Extended Internet Fax: Extended Facsimile Using Internet Mail
Fax over ordinary email, with required delivery confirmation and optional higher-quality documents.
Masinter, L. and D. Wing, "Extended Facsimile Using Internet Mail", RFC 2532, DOI 10.17487/RFC2532, March 1999, <https://www.rfc-editor.org/info/rfc2532>.
@misc{rfc2532,
series = {Request for Comments},
number = 2532,
howpublished = {RFC 2532},
publisher = {RFC Editor},
doi = {10.17487/RFC2532},
url = {https://www.rfc-editor.org/info/rfc2532},
author = {L. Masinter and D. Wing},
title = {{Extended Facsimile Using Internet Mail}},
pagetotal = 12,
year = 1999,
month = mar,
} RFC text
Contents
Network Working Group L. Masinter
Request for Comments: 2532 Xerox Corporation
Category: Standards Track D. Wing
Cisco Systems
March 1999
Extended Facsimile Using Internet Mail
Status of this Memo
This document specifies an Internet standards track protocol for the
Internet community, and requests discussion and suggestions for
improvements. Please refer to the current edition of the "Internet
Official Protocol Standards" (STD 1) for the standardization state
and status of this protocol. Distribution of this memo is unlimited.
Copyright Notice
Copyright (C) The Internet Society (1999). All Rights Reserved.
Abstract
This document describes extensions to "Simple Mode of Facsimile Using
Internet Mail" [RFC2305] and describes additional features, including
transmission of enhanced document characteristics (higher resolution,
color) and confirmation of delivery and processing.
These additional features are designed to provide the highest level
of interoperability with the existing and future standards-compliant
email infrastructure and mail user agents, while providing a level of
service that approximates the level currently enjoyed by fax users.
The IETF has been notified of intellectual property rights claimed in
regard to some or all of the specification contained in this
document. For more information consult the online list of claimed
rights in <http://www.ietf.org/ipr.html>.
1. Introduction
This document notes a number of enhancements to the "Simple Mode of
Facsimile Using Internet Mail" [RFC2305] that may be combined to
create an extended mode of facsimile using Internet mail.
The new features are designed to be interoperable with the existing
base of mail transfer agents (MTAs) and mail user agents (MUAs), and
take advantage of existing standards for advanced functionality such
as positive delivery confirmation and disposition notification. The
enhancements described in this document utilize the messaging
infrastructure, where possible, instead of creating fax-specific
features which are unlikely to be implemented in non-fax messaging
software.
This document standardizes the following two features.
* Delivery confirmation (Section 2) (required)
* Additional document features (Section 3) (optional)
These features are fully described in another document titled
"Terminology and Goals for Internet Fax" [RFC2542].
1.1. Definition of Terms
The term "processing" indicates the action of rendering or
transmitting the contents of the message to a printer, display
device, or fax machine.
The term "processing confirmation" is an indication by the recipient
of a message that it is able to process the contents of that message.
The term "recipient" indicates the device which performs the
processing function. For example, a recipient could be implemented
as a traditional Mail User Agent on a PC, a standalone device which
retrieves mail using POP3 or IMAP, an SMTP server which prints
incoming messages (similar to an LPR server).
The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT",
"SHOULD", SHOULD NOT", "RECOMMENDED", "MAY", and "OPTIONAL" in this
document are to be interpreted as described in [RFC2119].
1.2. GSTN Fax Gateways ("onramp"/"offramp")
The behavior of gateways from GSTN fax to SMTP ("onramps") and from
SMTP to GSTN fax ("offramps") are not described in this document.
However, such gateways SHOULD have the behavior characteristics of
senders and recipients as described in this document.
2. Delivery and Processing Confirmation
In traditional GSTN-based realtime facsimile, the receiving terminal
acknowledges successful receipt and processing of every page [T.30].
In Internet Mail, the operations of Delivery (to the mailbox) and
Disposition (to paper or a screen) may be separated in time (due to
store and forwarding of messages) and location (due to separation of
delivery agent (MTA) and user agent (MUA)). The confirmation of
these two operations are supplied by two different standards-track
mechanisms: Delivery Status Notifications (DSN) [RFC1891, RFC1894]
and Message Disposition Notifications (MDN) [RFC2298], respectively.
This section defines requirements for devices or services that are to
be considered compliant with this document.
2.1. Sender Requirements
Because delivery failure may occur (over disk quota, user no longer
exists, malconfigured mailer), a delivery failure message (in the
format described by [RFC1894] or otherwise) may be sent to the
envelope-from address specified by the sender. Thus, the envelope-
from address supplied by the sender MUST be able to properly handle
such delivery failure messages.
2.1.1. Delivery Confirmation
If the sender desires delivery confirmation, the sender MUST request
Delivery Status Notification by including the the esmtp-keyword
NOTIFY with the esmtp-value SUCCESS (section 5.1 of [RFC1891]).
2.1.2. Processing Confirmation
If the sender desires processing confirmation, the sender MUST
request Message Disposition Notification ([RFC2298] section 2) when
sending the message itself.
Because a recipient may silently ignore a request for an MDN (section
2.1 of [RFC2298]) at any time:
* MDNs MUST NOT be used for delivery confirmation, but are only
useful for disposition ("processing") notification.
* the sender MUST NOT assume the recipient will respond to an MDN
request in a subsequent message, even if the recipient has done
so in the past.
The address provided by the sender on the Disposition-Notification-To
field MUST be able to receive Message Disposition Notifications
messages [RFC2298] and SHOULD be able to receive messages that are
not in the Message Disposition Notification format (due to the
existence of legacy systems that generate non-RFC2298-compliant
responses to the Disposition-Notification-To field). The
Disposition-Notification-To address and the envelope-from address
SHOULD match to allow automated responses to MDN requests (section
2.1 of [RFC2298]).
2.2. Recipient Requirements
Recipients SHOULD implement Message Disposition Notifications
[RFC2298] and SHOULD indicate supported media features in DSN and MDN
messages per [RFC2530].
If the recipient is an SMTP server, it behaves as part of the
receiver infrastructure and is therefore subject to the "Receiver
Infrastructure" requirements of this document.
See also "Recipient Recommendations" in section 5.
2.2.1. MDN Recipient Requirements
Recipients MUST be configurable to silently ignore a request for an
MDN (section 2.1 of [RFC2298]).
If the recipient is an automated message processing system which is
not associated with a person, the device MAY be configurable to
always respond to MDN requests, but in all cases MUST be configurable
to never generate MDNs.
A recipient MUST NOT generate an unsolicited MDN to indicate
successful processing. A recipient MAY generate an unsolicited MDN
(sent to the envelope-from (Return-Path:) address) to indicate
processing failure, but subject to the [RFC2298] requirement that it
MUST always be possible for an operator to disable unsolicited MDN
generation.
2.2.2. Recipients Using Mailbox Access Protocols
A recipient using POP3 [RFC1939] or IMAP4 [RFC2060] to retrieve its
mail MUST NOT generate a Delivery Status Notification message
[RFC1894] because such a notification, if it was requested, would
have already been issued by the MTA on delivery to the POP3 or IMAP4
message store.
The recipient MUST NOT use the RFC822 "To:" fields, "Cc:" fields,
"Bcc:" fields, or any other fields containing header recipient
information to determine the ultimate destination mailbox or
addressee, and SHOULD NOT use other RFC822 or MIME fields for making
such determinations.
2.3. Messaging Infrastructure Requirements
This section explains the requirements of the SMTP messaging
infrastructure used by the sender and receiver. This infrastructure
is commonly provided by the ISP or a company's internal mailers but
can actually be provided by another organization with appropriate
service contracts.
2.3.1. Sender Infrastructure
Support for DSN [RFC1891] MUST be provided by the mail submission
server [RFC2476] used by the sender and MUST be provided up to the
mailer responsible for communicating with external (Internet)
mailers.
Also see section 5.1 of this document.
2.3.2. Receiver Infrastructure
Support for DSN [RFC1891] MUST be provided by the external
(Internet-accessible) mailer, and MUST be provided by each mailer
between the external mailer and the recipient. If the recipient is
implemented as an SMTP server it MUST also support DSN [RFC1891].
3. Additional Document Capabilities
Section 4 of "A Simple Mode of Facsimile Using Internet Mail"
[RFC2305] allows sending only the minimum subset of TIFF for
Facsimile "unless the sender has prior knowledge of other TIFF fields
or values supported by the recipient."
A recipient MAY support any or all (or any combination) of the TIFF
profiles defined in RFC 2301, in addition to profile S. A recipient
which supports additional profiles SHOULD indicate this support as
per section 3.2 or 3.3 of this document. As a consequence, a sender
MAY use those additional TIFF profiles when sending to a recipient
with the corresponding capabilities.
A sender SHOULD be able to recognize and process the feature tags as
defined in [RFC2531] when reviewing the capabilities presented by a
potential recipient. The capability matching rules indicated there
(by reference to [RFC2533]) allow for the introduction of new
features that may be unrecognized by older implementations.
A sender MAY send a message containing both the minimum subset of
TIFF for Facsimile (as specified in [RFC2305]) and a higher quality
TIFF using multipart/alternative.
Three methods for the sender to acquire such knowledge are described:
1. Sender manual configuration
2. Capabilities in Directory
3. Capabilities returned in MDN or DSN
Method (3) SHOULD be used.
An implementation may cache capabilities locally and lose
synchronization with the recipient's actual capabilities. A
mechanism SHOULD be provided to allow the sender to override the
locally-stored cache of capabilities. Also note section 4.1 of this
document.
3.1. Sender Manual Configuration
One way a sender can send a document which exceeds the minimum subset
allowed by [RFC2305] is for the user controlling the sender to
manually override the default settings, usually on a per-recipient
basis. For example, during transmission a user could indicate the
recipient is capable of receiving high resolution images or color
images.
While awkward and not automatic, this mechanism reflects the current
state of deployment of configuration for extended capabilities to
ordinary Internet email users.
3.2. Capabilities in Directory
A future direction for enhanced document features is to create a
directory structure of recipient capabilities, deployed, for example,
through LDAP or DNS. The directory would provide a mechanism by which
a sender could determine a recipient's capabilities before message
construction or transmission, using a directory lookup. Such
mechanisms are not defined in this document.
There is active investigation within the IETF to develop a solution
to this problem, which would resolve a wide range of issues with
store-and-forward messaging.
3.3. Capabilities Returned in MDN or DSN
As outlined in section 2 of this document, a sender may request a
positive DSN or an MDN.
If the recipient implements [RFC2530], the DSN or MDN that is
returned can contain information describing the recipient's
capabilities. The sender can use this information for subsequent
communications with that recipient.
The advantage of this approach is that additional infrastructure is
not required (unlike section 3.2), and the information is acquired
automatically (unlike section 3.1).
3.3.1. Restrictions and Recommendations
A sender MUST NOT send a message with no processable content to
attempt to elicit an MDN/DSN capability response. Doing so with a
message with no processable content (such as a message containing
only a request for capabilities or a blank message) will confuse a
recipient not already designed to understand the semantics of such a
message.
A recipient SHOULD indicate the profiles and features supported, even
if the recipient supports only Tiff Profile S (the minimum set for
fax as defined by [RFC2305]) [RFC2531]. This allows a sender to
determine that the recipient is compliant with this Extended
Facsimile Using Internet Mail specification.
4. Security Considerations
As this document is an extension of [RFC2305], the Security
Considerations section of [RFC2305] applies to this document.
The following additional security considerations are introduced by
the new features described in this document.
4.1. Inaccurate Capabilities Information
Inaccurate capability information (section 3) could cause a denial of
service. The capability information could be inaccurate due to many
reasons, including compromised or improperly configured directory
server, improper manual configuration of sender, compromised DNS, or
spoofed MDN. If a sender is using cached capability information,
there SHOULD be a mechanism to allow the cached information to be
ignored or overridden if necessary.
4.2. Forged MDNs or DSNs
Forged DSNs or MDNs, as described in [RFC1892, RFC1894, RFC2298] can
provide incorrect information to a sender.
5. Implementation Notes
This section contains notes to implementors.
5.1. Submit Mailer Does Not Support DSN
In some installations the generally available submit server may not
support DSNs. In such circumstances, it may be useful for the sender
to implement [RFC974] mail routing as well as additional submission
server functions [RFC2476] so that the installation is not
constrained by limitations of the incumbent submission server.
5.2. Recipient Recommendations
To provide a high degree of reliability, it is desirable for the
sender to know that a recipient could not process a message. The
inability to successfully process a message may be detectable by the
recipient's MTA or MUA.
If the recipient's MTA determines the message cannot be processed,
the recipient's MTA is strongly encouraged to reject the message with
a [RFC1893] status code of 5.6.1. This status code may be returned
in response to the end-of-mail-data indicator if the MTA supports
reporting of enhanced error codes [RFC2034], or after message
reception by generating a delivery failure DSN ("bounce").
Note: Providing this functionality in the MTA, via either of the
two mechanisms described above, is superior to providing the
function using MDNs because MDNs must generally be requested
by the sender (and the request may, at any time, be ignored by
the receiver). Message rejection performed by the MTA can
always occur without the sender requesting such behavior and
without the receiver circumventing the behavior.
If the message contains an MDN request and the recipient's MUA
determines the message cannot be processed, the recipient's MUA is
strongly encouraged to repond to an MDN request and indicate that
processing failed with the disposition-type "processed" or
"displayed" and disposition-modifier "error" or "warning" [RFC2298].
6. Acknowledgements
The authors would like to acknowledge the members of the IETF
Internet Fax working group, and especially the following contributors
who provided assistance and input during the development of this
document:
Vivian Cancio, Richard Coles, David Crocker, Ned Freed, Graham Klyne,
MAEDA Toru, Geoff Marshall, Lloyd McIntyre, Keith Moore, George
Pajari, James Rafferty, Mike Ruhl, Richard Shockey, Brian Stafford,
and Greg Vaudreuil.
7. References
[RFC2533] Klyne, G., "A Syntax for Describing Media Feature Sets",
RFC 2533, March 1999.
[RFC2531] McIntyre, L. and G. Klyne, "Content Feature Schema for
Internet Fax", RFC 2531, March 1999.
[RFC2530] Wing, D., "Indicating Supported Media Features Using
Extensions to DSN and MDN", RFC 2530, March 1999.
[RFC1891] Moore, K. "SMTP Service Extensions for Delivery Status
Notifications", RFC 1891, January 1996.
[RFC1893] Vaudreuil, G., "Enhanced Mail System Status Codes", RFC
1893, January 1996.
[RFC1894] Moore, K. and G. Vaudreuil, "An Extensible Message Format
for Delivery Status Notifications", RFC 1894, January 1996.
[RFC2034] Freed, N, "SMTP Service Extension for Returning Enhanced
Error Codes", RFC 2034, October 1996.
[RFC2119] Bradner, S., "Key words for use in RFCs to Indicate
Requirement Levels", BCP 14, RFC 2119, March 1997.
[RFC2298] Fajman, R., "An Extensible Message Format for Message
Disposition Notifications", RFC 2298, March 1998.
[RFC2301] McIntyre, L., Zilles, S., Buckley, R., Venable, D.,
Parsons, G. and J. Rafferty, "File Format for Internet
Fax", RFC 2301, March 1998.
[RFC2305] Toyoda, K., Ohno, H., Murai, J. and D. Wing, "A Simple
Mode of Facsimile Using Internet Mail", RFC 2305, March
1998.
[RFC974] Partridge. C., "Mail routing and the domain system", STD
14, RFC 974, January 1986.
[RFC2476] Gellens, R. and J. Klensin, "Message Submission", RFC 2476,
December 1998.
[RFC2542] Masinter, L., "Terminology and Goals for Internet Fax", RFC
2542, March 1999.
[T.30] "Procedures for Document Facsimile Transmission in the
General Switched Telephone Network", ITU-T (CCITT),
Recommendation T.30, July, 1996.
[RFC1939] Myers, J. and M. Rose, "Post Office Protocol - Version 3",
STD 53, RFC 1939, May 1996.
[RFC2060] Crispin, M., "Internet Message Access Protocol - Version
4Rev1", RFC 2060, December 1996.
8. Authors' Addresses
Larry Masinter
Xerox Palo Alto Research Center
3333 Coyote Hill Road
Palo Alto, CA 94304 USA
Fax: +1 650 812 4333
EMail: masinter@parc.xerox.com
Dan Wing
Cisco Systems, Inc.
101 Cooper Street
Santa Cruz, CA 95060 USA
Phone: +1 831 457 5200
Fax: +1 831 457 5208
EMail: dwing@cisco.com
9. Full Copyright Statement
Copyright (C) The Internet Society (1999). All Rights Reserved.
This document and translations of it may be copied and furnished to
others, and derivative works that comment on or otherwise explain it
or assist in its implementation may be prepared, copied, published
and distributed, in whole or in part, without restriction of any
kind, provided that the above copyright notice and this paragraph are
included on all such copies and derivative works. However, this
document itself may not be modified in any way, such as by removing
the copyright notice or references to the Internet Society or other
Internet organizations, except as needed for the purpose of
developing Internet standards in which case the procedures for
copyrights defined in the Internet Standards process must be
followed, or as required to translate it into languages other than
English.
The limited permissions granted above are perpetual and will not be
revoked by the Internet Society or its successors or assigns.
This document and the information contained herein is provided on an
"AS IS" basis and THE INTERNET SOCIETY AND THE INTERNET ENGINEERING
TASK FORCE DISCLAIMS 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.
AI ELI5
On this page
§1What This Extends
RFC 2532 builds on RFC 2305, the simple mode of facsimile using Internet mail. It describes enhancements that can be combined into an extended mode of Internet fax. The goal is to work with existing mail transfer agents (MTAs) and mail user agents (MUAs) and to reuse standard mail features, rather than creating fax-specific ones that ordinary mail software is unlikely to implement.
Two features are standardized: delivery confirmation (required) and additional document features (optional). The wider goals are described in RFC 2542.
The RFC defines processing as rendering or transmitting a message's contents to a printer, display or fax machine, and processing confirmation as the recipient's indication that it can process the contents. A recipient is the device that does the processing, for example a mail client on a PC, a standalone device that fetches mail using POP3 or IMAP, or an SMTP server that prints incoming messages. Gateways between the telephone fax network and SMTP are not described, though they SHOULD behave like the senders and recipients defined here.
This document adds two things to email-faxing: a receipt saying the fax arrived, and a way to send better pictures. It does so using email tools that already exist.
§2.1Sender Requirements for Confirmation
Traditional fax acknowledges receipt and processing of every page. In Internet mail, delivery to the mailbox and processing (to paper or a screen) can be separated in time and place, so two standards-track mechanisms are used: Delivery Status Notifications (DSN, RFC 1891 and RFC 1894) for delivery, and Message Disposition Notifications (MDN, RFC 2298) for processing.
The sender's envelope-from address MUST be able to handle delivery failure messages. A sender that wants delivery confirmation MUST request a DSN by including the ESMTP keyword NOTIFY with the value SUCCESS. A sender that wants processing confirmation MUST request an MDN when it sends the message.
Because a recipient may silently ignore an MDN request at any time, MDNs MUST NOT be used for delivery confirmation, and a sender MUST NOT assume a recipient will answer an MDN request, even if it has before. The Disposition-Notification-To address MUST be able to receive MDNs, SHOULD also accept non-MDN replies from legacy systems, and SHOULD match the envelope-from address so responses can be automated.
A sender can ask for two kinds of receipts: one saying the fax reached the mailbox, and one saying someone actually printed or viewed it. The second kind can be ignored, so it is only a bonus.
- DSN: deliveryConfirms delivery to the mailbox. Requested with NOTIFY=SUCCESS and issued by the mail system.
- MDN: processingConfirms disposition to paper or screen. The recipient may silently ignore the request.
§2.2Recipient and Infrastructure Requirements
Recipients SHOULD implement MDNs and SHOULD indicate supported media features in DSN and MDN messages as defined in RFC 2530. A recipient MUST be configurable to silently ignore an MDN request. An automated recipient not tied to a person MAY be configurable to always respond, but must always be configurable to never generate MDNs. A recipient MUST NOT send an unsolicited MDN to report success; it MAY send one to report failure, as long as an operator can always disable that.
A recipient that fetches mail with POP3 or IMAP MUST NOT generate a DSN, because the MTA already issued any requested notification when it put the message in the store. Such a recipient MUST NOT use the To, Cc, Bcc or other recipient header fields to decide the final mailbox or addressee.
The mail infrastructure must also help. The submission server used by the sender MUST support DSN, as must every mailer up to the one that talks to external mailers. On the receiving side, the Internet-accessible mailer and each mailer between it and the recipient MUST support DSN, and so MUST a recipient implemented as an SMTP server.
The person receiving can choose to ignore requests for a 'someone looked at it' receipt. The mail system along the way must be able to give 'it arrived' receipts, and only the first mail stop should send them.
§3Learning What a Recipient Can Handle
The simple mode allows only the minimum TIFF subset (profile S) unless the sender knows more about the recipient. Under RFC 2532 a recipient MAY support any of the TIFF profiles in RFC 2301 in addition to profile S, and SHOULD say so. A sender should recognize the feature tags defined in RFC 2531, whose matching rules (via RFC 2533) allow new features that older implementations do not know. A sender MAY send one message containing both the minimum TIFF and a higher quality TIFF using multipart/alternative.
Three ways to learn a recipient's abilities are described, and method 3 SHOULD be used: (1) the user configures it manually, (2) a directory such as LDAP or DNS holds it, and (3) it is returned in a DSN or MDN. A directory approach is only a future direction and is not defined here. The DSN/MDN approach needs no extra infrastructure and is automatic.
A sender MUST NOT send a message with no processable content just to provoke a capability response, since that would confuse a recipient not designed for it. A recipient SHOULD report its profiles and features even if it supports only profile S, so the sender can tell it complies with this specification. Senders that cache capabilities SHOULD provide a way to override the cache.
Before sending a fancy color picture, you want to know the other side can show it. You can be told by hand, look it up in a phone-book-like directory (not defined yet), or learn it from the receipts they send back.
| Method | What the RFC says |
|---|---|
| Sender manual configuration | Awkward and not automatic, but reflects current deployment. |
| Capabilities in directory | A future direction through LDAP or DNS; not defined in this document. |
| Capabilities returned in DSN or MDN THIS RFC | SHOULD be used; no extra infrastructure and automatic. |
§4Security Considerations
The security considerations of RFC 2305 apply here too. The new features add two more concerns.
Inaccurate capabilities. Wrong capability information could cause a denial of service. It can arise from a compromised or misconfigured directory server, a manual configuration mistake, compromised DNS or a spoofed MDN. A sender that caches capabilities SHOULD have a way to ignore or override the cached data.
Forged notifications. Forged DSNs or MDNs can give a sender incorrect information.
If someone gives you the wrong note about what the other side can handle, your faxes might fail. A fake receipt can also trick you, so you should be able to throw old notes away.
§5Notes for Implementers
If the usual submission server does not support DSN, the sender can implement mail routing as in RFC 974 and extra submission server functions from RFC 2476, so it is not limited by that server.
It is desirable for a sender to learn that a recipient could not process a message. If the recipient's MTA decides the message cannot be processed, it is strongly encouraged to reject it with status code 5.6.1 from RFC 1893, either in response to the end-of-mail-data indicator when enhanced error codes are supported, or afterward with a delivery failure DSN. The RFC notes this is better than using MDNs, because MDN requests may be ignored while MTA rejection happens whether or not the sender asks.
If the message has an MDN request and the recipient's MUA finds it cannot process the message, the MUA is strongly encouraged to answer, with disposition-type processed or displayed and disposition-modifier error or warning.
The best time to say 'we cannot handle this' is as the mail first arrives, because that works even when the sender did not ask. If a person's program finds a problem later, it should say so in its reply.