Docs: fix smtp transport TFO option indexing
[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
f59aaaaa
PP
29Exim version 4.88
30-----------------
31
32 * The "demime" ACL condition, deprecated for the past 10 years, has
33 now been removed.
34
35 * Old GnuTLS configuration options "gnutls_require_kx", "gnutls_require_mac",
36 and "gnutls_require_protocols" have now been removed. (Inoperative from
37 4.80, per below; logging warnings since 4.83, again per below).
38
39
c50e5871
JH
40Exim version 4.83
41-----------------
42
43 * SPF condition results renamed "permerror" and "temperror". The old
44 names are still accepted for back-compatability, for this release.
45
46 * TLS details are now logged on rejects, subject to log selectors.
47
48 * Items in headers_remove lists must now have any embedded list-separators
49 doubled.
50
51 * Attempted use of the deprecated options "gnutls_require_kx" et. al.
52 now result in logged warning.
53
54
2c422e6f 55Exim version 4.82
a5f239e4
PP
56-----------------
57
2519e60d 58 * New option gnutls_allow_auto_pkcs11 defaults false; if you have GnuTLS 2.12.0
a5f239e4
PP
59 or later and do want PKCS11 modules to be autoloaded, then set this option.
60
8b260705
PP
61 * A per-transport wait-<name> database is no longer updated if the transport
62 sets "connection_max_messages" to 1, as it can not be used and causes
63 unnecessary serialisation and load. External tools tracking the state of
64 Exim by the hints databases may need modification to take this into account.
65
b10e4ec2
TL
66 * The av_scanner option can now accept multiple clamd TCP targets, all other
67 setting limitations remain.
68
a5f239e4 69
83e2f8a2 70Exim version 4.80
dec5017e
PP
71-----------------
72
83e2f8a2
PP
73 * BEWARE backwards-incompatible changes in SSL libraries, thus the version
74 bump. See points below for details.
c2b063d4 75 Also an LDAP data returned format change.
83e2f8a2 76
dec5017e
PP
77 * The value of $tls_peerdn is now print-escaped when written to the spool file
78 in a -tls_peerdn line, and unescaped when read back in. We received reports
79 of values with embedded newlines, which caused spool file corruption.
80
81 If you have a corrupt spool file and you wish to recover the contents after
82 upgrading, then lock the message, replace the new-lines that should be part
83 of the -tls_peerdn line with the two-character sequence \n and then unlock
84 the message. No tool has been provided as we believe this is a rare
85 occurence.
86
f0f5a555
PP
87 * For OpenSSL, SSLv2 is now disabled by default. (GnuTLS does not support
88 SSLv2). RFC 6176 prohibits SSLv2 and some informal surveys suggest no
89 actual usage. You can re-enable with the "openssl_options" Exim option,
90 in the main configuration section. Note that supporting SSLv2 exposes
91 you to ciphersuite downgrade attacks.
92
c80c5570
PP
93 * With OpenSSL 1.0.1+, Exim now supports TLS 1.1 and TLS 1.2. If built
94 against 1.0.1a then you will get a warning message and the
95 "openssl_options" value will not parse "no_tlsv1_1": the value changes
96 incompatibly between 1.0.1a and 1.0.1b, because the value chosen for 1.0.1a
97 is infelicitous. We advise avoiding 1.0.1a.
98
99 "openssl_options" gains "no_tlsv1_1", "no_tlsv1_2" and "no_compression".
100
da3ad30d 101 COMPATIBILITY WARNING: The default value of "openssl_options" is no longer
f0f5a555
PP
102 "+dont_insert_empty_fragments". We default to "+no_sslv2".
103 That old default was grandfathered in from before openssl_options became a
104 configuration option.
da3ad30d
PP
105 Empty fragments are inserted by default through TLS1.0, to partially defend
106 against certain attacks; TLS1.1+ change the protocol so that this is not
107 needed. The DIEF SSL option was required for some old releases of mail
108 clients which did not gracefully handle the empty fragments, and was
109 initially set in Exim release 4.31 (see ChangeLog, item 37).
110
111 If you still have affected mail-clients, and you see SSL protocol failures
112 with this release of Exim, set:
113 openssl_options = +dont_insert_empty_fragments
114 in the main section of your Exim configuration file. You're trading off
115 security for compatibility. Exim is now defaulting to higher security and
116 rewarding more modern clients.
117
3375e053
PP
118 If the option tls_dhparams is set and the parameters loaded from the file
119 have a bit-count greater than the new option tls_dh_max_bits, then the file
120 will now be ignored. If this affects you, raise the tls_dh_max_bits limit.
121 We suspect that most folks are using dated defaults and will not be affected.
122
5407bfff
JH
123 * Ldap lookups returning multi-valued attributes now separate the attributes
124 with only a comma, not a comma-space sequence. Also, an actual comma within
125 a returned attribute is doubled. This makes it possible to parse the
126 attribute as a comma-separated list. Note the distinction from multiple
127 attributes being returned, where each one is a name=value pair.
c80c5570 128
c2b063d4
PP
129 If you are currently splitting the results from LDAP upon a comma, then you
130 should check carefully to see if adjustments are needed.
131
132 This change lets cautious folks distinguish "comma used as separator for
133 joining values" from "comma inside the data".
134
9cbad13b
PP
135 * accept_8bitmime now defaults on, which is not RFC compliant but is better
136 suited to today's Internet. See http://cr.yp.to/smtp/8bitmime.html for a
137 sane rationale. Those who wish to be strictly RFC compliant, or know that
138 they need to talk to servers that are not 8-bit-clean, now need to take
139 explicit configuration action to default this option off. This is not a
140 new option, you can safely force it off before upgrading, to decouple
141 configuration changes from the binary upgrade while remaining RFC compliant.
142
f90a9fd1
PP
143 * The GnuTLS support has been mostly rewritten, to use APIs which don't cause
144 deprecation warnings in GnuTLS 2.12.x. As part of this, these three options
145 are no longer supported:
17c76198
PP
146
147 gnutls_require_kx
148 gnutls_require_mac
149 gnutls_require_protocols
150
f90a9fd1
PP
151 Their functionality is entirely subsumed into tls_require_ciphers. In turn,
152 tls_require_ciphers is no longer an Exim list and is not parsed by Exim, but
153 is instead given to gnutls_priority_init(3), which expects a priority string;
154 this behaviour is much closer to the OpenSSL behaviour. See:
17c76198 155
42bfef1e 156 http://www.gnutls.org/manual/html_node/Priority-Strings.html
17c76198
PP
157
158 for fuller documentation of the strings parsed. The three gnutls_require_*
159 options are still parsed by Exim and, for this release, silently ignored.
160 A future release will add warnings, before a later still release removes
161 parsing entirely and the presence of the options will be a configuration
162 error.
163
83e2f8a2
PP
164 Note that by default, GnuTLS will not accept RSA-MD5 signatures in chains.
165 A tls_require_ciphers value of NORMAL:%VERIFY_ALLOW_SIGN_RSA_MD5 may
166 re-enable support, but this is not supported by the Exim maintainers.
167 Our test suite no longer includes MD5-based certificates.
168
17c76198
PP
169 This rewrite means that Exim will continue to build against GnuTLS in the
170 future, brings Exim closer to other GnuTLS applications and lets us add
171 support for SNI and other features more readily. We regret that it wasn't
172 feasible to retain the three dropped options.
173
3375e053
PP
174 * If built with TLS support, then Exim will now validate the value of
175 the main section tls_require_ciphers option at start-up. Before, this
176 would cause a STARTTLS 4xx failure, now it causes a failure to start.
177 Running with a broken configuration which causes failures that may only
178 be left in the logs has been traded off for something more visible. This
179 change makes an existing problem more prominent, but we do not believe
180 anyone would deliberately be running with an invalid tls_require_ciphers
181 option.
182
183 This also means that library linkage issues caused by conflicts of some
184 kind might take out the main daemon, not just the delivery or receiving
185 process. Conceivably some folks might prefer to continue delivering
186 mail plaintext when their binary is broken in this way, if there is a
187 server that is a candidate to receive such mails that does not advertise
188 STARTTLS. Note that Exim is typically a setuid root binary and given
189 broken linkage problems that cause segfaults, we feel it is safer to
190 fail completely. (The check is not done as root, to ensure that problems
191 here are not made worse by the check).
192
a799883d
PP
193 * The "tls_dhparam" option has been updated, so that it can now specify a
194 path or an identifier for a standard DH prime from one of a few RFCs.
195 The default for OpenSSL is no longer to not use DH but instead to use
196 one of these standard primes. The default for GnuTLS is no longer to use
197 a file in the spool directory, but to use that same standard prime.
198 The option is now used by GnuTLS too. If it points to a path, then
199 GnuTLS will use that path, instead of a file in the spool directory;
200 GnuTLS will attempt to create it if it does not exist.
201
202 To preserve the previous behaviour of generating files in the spool
203 directory, set "tls_dhparam = historic". Since prior releases of Exim
204 ignored tls_dhparam when using GnuTLS, this can safely be done before
205 the upgrade.
206
207
dec5017e 208
c566dd90
PP
209Exim version 4.77
210-----------------
211
212 * GnuTLS will now attempt to use TLS 1.2 and TLS 1.1 before TLS 1.0 and SSL3,
213 if supported by your GnuTLS library. Use the existing
214 "gnutls_require_protocols" option to downgrade this if that will be a
215 problem. Prior to this release, supported values were "TLS1" and "SSL3",
216 so you should be able to update configuration prior to update.
217
0215ec07
PP
218 [nb: gnutls_require_protocols removed in Exim 4.80, instead use
219 tls_require_ciphers to provide a priority string; see notes above]
220
39257585
PP
221 * The match_<type>{string1}{string2} expansion conditions no longer subject
222 string2 to string expansion, unless Exim was built with the new
223 "EXPAND_LISTMATCH_RHS" option. Too many people have inadvertently created
224 insecure configurations that way. If you need the functionality and turn on
225 that build option, please let the developers know, and know why, so we can
226 try to provide a safer mechanism for you.
227
a7ee5955
PP
228 The match{}{} expansion condition (for regular expressions) is NOT affected.
229 For match_<type>{s1}{s2}, all list functionality is unchanged. The only
230 change is that a '$' appearing in s2 will not trigger expansion, but instead
231 will be treated as a literal $ sign; the effect is very similar to having
232 wrapped s2 with \N...\N. If s2 contains a named list and the list definition
233 uses $expansions then those _will_ be processed as normal. It is only the
234 point at which s2 is read where expansion is inhibited.
235
236 If you are trying to test if two email addresses are equal, use eqi{s1}{s2}.
237 If you are testing if the address in s1 occurs in the list of items given
238 in s2, either use the new inlisti{s1}{s2} condition (added in 4.77) or use
239 the pre-existing forany{s2}{eqi{$item}{s1}} condition.
240
c566dd90 241
e6acf885
PP
242Exim version 4.74
243-----------------
244
4050a044
PP
245 * The integrated support for dynamically loadable lookup modules has an ABI
246 change from the modules supported by some OS vendors through an unofficial
247 patch. Don't try to mix & match.
248
249 * Some parts of the build system are now beginning to assume that the host
250 environment is POSIX. If you're building on a system where POSIX tools are
251 not the default, you might have an easier time if you switch to the POSIX
252 tools. Feel free to report non-POSIX issues as a request for a feature
253 enhancement, but if the POSIX variants are available then the fix will
254 probably just involve some coercion. See the README instructions for
255 building on such hosts.
e6acf885
PP
256
257
258Exim version 4.73
259-----------------
260
261 * The Exim run-time user can no longer be root; this was always
262 strongly discouraged, but is now prohibited both at build and
263 run-time. If you need Exim to run routinely as root, you'll need to
264 patch the source and accept the risk. Here be dragons.
265
266 * Exim will no longer accept a configuration file owned by the Exim
267 run-time user, unless that account is explicitly the value in
268 CONFIGURE_OWNER, which we discourage. Exim now checks to ensure that
a7ee5955 269 files are not writeable by other accounts.
e6acf885
PP
270
271 * The ALT_CONFIG_ROOT_ONLY build option is no longer optional and is forced
272 on; the Exim user can, by default, no longer use -C/-D and retain privilege.
273 Two new build options mitigate this.
274
275 * TRUSTED_CONFIG_LIST defines a file containing a whitelist of config
276 files that are trusted to be selected by the Exim user; one per line.
277 This is the recommended approach going forward.
278
279 * WHITELIST_D_MACROS defines a colon-separated list of macro names which
280 the Exim run-time user may safely pass without dropping privileges.
281 Because changes to this involve a recompile, this is not the recommended
282 approach but may ease transition. The values of the macros, when
43236f35 283 overridden, are constrained to match this regex: ^[A-Za-z0-9_/.-]*$
e6acf885
PP
284
285 * The system_filter_user option now defaults to the Exim run-time user,
286 rather than root. You can still set it explicitly to root and this
287 can be done with prior versions too, letting you roll versions
288 without needing to change this configuration option.
289
290 * ClamAV must be at least version 0.95 unless WITH_OLD_CLAMAV_STREAM is
291 defined at build time.
292
293
2eda48c6
TK
294Exim version 4.70
295-----------------
296
2971. Experimental Yahoo! Domainkeys support has been dropped in this release.
298It has been superceded by a native implementation of its successor DKIM.
299
3002. Up to version 4.69, Exim came with an embedded version of the PCRE library.
301As of 4.70, this is no longer the case. To compile Exim, you will need PCRE
a7ee5955 302installed. Most OS distributions have ready-made library and development
2eda48c6 303packages.
0f4f2a88
PH
304
305
3c223874
PH
306Exim version 4.68
307-----------------
308
3091. The internal implementation of the database keys that are used for ACL
310ratelimiting has been tidied up. This means that an update to 4.68 might cause
311Exim to "forget" previous rates that it had calculated, and reset them to zero.
312
313
26da7e20
PH
314Exim version 4.64
315-----------------
316
3171. Callouts were setting the name used for EHLO/HELO from $smtp_active_
318hostname. This is wrong, because it relates to the incoming message (and
319probably the interface on which it is arriving) and not to the outgoing
320callout (which could be using a different interface). This has been
321changed to use the value of the helo_data option from the smtp transport
322instead - this is what is used when a message is actually being sent. If
323there is no remote transport (possible with a router that sets up host
324addresses), $smtp_active_hostname is used. This change is mentioned here in
325case somebody is relying on the use of $smtp_active_hostname.
326
ebc9d865
PH
3272. A bug has been fixed that might just possibly be something that is relied on
328in some configurations. In expansion items such as ${if >{xxx}{yyy}...} an
329empty string (that is {}) was being interpreted as if it was {0} and therefore
330treated as the number zero. From release 4.64, such strings cause an error
331because a decimal number, possibly followed by K or M, is required (as has
332always been documented).
333
b6c27e28
PH
3343. There has been a change to the GnuTLS support (ChangeLog/PH/20) to improve
335Exim's performance. Unfortunately, this has the side effect of being slightly
336non-upwards compatible for versions 4.50 and earlier. If you are upgrading from
337one of these earlier versions and you use GnuTLS, you must remove the file
338called gnutls-params in Exim's spool directory. If you don't do this, you will
339see this error:
340
341 TLS error on connection from ... (DH params import): Base64 decoding error.
342
343Removing the file causes Exim to recompute the relevant encryption parameters
344and cache them in the new format that was introduced for release 4.51 (May
3452005). If you are upgrading from release 4.51 or later, there should be no
346problem.
347
26da7e20 348
a5bd321b
PH
349Exim version 4.63
350-----------------
351
352When an SMTP error message is specified in a "message" modifier in an ACL, or
353in a :fail: or :defer: message in a redirect router, Exim now checks the start
354of the message for an SMTP error code. This consists of three digits followed
355by a space, optionally followed by an extended code of the form n.n.n, also
356followed by a space. If this is the case and the very first digit is the same
357as the default error code, the code from the message is used instead. If the
358very first digit is incorrect, a panic error is logged, and the default code is
359used. This is an incompatible change, but it is not expected to affect many (if
360any) configurations. It is possible to suppress the use of the supplied code in
361a redirect router by setting the smtp_error_code option false. In this case,
362any SMTP code is quietly ignored.
363
364
47ca6d6c
PH
365Exim version 4.61
366-----------------
367
8def5aaf
PH
3681. The default number of ACL variables of each type has been increased to 20,
369and it's possible to compile Exim with more. You can safely upgrade to this
370release if you already have messages on the queue with saved ACL variable
371values. However, if you downgrade from this release with messages on the queue,
372any saved ACL values they may have will be lost.
373
3742. The default value for rfc1413_query_timeout has been changed from 30s to 5s.
47ca6d6c
PH
375
376
8857ccfd
PH
377Exim version 4.54
378-----------------
379
380There was a problem with 4.52/TF/02 in that a "name=" option on control=
381submission terminated at the next slash, thereby not allowing for slashes in
382the name. This has been changed so that "name=" takes the rest of the string as
383its data. It must therefore be the last option.
384
385
f3f0a1f0
PH
386Version 4.53
387------------
388
389If you are using the experimental Domain Keys support, you must upgrade to
390at least libdomainkeys 0.67 in order to run this release of Exim.
391
392
b5aea5e1
PH
393Version 4.51
394------------
395
c688b954
PH
3961. The format in which GnuTLS parameters are cached (in the file gnutls-params
397in the spool directory) has been changed. The new format can also be generated
b5aea5e1
PH
398externally, so it is now possible to update the values from outside Exim. This
399has been implemented in an upwards, BUT NOT downwards, compatible manner.
400Upgrading should be seamless: when Exim finds that it cannot understand an
401existing cache file, it generates new parameters and writes them to the cache
402in the new format. If, however, you downgrade from 4.51 to a previous release,
403you MUST delete the gnutls-params file in the spool directory, because the
404older Exim will not recognize the new format.
405
c688b954
PH
4062. When doing a callout as part of verifying an address, Exim was not paying
407attention to any local part prefix or suffix that was matched by the router
408that accepted the address. It now behaves in the same way as it does for
409delivery: the affixes are removed from the local part unless
410rcpt_include_affixes is set on the transport. If you have a configuration that
411uses prefixes or suffixes on addresses that could be used for callouts, and you
412want the affixes to be retained, you must make sure that rcpt_include_affixes
413is set on the transport.
414
447d236c
PH
4153. Bounce and delay warning messages no longer contain details of delivery
416errors, except for explicit messages (e.g. generated by :fail:) and SMTP
417responses from remote hosts.
418
b5aea5e1 419
8b5af54d
PH
420Version 4.50
421------------
422
4964e932
PH
423The exicyclog script has been updated to use three-digit numbers in rotated log
424files if the maximum number to keep is greater than 99. If you are already
425keeping more than 99, there will be an incompatible change when you upgrade.
426You will probably want to rename your old log files to the new form before
c3af992c 427running the new exicyclog.
8b5af54d
PH
428
429
0f4f2a88
PH
430Version 4.42
431------------
432
433RFC 3848 specifies standard names for the "with" phrase in Received: header
434lines when AUTH and/or TLS are in use. This is the "received protocol"
435field. Exim used to use "asmtp" for authenticated SMTP, without any
436indication (in the protocol name) for TLS use. Now it follows the RFC and
437uses "esmtpa" if the connection is authenticated, "esmtps" if it is
438encrypted, and "esmtpsa" if it is both encrypted and authenticated. These names
439appear in log lines as well as in Received: header lines.
440
441
442Version 4.34
443------------
444
445Change 4.31/2 gave problems to data ACLs and local_scan() functions that
446expected to see a Received: header. I have changed to yet another scheme. The
447Received: header is now generated after the body is received, but before the
448ACL or local_scan() is called. After they have run, the timestamp in the
449Received: header is updated.
450
451Thus, change (a) of 4.31/2 has been reversed, but change (b) is still true,
452which is lucky, since I decided it was a bug fix.
453
454
455Version 4.33
456------------
457
458If an expansion in a condition on a "warn" statement fails because a lookup
459defers, the "warn" statement is abandoned, and the next ACL statement is
460processed. Previously this caused the whole ACL to be aborted.
461
462
463Version 4.32
464------------
465
466Change 4.31/2 has been reversed, as it proved contentious. Recipient callout
467verification now uses <> in the MAIL command by default, as it did before. A
468new callout option, "use_sender", has been added to request the other
469behaviour.
470
471
472Version 4.31
473------------
474
4751. If you compile Exim to use GnuTLS, it now requires the use of release 1.0.0
476 or greater. The interface to the obsolete 0.8.x releases is no longer
477 supported. There is one externally visible change: the format for the
478 display of Distinguished Names now uses commas as a separator rather than a
479 slash. This is to comply with RFC 2253.
480
4812. When a message is received, the Received: header line is now generated when
482 reception is complete, instead of at the start of reception. For messages
483 that take a long time to come in, this changes the meaning of the timestamp.
484 There are several side-effects of this change:
485
486 (a) If a message is rejected by a DATA or non-SMTP ACL, or by local_scan(),
487 the logged header lines no longer include the local Received: line,
488 because it has not yet been created. If the message is a non-SMTP one,
489 and the error is processed by sending a message to the sender, the copy
490 of the original message that is returned does not have an added
491 Received: line.
492
493 (b) When a filter file is tested using -bf, no additional Received: header
494 is added to the test message. After some thought, I decided that this
495 is a bug fix.
496
497 The contents of $received_for are not affected by this change. This
498 variable still contains the single recipient of a message, copied after
499 addresses have been rewritten, but before local_scan() is run.
500
5012. Recipient callout verification, like sender verification, was using <> in
502 the MAIL FROM command. This isn't really the right thing, since the actual
503 sender may affect whether the remote host accepts the recipient or not. I
504 have changed it to use the actual sender in the callout; this means that
505 the cache record is now keyed on a recipient/sender pair, not just the
506 recipient address. There doesn't seem to be a real danger of callout loops,
507 since a callout by the remote host to check the sender would use <>.
508
509
510Version 4.30
511------------
512
5131. I have abolished timeout_DNS as an error that can be detected in retry
514 rules, because it has never worked. Despite the fact that it has been
515 documented since at least release 1.62, there was no code to support it.
516 If you have used it in your retry rules, you will now get a warning message
517 to the log and panic log. It is now treated as plain "timeout".
518
5192. After discussion on the mailing list, Exim no longer adds From:, Date:, or
520 Message-Id: header lines to messages that do not originate locally, that is,
521 messages that have an associated sending host address.
522
5233. When looking up a host name from an IP address, Exim now tries the DNS
524 first, and only if that fails does it use gethostbyaddr() (or equivalent).
525 This change was made because on some OS, not all the names are given for
526 addresses with multiple PTR records via the gethostbyaddr() interface. The
527 order of lookup can be changed by setting host_lookup_order.
528
529
530Version 4.23
531------------
532
5331. The new FIXED_NEVER_USERS build-time option creates a list of "never users"
534 that cannot be overridden. The default in the distributed EDITME is "root".
535 If for some reason you were (against advice) running deliveries as root, you
536 will have to ensure that FIXED_NEVER_USERS is not set in your
537 Local/Makefile.
538
5392. The ${quote: operator now quotes an empty string, which it did not before.
540
5413. Version 4.23 saves the contents of the ACL variables with the message, so
542 that they can be used later. If one of these variables contains a newline,
543 there will be a newline character in the spool that will not be interpreted
544 correctely by a previous version of Exim. (Exim ignores keyed spool file
545 items that it doesn't understand - precisely for this kind of problem - but
546 it expects them all to be on one line.)
547
548 So the bottom line is: if you have newlines in your ACL variables, you
549 cannot retreat from 4.23.
550
551
552Version 4.21
553------------
554
5551. The idea of the "warn" ACL verb is that it adds a header or writes to the
556 log only when "message" or "log_message" are set. However, if one of the
557 conditions was an address verification, or a call to a nested ACL, the
558 messages generated by the underlying test were being passed through. This
559 no longer happens. The underlying message is available in $acl_verify_
560 message for both "message" and "log_message" expansions, so it can be
561 passed through if needed.
562
5632. The way that the $h_ (and $header_) expansions work has been changed by the
564 addition of RFC 2047 decoding. See the main documentation (the NewStuff file
565 until release 4.30, then the manual) for full details. Briefly, there are
566 now three forms:
567
568 $rh_xxx: and $rheader_xxx: give the original content of the header
569 line(s), with no processing at all.
570
571 $bh_xxx: and $bheader_xxx: remove leading and trailing white space, and
572 then decode base64 or quoted-printable "words" within the header text,
573 but do not do charset translation.
574
575 $h_xxx: and $header_xxx: attempt to translate the $bh_ string to a
576 standard character set.
577
578 If you have previously been using $h_ expansions to access the raw
579 characters, you should change to $rh_ instead.
580
5813. When Exim creates an RFC 2047 encoded word in a header line, it labels it
582 with the default character set from the headers_charset option instead of
583 always using iso-8859-1.
584
5854. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
586 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
587 called TMPDIR, and if it finds it is different, it changes its value.
588
5895. Following a discussion on the list, the rules by which Exim recognises line
590 endings on incoming messages have been changed. The -dropcr and drop_cr
591 options are now no-ops, retained only for backwards compatibility. The
592 following line terminators are recognized: LF CRLF CR. However, special
593 processing applies to CR:
594
595 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
596 nor a local message in the state where . is a terminator.
597
598 (ii) If a bare CR is encountered in a header line, an extra space is added
599 after the line terminator so as not to end the header. The reasoning
600 behind this is that bare CRs in header lines are most likely either
601 to be mistakes, or people trying to play silly games.
602
6036. The code for using daemon_smtp_port, local_interfaces, and the -oX options
604 has been reorganized. It is supposed to be backwards compatible, but it is
605 mentioned here just in case I've screwed up.
606
607
608
609Version 4.20
610------------
611
6121. I have tidied and re-organized the code that uses alarm() for imposing time
613 limits on various things. It shouldn't affect anything, but if you notice
614 processes getting stuck, it may be that I've broken something.
615
6162. The "arguments" log selector now also logs the current working directory
617 when Exim is called.
618
6193. An incompatible change has been made to the appendfile transport. This
620 affects the case when it is used for file deliveries that are set up by
621 .forward and filter files. Previously, any settings of the "file" or
622 "directory" options were ignored. It is hoped that, like the address_file
623 transport in the default configuration, these options were never in fact set
624 on such transports, because they were of no use.
625
626 Now, if either of these options is set, it is used. The path that is passed
627 by the router is in $address_file (this is not new), so it can be used as
628 part of a longer path, or modified in any other way that expansion permits.
629
630 If neither "file" nor "directory" is set, the behaviour is unchanged.
631
6324. Related to the above: in a filter, if a "save" command specifies a non-
633 absolute path, the value of $home/ is pre-pended. This no longer happens if
634 $home is unset or is set to an empty string.
635
6365. Multiple file deliveries from a filter or .forward file can never be
637 batched; the value of batch_max on the transport is ignored for file
638 deliveries. I'm assuming that nobody ever actually set batch_max on the
639 address_file transport - it would have had odd effects previously.
640
6416. DESTDIR is the more common variable that ROOT for use when installing
642 software under a different root filing system. The Exim install script now
643 recognizes DESTDIR first; if it is not set, ROOT is used.
644
6457. If DESTDIR is set when installing Exim, it no longer prepends its value to
646 the path of the system aliases file that appears in the default
647 configuration (when a default configuration is installed). If an aliases
648 file is actually created, its name *does* use the prefix.
649
650
651Version 4.14
652------------
653
6541. The default for the maximum number of unknown SMTP commands that Exim will
655accept before dropping a connection has been reduced from 5 to 3. However, you
656can now change the value by setting smtp_max_unknown_commands.
657
6582. The ${quote: operator has been changed so that it turns newline and carriage
659return characters into \n and \r, respectively.
660
6613. The file names used for maildir messages now include the microsecond time
662fraction as well as the time in seconds, to cope with systems where the process
663id can be re-used within the same second. The format is now
664
665 <time>.H<microsec>P<pid>.<host>
666
667This should be a compatible change, but is noted here just in case.
668
6694. The rules for creating message ids have changed, to cope with systems where
670the process id can be re-used within the same second. The format, however, is
671unchanged, so this should not cause any problems, except as noted in the next
672item.
673
6745. The maximum value for localhost_number has been reduced from 255 to 16, in
675order to implement the new message id rules. For operating systems that have
676case-insensitive file systems (Cygwin and Darwin), the limit is 10.
677
6786. verify = header_syntax was allowing unqualified addresses in all cases. Now
679it allows them only for locally generated messages and from hosts that match
680sender_unqualified_hosts or recipient_unqualified_hosts, respectively.
681
6827. For reasons lost in the mists of time, when a pipe transport was run, the
683environment variable MESSAGE_ID was set to the message ID preceded by 'E' (the
684form used in Message-ID: header lines). The 'E' has been removed.
685
686
687Version 4.11
688------------
689
6901. The handling of lines in the configuration file has changed. Previously,
691macro expansion was applied to logical lines, after continuations had been
692joined on. This meant that it could not be used in .include lines, which are
693handled as physical rather than logical lines. Macro expansion is now done on
694physical lines rather than logical lines. This means there are two
695incompatibilities:
696
697 (a) A macro that expands to # to turn a line into a comment now applies only
698 to the physical line where it appears. Previously, it would have caused
699 any following continuations also to be ignored.
700
701 (b) A macro name can no longer be split over the boundary between a line and
702 its continuation. Actually, this is more of a bug fix. :-)
703
7042. The -D command line option must now all be within one command line item.
705This makes it possible to use -D to set a macro to the empty string by commands
706such as
707
708 exim -DABC ...
709 exim -DABC= ...
710
711Previously, these items would have moved on to the next item on the command
712line. To include spaces in a macro definition item, quotes must be used, in
713which case you can also have spaces after -D and surrounding the equals. For
714example:
715
716 exim '-D ABC = something' ...
717
7183. The way that addresses that redirect to themselves are handled has been
719changed, in order to fix an obscure bug. This should not cause any problems
720except in the case of wanting to go back from a 4.11 (or later) release to an
721earlier release. If there are undelivered messages on the spool that contain
722addresses which redirect to themselves, and the redirected addresses have
723already been delivered, you might get a duplicate delivery if you revert to an
724earlier Exim.
725
7264. The default way of looking up IP addresses for hosts in the manualroute and
727queryprogram routers has been changed. If "byname" or "bydns" is explicitly
728specified, there is no change, but if no method is specified, Exim now behaves
729as follows:
730
731 First, a DNS lookup is done. If this yields anything other than
732 HOST_NOT_FOUND, that result is used. Otherwise, Exim goes on to try a call to
733 getipnodebyname() (or gethostbyname() on older systems) and the result of the
734 lookup is the result of that call.
735
736This change has been made because it has been discovered that on some systems,
737if a DNS lookup called via getipnodebyname() times out, HOST_NOT_FOUND is
738returned instead of TRY_AGAIN. Thus, it is safest to try a DNS lookup directly
739first, and only if that gives a definite "no such host" to try the local
740function.
741
7425. In fixing the minor security problem with pid_file_path, I have removed some
743backwards-compatible (undocumented) code which was present to ease conversion
744from Exim 3. In Exim 4, pid_file_path is a literal; in Exim 3 it was allowed to
745contain "%s", which was replaced by the port number for daemons listening on
746non-standard ports. In Exim 4, such daemons do not write a pid file. The
747backwards compatibility feature was to replace "%s" by nothing if it occurred
748in an Exim 4 setting of pid_file_path. The bug was in this code. I have solved
749the problem by removing the backwards compatibility feature. Thus, if you still
750have "%s" somewhere in a setting of pid_file_path, you should remove it.
751
7526. There has been an extension to lsearch files. The keys in these files may
753now be quoted in order to allow for whitespace and colons in them. This means
754that if you were previously using keys that began with a doublequote, you will
755now have to wrap them with extra quotes and escape the internal quotes. The
756possibility that anybody is actually doing this seems extremely remote, but it
757is documented just in case.
758
759
760Version 4.10
761------------
762
763The build-time parameter EXIWHAT_KILL_ARG has been renamed EXIWHAT_KILL_SIGNAL
764to better reflect its function. The OS-specific files have been updated. Only
765if you have explicitly set this in your Makefile (highly unlikely) do you need
766to change anything.
767
768****