RFC 8615






Internet Engineering Task Force (IETF)                     M. Nottingham
Request for Comments: 8615                                      May 2019
Obsoletes: 5785
Updates: 7230, 7595
Category: Standards Track
ISSN: 2070-1721


             Well-Known Uniform Resource Identifiers (URIs)

Abstract



   This memo defines a path prefix for "well-known locations",
   "/.well-known/", in selected Uniform Resource Identifier (URI)
   schemes.

   In doing so, it obsoletes RFC 5785 and updates the URI schemes
   defined in RFC 7230 to reserve that space.  It also updates RFC 7595
   to track URI schemes that support well-known URIs in their registry.

Status of This Memo



   This is an Internet Standards Track document.

   This document is a product of the Internet Engineering Task Force
   (IETF).  It represents the consensus of the IETF community.  It has
   received public review and has been approved for publication by the
   Internet Engineering Steering Group (IESG).  Further information on
   Internet Standards is available in Section 2 of RFC 7841.

   Information about the current status of this document, any errata,
   and how to provide feedback on it may be obtained at
   https://www.rfc-editor.org/info/rfc8615.

Copyright Notice



   Copyright (c) 2019 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
   (https://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.



Nottingham                   Standards Track                    [Page 1]

RFC 8615                     Well-Known URIs                    May 2019


Table of Contents



   1.  Introduction  . . . . . . . . . . . . . . . . . . . . . . . .   2
   2.  Notational Conventions  . . . . . . . . . . . . . . . . . . .   3
   3.  Well-Known URIs . . . . . . . . . . . . . . . . . . . . . . .   3
     3.1.  Registering Well-Known URIs . . . . . . . . . . . . . . .   4
   4.  Security Considerations . . . . . . . . . . . . . . . . . . .   5
     4.1.  Protecting Well-Known Resources . . . . . . . . . . . . .   6
     4.2.  Interaction with Web Browsing . . . . . . . . . . . . . .   6
     4.3.  Scoping Applications  . . . . . . . . . . . . . . . . . .   7
     4.4.  Hidden Capabilities . . . . . . . . . . . . . . . . . . .   8
   5.  IANA Considerations . . . . . . . . . . . . . . . . . . . . .   8
     5.1.  The Well-Known URI Registry . . . . . . . . . . . . . . .   8
     5.2.  The Uniform Resource Identifier (URI) Schemes Registry  .   9
   6.  References  . . . . . . . . . . . . . . . . . . . . . . . . .   9
     6.1.  Normative References  . . . . . . . . . . . . . . . . . .   9
     6.2.  Informative References  . . . . . . . . . . . . . . . . .  10
   Appendix A.  Frequently Asked Questions . . . . . . . . . . . . .  12
   Appendix B.  Changes from RFC 5785  . . . . . . . . . . . . . . .  12
   Author's Address  . . . . . . . . . . . . . . . . . . . . . . . .  12

1.  Introduction



   Some applications on the Web require the discovery of information
   about an origin [RFC6454] (sometimes called "site-wide metadata")
   before making a request.  For example, the Robots Exclusion Protocol
   (http://www.robotstxt.org) specifies a way for automated processes to
   obtain permission to access resources; likewise, the Platform for
   Privacy Preferences [P3P] tells user agents how to discover privacy
   policy before interacting with an origin server.

   While there are several ways to access per-resource metadata (e.g.,
   HTTP header fields, PROPFIND in Web Distributed Authoring and
   Versioning (WebDAV) [RFC4918]), the perceived overhead (either in
   terms of client-perceived latency and/or deployment difficulties)
   associated with them often precludes their use in these scenarios.

   At the same time, it has become more popular to use HTTP as a
   substrate for non-Web protocols.  Sometimes, such protocols need a
   way to locate one or more resources on a given host.

   When this happens, one solution is to designate a "well-known
   location" for data or services related to the origin overall, so that
   it can be easily located.  However, this approach has the drawback of
   risking collisions, both with other such designated "well-known
   locations" and with resources that the origin has created (or wishes
   to create).  Furthermore, defining well-known locations usurps the
   origin's control over its own URI space [RFC7320].



Nottingham                   Standards Track                    [Page 2]

RFC 8615                     Well-Known URIs                    May 2019


   To address these uses, this memo reserves a path prefix in HTTP,
   HTTPS, WebSocket (WS), and Secure WebSocket (WSS) URIs for these
   "well-known locations", "/.well-known/".  Future specifications that
   need to define a resource for such metadata can register their use to
   avoid collisions and minimise impingement upon origins' URI space.

   Well-known URIs can also be used with other URI schemes, but only
   when those schemes' definitions explicitly allow it.

2.  Notational Conventions



   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
   BCP 14 [RFC2119] [RFC8174] when, and only when, they appear in all
   capitals, as shown here.

3.  Well-Known URIs



   A well-known URI is a URI [RFC3986] whose path component begins with
   the characters "/.well-known/", provided that the scheme is
   explicitly defined to support well-known URIs.

   For example, if an application registers the name 'example', the
   corresponding well-known URI on 'http://www.example.com/' would be
   'http://www.example.com/.well-known/example'.

   This specification updates the "http" [RFC7230] and "https" [RFC7230]
   schemes to support well-known URIs.  Other existing schemes can use
   the appropriate process for updating their definitions; for example,
   [RFC8307] does so for the "ws" and "wss" schemes.  The "Uniform
   Resource Identifier (URI) Schemes" registry tracks which schemes
   support well-known URIs; see Section 5.2.

   Applications that wish to mint new well-known URIs MUST register
   them, following the procedures in Section 5.1, subject to the
   following requirements.

   Registered names MUST conform to the "segment-nz" production in
   [RFC3986].  This means they cannot contain the "/" character.

   Registered names for a specific application SHOULD be correspondingly
   precise; "squatting" on generic terms is not encouraged.  For
   example, if the Example application wants a well-known location for
   metadata, an appropriate registered name might be "example-metadata"
   or even "example.com-metadata", not "metadata".





Nottingham                   Standards Track                    [Page 3]

RFC 8615                     Well-Known URIs                    May 2019


   At a minimum, a registration will reference a specification that
   defines the format and associated media type(s) to be obtained by
   dereferencing the well-known URI, along with the URI scheme(s) that
   the well-known URI can be used with.  If no URI schemes are
   explicitly specified, "http" and "https" are assumed.

   Typically, applications will use the default port for the given
   scheme; if an alternative port is used, it MUST be explicitly
   specified by the application in question.

   Registrations MAY also contain additional information, such as the
   syntax of additional path components, query strings, and/or fragment
   identifiers to be appended to the well-known URI, or protocol-
   specific details (e.g., HTTP [RFC7231] method handling).

   Note that this specification defines neither how to determine the
   hostname to use to find the well-known URI for a particular
   application, nor the scope of the metadata discovered by
   dereferencing the well-known URI; both should be defined by the
   application itself.

   Also, this specification does not define a format or media type for
   the resource located at "/.well-known/", and clients should not
   expect a resource to exist at that location.

   Well-known URIs are rooted in the top of the path's hierarchy; they
   are not well-known by definition in other parts of the path.  For
   example, "/.well-known/example" is a well-known URI, whereas
   "/foo/.well-known/example" is not.

   See also Section 4 for Security Considerations regarding well-known
   locations.

3.1.  Registering Well-Known URIs



   The "Well-Known URIs" registry is located at
   <https://www.iana.org/assignments/well-known-uris/>.  Registration
   requests can be made by following the instructions located there or
   by sending an email to the <wellknown-uri-review@ietf.org> mailing
   list.

   Registration requests consist of at least the following information:

   URI suffix:  The name requested for the well-known URI, relative to
      "/.well-known/"; e.g., "example".






Nottingham                   Standards Track                    [Page 4]

RFC 8615                     Well-Known URIs                    May 2019


   Change controller:  For Standards Track RFCs, state "IETF".  For
      others, give the name of the responsible party.  Other details
      (e.g., email address, home page URI) may also be included.

   Specification document(s):  Reference to the document that specifies
      the field, preferably including a URI that can be used to retrieve
      a copy of the document.  An indication of the relevant sections
      may also be included, but is not required.

   Status:  One of "permanent" or "provisional".  See guidance below.

   Related information:  Optionally, citations to additional documents
      containing further relevant information.

   General requirements for registered values are described in
   Section 3.

   Values defined by Standards Track RFCs and other open standards (in
   the sense of [RFC2026], Section 7.1.1) have a status of "permanent".
   Other values can also be registered as permanent, if the experts find
   that they are in use, in consultation with the community.  Other
   values should be registered as "provisional".

   Provisional entries can be removed by the experts if -- in
   consultation with the community -- the experts find that they are not
   in use.  The experts can change a provisional entry's status to
   permanent; in doing so, the experts should consider how widely used a
   value is and consult the community beforehand.

   Note that "consult the community" above refers to those responsible
   for the URI scheme(s) in question.  Generally, this would take place
   on the mailing list(s) of the appropriate Working Group(s) (possibly
   concluded), or on <art@ietf.org> if no such list exists.

   Well-known URIs can be registered by third parties (including the
   expert(s)), if the expert(s) determine that an unregistered well-
   known URI is widely deployed and not likely to be registered in a
   timely manner otherwise.  Such registrations still are subject to the
   requirements defined, including the need to reference a
   specification.

4.  Security Considerations



   Applications minting new well-known URIs, as well as administrators
   deploying them, will need to consider several security-related
   issues, including (but not limited to) exposure of sensitive data,
   denial-of-service attacks (in addition to normal load issues), server




Nottingham                   Standards Track                    [Page 5]

RFC 8615                     Well-Known URIs                    May 2019


   and client authentication, vulnerability to DNS rebinding attacks,
   and attacks where limited access to a server grants the ability to
   affect how well-known URIs are served.

   [RFC3552] contains some examples of potential security considerations
   that may be relevant to application protocols and administrators
   deploying them.

4.1.  Protecting Well-Known Resources



   Because well-known locations effectively represent the entire origin,
   server operators should appropriately control the ability to write to
   them.  This is especially true when more than one entity is
   co-located on the same origin.  Even for origins that are controlled
   by and represent a single entity, due care should be taken to assure
   that write access to well-known locations is not granted unwittingly,
   either externally through server configuration or locally through
   implementation permissions (e.g., on a filesystem).

4.2.  Interaction with Web Browsing



   Applications using well-known URIs for "http" or "https" URLs need to
   be aware that well-known resources will be accessible to Web
   browsers, and therefore are able to be manipulated by content
   obtained from other parts of that origin.  If an attacker is able to
   inject content (e.g., through a Cross-Site Scripting vulnerability),
   they will be able to make potentially arbitrary requests to the well-
   known resource.

   HTTP and HTTPS also use origins as a security boundary for many other
   mechanisms, including (but not limited to) cookies [RFC6265], Web
   Storage [WEBSTORAGE], and various capabilities.

   An application that defines well-known locations should not assume
   that it has sole access to these mechanisms or that it is the only
   application using the origin.  Depending on the nature of the
   application, mitigations can include:

   o  Encrypting sensitive information

   o  Allowing flexibility in the use of identifiers (e.g., cookie
      names) to avoid collisions with other applications

   o  Using the 'HttpOnly' flag on cookies to assure that cookies are
      not exposed to browser scripting languages [RFC6265]

   o  Using the 'Path' parameter on cookies to assure that they are not
      available to other parts of the origin [RFC6265]



Nottingham                   Standards Track                    [Page 6]

RFC 8615                     Well-Known URIs                    May 2019


   o  Using X-Content-Type-Options: nosniff [FETCH] to assure that
      content under attacker control can't be coaxed into a form that is
      interpreted as active content by a Web browser

   Other good practices include:

   o  Using an application-specific media type in the Content-Type
      header field, and requiring clients to fail if it is not used

   o  Using Content-Security-Policy [CSP] to constrain the capabilities
      of active content (such as HTML [HTML5]), thereby mitigating
      Cross-Site Scripting attacks

   o  Using Referrer-Policy [REFERRER-POLICY] to prevent sensitive data
      in URLs from being leaked in the Referer request header field

   o  Avoiding use of compression on any sensitive information (e.g.,
      authentication tokens, passwords), as the scripting environment
      offered by Web browsers allows an attacker to repeatedly probe the
      compression space; if the attacker has access to the path of the
      communication, they can use this capability to recover that
      information.

4.3.  Scoping Applications



   This memo does not specify the scope of applicability for the
   information obtained from a well-known URI, and does not specify how
   to discover a well-known URI for a particular application.

   Individual applications using this mechanism must define both
   aspects; if this is not specified, security issues can arise from
   implementation deviations and confusion about boundaries between
   applications.

   Applying metadata discovered in a well-known URI to resources other
   than those co-located on the same origin risks administrative as well
   as security issues.  For example, allowing
   "https://example.com/.well-known/example" to apply policy to
   "https://department.example.com", "https://www.example.com", or even
   "https://www.example.com:8000" assumes a relationship between hosts
   where there might be none, thereby giving control to a potential
   attacker.

   Likewise, specifying that a well-known URI on a particular hostname
   is to be used to bootstrap a protocol can cause a large number of
   undesired requests.  For example, if a well-known HTTPS URI is used
   to find policy about a separate service such as email, it can result




Nottingham                   Standards Track                    [Page 7]

RFC 8615                     Well-Known URIs                    May 2019


   in a flood of requests to Web servers, even if they don't implement
   the well-known URI.  Such undesired requests can resemble a denial-
   of-service attack.

4.4.  Hidden Capabilities



   Applications using well-known locations should consider that some
   server administrators might be unaware of their existence (especially
   on operating systems that hide directories whose names begin with
   ".").  This means that if an attacker has write access to the
   .well-known directory, they would be able to control its contents,
   possibly without the administrator realising it.

5.  IANA Considerations



5.1.  The Well-Known URI Registry



   This specification updates the registration procedures for the "Well-
   Known URI" registry, first defined in [RFC5785]; see Section 3.1.

   Well-known URIs are registered on the advice of one or more experts,
   with a Specification Required (using terminology from [RFC8126]).

   The experts' primary considerations in evaluating registration
   requests are:

   o  Conformance to the requirements in Section 3

   o  The availability and stability of the specifying document

   o  The considerations outlined in Section 4

   IANA will direct the senders of any incoming registry requests to
   this document and, if defined, the processes established by the
   expert(s); typically, this will mean referring them to the registry
   Web page.

   Per this document, IANA has:

   o  Updated the registration procedure to Specification Required.

   o  Added a "Status" column to the registry and marked all of the
      existing registrations as "permanent".








Nottingham                   Standards Track                    [Page 8]

RFC 8615                     Well-Known URIs                    May 2019


5.2.  The Uniform Resource Identifier (URI) Schemes Registry



   This specification adds a field to the registration template of the
   "Uniform Resource Identifier (URI) Schemes" registry, with the name
   "Well-Known URI Support" and a default value of "-".

   If a URI scheme explicitly has been specified to use well-known URIs
   as per Section 3, the value changes to a reference to that
   specification.  Initial values not equal to "-" are given in Table 1.

                  +------------+------------------------+
                  | URI Scheme | Well-Known URI Support |
                  +------------+------------------------+
                  | coap       | [RFC7252]              |
                  | coap+tcp   | [RFC8323]              |
                  | coap+ws    | [RFC8323]              |
                  | coaps      | [RFC7252]              |
                  | coaps+tcp  | [RFC8323]              |
                  | coaps+ws   | [RFC8323]              |
                  | http       | [RFC8615]              |
                  | https      | [RFC8615]              |
                  | ws         | [RFC8307]              |
                  | wss        | [RFC8307]              |
                  +------------+------------------------+

       Table 1: Rows in URI Scheme Registry with Nonempty New Column

6.  References



6.1.  Normative References



   [RFC2119]  Bradner, S., "Key words for use in RFCs to Indicate
              Requirement Levels", BCP 14, RFC 2119,
              DOI 10.17487/RFC2119, March 1997,
              <https://www.rfc-editor.org/info/rfc2119>.

   [RFC3986]  Berners-Lee, T., Fielding, R., and L. Masinter, "Uniform
              Resource Identifier (URI): Generic Syntax", STD 66,
              RFC 3986, DOI 10.17487/RFC3986, January 2005,
              <https://www.rfc-editor.org/info/rfc3986>.

   [RFC6454]  Barth, A., "The Web Origin Concept", RFC 6454,
              DOI 10.17487/RFC6454, December 2011,
              <https://www.rfc-editor.org/info/rfc6454>.







Nottingham                   Standards Track                    [Page 9]

RFC 8615                     Well-Known URIs                    May 2019


   [RFC7230]  Fielding, R., Ed. and J. Reschke, Ed., "Hypertext Transfer
              Protocol (HTTP/1.1): Message Syntax and Routing",
              RFC 7230, DOI 10.17487/RFC7230, June 2014,
              <https://www.rfc-editor.org/info/rfc7230>.

   [RFC8126]  Cotton, M., Leiba, B., and T. Narten, "Guidelines for
              Writing an IANA Considerations Section in RFCs", BCP 26,
              RFC 8126, DOI 10.17487/RFC8126, June 2017,
              <https://www.rfc-editor.org/info/rfc8126>.

   [RFC8174]  Leiba, B., "Ambiguity of Uppercase vs Lowercase in RFC
              2119 Key Words", BCP 14, RFC 8174, DOI 10.17487/RFC8174,
              May 2017, <https://www.rfc-editor.org/info/rfc8174>.

6.2.  Informative References



   [CSP]      West, M., "Content Security Policy Level 3", World Wide
              Web Consortium WD WD-CSP3-20160913, September 2016,
              <https://www.w3.org/TR/2016/WD-CSP3-20160913>.

   [FETCH]    WHATWG, "Fetch - Living Standard",
              <https://fetch.spec.whatwg.org>.

   [HTML5]    WHATWG, "HTML - Living Standard",
              <https://html.spec.whatwg.org>.

   [P3P]      Marchiori, M., "The Platform for Privacy Preferences 1.0
              (P3P1.0) Specification", World Wide Web Consortium
              Recommendation REC-P3P-20020416, April 2002,
              <http://www.w3.org/TR/2002/REC-P3P-20020416>.

   [REFERRER-POLICY]
              Eisinger, J. and E. Stark, "Referrer Policy", World Wide
              Web Consortium CR CR-referrer-policy-20170126, January
              2017,
              <https://www.w3.org/TR/2017/CR-referrer-policy-20170126>.

   [RFC2026]  Bradner, S., "The Internet Standards Process -- Revision
              3", BCP 9, RFC 2026, DOI 10.17487/RFC2026, October 1996,
              <https://www.rfc-editor.org/info/rfc2026>.

   [RFC3552]  Rescorla, E. and B. Korver, "Guidelines for Writing RFC
              Text on Security Considerations", BCP 72, RFC 3552,
              DOI 10.17487/RFC3552, July 2003,
              <https://www.rfc-editor.org/info/rfc3552>.






Nottingham                   Standards Track                   [Page 10]

RFC 8615                     Well-Known URIs                    May 2019


   [RFC4918]  Dusseault, L., Ed., "HTTP Extensions for Web Distributed
              Authoring and Versioning (WebDAV)", RFC 4918,
              DOI 10.17487/RFC4918, June 2007,
              <https://www.rfc-editor.org/info/rfc4918>.

   [RFC5785]  Nottingham, M. and E. Hammer-Lahav, "Defining Well-Known
              Uniform Resource Identifiers (URIs)", RFC 5785,
              DOI 10.17487/RFC5785, April 2010,
              <https://www.rfc-editor.org/info/rfc5785>.

   [RFC6265]  Barth, A., "HTTP State Management Mechanism", RFC 6265,
              DOI 10.17487/RFC6265, April 2011,
              <https://www.rfc-editor.org/info/rfc6265>.

   [RFC7231]  Fielding, R., Ed. and J. Reschke, Ed., "Hypertext Transfer
              Protocol (HTTP/1.1): Semantics and Content", RFC 7231,
              DOI 10.17487/RFC7231, June 2014,
              <https://www.rfc-editor.org/info/rfc7231>.

   [RFC7252]  Shelby, Z., Hartke, K., and C. Bormann, "The Constrained
              Application Protocol (CoAP)", RFC 7252,
              DOI 10.17487/RFC7252, June 2014,
              <https://www.rfc-editor.org/info/rfc7252>.

   [RFC7320]  Nottingham, M., "URI Design and Ownership", BCP 190,
              RFC 7320, DOI 10.17487/RFC7320, July 2014,
              <https://www.rfc-editor.org/info/rfc7320>.

   [RFC8307]  Bormann, C., "Well-Known URIs for the WebSocket Protocol",
              RFC 8307, DOI 10.17487/RFC8307, January 2018,
              <https://www.rfc-editor.org/info/rfc8307>.

   [RFC8323]  Bormann, C., Lemay, S., Tschofenig, H., Hartke, K.,
              Silverajan, B., and B. Raymor, Ed., "CoAP (Constrained
              Application Protocol) over TCP, TLS, and WebSockets",
              RFC 8323, DOI 10.17487/RFC8323, February 2018,
              <https://www.rfc-editor.org/info/rfc8323>.

   [WEBSTORAGE]
              Hickson, I., "Web Storage (Second Edition)", World Wide
              Web Consortium Recommendation REC-webstorage-20160419,
              April 2016,
              <http://www.w3.org/TR/2016/REC-webstorage-20160419>.








Nottingham                   Standards Track                   [Page 11]

RFC 8615                     Well-Known URIs                    May 2019


Appendix A.  Frequently Asked Questions



   Aren't well-known locations bad for the Web?
      They are, but for various reasons -- both technical and social --
      they are sometimes necessary.  This memo defines a "sandbox" for
      them, to reduce the risks of collision and to minimise the impact
      upon preexisting URIs on sites.

   Why "/.well-known?"
      It's short, descriptive, and according to search indices, not
      widely used.

   What impact does this have on existing mechanisms, such as P3P and
   robots.txt?
      None, until they choose to use this mechanism.

   Why aren't per-directory well-known locations defined?
      Allowing every URI path segment to have a well-known location
      (e.g., "/images/.well-known/") would increase the risks of
      colliding with a preexisting URI on a site, and generally these
      solutions are found not to scale well because they're too
      "chatty".

Appendix B.  Changes from RFC 5785



   o  Allowed non-Web well-known locations

   o  Adjusted IANA instructions

   o  Updated references

   o  Made various other clarifications

   o  Tracked supporting schemes in the "Uniform Resource Identifier
      (URI) Schemes" registry

Author's Address



   Mark Nottingham

   Email: mnot@mnot.net
   URI:   https://www.mnot.net/









Nottingham                   Standards Track                   [Page 12]