<?xml version="1.0" encoding="UTF-8"?> <!DOCTYPE rfcSYSTEM "rfc2629-xhtml.ent">[ <!ENTITY nbsp " "> <!ENTITY zwsp "​"> <!ENTITY nbhy "‑"> <!ENTITY wj "⁠"> ]> <rfc category="std" submissionType="IETF" docName="draft-ietf-dnssd-srp-25" number="9665" ipr="trust200902" xmlns:xi="http://www.w3.org/2001/XInclude" obsoletes="" updates="" version="3"scripts="Common,Latin" sortRefs="false"sortRefs="true" consensus="true" symRefs="true" tocDepth="4" tocInclude="true" xml:lang="en"> <front> <titleabbrev='Serviceabbrev="Service RegistrationProtocol'>ServiceProtocol">Service Registration Protocol for DNS-Based Service Discovery</title> <seriesInfo name="RFC" value="9665"/> <authorinitials="T"initials="T." surname="Lemon" fullname="Ted Lemon"> <organization>Apple Inc.</organization> <address> <postal> <street>One Apple Park Way</street> <city>Cupertino</city><region>California</region><region>CA</region> <code>95014</code><country>USA</country><country>United States of America</country> </postal> <email>mellon@fugue.com</email> </address> </author> <authorinitials='S' surname='Cheshire' fullname='Stuart Cheshire'>initials="S." surname="Cheshire" fullname="Stuart Cheshire"> <organization>Apple Inc.</organization> <address> <postal> <street>One Apple Park Way</street> <city>Cupertino</city><region>California</region><region>CA</region> <code>95014</code><country>USA</country><country>United States of America</country> </postal> <phone>+1 408 974 3207</phone> <email>cheshire@apple.com</email> </address> </author><date>March 4, 2024</date> <area>Internet</area> <workgroup>Internet Engineering Task Force</workgroup><date month="October" year="2024"/> <area>INT</area> <workgroup>dnssd</workgroup> <keyword>Multicast DNS</keyword> <keyword>DNS-Based Service Discovery</keyword> <keyword>DNS Update</keyword> <keyword>SIG(0)</keyword> <abstract><t> The<t>The Service Registration Protocol (SRP) forDNS-BasedDNS-based Service Discovery (DNS-SD) uses the standard DNS Update mechanism to enableDNS-Based Service DiscoveryDNS-SD using only unicast packets. This makes it possible to deployDNS Service DiscoveryDNS-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&nbhy;Fi)(Wi-Fi) and IEEE 802.15.4 networks.DNS&nbhy;SDDNS-SD Service registration uses public keys and SIG(0) to allow services to defend their registrations. </t> </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/browse/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> <middle> <section> <name>Introduction</name> <t> DNS-SD (see <xreftarget="RFC6763">DNS-Based Service Discovery</xref>target="RFC6763"></xref>) is a component of Zero Configuration Networking (see <xreftarget="RFC6760"/>target="RFC6760"/>, <xreftarget="ZC"/>target="ZC"/>, and <xreftarget="I-D.cheshire-dnssd-roadmap"/>.</t>target="I-D.cheshire-dnssd-roadmap"/>).</t> <t> This document describes an enhancement to<xref target="RFC6763">DNS-Based Service Discovery</xref> (DNS&nbhy;SD)DNS-SD that allows servers to register the services they offer using the DNS protocol rather than using Multicast DNS (mDNS) (see <xreftarget="RFC6762">Multicast DNS</xref> (mDNS).target="RFC6762"></xref>). There is already a large installed base of DNS&nbhy;SD clients that can discover services using the DNS protocol(e.g.(e.g., Android, Windows, Linux, Apple).</t> <t> This document is intended for three audiences: implementors of software that provides services that should be advertised using DNS&nbhy;SD, implementors of DNS servers that will be used in contexts where DNS&nbhy;SD registration is needed, and administrators of networks where DNS&nbhy;SDserviceis required. The document is expected to provide sufficient information to allow interoperable implementation of the 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&nbhy;SD)(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. --> DNS&nbhy;SD allows services to advertise the fact that they provide service and to provide the information required to access that service. DNS&nbhy;SD clients can then discover the set of services of a particular type that are available. They can then select a service from among those that are available and obtain the information required to use it. AlthoughDNS Service Discovery (DNS-SD)DNS-SD using the DNS protocol (as opposed to mDNS) can be more efficient and versatile, it is not common inpractice,practice because of the difficulties associated with updating authoritative DNS services with service information.</t> <t>ExistingThe existing practice for updating DNS zones istoeither to manually enter newdata,data orelseto use a DNS Update (see <xreftarget="RFC2136"/>. Unfortunatelytarget="RFC2136"/>). Unfortunately, a DNS Update requireseither thateither:</t> <ul> <li>that the authoritative DNS server automatically trustupdates, or else thatupdates or</li> <li>that the DNS Update requestor have some kind of shared secret or public key that is known to the DNS server and can be used to authenticate theupdate. Furthermore,update.</li></ul> <t>Furthermore, the DNS Update can be a fairly chatty process, requiring multipleround tripsroundtrips with different conditional predicates to complete the update process.</t> <t> The Service Registration Protocol (SRP) adds a set of default heuristics for processing DNS updates that eliminates the need forDNS update conditional predicates: instead,DNS-update-conditional predicates. Instead, the SRP registrar (a DNS server that supports SRP updates) has a set of default predicates that are applied to theupdate,update; and the update either succeedsentirely,entirely or fails in a way that allows the requestor to know what went wrong and construct a new update.</t> <t> SRP also adds a feature calledFirst-Come, First-Served (FCFS) Naming,"First Come, First Served Naming" (or "FCFS Naming"), which allows the requestorto claimto:</t> <ul><li>claim a name that is not yet in use,and, usingand</li> <li>using SIG(0)<xref target="RFC2931"/>, to(<xref target="RFC2931"/>), authenticate both the initial claim and subsequentupdates. Thisupdates.</li></ul> <t>This prevents name conflicts, since a second SRP requestor attempting to claim the same name will not possess the SIG(0) key used by the first requestor to claimit, andit: so its claim will berejectedrejected, and the second requestor will have to choose a new name.</t> <t> 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 important implications for what we can and can't do with data the client sends us. 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> <t> The reason for this is precisely that we have not established trust.SoSo, we can only publish information that we feel safe in publishing even though we do not have any basis for trusting the requestor. We reason that mDNS<xref target="RFC6762"/>(<xref target="RFC6762"/>) allows arbitrary hosts on a single IP link to advertise services<xref target="RFC6763"/>,(<xref target="RFC6763"/>), relying on whatever service is advertised to provide authentication as a part of its protocol rather than in the service advertisement.</t> <t> 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 to impose similar constraints.Because of thisTherefore, you will see in <xref target="add_validation"/> that a very restricted set of records 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> <t> 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 example, an SOA record, an MX record, or a CNAMErecord, and sorecord; therefore, these are forbidden. A future protocol specification might include analyses for otherrecords,records and extend the set of records that can be registered here. Or it might require establishment of trust, and add an authorization model to the authentication model we now have. But this is work for a future document.</t> <t> Finally, SRP adds the concept of a'lease,'"lease", similar to leases inDynamic Host Configuration Protocol <xref target="RFC8415"/>.DHCP (<xref target="RFC8415"/>). The SRP registration itself has a leasewhichthat may be on the order of an hour; if the requestor does not renew the lease before it has elapsed, the registration is removed. The claim on the name can have a longerlease,lease so that another requestor cannot claim the name, even though the registration has expired.</t> <t> TheService Registration ProtocolSRP forDNS&nbhy;SD (SRP),DNS-SD specified in thisdocument,document provides a reasonably secure mechanism for publishing this information. Once published, these services can be readily discovered byDNS&nbhy;SDDNS-SD clients using standard DNS lookups.</t> <t> TheDNS&nbhy;SDDNS-SD specification(<xref(see <xref target="RFC6763" section="10"sectionFormat="comma"/>, “Populating the DNS with Information”),sectionFormat="of"/> briefly discusses ways that servers can publish their information in the DNS namespace. In the case of mDNS, it allows servers to publish their information on the local link, using names in the ".local" namespace, which makes their services directly discoverable by peers attached to that same local link.</t> <t>RFC6763RFC 6763 also allows clients to discover services using<xref target="RFC1035">thethe DNSprotocol</xref>.protocol (see <xref target="RFC1035"></xref>). This can be done by having a system administrator manually configure service information in theDNS, butDNS; however, manually populating DNS authoritative server databases is costly and potentiallyerror-prone,error-prone and requires a knowledgeable network administrator. Consequently, although allDNS&nbhy;SDDNS-SD client implementations of which we are aware supportDNS&nbhy;SDDNS-SD using DNS queries, inpracticepractice, it is used much less frequently than mDNS.</t> <t> The Discovery Proxy (see <xreftarget="RFC8766">Discovery Proxy</xref>target="RFC8766"></xref>) provides one way to automatically populate the DNSnamespace,namespace but is only appropriate on networks where services are easily advertised using mDNS.ThisThe present document describes a solution more suitable for networks where multicast isinefficient,inefficient or where sleepy devices arecommon,common by supporting both the offering ofservices,services and the discovery ofservices,services using unicast.</t> </section> <section> <name>Conventions and Terminology Used in This Document</name> <t> The key words"MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", "SHOULD", "SHOULD NOT", "RECOMMENDED", "NOT RECOMMENDED", "MAY","<bcp14>MUST</bcp14>", "<bcp14>MUST NOT</bcp14>", "<bcp14>REQUIRED</bcp14>", "<bcp14>SHALL</bcp14>", "<bcp14>SHALL NOT</bcp14>", "<bcp14>SHOULD</bcp14>", "<bcp14>SHOULD NOT</bcp14>", "<bcp14>RECOMMENDED</bcp14>", "<bcp14>NOT RECOMMENDED</bcp14>", "<bcp14>MAY</bcp14>", and"OPTIONAL""<bcp14>OPTIONAL</bcp14>" in this document are to be interpreted as described inBCP 14BCP 14 <xref target="RFC2119"/> <xref target="RFC8174"/> when, and only when, they appear in all capitals, as shown here. </t> </section> <section> <name>Service Registration Protocol</name> <t> Services that implement SRP use DNS Update (see <xref target="RFC2136"/> and <xreftarget="RFC3007"/>target="RFC3007"/>) to publish service information in the DNS. Two variantsexist,exist: one for full-featuredhosts,hosts and one for devices designed for"Constrained-Node Networks" <xref target="RFC7228"/>.Constrained-Node Networks (CNNs) (<xref target="RFC7228"/>). An SRP registrar is most likely an authoritative DNSserver,server orelseis updating an authoritative DNS server. There is no requirement that the server that is receiving SRP updates be the same server that is answering queries that return records that have been registered.</t> <section> <name>Protocol Variants</name> <section><name>Full-featured<name>Full-Featured Hosts</name> <t> Full-featured hosts either are configured manually with a registrationdomain,domain or discover the default registration domain as described in <xref target="RFC6763" section="11" sectionFormat="of"/>. If this process does not produce a default registration domain, theService Registration protocolSRP 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> Manual configuration of the registration domain can be doneeither byeither:</t> <ul><li>by querying the list of available registration domains ("r._dns&nbhy;sd._udp") and allowing the user to select one from theUI, or byUI or</li> <li>by any other means appropriate to the particular use case beingaddressed. Full-featuredaddressed.</li></ul> <t>Full-featured devices construct the names of the SRV, TXT, and PTR records describing theirservice(s)service or services as subdomains of the chosen service registration domain. For thesenamesnames, they then discover the zone apex of the closest enclosing DNS zone using SOA queries (see <xref target="RFC8765"section="6.1"/>.section="6.1"/>). Having discovered the enclosing DNS zone, they query for the "_dnssd&nbhy;srp._tcp.<zone>" SRV record to discover the server to which they can send SRP updates. Hosts that support SRP Updates using TLS use the "_dnssd&nbhy;srp&nbhy;tls._tcp.<zone>" SRV record instead.</t> <t> Examples of full-featured hosts include devices such as home computers, laptops, powered peripherals with network connectionssuch(such as printers, home routers, and even battery-operated devices such as mobile phones that have long batterylives.lives). </t> </section> <section> <name>Constrained Hosts</name> <t> For devices designed forConstrained-Node Networks <xref target="RFC7228"/>CNNs (<xref target="RFC7228"/>), some simplifications are available. Instead of being configured with (or discovering) the service registration domain, the special-use domain name (see <xref target="RFC6761"/>) "default.service.arpa" is used. The details of how SRPregistrar(s)registrars are discovered will be specific to the constrainednetwork, and thereforenetwork; therefore, we do not suggest a specific mechanism here.</t> <t> SRP requestors on constrained networks are expected toreceivereceive, from thenetworknetwork, a list of SRP registrars with which to register. It is the responsibility of aConstrained-Node NetworkCNN supporting SRP to provide one or more registrar addresses. It is the responsibility of the registrar supporting aConstrained-Node NetworkCNN to handle the updates appropriately. In some network environments, updates may be accepted directly into a local "default.service.arpa" zone, which has only local visibility. In other network environments, updates for names ending in "default.service.arpa" may be rewritten by the registrar to names with broader visibility.</t> </section> <section> <name>Why two variants?</name> <t> The reason for these different variants is that low-power devices that typically useConstrained-Node NetworksCNNs may have very limited battery storage. The series of DNS lookups required to 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 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 thenetwork, and sonetwork; thus, this can be relied upon to provide nodes with the information they need.</t> <t> Networks that are not constrainednetworkscan have more complicated topologies at the IP layer. Nodes connected to such networks can be assumed to be able to do DNS-SD service registration domain discovery. Such networks are 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 on the local network in order to attack devices on the local network. To prevent such spoofing, TCP is required for such networks. </t> </section> </section> <section> <name>Protocol Details</name> <t> We will discuss several parts to thisprocess: howprocess:</t> <ul> <li>how to know what topublish, howpublish (see <xref target="what"/>),</li> <li>how to know where to publish it (under whatname), howname) (see <xref target="where"/>),</li> <li>how to publishit, and howit (see <xref target="how"/>),</li> <li>how to secure itspublication. Inpublication (see <xreftarget="maintenance"/>, we specify howtarget="how-to-secure"/>), and</li> <li>how to maintain the information oncepublished.</t> <section>published (see <xref target="maintenance"/>).</li></ul> <section anchor="what"> <name>What topublish</name>Publish</name> <t> SRP Updates are sent by SRP requestors to SRP registrars. Three types of instructions appear in an SRP update: Service Discovery instructions, Service Description instructions, and Host Description instructions. These instructions are made up of DNS UpdateRRsResource Records (RRs) that are either adds or deletes. The types of records that are added,updatedupdated, and removed in each 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 meet the constraints described in that section. The following is a brief overview of what is included in a typical SRP Update: </t> <ulspacing="compact">spacing="normal"> <li> PTRResource Record (RR)RR for services, which map from a generic service type (or subtype) name to a specific Service InstanceName.</li>Name (<xref target="RFC6763" section="4.1" sectionFormat="of"/>).</li> <li> For any Service InstanceName (<xref target="RFC6763" section="4.1" sectionFormat="comma"/>),Name, an SRV RR, one or more TXT RRs, and a KEY RR.AlthoughAlthough, inprincipleprinciple, DNS-SD Service Description records can include other record types with the same Service Instance Name, inpracticepractice, they rarely do. SRP does not permit other record types. The KEY RR is used to support FCFSnaming,naming and has no specific meaning for DNS-SD lookups. SRV records for all services described in an SRP update point to the same hostname.</li> <li> There is never more than one hostname in a single SRP update. 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 requestor may be required to suppress some addresses that would not be usable by hosts discovering the service through the SRP registrar. The exact address record suppression behavior required may vary for different types of SRP requestors. An example of such advice can be found in <xref target="RFC8766" section="5.5.2" sectionFormat="of"/>. </li> </ul> <t> <xref target="RFC6763"/> describes the details of what each of these types ofRRRRs mean, with the exception of the KEY RR, which is defined in <xref target="RFC2539"/>. These RFCs should be considered the definitivesourcesources for information about what to publish; the reason for summarizing this here is to provide the reader with enough information about what will be published that the service registration process can be understood at a high level without first learning the full details ofDNS&nbhy;SD.DNS-SD. Also, the "Service Instance Name" is an important aspect of FCFS naming, which we describe later on in this document.</t> </section><section><section anchor="where"> <name>Where topublish it</name>Publish It</name> <t> Multicast DNS (mDNS) uses a singlenamespace, ".local", whichnamespace that is valid on the locallink.link called ".local". This convenience is not available forDNS&nbhy;SDDNS-SD using the DNS protocol: services must exist in some specific DNS namespace that is chosen either by the networkoperator,operator or automatically.</t> <t> As described above, full-featured devices are responsible for knowing the domain in which to register their services. Such devicesMAY<bcp14>MAY</bcp14> optionally support configuration of a registration domain by the operator of the device. However, such devicesMUST<bcp14>MUST</bcp14> support registration domain discovery as described in <xref target="RFC6763" section="11"sectionFormat="of"/>, "Discovery of Browsing and Registration Domains".sectionFormat="of"/>. </t> <t> Devices made forConstrained-Node NetworksCNNs register in thespecial usespecial-use domain name<xref target="RFC6761"/> "default.service.arpa",(<xref target="RFC6761"/>) "default.service.arpa" and let the SRP registrar handle rewriting that to a different domain if necessary.</t> </section><section><section anchor="how"> <name>How topublish it</name>Publish It</name> <t> It is possible to issue a DNS Update that does several things atonce; this meansonce: meaning that it's possible to do all the work of adding a PTRresource recordRR to the PTR RRset on the ServiceName,Name and creating or updating the Service Instance Name and HostDescription,Description in a single transaction.</t> <t> 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> <t> Updates done according to this specification are somewhat different than regular DNS Updates as defined in <xreftarget="RFC2136"/>. The <xreftarget="RFC2136"/> where the update processcancould involve many updateattempts: youattempts. You might first attempt to add a name if it doesn't exist; if that fails, then in a second message you might update the name if it does exist but matches certain preconditions. Because the registration protocol described in this document uses a single transaction, some of this adaptability is lost.</t> <t> In order to allow updates to happen in a single transaction, SRP Updates do not include update prerequisites. The requirements specified in <xref target="server_behavior"/> are implicit in the processing of SRPUpdates, and soUpdates; thus, there is no need for the SRP requestor to put in any explicit prerequisites.</t> <section> <name>How theDNS&nbhy;SDDNS-SD Service Registrationprocess differsProcess Differs from the DNS Updateas specifiedSpecified inRFC2136</name>RFC 2136</name> <t>DNS&nbhy;SDDNS-SD Service Registration is based on the standardRFC2136DNSUpdate,Update specified in <xref target="RFC2136"/>, with some differences:</t> <ulspacing="compact">spacing="normal"> <li> It implementsfirst-come first-servedFCFS name allocation, protected using SIG(0)<xref target="RFC2931"/>.</li>(<xref target="RFC2931"/>).</li> <li> It enforces policy about what updates are allowed.</li> <li> It optionally performs rewriting of "default.service.arpa" to some other domain.</li> <li> It optionally performs automatic population of the address-to-name reverse mapping domains.</li> <li> An SRP registrar is not required to implement general DNS Update prerequisite processing.</li> <li> Constrained-Node SRP requestors are allowed to send updates to the generic domain"default.service.arpa."</li>"default.service.arpa.".</li> </ul> </section> <section> <name>Retransmission Strategy</name> <t>The DNS protocol, including DNS updates, can operate over UDP or TCP. When using UDP, reliable transmission must be guaranteed by retransmitting if a DNS UDP message is not acknowledged in a reasonable interval. <xref target="RFC1035" section="4.2.1" sectionFormat="of"/> provides some guidance on this topic, as does <xref target="RFC1536" section="1" sectionFormat="of"/>. <xref target="RFC8085" section="3.1.3" sectionFormat="of"/> also provides useful guidance that is particularly relevant to DNS.</t> </section> <section> <name>Successive Updates</name><t>Service Registration Protocol<t>SRP does not require that every update contain the same information. When an SRP requestor needs to send more than one SRP update to the SRP registrar, itMUST<bcp14>MUST</bcp14> send these sequentially: until an earlier update has been successfully acknowledged, the requestorMUST NOT<bcp14>MUST NOT</bcp14> begin sending a subsequent update.</t> </section> </section> <section anchor="how-to-secure"> <name>How tosecure it</name>Secure It</name> <t> A DNSupdateupdate, as described in <xreftarget="RFC2136"/>target="RFC2136"/>, is secured usingSecret Key Transaction Signatures, <xref target="RFC8945"/>, whichsecret key transaction signatures (<xref target="RFC8945"/>) that uses a secret key shared between the DNS Update requestor (which issues the update) and the server (which authenticates it). This model does not work for automatic service registration.</t> <t> The goal of securing theDNS&nbhy;SDDNS-SD Registration Protocol is to provide the best possible security given the constraint that service registration has to be automatic. It is possible to layer more operational security on top of what we describe here, but FCFS naming is already an improvement over the security of mDNS.</t> <section anchor="fcfs"><name>First-Come First-Served<name>FCFS Naming</name> <t>First-Come First-Serve<!--[rfced] To what does "that" refer in this sentence? Original: 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. 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 them. 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. --> FCFS naming provides a limited degree ofsecurity: asecurity. A server that registers its service usingDNS&nbhy;SDthe DNS-SD RegistrationprotocolProtocol is given ownership of a name for an extended period of time based on a lease specific to the key used to authenticate the DNS Update, which may be longer than the lease associated with the registered records. 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. If the server fails to renew its service registration before the KEY lease(<xref target="I-D.ietf-dnssd-update-lease"(see <xref target="RFC9664" section="4"/>) expires, its name is no longer protected. FCFS naming is used to protect both the Service Description and the Host Description.</t> </section> </section> <section> <name>SRP Requestor Behavior</name> <section> <name>Public/Privatekey pair generationKey Pair Generation andstorage</name>Storage</name> <t> The requestor generates a public/private key pair(See(see <xref target="rsa"/>). This key pairMUST<bcp14>MUST</bcp14> be stored in stable storage; if there is no writable stable storage on the SRP requestor, the SRP requestorMUST<bcp14>MUST</bcp14> bepre-configuredpreconfigured with a public/private key pair in read-only storage that can be used. This key pairMUST<bcp14>MUST</bcp14> be unique to the device. A device with rewritable storageSHOULD<bcp14>SHOULD</bcp14> retain this key indefinitely. When the device changes ownership, it may be appropriate for the former owner to erase the old key pair, which would then require the new owner to install a new one. Therefore, the SRP requestor on the deviceSHOULD<bcp14>SHOULD</bcp14> provide a mechanism to erase thekey, for examplekey (for example, as the result of a "factoryreset,"reset") and to generate a new key.</t> <t> The policy described here for managing keys assumes that the keys are only used for SRP. If a key that is used for SRP is also used for other purposes, the policy described here is likely to be insufficient. The policy stated here isNOT RECOMMENDED<bcp14>NOT RECOMMENDED</bcp14> in such a situation: a policy appropriate to the full set of uses for the key must be chosen. Specifying such a policy is out of scope for this document.</t> <t> When sending DNS updates, the requestor includes a KEY record containing the public portion of the key in each Host Description Instruction and each Service Description Instruction. Each KEY recordMUST<bcp14>MUST</bcp14> contain the same public key. The update is signed using SIG(0), using the private key that corresponds to the public key in the KEY record. The lifetimes of the records in the update is set using theEDNS(0)Extension Mechanisms for DNS (EDNS(0)) Update Lease option (see <xreftarget="I-D.ietf-dnssd-update-lease"/>.</t>target="RFC9664"/>).</t> <t> The format of the KEY resource record in the SRP Update is defined in <xref target="RFC3445"/>. Because the KEY RR used in TSIG is not a zone-signing key, the flags field in the KEY RRMUST<bcp14>MUST</bcp14> be all zeroes.</t> <t> The KEY record in Service Description updatesMAY<bcp14>MAY</bcp14> be omitted for brevity; if it is omitted, the SRP registrarMUST<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. Omitted KEY records are not used when computing the SIG(0) signature.</t> </section> <section> <name>Name Conflict Handling</name> <t>BothAdds for both Host DescriptionRR addsRRs and Service DescriptionRR addsRRs can have names that result in name conflicts. Service Discovery record adds 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 name, the registrar will respond to the SRP Update with a YXDomain RCODE (<xref target="RFC2136" section="2.2" sectionFormat="of"/>). In this case, the requestorMUST<bcp14>MUST</bcp14> choose a new name or give up.</t> <t> There is no specific requirement for how this isdone; typically,done. Typically, however, the requestor will append a number to the preferred name. This number could be sequentiallyincreasing,increasing or could be chosen randomly. One existing implementation attempts several sequential numbers before choosing randomly.So forFor instance, it might try host.default.service.arpa, then host-1.default.service.arpa, then host-2.default.service.arpa, then host-31773.default.service.arpa.</t> </section> <section> <name>Record Lifetimes</name> <t> The lifetime of the<xref target="RFC6763">DNS&nbhy;SDDNS-SD PTR, SRV, A,AAAAAAAA, and TXTrecords</xref>records (see <xref target="RFC6763"></xref>) uses the LEASE field of the Update Leaseoption,option and is typically set to two hours.This means thatThus, if a device is disconnected from the network, it does not appear in the user interfaces of devices looking for services of that type for too long.</t> <t> The lifetime of the KEY records is set using the KEY-LEASE field of the Update LeaseOption,Option andSHOULD<bcp14>SHOULD</bcp14> be set to a much longer time, typically 14 days. The resultof this isbeing that even though a device may be temporarilyunplugged,unplugged -- disappearing from the network for a fewdays,days -- it makes a claim on its name that lasts much longer.</t> <t>This means thatTherefore, 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 moment it disappears from the network. In the event that a device is unplugged from the network and permanently discarded, then its name is eventually cleaned up and made available forre-use.</t>reuse.</t> </section> <section> <name>Compression in SRVrecords</name>Records</name> <t> Although <xref target="RFC2782"/> requires that the target name in the SRV record not be compressed, an SRP requestorMAY<bcp14>MAY</bcp14> compress the target in the SRV record. The motivation for <em>not</em> compressing in <xref target="RFC2782"/> is notstated,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 ofSRP: anSRP. An SRP registrar needs to understand SRV records in order to validate the SRP Update. Compression of the target can save space in the SRP Update, so we want clients to be able to assume that the registrar will handle this. Therefore, SRP registrarsMUST<bcp14>MUST</bcp14> support compression 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<xref target="RFC2782"/>:[RFC2782]: DNS servers still MUST NOT compress SRV record targets. --> Note that this does not update <xref target="RFC2782"/>: DNS servers still <bcp14>MUST NOT</bcp14> compress SRV record targets. The requirement to accept compressed SRV records in updates only applies to SRP registrars, and SRP registrars that are also DNS servers stillMUST NOT<bcp14>MUST NOT</bcp14> compress SRV record targets in DNS responses. We note also that <xref target="RFC6762"/>recommentsrecommends that SRV records be compressed in mDNS messages, so <xref target="RFC2782"/> does not apply to mDNS messages.</t> <t> In addition, we note that an implementor of an SRP requestor might update existing code that creates SRV records or compresses DNS messages so that it compresses the target of an SRV record. Care must be taken if such code is used both in requestors and in DNS servers that the code only compresses in the case where a requestor is generating an SRP update.</t> </section> <section anchor="remove"> <name>Removingpublished services</name>Published Services</name> <section anchor="zero-lease"> <name>Removingall published services</name>All Published Services</name> <t> To remove all the services registered to a particular host, the SRP requestor transmits an SRP update for that host with an Update Lease option that has a LEASE value of zero. If the registration is to be permanently removed, KEY-LEASESHOULD<bcp14>SHOULD</bcp14> also be zero. Otherwise, itSHOULD<bcp14>SHOULD</bcp14> be set to the same value it had previously; this holds the name in reserve for when the SRP requestor is once again able to provide the service.</t> <t> SRP requestors are normally expected to remove all service instances when removing a host. However, in somecasescases, an SRP requestor may not have retained sufficient state to know that some service instance is pointing to a host that it is removing. This method of removing services is intended for the case 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">HostHost DescriptionInstruction</xref>.Instruction (see <xref target="hdi"></xref>). </t> <t> To support this, when removing services based on the lease time being zero, an SRP registrarMUST<bcp14>MUST</bcp14> remove all service instances pointing to a host when a host is removed, even if the SRP requestor doesn't list them explicitly. If the KEY lease time is nonzero, the SRP registrarMUST NOT<bcp14>MUST NOT</bcp14> delete the KEY records for these SRP requestors. </t> </section> <section> <name>Removingsome published services</name>Some Published Services</name> <t> In some usecasescases, a requestor may need to removesomea specificservice,service without removing its other services. This can be accomplished in one of twoways. Toways:</t> <ol><li>To simply remove a specific service, the requestor sends a valid SRP Update where the<xref target="servdis">ServiceService DiscoveryInstruction</xref>Instruction (see <xref target="servdis"></xref>) contains a singleDelete an"Delete An RRfrom an RRsetFrom An RRset" update (<xref target="RFC2136" section="2.5.4"sectionFormat="comma"/>) updatesectionFormat="of"/>) that deletes the PTR record whose target is the service instance name.The <xref target="servdesc">Service Description Instruction</xref> inIn thiscasecase, the Service Description Instruction (see <xref target="servdesc"></xref>) contains a singleDelete all"Delete All RRsetsfrom a NameFrom A Name" update (<xref target="RFC2136" section="2.5.3"sectionFormat="comma"/>) updatesectionFormat="of"/>) to the service instance name.</t> <t> The second</li> <li> This alternative is used when some service is being replaced by a different service with a different service instance name. 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 consist of a valid Service Discovery Instruction and a valid Service Description Instruction, the update as a whole is a valid SRPUpdate,Update and will result in the old service being removed and the new oneadded,added; or, to put it differently, the update will result in the old service being replaced by the new service.</t></li></ol> <t> It is perhaps worth notingthatthat, if a service is being updated without the service instance name changing, that situation will look very much like the second alternative above. The difference is that because the target for the PTR record in the Service Discovery Instruction is the same for both theDelete"Delete An RR From AnRRsetRRset" update and theAdd"Add To AnRRSet update,RRset" update (<xref target="RFC2136" section="2.5.1" sectionFormat="of"/>), there is no way to tell whether they were intended to be one or two Instructions. The same would be true of the Service Description Instruction. </t> <t> Whichever of these two alternatives is used, the host lease will be updated with the lease time provided in the SRP update. In neither of these cases is it permissible to delete the host. All services must point to a host. If a host is to be deleted, this must be done using the method described in <xref target="zero-lease"/>, which deletes the host and all services that have that host as their target. </t> </section> </section> </section></section> <section anchor="server_behavior"> <name>Validation and Processing of SRP Updates</name> <section anchor="add_validation"> <name>Validation of DNS Update Add and Delete RRs</name> <t> The SRP registrar first validates that the DNS Update is a syntactically and semantically valid DNS Update according to the rules specified in <xref target="RFC2136"/>.</t> <t> SRP Updates consist of a set of <em>instructions</em> that together add or remove one or more services. Each instruction consists of some combination of delete updates and add updates. When an instruction contains a delete and an add, the deleteMUST<bcp14>MUST</bcp14> precede the add.</t> <t> The SRP registrar checks each instruction in the SRP Update to see that it is either a Service Discovery Instruction, a 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;otherwiseotherwise, the add will have no effect. This is the only orderingconstraint;constraint: aside from this constraint, updates may appear in whatever order is convenient when constructing the update.</t> <t> Because the SRP Update is a DNS update, itMUST<bcp14>MUST</bcp14> contain a single question that indicates the zone to be updated. Every delete and update in an SRP UpdateMUST<bcp14>MUST</bcp14> be within the zone that is specified for the SRP Update.</t> <section anchor="servdis"> <name>Service Discovery Instruction</name> <t>An instruction is a Service Discovery Instruction if it:</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 itcontains</t> <ul spacing="compact"> <li>exactlycontains * exactly one "Add to an RRSet"(<xref target="RFC2136" section="2.5.1" sectionFormat="comma"/>)([RFC2136], Section 2.5.1) or exactly one "Delete an RR from an RRSet"(<xref target="RFC2136" section="2.5.4" sectionFormat="comma"/>)([RFC2136], Section 2.5.4) RRupdate,</li> <li>whichupdate, * which updates a PTRRR,</li> <li>theRR, * the target of which is a Service InstanceName</li> <li><t>forName * for which name a Service Description Instruction is present in the SRP Update,and:</t> <ul spacing="compact"> <li>ifand: - 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</li> <li>ifor - 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 InstanceName.</li></ul></li> <li>andName. * 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. --> <ul spacing="normal"> <li><t>Contains exactly one "Add To An RRset" RR update (<xref target="RFC2136" section="2.5.1" sectionFormat="of"/>) or exactly one "Delete An RR From An RRset" RR update (<xref target="RFC2136" section="2.5.4" sectionFormat="of"/>), 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:</t> <ul spacing="compact"> <li>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.</li> <li>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.</li></ul></li> <li>Contains no other add or delete RR updates for the same name as the PTR RR Update.</li> </ul> <t> Note that there can be more than one Service Discovery Instruction for the same name if the SRP requestor is advertising more than one service of the sametype,type or is changing the target of a PTR RR. This is also true for SRP subtypes (<xref target="RFC6763"section="7.1"/>).section="7.1" sectionFormat="of"/>). For each such PTR RR add or delete, the above constraints must be met.</t> </section> <section anchor="servdesc"> <name>Service Description Instruction</name> <t>An instruction is a Service Description Instruction if, for the appropriate Service Instance Name, the following are true:</t> <ulspacing="compact">spacing="normal"> <li> It contains exactly one "DeleteallAll RRsetsfrom a name"From A Name" update for the service instance name(<xref(see <xref target="RFC2136" section="2.5.3"sectionFormat="comma"/>),</li>sectionFormat="of"/>).</li> <li> It contains zero or one "Addto anTo An RRset" SRVRR,</li>RR.</li> <li> It contains zero or one "Addto anTo 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 KEYMUST<bcp14>MUST</bcp14> match the KEY RR given in the HostDescription),</li>Description).</li> <li> It contains zero or more "Addto anTo An RRset" TXTRRs,</li>RRs.</li> <li> If there is one "Addto anTo An RRset" SRV update, thereMUST<bcp14>MUST</bcp14> be at least one "Addto anTo An RRset" TXT update.</li> <li>The<t>The target of the SRV RR Add, ifpresentpresent, points to a hostname for which there is a Host Description Instruction in the SRPUpdate, or</li> <li> IfUpdate; or if there is no "Addto anTo An RRset" SRV RR, theneither:</li> <li><ul>either:</t> <ul spacing="normal"> <li>the name to which the "DeleteallAll RRsetsfrom a name"From A Name" applies does not exist, or</li> <li>there is an existing KEY RR on thatname, whichname that matches the key with which the SRP Update was signed.</li></ul></li> <li> No other resource records on the Service Instance Name are modified.</li> </ul> <t>An SRP registrarMUST<bcp14>MUST</bcp14> correctly handle compressed names in the SRV target.</t> </section> <section anchor="hdi"> <name>Host Description Instruction</name> <t>An instruction is a Host Description Instruction if, for the appropriate hostname, itcontains</t>contains the following:</t> <ulspacing="compact">spacing="normal"> <li> exactly one "DeleteallAll RRsetsfrom a name"From A Name" RR,</li> <li> one or more "Addto anTo An RRset" RRs of type A and/orAAAA,</li>AAAA, and </li> <li> exactly one "Addto anTo An RRset" RR that adds a KEY RR that contains the public key corresponding to the private key that was used to sign themessage,</li> <li> Hostmessage</li> </ul> <t>Host Description Instructions do not modify any other resourcerecords.</li> </ul>records.</t> <t> A and/or AAAA records that are not of sufficient scope to be validly published in a DNS zoneMAY<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> <t> For example, if a link-scope address or IPv4 autoconfiguration address is provided by the SRP requestor, the SRP registrar could not publish this in a DNS zone. However, in some situations, the registrar might make the records available through a mechanism such as an advertising proxy only on the specific link from which the SRP updateoriginated; inoriginated. In such a situation,locally-scopedlocally scoped records are still valid.</t> </section> </section> <section> <name>Valid SRP Update Requirements</name> <t> An SRP UpdateMUST<bcp14>MUST</bcp14> contain exactly one Host Description Instruction. In addition, thereMUST NOT<bcp14>MUST NOT</bcp14> be any Service Description Instruction to which no Service Discovery Instruction points. A DNS Update that contains any additional adds or deletes that cannot be identified as Service Discovery, ServiceDescriptionDescription, or Host Description Instructions is not an SRP Update. A DNS update that contains any prerequisites is not an SRP Update.</t> <t>An SRP UpdateMUST<bcp14>MUST</bcp14> include an EDNS(0) Update Lease option (see <xreftarget="I-D.ietf-dnssd-update-lease"/>.target="RFC9664"/>). The LEASE time specified in the Update Lease optionMUST<bcp14>MUST</bcp14> be less than or equal to the KEY-LEASE time. A DNS update that does not include the Update Lease option, or that includes a KEY-LEASE value that is less than the LEASE value, is not an SRP update.</t> <t>When an SRP registrar receives a DNS Update that is not an SRP update, itMAY<bcp14>MAY</bcp14> process the update as regularRFC2136 updates,updates described in <xref target="RFC2136" format="default"/>, including access control checks and constraint checks, if supported.OtherwiseOtherwise, the SRP registrarMUST<bcp14>MUST</bcp14> reject the DNS Update with the Refused RCODE.</t> <t> If the definitions of each of these instructions are followed carefully and the update requirements are validated correctly, many DNS Updates that look very much like SRP Updates nevertheless will fail to validate. For example, a DNS update that contains anAdd to an RRset"Add To An RRset" instruction for a Service Name and an Add to an RRset instruction for a Service Instance Name, where the PTR record added to the Service Name does not reference the Service Instance Name, is not a valid SRP Updatemessage,message but may be a validRFC2136 update.</t>update as described in <xref target="RFC2136" format="default"/>.</t> </section> <section> <name>FCFS NameAndand 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. --> <t> 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. 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 Description Instruction. The registrar performs the same check for the KEY records in any Service Description Instructions. For KEY records that were omitted from Service Description Instructions, the KEY from the Host Description Instruction is used. If any existing KEY record corresponding to a KEY record in the SRP Update does not match the KEY record in the SRP Update (whether provided or taken from the Host Description Instruction), then the SRP registrarMUST<bcp14>MUST</bcp14> reject the SRP Update with the YXDomain RCODE.</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. --> <t> Otherwise, the SRP registrar validates the SRP Update using SIG(0) against the public key in the KEY record of the Host Description Instruction. If the validation fails, the registrarMUST<bcp14>MUST</bcp14> reject the SRP Update with the Refused RCODE. Otherwise, the SRP Update is considered valid andauthentic,authentic and is processed according to the method described inRFC2136.</t><xref target="RFC2136" format="default"/>.</t> <t> KEY record updates omitted from Service Description Instruction are processed as if they had been explicitlypresent: every Service Description that is updated MUST, afterpresent. After the SRP Update has been applied, every Service Description that is updated <bcp14>MUST</bcp14> have a KEYRR,RR: and it must be the same KEY RR that is present in the Host Description to which the Service Description refers.</t> <t> <xref target="RFC3445"/> states that the flags field in the KEY RRMUST<bcp14>MUST</bcp14> be zero except for bit 7, which can be one in the case of a zone key. However, the SRP registrarMUST NOT<bcp14>MUST NOT</bcp14> validate the flags field.</t> </section> <section> <name>Handling of Service Subtypes</name> <t> SRP registrarsMUST<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 appears in the SRP Update is the entirety of information about that service and its subtypes. If any subtype appeared in a previous update but does not appear in the current update, then the SRP registrarMUST<bcp14>MUST</bcp14> remove that subtype. </t> <t> Similarly, there is no mechanism for deleting subtypes. A delete of a service deletes all of its subtypes. To delete an individual subtype, an SRP Update must be constructed that contains the service type and all subtypes for that service except for the one to be deleted. </t> </section> <section> <name>SRP Updateresponse</name>Response</name> <t> The status that is returned depends on the result of processing theupdate,update and can be either NoError, ServFail,RefusedRefused, orYXDomain: allYXDomain. All other possible outcomes will already have been accounted for when applying the constraints that qualify the update as an SRP Update. The meanings of these responses are explained in <xref target="RFC2136" section="2.2"/>.</t> <t> In the case of a response other than NoError, <xref target="RFC2136" section="3.8"/> specifies that the server is permitted to respond either with no RRs or to copy the RRs sent by the client into the response. The SRPRequestor MUST NOTrequestor <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 indications as to why the update failed, but atpresentthe time of writing this is notspecified, sospecified. So, if a client were to attempt to validate the RRs in the response, it might reject such aresponse,response since it would containRRs,RRs but probably not a set of RRs identical to what was sent in the SRP Update.</t> </section> <section> <name>Optional Behavior</name> <t> The SRP registrarMAY<bcp14>MAY</bcp14> add a Reverse Mapping(<xref(see <xref target="RFC1035"section="3.5"/>,section="3.5"/> and <xref target="RFC3596" section="2.5"/>) that corresponds to the Host Description. This is not required because theReverse Mappingreverse mapping serves no protocol function, but it may be useful for debugging,e.g.e.g., in annotating network packet traces or logs. In order for the registrar to do a reverse mapping update, it must be authoritative for the zone that would need to beupdated,updated or have credentials to do the update. The SRP requestorMAY<bcp14>MAY</bcp14> also do a reverse mapping update if it has credentials to do so.</t> <t> The SRP registrarMAY<bcp14>MAY</bcp14> apply additional criteria when accepting updates. In some networks, it may be possible to do out-of-band registration ofkeys,keys and only accept updates frompre-registeredpreregistered keys. In this case, an update for a key that has not been registeredSHOULD<bcp14>SHOULD</bcp14> be rejected with the Refused RCODE.</t> <t> There are at least two benefits to doing this rather than simply using normal SIG(0) DNSupdates. First, theupdates:</t> <ol><li>The same registration protocol can be used in both cases, so both use cases can be addressed by the same SRP requestorimplementation. Second, theimplementation.</li> <li>The registration protocol includes maintenance functionality not present with normal DNSupdates.</t>updates.</li></ol> <t> Note that the semantics of using SRP in this way are different than for typicalRFC2136 implementations: theimplementations described in <xref target="RFC2136" format="default"/>. The KEY used to sign the SRP Update only allows the SRP requestor to update records that refer to its Host Description.RFC2136 implementationsImplementations specific to <xref target="RFC2136" format="default"/> do not normally provide a way to enforce a constraint of this type.</t> <t> The SRP registrar could also have a dictionary of names or name patterns that are not permitted. If such a list is used, updates for Service Instance Names that match entries in the dictionary are rejected with a Refused RCODE.</t> </section> </section> </section> <section> <name>TTL Consistency</name> <t> All RRs within an RRset are required to have the same TTL(<xref(see <xref target="RFC2181" section="5.2"sectionFormat="comma"> Clarifications to the DNS Specification</xref>).sectionFormat="of"/>). In order to avoid inconsistencies, SRP places restrictions on TTLs sent by requestors and requires that SRP registrars enforce consistency.</t> <t> Requestors sending SRP UpdatesMUST<bcp14>MUST</bcp14> use consistent TTLs in all RRs within the SRP Update.</t> <t> SRP registrarsMUST<bcp14>MUST</bcp14> check that the TTLs for all RRs within the SRP Update are the same. If they are not, the SRP updateMUST<bcp14>MUST</bcp14> be rejected with a Refused RCODE.</t> <t> Additionally, when adding RRs to anRRset, for exampleRRset (for example, when processing Service Discoveryrecords,records), the SRP registrarMUST<bcp14>MUST</bcp14> use the same TTL on all RRs in the RRset. How this consistency is enforced is up to the implementation.</t> <t> TTLs sent in SRP Updates are advisory: they indicate the SRP requestor's guess as to what a good TTL would be. SRP registrars may override these TTLs. SRP registrarsSHOULD<bcp14>SHOULD</bcp14> ensure that TTLs are reasonable: neither too long nor too short. The TTLSHOULD NOT<bcp14>SHOULD NOT</bcp14> ever be longer than the lease time (<xref target="stale"/>). Shorter TTLs will result in more frequent data refreshes; this increases latency on the DNS-SD client side, increases load on any caching resolvers and on the authoritative server, and also increases network load, which may be an issue for constrained networks. Longer TTLs will increase the likelihood that data in caches will be stale. TTL minimums and maximumsSHOULD<bcp14>SHOULD</bcp14> be configurable by the operator of the SRP registrar. </t> </section> <section anchor="maintenance"> <name>Maintenance</name> <section anchor="stale"> <name>Cleaningup stale data</name>Up Stale Data</name> <t>Because theDNS&nbhy;SDDNS-SD registration protocol isautomatic,automatic and not managed by humans, some additional bookkeeping is required. When an update is constructed by the SRP requestor, itMUST<bcp14>MUST</bcp14> include an EDNS(0) Update Lease Option (see <xreftarget="I-D.ietf-dnssd-update-lease"/>.target="RFC9664"/>). The Update Lease Option contains two lease times: the Lease Time and the KEY Lease Time.</t><t>These leases are promises, similar<t>Similar to DHCP leases (see <xreftarget="RFC2131">DHCP leases</xref>,target="RFC2131"></xref>), these leases are promises from the SRP requestor that it will send a new update for the service registration before the lease time expires. The Lease time is chosen to represent the time after the update during which the registered records other than the KEY record can be assumed to be valid. The KEY lease time represents the time after the update during which the KEY record can be assumed to be valid.</t> <t>The reasoning behind the different lease times is discussed inthe section on FCFS naming (<xref target="fcfs"/>).<xref target="fcfs" format="default"/>. SRP registrars may be configured with limits for these values.AAt the time of writing, a default limit of two hours for the Lease and 14 days for the SIG(0) KEY arecurrentlythought to be good choices. Constrained devices with limited battery that wake infrequently are likely to request longer leases; registrars that support such devices may need to set higher limits. SRP requestors that are going to continue to use names on which they hold leasesSHOULD<bcp14>SHOULD</bcp14> update well before the leaseends,ends in case the registrar is unavailable or under heavy load.</t> <t> The lease time applies specifically to the host. All service instances, and all service entries for such service instances, depend on the host. When the lease on a host expires, the host and all services that reference itMUST<bcp14>MUST</bcp14> be removed at the sametime—ittime: it is never valid for a service instance to remain when the host it references has been removed. If the KEY record for the host is to remain, the KEY record for any services that reference itMUST<bcp14>MUST</bcp14> also remain. However, the service PTR recordMUST<bcp14>MUST</bcp14> beremoved,removed since it has no key associated withit,it and since it is never valid to have a service PTR record for which there is no service instance on the target of the PTR record. </t> <t> SRP registrarsMUST<bcp14>MUST</bcp14> also track a lease time per service instance. The reasonfor doing this isbeing that a requestor may re-register a host with a different set ofservices,services and not remember that some different service instance had previously been registered. In this case, when that service instance lease expires, the SRP registrarMUST<bcp14>MUST</bcp14> remove the service instance (although the KEY record for the service instanceSHOULD<bcp14>SHOULD</bcp14> be retained until the KEY lease on that service expires). This is beneficialbecause otherwisebecause, otherwise, if the SRP requestor continues to renew thehost,host but never mentions the stale service again, the stale service will continue to be advertised. </t> <t>The SRP registrarMUST<bcp14>MUST</bcp14> include an EDNS(0) Update Lease option in the response if the lease time proposed by the requestor has been shortened or lengthened by the registrar. The requestorMUST<bcp14>MUST</bcp14> check for the EDNS(0) Update Lease option in the response andMUST<bcp14>MUST</bcp14> use the lease times from that option in place of the options that it sent to the registrar when deciding when to renew its registration. The times may be shorter or longer than those specified in the SRPUpdate;Update: the SRP requestor must honor them in either case.</t><t>SRP<!-- [rfced] In Section 5.1, we see both "N" and "'n'". Please review and let us know if/how we may update for uniformity. Original "N": SRP requestors SHOULD assume that each lease ends N seconds after the update was first transmitted, where N is the lease duration. Original "'n'": 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 'n' when the remaining time on the lease is less than 'n'. --> <t>SRP requestors <bcp14>SHOULD</bcp14> assume that each lease ends N seconds after the update was first transmitted (where N is the lease duration). SRPRegistrars SHOULDregistrars <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 requestor sent it, this avoids the possibility of misunderstandings.</t> <t>SRP registrarsMUST<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 updatesMAY<bcp14>MAY</bcp14> accept updates that don't include leases, butSHOULDthey <bcp14>SHOULD</bcp14> differentiate between SRP Updates and otherupdates,updates andMUST<bcp14>MUST</bcp14> reject updates that would otherwise be SRP Updates if they do not include leases.</t> <t>Lease times have a completely different function than TTLs. On an authoritative DNS server, the TTL on a resource record is aconstant: wheneverconstant. 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 'n' when the remaining time on the lease is less than 'n'.</t> </section> </section> <section> <name>Security Considerations</name> <section anchor="source_validation"> <name>Source Validation</name> <t>SRP Updates have no authorization semantics other than FCFS.This means thatThus, if an attacker from outsideofthe administrative domain of the SRP registrar knows the registrar's IP address, itcancan, inprincipleprinciple, send updates to the registrar that will be processed successfully. Therefore, SRPRegistrars SHOULD thereforeregistrars <bcp14>SHOULD</bcp14> be configured to reject updates from source addresses outside of the administrative domain of the registrar.</t> <t>For TCP updates, the initial SYN-SYN+ACK handshake prevents updates being forged by an off-network attacker. In order to ensure that this handshake happens, SRP registrars relying on three-way-handshake validationMUST NOT<bcp14>MUST NOT</bcp14> accept TCP Fast Open payloads (see <xreftarget="RFC7413"/> payloads.target="RFC7413"/>). If the network infrastructure allows it, an SRP registrarMAY<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 type of spoofing at all ingress points.</t> <t>For UDP updates from constrained devices, spoofing would have to be prevented with appropriate source address filtration on routers<xref target="RFC2827"/>.(<xref target="RFC2827"/>). This would ordinarily be accomplished by measures such asarethose described in<xref(<xref target="RFC7084" section="4.5"sectionFormat="of"/>.sectionFormat="of"/>). For example, a stub router<xref target="I-D.ietf-snac-simple"/>(<xref target="I-D.ietf-snac-simple"/>) for a constrained network might only accept UDP updates from source addresses known to be on-link on that stubnetwork,network and might 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> </section> <section> <name>Other DNSupdates</name>Updates</name> <t>Note that these rules only apply to the validation of SRP Updates. A server that accepts updates from SRP requestors may also accept other DNS updates, and those DNS updates may be validated using different rules. However, in the case of a DNS server that accepts SRP updates, the intersection of the SRP Update rules and whatever other update rules are present must be considered very carefully.</t> <t>For example, anormal,normal authenticated DNS update to any RR that was added using SRP, but that is authenticated using a different key, could be used to override a promise made by the SRP registrar to an SRPrequestor,requestor by replacing all or part of the service registration information with information provided by an authenticated DNS update requestor. An implementation that allows both kinds of updatesSHOULD NOT<bcp14>SHOULD NOT</bcp14> allow DNS Update requestors that are using different authentication and authorization credentials to update records added by SRP requestors.</t> </section> <section> <name>Risks ofallowing arbitrary namesAllowing Arbitrary Names to beregisteredRegistered in SRPupdates</name>Updates</name> <t>It is possible to set up SRP updates for a zone that is used for non-DNSSD services. For example, imagine that you set up SRP service for example.com. SRP hosts can now register names like "www" or "mail" or "smtp" in this domain. In addition, SRP updates using FCFS naming can insert names that are obscene or offensive into the zone. There is no simple solution to these problems.WeHowever, we have two recommendations to address thisproblem, however:</t>problem:</t> <ulspacing="compact">spacing="normal"> <li>Do not provide SRP service in organization-level zones. Use subdomains of the organizational domain forDNS service discovery.DNS-SD. This does not prevent registering names as mentionedabove,above but does ensure that genuinely important names are not accidentally reserved for SRP clients.SoSo, for example, the zone "dnssd.example.com" could be used instead of "example.com" for SRP updates. Because of the way thatDNS browsingDNS-browsing domains are discovered, there is no need for the DNSSD 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 of common obscene and offensive names are no doubtavailable,available and can be augmented with a list of typical "special" names like "www", "mail","smtp""smtp", and so on. Lists of names are generallyavailable,available or can be constructed manually.</li> </ul> </section> <section> <name>Security oflocal service discovery</name>Local Service Discovery</name> <t>Local links can be protected by managed services such asRARouter Advertisement Guard (see <xreftarget="RFC6105"/>,target="RFC6105"/>), but multicast services likeDHCP <xref target="RFC2131"/>, DHCPv6 <xref target="RFC8415"/>DHCP, DHCPv6, and IPv6 Neighbor Discovery (see <xreftarget="RFC4861"/> aretarget="RFC2131"/>, <xref target="RFC8415"/>, and <xref target="RFC4861"/>, respectively) are, in mostcasescases, not authenticated and can't be controlled on unmanaged networks, such as home networks andsmall-officesmall office networks where no network management staff are present. In such situations, the SRP service has comparatively fewer potential security exposuresand henceand, hence, is not the weak link. This is discussed in more detail in <xref target="how-to-secure"/>.</t> <t>The fundamental protection for networks of this type is the user's choice of what devices to add to the network. Work is being done in other working groups and standards bodies to improve the state of the art for network on-boarding and device isolation (e.g., <xref target="RFC8520"/> provides a means for constraining what behaviors are allowed for a device in an automatic way), but such work is out of scope for this document.</t> </section> <section> <name>SRP Registrar Authentication</name> <t>This specification does not provide a mechanism for validating responses from SRPRegistrarsregistrars to SRP requestors. In principle, a KEY RR could be used by a non-constrained SRP requestor to validate responses from the registrar, but this is not required, 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 <xref target="RFC7858" section="4.2"sectionFormat="comma"/>sectionFormat="of"/> could be used for authentication of the SRP registrar,e.g.e.g., to prevent man-in-the-middle attacks.HoweverHowever, the use of such keys is impractical for an unmanaged service registrationprotocol, and henceprotocol; hence, it is out of scope for this document.</t> </section> <section anchor="rsa"> <name>Required Signature Algorithm</name> <t> For validation, SRP registrarsMUST<bcp14>MUST</bcp14> implement the ECDSAP256SHA256 signature algorithm. SRP registrarsSHOULD<bcp14>SHOULD</bcp14> implement the algorithms that are specified in <xref target="RFC8624" section="3.1"sectionFormat="comma"/>,sectionFormat="of"/>, in the validation column of the table, that are numbered 13 orhigherhigher, and that have a"MUST", "RECOMMENDED","<bcp14>MUST</bcp14>", "<bcp14>RECOMMENDED</bcp14>", or"MAY""<bcp14>MAY</bcp14>" designation in the validation column of the table. SRP requestorsMUST NOT<bcp14>MUST NOT</bcp14> assume that any algorithm numbered lower than 13 is available for use in validating SIG(0) signatures.</t> </section> </section> <section> <name>Privacy Considerations</name> <t> Because DNS-SD SRP Updates can be sent off-link, the privacy implications of SRP are different than formulticast DNSmDNS responses. Host implementations that are using TCPSHOULD<bcp14>SHOULD</bcp14> also use TLS if available. SRPRegistrarregistrar implementationsMUST<bcp14>MUST</bcp14> offer TLS support. The use of TLS with DNS is described in <xref target="RFC7858"/>. Because there is no mechanism for sharing keys, validation of DNS-over-TLS keys is not possible; DNS-over-TLS is used only as described in <xref target="RFC7858" section="4.1"sectionFormat="comma"/>sectionFormat="of"/>. </t> <t> Hosts that implement TLS supportSHOULD NOT<bcp14>SHOULD NOT</bcp14> fall back toTCP; sinceTCP. Since SRP registrars are required to support TLS, it is entirely up to the host implementation whether to use it. </t> <t> Public keys can be used as identifiers to track hosts. SRP registrarsMAY<bcp14>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 a KEY recordMUST NOT<bcp14>MUST NOT</bcp14> store the KEY record in the zone itself. Because the KEY record isn't in the zone, thenonexistancenonexistence of the KEY record can be validated. If the zone is not signed, the serverMAY<bcp14>MAY</bcp14> instead return a negative non-error response (either NXDOMAIN or no data). </t> </section> <section> <name>Domain Name Reservation Considerations</name> <t>This section specifies considerations for systems involved in domain name resolution when resolving queries for names ending with'.service.arpa.'.".service.arpa.". Each item in this section addresses some aspect of the DNS or the process of resolving domain names that would be affected by this special-use allocation. Detailed explanations of these items can be found inSection 5 of<xreftarget="RFC6761"/>.</t>target="RFC6761" sectionFormat="of" section="5"/>.</t> <section> <name>Users</name> <t>The current proposed use for'service.arpa'"service.arpa" does not require special knowledge on the part of the user. While the'default.service.arpa.'"default.service.arpa." subdomain is used as a generic name for registration, users are not expected to see this name in user interfaces. In the event that it does show up in a user interface, it is just a domainname,name and requires no special treatment by the user. Users are not expected to see this name in user interfaces, although it's certainly possible that they might. If they do, they are not expected to treat it specially.</t> </section> <section> <name>Application Software</name> <t> Application software does not need to handle subdomains of'service.arpa'"service.arpa" specially.'service.arpa' SHOULD NOT"service.arpa" <bcp14>SHOULD NOT</bcp14> be treated as more trustworthy than any other insecure DNS domain, simply because it islocally-servedlocally served (or for any other reason). It is not possible to register a PKI certificate for a subdomain of'service.arpa.'"service.arpa." because it is alocally-servedlocally served domain name.SoSo, no such subdomain can be consideredasto be uniquely identifying a particular host, as would be required for such a PKIcertcertificate to be issued. If a subdomain of'service.arpa.'"service.arpa." is returned 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 and practices for authenticating such endpoints are out of scope for this document.</t> </section> <section> <name>Name Resolution APIs and Libraries</name> <t>Name resolution APIs and librariesMUST NOT<bcp14>MUST NOT</bcp14> recognize names that end in'.service.arpa.'"service.arpa." as special andMUST NOT<bcp14>MUST NOT</bcp14> treat them as having special significance, except that it may be necessary that such APIs not bypass the locally configured recursive resolvers.</t> <t>One or more IP addresses for recursive DNS servers will usually be supplied to the client through router advertisements or DHCP. For an administrative domain that uses subdomains of'service.arpa.',"service.arpa.", the recursive resolvers provided by that domain will be able to answer queries for subdomains of'service.arpa.'; other"service.arpa.". Other (non-local) resolvers will not, or they will provide answers that are not correct within that administrative domain.</t> <t>A host that is configured to use a resolver other than one that has been provided by the local network may not beunableable toresolve,resolve or may receive incorrect resultsfor,for subdomains of'service.arpa.'."service.arpa.". In order to avoid this, it is permissible that hosts use the resolvers that are locally provided for resolving'service.arpa.',"service.arpa.", even when they are configured to use other resolvers.</t> </section> <section> <name>Caching DNS Servers</name><t>There<!-- [rfced] In the following text, before the two numbered points, the text reads "There are three considerations". Should we update "three" to "two", or is there another point that the text is missing? Current: There are three considerations for caching DNS servers that follow thisspecification:</t> <ol> <li>Forspecification: 1. For correctness, recursive resolvers at sites using 'service.arpa.'mustmust, inpracticepractice, transparently support DNSSEC queries: queries for DNSSEC records and queries with the DNSSEC OK (DO) bit set(<xref target="RFC4035" section="3.2.1" sectionFormat="of"/>).(Section 3.2.1 of [RFC4035]). DNSSEC validation is a Best Current Practice<xref target="RFC9364"/>:([RFC9364]): although validation is not required, a caching recursive resolver that does not validate answers that can be validated may cache invalid data.This, inIn 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 DNSSECvalidation.</li> <li> <t>Unlessvalidation. 2. Unless configured otherwise, recursive resolvers and DNS proxies MUST behave as described in Locally ServedZones, <xref target="RFC6303" section="3" sectionFormat="of"/>.Zones (Section 3 of [RFC6303]). That is, queries for 'service.arpa.' and subdomains of 'service.arpa.' MUST NOT be forwarded, with one important exception: a query for a DS record with the DO bit set MUST return the correct answer for that question, including correct information in the authority section that proves that the record isnonexistent.</t> <t>So,nonexistent. So, for example, a query for the NS record for 'service.arpa.' MUST NOT result in that query being forwarded to an upstream cache nor to the authoritative DNS server for '.arpa.'. However, as necessary to provide accurate authority information, a query for the DS record MUST 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>There are three considerations for caching 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: --> <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 (<xref target="RFC4035" section="3.2.1" sectionFormat="of"/>). DNSSEC validation is a Best Current Practice (<xref target="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.</li> <li> <t>Unless configured otherwise, recursive resolvers and DNS proxies <bcp14>MUST</bcp14> behave as described in Locally Served Zones (<xref target="RFC6303" section="3" sectionFormat="of"/>). That is, queries for "service.arpa." and subdomains of "service.arpa." <bcp14>MUST NOT</bcp14> be forwarded, with one important 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.</t> <!--[rfced] Is this text redundant (with two uses of necessary)? Does our suggestion change your intended meaning? Original: However, as necessary to provide accurate authority information, a query for the DS record MUST result in forwarding whatever queries are necessary; typically, ... Perhaps: However, to provide accurate authority information, a query for the DS record MUST result in forwarding whatever queries are necessary. --> <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, as necessary 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 DSrecord,record since the necessary authority information will be included in the authority section of the response if the DO bit is set.</t> </li> </ol> </section> <section> <name>Authoritative DNS Servers</name> <t>No special processing of'service.arpa.'"service.arpa." is required for authoritative DNS server implementations. It is possible that an authoritative DNS server might attempt to check the authoritative servers for'service.arpa.'"service.arpa." for a delegation beneath that name before answering authoritatively for such a delegated name. In such a case, because the name always has only local significance, there will be no such delegation in the'service.arpa.' zone, and so"service.arpa." zone; therefore, the server would refuse to answer authoritatively for such a zone. A server that implements this sort of checkMUST<bcp14>MUST</bcp14> be configurable so that either it does not do this check for the'service.arpa.'"service.arpa." domain or it ignores the results of the check.</t> </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. --> <name>DNS Server Operators</name> <t>DNS server operatorsMAY<bcp14>MAY</bcp14> configure an authoritative server for'service.arpa.'"service.arpa." for use with SRP. The operator for the DNS servers authoritative for'service.arpa.'"service.arpa." in the global DNS will configure any such servers as described in <xref target="delegation"/>.</t> </section> <section> <name>DNS Registries/Registrars</name><t>'service.arpa.'<t>"service.arpa." is a subdomain of the'arpa'"arpa" top-level domain, which is operated by IANA under the authority of the Internet Architecture Board (IAB) according to the rules established in[RFC3172].<xref target="RFC3172" format="default"/>. There are no other DNS registrars for'.arpa'.</t>".arpa".</t> </section> </section> <section anchor="delegation"> <name>Delegation of'service.arpa.'</name>"service.arpa."</name> <t>In order to be fully functional, the owner of the'arpa.'"arpa." zone must add a delegation of'service.arpa.'"service.arpa." in the'.arpa.'".arpa." zone (see <xreftarget="RFC3172"/>.target="RFC3172"/>). This delegation is to be set up as was done for'home.arpa',"home.arpa", as a result of the specification in <xref target="RFC8375" section="7" sectionFormat="of"/>. This is currently the responsibility of the IAB (see <xreftarget="IAB-ARPA"/></t>target="IAB-ARPA"/>).</t> </section> <section> <name>IANA Considerations</name> <section><name>Registration<!--[rfced] We have some questions about Section 10.1 in the IANA Considerations: a) We see the title of the section is related to the first paragraph 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 DomainName</name> <t>IANAName IANA is requested to record the domain name 'service.arpa.' in the Special-Use Domain Names registry<xref target="SUDN"/>.[SUDN]. IANA is requested, with the approval of IAB, to implement the delegation requested in<xref target="delegation"/>.</t> <t>IANASection 9. IANA is further requested to add a new entry to the"Transport-Independent"Transport- Independent Locally-Served Zones" subregistry of the "Locally-Served DNS Zones" registry<xref target="LSDZ"/>.[LSDZ]. The entry will be for the domain 'service.arpa.' with the description"DNS&nbhy;SD"DNS-SD Service Registration Protocol Special-Use Domain", and listing this document as the reference. 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]. --> <name>Registration and Delegation of "service.arpa" as a Special-Use Domain Name</name> <t>IANA has recorded the domain name "service.arpa." in the "Special-Use Domain Names" registry (see <xref target="SUDN"/>). IANA has implemented the delegation requested in <xref target="delegation"/>.</t> <t>IANA has also added a new entry to the "Transport-Independent Locally-Served Zones Registry" registry of the "Locally-Served DNS Zones" group (see <xref target="LSDZ"/>). The entry is for the domain "SERVICE.ARPA" with the description "DNS-SD Service Registration Protocol Special-Use Domain" and lists this document as the reference.</t> </section> <section anchor="subdomains"> <name>Subdomains of'service.arpa.'</name>"service.arpa."</name> <t>This document only makes use of the'default.service.arpa'"default.service.arpa" subdomain of'service.arpa.'"service.arpa." Other subdomains are reserved for future use by DNS-SD or related work.TheIANAis requested to create a registry,has created the "service.arpa Subdomain"registry.registry (see <xref target="SUB"/>). The IETFshall havehas change control for this registry. New entries may be added either as a result of Standards Action<xref(<xref target="RFC8126"section="4.9"/>section="4.9" sectionFormat="of"/>) or with IESGapproval <xrefApproval (<xref target="RFC8126"section="4.10"/>,section="4.10" sectionFormat="of"/>), provided that a specification exists<xref(<xref target="RFC8126"section="4.6"/>.section="4.6"/>). </t> <t>TheIANAshall grouphas grouped the "service.arpa Subdomain" registry with the "Locally-Served DNS Zones"registry.group. The registryshall beis a table with three columns: the subdomain name (expressed as afully-qualifiedfully qualified domain name), a brief description of how it is used, and a reference to the document that describes its use in detail. </t> <t> This initial contents of this registryshall beginare asthe following table:follows: </t> <table> <thead> <tr> <th>Subdomain Name</th> <th>Description</th><th>reference</th><th>Reference</th> </tr> </thead> <tbody> <tr> <td>default.service.arpa.</td> <td>Default domain for SRP updates</td><td>[THIS DOCUMENT]</td><td>RFC 9665</td> </tr> </tbody> </table> </section> <section> <name>Service Nameregistrations</name>Registrations</name> <t>IANAis requested to addhas added two new entries to theService Names"Service Name and Transport Protocol PortNumbers registry.Number Registry" (see <xref target="PORT"/>). The followingsectionssubsections contain tables with the fields required by <xref target="RFC6335" section="8.1.1" sectionFormat="of"/>.</t></section><section> <name>'dnssd-srp' Service Name</name> <table><thead><tr><td>Field Name</td><td>Value</td></tr></thead><thead><tr><th>Field Name</th><th>Value</th></tr></thead> <tbody> <tr><td> Service Name </td><td> dnssd-srp </td></tr> <tr><td> Transport Protocol </td><td>TCPtcp </td></tr> <tr><td> Assignee </td><td> IESG <iesg@ietf.org> </td></tr> <tr><td> Contact </td><td> IETF Chair <chair@ietf.org> </td></tr> <tr><td> Description </td><td> DNS-SD ServiceRegistrationDiscovery </td></tr> <tr><td> Reference </td><td>this documentRFC 9665 </td></tr> <tr><td> Port Number </td><td> None </td></tr> <tr><td> Service Code </td><td> None </td></tr> </tbody> </table> </section> <section> <name>'dnssd-srp-tls' Service Name</name> <table><thead><tr><td>Field Name</td><td>Value</td></tr></thead><thead><tr><th>Field Name</th><th>Value</th></tr></thead> <tbody> <tr><td> Service Name </td><td> dnssd-srp-tls </td></tr> <tr><td> Transport Protocol </td><td>TCPtcp </td></tr> <tr><td> Assignee </td><td> IESG <iesg@ietf.org> </td></tr> <tr><td> Contact </td><td> IETFChairChair<chair@ietf.org> </td></tr> <tr><td> Description </td><td> DNS-SD ServiceRegistrationDiscovery (TLS) </td></tr> <tr><td> Reference </td><td>this documentRFC 9665 </td></tr> <tr><td> Port Number </td><td> None </td></tr> <tr><td> Service Code </td><td> None </td></tr> </tbody> </table> </section> </section> <section> <name>Anycast Address</name> <t>IANAis requested to allocatehas allocated an IPv6 Anycast address from the "IANA IPv6 Special-Purpose AddressRegistry,Registry" (see <xref target="IPv6"/>), similar to the Port Control Protocol anycastaddress,address: 2001:1::1. Thevalue TBD is to be replaced with the actual allocation in the table that follows. Thepurpose of this allocation is to provide a fixed anycast address that can be commonly used as a destination for SRP updates when no SRP registrar is explicitly configured. The initial values for the registryare:</t>are as follows:</t> <table> <thead><tr><td>Attribute</td> <td>value</td></tr><tr><th>Attribute</th> <th>Value</th></tr> </thead> <tbody> <tr><td>Address Block</td><td>2001:1::TBD/128</td></tr><td>2001:1::3/128</td></tr> <tr><td>Name</td> <td>DNS-SD Service Registration Protocol Anycast Address</td></tr> <tr><td>RFC</td><td>[this document]</td></tr><td>RFC 9665</td></tr> <tr><td>Allocation Date</td><td>[date of allocation]</td></tr><td>2024-04</td></tr> <tr><td>Termination Date</td> <td>N/A</td></tr> <tr><td>Source</td> <td>True</td></tr> <tr><td>Destination</td> <td>True</td></tr> <tr><td>Forwardable</td> <td>True</td></tr><tr><td>Global</td><tr><td>Globally Reachable</td> <td>True</td></tr><tr><td>Reserved-by-protocol</td><tr><td>Reserved-by-Protocol</td> <td>False</td></tr> </tbody> </table> </section> </section><section> <name>Implementation Status</name> <t>[Note to the RFC Editor: please remove this section prior to publication.]</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 RFC 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 contributors. 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 assign due consideration to documents that have the benefit of running code, which may serve as evidence of valuable experimentation 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/pull/6038</li> <li>mDNSResponder open source project: https://github.com/Abhayakara/mdnsresponder</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 Advertising Proxy <xref target="I-D.ietf-dnssd-advertising-proxy"/>. Both are included in the mDNSResponder open source project mentioned above. </t> </section> <section> <name>Acknowledgments</name> <t>Thanks to <contact fullname="Toke Høiland-Jørgensen"/>, Jonathan Hui, Esko Dijk, Kangping Dong and Abtin Keshavarzian for their thorough technical reviews. Thanks to Kangping and Abtin as well for testing the document by doing an independent implementation. Thanks to Tamara Kemper for doing a nice developmental edit, Tim Wattenberg for doing an SRP requestor proof-of-concept implementation at the Montreal Hackathon at IETF 102, and Tom Pusateri for reviewing during the hackathon and afterwards. Thanks to Esko for a really thorough second last call review. Thanks also to Nathan Dyck, Gabriel Montenegro, Kangping Dong, Martin Turon, and Michael Cowan for their detailed 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 review! Thanks also to the other IESG members who provided comments or simply took the time to review the document.</t> </section></middle> <back> <displayreference target="I-D.cheshire-dnssd-roadmap" to="ROADMAP"/> <displayreferencetarget="I-D.ietf-dnssd-advertising-proxy" to="AP"/> <!-- <displayreference target="I-D.ietf-dnssd-hybrid" to="I-D.ietf-dnssd-hybrid"/> appears to not work in xml2rfc 2.6.2 -->target="I-D.ietf-snac-simple" to="SNAC-SIMPLE"/> <references> <name>References</name> <references> <name>Normative References</name> <!-- [I-D.ietf-dnssd-update-lease] companion document RFC 9664--> <reference anchor="RFC9664" target="https://www.rfc-editor.org/info/rfc9664"> <front> <title>An EDNS(0) Option to Negotiate Leases on DNS Updates</title> <author fullname="Stuart Cheshire" initials="S." surname="Cheshire"> <organization>Apple Inc.</organization> </author> <author fullname="Ted Lemon" initials="T." surname="Lemon"> <organization>Apple Inc</organization> </author> <date month="October" year="2024"/> </front> <seriesInfo name="RFC" value="9664"/> <seriesInfo name="DOI" value="10.17487/RFC9664"/> </reference> <xi:includehref="https://xml2rfc.ietf.org/public/rfc/bibxml3/reference.I-D.ietf-dnssd-update-lease.xml"/> <xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.1035.xml"href="https://bib.ietf.org/public/rfc/bibxml/reference.RFC.1035.xml" /> <xi:includehref="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.1536.xml"href="https://bib.ietf.org/public/rfc/bibxml/reference.RFC.1536.xml" /> <xi:includehref="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.2119.xml"href="https://bib.ietf.org/public/rfc/bibxml/reference.RFC.2119.xml" /> <xi:includehref="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.2136.xml"href="https://bib.ietf.org/public/rfc/bibxml/reference.RFC.2136.xml" /> <xi:includehref="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.2181.xml"href="https://bib.ietf.org/public/rfc/bibxml/reference.RFC.2181.xml" /> <xi:includehref="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.2539.xml"href="https://bib.ietf.org/public/rfc/bibxml/reference.RFC.2539.xml" /> <xi:includehref="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.2782.xml"href="https://bib.ietf.org/public/rfc/bibxml/reference.RFC.2782.xml" /> <xi:includehref="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.2931.xml"href="https://bib.ietf.org/public/rfc/bibxml/reference.RFC.2931.xml" /> <xi:includehref="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.3172.xml"/>href="https://bib.ietf.org/public/rfc/bibxml/reference.RFC.3172.xml"/> <!-- [rfced] Normative reference RFC 3445 has been obsoleted by RFC 4033. We will update to the latter unless we hear objection. --> <xi:includehref="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.3445.xml"/>href="https://bib.ietf.org/public/rfc/bibxml/reference.RFC.3445.xml"/> <xi:includehref="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.3596.xml"/>href="https://bib.ietf.org/public/rfc/bibxml/reference.RFC.3596.xml"/> <xi:includehref="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.4035.xml"/>href="https://bib.ietf.org/public/rfc/bibxml/reference.RFC.4035.xml"/> <xi:includehref="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.6303.xml"/>href="https://bib.ietf.org/public/rfc/bibxml/reference.RFC.6303.xml"/> <xi:includehref="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.6763.xml"/>href="https://bib.ietf.org/public/rfc/bibxml/reference.RFC.6763.xml"/> <xi:includehref="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.7858.xml"href="https://bib.ietf.org/public/rfc/bibxml/reference.RFC.7858.xml" /> <xi:includehref="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.8085.xml"href="https://bib.ietf.org/public/rfc/bibxml/reference.RFC.8085.xml" /> <xi:includehref="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.8126.xml"/>href="https://bib.ietf.org/public/rfc/bibxml/reference.RFC.8126.xml"/> <xi:includehref="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.8174.xml"/>href="https://bib.ietf.org/public/rfc/bibxml/reference.RFC.8174.xml"/> <xi:includehref="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.8375.xml"/>href="https://bib.ietf.org/public/rfc/bibxml/reference.RFC.8375.xml"/> <xi:includehref="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.8624.xml"/>href="https://bib.ietf.org/public/rfc/bibxml/reference.RFC.8624.xml"/> <xi:includehref="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.8765.xml"href="https://bib.ietf.org/public/rfc/bibxml/reference.RFC.8765.xml" /> <xi:includehref="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.9364.xml"href="https://bib.ietf.org/public/rfc/bibxml/reference.RFC.9364.xml" /> </references> <references> <name>Informative References</name> <xi:includehref="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.2131.xml"href="https://bib.ietf.org/public/rfc/bibxml/reference.RFC.2131.xml" /> <xi:includehref="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.2827.xml"href="https://bib.ietf.org/public/rfc/bibxml/reference.RFC.2827.xml" /> <xi:includehref="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.3007.xml"href="https://bib.ietf.org/public/rfc/bibxml/reference.RFC.3007.xml" /> <xi:includehref="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.4861.xml"href="https://bib.ietf.org/public/rfc/bibxml/reference.RFC.4861.xml" /> <xi:includehref="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.6105.xml"href="https://bib.ietf.org/public/rfc/bibxml/reference.RFC.6105.xml" /> <xi:includehref="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.6335.xml"href="https://bib.ietf.org/public/rfc/bibxml/reference.RFC.6335.xml" /> <xi:includehref="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.6760.xml"href="https://bib.ietf.org/public/rfc/bibxml/reference.RFC.6760.xml" /> <xi:includehref="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.6761.xml"href="https://bib.ietf.org/public/rfc/bibxml/reference.RFC.6761.xml" /> <xi:includehref="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.6762.xml"href="https://bib.ietf.org/public/rfc/bibxml/reference.RFC.6762.xml" /> <xi:includehref="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.7084.xml"href="https://bib.ietf.org/public/rfc/bibxml/reference.RFC.7084.xml" /> <xi:includehref="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.7228.xml"href="https://bib.ietf.org/public/rfc/bibxml/reference.RFC.7228.xml" /> <xi:includehref="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.7413.xml"href="https://bib.ietf.org/public/rfc/bibxml/reference.RFC.7413.xml" /> <xi:includehref="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.8415.xml"href="https://bib.ietf.org/public/rfc/bibxml/reference.RFC.8415.xml" /> <xi:includehref="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.8520.xml"href="https://bib.ietf.org/public/rfc/bibxml/reference.RFC.8520.xml" /> <xi:includehref="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.8766.xml"href="https://bib.ietf.org/public/rfc/bibxml/reference.RFC.8766.xml" /> <xi:includehref="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.8945.xml"href="https://bib.ietf.org/public/rfc/bibxml/reference.RFC.8945.xml" /> <!-- [I-D.cheshire-dnssd-roadmap] IESG state: Expired as of 07/15/24--> <xi:includehref="https://xml2rfc.ietf.org/public/rfc/bibxml3/reference.I-D.cheshire-dnssd-roadmap.xml"/> <xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml3/reference.I-D.ietf-dnssd-advertising-proxy.xml"/>href="https://datatracker.ietf.org/doc/bibxml3/reference.I-D.cheshire-dnssd-roadmap.xml"/> <!-- [I-D.ietf-snac-simple] IESG state: I-D Exists as of 07/15/24--> <xi:includehref="https://xml2rfc.ietf.org/public/rfc/bibxml3/reference.I-D.ietf-snac-simple.xml"/>href="https://datatracker.ietf.org/doc/bibxml3/reference.I-D.ietf-snac-simple.xml"/> <reference anchor="SUDN"target="https://www.iana.org/assignments/special-use-domain-names/special-use-domain-names.xhtml">target="https://www.iana.org/assignments/special-use-domain-names"> <front> <title>Special-Use DomainNames Registry</title> <author/> <date month="July" year="2012"/>Names</title> <author> <organization>IANA</organization> </author> </front> </reference> <reference anchor="LSDZ"target="https://www.iana.org/assignments/locally-served-dns-zones/locally-served-dns-zones.xhtml">target="https://www.iana.org/assignments/locally-served-dns-zones"> <front> <title>Locally-Served DNSZonesZones</title> <author> <organization>IANA</organization> </author> </front> </reference> <reference anchor="SUB" target="https://www.iana.org/assignments/locally-served-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/> <date month="July" year="2011"/><author> <organization>IANA</organization> </author> </front> </reference> <reference anchor="IPv6" target="https://www.iana.org/assignments/iana-ipv6-special-registry"> <front> <title>IANA IPv6 Special-Purpose Address Registry</title> <author> <organization>IANA</organization> </author> </front> </reference> <reference anchor="IAB-ARPA" target="https://www.iab.org/documents/correspondence-reports-documents/2017-2/iab-statement-on-the-registration-of-special-use-names-in-the-arpa-domain/"> <front> <title>Internet Architecture Board statement on the registration of special use names in the ARPA domain</title> <author/> <date month="March" year="2017"/> </front> </reference> <reference anchor="ZC"> <front> <title>Zero Configuration Networking: The Definitive Guide</title> <authorinitials="S." surname="Cheshire" fullname="Stuart Cheshire"/> <authorinitials="D.H." surname="Steinberg" fullname="Daniel H. Steinberg"/> <author initials="S." surname="Cheshire" fullname="Stuart Cheshire"/> <date year="2005" month="December"/> </front><seriesInfo name="O'Reilly<refcontent>O'Reilly Media,Inc." value=""/>Inc.</refcontent> <seriesInfo name="ISBN"value="0-596-10100-7"/>value="9780596101008"/> </reference> </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 --> <section> <name>Testingusing standard RFC2136-compliantUsing Standard DNSservers</name>Servers Compliant with RFC 2136</name> <t> It may be useful to set up an authoritative DNS server for testing that does not implement SRP. This can be done by configuring the server to listen on the anycastaddress,address or by advertising it in the _dnssd&nbhy;srp._tcp.<zone> SRV and _dnssd&nbhy;srp&nbhy;tls._tcp.<zone> record. It must be configured to be authoritative for"default.service.arpa","default.service.arpa" and to accept updates from hosts on local networks for names under "default.service.arpa" withoutauthentication,authentication since such servers will not have support for FCFS authentication (<xref target="fcfs"/>).</t> <t> An authoritative DNS server configured in this way will be able to successfully accept and process SRP Updates from requestors that send SRP updates. However, no prerequisites will beapplied, andapplied; this means that the test server will accept internally inconsistent SRPUpdates,Updates and will not stop two SRPUpdates,Updates sent by differentservices,services that claim the samename(s),name or names from overwriting each other.</t> <t> Since SRP Updates are signed with keys, validation of the SIG(0) algorithm used by the requestor can be done by manually installing the requestor's public key on the DNS server that will be receiving the updates. The key can then be used to authenticate the SRPupdate,update and can be used as a requirement for the update. An example configuration for testing SRP using BIND 9 is given in <xref target="bind-example"/>.</t> </section> <section> <name>How toallowAllow SRPrequestorsRequestors toupdate standard RFC2136-compliant servers</name>Update Standard Servers Compliant with RFC 2136</name> <t>OrdinarilyOrdinarily, SRP Updates will fail when sent toan RFC 2136-complianta server compliant with <xref target="RFC2136"/> that does not implement SRP because the zone being updated is"default.service.arpa","default.service.arpa" and because no DNS server that is not an SRP registrar would normally be configured to be authoritative for "default.service.arpa". Therefore, a requestor that sends an SRP Update can tell that the receiving server does not supportSRP,SRP but does supportRFC2136,<xref target="RFC2136"/> because the RCODE will either be NotZone,NotAuthNotAuth, orRefused,Refused or because there is no response to the update request (when using the anycastaddress)</t>address).</t> <t> In thiscasecase, a requestorMAY<bcp14>MAY</bcp14> attempt to register itself using regularRFC2136DNSupdates.updates described in <xref target="RFC2136"/>. To do so, it must discover the default registration zone and the DNS server designated 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 port 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, and a requestor that implements SRPMUST<bcp14>MUST</bcp14> first attempt to use SRP to registeritself,itself and only attempt to useRFC2136backwardscompatibilitycapability with <xref target="RFC2136"/> if that fails. Although the owner name for the SRV record specifiestheUDPprotocolfor updates, it is also possible to use TCP, and TCPSHOULD<bcp14>SHOULD</bcp14> be required to prevent spoofing.</t> </section> <section anchor="bind-example"> <name>Sample BIND9configurationConfiguration fordefault.service.arpa.</name>"default.service.arpa."</name> <figure title="Zone Configuration innamed.conf"><artwork><![CDATA[named.conf"> <artwork><![CDATA[ zone "default.service.arpa." { type primary; file "/etc/bind/primary/service.db"; allow-update { key demo.default.service.arpa.; };}; ]]></artwork></figure>};]]></artwork> </figure> <figure title="Example Zonefile"><artwork><![CDATA[File"> <artwork><![CDATA[ $ORIGIN . $TTL 57600 ; 16 hours default.service.arpa IN SOA ns3.default.service.arpa. postmaster.default.service.arpa. ( 2951053287 ; serial 3600 ; refresh (1 hour) 1800 ; retry (30 minutes) 604800 ; expire (1 week) 3600 ; minimum (1 hour) ) NS ns3.default.service.arpa. 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 demo AAAA 2001:db8:0:2::1 KEY 0 3 13 ( qweEmaaq0FAWok5//ftuQtZgiZoiFSUsm0srWREdywQU 9dpvtOhrdKWUuPT3uEFF5TZU6B4q1z1I662GdaUwqg== ); alg = ECDSAP256SHA256 ; key id = 15008 AAAA ::1]]></artwork></figure>]]></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 requestor 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> </back></rfc><!--Keep this comment[rfced] We had some questions about abbreviations: a) Should "DNSSD" (in "non-DNSSD services" and "DNSSD discovery zone") be updated to "DNS-SD" (hyphen) or "dnssd" (lowercase) to match prior usage in the document? b) Is the "Service" (or "Service Description") redundant here and in similar cases throughout the document (as SD = Service Discovery)? That is, just examples below, more cases exist. Original: DNS-SD Service registration uses public keys and SIG(0) to allow services to defend their registrations. 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. 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 MUST be all zeroes. 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 MUST be at least one "Add to an RRset" TXT update. 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. --> <!-- [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 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" service instance Service Name c) We see the following similar terms. Please let us know how to update for consistency. BIND 9 vs. BIND9 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. 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 Please review other terms similar to these titles if they exist and let us know if any further changes should be made. 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 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 MAY be omitted for brevity; if it is omitted, the SRP registrar MUST 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. 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 thefile Local variables: mode: sgml fill-column:132 sgml-omittag:t sgml-shorttag:t sgml-namecase-general:t sgml-general-insert-case:lower sgml-minimize-attributes:nil sgml-always-quote-attributes:t sgml-indent-step:2 sgml-indent-data:t sgml-parent-document:nil sgml-exposed-tags:nil sgml-local-catalogs:nil sgml-local-ecat-files:nil End:original consistently. Please review and let us know if further updates are necessary. 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 --> <!-- [rfced] Please review each artwork element in Appendix C in case they should be tagged as sourcecode or another element. --> <!-- [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>