Skip to main content

1. Introduction

1. Introduction​

1.1. Motivation​

The visible face of the Internet largely consists of services that employ a client-server architecture in which an interactive or automated client communicates with an application service in order to retrieve or upload information, communicate with other entities, or access a broader network of services. When a client communicates with an application service using Transport Layer Security [TLS] or Datagram Transport Layer Security [DTLS], it references some notion of the server's identity (e.g., "the website at example.com") while attempting to establish secure communication. Likewise, during TLS negotiation, the server presents its notion of the service's identity in the form of a public-key certificate that was issued by a certification authority (CA) in the context of the Internet Public Key Infrastructure using X.509 [PKIX]. Informally, we can think of these identities as the client's "reference identity" and the server's "presented identity" (these rough ideas are defined more precisely later in this document through the concept of particular identifiers). In general, a client needs to verify that the server's presented identity matches its reference identity so it can authenticate the communication.

Many application technologies adhere to the pattern just outlined. Such protocols have traditionally specified their own rules for representing and verifying application service identity. Unfortunately, this divergence of approaches has caused some confusion among certification authorities, application developers, and protocol designers.

Therefore, to codify secure procedures for the implementation and deployment of PKIX-based authentication, this document specifies recommended procedures for representing and verifying application service identity in certificates intended for use in application protocols employing TLS.

1.2. Audience​

The primary audience for this document consists of application protocol designers, who can reference this document instead of defining their own rules for the representation and verification of application service identity. Secondarily, the audience consists of certification authorities, service providers, and client developers from technology communities that might reuse the recommendations in this document when defining certificate issuance policies, generating certificate signing requests, or writing software algorithms for identity matching.

1.3. How to Read This Document​

This document is longer than the authors would have liked because it was necessary to carefully define terminology, explain the underlying concepts, define the scope, and specify recommended behavior for both certification authorities and application software implementations. The following sections are of special interest to various audiences:

  • Protocol designers might want to first read the checklist in Section 3.

  • Certification authorities might want to first read the recommendations for representation of server identity in Section 4.

  • Service providers might want to first read the recommendations for requesting of server certificates in Section 5.

  • Software implementers might want to first read the recommendations for verification of server identity in Section 6.

The sections on terminology (Section 1.8), naming of application services (Section 2), document scope (Section 1.7), and the like provide useful background information regarding the recommendations and guidelines that are contained in the above-referenced sections, but are not absolutely necessary for a first reading of this document.

1.4. Applicability​

This document does not supersede the rules for certificate issuance or validation provided in [PKIX]. Therefore, [PKIX] is authoritative on any point that might also be discussed in this document. Furthermore, [PKIX] also governs any certificate-related topic on which this document is silent, including but not limited to certificate syntax, certificate extensions such as name constraints and extended key usage, and handling of certification paths.

This document addresses only name forms in the leaf "end entity" server certificate, not any name forms in the chain of certificates used to validate the server certificate. Therefore, in order to ensure proper authentication, application clients need to verify the entire certification path per [PKIX].

This document also does not supersede the rules for verifying service identity provided in specifications for existing application protocols published prior to this document, such as those excerpted under Appendix B. However, the procedures described here can be referenced by future specifications, including updates to specifications for existing application protocols if the relevant technology communities agree to do so.

1.5. Overview of Recommendations​

To orient the reader, this section provides an informational overview of the recommendations contained in this document.

For the primary audience of application protocol designers, this document provides recommended procedures for the representation and verification of application service identity within PKIX certificates used in the context of TLS.

For the secondary audiences, in essence this document encourages certification authorities, application service providers, and application client developers to coalesce on the following practices:

  • Move away from including and checking strings that look like domain names in the subject's Common Name.

  • Move toward including and checking DNS domain names via the subjectAlternativeName extension designed for that purpose: dNSName.

  • Move toward including and checking even more specific subjectAlternativeName extensions where appropriate for using the protocol (e.g., uniformResourceIdentifier and the otherName form SRVName).

  • Move away from the issuance of so-called wildcard certificates (e.g., a certificate containing an identifier for "*.example.com").

These suggestions are not entirely consistent with all practices that are currently followed by certification authorities, client developers, and service providers. However, they reflect the best aspects of current practices and are expected to become more widely adopted in the coming years.

1.6. Generalization from Current Technologies​

This document attempts to generalize best practices from the many application technologies that currently use PKIX certificates with TLS. Such technologies include, but are not limited to:

  • The Internet Message Access Protocol [IMAP] and the Post Office Protocol [POP3]; see also [USINGTLS]

  • The Hypertext Transfer Protocol [HTTP]; see also [HTTP-TLS]

  • The Lightweight Directory Access Protocol [LDAP]; see also [LDAP-AUTH] and its predecessor [LDAP-TLS]

  • The Simple Mail Transfer Protocol [SMTP]; see also [SMTP-AUTH] and [SMTP-TLS]

  • The Extensible Messaging and Presence Protocol [XMPP]; see also [XMPP-OLD]

  • The Network News Transfer Protocol [NNTP]; see also [NNTP-TLS]

  • The NETCONF Configuration Protocol [NETCONF]; see also [NETCONF-SSH] and [NETCONF-TLS]

  • The Syslog Protocol [SYSLOG]; see also [SYSLOG-TLS] and [SYSLOG-DTLS]

  • The Session Initiation Protocol [SIP]; see also [SIP-CERTS]

  • The Simple Network Management Protocol [SNMP]; see also [SNMP-TLS]

  • The General Internet Signalling Transport [GIST]

However, as noted, this document does not supersede the rules for verifying service identity provided in specifications for those application protocols.

1.7. Scope​

1.7.1. In Scope​

This document applies only to service identities associated with fully qualified DNS domain names, only to TLS and DTLS (or the older Secure Sockets Layer (SSL) technology), and only to PKIX-based systems. As a result, the scenarios described in the following section are out of scope for this specification (although they might be addressed by future specifications).

1.7.2. Out of Scope​

The following topics are out of scope for this specification:

  • Client or end-user identities.

    Certificates representing client or end-user identities (e.g., the rfc822Name identifier) can be used for mutual authentication between a client and server or between two clients, thus enabling stronger client-server security or end-to-end security. However, certification authorities, application developers, and service operators have less experience with client certificates than with server certificates, thus giving us fewer models from which to generalize and a less solid basis for defining best practices.

  • Identifiers other than fully qualified DNS domain names.

    Some certification authorities issue server certificates based on IP addresses, but preliminary evidence indicates that such certificates are a very small percentage (less than 1%) of issued certificates. Furthermore, IP addresses are not necessarily reliable identifiers for application services because of the existence of private internets [PRIVATE], host mobility, multiple interfaces on a given host, Network Address Translators (NATs) resulting in different addresses for a host from different locations on the network, the practice of grouping many hosts together behind a single IP address, etc. Most fundamentally, most users find DNS domain names much easier to work with than IP addresses, which is why the domain name system was designed in the first place. We prefer to define best practices for the much more common use case and not to complicate the rules in this specification.

    Furthermore, we focus here on application service identities, not specific resources located at such services. Therefore this document discusses Uniform Resource Identifiers [URI] only as a way to communicate a DNS domain name (via the URI "host" component or its equivalent), not as a way to communicate other aspects of a service such as a specific resource (via the URI "path" component) or parameters (via the URI "query" component).

    We also do not discuss attributes unrelated to DNS domain names, such as those defined in [X.520] and other such specifications (e.g., organizational attributes, geographical attributes, company logos, and the like).

  • Security protocols other than [TLS], [DTLS], or the older Secure Sockets Layer (SSL) technology.

    Although other secure, lower-layer protocols exist and even employ PKIX certificates at times (e.g., IPsec [IPSEC]), their use cases can differ from those of TLS-based and DTLS-based application technologies. Furthermore, application technologies have less experience with IPsec than with TLS, thus making it more difficult to gather feedback on proposed best practices.

  • Keys or certificates employed outside the context of PKIX-based systems.

    Some deployed application technologies use a web of trust model based on or similar to OpenPGP [OPENPGP], or use self-signed certificates, or are deployed on networks that are not directly connected to the public Internet and therefore cannot depend on Certificate Revocation Lists (CRLs) or the Online Certificate Status Protocol [OCSP] to check CA-issued certificates. However, the method for binding a public key to an identifier in OpenPGP differs essentially from the method in X.509, the data in self-signed certificates has not been certified by a third party in any way, and checking of CA-issued certificates via CRLs or OCSP is critically important to maintaining the security of PKIX-based systems. Attempting to define best practices for such technologies would unduly complicate the rules defined in this specification.

  • Certification authority policies, such as:

    • What types or "classes" of certificates to issue and whether to apply different policies for them (e.g., allow the wildcard character in certificates issued to individuals who have provided proof of identity but do not allow the wildcard character in "Extended Validation" certificates [EV-CERTS]).

    • Whether to issue certificates based on IP addresses (or some other form, such as relative domain names) in addition to fully qualified DNS domain names.

    • Which identifiers to include (e.g., whether to include SRV-IDs or URI-IDs as defined in the body of this specification).

    • How to certify or validate fully qualified DNS domain names and application service types.

    • How to certify or validate other kinds of information that might be included in a certificate (e.g., organization name).

  • Resolution of DNS domain names.

    Although the process whereby a client resolves the DNS domain name of an application service can involve several steps (e.g., this is true of resolutions that depend on DNS SRV resource records, Naming Authority Pointer (NAPTR) DNS resource records [NAPTR], and related technologies such as [S-NAPTR]), for our purposes we care only about the fact that the client needs to verify the identity of the entity with which it communicates as a result of the resolution process. Thus the resolution process itself is out of scope for this specification.

  • User interface issues.

    In general, such issues are properly the responsibility of client software developers and standards development organizations dedicated to particular application technologies (see, for example, [WSC-UI]).

1.8. Terminology​

Because many concepts related to "identity" are often too vague to be actionable in application protocols, we define a set of more concrete terms for use in this specification.

  • application service: A service on the Internet that enables interactive and automated clients to connect for the purpose of retrieving or uploading information, communicating with other entities, or connecting to a broader network of services.

  • application service provider: An organization or individual that hosts or deploys an application service.

  • application service type: A formal identifier for the application protocol used to provide a particular kind of application service at a domain; the application service type typically takes the form of a Uniform Resource Identifier scheme [URI] or a DNS SRV Service [DNS-SRV].

  • attribute-type-and-value pair: A colloquial name for the ASN.1-based construction comprising a Relative Distinguished Name (RDN), which itself is a building-block component of Distinguished Names. See Section 2 of [LDAP-DN].

  • automated client: A software agent or device that is not directly controlled by a human user.

  • delegated domain: A domain name or host name that is explicitly configured for communicating with the source domain, by either (a) the human user controlling an interactive client or (b) a trusted administrator. In case (a), one example of delegation is an account setup that specifies the domain name of a particular host to be used for retrieving information or connecting to a network, which might be different from the server portion of the user's account name (e.g., a server at mailhost.example.com for connecting to an IMAP server hosting an email address of [email protected]). In case (b), one example of delegation is an admin-configured host-to-address/address-to-host lookup table.

  • derived domain: A domain name or host name that a client has derived from the source domain in an automated fashion (e.g., by means of a [DNS-SRV] lookup).

  • identifier: A particular instance of an identifier type that is either presented by a server in a certificate or referenced by a client for matching purposes.

  • identifier type: A formally defined category of identifier that can be included in a certificate and therefore that can also be used for matching purposes. For conciseness and convenience, we define the following identifier types of interest, which are based on those found in the PKIX specification [PKIX] and various PKIX extensions.

    • CN-ID = a Relative Distinguished Name (RDN) in the certificate subject field that contains one and only one attribute-type-and-value pair of type Common Name (CN), where the value matches the overall form of a domain name (informally, dot-separated letter-digit-hyphen labels); see [PKIX] and also [LDAP-SCHEMA]

    • DNS-ID = a subjectAltName entry of type dNSName; see [PKIX]

    • SRV-ID = a subjectAltName entry of type otherName whose name form is SRVName; see [SRVNAME]

    • URI-ID = a subjectAltName entry of type uniformResourceIdentifier whose value includes both (i) a "scheme" and (ii) a "host" component (or its equivalent) that matches the "reg-name" rule (where the quoted terms represent the associated [ABNF] productions from [URI]); see [PKIX] and [URI]

  • interactive client: A software agent or device that is directly controlled by a human user. (Other specifications related to security and application protocols, such as [WSC-UI], often refer to this entity as a "user agent".)

  • pinning: The act of establishing a cached name association between the application service's certificate and one of the client's reference identifiers, despite the fact that none of the presented identifiers matches the given reference identifier. Pinning is accomplished by allowing a human user to positively accept the mismatch during an attempt to communicate with the application service. Once a cached name association is established, the certificate is said to be pinned to the reference identifier and in future communication attempts the client simply verifies that the service's presented certificate matches the pinned certificate, as described under Section 6.6.2. (A similar definition of "pinning" is provided in [WSC-UI].)

  • PKIX: PKIX is a short name for the Internet Public Key Infrastructure using X.509 defined in RFC 5280 [PKIX], which comprises a profile of the X.509v3 certificate specifications and X.509v2 certificate revocation list (CRL) specifications for use in the Internet.

  • PKIX-based system: A software implementation or deployed service that makes use of X.509v3 certificates and X.509v2 certificate revocation lists (CRLs).

  • PKIX certificate: An X.509v3 certificate generated and employed in the context of PKIX.

  • presented identifier: An identifier that is presented by a server to a client within a PKIX certificate when the client attempts to establish secure communication with the server; the certificate can include one or more presented identifiers of different types, and if the server hosts more than one domain then the certificate might present distinct identifiers for each domain.

  • reference identifier: An identifier, constructed from a source domain and optionally an application service type, used by the client for matching purposes when examining presented identifiers.

  • source domain: The fully qualified DNS domain name that a client expects an application service to present in the certificate (e.g., "www.example.com"), typically input by a human user, configured into a client, or provided by reference such as in a hyperlink. The combination of a source domain and, optionally, an application service type enables a client to construct one or more reference identifiers.

  • subjectAltName entry: An identifier placed in a subjectAltName extension.

  • subjectAltName extension: A standard PKIX certificate extension [PKIX] enabling identifiers of various types to be bound to the certificate subject -- in addition to, or in place of, identifiers that may be embedded within or provided as a certificate's subject field.

  • subject field: The subject field of a PKIX certificate identifies the entity associated with the public key stored in the subject public key field (see Section 4.1.2.6 of [PKIX]).

  • subject name: In an overall sense, a subject's name(s) can be represented by or in the subject field, the subjectAltName extension, or both (see [PKIX] for details). More specifically, the term often refers to the name of a PKIX certificate's subject, encoded as the X.501 type Name and conveyed in a certificate's subject field (see Section 4.1.2.6 of [PKIX]).

  • TLS client: An entity that assumes the role of a client in a Transport Layer Security [TLS] negotiation. In this specification we generally assume that the TLS client is an (interactive or automated) application client; however, in application protocols that enable server-to-server communication, the TLS client could be a peer application service.

  • TLS server: An entity that assumes the role of a server in a Transport Layer Security [TLS] negotiation; in this specification we assume that the TLS server is an application service.

Most security-related terms in this document are to be understood in the sense defined in [SECTERMS]; such terms include, but are not limited to, "attack", "authentication", "authorization", "certification authority", "certification path", "certificate", "credential", "identity", "self-signed certificate", "trust", "trust anchor", "trust chain", "validate", and "verify".

The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", "SHOULD", "SHOULD NOT", "RECOMMENDED", "NOT RECOMMENDED", "MAY", and "OPTIONAL" in this document are to be interpreted as described in RFC 2119 [KEYWORDS].

Last updated on