Network Working Group D. Crocker
Internet-Draft Brandenburg InternetWorking
Intended status: Best Current Practice March 30, 2011
Expires: October 01, 2011

DNS Scoped Data Through Attribute Leaves


Historically, any DNS RR may occur for any domain name. Recent additions have defined DNS leaf nodes that contain a reserved node name, beginning with an underscore. The underscore construct is used to define a semantic scope for DNS records associated with the parent domain. This note explores the nature of this DNS usage and defines the "underscore names" registry with IANA.

Status of this Memo

This Internet-Draft is submitted in full conformance with the provisions of BCP 78 and BCP 79.

Internet-Drafts are working documents of the Internet Engineering Task Force (IETF). Note that other groups may also distribute working documents as Internet-Drafts. The list of current Internet- Drafts is at

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."

This Internet-Draft will expire on October 01, 2011.

Copyright Notice

Copyright (c) 2011 IETF Trust and the persons identified as the document authors. All rights reserved.

This document is subject to BCP 78 and the IETF Trust's Legal Provisions Relating to IETF Documents ( in effect on the date of publication of this document. Please review these documents carefully, as they describe your rights and restrictions with respect to this document. Code Components extracted from this document must include Simplified BSD License text as described in Section 4.e of the Trust Legal Provisions and are provided without warranty as described in the Simplified BSD License.

Table of Contents

1. Introduction

The core DNS technical specifications assign no semantics to domain names or their parts, and no constraints upon which resource records (RRs) may be associated with particular names. Over time, some leaf node names, such as "www" and "ftp" have come to imply support for particular services, but this is a matter of operational convention, rather than defined protocol semantics. This freedom in the basic technology has permitted a wide range of administrative and semantic policies to be used -- in parallel -- with the DNS. Data semantics have been limited to the specification of particular resource records, on the expectation that new ones would be added as needed.

Some recent service enhancements have defined define a restricted scope for the occurrence of particular resource records. That scope is a leaf node, within which the uses of specific resource records can be formally defined and constrained. This leaf has a distinguished naming convention: It uses a reserved DNS node name that begins with an underscore. Because host names are not allowed to use the underscore character, this distinguishes the name from all legal host name. Effectively, this convention creates a space for attributes that are associated with the parent domain, one level up.

An established example is the SRV record [RFC2782] which generalizes concepts long-used for email routing by the MX record [RFC0974][RFC2821]. The use of special DNS names has significant benefits and detriments. Some of these are explored in [RFC5507].

The terms "resolution context" and "scoping rules" have been suggested, in place of "semantic scope". In order to avoid concern for matters of semantics, this specification uses the term "scoping rules", to create a focus on the mechanics being defined, rather than nuances of interpretation for the mechanism.

The scoping feature is particularly useful when generalized resource records are used -- notably TXT and SRV. It provides efficient separation of one use of them from another. Absent this separation, an undifferentiated mass of these RRs are returned to the client which then must parse through the internals of the records in the hope of finding ones that are relevant. With underscore-based scoping, only the relevant RRs are returns.

This specification discusses this enhancement, provides an explicit definition of it, and establishes an IANA registry for the reserved names that begin with underscore.

Discussion Venue:
Discussion about this draft is directed to the dnsop@lists.uoregon.edumailing list of the IETF DNSOP Working Group.

2. Scaling Benefits and TXT and SRV Resource Records

Some resource records have a generic form, and support a variety of uses. Each additional use defines its own rules and, possibly, its own internal syntax and node-naming conventions to distinguish among particular types. The TXT and SRV records are the notable concern for this. Some of these approaches scale poorly, particularly when the same RR can be present in the same leaf node, but with different uses. An increasingly-popular approach, with excellent scaling properties, uses an underscore-based name to a define place in the DNS that is constrained to particular uses for particular RRs. This means that a direct lookup produces only the desired records, at no greater cost than a typical lookup.

In the case of TXT records, use for different scoping rules has developed organically and largely without coordination. One side-effect of this is no consistently distinguishable internal syntax for the records; even internal inspection might not be a reliable means of distinguishing among them. Underscore-based names therefore provide an administrative way of separating TXT records that might have different uses, but otherwise would have no syntactic markers for distinguishing among them.

In the case of the SRV RR this method of distinguishing among uses was part of the design. [RFC2782] In reality, the SRV specification defines an RR that may only be used for specific applications when there is an additional specification. So the SRV specification is best thought of as a template for future specifications. The template definition includes reference to tables of names from which underscore-names should be drawn. So, the set of <service> names is defined in terms of other IANA tables, namely any table with symbolic names. The other SRV naming field is <proto>, although its pool of names is not explicitly defined.

3. Underscore DNS Registry Function

This specification defines a registry for DNS nodes names, used to specify scope of use for specific resource records (RR). That is, a given names defines a specific, constrained context for the use of such records. This does not constrain the use of other resource records that are not specified. The purpose of the registries is to avoid collisions resulting from the use of the same underscore name, for different applications.

Structurally, the registry is defined as a single, flat table of names that begin with underscore. In some cases, such as for SRV, an underscore names might have further constraints, such as being valid only "under" some other underscore name. Semantically, this is a hierarchical model, thereby making a flat registry unexpected.

The registry requires such hierarchies to be registered as a combinatorial case analysis set, with each entry being a full sequence of underscore names. Given a naming scheme that is actually structured, this flat design is inelegant. However it has the benefit of being extremely simple, with the added advantage of being easier for readers to understand, as long as these cases are small and few.

Example of Underscore Names

The reasons for choosing a simplified registry design are:

4. DNS Underscore Registry Definition

A registry entry MUST contain:

5. IANA Considerations

IANA is requested to establish a DNS Underscore Name Registry, for DNS node names that begin with the underscore character (_) and have been specified in any published RFC, or are documented by a specification published by another standards organization. The contents of each entry are defined in Section 3.

DNS Underscore SCOPE Name Registry (with initial values)
SIP TCP _sip._tcp NAPTR [RFC3263] Locating SIP Servers
SIPS TCP _sips._tcp NAPTR [RFC3263] Locating SIP Servers
SIP UDP _sip._udp SRV [RFC3263] Locating SIP servers.
SPF _spf TXT [RFC4408]
DKIM _domainkey TXT [RFC4871] Public key for verifying DKIM signature.
ADSP _adsp._domainkey TXT [RFC5617]
PKI LDAP _PKIXREP._ldap SRV [RFC4386] LDAP PKI Repository
PKI HTTP _PKIXREP._http SRV [RFC4386] HTTP PKI Repository
PKI OCSP _PKIXREP._ocsp SRV [RFC4386] OCSP PKI Repository

6. Security Considerations

This memo raises no security issues.

7. References

7.1. References -- Normative

[RFC2782] Gulbrandsen, A., Vixie, P. and L. Esibov, "A DNS RR for specifying the location of services (DNS SRV)", RFC 2782, February 2000.
[RFC3263] Rosenberg, J. and H. Schulzrinne, "Session Initiation Protocol (SIP): Locating SIP Servers", RFC 3263, June 2002.
[RFC4408] Wong, M. and W. Schlitt, "Sender Policy Framework (SPF) for Authorizing Use of Domains in E-Mail, Version 1", RFC 4408, April 2006.

7.2. References -- Informative

[RFC2821] Klensin, J., "Simple Mail Transfer Protocol", RFC 2821, April 2001.
[RFC0974] Partridge, C., "Mail routing and the domain system", RFC 974, January 1986.
[RFC4871] Allman, E., Callas, J., Delany, M., Libbey, M., Fenton, J. and M. Thomas, "DomainKeys Identified Mail (DKIM) Signatures", RFC 4871, May 2007.
[RFC5507] Faltstrom, P. and R. Austein, "", RFC 5507, April 2009.
[RFC4386] Boeyen, S. and P. Hallam-Baker, "Internet X.509 Public Key Infrastructure: Repository Locator Service", February 2006.
[RFC5617] , , , , , and , "DomainKeys Identified Mail (DKIM) Author Domain Signing Practices (ADSP)", August 2009.

Appendix A. Acknowledgements

Thanks go to Bill Fenner, Tony Hansen, Peter Koch, Olaf Kolkman, and Andrew Sullivan for diligent review of the earlier drafts.

Author's Address

Dave Crocker Brandenburg InternetWorking 675 Spruce Dr. Sunnyvale, CA 94086 USA Phone: +1.408.246.8253 EMail: URI: