Add notification of OCSP-stapling facility inclusion.
[exim.git] / doc / doc-misc / RFC.conform
1 Conformance with RFCs
2 ---------------------
3
4 Exim is written to follow the rules laid down in the RFCs. However, there are
5 some circumstances where it either extends what is specified, or chooses not to
6 follow them strictly, for various reasons. Sometimes variations are controlled
7 by an option, which may default on or off. This document lists the variations
8 from the latest email RFCs, and discusses their background and implications.
9
10 Last Updated: 25 January 1999
11
12
13 1. RFC 822
14 ----------
15
16 The original specification of the format of Internet mail messages is RFC 822,
17 later clarified and modified by RFC 1123. At the time of writing (January 1999)
18 a new RFC (currently known as draft-ietf-drums-msg-fmt-07) which updates and
19 consolidates all the material related to the message format is at a late stage
20 of drafting, and is expected to become an Internet Standard in due course.
21
22 The following is (I hope) a complete list of major variations from the draft
23 RFC. References in square brackets are to the -07 draft.
24
25
26 1.1 Line termination [2.1, 2.3]
27 -------------------------------
28
29 [Lines are terminated by CRLF; isolated CR and LF are not permitted.]
30
31 The CRLF requirement has to be interpreted carefully, because the RFC also says
32 that it does not cover the internal format "used by sites". Exim keeps messages
33 on its spool in Unix format, using only LF as the line terminator, and also
34 does local deliveries using only LF. I believe this is compliant with the RFC,
35 as these are both "internal formats".
36
37 Messages sent out by SMTP have CRLF line terminators. However, isolated CR
38 characters are treated as any other data characters, because Exim is eight-bit
39 clean (see 1.2 below).
40
41 See 2.1 below for a discussion of line terminators in incoming messages.
42
43
44 1.2 Eight-bit characters [2.1]
45 ------------------------------
46
47 [Messages consist of 7-bit characters.]
48
49 Exim is eight-bit clean. It does not do any processing of the characters in the
50 body of a message.
51
52
53 1.3 Maximum line length [2.1, 2.3]
54 ----------------------------------
55
56 [The maximum length of a line is 998 characters.]
57
58 Exim does not enforce any limit on line length.
59
60
61 1.4 The "phrase" part of an address [3.4]
62 -----------------------------------------
63
64 [The phrase is a sequence of "words"; a word is an "atom" or a quoted string.]
65
66 The characters that can be used in an "atom" do not include the full stop
67 (dot, period). Thus a header line such as
68
69 To: John Q. Public <jqp@anywhere.org>
70
71 is syntactically invalid under a strict interpretation of the RFC because the
72 dot in the phrase part is not quoted. However, many MTAs do not enforce this
73 restriction, so Exim was changed to be relaxed about it as well. In fact, the
74 draft RFC is moving towards allowing this. In section [4.1], which is defining
75 "obsolete" syntax that programs must accept (but not generate), it says this:
76
77 The period character is added to obs-phrase.
78
79 Note: The period character in obs-phrase is not a form that was allowed
80 in earlier versions of this or any other standard. Period (nor any other
81 character from specials) was not allowed in phrase because it introduced
82 a parsing difficulty distinguishing between phrases and portions of an
83 addr-spec (see section 4.4). It appears here because the period
84 character is currently used in many messages in the display-name portion
85 of addresses, especially for initials in names, and therefore must be
86 interpreted properly. In the future, period may appear in the regular
87 syntax of phrase.
88
89
90 1.5 Source routed addresses [4.4]
91 ---------------------------------
92
93 [Source routed addresses are always enclosed in <>.]
94
95 Source routed addresses are declared obsolete in the draft RFC, but MTAs are
96 still required to handle them. Strictly, a source-routed address must be
97 enclosed in <> characters, so a header such as
98
99 From: @a,@b:c@d
100
101 is syntactally invalid. Exim does not enforce this restriction.
102
103
104 1.6 Local parts [3.4.1]
105 -----------------------
106
107 [Dots in unquoted local parts may not be consecutive or at either end.]
108
109 Exim allows unquoted local parts to begin or end with a dot (period, full
110 stop), and it also permits two consecutive dots in a local part.
111
112
113
114 2. RFC 821
115 ----------
116
117 The original specification of SMTP is RFC 821, later clarified and modified by
118 RFC 1123. Domain name system requirements and their implications for mail are
119 covered in RFCs 1035 and 974. A scheme for extending the SMTP protocol is
120 described in RFC 1869, and there are subsequent RFCs specifying particular
121 extensions.
122
123 At the time of writing (January 1999) a new RFC (currently known as
124 draft-ietf-drums-smtpupd-09) which updates and consolidates all the material
125 connected with SMTP message transmission is at a late stage of drafting, and is
126 expected to become an Internet Standard in due course.
127
128 The new draft is written using the terms MUST, SHOULD, and MAY, which, when
129 written in capital letters, have precise meanings. To quote from the draft:
130
131 "MUST" or "MUST NOT" identify absolute requirements for conformance to
132 this specification. Implementations that do not conform to them lie
133 outside the scope of this specification and often will not
134 interoperate properly with SMTP implementations that do conform.
135 Implementations that are fully conforming also adhere to all "SHOULD"
136 and "SHOULD NOT" requirements. Implementations that adhere to all
137 "MUST" ("MUST NOT") but not to all of these are considered to be
138 partially conforming. Such implementations may interoperate properly
139 with fully conforming ones and with each other, but this will
140 typically be the case only if great care is taken. Consequently, an
141 implementation should violate "SHOULD" ("SHOULD NOT") requirements
142 only under exceptional and well-understood circumstances.
143
144 The implementation of Exim is intended to conform to the spirit of this
145 paragraph. The following is (I hope) a complete list of major variations
146 from the draft RFC. In addition to the items listed here, there are other minor
147 extensions such as the tolerance of white space in places where it is not
148 strictly permitted by the RFC. References in square brackets are to the -09
149 draft sections, and brief summaries of the RFC requirement are also given in
150 square brackets.
151
152
153 2.1 Line termination [2.3.7, 4.1.1.4]
154 -------------------------------------
155
156 [SMTP lines are terminated by CRLF.]
157
158 Exim recognizes LF without CR as a line terminator in all forms of input. For
159 SMTP input, any preceding CR is discarded. An early version of Exim followed
160 the RFC strictly, and did not recognize LF without CR in SMTP input. However,
161 it seems that sites on the net send out messages with just LF terminators,
162 despite the warnings in the RFCs, and other MTAs handle this, so Exim was
163 changed. However, there is a compile time macro called STRICT_CRLF which can be
164 set to restore the strict behaviour, though this is undocumented.
165
166
167 2.2 Eight-bit characters [2.4.1]
168 --------------------------------
169
170 [SMTP transmits only 7-bit characters.]
171
172 Exim is eight-bit clean, and makes no attempt to modify the data in a message
173 in any way. In particular, for messages containing characters with the top bit
174 set, it neither tries to negotiate 8-bit transmission, nor converts such
175 characters into an encoded form. In other words, it adopts the "just send 8"
176 strategy. It can be configured to send out 8BITMIME in its response to EHLO
177 (which it does not do by default), and it recognizes the 8BITMIME keyword on
178 incoming messages, but neither of these affect its handling of message data.
179 "Just send 8" is the strategy of a number of MTAs; it is argued that it
180 achieves what the user wants more often than other strategies.
181
182
183 2.3 Use of EHLO/HELO [3.2]
184 --------------------------
185
186 [Client MTAs should always start with EHLO, not HELO.]
187
188 Exim sends EHLO only when it finds the string "ESMTP" in an SMTP greeting
189 message. If EHLO is refused with a 5xx return code, it then reverts to HELO as
190 required, but it does not contain logic for converting to HELO on other errors
191 such as loss of connection or timeout after EHLO. That is one reason why it
192 doesn't always send EHLO; there are reported to be ancient SMTP servers out
193 there which collapse on receiving EHLO. (There is also at least one server
194 whose banner reads "<host name> ignores ESMTP", but it is RFC 821 compliant in
195 that it responds with 5O0 to EHLO, so Exim successfully reverts to HELO.)
196
197
198 2.4 Closing the connection [4.1.1.10]
199 -------------------------------------
200
201 [Client must wait for response to QUIT before closing the connection.]
202
203 Exim closes the connection immediately after sending QUIT, without waiting for
204 the reply. There was a lot of discussion about this on one of the mailing
205 lists. The conclusion was that this behaviour is fine on Unix systems, which
206 have TCP/IP implementations that close down the underlying channel tidily even
207 when the associated process has terminated. Indeed, not waiting may be
208 beneficial, as it moves the TIME_WAIT state (waiting to ensure there's no more
209 data in transit) from the server to the client system. On some other operating
210 systems (I understand) it is a disaster to terminate the sending process
211 without waiting for the QUIT response, because all the data about the
212 connection lives in the client's process space, and is therefore thrown away
213 before the response arrives. The subsequent arrival of the response then causes
214 bad behaviour.
215
216
217 2.5 IPv6 address literals [4.1.2]
218 ---------------------------------
219
220 [IPv6 address literals are introduced by "IPv6".]
221
222 Exim recognizes IPv6 literals as just the colon-separated hexadecimal form of
223 an IPv6 address, for example 1080:0:0:0:8:800:200C:417A, without the need for a
224 prefix. At present, it does not even recognize the prefix. When IPv6 becomes
225 more widespread, Exim will follow whatever the common usage is.
226
227
228 2.6 Underscores in domain names [4.1.2]
229 ---------------------------------------
230
231 [Underscores are not legal in domain names.]
232
233 RFC 822 allows all characters except specials, space, and controls in domain
234 names, but the SMTP RFCs are stricter, allowing only letters, digits, and
235 hyphen. Exim is compliant when checking incoming addresses in SMTP commands,
236 but it is more relaxed by default when checking domain names that are supplied
237 by EHLO or HELO commands, because many client workstations get set up with
238 underscores in their names. There is an option that can be set to cause Exim to
239 refuse underscores. (There are also options to specify certain hosts from which
240 it will accept any old junk after EHLO or HELO. Such is the woeful state of
241 some SMTP clients.)
242
243
244 2.7 Removal of return-path headers [4.4]
245 ----------------------------------------
246
247 [Relaying MTAs should not remove return-path.]
248
249 Exim removes Return-Path: headers from all messages, if return_path_remove is
250 set (the default). It does not attempt to determine if it is being a relay or
251 not. Indeed, for some messages it might be both a relay and a final destination
252 MTA for the same message.
253
254
255 2.8 Randomizing the order of addresses of multihomed hosts [5]
256 --------------------------------------------------------------
257
258 [Multihomed host addresses should not be randomized.]
259
260 Exim does randomize a list of several addresses for a single host, because
261 caching in resolvers will defeat the round-robinning that many namerservers
262 use. (Note: this is not the same as randomizing equal-valued MX records. That
263 is required by the RFC.)
264
265
266 2.9 Handling "MX points to self" [5]
267 ------------------------------------
268
269 [MX points to self must be treated as an error.]
270
271 The RFC doesn't allow for the possibility of special-purpose routing in the
272 case when the lowest numbered MX record points to the local host. The default
273 Exim configuration is compliant, but it is possible to configure Exim to behave
274 differently, and there are several situations where this can be useful.
275
276
277 2.10 Source routing [6.1]
278 -------------------------
279
280 [Source routes should be stripped.]
281
282 The new RFC has moved forward in deprecating source-routed email addresses.
283 Exim does not strip them down by default, but can be made to do so by setting
284 collapse_source_routes. However, even when it is not stripping them down, it
285 does not add host routing to reverse-paths when processing a source-routed
286 forward-path.
287
288
289 2.11 Loop detection [6.2]
290 -------------------------
291
292 [Loop count for Received: headers should be at least 100.]
293
294 Exim's default setting of the received_headers_max option is 30. Most messages
295 these days seem to accumulate less than half a dozen Received: headers, and
296 even a couple of forwardings don't bring this anywhere near 30.
297
298
299 2.12 Addition of missing headers [6.3]
300 --------------------------------------
301
302 [Missing headers may be added, and domains qualified, only if client is
303 identified.]
304
305 Exim always adds Message-Id: and Date: headers if these are missing, whatever
306 the source of the message, and likewise when it expands non-fully-qualified
307 domains, it does so independently of the message's source.
308
309
310 2.13 Syntax of MAIL and RCPT commands [4.1.1.2, 4.1.1.3]
311 --------------------------------------------------------
312
313 Exim is more relaxed than the RFC requires:
314
315 (1) Trailing white space is ignored.
316
317 (2) It permits white space after the "FROM" and "TO" keywords.
318
319 (3) It does not insist on the address being enclosed in <> characters. In fact,
320 it recognizes addresses in RFC 822 format here, except that domain
321 components are restricted to containing only letters, digits, and hyphens.
322
323 (4) Local parts are permitted to contain null components, that is, may start or
324 end with an unquoted full stop (period) or contain two consecutive
325 unquoted full stops.
326
327
328 2.14 Non-fully-qualified domains [2.3.5]
329 ----------------------------------------
330
331 [All domains must be fully qualified.]
332
333 A domain that is not fully qualified has some of its trailing components
334 missing, and is normally a local alias of some sort, for example, just a
335 single-component host name.
336
337 Exim can be configured to "widen" non-fully-qualified domains, either by using
338 the facilities of the DNS resolver, or by an explicit list of widening strings.
339 When this is done, it applies to addresses received by SMTP from other hosts,
340 as well as to locally-originated addresses. Address re-writing could also be
341 used for this purpose.
342
343
344 2.15 Unqualified addresses [4.1.2]
345 ----------------------------------
346
347 [Addresses in SMTP commands must include domains.]
348
349 An unqualified address consists of a local part without a domain. Do not
350 confuse "qualified address" and "qualified domain". A qualified address may
351 include a non-fully-qualified domain.
352
353 There is one exception to the RFC rule: it is required that the unqualified
354 address "<postmaster>" always be accepted. Apart from this, Exim rejects
355 domainless addresses in SMTP commands by default, but it can be configured with
356 a list of hosts and/or networks that are permitted to send addresses without
357 domains in SMTP commands. Any such address that is accepted (including
358 <postmaster>) is qualified by adding the value of the qualify_domain option.
359
360
361 2.16 VRFY and EXPN [3.5.1, 3.5.2, 3.5.3, 7.3]
362 ---------------------------------------------
363
364 [VRFY and EXPN should be supported.]
365
366 Exim does not support VRFY and EXPN by default, but a list of hosts and
367 networks for which they are permitted can be given.
368
369
370 2.17 Checking of EHLO/HELO commands [4.1.4]
371 -------------------------------------------
372
373 [Client must send EHLO. Server must not refuse message if EHLO/HELO check
374 fails.]
375
376 Exim, as a client, always sends EHLO or HELO (see 2.3 above). As a server, it
377 does not insist on there having been a valid EHLO or HELO command before the
378 start of a message transaction. Any EHLO or HELO command that is received is
379 rejected only if it contains a syntax error. That is, it is never rejected on
380 the basis of any validation checking that may be performed on the data it
381 contains.
382
383 However, Exim can be configured to insist that (a) there is valid EHLO/HELO
384 command before any message transaction and (b) the domain in that command
385 matches the domain obtained by looking up the IP address of the sending host.
386 It is possible to specify exception lists of hosts and/or networks for which
387 this check does not apply.
388
389
390 2.18 Format of delivery error messages [3.7]
391 --------------------------------------------
392
393 [Standard report formats should be used if possible.]
394
395 Exim's delivery failure reports do not conform to the format described in RFC
396 1894.
397
398
399 ## End ##