Network Working Group M. Crispin INTERNET-DRAFT: IMAP URLAUTH University of Washington October 2004 Document: internet-drafts/draft-ietf-lemonade-urlauth-03.txt Internet Message Access Protocol (IMAP) - URLAUTH Extension Status of this Memo By submitting this Internet-Draft, I certify that any applicable patent or other IPR claims of which I am aware have been disclosed, or will be disclosed, and any of which I become aware will be disclosed, in accordance with RFC 3668. Internet-Drafts are working documents of the Internet Engineering Task Force (IETF), its areas, and its working groups. Note that other groups may also distribute working documents as Internet-Drafts. 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." The list of current Internet-Drafts can be accessed at http://www.ietf.org/ietf/1id-abstracts.txt To view the list Internet-Draft Shadow Directories, see http://www.ietf.org/shadow.html. A revised version of this document will be submitted to the RFC editor as an Informational Document for the Internet Community. A revised version of this draft document will be submitted to the RFC editor as a Proposed Standard for the Internet Community. Discussion and suggestions for improvement are requested, and should be sent to lemonade@IETF.ORG. This document will expire before 15 April 2005. Distribution of this memo is unlimited. ;;; POINTS FOR DISCUSSION: ;;; ;;; This document was formerly known as draft-crispin-imap-urlauth. The ;;; 00 version of this document is identical to the 09 version of that ;;; document. ;;; ;;; Earlier versions of this document had access identifiers for "any ;;; authorized user" and anonymous. These were not needed for Lemonade ;;; submit-without-download and were removed. Should something of this ;;; nature be put back as extensions? Abstract This document describes the URLAUTH extension to the Internet Message Access Protocol (IMAP) (RFC 3501) and the IMAP URL Scheme (IMAPURL) (RFC 2192). This extension provides a means by which an IMAP client can use URLs carrying authorization to access limited message data on the IMAP server. An IMAP server which supports this extension indicates this with a capability name of "URLAUTH". Conventions Used in this Document The key words "MUST", "MUST NOT", "SHOULD", "SHOULD NOT", and "MAY" in this document are to be interpreted as defined in [KEYWORDS]. The formal syntax uses the Augmented Backus-Naur Form (ABNF) notation including the core rules defined in Appendix A of [ABNF]. In examples, "C:" and "S:" indicate lines sent by the client and server respectively. If a single "C:" or "S:" label applies to multiple lines, then the line breaks between those lines are for editorial clarity only and are not part of the actual protocol exchange. Introduction In [IMAPURL], a URL of the form imap://fred@example.com/INBOX/;uid=20 requires authorization as userid "fred". The URLAUTH extension provides a means by which an authorized user of an IMAP server can create URLAUTH authorized IMAP URLs. A URLAUTH authorized URL conveys authorization (not authentication) to the data addressed by that URL, and can be used in another IMAP session to access specific content on the IMAP server without otherwise providing authorization to any other data owned by the authorizing user (including other data in the mailbox specified in the URL). A URLAUTH authorized URL can be used in the argument to the BURL command in message composition, as described in [BURL], for such purposes as a memory (or other resource) constrainted client submitting a message forward or resend from an IMAP mailbox without requiring the client to fetch that message data. The URLAUTH is generated using an authorization mechanism name and an authorization token, which is generated using a secret mailbox access key. An IMAP client can request the server to generate and assign a new mailbox access key (thus effectively revoking all current URLs using URLAUTH with the old mailbox access key) but can not set the mailbox access key to a key of its own choosing. 1. Concepts 1.1. URLAUTH The URLAUTH is a component, appended at the end of a URL, which conveys authorization to access the data addressed by that URL. It contains an authorized access identifier, an authorization mechanism name, and an authorization token which is generated from the URL, the authorized access identifer, authorization mechanism name, and a mailbox access key. 1.2. Mailbox Access Key The mailbox access key is a random string with at least 128 bits of entropy. It is generated by software (not by the human user), and MUST be unpredictable. Each user has a table of mailboxes and an associated mailbox access key for each mailbox. Consequently, the mailbox access key is per-user and per-mailbox. In other words, two users sharing the same mailbox each have a different mailbox access key for that mailbox; and each mailbox accessed by a single user also has a different mailbox access key. 1.3. Authorized Access Identifier The authorized access identifier restricts use of the URLAUTH authorized URL to certain users authorized on the server, as described in section 2. 1.4. Authorization Mechanism The authorization mechanism is the algorithm by which the URLAUTH is generated and subsequently verified, using the mailbox access key. This specification defines the INTERNAL mechanism, which uses a token generation algorithm of the server's choosing (a modern and reasonably secure [HMAC] such as HMAC-SHA1 is recommended) and does not involve disclosure of the mailbox access key to the client. Note: If, for any reason, it is necesary to change the token generation algorithm of the INTERNAL mechanism (e.g. because an attack against the current algorithm has been discovered), all currently existing URLAUTH authorized URLs are invalidated by the change in algorithm. Since this would be an unpleasant surprise to applications which depend upon the validity of a URLAUTH authorized URL, and there is no good way to do a bulk update of existing deployed URLs, it is best to avoid this situation by using a secure algorithm as opposed to one that is "good enough". Although this specification is extensible for other mechanisms, none are defined in this document. In addition to the mechanism name itself, other mechanisms may have mechanism specific data, which is to be interpreted according to the definition of that mechanism. 1.5. Authorization Token The authorization token is a deterministic string of at least 128 bits which an entity with knowledge of the secret mailbox access key and URL authorization mechanism can use to verify the URL. 2. IMAP URL Extensions [IMAPURL] is extended by allowing the addition of ;EXPIRE=" and ";URLAUTH=::" to IMAP URLs which refer to specific message or message parts. ";URLAUTH=::" (the URLAUTH) MUST be at the end of the URL. URLAUTH does not apply to, and MUST NOT be used with, any IMAP URL which refers to an entire IMAP server, list of mailboxes, an entire IMAP mailbox, or IMAP search results. When ";EXPIRE=" is used, this indicates the latest date and time that the URL is valid. After that date and time, the URL has expired and server implementations MUST reject the URL. If ";EXPIRE=" is not used, the URL has no expiration, but still can be revoked as discussed below. ";URLAUTH=::" indicates the access identifiers which are permitted to use this URL, the authorization mechanism, and the authorization token. The "submit+" access identifier, followed by a userid, indicates that only a userid authorized as a message submission entity on behalf of the specified userid is permitted to use this URL. The IMAP server does not validate the specified userid but does validate that the IMAP session has an authorization identity that is authorized as a message submission entity. The authorized message submission entity MUST validate the userid prior to contacting the IMAP server. The "user+" access identifier, followed by a userid, indicates that use of this URL is limited to IMAP sessions which are logged in as the specified userid (that is, have authorization identity as that userid). Note: if a SASL mechanism which provides both authorization and authentication identifiers is used to authenticate to the IMAP server, the "user+" access identifier must match the authorization identifier. The authorization token is represented as an ASCII-encoded hexadecimal string, which is used to authorize the URL. The length and the calculation of the authorization token depends upon the mechanism used; but in all cases the authorization token is at least 128 bits (and therefore at least 32 hexadecimal digits). 3. Discussion of URLAUTH Authorization Issues In [IMAPURL], the userid before the "@" in the URL has two purposes: 1) It provides context for user-specific mailbox paths such as "INBOX". 2) It specifies that resolution of the URL requires logging in as that user and limits use of that URL to only that user. An obvious limitation of using the same field for both purposes is that the URL can only be resolved by the mailbox owner. URLAUTH overrides the second purpose of the userid in the IMAP URL and by default permits the URL to be resolved by any user permitted by the access identifier. The "user+" access identifier limits resolution of that URL to a particular userid, whereas the "submit+" access identifier is more general and simply requires the session be authorized by a user that has been granted a "submit" role within the authentication system. This makes it impossible for an attacker, spying on the session, to use the same URL, either directly or by submission to a message submission entity. 4. Generation of URLAUTH authorized URLs A URLAUTH authorized URL is generated from an initial URL as follows. An initial URL is built, ending with ";URLAUTH=" but without the "::" components. An authorization mechanism is selected and used to calculate the authorization token, with the initial URL as the data and a secret known to the IMAP server as the key. The URLAUTH authorized URL is generated by taking the initial URL and appending ":", the URL authorization mechanism name, ":", and the ASCII-encoded hexadecimal representation of the authorization token. Note: ASCII-encoded hexadecimal is used instead of BASE64 because a BASE64 representation may have "=" padding characters which would be problematic in a URL. In the INTERNAL mechanism, the mailbox access key for that mailbox is the secret known to the IMAP server, and a server-selected algorithm such as HMAC-SHA1 is used to to calculate the authorization token. 5. Validation of URLAUTH authorized URLs A URLAUTH authorized URL is validated as follows. The URL is split at the ":" which separates "" from ":" in the ";URLAUTH=::" portion of the URL. The ":" portion is first parsed and saved as the authorization mechanism and the authorization token. The URL is truncated, discarding the ":" described above, to create a "rump URL" (the URL minus the ":" and the ":" portion). The rump URL is then analyzed to identify the mailbox. If the mailbox cannot be identified, an authorization token is calculated on the rump URL, using random "plausible" keys (selected by the server) as needed, before returning a validation failure. This prevents timing attacks aimed at identifying mailbox names. If the mailbox can be identified, the authorization token is calculated on the rump URL and a secret known to the IMAP server using the given URL authorization mechanism. Validation is successful if, and only if, the calculated authorization token for that mechanism matches the authorization token supplied in ";URLAUTH=::". Removal of the "::" portion of the URL MUST be the only operation applied to the URLAUTH authorized URL to get the rump URL. In particular, URL percent escape decoding and case-folding MUST NOT occur. In the INTERNAL mechanism, the mailbox access key for that mailbox is used as the secret known to the IMAP server, and the same server-selected algorithm used for generating URLs is used to calculate the authorization token for verification. 6. Additional Commands These commands are extension to the [IMAP] base protocol. The section headings of these commands are intended to correspond with where they would be located in the base protocol document if they were part of that document. BASE.6.3.RESETKEY. RESETKEY Command Arguments: optional mailbox name optional mechanism name(s) Responses: none other than in result Result: OK - RESETKEY completed, URLMECH containing new data NO - RESETKEY error: can't change key of that mailbox BAD - command unknown or arguments invalid The RESETKEY command has two forms. The first form accepts a mailbox name as an argument, and generates a new mailbox access key for the given mailbox in the user's mailbox access key table, replacing any previous mailbox access key (and revoking any URLs that were authorized with a URLAUTH using that key) in that table. By default, the mailbox access key is generated for the INTERNAL mechanism; other mechanisms can be specified with the optional mechanism argument. The second form, with no arguments, removes all mailbox access keys in the user's mailbox access key table, revoking all URLs currently authorized using URLAUTH by the user. Any current IMAP session logged in as the user which has the mailbox selected will receive an untagged OK response with the URLMECH status response code (see below for more details about the URLMECH status response code). Example: C: a31 RESETKEY S: a31 OK All keys removed C: a32 RESETKEY INBOX S: a32 OK [URLMECH INTERNAL] mechs C: a33 RESETKEY INBOX XSAMPLE S: a33 OK [URLMECH INTERNAL XSAMPLE=P34OKhO7VEkCbsiYY8rGEg==] mechs BASE.6.3.GENURLAUTH. GENURLAUTH Command Argument: one or more URL/mechanism pairs Response: untagged response: GENURLAUTH Result: OK - GENURLAUTH completed NO - GENURLAUTH error: can't generate a URLAUTH BAD - command unknown or arguments invalid The GENURLAUTH command requests the server to generate a URLAUTH authorized URL for each of the given URLs using the given URL authorization mechanism. If the command is successful, a GENURLAUTH response code is returned listing the requested URLs as URLAUTH authorized URLs. Example: Note that lines which do not begin with "C:" or "S:" are continuations of the previous line for typographical clarity and are not separate lines in the protocol. C: a777 GENURLAUTH "imap://joe@example.com/INBOX/;uid=20/;section=1.2 ;urlauth=submit+fred" INTERNAL S: * GENURLAUTH "imap://joe@example.com/INBOX/;uid=20/;section=1.2 ;urlauth=submit+fred:internal:91354a473744909de610943775f92038" S: a777 OK GENURLAUTH completed BASE.6.3.URLFETCH. URLFETCH Command Argument: one or more URLs Response: untagged response: URLFETCH Result: OK - urlfetch completed NO - urlfetch failed due to server internal error BAD - command unknown or arguments invalid The URLFETCH command requests that the server return the text data associated with the specified IMAP URLs, as described in [IMAPURL] and extended by this document. The data is returned for all validated URLs, regardless of whether or not the session would otherwise be able to access the mailbox containing that data via SELECT or examine. Note: This command does not require that the URL refer to the selected mailbox; nor does it require that any mailbox be selected. It also does not in any way interfere with any selected mailbox. The URLFETCH command MUST return an untagged URLFETCH response and a tagged OK response to any URLFETCH command that is syntactically valid. A NO response indicates a server internal failure which may be resolved on later retry. Note: the possibility of a NO response is to accommodate implementations which would otherwise have to issue an untagged BYE with a fatal error due to an inability to respond to a valid request. In an ideal world, a server SHOULD NOT issue a NO response. The server MUST return NIL for any IMAP URL which references an entire IMAP server, list of mailboxes, an entire IMAP mailbox, or IMAP search results. Example Note that lines which do not begin with "C:" or "S:" are continuations of the previous line for typographical clarity and are not separate lines in the protocol. For clarity, this example uses the LOGIN command which SHOULD NOT be used over a non-encrypted communication path. This example is of a submit server, obtaining a message segment for a message that it has already validated was submitted by "fred". S: * OK [CAPABILITY IMAP4REV1 URLAUTH] example.com IMAP server ready C: a001 LOGIN submitserver secret S: a001 OK submitserver logged in C: a002 URLFETCH "imap://joe@example.com/INBOX/;uid=20/;section=1.2 ;urlauth=submit+fred:internal: 91354a473744909de610943775f92038" S: * URLFETCH "imap://joe@example.com/INBOX/;uid=20/;section=1.2 ;urlauth=submit+fred:internal:91354a473744909de610943775f92038" {28} S: Si vis pacem, para bellum. S: S: a002 OK URLFETCH completed 7. Additional Responses These responses are extensions to the [IMAP] base protocol. The section headings of these responses are intended to correspond with where they would be located in the base protocol document if they were part of that document. BASE.7.1.URLMECH. URLMECH Status Response Code The URLMECH status response code is followed by a list of URL authorization mechanism names. Mechanism names other than INTERNAL may be appended with an "=" and BASE64 encoded form of mechanism specific data. This status response code is returned in an untagged OK response in response to a RESETKEY, SELECT, or EXAMINE command. Example: C: a33 RESETKEY INBOX XSAMPLE S: a33 OK [URLMECH INTERNAL XSAMPLE=P34OKhO7VEkCbsiYY8rGEg==] mechs In this example, the server supports the INTERNAL mechanism and an experimental mechanism called XSAMPLE which also holds some mechanism specific data (the name "XSAMPLE" is for illustrative purposes only). BASE.7.4.GENURLAUTH. GENURLAUTH Response Contents: One or more URLs The GENURLAUTH response returns the URLAUTH authorized URL(s) requested by a GENURLAUTH command. Example: C: a777 GENURLAUTH "imap://joe@example.com/INBOX/;uid=20/;section=1.2 ;urlauth=submit+fred" INTERNAL S: * GENURLAUTH "imap://joe@example.com/INBOX/;uid=20/;section=1.2 ;urlauth=submit+fred:internal:91354a473744909de610943775f92038" S: a777 OK GENURLAUTH completed BASE.7.4.URLFETCH. URLFETCH Response Contents: URL/nstring pair The URLFETCH response returns the message text data associated with an IMAP URLs, as described in [IMAPURL] and extended by this document. This response occurs as the result of a URLFETCH command. The returned data string is NIL if the URL is invalid for any reason (including validation failure). If the URL is valid, but the IMAP fetch of the body part returned NIL (this should not happen), the returned data string should be the empty string ("") and not NIL. Example: C: a002 URLFETCH "imap://joe@example.com/INBOX/;uid=20/;section=1.2 ;urlauth=submit+fred:internal: 91354a473744909de610943775f92038" S: * URLFETCH "imap://joe@example.com/INBOX/;uid=20/;section=1.2 ;urlauth=submit+fred:internal:91354a473744909de610943775f92038" {28} S: Si vis pacem, para bellum. S: S: a002 OK URLFETCH completed 8. Formal Syntax The following syntax specification uses the Augmented Backus-Naur Form (ABNF) notation as specified in [ABNF]. The following modifications are made to the Formal Syntax in [IMAP]: resetkey = "RESETKEY" [SP mailbox *(SP mechanism)] command-auth =/ resetkey / genurlauth / urlfetch resp-text-code =/ "URLMECH" SP "INTERNAL" *(SP mechanism ["=" base64]) genurlauth = "GENURLAUTH" 1*(SP url SP mechanism) genurlauth-data = "*" SP "GENURLAUTH" 1*(SP url) url = astring ; contains imapurl as defined below urlfetch = "URLFETCH" 1*(SP url) urlfetch-data = "*" SP "URLFETCH" 1*(SP url SP nstring) The following modifications are made to the Formal Syntax in [IMAPURL]: enc-urlauth = 32*HEXDIG imessagepart = enc_mailbox [uidvalidity] iuid [isection] [iurlauth] iurlauth = [expire] ";URLAUTH=" access ":" mechanism ":" enc-urlauth access = ("submit+" iuserauth) / ("user+" iuserauth) expire = ";EXPIRE=" date-time ; date-time defined in [DATETIME] mechanism = "INTERNAL" / 1*uchar ; new mechanisms MUST be registered with IANA 9. Security Considerations Security considerations are discussed throughout this memo. The mailbox access key SHOULD have at least 128 bits of entropy (refer to [RANDOM] for more details) and MUST be unpredictable. The URLMECH status response code may expose sensitive data in the mechanism specific data for mechanisms other than INTERNAL. A server implementation MUST implement a configuration that will not return a URLMECH status response code unless some mechanism is provided that protects the session from snooping, such as a TLS or SASL security layer that provides confidentiality protection. The calculation of a authorization token with a "plausible" key if the mailbox can not be identified is necessary to avoid attacks in which the server is probed to see if a particular mailbox exists on the server by measuring the amount of time taken to reject a known bad name vs. some other name. To protect against a computational denial-of-service attack, a server MAY impose progressively longer delays on multiple URL requests that fail validation. IANA Considerations URLAUTH authorization mechanisms are registered by publishing a standards track or IESG approved experimental RFC. The registry is currently located at: [to be defined by IANA] This document consitutes registration of the INTERNAL URLAUTH authorization mechanism. IMAP URLAUTH Authorization Mechanism Registry Mechanism Name Reference -------------- --------- INTERNAL [this document, to be filled in by IANA] References The following references are normative: [ABNF] Crocker, D. and P. Overell, "Augmented BNF for Syntax Specifications: ABNF", RFC 2234, November 1997. [BURL] Newman, C., "Message Submission BURL Extension", draft-newman-lemonade-burl-00.txt (work in progress), March 2004. [DATETIME] Klyne, G., and Newman, C., "Date and Time on the Internet: Timestamps", RFC 3339, July 2002. [IMAP] Crispin, M., "Internet Message Access Protocol - Version 4rev1", RFC 3501, March 2003. [IMAPURL] Newman, C., "IMAP URL Scheme", RFC 2192, September 1997. [KEYWORDS] Bradner, S., "Key words for use in RFCs to Indicate Requirement Levels", BCP 14, RFC 2119, March 1997. The following references are informative: [HMAC] Krawczyk, H., Bellare, M., and Canetti, R., "HMAC: Keyed-Hashing for Message Authentication", RFC 2104, February 1997. [RANDOM] Eastlake, D., Crocker, S., and Schiller, J., "Randomness Recommendations for Security", RFC 1750, December 1994. Author's Addresses Mark R. Crispin Networks and Distributed Computing University of Washington 4545 15th Avenue NE Seattle, WA 98105-4527 Phone: (206) 543-5762 EMail: MRC@CAC.Washington.EDU Chris Newman Sun Microsystems 1050 Lakes Drive West Covina, CA 91790 EMail: chris.newman@sun.com IPR Disclosure Acknowledgement By submitting this Internet-Draft, I certify that any applicable patent or other IPR claims of which I am aware have been disclosed, and any of which I become aware will be disclosed, in accordance with RFC 3668. Intellectual Property Statement The IETF takes no position regarding the validity or scope of any intellectual property or other rights that might be claimed to pertain to the implementation or use of the technology described in this document or the extent to which any license under such rights might or might not be available; neither does it represent that it has made any effort to identify any such rights. Information on the IETF's procedures with respect to rights in standards-track and standards-related documentation can be found in BCP-11. Copies of claims of rights made available for publication and any assurances of licenses to be made available, or the result of an attempt made to obtain a general license or permission for the use of such proprietary rights by implementors or users of this specification can be obtained from the IETF Secretariat. The IETF invites any interested party to bring to its attention any copyrights, patents or patent applications, or other proprietary rights which may cover technology that may be required to practice this standard. Please address the information to the IETF Executive Director. Full Copyright Statement Copyright (C) The Internet Society (2004). This document is subject to the rights, licenses and restrictions contained in BCP 78 and except as set forth therein, the authors retain all their rights. This document and the information contained herein are provided on an "AS IS" basis and THE CONTRIBUTOR, THE ORGANIZATION HE/SHE REPRESENTS OR IS SPONSORED BY (IF ANY), THE INTERNET SOCIETY AND THE INTERNET ENGINEERING TASK FORCE DISCLAIM ALL WARRANTIES, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO ANY WARRANTY THAT THE USE OF THE INFORMATION HEREIN WILL NOT INFRINGE ANY RIGHTS OR ANY IMPLIED WARRANTIES OF MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE. Intellectual Property The IETF takes no position regarding the validity or scope of any Intellectual Property Rights or other rights that might be claimed to pertain to the implementation or use of the technology described in this document or the extent to which any license under such rights might or might not be available; nor does it represent that it has made any independent effort to identify any such rights. Information on the procedures with respect to rights in RFC documents can be found in BCP 78 and BCP 79. Copies of IPR disclosures made to the IETF Secretariat and any assurances of licenses to be made available, or the result of an attempt made to obtain a general license or permission for the use of such proprietary rights by implementers or users of this specification can be obtained from the IETF on-line IPR repository at http://www.ietf.org/ipr. The IETF invites any interested party to bring to its attention any copyrights, patents or patent applications, or other proprietary rights that may cover technology that may be required to implement this standard. Please address the information to the IETF at ietf-ipr@ietf.org. Acknowledgement Funding for the RFC Editor function is currently provided by the Internet Society.