPv6 support, July 2005

considerations for DNS updates for both the forward and reverse
trees, and miscellaneous issues. This memo is aimed to include a
summary of information about IPv6 DNS considerations for those who
have experience with IPv4 DNS.

Table of Contents

1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . 4
1.1 Representing IPv6 Addresses in DNS Records . . . . . . . . 4
1.2 Independence of DNS Transport and DNS Records . . . . . . 4
1.3 Avoiding IPv4/IPv6 Name Space Fragmentation . . . . . . . 5
1.4 Query Type '*' and A/AAAA Records . . . . . . . . . . . . 5
2. DNS Considerations about Special IPv6 Addresses . . . . . . . 5
2.1 Limited-scope Addresses . . . . . . . . . . . . . . . . . 6
2.2 Temporary Addresses . . . . . . . . . . . . . . . . . . . 6
2.3 6to4 Addresses . . . . . . . . . . . . . . . . . . . . . . 6
2.4 Other Transition Mechanisms . . . . . . . . . . . . . . . 6
3. Observed DNS Implementation Misbehaviour . . . . . . . . . . . 7
3.1 Misbehaviour of DNS Servers and Load-balancers . . . . . . 7
3.2 Misbehaviour of DNS Resolvers . . . . . . . . . . . . . . 7
4. Recommendations for Service Provisioning using DNS . . . . . . 8 7
4.1 Use of Service Names instead of Node Names . . . . . . . . 8
4.2 Separate vs the Same Service Names for IPv4 and IPv6 . . . 8
4.3 Adding the Records Only when Fully IPv6-enabled . . . . . 9
4.4 Behaviour The Use of Additional Data in IPv4/IPv6 Environments TTL for IPv4 and IPv6 RRs . . . . 10
4.4.1 Description of Additional Data Scenarios . . . . . . . 10
4.4.2 Which
4.4.1 TTL With Courtesy Additional Data to Keep, If Any? . . . . . . . . 11
4.4.3 Discussion of the Problems . . . . . . . . . . . . . . 12
4.5 The Use of 10
4.4.2 TTL for IPv4 and IPv6 RRs . With Critical Additional Data . . . . . . . . . . 13
4.6 10
4.5 IPv6 Transport Guidelines for DNS Servers . . . . . . . . 14 11
5. Recommendations for DNS Resolver IPv6 Support . . . . . . . . 15 11
5.1 DNS Lookups May Query IPv6 Records Prematurely . . . . . . 15 11
5.2 Obtaining a List of DNS Recursive Resolvers . . . . . . . 16 13
5.3 IPv6 Transport Guidelines for Resolvers . . . . . . . . . 17 13
6. Considerations about Forward DNS Updating . . . . . . . . . . 17 13
6.1 Manual or Custom DNS Updates . . . . . . . . . . . . . . . 17 14
6.2 Dynamic DNS . . . . . . . . . . . . . . . . . . . . . . . 18 14
7. Considerations about Reverse DNS Updating . . . . . . . . . . 19 15
7.1 Applicability of Reverse DNS . . . . . . . . . . . . . . . 19 15
7.2 Manual or Custom DNS Updates . . . . . . . . . . . . . . . 20 16
7.3 DDNS with Stateless Address Autoconfiguration . . . . . . 20 16
7.4 DDNS with DHCP . . . . . . . . . . . . . . . . . . . . . . 21 18
7.5 DDNS with Dynamic Prefix Delegation . . . . . . . . . . . 22 18
8. Miscellaneous DNS Considerations . . . . . . . . . . . . . . . 23 19
8.1 NAT-PT with DNS-ALG . . . . . . . . . . . . . . . . . . . 23 19
8.2 Renumbering Procedures and Applications' Use of DNS . . . 23 19
9. Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . 23

Durand, et al. Expires April 24, 2005 [Page 2]
Internet-Draft Considerations and Issues with IPv6 DNS October 2004 20
10. Security Considerations . . . . . . . . . . . . . . . . . . 24 20
11. References . . . . . . . . . . . . . . . . . . . . . . . . . 24 20
11.1 Normative References . . . . . . . . . . . . . . . . . . . . 24 20

Durand, et al. Expires January 17, 2006 [Page 2]

Internet-Draft Considerations with IPv6 DNS July 2005

11.2 Informative References . . . . . . . . . . . . . . . . . . . 26 22
Authors' Addresses . . . . . . . . . . . . . . . . . . . . . . 28 24
A. Site-local Unique Local Addressing Considerations for DNS . . . . . . . . 25
B. Behaviour of Additional Data in IPv4/IPv6 Environments . 29
Intellectual Property and Copyright Statements . . . 25
B.1 Description of Additional Data Scenarios . . . . . 30

Durand, et al. Expires April 24, 2005 [Page 3]
Internet-Draft Considerations and Issues with IPv6 DNS October 2004

1. Introduction

This memo presents operational considerations and issues with IPv6
DNS; it is meant to be an extensive summary and . . . . 26
B.2 Which Additional Data to Keep, If Any? . . . . . . . . . . 27
B.3 Discussion of the Potential Problems . . . . . . . . . . . 28
Intellectual Property and Copyright Statements . . . . . . . . 30

Durand, et al. Expires January 17, 2006 [Page 3]

Internet-Draft Considerations with IPv6 DNS July 2005

1. Introduction

This memo presents operational considerations and issues with IPv6
DNS; it is meant to be an extensive summary and a list of pointers
for more information about IPv6 DNS considerations for those with
experience with IPv4 DNS.

The purpose of this document is to give information about various
issues and considerations related to DNS operations with IPv6; it is
not meant to be a normative specification or standard for IPv6 DNS.

The first section gives a brief overview of how IPv6 addresses and
names are represented in the DNS, how transport protocols and
resource records (don't) relate, and what IPv4/IPv6 name space
fragmentation means and how to avoid it; all of these are described
at more length in other documents.

The second section summarizes the special IPv6 address types and how
they relate to DNS. The third section describes observed DNS
implementation misbehaviours which have a varying effect on the use
of IPv6 records with DNS. The fourth section lists recommendations
and considerations for provisioning services with DNS. The fifth
section in turn looks at recommendations and considerations about
providing IPv6 support in the resolvers. The sixth and seventh
sections describe considerations with forward and reverse DNS
updates, respectively. The eighth section introduces several
miscellaneous IPv6 issues relating to DNS for which no better place
has been found in this memo. Appendix A looks briefly at the
requirements for site-local unique local addressing.

1.1 Representing IPv6 Addresses in DNS Records

In the forward zones, IPv6 addresses are represented using AAAA
records. In the reverse zones, IPv6 address are represented using
PTR records in the nibble format under the ip6.arpa. tree. See
[RFC3596] for more about IPv6 DNS usage, and [RFC3363] or [RFC3152]
for background information.

In particular one should note that the use of A6 records in the
forward tree or Bitlabels in the reverse tree is not recommended
[RFC3363]. Using DNAME records is not recommended in the reverse
tree in conjunction with A6 records; the document did not mean to
take a stance on any other use of DNAME records [RFC3364].

1.2 Independence of DNS Transport and DNS Records

DNS has been designed to present a single, globally unique name space
[RFC2826]. This property should be maintained, as described here and

Durand, et al. Expires April 24, 2005 January 17, 2006 [Page 4]

Internet-Draft Considerations and Issues with IPv6 DNS October 2004 July 2005

in Section 1.3.

The IP version used to transport the DNS queries and responses is
independent of the records being queried: AAAA records can be queried
over IPv4, and A records over IPv6. The DNS servers must not make
any assumptions about what data to return for Answer and Authority
sections based on the underlying transport used in a query.

However, there is some debate whether the addresses in Additional
section could be selected or filtered using hints obtained from which
transport was being used; this has some obvious problems because in
many cases the transport protocol does not correlate with the
requests, and because a "bad" answer is in a way worse than no answer
at all (consider the case where the client is led to believe that a
name received in the additional record does not have any AAAA records
at all).

As stated in [RFC3596]:

The IP protocol version used for querying resource records is
independent of the protocol version of the resource records; e.g.,
IPv4 transport can be used to query IPv6 records and vice versa.

1.3 Avoiding IPv4/IPv6 Name Space Fragmentation

To avoid the DNS name space from fragmenting into parts where some
parts of DNS are only visible using IPv4 (or IPv6) transport, the
recommendation is to always keep at least one authoritative server
IPv4-enabled, and to ensure that recursive DNS servers support IPv4.
See DNS IPv6 transport guidelines [RFC3901] for more information.

1.4 Query Type '*' and A/AAAA Records

QTYPE=* is typically only used for debugging or management purposes;
it is worth keeping in mind that QTYPE=* ("ANY" queries) only return
any available RRsets, not *all* the RRsets, because the caches do not
necessarily have all the RRsets and have no way of guaranteeing that
they have all the RRsets. Therefore, to get both A and AAAA records
reliably, two separate queries must be made.

2. DNS Considerations about Special IPv6 Addresses

There are a couple of IPv6 address types which are somewhat special;
these are considered here.

Durand, et al. Expires April 24, 2005 January 17, 2006 [Page 5]

Internet-Draft Considerations and Issues with IPv6 DNS October 2004 July 2005

2.1 Limited-scope Addresses

The IPv6 addressing architecture [RFC3513] includes two kinds of
local-use addresses: link-local (fe80::/10) and site-local
(fec0::/10). The site-local addresses have been deprecated
[RFC3879], and [RFC3879]
but are only discussed with unique local addresses in Appendix A.

Link-local addresses should never be published in DNS (whether in
forward or reverse tree), because they have only local (to the
connected link) significance
[I-D.ietf-dnsop-dontpublish-unreachable]. [I-D.durand-dnsop-dont-publish].

2.2 Temporary Addresses

Temporary addresses defined in RFC3041 [RFC3041] (sometimes called
"privacy addresses") use a random number as the interface identifier.
Having DNS AAAA records that are updated to always contain the
current value of a node's temporary address would defeat the purpose
of the mechanism and is not recommended. However, it would still be
possible to return a non-identifiable name (e.g., the IPv6 address in
hexadecimal format), as described in [RFC3041].

2.3 6to4 Addresses

6to4 [RFC3056] specifies an automatic tunneling mechanism which maps
a public IPv4 address V4ADDR to an IPv6 prefix 2002:V4ADDR::/48.

If the reverse DNS population would be desirable (see Section 7.1 for
applicability), there are a number of possible ways to do so
[I-D.moore-6to4-dns], some more applicable than the others. so.

The main proposal [I-D.huston-6to4-reverse-dns] aims to design an
autonomous reverse-delegation system that anyone being capable of
communicating using a specific 6to4 address would be able to set up a
reverse delegation to the corresponding 6to4 prefix. This could be
deployed by e.g., Regional Internet Registries (RIRs). This is a
practical solution, but may have some scalability concerns.

2.4 Other Transition Mechanisms

6to4, above,

6to4 is mentioned as a case of an IPv6 transition mechanism requiring
special considerations. In general, mechanisms which include a
special prefix may need a custom solution; otherwise, for example
when IPv4 address is embedded as the suffix or not embedded at all,
special solutions are likely not needed. This is why only
6to4 and Teredo [I-D.huitema-v6ops-teredo] are described.

Note that it does not seem feasible to provide reverse DNS with

Durand, et al. Expires April 24, 2005 [Page 6]
Internet-Draft Considerations and Issues with IPv6 DNS October 2004
another automatic tunneling mechanism, Teredo; Teredo [I-D.huitema-v6ops-
teredo]; this is because the IPv6 address is based on the IPv4
address and UDP port of the current NAT mapping which is likely to be

Durand, et al. Expires January 17, 2006 [Page 6]

Internet-Draft Considerations with IPv6 DNS July 2005

relatively short-lived.

3. Observed DNS Implementation Misbehaviour

Several classes of misbehaviour in DNS servers, load-balancers and
resolvers have been observed. Most of these are rather generic, not
only applicable to IPv6 -- but in some cases, the consequences of
this misbehaviour are extremely severe in IPv6 environments and
deserve to be mentioned.

3.1 Misbehaviour of DNS Servers and Load-balancers

There are several classes of misbehaviour in certain DNS servers and
load-balancers which have been noticed and documented
[I-D.ietf-dnsop-misbehavior-against-aaaa]: [RFC4074]: some
implementations silently drop queries for unimplemented DNS records
types, or provide wrong answers to such queries (instead of a proper
negative reply). While typically these issues are not limited to
AAAA records, the problems are aggravated by the fact that AAAA
records are being queried instead of (mainly) A records.

The problems are serious because when looking up a DNS name, typical
getaddrinfo() implementations, with AF_UNSPEC hint given, first try
to query the AAAA records of the name, and after receiving a
response, query the A records. This is done in a serial fashion --
if the first query is never responded to (instead of properly
returning a negative answer), significant timeouts will occur.

In consequence, this is an enormous problem for IPv6 deployments, and
in some cases, IPv6 support in the software has even been disabled
due to these problems.

The solution is to fix or retire those misbehaving implementations,
but that is likely not going to be effective. There are some
possible ways to mitigate the problem, e.g., by performing the
lookups somewhat in parallel and reducing the timeout as long as at
least one answer has been received; but such methods remain to be
investigated; slightly more on this is included in Section 5.

3.2 Misbehaviour of DNS Resolvers

Several classes of misbehaviour have also been noticed in DNS
resolvers [I-D.ietf-dnsop-bad-dns-res]. However, these do not seem
to directly impair IPv6 use, and are only referred to for
completeness.

Durand, et al. Expires April 24, 2005 [Page 7]
Internet-Draft Considerations and Issues with IPv6 DNS October 2004

4. Recommendations for Service Provisioning using DNS

When names are added in the DNS to facilitate a service, there are

Durand, et al. Expires January 17, 2006 [Page 7]

Internet-Draft Considerations with IPv6 DNS July 2005

several general guidelines to consider to be able to do it as
smoothly as possible.

4.1 Use of Service Names instead of Node Names

It makes sense to keep logically information about separate services by a node logically
separate in the DNS, due to DNS by using a number of reasons, different DNS hostname for each
service. There are several reasons for doing this, for example:

o It allows more flexibility and ease for migration of (only a part
of) services from one node to another,

o It allows configuring different properties (e.g., TTL) for each
service, and

o It allows deciding separately for each service whether to publish
the IPv6 addresses or not (in cases if where some services are more
IPv6-ready than others).

Using SRV records [RFC2782] would avoid these problems.
Unfortunately, those are not sufficiently widely used to be
applicable in most cases. Hence an operation technique is to use
service names instead of node names (or, "hostnames"). This
operational technique is not specific to IPv6, but required to
understand the considerations described in Section 4.2 and
Section 4.3.

For example, assume a node named "pobox.example.com" provides both
SMTP and IMAP service. Instead of configuring the MX records to
point at "pobox.example.com", and configuring the mail clients to
look up the mail via IMAP from "pobox.example.com", one could use
e.g., "smtp.example.com" for SMTP (for both message submission and
mail relaying between SMTP servers) and "imap.example.com" for IMAP.
Note that in the specific case of SMTP relaying, the server itself
must typically also be configured to know all its names to ensure
loops do not occur. DNS can provide a layer of indirection between
service names and where the service actually is, and using which
addresses. (Obviously, when wanting to reach a specific node, one
should use the hostname rather than a service name.)

4.2 Separate vs the Same Service Names for IPv4 and IPv6

The service naming can be achieved in basically two ways: when a
service is named "service.example.com" for IPv4, the IPv6-enabled
service could be either be added to "service.example.com", or added
separately under a different name, e.g., in a sub-domain, like,
"service.ipv6.example.com".

Durand, et al. Expires April 24, 2005 January 17, 2006 [Page 8]

Internet-Draft Considerations and Issues with IPv6 DNS October 2004

"service.ipv6.example.com". July 2005

These two methods have different characteristics. Using a different
name allows for easier service piloting, minimizing the disturbance
to the "regular" users of IPv4 service; however, the service would
not be used transparently, without the user/application explicitly
finding it and asking for it -- which would be a disadvantage in most
cases. When the different name is under a sub-domain, if the
services are deployed within a restricted network (e.g., inside an
enterprise), it's possible to prefer them transparently, at least to
a degree, by modifying the DNS search path; however, this is a
suboptimal solution. Using the same service name is the "long-term"
solution, but may degrade performance for those clients whose IPv6
performance is lower than IPv4, or does not work as well (see
Section 4.3 for more).

In most cases, it makes sense to pilot or test a service using
separate service names, and move to the use of the same name when
confident enough that the service level will not degrade for the
users unaware of IPv6.

4.3 Adding the Records Only when Fully IPv6-enabled

The recommendation is that AAAA records for a service should not be
added to the DNS until all of following are true:

1. The address is assigned to the interface on the node.

2. The address is configured on the interface.

3. The interface is on a link which is connected to the IPv6
infrastructure.

In addition, if the AAAA record is added for the node, instead of
service as recommended, all the services of the node should be
IPv6-enabled IPv6-
enabled prior to adding the resource record.

For example, if an IPv6 node is isolated from an IPv6 perspective
(e.g., it is not connected to IPv6 Internet) constraint #3 would mean
that it should not have an address in the DNS.

Consider the case of two dual-stack nodes, which both have IPv6
enabled, but the server does not have (global) IPv6 connectivity. As
the client looks up the server's name, only A records are returned
(if the recommendations above are followed), and no IPv6
communication, which would have been unsuccessful, is even attempted.

The issues are not always so black-and-white. Usually it's important

Durand, et al. Expires April 24, 2005 [Page 9]
Internet-Draft Considerations and Issues with IPv6 DNS October 2004

if
that the service offered using both protocols is of roughly equal
quality, using the appropriate metrics for the service (e.g.,
latency, throughput, low packet

Durand, et al. Expires January 17, 2006 [Page 9]

Internet-Draft Considerations with IPv6 DNS July 2005

latency, throughput, low packet loss, general reliability, etc.) --
this is typically very important especially for interactive or
real-time real-
time services. In many cases, the quality of IPv6 connectivity may
not yet be equal to that of IPv4, at least globally -- this has to be
taken into consideration when enabling services
[I-D.savola-v6ops-6bone-mess]. services.

4.4 Behaviour The Use of TTL for IPv4 and IPv6 RRs

The behaviour of Additional Data in IPv4/IPv6 Environments DNS responses do not always fit in a single UDP packet. We'll
examine the cases which happen caching when this is due to too much data in
the Additional Section.

4.4.1 Description of Additional Data Scenarios

There different TTL values are two kinds used for
different RRsets of additional data:

1. "critical" the same name calls for explicit discussion. For
example, let's consider two unrelated zone fragments:

example.com. 300 IN MX foo.example.com.
foo.example.com. 300 IN A 192.0.2.1
foo.example.com. 100 IN AAAA 2001:db8::1

...

child.example.com. 300 IN NS ns.child.example.com.
ns.child.example.com. 300 IN A 192.0.2.1
ns.child.example.com. 100 IN AAAA 2001:db8::1

In the former case, we have "courtesy" additional data; this must be included in all
scenarios, with all the RRsets, and

2. "courtesy"
latter, we have "critical" additional data; this could be sent data. See more extensive
background discussion of additional data handling in full, with only Appendix B.

4.4.1 TTL With Courtesy Additional Data

When a few RRsets, caching resolver asks for the MX record of example.com, it
gets back "foo.example.com". It may also get back either one or with no RRsets, both
of the A and can be fetched separately as
well, but at AAAA records in the cost of additional queries. section. The responding server can algorithmically determine which type resolver
must explicitly query for both A and AAAA records [RFC2821].

After 100 seconds, the
additional data AAAA record is by checking whether it's at or below a zone cut.

Only those additional data records (even if sometimes carelessly
termed "glue") are considered "critical" or real "glue" if and only
if they meet removed from the abovementioned condition, as specified in Section
4.2.1 of [RFC1034].

Remember that resource record sets (RRsets) are never "broken up", so
if a name has 4 A records and 5 AAAA records, you can either return
all 9, all 4 A records, all 5 AAAA records or nothing. In
particular, notice that cache(s)
because its TTL expired. It could be argued to be useful for the "critical" additional data getting
all
caching resolvers to discard the RRsets can be critical.

In particular, [RFC2181] specifies A record when the shorter TTL (in Section 9) that:

a. if all
this case, for the "critical" RRsets do not fit, AAAA record) expires; this would avoid the sender should set
situation where there would be a window of 200 seconds when
incomplete information is returned from the TC bit, and cache. Further argument
for discarding is that in the recipient should discard normal operation, the whole response
and retry using mechanism allowing larger responses such as TCP.

b. "courtesy" TTL values are so
high that very likely the incurred additional data should queries would not cause the setting of TC
bit, but instead all be
noticeable, compared to the non-fitting obtained performance optimization. The
behaviour in this scenario is unspecified.

4.4.2 TTL With Critical Additional Data

The difference to courtesy additional data RRsets is that the A/AAAA records
served by the parent zone cannot be queried explicitly. Therefore

Durand, et al. Expires April 24, 2005 January 17, 2006 [Page 10]

Internet-Draft Considerations and Issues with IPv6 DNS October 2004

should be removed.

An example of July 2005

after 100 seconds the "courtesy" additional data is A/AAAA records in
conjunction of MX records is shown in Section 4.5; an example of the
"critical" additional data is shown below (where getting both the A
and AAAA RRsets is critical):

child.example.com. IN NS ns.child.example.com.
ns.child.example.com. IN A 192.0.2.1
ns.child.example.com. IN AAAA 2001:db8::1

When there record is too much courtesy additional data, at least the
non-fitting RRsets should be removed [RFC2181]; however, as the
additional data is not critical, even all of it could be safely
removed.

When there is too much critical additional data, TC bit will have to
be set, and from the recipient should ignore cache(s), but
the response and retry using
TCP; if some data were to be left in A record remains. Queries for the UDP response, remaining 200 seconds
(provided that there are no further queries from the issue is parent which data
could be retained.

Failing to discard refresh the response with TC bit set leads to a protocol
problem. Omitting caches) only some of return the RRsets if all would not fit leads A record, leading to a performance problem. These are discussed
potential opererational situation with unreachable servers.

Similar cache flushing strategies apply in Section 4.4.3.

4.4.2 Which Additional Data to Keep, If Any?

If this scenario; the implementation decides to keep as much data (whether
"critical" or "courtesy") as possible record.

4.5 IPv6 Transport Guidelines for DNS Servers

As described in the UDP responses, it might
be tempting Section 1.3 and [RFC3901], there should continue to use the transport of the
be at least one authoritative IPv4 DNS query as a hint in either
of these cases: return the AAAA records if the query was done over
IPv6, or return the A records server for every zone, even if
the query was done over IPv4.
However, zone has only IPv6 records. (Note that obviously, having more
servers with robust connectivity would be preferable, but this breaks is the model of independence of
minimum recommendation; also see [RFC2182].)

5. Recommendations for DNS transport and
resource records, as noted in Section 1.2.

With courtesy additional data, as long as enough RRsets will be
removed so that TC will not be set, it Resolver IPv6 Support

When IPv6 is allowed enabled on a node, there are several things to send consider
to ensure that the process is as many
complete RRsets smooth as possible.

5.1 DNS Lookups May Query IPv6 Records Prematurely

The system library that implements the implementations prefers. However, the
implementations are also free to omit all such RRsets, even if
complete. Removing all getaddrinfo() function for
looking up names is a critical piece when considering the RRsets if some would robustness
of enabling IPv6; it may come in basically three flavours:

1. The system library does not fit obviates know whether IPv6 has been enabled in
the kernel of the operating system: it may start looking up AAAA
records with getaddrinfo() and AF_UNSPEC hint when the system is
upgraded to a
performance problem, system library version which would require the users supports IPv6.

2. The system library might start to issue second perform IPv6 queries to obtain consistent information.

With critical additional data, the alternatives are either returning
nothing (and absolutely requiring a retry with TCP) or returning
something (working also
getaddrinfo() only when IPv6 has been enabled in the case if the recipient kernel.
However, this does not discard guarantee that there exists any useful
IPv6 connectivity (e.g., the response and retry using TCP) in addition node could be isolated from the
other IPv6 networks, only having link-local addresses).

3. The system library might implement a toggle which would apply
some heuristics to setting the TC bit.
If "IPv6-readiness" of the process node before
starting to perform queries; for selecting "something" from example, it could check whether
only link-local IPv6 address(es) exists, or if at least one
global IPv6 address exists.

First, let us consider generic implications of unnecessary queries
for AAAA records: when looking up all the critical data would records in the DNS, AAAA
records are typically tried first, and then A records. These are
done in serial, and the A query is not performed until a response is

Durand, et al. Expires April 24, 2005 January 17, 2006 [Page 11]

Internet-Draft Considerations and Issues with IPv6 DNS October 2004

otherwise be practically "flipping July 2005

received to the coin" between A and AAAA
records, it could be argued that if one looked at query. Considering the transport misbehaviour of DNS
servers and load-balancers, as described in Section 3.1, the query, it would have look-up
delay for AAAA may incur additional unnecessary latency, and
introduce a larger possibility component of being right than
just 50/50. In other words, unreliability.

One option here could be to do the queries partially in parallel; for
example, if the returned critical additional data
would have final response to be selected somehow, using something more sophisticated
than a random process would seem justifiable.

That is, leaving in some intelligently selected critical additional
data the AAAA query is a tradeoff between creating an optimization not received in
0.5 seconds, start performing the A query while waiting for those
resolvers which ignore the "should discard" recommendation, and a
causing a protocol problem by propagating inconsistent
result (immediate parallelism might be unoptimal, at least without
information
about "critical" records in the caches.

Similarly, leaving in sharing between the complete courtesy look-up threads, as that would
probably lead to duplicate non-cached delegation chain lookups).

An additional data RRsets
instead of removing all the RRsets concern is a performance tradeoff as
described in the next section.

4.4.3 Discussion of the Problems

As noted above, address selection, which may, in some
circumstances, prefer AAAA records over A records even when the temptation for omitting only node
does not have any IPv6 connectivity [I-D.ietf-v6ops-v6onbydefault].
In some of cases, the
additional data based implementation may attempt to connect or send a
datagram on a physical link [I-D.ietf-v6ops-onlinkassumption],
incurring very long protocol timeouts, instead of quickly failing
back to IPv4.

Now, we can consider the transport issues specific to each of the query could be
problematic. three
possibilities:

In particular, there appears to be little justification
for doing so in the case of "courtesy" data.

For courtesy additional data, this causes first case, the node performs a performance problem number of completely useless
DNS lookups as
this requires that it will not be able to use the clients issue re-query for returned AAAA records
anyway. (The only exception is where the potentially
omitted RRsets. For critical additional data, this causes a
potential protocol problem if application desires to know
what's in the response is DNS, but not discarded and use the
query not re-tried with TCP, result for communication.) One
should be able to disable these unnecessary queries, for both latency
and reliability reasons. However, as IPv6 has not been enabled, the nameservers might be reachable
only through the omitted RRsets.

If an implementation would look at the transport used for
connections to IPv6 addresses fail immediately, and if the query,
it
application is worth remembering that often the host using programmed properly, the records application can fall
gracefully back to IPv4 [RFC4038].

The second case is
different from the node requesting them from the authoritative DNS
server (or even a caching resolver). So, whichever version the
requestor (e.g., a recursive server in the middle) uses makes no
difference similar to the ultimate user first, except it happens to a
smaller set of nodes when IPv6 has been enabled but connectivity has
not been provided yet; similar considerations apply, with the
exception that IPv6 records, whose transport
capabilities might differ from those of the requestor. This might
result in e.g., inappropriately returning A records when returned, will be actually tried
first which may typically lead to an IPv6-only
node, going through a translation, or opening up another IP-level
session (e.g., long timeouts.

The third case is a PDP context [I-D.ietf-v6ops-3gpp-analysis]).
Therefore, at least in many scenarios, it would be very useful if bit more complex: optimizing away the
information returned would DNS lookups
with only link-locals is probably safe (but may be consistent and complete -- or if that desirable with
different lookup services which getaddrinfo() may support), as the
link-locals are typically automatically generated when IPv6 is
enabled, and do not feasible, return no misleading information but rather leave it
to the client to query again.

The problem indicate any form of too much additional data seems to be an operational
one: the zone administrator entering too many records which will IPv6 connectivity. That is,
performing DNS lookups only when a non-link-local address has been
configured on any interface could be beneficial -- this would be an
indication that either the address has been configured either from a
router advertisement, DHCPv6 [RFC3315], or manually. Each would

Durand, et al. Expires April 24, 2005 January 17, 2006 [Page 12]

Internet-Draft Considerations and Issues with IPv6 DNS October 2004

returned either truncated (or missing some RRsets, depending on
implementations) to the users. A protocol fix for this is using
EDNS0 [RFC2671] to signal the capacity for larger UDP packet sizes,
pushing up the relevant threshold. Further, DNS server
implementations should rather omit courtesy additional data
completely rather than including only July 2005

indicate at least some RRsets [RFC2181]. An
operational fix for this is having the DNS server implementations
return a warning when the administrators create zones which would
result in too much additional data being returned. Further, DNS
server implementations should warn form of or disallow such zone
configurations which are recursive or otherwise difficult to manage
by the protocol.

Additionally, to avoid the case where an application IPv6 connectivity, even though there
would not get an
address at all due to some be guarantees of "courtesy" additional data being
omitted, the resolvers it.

These issues should be able to query the specific records
of the desired protocol, not just rely on getting all analyzed at more depth, and the required
RRsets fixes found
consensus on, perhaps in the additional section.

4.5 The Use a separate document.

5.2 Obtaining a List of TTL for IPv4 and IPv6 RRs DNS Recursive Resolvers

In the previous section, we discussed scenarios where DHCPv6 is available, a danger with queries,
potentially leading to omitting RRsets from the additional section;
this could happen to both critical and "courtesy" additional data
(however, both host can discover a list of these are recommended against in [RFC2181]).
DNS recursive resolvers through DHCPv6 "DNS Recursive Name Server"
option [RFC3646]. This
section discusses another problem with courtesy additional data,
leading option can be passed to omitting RRsets in cached data, highlighted in the
IPv4/IPv6 environment.

The behaviour of DNS caching when different TTL values are used for
different RRsets a host through a
subset of DHCPv6 [RFC3736].

The IETF is considering the same name requires explicit discussion. For
example, let's consider a part development of a zone:

example.com. 300 IN MX foo.example.com.
foo.example.com. 300 IN A 192.0.2.1
foo.example.com. 100 IN AAAA 2001:db8::1

When a caching resolver asks alternative mechanisms for
obtaining the MX record list of example.com, it
gets back "foo.example.com". It may also get back either one DNS recursive name servers when DHCPv6 is
unavailable or both
of the A and AAAA records in the additional section. So, there are
three cases inappropriate. No decision about returning records for the MX in the additional
section:

1. We get back no A or AAAA RRsets: taking on this
development work has been reached as of this writing (Aug 2004)
[I-D.ietf-dnsop-ipv6-dns-configuration].

In scenarios where DHCPv6 is unavailable or inappropriate, mechanisms
under consideration for development include the simplest case,
because then we have use of well-known
addresses [I-D.ohta-preconfigured-dns] and the use of Router
Advertisements to query which information is required
explicitly, guaranteeing that we get all convey the information we're
interested in.

Durand, et al. Expires April 24, 2005 [Page 13]
Internet-Draft Considerations and Issues with [I-D.jeong-dnsop-ipv6-dns-
discovery].

Note that even though IPv6 DNS October 2004

2. We get back all the RRsets: this is an optimization as there resolver discovery is
no need to perform more queries, causing lower latency. However, a recommended
procedure, it is impossible to guarantee that in fact we would always get
back all the records (the only way to ensure that is to send a
AAAA query not required for the name after getting the cached reply with A
records or vice versa).

3. We only get back A or AAAA RRsets even if both existed: this is
indistinguishable from the previous case, and may have
performance problems at least dual-stack nodes in certain environments dual-stack
networks as
described IPv6 DNS records can be queried over IPv4 as well as
IPv6. Obviously, nodes which are meant to function without manual
configuration in IPv6-only networks must implement the previous section. DNS resolver
discovery function.

5.3 IPv6 Transport Guidelines for Resolvers

As the third case was considered described in the previous section, we assume
we get back both A Section 1.3 and AAAA records of foo.example.com, or [RFC3901], the recursive resolvers
should be IPv4-only or dual-stack to be able to reach any IPv4-only
DNS server. Note that this requirement is also fulfilled by an IPv6-
only stub resolver explicitly asks, in two separate queries, both A and AAAA
records.

After 100 seconds, pointing to a dual-stack recursive DNS resolver.

6. Considerations about Forward DNS Updating

While the AAAA record is removed topic of how to enable updating the forward DNS, i.e., the
mapping from names to the cache(s)
because its TTL expired. It could be argued correct new addresses, is not specific to
IPv6, it should be useful for the
caching resolvers considered especially due to discard the A record when the shorter TTL (in
this case, for the AAAA record) expires; this would avoid the
situation where there would be a window advent of 200 seconds when
incomplete information is returned from the cache. Further argument
for discarding is that
Stateless Address Autoconfiguration [RFC2462].

Typically forward DNS updates are more manageable than doing them in
the normal operation, the TTL values are so
high that very likely reverse DNS, because the incurred additional queries would not updater can often be
noticeable, compared assumed to "own" a

Durand, et al. Expires January 17, 2006 [Page 13]

Internet-Draft Considerations with IPv6 DNS July 2005

certain DNS name -- and we can create a form of security relationship
with the obtained performance optimization. The
behaviour in this scenario is unspecified.

To simplify DNS name and the situation, node which is allowed to update it might help to use the same TTL for all
the resource record sets referring point
to the same name, unless there is a particular reason for not doing so. However, there are some
scenarios (e.g., when renumbering IPv6 but keeping IPv4 intact) where new address.

A more complex form of DNS updates -- adding a different strategy is preferable.

Thus, applications that use the response should not rely on whole new name into a
particular TTL configuration. For example, even if
DNS zone, instead of updating an application
gets existing name -- is considered out
of scope for this memo as it could require zone-wide authentication.
Adding a response that only has the A record new name in the example described
above, it should be forward zone is a problem which is still aware
being explored with IPv4, and IPv6 does not seem to add much new in
that there could be area.

6.1 Manual or Custom DNS Updates

The DNS mappings can also be maintained by hand, in a AAAA record semi-automatic
fashion or by running non-standardized protocols. These are not
considered at more length in this memo.

6.2 Dynamic DNS

Dynamic DNS updates (DDNS) [RFC2136] [RFC3007] is a standardized
mechanism for
"foo.example.com". That is, dynamically updating the application should try DNS. It works equally well
with stateless address autoconfiguration (SLAAC), DHCPv6 or manual
address configuration. It is important to fetch the
missing records itself consider how each of these
behave if it needs the record.

4.6 IPv6 Transport Guidelines for DNS Servers

As described IP address-based authentication, instead of stronger
mechanisms [RFC3007], was used in Section 1.3 the updates.

1. manual addresses are static and [RFC3901], there should continue to can be at least one authoritative IPv4 configured

2. DHCPv6 addresses could be reasonably static or dynamic, depending
on the deployment, and could or could not be configured on the
DNS server for every zone, even if the zone has only IPv6 records. (Note that obviously, having more
servers with robust connectivity would be preferable, long term

3. SLAAC addresses are typically stable for a long time, but this is the
minimum recommendation; also see [RFC2182].)

Durand, et al. Expires April 24, 2005 [Page 14]
Internet-Draft Considerations could
require work to be configured and Issues with IPv6 DNS October 2004

5. Recommendations maintained.

As relying on IP addresses for Dynamic DNS Resolver IPv6 Support

When IPv6 is enabled on a node, there are several things to consider
to ensure rather insecure at
best, stronger authentication should always be used; however, this
requires that the process is as smooth as possible.

5.1 DNS Lookups May Query IPv6 Records Prematurely

The system library authorization keying will be explicitly configured
using unspecified operational methods.

Note that implements the getaddrinfo() function for
looking up names is a critical piece when considering the robustness
of enabling IPv6; with DHCP it may come in basically three flavours:

1. The system library does not know whether IPv6 has been enabled in is also possible that the kernel of DHCP server updates
the operating system: it may start looking up AAAA
records with getaddrinfo() and AF_UNSPEC hint when DNS, not the system is
upgraded to a system library version which supports IPv6.

2. host. The system library host might start to perform IPv6 queries with
getaddrinfo() only when IPv6 has been enabled indicate in the kernel.
However, this does not guarantee that there exists any useful
IPv6 connectivity (e.g., DHCP
exchange which hostname it would prefer, and the node could be isolated from DHCP server would
make the
other IPv6 networks, only having link-local addresses).

3. The system library might implement appropriate updates. Nonetheless, while this makes setting
up a toggle which would apply
some heuristics to secure channel between the "IPv6-readiness" of updater and the node before
starting to perform queries; for example, DNS server easier, it could check
does not help much with "content" security, i.e., whether
only link-local IPv6 address(es) exists, or the
hostname was acceptable -- if at least one
global IPv6 address exists.

First, let us consider generic implications of unnecessary queries
for AAAA records: when looking up all the records DNS server does not include
policies, they must be included in the DNS, AAAA
records are typically tried first, and then A records. These are
done in serial, and the A query is not performed until DHCP server (e.g., a response is
received to the AAAA query. Considering the misbehaviour of regular

Durand, et al. Expires January 17, 2006 [Page 14]

Internet-Draft Considerations with IPv6 DNS
servers and load-balancers, as July 2005

host should not be able to state that its name is "www.example.com").
DHCP-initiated DDNS updates have been extensively described in Section 3.1,
[I-D.ietf-dhc-ddns-resolution], [I-D.ietf-dhc-fqdn-option] and
[I-D.ietf-dnsext-dhcid-rr].

The nodes must somehow be configured with the look-up
delay information about the
servers where they will attempt to update their addresses, sufficient
security material for AAAA may incur additional unnecessary latency, authenticating themselves to the server, and
introduce a component of unreliability.

One option here
the hostname they will be updating. Unless otherwise configured, the
first could be to do obtained by looking up the queries partially in parallel; authoritative name servers
for
example, if the final response hostname; the second must be configured explicitly unless one
chooses to trust the AAAA query is not received in
0.5 seconds, start performing IP address-based authentication (not a good
idea); and lastly, the A query while waiting for nodename is typically pre-configured somehow
on the
result (immediate parallelism might be unoptimal, node, e.g., at least without
information sharing between install time.

Care should be observed when updating the look-up threads, as that would
probably lead addresses not to duplicate non-cached delegation chain lookups).

An additional concern is use longer
TTLs for addresses than are preferred lifetimes for the address selection, which may, in some
circumstances, prefer AAAA records over A records even when addresses, so
that if the node
does not have any IPv6 connectivity [I-D.ietf-v6ops-v6onbydefault].
In some cases, the implementation may attempt to connect or send a

Durand, et al. Expires April 24, 2005 [Page 15]
Internet-Draft Considerations and Issues with IPv6 DNS October 2004

datagram on is renumbered in a physical link [I-D.ietf-v6ops-onlinkassumption],
incurring very long protocol timeouts, instead managed fashion, the amount of quickly failing
back
stale DNS information is kept to IPv4.

Now, we can consider the issues specific to each minimum. That is, if the
preferred lifetime of an address expires, the three
possibilities:

In TTL of the first case, record needs
be modified unless it was already done before the node performs expiration. For
better flexibility, the DNS TTL should be much shorter (e.g., a number half
or a third) than the lifetime of completely useless an address; that way, the node can
start lowering the DNS lookups as TTL if it will seems like the address has not be able to use been
renewed/refreshed in a while. Some discussion on how an
administrator could manage the returned AAAA records
anyway. (The only exception DNS TTL is where the application desires to know
what's included in the DNS, but not use the result for communication.) One
should [I-D.ietf-
v6ops-renumbering-procedure]; this could be able applied to disable these unnecessary queries, for both latency
and reliability reasons. However, (smart) hosts
as IPv6 has not been enabled, well.

7. Considerations about Reverse DNS Updating

Updating the
connections reverse DNS zone may be difficult because of the split
authority over an address. However, first we have to IPv6 addresses fail immediately, and if consider the
application is programmed properly,
applicability of reverse DNS in the application can fall
gracefully back to IPv4 [I-D.ietf-v6ops-application-transition].

The second case is similar to the first, except it happens to a
smaller set of nodes when IPv6 has been enabled but connectivity has
not been provided yet; similar considerations apply, with the
exception that IPv6 records, when returned, will be actually tried first which may typically lead place.

7.1 Applicability of Reverse DNS

Today, some applications use reverse DNS to long timeouts.

The third case is a bit more complex: optimizing away either look up some hints
about the DNS lookups
with only link-locals is probably safe (but may be desirable topological information associated with
different lookup services which getaddrinfo() may support), an address (e.g.
resolving web server access logs), or as the
link-locals are typically automatically generated when IPv6 is
enabled, and do not indicate any a weak form of IPv6 connectivity. That is,
performing DNS lookups only when a non-link-local address security
check, to get a feel whether the user's network administrator has been
configured on any interface could be beneficial -- this would be an
indication that either
"authorized" the use of the address has been configured either from (on the premises that adding a
router advertisement, DHCPv6 [RFC3315], or manually. Each
reverse record for an address would
indicate at least signal some form of IPv6 connectivity, even though there
would not be guarantees of it.

These issues should be analyzed at
authorization).

One additional, maybe slightly more depth, useful usage is ensuring that the
reverse and forward DNS contents match (by looking up the fixes found
consensus on, perhaps in pointer to
the name by the IP address from the reverse tree, and ensuring that a separate document.

5.2 Obtaining a List of DNS Recursive Resolvers

In scenarios where DHCPv6 is available, a host can discover a list of
DNS recursive resolvers through DHCPv6 "DNS Recursive Name Server"
option [RFC3646]. This option can be passed to a host through a
subset of DHCPv6 [RFC3736].

The IETF is considering the development of alternative mechanisms for
obtaining the list of DNS recursive name servers when DHCPv6 is
unavailable or inappropriate. No decision about taking on this

Durand, et al. Expires April 24, 2005 January 17, 2006 [Page 16] 15]

Internet-Draft Considerations and Issues with IPv6 DNS October 2004

development work has been reached as of this writing (Aug 2004)
[I-D.ietf-dnsop-ipv6-dns-configuration].

In scenarios where DHCPv6 is unavailable or inappropriate, mechanisms July 2005

record under consideration for development include the use of well-known
addresses [I-D.ohta-preconfigured-dns] and name in the use of Router
Advertisements forward tree points to convey the information
[I-D.jeong-dnsop-ipv6-dns-discovery].

Note that even though IPv6 DNS resolver discovery is IP address)
and correspond to a recommended
procedure, configured name or domain. As a security check,
it is not required for dual-stack nodes in dual-stack
networks typically accompanied by other mechanisms, such as IPv6 a user/
password login; the main purpose of the reverse+forward DNS records can be queried over IPv4 as well as
IPv6. Obviously, nodes which are meant check is
to function without manual
configuration in IPv6-only networks must implement weed out the DNS resolver
discovery function.

5.3 IPv6 Transport Guidelines for Resolvers

As described in Section 1.3 majority of unauthorized users, and [RFC3901], if someone
managed to bypass the recursive resolvers
should be IPv4-only or dual-stack checks, he would still need to authenticate
"properly".

It may also be able desirable to reach any IPv4-only
DNS server. Note that this requirement is also fulfilled by an
IPv6-only stub resolver pointing to a dual-stack recursive DNS
resolver.

6. Considerations about Forward DNS Updating

While the topic how store IPsec keying material corresponding
to enable updating an IP address in the forward reverse DNS, i.e., the
mapping from names to the correct new addresses, as justified and described in
[RFC4025].

It is not specific to
IPv6, clear whether it should be considered especially due makes sense to the advent of
Stateless Address Autoconfiguration [RFC2462].

Typically forward DNS updates are more manageable than doing them in
the require or recommend that
reverse DNS, because the updater can often DNS records be assumed updated. In many cases, it would just make
more sense to "own" a
certain DNS name -- and we can create a form of use proper mechanisms for security relationship
with (or topological
information lookup) in the DNS name and first place. At minimum, the node applications
which is allowed to update use it to point
to a new address.

A more complex form of DNS updates -- adding a whole new name into a
DNS zone, instead of updating an existing name -- is considered out
of scope for this memo as it could require zone-wide authentication.
Adding a new name in generic authorization (in the forward zone is sense that a problem which is still
being explored with IPv4, and IPv6 does not seem record
exists at all) should be modified as soon as possible to add much new avoid such
lookups completely.

The applicability is discussed at more length in
that area.

6.1 [I-D.ietf-dnsop-
inaddr-required].

7.2 Manual or Custom DNS Updates

The

Reverse DNS mappings can also of course be maintained by hand, in a semi-automatic
fashion or by running non-standardized protocols. updated using manual or custom methods.
These are not

Durand, et al. Expires April 24, 2005 [Page 17]
Internet-Draft Considerations and Issues with IPv6 DNS October 2004

considered at more length in this memo.

6.2 Dynamic DNS

Dynamic DNS updates (DDNS) [RFC2136][RFC3007] is a standardized
mechanism further described here, except for dynamically updating the DNS. It works equally well
with stateless address autoconfiguration (SLAAC), DHCPv6 or manual
address configuration. It is important one special case.

One way to consider how each of these
behave if IP address-based authentication, instead of stronger
mechanisms [RFC3007], was used in the updates.

1. manual addresses are static and can be configured

2. DHCPv6 addresses could deploy reverse DNS would be reasonably static to use wildcard records, for
example, by configuring one name for a subnet (/64) or dynamic, depending
on a site (/48).
As a concrete example, a site (or the deployment, and could or site's ISP) could not be configured on configure the
DNS server for
reverses of the long term

3. SLAAC addresses are typically stable for prefix 2001:db8:f00::/48 to point to one name using a long time, but
wildcard record like "*.0.0.f.0.8.b.d.0.1.0.0.2.ip6.arpa. IN PTR
site.example.com." Naturally, such a name could
require work to not be configured and maintained.

As relying on IP addresses for Dynamic DNS is rather insecure verified from
the forward DNS, but would at
best, stronger authentication should always be used; however, this
requires least provide some form of "topological
information" or "weak authorization" if that the authorization keying will is really considered to
be explicitly configured
using unspecified operational methods. useful. Note that with DHCP it this is also possible that the DHCP server updates
the DNS, not the host. The host might only indicate in the DHCP
exchange which hostname it would prefer, and the DHCP server would
make the appropriate updates. Nonetheless, while this makes setting
up a secure channel between the updater and actually updating the DNS server easier, it
does not help much with "content" security, i.e., whether the
hostname was acceptable -- if as such,
as the whole point is to avoid DNS server does not include
policies, they must be included in the DHCP server (e.g., updates completely by manually
configuring a regular
host should not be able to state that its name is "www.example.com").
DHCP-initiated generic name.

7.3 DDNS with Stateless Address Autoconfiguration

Dynamic reverse DNS with SLAAC is simpler than forward DNS updates have been extensively described in
[I-D.ietf-dhc-ddns-resolution], [I-D.ietf-dhc-fqdn-option] and
[I-D.ietf-dnsext-dhcid-rr].
some regard, while being more difficult in another, as described
below.

The nodes must somehow be configured with the information about address space administrator decides whether the
servers where they will attempt hosts are trusted
to update their addresses, sufficient
security material for authenticating themselves to the server, and
the hostname reverse DNS records or not. If they will be updating. Unless otherwise configured, the
first could be obtained by looking up the authoritative name servers
for the hostname; the second must be configured explicitly unless one
chooses to trust the IP address-based authentication (not a good
idea); are trusted and lastly, the nodename is typically pre-configured somehow
on the node, e.g., at install time.

Durand, et al. Expires April 24, 2005 January 17, 2006 [Page 18] 16]

Internet-Draft Considerations and Issues with IPv6 DNS October 2004

Care should be observed when updating July 2005

deployed at the addresses same site (e.g., not to use longer
TTLs for addresses than are preferred lifetimes for the addresses, so
that if across the node is renumbered in Internet), a managed fashion, simple
address-based authorization is typically sufficient (i.e., check that
the amount of
stale DNS information update is kept to the minimum. That is, if done from the
preferred lifetime of an same IP address expires, the TTL of as the record needs
be modified unless it was already done before the expiration. For
better flexibility, the DNS TTL should being
updated); stronger security can also be much shorter (e.g., a half
or a third) than the lifetime of an address; that way, used [RFC3007]. If they
aren't allowed to update the node reverses, no update can
start lowering the DNS TTL if it seems like the address occur. However,
such address-based update authorization operationally requires that
ingress filtering [RFC3704] has not been
renewed/refreshed in a while. Some discussion on how an
administrator could manage the DNS TTL is included in
[I-D.ietf-v6ops-renumbering-procedure]; this could be applied to
(smart) hosts as well.

7. Considerations about Reverse DNS Updating

Updating set up at the reverse DNS zone may be difficult because border of the split
authority over an address. However, first we have site
where the updates occur, and as close to consider the
applicability of updater as possible.

Address-based authorization is simpler with reverse DNS in (as there is
a connection between the first place.

7.1 Applicability record and the address) than with forward
DNS. However, when a stronger form of Reverse DNS

Today, some applications use reverse security is used, forward DNS
updates are simpler to either look up some hints
about manage because the topological information associated with an address (e.g.
resolving web server access logs), or as a weak form of a security
check, host can be assumed to get a feel whether the user's network administrator has
"authorized" have
an association with the use of domain. Note that the address (on user may roam to
different networks, and does not necessarily have any association
with the premises owner of that adding a
reverse record for an address would signal some space -- so, assuming stronger form of
authorization).

One additional, maybe slightly more useful usage
authorization for reverse DNS updates than an address association is ensuring that
generally infeasible.

Moreover, the reverse and forward DNS contents match (by looking zones must be cleaned up the pointer to
the name by an unspecified
janitorial process: the IP address from the reverse tree, and ensuring node does not typically know a priori that it
will be disconnected, and cannot send a
record under the name in the forward tree points to DNS update using the IP address)
and correspond correct
source address to remove a configured name or domain. As a security check, record.

A problem with defining the clean-up process is that it is typically accompanied by other mechanisms, such as difficult
to ensure that a
user/password login; specific IP address and the main purpose of corresponding record are
no longer being used. Considering the reverse+forward DNS
check is to weed out huge address space, and the majority
unlikelihood of collision within 64 bits of unauthorized users, and if
someone managed to bypass the checks, he interface
identifiers, a process which would still need to
authenticate "properly".

It may also be desirable to store IPsec keying material corresponding
to an IP address remove the record after no traffic
has been seen from a node in a long period of time (e.g., a month or
year) might be one possible approach.

To insert or update the record, the node must discover the DNS server
to send the reverse DNS, update to somehow, similar to as justified and described discussed in
[I-D.ietf-ipseckey-rr].

It is not clear whether it makes sense
Section 6.2. One way to automate this is looking up the DNS server
authoritative (e.g., through SOA record) for the IP address being
updated, but the security material (unless the IP address-based
authorization is trusted) must also be established by some other
means.

One should note that Cryptographically Generated Addresses [RFC3972]
(CGAs) may require a slightly different kind of treatment. CGAs are
addresses where the interface identifier is calculated from a public
key, a modifier (used as a nonce), the subnet prefix, and other data.
Depending on the usage profile, CGAs might or recommend that
reverse DNS records might not be updated. In many cases, it would just make
more sense changed
periodically due to use proper mechanisms for security (or topological e.g., privacy reasons. As the CGA address is not
predicatable, a reverse record can only reasonably be inserted in the
DNS by the node which generates the address.

Durand, et al. Expires April 24, 2005 January 17, 2006 [Page 19] 17]

Internet-Draft Considerations and Issues with IPv6 DNS October 2004

information lookup) in the first place. At minimum, July 2005

7.4 DDNS with DHCP

With DHCPv4, the applications
which use it as a generic authorization (in reverse DNS name is typically already inserted to
the sense DNS that a record
exists at all) should be modified as soon as possible reflects to avoid such
lookups completely.

The applicability is discussed at more length in
[I-D.ietf-dnsop-inaddr-required].

7.2 Manual or Custom DNS Updates

Reverse DNS the name (e.g., "dhcp-67.example.com"). One
can of course assume similar practice may become commonplace with DHCPv6 as
well; all such mappings would be updated using manual or custom methods.
These are not further described here, pre-configured, and would require no
updating.

If a more explicit control is required, similar considerations as
with SLAAC apply, except for the fact that typically one special case.

One way to deploy must update
a reverse DNS would be to use wildcard records, for
example, by configuring one name for a subnet (/64) or a site (/48).
As a concrete example, a site (or the site's ISP) could configure the
reverses record instead of the prefix 2001:db8:f00::/48 to point to inserting one name using (if an address
assignment policy that reassigns disused addresses is adopted) and
updating a
wildcard record seems like "*.0.0.f.0.8.b.d.0.1.0.0.2.ip6.arpa. IN PTR
site.example.com." Naturally, such a name could not be verified from
the forward DNS, but would at least provide some form of "topological
information" or "weak authorization" if that slightly more difficult thing to
secure. However, it is really considered yet uncertain how DHCPv6 is going to be useful. used
for address assignment.

Note that this is not actually updating when using DHCP, either the host or the DHCP server could
perform the DNS as such,
as updates; see the whole point is implications in Section 6.2.

If disused addresses were to avoid DNS updates completely by manually
configuring a generic name.

7.3 be reassigned, host-based DDNS with Stateless Address Autoconfiguration

Dynamic reverse DNS with SLAAC is simpler than forward DNS
updates in
some regard, while being more difficult in another, would need policy considerations for DNS record modification,
as described
below.

The address space administrator decides whether noted above. On the hosts are trusted other hand, if disused address were not to update their reverse be
assigned, host-based DNS records or not. If they are trusted and
deployed at the same site (e.g., not across reverse updates would have similar
considerations as SLAAC in Section 7.3. Server-based updates have
similar properties except that the Internet), janitorial process could be
integrated with DHCP address assignment.

7.5 DDNS with Dynamic Prefix Delegation

In cases where a simple
address-based authorization prefix, instead of an address, is typically sufficient (i.e., check that being used and
updated, one should consider what is the location of the server where
DDNS updates are made. That is, where the DNS update server is done from located:

1. At the same IP address organization as the record being
updated); stronger security can also be used [RFC3007]. If they
aren't allowed to update the reverses, no update can occur. However,
such address-based update authorization operationally requires that
ingress filtering [RFC3704] has been set up at the border of prefix delegator.

2. At the site where the updates occur, and as close to prefixes are delegated to. In this case,
the authority of the updater as possible.

Address-based authorization is simpler with reverse DNS (as there reverse zone corresponding to the
delegated prefix is also delegated to the site.

3. Elsewhere; this implies a connection relationship between the record site and the address) than with forward
DNS. However, when a stronger form of security is used, forward where
DNS
updates are simpler to manage because the host can server is located, and such a relationship should be assumed rather
straightforward to have
an association with secure as well. Like in the domain. Note that previous case,
the user may roam to
different networks, authority of the DNS reverse zone is also delegated.

In the first case, managing the reverse DNS (delegation) is simpler
as the DNS server and does not necessarily have any association the prefix delegator are in the same
administrative domain (as there is no need to delegate anything at
all); alternatively, the prefix delegator might forgo DDNS reverse

Durand, et al. Expires April 24, 2005 January 17, 2006 [Page 20] 18]

Internet-Draft Considerations and Issues with IPv6 DNS October 2004

with July 2005

capability altogether, and use e.g., wildcard records (as described
in Section 7.2). In the owner of that address space -- so, assuming stronger form of
authorization other cases, it can be slighly more
difficult, particularly as the site will have to configure the DNS
server to be authoritative for the delegated reverse zone, implying
automatic configuration of the DNS updates than an address association is
generally unfeasible.

Moreover, server -- as the reverse zones must prefix may be cleaned up by an unspecified
janitorial process:
dynamic.

Managing the node does not DDNS reverse updates is typically know a priori that it
will be disconnected, and cannot send a DNS update using simple in the correct
source address to remove a record.

A problem with defining second
case, as the clean-up process is that it updated server is difficult
to ensure that a specific IP address and the corresponding record are
no longer being used. Considering located at the huge address space, local site, and
arguably IP address-based authentication could be sufficient (or if
not, setting up security relationships would be simpler). As there
is an explicit (security) relationship between the
unlikelihood of collision within 64 bits of parties in the interface
identifiers,
third case, setting up the security relationships to allow reverse
DDNS updates should be rather straightforward as well (but IP
address-based authentication might not be acceptable). In the first
case, however, setting up and managing such relationships might be a process
lot more difficult.

8. Miscellaneous DNS Considerations

This section describes miscellaneous considerations about DNS which
seem related to IPv6, for which would remove the record after no traffic better place has been seen from a node found in a long period
this document.

8.1 NAT-PT with DNS-ALG

The DNS-ALG component of time (e.g., a month or
year) might be one possible approach.

To insert or update the record, the node must discover the DNS server NAT-PT mangles A records to send the update look like AAAA
records to somehow, similar the IPv6-only nodes. Numerous problems have been
identified with DNS-ALG [I-D.ietf-v6ops-natpt-to-exprmntl]. This is
a strong reason not to as discussed use NAT-PT in Section
6.2. One way to automate this is looking up the first place.

8.2 Renumbering Procedures and Applications' Use of DNS server
authoritative (e.g., through SOA record) for

One of the most difficult problems of systematic IP address being
updated, but the security material (unless the IP address-based
authorization
renumbering procedures [I-D.ietf-v6ops-renumbering-procedure] is trusted) must also be established by some other
means.

One should note that Cryptographically Generated Addresses
[I-D.ietf-send-cga] (CGAs) may require
an application which looks up a slightly different kind of
treatment. CGAs are addresses where DNS name disregards information such
as TTL, and uses the interface identifier is
calculated result obtained from a public key, a modifier (used DNS as a nonce), long as it happens
to be stored in the
subnet prefix, and other data. Depending on memory of the usage profile, CGAs
might application. For applications
which run for a long time, this could be days, weeks or might not even months;
some applications may be changed periodically due clever enough to e.g., privacy
reasons. As organize the CGA address is not predicatable, a reverse record
can only reasonably be inserted data
structures and functions in such a manner that look-ups get refreshed
now and then.

While the DNS by the node which
generates issue appears to have a clear solution, "fix the address.

7.4 DDNS with DHCP

With DHCPv4,
applications", practically this is not reasonable immediate advice;
the reverse DNS name TTL information is not typically already inserted to
the DNS that reflects to available in the name (e.g., "dhcp-67.example.com"). One
can assume similar practice may become commonplace with DHCPv6 as
well; all such mappings would be pre-configured, APIs and would require no
updating.

If a more explicit control is required, similar considerations as
with SLAAC apply, except for
libraries (so, the fact that typically one must update
a reverse DNS record instead of inserting one (if an address
assignment policy that reassigns disused addresses is adopted) advice becomes "fix the applications, APIs and
libraries"), and
updating a record seems like a slightly lot more difficult thing analysis is needed on how to practically

Durand, et al. Expires April 24, 2005 January 17, 2006 [Page 21] 19]

Internet-Draft Considerations and Issues with IPv6 DNS October 2004

secure. However, it is yet uncertain how DHCPv6 is going July 2005

go about to be used
for address assignment.

Note that when achieve the ultimate goal of avoiding using DHCP, either the host or names
longer than expected.

9. Acknowledgements

Some recommendations (Section 4.3, Section 5.1) about IPv6 service
provisioning were moved here from [I-D.ietf-v6ops-mech-v2] by Erik
Nordmark and Bob Gilligan. Havard Eidnes and Michael Patton provided
useful feedback and improvements. Scott Rose, Rob Austein, Masataka
Ohta, and Mark Andrews helped in clarifying the DHCP server could
perform issues regarding
additional data and the DNS updates; see use of TTL. Jefsey Morfin, Ralph Droms,
Peter Koch, Jinmei Tatuya, Iljitsch van Beijnum, Edward Lewis, and
Rob Austein provided useful feedback during the implications in Section 6.2.

If disused addresses were to be reassigned, host-based DDNS reverse
updates would need policy considerations WG last call. Thomas
Narten provided extensive feedback during the IESG evaluation.

10. Security Considerations

This document reviews the operational procedures for IPv6 DNS record modification,
as noted above. On the other hand, if disused address were
operations and does not to be
assigned, host-based DNS reverse updates would have similar security considerations as SLAAC in Section 7.3. Server-based updates have
similar properties except itself.

However, it is worth noting that the janitorial process could be
integrated with DHCP address assignment.

7.5 DDNS in particular with Dynamic Prefix Delegation

In cases where a prefix, instead of an address, is being used and
updated, one should consider what is the location of the server where
DDNS updates are made. That is, where the DNS server is located:

1. At
Updates, security models based on the same organization as the prefix delegator.

2. At the site where the prefixes source address validation are delegated to. In this case,
the authority of the DNS reverse zone corresponding to the
delegated prefix is also delegated to the site.

3. Elsewhere; this implies a relationship between the site and where
DNS server is located,
very weak and such a relationship should cannot be rather
straightforward to secure as well. Like in the previous case,
the authority of the DNS reverse zone is also delegated.

In the first case, managing the reverse DNS (delegation) is simpler
as the DNS server and the prefix delegator are recommended -- they could only be considered
in the same
administrative domain (as there is no need to delegate anything at
all); alternatively, the prefix delegator might forgo DDNS reverse
capability altogether, and use e.g., wildcard records (as described
in Section 7.2). In environments where ingress filtering [RFC3704] has been
deployed. On the other cases, hand, it can should be slighly more
difficult, particularly as the site will have to configure noted that setting up an
authorization mechanism (e.g., a shared secret, or public-private
keys) between a node and the DNS server has to be authoritative for the delegated reverse zone, implying
automatic configuration done manually, and
may require quite a bit of time and expertise.

To re-emphasize what was already stated, the reverse+forward DNS server -- as
check provides very weak security at best, and the prefix only
(questionable) security-related use for them may be
dynamic.

Managing the DDNS reverse updates is typically simple in the second
case, as the updated server is located at the local site, and
arguably IP address-based authentication could be sufficient (or if
not, setting up security relationships would be simpler). As there
is an explicit (security) relationship between the parties conjunction
with other mechanisms when authenticating a user.

11. References

11.1 Normative References

[I-D.ietf-dnsop-ipv6-dns-configuration]
Jeong, J., "IPv6 Host Configuration of DNS Server
Information Approaches",
draft-ietf-dnsop-ipv6-dns-configuration-06 (work in the
third case, setting up the security relationships to allow reverse
progress), May 2005.

[I-D.ietf-ipv6-unique-local-addr]
Hinden, R. and B. Haberman, "Unique Local IPv6 Unicast
Addresses", draft-ietf-ipv6-unique-local-addr-09 (work in
progress), January 2005.

Durand, et al. Expires April 24, 2005 January 17, 2006 [Page 22] 20]

Internet-Draft Considerations and Issues with IPv6 DNS October 2004

DDNS updates should be rather straightforward as well (but IP
address-based authentication might not be acceptable). In the first
case, however, setting up and managing such relationships might be a
lot more difficult.

8. Miscellaneous DNS Considerations

This section describes miscellaneous considerations about DNS which
seem related to IPv6, July 2005

[I-D.ietf-v6ops-renumbering-procedure]
Baker, F., "Procedures for which no better place has been found in
this document.

8.1 NAT-PT with DNS-ALG

The DNS-ALG component of NAT-PT mangles A records to look like AAAA
records to the IPv6-only nodes. Numerous problems have been
identified with DNS-ALG [I-D.durand-v6ops-natpt-dns-alg-issues].
This is Renumbering an IPv6 Network
without a strong reason not to use NAT-PT Flag Day",
draft-ietf-v6ops-renumbering-procedure-05 (work in
progress), March 2005.

[RFC1034] Mockapetris, P., "Domain names - concepts and facilities",
STD 13, RFC 1034, November 1987.

[RFC2136] Vixie, P., Thomson, S., Rekhter, Y., and J. Bound,
"Dynamic Updates in the first place.

8.2 Renumbering Procedures Domain Name System (DNS UPDATE)",
RFC 2136, April 1997.

[RFC2181] Elz, R. and Applications' Use of DNS

One of R. Bush, "Clarifications to the most difficult problems DNS
Specification", RFC 2181, July 1997.

[RFC2182] Elz, R., Bush, R., Bradner, S., and M. Patton, "Selection
and Operation of systematic IP address
renumbering procedures [I-D.ietf-v6ops-renumbering-procedure] is that
an application which looks up a Secondary DNS name disregards information such
as TTL, Servers", BCP 16, RFC 2182,
July 1997.

[RFC2462] Thomson, S. and uses the result obtained from T. Narten, "IPv6 Stateless Address
Autoconfiguration", RFC 2462, December 1998.

[RFC2671] Vixie, P., "Extension Mechanisms for DNS as long as it happens
to be stored (EDNS0)",
RFC 2671, August 1999.

[RFC2821] Klensin, J., "Simple Mail Transfer Protocol", RFC 2821,
April 2001.

[RFC3007] Wellington, B., "Secure Domain Name System (DNS) Dynamic
Update", RFC 3007, November 2000.

[RFC3041] Narten, T. and R. Draves, "Privacy Extensions for
Stateless Address Autoconfiguration in the memory IPv6", RFC 3041,
January 2001.

[RFC3056] Carpenter, B. and K. Moore, "Connection of the application. For applications
which run IPv6 Domains
via IPv4 Clouds", RFC 3056, February 2001.

[RFC3152] Bush, R., "Delegation of IP6.ARPA", BCP 49, RFC 3152,
August 2001.

[RFC3315] Droms, R., Bound, J., Volz, B., Lemon, T., Perkins, C.,
and M. Carney, "Dynamic Host Configuration Protocol for a long time, this could be days, weeks or even months;
some applications may be clever enough to organize the data
structures
IPv6 (DHCPv6)", RFC 3315, July 2003.

[RFC3363] Bush, R., Durand, A., Fink, B., Gudmundsson, O., and functions T.
Hain, "Representing Internet Protocol version 6 (IPv6)

Durand, et al. Expires January 17, 2006 [Page 21]

Internet-Draft Considerations with IPv6 DNS July 2005

Addresses in such a manner that look-ups get refreshed
now and then.

While the issue appears to have a clear solution, "fix the
applications", practically this is not reasonable immediate advice;
the TTL information is not typically available Domain Name System (DNS)", RFC 3363,
August 2002.

[RFC3364] Austein, R., "Tradeoffs in the APIs and
libraries (so, the advice becomes "fix the applications, APIs Domain Name System (DNS)
Support for Internet Protocol version 6 (IPv6)", RFC 3364,
August 2002.

[RFC3513] Hinden, R. and
libraries"), S. Deering, "Internet Protocol Version 6
(IPv6) Addressing Architecture", RFC 3513, April 2003.

[RFC3596] Thomson, S., Huitema, C., Ksinant, V., and a lot more analysis is needed on how to practically
go about M. Souissi,
"DNS Extensions to achieve the ultimate goal of avoiding using the names
longer than expected.

9. Acknowledgements

Some recommendations (Section 4.3, Section 5.1) about Support IP Version 6", RFC 3596,
October 2003.

[RFC3646] Droms, R., "DNS Configuration options for Dynamic Host
Configuration Protocol for IPv6 service
provisioning were moved here from [I-D.ietf-v6ops-mech-v2] by Erik
Nordmark (DHCPv6)", RFC 3646,
December 2003.

[RFC3736] Droms, R., "Stateless Dynamic Host Configuration Protocol
(DHCP) Service for IPv6", RFC 3736, April 2004.

[RFC3879] Huitema, C. and Bob Gilligan. Havard Eidnes B. Carpenter, "Deprecating Site Local
Addresses", RFC 3879, September 2004.

[RFC3901] Durand, A. and Michael Patton provided
useful feedback J. Ihren, "DNS IPv6 Transport Operational
Guidelines", BCP 91, RFC 3901, September 2004.

[RFC4038] Shin, M-K., Hong, Y-G., Hagino, J., Savola, P., and improvements. Scott Rose, Rob Austein, Masataka
Ohta, E.
Castro, "Application Aspects of IPv6 Transition",
RFC 4038, March 2005.

[RFC4074] Morishita, Y. and Mark Andrews helped in clarifying the issues regarding
additional data T. Jinmei, "Common Misbehavior Against
DNS Queries for IPv6 Addresses", RFC 4074, May 2005.

11.2 Informative References

[I-D.durand-dnsop-dont-publish]
Durand, A. and T. Chown, "To publish, or not to publish,
that is the use of TTL. Jefsey Morfin, Ralph Droms,
Peter Koch, Jinmei Tatuya, Iljitsch van Beijnum, Edward Lewis, and
Rob Austein provided useful feedback during the WG last call. Thomas question.", draft-durand-dnsop-dont-publish-00
(work in progress), February 2005.

[I-D.huitema-v6ops-teredo]
Huitema, C., "Teredo: Tunneling IPv6 over UDP through
NATs", draft-huitema-v6ops-teredo-05 (work in progress),
April 2005.

[I-D.huston-6to4-reverse-dns]
Huston, G., "6to4 Reverse DNS Delegation",

Durand, et al. Expires April 24, 2005 January 17, 2006 [Page 23] 22]

Internet-Draft Considerations and Issues with IPv6 DNS July 2005

draft-huston-6to4-reverse-dns-03 (work in progress),
October 2004

Narten provided extensive feedback during the IESG evaluation.

10. Security Considerations

This document reviews the operational procedures for IPv6 DNS
operations 2004.

[I-D.ietf-dhc-ddns-resolution]
Stapp, M. and does not have security considerations in itself.

However, it is worth noting that B. Volz, "Resolution of FQDN Conflicts among
DHCP Clients", draft-ietf-dhc-ddns-resolution-09 (work in particular with Dynamic DNS
Updates, security models based on the source address validation are
very weak
progress), June 2005.

[I-D.ietf-dhc-fqdn-option]
Stapp, M. and cannot be recommended -- they could only be considered Y. Rekhter, "The DHCP Client FQDN Option",
draft-ietf-dhc-fqdn-option-10 (work in the environments where ingress filtering [RFC3704] has been
deployed. On the other hand, it should be noted that setting up an
authorization mechanism (e.g., a shared secret, or public-private
keys) between a node progress),
February 2005.

[I-D.ietf-dnsext-dhcid-rr]
Stapp, M., Lemon, T., and the A. Gustafsson, "A DNS server has to be done manually, and
may require quite a bit of time RR for
encoding DHCP information (DHCID RR)",
draft-ietf-dnsext-dhcid-rr-09 (work in progress),
February 2005.

[I-D.ietf-dnsop-bad-dns-res]
Larson, M. and expertise.

To re-emphasize which was already stated, P. Barber, "Observed DNS Resolution
Misbehavior", draft-ietf-dnsop-bad-dns-res-03 (work in
progress), October 2004.

[I-D.ietf-dnsop-inaddr-required]
Senie, D., "Encouraging the reverse+forward use of DNS
check provides very weak security at best, and the only
(questionable) security-related use for them may be in conjunction
with other mechanisms when authenticating a user.

11. References

11.1 Normative References

[I-D.ietf-dnsop-ipv6-dns-configuration]
Jeong, J., "IPv6 Host Configuration of DNS Server
Information Approaches",
draft-ietf-dnsop-ipv6-dns-configuration-04 IN-ADDR Mapping",
draft-ietf-dnsop-inaddr-required-06 (work in progress), September 2004.

[I-D.ietf-dnsop-misbehavior-against-aaaa]
Morishita, Y. and T. Jinmei, "Common Misbehavior against
DNS Queries for
February 2005.

[I-D.ietf-v6ops-3gpp-analysis]
Wiljakka, J., "Analysis on IPv6 Addresses",
draft-ietf-dnsop-misbehavior-against-aaaa-01 (work Transition in
progress), April 2004.

[I-D.ietf-v6ops-application-transition]
Shin, M., "Application Aspects of IPv6 Transition",
draft-ietf-v6ops-application-transition-03 3GPP
Networks", draft-ietf-v6ops-3gpp-analysis-11 (work in
progress), June October 2004.

[I-D.ietf-v6ops-renumbering-procedure]
Baker, F., Lear,

[I-D.ietf-v6ops-mech-v2]
Nordmark, E. and R. Droms, "Procedures Gilligan, "Basic Transition Mechanisms
for
Renumbering an IPv6 Network without a Flag Day",
draft-ietf-v6ops-renumbering-procedure-01 Hosts and Routers", draft-ietf-v6ops-mech-v2-07
(work in progress), March 2005.

[I-D.ietf-v6ops-natpt-to-exprmntl]
Aoun, C. and E. Davies, "Reasons to Move NAT-PT to
Experimental", draft-ietf-v6ops-natpt-to-exprmntl-01 (work
in progress), July 2004. 2005.

[I-D.ietf-v6ops-onlinkassumption]
Roy, S., "IPv6 Neighbor Discovery On-Link Assumption
Considered Harmful", draft-ietf-v6ops-onlinkassumption-03
(work in progress), May 2005.

Durand, et al. Expires April 24, 2005 January 17, 2006 [Page 24] 23]

Internet-Draft Considerations with IPv6 DNS July 2005

[I-D.ietf-v6ops-v6onbydefault]
Roy, S., Durand, A., and Issues J. Paugh, "Issues with Dual Stack
IPv6 on by Default", draft-ietf-v6ops-v6onbydefault-03
(work in progress), July 2004.

[I-D.jeong-dnsop-ipv6-dns-discovery]
Jeong, J., "IPv6 DNS October 2004

[RFC1034] Mockapetris, P., "Domain names - concepts Configuration based on Router
Advertisement", draft-jeong-dnsop-ipv6-dns-discovery-04
(work in progress), February 2005.

[I-D.ohta-preconfigured-dns]
Ohta, M., "Preconfigured DNS Server Addresses",
draft-ohta-preconfigured-dns-01 (work in progress),
February 2004.

[RFC2766] Tsirtsis, G. and facilities",
STD 13, P. Srisuresh, "Network Address
Translation - Protocol Translation (NAT-PT)", RFC 1034, November 1987.

[RFC2136] 2766,
February 2000.

[RFC2782] Gulbrandsen, A., Vixie, P., Thomson, S., Rekhter, Y. and J. Bound, "Dynamic
Updates in L. Esibov, "A DNS RR for
specifying the Domain Name System location of services (DNS UPDATE)", RFC 2136,
April 1997.

[RFC2181] Elz, R. and R. Bush, "Clarifications to the DNS
Specification", SRV)", RFC 2181, July 1997.

[RFC2182] Elz, R., Bush, R., Bradner, S. and M. Patton, "Selection
and Operation of Secondary DNS Servers", BCP 16, RFC 2182,
July 1997.

[RFC2462] Thomson, S. and T. Narten, "IPv6 Stateless Address
Autoconfiguration", RFC 2462, December 1998.

[RFC2671] Vixie, P., "Extension Mechanisms for 2782,
February 2000.

[RFC2826] Internet Architecture Board, "IAB Technical Comment on the
Unique DNS (EDNS0)", RFC
2671, August 1999.

[RFC3007] Wellington, B., "Secure Domain Name System (DNS) Dynamic
Update", Root", RFC 3007, November 2826, May 2000.

[RFC3041] Narten, T.

[RFC3704] Baker, F. and R. Draves, "Privacy Extensions P. Savola, "Ingress Filtering for
Stateless Address Autoconfiguration in IPv6", RFC 3041,
January 2001.

[RFC3056] Carpenter, B. and K. Moore, "Connection of IPv6 Domains
via IPv4 Clouds", RFC 3056, February 2001.

[RFC3152] Bush, R., "Delegation of IP6.ARPA", Multihomed
Networks", BCP 49, 84, RFC 3152,
August 2001.

[RFC3315] Droms, R., Bound, J., Volz, B., Lemon, 3704, March 2004.

[RFC3972] Aura, T., Perkins, C. and
M. Carney, "Dynamic Host Configuration Protocol for IPv6
(DHCPv6)", RFC 3315, July 2003.

[RFC3363] Bush, R., Durand, A., Fink, B., Gudmundsson, O. and T.
Hain, "Representing Internet Protocol version 6 (IPv6) "Cryptographically Generated Addresses in the Domain Name System (DNS)", (CGA)",
RFC 3363,
August 2002.

[RFC3364] Austein, R., "Tradeoffs in Domain Name System (DNS)
Support 3972, March 2005.

[RFC4025] Richardson, M., "A Method for Internet Protocol version 6 (IPv6)", RFC 3364,
August 2002.

[RFC3513] Hinden, R. and S. Deering, "Internet Protocol Version 6
(IPv6) Addressing Architecture", Storing IPsec Keying
Material in DNS", RFC 3513, April 2003. 4025, March 2005.

Authors' Addresses

Alain Durand
SUN Microsystems, Inc.
17 Network circle UMPL17-202
Menlo Park, CA 94025
USA

Email: Alain.Durand@sun.com

Durand, et al. Expires April 24, 2005 January 17, 2006 [Page 25] 24]

Internet-Draft Considerations and Issues with IPv6 DNS October 2004

[RFC3596] Thomson, S., Huitema, C., Ksinant, V. and M. Souissi, "DNS
Extensions to Support IP Version 6", RFC 3596, October
2003.

[RFC3646] Droms, R., "DNS Configuration options for Dynamic Host
Configuration Protocol for IPv6 (DHCPv6)", RFC 3646,
December 2003.

[RFC3736] Droms, R., "Stateless Dynamic Host Configuration Protocol
(DHCP) Service for IPv6", RFC 3736, April 2004.

[RFC3879] Huitema, C. and B. Carpenter, "Deprecating Site Local
Addresses", RFC 3879, September 2004.

[RFC3901] Durand, July 2005

Johan Ihren
Autonomica
Bellmansgatan 30
SE-118 47 Stockholm
Sweden

Email: johani@autonomica.se

Pekka Savola
CSC/FUNET
Espoo
Finland

Email: psavola@funet.fi

Appendix A. Unique Local Addressing Considerations for DNS

Unique local addresses [I-D.ietf-ipv6-unique-local-addr] have
replaced the now-deprecated site-local addresses [RFC3879]. From the
perspective of the DNS, the locally generated unique local addresses
(LUL) and J. Ihren, "DNS IPv6 Transport Operational
Guidelines", BCP 91, RFC 3901, September 2004.

11.2 Informative References

[I-D.durand-v6ops-natpt-dns-alg-issues]
Durand, A., "Issues site-local addresses have similar properties.

The interactions with NAT-PT DNS ALG in RFC2766",
draft-durand-v6ops-natpt-dns-alg-issues-00 (work in
progress), February 2003.

[I-D.huitema-v6ops-teredo]
Huitema, C., "Teredo: Tunneling IPv6 over UDP through
NATs", draft-huitema-v6ops-teredo-02 (work in progress),
June 2004.

[I-D.huston-6to4-reverse-dns]
Huston, G., "6to4 Reverse DNS Delegation",
draft-huston-6to4-reverse-dns-03 (work come in progress),
October 2004.

[I-D.ietf-dhc-ddns-resolution]
Stapp, M., "Resolution two flavors: forward and reverse
DNS.

To actually use local addresses within a site, this implies the
deployment of a "split-faced" or a fragmented DNS Name Conflicts Among DHCP
Clients", draft-ietf-dhc-ddns-resolution-08 (work in
progress), October 2004.

[I-D.ietf-dhc-fqdn-option]
Stapp, M. name space, for the
zones internal to the site, and Y. Rekhter, "The DHCP Client FQDN Option",
draft-ietf-dhc-fqdn-option-07 (work the outsiders' view to it. The
procedures to achieve this are not elaborated here. The implication
is that local addresses must not be published in progress), July
2004.

[I-D.ietf-dnsext-dhcid-rr]
Stapp, M., Lemon, T. and A. Gustafsson, "A the public DNS.

To faciliate reverse DNS RR (if desired) with local addresses, the stub
resolvers must look for
encoding DHCP DNS information (DHCID RR)",
draft-ietf-dnsext-dhcid-rr-08 (work from the local DNS servers,
not e.g. starting from the root servers, so that the local
information may be provided locally. Note that the experience of
private addresses in progress), July IPv4 has shown that the root servers get loaded
for requests for private address lookups in any case. This
requirement is discussed in [I-D.ietf-ipv6-unique-local-addr].

Appendix B. Behaviour of Additional Data in IPv4/IPv6 Environments

DNS responses do not always fit in a single UDP packet. We'll
examine the cases which happen when this is due to too much data in
the Additional Section.

Durand, et al. Expires April 24, 2005 January 17, 2006 [Page 26] 25]

Internet-Draft Considerations and Issues with IPv6 DNS October 2004

2004.

[I-D.ietf-dnsop-bad-dns-res]
Larson, M. and P. Barber, "Observed DNS Resolution
Misbehavior", draft-ietf-dnsop-bad-dns-res-02 (work in
progress), July 2004.

[I-D.ietf-dnsop-dontpublish-unreachable]
Hazel, P., "IP Addresses that should never appear 2005

B.1 Description of Additional Data Scenarios

There are two kinds of additional data:

1. "critical" additional data; this must be included in all
scenarios, with all the
public DNS", draft-ietf-dnsop-dontpublish-unreachable-03
(work in progress), February 2002.

[I-D.ietf-dnsop-inaddr-required]
Senie, D., "Requiring DNS IN-ADDR Mapping",
draft-ietf-dnsop-inaddr-required-05 (work in progress),
April 2004.

[I-D.ietf-ipseckey-rr]
Richardson, M., "A method for storing IPsec keying
material in DNS", draft-ietf-ipseckey-rr-11 (work in
progress), July 2004.

[I-D.ietf-ipv6-unique-local-addr]
Hinden, R. RRsets, and B. Haberman, "Unique Local IPv6 Unicast
Addresses", draft-ietf-ipv6-unique-local-addr-06 (work in
progress), September 2004.

[I-D.ietf-send-cga]
Aura, T., "Cryptographically Generated Addresses (CGA)",
draft-ietf-send-cga-06 (work in progress), April 2004.

[I-D.ietf-v6ops-3gpp-analysis]
Wiljakka, J., "Analysis on IPv6 Transition in 3GPP
Networks", draft-ietf-v6ops-3gpp-analysis-10 (work

2. "courtesy" additional data; this could be sent in
progress), May 2004.

[I-D.ietf-v6ops-mech-v2]
Nordmark, E. full, with only
a few RRsets, or with no RRsets, and R. Gilligan, "Basic Transition Mechanisms can be fetched separately as
well, but at the cost of additional queries.

The responding server can algorithmically determine which type the
additional data is by checking whether it's at or below a zone cut.

Only those additional data records (even if sometimes carelessly
termed "glue") are considered "critical" or real "glue" if and only
if they meet the abovementioned condition, as specified in Section
4.2.1 of [RFC1034].

Remember that resource record sets (RRsets) are never "broken up", so
if a name has 4 A records and 5 AAAA records, you can either return
all 9, all 4 A records, all 5 AAAA records or nothing. In
particular, notice that for IPv6 Hosts the "critical" additional data getting
all the RRsets can be critical.

In particular, [RFC2181] specifies (in Section 9) that:

a. if all the "critical" RRsets do not fit, the sender should set
the TC bit, and Routers", draft-ietf-v6ops-mech-v2-06
(work the recipient should discard the whole response
and retry using mechanism allowing larger responses such as TCP.

b. "courtesy" additional data should not cause the setting of TC
bit, but instead all the non-fitting additional data RRsets
should be removed.

An example of the "courtesy" additional data is A/AAAA records in progress), September 2004.

[I-D.ietf-v6ops-onlinkassumption]
Roy, S., Durand, A.
conjunction with MX records as shown in Section 4.4; an example of
the "critical" additional data is shown below (where getting both the
A and J. Paugh, "IPv6 Neighbor Discovery
On-Link Assumption Considered Harmful",
draft-ietf-v6ops-onlinkassumption-02 (work AAAA RRsets is critical w.r.t. to the NS RR):

child.example.com. IN NS ns.child.example.com.
ns.child.example.com. IN A 192.0.2.1
ns.child.example.com. IN AAAA 2001:db8::1

When there is too much "courtesy" additional data, at least the non-
fitting RRsets should be removed [RFC2181]; however, as the
additional data is not critical, even all of it could be safely
removed.

Durand, et al. Expires January 17, 2006 [Page 26]

Internet-Draft Considerations with IPv6 DNS July 2005

When there is too much "critical" additional data, TC bit will have
to be set, and the recipient should ignore the response and retry
using TCP; if some data were to be left in the UDP response, the
issue is which data could be retained.

Failing to discard the response with TC bit or omitting critical
information but not setting TC bit lead to an unrecoverable problem.
Omitting only some of the RRsets if all would not fit (but not
setting TC bit) leads to a performance problem. These are discussed
in the next two subsections.

B.2 Which Additional Data to Keep, If Any?

If the implementation decides to keep as much data (whether
"critical" or "courtesy") as possible in the UDP responses, it might
be tempting to use the transport of the DNS query as a hint in either
of these cases: return the AAAA records if the query was done over
IPv6, or return the A records if the query was done over IPv4.
However, this breaks the model of independence of DNS transport and
resource records, as noted in Section 1.2.

With courtesy additional data, as long as enough RRsets will be
removed so that TC will not be set, it is allowed to send as many
complete RRsets as the implementations prefers. However, the
implementations are also free to omit all such RRsets, even if
complete. Omitting all the RRsets (when removing only some would
suffice) may create a performance penalty, whereby the client may
need to issue one or more additional queries to obtain necessary
and/or consistent information.

With critical additional data, the alternatives are either returning
nothing (and absolutely requiring a retry with TCP) or returning
something (working also in the case if the recipient does not discard
the response and retry using TCP) in addition to setting the TC bit.
If the process for selecting "something" from the critical data would
otherwise be practically "flipping the coin" between A and AAAA
records, it could be argued that if one looked at the transport of
the query, it would have a larger possibility of being right than
just 50/50. In other words, if the returned critical additional data
would have to be selected somehow, using something more sophisticated
than a random process would seem justifiable.

That is, leaving in some intelligently selected critical additional
data is a tradeoff between creating an optimization for those
resolvers which ignore the "should discard" recommendation, and
causing a protocol problem by propagating inconsistent information
about "critical" records in progress),
May 2004.

[I-D.ietf-v6ops-v6onbydefault] the caches.

Durand, et al. Expires April 24, 2005 January 17, 2006 [Page 27]

Internet-Draft Considerations and Issues with IPv6 DNS October 2004

Roy, S., Durand, A. and J. Paugh, "Issues with Dual Stack IPv6 on by Default", draft-ietf-v6ops-v6onbydefault-03
(work in progress), July 2004.

[I-D.jeong-dnsop-ipv6-dns-discovery]
Jeong, J., "IPv6 DNS Discovery based on Router
Advertisement", draft-jeong-dnsop-ipv6-dns-discovery-02
(work in progress), July 2004.

[I-D.moore-6to4-dns]
Moore, K., "6to4 and DNS", draft-moore-6to4-dns-03 (work
in progress), October 2002.

[I-D.ohta-preconfigured-dns]
Ohta, M., "Preconfigured DNS Server Addresses",
draft-ohta-preconfigured-dns-01 (work 2005

Similarly, leaving in progress),
February 2004.

[I-D.savola-v6ops-6bone-mess]
Savola, P., "Moving from 6bone to IPv6 Internet",
draft-savola-v6ops-6bone-mess-01 (work the complete courtesy additional data RRsets
instead of removing all the RRsets is a performance tradeoff as
described in progress),
November 2002.

[RFC2766] Tsirtsis, G. and P. Srisuresh, "Network Address
Translation - Protocol Translation (NAT-PT)", RFC 2766,
February 2000.

[RFC2782] Gulbrandsen, A., Vixie, P. and L. Esibov, "A DNS RR for
specifying the location next section.

B.3 Discussion of services (DNS SRV)", RFC 2782,
February 2000.

[RFC2826] Internet Architecture Board, "IAB Technical Comment on the
Unique DNS Root", RFC 2826, May 2000.

[RFC3704] Baker, F. and P. Savola, "Ingress Filtering for Multihomed
Networks", BCP 84, RFC 3704, March 2004.

Authors' Addresses

Alain Durand
SUN Microsystems, Inc.
17 Network circle UMPL17-202
Menlo Park, CA 94025
USA

EMail: Alain.Durand@sun.com

Durand, et al. Expires April 24, 2005 [Page 28]
Internet-Draft Considerations and Issues with IPv6 DNS October 2004

Johan Ihren
Autonomica
Bellmansgatan 30
SE-118 47 Stockholm
Sweden

EMail: johani@autonomica.se

Pekka Savola
CSC/FUNET
Espoo
Finland

EMail: psavola@funet.fi

Appendix A. Site-local Addressing Considerations for DNS Potential Problems

As site-local addressing has been deprecated, noted above, the temptation for omitting only some of the considerations for
site-local addressing are
additional data could be problematic. This is discussed briefly here. Unique local
addressing format [I-D.ietf-ipv6-unique-local-addr] has been proposed more below.

For courtesy additional data, this causes a potential performance
problem as this requires that the clients issue re-queries for the
potentially omitted RRsets. For critical additional data, this
causes a replacement, but being work-in-progress, it potential unrecoverable problem if the response is not considered
further.

The interactions
discarded and the query not re-tried with TCP, as the nameservers
might be reachable only through the omitted RRsets.

If an implementation would look at the transport used for the query,
it is worth remembering that often the host using the records is
different from the node requesting them from the authoritative DNS come
server (or even a caching resolver). So, whichever version the
requestor (e.g., a recursive server in two flavors: forward and reverse
DNS.

To actually use site-local addresses within the middle) uses makes no
difference to the ultimate user of the records, whose transport
capabilities might differ from those of the requestor. This might
result in e.g., inappropriately returning A records to an IPv6-only
node, going through a site, this implies translation, or opening up another IP-level
session (e.g., a PDP context [I-D.ietf-v6ops-3gpp-analysis]).
Therefore, at least in many scenarios, it would be very useful if the
information returned would be consistent and complete -- or if that
is not feasible, return no misleading information but rather leave it
to the client to query again.

The problem of too much additional data seems to be an operational
one: the zone administrator entering too many records which will be
returned either truncated (or missing some RRsets, depending on
implementations) to the
deployment of a "split-faced" or a fragmented DNS name space, users. A protocol fix for the
zones internal this is using
EDNS0 [RFC2671] to signal the site, and capacity for larger UDP packet sizes,
pushing up the outsiders' view to it. The
procedures to achieve relevant threshold. Further, DNS server
implementations should rather omit courtesy additional data
completely rather than including only some RRsets [RFC2181]. An
operational fix for this are not elaborated here. The implication is that site-local addresses must not be published in having the public DNS.

To faciliate reverse DNS (if desired) with site-local addresses, server implementations
return a warning when the
stub resolvers must look for administrators create zones which would
result in too much additional data being returned. Further, DNS information from
server implementations should warn of or disallow such zone
configurations which are recursive or otherwise difficult to manage
by the local DNS
servers, not e.g. starting from protocol.

Additionally, to avoid the root servers, so that case where an application would not get an
address at all due to some of courtesy additional data being omitted,

Durand, et al. Expires January 17, 2006 [Page 28]

Internet-Draft Considerations with IPv6 DNS July 2005

the
site-local information may resolvers should be provided locally. Note that able to query the
experience specific records of private addresses in IPv4 has shown that the root
servers get loaded for requests for private address lookups
desired protocol, not just rely on getting all the required RRsets in any
case.
the additional section.

Durand, et al. Expires April 24, 2005 January 17, 2006 [Page 29]

Internet-Draft Considerations and Issues with IPv6 DNS October 2004 July 2005

Intellectual Property Statement

The IETF takes no position regarding the validity or scope of any
Intellectual Property Rights or other rights that might be claimed to
pertain to the implementation or use of the technology described in
this document or the extent to which any license under such rights
might or might not be available; nor does it represent that it has
made any independent effort to identify any such rights. Information
on the procedures with respect to rights in RFC documents can be
found in BCP 78 and BCP 79.

Copies of IPR disclosures made to the IETF Secretariat and any
assurances of licenses to be made available, or the result of an
attempt made to obtain a general license or permission for the use of
such proprietary rights by implementers or users of this
specification can be obtained from the IETF on-line IPR repository at
http://www.ietf.org/ipr.

The IETF invites any interested party to bring to its attention any
copyrights, patents or patent applications, or other proprietary
rights that may cover technology that may be required to implement
this standard. Please address the information to the IETF at
ietf-ipr@ietf.org.

Disclaimer of Validity

This document and the information contained herein are provided on an
"AS IS" basis and THE CONTRIBUTOR, THE ORGANIZATION HE/SHE REPRESENTS
OR IS SPONSORED BY (IF ANY), THE INTERNET SOCIETY AND THE INTERNET
ENGINEERING TASK FORCE DISCLAIM 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.

Copyright (C) The Internet Society (2004). (2005). This document is subject
to the rights, licenses and restrictions contained in BCP 78, and
except as set forth therein, the authors retain all their rights.

Acknowledgment

Funding for the RFC Editor function is currently provided by the
Internet Society.

Durand, et al. Expires April 24, 2005 January 17, 2006 [Page 30]

----