RAINS (Another Internet Naming Service) Protocol Specification
draft-trammell-rains-protocol-00

Abstract

This document defines an alternate protocol for Internet name resolution, designed as a prototype to facilitate conversation about the evolution or replacement of the Domain Name System protocol. It attempts to answer the question: “how would we design the DNS knowing what we do now,” on the background of the properties of an ideal naming service described in [I-D.trammell-inip-pins].

Status of This Memo

This Internet-Draft is submitted in full conformance with the provisions of BCP 78 and BCP 79.

Internet-Drafts are working documents of the Internet Engineering Task Force (IETF). Note that other groups may also distribute working documents as Internet-Drafts. The list of current Internet-Drafts is at http://datatracker.ietf.org/drafts/current/.

Internet-Drafts are draft documents valid for a maximum of six months and may be updated, replaced, or obsoleted by other documents at any time. It is inappropriate to use Internet-Drafts as reference material or to cite them other than as "work in progress."

This Internet-Draft will expire on April 20, 2017.

Copyright Notice

Copyright (c) 2016 IETF Trust and the persons identified as the document authors. All rights reserved.

This document is subject to BCP 78 and the IETF Trust's Legal Provisions Relating to IETF Documents (http://trustee.ietf.org/license-info) in effect on the date of publication of this document. Please review these documents carefully, as they describe your rights and restrictions with respect to this document. Code Components extracted from this document must include Simplified BSD License text as described in Section 4.e of the Trust Legal Provisions and are provided without warranty as described in the Simplified BSD License.

Table of Contents

• 1. Introduction
• 2. Terminology
• 3. Architecture
• 4. Information Model
• 4.1. Assertion
• 4.1.1. Context in Assertions
• 4.1.2. Signatures in Assertions
• 4.1.3. Shards and Zones
• 4.2. Query
• 4.2.1. Context in Queries
• 4.3. Answer
• 5. Data Model
• 5.1. Symbol Table
• 5.2. Message
• 5.3. Message Section header
• 5.4. Assertion body
• 5.5. Shard body
• 5.6. Zone Message Section body
• 5.7. Query Message Section body
• 5.8. Notification Message Section body
• 5.9. Object
• 5.10. Signatures, delegation keys, and RAINS infrastructure keys
• 5.10.1. ECDSA signature and public key format
• 5.10.2. Hash-chain based revocation
• 5.11. Certificate information format
• 5.12. Name expression format
• 5.13. Capabilities
• 6. RAINS Protocol Definition
• 6.1. Message processing
• 6.2. Message Transmission
• 6.3. Message Limits
• 6.4. Runtime consistency checking
• 6.5. Integrity and Confidentiality Protection
• 7. RAINS Client Protocol
• 8. Deployment Considerations
• 8.1. Assertion lifetime management
• 8.2. Secret Key Management
• 8.3. Query Service Discovery
• 8.4. Transition using translation between RAINS and DNS information models
• 9. Experimental Design and Evaluation
• 10. IANA Considerations
• 11. Security Considerations
• 12. Acknowledgments
• 13. References
• 13.1. Normative References
• 13.2. Informative References
• Appendix A. Open Issues
• Author's Address

1. Introduction

This document defines an experimental protocol for providing Internet name resolution services, as a replacement for DNS, called RAINS (RAINS, Another Internet Naming Service). It is designed as a prototype to facilitate conversation about the evolution or replacement of the Domain Name System protocol, and was developed as a name resolution system for the SCION (“Scalability, Control, and Isolation on Next-Generation Networks”) future Internet architecture [SCION]. It attempts to answer the question: “how would we design the DNS knowing what we do now,” on the background of the properties of an ideal naming service described in [I-D.trammell-inip-pins].

Its architecture (Section 3) and information model ({{information- model}}) are largely compatible with the existing Domain Name System. However, it does take several radical departures from DNS as presently defined and implemented:

• Delegation from a superordinate zone to a subordinate zone is done solely with cryptography: a superordinate defines the key(s) that are valid for signing assertions in the subordinate during a particular time interval. Assertions about names can therefore safely be served from any infrastructure.
• All time references in RAINS are absolute: instead of a time to live, each assertion’s temporal validity is defined by the temporal validity of the signature(s) on it.
• All assertions have validity within a specific context. A context determines the rules for chaining signatures to verify validity of an assertion. The global context is a special case of context, which uses chains from the global naming root key. The use of context explicitly separates global usage of the DNS from local usage thereof, and allows other application-specific naming constraints to be bound to names; see Section 4.1.1. Queries are valid in one or more contexts, with specific rules for determining which assertions answer which queries; see {{context-in- queries}}.
• There is an explicit separation between registrant-level names and sub- registrant-level names, and explicit information about registrars and registrants available in the naming system at runtime.
• Sets of valid characters and rules for valid names are defined on a per-zone basis, and can be verified at runtime.

Instead of using a custom binary framing as DNS, RAINS uses Concise Binary Object Representation [RFC7049], partially in an effort to make implementations easier to verify and less likely to contain potentially dangerous parser bugs [PARSER-BUGS]. Like DNS, CBOR messages can be carried atop any number of substrate protocols; RAINS is presently defined to use TLS over persistent TCP connections (see Section 6).

2. Terminology

The terms MUST, MUST NOT, SHOULD, SHOULD NOT, and MAY, when they appear in all-capitals, are to be interpreted as defined in [RFC2119].

In addition, the following terms are used in this document as defined:

• Authority: An entity which may make assertions about names in a zone, by virtue of holding a secret key which can generate signatures verifiable using a public key associated with a delegation to the zone.
• Assertion: A mapping between a name and object(s) of specified types describing the name, signed by an authority for the zone containing the subject name. See Section 4.1.
• Subject: The name to which an assertion pertains.
• Object: A type/value pair of information about a name within an assertion.
• Query: An expression of interest in certain types of objects pertaining to a subject name in one or more contexts. See Section 4.2.
• Context: Additional information about the scope in which an assertion or query is valid. See Section 4.1.1 and Section 4.2.1.
• Shard: A group of assertions common to a zone, with common signatures, which may be lexicographically complete for purposes of proving nonexistence of an assertion. See Section 4.1.3.
• Zone: A group of all assertions valid at a given point in time, with common signatures, for a given level of delegation and context within the namespace. See Section 4.1.3.
• RAINS Message: Unit of exchange in the RAINS protocol, containing assertions, shards, zones, queries, and notifications. See Section 5.2.
• Notification: A RAINS-internal message section carrying information about the operation of the protocol itself. See Section 5.8.
• Authority Service: A service provided by a RAINS Server for publishing assertions by an authority. See Section 3.
• Query Service: A service provided by a RAINS Server for answering queries on behalf of a RAINS Client. See Section 3.
• Intermediary Service: A service provided by a RAINS Server for answering queries and providing temporary storage for assertions on behalf of other RAINS Servers. See Section 3.
• RAINS Server: A server that speaks the RAINS Protocol, and provides on or more services on behalf of other RAINS Servers and/or RAINS Clients. See Section 3.
• RAINS Client: A client that uses the Query Service of one or more RAINS Servers to retrieve assertions on behalf of applications that wish to connect to named services in the Internet.

3. Architecture

The RAINS architecture is simple, and resembles the architecture of DNS. A RAINS Server is an entity that provides transient and/or permanent storage for assertions about names, and a lookup function that finds assertions for a given query about a name, either by searching local storage or by delegating to another RAINS server. RAINS servers can take on any or all of three roles:

• authority service, acting on behalf of an authority to ensure properly signed assertions are made available to the system (equivalent to an authoritative server in DNS);
• query service, acting on behalf of a client to answer queries with relevant assertions (equivalent to a recursive resolver in DNS), and to validate assertions on the client’s behalf; and/or
• intermediary service, acting on behalf of neither but providing storage and lookup for assertions with certain properties for query and authority servers (partially replacing, but not really equivalent to, caching resolvers in DNS).

RAINS Servers use the RAINS Protocol defined in this document to exchange queries and assertions. RAINS Clients use a subset variant of the RAINS Protocol (called the RAINS Client Protocol) to interact with RAINS Servers providing query services on their behalf.

4. Information Model

Messages in the RAINS Protocol are made up of two kinds of elements: Assertion and Query. A third type of element, Answer, binds a Query to a set of Assertions in response to a Query.

The information model in this section omits information elements required by the resolution mechanism itself; these are defined in more detail in Section 5 and Section 6.

4.1. Assertion

An Assertion is a signed statement about a mapping from a subject name to an object value, and consists of the following elements:

• Context: name of the context in which the assertion is valid; see Section 4.1.1 below.
• Subject: name about which the assertion is made.
• Zone: name of the zone in which the assertion is made. The fully qualified name of the subject is made by appending the zone name to the subject name with a domain name separator (‘.’).
• Type: the type of information about the Subject contained in the assertion. Each Assertion is about a single type of data.
• Object: the data of the indicated type associated with the Subject
• Signatures: one or more signatures generated by the authority for the Assertion. Signatures contain a time interval during which they are considered valid, and may contain a revocation token allowing them to be revoked before the end of the time interval. See Section 4.1.2 below.

The Types supported for each assertion are:

• Delegation: the authority associated with the zone identified by the name (roughly equivalent to the DNSSEC DS RRTYPE). The Object contains a public key by which the authority can be identified.
• Redirection: The name(s) of one or more a RAINS servers providing authority service for the authority associated with the zone (roughly equivalent to the DNSSEC NS RRTYPE, but not always consulted directly during resolution). The Object contains a set of names.
• Address: one or more addresses associated with the name (replaces DNS A and AAAA RTYPEs). The Object contains a set of Addresses. An Address is an {address-family, value} tuple.
• Service-Info: one or more layer 4 ports and hostnames associated with a service name (replaces DNS SRV RRTYPE). The object contains a {hostname, port-number, priority tuple}.
• Name: one or more names associated with the name (roughly equivalent to DNS CNAME). The Object contains a set of names.
• Certificate: a certificate which must appear at a specified location in the certificate chain presented on a connection attempt with the named entity (roughly equivalent to DNS TLSA). The details of this type will be described in a separate document.
• Zone-Nameset: an expression of the set of names allowed within a zone; e.g. Unicode scripts or codepages in which names in the zone may be issued. This allows a zone to set policy on names in support of the distinguishability property in [I-D.trammell-inip-pins] that can be checked by RAINS servers at runtime. An assertion about a Subject within a Zone whose name is not allowed by a valid signed Zone-Nameset expression is taken to be invalid, even if it has a valid signature. The details of this type will be described in a separate document.
• Registrar: Information about the organization that caused a Subject name to exist, for registrant-level names.
• Registrant: Information about the organization responsible for a Subject name, for registrant-level names.

For a given {subject, type} tuple, multiple assertions can be valid at a given point in time; the union of the object values of all of these assertions is considered to be the set of valid values at that point in time.

4.1.1. Context in Assertions

Assertion contexts are used to determine the validity of the signature by the declared authority as follows:

• The global context is identified by the special context name `.’. Assertions in the global context are signed by the authority for the subject name. For example, assertions about the name simplon.inf.ethz.ch in the global context are only valid if signed by the relevant authority inf.ethz.ch.
• A local context is associated with a given authority. The authority-part and the context-part of a local context name are divided by a context marker (‘cx–’). The authority-part directly identifies the authority whose key was used to sign the assertion; assertions within a local context are only valid if signed by the identified authority. Authorities have complete control over how the contexts under their namespaces are arranged, and over the names within those contexts.

Assertion context is the mechanism by which RAINS provides explicit inconsistency (see section 5.3.2 of [I-D.trammell-inip-pins]). Some examples illustrate how context works:

• For the common split-DNS case, an enterprise could place names for machines on its local networks within a separate context. E.g., a workstation could be named simplon.cab.inf.ethz.ch within the context staff-workstations.cx–.inf.ethz.ch. Assertions about this name would be signed by the authority for inf.ethz.ch. Here, the context serves simply as a marker, without enabling an alternate signature chain: note that the name simplon.cab.inf.ethz.ch can be validly signed by the authority for inf.ethz.ch if no delegation exists for cab.inf.ethz.ch. The context simply marks this assertion as internal. This allows a client making requests of local names to know they are local, and for local resolvers to manage visibility of assertions outside the enterprise: explicit context makes accidental leakage of both queries and assertions easier to detect and avoid.
• Contexts make captive-portal interactions more explicit: a captive portal resolver could respond to a query for a common website (e.g. www.google.ch) with a signed response directed at the captive portal, but within a context identifying the location as well as the ISP (e.g. sihlquai.zurich.ch.cx–.starbucks.access.some-isp.net.). This response will be signed by the authority for starbucks.access.some-isp.net. This signature achieves two things: first, the client knows the result for www.google.ch is not globally valid; second, it can present the user with some indication as to the identity of the captive portal it is connected to.

Further examples showing how context can be used in queries as well are given in Section 4.2.1 below.

Developing conventions for assertion contexts for different situations will require implementation and deployment experience, and is a subject for future work.

4.1.2. Signatures in Assertions

A signature over an assertion contains the following information elements:

• Algorithm: identifier of the algorithm used to generate the signature.
• Valid-Since: a timestamp of the start of validity of this signature.
• Valid-Until: a timestamp of the end of validity of this signature.
• Signature: the cryptographic signature itself, whose format is determined by the algorithm used.
• Revocation-Token: an optional revocation token for this signature, which allows a signature to be replaced or removed before the end of its validity. Revocation tokens are generally based on hash chains, meaning that a signature with a revocation token “down” the chain from a given token supercedes it. The format and mechanism used by the revocation token is determined by the alogrithm used.

The signature protects all the information in an assertion as well as its own valid-since and valid-until values and the revocation token; it does not protect other signatures on the assertion.

4.1.3. Shards and Zones

Assertions may also be grouped and signed as a group. A shard is a set of assertions subject to the same authority within the same context, protected by one or more signatures over all assertions within the shard. A shard may have an additional property that given a subject and an authenticated shard, it can be shown that either an assertion with a given name and type exists within the shard or does not exist at all.

A shard has the following information elements:

• Context: name of the context in which the assertions in the shard are valid; see Section 4.1.1 above.
• Zone: name of the zone in which the assertions are made.
• Content: a set of assertions sharing the context and zone.
• Signatures: one or more signatures generated by the authority for the shard; see Section 4.1.2.
• Complete-Flag: if true, the shard is lexicographically complete, and subject names that sort such that they would be within the shard if they existed, but are not in the shard, can be assumed not to exist.

For efficiency’s sake, information elements within a shard common to all assertions (zone, context, signature) within the shard may be omitted from the assertions themselves.

A zone is the entire set of shards subject to a given authority within a given context. There are three kinds of zones; treating these zones differently may allow lookup protocol optimizations:

• Zones containing only delegation assertions are delegation-only zones. Delegation-only zones are not relevant as part of an assertion lookup, other than for discovering and verifying authority. Top-level domains are generally delegation-only.
• Zones containing no delegation assertions are final zones. Final zones are not relevant as part of an authority discovery.
• Zones containing at least one delegation assertion and at least one assertion that is not a delegation assertion are mixed zones. No optimizations are available for mixed zones.

A zone has the following information elements:

• Context: name of the context in which the assertions in the zone are valid; see Section 4.1.1 above.
• Zone: name of the zone.
• Content: a set of assertions and/or shards sharing the context and zone.
• Signatures: one or more signatures generated by the authority for the shard; see Section 4.1.2.
• Kind: delegation-only, final, or mixed; see above.

4.2. Query

A query is a request for a set of assertions supporting a conclusion about a given subject-object mapping. It consists of the following information elements:

• Contexts: an expression of the context(s) in which assertions answering the query will be accepted; see Section 4.2.1 below.
• Qualified-Subject: the name about which the query is made. The subject name in a query must be fully-qualified.
• Types: a set of assertion types the querier is interested in.
• Valid-Until: an optional client-generated timestamp for the query after which it expires and should not be answered.
• Token: an optional client-generated token for the query, which can be used in the answer to refer to the query (instead of the answer containing the query itself).

A query expresses interest about all the given types of assertion in all the specified contexts; more complex expressions of which types in which contexts must be asked using multiple queries. Preferences for tradeoffs (freshness, bandwidth efficiency, latency, privacy preservation) in servicing a query may be bound to the query using query options.

4.2.1. Context in Queries

Contexts are used in queries as they are in assertions (see {{context-in- assertions}}). Assertion contexts in an answer to a query have to match some context in the query in order to respond to a query. However, there are a few additional considerations. An assertion can only exist with a specific context, while queries may accept answers in multiple contexts. The Contexts part of a query is a sequence of context specifiers taken to be in order of decreasing priority. A special null context (represented by the empty string) indicates that assertions in any context will be accepted. Any context in the Contexts part of a query may additionally be negated, in order to note that assertions in those contexts are not acceptable. Negated context name appearing in the Contexts part of a query before the null context expresses “any context except these”.

Query contexts can also be used to provide additional information to RAINS servers about the query. For example, contexts can provide a method for explicit selection of a CDN servers not based on either the client’s or the resolver’s address (see [RFC7871]). Here, the CDN creates a context for each of its content zones, and an external service selects appropriate contexts for the client based not just on client source address but passive and active measurement of performance. Queries for names at which content resides can then be made within these contexts, with the priority order of the contexts reflecting the goodness of the zone for the client. Here, a context might be zrh.cx–.cdn-zones.some-cdn.com for names of servers hosting content in a CDN’s Zurich data center, and a client could represent its desire to find content nearby by making queries in the zrh.cx–, fra.cx– (Frankfurt), and ams.cx– (Amsterdam) contexts within cdn-zones .some-cdn.com. In all cases, the assertions themselves will be signed by the authority for cdn-zones.some-cdn.com, accurately representing that it is the CDN, not the owner of the related name in the global context, that is making the assertion.

As with assertion contexts, developing conventions for query contexts for different situations will require implementation and deployment experience, and is a subject for future work.

4.3. Answer

An answer consists of a set of assertions, shards, and/or zones which respond to a query, bound to that query. It consists of the following information elements:

• Query: the query this answer applies to. If the query was issued with a token, the query in the answer may omit all content except the token.
• Content: a set of assertions and/or shards answering the query.

The content of an answer content depends on whether the answer is positive or negative. A positive answer contains the information requested in the smallest atomic container that can be found, usually a single assertion. A negative answer contains the information used to verify it; either a shard with the Complete-Flag set, an entire Zone, or a Zone-Nameset assertion showing the name is illegal within the zone.

A query is taken to have an inconclusive answer when no answer returns to the querier before the query’s Valid-Until time.

5. Data Model

The RAINS data model is a relatively straightforward mapping of the information model in Section 4 to the Concise Binary Object Representation (CBOR) [RFC7049], with an outer message type providing a mechanism for future capabilities-based versioning and recognition of a message as a RAINS message.

5.1. Symbol Table

Each CBOR object in a RAINS message is implemented as maps of integer keys to values, which implements a good tradeoff between efficiency of representation and flexibility. The meaning of each of the integer keys is given in the symbol table below:

5.2. Message

All interactions in RAINS take place in an outer envelope called a Message, which is a CBOR map tagged with the RAINS Message tag (hex 0xE99BA8, decimal 15309736).

A Message map MUST contain a content (0) key, whose value is an array of Message Sections; a Message Section is either an Assertion, Shard, Zone, or Query.

A Message map MAY contain a capabilities (1) key, whose value is described in {#cbor-capabilities}.

A Message map MAY contain a signatures (2) key, whose value is an array of Signatures over the entire message as defined in Section 5.10, to be verified against the

A Message map MAY contain a token (8) key, whose value is either an integer or a UTF-8 string of maximum byte length 32. The token key may be used to refer to the message in future messages, or may refer to a past message or query by token. If the message is in response to a message or query containing a token, the message MUST contain that token.

5.3. Message Section header

Each Message Section in the Message’s content value MUST be a two-element array. The first element in the array is the message section type, encoded as an integer as in Section 5.1. The second element in the array is the message section body, defined as in Section 5.4, Section 5.5, Section 5.6, Section 5.7, or Section 5.8.

Section types are as in the following table, taken from Section 5.1:

5.4. Assertion body

An Assertion body is a map. The keys present in this map depend on whether the Assertion is contained in a Message Section or in a Shard or Zone.

Assertions contained in Message Sections are “bare Assertions”. Since they cannot inherit any values from their containers, they MUST contain the signatures (2), subject-name (3), subject-zone (4), context (6), and objects (7) keys.

Assertions within a Shard or Zone are “contained Assertions”, and can inherit values from their containers. A contained Assertion MAY contain the signatures (2) key and MUST contain the subject-name (3) and objects (7) keys. It MAY contain subject-zone (4) and context (6) keys, but in this case the values of these keys MUST be identical to the values in the containing Shard or Zone.

The value of the signatures (2) key, if present, is an array of one or more Signatures as defined in Section 5.10. If not present, the containing Shard or Zone MUST be signed. Signatures on a contained Assertion are generated as if the inherited values are present in the Assertion, whether actually present or not.

The value of the subject-name (3) key is a UTF-8 encoded [RFC3629] string containing the name of the subject of the assertion. The subject name never contains the zone in which the subject name; the fully-qualified name is obtained by joining the subject-name to the subject-zone with a ‘.’ character. The subject-name must be valid according to the nameset expression for the zone, if any.

The value of the subject-zone (4) key, if present, is a UTF-8 encoded string containing the name of the zone in which the assertion is made. If not present, the zone of the assertion is inherited from the containing Shard or Zone.

The value of the context (6) key, if present, is a UTF-8 encoded string containing the name of the context in which the assertion is valid. If not present, the context of the assertion is inherited from the containing Shard or Zone.

The value of the objects (7) key is an array of objects, as defined in {{cbor- object}}.

5.5. Shard body

A Shard body is a map. The keys present in the map depend on whether the Shard is contained in a Message Section or in a Zone.

Shards contained in Message Sections are “bare Shards”. Since they cannot inherit any values from their contained Zone, they MUST contain the content (0), signatures (2), subject-zone (4), and context (6) keys.

Shards within a Zone are “contained Shards”, and can inherit values from their containing Zone. A contained Shard MUST contain the content (0) key, and MAY contain the signatures (2) key and shard-range (11) keys. It MAY contain subject-zone (4) and context (6) keys, but in this case the values of these keys MUST be identical to the values in the containing Zone.

The value of the content (0) key is an array of Assertion bodies as defined in {#cbor-assertion}.

The value of the signatures (2) key, if present, is an array of one or more Signatures as defined in Section 5.10. If not present, the containing Zone MUST be signed. Signatures on a contained Shard are generated as if the inherited values are present in the Shard, whether actually present or not.

The value of the subject-zone (4) key, if present, is a UTF-8 encoded string containing the name of the zone in which the Assertions within the Shard is made. If not present, the zone of the assertion is inherited from the containing Zone.

The value of the context (6) key, if present, is a UTF-8 encoded string containing the name of the context in which the Assertions within the Shard are valid. If not present, the context of the assertion is inherited from the containing Zone.

If the shard-range (11) key is present, the shard is lexicographically complete within the range described in its value: a mapping for a (subject- name, object-type) pair that should be between the two values given in the range but is not is asserted to not exist. Lexicographic sorting is done on subject names by ordering Unicode codepoints in ascending order; ordering on object types is done via their code values in Section 5.9 in ascending order

The shard-range value MUST be a four element array of (subject-name A, object- type A, subject-name B, object type B) where A does not necessarily need to sort before B, and the (subject-name, object-type) pairs need not exist in the shard. The shard MUST NOT contain any assertions for subject-names outside the range.

If the shard-range key is not present, the shard is not lexicographically complete and MUST NOT be used to make assertions about nonexistance.

5.6. Zone Message Section body

A Zone body is a map. Zones MUST contain the content (0), signatures (2), subject-zone (4), and context (6) keys.

The value of the content (0) key is an array of Shard bodies as defined in {#cbor-shard} and/or Assertion bodies as defined in {#cbor-assertion}.

The value of the subject-zone (4) key is a UTF-8 encoded string containing the name of the Zone.

The value of the context (6) key is a UTF-8 encoded string containing the name of the context for which the Zone is valid.

5.7. Query Message Section body

A Query body is a map. Queries MUST contain the query-name (5), query-contexts (13), and query-type (14) keys. Queries MAY contain the token(8) key and the query-opts (22) key.

The value of the query-name (5) key is a UTF-8 encoded string containing the fully qualified name that is the subject of the query.

The value of the query-contexts (13) key is an allowable context expression, as an array of context names as UTF-8 encoded strings. The allowable context expression is evaluated in-order, as follows:

• Context names appearing earlier in the expression are given priority over context names appearing later in the expression.
• A context name may be negated by prepending the context negation marker ‘cx–0-.’ to the context name; a negated context name means the named context is not acceptable in answers to this query.
• The special context name ‘.’ refers to the global context.
• The special context name ‘cx–any-‘ means ‘any context is acceptable’.

Some examples:

• [‘cx–.inf.ethz.ch’, ‘cx–any-‘] means that answers in the ‘cx–.inf.ethz.ch’ context are preferred, but any context is acceptable;
• [’.’, ‘cx–.inf.ethz.ch’] means that only answers in the ‘cx–.inf.ethz.ch’ or global contexts are acceptable, with the global context preferred;
• [’.’, cx–0-.cx–.inf.ethz.ch’, ‘cx–any-‘] means that answers in any context except ‘cx–.inf.ethz.ch’ are acceptable, with the global context preferred.

An empty context array in a query is taken to be equivalent to an array containing only [’.’, ‘cx–any-‘]; i.e. any context acceptable, global context preferred.

The value of the query-type (14) key is an array of integers encoding the type(s) of objects (as in Section 5.9) acceptable in answers to the query. All values in the query-type array are treated at equal priority: [2,3] means the querier is equally interested in both IPv4 and IPv6 addresses for the query-name.

The value of the token (8) key, if present, is either an integer or a UTF-8 string of maximum byte length 32. Future messages or notifications containing answers to this query MUST contain this token, if present.

The value of the query-opts (22) key, if present, is an array of integers in priority order of the querier’s preferences in tradeoffs in answering the query.

Each server is free to determine how to minimize each performance metric requested; however, servers MUST NOT generate queries to other servers if “no information leakage” is specified, and servers MUST NOT return expired assertions unless “expired assertions acceptable” is specified.

5.8. Notification Message Section body

Notification Message Sections contain information about the operation of the RAINS protocol itself. A Notification Message Section body is a map which MUST contain the note-type (17) key and MAY contain the token (8) and note-data (23) keys. The value of the note-type key is encoded as an integer as in the following table:

Note that the status codes are chosen to be mnemonically similar to status codes for HTTP [RFC7231]. Details of the meaning of each status code are given in Section 6.

The value of the token (8) key, if present, is either an integer or a UTF-8 string of maximum byte length 32. If the notification is in response to a message or query containing a token, the notification MUST contain that token.

The value of the note-data (23) key, if present, is a UTF-8 encoded string with additional information about the notification, intended to be displayed to an administrator to help debug the issue identified by the negotiation.

5.9. Object

Objects are encoded as arrays in CBOR, where the first element is the type of the object, encoded as an integer in the following table:

A name (1) object contains a name associated with a name as an alias. It is represented as a two-element array. The second element is a fully-qualified name as a UTF-8 encoded string.

An ip6-addr (2) object contains an IPv6 address associated with a name. It is represented as a two element array. The second element is a byte array of length 16 containing an IPv6 address in network byte order.

An ip4-addr (3) object contains an IPv4 address associated with a name. It is represented as a two element array. The second element is a byte array of length 4 containing an IPv4 address in network byte order.

A redirection (4) object contains the fully-qualified name of a RAINS authority server for a named zone. It is represented as a two-element array. The second element is a fully-qualified name of an RAINS authority server as a UTF-8 encoded string.

A delegation (5) object contains the public key used to generate signatures on assertions in a named zone, and by which a delegation of a name within a zone to a subordinate zone may be verified. It is represented as an N-element array. The second element is a signature algorithm identifier as in Section 5.10. Additional elements are as defined in Section 5.10 for the given algorithm identifier.

A nameset (6) object contains an expression defining which names are allowed and which names are disallowed in a given zone. It is represented as an N- element array, as defined in Section 5.12

A cert-info (7) object contains an expression binding a certificate or certificate authority to a name, such that connections to the name must either use the bound certificate or a certificate signed by a bound authority. It is represented as an N-element array, as defined in Section 5.11.

A service-info (8) object gives information about a named service. Services are named as in [RFC2782]. It is represented as a four-element array. The second element is a fully-qualified name of a host providing the named service as a UTF-8 string. The third element is a transport port number as a positive integer in the range 0-65535. The fourth element is a priority as a positive integer, with lower numbers having higher priority.

A registrar (9) object gives the name and other identifying information of the registrar (the organization which caused the name to be added to the namespace) for organization-level names. It is represented as a UTF-8 string containing identifying information chosen by the registrar according to the registry’s policy.

A registrant (10) object gives information about the registrant of an organization-level name. It is represented as a UTF-8 string containing this information, with a format chosen by the registrar according to the registry’s policy.

An infrakey (11) object contains the public key used to generate signatures on messages by a named RAINS server, by which a RAINS message signature may be verified by a receiver. It is identical in structure to a delegation object, as defined in Section 5.10.

5.10. Signatures, delegation keys, and RAINS infrastructure keys

RAINS supports multiple signature algorithms and hash functions for signing assertions for cryptographic algorithm agility [RFC7696]. A RAINS signature algorithm identifier specifies the signature algorithm; a hash function for generating the HMAC; a method for generating, verifying, and interpreting hash chain tokens in signatures; and the format of the encodings of the signature values in Assertions, Shards, Zones, and Messages, as well as of public key values in delegation objects.

RAINS signatures have four common elements: the algorithm identifier, a valid- since timestamp, a valid-until timestamp, and a hash chain token. Signatures are represented as an array of these four values followed by additional elements containing the signature data itself, according to the algorithm identifier.

Valid-since and valid-until timestamps are represented as CBOR integers counting seconds since the UNIX epoch UTC, identified with tag value 1 and encoded as in section 2.4.1 of [RFC7049]. A signature MUST have a valid- until timestamp. If a signature has no specified valid-since time (i.e., is valid from the beginning of time until its valid-until timestamp), the valid- since time MAY be null (as in Table 2 in Section 2.3 of [RFC7049]).

Hash chain tokens and their use are specified in the appropriate subsection of this section for the given algorithm identifier.

Signatures in RAINS are generated over a normalized serialized CBOR object (a Message; or an Assertion, Shard, or Zone section body). To normalize and serialize an object for sigining or verifying:

• Remove all signatures from the object.
• Add a new “blank” signature to the object, containing Null in the place of any elements containing signature data.
• Serialize the object, emitting all keys in CBOR maps in ascending numerical order. Note that when serializing anything with a Content array, the order of the content array is preserved. If the serialized object is a Message, it should be tagged with the RAINS tag.
• Generate a signature on the resulting byte stream according to the algorithm selected, and fill the resulting values into the “blank” signature; or verify the signature of the resulting byte stream according to the algorithm selected.

The following algorithms are supported:

5.10.1. ECDSA signature and public key format

ECDSA public keys consist of a single value, called “Q” in [FIPS-186-3]. Q is a simple bit string that represents the uncompressed form of a curve point, concatenated together as “x | y”. The third element in a RAINS delegation object is the Q bit string encoded as a CBOR byte array. RAINS delegation objects for ECDSA-256 public keys are therefore represented as the array [5, 2, Q]; and for ECDSA-384 public keys as [5, 3, Q].

ECDSA signatures are a combination of two non-negative integers, called “r” and “s” in [FIPS-186-3]. A Signature using ECDSA is represented using a six element CBOR array, with the fifth element being r represented as a byte array as described in Section C.2 of [FIPS-186-3], and the sixth being s represented as a byte array as described in Section C.2 of [FIPS-186-3]. For ECDSA-256 signatures, each integer MUST be represented as a 32-byte array. For ECDSA-384 signatures, each integer MUST be represented as a 48-byte array. RAINS signatures using ECDSA-256 are therefore the array [2, valid-from, valid-until, token, r, s]; and for ECDSA-384 the array [3, valid-from, valid-until, token, r, s].

ECDSA-256 signatures and public keys use the P-256 curve as defined in [FIPS-186-3]. ECDSA-384 signatures and public keys use the P-384 curve as defined in [FIPS-186-3].

All RAINS servers MUST implement ECDSA-256 and ECDSA-384.

5.10.2. Hash-chain based revocation

Hash-chain based revocation allows a signature (and the Assertion, Shard, or Zone it protects) to be replaced before it expires. To use hash-chain based revocation, a signing entity generates a hash chain from a known seed using the hash function specified by the signature algorithm in use, and places the Nth value derived therefrom in the hash chain revocation token on a signature.

A revocation can be issued by generating a new section and signing it, revealing the N-1st value from the hash chain in the revocation token. To allow a recipient of a revoked section to verify the revocation, the following restrictions on what can replace what apply:

• An Assertion can only be replaced by another Assertion with the same Subject within the same Context and Zone, containing an Objects array of the same length containing the same types of Objects. To delete Object values, those values can be replaced with Null in the replacing Assertion.
• A Shard can only be replaced by another Shard with an identical shard-range key, within the same Context and Zone. Incomplete Shards cannot be replaced.
• A Zone can only be replaced by another Zone with an identical name within the same Context.

Signing entities can decline to use hash-chain based revocation by replacing the revocation token with Null.

5.11. Certificate information format

[EDITOR’S NOTE: Not yet defined. Write me. This should be largely in line with TLSA; determine if there’s any guidance from implementation experience we should consider, as well.]

5.12. Name expression format

[EDITOR’S NOTE: Not yet defined. Write me. For discussion within the INIP. The expression language is primarily intended to solve the same problem the IDNA tables to – acceptable character sets and character set mixing – although a general-purpose expression language that allows arbitrary matching of whole names would be more flexible.]

5.13. Capabilities

When a RAINS server or client sends the first message in a stream to a peer, it MAY expose optional capabilities to its peer using the capabilities (1) key. This key contains either:

• an array of uniform resource names specifying capabilities supported by the sending server, taken from the table below, with each name encoded as a UTF-8 string.
• a SHA-256 hash of the CBOR byte stream derived from normalizing such an array by sorting it in lexicographically increasing order, then serializing it.

This mechanism is inspired by [XEP0115], and is intended to be used to reduce the overhead in exposing common sets of capabilities. Each RAINS server can cache a set of recently-seen or common hashes, and only request the full URN set (using notification code 399) on a cache miss.

The following URNs are presently defined; other URNs will specify future optional features, support for alternate transport protocols and new signature algorithms, etc.

Since there are only two defined capabilities at this time, RAINS servers can be implemented with two hard-coded hashes to determine whether a peer is listening or not. The hash presented by a server supporting urn:x-rains:tlssrv is e5365a09be554ae55b855f15264dbc837b04f5831daeb321359e18cdabab5745; the hash presented by a server supporting no capabilities is 76be8b528d0075f7aae98d6fa57a6d3c83ae480a8469e668d7b0af968995ac71.

A RAINS server MUST NOT assume that a peer server supports a given capability unless it has received a message containing that capability from that server. An exception are the capabilities indicating that a server listens for connections using a given transport protocol; servers and clients can also learn this information from RAINS itself (given a redirection assertion for a named zone) or from external configuration values.

6. RAINS Protocol Definition

As noted in Section 5, RAINS is a message-exchange protocol that uses CBOR [RFC7049] as its framing. Since CBOR is self-framing – a CBOR parser can determine when a CBOR object is complete at the point at which it has read its final byte – RAINS requires no external framing. It can therefore run over any streaming, multistreaming, or message-oriented transport protocol. In order to protect query confidentiality, and support rapid deployment over a ubiquitously implemented transport, RAINS is defined in this document to run over persistent TLS 1.2 connections [RFC5246] over TCP [RFC0793] with mutual authentication. The TLS certificates of RAINS server peers can be verified as specified in the certificate assertions for those servers.

RAINS servers MUST support this transport; future transports can be negotiated using the capabilities mechanism after bootstrapping using TLS 1.2. As RAINS is an experimental protocol, RAINS servers listen on port 1022 [RFC4727] for connections from other RAINS servers and clients. RAINS servers should strive to keep connections open to peer servers, unless it is clear that no future messages will be exchanged with those peers, or in the face of resource limitations at either peer. If a RAINS server needs to send a message to another RAINS server to which it does not have an open connection, it attempts to open a connection with that server.

This section describes the operation of the protocol as used among RAINS servers. A simplified version of the protocol for client access is described in Section 7.

6.1. Message processing

Once a transport is established, any server may validly send a message with any content to any other server. Upon receipt of a message, a server parses it, and:

• notes the token on the message. If present, the token MUST be present on any messages sent in reply to this message.
• processes any capabilities present, replacing the set of capabilities known for the peer with the set present in the message. If the present capabilities are represented by a hash that the server does not have in its cache, it prepares a notification of type 399 “Capability hash not understood” to send to its peer.
• splits the contents into its constituent message sections, processing them independently.

On receipt of an assertion, shard, or zone message section, the server:

• verifies its consistency (see Section 6.4). If the section is not consistent, it prepares to send a notification of type 403 Inconsistent Message to the peer, and discards the section. Otherwise, it:
• determines whether it answers an outstanding query; if so, it prepares to forward the section to the server that issued the query.
• determines whether it replaces any information presently in its cache (see Section 5.10.2). If so, it discards the old information, and caches the new section.
• determines whether it is likely to answer a future query, according to its configuration, policy, and query history; if so, it caches the section.

On receipt of a query, the server:

• determines whether it has a stored assertion, shard, and/or zone message section which answers the query. If so, it prepares to return the most specific such section with the signature of the longest remaining validity to the peer that issued the query. If not, it:
• checks to see whether the query specifies option 4 (cached answers only). If so, and if option 5 (expired assertions acceptable) is also specified, it then checks to see if it has any cached sections that answer the query on which signatures are expired; otherwise, processing stops. If the query does not specify option 4, delegation proceeds as follows:
• determines whether it has other non-authoritative servers it can forward the query to, according to its configuration and policy, and in compliance with any query options (see Section 5.7). If so, it prepares to forward the query to those servers, noting the reply for the received query depends on the replies for the forwarded query. If not, it:
• determines the responsible authority servers for the zone containing the query name in the query for contexts requested, and forwards the query to those authority servers, noting the reply for the received query depends on the reply for the forwarded query.

If query delegation fails to return an answer within a configured timeout for a delegated query, the server prepares to send a 504 No assertion available response to the peer from which it received the query.

When a server creates a new query to forward to another server in response to a query it received, it SHOULD NOT use the same token on the delegated query as on the received query, unless option 6 Enable Tracing is present in the received , in which case it MUST use the same token. The Enable Tracing option is designed to allow debugging of query processing across multiple servers, and MUST NOT be enabled by default.

On receipt of a notification, the server’s behavior depends on the notification type:

• For type 100 “Connection Heartbeat”, the server does nothing: these null messages are used to keep long-lived connections open in the presence of network behaviors that may drop state for idle connections.
• For type 399 “Capability hash not understood”, the server prepares to send a full capabilities list on the next message it sends to the peer.
• For type 504 “No assertion available”, the server checks the token on the message, and prepares to forward the assertion to the associated query.
• For type 413 “Message too large” the server notes that large messages may not be sent to a peer and tries again (see Section 6.3), or logs the error along with the note-data content.
• For type 400 “Malformed message”, type 403 “Inconsistent message”, type 500 “Server error”, or type 501 “Server not capable”, the server logs the error along with the note-data content, as these notifications generally represent implementation or configuration error conditions which will require human intervention to mitigate.

The first message a server sends to a peer after a new connection is established SHOULD contain a capabilities section, if the server supports any optional capabilities. See Section 5.13.

If the server is configured to keep long-running connections open, due to the presence of network behaviors that may drop state for idle connections, it SHOULD send a message containing a type 100 Connection Heartbeat notification after a configured idle time without any messages containing other content being sent.

6.2. Message Transmission

As noted in Section 6.1 many messages are sent in reply to messages received from peers. Servers may also originate messages on their own, based on their configuration and policy:

• Proactive queries to retrieve assertions, shards, and zones for which all signatures have expired or will soon expire, for cache management purposes.
• Proactive push of assertions, shards, and zones to other servers, based on query history or other information indicating those servers may query for the assertions they contain.

6.3. Message Limits

RAINS servers MUST accept messages up to 65536 bytes in length, but MAY accept messages of greater length, subject to resource limitations of the server. A server with resource limitations MUST respond to a message rejected due to length restrictions with a notification of type 414 (Message Too Large). A server that receives a type 413 notification must note that the peer sending the message only accepts messages smaller than the largest message it’s successfully sent that peer, or cap messages to that peer to 65536 bytes in length.

Since a bare assertion with a single ECDSA signature requires on the order of 180 bytes, it is clear that many full zones won’t fit into a single minimum maximum-size message. Authorities are therefore encouraged to publish zones grouped into shards that will fit into 65536-byte messages, to allow servers to reply using these shards when full-zone transfers are not possible due to message size limitations.

6.4. Runtime consistency checking

The data model used by the RAINS protocol allows inconsistent information to be asserted, all resulting from misconfigured or misbehaving authority servers. For example, a shard valid at a given point in time can be marked lexically complete, but not contain an assertion within its lexical range, which is also valid at that point. This would allow both proof of the presence and absence of an assertion at the same time, which is clearly nonsensical.

RAINS relies on runtime consistency checking to mitigate this problem: each server receiving an assertion, shard, or zone SHOULD, subject to resource constraints, ensure that it is consistent with other information it has, and if not, discard all assertions, shards, and zones in its cache, log the error, and send a 403 Inconsistent Message to the source of the message.

6.5. Integrity and Confidentiality Protection

Assertions are not valid unless they contain at least one signature that can be verified from the chain of authorities specified by the name and context on the assertion; integrity protection is built into the information model. The infrastructure key object type allows keys to be associated with RAINS servers in addition to zone authorities, which allows a client to delegate integrity verification of assertions to a trusted query service (see Section 7).

Since the job of an Internet naming service is to provide publicly-available information mapping names to information needed to connect to the services they name, confidentiality protection for assertions is not a goal of the system. Specifically, the information model and the mechanism for proving non- existence of an assertion is not designed to provide resistance against zone enumeration.

On the other hand, confidentiality protection of query information in crucial. Linking naming queries to a specific user can be nearly as useful to build a profile of that user for surveillance purposes as full access to the clear text of that client’s communications [RFC7624]. In this revision, RAINS uses TLS to protect communications between servers and between servers and clients, with certificate information for RAINS infrastructure stored in RAINS itself. Together with hop-by-hop confidentiality protection, query options, proactive caching, default use of non-persistent tokens, and redirection among servers can be used to mix queries and reduce the linkability of query information to specific clients.

7. RAINS Client Protocol

The protocol used by clients to issue queries to and receive responses from an query service is a subset of the full RAINS protocol, with the following differences:

• Clients only process assertion, shard, zone, and notification sections; sending a query to a client results in a 400 Malformed Message notification.
• Clients never listen for connections; a client must initiate and maintain a transport session to the query server(s) it uses for name resolution.
• Servers only process query and notification sections when connected to clients; a client sending assertions to a server results in a 400 Malformed Message notification.

Since signature verification is resource-intensive, clients delegate signature verification to query servers by default. The query server signs the message containing results for a query using its own key (published as an infrakey object associated with the query server’s name), and a validity time corresponding to the signature it verified with the longest lifetime, stripping other signatures from the reply. This behavior can be disabled by a client by specifying query option 7, allowing the client to do its own verification.

8. Deployment Considerations

The following subsections discuss issues that must be considered in any deployment of RAINS at scale.

8.1. Assertion lifetime management

An assertion can contain multiple signatures, each with a different lifetime. Signature lifetimes are equivalent to a time to live in the present DNS: authorities should compute a new signature for each validity period, and make these new signatures available when old ones are expiring.

If an unexpected change to an assertion is necessary, the hash chain based replacement mechanism described in Section 5.10.2 provides a way for an authority to replace signed assertions with new information or with Null objects, in the case of deletion.

Since assertion lifetime management is based on a real-time clock expressed in UTC, RAINS servers MUST use a clock synchronization protocol such as NTP [RFC5905].

8.2. Secret Key Management

The secret keys associated with public keys for each RAINS server (via infrakey objects) must be available on that server, whether through a hardware or software security device, so they can sign messages on demand; this is particularly important for query servers. In addition, the secret keys associated with TLS certificates for each server (published via certinfo objects) must be available as well in order to establish TLS sessions.

However, storing zone secret keys (associated via delegation objects) on RAINS servers would represent a more serious operational risk. To keep this from being necessary, authority servers have an additional signer interface, from which they will accept and cache any assertion, shard, or zone for which they are authority servers until at least the end of validity of the last signature, provided the signature is verifiable.

8.3. Query Service Discovery

A client that will not do its own verification must be able to discover the oracle server(s) it should trust for resolution. Integration with e.g. DHCP or selection of a local multicast discovery method are left to a future revision of this document.

In any case, clients MUST provide a configuration interface to allow a user to specify (by address or name) and/or constrain (by certificate property) a preferred/trusted oracle. This would allow client on an untrusted network to use an untrusted locally-available oracle to discover a preferred oracle (doing key verification on its own for bootstrapping), before connecting to that oracle for normal name resolution.

8.4. Transition using translation between RAINS and DNS information models

Full adoption of RAINS would require changes to every client device (replacing DNS stub resolvers with RAINS clients) and name server on the Internet. In addition, most client software would need to change, as well, to get the full benefits of explicit context in name resolution. This is a wholly unrealistic goal.

RAINS servers can, however, coexist with Domain Name System servers and clients during an indefinite transition period. RAINS assertions can be algorithmically translated into DNS answers, and RAINS queries can be algorithmically translated into DNS queries, by RAINS to DNS gateways, given the mostly compatible information models used by the two.

While DNSSEC and RAINS keys for equivalent ciphersuites are compatible with each other, there is no equivalent to query option 7 for gateways, since the RAINS signatures are generated over the RAINS bytestream for an assertion, not the DNS bytestream. Therefore, RAINS to DNS gateways must provide verification services for DNS clients. DNS over TLS [RFC7858] SHOULD be used between the DNS client and gateway to ensure confidentiality and integrity for queries and answers.

Object type mappings are as follows:

• Objects of type name can (largely) be represented as CNAME RRs.
• Objects of type ip6-addr can be represented as AAAA RRs.
• Objects of type ip4-addr can be represented as A RRs.
• Objects of type redirection can be represented as NS RRs.
• Objects of type cert-info can be represented as TLSA RRs
• Objects of type service-info can be represented as SRV RRs.

There are a few object types without mappings:

• Objects of type delegation can be represented as DS RRs, and signatures as RRSIG RRs, but since these keys are verified by the gateway, there is no need to represent this information to the client.
• Objects of type infrakey cannot be represented in DNS, but are irrelevant for DNS translation of RAINS messages, since DNS does not support server signing of responses.
• Objects of type registrar and registrant cannot be represented in DNS; clients can use WHOIS instead. In addition, RRTYPEs could be added for them in the future if RAINS sees significant deployment with DNS as a front-end protocol.
• Objects of type nameset cannot be represented in DNS; the current equivalent are the IDNA parameters maintained by IANA (for the DNS root zone only) at https://www.iana.org/assignments/idna-tables-6.3.0/idna-tables-6.3.0.xhtml.

When translating a DNS query from a client to a RAINS query for that client, client options can be set on a per-server, per-client, or per-query basis using some out of band configuration options.

When translating a RAINS assertion to a DNS answer, the gateway can use the time to expiry for the verified signature as the TTL.

There is no method for exposing context information in a DNS query or answer. Therefore, queries and answers at a RAINS gateway are only supported for the global context “.”.

9. Experimental Design and Evaluation

The protocol described in this document is intended primarily as a prototype for discussion, though the goal of the document is to specify RAINS completely enough to allow independent, interoperable implementation of clients an servers. The massive inertia behind the deployment of the present domain name system makes full deployment as a replacement for DNS unlikely. Despite this, there are some criteria by which the success of the RAINS experiment may be judged:

First, deployment in simulated or closed networks, or in alternate Internet architectures such as SCION, allows implementation experience with the features of RAINS which DNS lacks (signatures as a first-order delegation primitive, support for explicit contexts, explicit tradeoffs in queries, runtime availability of registrar/registrant data, and nameset support), which in turn may inform the specification and deployment of these features on the present DNS.

Second, deployment of RAINS “islands” in the present Internet alongside DNS on a per-domain basis would allow for comparison between operational and implementation complexity and efficiency and benefits derived from RAINS’ features, as information for future development of the DNS protocol.

10. IANA Considerations

The present revision of this document has no actions for IANA.

The authors have registered the CBOR tag 15309736 to identify RAINS messages in the CBOR tag registry at https://www.iana.org/assignments/cbor-tags/cbor-tags.xhtml.

RAINS servers currently listen for connections from other servers on Port 1022. Future revisions of this document may specify a different port, registered with IANA via Expert Review [RFC5226].

The symbol table in this document in Section 5.1, the notification code table in Section 5.8, and the signature algorithm table in Section 5.10 may be candidates for IANA registries in future revisions of this document.

The urn:x-rains namespace used by the RAINS capability mechanism in {{cbor- capabilities}} may be a candidate for replacement with an IANA-registered namespace in a future revision of this document.

11. Security Considerations

This document specifies a new, experimental protocol for Internet name resolution, with mandatory integrity protection for assertions about names built into the information model, and confidentiality for query information protected on a hop-by-hop basis. See especially Section 4.1.2, Section 6.5, Section 5.10, Section 5.11, and Section 8.2 for security-relevant details.

12. Acknowledgments

Thanks to Daniele Asoni, Laurent Chuat, Ted Hardie, Joe Hildebrand, Steve Matsumoto, Adrian Perrig, Raphael Reischuk, Stephen Shirley, Andrew Sullivan, and Suzanne Woolf for the discussions leading to the design of this protocol.

13. References

13.1. Normative References

13.2. Informative References

Appendix A. Open Issues

• The format of the nameset and certinfo object types needs to be specified.
• A method for clients to discover local oracles needs to be specified.

Author's Address