rfc9076.original   rfc9076.txt 
dprive T. Wicinski, Ed. Internet Engineering Task Force (IETF) T. Wicinski, Ed.
Internet-Draft March 9, 2021 Request for Comments: 9076 July 2021
Obsoletes: 7626 (if approved) Obsoletes: 7626
Intended status: Informational Category: Informational
Expires: September 10, 2021 ISSN: 2070-1721
DNS Privacy Considerations DNS Privacy Considerations
draft-ietf-dprive-rfc7626-bis-09
Abstract Abstract
This document describes the privacy issues associated with the use of This document describes the privacy issues associated with the use of
the DNS by Internet users. It provides general observations about the DNS by Internet users. It provides general observations about
typical current privacy practices. It is intended to be an analysis typical current privacy practices. It is intended to be an analysis
of the present situation and does not prescribe solutions. This of the present situation and does not prescribe solutions. This
document obsoletes RFC 7626. document obsoletes RFC 7626.
Status of This Memo Status of This Memo
This Internet-Draft is submitted in full conformance with the This document is not an Internet Standards Track specification; it is
provisions of BCP 78 and BCP 79. published for informational purposes.
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 This document is a product of the Internet Engineering Task Force
and may be updated, replaced, or obsoleted by other documents at any (IETF). It represents the consensus of the IETF community. It has
time. It is inappropriate to use Internet-Drafts as reference received public review and has been approved for publication by the
material or to cite them other than as "work in progress." Internet Engineering Steering Group (IESG). Not all documents
approved by the IESG are candidates for any level of Internet
Standard; see Section 2 of RFC 7841.
This Internet-Draft will expire on September 10, 2021. 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/rfc9076.
Copyright Notice Copyright Notice
Copyright (c) 2021 IETF Trust and the persons identified as the Copyright (c) 2021 IETF Trust and the persons identified as the
document authors. All rights reserved. document authors. All rights reserved.
This document is subject to BCP 78 and the IETF Trust's Legal This document is subject to BCP 78 and the IETF Trust's Legal
Provisions Relating to IETF Documents Provisions Relating to IETF Documents
(https://trustee.ietf.org/license-info) in effect on the date of (https://trustee.ietf.org/license-info) in effect on the date of
publication of this document. Please review these documents publication of this document. Please review these documents
carefully, as they describe your rights and restrictions with respect carefully, as they describe your rights and restrictions with respect
to this document. Code Components extracted from this document must to this document. Code Components extracted from this document must
include Simplified BSD License text as described in Section 4.e of include Simplified BSD License text as described in Section 4.e of
the Trust Legal Provisions and are provided without warranty as the Trust Legal Provisions and are provided without warranty as
described in the Simplified BSD License. described in the Simplified BSD License.
Table of Contents Table of Contents
1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . 2 1. Introduction
2. Scope . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5 2. Scope
3. Risks . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5 3. Risks
4. Risks in the DNS Data . . . . . . . . . . . . . . . . . . . . 6 4. Risks in the DNS Data
4.1. The Public Nature of DNS Data . . . . . . . . . . . . . . 6 4.1. The Public Nature of DNS Data
4.2. Data in the DNS Request . . . . . . . . . . . . . . . . . 6 4.2. Data in the DNS Request
4.2.1. Data in the DNS Payload . . . . . . . . . . . . . . . 8 4.2.1. Data in the DNS Payload
4.3. Cache Snooping . . . . . . . . . . . . . . . . . . . . . 8 4.3. Cache Snooping
5. Risks On the Wire . . . . . . . . . . . . . . . . . . . . . . 8 5. Risks on the Wire
5.1. Unencrypted Transports . . . . . . . . . . . . . . . . . 8 5.1. Unencrypted Transports
5.2. Encrypted Transports . . . . . . . . . . . . . . . . . . 10 5.2. Encrypted Transports
6. Risks in the Servers . . . . . . . . . . . . . . . . . . . . 11 6. Risks in the Servers
6.1. In the Recursive Resolvers . . . . . . . . . . . . . . . 12 6.1. In the Recursive Resolvers
6.1.1. Resolver Selection . . . . . . . . . . . . . . . . . 12 6.1.1. Resolver Selection
6.1.2. Active Attacks on Resolver Configuration . . . . . . 14 6.1.2. Active Attacks on Resolver Configuration
6.1.3. Blocking of DNS Resolution Services . . . . . . . . . 15 6.1.3. Blocking of DNS Resolution Services
6.1.4. Encrypted Transports and Recursive Resolvers . . . . 15 6.1.4. Encrypted Transports and Recursive Resolvers
6.2. In the Authoritative Name Servers . . . . . . . . . . . . 16 6.2. In the Authoritative Name Servers
7. Other risks . . . . . . . . . . . . . . . . . . . . . . . . . 17 7. Other Risks
7.1. Re-identification and Other Inferences . . . . . . . . . 17 7.1. Re-identification and Other Inferences
7.2. More Information . . . . . . . . . . . . . . . . . . . . 18 7.2. More Information
8. Actual "Attacks" . . . . . . . . . . . . . . . . . . . . . . 18 8. Actual "Attacks"
9. Legalities . . . . . . . . . . . . . . . . . . . . . . . . . 19 9. Legalities
10. Security Considerations . . . . . . . . . . . . . . . . . . . 19 10. Security Considerations
11. IANA Considerations . . . . . . . . . . . . . . . . . . . . . 19 11. IANA Considerations
12. Contributions . . . . . . . . . . . . . . . . . . . . . . . . 19 12. References
13. Acknowledgments . . . . . . . . . . . . . . . . . . . . . . . 19 12.1. Normative References
14. References . . . . . . . . . . . . . . . . . . . . . . . . . 20 12.2. Informative References
14.1. Normative References . . . . . . . . . . . . . . . . . . 20 Appendix A. Updates since RFC 7626
14.2. Informative References . . . . . . . . . . . . . . . . . 20 Acknowledgments
14.3. URIs . . . . . . . . . . . . . . . . . . . . . . . . . . 26 Contributions
Appendix A. Updates since RFC7626 . . . . . . . . . . . . . . . 27 Author's Address
Appendix B. Changelog . . . . . . . . . . . . . . . . . . . . . 27
Author's Address . . . . . . . . . . . . . . . . . . . . . . . . 30
1. Introduction 1. Introduction
This document is an analysis of the DNS privacy issues, in the spirit This document is an analysis of the DNS privacy issues, in the spirit
of Section 8 of [RFC6973]. of Section 8 of [RFC6973].
The Domain Name System (DNS) is specified in [RFC1034], [RFC1035], The Domain Name System (DNS) is specified in [RFC1034], [RFC1035],
and many later RFCs, which have never been consolidated. It is one and many later RFCs, which have never been consolidated. It is one
of the most important infrastructure components of the Internet and of the most important infrastructure components of the Internet and
often ignored or misunderstood by Internet users (and even by many is often ignored or misunderstood by Internet users (and even by many
professionals). Almost every activity on the Internet starts with a professionals). Almost every activity on the Internet starts with a
DNS query (and often several). Its use has many privacy implications DNS query (and often several). Its use has many privacy
and this document is an attempt at a comprehensive and accurate list. implications, and this document is an attempt at a comprehensive and
accurate list.
Let us begin with a simplified reminder of how the DNS works (See Let us begin with a simplified reminder of how the DNS works (see
also [RFC8499]). A client, the stub resolver, issues a DNS query to also [RFC8499]). A client, the stub resolver, issues a DNS query to
a server, called the recursive resolver (also called caching resolver a server called the recursive resolver (also called caching resolver,
or full resolver or recursive name server). Let's use the query full resolver, or recursive name server). Let's use the query "What
"What are the AAAA records for www.example.com?" as an example. AAAA are the AAAA records for www.example.com?" as an example. AAAA is
is the QTYPE (Query Type), and www.example.com is the QNAME (Query the QTYPE (Query Type), and www.example.com is the QNAME (Query
Name). (The description that follows assumes a cold cache, for Name). (The description that follows assumes a cold cache, for
instance, because the server just started.) The recursive resolver instance, because the server just started.) The recursive resolver
will first query the root name servers. In most cases, the root name will first query the root name servers. In most cases, the root name
servers will send a referral. In this example, the referral will be servers will send a referral. In this example, the referral will be
to the .com name servers. The resolver repeats the query to one of to the .com name servers. The resolver repeats the query to one of
the .com name servers. The .com name servers, in turn, will refer to the .com name servers. The .com name servers, in turn, will refer to
the example.com name servers. The example.com name server will then the example.com name servers. The example.com name servers will then
return the answer. The root name servers, the name servers of .com, return the answers. The root name servers, the name servers of .com,
and the name servers of example.com are called authoritative name and the name servers of example.com are called authoritative name
servers. It is important, when analyzing the privacy issues, to servers. It is important, when analyzing the privacy issues, to
remember that the question asked to all these name servers is always remember that the question asked to all these name servers is always
the original question, not a derived question. The question sent to the original question, not a derived question. The question sent to
the root name servers is "What are the AAAA records for the root name servers is "What are the AAAA records for
www.example.com?", not "What are the name servers of .com?". By www.example.com?", not "What are the name servers of .com?". By
repeating the full question, instead of just the relevant part of the repeating the full question, instead of just the relevant part of the
question to the next in line, the DNS provides more information than question to the next in line, the DNS provides more information than
necessary to the name server. In this simplified description, necessary to the name server. In this simplified description,
recursive resolvers do not implement QNAME minimization as described recursive resolvers do not implement QNAME minimization as described
in [RFC7816], which will only send the relevant part of the question in [RFC7816], which will only send the relevant part of the question
to the upstream name server. to the upstream name server.
DNS relies heavily on caching, so the algorithm described above is DNS relies heavily on caching, so the algorithm described above is
actually a bit more complicated, and not all questions are sent to actually a bit more complicated, and not all questions are sent to
the authoritative name servers. If a few seconds later the stub the authoritative name servers. If the stub resolver asks the
resolver asks the recursive resolver, "What are the SRV records of recursive resolver a few seconds later, "What are the SRV records of
_xmpp-server._tcp.example.com?", the recursive resolver will remember _xmpp-server._tcp.example.com?", the recursive resolver will remember
that it knows the name servers of example.com and will just query that it knows the name servers of example.com and will just query
them, bypassing the root and .com. Because there is typically no them, bypassing the root and .com. Because there is typically no
caching in the stub resolver, the recursive resolver, unlike the caching in the stub resolver, the recursive resolver, unlike the
authoritative servers, sees all the DNS traffic. (Applications, like authoritative servers, sees all the DNS traffic. (Applications, like
web browsers, may have some form of caching that does not follow DNS web browsers, may have some form of caching that does not follow DNS
rules, for instance, because it may ignore the TTL. So, the rules, for instance, because it may ignore the TTL. So, the
recursive resolver does not see all the name resolution activity.) recursive resolver does not see all the name resolution activity.)
It should be noted that DNS recursive resolvers sometimes forward It should be noted that DNS recursive resolvers sometimes forward
requests to other recursive resolvers, typically bigger machines, requests to other recursive resolvers, typically bigger machines,
with a larger and more shared cache (and the query hierarchy can be with a larger and more shared cache (and the query hierarchy can be
even deeper, with more than two levels of recursive resolvers). From even deeper, with more than two levels of recursive resolvers). From
the point of view of privacy, these forwarders are like resolvers, the point of view of privacy, these forwarders are like resolvers
except that they do not see all of the requests being made (due to except that they do not see all of the requests being made (due to
caching in the first resolver). caching in the first resolver).
At the time of writing, almost all this DNS traffic is currently sent At the time of writing, almost all this DNS traffic is currently sent
unencrypted. However, there is increasing deployment of DNS-over-TLS unencrypted. However, there is increasing deployment of DNS over TLS
(DoT) [RFC7858] and DNS-over-HTTPS (DoH) [RFC8484], particularly in (DoT) [RFC7858] and DNS over HTTPS (DoH) [RFC8484], particularly in
mobile devices, browsers, and by providers of anycast recursive DNS mobile devices, browsers, and by providers of anycast recursive DNS
resolution services. There are a few cases where there is some resolution services. There are a few cases where there is some
alternative channel encryption, for instance, in an IPsec VPN tunnel, alternative channel encryption, for instance, in an IPsec VPN tunnel,
at least between the stub resolver and the resolver. Some recent at least between the stub resolver and the resolver. Some recent
analysis on service quality of encrypted DNS traffic can be found in analysis on the service quality of encrypted DNS traffic can be found
[dns-over-encryption]. in [dns-over-encryption].
Today, almost all DNS queries are sent over UDP [thomas-ditl-tcp]. Today, almost all DNS queries are sent over UDP [thomas-ditl-tcp].
This has practical consequences when considering encryption of the This has practical consequences when considering encryption of the
traffic as a possible privacy technique. Some encryption solutions traffic as a possible privacy technique. Some encryption solutions
are only designed for TCP, not UDP, although new solutions are still are only designed for TCP, not UDP, although new solutions are still
emerging [I-D.ietf-quic-transport] [I-D.ietf-dprive-dnsoquic]. emerging [RFC9000] [DPRIVE-DNSOQUIC].
Another important point to keep in mind when analyzing the privacy Another important point to keep in mind when analyzing the privacy
issues of DNS is the fact that DNS requests received by a server are issues of DNS is the fact that DNS requests received by a server are
triggered by different reasons. Let's assume an eavesdropper wants triggered for different reasons. Let's assume an eavesdropper wants
to know which web page is viewed by a user. For a typical web page, to know which web page is viewed by a user. For a typical web page,
there are three sorts of DNS requests being issued: there are three sorts of DNS requests being issued:
o Primary request: this is the domain name in the URL that the user Primary request:
typed, selected from a bookmark, or chose by clicking on an This is the domain name in the URL that the user typed, selected
hyperlink. Presumably, this is what is of interest for the from a bookmark, or chose by clicking on a hyperlink. Presumably,
eavesdropper. this is what is of interest for the eavesdropper.
o Secondary requests: these are the additional requests performed by Secondary requests:
the user agent (here, the web browser) without any direct These are the additional requests performed by the user agent
involvement or knowledge of the user. For the Web, they are (here, the web browser) without any direct involvement or
triggered by embedded content, Cascading Style Sheets (CSS), knowledge of the user. For the Web, they are triggered by
JavaScript code, embedded images, etc. In some cases, there can embedded content, Cascading Style Sheets (CSS), JavaScript code,
be dozens of domain names in different contexts on a single web embedded images, etc. In some cases, there can be dozens of
page. domain names in different contexts on a single web page.
o Tertiary requests: these are the additional requests performed by Tertiary requests:
the DNS system itself. For instance, if the answer to a query is These are the additional requests performed by the DNS service
a referral to a set of name servers, and the glue records are not itself. For instance, if the answer to a query is a referral to a
returned, the resolver will have to do additional requests to turn set of name servers and the glue records are not returned, the
the name servers' names into IP addresses. Similarly, even if resolver will have to send additional requests to turn the name
glue records are returned, a careful recursive server will do servers' names into IP addresses. Similarly, even if glue records
tertiary requests to verify the IP addresses of those records. are returned, a careful recursive server will send tertiary
requests to verify the IP addresses of those records.
It can also be noted that, in the case of a typical web browser, more It can also be noted that, in the case of a typical web browser, more
DNS requests than strictly necessary are sent, for instance, to DNS requests than strictly necessary are sent, for instance, to
prefetch resources that the user may query later or when prefetch resources that the user may query later or when
autocompleting the URL in the address bar. Both are a significant autocompleting the URL in the address bar. Both are a significant
privacy concern since they may leak information even about non- privacy concern since they may leak information even about non-
explicit actions. For instance, just reading a local HTML page, even explicit actions. For instance, just reading a local HTML page, even
without selecting the hyperlinks, may trigger DNS requests. without selecting the hyperlinks, may trigger DNS requests.
For privacy-related terms, the terminology is from [RFC6973]. Privacy-related terminology is from [RFC6973]. This document
obsoletes [RFC7626].
2. Scope 2. Scope
This document focuses mostly on the study of privacy risks for the This document focuses mostly on the study of privacy risks for the
end user (the one performing DNS requests). The risks of pervasive end user (the one performing DNS requests). The risks of pervasive
surveillance [RFC7258] are considered as well as risks coming from a surveillance [RFC7258] are considered as well as risks coming from a
more focused surveillance. In this document, the term 'end user' is more focused surveillance. In this document, the term "end user" is
used as defined in [RFC8890]. used as defined in [RFC8890].
This document does not attempt a comparison of specific privacy This document does not attempt a comparison of specific privacy
protections provided by individual networks or organizations, it protections provided by individual networks or organizations; it
makes only general observations about typical current practices. makes only general observations about typical current practices.
Privacy risks for the holder of a zone (the risk that someone gets Privacy risks for the holder of a zone (the risk that someone gets
the data) are discussed in [RFC5936] and [RFC5155]. the data) are discussed in [RFC5155] and [RFC5936].
Privacy risks for recursive operators (including access providers and Privacy risks for recursive operators (including access providers and
operators in enterprise networks) such as leakage of private operators in enterprise networks) such as leakage of private
namespaces or blocklists are out of scope for this document. namespaces or blocklists are out of scope for this document.
Non-privacy risks (e.g security related considerations such as cache Non-privacy risks (e.g., security-related considerations such as
poisoning) are also out of scope. cache poisoning) are also out of scope.
The privacy risks associated with the use of other protocols that The privacy risks associated with the use of other protocols that
make use of DNS information are not considered here. make use of DNS information are not considered here.
3. Risks 3. Risks
The following four sections outline the privacy considerations The following four sections outline the privacy considerations
associated with different aspects of the DNS for the end user. When associated with different aspects of the DNS for the end user. When
reading these sections it needs to be kept in mind that many of the reading these sections, it needs to be kept in mind that many of the
considerations (for example, recursive resolver and transport considerations (for example, recursive resolver and transport
protocol) can be specific to the network context that a device is protocol) can be specific to the network context that a device is
using at a given point in time. A user may have many devices and using at a given point in time. A user may have many devices, and
each device might utilize many different networks (e.g. home, work, each device might utilize many different networks (e.g., home, work,
public or cellular) over a period of time or even concurrently. An public, or cellular) over a period of time or even concurrently. An
exhaustive analysis of the privacy considerations for an individual exhaustive analysis of the privacy considerations for an individual
user would need to take into account the set of devices used and the user would need to take into account the set of devices used and the
multiple dynamic contexts of each device. This document does not multiple dynamic contexts of each device. This document does not
attempt such a complex analysis, but instead it presents an overview attempt such a complex analysis; instead, it presents an overview of
of the various considerations that could form the basis of such an the various considerations that could form the basis of such an
analysis. analysis.
4. Risks in the DNS Data 4. Risks in the DNS Data
4.1. The Public Nature of DNS Data 4.1. The Public Nature of DNS Data
It has been stated that "the data in the DNS is public". This It has been stated that "the data in the DNS is public". This
sentence makes sense for an Internet-wide lookup system, and there sentence makes sense for an Internet-wide lookup system, and there
are multiple facets to the data and metadata involved that deserve a are multiple facets to the data and metadata involved that deserve a
more detailed look. First, access control lists (ACLs) and private more detailed look. First, access control lists (ACLs) and private
namespaces notwithstanding, the DNS operates under the assumption namespaces notwithstanding, the DNS operates under the assumption
that public-facing authoritative name servers will respond to "usual" that public-facing authoritative name servers will respond to "usual"
DNS queries for any zone they are authoritative for without further DNS queries for any zone they are authoritative for, without further
authentication or authorization of the client (resolver). Due to the authentication or authorization of the client (resolver). Due to the
lack of search capabilities, only a given QNAME will reveal the lack of search capabilities, only a given QNAME will reveal the
resource records associated with that name (or that name's non- resource records associated with that name (or that name's
existence). In other words: one needs to know what to ask for, in nonexistence). In other words: one needs to know what to ask for in
order to receive a response. There are many ways in which supposedly order to receive a response. There are many ways in which supposedly
"private" resources currently leak. A few examples are DNSSEC NSEC "private" resources currently leak. A few examples are DNSSEC NSEC
zone walking[RFC4470]; passive-DNS services[passive-dns]; etc. The zone walking [RFC4470], passive DNS services [passive-dns], etc. The
zone transfer QTYPE [RFC5936] is often blocked or restricted to zone transfer QTYPE [RFC5936] is often blocked or restricted to
authenticated/authorized access to enforce this difference (and maybe authenticated/authorized access to enforce this difference (and maybe
for other reasons). for other reasons).
Another difference between the DNS data and a particular DNS Another difference between the DNS data and a particular DNS
transaction (i.e., a DNS name lookup). DNS data and the results of a transaction (i.e., a DNS name lookup): DNS data and the results of a
DNS query are public, within the boundaries described above, and may DNS query are public, within the boundaries described above, and may
not have any confidentiality requirements. However, the same is not not have any confidentiality requirements. However, the same is not
true of a single transaction or a sequence of transactions; those true of a single transaction or a sequence of transactions; those
transactions are not / should not be public. A single transaction transactions are not / should not be public. A single transaction
reveals both the originator of the query and the query contents which reveals both the originator of the query and the query contents; this
potentially leaks sensitive information about a specific user. A potentially leaks sensitive information about a specific user. A
typical example from outside the DNS world is: the web site of typical example from outside the DNS world is that the website of
Alcoholics Anonymous is public; the fact that you visit it should not Alcoholics Anonymous is public but the fact that you visit it should
be. Furthermore, the ability to link queries reveals information not be. Furthermore, the ability to link queries reveals information
about individual use patterns. about individual use patterns.
4.2. Data in the DNS Request 4.2. Data in the DNS Request
The DNS request includes many fields, but two of them seem The DNS request includes many fields, but two of them seem
particularly relevant for the privacy issues: the QNAME and the particularly relevant for the privacy issues: the QNAME and the
source IP address. "source IP address" is used in a loose sense of source IP address. "Source IP address" is used in a loose sense of
"source IP address + maybe source port number", because the port "source IP address + maybe source port number", because the port
number is also in the request and can be used to differentiate number is also in the request and can be used to differentiate
between several users sharing an IP address (behind a Carrier-Grade between several users sharing an IP address (behind a Carrier-Grade
NAT (CGN), for instance [RFC6269]). NAT (CGN), for instance [RFC6269]).
The QNAME is the full name sent by the user. It gives information The QNAME is the full name sent by the user. It gives information
about what the user does ("What are the MX records of example.net?" about what the user does ("What are the MX records of example.net?"
means he probably wants to send email to someone at example.net, means they probably want to send email to someone at example.net,
which may be a domain used by only a few persons and is therefore which may be a domain used by only a few persons and is therefore
very revealing about communication relationships). Some QNAMEs are very revealing about communication relationships). Some QNAMEs are
more sensitive than others. For instance, querying the A record of a more sensitive than others. For instance, querying the A record of a
well-known web statistics domain reveals very little (everybody well-known web statistics domain reveals very little (everybody
visits web sites that use this analytics service), but querying the A visits websites that use this analytics service), but querying the A
record of www.verybad.example where verybad.example is the domain of record of www.verybad.example where verybad.example is the domain of
an organization that some people find offensive or objectionable may an organization that some people find offensive or objectionable may
create more problems for the user. Also, sometimes, the QNAME embeds create more problems for the user. Also, sometimes, the QNAME embeds
the software one uses, which could be a privacy issue. For instance, the software one uses, which could be a privacy issue (for instance,
_ldap._tcp.Default-First-Site-Name._sites.gc._msdcs.example.org. _ldap._tcp.Default-First-Site-Name._sites.gc._msdcs.example.org.
There are also some BitTorrent clients that query an SRV record for There are also some BitTorrent clients that query an SRV record for
_bittorrent-tracker._tcp.domain.example. _bittorrent-tracker._tcp.domain.example.
Another important thing about the privacy of the QNAME is the future Another important thing about the privacy of the QNAME is future
usages. Today, the lack of privacy is an obstacle to putting usages. Today, the lack of privacy is an obstacle to putting
potentially sensitive or personally identifiable data in the DNS. At potentially sensitive or personally identifiable data in the DNS. At
the moment, your DNS traffic might reveal that you are doing email the moment, your DNS traffic might reveal that you are exchanging
but not with whom. If your Mail User Agent (MUA) starts looking up emails but not with whom. If your Mail User Agent (MUA) starts
Pretty Good Privacy (PGP) keys in the DNS [RFC7929], then privacy looking up Pretty Good Privacy (PGP) keys in the DNS [RFC7929], then
becomes a lot more important. And email is just an example; there privacy becomes a lot more important. And email is just an example;
would be other really interesting uses for a more privacy-friendly there would be other really interesting uses for a more privacy-
DNS. friendly DNS.
For the communication between the stub resolver and the recursive For the communication between the stub resolver and the recursive
resolver, the source IP address is the address of the user's machine. resolver, the source IP address is the address of the user's machine.
Therefore, all the issues and warnings about collection of IP Therefore, all the issues and warnings about collection of IP
addresses apply here. For the communication between the recursive addresses apply here. For the communication between the recursive
resolver and the authoritative name servers, the source IP address resolver and the authoritative name servers, the source IP address
has a different meaning; it does not have the same status as the has a different meaning; it does not have the same status as the
source address in an HTTP connection. It can be typically the IP source address in an HTTP connection. It is typically the IP address
address of the recursive resolver that, in a way, "hides" the real of the recursive resolver that, in a way, "hides" the real user.
user. However, hiding does not always work. Sometimes EDNS(0) However, hiding does not always work. The edns-client-subnet (ECS)
Client subnet [RFC7871] is used (see one privacy analysis in EDNS0 option [RFC7871] is sometimes used (see one privacy analysis in
[denis-edns-client-subnet]). Sometimes the end user has a personal [denis-edns-client-subnet]). Sometimes the end user has a personal
recursive resolver on their machine. In both cases, the IP address recursive resolver on their machine. In both cases, the IP address
originating queries to the authoritative server is as sensitive as it originating queries to the authoritative server is as sensitive as it
is for HTTP [sidn-entrada]. is for HTTP [sidn-entrada].
A note about IP addresses: there is currently no IETF document that A note about IP addresses: there is currently no IETF document that
describes in detail all the privacy issues around IP addressing in describes in detail all the privacy issues around IP addressing in
general, although [RFC7721] does discuss privacy considerations for general, although [RFC7721] does discuss privacy considerations for
IPv6 address generation mechanisms. In the meantime, the discussion IPv6 address generation mechanisms. In the meantime, the discussion
here is intended to include both IPv4 and IPv6 source addresses. For here is intended to include both IPv4 and IPv6 source addresses. For
a number of reasons, their assignment and utilization characteristics a number of reasons, their assignment and utilization characteristics
are different, which may have implications for details of information are different, which may have implications for details of information
leakage associated with the collection of source addresses. (For leakage associated with the collection of source addresses. (For
example, a specific IPv6 source address seen on the public Internet example, a specific IPv6 source address seen on the public Internet
is less likely than an IPv4 address to originate behind an address is less likely than an IPv4 address to originate behind an address-
sharing scheme.) However, for both IPv4 and IPv6 addresses, it is sharing scheme.) However, for both IPv4 and IPv6 addresses, it is
important to note that source addresses are propagated with queries important to note that source addresses are propagated with queries
via EDNS(0) Client subnet and comprise metadata about the host, user, via the ECS option and comprise metadata about the host, user, or
or application that originated them. application that originated them.
4.2.1. Data in the DNS Payload 4.2.1. Data in the DNS Payload
At the time of writing there are no standardized client identifiers At the time of writing, there are no standardized client identifiers
contained in the DNS payload itself (ECS [RFC7871] while widely used contained in the DNS payload itself (ECS, as described in [RFC7871],
is only of Category Informational). is widely used; however, [RFC7871] is only an Informational RFC).
DNS Cookies [RFC7873] are a lightweight DNS transaction security DNS Cookies [RFC7873] are a lightweight DNS transaction security
mechanism that provides limited protection against a variety of mechanism that provides limited protection against a variety of
increasingly common denial-of-service and amplification/forgery or increasingly common denial-of-service and amplification/forgery or
cache poisoning attacks by off-path attackers. It is noted, however, cache poisoning attacks by off-path attackers. It is noted, however,
that they are designed to just verify IP addresses (and should change that they are designed to just verify IP addresses (and should change
once a client's IP address changes), but they are not designed to once a client's IP address changes), but they are not designed to
actively track users (like HTTP cookies). actively track users (like HTTP cookies).
There are anecdotal accounts of MAC addresses [1] and even user names There are anecdotal accounts of Media Access Control (MAC) addresses
being inserted in non-standard EDNS(0) options [RFC6891] for stub to (https://lists.dns-oarc.net/pipermail/dns-
operations/2016-January/014143.html) and even user names being
inserted in nonstandard EDNS(0) options [RFC6891] for stub-to-
resolver communications to support proprietary functionality resolver communications to support proprietary functionality
implemented at the resolver (e.g., parental filtering). implemented at the resolver (e.g., parental filtering).
4.3. Cache Snooping 4.3. Cache Snooping
The content of recursive resolvers' caches can reveal data about the The content of recursive resolvers' caches can reveal data about the
clients using it (the privacy risks depend on the number of clients). clients using it (the privacy risks depend on the number of clients).
This information can sometimes be examined by sending DNS queries This information can sometimes be examined by sending DNS queries
with RD=0 to inspect cache content, particularly looking at the DNS with RD=0 to inspect cache content, particularly looking at the DNS
TTLs [grangeia.snooping]. Since this also is a reconnaissance TTLs [grangeia.snooping]. Since this also is a reconnaissance
technique for subsequent cache poisoning attacks, some counter technique for subsequent cache poisoning attacks, some
measures have already been developed and deployed countermeasures have already been developed and deployed
[cache-snooping-defence]. [cache-snooping-defence].
5. Risks On the Wire 5. Risks on the Wire
5.1. Unencrypted Transports 5.1. Unencrypted Transports
For unencrypted transports, DNS traffic can be seen by an For unencrypted transports, DNS traffic can be seen by an
eavesdropper like any other traffic. (DNSSEC, specified in eavesdropper like any other traffic. (DNSSEC, specified in
[RFC4033], explicitly excludes confidentiality from its goals.) So, [RFC4033], explicitly excludes confidentiality from its goals.) So,
if an initiator starts an HTTPS communication with a recipient, while if an initiator starts an HTTPS communication with a recipient, the
the HTTP traffic will be encrypted, the DNS exchange prior to it will HTTP traffic will be encrypted, but the DNS exchange prior to it will
not be. When other protocols will become more and more privacy-aware not be. When other protocols become more and more privacy aware and
and secured against surveillance (e.g., [RFC8446], secured against surveillance (e.g., [RFC8446], [RFC9000]), the use of
[I-D.ietf-quic-transport]), the use of unencrypted transports for DNS unencrypted transports for DNS may become "the weakest link" in
may become "the weakest link" in privacy. It is noted that at the privacy. It is noted that, at the time of writing, there is ongoing
time of writing there is on-going work attempting to encrypt the SNI work attempting to encrypt the Server Name Identification (SNI) in
in the TLS handshake [RFC8744], which is one of the last remaining the TLS handshake [RFC8744], which is one of the last remaining non-
non-DNS cleartext identifiers of a connection target. DNS cleartext identifiers of a connection target.
An important specificity of the DNS traffic is that it may take a An important specificity of the DNS traffic is that it may take a
different path than the communication between the initiator and the different path than the communication between the initiator and the
recipient. For instance, an eavesdropper may be unable to tap the recipient. For instance, an eavesdropper may be unable to tap the
wire between the initiator and the recipient but may have access to wire between the initiator and the recipient but may have access to
the wire going to the recursive resolver, or to the authoritative the wire going to the recursive resolver or to the authoritative name
name servers. servers.
The best place to tap, from an eavesdropper's point of view, is The best place to tap, from an eavesdropper's point of view, is
clearly between the stub resolvers and the recursive resolvers, clearly between the stub resolvers and the recursive resolvers,
because traffic is not limited by DNS caching. because traffic is not limited by DNS caching.
The attack surface between the stub resolver and the rest of the The attack surface between the stub resolver and the rest of the
world can vary widely depending upon how the end user's device is world can vary widely depending upon how the end user's device is
configured. By order of increasing attack surface: configured. By order of increasing attack surface:
o The recursive resolver can be on the end user's device. In * The recursive resolver can be on the end user's device. In
(currently) a small number of cases, individuals may choose to (currently) a small number of cases, individuals may choose to
operate their own DNS resolver on their local machine. In this operate their own DNS resolver on their local machine. In this
case, the attack surface for the connection between the stub case, the attack surface for the connection between the stub
resolver and the caching resolver is limited to that single resolver and the caching resolver is limited to that single
machine. The recursive resolver will expose data to authoritative machine. The recursive resolver will expose data to authoritative
resolvers as discussed in Section 6.2. resolvers as discussed in Section 6.2.
o The recursive resolver may be at the local network edge. For * The recursive resolver may be at the local network edge. For
many/most enterprise networks and for some residential networks, many/most enterprise networks and for some residential networks,
the caching resolver may exist on a server at the edge of the the caching resolver may exist on a server at the edge of the
local network. In this case, the attack surface is the local local network. In this case, the attack surface is the local
network. Note that in large enterprise networks, the DNS resolver network. Note that in large enterprise networks, the DNS resolver
may not be located at the edge of the local network but rather at may not be located at the edge of the local network but rather at
the edge of the overall enterprise network. In this case, the the edge of the overall enterprise network. In this case, the
enterprise network could be thought of as similar to the Internet enterprise network could be thought of as similar to the Internet
Access Provider (IAP) network referenced below. Access Provider (IAP) network referenced below.
o The recursive resolver can be in the IAP network. For most * The recursive resolver can be in the IAP network. For most
residential networks and potentially other networks, the typical residential networks and potentially other networks, the typical
case is for the user's device to be configured (typically case is for the user's device to be configured (typically
automatically through DHCP or RA options) with the addresses of automatically through DHCP or relay agent options) with the
the DNS proxy in the Customer Premise Equipment (CPE), which in addresses of the DNS proxy in the Customer Premises Equipment
turns points to the DNS recursive resolvers at the IAP. The (CPE), which in turn points to the DNS recursive resolvers at the
attack surface for on-the-wire attacks is therefore from the end IAP. The attack surface for on-the-wire attacks is therefore from
user system across the local network and across the IAP network to the end user system across the local network and across the IAP
the IAP's recursive resolvers. network to the IAP's recursive resolvers.
o The recursive resolver can be a public DNS service (or a privately * The recursive resolver can be a public DNS service (or a privately
run DNS resolver hosted on the public internet). Some machines run DNS resolver hosted on the public Internet). Some machines
may be configured to use public DNS resolvers such as those may be configured to use public DNS resolvers such as those
operated by Google Public DNS or OpenDNS. The user may have operated by Google Public DNS or OpenDNS. The user may have
configured their machine to use these DNS recursive resolvers configured their machine to use these DNS recursive resolvers
themselves -- or their IAP may have chosen to use the public DNS themselves -- or their IAP may have chosen to use the public DNS
resolvers rather than operating their own resolvers. In this resolvers rather than operating their own resolvers. In this
case, the attack surface is the entire public Internet between the case, the attack surface is the entire public Internet between the
user's connection and the public DNS service. It can be noted user's connection and the public DNS service. It can be noted
that if the user selects a single resolver with a small client that if the user selects a single resolver with a small client
population (even when using an encrypted transport) it can population (even when using an encrypted transport), it can
actually serve to aid tracking of that user as they move across actually serve to aid tracking of that user as they move across
network environments. network environments.
It is also noted that typically a device connected _only_ to a modern It is also noted that, typically, a device connected _only_ to a
cellular network is modern cellular network is
o directly configured with only the recursive resolvers of the IAP * directly configured with only the recursive resolvers of the IAP
and and
o afforded some level of protection against some types of * afforded some level of protection against some types of
eavesdropping for all traffic (including DNS traffic) due to the eavesdropping for all traffic (including DNS traffic) due to the
cellular network link-layer encryption. cellular network link-layer encryption.
The attack surface for this specific scenario is not considered here. The attack surface for this specific scenario is not considered here.
5.2. Encrypted Transports 5.2. Encrypted Transports
The use of encrypted transports directly mitigates passive The use of encrypted transports directly mitigates passive
surveillance of the DNS payload, however there are still some privacy surveillance of the DNS payload; however, some privacy attacks are
attacks possible. This section enumerates the residual privacy risks still possible. This section enumerates the residual privacy risks
to an end user when an attacker can passively monitor encrypted DNS to an end user when an attacker can passively monitor encrypted DNS
traffic flows on the wire. traffic flows on the wire.
These are cases where user identification, fingerprinting or These are cases where user identification, fingerprinting, or
correlations may be possible due to the use of certain transport correlations may be possible due to the use of certain transport
layers or clear text/observable features. These issues are not layers or cleartext/observable features. These issues are not
specific to DNS, but DNS traffic is susceptible to these attacks when specific to DNS, but DNS traffic is susceptible to these attacks when
using specific transports. using specific transports.
There are some general examples, for example, certain studies have Some general examples exist; for example, certain studies highlight
highlighted that IPv4 TTL, IPv6 Hop Limit, or TCP Window sizes os- that the OS fingerprint values (http://netres.ec/?b=11B99BD) of IPv4
fingerprint [2] values can be used to fingerprint client OS's or that TTL, IPv6 Hop Limit, or TCP Window size can be used to fingerprint
various techniques can be used to de-NAT DNS queries [dns-de-nat]. client OSes or that various techniques can be used to de-NAT DNS
queries [dns-de-nat].
Note that even when using encrypted transports, the use of clear text Note that even when using encrypted transports, the use of cleartext
transport options to decrease latency can provide correlation of a transport options to decrease latency can provide correlation of a
users' connections, e.g. using TCP Fast Open [RFC7413]. user's connections, e.g., using TCP Fast Open [RFC7413].
Implementations that support encrypted transports also commonly re- Implementations that support encrypted transports also commonly reuse
use connections for multiple DNS queries to optimize performance connections for multiple DNS queries to optimize performance (e.g.,
(e.g. via DNS pipelining or HTTPS multiplexing). Default via DNS pipelining or HTTPS multiplexing). Default configuration
configuration options for encrypted transports could in principle options for encrypted transports could, in principle, fingerprint a
fingerprint a specific client application. For example: specific client application. For example:
o TLS version or cipher suite selection * TLS version or cipher suite selection
o session resumption * session resumption
o the maximum number of messages to send or * the maximum number of messages to send and
o a maximum connection time before closing a connections and re- * a maximum connection time before closing a connections and
opening. reopening.
If libraries or applications offer user configuration of such options If libraries or applications offer user configuration of such options
(e.g. [getdns]) then they could in principle help to identify a (e.g., [getdns]), then they could, in principle, help to identify a
specific user. Users may want to use only the defaults to avoid this specific user. Users may want to use only the defaults to avoid this
issue. issue.
Whilst there are known attacks on older versions of TLS, the most While there are known attacks on older versions of TLS, the most
recent recommendations [RFC7525] and the development of TLS 1.3 recent recommendations [RFC7525] and the development of TLS 1.3
[RFC8446] largely mitigate those. [RFC8446] largely mitigate those.
Traffic analysis of unpadded encrypted traffic is also possible Traffic analysis of unpadded encrypted traffic is also possible
[pitfalls-of-dns-encryption] because the sizes and timing of [pitfalls-of-dns-encryption] because the sizes and timing of
encrypted DNS requests and responses can be correlated to unencrypted encrypted DNS requests and responses can be correlated to unencrypted
DNS requests upstream of a recursive resolver. DNS requests upstream of a recursive resolver.
6. Risks in the Servers 6. Risks in the Servers
Using the terminology of [RFC6973], the DNS servers (recursive Using the terminology of [RFC6973], the DNS servers (recursive
resolvers and authoritative servers) are enablers: they facilitate resolvers and authoritative servers) are enablers: "they facilitate
communication between an initiator and a recipient without being communication between an initiator and a recipient without being
directly in the communications path. As a result, they are often directly in the communications path". As a result, they are often
forgotten in risk analysis. But, to quote again [RFC6973], "Although forgotten in risk analysis. But, to quote [RFC6973] again, "Although
[...] enablers may not generally be considered as attackers, they may [...] enablers may not generally be considered as attackers, they may
all pose privacy threats (depending on the context) because they are all pose privacy threats (depending on the context) because they are
able to observe, collect, process, and transfer privacy-relevant able to observe, collect, process, and transfer privacy-relevant
data." In [RFC6973] parlance, enablers become observers when they data". In [RFC6973] parlance, enablers become observers when they
start collecting data. start collecting data.
Many programs exist to collect and analyze DNS data at the servers -- Many programs exist to collect and analyze DNS data at the servers --
from the "query log" of some programs like BIND to tcpdump and more from the "query log" of some programs like BIND to tcpdump and more
sophisticated programs like PacketQ [packetq] and DNSmezzo sophisticated programs like PacketQ [packetq] and DNSmezzo
[dnsmezzo]. The organization managing the DNS server can use this [dnsmezzo]. The organization managing the DNS server can use this
data itself, or it can be part of a surveillance program like PRISM data itself, or it can be part of a surveillance program like PRISM
[prism] and pass data to an outside observer. [prism] and pass data to an outside observer.
Sometimes, this data is kept for a long time and/or distributed to Sometimes this data is kept for a long time and/or distributed to
third parties for research purposes [ditl] [day-at-root], security third parties for research purposes [ditl] [day-at-root], security
analysis, or surveillance tasks. These uses are sometimes under some analysis, or surveillance tasks. These uses are sometimes under some
sort of contract, with various limitations, for instance, on sort of contract, with various limitations, for instance, on
redistribution, given the sensitive nature of the data. Also, there redistribution, given the sensitive nature of the data. Also, there
are observation points in the network that gather DNS data and then are observation points in the network that gather DNS data and then
make it accessible to third parties for research or security purposes make it accessible to third parties for research or security purposes
("passive DNS" [passive-dns]). ("passive DNS" [passive-dns]).
6.1. In the Recursive Resolvers 6.1. In the Recursive Resolvers
Recursive Resolvers see all the traffic since there is typically no Recursive resolvers see all the traffic since there is typically no
caching before them. To summarize: your recursive resolver knows a caching before them. To summarize: your recursive resolver knows a
lot about you. The resolver of a large IAP, or a large public lot about you. The resolver of a large IAP, or a large public
resolver, can collect data from many users. resolver, can collect data from many users.
6.1.1. Resolver Selection 6.1.1. Resolver Selection
Given all the above considerations, the choice of recursive resolver Given all the above considerations, the choice of recursive resolver
has direct privacy considerations for end users. Historically, end has direct privacy considerations for end users. Historically, end
user devices have used the DHCP-provided local network recursive user devices have used the DHCP-provided local network recursive
resolver. The choice by a user to join a particular network (e.g. by resolver. The choice by a user to join a particular network (e.g.,
physically plugging in a cable or selecting a network in a OS by physically plugging in a cable or selecting a network in an OS
dialogue) typically updates a number of system resources - these can dialogue) typically updates a number of system resources -- these can
include IP addresses, availability of IPv4/IPv6, DHCP server, and DNS include IP addresses, the availability of IPv4/IPv6, DHCP server, and
resolver. These individual changes, including the change in DNS DNS resolver. These individual changes, including the change in DNS
resolver, are not normally communicated directly to the user by the resolver, are not normally communicated directly to the user by the
OS when the network is joined. The choice of network has OS when the network is joined. The choice of network has
historically determined the default system DNS resolver selection; historically determined the default system DNS resolver selection;
the two are directly coupled in this model. the two are directly coupled in this model.
The vast majority of users do not change their default system DNS The vast majority of users do not change their default system DNS
settings and so implicitly accept the network settings for DNS. The settings and so implicitly accept the network settings for the DNS.
network resolvers have therefore historically been the sole The network resolvers have therefore historically been the sole
destination for all of the DNS queries from a device. These destination for all of the DNS queries from a device. These
resolvers may have varied privacy policies depending on the network. resolvers may have varied privacy policies depending on the network.
Privacy policies for these servers may or may not be available and Privacy policies for these servers may or may not be available, and
users need to be aware that privacy guarantees will vary with the users need to be aware that privacy guarantees will vary with the
network. network.
All major OS's expose the system DNS settings and allow users to All major OSes expose the system DNS settings and allow users to
manually override them if desired. manually override them if desired.
More recently, some networks and users have actively chosen to use a More recently, some networks and users have actively chosen to use a
large public resolver, e.g., Google Public DNS [3], Cloudflare [4], large public resolver, e.g., Google Public DNS
or Quad9 [5]. There can be many reasons: cost considerations for (https://developers.google.com/speed/public-dns), Cloudflare
network operators, better reliability or anti-censorship (https://developers.cloudflare.com/1.1.1.1/setting-up-1.1.1.1/), or
considerations are just a few. Such services typically do provide a Quad9 (https://www.quad9.net). There can be many reasons: cost
privacy policy and the user can get an idea of the data collected by considerations for network operators, better reliability, or anti-
such operators by reading one e.g., Google Public DNS - Your Privacy censorship considerations are just a few. Such services typically do
[6]. provide a privacy policy, and the user can get an idea of the data
collected by such operators by reading one, e.g., Google Public DNS -
Your Privacy (https://developers.google.com/speed/public-dns/
privacy).
In general, as with many other protocols, issues around In general, as with many other protocols, issues around
centralization also arise with DNS. The picture is fluid with centralization also arise with DNS. The picture is fluid with
several competing factors contributing which can also vary by several competing factors contributing, where these factors can also
geographic region. These include: vary by geographic region. These include:
o ISP outsourcing, including to third party and public resolvers * ISP outsourcing, including to third-party and public resolvers
o regional market domination by one or only a few ISPs * regional market domination by one or only a few ISPs
o applications directing DNS traffic by default to a limited subset * applications directing DNS traffic by default to a limited subset
of resolvers, see Section 6.1.1.2 of resolvers (see Section 6.1.1.2)
An increased proportion of the global DNS resolution traffic being An increased proportion of the global DNS resolution traffic being
served by only a few entities means that the privacy considerations served by only a few entities means that the privacy considerations
for users are highly dependent on the privacy policies and practices for users are highly dependent on the privacy policies and practices
of those entities. Many of the issues around centralization are of those entities. Many of the issues around centralization are
discussed in [centralisation-and-data-sovereignty]. discussed in [centralisation-and-data-sovereignty].
6.1.1.1. Dynamic Discovery of DoH and Strict DoT 6.1.1.1. Dynamic Discovery of DoH and Strict DoT
Whilst support for opportunistic DoT can be determined by probing a While support for opportunistic DoT can be determined by probing a
resolver on port 853, there is currently no standardized discovery resolver on port 853, there is currently no standardized discovery
mechanism for DoH and Strict DoT servers. mechanism for DoH and Strict DoT servers.
This means that clients which might want to dynamically discover such This means that clients that might want to dynamically discover such
encrypted services, and where users are willing to trust such encrypted services, and where users are willing to trust such
services, are not able to do so. At the time of writing, efforts to services, are not able to do so. At the time of writing, efforts to
provide standardized signaling mechanisms to discover the services provide standardized signaling mechanisms to discover the services
offered by local resolvers are in progress offered by local resolvers are in progress [DNSOP-RESOLVER]. Note
[I-D.ietf-dnsop-resolver-information]. Note that an increasing that an increasing number of ISPs are deploying encrypted DNS; for
numbers of ISPs are deploying encrypted DNS, for example see the example, see the Encrypted DNS Deployment Initiative [EDDI].
Encrypted DNS Deployment Initiative [EDDI].
6.1.1.2. Application-specific Resolver Selection 6.1.1.2. Application-Specific Resolver Selection
An increasing number of applications are offering application- An increasing number of applications are offering application-
specific encrypted DNS resolution settings, rather than defaulting to specific encrypted DNS resolution settings, rather than defaulting to
using only the system resolver. A variety of heuristics and using only the system resolver. A variety of heuristics and
resolvers are available in different applications including hard- resolvers are available in different applications, including hard-
coded lists of recognized DoH/DoT servers. coded lists of recognized DoH/DoT servers.
Generally, users are not aware of application specific DNS settings, Generally, users are not aware of application-specific DNS settings
and may not have control over those settings. To address these and may not have control over those settings. To address these
limitations, users will only be aware of and have the ability to limitations, users will only be aware of and have the ability to
control such settings if applications provide the following control such settings if applications provide the following
functions: functions:
o communicate clearly to users the change when the default * communicate the change clearly to users when the default
application resolver changes away from the system resolver application resolver changes away from the system resolver
o provide configuration options to change the default application * provide configuration options to change the default application
resolver, including a choice to always use the system resolver resolver, including a choice to always use the system resolver
o provide mechanisms for users to locally inspect, selectively * provide mechanisms for users to locally inspect, selectively
forward, and filter queries (either via the application itself or use forward, and filter queries (either via the application itself or
of the system resolver) use of the system resolver)
Application-specific changes to default destinations for users' DNS Application-specific changes to default destinations for users' DNS
queries might increase or decrease user privacy - it is highly queries might increase or decrease user privacy; it is highly
dependent on the network context and the application-specific dependent on the network context and the application-specific
default. This is an area of active debate and the IETF is working on default. This is an area of active debate, and the IETF is working
a number of issues related to application-specific DNS settings. on a number of issues related to application-specific DNS settings.
6.1.2. Active Attacks on Resolver Configuration 6.1.2. Active Attacks on Resolver Configuration
The previous section discussed DNS privacy, assuming that all the The previous section discussed DNS privacy, assuming that all the
traffic was directed to the intended servers (i.e those that would be traffic was directed to the intended servers (i.e., those that would
used in the absence of an active attack) and that the potential be used in the absence of an active attack) and that the potential
attacker was purely passive. But, in reality, there can be active attacker was purely passive. But, in reality, there can be active
attackers in the network. attackers in the network.
The Internet Threat model, as described in [RFC3552], assumes that The Internet Threat model, as described in [RFC3552], assumes that
the attacker controls the network. Such an attacker can completely the attacker controls the network. Such an attacker can completely
control any insecure DNS resolution, both passively monitoring the control any insecure DNS resolution, both passively monitoring the
queries and responses and substituting their own responses. Even if queries and responses and substituting their own responses. Even if
encrypted DNS such as DoH or DoT is used, unless the client has been encrypted DNS such as DoH or DoT is used, unless the client has been
configured in a secure way with the server identity, an active configured in a secure way with the server identity, an active
attacker can impersonate the server. This implies that opportunistic attacker can impersonate the server. This implies that opportunistic
modes of DoH/DoT as well as modes where the client learns of the DoH/ modes of DoH/DoT as well as modes where the client learns of the DoH/
DoT server via in-network mechanisms such as DHCP are vulnerable to DoT server via in-network mechanisms such as DHCP are vulnerable to
attack. In addition, if the client is compromised, the attacker can attack. In addition, if the client is compromised, the attacker can
replace the DNS configuration with one of its own choosing. replace the DNS configuration with one of its own choosing.
6.1.3. Blocking of DNS Resolution Services 6.1.3. Blocking of DNS Resolution Services
User privacy can also be at risk if there is blocking of access to User privacy can also be at risk if there is blocking of access to
remote recursive servers that offer encrypted transports e.g., when remote recursive servers that offer encrypted transports, e.g., when
the local resolver does not offer encryption and/or has very poor the local resolver does not offer encryption and/or has very poor
privacy policies. For example, active blocking of port 853 for DoT privacy policies. For example, active blocking of port 853 for DoT
or of specific IP addresses could restrict the resolvers available to or blocking of specific IP addresses could restrict the resolvers
the user. The extent of the risk to user privacy is highly dependent available to the user. The extent of the risk to user privacy is
on the specific network and user context; a user on a network that is highly dependent on the specific network and user context; a user on
known to perform surveillance would be compromised if they could not a network that is known to perform surveillance would be compromised
access such services, whereas a user on a trusted network might have if they could not access such services, whereas a user on a trusted
no privacy motivation to do so. network might have no privacy motivation to do so.
As a matter of policy, some recursive resolvers use their position in As a matter of policy, some recursive resolvers use their position in
the query path to selectively block access to certain DNS records. the query path to selectively block access to certain DNS records.
This is a form of Rendezvous-Based Blocking as described in This is a form of rendezvous-based blocking as described in
Section 4.3 of [RFC7754]. Such blocklists often include servers Section 4.3 of [RFC7754]. Such blocklists often include servers
known to be used for malware, bots or other security risks. In order known to be used for malware, bots, or other security risks. In
to prevent circumvention of their blocking policies, some networks order to prevent circumvention of their blocking policies, some
also block access to resolvers with incompatible policies. networks also block access to resolvers with incompatible policies.
It is also noted that attacks on remote resolver services, e.g., It is also noted that attacks on remote resolver services, e.g.,
DDoS, could force users to switch to other services that do not offer DDoS, could force users to switch to other services that do not offer
encrypted transports for DNS. encrypted transports for DNS.
6.1.4. Encrypted Transports and Recursive Resolvers 6.1.4. Encrypted Transports and Recursive Resolvers
6.1.4.1. DoT and DoH 6.1.4.1. DoT and DoH
Use of encrypted transports does not reduce the data available in the Use of encrypted transports does not reduce the data available in the
recursive resolver and ironically can actually expose more recursive resolver and ironically can actually expose more
information about users to operators. As described in Section 5.2 information about users to operators. As described in Section 5.2,
use of session based encrypted transports (TCP/TLS) can expose use of session-based encrypted transports (TCP/TLS) can expose
correlation data about users. correlation data about users.
6.1.4.2. DoH Specific Considerations 6.1.4.2. DoH-Specific Considerations
DoH inherits the full privacy properties of the HTTPS stack and as a DoH inherits the full privacy properties of the HTTPS stack and as a
consequence introduces new privacy considerations when compared with consequence introduces new privacy considerations when compared with
DNS over UDP, TCP or TLS [RFC7858]. Section 8.2 of [RFC8484] DNS over UDP, TCP, or TLS [RFC7858]. Section 8.2 of [RFC8484]
describes the privacy consideration in the server of the DoH describes the privacy considerations in the server of the DoH
protocol. protocol.
A brief summary of some of the issues includes: A brief summary of some of the issues includes the following:
o HTTPS presents new considerations for correlation, such as * HTTPS presents new considerations for correlation, such as
explicit HTTP cookies and implicit fingerprinting of the unique explicit HTTP cookies and implicit fingerprinting of the unique
set and ordering of HTTP request header fields. set and ordering of HTTP request header fields.
o The User-Agent and Accept-Language request header fields often * The User-Agent and Accept-Language request header fields often
convey specific information about the client version or locale. convey specific information about the client version or locale.
o Utilizing the full set of HTTP features enables DoH to be more * Utilizing the full set of HTTP features enables DoH to be more
than an HTTP tunnel, but it is at the cost of opening up than an HTTP tunnel, but it is at the cost of opening up
implementations to the full set of privacy considerations of HTTP. implementations to the full set of privacy considerations of HTTP.
o Implementations are advised to expose the minimal set of data * Implementations are advised to expose the minimal set of data
needed to achieve the desired feature set. needed to achieve the desired feature set.
[RFC8484] specifically makes selection of HTTPS functionality vs [RFC8484] specifically makes selection of HTTPS functionality vs.
privacy an implementation choice. At the extremes, there may be privacy an implementation choice. At the extremes, there may be
implementations that attempt to achieve parity with DoT from a implementations that attempt to achieve parity with DoT from a
privacy perspective at the cost of using no identifiable HTTP privacy perspective at the cost of using no identifiable HTTP
headers, there might be others that provide feature rich data flows headers, and there might be others that provide feature-rich data
where the low-level origin of the DNS query is easily identifiable. flows where the low-level origin of the DNS query is easily
Some implementations have, in fact, chosen to restrict the use of the identifiable. Some implementations have, in fact, chosen to restrict
'User-Agent' header so that resolver operators cannot identify the the use of the User-Agent header so that resolver operators cannot
specific application that is originating the DNS queries. identify the specific application that is originating the DNS
queries.
Privacy focused users should be aware of the potential for additional Privacy-focused users should be aware of the potential for additional
client identifiers in DoH compared to DoT and may want to only use client identifiers in DoH compared to DoT and may want to only use
DoH client implementations that provide clear guidance on what DoH client implementations that provide clear guidance on what
identifiers they add. identifiers they add.
6.2. In the Authoritative Name Servers 6.2. In the Authoritative Name Servers
Unlike what happens for recursive resolvers, observation capabilities Unlike what happens for recursive resolvers, the observation
of authoritative name servers are limited by caching; they see only capabilities of authoritative name servers are limited by caching;
the requests for which the answer was not in the cache. For they see only the requests for which the answer was not in the cache.
aggregated statistics ("What is the percentage of LOC queries?"), For aggregated statistics ("What is the percentage of LOC queries?"),
this is sufficient, but it prevents an observer from seeing this is sufficient, but it prevents an observer from seeing
everything. Similarly the increasing deployment of QNAME everything. Similarly, the increasing deployment of QNAME
minimisation [ripe-qname-measurements] reduces the data visible at minimization [ripe-qname-measurements] reduces the data visible at
the authoritative name server. Still, the authoritative name servers the authoritative name server. Still, the authoritative name servers
see a part of the traffic, and this subset may be sufficient to see a part of the traffic, and this subset may be sufficient to
violate some privacy expectations. violate some privacy expectations.
Also, the user often has some legal/contractual link with the Also, the user often has some legal/contractual link with the
recursive resolver (he has chosen the IAP, or he has chosen to use a recursive resolver (they have chosen the IAP, or they have chosen to
given public resolver), while having no control and perhaps no use a given public resolver) while having no control and perhaps no
awareness of the role of the authoritative name servers and their awareness of the role of the authoritative name servers and their
observation abilities. observation abilities.
As noted before, using a local resolver or a resolver close to the As noted before, using a local resolver or a resolver close to the
machine decreases the attack surface for an on-the-wire eavesdropper. machine decreases the attack surface for an on-the-wire eavesdropper.
But it may decrease privacy against an observer located on an But it may decrease privacy against an observer located on an
authoritative name server. This authoritative name server will see authoritative name server. This authoritative name server will see
the IP address of the end client instead of the address of a big the IP address of the end client instead of the address of a big
recursive resolver shared by many users. recursive resolver shared by many users.
skipping to change at page 17, line 23 skipping to change at line 789
receive together around 50,000 queries per second. While most of it receive together around 50,000 queries per second. While most of it
is "junk" (errors on the Top-Level Domain (TLD) name), it gives an is "junk" (errors on the Top-Level Domain (TLD) name), it gives an
idea of the amount of big data that pours into name servers. (And idea of the amount of big data that pours into name servers. (And
even "junk" can leak information; for instance, if there is a typing even "junk" can leak information; for instance, if there is a typing
error in the TLD, the user will send data to a TLD that is not the error in the TLD, the user will send data to a TLD that is not the
usual one.) usual one.)
Many domains, including TLDs, are partially hosted by third-party Many domains, including TLDs, are partially hosted by third-party
servers, sometimes in a different country. The contracts between the servers, sometimes in a different country. The contracts between the
domain manager and these servers may or may not take privacy into domain manager and these servers may or may not take privacy into
account. Whatever the contract, the third-party hoster may be honest account. Whatever the contract, the third-party hoster may or may
or not but, in any case, it will have to follow its local laws. For not be honest; in any case, it will have to follow its local laws.
example, requests to a given ccTLD may go to servers managed by For example, requests to a given ccTLD may go to servers managed by
organizations outside of the ccTLD's country. Users may not organizations outside of the ccTLD's country. Users may not
anticipate that, when doing a security analysis. anticipate that when doing a security analysis.
Also, it seems (see the survey described in [aeris-dns]) that there Also, it seems (see the survey described in [aeris-dns]) that there
is a strong concentration of authoritative name servers among is a strong concentration of authoritative name servers among
"popular" domains (such as the Alexa Top N list). For instance, "popular" domains (such as the Alexa Top N list). For instance,
among the Alexa Top 100K [7], one DNS provider hosts today 10% of the among the Alexa Top 100K (https://www.alexa.com/topsites), one DNS
domains. The ten most important DNS providers host together one provider hosts 10% of the domains today. The ten most important DNS
third of the domains. With the control (or the ability to sniff the providers together host one-third of all domains. With the control
traffic) of a few name servers, you can gather a lot of information. (or the ability to sniff the traffic) of a few name servers, you can
gather a lot of information.
7. Other risks 7. Other Risks
7.1. Re-identification and Other Inferences 7.1. Re-identification and Other Inferences
An observer has access not only to the data he/she directly collects An observer has access not only to the data they directly collect but
but also to the results of various inferences about this data. The also to the results of various inferences about this data. The term
term 'observer' here is used very generally, it might be one that is "observer" here is used very generally; for example, the observer
passively observing cleartext DNS traffic, one in the network that is might passively observe cleartext DNS traffic or be in the network
actively attacking the user by re-directing DNS resolution, or it that is actively attacking the user by redirecting DNS resolution, or
might be a local or remote resolver operator. it might be a local or remote resolver operator.
For instance, a user can be re-identified via DNS queries. If the For instance, a user can be re-identified via DNS queries. If the
adversary knows a user's identity and can watch their DNS queries for adversary knows a user's identity and can watch their DNS queries for
a period, then that same adversary may be able to re-identify the a period, then that same adversary may be able to re-identify the
user solely based on their pattern of DNS queries later on regardless user solely based on their pattern of DNS queries later on regardless
of the location from which the user makes those queries. For of the location from which the user makes those queries. For
example, one study [herrmann-reidentification] found that such re- example, one study [herrmann-reidentification] found that such re-
identification is possible so that "73.1% of all day-to-day links identification is possible so that "73.1% of all day-to-day links
were correctly established, i.e., user u was either re-identified were correctly established, i.e. user u was either re-identified
unambiguously (1) or the classifier correctly reported that u was not unambiguously (1) or the classifier correctly reported that u was not
present on day t+1 any more (2)." While that study related to web present on day t + 1 any more (2)". While that study related to web
browsing behavior, equally characteristic patterns may be produced browsing behavior, equally characteristic patterns may be produced
even in machine-to-machine communications or without a user taking even in machine-to-machine communications or without a user taking
specific actions, e.g., at reboot time if a characteristic set of specific actions, e.g., at reboot time if a characteristic set of
services are accessed by the device. services are accessed by the device.
For instance, one could imagine that an intelligence agency For instance, one could imagine that an intelligence agency
identifies people going to a site by putting in a very long DNS name identifies people going to a site by putting in a very long DNS name
and looking for queries of a specific length. Such traffic analysis and looking for queries of a specific length. Such traffic analysis
could weaken some privacy solutions. could weaken some privacy solutions.
The IAB privacy and security program also have a document [RFC7624] The IAB Privacy and Security Program also has a document [RFC7624]
that considers such inference-based attacks in a more general that considers such inference-based attacks in a more general
framework. framework.
7.2. More Information 7.2. More Information
Useful background information can also be found in [tor-leak] (about Useful background information can also be found in [tor-leak]
the risk of privacy leak through DNS) and in a few academic papers: (regarding the risk of privacy leaks through DNS) and in a few
[yanbin-tsudik], [castillo-garcia], [fangming-hori-sakurai], and academic papers: [yanbin-tsudik], [castillo-garcia],
[federrath-fuchs-herrmann-piosecny]. [fangming-hori-sakurai], and [federrath-fuchs-herrmann-piosecny].
8. Actual "Attacks" 8. Actual "Attacks"
A very quick examination of DNS traffic may lead to the false A very quick examination of DNS traffic may lead to the false
conclusion that extracting the needle from the haystack is difficult. conclusion that extracting the needle from the haystack is difficult.
"Interesting" primary DNS requests are mixed with useless (for the "Interesting" primary DNS requests are mixed with useless (for the
eavesdropper) secondary and tertiary requests (see the terminology in eavesdropper) secondary and tertiary requests (see the terminology in
Section 1). But, in this time of "big data" processing, powerful Section 1). But, in this time of "big data" processing, powerful
techniques now exist to get from the raw data to what the techniques now exist to get from the raw data to what the
eavesdropper is actually interested in. eavesdropper is actually interested in.
Many research papers about malware detection use DNS traffic to Many research papers about malware detection use DNS traffic to
detect "abnormal" behavior that can be traced back to the activity of detect "abnormal" behavior that can be traced back to the activity of
malware on infected machines. Yes, this research was done for the malware on infected machines. Yes, this research was done for the
good, but technically it is a privacy attack and it demonstrates the greater good, but technically it is a privacy attack and it
power of the observation of DNS traffic. See [dns-footprint], demonstrates the power of the observation of DNS traffic. See
[dagon-malware], and [darkreading-dns]. [dns-footprint], [dagon-malware], and [darkreading-dns].
Passive DNS systems [passive-dns] allow reconstruction of the data of Passive DNS services [passive-dns] allow reconstruction of the data
sometimes an entire zone. Well-known passive DNS systems keep only of sometimes an entire zone. Well-known passive DNS services keep
the DNS responses, and not the source IP address of the client, only the DNS responses and not the source IP address of the client,
precisely for privacy reasons. Other passive DNS systems may not be precisely for privacy reasons. Other passive DNS services may not be
so careful. And there is still the potential problems with revealing so careful. And there are still potential problems with revealing
QNAMEs. QNAMEs.
The revelations from the Edward Snowden documents, which were leaked The revelations from the Edward Snowden documents, which were leaked
from the National Security Agency (NSA), provide evidence of the use from the National Security Agency (NSA), provide evidence of the use
of the DNS in mass surveillance operations [morecowbell]. For of the DNS in mass surveillance operations [morecowbell]. For
example the MORECOWBELL surveillance program, which uses a dedicated example, the MORECOWBELL surveillance program uses a dedicated covert
covert monitoring infrastructure to actively query DNS servers and monitoring infrastructure to actively query DNS servers and perform
perform HTTP requests to obtain meta information about services and HTTP requests to obtain meta-information about services and to check
to check their availability. Also the QUANTUMTHEORY [8] project their availability. Also, the QUANTUMTHEORY
which includes detecting lookups for certain addresses and injecting (https://theintercept.com/document/2014/03/12/nsa-gchqs-
bogus replies is another good example showing that the lack of quantumtheory-hacking-tactics/) project, which includes detecting
privacy protections in the DNS is actively exploited. lookups for certain addresses and injecting bogus replies, is another
good example showing that the lack of privacy protections in the DNS
is actively exploited.
9. Legalities 9. Legalities
To our knowledge, there are no specific privacy laws for DNS data, in To our knowledge, there are no specific privacy laws for DNS data in
any country. Interpreting general privacy laws like any country. Interpreting general privacy laws, like the European
[data-protection-directive] or GDPR [9] applicable in the European Union's [data-protection-directive] or GDPR (https://gdpr.eu/tag/
Union in the context of DNS traffic data is not an easy task, and gdpr/), in the context of DNS traffic data is not an easy task, and
there is no known court precedent. See an interesting analysis in there is no known court precedent. See an interesting analysis in
[sidn-entrada]. [sidn-entrada].
10. Security Considerations 10. Security Considerations
This document is entirely about security, more precisely privacy. It This document is entirely about security -- more precisely, privacy.
just lays out the problem; it does not try to set requirements (with It just lays out the problem; it does not try to set requirements
the choices and compromises they imply), much less define solutions. (with the choices and compromises they imply), much less define
Possible solutions to the issues described here are discussed in solutions. Possible solutions to the issues described here are
other documents (currently too many to all be mentioned); see, for discussed in other documents (currently too many to all be
instance, 'Recommendations for DNS Privacy Operators' mentioned); see, for instance, "Recommendations for DNS Privacy
[I-D.ietf-dprive-bcp-op]. Operators" [RFC8932].
11. IANA Considerations 11. IANA Considerations
This document makes no requests of the IANA. This document has no IANA actions.
12. Contributions
Sara Dickinson and Stephane Bortzmeyer were the original authors on
the document, and their contribution on the initial version is
greatly appreciated.
13. Acknowledgments
Thanks to Nathalie Boulvard and to the CENTR members for the original
work that led to this document. Thanks to Ondrej Sury for the
interesting discussions. Thanks to Mohsen Souissi and John Heidemann
for proofreading and to Paul Hoffman, Matthijs Mekking, Marcos Sanz,
Tim Wicinski, Francis Dupont, Allison Mankin, and Warren Kumari for
proofreading, providing technical remarks, and making many
readability improvements. Thanks to Dan York, Suzanne Woolf, Tony
Finch, Stephen Farrell, Peter Koch, Simon Josefsson, and Frank Denis
for good written contributions. Thanks to Vittorio Bertola and
Mohamed Boucadair for a detailed review of the -bis. And thanks to
the IESG members for the last remarks.
14. References 12. References
14.1. Normative References 12.1. Normative References
[RFC1034] Mockapetris, P., "Domain names - concepts and facilities", [RFC1034] Mockapetris, P., "Domain names - concepts and facilities",
STD 13, RFC 1034, DOI 10.17487/RFC1034, November 1987, STD 13, RFC 1034, DOI 10.17487/RFC1034, November 1987,
<https://www.rfc-editor.org/info/rfc1034>. <https://www.rfc-editor.org/info/rfc1034>.
[RFC1035] Mockapetris, P., "Domain names - implementation and [RFC1035] Mockapetris, P., "Domain names - implementation and
specification", STD 13, RFC 1035, DOI 10.17487/RFC1035, specification", STD 13, RFC 1035, DOI 10.17487/RFC1035,
November 1987, <https://www.rfc-editor.org/info/rfc1035>. November 1987, <https://www.rfc-editor.org/info/rfc1035>.
[RFC6973] Cooper, A., Tschofenig, H., Aboba, B., Peterson, J., [RFC6973] Cooper, A., Tschofenig, H., Aboba, B., Peterson, J.,
Morris, J., Hansen, M., and R. Smith, "Privacy Morris, J., Hansen, M., and R. Smith, "Privacy
Considerations for Internet Protocols", RFC 6973, Considerations for Internet Protocols", RFC 6973,
DOI 10.17487/RFC6973, July 2013, DOI 10.17487/RFC6973, July 2013,
<https://www.rfc-editor.org/info/rfc6973>. <https://www.rfc-editor.org/info/rfc6973>.
[RFC7258] Farrell, S. and H. Tschofenig, "Pervasive Monitoring Is an [RFC7258] Farrell, S. and H. Tschofenig, "Pervasive Monitoring Is an
Attack", BCP 188, RFC 7258, DOI 10.17487/RFC7258, May Attack", BCP 188, RFC 7258, DOI 10.17487/RFC7258, May
2014, <https://www.rfc-editor.org/info/rfc7258>. 2014, <https://www.rfc-editor.org/info/rfc7258>.
14.2. Informative References 12.2. Informative References
[aeris-dns] [aeris-dns]
Vinot, N., "Vie privee: et le DNS alors?", (In French), Vinot, N., "Vie privée: et le DNS alors? [Privacy: what
2015, about DNS?]", February 2015,
<https://blog.imirhil.fr/vie-privee-et-le-dns-alors.html>. <https://blog.imirhil.fr/vie-privee-et-le-dns-alors.html>.
[cache-snooping-defence] [cache-snooping-defence]
ISC, "ISC Knowledge Database: DNS Cache snooping - should ISC, "DNS Cache snooping - should I be concerned?",
I be concerned?", 2018, October 2018, <https://kb.isc.org/docs/aa-00482>.
<https://kb.isc.org/docs/aa-00482>.
[castillo-garcia] [castillo-garcia]
Castillo-Perez, S. and J. Garcia-Alfaro, "Anonymous Castillo-Perez, S. and J. Garcia-Alfaro, "Anonymous
Resolution of DNS Queries", 2008, Resolution of DNS Queries", Lecture Notes in Computer
<http://deic.uab.es/~joaquin/papers/is08.pdf>. Science, Vol. 5332, DOI 10.1007/978-3-540-88873-4_5, 2008,
<https://dl.acm.org/doi/10.1007/978-3-540-88873-4_5>.
[centralisation-and-data-sovereignty] [centralisation-and-data-sovereignty]
De Filippi, P. and S. McCarthy, "Cloud Computing: De Filippi, P. and S. McCarthy, "Cloud Computing:
Centralization and Data Sovereignty", October 2012, Centralization and Data Sovereignty", European Journal of
Law and Technology, Vol. 3, No. 2, October 2012,
<https://papers.ssrn.com/sol3/ <https://papers.ssrn.com/sol3/
papers.cfm?abstract_id=2167372>. papers.cfm?abstract_id=2167372>.
[dagon-malware] [dagon-malware]
Dagon, D., "Corrupted DNS Resolution Paths: The Rise of a Dagon, D., "Corrupted DNS Resolution Paths: The Rise of a
Malicious Resolution Authority", ISC/OARC Workshop, 2007, Malicious Resolution Authority", ISC/OARC Workshop, 2007,
<https://www.dns-oarc.net/files/workshop-2007/Dagon- <https://www.dns-oarc.net/files/workshop-2007/Dagon-
Resolution-corruption.pdf>. Resolution-corruption.pdf>.
[darkreading-dns] [darkreading-dns]
Lemos, R., "Got Malware? Three Signs Revealed In DNS Lemos, R., "Got Malware? Three Signs Revealed In DNS
Traffic", InformationWeek Dark Reading, May 2013, Traffic", May 2013,
<http://www.darkreading.com/analytics/security-monitoring/ <https://www.darkreading.com/analytics/security-
got-malware-three-signs-revealed-in-dns-traffic/d/ monitoring/got-malware-three-signs-revealed-in-dns-
d-id/1139680>. traffic/d/d-id/1139680>.
[data-protection-directive] [data-protection-directive]
European Parliament, "Directive 95/46/EC of the European European Parliament, "Directive 95/46/EC of the European
Parliament and of the council on the protection of Parliament and of the Council of 24 October 1995 on the
individuals with regard to the processing of personal data protection of individuals with regard to the processing of
and on the free movement of such data", Official Journal L personal data and on the free movement of such data",
281, pp. 0031 - 0050, November 1995, <http://eur- Official Journal L 281, pp. 31-50, November 1995,
lex.europa.eu/LexUriServ/ <https://eur-lex.europa.eu/LexUriServ/
LexUriServ.do?uri=CELEX:31995L0046:EN:HTML>. LexUriServ.do?uri=CELEX:31995L0046:EN:HTML>.
[day-at-root] [day-at-root]
Castro, S., Wessels, D., Fomenkov, M., and K. Claffy, "A Castro, S., Wessels, D., Fomenkov, M., and K. Claffy, "A
Day at the Root of the Internet", ACM SIGCOMM Computer Day at the Root of the Internet", ACM SIGCOMM Computer
Communication Review, Vol. 38, Number 5, Communication Review, Vol. 38, No. 5,
DOI 10.1145/1452335.1452341, October 2008, DOI 10.1145/1452335.1452341, October 2008,
<http://www.sigcomm.org/sites/default/files/ccr/ <https://www.sigcomm.org/sites/default/files/ccr/
papers/2008/October/1452335-1452341.pdf>. papers/2008/October/1452335-1452341.pdf>.
[denis-edns-client-subnet] [denis-edns-client-subnet]
Denis, F., "Security and privacy issues of edns-client- Denis, F., "Security and privacy issues of edns-client-
subnet", August 2013, subnet", August 2013,
<https://00f.net/2013/08/07/edns-client-subnet/>. <https://00f.net/2013/08/07/edns-client-subnet/>.
[ditl] CAIDA, "A Day in the Life of the Internet (DITL)", 2002, [ditl] CAIDA, "A Day in the Life of the Internet (DITL)",
<http://www.caida.org/projects/ditl/>. <https://www.caida.org/projects/ditl/>.
[dns-de-nat]
Orevi, L., Herzberg, A., Zlatokrilov, H., and D. Sigron,
"DNS-DNS: DNS-based De-NAT Scheme", January 2017,
<https://www.researchgate.net/publication/320322146_DNS-
DNS_DNS-based_De-NAT_Scheme>.
[dns-footprint] [dns-footprint]
Stoner, E., "DNS Footprint of Malware", OARC Workshop, Stoner, E., "DNS Footprint of Malware", OARC Workshop,
October 2010, <https://www.dns-oarc.net/files/workshop- October 2010, <https://www.dns-oarc.net/files/workshop-
201010/OARC-ers-20101012.pdf>. 201010/OARC-ers-20101012.pdf>.
[dns-over-encryption] [dns-over-encryption]
Lu, C., Liu, B., Li, Z., Hao, S., Duan, H., Zhang, M., Lu, C., Liu, B., Li, Z., Hao, S., Duan, H., Zhang, M.,
Leng, C., Liu, Y., Zhang, Z., and J. Wu, "An End-to-End, Leng, C., Liu, Y., Zhang, Z., and J. Wu, "An End-to-End,
Large-Scale Measurement of DNS-over-Encryption", IMC Large-Scale Measurement of DNS-over-Encryption: How Far
'19 Amsterdam, Netherlands, DOI 10.1145/3355369.3355580, Have We Come?", IMC '19: Proceedings of the Internet
October 2019, Measurement Conference, pp. 22-35,
<http://dl.acm.org/citation.cfm?id=3355369.3355580>. DOI 10.1145/3355369.3355580, October 2019,
<https://dl.acm.org/citation.cfm?id=3355369.3355580>.
[dnsmezzo] [dnsmezzo] Bortzmeyer, S., "DNSmezzo", <http://www.dnsmezzo.net/>.
Bortzmeyer, S., "DNSmezzo", 2009,
<http://www.dnsmezzo.net/>.
[EDDI] EDDI, "Encrypted DNS Deployment Initiative", 2020, [DNSOP-RESOLVER]
Sood, P., Arends, R., and P. Hoffman, "DNS Resolver
Information Self-publication", Work in Progress, Internet-
Draft, draft-ietf-dnsop-resolver-information-01, 11
February 2020, <https://datatracker.ietf.org/doc/html/
draft-ietf-dnsop-resolver-information-01>.
[DPRIVE-DNSOQUIC]
Huitema, C., Dickinson, S., and A. Mankin, "Specification
of DNS over Dedicated QUIC Connections", Work in Progress,
Internet-Draft, draft-ietf-dprive-dnsoquic-03, 12 July
2021, <https://datatracker.ietf.org/doc/html/draft-ietf-
dprive-dnsoquic-03>.
[EDDI] EDDI, "Encrypted DNS Deployment Initiative",
<https://www.encrypted-dns.org>. <https://www.encrypted-dns.org>.
[fangming-hori-sakurai] [fangming-hori-sakurai]
Fangming, Z., Hori, Y., and K. Sakurai, "Analysis of Zhao, F., Hori, Y., and K. Sakurai, "Analysis of Privacy
Privacy Disclosure in DNS Query", 2007 International Disclosure in DNS Query", MUE '07: Proceedings of the 2007
Conference on Multimedia and Ubiquitous Engineering (MUE International Conference on Multimedia and Ubiquitous
2007), Seoul, Korea, ISBN: 0-7695-2777-9, pp. 952-957, Engineering, pp. 952-957, DOI 10.1109/MUE.2007.84,
DOI 10.1109/MUE.2007.84, April 2007, ISBN 0-7695-2777-9, April 2007,
<http://dl.acm.org/citation.cfm?id=1262690.1262986>. <https://dl.acm.org/citation.cfm?id=1262690.1262986>.
[federrath-fuchs-herrmann-piosecny] [federrath-fuchs-herrmann-piosecny]
Federrath, H., Fuchs, K., Herrmann, D., and C. Piosecny, Federrath, H., Fuchs, K.-P., Herrmann, D., and C.
"Privacy-Preserving DNS: Analysis of Broadcast, Range Piosecny, "Privacy-Preserving DNS: Analysis of Broadcast,
Queries and Mix-based Protection Methods", Computer Range Queries and Mix-based Protection Methods", ESORICS
Security ESORICS 2011, Springer, page(s) 665-683, 2011, pp. 665-683, DOI 10.1007/978-3-642-23822-2_36,
ISBN 978-3-642-23821-5, 2011, <https://svs.informatik.uni- ISBN 978-3-642-23822-2, 2011, <https://svs.informatik.uni-
hamburg.de/publications/2011/2011-09-14_FFHP_PrivacyPreser hamburg.de/publications/2011/2011-09-14_FFHP_PrivacyPreser
vingDNS_ESORICS2011.pdf>. vingDNS_ESORICS2011.pdf>.
[getdns] getdns, "getdns - A modern asynchronous DNS API", January [getdns] "getdns", <https://getdnsapi.net>.
2020, <https://getdnsapi.net>.
[grangeia.snooping] [grangeia.snooping]
Grangeia, L., "DNS Cache Snooping or Snooping the Cache Grangeia, L., "Cache Snooping or Snooping the Cache for
for Fun and Profit", 2005, Fun and Profit", 2005,
<https://www.semanticscholar.org/paper/Cache-Snooping-or- <https://www.semanticscholar.org/paper/Cache-Snooping-or-
Snooping-the-Cache-for-Fun-and- Snooping-the-Cache-for-Fun-and-
1-Grangeia/9b22f606e10b3609eafbdcbfc9090b63be8778c3>. 1-Grangeia/9b22f606e10b3609eafbdcbfc9090b63be8778c3>.
[herrmann-reidentification] [herrmann-reidentification]
Herrmann, D., Gerber, C., Banse, C., and H. Federrath, Herrmann, D., Gerber, C., Banse, C., and H. Federrath,
"Analyzing Characteristic Host Access Patterns for Re- "Analyzing Characteristic Host Access Patterns for Re-
Identification of Web User Sessions", Identification of Web User Sessions", Lecture Notes in
DOI 10.1007/978-3-642-27937-9_10, 2012, <http://epub.uni- Computer Science, Vol. 7127,
DOI 10.1007/978-3-642-27937-9_10, 2012, <https://epub.uni-
regensburg.de/21103/1/Paper_PUL_nordsec_published.pdf>. regensburg.de/21103/1/Paper_PUL_nordsec_published.pdf>.
[I-D.ietf-dnsop-resolver-information]
Sood, P., Arends, R., and P. Hoffman, "DNS Resolver
Information Self-publication", draft-ietf-dnsop-resolver-
information-01 (work in progress), February 2020.
[I-D.ietf-dprive-bcp-op]
Dickinson, S., Overeinder, B., Rijswijk-Deij, R., and A.
Mankin, "Recommendations for DNS Privacy Service
Operators", draft-ietf-dprive-bcp-op-14 (work in
progress), July 2020.
[I-D.ietf-dprive-dnsoquic]
Huitema, C., Mankin, A., and S. Dickinson, "Specification
of DNS over Dedicated QUIC Connections", draft-ietf-
dprive-dnsoquic-01 (work in progress), October 2020.
[I-D.ietf-quic-transport]
Iyengar, J. and M. Thomson, "QUIC: A UDP-Based Multiplexed
and Secure Transport", draft-ietf-quic-transport-34 (work
in progress), January 2021.
[morecowbell] [morecowbell]
Grothoff, C., Wachs, M., Ermert, M., and J. Appelbaum, Grothoff, C., Wachs, M., Ermert, M., and J. Appelbaum,
"NSA's MORECOWBELL: Knell for DNS", GNUnet e.V., January "NSA's MORECOWBELL: Knell for DNS", January 2015, <https:/
2015, <https://pdfs.semanticscholar.org/2610/2b99bdd6a258a /pdfs.semanticscholar.org/2610/2b99bdd6a258a98740af8217ba8
98740af8217ba8da8a1e4fa.pdf>. da8a1e4fa.pdf>.
[packetq] DNS-OARC, "PacketQ, a simple tool to make SQL-queries [packetq] DNS-OARC, "A tool that provides a basic SQL-frontend to
against PCAP-files", 2011, PCAP-files", Release 1.4.3, commit 29a8288, October 2020,
<https://github.com/DNS-OARC/PacketQ>. <https://github.com/DNS-OARC/PacketQ>.
[passive-dns] [passive-dns]
Weimer, F., "Passive DNS Replication", April 2005, Weimer, F., "Passive DNS Replication", 17th Annual FIRST
Conference, April 2005,
<https://www.first.org/conference/2005/papers/florian- <https://www.first.org/conference/2005/papers/florian-
weimer-slides-1.pdf>. weimer-slides-1.pdf>.
[pitfalls-of-dns-encryption] [pitfalls-of-dns-encryption]
Shulman, H., "Pretty Bad Privacy:Pitfalls of DNS Shulman, H., "Pretty Bad Privacy: Pitfalls of DNS
Encryption", <https://dl.acm.org/citation.cfm?id=2665959>. Encryption", WPES '14: Proceedings of the 13th Workshop on
Privacy in the Electronic Society, pp. 191-200,
DOI 10.1145/2665943.2665959, November 2014,
<https://dl.acm.org/citation.cfm?id=2665959>.
[prism] Wikipedia, "PRISM (surveillance program)", July 2015, [prism] Wikipedia, "PRISM (surveillance program)", July 2015,
<https://en.wikipedia.org/w/index.php?title=PRISM_(surveil <https://en.wikipedia.org/w/index.php?title=PRISM_(surveil
lance_program)&oldid=673789455>. lance_program)&oldid=673789455>.
[RFC3552] Rescorla, E. and B. Korver, "Guidelines for Writing RFC [RFC3552] Rescorla, E. and B. Korver, "Guidelines for Writing RFC
Text on Security Considerations", BCP 72, RFC 3552, Text on Security Considerations", BCP 72, RFC 3552,
DOI 10.17487/RFC3552, July 2003, DOI 10.17487/RFC3552, July 2003,
<https://www.rfc-editor.org/info/rfc3552>. <https://www.rfc-editor.org/info/rfc3552>.
skipping to change at page 25, line 5 skipping to change at line 1141
(DTLS)", BCP 195, RFC 7525, DOI 10.17487/RFC7525, May (DTLS)", BCP 195, RFC 7525, DOI 10.17487/RFC7525, May
2015, <https://www.rfc-editor.org/info/rfc7525>. 2015, <https://www.rfc-editor.org/info/rfc7525>.
[RFC7624] Barnes, R., Schneier, B., Jennings, C., Hardie, T., [RFC7624] Barnes, R., Schneier, B., Jennings, C., Hardie, T.,
Trammell, B., Huitema, C., and D. Borkmann, Trammell, B., Huitema, C., and D. Borkmann,
"Confidentiality in the Face of Pervasive Surveillance: A "Confidentiality in the Face of Pervasive Surveillance: A
Threat Model and Problem Statement", RFC 7624, Threat Model and Problem Statement", RFC 7624,
DOI 10.17487/RFC7624, August 2015, DOI 10.17487/RFC7624, August 2015,
<https://www.rfc-editor.org/info/rfc7624>. <https://www.rfc-editor.org/info/rfc7624>.
[RFC7626] Bortzmeyer, S., "DNS Privacy Considerations", RFC 7626,
DOI 10.17487/RFC7626, August 2015,
<https://www.rfc-editor.org/info/rfc7626>.
[RFC7721] Cooper, A., Gont, F., and D. Thaler, "Security and Privacy [RFC7721] Cooper, A., Gont, F., and D. Thaler, "Security and Privacy
Considerations for IPv6 Address Generation Mechanisms", Considerations for IPv6 Address Generation Mechanisms",
RFC 7721, DOI 10.17487/RFC7721, March 2016, RFC 7721, DOI 10.17487/RFC7721, March 2016,
<https://www.rfc-editor.org/info/rfc7721>. <https://www.rfc-editor.org/info/rfc7721>.
[RFC7754] Barnes, R., Cooper, A., Kolkman, O., Thaler, D., and E. [RFC7754] Barnes, R., Cooper, A., Kolkman, O., Thaler, D., and E.
Nordmark, "Technical Considerations for Internet Service Nordmark, "Technical Considerations for Internet Service
Blocking and Filtering", RFC 7754, DOI 10.17487/RFC7754, Blocking and Filtering", RFC 7754, DOI 10.17487/RFC7754,
March 2016, <https://www.rfc-editor.org/info/rfc7754>. March 2016, <https://www.rfc-editor.org/info/rfc7754>.
skipping to change at page 26, line 14 skipping to change at line 1199
[RFC8744] Huitema, C., "Issues and Requirements for Server Name [RFC8744] Huitema, C., "Issues and Requirements for Server Name
Identification (SNI) Encryption in TLS", RFC 8744, Identification (SNI) Encryption in TLS", RFC 8744,
DOI 10.17487/RFC8744, July 2020, DOI 10.17487/RFC8744, July 2020,
<https://www.rfc-editor.org/info/rfc8744>. <https://www.rfc-editor.org/info/rfc8744>.
[RFC8890] Nottingham, M., "The Internet is for End Users", RFC 8890, [RFC8890] Nottingham, M., "The Internet is for End Users", RFC 8890,
DOI 10.17487/RFC8890, August 2020, DOI 10.17487/RFC8890, August 2020,
<https://www.rfc-editor.org/info/rfc8890>. <https://www.rfc-editor.org/info/rfc8890>.
[RFC8932] Dickinson, S., Overeinder, B., van Rijswijk-Deij, R., and
A. Mankin, "Recommendations for DNS Privacy Service
Operators", BCP 232, RFC 8932, DOI 10.17487/RFC8932,
October 2020, <https://www.rfc-editor.org/info/rfc8932>.
[RFC9000] Iyengar, J., Ed. and M. Thomson, Ed., "QUIC: A UDP-Based
Multiplexed and Secure Transport", RFC 9000,
DOI 10.17487/RFC9000, May 2021,
<https://www.rfc-editor.org/info/rfc9000>.
[ripe-qname-measurements] [ripe-qname-measurements]
Vries, W., "Making the DNS More Private with QNAME de Vries, W., "Making the DNS More Private with QNAME
Minimisation", April 2019, Minimisation", April 2019,
<https://labs.ripe.net/Members/wouter_de_vries/make-dns-a- <https://labs.ripe.net/Members/wouter_de_vries/make-dns-a-
bit-more-private-with-qname-minimisation>. bit-more-private-with-qname-minimisation>.
[sidn-entrada] [sidn-entrada]
Hesselman, C., Jansen, J., Wullink, M., Vink, K., and M. Hesselman, C., Jansen, J., Wullink, M., Vink, K., and M.
Simon, "A privacy framework for 'DNS big data' Simon, "A privacy framework for 'DNS big data'
applications", November 2014, applications", November 2014,
<https://www.sidnlabs.nl/downloads/ <https://www.sidnlabs.nl/downloads/
yBW6hBoaSZe4m6GJc_0b7w/2211058ab6330c7f3788141ea19d3db7/ yBW6hBoaSZe4m6GJc_0b7w/2211058ab6330c7f3788141ea19d3db7/
SIDN_Labs_Privacyraamwerk_Position_Paper_V1.4_ENG.pdf>. SIDN_Labs_Privacyraamwerk_Position_Paper_V1.4_ENG.pdf>.
[thomas-ditl-tcp] [thomas-ditl-tcp]
Thomas, M. and D. Wessels, "An Analysis of TCP Traffic in Thomas, M. and D. Wessels, "An Analysis of TCP Traffic in
Root Server DITL Data", DNS-OARC 2014 Fall Workshop, Root Server DITL Data", DNS-OARC 2014 Fall Workshop,
October 2014, <https://indico.dns- October 2014, <https://indico.dns-
oarc.net/event/20/session/2/contribution/15/material/ oarc.net/event/20/session/2/contribution/15/material/
slides/1.pdf>. slides/1.pdf>.
[tor-leak] [tor-leak] Tor, "Tor FAQs: I keep seeing these warnings about SOCKS
Tor, "DNS leaks in Tor", 2013, and DNS information leaks. Should I worry?",
<https://www.torproject.org/docs/ <https://www.torproject.org/docs/
faq.html.en#WarningsAboutSOCKSandDNSInformationLeaks>. faq.html.en#WarningsAboutSOCKSandDNSInformationLeaks>.
[yanbin-tsudik] [yanbin-tsudik]
Yanbin, L. and G. Tsudik, "Towards Plugging Privacy Leaks Yanbin, L. and G. Tsudik, "Towards Plugging Privacy Leaks
in the Domain Name System", October 2009, in Domain Name System", June 2010,
<http://arxiv.org/abs/0910.2472>. <https://arxiv.org/abs/0910.2472>.
14.3. URIs
[1] https://lists.dns-oarc.net/pipermail/dns-
operations/2016-January/014143.html
[2] http://netres.ec/?b=11B99BD
[3] https://developers.google.com/speed/public-dns
[4] https://developers.cloudflare.com/1.1.1.1/setting-up-1.1.1.1/
[5] https://www.quad9.net
[6] https://developers.google.com/speed/public-dns/privacy
[7] https://www.alexa.com/topsites
[8] https://theintercept.com/document/2014/03/12/nsa-gchqs-
quantumtheory-hacking-tactics/
[9] https://www.eugdpr.org/the-regulation.html
Appendix A. Updates since RFC7626
Update many references; Added discussions of encrypted transports
including DoT and DoH; Added section on DNS payload; Added section on
authentication of servers; Added section on blocking of services.
With the publishing of RFC7816 on QNAME minimisation, text,
references, and initial attempts to measure deployment were added to
reflect this. The text and references on the Snowden revelations
were updated.
The "Risks overview" section was changed to "Scope" to help clarify
the risks being considered. Text was adding on cellular network DNS,
blocking and security. Considerations for recursive resolvers were
collected and placed together. Addded a discussion on resolver
selection.
Appendix B. Changelog
draft-ietf-dprive-rfc7626-bis-08
o Second batch of Editorial updates from IESG last call
draft-ietf-dprive-rfc7626-bis-07
o First batch of Editorial updates from IESG last call
draft-ietf-dprive-rfc7626-bis-06
o Removed Sara and Stephane as editors, made chairs as Editor.
o Replaced the text in 6.1.1.2 with the text from the -04 version.
o Clarified text about resolver selection in 6.1.1.
draft-ietf-dprive-rfc7626-bis-05
o Editorial updates from second IESG last call
o Section renumbering as suggested by Vittorio Bertola
draft-ietf-dprive-rfc7626-bis-04
o Tsvart review: Add reference to DNS-over-QUIC, fix typo.
o Secdir review: Add text in Section 3 on devices using many
networks.
o Update bullet in 3.4.1 on cellular encryption.
o Section 3.5.1.1 - re-work the section to try to address multiple
comments.
o Section 3.5.1.4 - remove this section as now covered by 3.5.1.1.
o Section 3.5.1.5.2 - Remove several paragraphs and more directly
reference RFC8484 by including bullet points quoting text from
Section 8.2 of RFC8484. Retain the last 2 paragraphs as they are
information for users, not implementors.
o Section 3.4.2 - some minor updates made based on specific
comments.
draft-ietf-dprive-rfc7626-bis-03
o Address 2 minor nits (typo in section 3.4.1 and adding an IANA
section)
o Minor updates from AD review
draft-ietf-dprive-rfc7626-bis-02
o Numerous editorial corrections thanks to Mohamed Boucadair and
* Minor additions to Scope section
* New text on cellular network DNS
o Additional text from Vittorio Bertola on blocking and security
draft-ietf-dprive-rfc7626-bis-01
o Re-structure section 3.5 (was 2.5)
* Collect considerations for recursive resolvers together
* Re-work several sections here to clarify their context (e.g.,
'Rogue servers' becomes 'Active attacks on resolver
configuration')
* Add discussion of resolver selection
o Update text and old reference on Snowdon revelations.
o Add text on and references to QNAME minimisation RFC and
deployment measurements
o Correct outdated references
o Clarify scope by adding a Scope section (was Risks overview)
o Clarify what risks are considered in section 3.4.2
draft-ietf-dprive-rfc7626-bis-00
o Rename after WG adoption
o Use DoT acronym throughout
o Minor updates to status of deployment and other drafts
draft-bortzmeyer-dprive-rfc7626-bis-02
o Update various references and fix some nits.
draft-bortzmeyer-dprive-rfc7626-bis-01
o Update reference for dickinson-bcp-op to draft-dickinson-dprive-
bcp-op
draft-borztmeyer-dprive-rfc7626-bis-00: Appendix A. Updates since RFC 7626
Initial commit. Differences to RFC7626: Many references were updated. Discussions of encrypted transports,
including DoT and DoH, and sections on DNS payload, authentication of
servers, and blocking of services were added. With the publishing of
[RFC7816] on QNAME minimization, text, references, and initial
attempts to measure deployment were added to reflect this. The text
and references on the Snowden revelations were updated.
o Update many references The "Risks Overview" section was changed to "Scope" to help clarify
the risks being considered. Text on cellular network DNS, blocking,
and security was added. Considerations for recursive resolvers were
collected and placed together. A discussion on resolver selection
was added.
o Add discussions of encrypted transports including DoT and DoH Acknowledgments
o Add section on DNS payload Thanks to Nathalie Boulvard and to the CENTR members for the original
work that led to this document. Thanks to Ondrej Sury for the
interesting discussions. Thanks to Mohsen Souissi and John Heidemann
for proofreading and to Paul Hoffman, Matthijs Mekking, Marcos Sanz,
Francis Dupont, Allison Mankin, and Warren Kumari for proofreading,
providing technical remarks, and making many readability
improvements. Thanks to Dan York, Suzanne Woolf, Tony Finch, Stephen
Farrell, Peter Koch, Simon Josefsson, and Frank Denis for good
written contributions. Thanks to Vittorio Bertola and Mohamed
Boucadair for a detailed review of the -bis. And thanks to the IESG
members for the last remarks.
o Add section on authentication of servers Contributions
o Add section on blocking of services Sara Dickinson and Stephane Bortzmeyer were the original authors of
the document, and their contribution to the initial draft of this
document is greatly appreciated.
Author's Address Author's Address
Tim Wicinski (editor) Tim Wicinski (editor)
Elkins, WV 26241 Elkins, WV 26241
USA United States of America
Email: tjw.ietf@gmail.com Email: tjw.ietf@gmail.com
 End of changes. 168 change blocks. 
570 lines changed or deleted 457 lines changed or added

This html diff was produced by rfcdiff 1.48. The latest version is available from http://tools.ietf.org/tools/rfcdiff/