ARC: Fix signing for case when DKIM signing failed
[exim.git] / src / README.UPDATING
CommitLineData
0f4f2a88
PH
1This document contains detailed information about incompatibilities that might
2be encountered when upgrading from one release of Exim to another. The
3information is in reverse order of release numbers. Mostly these are relatively
4small points, and the configuration file is normally upwards compatible, but
5there have been two big upheavals...
6
7
8**************************************************************************
9* There was a big reworking of the way mail routing works for release *
10* 4.00. Previously used "directors" were abolished, and all routing is *
11* now done by routers. Policy controls for incoming mail are now done by *
12* Access Control Lists instead of separate options. All this means that *
13* pre-4.00 configuration files have to be massively converted. If you *
14* are coming from a 3.xx release, please read the document in the file *
15* doc/Exim4.upgrade, and allow some time to complete the upgrade. *
16* *
17* There was a big reworking of the way domain/host/net/address lists are *
18* handled at release 3.00. If you are coming from a pre-3.00 release, it *
19* might be easier to start again from a default configuration. Otherwise *
20* you need to read doc/Exim3.upgrade and do a double conversion of your *
21* configuration file. *
22**************************************************************************
23
24
25The rest of this document contains information about changes in 4.xx releases
26that might affect a running system.
2eda48c6 27
e6acf885 28
1f791335
PP
29Exim version 4.91
30-----------------
31
32 * DANE and SPF have been promoted from Experimental to Supported status, thus
33 the options to enable them in Local/Makefile have been renamed.
34 See current src/EDITME for full details, including changes in dependencies,
35 but loosely: replace EXPERIMENTAL_SPF with SUPPORT_SPF and replace
36 EXPERIMENTAL_DANE with SUPPORT_DANE.
37
38 * Ancient ClamAV stream support, long deprecated by ClamAV, has been removed;
39 if you were building with WITH_OLD_CLAMAV_STREAM enabled then your problems
40 have marginally increased.
41
42 * A number of logging changes; if relying upon the previous DKIM additional
43 log-line, explicit log_selector configuration is needed to keep it.
44
45 * Other incompatible changes in EXPERIMENTAL_* features, read NewStuff and
46 ChangeLog carefully if relying upon an experimental feature such as DMARC.
47 Note that this includes changes to SPF as it was promoted into Supported.
48
49
2eec84ca
PP
50Exim version 4.89
51-----------------
52
7dc5f827
PP
53 * SMTP CHUNKING in Exim 4.88 did not ensure that received mails had a final
54 newline; attempts to deliver such messages onwards to non-chunking hosts
55 would probably hang, as Exim does not insert the newline before a ".".
56 In 4.89, the newline is added upon receipt. For already-received messages
57 in your queue, try util/chunking_fixqueue_finalnewlines.pl
58 to walk the queue, fixing any affected messages. Note that because a
59 delivery attempt will be hanging, attempts to lock the messages for fixing
60 them will stall; stopping all queue-runners temporarily is recommended.
61
2eec84ca
PP
62 * OpenSSL: oldest supported release series is now 1.0.2, which is the oldest
63 supported by the OpenSSL project. If you can build Exim with an older
64 release series, congratulations. If you can't, then upgrade.
65 The file doc/openssl.txt contains instructions for installing a current
66 OpenSSL outside the system library paths and building Exim to use it.
67
0aaf76a4
PP
68 * FreeBSD: we now always use the system iconv in libc, as all versions of
69 FreeBSD supported by the FreeBSD project provide this functionality.
70
2eec84ca 71
f59aaaaa
PP
72Exim version 4.88
73-----------------
74
75 * The "demime" ACL condition, deprecated for the past 10 years, has
76 now been removed.
77
78 * Old GnuTLS configuration options "gnutls_require_kx", "gnutls_require_mac",
79 and "gnutls_require_protocols" have now been removed. (Inoperative from
80 4.80, per below; logging warnings since 4.83, again per below).
81
82
c50e5871
JH
83Exim version 4.83
84-----------------
85
86 * SPF condition results renamed "permerror" and "temperror". The old
87 names are still accepted for back-compatability, for this release.
88
89 * TLS details are now logged on rejects, subject to log selectors.
90
91 * Items in headers_remove lists must now have any embedded list-separators
92 doubled.
93
94 * Attempted use of the deprecated options "gnutls_require_kx" et. al.
95 now result in logged warning.
96
97
2c422e6f 98Exim version 4.82
a5f239e4
PP
99-----------------
100
2519e60d 101 * New option gnutls_allow_auto_pkcs11 defaults false; if you have GnuTLS 2.12.0
a5f239e4
PP
102 or later and do want PKCS11 modules to be autoloaded, then set this option.
103
8b260705
PP
104 * A per-transport wait-<name> database is no longer updated if the transport
105 sets "connection_max_messages" to 1, as it can not be used and causes
106 unnecessary serialisation and load. External tools tracking the state of
107 Exim by the hints databases may need modification to take this into account.
108
b10e4ec2
TL
109 * The av_scanner option can now accept multiple clamd TCP targets, all other
110 setting limitations remain.
111
a5f239e4 112
83e2f8a2 113Exim version 4.80
dec5017e
PP
114-----------------
115
83e2f8a2
PP
116 * BEWARE backwards-incompatible changes in SSL libraries, thus the version
117 bump. See points below for details.
c2b063d4 118 Also an LDAP data returned format change.
83e2f8a2 119
dec5017e
PP
120 * The value of $tls_peerdn is now print-escaped when written to the spool file
121 in a -tls_peerdn line, and unescaped when read back in. We received reports
122 of values with embedded newlines, which caused spool file corruption.
123
124 If you have a corrupt spool file and you wish to recover the contents after
125 upgrading, then lock the message, replace the new-lines that should be part
126 of the -tls_peerdn line with the two-character sequence \n and then unlock
127 the message. No tool has been provided as we believe this is a rare
128 occurence.
129
f0f5a555
PP
130 * For OpenSSL, SSLv2 is now disabled by default. (GnuTLS does not support
131 SSLv2). RFC 6176 prohibits SSLv2 and some informal surveys suggest no
132 actual usage. You can re-enable with the "openssl_options" Exim option,
133 in the main configuration section. Note that supporting SSLv2 exposes
134 you to ciphersuite downgrade attacks.
135
c80c5570
PP
136 * With OpenSSL 1.0.1+, Exim now supports TLS 1.1 and TLS 1.2. If built
137 against 1.0.1a then you will get a warning message and the
138 "openssl_options" value will not parse "no_tlsv1_1": the value changes
139 incompatibly between 1.0.1a and 1.0.1b, because the value chosen for 1.0.1a
140 is infelicitous. We advise avoiding 1.0.1a.
141
142 "openssl_options" gains "no_tlsv1_1", "no_tlsv1_2" and "no_compression".
143
da3ad30d 144 COMPATIBILITY WARNING: The default value of "openssl_options" is no longer
f0f5a555
PP
145 "+dont_insert_empty_fragments". We default to "+no_sslv2".
146 That old default was grandfathered in from before openssl_options became a
147 configuration option.
da3ad30d
PP
148 Empty fragments are inserted by default through TLS1.0, to partially defend
149 against certain attacks; TLS1.1+ change the protocol so that this is not
150 needed. The DIEF SSL option was required for some old releases of mail
151 clients which did not gracefully handle the empty fragments, and was
152 initially set in Exim release 4.31 (see ChangeLog, item 37).
153
154 If you still have affected mail-clients, and you see SSL protocol failures
155 with this release of Exim, set:
156 openssl_options = +dont_insert_empty_fragments
157 in the main section of your Exim configuration file. You're trading off
158 security for compatibility. Exim is now defaulting to higher security and
159 rewarding more modern clients.
160
3375e053
PP
161 If the option tls_dhparams is set and the parameters loaded from the file
162 have a bit-count greater than the new option tls_dh_max_bits, then the file
163 will now be ignored. If this affects you, raise the tls_dh_max_bits limit.
164 We suspect that most folks are using dated defaults and will not be affected.
165
5407bfff
JH
166 * Ldap lookups returning multi-valued attributes now separate the attributes
167 with only a comma, not a comma-space sequence. Also, an actual comma within
168 a returned attribute is doubled. This makes it possible to parse the
169 attribute as a comma-separated list. Note the distinction from multiple
170 attributes being returned, where each one is a name=value pair.
c80c5570 171
c2b063d4
PP
172 If you are currently splitting the results from LDAP upon a comma, then you
173 should check carefully to see if adjustments are needed.
174
175 This change lets cautious folks distinguish "comma used as separator for
176 joining values" from "comma inside the data".
177
9cbad13b
PP
178 * accept_8bitmime now defaults on, which is not RFC compliant but is better
179 suited to today's Internet. See http://cr.yp.to/smtp/8bitmime.html for a
180 sane rationale. Those who wish to be strictly RFC compliant, or know that
181 they need to talk to servers that are not 8-bit-clean, now need to take
182 explicit configuration action to default this option off. This is not a
183 new option, you can safely force it off before upgrading, to decouple
184 configuration changes from the binary upgrade while remaining RFC compliant.
185
f90a9fd1
PP
186 * The GnuTLS support has been mostly rewritten, to use APIs which don't cause
187 deprecation warnings in GnuTLS 2.12.x. As part of this, these three options
188 are no longer supported:
17c76198
PP
189
190 gnutls_require_kx
191 gnutls_require_mac
192 gnutls_require_protocols
193
f90a9fd1
PP
194 Their functionality is entirely subsumed into tls_require_ciphers. In turn,
195 tls_require_ciphers is no longer an Exim list and is not parsed by Exim, but
196 is instead given to gnutls_priority_init(3), which expects a priority string;
197 this behaviour is much closer to the OpenSSL behaviour. See:
17c76198 198
42bfef1e 199 http://www.gnutls.org/manual/html_node/Priority-Strings.html
17c76198
PP
200
201 for fuller documentation of the strings parsed. The three gnutls_require_*
202 options are still parsed by Exim and, for this release, silently ignored.
203 A future release will add warnings, before a later still release removes
204 parsing entirely and the presence of the options will be a configuration
205 error.
206
83e2f8a2
PP
207 Note that by default, GnuTLS will not accept RSA-MD5 signatures in chains.
208 A tls_require_ciphers value of NORMAL:%VERIFY_ALLOW_SIGN_RSA_MD5 may
209 re-enable support, but this is not supported by the Exim maintainers.
210 Our test suite no longer includes MD5-based certificates.
211
17c76198
PP
212 This rewrite means that Exim will continue to build against GnuTLS in the
213 future, brings Exim closer to other GnuTLS applications and lets us add
214 support for SNI and other features more readily. We regret that it wasn't
215 feasible to retain the three dropped options.
216
3375e053
PP
217 * If built with TLS support, then Exim will now validate the value of
218 the main section tls_require_ciphers option at start-up. Before, this
219 would cause a STARTTLS 4xx failure, now it causes a failure to start.
220 Running with a broken configuration which causes failures that may only
221 be left in the logs has been traded off for something more visible. This
222 change makes an existing problem more prominent, but we do not believe
223 anyone would deliberately be running with an invalid tls_require_ciphers
224 option.
225
226 This also means that library linkage issues caused by conflicts of some
227 kind might take out the main daemon, not just the delivery or receiving
228 process. Conceivably some folks might prefer to continue delivering
229 mail plaintext when their binary is broken in this way, if there is a
230 server that is a candidate to receive such mails that does not advertise
231 STARTTLS. Note that Exim is typically a setuid root binary and given
232 broken linkage problems that cause segfaults, we feel it is safer to
233 fail completely. (The check is not done as root, to ensure that problems
234 here are not made worse by the check).
235
a799883d
PP
236 * The "tls_dhparam" option has been updated, so that it can now specify a
237 path or an identifier for a standard DH prime from one of a few RFCs.
238 The default for OpenSSL is no longer to not use DH but instead to use
239 one of these standard primes. The default for GnuTLS is no longer to use
240 a file in the spool directory, but to use that same standard prime.
241 The option is now used by GnuTLS too. If it points to a path, then
242 GnuTLS will use that path, instead of a file in the spool directory;
243 GnuTLS will attempt to create it if it does not exist.
244
245 To preserve the previous behaviour of generating files in the spool
246 directory, set "tls_dhparam = historic". Since prior releases of Exim
247 ignored tls_dhparam when using GnuTLS, this can safely be done before
248 the upgrade.
249
250
dec5017e 251
c566dd90
PP
252Exim version 4.77
253-----------------
254
255 * GnuTLS will now attempt to use TLS 1.2 and TLS 1.1 before TLS 1.0 and SSL3,
256 if supported by your GnuTLS library. Use the existing
257 "gnutls_require_protocols" option to downgrade this if that will be a
258 problem. Prior to this release, supported values were "TLS1" and "SSL3",
259 so you should be able to update configuration prior to update.
260
0215ec07
PP
261 [nb: gnutls_require_protocols removed in Exim 4.80, instead use
262 tls_require_ciphers to provide a priority string; see notes above]
263
39257585
PP
264 * The match_<type>{string1}{string2} expansion conditions no longer subject
265 string2 to string expansion, unless Exim was built with the new
266 "EXPAND_LISTMATCH_RHS" option. Too many people have inadvertently created
267 insecure configurations that way. If you need the functionality and turn on
268 that build option, please let the developers know, and know why, so we can
269 try to provide a safer mechanism for you.
270
a7ee5955
PP
271 The match{}{} expansion condition (for regular expressions) is NOT affected.
272 For match_<type>{s1}{s2}, all list functionality is unchanged. The only
273 change is that a '$' appearing in s2 will not trigger expansion, but instead
274 will be treated as a literal $ sign; the effect is very similar to having
275 wrapped s2 with \N...\N. If s2 contains a named list and the list definition
276 uses $expansions then those _will_ be processed as normal. It is only the
277 point at which s2 is read where expansion is inhibited.
278
279 If you are trying to test if two email addresses are equal, use eqi{s1}{s2}.
280 If you are testing if the address in s1 occurs in the list of items given
281 in s2, either use the new inlisti{s1}{s2} condition (added in 4.77) or use
282 the pre-existing forany{s2}{eqi{$item}{s1}} condition.
283
c566dd90 284
e6acf885
PP
285Exim version 4.74
286-----------------
287
4050a044
PP
288 * The integrated support for dynamically loadable lookup modules has an ABI
289 change from the modules supported by some OS vendors through an unofficial
290 patch. Don't try to mix & match.
291
292 * Some parts of the build system are now beginning to assume that the host
293 environment is POSIX. If you're building on a system where POSIX tools are
294 not the default, you might have an easier time if you switch to the POSIX
295 tools. Feel free to report non-POSIX issues as a request for a feature
296 enhancement, but if the POSIX variants are available then the fix will
297 probably just involve some coercion. See the README instructions for
298 building on such hosts.
e6acf885
PP
299
300
301Exim version 4.73
302-----------------
303
304 * The Exim run-time user can no longer be root; this was always
305 strongly discouraged, but is now prohibited both at build and
306 run-time. If you need Exim to run routinely as root, you'll need to
307 patch the source and accept the risk. Here be dragons.
308
309 * Exim will no longer accept a configuration file owned by the Exim
310 run-time user, unless that account is explicitly the value in
311 CONFIGURE_OWNER, which we discourage. Exim now checks to ensure that
a7ee5955 312 files are not writeable by other accounts.
e6acf885
PP
313
314 * The ALT_CONFIG_ROOT_ONLY build option is no longer optional and is forced
315 on; the Exim user can, by default, no longer use -C/-D and retain privilege.
316 Two new build options mitigate this.
317
318 * TRUSTED_CONFIG_LIST defines a file containing a whitelist of config
319 files that are trusted to be selected by the Exim user; one per line.
320 This is the recommended approach going forward.
321
322 * WHITELIST_D_MACROS defines a colon-separated list of macro names which
323 the Exim run-time user may safely pass without dropping privileges.
324 Because changes to this involve a recompile, this is not the recommended
325 approach but may ease transition. The values of the macros, when
43236f35 326 overridden, are constrained to match this regex: ^[A-Za-z0-9_/.-]*$
e6acf885
PP
327
328 * The system_filter_user option now defaults to the Exim run-time user,
329 rather than root. You can still set it explicitly to root and this
330 can be done with prior versions too, letting you roll versions
331 without needing to change this configuration option.
332
333 * ClamAV must be at least version 0.95 unless WITH_OLD_CLAMAV_STREAM is
334 defined at build time.
335
336
2eda48c6
TK
337Exim version 4.70
338-----------------
339
3401. Experimental Yahoo! Domainkeys support has been dropped in this release.
341It has been superceded by a native implementation of its successor DKIM.
342
3432. Up to version 4.69, Exim came with an embedded version of the PCRE library.
344As of 4.70, this is no longer the case. To compile Exim, you will need PCRE
a7ee5955 345installed. Most OS distributions have ready-made library and development
2eda48c6 346packages.
0f4f2a88
PH
347
348
3c223874
PH
349Exim version 4.68
350-----------------
351
3521. The internal implementation of the database keys that are used for ACL
353ratelimiting has been tidied up. This means that an update to 4.68 might cause
354Exim to "forget" previous rates that it had calculated, and reset them to zero.
355
356
26da7e20
PH
357Exim version 4.64
358-----------------
359
3601. Callouts were setting the name used for EHLO/HELO from $smtp_active_
361hostname. This is wrong, because it relates to the incoming message (and
362probably the interface on which it is arriving) and not to the outgoing
363callout (which could be using a different interface). This has been
364changed to use the value of the helo_data option from the smtp transport
365instead - this is what is used when a message is actually being sent. If
366there is no remote transport (possible with a router that sets up host
367addresses), $smtp_active_hostname is used. This change is mentioned here in
368case somebody is relying on the use of $smtp_active_hostname.
369
ebc9d865
PH
3702. A bug has been fixed that might just possibly be something that is relied on
371in some configurations. In expansion items such as ${if >{xxx}{yyy}...} an
372empty string (that is {}) was being interpreted as if it was {0} and therefore
373treated as the number zero. From release 4.64, such strings cause an error
374because a decimal number, possibly followed by K or M, is required (as has
375always been documented).
376
b6c27e28
PH
3773. There has been a change to the GnuTLS support (ChangeLog/PH/20) to improve
378Exim's performance. Unfortunately, this has the side effect of being slightly
379non-upwards compatible for versions 4.50 and earlier. If you are upgrading from
380one of these earlier versions and you use GnuTLS, you must remove the file
381called gnutls-params in Exim's spool directory. If you don't do this, you will
382see this error:
383
384 TLS error on connection from ... (DH params import): Base64 decoding error.
385
386Removing the file causes Exim to recompute the relevant encryption parameters
387and cache them in the new format that was introduced for release 4.51 (May
3882005). If you are upgrading from release 4.51 or later, there should be no
389problem.
390
26da7e20 391
a5bd321b
PH
392Exim version 4.63
393-----------------
394
395When an SMTP error message is specified in a "message" modifier in an ACL, or
396in a :fail: or :defer: message in a redirect router, Exim now checks the start
397of the message for an SMTP error code. This consists of three digits followed
398by a space, optionally followed by an extended code of the form n.n.n, also
399followed by a space. If this is the case and the very first digit is the same
400as the default error code, the code from the message is used instead. If the
401very first digit is incorrect, a panic error is logged, and the default code is
402used. This is an incompatible change, but it is not expected to affect many (if
403any) configurations. It is possible to suppress the use of the supplied code in
404a redirect router by setting the smtp_error_code option false. In this case,
405any SMTP code is quietly ignored.
406
407
47ca6d6c
PH
408Exim version 4.61
409-----------------
410
8def5aaf
PH
4111. The default number of ACL variables of each type has been increased to 20,
412and it's possible to compile Exim with more. You can safely upgrade to this
413release if you already have messages on the queue with saved ACL variable
414values. However, if you downgrade from this release with messages on the queue,
415any saved ACL values they may have will be lost.
416
4172. The default value for rfc1413_query_timeout has been changed from 30s to 5s.
47ca6d6c
PH
418
419
8857ccfd
PH
420Exim version 4.54
421-----------------
422
423There was a problem with 4.52/TF/02 in that a "name=" option on control=
424submission terminated at the next slash, thereby not allowing for slashes in
425the name. This has been changed so that "name=" takes the rest of the string as
426its data. It must therefore be the last option.
427
428
f3f0a1f0
PH
429Version 4.53
430------------
431
432If you are using the experimental Domain Keys support, you must upgrade to
433at least libdomainkeys 0.67 in order to run this release of Exim.
434
435
b5aea5e1
PH
436Version 4.51
437------------
438
c688b954
PH
4391. The format in which GnuTLS parameters are cached (in the file gnutls-params
440in the spool directory) has been changed. The new format can also be generated
b5aea5e1
PH
441externally, so it is now possible to update the values from outside Exim. This
442has been implemented in an upwards, BUT NOT downwards, compatible manner.
443Upgrading should be seamless: when Exim finds that it cannot understand an
444existing cache file, it generates new parameters and writes them to the cache
445in the new format. If, however, you downgrade from 4.51 to a previous release,
446you MUST delete the gnutls-params file in the spool directory, because the
447older Exim will not recognize the new format.
448
c688b954
PH
4492. When doing a callout as part of verifying an address, Exim was not paying
450attention to any local part prefix or suffix that was matched by the router
451that accepted the address. It now behaves in the same way as it does for
452delivery: the affixes are removed from the local part unless
453rcpt_include_affixes is set on the transport. If you have a configuration that
454uses prefixes or suffixes on addresses that could be used for callouts, and you
455want the affixes to be retained, you must make sure that rcpt_include_affixes
456is set on the transport.
457
447d236c
PH
4583. Bounce and delay warning messages no longer contain details of delivery
459errors, except for explicit messages (e.g. generated by :fail:) and SMTP
460responses from remote hosts.
461
b5aea5e1 462
8b5af54d
PH
463Version 4.50
464------------
465
4964e932
PH
466The exicyclog script has been updated to use three-digit numbers in rotated log
467files if the maximum number to keep is greater than 99. If you are already
468keeping more than 99, there will be an incompatible change when you upgrade.
469You will probably want to rename your old log files to the new form before
c3af992c 470running the new exicyclog.
8b5af54d
PH
471
472
0f4f2a88
PH
473Version 4.42
474------------
475
476RFC 3848 specifies standard names for the "with" phrase in Received: header
477lines when AUTH and/or TLS are in use. This is the "received protocol"
478field. Exim used to use "asmtp" for authenticated SMTP, without any
479indication (in the protocol name) for TLS use. Now it follows the RFC and
480uses "esmtpa" if the connection is authenticated, "esmtps" if it is
481encrypted, and "esmtpsa" if it is both encrypted and authenticated. These names
482appear in log lines as well as in Received: header lines.
483
484
485Version 4.34
486------------
487
488Change 4.31/2 gave problems to data ACLs and local_scan() functions that
489expected to see a Received: header. I have changed to yet another scheme. The
490Received: header is now generated after the body is received, but before the
491ACL or local_scan() is called. After they have run, the timestamp in the
492Received: header is updated.
493
494Thus, change (a) of 4.31/2 has been reversed, but change (b) is still true,
495which is lucky, since I decided it was a bug fix.
496
497
498Version 4.33
499------------
500
501If an expansion in a condition on a "warn" statement fails because a lookup
502defers, the "warn" statement is abandoned, and the next ACL statement is
503processed. Previously this caused the whole ACL to be aborted.
504
505
506Version 4.32
507------------
508
509Change 4.31/2 has been reversed, as it proved contentious. Recipient callout
510verification now uses <> in the MAIL command by default, as it did before. A
511new callout option, "use_sender", has been added to request the other
512behaviour.
513
514
515Version 4.31
516------------
517
5181. If you compile Exim to use GnuTLS, it now requires the use of release 1.0.0
519 or greater. The interface to the obsolete 0.8.x releases is no longer
520 supported. There is one externally visible change: the format for the
521 display of Distinguished Names now uses commas as a separator rather than a
522 slash. This is to comply with RFC 2253.
523
5242. When a message is received, the Received: header line is now generated when
525 reception is complete, instead of at the start of reception. For messages
526 that take a long time to come in, this changes the meaning of the timestamp.
527 There are several side-effects of this change:
528
529 (a) If a message is rejected by a DATA or non-SMTP ACL, or by local_scan(),
530 the logged header lines no longer include the local Received: line,
531 because it has not yet been created. If the message is a non-SMTP one,
532 and the error is processed by sending a message to the sender, the copy
533 of the original message that is returned does not have an added
534 Received: line.
535
536 (b) When a filter file is tested using -bf, no additional Received: header
537 is added to the test message. After some thought, I decided that this
538 is a bug fix.
539
540 The contents of $received_for are not affected by this change. This
541 variable still contains the single recipient of a message, copied after
542 addresses have been rewritten, but before local_scan() is run.
543
5442. Recipient callout verification, like sender verification, was using <> in
545 the MAIL FROM command. This isn't really the right thing, since the actual
546 sender may affect whether the remote host accepts the recipient or not. I
547 have changed it to use the actual sender in the callout; this means that
548 the cache record is now keyed on a recipient/sender pair, not just the
549 recipient address. There doesn't seem to be a real danger of callout loops,
550 since a callout by the remote host to check the sender would use <>.
551
552
553Version 4.30
554------------
555
5561. I have abolished timeout_DNS as an error that can be detected in retry
557 rules, because it has never worked. Despite the fact that it has been
558 documented since at least release 1.62, there was no code to support it.
559 If you have used it in your retry rules, you will now get a warning message
560 to the log and panic log. It is now treated as plain "timeout".
561
5622. After discussion on the mailing list, Exim no longer adds From:, Date:, or
563 Message-Id: header lines to messages that do not originate locally, that is,
564 messages that have an associated sending host address.
565
5663. When looking up a host name from an IP address, Exim now tries the DNS
567 first, and only if that fails does it use gethostbyaddr() (or equivalent).
568 This change was made because on some OS, not all the names are given for
569 addresses with multiple PTR records via the gethostbyaddr() interface. The
570 order of lookup can be changed by setting host_lookup_order.
571
572
573Version 4.23
574------------
575
5761. The new FIXED_NEVER_USERS build-time option creates a list of "never users"
577 that cannot be overridden. The default in the distributed EDITME is "root".
578 If for some reason you were (against advice) running deliveries as root, you
579 will have to ensure that FIXED_NEVER_USERS is not set in your
580 Local/Makefile.
581
5822. The ${quote: operator now quotes an empty string, which it did not before.
583
5843. Version 4.23 saves the contents of the ACL variables with the message, so
585 that they can be used later. If one of these variables contains a newline,
586 there will be a newline character in the spool that will not be interpreted
4c04137d 587 correctly by a previous version of Exim. (Exim ignores keyed spool file
0f4f2a88
PH
588 items that it doesn't understand - precisely for this kind of problem - but
589 it expects them all to be on one line.)
590
591 So the bottom line is: if you have newlines in your ACL variables, you
592 cannot retreat from 4.23.
593
594
595Version 4.21
596------------
597
5981. The idea of the "warn" ACL verb is that it adds a header or writes to the
599 log only when "message" or "log_message" are set. However, if one of the
600 conditions was an address verification, or a call to a nested ACL, the
601 messages generated by the underlying test were being passed through. This
602 no longer happens. The underlying message is available in $acl_verify_
603 message for both "message" and "log_message" expansions, so it can be
604 passed through if needed.
605
6062. The way that the $h_ (and $header_) expansions work has been changed by the
607 addition of RFC 2047 decoding. See the main documentation (the NewStuff file
608 until release 4.30, then the manual) for full details. Briefly, there are
609 now three forms:
610
611 $rh_xxx: and $rheader_xxx: give the original content of the header
612 line(s), with no processing at all.
613
614 $bh_xxx: and $bheader_xxx: remove leading and trailing white space, and
615 then decode base64 or quoted-printable "words" within the header text,
616 but do not do charset translation.
617
618 $h_xxx: and $header_xxx: attempt to translate the $bh_ string to a
619 standard character set.
620
621 If you have previously been using $h_ expansions to access the raw
622 characters, you should change to $rh_ instead.
623
6243. When Exim creates an RFC 2047 encoded word in a header line, it labels it
625 with the default character set from the headers_charset option instead of
626 always using iso-8859-1.
627
6284. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
629 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
630 called TMPDIR, and if it finds it is different, it changes its value.
631
6325. Following a discussion on the list, the rules by which Exim recognises line
633 endings on incoming messages have been changed. The -dropcr and drop_cr
634 options are now no-ops, retained only for backwards compatibility. The
635 following line terminators are recognized: LF CRLF CR. However, special
636 processing applies to CR:
637
638 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
639 nor a local message in the state where . is a terminator.
640
641 (ii) If a bare CR is encountered in a header line, an extra space is added
642 after the line terminator so as not to end the header. The reasoning
643 behind this is that bare CRs in header lines are most likely either
644 to be mistakes, or people trying to play silly games.
645
6466. The code for using daemon_smtp_port, local_interfaces, and the -oX options
647 has been reorganized. It is supposed to be backwards compatible, but it is
648 mentioned here just in case I've screwed up.
649
650
651
652Version 4.20
653------------
654
6551. I have tidied and re-organized the code that uses alarm() for imposing time
656 limits on various things. It shouldn't affect anything, but if you notice
657 processes getting stuck, it may be that I've broken something.
658
6592. The "arguments" log selector now also logs the current working directory
660 when Exim is called.
661
6623. An incompatible change has been made to the appendfile transport. This
663 affects the case when it is used for file deliveries that are set up by
664 .forward and filter files. Previously, any settings of the "file" or
665 "directory" options were ignored. It is hoped that, like the address_file
666 transport in the default configuration, these options were never in fact set
667 on such transports, because they were of no use.
668
669 Now, if either of these options is set, it is used. The path that is passed
670 by the router is in $address_file (this is not new), so it can be used as
671 part of a longer path, or modified in any other way that expansion permits.
672
673 If neither "file" nor "directory" is set, the behaviour is unchanged.
674
6754. Related to the above: in a filter, if a "save" command specifies a non-
676 absolute path, the value of $home/ is pre-pended. This no longer happens if
677 $home is unset or is set to an empty string.
678
6795. Multiple file deliveries from a filter or .forward file can never be
680 batched; the value of batch_max on the transport is ignored for file
681 deliveries. I'm assuming that nobody ever actually set batch_max on the
682 address_file transport - it would have had odd effects previously.
683
6846. DESTDIR is the more common variable that ROOT for use when installing
685 software under a different root filing system. The Exim install script now
686 recognizes DESTDIR first; if it is not set, ROOT is used.
687
6887. If DESTDIR is set when installing Exim, it no longer prepends its value to
689 the path of the system aliases file that appears in the default
690 configuration (when a default configuration is installed). If an aliases
691 file is actually created, its name *does* use the prefix.
692
693
694Version 4.14
695------------
696
6971. The default for the maximum number of unknown SMTP commands that Exim will
698accept before dropping a connection has been reduced from 5 to 3. However, you
699can now change the value by setting smtp_max_unknown_commands.
700
7012. The ${quote: operator has been changed so that it turns newline and carriage
702return characters into \n and \r, respectively.
703
7043. The file names used for maildir messages now include the microsecond time
705fraction as well as the time in seconds, to cope with systems where the process
706id can be re-used within the same second. The format is now
707
708 <time>.H<microsec>P<pid>.<host>
709
710This should be a compatible change, but is noted here just in case.
711
7124. The rules for creating message ids have changed, to cope with systems where
713the process id can be re-used within the same second. The format, however, is
714unchanged, so this should not cause any problems, except as noted in the next
715item.
716
7175. The maximum value for localhost_number has been reduced from 255 to 16, in
718order to implement the new message id rules. For operating systems that have
719case-insensitive file systems (Cygwin and Darwin), the limit is 10.
720
7216. verify = header_syntax was allowing unqualified addresses in all cases. Now
722it allows them only for locally generated messages and from hosts that match
723sender_unqualified_hosts or recipient_unqualified_hosts, respectively.
724
7257. For reasons lost in the mists of time, when a pipe transport was run, the
726environment variable MESSAGE_ID was set to the message ID preceded by 'E' (the
727form used in Message-ID: header lines). The 'E' has been removed.
728
729
730Version 4.11
731------------
732
7331. The handling of lines in the configuration file has changed. Previously,
734macro expansion was applied to logical lines, after continuations had been
735joined on. This meant that it could not be used in .include lines, which are
736handled as physical rather than logical lines. Macro expansion is now done on
737physical lines rather than logical lines. This means there are two
738incompatibilities:
739
740 (a) A macro that expands to # to turn a line into a comment now applies only
741 to the physical line where it appears. Previously, it would have caused
742 any following continuations also to be ignored.
743
744 (b) A macro name can no longer be split over the boundary between a line and
745 its continuation. Actually, this is more of a bug fix. :-)
746
7472. The -D command line option must now all be within one command line item.
748This makes it possible to use -D to set a macro to the empty string by commands
749such as
750
751 exim -DABC ...
752 exim -DABC= ...
753
754Previously, these items would have moved on to the next item on the command
755line. To include spaces in a macro definition item, quotes must be used, in
756which case you can also have spaces after -D and surrounding the equals. For
757example:
758
759 exim '-D ABC = something' ...
760
7613. The way that addresses that redirect to themselves are handled has been
762changed, in order to fix an obscure bug. This should not cause any problems
763except in the case of wanting to go back from a 4.11 (or later) release to an
764earlier release. If there are undelivered messages on the spool that contain
765addresses which redirect to themselves, and the redirected addresses have
766already been delivered, you might get a duplicate delivery if you revert to an
767earlier Exim.
768
7694. The default way of looking up IP addresses for hosts in the manualroute and
770queryprogram routers has been changed. If "byname" or "bydns" is explicitly
771specified, there is no change, but if no method is specified, Exim now behaves
772as follows:
773
774 First, a DNS lookup is done. If this yields anything other than
775 HOST_NOT_FOUND, that result is used. Otherwise, Exim goes on to try a call to
776 getipnodebyname() (or gethostbyname() on older systems) and the result of the
777 lookup is the result of that call.
778
779This change has been made because it has been discovered that on some systems,
780if a DNS lookup called via getipnodebyname() times out, HOST_NOT_FOUND is
781returned instead of TRY_AGAIN. Thus, it is safest to try a DNS lookup directly
782first, and only if that gives a definite "no such host" to try the local
783function.
784
7855. In fixing the minor security problem with pid_file_path, I have removed some
786backwards-compatible (undocumented) code which was present to ease conversion
787from Exim 3. In Exim 4, pid_file_path is a literal; in Exim 3 it was allowed to
788contain "%s", which was replaced by the port number for daemons listening on
789non-standard ports. In Exim 4, such daemons do not write a pid file. The
790backwards compatibility feature was to replace "%s" by nothing if it occurred
791in an Exim 4 setting of pid_file_path. The bug was in this code. I have solved
792the problem by removing the backwards compatibility feature. Thus, if you still
793have "%s" somewhere in a setting of pid_file_path, you should remove it.
794
7956. There has been an extension to lsearch files. The keys in these files may
796now be quoted in order to allow for whitespace and colons in them. This means
797that if you were previously using keys that began with a doublequote, you will
798now have to wrap them with extra quotes and escape the internal quotes. The
799possibility that anybody is actually doing this seems extremely remote, but it
800is documented just in case.
801
802
803Version 4.10
804------------
805
806The build-time parameter EXIWHAT_KILL_ARG has been renamed EXIWHAT_KILL_SIGNAL
807to better reflect its function. The OS-specific files have been updated. Only
808if you have explicitly set this in your Makefile (highly unlikely) do you need
809to change anything.
810
811****