rfc9665.original.xml   rfc9665.xml 
<?xml version="1.0" encoding="UTF-8"?> <?xml version="1.0" encoding="UTF-8"?>
<!DOCTYPE rfc SYSTEM "rfc2629-xhtml.ent">
<rfc category="std" submissionType="IETF" docName="draft-ietf-dnssd-srp-25" ipr= <!DOCTYPE rfc [
"trust200902" <!ENTITY nbsp "&#160;">
xmlns:xi="http://www.w3.org/2001/XInclude" version="3" <!ENTITY zwsp "&#8203;">
scripts="Common,Latin" sortRefs="false" consensus="true" <!ENTITY nbhy "&#8209;">
symRefs="true" tocDepth="4" tocInclude="true" xml:lang="en"> <!ENTITY wj "&#8288;">
]>
<rfc category="std" submissionType="IETF" docName="draft-ietf-dnssd-srp-latest"
number="9665" ipr="trust200902" xmlns:xi="http://www.w3.org/2001/XInclude" obsol
etes="" updates="" version="3" sortRefs="true" consensus="true" symRefs="true" t
ocDepth="4" tocInclude="true" xml:lang="en">
<front> <front>
<title abbrev='Service Registration Protocol'>Service Registration Protocol <title abbrev="Service Registration Protocol">Service Registration Protocol
for DNS-Based Service Discovery</title> for DNS-Based Service Discovery</title>
<author initials="T" surname="Lemon" fullname="Ted Lemon"> <seriesInfo name="RFC" value="9665"/>
<author initials="T." surname="Lemon" fullname="Ted Lemon">
<organization>Apple Inc.</organization> <organization>Apple Inc.</organization>
<address> <address>
<postal> <postal>
<street>One Apple Park Way</street> <street>One Apple Park Way</street>
<city>Cupertino</city> <city>Cupertino</city>
<region>California</region> <region>CA</region>
<code>95014</code> <code>95014</code>
<country>USA</country> <country>United States of America</country>
</postal> </postal>
<email>mellon@fugue.com</email> <email>mellon@fugue.com</email>
</address> </address>
</author> </author>
<author initials='S' surname='Cheshire' fullname='Stuart Cheshire'> <author initials="S." surname="Cheshire" fullname="Stuart Cheshire">
<organization>Apple Inc.</organization> <organization>Apple Inc.</organization>
<address> <address>
<postal> <postal>
<street>One Apple Park Way</street> <street>One Apple Park Way</street>
<city>Cupertino</city> <city>Cupertino</city>
<region>California</region> <region>CA</region>
<code>95014</code> <code>95014</code>
<country>USA</country> <country>United States of America</country>
</postal> </postal>
<phone>+1 408 974 3207</phone> <phone>+1 408 974 3207</phone>
<email>cheshire@apple.com</email> <email>cheshire@apple.com</email>
</address> </address>
</author> </author>
<date>March 4, 2024</date> <date month="October" year="2024"/>
<area>Internet</area> <area>INT</area>
<workgroup>Internet Engineering Task Force</workgroup> <workgroup>dnssd</workgroup>
<keyword>Multicast DNS</keyword> <keyword>Multicast DNS</keyword>
<keyword>DNS-Based Service Discovery</keyword> <keyword>DNS-Based Service Discovery</keyword>
<keyword>DNS Update</keyword> <keyword>DNS Update</keyword>
<keyword>SIG(0)</keyword> <keyword>SIG(0)</keyword>
<abstract>
<t>
The Service Registration Protocol for DNS-Based Service Discovery uses t
he standard DNS Update mechanism to enable DNS-Based
Service Discovery using only unicast packets. This makes it possible to
deploy DNS Service Discovery without multicast,
which greatly improves scalability and improves performance on networks
where multicast service is not an optimal choice,
particularly IEEE 802.11 (Wi&nbhy;Fi) and IEEE 802.15.4 networks. DNS&n
bhy;SD Service registration
uses public keys and SIG(0) to allow services to defend their registrati
ons.
<abstract>
<t>The Service Registration Protocol (SRP) for DNS-based Service
Discovery (DNS&nbhy;SD) uses the standard DNS Update mechanism to
enable DNS&nbhy;SD using only unicast packets. This makes it possible
to deploy DNS&nbhy;SD without multicast, which greatly improves
scalability and improves performance on networks where multicast
service is not an optimal choice, particularly IEEE 802.11
(Wi-Fi) and IEEE 802.15.4 networks. DNS&nbhy;SD Service
registration uses public keys and SIG(0) to allow services to defend
their registrations.
</t> </t>
</abstract> </abstract>
<note removeInRFC="true">
<name>About This Document</name>
<t>
The latest revision of this draft can be found at <eref target="https://
dnssd-wg.github.io/draft-ietf-dnssd-srp/draft-ietf-dnssd-srp.html"/>.
Status information for this document may be found at <eref target="https
://datatracker.ietf.org/doc/draft-ietf-dnssd-srp/"/>.
</t>
<t>
Discussion of this document takes place on the
DNS-SD Working Group mailing list (<eref target="mailto:dnssd@ietf.org"/
>),
which is archived at <eref target="https://mailarchive.ietf.org/arch/bro
wse/dnssd/"/>.
Subscribe at <eref target="https://www.ietf.org/mailman/listinfo/dnssd/"
/>.
</t>
<t>Source for this draft and an issue tracker can be found at
<eref target="https://github.com/dnssd-wg/draft-ietf-dnssd-srp"/>.</t>
</note>
</front> </front>
<middle> <middle>
<section> <section>
<name>Introduction</name> <name>Introduction</name>
<t> <t>
DNS&nbhy;SD <xref target="RFC6763"/>
<xref target="RFC6763">DNS-Based Service Discovery</xref> is a component is a component of Zero Configuration Networking
of Zero Configuration Networking <xref target="RFC6760"/>
<xref target="RFC6760"/> <xref target="ZC"/> <xref target="I-D.cheshire- <xref target="ZC"/>
dnssd-roadmap"/>.</t> <xref target="I-D.cheshire-dnssd-roadmap"/>.</t>
<t>
This document describes an enhancement to <xref target="RFC6763">DNS-Bas
ed Service Discovery</xref> (DNS&nbhy;SD) that
allows servers to register the services they offer using the DNS protocol
rather than using <xref target="RFC6762">Multicast
DNS</xref> (mDNS). There is already a large installed base of DNS&nbhy;S
D clients that can discover services using the DNS
protocol (e.g. Android, Windows, Linux, Apple).</t>
<t> <t>
This document is intended for three audiences: implementors of software This document describes an enhancement to DNS&nbhy;SD that
that provides services that should be advertised allows servers to register the services they offer using the DNS protocol
using DNS&nbhy;SD, implementors of DNS servers that will be used in cont over unicast rather than using Multicast DNS (mDNS) <xref target="RFC6762
exts where DNS&nbhy;SD registration is needed, and "/>.
administrators of networks where DNS&nbhy;SD service is required. The d There is already a large installed base of DNS&nbhy;SD
ocument is expected to provide sufficient clients that can discover services using the DNS
information to allow interoperable implementation of the registration pr protocol (e.g., Android, Windows, Linux, Apple operating systems).</t>
otocol.</t>
<t> <t>
DNS-Based Service Discovery (DNS&nbhy;SD) allows services to advertise t This document is intended for three audiences: Implementers of software
he fact that they provide service, and to provide that provides services that should be advertised
the information required to access that service. DNS&nbhy;SD clients ca using DNS&nbhy;SD, implementers of authoritative DNS servers that will
n then discover the set of services of a particular be used in contexts where DNS&nbhy;SD registration is needed, and
administrators of networks where DNS&nbhy;SD service is required.
The document is expected to provide sufficient
information to allow interoperable implementation
of the Service Registration Protocol.</t>
<t>
<!--[rfced] Should "services" be "servers" here to match previous,
similar text? And perhaps avoiding the two "provide" uses so
close together would be helpful for the reader?
Original:
DNS-Based Service Discovery (DNS-SD) allows services to advertise
the fact that they provide service, and to provide the
information required to access that service.
Perhaps:
DNS-SD allows servers to advertise the fact that they provide
service and to share the information required to access that
service.
[Ted]
The term "server" would be confusing here. We really do mean services. That's wh
at is discovered
using service discovery. I've reworded the sentence to make this (I hope!) clear
er.
A server is the thing that provides the service. So servers register services, o
r more correctly the SRP requester registers a service on behalf of a server.
Another point of confusion is "server" versus "DNS server" so I've clarified tha
t
throughout by replacing "server" with "authoritative DNS server" when it's a DNS
server and
not a server publishing a service.
-->
DNS&nbhy;SD allows servers to publish
the information required to access the services they provide.
DNS&nbhy;SD clients can then discover the set of services of a particula
r
type that are available. They can then select a service from among thos e that are available and obtain the information type that are available. They can then select a service from among thos e that are available and obtain the information
required to use it. Although DNS Service Discovery (DNS-SD) using the D required to use it. Although DNS&nbhy;SD using the DNS protocol
NS protocol (as opposed to mDNS) can be more efficient and versatile, it is can be more efficient and versatile than using mDNS, it is
not common in practice, because of the difficulties associated with upda not common in practice because of the difficulties associated with updat
ting authoritative DNS services with service ing authoritative DNS services with service
information.</t> information.</t>
<t> <t>
Existing practice for updating DNS zones is to either manually enter new The existing practice for updating DNS zones is either to enter new data
data, or else use DNS Update manually or to use DNS Update
<xref target="RFC2136"/>. Unfortunately DNS Update requires either that t <xref target="RFC2136"/>. Unfortunately, DNS Update requires either:</t>
he authoritative DNS server automatically trust <ul>
updates, or else that the DNS Update requestor have some kind of shared s <li>that the authoritative DNS server automatically trust
ecret or public key that is known to the DNS server updates or</li>
and can be used to authenticate the update. Furthermore, DNS Update can <li>that the DNS Update requester have some kind of shared secret
be a fairly chatty process, requiring multiple or public key that is known to the authoritative DNS server
round trips with different conditional predicates to complete the update and can be used to authenticate the update.</li></ul>
process.</t> <t>Furthermore, DNS Update can be a fairly chatty process, requiring mult
iple
roundtrips with different conditional predicates to complete the update p
rocess.</t>
<!-- [Ted] conditional predicates are a thing in DNS Update. It's not that there
are
predicates that are conditional on the use of DNS update. So putting a '-'
here
changes the meaning in a particularly confusing way. I think saying DNS Upd
ate
here is actually redundant anyway, and the sentence is clearer without it.
-->
<t> <t>
The Service Registration Protocol (SRP) adds a set of default heuristics The Service Registration Protocol (SRP) adds a set of default
for processing DNS updates that eliminates the need for DNS update heuristics for processing DNS updates that eliminates the need for
conditional predicates: instead, the SRP registrar (a DNS server that sup conditional predicates.
ports SRP updates) has a set of default predicates Instead, the SRP registrar (an authoritative DNS server
that are applied to the update, and the update either succeeds entirely, that supports SRP Updates) has a set of default predicates
or fails in a way that allows the requestor to know that are applied to the update; and the update either succeeds entirely o
r fails in a way that allows the requester to know
what went wrong and construct a new update.</t> what went wrong and construct a new update.</t>
<t> <t>
SRP also adds a feature called First-Come, First-Served (FCFS) Naming, wh SRP also adds a feature called "First Come, First Served Naming" (or "FCF
ich allows the requestor to claim a name that is S Naming"), which allows the requester to:</t>
not yet in use, and, using SIG(0) <xref target="RFC2931"/>, to authentica <ul><li>claim a name that is
te both the initial claim and subsequent not yet in use, and</li>
updates. This prevents name conflicts, since a second SRP requestor attem <li>authenticate, using SIG(0) <xref target="RFC2931"/>,
pting to claim the same name will not possess the both the initial claim
SIG(0) key used by the first requestor to claim it, and so its claim will (to ensure it has not been modified in transit)
be rejected and the second requestor will have to and subsequent updates
(to ensure they come from the same entity that performed the initial clai
m).</li></ul>
<t>This prevents a new service instance from "stealing" a name that is al
ready in use:
A second SRP requester attempting to claim an existing name will not poss
ess the
SIG(0) key used by the first requester to claim it. Because of this, its
claim will be rejected. This will force it to
choose a new name.</t> choose a new name.</t>
<t> <t>
It is important to understand that "authenticate" here just means that we can tell that an update came from the same source It is important to understand that "authenticate" here just means that we can tell that an update came from the same source
as the original registration. We have not established trust. This has imp ortant implications for what we can and can't do as the original registration. We have not established trust. This has imp ortant implications for what we can and can't do
with data the client sends us. You will notice as you read this document with data the SRP requester sends us.
that we only support adding a very restricted set You will notice as you read this document that
we only support adding a very restricted set
of records, and the content of those records is further constrained.</t> of records, and the content of those records is further constrained.</t>
<t> <t>
The reason for this is precisely that we have not established trust. So w The reason for this is precisely that we have not established trust. So,
e can only publish information that we feel safe in we can only publish information that we feel safe in
publishing even though we do not have any basis for trusting the requesto publishing even though we do not have any basis for trusting the requeste
r. We reason that mDNS <xref target="RFC6762"/> r.
allows arbitrary hosts on a single IP link to advertise services <xref ta We reason that mDNS <xref target="RFC6762"/> allows
rget="RFC6763"/>, relying on whatever service is arbitrary hosts on a single IP link to advertise services
<xref target="RFC6763"/>, relying on whatever service is
advertised to provide authentication as a part of its protocol rather tha n in the service advertisement.</t> advertised to provide authentication as a part of its protocol rather tha n in the service advertisement.</t>
<t> <t>
This is considered reasonably safe because it requires physical presence on the network in order to advertise. An off-network This is considered reasonably safe because it requires physical presence on the network in order to advertise. An off-network
mDNS attack is simply not possible. Our goal with this specification is t o impose similar constraints. Because of this you will mDNS attack is simply not possible. Our goal with this specification is t o impose similar constraints. Therefore, you will
see in <xref target="add_validation"/> that a very restricted set of reco rds with a very restricted set of relationships are see in <xref target="add_validation"/> that a very restricted set of reco rds with a very restricted set of relationships are
allowed. You will also see in <xref target="source_validation"/> that we give advice on how to prevent off-network attacks.</t> allowed. You will also see in <xref target="source_validation"/> that we give advice on how to prevent off-network attacks.</t>
<t> <t>
This leads us to the disappointing observation that this protocol is not a mechanism for adding arbitrary information to This leads us to the disappointing observation that this protocol is not a mechanism for adding arbitrary information to
DNS zones. We have not evaluated the security properties of adding, for e xample, an SOA record, an MX record, or a CNAME DNS zones. We have not evaluated the security properties of adding, for e xample, an SOA record, an MX record, or a CNAME
record, and so these are forbidden. A future protocol specification might record; therefore, these are forbidden.
include analyses for other records, and extend Future updates to this specification might include analyses for other rec
the set of records that can be registered here. Or it might require estab ords
lishment of trust, and add an authorization model and extend the set of records and/or record content that can be registere
to the authentication model we now have. But this is work for a future do d here.
cument.</t> Or it might require establishment of trust, and add an authorization mode
l
to the authentication model we now have.
But that is work for a future document.</t>
<t> <t>
Finally, SRP adds the concept of a 'lease,' similar to leases in Dynamic Finally, SRP adds the concept of a "lease" <xref target="RFC9664"/>,
Host Configuration Protocol analogous to leases in DHCP
<xref target="RFC8415"/>. The SRP registration itself has a lease which <xref target="RFC2131"/> <xref target="RFC8415"/>.
may be on the order of an hour; if the requestor The SRP registration itself has a lease that
may be on the order of two hours; if the requester
does not renew the lease before it has elapsed, the registration is remov ed. The claim on the name can have a longer does not renew the lease before it has elapsed, the registration is remov ed. The claim on the name can have a longer
lease, so that another requestor cannot claim the name, even though the r lease so that another requester cannot claim the name, even though the re
egistration has expired.</t> gistration has expired.</t>
<!-- [Ted] This edit makes the sentence not make sense. What is "The SRP for DNS
-SD?" I don't know. This is not a good time to use
an acronym. We could say "SRP provides a..." but for whatever reason we wan
ted to spell out the protocol that is described
in this document, so I think we need to fully spell it out. -->
<t> <t>
The Service Registration Protocol for DNS&nbhy;SD (SRP), specified in th The Service Registration Protocol for DNS&nbhy;SD
is document, provides a reasonably secure mechanism specified in this document provides a reasonably secure
for publishing this information. Once published, these services can be mechanism for publishing this information.
readily discovered by DNS&nbhy;SD clients using Once published, these services can be readily
discovered by DNS&nbhy;SD clients using
standard DNS lookups.</t> standard DNS lookups.</t>
<t> <t>
The DNS&nbhy;SD specification (<xref target="RFC6763" section="10" secti Section <xref target="RFC6763" section="10" sectionFormat="bare"/>
onFormat="comma"/>, “Populating the DNS with of the DNS&nbhy;SD specification <xref target="RFC6763"/>
Information”), briefly discusses ways that servers can publish their inf briefly discusses ways that servers can advertise the services
ormation in the DNS namespace. In the case of they provide in the DNS namespace. In the case of
mDNS, it allows servers to publish their information on the local link, mDNS, it allows servers to advertise their services on the local link,
using names in the ".local" namespace, which makes using names in the "local." namespace, which makes
their services directly discoverable by peers attached to that same loca l link.</t> their services directly discoverable by peers attached to that same loca l link.</t>
<t> <t>
RFC6763 also allows clients to discover services using <xref target="RFC DNS&nbhy;SD <xref target="RFC6763"/> also allows clients to discover ser
1035">the DNS protocol</xref>. This can be done by vices
having a system administrator manually configure service information in by using the DNS protocol over traditional unicast <xref target="RFC1035
the DNS, but manually populating DNS authoritative "/>.
server databases is costly and potentially error-prone, and requires a k This can be done by
nowledgeable network administrator. Consequently, having a system administrator manually configure service information in
although all DNS&nbhy;SD client implementations of which we are aware su the DNS; however, manually populating DNS authoritative
pport DNS&nbhy;SD using DNS queries, in practice it server databases is costly and potentially error-prone and requires a kn
owledgeable network administrator. Consequently,
although all DNS&nbhy;SD client implementations of which we are
aware support DNS&nbhy;SD using DNS queries, in practice it
is used much less frequently than mDNS.</t> is used much less frequently than mDNS.</t>
<t> <t>
The <xref target="RFC8766">Discovery Proxy</xref> provides one way to au The Discovery Proxy <xref target="RFC8766"/> provides one way to automat
tomatically populate the DNS ically populate the DNS
namespace, but is only appropriate on networks where services are easily namespace but is only appropriate on networks where services are easily
advertised using mDNS. This document describes a advertised using mDNS. The present document describes a
solution more suitable for networks where multicast is inefficient, or w solution more suitable for networks where multicast is inefficient,
here sleepy devices are common, by supporting both or where sleepy devices are common, by supporting the use of unicast
offering of services, and discovery of services, using unicast.</t> for both the offering of and the discovery of services.</t>
</section> </section>
<section> <section>
<name>Conventions and Terminology Used in This Document</name> <name>Conventions and Terminology Used in This Document</name>
<t> <t>
The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", "SHOU The key words "<bcp14>MUST</bcp14>", "<bcp14>MUST NOT</bcp14>", "<bcp14>REQU
LD", "SHOULD NOT", "RECOMMENDED", "NOT RECOMMENDED", IRED</bcp14>", "<bcp14>SHALL</bcp14>", "<bcp14>SHALL
"MAY", and "OPTIONAL" in this document are to be interpreted as described NOT</bcp14>", "<bcp14>SHOULD</bcp14>", "<bcp14>SHOULD NOT</bcp14>", "<bcp14>
in BCP 14 <xref target="RFC2119"/> RECOMMENDED</bcp14>", "<bcp14>NOT RECOMMENDED</bcp14>",
<xref target="RFC8174"/> when, and only when, they appear in all capital "<bcp14>MAY</bcp14>", and "<bcp14>OPTIONAL</bcp14>" in this document are to
s, as shown here. be interpreted as
</t> described in BCP&nbsp;14 <xref target="RFC2119"/> <xref target="RFC8174"/>
when, and only when, they appear in all capitals, as shown here.
</t>
<t>
Strictly speaking, fully qualified domain names end with a period.
In DNS zone files and other similar contexts, if the final period
is omitted, then a name may be treated incorrectly as relative
to some other parent domain. This document follows the formal
DNS convention, ending fully qualified domain names with a period (".").
When this document mentions domain names such as "local." and
"default.service.arpa.", the final period is part of the domain name and
does not indicate the end of a sentence as it would in normal prose.
</t>
</section> </section>
<section> <section>
<name>Service Registration Protocol</name> <name>Service Registration Protocol</name>
<t> <t>
Services that implement SRP use DNS Update <xref target="RFC2136"/> <xre Services that implement SRP use DNS Update <xref target="RFC2136"/> with
f target="RFC3007"/> to publish service information SIG(0) <xref target="RFC3007"/> to publish service information
in the DNS. Two variants exist, one for full-featured hosts, and one fo in the DNS. Two variants exist: One for full-featured hosts and one for
r devices designed for "Constrained-Node Networks" devices designed for Constrained-Node Networks (CNNs)
<xref target="RFC7228"/>. An SRP registrar is most likely an authoritati <xref target="RFC7228"/>.
ve DNS server, or else is updating an authoritative An SRP registrar is most likely an authoritative DNS server
DNS server. There is no requirement that the server that is receiving SRP or is a source of data for one or more authoritative DNS servers.
updates be the same server that is answering There is no requirement that the authoritative DNS server that is
queries that return records that have been registered.</t> receiving SRP Updates be the same authoritative DNS server that is
answering queries that return records that have been registered.
For example, an SRP registrar could be the "hidden primary" that is
the source of data for a fleet of secondary authoritative DNS servers.</
t>
<section> <section>
<name>Protocol Variants</name> <name>Protocol Variants</name>
<section> <section>
<name>Full-featured Hosts</name> <name>Full-Featured Hosts</name>
<t> <t>
Full-featured hosts either are configured manually with a registrati Full-featured hosts either are configured manually with a registrati
on domain, or discover the default registration on domain or discover the default registration
domain as described in <xref target="RFC6763" section="11" sectionFor domain automatically using the Domain Enumeration process described i
mat="of"/>. If this process does not produce a n
default registration domain, the Service Registration protocol is not Section <xref target="RFC6763" section="11" sectionFormat="bare"/>
discoverable on the local network using this of the DNS&nbhy;SD specification <xref target="RFC6763"/>.
mechanism. Other discovery mechanisms are possible, but are out of sc If this process does not produce a
ope for this document.</t> default registration domain, the SRP registrar
is not discoverable on the local network using this
mechanism. Other discovery mechanisms are possible, but they are out
of scope for this document.</t>
<t> <t>
Manual configuration of the registration domain can be done either b Configuration of the registration domain can be done either:</t>
y querying the list of available registration <ul><li>by querying the list of available registration
domains ("r._dns&nbhy;sd._udp") and allowing the user to select one domains ("r._dns&nbhy;sd._udp") and allowing the user to select one
from the UI, or by any other means appropriate to from the UI, or</li>
the particular use case being addressed. Full-featured devices cons <li>by any other means appropriate to
truct the names of the SRV, TXT, and PTR records the particular use case being addressed.</li></ul>
describing their service(s) as subdomains of the chosen service regi <t>Full-featured devices construct the names of the SRV, TXT, and PTR
stration domain. For these names they then discover records
the zone apex of the closest enclosing DNS zone using SOA queries <x describing their service or services as subdomains of the chosen ser
ref target="RFC8765" section="6.1"/>. Having vice registration domain. For these names, they then discover
the zone apex of the closest enclosing DNS zone using SOA queries
as described in
Section <xref target="RFC8765" section="6.1" sectionFormat="bare"/>
of the DNS Push Notification specification <xref target="RFC8765"/>.
Having
discovered the enclosing DNS zone, they query for the "_dnssd&nbhy;s rp._tcp.&lt;zone&gt;" SRV record to discover the discovered the enclosing DNS zone, they query for the "_dnssd&nbhy;s rp._tcp.&lt;zone&gt;" SRV record to discover the
server to which they can send SRP updates. Hosts that support SRP U pdates using TLS use the SRP registrar to which they can send SRP Updates. Hosts that suppor t SRP Updates using TLS use the
"_dnssd&nbhy;srp&nbhy;tls._tcp.&lt;zone&gt;" SRV record instead.</t> "_dnssd&nbhy;srp&nbhy;tls._tcp.&lt;zone&gt;" SRV record instead.</t>
<t> <t>
Examples of full-featured hosts include devices such as home computer s, laptops, powered peripherals with network Examples of full-featured hosts include devices such as home computer s, laptops, powered peripherals with network
connections such as printers, home routers, and even battery-operated devices such as mobile phones that have connections (such as printers and home routers), and even battery-ope rated devices such as mobile phones that have
long battery lives. long battery lives.
</t> </t>
</section> </section>
<section> <section anchor="constrained_hosts">
<name>Constrained Hosts</name> <name>Constrained Hosts</name>
<t> <t>
For devices designed for Constrained-Node Networks <xref target="RFC For devices designed for CNNs <xref target="RFC7228"/>,
7228"/> some simplifications are available. Instead of some simplifications are available. Instead of
being configured with (or discovering) the service registration doma being configured with (or discovering) the service registration doma
in, the special-use domain name (see in,
<xref target="RFC6761"/>) "default.service.arpa" is used. The detai the special-use domain name <xref target="RFC6761"/> "default.servic
ls of how SRP registrar(s) are discovered will be specific e.arpa." is used.
to the constrained network, and therefore we do not suggest a specif The details of how SRP registrars are discovered will be specific
ic mechanism here.</t> to the constrained network; therefore, we do not suggest a specific
mechanism here.</t>
<t> <t>
SRP requestors on constrained networks are expected to receive from SRP requesters on CNNs are expected to receive, from the network, a
the network a list of SRP registrars with which to register. list of SRP registrars with which to register.
It is the responsibility of a Constrained-Node Network supporting SR It is the responsibility of a CNN supporting SRP to provide one or m
P to provide one or more registrar addresses. It is ore registrar addresses. It is
the responsibility of the registrar supporting a Constrained-Node Ne the responsibility of the registrar supporting a CNN to handle the u
twork to handle the updates appropriately. In some pdates appropriately. In some
network environments, updates may be accepted directly into a local network environments, updates may be accepted directly into
"default.service.arpa" zone, which has only local a local "default.service.arpa." zone, which has only local
visibility. In other network environments, updates for names ending visibility. In other network environments, updates for names
in "default.service.arpa" may be rewritten by the registrar ending in "default.service.arpa." may be rewritten by the registrar
to names with broader visibility.</t> to names with broader visibility.
Domain name rewriting should be performed as appropriate
for the network environment in question.
Some suggested techniques for how domain names can be
translated from a locally scoped name to a domain name
with larger scope can be found in the discussion of
data translation for names in Multicast DNS answers in
Section <xref target="RFC8766" section="5.5" sectionFormat="bare"/>
of the Discovery Proxy specification <xref target="RFC8766"/>.</t>
</section> </section>
<section> <section>
<name>Why two variants?</name> <name>Why two variants?</name>
<t> <t>
The reason for these different variants is that low-power devices th The reason for these different variants is that low-power devices th
at typically use Constrained-Node Networks may have at typically use CNNs may have
very limited battery storage. The series of DNS lookups required to very limited battery capacity. The series of DNS lookups required t
discover an SRP registrar and then communicate with o discover an SRP registrar and then communicate with
it will increase the energy required to advertise a service; for low -power devices, the additional flexibility this it will increase the energy required to advertise a service; for low -power devices, the additional flexibility this
provides does not justify the additional use of energy. It is also fairly typical of such networks that some network provides does not justify the additional use of energy. It is also fairly typical of such networks that some network
service information is obtained as part of the process of joining th e network, and so this can be relied upon to provide service information is obtained as part of the process of joining th e network; thus, this can be relied upon to provide
nodes with the information they need.</t> nodes with the information they need.</t>
<t> <t>
Networks that are not constrained networks can have more complicated Networks that are not CNNs can have more complicated topologies at t
topologies at the IP layer. Nodes connected he IP layer. Nodes connected
to such networks can be assumed to be able to do DNS-SD service regi to such networks can be assumed to be able to do DNS&nbhy;SD
stration domain discovery. Such networks are service registration domain discovery. Such networks are
generally able to provide registration domain discovery and routing. This creates the possibility of off-network generally able to provide registration domain discovery and routing. This creates the possibility of off-network
spoofing, where a device from a foreign network registers a service o n the local network in order to attack devices spoofing, where a device from a foreign network registers a service o n the local network in order to attack devices
on the local network. To prevent such spoofing, TCP is required for s uch networks. on the local network. To prevent such spoofing, TCP is required for s uch networks.
</t> </t>
</section> </section>
</section> </section>
<section> <section>
<name>Protocol Details</name> <name>Protocol Details</name>
<t> <t>
We will discuss several parts to this process: how to know what to pub We will discuss several parts to this process:</t>
lish, how to know where to publish it (under what
name), how to publish it, and how to secure its publication. In <xref
target="maintenance"/>, we specify how to maintain
the information once published.</t>
<section> <ul spacing="compact">
<name>What to publish</name> <li>how to know what to publish (<xref target="what"/>),</li>
<li>how to know where to publish it (under what name) (<xref target="where"/>)
,</li>
<li>how to publish it (<xref target="how"/>),</li>
<li>how to secure its publication (<xref target="how-to-secure"/>), and</li>
<li>how to maintain
the information once published (<xref target="maintenance"/>).</li></u
l>
<section anchor="what">
<name>What to Publish</name>
<t> <t>
SRP Updates are sent by SRP requestors to SRP registrars. Three typ es of instructions appear in an SRP update: Service SRP Updates are sent by SRP requesters to SRP registrars. Three typ es of instructions appear in an SRP Update: Service
Discovery instructions, Service Description instructions, and Host De scription instructions. These instructions are made Discovery instructions, Service Description instructions, and Host De scription instructions. These instructions are made
up of DNS Update RRs that are either adds or deletes. The types of re cords that are added, updated and removed in each up of DNS Update Resource Records (RRs) that are either adds or delet es. The types of records that are added, updated, and removed in each
of these instructions, as well as the constraints that apply to them, are described in <xref target="server_behavior"/>. of these instructions, as well as the constraints that apply to them, are described in <xref target="server_behavior"/>.
An SRP Update is a DNS Update message that is constructed so as to me An SRP Update is a DNS Update message <xref target="RFC2136"/> that i
et the constraints described in that section. The s
constructed so as to meet the constraints described in that section.
The
following is a brief overview of what is included in a typical SRP Up date: following is a brief overview of what is included in a typical SRP Up date:
</t> </t>
<ul spacing="compact"> <ul spacing="normal">
<li> <li>
PTR Resource Record (RR) for services, which map from a generic se Service Discovery PTR RR(s) for service(s), which map from a
rvice type (or subtype) name to a specific generic service type (or subtype(s)) to a specific
Service Instance Name.</li> service instance name <xref target="RFC6763"/>.</li>
<li> <li>
For any Service Instance Name (<xref target="RFC6763" section="4.1" For each service instance name, an SRV RR, one or more
sectionFormat="comma"/>), an SRV RR, one or more TXT RRs, and a KEY RR. Although, in principle, DNS&nbhy;SD
TXT RRs, and a KEY RR. Although in principle DNS-SD Service Descrip Service Description records can include other record types with
tion records can include other record types with the same service instance name, in practice, they rarely do.
the same Service Instance Name, in practice they rarely do. SRP doe Currently, SRP does not permit other record types. The KEY RR is us
s not permit other record types. The KEY RR is used ed
to support FCFS naming, and has no specific meaning for DNS-SD look to support FCFS Naming and has no specific meaning for DNS&nbhy;SD
ups. SRV records for all services described in an lookups. SRV records for all services described in an
SRP update point to the same hostname.</li> SRP Update point to the same hostname.</li>
<li> <li>
There is never more than one hostname in a single SRP update. The h There is always exactly one hostname in a single SRP Update.
ostname has one or more address RRs (AAAA or A) and A DNS Update containing more than one hostname is not an SRP Updat
a KEY RR (used for FCFS naming). Depending on the use case, an SRP e.
requestor may be required to suppress some The hostname has one or more address RRs (AAAA or A) and
a KEY RR (used for FCFS Naming). Depending on the use case, an SRP
requester may be required to suppress some
addresses that would not be usable by hosts discovering the servic e through the SRP registrar. The exact address addresses that would not be usable by hosts discovering the servic e through the SRP registrar. The exact address
record suppression behavior required may vary for different types record suppression behavior required may vary for different types
of SRP requestors. An example of such advice can be of SRP requesters.
found in <xref target="RFC8766" section="5.5.2" sectionFormat="of" Some suggested policies for suppressing unusable records can be fo
/>. und in
Section <xref target="RFC8766" section="5.5.2" sectionFormat="bare
"/>
of the Discovery Proxy specification <xref target="RFC8766"/>.
</li> </li>
</ul> </ul>
<!-- [Ted] An RRtype is a type of RR, not a type of RRs, so the plural there doe
sn't make sense. I've tweaked the sentence so that
the confusion is eliminated (I hope). -->
<t> <t>
<xref target="RFC6763"/> describes the details of what each of these The DNS-Based Service Discovery specification
types of RR mean, with the exception of <xref target="RFC6763"/> describes the details of
the KEY RR, which is defined in <xref target="RFC2539"/>. These RFCs what each of these RR types mean, with the exception of
should be considered the definitive source for the KEY RR, which was defined in the
specification for how to store Diffie-Hellman
Keys in the DNS <xref target="RFC2539"/>.
These specifications should be considered
the definitive sources for
information about what to publish; the reason for summarizing this h ere is to provide the reader with enough information information about what to publish; the reason for summarizing this h ere is to provide the reader with enough information
about what will be published that the service registration process c an be understood at a high level without first about what will be published that the service registration process c an be understood at a high level without first
learning the full details of DNS&nbhy;SD. Also, the "Service Instan learning the full details of DNS&nbhy;SD.
ce Name" is an important aspect of FCFS Also, the "service instance name" is an important aspect of FCFS
naming, which we describe later on in this document.</t> Naming, which we describe later on in this document.</t>
</section> </section>
<section> <section anchor="where">
<name>Where to publish it</name> <name>Where to Publish It</name>
<!-- [Ted] Sentence was vague, edit didn't help, I've reworded it. -->
<t> <t>
Multicast DNS uses a single namespace, ".local", which is valid on t Multicast DNS (mDNS) uses a single namespace, "local.".
he local link. This convenience is not available for Subdomains of "local." are specific to the local link on which they
DNS&nbhy;SD using the DNS protocol: services must exist in some spec are advertised.
ific DNS namespace that is chosen either by the This convenience is not available for
network operator, or automatically.</t> DNS&nbhy;SD using the DNS protocol: Services must exist in
some specific DNS namespace that is chosen either by the
network operator or automatically.</t>
<t> <t>
As described above, full-featured devices are responsible for knowin g the domain in which to register their services. As described above, full-featured devices are responsible for knowin g the domain in which to register their services.
Such devices MAY optionally support configuration of a registration d Such devices <bcp14>MAY</bcp14> optionally support configuration of a
omain by the operator of the device. However, registration domain by the operator of the device. However,
such devices MUST support registration domain discovery as described such devices <bcp14>MUST</bcp14> support registration domain discover
in <xref target="RFC6763" section="11" sectionFormat="of"/>, y as described in
"Discovery of Browsing and Registration Domains". Section <xref target="RFC6763" section="11" sectionFormat="bare"/>
of the DNS&nbhy;SD specification <xref target="RFC6763"/>.
</t> </t>
<t> <t>
Devices made for Constrained-Node Networks register in the special u Devices made for CNNs register in the
se domain name <xref target="RFC6761"/> special-use domain name <xref target="RFC6761"/>
"default.service.arpa", and let the SRP registrar handle rewriting t "default.service.arpa." and let the SRP registrar handle
hat to a different domain if necessary.</t> rewriting that to a different domain if necessary,
as described in <xref target="constrained_hosts"/>.</t>
</section> </section>
<section> <section anchor="how">
<name>How to publish it</name> <name>How to Publish It</name>
<t> <t>
It is possible to issue a DNS Update that does several things at onc It is possible to send a DNS Update message that does several things
e; this means that it's possible to do all the work of at once:
adding a PTR resource record to the PTR RRset on the Service Name, a For example, it's possible in a single transaction
nd creating or updating the Service Instance Name and to add or update a single Host Description
Host Description, in a single transaction.</t> while also adding or updating the RRs comprising the Service Descrip
tion(s)
for one or more service instance(s) available on that host
and adding or updating the RRs comprising the Service Discovery inst
ruction(s)
for those service instance(s).</t>
<t> <t>
An SRP Update takes advantage of this: it is implemented as a single DNS Update message that contains a service's Service An SRP Update takes advantage of this: It is implemented as a single DNS Update message that contains a service's Service
Discovery records, Service Description records, and Host Description records.</t> Discovery records, Service Description records, and Host Description records.</t>
<t> <t>
Updates done according to this specification are somewhat different Updates done according to this specification
than regular DNS Updates as defined in are somewhat different from normal DNS Updates
<xref target="RFC2136"/>. The <xref target="RFC2136"/> update proces <xref target="RFC2136"/> where the update process could involve many
s can involve many update attempts: you might first update attempts. You might first
attempt to add a name if it doesn't exist; if that fails, then in a s econd message you might update the name if it does attempt to add a name if it doesn't exist; if that fails, then in a s econd message you might update the name if it does
exist but matches certain preconditions. Because the registration pr exist but matches certain preconditions.
otocol uses a single transaction, some of this Because the Service Registration Protocol described in
this document uses a single transaction, some of this
adaptability is lost.</t> adaptability is lost.</t>
<t> <t>
In order to allow updates to happen in a single transaction, SRP Upd ates do not include update prerequisites. The In order to allow updates to happen in a single transaction, SRP Upd ates do not include update prerequisites. The
requirements specified in <xref target="server_behavior"/> are impli requirements specified in <xref target="server_behavior"/> are impli
cit in the processing of SRP Updates, and so there is cit in the processing of SRP Updates; thus, there is
no need for the SRP requestor to put in any explicit prerequisites.< no need for the SRP requester to put in any explicit prerequisites.<
/t> /t>
<section> <section>
<name>How the DNS&nbhy;SD Service Registration process differs from D NS Update as specified in RFC2136</name> <name>How the DNS&nbhy;SD Service Registration Process Differs from D NS Update</name>
<t> <t>
DNS&nbhy;SD Service Registration is based on standard RFC2136 DNS DNS&nbhy;SD Service Registration uses the DNS Update specification
Update, with some differences:</t> <xref target="RFC2136"/>
<ul spacing="compact"> with some additions:</t>
<ul spacing="normal">
<li> <li>
It implements first-come first-served name allocation, protected using SIG(0) <xref target="RFC2931"/>.</li> It implements FCFS Naming, protected using SIG(0) <xref target="R FC2931"/>.</li>
<li> <li>
It enforces policy about what updates are allowed.</li> It enforces policy about what updates are allowed.</li>
<li> <li>
It optionally performs rewriting of "default.service.arpa" to som e other domain.</li> It optionally performs rewriting of "default.service.arpa." to so me other domain.</li>
<li> <li>
It optionally performs automatic population of the address-to-nam e reverse mapping domains.</li> It optionally performs automatic population of the address-to-nam e reverse mapping domains.</li>
<li> <li>
An SRP registrar is not required to implement general DNS Update prerequisite processing.</li> An SRP registrar is not required to implement general DNS Update prerequisite processing.</li>
<li> <li>
Constrained-Node SRP requestors are allowed to send updates to th e generic domain "default.service.arpa."</li> CNN SRP requesters are allowed to send updates to the generic dom ain "default.service.arpa.".</li>
</ul> </ul>
</section> </section>
<section> <section>
<name>Retransmission Strategy</name> <name>Retransmission Strategy</name>
<t>The DNS protocol, including DNS updates, can operate over UDP or T CP. When using UDP, reliable <t>The DNS protocol, including DNS updates, can operate over UDP or T CP. When using UDP, reliable
transmission must be guaranteed by retransmitting if a DNS UDP mess age is not acknowledged in a transmission must be guaranteed by retransmitting if a DNS UDP mess age is not acknowledged in a
reasonable interval. <xref target="RFC1035" section="4.2.1" section reasonable interval.
Format="of"/> provides some Section <xref target="RFC1035" section="4.2.1" sectionFormat="bare"
guidance on this topic, as does <xref target="RFC1536" section="1" />
sectionFormat="of"/>. of the DNS specification <xref target="RFC1035"/> provides some
<xref target="RFC8085" section="3.1.3" sectionFormat="of"/> also pr guidance on this topic, as does
ovides useful guidance that Section <xref target="RFC1536" section="1" sectionFormat="bare"/>
is particularly relevant to DNS.</t> of the IETF document describing common DNS implementation errors <x
ref target="RFC1536"/>.
Section <xref target="RFC8085" section="3.1.3" sectionFormat="bare"
/>
of the UDP Usage Guidelines document <xref target="RFC8085"/>
also provides useful guidance that is particularly relevant to DNS.
</t>
</section> </section>
<section> <section>
<name>Successive Updates</name> <name>Successive Updates</name>
<t>Service Registration Protocol does not require that every update c <t>SRP does not require that every update contain the same informatio
ontain the same information. n.
When an SRP requestor needs to send more than one SRP update to the When an SRP requester needs to send more than one
SRP registrar, it MUST send SRP Update to the SRP registrar, it <bcp14>SHOULD</bcp14> combine
these sequentially: until an earlier update has been successfully a these into a single SRP Update,
cknowledged, the requestor when possible, subject to DNS message size limits and link-specific
MUST NOT begin sending a subsequent update.</t> size limits (e.g., an IEEE 802.15.4 network will perform poorly when a
sked
to deliver a packet larger than about 500 bytes).
If the updates do not fit into a single SRP Update,
then the SRP requester <bcp14>MUST</bcp14>
send subsequent SRP Updates sequentially:
Until an earlier SRP Update has been acknowledged,
the requester <bcp14>MUST NOT</bcp14>
send any subsequent SRP Updates.
If a configuration change occurs while an outstanding
SRP Update is in flight, the SRP registrar
<bcp14>MUST</bcp14> defer sending a new SRP Update for
that change until the previous SRP Update has completed.</t>
</section> </section>
</section> </section>
<section anchor="how-to-secure"> <section anchor="how-to-secure">
<name>How to secure it</name> <name>How to Secure It</name>
<t> <t>
DNS update as described in <xref target="RFC2136"/> is secured using DNS Update messages can be secured using secret key transaction sign
Secret Key Transaction Signatures, atures (TSIG)
<xref target="RFC8945"/>, which uses a secret key shared between the <xref target="RFC8945"/>.
DNS Update requestor (which issues the update) and This approach uses a secret key shared between
the server (which authenticates it). This model does not work for a the DNS Update requester (which issues the update) and
utomatic service registration.</t> the authoritative DNS server (which authenticates it).
This model does not work for automatic service registration.</t>
<t> <t>
The goal of securing the DNS&nbhy;SD Registration Protocol is to pro The goal of securing the DNS&nbhy;SD Registration Protocol
vide the best possible security given the constraint is to provide the best possible security given the constraint
that service registration has to be automatic. It is possible to la yer more operational security on top of what we that service registration has to be automatic. It is possible to la yer more operational security on top of what we
describe here, but FCFS naming is already an improvement over the se curity of mDNS.</t> describe here, but FCFS Naming is already an improvement over the se curity of mDNS.</t>
<section anchor="fcfs"> <section anchor="fcfs">
<name>First-Come First-Served Naming</name> <name>FCFS Naming</name>
<t> <t>
First-Come First-Serve naming provides a limited degree of securit
y: a server that registers its service using <!--[rfced] To what does "that" refer in this sentence?
DNS&nbhy;SD Registration protocol is given ownership of a name for
an extended period of time based on a lease Original:
specific to the key used to authenticate the DNS Update, which may As long as the registration service remembers the name and the key
be longer than the lease associated with the used to register that name, no other server can add or update the
registered records. As long as the registration service remembers information associated with that.
the name and
the key used to register that name, no other server can add or upd Perhaps:
ate the information associated with that. If the As long as the registration service remembers the name and the key
server fails to renew its service registration before the KEY leas used to register that name, no other server can add or update the
e (<xref target="I-D.ietf-dnssd-update-lease" information associated with them.
section="4"/>) expires, its name is no longer protected. FCFS nam
ing is used to protect both the Service Description Perhaps:
As long as the registration service remembers the name and the key
used to register that name, no other server can add or update the
information associated with that pair.
[Ted] Updated a bit.
-->
FCFS Naming provides a limited degree of security. A server that r
egisters its service using SRP
is given ownership of a name for an extended period of time based
on a lease
specific to the key used to authenticate the SRP Update, which may
be longer than the lease associated with the
registered RRs. As long as the registrar remembers the name
and the public key corresponding to the private key used to regist
er RRs on that name,
no other SRP requester can add or update the
information associated with that name.
If the SRP requester fails to renew its
service registration before the KEY lease expires
(Section <xref target="RFC9664" section="4" sectionFormat="bare"/>
of the DNS Update Lease specification <xref target="RFC9664"/>)
its name is no longer protected.
FCFS Naming is used to protect both the Service Description
and the Host Description.</t> and the Host Description.</t>
</section> </section>
</section> </section>
<section> <section>
<name>SRP Requestor Behavior</name> <name>SRP Requester Behavior</name>
<section> <section>
<name>Public/Private key pair generation and storage</name> <name>Public/Private Key Pair Generation and Storage</name>
<t> <t>
The requestor generates a public/private key pair (See <xref target The requester generates a public/private key pair (<xref target="rs
="rsa"/>). This key pair MUST be stored in stable a"/>).
storage; if there is no writable stable storage on the SRP requesto This key pair <bcp14>MUST</bcp14> be stored in stable
r, the SRP requestor MUST be pre-configured with a storage; if there is no writable stable storage on the SRP requeste
public/private key pair in read-only storage that can be used. Thi r, the SRP requester <bcp14>MUST</bcp14> be preconfigured with a
s key pair MUST be unique to the device. A device public/private key pair in read-only storage.
with rewritable storage SHOULD retain this key indefinitely. When This key pair <bcp14>MUST</bcp14> be unique to the device.
the device changes ownership, it may be appropriate A device
with rewritable storage <bcp14>SHOULD</bcp14> retain this key indef
initely. When the device changes ownership, it may be appropriate
for the former owner to erase the old key pair, which would then re quire the new owner to install a new for the former owner to erase the old key pair, which would then re quire the new owner to install a new
one. Therefore, the SRP requestor on the device SHOULD provide a me one. Therefore, the SRP requester on the device <bcp14>SHOULD</bcp
chanism to erase the key, for example as the 14> provide a mechanism to erase the key (for example, as the
result of a "factory reset," and to generate a new key.</t> result of a "factory reset") and to generate a new key.</t>
<t>
Note that when a new key is generated, this will prevent the
device from registering with the name associated with the old
key in the same domain where it had previously registered. So,
implicit in the generation of a new key is the generation of a
new name; this can be done either proactively when regenerating
a key or when the SRP update produces a name conflict.
</t>
<t> <t>
The policy described here for managing keys assumes that the keys a re only used for SRP. If a key that is used for SRP The policy described here for managing keys assumes that the keys a re only used for SRP. If a key that is used for SRP
is also used for other purposes, the policy described here is likel is also used for other purposes, the policy described here is likel
y to be insufficient. The policy stated here is NOT y to be insufficient. The policy stated here is <bcp14>NOT
RECOMMENDED in such a situation: a policy appropriate to the full s RECOMMENDED</bcp14> in such a situation: a policy appropriate to th
et of uses for the key must be chosen. Specifying e full set of uses for the key must be chosen. Specifying
such a policy is out of scope for this document.</t> such a policy is out of scope for this document.</t>
<t> <t>
When sending DNS updates, the requestor includes a KEY record conta When sending DNS updates, the requester includes a KEY record conta
ining the public portion of the key in each Host ining the public portion of the key in each Host
Description Instruction and each Service Description Instruction. Description Instruction and each Service Description Instruction.
Each KEY record MUST contain the same public key. Each KEY record <bcp14>MUST</bcp14> contain the same public key.
The update is signed using SIG(0), using the private key that corre sponds to the public key in the KEY record. The The update is signed using SIG(0), using the private key that corre sponds to the public key in the KEY record. The
lifetimes of the records in the update is set using the EDNS(0) Upd lifetimes of the records in the update are
ate Lease option set using the EDNS(0) Update Lease option
<xref target="I-D.ietf-dnssd-update-lease"/>.</t> <xref target="RFC9664"/>.</t>
<t> <t>
The format of the KEY resource record in the SRP Update is defined The format of the KEY resource record in the SRP Update is
in <xref target="RFC3445"/>. Because the KEY RR defined in the IETF specification for DNSSEC Resource Records
used in TSIG is not a zone-signing key, the flags field in the KEY <xref target="RFC4034"/>. Because the KEY RR
RR MUST be all zeroes.</t> used in SIG(0) is not a zone-signing key, the flags field in the KE
Y RR <bcp14>MUST</bcp14> be all zeroes.</t>
<t> <t>
The KEY record in Service Description updates MAY be omitted for br evity; if it is omitted, the SRP registrar MUST behave The KEY record in Service Description updates <bcp14>MAY</bcp14> be omitted for brevity; if it is omitted, the SRP registrar <bcp14>MUST</bcp14> be have
as if the same KEY record that is given for the Host Description is also given for each Service Description for which as if the same KEY record that is given for the Host Description is also given for each Service Description for which
no KEY record is provided. Omitted KEY records are not used when c omputing the SIG(0) signature.</t> no KEY record is provided. Omitted KEY records are not used when c omputing the SIG(0) signature.</t>
</section> </section>
<section> <section>
<name>Name Conflict Handling</name> <name>Name Conflict Handling</name>
<t> <t>
Both Host Description RR adds and Service Description RR adds can h "Add" operations for both Host Description RRs and
ave names that result in name conflicts. Service Description RRs can have names that result in name conflict
Service Discovery record adds cannot have name conflicts. If any Ho s.
st Description or Service Description record Service Discovery record "Add" operations
cannot have name conflicts.
If any Host Description or Service Description record
is found by the SRP registrar to have a conflict with an existing n ame, the registrar will respond to the SRP Update is found by the SRP registrar to have a conflict with an existing n ame, the registrar will respond to the SRP Update
with a YXDomain RCODE (<xref target="RFC2136" section="2.2" section with a YXDomain RCODE <xref target="RFC2136"/>, indicating that the
Format="of"/>). In this case, the desired name is already owned by a different SIG(0) key. In this ca
requestor MUST choose a new name or give up.</t> se, the
SRP requester <bcp14>MUST</bcp14> choose a new name or give up.</t>
<t> <t>
There is no specific requirement for how this is done; typically, h There is no specific requirement for how the SRP
owever, the requestor will append a number to the requester should choose a new name. Typically,
preferred name. This number could be sequentially increasing, or co however, the requester will append a number to the
uld be chosen randomly. One existing implementation preferred name. This number could be sequentially increasing or cou
attempts several sequential numbers before choosing randomly. So fo ld be chosen randomly. One existing implementation
r instance, it might try host.default.service.arpa, attempts several sequential numbers before choosing randomly.
then host-1.default.service.arpa, then host-2.default.service.arpa, For instance, it might try host.default.service.arpa.,
then host-31773.default.service.arpa.</t> then host&nbhy;1.default.service.arpa.,
then host&nbhy;2.default.service.arpa.,
then host&nbhy;31773.default.service.arpa.</t>
</section> </section>
<section> <section anchor="lifetimes">
<name>Record Lifetimes</name> <name>Record Lifetimes</name>
<t> <t>
The lifetime of the <xref target="RFC6763">DNS&nbhy;SD PTR, SRV, A, The lifetime of the DNS&nbhy;SD PTR, SRV, A, AAAA, and TXT
AAAA and TXT records</xref> uses the LEASE field records <xref target="RFC6763"/> uses the LEASE field
of the Update Lease option, and is typically set to two hours. Thi of the Update Lease option and is typically set to two hours. Thus
s means that if a device is disconnected from the , if a device is disconnected from the
network, it does not appear in the user interfaces of devices looki network, it does not continue to appear for too long in the
ng for services of that type for too long.</t> user interfaces of devices looking for instances of that service ty
pe.</t>
<t> <t>
The lifetime of the KEY records is set using the KEY-LEASE field of The lifetime of the KEY records is set using the KEY-LEASE field of
the Update Lease Option, and SHOULD be set to a the Update Lease Option and <bcp14>SHOULD</bcp14> be set to a
much longer time, typically 14 days. The result of this is that ev much longer time, typically 14 days. The result being that even th
en though a device may be temporarily unplugged, ough a device may be temporarily unplugged --
disappearing from the network for a few days, it makes a claim on i disappearing from the network for a few days -- it makes a claim on
ts name that lasts much longer.</t> its name that lasts much longer.</t>
<t> <t>
This means that even if a device is unplugged from the network for a few days, and its services are not available for Therefore, even if a device is unplugged from the network for a few days, and its services are not available for
that time, no other device can come along and claim its name the mo ment it disappears from the network. In the event that time, no other device can come along and claim its name the mo ment it disappears from the network. In the event
that a device is unplugged from the network and permanently discard ed, then its name is eventually cleaned up and made that a device is unplugged from the network and permanently discard ed, then its name is eventually cleaned up and made
available for re-use.</t> available for reuse.</t>
</section> </section>
<section> <section>
<name>Compression in SRV records</name> <name>Compression in SRV Records</name>
<t>
Although <xref target="RFC2782"/> requires that the target name in
the SRV record not be compressed, an SRP requestor
MAY compress the target in the SRV record. The motivation for <em>n
ot</em> compressing in <xref target="RFC2782"/>
is not stated, but is assumed to be because a caching resolver that
does not understand the format of the SRV record
might store it as binary data and thus return an invalid pointer in
response to a query. This does not apply in the
case of SRP: an SRP registrar needs to understand SRV records in or
der to validate the SRP Update. Compression of the
target can save space in the SRP Update, so we want clients to be a
ble to assume that the registrar will handle
this. Therefore, SRP registrars MUST support compression of SRV RR
targets.</t>
<t> <t>
Note that this does not update <xref target="RFC2782"/>: DNS server Although the original SRV specification <xref target="RFC2782"/>
s still MUST NOT compress SRV record targets. The requires that the target hostname in the rdata of an SRV record
requirement to accept compressed SRV records in updates only applie not be compressed in DNS queries and responses, an SRP requester
s to SRP registrars, and SRP registrars that are <bcp14>MAY</bcp14> compress the target in the SRV record,
also DNS servers still MUST NOT compress SRV record targets in DNS since an SRP Update is neither a DNS query nor a DNS response.
responses. We note also that The motivation for <em>not</em> compressing
<xref target="RFC6762"/> recomments that SRV records be compressed is not stated in the SRV specification
in mDNS messages, so <xref target="RFC2782"/> does but is assumed to be because a recursive resolver
not apply to mDNS messages.</t> (caching server) that does not understand the format of the
SRV record might store it as binary data without decoding a
compression pointer embedded with the target hostname field
and thus return nonsensical rdata in response to a query.
This concern does not apply in the
case of SRP. An SRP registrar needs to understand SRV records in or
der to validate the SRP Update. Compression of the
target can save space in the SRP Update,
so we want SRP requesters to be able to
assume that the registrar will handle
this. Therefore, SRP registrars <bcp14>MUST</bcp14> support compres
sion of SRV RR targets.</t>
<t>
<!--[rfced] How might we clarify "this" for the ease of the reader
(especially as this sentence is the first of the paragraph)?
Original:
Note that this does not update [RFC2782]: DNS servers still <bcp14>MUST
NOT</bcp14> compress SRV record targets.
[Ted] "this" refers to this document, RFC9665.
-->
Note that this document does not update
the SRV specification <xref target="RFC2782"/>:
Authoritative DNS servers still <bcp14>MUST NOT</bcp14> compress SR
V record targets.
The requirement to accept compressed SRV records in updates only ap
plies to SRP
registrars, and SRP registrars that are also authoritative DNS serv
ers still
<bcp14>MUST NOT</bcp14> compress SRV record targets in DNS response
s.
We note also that Multicast DNS <xref target="RFC6762"/>
similarly compresses SRV records in mDNS messages.</t>
<t> <t>
In addition, we note that an implementor of an SRP requestor might update existing code that creates SRV records In addition, we note that an implementer of an SRP requester might update existing code that creates SRV records
or compresses DNS messages so that it compresses the target of an S RV record. Care must be taken if such code is or compresses DNS messages so that it compresses the target of an S RV record. Care must be taken if such code is
used both in requestors and in DNS servers that the code only compr used both in requesters and in authoritative DNS servers that the c
esses in the case where a requestor is generating ode only
an SRP update.</t> compresses in the case where a requester is generating an SRP Updat
e.</t>
</section> </section>
<section anchor="remove"> <section anchor="remove">
<name>Removing published services</name> <name>Removing Published Services</name>
<section anchor="zero-lease"> <section anchor="zero-lease">
<name>Removing all published services</name> <name>Removing All Published Services</name>
<t> <t>
To remove all the services registered to a particular host, the S To remove all the services registered to a particular hostname,
RP requestor transmits an SRP update for that host the SRP requester transmits an SRP Update for that hostname
with an Update Lease option that has a LEASE value of zero. If th with an Update Lease option that has a LEASE value of zero.
e registration is to be permanently removed, The SRP Update <bcp14>MUST</bcp14> contain
KEY-LEASE SHOULD also be zero. Otherwise, it SHOULD be set to the exactly one Host Description Instruction
same value it had previously; this holds the name that contains exactly one "Delete All RRsets From A Name" instruc
in reserve for when the SRP requestor is once again able to provi tion for the hostname
de the service.</t> and no "Add to an RRSet" instructions for that hostname.
If the registration is to be permanently removed,
KEY-LEASE <bcp14>SHOULD</bcp14> also be zero. Otherwise, it <bcp1
4>SHOULD</bcp14> be set to the same value it had previously; this holds the name
in reserve for when the SRP requester is once again able to provi
de the service.</t>
<t> <t>
SRP requestors are normally expected to remove all service instan This method of removing services is intended for the case
ces when removing a host. However, in some cases an SRP where the requester is going offline and does not want
requestor may not have retained sufficient state to know that som any of its services to continue being advertised.
e service instance is pointing to a host that it is
removing. This method of removing services is intended for the c
ase where the requestor is going offline and does
not want its services advertised. Therefore, it is sufficient for
the requestor to send the
<xref target="hdi">Host Description Instruction</xref>.
</t> </t>
<t> <t>
To support this, when removing services based on the lease time b To support this, when removing a hostname,
eing zero, an SRP registrar MUST remove all service an SRP registrar <bcp14>MUST</bcp14> remove all service
instances pointing to a host when a host is removed, even if the instances pointing to that hostname
SRP requestor doesn't list them explicitly. If the and all Service Discovery PTR records pointing to those service i
KEY lease time is nonzero, the SRP registrar MUST NOT delete the nstances,
KEY records for these SRP requestors. even if the SRP requester doesn't list them explicitly. If the
KEY lease time is nonzero, the SRP registrar <bcp14>MUST NOT</bcp
14> delete the KEY records for these SRP requesters.
</t> </t>
</section> </section>
<section> <section>
<name>Removing some published services</name> <name>Removing Some Published Services</name>
<t>
In some use cases a requestor may need to remove some specific se
rvice, without removing its other services. This can
be accomplished in one of two ways. To simply remove a specific s
ervice, the requestor sends a valid SRP Update where
the <xref target="servdis">Service Discovery Instruction</xref> c
ontains a single Delete an RR from an RRset
(<xref target="RFC2136" section="2.5.4" sectionFormat="comma"/>)
update that deletes the PTR record whose target is
the service instance name. The <xref target="servdesc">Service De
scription Instruction</xref> in this case contains
a single Delete all RRsets from a Name (<xref target="RFC2136" se
ction="2.5.3" sectionFormat="comma"/>) update to
the service instance name.
</t>
<t> <t>
The second alternative is used when some service is being replace In some use cases, a requester may need to remove a
d by a different service with a different service specific service without removing its other services.
instance name. In this case, the old service is deleted as in the For example, a device may
first alternative. The new service is added, just shut down its remote screen access (_rfb._tcp) service
as it would be in an update that wasn't deleting the old service. while retaining its command-line login (_ssh._tcp) service.
Because both the removal of the old service and This can be accomplished in one of two ways:</t>
the add of the new service consist of a valid Service Discovery I
nstruction and a valid Service Description <ol type="1" spacing="normal">
Instruction, the update as a whole is a valid SRP Update, and wil <li>To simply remove a specific service, the requester sends
l result in the old service being removed and the a valid SRP Update with a Service Description Instruction
new one added, or, to put it differently, in the old service bein (<xref target="servdesc"/>) containing a single "Delete All
g replaced by the new service. RRsets From A Name" update to the service instance name.
</t> The SRP Update <bcp14>SHOULD</bcp14> include Service
Discovery Instructions (<xref target="servdis"/>) consisting
of "Delete An RR From An RRset" updates <xref
target="RFC2136"/> that delete any Service Discovery PTR
records whose target is the service instance name. However,
even in the absence of such Service Discovery Instructions,
the SRP registrar <bcp14>MUST</bcp14> delete any Service
Discovery PTR records that point to the deleted service
instance name.
</li>
<li>When deleting one service instance while simultaneously
creating a new service instance with a different service
instance name, an alternative is to perform both operations
using a single SRP Update. In this case, the old service is
deleted as in the first alternative. The new service is
added, just as it would be in an update that wasn't deleting
the old service. Because both the removal of the old service
and the add of the new service consists of a valid Service
Discovery Instruction and a valid Service Description
Instruction, the update as a whole is a valid SRP Update and
will result in the old service being removed and the new one
added; or, to put it differently, the SRP Update will result
in the old service being replaced by the new service.
</li>
</ol>
<t> <t>
It is perhaps worth noting that if a service is being updated wit It is perhaps worth noting that if a service is being updated wit
hout the service instance name changing, that will hout
look very much like the second alternative above. The difference the service instance name changing (for example, when only the ta
is that because the target for the PTR record in rget
the Service Discovery Instruction is the same for both the Delete port in the SRV record is being updated), then that SRP Update wi
An RR From An RRset update and the Add To An RRSet ll
update, there is no way to tell whether they were intended to be look very much like the second alternative above.
one or two Instructions. The same would be true of The PTR record in the Service Discovery Instruction will be the s
the Service Description Instruction. ame for
both the "Delete An RR From An RRset" update and the "Add To An R
Rset" update
<xref target="RFC2136"/>.
Since the removal of the old service and the addition
of the new service are both valid SRP Update operations,
the combined operation is a valid SRP Update operation.
The SRP registrar does not need to include code to
recognize this special case and does not need to
take any special actions to handle it correctly.
</t> </t>
<t> <t>
Whichever of these two alternatives is used, the host lease will Whichever of these two alternatives is used, the hostname lease
be updated with the lease time provided in the SRP will be updated with the lease time provided in the SRP update.
update. In neither of these cases is it permissible to delete the In neither of these cases is it permissible to delete the hostnam
host. All services must point to a host. If a host e.
is to be deleted, this must be done using the method described in All services must point to a hostname. If a hostname
<xref target="zero-lease"/>, which deletes the is to be deleted, this must be done using the method
host and all services that have that host as their target. described in <xref target="zero-lease"/>, which deletes the
hostname and all services that have that hostname as their target
.
</t> </t>
</section> </section>
</section> </section>
</section></section> </section></section>
<section anchor="server_behavior"> <section anchor="server_behavior">
<name>Validation and Processing of SRP Updates</name> <name>Validation and Processing of SRP Updates</name>
<section anchor="add_validation"> <section anchor="add_validation">
<name>Validation of DNS Update Add and Delete RRs</name> <name>Validation of DNS Update Add and Delete RRs</name>
<t> <t>
The SRP registrar first validates that the DNS Update is a syntactica The SRP registrar first validates that the DNS Update message is a sy
lly and semantically valid DNS Update according to ntactically and semantically valid DNS Update message according to
the rules specified in <xref target="RFC2136"/>.</t> the usual DNS Update rules <xref target="RFC2136"/>.</t>
<t> <t>
SRP Updates consist of a set of <em>instructions</em> that together a SRP Updates consist of a set of <em>instructions</em>
dd or remove one or more services. Each instruction that together add or remove one or more services.
consists of some combination of delete updates and add updates. When Each <em>instruction</em> consists of
an instruction contains a delete and an add, the one or more delete update(s), or one or more add update(s),
delete MUST precede the add.</t> or some combination of both delete updates and add updates.</t>
<t> <t>
The SRP registrar checks each instruction in the SRP Update to see th at it is either a Service Discovery Instruction, a The SRP registrar checks each instruction in the SRP Update to see th at it is either a Service Discovery Instruction, a
Service Description Instruction, or a Host Description Instruction. Order matters in DNS updates. Specifically, Service Description Instruction, or a Host Description Instruction. Order matters in DNS updates. Specifically,
deletes must precede adds for records that the deletes would affect; deletes must precede adds for records that the deletes would affect;
otherwise the add will have no effect. This is the otherwise, the add will have no effect. This is the
only ordering constraint; aside from this constraint, updates may app only ordering constraint: Aside from this constraint, updates may app
ear in whatever order is convenient when ear in whatever order is convenient when
constructing the update.</t> constructing the update.</t>
<t> <t>
Because the SRP Update is a DNS update, it MUST contain a single ques Because the SRP Update is a DNS update, it <bcp14>MUST</bcp14> contai
tion that indicates the zone to be updated. n
Every delete and update in an SRP Update MUST be within the zone that a single entry in the Zone Section (what would be the Question Sectio
is specified for the SRP Update.</t> n
in a traditional DNS message) that indicates the zone to be updated.
Every delete and update in an SRP Update <bcp14>MUST</bcp14> be withi
n the zone that is specified for the SRP Update.</t>
<section anchor="servdis"> <section anchor="servdis">
<name>Service Discovery Instruction</name> <name>Service Discovery Instruction</name>
<t>An instruction is a Service Discovery Instruction if it contains< <t>An instruction is a Service Discovery Instruction if it:</t>
/t>
<!-- [rfced] FYI - we updated the list as follows for clarity. Please let us
know if there are any objections.
Original:
An instruction is a Service Discovery Instruction if it contains
* exactly one "Add to an RRSet" ([RFC2136], Section 2.5.1) or
exactly one "Delete an RR from an RRSet" ([RFC2136],
Section 2.5.4) RR update,
* which updates a PTR RR,
* the target of which is a Service Instance Name
* for which name a Service Description Instruction is present in the
SRP Update, and:
- if the RR Update is an "Add to an RRSet" instruction, that
Service Description Instruction contains an "Add to an RRset"
RR update for the SRV RR describing that service and no other
"Delete from an RRset" instructions for that Service Instance
Name; or
- if the RR Update is a "Delete an RR from an RRSet" instruction,
that Service Description Instruction contains a "Delete from an
RRset" RR update and no other "Add to an RRset" instructions
for that Service Instance Name.
* and contains no other add or delete RR updates for the same name
as the PTR RR Update.
Current:
An instruction is a Service Discovery Instruction if it:
* Contains exactly one "Add to an RRSet" (Section 2.5.1 of
[RFC2136]) or exactly one "Delete an RR from an RRSet"
(Section 2.5.4 of [RFC2136]) RR update, which updates a PTR RR;
the target of which is a Service Instance Name for which name a
Service Description Instruction is present in the SRP Update.
Additionally:
- If the RR Update is an "Add to an RRSet" instruction, that
Service Description Instruction contains an "Add to an RRset"
RR update for the SRV RR describing that service and no other
"Delete from an RRset" instructions for that Service Instance
Name.
- If the RR Update is a "Delete an RR from an RRSet" instruction,
that Service Description Instruction contains a "Delete from an
RRset" RR update and no other "Add to an RRset" instructions
for that Service Instance Name.
* Contains no other add or delete RR updates for the same name as
the PTR RR Update.
[Ted] The working group worked very hard on the text and the organization and st
ructure of the
bullets here, and this change completely goes against that consensus, so I
am reverting it. I
don't mean to suggest that there was no point to this change, but it was r
eally hard to get
agreement on this precise way of formatting the text, and I do not want to
repeat that process.
I actually tried to replicate some of your changes, and wasn't sure if I w
as changing the
meaning of the text, so I've only retained your capitalization changes. Pl
ease think of this
text as more like computer code than human language, if that helps.
-->
<ul spacing="compact"> <ul spacing="compact">
<li>exactly one "Add to an RRSet" (<xref target="RFC2136" section=" <li>consists of
2.5.1" sectionFormat="comma"/>) or exactly one exactly one "Add To An RRSet" or
"Delete an RR from an RRSet" (<xref target="RFC2136" section="2.5 exactly one "Delete An RR From An RRSet"
.4" sectionFormat="comma"/>) RR update,</li> RR update
(Section <xref target="RFC2136" section="2.5" sectionFormat="bare"/
>
of the DNS Update specification <xref target="RFC2136"/>),</li>
<li>which updates a PTR RR,</li> <li>which updates a PTR RR,</li>
<li>the target of which is a Service Instance Name</li> <li>the target of which is a service instance name</li>
<li><t>for which name a Service Description Instruction is present in the SRP Update, and:</t> <li><t>for which name a Service Description Instruction is present in the SRP Update, and:</t>
<ul spacing="compact"> <ul spacing="compact">
<li>if the RR Update is an "Add to an RRSet" instruction, that <li>if the Service Discovery Instruction is an "Add To An RRSet
Service Description Instruction contains an "Add to " instruction,
an RRset" RR update for the SRV RR describing that service an that Service Description Instruction contains
d no other "Delete from an RRset" instructions for a "Delete All RRsets From A Name" instruction for that service
that Service Instance Name; or</li> instance name
<li>if the RR Update is a "Delete an RR from an RRSet" instruct followed by "Add To An RRset" instructions
ion, that Service Description Instruction contains for the SRV and TXT records describing that service; or</li>
a "Delete from an RRset" RR update and no other "Add to an RR <li>if the Service Discovery Instruction is a "Delete An RR Fro
set" instructions for that Service Instance m An RRSet" instruction,
Name.</li></ul></li> that Service Description Instruction contains
<li>and contains no other add or delete RR updates for the same nam a "Delete All RRsets From A Name" instruction for that service
e as the PTR RR Update.</li> instance name
with no following "Add To An RRset" instructions
for the SRV and TXT records describing that service.</li></ul><
/li>
</ul> </ul>
<t> <t>
Note that there can be more than one Service Discovery Instruction Note that there can be more than one Service Discovery
for the same name if the SRP requestor is Instruction for the same service name
advertising more than one service of the same type, or is changing (the owner name of the Service Discovery PTR record)
the target of a PTR RR. This is also true for SRP if the SRP requester is advertising more than one instance
subtypes (<xref target="RFC6763" section="7.1"/>). For each such PT of the same service type or is changing the target of a PTR RR.
R RR add or delete, the above constraints must be When subtypes are being used
met.</t> (Section <xref target="RFC6763" section="7.1" sectionFormat="bare"/
>
of the DNS&nbhy;SD specification <xref target="RFC6763"/>),
each subtype is a separate Service Discovery Instruction.
For each such PTR RR add or delete, the above constraints must be m
et.</t>
</section> </section>
<section anchor="servdesc"> <section anchor="servdesc">
<name>Service Description Instruction</name> <name>Service Description Instruction</name>
<t>An instruction is a Service Description Instruction if, for the a <t>An instruction is a Service Description Instruction if, for the
ppropriate Service Instance Name, the following are true:</t> given service instance name, all of the following are true:</t>
<ul spacing="compact"> <ul spacing="compact">
<li> <li>
It contains exactly one "Delete all RRsets from a name" update fo It contains exactly one "Delete All RRsets From A Name" update fo
r the service instance name r the service instance name
(<xref target="RFC2136" section="2.5.3" sectionFormat="comma"/>), (Section <xref target="RFC2136" section="2.5.3" sectionFormat="ba
</li> re"/>
<li> of the DNS Update specification <xref target="RFC2136"/>).</li>
It contains zero or one "Add to an RRset" SRV RR,</li>
<li>
It contains zero or one "Add to an RRset" KEY RR that, if present
, contains the public key corresponding to the private key
that was used to sign the message (if present, the KEY MUST match
the KEY RR given in the Host Description),</li>
<li> <li>
It contains zero or more "Add to an RRset" TXT RRs,</li> It contains zero or one "Add To An RRset" KEY RRs that, if presen
t, contains the public key corresponding to the private key
that was used to sign the message (if present, the KEY RR <bcp14>
MUST</bcp14> match the KEY RR given in the Host Description).</li>
<li> <li>
If there is one "Add to an RRset" SRV update, there MUST be at le ast one "Add to an RRset" TXT update.</li> It contains zero or one "Add To An RRset" SRV RR.</li>
<li> <li>
The target of the SRV RR Add, if present points to a hostname for If an "Add To An RRSet" update for an SRV RR is present,
which there is a Host Description Instruction in there <bcp14>MUST</bcp14> be at least one "Add To An RRset"
update for the corresponding TXT RR, and
the target of the SRV RR <bcp14>MUST</bcp14> be the hostname give
n in the Host Description Instruction in
the SRP Update, or</li> the SRP Update, or</li>
<li> <li>
If there is no "Add to an RRset" SRV RR, then either:</li> If there is no "Add To An RRset" update for an SRV RR, then
<li><ul> there <bcp14>MUST</bcp14> be no "Add To An RRset" updates for the
<li>the name to which the "Delete all RRsets from a name" applies corresponding TXT RR,
does not exist, or</li> and either:</li>
<li>there is an existing KEY RR on that name, which matches the k <li><ul spacing="compact">
ey with which the SRP Update was <li>the name to which the "Delete All RRsets From A Name" applies
does not exist, or</li>
<li>there is an existing KEY RR on that name that matches the key
with which the SRP Update was
signed.</li></ul></li> signed.</li></ul></li>
<li>
No other resource records on the Service Instance Name are modifi
ed.</li>
</ul> </ul>
<t>An SRP registrar MUST correctly handle compressed names in the SRV <t>Service Description Instructions do not modify any other resource
target.</t> records.</t>
<t>An SRP registrar <bcp14>MUST</bcp14> correctly handle compressed n
ames in the SRV target.</t>
</section> </section>
<section anchor="hdi"> <section anchor="hdi">
<name>Host Description Instruction</name> <name>Host Description Instruction</name>
<t>An instruction is a Host Description Instruction if, for the appr <t>Every SRP Update alway contains exactly one Host Description Inst
opriate hostname, it contains</t> ruction.</t>
<ul spacing="compact">
<li> <t>An instruction is a Host Description Instruction if, for the appr
exactly one "Delete all RRsets from a name" RR,</li> opriate hostname, it contains the following:</t>
<ul spacing="normal">
<li> <li>
one or more "Add to an RRset" RRs of type A and/or AAAA,</li> exactly one "Delete All RRsets From A Name" RR</li>
<li> <li>
exactly one "Add to an RRset" RR that adds a KEY RR that contains exactly one "Add To An RRset" RR that adds a KEY RR that
the public key corresponding to the private key contains the public key corresponding to the private key that
that was used to sign the message,</li> was used to sign the message</li>
<li> <li>
Host Description Instructions do not modify any other resource re zero "Add To An RRset" operations (in the case of deleting a regi
cords.</li> stration)
</ul> or one or more "Add To An RRset" RRs of type A and/or AAAA
(in the case of creating or updating a registration)</li>
</ul>
<t> <t>
A and/or AAAA records that are not of sufficient scope to be validl Host Description Instructions do not modify any other resource reco
y published in a DNS zone MAY be ignored by the rds.</t>
SRP registrar, which could result in a host description effectively <t>
containing zero reachable addresses even when it A and/or AAAA records that are not of sufficient scope to be
validly published in a DNS zone <bcp14>MAY</bcp14> be ignored by
the SRP registrar, which could result in a Host Description
effectively containing zero reachable addresses even when it
contains one or more addresses.</t> contains one or more addresses.</t>
<t> <t>
For example, if a link-scope address or IPv4 autoconfiguration addr For example, if an IPv4 link-local address <xref target="RFC3927"/>
ess is provided by the SRP requestor, the SRP or an IPv6 link-local address <xref target="RFC4862"/>
registrar could not publish this in a DNS zone. However, in some si is provided by the SRP requester, the SRP
tuations, the registrar might make the records registrar could elect not to publish this in a DNS zone.
available through a mechanism such as an advertising proxy only on However, in some situations, the registrar might make the records
the specific link from which the SRP update available through a mechanism such as an advertising proxy only on
originated; in such a situation, locally-scoped records are still v the specific link from which the SRP Update
alid.</t> originated. In such a situation, locally scoped records are still v
alid.</t>
</section> </section>
</section> </section>
<section> <section anchor="validation">
<name>Valid SRP Update Requirements</name> <name>Valid SRP Update Requirements</name>
<t> <t>
An SRP Update MUST contain exactly one Host Description Instruction. An SRP Update <bcp14>MUST</bcp14> contain exactly one Host Descriptio
In addition, there MUST NOT be any Service n Instruction.
Description Instruction to which no Service Discovery Instruction poi Multiple Service Discovery updates and Service Description update
nts. A DNS Update that contains any additional s
adds or deletes that cannot be identified as Service Discovery, Servi may be combined into a single single SRP Update
ce Description or Host Description Instructions is along with a single Host Description update,
as described in <xref target="how"/>.
A DNS Update message that contains any additional
adds or deletes that cannot be identified as Service Discovery, Servi
ce Description, or Host Description Instructions is
not an SRP Update. A DNS update that contains any prerequisites is no t an SRP Update.</t> not an SRP Update. A DNS update that contains any prerequisites is no t an SRP Update.</t>
<t>An SRP Update MUST include an EDNS(0) Update Lease option <t>An SRP Update <bcp14>MUST</bcp14> include an EDNS(0) Update Lease op
<xref target="I-D.ietf-dnssd-update-lease"/>. The LEASE time specifie tion
d in the Update Lease option MUST be less than <xref target="RFC9664"/>.
The LEASE time specified in the Update Lease
option <bcp14>MUST</bcp14> be less than
or equal to the KEY-LEASE time. A DNS update that does not include th e Update Lease option, or that includes a or equal to the KEY-LEASE time. A DNS update that does not include th e Update Lease option, or that includes a
KEY-LEASE value that is less than the LEASE value, is not an SRP upda KEY-LEASE value that is less than the LEASE value, is not an SRP Upda
te.</t> te.</t>
<t>When an SRP registrar receives a DNS Update that is not an SRP updat <t>When an SRP registrar receives a DNS Update message that is not an S
e, it MAY RP
process the update as regular RFC2136 updates, including access contr update, it <bcp14>MAY</bcp14> process the update as normal DNS Update
ol checks and constraint <xref target="RFC2136"/>, including
checks, if supported. Otherwise the SRP registrar MUST reject the DNS access control checks and constraint checks, if supported. Otherwise,
Update with the Refused RCODE.</t> the SRP registrar <bcp14>MUST</bcp14> reject the DNS Update with the
Refused RCODE.</t>
<t> <t>
If the definitions of each of these instructions are followed careful If the definitions of each of these instructions are followed
ly and the update requirements are validated carefully and the update requirements are validated correctly,
correctly, many DNS Updates that look very much like SRP Updates neve many DNS Update messages that look very much like SRP Updates
rtheless will fail to validate. For example, a DNS nevertheless will fail to validate. For example, a DNS update
update that contains an Add to an RRset instruction for a Service Nam that contains an "Add To An RRset" instruction for a Service Name
e and an Add to an RRset instruction for a Service and an "Add to an RRset" instruction for a service instance name
Instance Name, where the PTR record added to the Service Name does no where the PTR record added to the Service Name does not reference
t reference the Service Instance Name, is not a the service instance name is not a valid SRP Update but may be a
valid SRP Update message, but may be a valid RFC2136 update.</t> valid DNS Update.</t>
</section> </section>
<section> <section>
<name>FCFS Name And Signature Validation</name> <name>FCFS Name and Signature Validation</name>
<!--[rfced] For the ease of the reader, might we clarify what "these
conditions" are?
Original:
Assuming that a DNS Update message has been validated with these
conditions and is a valid SRP Update, the SRP registrar checks that
the name in the Host Description Instruction exists.
Perhaps:
Assuming that a DNS Update message has been validated with an FCFS name
and signature and is a valid SRP Update, the SRP registrar checks that
the name in the Host Description Instruction exists.
[Ted] No, that's not what "these conditions" means. How about:
-->
<t> <t>
Assuming that a DNS Update message has been validated with these cond Assuming that the SRP registrar has confirmed that a DNS Update messa
itions and is a valid SRP Update, the SRP registrar ge
checks that the name in the Host Description Instruction exists. If is a valid SRP Update (<xref target="validation"/>), it
so, then the registrar checks to see if the KEY then checks that the name in the Host Description Instruction exists
in the zone being updated. If so, then the registrar checks to see if the KEY
record on that name is the same as the KEY record in the Host Descrip tion Instruction. The registrar performs the same record on that name is the same as the KEY record in the Host Descrip tion Instruction. The registrar performs the same
check for the KEY records in any Service Description Instructions. F or KEY records that were omitted from Service check for the KEY records in any Service Description Instructions. F or KEY records that were omitted from Service
Description Instructions, the KEY from the Host Description Instructi on is used. If any existing KEY record Description Instructions, the KEY from the Host Description Instructi on is used. If any existing KEY record
corresponding to a KEY record in the SRP Update does not match the KE Y record in the SRP Update (whether provided corresponding to a KEY record in the SRP Update does not match the KE Y record in the SRP Update (whether provided
or taken from the Host Description Instruction), then the SRP registr or taken from the Host Description Instruction), then the SRP registr
ar MUST reject the SRP Update with the YXDomain ar
RCODE.</t> <bcp14>MUST</bcp14> reject the SRP Update with an YXDomain
RCODE indicating that the desired name is already owned by a differen
t SIG(0) key.
This informs the SRP requester that it should select a different name
and try again.</t>
<!--[rfced] Please review this transition sentence. Because it is
placed at the beginning of a new paragraph, the "Otherwise" might
be a bit jarring to the reader. (Our suggestion is likely weak,
but for demonstrative purposes...)
Original:
Otherwise, the SRP registrar validates the SRP Update using SIG(0)
against the public key in the KEY record of the Host Description
Instruction.
Perhaps:
If the above steps are not taken, the SRP registrar validates the
SRP Update using SIG(0) against the public key in the KEY record of
the Host Description Instruction.
[Ted] Hm, not easy. See edit below.
-->
<t> <t>
Otherwise, the SRP registrar validates the SRP Update using SIG(0) ag If the SRP Update is not in conflict with existing data in the zone b
ainst the public key in the KEY record of the Host eing updated, the SRP registrar validates the SRP Update using SIG(0) against th
Description Instruction. If the validation fails, the registrar MUST e public key in the KEY record of the Host
reject the SRP Update with the Refused RCODE. Description Instruction. If the validation fails,
Otherwise, the SRP Update is considered valid and authentic, and is p the SRP Update is malformed, and the registrar
rocessed according to the method described in <bcp14>MUST</bcp14> reject the SRP Update with the Refused RCODE.
RFC2136.</t> Otherwise, the SRP Update is considered valid and authentic and
is processed as for a normal DNS Update <xref target="RFC2136"/>.</t>
<!-- [Ted] The comma after KEY RR was changed to a colon, and I can't see why. H
ow about: -->
<t> <t>
KEY record updates omitted from Service Description Instruction are p KEY record updates omitted from Service Description Instruction(s) ar
rocessed as if they had been explicitly present: e processed as if they had been explicitly present.
every Service Description that is updated MUST, after the SRP Update After the SRP Update has been applied, every Service Description that
has been applied, have a KEY RR, and it must be the is updated <bcp14>MUST</bcp14> have a KEY RR, which <bcp14>MUST</bcp14> have th
same KEY RR that is present in the Host Description to which the Serv e
ice Description refers.</t> same valua as the KEY RR that is present in the Host Description to w
hich the Service Description refers.</t>
<t> <t>
<xref target="RFC3445"/> states that the flags field in the KEY RR MU The IETF specification for
ST be zero except for bit 7, which can DNSSEC Resource Records <xref target="RFC4034"/>
be one in the case of a zone key. However, the SRP registrar MUST NOT states that the flags field in the KEY RR
validate the flags field.</t> <bcp14>MUST</bcp14> be zero except for bit 7, which can
be one in the case of a zone key.
SRP requesters implementing this version of the SRP specification
<bcp14>MUST</bcp14> set the flags field in the KEY RR to all zeroes.
SRP registrars implementing this version of the SRP specification
<bcp14>MUST</bcp14> accept and store the flags field in the KEY RR
as received, without checking or modifying its value.</t>
</section> </section>
<section> <section>
<name>Handling of Service Subtypes</name> <name>Handling of Service Subtypes</name>
<t> <t>
SRP registrars MUST treat the update instructions for a service type and all its subtypes as atomic. That is, when a SRP registrars <bcp14>MUST</bcp14> treat the update instructions for a service type and all its subtypes as atomic. That is, when a
service and its subtypes are being updated, whatever information appe ars in the SRP Update is the entirety of service and its subtypes are being updated, whatever information appe ars in the SRP Update is the entirety of
information about that service and its subtypes. If any subtype appea red in a previous update but does not appear in information about that service and its subtypes. If any subtype appea red in a previous update but does not appear in
the current update, then the SRP registrar MUST remove that subtype. the current update, then the SRP registrar <bcp14>MUST</bcp14> remove that subtype.
</t> </t>
<t> <t>
Similarly, there is no mechanism for deleting subtypes. A delete of a There is intentionally no mechanism for deleting a single subtype
service deletes all of its subtypes. To delete an individually. A delete of a service deletes all of its subtypes.
individual subtype, an SRP Update must be constructed that contains t To delete a single subtype individually, an SRP Update must
he service type and all subtypes for that service be constructed that contains the service type and all subtypes
except for the one to be deleted. for that service except for the subtype(s) to be deleted.
</t> </t>
</section> </section>
<section> <section>
<name>SRP Update response</name> <name>SRP Update Response</name>
<t> <t>
The status that is returned depends on the result of processing the u The status that is returned depends on the result of processing the u
pdate, and can be either NoError, ServFail, Refused pdate and can be either NoError, ServFail, Refused,
or YXDomain: all other possible outcomes will already have been accou or YXDomain. All other possible outcomes will already have been accou
nted for when applying the constraints that nted for when applying the constraints that
qualify the update as an SRP Update. The meanings of these responses qualify the update as an SRP Update. The meanings of these responses
are explained in <xref target="RFC2136" are explained in
section="2.2"/>.</t> Section <xref target="RFC2136" section="2.2" sectionFormat="bare"/>
of the DNS Update specification <xref target="RFC2136"/>.</t>
<t> <t>
In the case of a response other than NoError, <xref target="RFC2136" In the case of a response other than NoError,
section="3.8"/> specifies that the server is permitted Section <xref target="RFC2136" section="3.8" sectionFormat="bare"/>
to respond either with no RRs or to copy the RRs sent by the client of the DNS Update specification <xref target="RFC2136"/>
into the response. The SRP Requestor MUST NOT attempt states that
the authoritative DNS server is permitted
to respond either with no RRs or to copy the RRs
sent by the DNS Update client into the response.
The SRP requester <bcp14>MUST NOT</bcp14> attempt
to validate any RRs that are included in the response. It is possible that a future SRP extension may include per-RR to validate any RRs that are included in the response. It is possible that a future SRP extension may include per-RR
indications as to why the update failed, but at present this is not s indications as to why the update failed, but
pecified, so if a client were to attempt to validate at the time of writing this is not specified.
the RRs in the response, it might reject such a response, since it w So, if an SRP requester were to attempt to validate
ould contain RRs, but probably not a set of RRs the RRs in the response, it might reject such a response, since it w
ould contain RRs but probably not a set of RRs
identical to what was sent in the SRP Update.</t> identical to what was sent in the SRP Update.</t>
</section> </section>
<section> <section>
<name>Optional Behavior</name> <name>Optional Behavior</name>
<!-- [Ted] I don't think e.g. should have a comma after it. I changed it to "for
example" to
illustrate why I think this, but my Latin is rusty, so maybe it does make s
ense when the
abbreviation is used? Ah, I see why I'm confused. In most of the cases wher
e e.g. or for example
is being used, it's being used like this: If we use foo, for example, then
BAR. But here debugging
isn't the example, so the extra comma changes the meaning. -->
<t> <t>
The SRP registrar MAY add a Reverse Mapping (<xref target="RFC1035" s The SRP registrar <bcp14>MAY</bcp14> add a Reverse Mapping PTR record
ection="3.5"/>, <xref target="RFC3596" section="2.5"/>) (described for IPv4 in
that corresponds to the Host Description. This is not required becau <xref target="RFC1035" section="3.5" sectionFormat="of"/>
se the Reverse Mapping serves no protocol function, of the DNS specification <xref target="RFC1035"/>
but it may be useful for debugging, e.g. in annotating network packet and for IPv6 in
traces or logs. In order for the registrar to do <xref target="RFC3596" section="2.5" sectionFormat="of"/>
a reverse mapping update, it must be authoritative for the zone that of the later document updating DNS for IPv6 <xref
would need to be updated, or have credentials to do target="RFC3596"/>) that corresponds to the Host Description.
the update. The SRP requestor MAY also do a reverse mapping update i This is optional: The reverse mapping PTR record serves no
f it has credentials to do so.</t> essential protocol function. One reason to provide reverse
<t> mappings is that they can be used to annotate logs and network
The SRP registrar MAY apply additional criteria when accepting update packet traces. In order for the registrar to do a reverse mapping
s. In some networks, it may be possible to do update, it must be authoritative for the zone that would need to
out-of-band registration of keys, and only accept updates from pre-re be updated or have credentials to do the update. The SRP
gistered keys. In this case, an update for a key requester <bcp14>MAY</bcp14> also do a reverse mapping update if
that has not been registered SHOULD be rejected with the Refused RCOD it has credentials to do so.</t>
E.</t>
<t> <t>
There are at least two benefits to doing this rather than simply usin The SRP registrar <bcp14>MAY</bcp14> apply additional criteria when a
g normal SIG(0) DNS updates. First, the same ccepting updates. In some networks, it may be possible to do
registration protocol can be used in both cases, so both use cases ca out-of-band registration of keys and only accept updates from preregi
n be addressed by the same SRP requestor stered keys. In this case, an update for a key
implementation. Second, the registration protocol includes maintenan that has not been registered <bcp14>SHOULD</bcp14> be rejected with t
ce functionality not present with normal DNS he Refused RCODE.
updates.</t> When use of managed keys is desired,
there are at least two benefits to doing this in conjunction with SRP
rather than simply performing traditional DNS Updates using SIG(0) ke
ys:</t>
<ol><li>The same
over-the-air registration protocol is used in both cases,
so both use cases can be addressed by the same SRP requester
implementation.</li>
<li>The Service Registration Protocol includes
maintenance functionality not present with normal DNS
updates.</li></ol>
<t> <t>
Note that the semantics of using SRP in this way are different than f Note that the semantics of using SRP in this way
or typical RFC2136 implementations: the KEY used are different from the semantics of typical
to sign the SRP Update only allows the SRP requestor to update record implementations of DNS Update. The KEY used
s that refer to its Host Description. RFC2136 to sign the SRP Update only allows the SRP requester to update record
implementations do not normally provide a way to enforce a constraint s that refer to its Host Description.
of this type.</t> Implementations of a traditional DNS Update
<xref target="RFC2136"/> do not normally provide
a way to enforce a constraint of this type.</t>
<t> <t>
The SRP registrar could also have a dictionary of names or name patte rns that are not permitted. If such a list is used, The SRP registrar could also have a dictionary of names or name patte rns that are not permitted. If such a list is used,
updates for Service Instance Names that match entries in the dictiona ry are rejected with a Refused RCODE.</t> updates for service instance names that match entries in the dictiona ry are rejected with a Refused RCODE.</t>
</section> </section>
</section> </section>
</section> </section>
<section> <section>
<name>TTL Consistency</name> <name>TTL Consistency</name>
<t> <t>
All RRs within an RRset are required to have the same TTL All RRs within an RRset are required to have the same TTL
(<xref target="RFC2181" section="5.2" sectionFormat="comma"> Clarificatio (required by
ns to the DNS Specification</xref>). Section <xref target="RFC2181" section="5.2" sectionFormat="bare"/>
In order to avoid inconsistencies, SRP places restrictions on TTLs sent b of the DNS Clarifications document <xref target="RFC2181"/>).
y requestors and requires that SRP registrars enforce In order to avoid inconsistencies, SRP places restrictions on TTLs sent b
y requesters and requires that SRP registrars enforce
consistency.</t> consistency.</t>
<t> <t>
Requestors sending SRP Updates MUST use consistent TTLs in all RRs within Requesters sending SRP Updates <bcp14>MUST</bcp14> use consistent
the SRP Update.</t> TTLs in all RRs within each RRset contained within an SRP Update.</t>
<t> <t>
SRP registrars MUST check that the TTLs for all RRs within the SRP Update SRP registrars <bcp14>MUST</bcp14> check that the TTLs for all RRs
are the same. If they are not, the SRP within each RRset contained within an SRP Update are the same.
update MUST be rejected with a Refused RCODE.</t> If they are not, the SRP
update <bcp14>MUST</bcp14> be rejected with a Refused RCODE.</t>
<t> <t>
Additionally, when adding RRs to an RRset, for example when processing Se rvice Discovery records, the SRP registrar MUST use the Additionally, when adding RRs to an RRset (for example, when processing S ervice Discovery records), the SRP registrar <bcp14>MUST</bcp14> use the
same TTL on all RRs in the RRset. How this consistency is enforced is up to the implementation.</t> same TTL on all RRs in the RRset. How this consistency is enforced is up to the implementation.</t>
<t> <t>
TTLs sent in SRP Updates are advisory: they indicate the SRP requestor's TTLs sent in SRP Updates are advisory: they indicate the SRP requester's
guess as to what a good TTL would be. SRP registrars may guess as to what a good TTL would be. SRP registrars may
override these TTLs. SRP registrars SHOULD ensure that TTLs are reasonab override these TTLs. SRP registrars <bcp14>SHOULD</bcp14> ensure that TT
le: neither too long nor too short. The TTL SHOULD NOT Ls are reasonable: neither too long nor too short. The TTL <bcp14>SHOULD NOT</b
cp14>
ever be longer than the lease time (<xref target="stale"/>). Shorter TTL s will result in more frequent data refreshes; ever be longer than the lease time (<xref target="stale"/>). Shorter TTL s will result in more frequent data refreshes;
this increases latency on the DNS-SD client side, increases load on any c this increases latency on the DNS&nbhy;SD client side, increases
aching resolvers and on the authoritative server, load on any caching resolvers and on the authoritative DNS server,
and also increases network load, which may be an issue for constrained ne and also increases network load, which may be an issue for CNNs. Longer
tworks. Longer TTLs will increase the likelihood TTLs will increase the likelihood
that data in caches will be stale. TTL minimums and maximums SHOULD be c that data in caches will be stale. TTL minimums and maximums <bcp14>SHOU
onfigurable by the operator of the SRP registrar. LD</bcp14> be configurable by the operator of the SRP registrar.
</t> </t>
</section> </section>
<section anchor="maintenance"> <section anchor="maintenance">
<name>Maintenance</name> <name>Maintenance</name>
<section anchor="stale"> <section anchor="stale">
<name>Cleaning up stale data</name> <name>Cleaning Up Stale Data</name>
<t>Because the DNS&nbhy;SD registration protocol is automatic, and not ma <t>Because the DNS&nbhy;SD Service Registration Protocol
naged by humans, is automatic and not managed by humans,
some additional bookkeeping is required. When an update is constructe some additional bookkeeping is required. When an update is constructe
d by the SRP requestor, d by the SRP requester,
it MUST include an EDNS(0) Update Lease Option <xref target="I-D.ietf- it <bcp14>MUST</bcp14> include an EDNS(0) Update Lease Option <xref ta
dnssd-update-lease"/>. rget="RFC9664"/>.
The Update Lease Option contains two lease times: the Lease Time and t he KEY The Update Lease Option contains two lease times: the Lease Time and t he KEY
Lease Time.</t> Lease Time.</t>
<t>These leases are promises, similar to <xref target="RFC2131">DHCP leas <t>Similar to DHCP leases <xref target="RFC2131"/>,
es</xref>, these leases are promises from the SRP requester that it will
from the SRP requestor that it will send a new update for the service send a new update for the service registration before the
registration before the lease time expires.
lease time expires. The Lease time is chosen to represent the time af The Lease time is chosen to represent the duration after the update
ter the during which the registered records other than the KEY record
update during which the registered records other than the KEY record c can be assumed to be valid.
an be assumed The KEY lease time represents the duration after the update
to be valid. The KEY lease time represents the time after the update during which the KEY record can be assumed to be valid.
during The reasoning behind the different lease times is discussed in
which the KEY record can be assumed to be valid.</t> Sections <xref target="fcfs" format="counter"/> and
<xref target="lifetimes" format="counter"/>.</t>
<t>The reasoning behind the different lease times is discussed in the sec <t>SRP registrars may be configured with limits for these values.
tion on FCFS naming At the time of writing, a default limit of two hours for
(<xref target="fcfs"/>). SRP registrars may be configured with limits the Lease and 14 days for the SIG(0) KEY are thought to be good choice
for these values. A default limit of two hours for s. Devices with limited
the Lease and 14 days for the SIG(0) KEY are currently thought to be g
ood choices. Constrained devices with limited
battery that wake infrequently are likely to request longer leases; re gistrars that support such devices may need to set battery that wake infrequently are likely to request longer leases; re gistrars that support such devices may need to set
higher limits. SRP requestors that are going to continue to use names higher limits. SRP requesters that are going to
on which they hold leases SHOULD update well before continue to use names on which they hold leases
the lease ends, in case the registrar is unavailable or under heavy lo <bcp14>SHOULD</bcp14> refresh them well before
ad.</t> the lease ends in case the registrar is
temporarily unavailable or under heavy load.</t>
<t> <t>
The lease time applies specifically to the host. All service instances, The lease time applies specifically to the hostname.
and all service entries for such service All service instances, and all service entries for such service
instances, depend on the host. When the lease on a host expires, the ho instances, depend on the hostname. When the lease on a
st and all services that reference it MUST be hostname expires, the hostname and all services that
removed at the same time&mdash;it is never valid for a service instance reference it <bcp14>MUST</bcp14> be removed at the same
to remain when the host it references has been time: It is never valid for a service instance to remain
removed. If the KEY record for the host is to remain, the KEY record fo when the hostname it references has been removed.
r any services that reference it MUST also If the KEY record for the hostname is to remain, the KEY record
remain. However, the service PTR record MUST be removed, since it has n for any services that reference it <bcp14>MUST</bcp14> also
o key associated with it, and since it is never remain. However, the Service Discovery PTR record <bcp14>MUST</bcp14>
valid to have a service PTR record for which there is no service instan be removed since it has no key associated with it and since it
ce on the target of the PTR record. is never valid to have a Service Discovery PTR record for which
there is no service instance on the target of the PTR record.
</t> </t>
<t> <t>
SRP registrars MUST also track a lease time per service instance. The r SRP registrars <bcp14>MUST</bcp14> also track a lease time per service
eason for doing this is that a requestor may instance. The reason being that a requester may
re-register a host with a different set of services, and not remember t re-register a hostname with a different set of services and
hat some different service instance had previously not remember that some different service instance had previously
been registered. In this case, when that service instance lease expires been registered. In this case, when that service instance lease expires
, the SRP registrar MUST remove the service , the SRP registrar <bcp14>MUST</bcp14> remove the service
instance (although the KEY record for the service instance SHOULD be re instance,
tained until the KEY lease on that service and any associated Service Discovery PTR records pointing to that servi
expires). This is beneficial because otherwise if the SRP requestor con ce instance,
tinues to renew the host, but never mentions the (although the KEY record for the service instance
stale service again, the stale service will continue to be advertised. <bcp14>SHOULD</bcp14> be retained until the
KEY lease on that service
expires).
This is beneficial because it avoids stale services continuing
to be advertised after the SRP requester has forgotten about them.
</t> </t>
<t>The SRP registrar MUST include an EDNS(0) Update Lease option in the <t>The SRP registrar <bcp14>MUST</bcp14> include an EDNS(0) Update Lease
response if the lease time proposed by the requestor has been shortene option in the
d or lengthened by the registrar. The requestor response. The requester
MUST check for the EDNS(0) Update Lease option in the response and MUS <bcp14>MUST</bcp14> check for the EDNS(0) Update Lease option
T use the lease in the response, and when deciding when to renew its
times from that option in place of the options that it sent to the reg registration the requester <bcp14>MUST</bcp14> use the
istrar when lease times from that received option in place of the
deciding when to renew its registration. The times may be shorter or lease times that it originally requested from the registrar.
longer than The times may be shorter or longer than
those specified in the SRP Update; the SRP requestor must honor them i those specified in the SRP Update. The SRP requester must honor them i
n either case.</t> n either case.</t>
<t>SRP requestors SHOULD assume that each lease ends N seconds after the <!-- [rfced] In Section 5.1, we see both "N" and "'n'". Please review
update was first and let us know if/how we may update for uniformity.
transmitted, where N is the lease duration. SRP Registrars SHOULD ass
ume that each lease
ends N seconds after the update that was successfully processed was re
ceived. Because
the registrar will always receive the update after the SRP requestor s
ent it, this avoids the
possibility of misunderstandings.</t>
<t>SRP registrars MUST reject updates that do not include an Original "N":
EDNS(0) Update Lease option. DNS authoritative servers that allow bot SRP requesters <bcp14>SHOULD</bcp14> assume that each lease ends N seconds af
h SRP and non-SRP DNS updates MAY accept updates that don't include ter the
leases, but SHOULD differentiate between SRP Updates and update was first transmitted, where N is the lease duration.
other updates, and MUST reject updates that would otherwise be SRP Upd
ates
if they do not include leases.</t>
<t>Lease times have a completely different function than TTLs. On an aut Original "'n'":
horitative The lease time is never sent as a TTL; its
DNS server, the TTL on a resource record is a constant: whenever that sole purpose is to determine when the authoritative DNS server will
RR is served in delete stale records. It is not an error to send a DNS response with
a DNS response, the TTL value sent in the answer is the same. The lea a TTL of 'n' when the remaining time on the lease is less than 'n'.
se time is never
sent as a TTL; its sole purpose is to determine when the authoritative [Ted] These are actually two different quantities, but I agree with the correcti
DNS server will on generally, so have used M for the last paragraph.
delete stale records. It is not an error to send a DNS response with -->
a TTL of 'n' when
the remaining time on the lease is less than 'n'.</t> <t>SRP requesters <bcp14>SHOULD</bcp14> assume that each lease ends N
seconds after the update was first transmitted (where N is the granted le
ase
duration). SRP registrars <bcp14>SHOULD</bcp14> assume that each lease
ends N seconds after the update that was successfully processed was
received. Because the registrar will always receive the update after
the SRP requester sent it, this avoids the possibility of
a race condition where the SRP registrar prematurely removes
a service when the SRP requester thinks the lease has not yet expired.
In addition, the SRP requester <bcp14>MUST</bcp14> begin attempting to re
new
its lease in advance of the expected expiration time, as required
by the DNS Update Lease specification <xref target="RFC9664"/>,
to accomodate the situation where the clocks on the SRP requester
and the SRP registrar do not run at precisely the same rate.</t>
<t>SRP registrars <bcp14>MUST</bcp14> reject updates that do not
include an EDNS(0) Update Lease option. DNS authoritative servers
that allow both SRP and non-SRP DNS updates <bcp14>MAY</bcp14> accept
updates that don't include leases, but they <bcp14>SHOULD</bcp14>
differentiate between SRP Updates and other updates and
<bcp14>MUST</bcp14> reject updates that would otherwise be SRP Updates
if they do not include leases.</t>
<t>The function of Lease times and the
function of TTLs are completely different. On an
authoritative DNS server, the TTL on a resource record is a
constant. Whenever that RR is served in a DNS response, the TTL value
sent in the answer is the same. The lease time is never sent as a
TTL; its sole purpose is to determine when the authoritative DNS
server will delete stale records. It is not an error to send a DNS
response with a TTL of M when the remaining time on the lease is
less than M.</t>
</section> </section>
</section> </section>
<section> <section>
<name>Security Considerations</name> <name>Security Considerations</name>
<section anchor="source_validation"> <section anchor="source_validation">
<name>Source Validation</name> <name>Source Validation</name>
<t>SRP Updates have no authorization semantics other than <t>SRP Updates have no authorization semantics other than
FCFS. This means that if an attacker from outside of the administrati "First Come, First Served" (FCFS).
ve Thus, if an attacker from outside the administrative
domain of the SRP registrar knows the registrar's IP address, it can in domain of the SRP registrar knows the registrar's IP address, it can, i
principle send updates to the registrar n principle, send updates to the registrar
that will be processed successfully. SRP Registrars SHOULD therefore that will be processed successfully. Therefore, SRP registrars <bcp14
be configured to reject updates >SHOULD</bcp14> be configured to reject updates
from source addresses outside of the administrative domain of the regis trar.</t> from source addresses outside of the administrative domain of the regis trar.</t>
<t>For TCP updates, the initial SYN-SYN+ACK handshake prevents updates be <t>For TCP updates, the initial SYN-SYN+ACK handshake prevents
ing forged by an off-network attacker. In order to updates being forged by an off-path attacker. In order to
ensure that this handshake happens, SRP registrars relying on three-way ensure that this handshake happens, SRP registrars relying on three-way
-handshake validation MUST NOT accept TCP Fast Open -handshake validation <bcp14>MUST NOT</bcp14> accept TCP Fast Open payloads
<xref target="RFC7413"/> payloads. If the network infrastructure allow <xref target="RFC7413"/>.
s it, an SRP registrar MAY accept TCP Fast Open payloads if all such packets If the network infrastructure allows it, an SRP registrar
<bcp14>MAY</bcp14> accept TCP Fast Open payloads if all such packets
are validated along the path, and the network is able to reject this ty pe of spoofing at all ingress points.</t> are validated along the path, and the network is able to reject this ty pe of spoofing at all ingress points.</t>
<t>For UDP updates from constrained devices, spoofing would have to be pr <t>For UDP updates from CNN devices, spoofing would have to be prevented
evented with appropriate source address filtration with appropriate source address filtering
on routers <xref target="RFC2827"/>. This would ordinarily be accomplis on routers <xref target="RFC2827"/>.
hed by measures such as are described in This would ordinarily be accomplished by measures such as those describ
<xref target="RFC7084" section="4.5" sectionFormat="of"/>. For example, ed in
a stub router <xref target="I-D.ietf-snac-simple"/> Section <xref target="RFC7084" section="4.5" sectionFormat="bare"/>
for a constrained network might only accept UDP updates from source add of the IPv6 CE Router Requirements document <xref target="RFC7084"/>.
resses known to be on-link on that stub network, and might For example, a stub router <xref target="I-D.ietf-snac-simple"/>
for a CNN might only accept UDP updates from source addresses known to
be on-link on that stub network and might
further validate that the UDP update was actually received on the stub network interface and not the interface connected to further validate that the UDP update was actually received on the stub network interface and not the interface connected to
the adjacent infrastructure link.</t> the adjacent infrastructure link.</t>
</section> </section>
<section> <section>
<name>Other DNS updates</name> <name>Other DNS Updates</name>
<t>Note that these rules only apply to the validation of SRP Updates. <t>Note that these rules only apply to the validation of SRP Updates.
A server that accepts updates from SRP An authoritative DNS server that accepts updates from SRP
requestors may also accept other DNS updates, and those DNS updates may requesters may also accept other DNS Update messages, and those DNS Upd
be validated ate messages may be validated
using different rules. However, in the case of a DNS server that acce using different rules.
pts SRP However, in the case of an authoritative DNS server that accepts SRP
updates, the intersection of the SRP Update rules and updates, the intersection of the SRP Update rules and
whatever other update rules are present must be considered very careful ly.</t> whatever other update rules are present must be considered very careful ly.</t>
<t>For example, a normal, authenticated DNS update to any RR that was add <t>For example, a normal authenticated DNS update to any
ed using SRP, but that is authenticated using a RR that was added using SRP, but is authenticated using a
different key, could be used to override a promise made by the SRP regi different key, could be used to override a promise made by the SRP regi
strar to an SRP requestor, by replacing all or part of strar to an SRP requester by replacing all or part of
the service registration information with information provided by an au the service registration information with information provided by an au
thenticated DNS update requestor. An implementation thenticated DNS update requester. An implementation
that allows both kinds of updates SHOULD NOT allow DNS Update requestor that allows both kinds of updates <bcp14>SHOULD NOT</bcp14> allow DNS U
s that are using different authentication and pdate requesters that are using different authentication and
authorization credentials to update records added by SRP requestors.</t authorization credentials to update records added by SRP requesters.</t
> >
</section> </section>
<section> <section>
<name>Risks of allowing arbitrary names to be registered in SRP updates</ <name>Risks of Allowing Arbitrary Names to be Registered in SRP Updates</
name> name>
<t>It is possible to set up SRP updates for a zone that is used for non-D <t>It is possible to set up SRP Updates for a zone
NSSD services. For example, imagine that you set that is also used for non-DNS&nbhy;SD records.
up SRP service for example.com. SRP hosts can now register names like " For example, imagine that you set
www" or "mail" or "smtp" in this domain. In addition, up SRP service for example.com.
SRP updates using FCFS naming can insert names that are obscene or offe SRP requesters can now register names like "www"
nsive into the zone. There is no simple solution to or "mail" or "smtp" in this domain. In addition,
these problems. We have two recommendations to address this problem, ho SRP Updates using FCFS Naming can insert names that are obscene or offe
wever:</t> nsive into the zone. There is no simple solution to
<ul spacing="compact"> these problems. However, we have two recommendations to address this pr
<li>Do not provide SRP service in organization-level zones. Use subdoma oblem:</t>
ins of the organizational domain for DNS service <ul spacing="normal">
discovery. This does not prevent registering names as mentioned abov <li>Do not provide SRP service in organization-level zones.
e, but does ensure that genuinely important names Use subdomains of the organizational domain for DNS&nbhy;SD.
are not accidentally reserved for SRP clients. So for example, the zo This does not prevent registering names as mentioned above
ne "dnssd.example.com" could be used instead of but does ensure that genuinely important names
"example.com" for SRP updates. Because of the way that DNS browsing d are not accidentally claimed by SRP requesters.
omains are discovered, there is no need for the So, for example, the zone "dnssd.example.com." could be used
DNSSD discovery zone that is updated by SRP to have a user-friendly o instead of "example.com." for SRP Updates. Because of the way that
r important-sounding name.</li> DNS-browsing domains are discovered, there is no need for the
DNS&nbhy;SD discovery zone that is updated by SRP to
have a user-friendly or important-sounding name.</li>
<li>Configure a dictionary of names that are prohibited. Dictionaries o f common obscene and offensive names are no doubt <li>Configure a dictionary of names that are prohibited. Dictionaries o f common obscene and offensive names are no doubt
available, and can be augmented with a list of typical "special" name available and can be augmented with a list of typical "special" names
s like "www", "mail", "smtp" and so on. Lists of like "www", "mail", "smtp", and so on. Lists of
names are generally available, or can be constructed manually.</li> names are generally available or can be constructed manually.
Names rejected due to this should return a Refused
RCODE, indicating to the SRP requester that it
should not append or increment a number at the
end of the name and then try again, since this
would likely result in an infinite loop.
If a name is considered unacceptable because it is
obscene or offensive, adding a number on the end is
unlikely to make the name acceptable.</li>
</ul> </ul>
</section> </section>
<section> <section>
<name>Security of local service discovery</name> <name>Security of Local Service Discovery</name>
<t>Local links can be protected by managed services such as RA Guard <xre <t>Local links can be protected by managed services such as RA Guard
f target="RFC6105"/>, but multicast services like <xref target="RFC6105"/>, but multicast services like DHCP <xref
DHCP <xref target="RFC2131"/>, DHCPv6 <xref target="RFC8415"/> and IPv6 target="RFC2131"/>, DHCPv6 <xref target="RFC8415"/>, and IPv6 Neighbor
Neighbor Discovery <xref target="RFC4861"/> are Discovery <xref target="RFC4861"/> are, in most cases, not
in most cases not authenticated and can't be controlled on unmanaged ne authenticated and can't be controlled on unmanaged networks, such as
tworks, such as home networks and small-office home networks and small office networks where no network management
networks where no network management staff are present. In such situati staff are present. In such situations, the SRP service has
ons, the SRP service has comparatively fewer comparatively fewer potential security exposures and, hence, is not
potential security exposures and hence is not the weak link. This is di the weak link. This is discussed in more detail in <xref
scussed in more detail in target="how-to-secure"/>.</t>
<xref target="how-to-secure"/>.</t> <t>The fundamental protection for networks of this type is the user's
<t>The fundamental protection for networks of this type is the user's cho choice of what devices to add to the network. Work is being done in
ice of what devices to add to the network. Work is other working groups and standards bodies to improve the state of the
being done in other working groups and standards bodies to improve the art for network on-boarding and device isolation (e.g., Manufacturer
state of the art for network on-boarding and device Usage Descriptions <xref target="RFC8520"/> provide a means for
isolation (e.g., <xref target="RFC8520"/> provides a means for constrai constraining what behaviors are allowed for a device in an automatic
ning what behaviors are allowed for a device in an way), but such work is out of scope for this document.</t>
automatic way), but such work is out of scope for this document.</t>
</section> </section>
<section> <section>
<name>SRP Registrar Authentication</name> <name>SRP Registrar Authentication</name>
<t>This specification does not provide a mechanism for validating respons <t>This specification does not provide a mechanism for validating respons
es from SRP Registrars to es from SRP registrars to
SRP requestors. In principle, a KEY RR could be used by SRP requesters. In principle, a KEY RR could be used by
a non-constrained SRP requestor to validate responses from the registra a non-CNN SRP requester to validate responses from the registrar, but t
r, but this is not required, his is not required,
nor do we specify a mechanism for determining which key to use.</t> nor do we specify a mechanism for determining which key to use.</t>
<t>In addition, for DNS-over-TLS connections, out-of-band key pinning as described in <t>In addition, for DNS-over-TLS connections, out-of-band key pinning as described in
<xref target="RFC7858" section="4.2" sectionFormat="comma"/> could be u Section <xref target="RFC7858" section="4.2" sectionFormat="bare"/>
sed for authentication of the SRP registrar, of the DNS-over-TLS specification <xref target="RFC7858"/>
e.g. to prevent man-in-the-middle attacks. However the use of such keys could be used for authentication of the SRP registrar,
is impractical for an unmanaged service e.g., to prevent man-in-the-middle attacks. However, the use of such ke
registration protocol, and hence is out of scope for this document.</t> ys is impractical for an unmanaged service
registration protocol; hence, it is out of scope for this document.</t>
</section> </section>
<section anchor="rsa"> <section anchor="rsa">
<name>Required Signature Algorithm</name> <name>Required Signature Algorithm</name>
<t> <t>
For validation, SRP registrars MUST implement the ECDSAP256SHA256 signa For validation, SRP registrars <bcp14>MUST</bcp14> implement the ECDSAP
ture algorithm. SRP registrars SHOULD implement the 256SHA256 signature algorithm. SRP registrars <bcp14>SHOULD</bcp14> implement t
algorithms specified in <xref target="RFC8624" section="3.1" sectionFor he
mat="comma"/>, in the validation column of the algorithms that are listed in
table, that are numbered 13 or higher and have a "MUST", "RECOMMENDED", Section <xref target="RFC8624" section="3.1" sectionFormat="bare"/>
or "MAY" designation in the validation column of of the DNSSEC Cryptographic Algorithms specification <xref target="RFC8
624"/>,
in the validation column of the
table, that are numbered 13 or higher and that have a "<bcp14>MUST</bcp
14>", "<bcp14>RECOMMENDED</bcp14>", or "<bcp14>MAY</bcp14>" designation in the v
alidation column of
the table. the table.
SRP requestors MUST NOT assume that any algorithm numbered lower than 1 3 is SRP requesters <bcp14>MUST NOT</bcp14> assume that any algorithm number ed lower than 13 is
available for use in validating SIG(0) signatures.</t> available for use in validating SIG(0) signatures.</t>
</section> </section>
</section> </section>
<section> <section>
<name>Privacy Considerations</name> <name>Privacy Considerations</name>
<t> <t>
Because DNS-SD SRP Updates can be sent off-link, the privacy implications Because DNS&nbhy;SD SRP Updates can be sent off-link,
of SRP are different than for multicast DNS the privacy implications of SRP are
responses. Host implementations that are using TCP SHOULD also use TLS i different from those for mDNS responses.
f available. SRP Registrar implementations MUST offer SRP Requester implementations that are using TCP <bcp14>SHOULD</bcp14>
TLS support. The use of TLS with DNS is described in <xref target="RFC78 also use DNS-over-TLS <xref target="RFC7858"/> if available.
58"/>. Because there is no mechanism for sharing SRP registrar implementations <bcp14>MUST</bcp14> offer TLS support.
keys, validation of DNS-over-TLS keys is not possible; DNS-over-TLS is us Because there is no mechanism for sharing keys,
ed only as described in validation of DNS-over-TLS keys is not possible;
<xref target="RFC7858" section="4.1" sectionFormat="comma"/> DNS-over-TLS is used only for Opportunistic Privacy, as documented in
Section <xref target="RFC7858" section="4.1" sectionFormat="bare"/>
of the DNS-over-TLS specification <xref target="RFC7858"/>.
</t> </t>
<t> <t>
Hosts that implement TLS support SHOULD NOT fall back to TCP; since SRP r SRP requesters that are able to use TLS <bcp14>SHOULD NOT</bcp14>
egistrars are required to support fall back to TCP. Since all SRP registrars are required to support TLS,
TLS, it is entirely up to the host implementation whether to use it. whether to use TLS is entirely the decision of the SRP requester.
</t> </t>
<t> <t>
Public keys can be used as identifiers to track hosts. SRP registrars MAY elect not to return KEY records for queries for Public keys can be used as identifiers to track hosts. SRP registrars <bc p14>MAY</bcp14> elect not to return KEY records for queries for
SRP registrations. To avoid DNSSEC validation failures, an SRP registrar that signs the zone for DNSSEC but refuses to return SRP registrations. To avoid DNSSEC validation failures, an SRP registrar that signs the zone for DNSSEC but refuses to return
a KEY record MUST NOT store the KEY record in the zone itself. Because th a KEY record <bcp14>MUST NOT</bcp14> store the KEY record in the zone its
e KEY record isn't in the zone, the nonexistance of elf. Because the KEY record isn't in the zone, the nonexistence of
the KEY record can be validated. If the zone is not signed, the server MA the KEY record can be validated.
Y instead return a negative non-error response If the zone is not signed, the authoritative DNS server <bcp14>MAY</bcp14
(either NXDOMAIN or no data). >
instead return a negative non-error response (either NXDOMAIN or no data)
.
</t> </t>
</section> </section>
<section> <section>
<name>Domain Name Reservation Considerations</name> <name>Domain Name Reservation Considerations</name>
<t>This section specifies considerations for systems involved in domain na me resolution when resolving queries for names <t>This section specifies considerations for systems involved in domain na me resolution when resolving queries for names
ending with '.service.arpa.'. Each item in this section addresses some a ending with ".service.arpa.". Each item in this section addresses some a
spect of the DNS or the process of resolving domain spect of the DNS or the process of resolving domain
names that would be affected by this special-use allocation. Detailed ex names that would be affected by this special-use allocation.
planations of these items can be found in Section 5 Detailed explanations of these items can be found in
of <xref target="RFC6761"/>.</t> Section <xref target="RFC6761" section="5" sectionFormat="bare"/>
of the Special-Use Domain Names specification <xref target="RFC6761"/>.
</t>
<section> <section>
<name>Users</name> <name>Users</name>
<t>The current proposed use for 'service.arpa' does not require special k <t>The current proposed use for "service.arpa." does not
nowledge on the part of the user. While the require special knowledge on the part of the user. While the
'default.service.arpa.' subdomain is used as a generic name for registr "default.service.arpa." subdomain is used as a generic name for registr
ation, users are not expected to see this name in ation, users are not expected to see this name in
user interfaces. In the event that it does show up in a user interface, user interfaces. In the event that it does show up in a user interface,
it is just a domain name, and requires no special it is just a domain name and requires no special
treatment by the user. Users are not expected to see this name in user treatment by the user.</t>
interfaces, although it's certainly possible that
they might. If they do, they are not expected to treat it specially.</t
>
</section> </section>
<section> <section>
<name>Application Software</name> <name>Application Software</name>
<t> <t>
Application software does not need to handle subdomains of 'service.arp Application software does not need to handle
a' specially. 'service.arpa' SHOULD NOT be treated subdomains of "service.arpa." specially.
as more trustworthy than any other insecure DNS domain, simply because "service.arpa."&nbsp;<bcp14>SHOULD NOT</bcp14> be treated
it is locally-served (or for any other reason). It as more trustworthy than any other insecure DNS domain, simply because
is not possible to register a PKI certificate for a subdomain of 'servi it is locally served (or for any other reason). It
ce.arpa.' because it is a locally-served domain is not possible to register a PKI certificate for a subdomain of "servi
name. So no such subdomain can be considered as uniquely identifying a ce.arpa." because it is a locally served domain
particular host, as would be required for such a name. So, no such subdomain can be considered to be uniquely identifyin
PKI cert to be issued. If a subdomain of 'service.arpa.' is returned by g a particular host, as would be required for such a
an API or entered in an input field of an PKI certificate to be issued. If a subdomain of "service.arpa." is retu
rned by an API or entered in an input field of an
application, PKI authentication of the endpoint being identified by the name will not be possible. Alternative methods application, PKI authentication of the endpoint being identified by the name will not be possible. Alternative methods
and practices for authenticating such endpoints are out of scope for th is document.</t> and practices for authenticating such endpoints are out of scope for th is document.</t>
</section> </section>
<section> <section>
<name>Name Resolution APIs and Libraries</name> <name>Name Resolution APIs and Libraries</name>
<t>Name resolution APIs and libraries MUST NOT recognize names that end i <t>Name resolution APIs and libraries <bcp14>MUST NOT</bcp14> recognize n
n '.service.arpa.' as special and MUST NOT treat ames that end in "service.arpa." as special and <bcp14>MUST NOT</bcp14> treat
them as having special significance, except that it may be necessary th them as having special significance, except that it may be
at such APIs not bypass the locally configured necessary that such APIs not bypass the locally discovered
recursive resolvers.</t> recursive resolvers.</t>
<t>One or more IP addresses for recursive DNS servers will usually be sup <t>One or more IP addresses for recursive resolvers will usually
plied to the client through router advertisements be supplied to the SRP requester through router advertisements
or DHCP. For an administrative domain that uses subdomains of 'service or DHCP. For an administrative domain that uses subdomains of "service
.arpa.', the recursive resolvers provided by that .arpa.", the recursive resolvers provided by that
domain will be able to answer queries for subdomains of 'service.arpa.' domain will be able to answer queries for subdomains of "service.arpa."
; other (non-local) resolvers will not, or they . Other (non-local) resolvers will not, or they
will provide answers that are not correct within that administrative do main.</t> will provide answers that are not correct within that administrative do main.</t>
<t>A host that is configured to use a resolver other than one that has be <t>A host that is configured to use a resolver other than one that has be
en provided by the local network may be unable to en provided by the local network may not be able to
resolve, or may receive incorrect results for, subdomains of 'service.a resolve or may receive incorrect results for subdomains of
rpa.'. In order to avoid this, it is permissible "service.arpa.". In order to avoid this, hosts <bcp14>SHOULD</bcp14> u
that hosts use the resolvers that are locally provided for resolving 's se the
ervice.arpa.', even when they are configured to resolvers that are locally provided for resolving "service.arpa." names
use other resolvers.</t> ,
even when they are configured to use other resolvers for other names.</
t>
</section> </section>
<section> <section>
<name>Caching DNS Servers</name> <name>Recursive Resolvers</name>
<t>There are three considerations for caching DNS servers that
follow this specification:</t> <!-- [rfced] In the following text, before the two numbered points,
<ol> the text reads "There are three considerations". Should we update
<li>For correctness, recursive resolvers at sites using 'service.arpa.' "three" to "two", or is there another point that the text is
must in practice transparently support DNSSEC missing?
queries: queries for DNSSEC records and queries with the DNSSEC OK (
DO) bit set (<xref target="RFC4035" section="3.2.1" Current:
sectionFormat="of"/>). DNSSEC validation is a Best Current Practice There are three considerations for caching DNS servers that follow
<xref target="RFC9364"/>: although validation is this specification:
not required, a caching recursive resolver that does not validate an
swers that can be validated may cache invalid data. 1. For correctness, recursive resolvers at sites using
This, in turn, would prevent validating stub resolvers from successf 'service.arpa.' must, in practice, transparently support DNSSEC
ully validating answers. Hence, as a practical queries: queries for DNSSEC records and queries with the DNSSEC
matter, recursive resolvers at sites using 'service.arpa' should do OK (DO) bit set (Section 3.2.1 of [RFC4035]). DNSSEC validation
DNSSEC validation.</li> is a Best Current Practice ([RFC9364]): although validation is not
required, a caching recursive resolver that does not validate
answers that can be validated may cache invalid data. In turn,
this would prevent validating stub resolvers from successfully
validating answers. Hence, as a practical matter, recursive
resolvers at sites using 'service.arpa' should do DNSSEC
validation.
2. Unless configured otherwise, recursive resolvers and DNS proxies
<bcp14>MUST</bcp14> behave as described in Locally Served Zones (Section
3 of
[RFC6303]). That is, queries for 'service.arpa.' and subdomains
of 'service.arpa.' <bcp14>MUST NOT</bcp14> be forwarded, with one import
ant
exception: a query for a DS record with the DO bit set <bcp14>MUST</bcp14
>
return the correct answer for that question, including correct
information in the authority section that proves that the record
is nonexistent.
So, for example, a query for the NS record for 'service.arpa.'
<bcp14>MUST NOT</bcp14> result in that query being forwarded to an upstre
am
cache nor to the authoritative DNS server for '.arpa.'. However,
as necessary to provide accurate authority information, a query
for the DS record <bcp14>MUST</bcp14> result in forwarding whatever queri
es are
necessary. Typically, this will just be a query for the DS
record since the necessary authority information will be included
in the authority section of the response if the DO bit is set.
[Ted] NOOBODY expects the Spanish Inquisition [https://en.wikipedia.org/wiki/The
_Spanish_Inquisition_(Monty_Python)]
I think it should be two. Dunno what happened here.
-->
<t>There are two considerations for recursive resolvers
(also known as "caching DNS servers" or "recursive DNS servers") that
follow this specification:</t>
<!--[rfced In the following, is the intention to talk about the
document status of RFC 9365 or to talk about the concept of
DNSSEC validation as being a best current practice in the general
sense?
Original:
DNSSEC validation is a Best Current Practice [RFC9364]:
Perhaps A:
"DNS Security Extensions (DNSSEC)" is a Best Current Practice
([RFC9364]) that describes DNSSEC validation:
Perhaps B:
DNSSEC (see [RFC9364]) validation is a best current practice:
[Ted] I think B is better.
-->
<ol spacing="normal">
<li>For correctness, recursive resolvers at sites using
'service.arpa.' must, in practice, transparently support DNSSEC
queries: queries for DNSSEC records and queries with the DNSSEC OK
(DO) bit set
(Section <xref target="RFC4035" section="3.2.1" sectionFormat="bare"/>
of the DNSSEC specification <xref target="RFC4035"/>).
DNSSEC validation <xref target="RFC9364"/>
is a best current practice: Although validation is not required, a
caching recursive resolver that does not validate answers that can
be validated may cache invalid data. In turn, this would prevent
validating stub resolvers from successfully validating
answers. Hence, as a practical matter, recursive resolvers at sites
using "service.arpa." should do DNSSEC validation.</li>
<li> <li>
<t>Unless configured otherwise, recursive resolvers and DNS proxies M <t>Unless configured otherwise, recursive resolvers and DNS
UST behave as described in Locally Served Zones, proxies <bcp14>MUST</bcp14> behave following
<xref target="RFC6303" section="3" sectionFormat="of"/>. That is, the rules prescribed for Iterative Resolvers in
queries for 'service.arpa.' and subdomains of Section <xref target="RFC6303" section="3" sectionFormat="bare"/>
'service.arpa.' MUST NOT be forwarded, with one important exceptio of the IETF Locally Served DNS Zones document <xref target="RFC6303"/
n: a query for a DS record with the DO bit set MUST >.
return the correct answer for that question, including correct info That is, queries for "service.arpa." and subdomains of
rmation in the authority section that proves that "service.arpa."&nbsp;<bcp14>MUST NOT</bcp14> be forwarded, with one
the record is nonexistent.</t> important exception: a query for a DS record with the DO bit set
<t>So, for example, a query for the NS record for 'service.arpa.' M <bcp14>MUST</bcp14> return the correct answer for that question,
UST NOT result in that query being forwarded to an including correct information in the authority section that proves
upstream cache nor to the authoritative DNS server for '.arpa.'. H that the record is nonexistent.</t>
owever, as necessary to provide accurate
authority information, a query for the DS record MUST result in for <!--[rfced] Is this text redundant (with two uses of necessary)? Does our
warding whatever queries are necessary; suggestion change your intended meaning?
typically, this will just be a query for the DS record, since the n
ecessary authority information will be included Original:
in the authority section of the response if the DO bit is set.</t> However, as necessary to provide accurate authority information, a
query for the DS record <bcp14>MUST</bcp14> result in forwarding whatever querie
s are
necessary; typically, ...
Perhaps:
However, to provide accurate authority information, a
query for the DS record <bcp14>MUST</bcp14> result in forwarding whatever querie
s are
necessary.
[Ted] Yup.
-->
<t>So, for example, a query for the NS record for "service.arpa."
<bcp14>MUST NOT</bcp14> result in that query being forwarded to an
upstream cache nor to the authoritative DNS server for ".arpa.".
However, to provide accurate authority information, a
query for the DS record <bcp14>MUST</bcp14> result in forwarding
whatever queries are necessary. Typically, this will just be a
query for the DS record since the necessary authority information
will be included in the authority section of the response if the
DO bit is set.</t>
</li> </li>
</ol> </ol>
</section> </section>
<section> <section>
<name>Authoritative DNS Servers</name> <name>Authoritative DNS Servers</name>
<t>No special processing of 'service.arpa.' is required for authoritative <t>No special processing of "service.arpa." is required for authoritative
DNS server implementations. It is possible that an DNS server implementations. It is possible that an
authoritative DNS server might attempt to check the authoritative serve authoritative DNS server might attempt to check the authoritative DNS s
rs for 'service.arpa.' for a delegation beneath that ervers for "service.arpa." for a delegation beneath that
name before answering authoritatively for such a delegated name. In su ch a case, because the name always has only local name before answering authoritatively for such a delegated name. In su ch a case, because the name always has only local
significance, there will be no such delegation in the 'service.arpa.' z significance, there will be no such delegation in the "service.arpa." z
one, and so the server would refuse to answer one;
authoritatively for such a zone. A server that implements this sort of therefore, the authoritative DNS server would refuse to answer
check MUST be configurable so that either it does authoritatively for such a zone. An authoritative DNS server that impl
not do this check for the 'service.arpa.' domain or it ignores the resu ements
lts of the check.</t> this sort of check <bcp14>MUST</bcp14> be configurable so that either i
t does
not do this check for the "service.arpa." domain or it ignores the resu
lts of the check.</t>
</section> </section>
<section> <section>
<!--[rfced] We are having trouble parsing this sentence. Is there text
missing?
Original:
The operator for the DNS servers authoritative for 'service.arpa.' in
the global DNS will configure any such servers as described in Section
9.
Perhaps:
The operator for the DNS servers that are authoritative for "service.arpa." in
the global DNS will configure any such servers as described in Section
9.
[Ted] Yup.
-->
<name>DNS Server Operators</name> <name>DNS Server Operators</name>
<t>DNS server operators MAY configure an authoritative server for 'servic <t>DNS server operators <bcp14>MAY</bcp14> configure an authoritative DNS
e.arpa.' for use with SRP. The operator for the server for "service.arpa." for use with SRP. The operator for the
DNS servers authoritative for 'service.arpa.' in the global DNS will co DNS servers that are authoritative for "service.arpa." in the global DN
nfigure any such servers as described in S will configure any such DNS servers as described in
<xref target="delegation"/>.</t> <xref target="delegation"/>.</t>
</section> </section>
<section> <section>
<name>DNS Registries/Registrars</name> <name>DNS Registries/Registrars</name>
<t>'service.arpa.' is a subdomain of the 'arpa' top-level domain, which i <t>"service.arpa." is a subdomain of the "arpa." top-level domain,
s operated by IANA under the authority of the which is operated by IANA under the authority of the
Internet Architecture Board according to the rules established in [RFC3 Internet Architecture Board (IAB) <xref target="RFC3172"/>.
172]. There are no other DNS registrars for There are no other DNS registrars for "arpa.".</t>
'.arpa'.</t>
</section> </section>
</section> </section>
<section anchor="delegation"> <section anchor="delegation">
<name>Delegation of 'service.arpa.'</name> <name>Delegation of "service.arpa."</name>
<t>In order to be fully functional, the owner of the 'arpa.' zone must add <t>
a delegation of 'service.arpa.' in the '.arpa.' The owner of the "arpa." zone, at the time of writing the IAB <xref targe
zone <xref target="RFC3172"/>. This delegation is to be set up as was don t="IAB-ARPA"/>,
e for 'home.arpa', as a result of the has added a delegation of "service.arpa."
specification in <xref target="RFC8375" section="7" sectionFormat="of"/>. in the "arpa." zone <xref target="RFC3172"/>,
This is currently the responsibility of the IAB following the guidance provided in
<xref target="IAB-ARPA"/></t> Section <xref target="RFC8375" section="7" sectionFormat="bare"/>
of the "home.arpa." specification <xref target="RFC8375"/>.
</t>
</section> </section>
<section> <section>
<name>IANA Considerations</name> <name>IANA Considerations</name>
<section> <section>
<name>Registration and Delegation of 'service.arpa' as a Special-Use Doma
in Name</name> <!--[rfced] We have some questions about Section 10.1 in the IANA
<t>IANA is requested to record the domain name 'service.arpa.' in the Spe Considerations:
cial-Use Domain Names registry
<xref target="SUDN"/>. IANA is requested, with the approval of IAB, to a) We see the title of the section is related to the first paragraph
implement the delegation requested in only. May we move the second paragraph to its own subsection? If so,
please let us know how you would like the text to appear using
Old/New.
Original:
10.1. Registration and Delegation of 'service.arpa' as a Special-Use
Domain Name
IANA is requested to record the domain name 'service.arpa.' in the
Special-Use Domain Names registry [SUDN]. IANA is requested, with
the approval of IAB, to implement the delegation requested in
Section 9.
IANA is further requested to add a new entry to the "Transport-
Independent Locally-Served Zones" subregistry of the "Locally-Served
DNS Zones" registry [LSDZ]. The entry will be for the domain
'service.arpa.' with the description "DNS-SD Service Registration
Protocol Special-Use Domain", and listing this document as the
reference.
[Ted] I've proposed a fix below.
b) The first paragraph of Section 10.1 mentions Section 9, which
states:
Original:
9. Delegation of 'service.arpa.'
In order to be fully functional, the owner of the 'arpa.' zone must
add a delegation of 'service.arpa.' in the '.arpa.' zone [RFC3172].
This delegation is to be set up as was done for 'home.arpa', as a
result of the specification in Section 7 of [RFC8375]. This is
currently the responsibility of the IAB [IAB-ARPA]
Should Section 9 be updated as follows since this action has been
taken? Also, please review whether this information actually belongs
in the IANA section. If so, please let us know (using old/new) how to
update.
9. Delegation of "service.arpa."
The owner of the 'arpa.' zone, at the time of writing the IAB [IAB-ARPA],
has added a delegation of 'service.arpa.' in the '.arpa.' zone
[RFC3172], following the guidance provided in Section 7 of [RFC8375].
[Ted] This looks fine.
-->
<name>Registration and Delegation of "service.arpa." as a Special-Use Dom
ain Name</name>
<t>IANA has recorded the domain name "service.arpa." in the "Special-Use
Domain Names" registry
<xref target="SUDN"/>. IANA has implemented the delegation requested in
<xref target="delegation"/>.</t> <xref target="delegation"/>.</t>
<t>IANA is further requested to add a new entry to the "Transport-Indepen </section>
dent Locally-Served Zones" subregistry of <section>
the "Locally-Served DNS Zones" registry <xref target="LSDZ"/>. The ent <name>Addition of "service.arpa." to the Locally-Served Zones Registry</n
ry will be for the domain 'service.arpa.' with the ame>
description "DNS&nbhy;SD Service Registration Protocol Special-Use Doma <t>IANA has also added a new entry to the "Transport-Independent Locally-
in", listing this document as the reference.</t> Served Zones Registry" registry of
the "Locally-Served DNS Zones" group <xref target="LSDZ"/>.
The entry is for the domain "SERVICE.ARPA." with the
description "DNS&nbhy;SD Service Registration Protocol
Special-Use Domain" and lists this document as the reference.</t>
</section> </section>
<section anchor="subdomains"> <section anchor="subdomains">
<name>Subdomains of 'service.arpa.'</name> <name>Subdomains of "service.arpa."</name>
<t>This document only makes use of the 'default.service.arpa' subdomain o
f 'service.arpa.' Other subdomains are reserved for <t>This document only makes use of the "default.service.arpa."
future use by DNS-SD or related work. The IANA is requested to create a subdomain of "service.arpa." Other subdomains are reserved for future
registry, the "service.arpa Subdomain" registry. use by DNS&nbhy;SD or related work. IANA has created the
The IETF shall have change control for this registry. New entries may b "service.arpa. Subdomain" registry <xref target="SUB"/>. The IETF has
e added either as a result of Standards Action change control for this registry. New entries may be added either as
<xref target="RFC8126" section="4.9"/> or with IESG approval <xref targ a result of Standards Action (<xref target="RFC8126"
et="RFC8126" section="4.10"/>, provided that a sectionFormat="of" section="4.9"/>) or with IESG Approval (<xref
specification exists <xref target="RFC8126" section="4.6"/>. target="RFC8126" sectionFormat="of" section="4.10"/>), provided that
</t> the values and their meanings are documented in a permanent and
readily available public specification, in sufficient detail so that
interoperability between independent implementations is possible. </t>
<t> <t>
The IANA shall group the "service.arpa Subdomain" registry with the "Lo IANA has grouped the "service.arpa.&nbsp;Subdomain" registry
cally-Served DNS Zones" registry. with the "Locally-Served DNS Zones" group.
The registry shall be a table with three columns: the subdomain name ( The registry is a table with three columns: the subdomain name (expres
expressed as a fully-qualified domain sed as a fully qualified domain
name), a brief description of how it is used, and a reference to the do cument that describes its use in detail. name), a brief description of how it is used, and a reference to the do cument that describes its use in detail.
</t> </t>
<t> <t>
This registry shall begin as the following table: This initial contents of this registry are as follows:
</t> </t>
<table> <table>
<thead> <thead>
<tr> <tr>
<th>Subdomain Name</th> <th>Subdomain Name</th>
<th>Description</th> <th>Description</th>
<th>reference</th> <th>Reference</th>
</tr> </tr>
</thead> </thead>
<tbody> <tbody>
<tr> <tr>
<td>default.service.arpa.</td> <td>default.service.arpa.</td>
<td>Default domain for SRP updates</td> <td>Default domain for SRP Updates</td>
<td>[THIS DOCUMENT]</td> <td>RFC 9665</td>
</tr> </tr>
</tbody> </tbody>
</table> </table>
</section> </section>
<section> <section>
<name>Service Name registrations</name> <name>Service Name Registrations</name>
<t>IANA is requested to add two new entries to the Service Names and Port <t>IANA has added two new entries to the
Numbers registry. The following sections "Service Name and Transport Protocol Port Number Registry"
contain tables with the fields required by <xref target="RFC6335" secti <xref target="PORT"/>. The following subsections
on="8.1.1" sectionFormat="of"/>.</t> contain tables with the fields required by
</section> Section <xref target="RFC6335" section="8.1.1" sectionFormat="bare"/>
of IANA's Procedures for Service Name allocation <xref target="RFC6335"
/>.</t>
<section> <section>
<name>'dnssd-srp' Service Name</name> <name>"dnssd-srp" Service Name</name>
<table> <table>
<thead><tr><td>Field Name</td><td>Value</td></tr></thead> <thead><tr><th>Field Name</th><th>Value</th></tr></thead>
<tbody> <tbody>
<tr><td> Service Name </td><td> dnssd-srp </td></tr> <tr><td> Service Name </td><td> dnssd-srp </td></tr>
<tr><td> Transport Protocol </td><td> TCP </td></tr> <tr><td> Transport Protocol </td><td> tcp </td></tr>
<tr><td> Assignee </td><td> IESG &lt;iesg@ietf.org&gt; </td></tr> <tr><td> Assignee </td><td> IESG &lt;iesg@ietf.org&gt; </td></tr>
<tr><td> Contact </td><td> IETF Chair &lt;chair@ietf.org &gt; </td></tr> <tr><td> Contact </td><td> IETF Chair &lt;chair@ietf.org &gt; </td></tr>
<tr><td> Description </td><td> DNS-SD Service Registration <tr><td> Description </td><td> DNS&nbhy;SD Service Discovery
</td></tr> </td></tr>
<tr><td> Reference </td><td> this document <tr><td> Reference </td><td> RFC 9665
</td></tr> </td></tr>
<tr><td> Port Number </td><td> None </td></tr> <tr><td> Port Number </td><td> None </td></tr>
<tr><td> Service Code </td><td> None </td></tr> <tr><td> Service Code </td><td> None </td></tr>
</tbody> </tbody>
</table> </table>
</section> </section>
<section> <section>
<name>'dnssd-srp-tls' Service Name</name> <name>"dnssd-srp-tls" Service Name</name>
<table> <table>
<thead><tr><td>Field Name</td><td>Value</td></tr></thead> <thead><tr><th>Field Name</th><th>Value</th></tr></thead>
<tbody> <tbody>
<tr><td> Service Name </td><td> dnssd-srp-tls </td></tr> <tr><td> Service Name </td><td> dnssd-srp-tls </td></tr>
<tr><td> Transport Protocol </td><td> TCP <tr><td> Transport Protocol </td><td> tcp
</td></tr> </td></tr>
<tr><td> Assignee </td><td> IESG <tr><td> Assignee </td><td> IESG &lt;iesg@ietf.org&gt;
</td></tr> </td></tr>
<tr><td> Contact </td><td> IETF Chair <tr><td> Contact </td><td> IETF Chair &lt;chair@ietf.org
</td></tr> &gt; </td></tr>
<tr><td> Description </td><td> DNS-SD Service Registration ( <tr><td> Description </td><td> DNS&nbhy;SD Service Discovery
TLS) </td></tr> (TLS) </td></tr>
<tr><td> Reference </td><td> this document <tr><td> Reference </td><td> RFC 9665
</td></tr> </td></tr>
<tr><td> Port Number </td><td> None </td></tr> <tr><td> Port Number </td><td> None </td></tr>
<tr><td> Service Code </td><td> None </td></tr> <tr><td> Service Code </td><td> None </td></tr>
</tbody> </tbody>
</table> </table>
</section> </section>
</section>
<section> <section>
<name>Anycast Address</name> <name>Anycast Address</name>
<t>IANA is requested to allocate an IPv6 Anycast address from the IPv6 Sp <t>IANA has allocated an IPv6 anycast address from the
ecial-Purpose Address Registry, similar to the Port "IANA IPv6 Special-Purpose Address Registry" <xref target="IPv6"/>,
Control Protocol anycast address, 2001:1::1. The value TBD is to be rep similar to the Port
laced with the actual allocation in the table that Control Protocol <xref target="RFC6887"/>
follows. The purpose of this allocation is to provide a fixed anycast a anycast address <xref target="RFC7723"/>.
ddress that can be commonly used as a destination for The purpose of this allocation is to provide a fixed anycast
SRP updates when no SRP registrar is explicitly configured. The values address that can be commonly used as a destination for
for the registry are:</t> SRP Updates when no SRP registrar is explicitly configured. The initial
values for the registry are as follows:</t>
<table> <table>
<thead> <thead>
<tr><td>Attribute</td> <td>value</td></tr> <tr><th>Attribute</th> <th>Value</th></tr>
</thead> </thead>
<tbody> <tbody>
<tr><td>Address Block</td> <td>2001:1::TBD/128</td></t <tr><td>Address Block</td> <td>2001:1::3/128</td></tr>
r> <tr><td>Name</td> <td>DNS&nbhy;SD Service Reg
<tr><td>Name</td> <td>DNS-SD Service Registra istration Protocol Anycast Address</td></tr>
tion Protocol Anycast Address</td></tr> <tr><td>RFC</td> <td>RFC 9665</td></tr>
<tr><td>RFC</td> <td>[this document]</td></t <tr><td>Allocation Date</td> <td>2024-04</td></tr>
r>
<tr><td>Allocation Date</td> <td>[date of allocation]</t
d></tr>
<tr><td>Termination Date</td> <td>N/A</td></tr> <tr><td>Termination Date</td> <td>N/A</td></tr>
<tr><td>Source</td> <td>True</td></tr> <tr><td>Source</td> <td>True</td></tr>
<tr><td>Destination</td> <td>True</td></tr> <tr><td>Destination</td> <td>True</td></tr>
<tr><td>Forwardable</td> <td>True</td></tr> <tr><td>Forwardable</td> <td>True</td></tr>
<tr><td>Global</td> <td>True</td></tr> <tr><td>Globally Reachable</td> <td>True</td></
<tr><td>Reserved-by-protocol</td> <td>False</td></tr> tr>
<tr><td>Reserved-by-Protocol</td> <td>False</td></tr>
</tbody> </tbody>
</table> </table>
</section> </section>
</section> </section>
<section>
<name>Implementation Status</name>
<t>[Note to the RFC Editor: please remove this section prior to publicatio
n.]</t>
<t>
This section records the status of known implementations of the protocol
defined by this specification at the time of
posting of this Internet-Draft, and is based on a proposal described in R
FC 7942. The description of implementations in
this section is intended to assist the IETF in its decision processes in
progressing drafts to RFCs. Please note that the
listing of any individual implementation here does not imply endorsement
by the IETF. Furthermore, no effort has been spent
to verify the information presented here that was supplied by IETF contri
butors. This is not intended as, and must not be
construed to be, a catalog of available implementations or their features
. Readers are advised to note that other
implementations may exist.
</t>
<t>
According to RFC 7942, "this will allow reviewers and working groups to a
ssign due consideration to documents that have the
benefit of running code, which may serve as evidence of valuable experime
ntation and feedback that have made the implemented
protocols more mature. It is up to the individual working groups to use
this information as they see fit".
</t>
<t>
There are two known independent implementations of SRP requestors:
</t>
<ul>
<li>SRP Client for OpenThread: https://github.com/openthread/openthread/p
ull/6038</li>
<li>mDNSResponder open source project: https://github.com/Abhayakara/mdns
responder</li>
</ul>
<t>
There are two related implementations of an SRP registrar. One acts as a
DNS Update proxy, taking an SRP Update and applying it
to the specified DNS zone using DNS update. The other acts as an Advertis
ing Proxy
<xref target="I-D.ietf-dnssd-advertising-proxy"/>. Both are included in t
he mDNSResponder open source project mentioned above.
</t>
</section>
<section>
<name>Acknowledgments</name>
<t>Thanks to <contact fullname="Toke Høiland-Jørgensen"/>, Jonathan Hui, E
sko Dijk, Kangping Dong and Abtin Keshavarzian for
their thorough technical reviews. Thanks to Kangping and Abtin as well fo
r testing the document by doing an independent
implementation. Thanks to Tamara Kemper for doing a nice developmental ed
it, Tim Wattenberg for doing an SRP requestor
proof-of-concept implementation at the Montreal Hackathon at IETF 102, an
d Tom Pusateri for reviewing during the hackathon
and afterwards. Thanks to Esko for a really thorough second last call rev
iew. Thanks also to Nathan Dyck, Gabriel
Montenegro, Kangping Dong, Martin Turon, and Michael Cowan for their deta
iled second last call reviews. Thanks to Patrik
Fältström, Dhruv Dhody, David Dong, Joey Salazar, Jean-Michel Combes, and
Joerg Ott for their respective directorate
reviews. Thanks to Paul Wouters for a <em>really</em> detailed IESG revie
w! Thanks also to the other IESG members who
provided comments or simply took the time to review the document.</t>
</section>
</middle> </middle>
<back> <back>
<displayreference target="I-D.cheshire-dnssd-roadmap" to="ROADMAP"/> <displayreference target="I-D.cheshire-dnssd-roadmap" to="ROADMAP"/>
<displayreference target="I-D.ietf-dnssd-advertising-proxy" to="AP"/> <displayreference target="I-D.ietf-snac-simple" to="SNAC-SIMPLE"/>
<!-- <displayreference target="I-D.ietf-dnssd-hybrid" to="I-D.ietf-dnssd-hyb rid"/> appears to not work in xml2rfc 2.6.2 -->
<references> <references>
<name>References</name>
<references>
<name>Normative References</name> <name>Normative References</name>
<xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml3/reference.I-
D.ietf-dnssd-update-lease.xml"/> <!-- [I-D.ietf-dnssd-update-lease] companion document RFC 9664-->
<xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC <reference anchor="RFC9664" target="https://www.rfc-editor.org/info/rfc966
.1035.xml" /> 4">
<xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC <front>
.1536.xml" /> <title>An EDNS(0) Option to Negotiate Leases on DNS Updates</title>
<xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC <author fullname="Stuart Cheshire" initials="S." surname="Cheshire">
.2119.xml" /> <organization>Apple Inc.</organization>
<xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC </author>
.2136.xml" /> <author fullname="Ted Lemon" initials="T." surname="Lemon">
<xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC <organization>Apple Inc</organization>
.2181.xml" /> </author>
<xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC <date month="October" year="2024"/>
.2539.xml" /> </front>
<xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC <seriesInfo name="RFC" value="9664"/>
.2782.xml" /> <seriesInfo name="DOI" value="10.17487/RFC9664"/>
<xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC </reference>
.2931.xml" />
<xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC <xi:include href="https://bib.ietf.org/public/rfc/bibxml/reference.RFC.103
.3172.xml"/> 5.xml" />
<xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC <xi:include href="https://bib.ietf.org/public/rfc/bibxml/reference.RFC.153
.3445.xml"/> 6.xml" />
<xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC <xi:include href="https://bib.ietf.org/public/rfc/bibxml/reference.RFC.211
.3596.xml"/> 9.xml" />
<xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC <xi:include href="https://bib.ietf.org/public/rfc/bibxml/reference.RFC.213
.4035.xml"/> 6.xml" />
<xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC <xi:include href="https://bib.ietf.org/public/rfc/bibxml/reference.RFC.218
.6303.xml"/> 1.xml" />
<xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC <xi:include href="https://bib.ietf.org/public/rfc/bibxml/reference.RFC.253
.6763.xml"/> 9.xml" />
<xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC <xi:include href="https://bib.ietf.org/public/rfc/bibxml/reference.RFC.278
.7858.xml" /> 2.xml" />
<xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC <xi:include href="https://bib.ietf.org/public/rfc/bibxml/reference.RFC.293
.8085.xml" /> 1.xml" />
<xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC <xi:include href="https://bib.ietf.org/public/rfc/bibxml/reference.RFC.317
.8126.xml"/> 2.xml"/>
<xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC <xi:include href="https://bib.ietf.org/public/rfc/bibxml/reference.RFC.359
.8174.xml"/> 6.xml"/>
<xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC <xi:include href="https://bib.ietf.org/public/rfc/bibxml/reference.RFC.403
.8375.xml"/> 4.xml"/>
<xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC <xi:include href="https://bib.ietf.org/public/rfc/bibxml/reference.RFC.403
.8624.xml"/> 5.xml"/>
<xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC <xi:include href="https://bib.ietf.org/public/rfc/bibxml/reference.RFC.630
.8765.xml" /> 3.xml"/>
<xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC <xi:include href="https://bib.ietf.org/public/rfc/bibxml/reference.RFC.676
.9364.xml" /> 3.xml"/>
<xi:include href="https://bib.ietf.org/public/rfc/bibxml/reference.RFC.785
8.xml" />
<xi:include href="https://bib.ietf.org/public/rfc/bibxml/reference.RFC.808
5.xml" />
<xi:include href="https://bib.ietf.org/public/rfc/bibxml/reference.RFC.812
6.xml"/>
<xi:include href="https://bib.ietf.org/public/rfc/bibxml/reference.RFC.817
4.xml"/>
<xi:include href="https://bib.ietf.org/public/rfc/bibxml/reference.RFC.837
5.xml"/>
<xi:include href="https://bib.ietf.org/public/rfc/bibxml/reference.RFC.862
4.xml"/>
<xi:include href="https://bib.ietf.org/public/rfc/bibxml/reference.RFC.876
5.xml" />
<xi:include href="https://bib.ietf.org/public/rfc/bibxml/reference.RFC.936
4.xml" />
</references> </references>
<references> <references>
<name>Informative References</name> <name>Informative References</name>
<xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC <xi:include href="https://bib.ietf.org/public/rfc/bibxml/reference.RFC.213
.2131.xml" /> 1.xml" />
<xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC <xi:include href="https://bib.ietf.org/public/rfc/bibxml/reference.RFC.282
.2827.xml" /> 7.xml" />
<xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC <xi:include href="https://bib.ietf.org/public/rfc/bibxml/reference.RFC.300
.3007.xml" /> 7.xml" />
<xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC <xi:include href="https://bib.ietf.org/public/rfc/bibxml/reference.RFC.392
.4861.xml" /> 7.xml" />
<xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC <xi:include href="https://bib.ietf.org/public/rfc/bibxml/reference.RFC.486
.6105.xml" /> 1.xml" />
<xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC <xi:include href="https://bib.ietf.org/public/rfc/bibxml/reference.RFC.486
.6335.xml" /> 2.xml" />
<xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC <xi:include href="https://bib.ietf.org/public/rfc/bibxml/reference.RFC.610
.6760.xml" /> 5.xml" />
<xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC <xi:include href="https://bib.ietf.org/public/rfc/bibxml/reference.RFC.633
.6761.xml" /> 5.xml" />
<xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC <xi:include href="https://bib.ietf.org/public/rfc/bibxml/reference.RFC.676
.6762.xml" /> 0.xml" />
<xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC <xi:include href="https://bib.ietf.org/public/rfc/bibxml/reference.RFC.676
.7084.xml" /> 1.xml" />
<xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC <xi:include href="https://bib.ietf.org/public/rfc/bibxml/reference.RFC.676
.7228.xml" /> 2.xml" />
<xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC <xi:include href="https://bib.ietf.org/public/rfc/bibxml/reference.RFC.688
.7413.xml" /> 7.xml" />
<xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC <xi:include href="https://bib.ietf.org/public/rfc/bibxml/reference.RFC.708
.8415.xml" /> 4.xml" />
<xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC <xi:include href="https://bib.ietf.org/public/rfc/bibxml/reference.RFC.772
.8520.xml" /> 3.xml" />
<xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC <xi:include href="https://bib.ietf.org/public/rfc/bibxml/reference.RFC.722
.8766.xml" /> 8.xml" />
<xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC <xi:include href="https://bib.ietf.org/public/rfc/bibxml/reference.RFC.741
.8945.xml" /> 3.xml" />
<xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml3/reference.I- <xi:include href="https://bib.ietf.org/public/rfc/bibxml/reference.RFC.841
D.cheshire-dnssd-roadmap.xml"/> 5.xml" />
<xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml3/reference.I- <xi:include href="https://bib.ietf.org/public/rfc/bibxml/reference.RFC.852
D.ietf-dnssd-advertising-proxy.xml"/> 0.xml" />
<xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml3/reference.I- <xi:include href="https://bib.ietf.org/public/rfc/bibxml/reference.RFC.876
D.ietf-snac-simple.xml"/> 6.xml" />
<xi:include href="https://bib.ietf.org/public/rfc/bibxml/reference.RFC.894
5.xml" />
<reference anchor="SUDN" target="https://www.iana.org/assignments/special- <!-- [I-D.cheshire-dnssd-roadmap] IESG state: Expired as of 07/15/24
use-domain-names/special-use-domain-names.xhtml">
[Ted] I don't know if Stuart is going to update this. References to drafts are a
lways problematic in this way.
I think we need to either leave it as is, or take it out. I think it's oka
y to leave it as is, but will
defer to the experts. -->
<xi:include href="https://datatracker.ietf.org/doc/bibxml3/reference.I-D.c
heshire-dnssd-roadmap.xml"/>
<!-- [I-D.ietf-snac-simple] IESG state: I-D Exists as of 07/15/24
[Ted] I would not want to see this added to the cluster. It's informative, after
all. snac-simple should head to the IESG in the next couple of
months, but I think we should just leave the reference to the draft unless
there's a strong argument not to. This is just an informative
reference anyway. -->
<xi:include href="https://datatracker.ietf.org/doc/bibxml3/reference.I-D.i
etf-snac-simple.xml"/>
<reference anchor="SUDN" target="https://www.iana.org/assignments/special-
use-domain-names">
<front> <front>
<title>Special-Use Domain Names Registry</title> <title>Special-Use Domain Names</title>
<author/> <author>
<date month="July" year="2012"/> <organization>IANA</organization>
</author>
</front> </front>
</reference> </reference>
<reference anchor="LSDZ" target="https://www.iana.org/assignments/locally- served-dns-zones/locally-served-dns-zones.xhtml"> <reference anchor="LSDZ" target="https://www.iana.org/assignments/locally- served-dns-zones">
<front> <front>
<title>Locally-Served DNS Zones Registry</title> <title>Locally-Served DNS Zones</title>
<author/> <author>
<date month="July" year="2011"/> <organization>IANA</organization>
</author>
</front>
</reference>
<reference anchor="SUB" target="https://www.iana.org/assignments/locally-s
erved-dns-zones/locally-served-dns-zones">
<front>
<title>service.arpa Subdomain</title>
<author>
<organization>IANA</organization>
</author>
</front>
</reference>
<reference anchor="PORT" target="https://www.iana.org/assignments/service-names-
port-numbers">
<front>
<title>Service Name and Transport Protocol Port Number Registry</title
>
<author>
<organization>IANA</organization>
</author>
</front>
</reference>
<reference anchor="IPv6" target="https://www.iana.org/assignments/iana-ipv
6-special-registry">
<front>
<title>IANA IPv6 Special-Purpose Address Registry</title>
<author>
<organization>IANA</organization>
</author>
</front> </front>
</reference> </reference>
<reference anchor="IAB-ARPA" target="https://www.iab.org/documents/corresp ondence-reports-documents/2017-2/iab-statement-on-the-registration-of-special-us e-names-in-the-arpa-domain/"> <reference anchor="IAB-ARPA" target="https://www.iab.org/documents/corresp ondence-reports-documents/2017-2/iab-statement-on-the-registration-of-special-us e-names-in-the-arpa-domain/">
<front> <front>
<title>Internet Architecture Board statement on the registration of sp ecial use names in the ARPA domain</title> <title>Internet Architecture Board statement on the registration of sp ecial use names in the ARPA domain</title>
<author/> <author/>
<date month="March" year="2017"/> <date month="March" year="2017"/>
</front> </front>
</reference> </reference>
<reference anchor="ZC"> <reference anchor="ZC">
<front> <front>
<title>Zero Configuration Networking: The Definitive Guide</title> <title>Zero Configuration Networking: The Definitive Guide</title>
<author initials="S." surname="Cheshire" fullname="Stuart Cheshire"/>
<author initials="D.H." surname="Steinberg" fullname="Daniel H. Steinb erg"/> <author initials="D.H." surname="Steinberg" fullname="Daniel H. Steinb erg"/>
<author initials="S." surname="Cheshire" fullname="Stuart Cheshire"/>
<date year="2005" month="December"/> <date year="2005" month="December"/>
</front> </front>
<seriesInfo name="O'Reilly Media, Inc." value=""/> <refcontent>O'Reilly Media, Inc.</refcontent>
<seriesInfo name="ISBN" value="0-596-10100-7"/> <seriesInfo name="ISBN" value="9780596101008"/>
</reference> </reference>
</references> </references>
</references>
<!--[rfced] Might this be an agreeable update to the title of
Appendix A (to avoid double -ing words in the beginning?)?
Original:
Appendix A. Testing Using Standard DNS Servers Compliant with RFC
2136
Perhaps:
Appendix A. Testing the Use of Standard DNS Servers Compliant with RFC
2136
Perhaps:
Appendix A. Testing Standard DNS Servers Compliant with RFC 2136
[Ted] Nope. See below. :)
-->
<section> <section>
<name>Testing using standard RFC2136-compliant DNS servers</name> <name>Using Standard Authoritative DNS Servers Compliant with RFC 2136 to Test SRP Requesters</name>
<t> <t>
It may be useful to set up an authoritative DNS server for testing that For testing, it may be useful to set up an
does not implement SRP. This can be done by configuring the authoritative DNS server that does not implement SRP.
server to listen on the anycast address, or advertising it in the _dnssd This can be done by configuring the
&nbhy;srp._tcp.&lt;zone&gt; SRV and authoritative DNS server to listen on the anycast address or by
_dnssd&nbhy;srp&nbhy;tls._tcp.&lt;zone&gt; record. It must be configure advertising it in the "_dnssd&nbhy;srp._tcp.&lt;zone&gt;" and
d to be authoritative for "_dnssd&nbhy;srp&nbhy;tls._tcp.&lt;zone&gt;" SRV records.
"default.service.arpa", and to accept updates from hosts on local networ It must be configured to be authoritative for
ks for names under "default.service.arpa" "default.service.arpa." and to accept updates from hosts
without authentication, since such servers will not have support for FCF on local networks for names under "default.service.arpa."
S authentication (<xref target="fcfs"/>).</t> without authentication since such authoritative DNS servers will not
have support for FCFS authentication (<xref target="fcfs"/>).</t>
<t> <t>
An authoritative DNS server configured in this way will be able to succe An authoritative DNS server configured in this way will be able to succe
ssfully accept and process SRP Updates from requestors that send SRP ssfully accept and process SRP Updates from requesters that send SRP
updates. However, no prerequisites will be applied, and this means that updates. However, no prerequisites will be applied; this means
the test server will accept internally that the test authoritative DNS server will accept internally
inconsistent SRP Updates, and will not stop two SRP Updates, sent by dif inconsistent SRP Updates and will not stop two SRP Updates sent by diffe
ferent services, that claim the same name(s), rent services that claim the same name or names
from overwriting each other.</t> from overwriting each other.</t>
<t> <t>
Since SRP Updates are signed with keys, validation of the SIG(0) algorit Since SRP Updates are signed with keys, validation of the SIG(0) algorit
hm used by the requestor can be done by manually hm used by the requester can be done by manually
installing the requestor's public key on the DNS server that will be rec installing the requester's public key on the authoritative DNS server
eiving the updates. The key can then be used to that will be receiving the updates. The key can then be used to
authenticate the SRP update, and can be used as a requirement for the up authenticate the SRP Update and can be used as a requirement for the upd
date. An example configuration for testing SRP ate. An example configuration for testing SRP
using BIND 9 is given in <xref target="bind-example"/>.</t> using BIND 9 is given in <xref target="bind-example"/>.</t>
</section> </section>
<section> <section>
<name>How to allow SRP requestors to update standard RFC2136-compliant ser vers</name> <name>How to Allow SRP Requesters to Update Standard Servers Compliant wit h RFC 2136</name>
<t> <t>
Ordinarily SRP Updates will fail when sent to an RFC 2136-compliant serv Ordinarily, CNN SRP Updates sent to an authoritative DNS server
er that does not implement SRP because the zone that implements standard DNS Update <xref target="RFC2136"/> but not SRP
being updated is "default.service.arpa", and no DNS server that is not a will fail
n SRP registrar would normally be configured to be because the zone being updated is "default.service.arpa." and because
authoritative for "default.service.arpa". Therefore, a requestor that s no authoritative DNS server that is not an SRP registrar would normally
ends an SRP Update can tell that the receiving server be configured to be authoritative for "default.service.arpa.".
does not support SRP, but does support RFC2136, because the RCODE will e Therefore, a requester that sends an SRP Update can
ither be NotZone, NotAuth or Refused, or because tell that the receiving authoritative DNS server
there is no response to the update request (when using the anycast addre does not support SRP but does support
ss)</t> standard DNS Update <xref target="RFC2136"/>
because the RCODE will either be NotZone, NotAuth, or Refused or because
there is no response to the update request (when using the anycast addre
ss).</t>
<t> <t>
In this case a requestor MAY attempt to register itself using regular RF In this case, a requester <bcp14>MAY</bcp14>
C2136 DNS updates. To do so, it must discover the attempt to register itself using
default registration zone and the DNS server designated to receive updat normal DNS updates <xref target="RFC2136"/>.
es for that zone, as described earlier, using the To do so, it must discover the
_dns&nbhy;update._udp SRV record. It can then send the update to the po default registration zone and the authoritative DNS server designated
rt and host pointed to by the SRV record, and is to receive updates for that zone, as described earlier, using the
_dns&nbhy;update._udp SRV record. It can then send the update to the po
rt and host pointed to by the SRV record, and it is
expected to use appropriate prerequisites to avoid overwriting competing records. Such updates are out of scope for SRP, expected to use appropriate prerequisites to avoid overwriting competing records. Such updates are out of scope for SRP,
and a requestor that implements SRP MUST first attempt to use SRP to reg and a requester that implements SRP <bcp14>MUST</bcp14>
ister itself, and only attempt to use RFC2136 first attempt to use SRP to register itself and
backwards compatibility if that fails. Although the owner name for the only attempt to use backwards capability with
SRV record specifies the UDP protocol for updates, normal DNS Update <xref target="RFC2136"/>
it is also possible to use TCP, and TCP SHOULD be required to prevent sp if that fails.
oofing.</t> Although the owner name of the SRV record for
DNS Update (_dns-update._udp) specifies UDP,
it is also possible to use TCP, and TCP <bcp14>SHOULD</bcp14> be require
d to prevent spoofing.</t>
</section> </section>
<section anchor="bind-example"> <section anchor="bind-example">
<name>Sample BIND9 configuration for default.service.arpa.</name> <name>Sample BIND 9 Configuration for "default.service.arpa."</name>
<figure title="Zone Configuration in named.conf"><artwork><![CDATA[
<figure title="Zone Configuration in named.conf">
<artwork align="center"><![CDATA[
zone "default.service.arpa." { zone "default.service.arpa." {
type primary; type primary;
file "/etc/bind/primary/service.db"; file "/etc/bind/primary/service.db";
allow-update { key demo.default.service.arpa.; }; allow-update { key demo.default.service.arpa.; };
}; };]]></artwork>
]]></artwork></figure> </figure>
<figure title="Example Zone file"><artwork><![CDATA[
$ORIGIN . <figure title="Example Zone File">
<artwork align="center"><![CDATA[
$TTL 57600 ; 16 hours $TTL 57600 ; 16 hours
default.service.arpa IN SOA ns3.default.service.arpa. @ IN SOA ns postmaster (
postmaster.default.service.arpa. ( 2951053287 ; serial
2951053287 ; serial 3600 ; refresh (1 hour)
3600 ; refresh (1 hour) 1800 ; retry (30 minutes)
1800 ; retry (30 minutes) 604800 ; expire (1 week)
604800 ; expire (1 week) 3600 ; minimum (1 hour)
3600 ; minimum (1 hour) )
) NS ns
NS ns3.default.service.arpa. ns AAAA 2001:db8:0:2::1
SRV 0 0 53 ns3.default.service.arpa.
$ORIGIN default.service.arpa.
$TTL 3600 ; 1 hour
_ipps._tcp PTR demo._ipps._tcp
$ORIGIN _ipps._tcp.default.service.arpa.
demo TXT "0"
SRV 0 0 9992 demo.default.service.arpa.
$ORIGIN _udp.default.service.arpa.
$TTL 3600 ; 1 hour
_dns-update PTR ns3.default.service.arpa.
$ORIGIN _tcp.default.service.arpa.
_dnssd-srp PTR ns3.default.service.arpa.
$ORIGIN default.service.arpa.
$TTL 300 ; 5 minutes
ns3 AAAA 2001:db8:0:1::1
$TTL 3600 ; 1 hour $TTL 3600 ; 1 hour
demo AAAA 2001:db8:0:2::1
KEY 0 3 13 ( ; Autoconguration bootstrap records
qweEmaaq0FAWok5//ftuQtZgiZoiFSUsm0srWREdywQU _dnssd-srp._tcp SRV 0 0 53 ns
9dpvtOhrdKWUuPT3uEFF5TZU6B4q1z1I662GdaUwqg== _dnssd-srp-tls._tcp SRV 0 0 853 ns
); alg = ECDSAP256SHA256 ; key id = 15008
AAAA ::1 ; Service Discovery Instruction
]]></artwork></figure> _ipps._tcp PTR demo._ipps._tcp
; Service Description Instruction
demo._ipps._tcp SRV 0 0 631 demohost
TXT ""
; Host Description Instruction
demohost AAAA 2001:db8:0:2::2
KEY 0 3 13 (
qweEmaaq0FAWok5//ftuQtZgiZoiFSUsm0srWREdywQU
9dpvtOhrdKWUuPT3uEFF5TZU6B4q1z1I662GdaUwqg==
); alg = ECDSAP256SHA256 ; key id = 14495
]]></artwork>
</figure>
</section>
<section numbered="false">
<name>Acknowledgments</name>
<t>Thanks to <contact fullname="Toke Høiland-Jørgensen"/>, <contact
fullname="Jonathan Hui"/>, <contact fullname="Esko Dijk"/>, <contact
fullname="Kangping Dong"/>, and <contact fullname="Abtin Keshavarzian"/>
for their thorough technical reviews. Thanks to <contact
fullname="Kangping"/> and <contact fullname="Abtin"/> as well for
testing the document by doing an independent implementation. Thanks to
<contact fullname="Tamara Kemper"/> for doing a nice developmental edit,
<contact fullname="Tim Wattenberg"/> for doing an SRP requester
proof-of-concept implementation at the Montreal Hackathon at IETF 102,
and <contact fullname="Tom Pusateri"/> for reviewing during the
hackathon and afterwards. Thanks to <contact fullname="Esko"/> for a
really thorough second Last Call review. Thanks also to <contact
fullname="Nathan Dyck"/>, <contact fullname="Gabriel Montenegro"/>,
<contact fullname="Kangping Dong"/>, <contact fullname="Martin Turon"/>,
and <contact fullname="Michael Cowan"/> for their detailed second last
call reviews. Thanks to <contact fullname="Patrik Fältström"/>, <contact
fullname="Dhruv Dhody"/>, <contact fullname="David Dong"/>, <contact
fullname="Joey Salazar"/>, <contact fullname="Jean-Michel Combes"/>, and
<contact fullname="Joerg Ott"/> for their respective directorate
reviews. Thanks to <contact fullname="Paul Wouters"/> for a
<em>really</em> detailed IESG review! Thanks also to the other IESG
members who provided comments or simply took the time to review the
document.</t>
</section> </section>
</back> </back>
</rfc>
<!-- Keep this comment at the end of the file <!-- [rfced] We had some questions about abbreviations:
Local variables:
mode: sgml a) Should "DNSSD" (in "non-DNSSD services" and "DNSSD discovery zone")
fill-column:132 be updated to "DNS-SD" (hyphen) or "dnssd" (lowercase) to match prior
sgml-omittag:t usage in the document?
sgml-shorttag:t
sgml-namecase-general:t [Ted] Yes.
sgml-general-insert-case:lower
sgml-minimize-attributes:nil b) Is the "Service" (or "Service Description") redundant here and in
sgml-always-quote-attributes:t similar cases throughout the document (as SD = Service Discovery)?
sgml-indent-step:2 That is, just examples below, more cases exist.
sgml-indent-data:t
sgml-parent-document:nil Original:
sgml-exposed-tags:nil DNS-SD Service registration uses public keys and SIG(0) to allow
sgml-local-catalogs:nil services to defend their registrations.
sgml-local-ecat-files:nil
End: Original:
--> Although in principle DNS-SD Service Description records can
include other record types with the same Service Instance Name, in
practice they rarely do.
[Ted] No. Do not unpack the acronym! :)
c) For "TSIG", would you like us to expand to "transaction signature"
upon first usage to match RFC 8945?
Original:
The format of the KEY resource record in the SRP Update is defined in
[RFC3445]. Because the KEY RR used in TSIG is not a zone-signing
key, the flags field in the KEY RR <bcp14>MUST</bcp14> be all zeroes.
[Ted] Yikes! This should be SIG(0), not TSIG! That's the name of the protocol, a
nd we've given reference to the correct RFC, so unpacking it would just be confu
sing.
d) Throughout the document, "SRP Update" is used, and there is only
one instance of "SRV update". We wanted to make sure that "SRV" was
indeed intended and not "SRP".
Original:
* If there is one "Add to an RRset" SRV update, there <bcp14>MUST</bcp14> be
at
least one "Add to an RRset" TXT update.
[Ted] This is actually an update to an SRV RR, not an SRP Update, but I agree th
at it's confusing and have tweaked the text to address the concern.
e) We have updated to use the abbreviation CNN for Constrained-Node
Network (to match its use in RFC 7228). Please review and let us know
any objections. Further, please review uses of "constrained network"
and let us know if any of these should be updated to CNN as well.
[Ted] I'm vaguely on the fence about this, but I think consistency is
good, and the term is used enough that using the abbreviation pays off.
-->
<!-- [rfced] We had some questions regarding capitalization of certain terms:
a) We see instances of "Anycast" (capitalized) and "anycast"
(lowercase) throughout the document, but we are unsure if certain
instances are part of proper names while other instances are more
generic. Please let us know if these need to be made more consistent
or if they are accurate as they currently are. We've listed a few
instances below.
Anycast vs. anycast:
IPv6 Anycast address
Port Control Protocol anycast address
fixed anycast address
anycast address
[Ted] Should be lowercase, I've fixed in the text.
b) We see the following similar terms. Please review and let us know
if/how to make these terms consistent.
service instance name
Service Instance Name
"Service Instance Name"
[Ted] The above are all the same thing
service instance
[Ted] This is the data structure (a set of RRs) that the Service Instance Name p
oints to. It's not currently capitalized, and that's probably okay.
Service Name
[Ted] The Service Name has zero or more PTR RRs each of which has a Service Inst
ance Name as its target. IOW the Service Name is not the same as the Service Ins
tance Name.
c) We see the following similar terms. Please let us know how to
update for consistency.
BIND 9 vs. BIND9
[Ted] Fixed
d) We have updated the quoted terms that correspond to Sections 2.5.1
- 2.5.4 of RFC 2136 to appear consistently in double quotes and with
capitalization that matches those section titles. Please let us know
any objections.
[Ted] This seems fine.
We further wondered if the following update should be made:
Original:
The target of the SRV RR Add...
Perhaps:
The "Add To An RRset" SRV update
[Ted] That's indeed confusing, and I've updated the text to make it more clear.
Please review other terms similar to these titles if they exist and
let us know if any further changes should be made.
[Ted] I'm not sure what you're asking here, unfortunately.
e) The NoError status names are in all caps in Section 2.2 of RFC
2136. Should the following updates be made to match?
ServFail to SERVFAIL
Refused to REFUSED
YXDomain to YXDOMAIN
[Ted] I don't think this is necessary, and would rather we didn't.
f) Regarding the terms “Service Description”, Service Discovery, and
“Host Description”.
- We see both Instruction and instruction when following these terms.
If/How may we make this uniform?
- Should “instruction” or the like should be inserted after instances
of these terms? Sometimes they are followed by "record" or "update",
if they appear without a label, might this be confusing to the reader?
Example:
The KEY record in Service Description updates <bcp14>MAY</bcp14> be omitted f
or
brevity; if it is omitted, the SRP registrar <bcp14>MUST</bcp14> behave as if
the
same KEY record that is given for the Host Description is also given
for each Service Description for which no KEY record is provided.
[Ted] Please just leave this text alone. The working group reviewed it and belie
ves it is correct. Making editorial changes for consistency here might in some c
ases have merit, but our cache is blown on this work and I do not want to open t
his can of worms.
g) Please review the following similar terms and let us know if/how
they should be made uniform with regard to quotes and ending with a
period (note that this term would have IANA implications):
"default.service.arpa"
"default.service.arpa."
host.default.service.arpa
host-1.default.service.arpa
host-2.default.service.arpa
host-31773.default.service.arpa. (at end of sentence)
".service.arpa."
"service.arpa"
"service.arpa."
Further note that we have updated from single to double quotes around
terms that were quoted in the original consistently. Please review
and let us know if further updates are necessary.
[Ted] I've reviewed and don't see any issues.
h) Please review the following for the use of quotes and consistent
use of SRV record. Please let us know if/how to update.
"_dnssd-srp._tcp.<zone>" SRV record vs. _dnssd-srp._tcp.<zone> SRV
"_dnssd-srp-tls._tcp.<zone>" SRV record vs. _dnssd-srp-tls._tcp.<zone> record
_dns-update._udp SRV
[Ted] I've made changes for consistency here. Any inconsistency you still see, e
.g. with the zone file, is intentional.
-->
<!-- [rfced] Please review each artwork element in Appendix C in case
they should be tagged as sourcecode or another element.
[Ted] These are configuration files. Spacing should not be modified. I don't thi
nk there's a need to do anything fancy here. The HTML is currently correct.
-->
<!-- [rfced] Please review the "Inclusive Language" portion of the
online Style Guide
<https://www.rfc-editor.org/styleguide/part2/#inclusive_language>
and let us know if any changes are needed.
Note that our script did not flag any words in particular, but this
should still be reviewed as a best practice.
-->
</rfc>
 End of changes. 250 change blocks. 
1343 lines changed or deleted 2041 lines changed or added

This html diff was produced by rfcdiff 1.48.