Minimal ANY: Providing Minimal-Sized Responses to DNS Queries That Have QTYPE=ANY
How a DNS server can answer an ANY query with one small record instead of everything it knows.
Abley, J., Gudmundsson, O., Majkowski, M., and E. Hunt, "Providing Minimal-Sized Responses to DNS Queries That Have QTYPE=ANY", RFC 8482, DOI 10.17487/RFC8482, January 2019, <https://www.rfc-editor.org/info/rfc8482>.
@misc{rfc8482,
series = {Request for Comments},
number = 8482,
howpublished = {RFC 8482},
publisher = {RFC Editor},
doi = {10.17487/RFC8482},
url = {https://www.rfc-editor.org/info/rfc8482},
author = {J. Abley and O. Gudmundsson and M. Majkowski and E. Hunt},
title = {{Providing Minimal-Sized Responses to DNS Queries That Have QTYPE=ANY}},
pagetotal = 10,
year = 2019,
month = jan,
} RFC text
Contents
Internet Engineering Task Force (IETF) J. Abley
Request for Comments: 8482 Afilias
Updates: 1034, 1035 O. Gudmundsson
Category: Standards Track M. Majkowski
ISSN: 2070-1721 Cloudflare Inc.
E. Hunt
ISC
January 2019
Providing Minimal-Sized Responses to DNS Queries That Have QTYPE=ANY
Abstract
The Domain Name System (DNS) specifies a query type (QTYPE) "ANY".
The operator of an authoritative DNS server might choose not to
respond to such queries for reasons of local policy, motivated by
security, performance, or other reasons.
The DNS specification does not include specific guidance for the
behavior of DNS servers or clients in this situation. This document
aims to provide such guidance.
This document updates RFCs 1034 and 1035.
Status of This Memo
This is an Internet Standards Track document.
This document is a product of the Internet Engineering Task Force
(IETF). It represents the consensus of the IETF community. It has
received public review and has been approved for publication by the
Internet Engineering Steering Group (IESG). Further information on
Internet Standards is available in Section 2 of RFC 7841.
Information about the current status of this document, any errata,
and how to provide feedback on it may be obtained at
https://www.rfc-editor.org/info/rfc8482.
Copyright Notice
Copyright (c) 2019 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
(https://trustee.ietf.org/license-info) 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 ....................................................3
1.1. Terminology ................................................3
2. Motivations for Use of ANY Queries ..............................3
3. General Approach ................................................4
4. Behavior of DNS Responders ......................................5
4.1. Answer with a Subset of Available RRsets ...................5
4.2. Answer with a Synthesized HINFO RRset ......................5
4.3. Answer with Best Guess as to Intention .....................6
4.4. Transport Considerations ...................................6
5. Behavior of DNS Initiators ......................................7
6. HINFO Considerations ............................................7
7. Updates to RFCs 1034 and 1035 ...................................7
8. Implementation Experience .......................................8
9. Security Considerations .........................................8
10. IANA Considerations ............................................9
11. References .....................................................9
11.1. Normative References ......................................9
11.2. Informative References ....................................9
Acknowledgements ..................................................10
Authors' Addresses ................................................10
1. Introduction
The Domain Name System (DNS) specifies a query type (QTYPE) "ANY".
The operator of an authoritative DNS server might choose not to
respond to such queries for reasons of local policy, motivated by
security, performance, or other reasons.
The DNS specification [RFC1034] [RFC1035] does not include specific
guidance for the behavior of DNS servers or clients in this
situation. This document aims to provide such guidance.
1.1. Terminology
This document uses terminology specific to the Domain Name System
(DNS), descriptions of which can be found in [RFC8499].
[RFC1035] defined type 255 to be "*". However, DNS implementations
commonly use the keyword "ANY" to refer to that type code; this
document follows that common usage.
In this document, "ANY query" refers to a DNS meta-query with
QTYPE=ANY. An "ANY response" is a response to such a query.
In this document, "conventional ANY response" means an ANY response
that is constructed in accordance with the algorithm documented in
Section 4.3.2 of [RFC1034] and specifically without implementing any
of the mechanisms described in this document.
In an exchange of DNS messages between two hosts, this document
refers to the host sending a DNS request as the "initiator" and the
host sending a DNS response as the "responder".
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.
2. Motivations for Use of ANY Queries
ANY queries are legitimately used for debugging and checking the
state of a DNS server for a particular name.
ANY queries are sometimes used as an attempt to reduce the number of
queries needed to get information, e.g., to obtain MX, A, and AAAA
resource record sets (RRsets) for a mail domain in a single query.
However, there is no documented guidance available for this use case,
and some implementations have been observed not to function as their
developers expected. If implementers assume that an ANY query will
ultimately be received by an authoritative server and will fetch all
existing RRsets, they should include a fallback mechanism to use when
that does not happen.
ANY queries are frequently used to exploit the amplification
potential of DNS servers and resolvers using spoofed source addresses
and UDP transport (see [RFC5358]). Having the ability to return
small responses to such queries makes DNS servers less attractive
amplifiers.
ANY queries are sometimes used to help mine authoritative-only DNS
servers for zone data, since they are expected to return all RRsets
for a particular query name. If DNS operators prefer to reduce the
potential for information leaks, they might choose not to send large
ANY responses.
Some authoritative-only DNS server implementations require additional
processing in order to send a conventional ANY response; avoiding
that processing expense might be desirable.
3. General Approach
This proposal provides a mechanism for an authoritative DNS server to
signal that conventional ANY queries are not supported for a
particular QNAME. It does so in a way that is both compatible with
and triggers desirable behavior by unmodified clients (e.g., DNS
resolvers).
Alternative proposals for dealing with ANY queries have been
discussed. One approach proposes using a new RCODE to signal that an
authoritative server did not answer ANY queries in the standard way.
This approach was found to have an undesirable effect on both
resolvers and authoritative-only servers; resolvers receiving an
unknown RCODE would resend the same query to all available
authoritative servers rather than suppress future ANY queries for the
same QNAME.
The proposal described in this document avoids that outcome by
returning a non-empty RRset in the ANY response, which provides
resolvers with something to cache and effectively suppresses repeat
queries to the same or different authoritative DNS servers.
4. Behavior of DNS Responders
Below are the three different modes of behavior by DNS responders
when processing queries with QNAMEs that exist, QCLASS=IN, and
QTYPE=ANY. Operators and implementers are free to choose whichever
mechanism best suits their environment.
1. A DNS responder can choose to select one or a larger subset of
the available RRsets at the QNAME.
2. A DNS responder can return a synthesized HINFO resource record.
See Section 6 for discussion of the use of HINFO.
3. A resolver can try to give out the most likely records the
requester wants. This is not always possible, and the result
might well be a large response.
Except as described below in this section, the DNS responder MUST
follow the standard algorithms when constructing a response.
4.1. Answer with a Subset of Available RRsets
A DNS responder that receives an ANY query MAY decline to provide a
conventional ANY response or MAY instead send a response with a
single RRset (or a larger subset of available RRsets) in the answer
section.
The RRsets returned in the answer section of the response MAY consist
of a single RRset owned by the name specified in the QNAME. Where
multiple RRsets exist, the responder SHOULD choose a small subset of
those available to reduce the amplification potential of the
response.
If the zone is signed, appropriate RRSIG records MUST be included in
the answer.
Note that this mechanism does not provide any signaling to indicate
to a client that an incomplete subset of the available RRsets has
been returned.
4.2. Answer with a Synthesized HINFO RRset
If there is no CNAME present at the owner name matching the QNAME,
the resource record returned in the response MAY instead be
synthesized. In this case, a single HINFO resource record SHOULD be
returned. The CPU field of the HINFO RDATA SHOULD be set to
"RFC8482". The OS field of the HINFO RDATA SHOULD be set to the null
string to minimize the size of the response.
The TTL encoded for the synthesized HINFO resource record SHOULD be
chosen by the operator of the DNS responder to be large enough to
suppress frequent subsequent ANY queries from the same initiator with
the same QNAME, understanding that a TTL that is too long might make
policy changes relating to ANY queries difficult to change in the
future. The specific value used SHOULD be configurable by the
operator of the nameserver according to local policy, based on the
familiar considerations involved in choosing a TTL value for any
resource record in any zone.
If the DNS query includes DO=1 and the QNAME corresponds to a zone
that is known by the responder to be signed, a valid RRSIG for the
RRsets in the answer (or authority if answer is empty) section MUST
be returned. In the case of DO=0, the RRSIG SHOULD be omitted.
A system that receives an HINFO response SHOULD NOT infer that the
response was generated according to this specification and apply any
special processing of the response because, in general, it is not
possible to tell with certainty whether the HINFO RRset received was
synthesized. In particular, systems SHOULD NOT rely upon the HINFO
RDATA described in this section to distinguish between synthesized
and non-synthesized HINFO RRsets.
4.3. Answer with Best Guess as to Intention
In some cases, it is possible to guess what the initiator wants in
the answer (but not always). Some implementations have implemented
the spirit of this document by returning all RRsets of RRTYPE CNAME,
MX, A, and AAAA that are present at the owner name while suppressing
others. This heuristic seems to work well in practice; it satisfies
the needs of some applications whilst suppressing other RRsets such
as TXT and DNSKEY that can often contribute to large responses.
Whilst some applications may be satisfied by this behavior, the
resulting responses in the general case are larger than in the
approaches described in Sections 4.1 and 4.2.
As before, if the zone is signed and the DO bit is set on the
corresponding query, an RRSIG RRset MUST be included in the response.
4.4. Transport Considerations
A DNS responder MAY behave differently when processing ANY queries
received over different transports, e.g., by providing a conventional
ANY response over TCP whilst using one of the other mechanisms
specified in this document in the case where a query was received
using UDP.
Implementers MAY provide configuration options to allow operators to
specify different behavior over different transports.
5. Behavior of DNS Initiators
A DNS initiator that sends a query with QTYPE=ANY and receives a
response containing an HINFO resource record or a single RRset, as
described in Section 4, MAY cache the response in the normal way.
Such cached resource records SHOULD be retained in the cache
following normal caching semantics, as with any other response
received from a DNS responder.
A DNS initiator MAY suppress queries with QTYPE=ANY in the event that
the local cache contains a matching HINFO resource record with the
CPU field of the HINFO RDATA, as described in Section 4. A DNS
initiator MAY instead respond to such queries with the contents of
the local cache in the usual way.
6. HINFO Considerations
It is possible that the synthesized HINFO RRset in an ANY response,
once cached by the initiator, might suppress subsequent queries from
the same initiator with QTYPE=HINFO. Thus, the use of HINFO in this
proposal would effectively mask the HINFO RRset present in the zone.
Operators of authoritative servers who serve zones that rely upon
conventional use of the HINFO RRTYPE SHOULD sensibly choose the
"single RRset" method described in this document or select another
type.
The HINFO RRTYPE is believed to be rarely used in the DNS at the time
of writing, based on observations made in passive DNS and at
recursive and authoritative DNS servers.
7. Updates to RFCs 1034 and 1035
This document extends the specification for processing ANY queries
described in Section 4.3.2 of [RFC1034].
It is important to note that returning a subset of available RRsets
when processing an ANY query is legitimate and consistent with
[RFC1035]; it can be argued that ANY does not always mean ALL, as
used in Section 3.2.3 of [RFC1035]. The main difference here is that
the TC bit SHOULD NOT be set in the response, thus indicating that
this is not a complete answer.
This document describes optional behavior for both DNS initiators and
responders; implementation of the guidance provided by this document
is OPTIONAL.
RRSIG queries (i.e., queries with QTYPE=RRSIG) are similar to ANY
queries in the sense that they have the potential to generate large
responses as well as extra work for the responders that process them,
e.g., in the case where signatures are generated on the fly. RRSIG
RRsets are not usually obtained using such explicit queries but are
rather included in the responses for other RRsets that the RRSIGs
cover. This document does not specify appropriate behavior for RRSIG
queries; however, future such advice might well benefit from
consistency with and experience with the approaches for ANY queries
described here.
8. Implementation Experience
In October 2015, the Cloudflare authoritative nameserver
implementation implemented the HINFO response. A few minor problems
were reported and have since been resolved.
An implementation of the subset-mode response to ANY queries was
implemented in NSD 4.1 in 2016.
An implementation of a single RRset response to an ANY query was made
for BIND9 by Tony Finch, and that functionality was subsequently made
available in production releases starting in BIND 9.11.
9. Security Considerations
Queries with QTYPE=ANY are frequently observed as part of reflection
attacks, since a relatively small query can be used to elicit a large
response. This is a desirable characteristic if the goal is to
maximize the amplification potential of a DNS server as part of a
volumetric attack. The ability of a DNS operator to suppress such
responses on a particular server makes that server a less useful
amplifier.
The optional behavior described in this document to reduce the size
of responses to queries with QTYPE=ANY is compatible with the use of
DNSSEC by both initiator and responder.
10. IANA Considerations
IANA has updated the following entry in the "Resource Record (RR)
TYPEs" registry [RR_TYPES]:
+------+-------+-------------------------------+--------------------+
| TYPE | Value | Meaning | Reference |
+------+-------+-------------------------------+--------------------+
| * | 255 | A request for some or all | [RFC1035][RFC6895] |
| | | records the server has | [RFC8482] |
| | | available | |
+------+-------+-------------------------------+--------------------+
11. References
11.1. Normative References
[RFC1034] Mockapetris, P., "Domain names - concepts and facilities",
STD 13, RFC 1034, DOI 10.17487/RFC1034, November 1987,
<https://www.rfc-editor.org/info/rfc1034>.
[RFC1035] Mockapetris, P., "Domain names - implementation and
specification", STD 13, RFC 1035, DOI 10.17487/RFC1035,
November 1987, <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, March 1997,
<https://www.rfc-editor.org/info/rfc2119>.
[RFC8174] Leiba, B., "Ambiguity of Uppercase vs Lowercase in RFC
2119 Key Words", BCP 14, RFC 8174, DOI 10.17487/RFC8174,
May 2017, <https://www.rfc-editor.org/info/rfc8174>.
11.2. Informative References
[RFC5358] Damas, J. and F. Neves, "Preventing Use of Recursive
Nameservers in Reflector Attacks", BCP 140, RFC 5358,
DOI 10.17487/RFC5358, October 2008,
<https://www.rfc-editor.org/info/rfc5358>.
[RFC6895] Eastlake 3rd, D., "Domain Name System (DNS) IANA
Considerations", BCP 42, RFC 6895, DOI 10.17487/RFC6895,
April 2013, <https://www.rfc-editor.org/info/rfc6895>.
[RFC8499] Hoffman, P., Sullivan, A., and K. Fujiwara, "DNS
Terminology", BCP 219, RFC 8499, DOI 10.17487/RFC8499,
January 2019, <https://www.rfc-editor.org/info/rfc8499>.
[RR_TYPES] IANA, "Domain Name System (DNS) Parameters",
<https://www.iana.org/assignments/dns-parameters>.
Acknowledgements
David Lawrence provided valuable observations and concrete
suggestions. Jeremy Laidman helped make the document better. Tony
Finch realized that this document was valuable and implemented it
while under attack. Richard Gibson identified areas where more
detail and accuracy were useful. A large number of other people also
provided comments and suggestions; we thank them all for the
feedback.
Authors' Addresses
Joe Abley
Afilias
300-184 York Street
London, ON N6A 1B5
Canada
Phone: +1 519 670 9327
Email: jabley@afilias.info
Olafur Gudmundsson
Cloudflare Inc.
Email: olafur+ietf@cloudflare.com
Marek Majkowski
Cloudflare Inc.
Email: marek@cloudflare.com
Evan Hunt
ISC
950 Charter St
Redwood City, CA 94063
United States of America
Email: each@isc.org
AI ELI5
On this page
§1Introduction and Terminology
The Domain Name System (DNS) has a query type (QTYPE) called ANY, which is type code 255, written "*" in RFC 1035. The operator of an authoritative server might choose not to answer such queries in full, for reasons of local policy, security or performance.
The DNS specification (RFC 1034 and RFC 1035) gave no guidance for that situation. This document supplies it. It calls the host sending a request the initiator and the host sending the response the responder. A conventional ANY response is one built by the algorithm in Section 4.3.2 of RFC 1034 with none of this document's mechanisms.
Sometimes a computer asks a name server for everything it knows about a name. The rules did not say what to do if the server would rather not hand over a giant list. This document says.
§2Why ANY Queries Are Used
ANY queries have legitimate uses. They help with debugging and with checking the state of a server for a name. Some people use them to save queries, for example to get the MX, A and AAAA RRsets for a mail domain in one go. There is no documented guidance for that use, and the RFC says some implementations have been observed not to work as their developers expected. Anyone who relies on ANY should include a fallback.
They also have abusive uses:
- ANY queries are frequently used to exploit the amplification potential of DNS servers using spoofed source addresses and UDP. Small responses make a server a less attractive amplifier.
- They can be used to mine authoritative-only servers for zone data, since they are expected to return all RRsets for a name.
- Some authoritative-only servers need extra processing to build a conventional ANY response, which operators may want to avoid.
Asking for everything is handy for checking things. But bad people can use it to make a server send a huge pile of data at someone else, so servers want a way to keep the answer small.
§3The General Approach
The proposal lets an authoritative server signal that conventional ANY queries are not supported for a particular QNAME. It does this in a way that works with unmodified clients such as resolvers, and triggers the behavior wanted from them.
The RFC says that other approaches were discussed. One proposed a new RCODE to signal that the server did not answer ANY in the standard way. That had an undesirable effect: a resolver receiving an unknown RCODE would resend the same query to all available authoritative servers instead of suppressing future ANY queries for that name. Returning a non-empty RRset avoids this, because the resolver has something to cache, which suppresses repeat queries to the same or different servers.
Instead of saying "I will not answer", the server gives a small real answer. The asker writes it down and does not keep pestering anyone.
§4How Responders Answer
For queries where the QNAME exists, QCLASS=IN and QTYPE=ANY, the RFC describes three modes, and operators and implementers may choose whichever suits them. Except as described, the responder MUST follow the standard algorithms.
Subset of RRsets (4.1). The responder MAY decline to give a conventional response and MAY send a single RRset, or a larger subset, in the answer section. Where several RRsets exist, it SHOULD choose a small subset to reduce amplification. If the zone is signed, the appropriate RRSIG records MUST be included. There is no signal telling the client the answer is incomplete.
Synthesized HINFO (4.2). If no CNAME exists at the name, the responder MAY synthesize one HINFO record. Its CPU field SHOULD be "RFC8482" and its OS field SHOULD be the null string, to keep the response small. The TTL SHOULD be large enough to suppress frequent repeat ANY queries from the same initiator for the same QNAME, but not so long that changing the policy later is hard, and it SHOULD be configurable. If DO=1 and the zone is known to be signed, a valid RRSIG MUST be returned; with DO=0 it SHOULD be omitted. Receivers SHOULD NOT assume an HINFO answer was synthesized under this specification.
Best guess (4.3). A responder can try to return what the initiator probably wants. Some implementations return all CNAME, MX, A and AAAA RRsets at the name and suppress others such as TXT and DNSKEY, which can make responses large. The result is generally larger than with the first two approaches. If the zone is signed and the DO bit is set, an RRSIG RRset MUST be included.
Transport (4.4). A responder MAY behave differently per transport, for example a conventional response over TCP while using one of the small-response methods over UDP.
The server has three choices: give one small answer, give a made-up short note that says it is not answering, or guess the few things you most likely want. It may even give the full answer on a safer road, TCP, and the short one on UDP.
| Mode | What is returned | Response size |
|---|---|---|
| Subset (4.1) | A single RRset or a small subset | Small |
| Synthesized HINFO (4.2) THIS RFC | One HINFO record, CPU "RFC8482" | Minimal |
| Best guess (4.3) | CNAME, MX, A and AAAA RRsets | Larger than 4.1 and 4.2 |
§5Initiators and the HINFO Side Effect
An initiator that gets back an HINFO record or a single RRset MAY cache it in the normal way, following normal caching semantics. It MAY also suppress further ANY queries if its cache holds a matching HINFO record with the CPU field described in Section 4, or answer such queries from its cache.
There is a side effect. A cached synthesized HINFO RRset might suppress later queries for QTYPE=HINFO from the same initiator, effectively masking any real HINFO RRset in the zone. Operators whose zones use HINFO conventionally SHOULD choose the single-RRset method or another type. The RFC says HINFO is believed to be rarely used in the DNS at the time of writing, based on passive DNS and server observations.
The asker keeps the short answer in a notebook and does not ask again for a while. The small catch is that the note could hide a real answer to a different question about the same name.
- QTYPE=ANY queryAsks for all records at a name.
- Synthesized HINFOCPU field "RFC8482"; the initiator may cache it and suppress further ANY queries.
§6Changes to RFC 1034 and RFC 1035
This document extends the processing of ANY queries in Section 4.3.2 of RFC 1034. It argues that returning a subset is legitimate and consistent with RFC 1035: it can be argued that ANY does not always mean ALL, as used in Section 3.2.3 of RFC 1035. The main difference is that the TC bit SHOULD NOT be set in the response, which signals that it is not a complete answer.
All of this is optional for both initiators and responders. RRSIG queries (QTYPE=RRSIG) can also cause large responses and extra work, but this RFC does not specify behavior for them, though future advice might benefit from consistency with these approaches.
Nothing breaks the old rules. "Any" never really promised "all", and nobody has to follow the new advice.
§7Security and Implementation Experience
ANY queries are frequently seen in reflection attacks, since a relatively small query can bring a large response. Being able to suppress those responses makes a server a less useful amplifier. The RFC states the optional behavior is compatible with DNSSEC for both initiator and responder.
On implementation experience, the RFC reports that the Cloudflare authoritative nameserver implemented the HINFO response in October 2015, with a few minor problems since resolved. NSD 4.1 implemented the subset mode in 2016. BIND9 got a single-RRset response, available in production releases starting in BIND 9.11.
Bad guys like questions with tiny asks and giant answers. Making the answer small takes away the fun for them.