Internet-Draft DNS Multiple QTYPEs February 2023
Bellis Expires 20 August 2023 [Page]
Workgroup:
DNSEXT Working Group
Internet-Draft:
draft-bellis-dnsext-multi-qtypes-07
Published:
Intended Status:
Standards Track
Expires:
Author:
R. Bellis
ISC

DNS Multiple QTYPEs

Abstract

This document specifies a method for a DNS client to request additional DNS record types to be delivered alongside the primary record type specified in the question section of a DNS query.

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 https://datatracker.ietf.org/drafts/current/.

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 20 August 2023.

Table of Contents

1. Introduction

RFC EDITOR: PLEASE REMOVE THE FOLLOWING PARAGRAPH BEFORE PUBLISHING: The source for this draft is maintained in GitHub at: https://github.com/raybellis/draft-bellis-dnsext-multi-qtypes Please submit suggested changes as issues or pull requests there.

A commonly requested DNS [RFC1035] feature is the ability to receive multiple related resource records (RRs) in a single DNS response.

For example, it may be desirable to receive both the A and AAAA records for a domain name together, rather than having to issue multiple queries.

The DNS wire protocol in theory supports having multiple questions in a single packet, but in practise this does not work:

To mitigate these issues, this document constrains the problem to those cases where only the QTYPE varies by specifying a new option for the Extension Mechanisms for DNS (EDNS [RFC6891]) that contains an additional list of QTYPE values that the client wishes to receive in addition to the single QTYPE appearing in the question section.

TODO: why not "ANY" ?

2. Terminology used in this document

The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", "SHOULD", "SHOULD NOT", "RECOMMENDED", "NOT RECOMMENDED", "MAY", and "OPTIONAL" in this document are to be interpreted as described in BCP 14 [RFC2119] [RFC8174] when, and only when, they appear in all capitals, as shown here.

3. Description

3.1. Multiple QTYPE EDNS Option Format

The overall format of an EDNS option is shown for reference below, per [RFC6891], followed by the option specific data:

   +---+---+---+---+---+---+---+---+---+---+---+---+---+---+---+---+
0: |                          OPTION-CODE                          |
   +---+---+---+---+---+---+---+---+---+---+---+---+---+---+---+---+
2: |                         OPTION-LENGTH                         |
   +---+---+---+---+---+---+---+---+---+---+---+---+---+---+---+---+
4: |                                                               |
   /                          OPTION-DATA                          /
   /                                                               /
   +---+---+---+---+---+---+---+---+---+---+---+---+---+---+---+---+

OPTION-CODE: TBD by IANA

OPTION-LENGTH: Size (in octets) of OPTION-DATA.

OPTION-DATA: Option specific, as below:

                +0 (MSB)                            +1 (LSB)
   +---+---+---+---+---+---+---+---+---+---+---+---+---+---+---+---+
0: |QTD|   reserved    |  QTCOUNT  |           QT1 (MSB)           |
   +---+---+---+---+---+---+---+---+---+---+---+---+---+---+---+---+
2: |           QT1 (LSB)           |              ...              |
   +---+---+---+---+---+---+---+---+---+---+---+---+---+---+---+---+
   |              ...             ///          QTn (MSB)           |
   +---+---+---+---+---+---+---+---+---+---+---+---+---+---+---+---+
   |           QTn (LSB)           |
   +---+---+---+---+---+---+---+---+

QTD: this bit indicates the direction of the packet. It MUST be clear (0) in a request and set (1) in a response.

QTCOUNT: a 3 bit field with range 0 .. 7 specifying the number of QT fields to follow. NB: Whilst the QTCOUNT could in theory be calculated based on the OPTION-LENGTH field, having it explicitly specified ensures a sensible constraint on its range of values.

QTn: a 2 byte field (MSB first) specifying a DNS RR type. The RR type MUST be for a real resource record, and MUST NOT refer to a pseudo RR type such as "OPT", "IXFR", "TSIG", "*", etc.

3.2. Response Generation

3.2.1. Server Side Processing

A conforming server that receives a Multiple QTYPE Option in a query MUST return a Multiple QTYPE Option in its response.

The QTD bit in that response MUST be set (1) as protection against servers which simply echo unknown EDNS options verbatim. If the QTD bit in a response is zero the client MUST treat the response as if this option is unsupported.

The server SHOULD attempt to return any resource records known to it that match the additional (QNAME, QTn, QCLASS) tuples. These records MUST be returned in the Answer Section of the response, but the answer for the primary QTYPE from the Question Section MUST be included first.

For any particular QTn in the query, if the server provides additional answers, or has knowledge that the RR type type does not exist for that QNAME (a "negative answer"), it must include that QTn value in the Multiple QTYPE Option of its response.

A negative answer is therefore indicated by the combination of the presence of a QTn value in the Multiple QTYPE Option and the absence of a matching record in the Answer Section. This is necessary (in the absence of DNSSEC) to differentiate between absence of the record from the zone and absence of the record from the response.

A server that is authoritative for the specified QNAME on receipt of a Multiple QTYPE Option MUST attempt to return all specified RR types except where that would result in truncation in which case it may omit some (or all) of the records for the additional RR types. Those RR types MUST then also be omitted from the Multiple QTYPE Option in the response.

A caching recursive server receiving a Multiple QTYPE Option SHOULD attempt to fill its positive and negative caches with all of the specified RR types before returning its response to the client.

TODO: is there a case for mandatory answers, i.e. the client saying I really want all these?

3.2.2. Client Side Processing

Recursive resolvers MAY use this method to obtain multiple records from an authoritative server. For the purposes of Section 5.4.1 of [RFC2181] any authoritative answers received MUST be ranked the same as the answer for the primary question.

3.2.3. DNSSEC

If the DNS client sets the "DNSSEC OK" (DO) bit in the query then the server MUST also return the related DNSSEC records that would have been returned in a standalone query for the same QTYPE.

A negative answer from a signed zone MUST contain the appropriate authenticated denial of existence records, per [RFC4034] and [RFC5155].

In a signed zone there is a theoretical risk of valid signatures for one RR type and invalid signatures for another. This is the only case known to the author where the response code for any particular QNAME may be inconsistent across different RR types.

Should a validating resolver produce NOERROR for some RR types and SERVFAIL for others it MUST omit the RR types that failed to validate from its response and from the QTn fields on the Multiple QTYPE option. The client MAY then initiate standalone queries for those RR types.

4. Security Considerations

The method documented here does not change any of the security properties of the DNS protocol itself.

It should however be noted that this method does increase the potential amplification factor when the DNS protocol is used as a vector for a denial of service attack.

5. IANA Considerations

IANA is requested to assign a new value in the DNS EDNS0 Option Codes registry.

6. Acknowledgements

The author wishes to thank the following for their feedback and reviews during the initial development of this document: Michael Graff, Olafur Gudmundsson, Matthijs Mekking, Paul Vixie.

7. References

7.1. Normative References

[RFC1035]
Mockapetris, P., "Domain names - implementation and specification", STD 13, RFC 1035, DOI 10.17487/RFC1035, , <https://www.rfc-editor.org/info/rfc1035>.
[RFC2119]
Bradner, S., "Key words for use in RFCs to Indicate Requirement Levels", BCP 14, RFC 2119, DOI 10.17487/RFC2119, , <https://www.rfc-editor.org/info/rfc2119>.
[RFC2181]
Elz, R. and R. Bush, "Clarifications to the DNS Specification", RFC 2181, DOI 10.17487/RFC2181, , <https://www.rfc-editor.org/info/rfc2181>.
[RFC4034]
Arends, R., Austein, R., Larson, M., Massey, D., and S. Rose, "Resource Records for the DNS Security Extensions", RFC 4034, DOI 10.17487/RFC4034, , <https://www.rfc-editor.org/rfc/rfc4034>.
[RFC5155]
Laurie, B., Sisson, G., Arends, R., and D. Blacka, "DNS Security (DNSSEC) Hashed Authenticated Denial of Existence", RFC 5155, DOI 10.17487/RFC5155, , <https://www.rfc-editor.org/info/rfc5155>.
[RFC6891]
Damas, J., Graff, M., and P. Vixie, "Extension Mechanisms for DNS (EDNS(0))", STD 75, RFC 6891, DOI 10.17487/RFC6891, , <https://www.rfc-editor.org/rfc/rfc6891>.
[RFC8174]
Leiba, B., "Ambiguity of Uppercase vs Lowercase in RFC 2119 Key Words", BCP 14, RFC 8174, DOI 10.17487/RFC8174, , <https://www.rfc-editor.org/info/rfc8174>.

7.2. Informative References

[RFC3425]
Lawrence, D., "Obsoleting IQUERY", RFC 3425, DOI 10.17487/RFC3425, , <https://www.rfc-editor.org/info/rfc3425>.

Author's Address

Ray Bellis
Internet Systems Consortium, Inc.
PO Box 360
Redwood City, NH 03857
United States of America