-Network Working Group D. Blacka
+
+DNSEXT D. Blacka
Internet-Draft Verisign, Inc.
-Expires: August 3, 2005 February 2, 2005
+Expires: January 19, 2006 July 18, 2005
DNSSEC Experiments
- draft-ietf-dnsext-dnssec-experiments-00
+ draft-ietf-dnsext-dnssec-experiments-01
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 aware will be disclosed, in accordance with
- RFC 3668.
+ 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 becomes
+ aware will be disclosed, in accordance with 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.
+ other groups may also distribute working documents as Internet-
+ Drafts.
Internet-Drafts are draft documents valid for a maximum of six months
and may be updated, replaced, or obsoleted by other documents at any
The list of Internet-Draft Shadow Directories can be accessed at
http://www.ietf.org/shadow.html.
- This Internet-Draft will expire on August 3, 2005.
+ This Internet-Draft will expire on January 19, 2006.
Copyright Notice
Abstract
- In the long history of the development of the DNS security [1]
- extensions (DNSSEC), a number of alternate methodologies and
- modifications have been proposed and rejected for practical, rather
- than strictly technical, reasons. There is a desire to be able to
- experiment with these alternate methods in the public DNS. This
- document describes a methodology for deploying alternate,
- non-backwards-compatible, DNSSEC methodologies in an experimental
- fashion without disrupting the deployment of standard DNSSEC.
+ In the long history of the development of the DNS security extensions
+ [1] (DNSSEC), a number of alternate methodologies and modifications
+ have been proposed and rejected for practical, rather than strictly
+ technical, reasons. There is a desire to be able to experiment with
+ these alternate methods in the public DNS. This document describes a
+ methodology for deploying alternate, non-backwards-compatible, DNSSEC
+ methodologies in an experimental fashion without disrupting the
+ deployment of standard DNSSEC.
+
-Blacka Expires August 3, 2005 [Page 1]
+Blacka Expires January 19, 2006 [Page 1]
\f
-Internet-Draft DNSSEC Experiments February 2005
+Internet-Draft DNSSEC Experiments July 2005
Table of Contents
8. Security Considerations . . . . . . . . . . . . . . . . . . 11
9. IANA Considerations . . . . . . . . . . . . . . . . . . . . 12
10. References . . . . . . . . . . . . . . . . . . . . . . . . . 13
- 10.1 Normative References . . . . . . . . . . . . . . . . . . . 13
- 10.2 Informative References . . . . . . . . . . . . . . . . . . 13
- Editorial Comments . . . . . . . . . . . . . . . . . . . . . 14
- Author's Address . . . . . . . . . . . . . . . . . . . . . . 14
- Intellectual Property and Copyright Statements . . . . . . . 15
+ 10.1 Normative References . . . . . . . . . . . . . . . . . . 13
+ 10.2 Informative References . . . . . . . . . . . . . . . . . 13
+ Author's Address . . . . . . . . . . . . . . . . . . . . . . 13
+ Intellectual Property and Copyright Statements . . . . . . . 14
+
-Blacka Expires August 3, 2005 [Page 2]
+Blacka Expires January 19, 2006 [Page 2]
\f
-Internet-Draft DNSSEC Experiments February 2005
+Internet-Draft DNSSEC Experiments July 2005
1. Definitions and Terminology
-Blacka Expires August 3, 2005 [Page 3]
+Blacka Expires January 19, 2006 [Page 3]
\f
-Internet-Draft DNSSEC Experiments February 2005
+Internet-Draft DNSSEC Experiments July 2005
2. Overview
introduce non-backwards-compatible changes to DNSSEC, and to try
these changes on real zones in the public DNS. This creates a
problem when the change to DNSSEC would make all or part of the zone
- using those changes appear bogus or otherwise broken to existing
- DNSSEC-aware resolvers.
+ using those changes appear bogus (bad) or otherwise broken to
+ existing DNSSEC-aware resolvers.
This document describes a standard methodology for setting up public
- DNSSEC experiments. This methodology addresses the issue of
- co-existence with standard DNSSEC and DNS by using unknown algorithm
+ DNSSEC experiments. This methodology addresses the issue of co-
+ existence with standard DNSSEC and DNS by using unknown algorithm
identifiers to hide the experimental DNSSEC protocol modifications
from standard DNSSEC-aware resolvers.
-Blacka Expires August 3, 2005 [Page 4]
+Blacka Expires January 19, 2006 [Page 4]
\f
-Internet-Draft DNSSEC Experiments February 2005
+Internet-Draft DNSSEC Experiments July 2005
3. Experiments
strictly adhering to the DNSSEC standard, are nonetheless
interoperable with clients and server that do implement the DNSSEC
standard.
+
Non-Backwards-Compatible: describes experiments that would cause a
standard DNSSEC-aware resolver to (incorrectly) determine that all
or part of a zone is bogus, or to otherwise not interoperable with
-
-Blacka Expires August 3, 2005 [Page 5]
+Blacka Expires January 19, 2006 [Page 5]
\f
-Internet-Draft DNSSEC Experiments February 2005
+Internet-Draft DNSSEC Experiments July 2005
4. Method
- The core of the methodology is the use of only "unknown" algorithms
- to sign the experimental zone, and more importantly, having only
- unknown algorithm DS records for the delegation to the zone at the
- parent.
+ The core of the methodology is the use of strictly "unknown"
+ algorithms to sign the experimental zone, and more importantly,
+ having only unknown algorithm DS records for the delegation to the
+ zone at the parent.
This technique works because of the way DNSSEC-compliant validators
are expected to work in the presence of a DS set with only unknown
more to the point, it will not violate this behavior in an unsafe way
(see below (Section 6).)
- Because we are talking about experiments, it is recommended that
- private algorithm numbers be used (see [2], appendix A.1.1
- [Comment.1].) Normally, instead of actually inventing new signing
- algorithms, the recommended path is to create alternate algorithm
- identifiers that are aliases for the existing, known algorithms.
- While, strictly speaking, it is only necessary to create an alternate
- identifier for the mandatory algorithms (currently, this is only
- algorithm 5, RSASHA1), it is RECOMMENDED that all OPTIONAL defined
- algorithms be aliased as well.
+ Because we are talking about experiments, it is RECOMMENDED that
+ private algorithm numbers be used (see [2], appendix A.1.1. Note
+ that secure handling of private algorithms requires special handing
+ by the validator logic. See [6] for futher details.) Normally,
+ instead of actually inventing new signing algorithms, the recommended
+ path is to create alternate algorithm identifiers that are aliases
+ for the existing, known algorithms. While, strictly speaking, it is
+ only necessary to create an alternate identifier for the mandatory
+ algorithms, it is RECOMMENDED that all OPTIONAL defined algorithms be
+ aliased as well.
It is RECOMMENDED that for a particular DNSSEC experiment, a
particular domain name base is chosen for all new algorithms, then
the algorithm number (or name) is prepended to it. For example, for
experiment A, the base name of "dnssec-experiment-a.example.com" is
chosen. Then, aliases for algorithms 3 (DSA) and 5 (RSASHA1) are
- defined to be "3.dnssec-experiment-a.example.com" and
- "5.dnssec-experiment-a.example.com". However, any unique identifier
- will suffice.
+ defined to be "3.dnssec-experiment-a.example.com" and "5.dnssec-
+ experiment-a.example.com". However, any unique identifier will
-Blacka Expires August 3, 2005 [Page 6]
+Blacka Expires January 19, 2006 [Page 6]
\f
-Internet-Draft DNSSEC Experiments February 2005
+Internet-Draft DNSSEC Experiments July 2005
+ suffice.
+
Using this method, resolvers (or, more specificially, DNSSEC
validators) essentially indicate their ability to understand the
DNSSEC experiment's semantics by understanding what the new algorithm
experimental semantics), and servers and resolvers that are unware of
the experiment.
+ This method also precludes any zone from being both in an experiment
+ and in a classic DNSSEC island of security. That is, a zone is
+ either in an experiment and only experimentally validatable, or it
+ isn't.
-
-
-
-
-
-
-Blacka Expires August 3, 2005 [Page 7]
+Blacka Expires January 19, 2006 [Page 7]
\f
-Internet-Draft DNSSEC Experiments February 2005
+Internet-Draft DNSSEC Experiments July 2005
5. Defining an Experiment
this will be a mapping of private algorithm identifiers to existing,
known algorithms.
- Typically, the experiment will choose a DNS name as the algorithm
+ Normally the experiment will choose a DNS name as the algorithm
identifier base. This DNS name SHOULD be under the control of the
authors of the experiment. Then the experiment will define a mapping
between known mandatory and optional algorithms into this private
In general, however, resolvers involved in the experiment are
expected to understand both standard DNSSEC and the defined
- experimental DNSSEC protocol, although this isn't, strictly speaking,
- required.
+ experimental DNSSEC protocol, although this isn't required.
+
-Blacka Expires August 3, 2005 [Page 8]
+Blacka Expires January 19, 2006 [Page 8]
\f
-Internet-Draft DNSSEC Experiments February 2005
+Internet-Draft DNSSEC Experiments July 2005
6. Considerations
conclude that the response is bogus, either due to local policy
or implementation details. This is not expected to be the common
case, however.
- 2. It will, in general, not be possible for DNSSEC-aware resolvers
+
+ 2. In general, it will not be possible for DNSSEC-aware resolvers
not aware of the experiment to build a chain of trust through an
experimental zone.
-
-Blacka Expires August 3, 2005 [Page 9]
+Blacka Expires January 19, 2006 [Page 9]
\f
-Internet-Draft DNSSEC Experiments February 2005
+Internet-Draft DNSSEC Experiments July 2005
7. Transitions
-Blacka Expires August 3, 2005 [Page 10]
+Blacka Expires January 19, 2006 [Page 10]
\f
-Internet-Draft DNSSEC Experiments February 2005
+Internet-Draft DNSSEC Experiments July 2005
8. Security Considerations
-Blacka Expires August 3, 2005 [Page 11]
+Blacka Expires January 19, 2006 [Page 11]
\f
-Internet-Draft DNSSEC Experiments February 2005
+Internet-Draft DNSSEC Experiments July 2005
9. IANA Considerations
-Blacka Expires August 3, 2005 [Page 12]
+Blacka Expires January 19, 2006 [Page 12]
\f
-Internet-Draft DNSSEC Experiments February 2005
+Internet-Draft DNSSEC Experiments July 2005
10. References
10.1 Normative References
- [1] Arends, R., Austein, R., Massey, D., Larson, M. and S. Rose,
- "DNS Security Introduction and Requirements",
- draft-ietf-dnsext-dnssec-intro-13 (work in progress), October
- 2004.
+ [1] Arends, R., Austein, R., Larson, M., Massey, D., and S. Rose,
+ "DNS Security Introduction and Requirements", RFC 4033,
+ March 2005.
- [2] Arends, R., "Resource Records for the DNS Security Extensions",
- draft-ietf-dnsext-dnssec-records-11 (work in progress), October
- 2004.
+ [2] Arends, R., Austein, R., Larson, M., Massey, D., and S. Rose,
+ "Resource Records for the DNS Security Extensions", RFC 4034,
+ March 2005.
- [3] Arends, R., "Protocol Modifications for the DNS Security
- Extensions", draft-ietf-dnsext-dnssec-protocol-09 (work in
- progress), October 2004.
+ [3] Arends, R., Austein, R., Larson, M., Massey, D., and S. Rose,
+ "Protocol Modifications for the DNS Security Extensions",
+ RFC 4035, March 2005.
10.2 Informative References
[5] Bradner, S., "Key words for use in RFCs to Indicate Requirement
Levels", BCP 14, RFC 2119, March 1997.
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-Blacka Expires August 3, 2005 [Page 13]
-\f
-Internet-Draft DNSSEC Experiments February 2005
-
-
-Editorial Comments
-
- [Comment.1] Note: how private algorithms work in DNSSEC is not well
- explained in the DNSSECbis RFCs. In particular, how to
- validate that the DS records contain only unknown
- algorithms is not explained at all.
+ [6] Weiler, S., "Clarifications and Implementation Notes for
+ DNSSECbis", draft-weiler-dnsext-dnssec-bis-updates-00 (work in
+ progress), March 2005.
Author's Address
US
Phone: +1 703 948 3200
- EMail: davidb@verisign.com
+ Email: davidb@verisign.com
URI: http://www.verisignlabs.com
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-Blacka Expires August 3, 2005 [Page 14]
+Blacka Expires January 19, 2006 [Page 13]
\f
-Internet-Draft DNSSEC Experiments February 2005
+Internet-Draft DNSSEC Experiments July 2005
Intellectual Property Statement
-Blacka Expires August 3, 2005 [Page 15]
+Blacka Expires January 19, 2006 [Page 14]
\f
-
+
DNSEXT R. Arends
Internet-Draft Telematica Instituut
-Expires: August 4, 2005 M. Kosters
+Expires: January 19, 2006 M. Kosters
D. Blacka
Verisign, Inc.
- February 3, 2005
+ July 18, 2005
DNSSEC Opt-In
- draft-ietf-dnsext-dnssec-opt-in-06
+ draft-ietf-dnsext-dnssec-opt-in-07
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 aware will be disclosed, in accordance with
- RFC 3668.
+ 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 becomes
+ aware will be disclosed, in accordance with 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.
+ other groups may also distribute working documents as Internet-
+ Drafts.
Internet-Drafts are draft documents valid for a maximum of six months
and may be updated, replaced, or obsoleted by other documents at any
The list of Internet-Draft Shadow Directories can be accessed at
http://www.ietf.org/shadow.html.
- This Internet-Draft will expire on August 4, 2005.
+ This Internet-Draft will expire on January 19, 2006.
Copyright Notice
Abstract
- In the DNS security extensions (DNSSEC, defined in RFC 2535bis, [3],
- [4], and [5]), delegations to unsigned subzones are cryptographically
- secured. Maintaining this cryptography is not practical or
- necessary. This document describes an experimental "Opt-In" model
- that allows administrators to omit this cryptography and manage the
+ In the DNS security extensions (DNSSEC, defined in RFC 4033 [3], RFC
+ 4034 [4], and RFC 4035 [5]), delegations to unsigned subzones are
+ cryptographically secured. Maintaining this cryptography is not
+ practical or necessary. This document describes an experimental
+ "Opt-In" model that allows administrators to omit this cryptography
+ and manage the cost of adopting DNSSEC with large zones.
-Arends, et al. Expires August 4, 2005 [Page 1]
+Arends, et al. Expires January 19, 2006 [Page 1]
\f
-Internet-Draft DNSSEC Opt-In February 2005
-
+Internet-Draft DNSSEC Opt-In July 2005
- cost of adopting DNSSEC with large zones.
Table of Contents
1. Definitions and Terminology . . . . . . . . . . . . . . . . . 3
- 2. Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
- 3. Experimental Status . . . . . . . . . . . . . . . . . . . . . 5
- 4. Protocol Additions . . . . . . . . . . . . . . . . . . . . . . 6
- 4.1 Server Considerations . . . . . . . . . . . . . . . . . . 7
- 4.1.1 Delegations Only . . . . . . . . . . . . . . . . . . . 7
- 4.1.2 Insecure Delegation Responses . . . . . . . . . . . . 7
- 4.1.3 Wildcards and Opt-In . . . . . . . . . . . . . . . . . 7
- 4.1.4 Dynamic Update . . . . . . . . . . . . . . . . . . . . 8
- 4.2 Client Considerations . . . . . . . . . . . . . . . . . . 8
- 4.2.1 Delegations Only . . . . . . . . . . . . . . . . . . . 8
- 4.2.2 Validation Process Changes . . . . . . . . . . . . . . 8
- 4.2.3 NSEC Record Caching . . . . . . . . . . . . . . . . . 9
- 4.2.4 Use of the AD bit . . . . . . . . . . . . . . . . . . 9
- 5. Benefits . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
- 6. Example . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
- 7. Transition Issues . . . . . . . . . . . . . . . . . . . . . . 13
- 8. Security Considerations . . . . . . . . . . . . . . . . . . . 14
- 9. IANA Considerations . . . . . . . . . . . . . . . . . . . . . 16
- 10. Acknowledgments . . . . . . . . . . . . . . . . . . . . . . 17
- 11. References . . . . . . . . . . . . . . . . . . . . . . . . . 18
- 11.1 Normative References . . . . . . . . . . . . . . . . . . . . 18
- 11.2 Informative References . . . . . . . . . . . . . . . . . . . 18
- Authors' Addresses . . . . . . . . . . . . . . . . . . . . . . 19
- A. Implementing Opt-In using "Views" . . . . . . . . . . . . . . 20
- Intellectual Property and Copyright Statements . . . . . . . . 21
+ 2. Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
+ 3. Experimental Status . . . . . . . . . . . . . . . . . . . . . 4
+ 4. Protocol Additions . . . . . . . . . . . . . . . . . . . . . . 4
+ 4.1 Server Considerations . . . . . . . . . . . . . . . . . . 5
+ 4.1.1 Delegations Only . . . . . . . . . . . . . . . . . . . 5
+ 4.1.2 Insecure Delegation Responses . . . . . . . . . . . . 6
+ 4.1.3 Wildcards and Opt-In . . . . . . . . . . . . . . . . . 6
+ 4.1.4 Dynamic Update . . . . . . . . . . . . . . . . . . . . 7
+ 4.2 Client Considerations . . . . . . . . . . . . . . . . . . 7
+ 4.2.1 Delegations Only . . . . . . . . . . . . . . . . . . . 7
+ 4.2.2 Validation Process Changes . . . . . . . . . . . . . . 7
+ 4.2.3 NSEC Record Caching . . . . . . . . . . . . . . . . . 8
+ 4.2.4 Use of the AD bit . . . . . . . . . . . . . . . . . . 8
+ 5. Benefits . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
+ 6. Example . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
+ 7. Transition Issues . . . . . . . . . . . . . . . . . . . . . . 10
+ 8. Security Considerations . . . . . . . . . . . . . . . . . . . 11
+ 9. IANA Considerations . . . . . . . . . . . . . . . . . . . . . 12
+ 10. Acknowledgments . . . . . . . . . . . . . . . . . . . . . . 12
+ 11. References . . . . . . . . . . . . . . . . . . . . . . . . . 13
+ 11.1 Normative References . . . . . . . . . . . . . . . . . . . 13
+ 11.2 Informative References . . . . . . . . . . . . . . . . . . 13
+ Authors' Addresses . . . . . . . . . . . . . . . . . . . . . . 14
+ A. Implementing Opt-In using "Views" . . . . . . . . . . . . . . 14
+ Intellectual Property and Copyright Statements . . . . . . . . 16
+
+
-Arends, et al. Expires August 4, 2005 [Page 2]
+Arends, et al. Expires January 19, 2006 [Page 2]
\f
-Internet-Draft DNSSEC Opt-In February 2005
+Internet-Draft DNSSEC Opt-In July 2005
1. Definitions and Terminology
Throughout this document, familiarity with the DNS system (RFC 1035
[1]), DNS security extensions ([3], [4], and [5], referred to in this
- document as "RFC 2535bis"), and DNSSEC terminology (RFC 3090 [10]) is
- assumed.
+ document as "standard DNSSEC"), and DNSSEC terminology (RFC 3090
+ [10]) is assumed.
The following abbreviations and terms are used in this document:
"SHOULD", "SHOULD NOT", "RECOMMENDED", "MAY, and "OPTIONAL" in this
document are to be interpreted as described in RFC 2119 [7].
-
-
-
-
-
-
-
-
-
-
-
-
-
-Arends, et al. Expires August 4, 2005 [Page 3]
-\f
-Internet-Draft DNSSEC Opt-In February 2005
-
-
2. Overview
The cost to cryptographically secure delegations to unsigned zones is
zones.
This document describes an experimental method of eliminating the
- superfluous cryptography present in secure delegations to unsigned
- zones. Using "Opt-In", a zone administrator can choose to remove
- insecure delegations from the NSEC chain. This is accomplished by
- extending the semantics of the NSEC record by using a redundant bit
- in the type map.
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-Arends, et al. Expires August 4, 2005 [Page 4]
+Arends, et al. Expires January 19, 2006 [Page 3]
\f
-Internet-Draft DNSSEC Opt-In February 2005
+Internet-Draft DNSSEC Opt-In July 2005
+ superfluous cryptography present in secure delegations to unsigned
+ zones. Using "Opt-In", a zone administrator can choose to remove
+ insecure delegations from the NSEC chain. This is accomplished by
+ extending the semantics of the NSEC record by using a redundant bit
+ in the type map.
+
3. Experimental Status
This document describes an EXPERIMENTAL extension to DNSSEC. It
"3.optin.verisignlabs.com": is an alias for DNSSEC algorithm 3, DSA,
and
- "4.optin.verisignlabs.com": is an alias for DNSSEC algorithm 5,
+ "5.optin.verisignlabs.com": is an alias for DNSSEC algorithm 5,
RSASHA1.
Servers wishing to sign and serve zones that utilize Opt-In MUST sign
- the zone with one or more of these private algorithms. This requires
- the signing tools and servers to support private algorithms, as well
- as Opt-In.
+ the zone with only one or more of these private algorithms. This
+ requires the signing tools and servers to support private algorithms,
+ as well as Opt-In.
Resolvers wishing to validate Opt-In zones MUST only do so when the
- zone is signed using one or more of these private algorithms.
+ zone is only signed using one or more of these private algorithms.
The remainder of this document assumes that the servers and resolvers
involved are aware of and are involved in this experiment.
+4. Protocol Additions
+ In DNSSEC, delegation NS RRsets are not signed, but are instead
+ accompanied by a NSEC RRset of the same name and (possibly) a DS
+ record. The security status of the subzone is determined by the
+ presence or absence of the DS RRset, cryptographically proven by the
+ NSEC record. Opt-In expands this definition by allowing insecure
+ delegations to exist within an otherwise signed zone without the
+ corresponding NSEC record at the delegation's owner name. These
+ insecure delegations are proven insecure by using a covering NSEC
+ record.
+ Since this represents a change of the interpretation of NSEC records,
+ resolvers must be able to distinguish between RFC standard DNSSEC
+ NSEC records and Opt-In NSEC records. This is accomplished by
+ "tagging" the NSEC records that cover (or potentially cover) insecure
+ delegation nodes. This tag is indicated by the absence of the NSEC
+ bit in the type map. Since the NSEC bit in the type map merely
+ indicates the existence of the record itself, this bit is redundant
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-Arends, et al. Expires August 4, 2005 [Page 5]
+Arends, et al. Expires January 19, 2006 [Page 4]
\f
-Internet-Draft DNSSEC Opt-In February 2005
-
+Internet-Draft DNSSEC Opt-In July 2005
-4. Protocol Additions
-
- In RFC 2535bis, delegation NS RRsets are not signed, but are instead
- accompanied by a NSEC RRset of the same name and a DS record. The
- security status of the subzone is determined by the presence or
- absence of the DS RRset, cryptographically proven by the NSEC record.
- Opt-In expands this definition by allowing insecure delegations to
- exist within an otherwise signed zone without the corresponding NSEC
- record at the delegation's owner name. These insecure delegations
- are proven insecure by using a covering NSEC record.
- Since this represents a change of the interpretation of NSEC records,
- resolvers must be able to distinguish between RFC 2535bis NSEC
- records and Opt-In NSEC records. This is accomplished by "tagging"
- the NSEC records that cover (or potentially cover) insecure
- delegation nodes. This tag is indicated by the absence of the NSEC
- bit in the type map. Since the NSEC bit in the type map merely
- indicates the existence of the record itself, this bit is redundant
and safe for use as a tag.
An Opt-In tagged NSEC record does not assert the (non)existence of
the existence of the delegation.
Zones using Opt-In MAY contain a mixture of Opt-In tagged NSEC
- records and RFC 2535bis NSEC records. If a NSEC record is not
+ records and standard DNSSEC NSEC records. If a NSEC record is not
Opt-In, there MUST NOT be any insecure delegations (or any other
records) between it and the RRsets indicated by the 'next domain
name' in the NSEC RDATA. If it is Opt-In, there MUST only be
In summary,
- o An Opt-In NSEC type is identified by a zero-valued (or
- not-specified) NSEC bit in the type bit map of the NSEC record.
+ o An Opt-In NSEC type is identified by a zero-valued (or not-
+ specified) NSEC bit in the type bit map of the NSEC record.
o A RFC2535bis NSEC type is identified by a one-valued NSEC bit in
the type bit map of the NSEC record.
and,
-
-
-Arends, et al. Expires August 4, 2005 [Page 6]
-\f
-Internet-Draft DNSSEC Opt-In February 2005
-
-
o An Opt-In NSEC record does not assert the non-existence of a name
between its owner name and "next" name, although it does assert
that any name in this span MUST be an insecure delegation.
between the owner and "next" names of an Opt-In tagged NSEC record.
Signing tools SHOULD NOT generate signed zones that violate this
restriction. Servers SHOULD refuse to load and/or serve zones that
- violate this restriction.
+ violate this restriction. Servers also SHOULD reject AXFR or IXFR
+
+
+
+Arends, et al. Expires January 19, 2006 [Page 5]
+\f
+Internet-Draft DNSSEC Opt-In July 2005
+
+
+ responses that violate this restriction.
4.1.2 Insecure Delegation Responses
When returning an Opt-In insecure delegation, the server MUST return
the covering NSEC RRset in the Authority section.
- In RFC 2535bis, NSEC records already must be returned along with the
- insecure delegation. The primary difference that this proposal
+ In standard DNSSEC, NSEC records already must be returned along with
+ the insecure delegation. The primary difference that this proposal
introduces is that the Opt-In tagged NSEC record will have a
different owner name from the delegation RRset. This may require
- implementations to do a NSEC search on cached responses.
+ implementations to search for the covering NSEC RRset.
4.1.3 Wildcards and Opt-In
- RFC 2535bis describes the practice of returning NSEC records to prove
- the non-existence of an applicable wildcard in non-existent name
- responses. This NSEC record can be described as a "negative wildcard
- proof". The use of Opt-In NSEC records changes the necessity for
- this practice. For non-existent name responses when the query name
- (qname) is covered by an Opt-In tagged NSEC record, servers MAY
- choose to omit the wildcard proof record, and clients MUST NOT treat
- the absence of this NSEC record as a validation error.
+ Standard DNSSEC describes the practice of returning NSEC records to
+ prove the non-existence of an applicable wildcard in non-existent
+ name responses. This NSEC record can be described as a "negative
+ wildcard proof". The use of Opt-In NSEC records changes the
+ necessity for this practice. For non-existent name responses when
+ the query name (qname) is covered by an Opt-In tagged NSEC record,
+ servers MAY choose to omit the wildcard proof record, and clients
+ MUST NOT treat the absence of this NSEC record as a validation error.
- The intent of the RFC 2535bis negative wildcard proof requirement is
- to prevent malicious users from undetectably removing valid wildcard
- responses. In order for this cryptographic proof to work, the
- resolver must be able to prove:
+ The intent of the standard DNSSEC negative wildcard proof requirement
+ is to prevent malicious users from undetectably removing valid
+ wildcard responses. In order for this cryptographic proof to work,
+ the resolver must be able to prove:
1. The exact qname does not exist. This is done by the "normal"
NSEC record.
-
-
-
-
-Arends, et al. Expires August 4, 2005 [Page 7]
-\f
-Internet-Draft DNSSEC Opt-In February 2005
-
-
2. No applicable wildcard exists. This is done by returning a NSEC
record proving that the wildcard does not exist (this is the
negative wildcard proof).
The presence of an Opt-In tagged NSEC record does not change the
practice of returning a NSEC along with a wildcard expansion. Even
+
+
+
+Arends, et al. Expires January 19, 2006 [Page 6]
+\f
+Internet-Draft DNSSEC Opt-In July 2005
+
+
though the Opt-In NSEC will not be able to prove that the wildcard
expansion is valid, it will prove that the wildcard expansion is not
masking any signed records.
add or remove a delegation name from the NSEC chain. This document
does not attempt to define these rules. Until these rules are
defined, servers MUST NOT process DNS Dynamic Update requests against
- zones that use Opt-In NSEC records.
+ zones that use Opt-In NSEC records. Servers SHOULD return responses
+ to update requests with RCODE=REFUSED.
4.2 Client Considerations
This specification does not change the resolver's resolution
algorithm. However, it does change the DNSSEC validation process.
Resolvers MUST be able to use Opt-In tagged NSEC records to
-
-
-
-Arends, et al. Expires August 4, 2005 [Page 8]
-\f
-Internet-Draft DNSSEC Opt-In February 2005
-
-
cryptographically prove the validity and security status (as
insecure) of a referral. Resolvers determine the security status of
the referred-to zone as follows:
- o In RFC 2535bis, the security status is proven by the existence or
- absence of a DS RRset at the same name as the delegation. The
+ o In standard DNSSEC, the security status is proven by the existence
+ or absence of a DS RRset at the same name as the delegation. The
existence of the DS RRset indicates that the referred-to zone is
signed. The absence of the DS RRset is proven using a verified
NSEC record of the same name that does not have the DS bit set in
record does not have the NSEC bit set in the type map, and the
delegation name falls between the NSEC's owner and "next" name.
+
+
+
+Arends, et al. Expires January 19, 2006 [Page 7]
+\f
+Internet-Draft DNSSEC Opt-In July 2005
+
+
Using Opt-In does not substantially change the nature of following
referrals within DNSSEC. At every delegation point, the resolver
- will have cryptographic proof that the subzone is signed or unsigned.
+ will have cryptographic proof that the referred-to subzone is signed
+ or unsigned.
When receiving either an Opt-In insecure delegation response or a
non-existent name response where that name is covered by an Opt-In
Caching resolvers MUST be able to retrieve the appropriate covering
Opt-In NSEC record when returning referrals that need them. This
- requirement differs from RFC 2535bis in that the covering NSEC will
- not have the same owner name as the delegation. Some implementations
- may have to use new methods for finding these NSEC records.
+ requirement differs from standard DNSSEC in that the covering NSEC
+ will not have the same owner name as the delegation. Some
+ implementations may have to use new methods for finding these NSEC
+ records.
4.2.4 Use of the AD bit
The AD bit, as defined by [2] and [5], MUST NOT be set when:
- o sending a non-existent name (NXDOMAIN) response where the covering
- NSEC is tagged as Opt-In.
+ o sending a Name Error (RCODE=3) response where the covering NSEC is
+ tagged as Opt-In.
o sending an Opt-In insecure delegation response, unless the
covering (Opt-In) NSEC record's owner name equals the delegation
name.
or existence of the name. As such, not all data in the response has
been cryptographically verified, so the AD bit cannot be set.
-
-
-Arends, et al. Expires August 4, 2005 [Page 9]
-\f
-Internet-Draft DNSSEC Opt-In February 2005
-
-
5. Benefits
Using Opt-In allows administrators of large and/or changing
for the addition or removal of insecure delegations without modifying
the NSEC record chain. Zones that are frequently updating insecure
delegations (e.g., TLDs) can avoid the substantial overhead of
- modifying and resigning the affected NSEC records.
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-Arends, et al. Expires August 4, 2005 [Page 10]
+Arends, et al. Expires January 19, 2006 [Page 8]
\f
-Internet-Draft DNSSEC Opt-In February 2005
+Internet-Draft DNSSEC Opt-In July 2005
+
+ modifying and resigning the affected NSEC records.
6. Example
EXAMPLE. RRSIG NS ...
EXAMPLE. DNSKEY ...
EXAMPLE. RRSIG DNSKEY ...
- EXAMPLE. NSEC FIRST-SECURE.EXAMPLE. SOA NS RRSIG DNSKEY
+ EXAMPLE. NSEC FIRST-SECURE.EXAMPLE. (
+ SOA NS RRSIG DNSKEY )
EXAMPLE. RRSIG NSEC ...
FIRST-SECURE.EXAMPLE. A ...
NS.UNSIGNED.EXAMPLE. A ...
- In this example, a query for a signed RRset (e.g.,
- "FIRST-SECURE.EXAMPLE A"), or a secure delegation
- ("WWW.SECOND-SECURE.EXAMPLE A") will result in a normal RFC 2535bis
- response.
+ In this example, a query for a signed RRset (e.g., "FIRST-
+ SECURE.EXAMPLE A"), or a secure delegation ("WWW.SECOND-
+ SECURE.EXAMPLE A") will result in a standard DNSSEC response.
A query for a nonexistent RRset will result in a response that
- differs from RFC 2535bis by: the NSEC record will be tagged as
+ differs from standard DNSSEC by: the NSEC record will be tagged as
Opt-In, there may be no NSEC record proving the non-existence of a
- matching wildcard record, and the AD bit will not be set.
-
-Arends, et al. Expires August 4, 2005 [Page 11]
+Arends, et al. Expires January 19, 2006 [Page 9]
\f
-Internet-Draft DNSSEC Opt-In February 2005
+Internet-Draft DNSSEC Opt-In July 2005
+ matching wildcard record, and the AD bit will not be set.
+
A query for an insecure delegation RRset (or a referral) will return
both the answer (in the Authority section) and the corresponding
Opt-In NSEC record to prove that it is not secure.
Additional Section:
NS.UNSIGNED.EXAMPLE. A ...
- In the Example A.1 zone, the EXAMPLE. node MAY use either style of
+ In the Example A.1 zone, the EXAMPLE. node MAY use either style of
NSEC record, because there are no insecure delegations that occur
between it and the next node, FIRST-SECURE.EXAMPLE. In other words,
Example A would still be a valid zone if the NSEC record for EXAMPLE.
EXAMPLE. NSEC FIRST-SECURE.EXAMPLE. (SOA NS
RRSIG DNSKEY NSEC )
- However, the other NSEC records (FIRST-SECURE.EXAMPLE. and
- SECOND-SECURE.EXAMPLE.) MUST be tagged as Opt-In because there are
- insecure delegations in the range they define. (NOT-SECURE.EXAMPLE.
- and UNSIGNED.EXAMPLE., respectively).
-
- NOT-SECURE-2.EXAMPLE. is an example of an insecure delegation that
- is part of the NSEC chain and also covered by an Opt-In tagged NSEC
- record. Because NOT-SECURE-2.EXAMPLE. is a signed name, it cannot
- be removed from the zone without modifying and resigning the prior
- NSEC record. Delegations with names that fall between
- NOT-SECURE-2.EXAMPLE. and SECOND-SECURE.EXAMPLE. may be added or
- removed without resigning any NSEC records.
-
-
-
-
-
-
-
-
-
-
-
-Arends, et al. Expires August 4, 2005 [Page 12]
-\f
-Internet-Draft DNSSEC Opt-In February 2005
+ However, the other NSEC records (FIRST-SECURE.EXAMPLE. and SECOND-
+ SECURE.EXAMPLE.) MUST be tagged as Opt-In because there are insecure
+ delegations in the range they define. (NOT-SECURE.EXAMPLE. and
+ UNSIGNED.EXAMPLE., respectively).
+ NOT-SECURE-2.EXAMPLE. is an example of an insecure delegation that is
+ part of the NSEC chain and also covered by an Opt-In tagged NSEC
+ record. Because NOT-SECURE-2.EXAMPLE. is a signed name, it cannot be
+ removed from the zone without modifying and resigning the prior NSEC
+ record. Delegations with names that fall between NOT-SECURE-
+ 2.EXAMPLE. and SECOND-SECURE.EXAMPLE. may be added or removed without
+ resigning any NSEC records.
7. Transition Issues
- Opt-In is not backwards compatible with RFC 2535bis. RFC 2535bis
- compliant DNSSEC implementations will not recognize Opt-In tagged
- NSEC records as different from RFC 2535bis NSEC records. Because of
- this, RFC 2535bis implementations will reject all Opt-In insecure
- delegations within a zone as invalid.
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
+ Opt-In is not backwards compatible with standard DNSSEC and is
+ considered experimental. Standard DNSSEC compliant implementations
+ would not recognize Opt-In tagged NSEC records as different from
+Arends, et al. Expires January 19, 2006 [Page 10]
+\f
+Internet-Draft DNSSEC Opt-In July 2005
+ standard NSEC records. Because of this, standard DNSSEC
+ implementations, if they were to validate Opt-In style responses,
+ would reject all Opt-In insecure delegations within a zone as
+ invalid. However, by only signing with private algorithms, standard
+ DNSSEC implementations will treat Opt-In responses as unsigned.
-Arends, et al. Expires August 4, 2005 [Page 13]
-\f
-Internet-Draft DNSSEC Opt-In February 2005
-
+ It should be noted that all elements in the resolution path between
+ (and including) the validator and the authoritative name server must
+ be aware of the Opt-In experiment and implement the Opt-In semantics
+ for successful validation to be possible. In particular, this
+ includes any caching middleboxes between the validator and
+ authoritative name server.
8. Security Considerations
o Records with unsigned names (whether existing or not) suffer from
the same vulnerabilities as records in an unsigned zone. These
- vulnerabilites are described in more detail in [12] (note in
+ vulnerabilities are described in more detail in [12] (note in
particular sections 2.3, "Name Games" and 2.6, "Authenticated
Denial").
o Records with signed names have the same security whether or not
-
-
-
-
-
-
-
-
-
-
-
-
-
-Arends, et al. Expires August 4, 2005 [Page 14]
+Arends, et al. Expires January 19, 2006 [Page 11]
\f
-Internet-Draft DNSSEC Opt-In February 2005
+Internet-Draft DNSSEC Opt-In July 2005
Example S.1: Response to query for WWW.DOES-NOT-EXIST.EXAMPLE. A
The resolver would have no choice but to believe that the referral to
- NS.FORGED. is valid. If a wildcard existed that would have been
+ NS.FORGED. is valid. If a wildcard existed that would have been
expanded to cover "WWW.DOES-NOT-EXIST.EXAMPLE.", an attacker could
have undetectably removed it and replaced it with the forged
delegation.
In particular, zone signing tools SHOULD NOT default to Opt-In, and
MAY choose to not support Opt-In at all.
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-Arends, et al. Expires August 4, 2005 [Page 15]
-\f
-Internet-Draft DNSSEC Opt-In February 2005
-
-
9. IANA Considerations
None.
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-Arends, et al. Expires August 4, 2005 [Page 16]
-\f
-Internet-Draft DNSSEC Opt-In February 2005
-
-
10. Acknowledgments
The contributions, suggestions and remarks of the following persons
Dan Massey, Scott Rose, Mike Schiraldi, Jakob Schlyter, Brian
Wellington.
+11. References
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-Arends, et al. Expires August 4, 2005 [Page 17]
+Arends, et al. Expires January 19, 2006 [Page 12]
\f
-Internet-Draft DNSSEC Opt-In February 2005
-
+Internet-Draft DNSSEC Opt-In July 2005
-11. References
11.1 Normative References
[2] Wellington, B. and O. Gudmundsson, "Redefinition of DNS
Authenticated Data (AD) bit", RFC 3655, November 2003.
- [3] Arends, R., Austein, R., Massey, D., Larson, M. and S. Rose,
- "DNS Security Introduction and Requirements",
- draft-ietf-dnsext-dnssec-intro-13 (work in progress), October
- 2004.
+ [3] Arends, R., Austein, R., Larson, M., Massey, D., and S. Rose,
+ "DNS Security Introduction and Requirements", RFC 4033,
+ March 2005.
- [4] Arends, R., "Resource Records for the DNS Security Extensions",
- draft-ietf-dnsext-dnssec-records-11 (work in progress), October
- 2004.
+ [4] Arends, R., Austein, R., Larson, M., Massey, D., and S. Rose,
+ "Resource Records for the DNS Security Extensions", RFC 4034,
+ March 2005.
- [5] Arends, R., "Protocol Modifications for the DNS Security
- Extensions", draft-ietf-dnsext-dnssec-protocol-09 (work in
- progress), October 2004.
+ [5] Arends, R., Austein, R., Larson, M., Massey, D., and S. Rose,
+ "Protocol Modifications for the DNS Security Extensions",
+ RFC 4035, March 2005.
[6] Blacka, D., "DNSSEC Experiments",
- draft-blacka-dnssec-experiments-00 (work in progress), December
- 2004.
+ draft-ietf-dnsext-dnssec-experiments-01 (work in progress),
+ July 2005.
11.2 Informative References
[8] Elz, R. and R. Bush, "Clarifications to the DNS Specification",
RFC 2181, July 1997.
- [9] Eastlake, D., "Secure Domain Name System Dynamic Update", RFC
- 2137, April 1997.
+ [9] Eastlake, D., "Secure Domain Name System Dynamic Update",
+ RFC 2137, April 1997.
[10] Lewis, E., "DNS Security Extension Clarification on Zone
Status", RFC 3090, March 2001.
-Arends, et al. Expires August 4, 2005 [Page 18]
+
+
+
+Arends, et al. Expires January 19, 2006 [Page 13]
\f
-Internet-Draft DNSSEC Opt-In February 2005
+Internet-Draft DNSSEC Opt-In July 2005
Authors' Addresses
7522 NB Enschede
NL
- EMail: roy.arends@telin.nl
+ Email: roy.arends@telin.nl
Mark Kosters
US
Phone: +1 703 948 3200
- EMail: markk@verisign.com
+ Email: markk@verisign.com
URI: http://www.verisignlabs.com
US
Phone: +1 703 948 3200
- EMail: davidb@verisign.com
+ Email: davidb@verisign.com
URI: http://www.verisignlabs.com
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-Arends, et al. Expires August 4, 2005 [Page 19]
-\f
-Internet-Draft DNSSEC Opt-In February 2005
-
-
Appendix A. Implementing Opt-In using "Views"
In many cases, it may be convenient to implement an Opt-In zone by
In addition, the only RRsets that may solely exist in the insecure
view are non-zone-apex NS RRsets. That is, all non-NS RRsets (and
+
+
+
+Arends, et al. Expires January 19, 2006 [Page 14]
+\f
+Internet-Draft DNSSEC Opt-In July 2005
+
+
the zone apex NS RRset) MUST be signed and in the secure view.
These two views may be combined at request time to provide a virtual,
-Arends, et al. Expires August 4, 2005 [Page 20]
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+Arends, et al. Expires January 19, 2006 [Page 15]
\f
-Internet-Draft DNSSEC Opt-In February 2005
+Internet-Draft DNSSEC Opt-In July 2005
Intellectual Property Statement
-Arends, et al. Expires August 4, 2005 [Page 21]
+Arends, et al. Expires January 19, 2006 [Page 16]
\f
-
projects / thirdparty / bind9.git / commitdiff
