nit: document that `fout` must be open for reading too
[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
2eec84ca
PP
29Exim version 4.89
30-----------------
31
32 * OpenSSL: oldest supported release series is now 1.0.2, which is the oldest
33 supported by the OpenSSL project. If you can build Exim with an older
34 release series, congratulations. If you can't, then upgrade.
35 The file doc/openssl.txt contains instructions for installing a current
36 OpenSSL outside the system library paths and building Exim to use it.
37
0aaf76a4
PP
38 * FreeBSD: we now always use the system iconv in libc, as all versions of
39 FreeBSD supported by the FreeBSD project provide this functionality.
40
2eec84ca 41
f59aaaaa
PP
42Exim version 4.88
43-----------------
44
45 * The "demime" ACL condition, deprecated for the past 10 years, has
46 now been removed.
47
48 * Old GnuTLS configuration options "gnutls_require_kx", "gnutls_require_mac",
49 and "gnutls_require_protocols" have now been removed. (Inoperative from
50 4.80, per below; logging warnings since 4.83, again per below).
51
52
c50e5871
JH
53Exim version 4.83
54-----------------
55
56 * SPF condition results renamed "permerror" and "temperror". The old
57 names are still accepted for back-compatability, for this release.
58
59 * TLS details are now logged on rejects, subject to log selectors.
60
61 * Items in headers_remove lists must now have any embedded list-separators
62 doubled.
63
64 * Attempted use of the deprecated options "gnutls_require_kx" et. al.
65 now result in logged warning.
66
67
2c422e6f 68Exim version 4.82
a5f239e4
PP
69-----------------
70
2519e60d 71 * New option gnutls_allow_auto_pkcs11 defaults false; if you have GnuTLS 2.12.0
a5f239e4
PP
72 or later and do want PKCS11 modules to be autoloaded, then set this option.
73
8b260705
PP
74 * A per-transport wait-<name> database is no longer updated if the transport
75 sets "connection_max_messages" to 1, as it can not be used and causes
76 unnecessary serialisation and load. External tools tracking the state of
77 Exim by the hints databases may need modification to take this into account.
78
b10e4ec2
TL
79 * The av_scanner option can now accept multiple clamd TCP targets, all other
80 setting limitations remain.
81
a5f239e4 82
83e2f8a2 83Exim version 4.80
dec5017e
PP
84-----------------
85
83e2f8a2
PP
86 * BEWARE backwards-incompatible changes in SSL libraries, thus the version
87 bump. See points below for details.
c2b063d4 88 Also an LDAP data returned format change.
83e2f8a2 89
dec5017e
PP
90 * The value of $tls_peerdn is now print-escaped when written to the spool file
91 in a -tls_peerdn line, and unescaped when read back in. We received reports
92 of values with embedded newlines, which caused spool file corruption.
93
94 If you have a corrupt spool file and you wish to recover the contents after
95 upgrading, then lock the message, replace the new-lines that should be part
96 of the -tls_peerdn line with the two-character sequence \n and then unlock
97 the message. No tool has been provided as we believe this is a rare
98 occurence.
99
f0f5a555
PP
100 * For OpenSSL, SSLv2 is now disabled by default. (GnuTLS does not support
101 SSLv2). RFC 6176 prohibits SSLv2 and some informal surveys suggest no
102 actual usage. You can re-enable with the "openssl_options" Exim option,
103 in the main configuration section. Note that supporting SSLv2 exposes
104 you to ciphersuite downgrade attacks.
105
c80c5570
PP
106 * With OpenSSL 1.0.1+, Exim now supports TLS 1.1 and TLS 1.2. If built
107 against 1.0.1a then you will get a warning message and the
108 "openssl_options" value will not parse "no_tlsv1_1": the value changes
109 incompatibly between 1.0.1a and 1.0.1b, because the value chosen for 1.0.1a
110 is infelicitous. We advise avoiding 1.0.1a.
111
112 "openssl_options" gains "no_tlsv1_1", "no_tlsv1_2" and "no_compression".
113
da3ad30d 114 COMPATIBILITY WARNING: The default value of "openssl_options" is no longer
f0f5a555
PP
115 "+dont_insert_empty_fragments". We default to "+no_sslv2".
116 That old default was grandfathered in from before openssl_options became a
117 configuration option.
da3ad30d
PP
118 Empty fragments are inserted by default through TLS1.0, to partially defend
119 against certain attacks; TLS1.1+ change the protocol so that this is not
120 needed. The DIEF SSL option was required for some old releases of mail
121 clients which did not gracefully handle the empty fragments, and was
122 initially set in Exim release 4.31 (see ChangeLog, item 37).
123
124 If you still have affected mail-clients, and you see SSL protocol failures
125 with this release of Exim, set:
126 openssl_options = +dont_insert_empty_fragments
127 in the main section of your Exim configuration file. You're trading off
128 security for compatibility. Exim is now defaulting to higher security and
129 rewarding more modern clients.
130
3375e053
PP
131 If the option tls_dhparams is set and the parameters loaded from the file
132 have a bit-count greater than the new option tls_dh_max_bits, then the file
133 will now be ignored. If this affects you, raise the tls_dh_max_bits limit.
134 We suspect that most folks are using dated defaults and will not be affected.
135
5407bfff
JH
136 * Ldap lookups returning multi-valued attributes now separate the attributes
137 with only a comma, not a comma-space sequence. Also, an actual comma within
138 a returned attribute is doubled. This makes it possible to parse the
139 attribute as a comma-separated list. Note the distinction from multiple
140 attributes being returned, where each one is a name=value pair.
c80c5570 141
c2b063d4
PP
142 If you are currently splitting the results from LDAP upon a comma, then you
143 should check carefully to see if adjustments are needed.
144
145 This change lets cautious folks distinguish "comma used as separator for
146 joining values" from "comma inside the data".
147
9cbad13b
PP
148 * accept_8bitmime now defaults on, which is not RFC compliant but is better
149 suited to today's Internet. See http://cr.yp.to/smtp/8bitmime.html for a
150 sane rationale. Those who wish to be strictly RFC compliant, or know that
151 they need to talk to servers that are not 8-bit-clean, now need to take
152 explicit configuration action to default this option off. This is not a
153 new option, you can safely force it off before upgrading, to decouple
154 configuration changes from the binary upgrade while remaining RFC compliant.
155
f90a9fd1
PP
156 * The GnuTLS support has been mostly rewritten, to use APIs which don't cause
157 deprecation warnings in GnuTLS 2.12.x. As part of this, these three options
158 are no longer supported:
17c76198
PP
159
160 gnutls_require_kx
161 gnutls_require_mac
162 gnutls_require_protocols
163
f90a9fd1
PP
164 Their functionality is entirely subsumed into tls_require_ciphers. In turn,
165 tls_require_ciphers is no longer an Exim list and is not parsed by Exim, but
166 is instead given to gnutls_priority_init(3), which expects a priority string;
167 this behaviour is much closer to the OpenSSL behaviour. See:
17c76198 168
42bfef1e 169 http://www.gnutls.org/manual/html_node/Priority-Strings.html
17c76198
PP
170
171 for fuller documentation of the strings parsed. The three gnutls_require_*
172 options are still parsed by Exim and, for this release, silently ignored.
173 A future release will add warnings, before a later still release removes
174 parsing entirely and the presence of the options will be a configuration
175 error.
176
83e2f8a2
PP
177 Note that by default, GnuTLS will not accept RSA-MD5 signatures in chains.
178 A tls_require_ciphers value of NORMAL:%VERIFY_ALLOW_SIGN_RSA_MD5 may
179 re-enable support, but this is not supported by the Exim maintainers.
180 Our test suite no longer includes MD5-based certificates.
181
17c76198
PP
182 This rewrite means that Exim will continue to build against GnuTLS in the
183 future, brings Exim closer to other GnuTLS applications and lets us add
184 support for SNI and other features more readily. We regret that it wasn't
185 feasible to retain the three dropped options.
186
3375e053
PP
187 * If built with TLS support, then Exim will now validate the value of
188 the main section tls_require_ciphers option at start-up. Before, this
189 would cause a STARTTLS 4xx failure, now it causes a failure to start.
190 Running with a broken configuration which causes failures that may only
191 be left in the logs has been traded off for something more visible. This
192 change makes an existing problem more prominent, but we do not believe
193 anyone would deliberately be running with an invalid tls_require_ciphers
194 option.
195
196 This also means that library linkage issues caused by conflicts of some
197 kind might take out the main daemon, not just the delivery or receiving
198 process. Conceivably some folks might prefer to continue delivering
199 mail plaintext when their binary is broken in this way, if there is a
200 server that is a candidate to receive such mails that does not advertise
201 STARTTLS. Note that Exim is typically a setuid root binary and given
202 broken linkage problems that cause segfaults, we feel it is safer to
203 fail completely. (The check is not done as root, to ensure that problems
204 here are not made worse by the check).
205
a799883d
PP
206 * The "tls_dhparam" option has been updated, so that it can now specify a
207 path or an identifier for a standard DH prime from one of a few RFCs.
208 The default for OpenSSL is no longer to not use DH but instead to use
209 one of these standard primes. The default for GnuTLS is no longer to use
210 a file in the spool directory, but to use that same standard prime.
211 The option is now used by GnuTLS too. If it points to a path, then
212 GnuTLS will use that path, instead of a file in the spool directory;
213 GnuTLS will attempt to create it if it does not exist.
214
215 To preserve the previous behaviour of generating files in the spool
216 directory, set "tls_dhparam = historic". Since prior releases of Exim
217 ignored tls_dhparam when using GnuTLS, this can safely be done before
218 the upgrade.
219
220
dec5017e 221
c566dd90
PP
222Exim version 4.77
223-----------------
224
225 * GnuTLS will now attempt to use TLS 1.2 and TLS 1.1 before TLS 1.0 and SSL3,
226 if supported by your GnuTLS library. Use the existing
227 "gnutls_require_protocols" option to downgrade this if that will be a
228 problem. Prior to this release, supported values were "TLS1" and "SSL3",
229 so you should be able to update configuration prior to update.
230
0215ec07
PP
231 [nb: gnutls_require_protocols removed in Exim 4.80, instead use
232 tls_require_ciphers to provide a priority string; see notes above]
233
39257585
PP
234 * The match_<type>{string1}{string2} expansion conditions no longer subject
235 string2 to string expansion, unless Exim was built with the new
236 "EXPAND_LISTMATCH_RHS" option. Too many people have inadvertently created
237 insecure configurations that way. If you need the functionality and turn on
238 that build option, please let the developers know, and know why, so we can
239 try to provide a safer mechanism for you.
240
a7ee5955
PP
241 The match{}{} expansion condition (for regular expressions) is NOT affected.
242 For match_<type>{s1}{s2}, all list functionality is unchanged. The only
243 change is that a '$' appearing in s2 will not trigger expansion, but instead
244 will be treated as a literal $ sign; the effect is very similar to having
245 wrapped s2 with \N...\N. If s2 contains a named list and the list definition
246 uses $expansions then those _will_ be processed as normal. It is only the
247 point at which s2 is read where expansion is inhibited.
248
249 If you are trying to test if two email addresses are equal, use eqi{s1}{s2}.
250 If you are testing if the address in s1 occurs in the list of items given
251 in s2, either use the new inlisti{s1}{s2} condition (added in 4.77) or use
252 the pre-existing forany{s2}{eqi{$item}{s1}} condition.
253
c566dd90 254
e6acf885
PP
255Exim version 4.74
256-----------------
257
4050a044
PP
258 * The integrated support for dynamically loadable lookup modules has an ABI
259 change from the modules supported by some OS vendors through an unofficial
260 patch. Don't try to mix & match.
261
262 * Some parts of the build system are now beginning to assume that the host
263 environment is POSIX. If you're building on a system where POSIX tools are
264 not the default, you might have an easier time if you switch to the POSIX
265 tools. Feel free to report non-POSIX issues as a request for a feature
266 enhancement, but if the POSIX variants are available then the fix will
267 probably just involve some coercion. See the README instructions for
268 building on such hosts.
e6acf885
PP
269
270
271Exim version 4.73
272-----------------
273
274 * The Exim run-time user can no longer be root; this was always
275 strongly discouraged, but is now prohibited both at build and
276 run-time. If you need Exim to run routinely as root, you'll need to
277 patch the source and accept the risk. Here be dragons.
278
279 * Exim will no longer accept a configuration file owned by the Exim
280 run-time user, unless that account is explicitly the value in
281 CONFIGURE_OWNER, which we discourage. Exim now checks to ensure that
a7ee5955 282 files are not writeable by other accounts.
e6acf885
PP
283
284 * The ALT_CONFIG_ROOT_ONLY build option is no longer optional and is forced
285 on; the Exim user can, by default, no longer use -C/-D and retain privilege.
286 Two new build options mitigate this.
287
288 * TRUSTED_CONFIG_LIST defines a file containing a whitelist of config
289 files that are trusted to be selected by the Exim user; one per line.
290 This is the recommended approach going forward.
291
292 * WHITELIST_D_MACROS defines a colon-separated list of macro names which
293 the Exim run-time user may safely pass without dropping privileges.
294 Because changes to this involve a recompile, this is not the recommended
295 approach but may ease transition. The values of the macros, when
43236f35 296 overridden, are constrained to match this regex: ^[A-Za-z0-9_/.-]*$
e6acf885
PP
297
298 * The system_filter_user option now defaults to the Exim run-time user,
299 rather than root. You can still set it explicitly to root and this
300 can be done with prior versions too, letting you roll versions
301 without needing to change this configuration option.
302
303 * ClamAV must be at least version 0.95 unless WITH_OLD_CLAMAV_STREAM is
304 defined at build time.
305
306
2eda48c6
TK
307Exim version 4.70
308-----------------
309
3101. Experimental Yahoo! Domainkeys support has been dropped in this release.
311It has been superceded by a native implementation of its successor DKIM.
312
3132. Up to version 4.69, Exim came with an embedded version of the PCRE library.
314As of 4.70, this is no longer the case. To compile Exim, you will need PCRE
a7ee5955 315installed. Most OS distributions have ready-made library and development
2eda48c6 316packages.
0f4f2a88
PH
317
318
3c223874
PH
319Exim version 4.68
320-----------------
321
3221. The internal implementation of the database keys that are used for ACL
323ratelimiting has been tidied up. This means that an update to 4.68 might cause
324Exim to "forget" previous rates that it had calculated, and reset them to zero.
325
326
26da7e20
PH
327Exim version 4.64
328-----------------
329
3301. Callouts were setting the name used for EHLO/HELO from $smtp_active_
331hostname. This is wrong, because it relates to the incoming message (and
332probably the interface on which it is arriving) and not to the outgoing
333callout (which could be using a different interface). This has been
334changed to use the value of the helo_data option from the smtp transport
335instead - this is what is used when a message is actually being sent. If
336there is no remote transport (possible with a router that sets up host
337addresses), $smtp_active_hostname is used. This change is mentioned here in
338case somebody is relying on the use of $smtp_active_hostname.
339
ebc9d865
PH
3402. A bug has been fixed that might just possibly be something that is relied on
341in some configurations. In expansion items such as ${if >{xxx}{yyy}...} an
342empty string (that is {}) was being interpreted as if it was {0} and therefore
343treated as the number zero. From release 4.64, such strings cause an error
344because a decimal number, possibly followed by K or M, is required (as has
345always been documented).
346
b6c27e28
PH
3473. There has been a change to the GnuTLS support (ChangeLog/PH/20) to improve
348Exim's performance. Unfortunately, this has the side effect of being slightly
349non-upwards compatible for versions 4.50 and earlier. If you are upgrading from
350one of these earlier versions and you use GnuTLS, you must remove the file
351called gnutls-params in Exim's spool directory. If you don't do this, you will
352see this error:
353
354 TLS error on connection from ... (DH params import): Base64 decoding error.
355
356Removing the file causes Exim to recompute the relevant encryption parameters
357and cache them in the new format that was introduced for release 4.51 (May
3582005). If you are upgrading from release 4.51 or later, there should be no
359problem.
360
26da7e20 361
a5bd321b
PH
362Exim version 4.63
363-----------------
364
365When an SMTP error message is specified in a "message" modifier in an ACL, or
366in a :fail: or :defer: message in a redirect router, Exim now checks the start
367of the message for an SMTP error code. This consists of three digits followed
368by a space, optionally followed by an extended code of the form n.n.n, also
369followed by a space. If this is the case and the very first digit is the same
370as the default error code, the code from the message is used instead. If the
371very first digit is incorrect, a panic error is logged, and the default code is
372used. This is an incompatible change, but it is not expected to affect many (if
373any) configurations. It is possible to suppress the use of the supplied code in
374a redirect router by setting the smtp_error_code option false. In this case,
375any SMTP code is quietly ignored.
376
377
47ca6d6c
PH
378Exim version 4.61
379-----------------
380
8def5aaf
PH
3811. The default number of ACL variables of each type has been increased to 20,
382and it's possible to compile Exim with more. You can safely upgrade to this
383release if you already have messages on the queue with saved ACL variable
384values. However, if you downgrade from this release with messages on the queue,
385any saved ACL values they may have will be lost.
386
3872. The default value for rfc1413_query_timeout has been changed from 30s to 5s.
47ca6d6c
PH
388
389
8857ccfd
PH
390Exim version 4.54
391-----------------
392
393There was a problem with 4.52/TF/02 in that a "name=" option on control=
394submission terminated at the next slash, thereby not allowing for slashes in
395the name. This has been changed so that "name=" takes the rest of the string as
396its data. It must therefore be the last option.
397
398
f3f0a1f0
PH
399Version 4.53
400------------
401
402If you are using the experimental Domain Keys support, you must upgrade to
403at least libdomainkeys 0.67 in order to run this release of Exim.
404
405
b5aea5e1
PH
406Version 4.51
407------------
408
c688b954
PH
4091. The format in which GnuTLS parameters are cached (in the file gnutls-params
410in the spool directory) has been changed. The new format can also be generated
b5aea5e1
PH
411externally, so it is now possible to update the values from outside Exim. This
412has been implemented in an upwards, BUT NOT downwards, compatible manner.
413Upgrading should be seamless: when Exim finds that it cannot understand an
414existing cache file, it generates new parameters and writes them to the cache
415in the new format. If, however, you downgrade from 4.51 to a previous release,
416you MUST delete the gnutls-params file in the spool directory, because the
417older Exim will not recognize the new format.
418
c688b954
PH
4192. When doing a callout as part of verifying an address, Exim was not paying
420attention to any local part prefix or suffix that was matched by the router
421that accepted the address. It now behaves in the same way as it does for
422delivery: the affixes are removed from the local part unless
423rcpt_include_affixes is set on the transport. If you have a configuration that
424uses prefixes or suffixes on addresses that could be used for callouts, and you
425want the affixes to be retained, you must make sure that rcpt_include_affixes
426is set on the transport.
427
447d236c
PH
4283. Bounce and delay warning messages no longer contain details of delivery
429errors, except for explicit messages (e.g. generated by :fail:) and SMTP
430responses from remote hosts.
431
b5aea5e1 432
8b5af54d
PH
433Version 4.50
434------------
435
4964e932
PH
436The exicyclog script has been updated to use three-digit numbers in rotated log
437files if the maximum number to keep is greater than 99. If you are already
438keeping more than 99, there will be an incompatible change when you upgrade.
439You will probably want to rename your old log files to the new form before
c3af992c 440running the new exicyclog.
8b5af54d
PH
441
442
0f4f2a88
PH
443Version 4.42
444------------
445
446RFC 3848 specifies standard names for the "with" phrase in Received: header
447lines when AUTH and/or TLS are in use. This is the "received protocol"
448field. Exim used to use "asmtp" for authenticated SMTP, without any
449indication (in the protocol name) for TLS use. Now it follows the RFC and
450uses "esmtpa" if the connection is authenticated, "esmtps" if it is
451encrypted, and "esmtpsa" if it is both encrypted and authenticated. These names
452appear in log lines as well as in Received: header lines.
453
454
455Version 4.34
456------------
457
458Change 4.31/2 gave problems to data ACLs and local_scan() functions that
459expected to see a Received: header. I have changed to yet another scheme. The
460Received: header is now generated after the body is received, but before the
461ACL or local_scan() is called. After they have run, the timestamp in the
462Received: header is updated.
463
464Thus, change (a) of 4.31/2 has been reversed, but change (b) is still true,
465which is lucky, since I decided it was a bug fix.
466
467
468Version 4.33
469------------
470
471If an expansion in a condition on a "warn" statement fails because a lookup
472defers, the "warn" statement is abandoned, and the next ACL statement is
473processed. Previously this caused the whole ACL to be aborted.
474
475
476Version 4.32
477------------
478
479Change 4.31/2 has been reversed, as it proved contentious. Recipient callout
480verification now uses <> in the MAIL command by default, as it did before. A
481new callout option, "use_sender", has been added to request the other
482behaviour.
483
484
485Version 4.31
486------------
487
4881. If you compile Exim to use GnuTLS, it now requires the use of release 1.0.0
489 or greater. The interface to the obsolete 0.8.x releases is no longer
490 supported. There is one externally visible change: the format for the
491 display of Distinguished Names now uses commas as a separator rather than a
492 slash. This is to comply with RFC 2253.
493
4942. When a message is received, the Received: header line is now generated when
495 reception is complete, instead of at the start of reception. For messages
496 that take a long time to come in, this changes the meaning of the timestamp.
497 There are several side-effects of this change:
498
499 (a) If a message is rejected by a DATA or non-SMTP ACL, or by local_scan(),
500 the logged header lines no longer include the local Received: line,
501 because it has not yet been created. If the message is a non-SMTP one,
502 and the error is processed by sending a message to the sender, the copy
503 of the original message that is returned does not have an added
504 Received: line.
505
506 (b) When a filter file is tested using -bf, no additional Received: header
507 is added to the test message. After some thought, I decided that this
508 is a bug fix.
509
510 The contents of $received_for are not affected by this change. This
511 variable still contains the single recipient of a message, copied after
512 addresses have been rewritten, but before local_scan() is run.
513
5142. Recipient callout verification, like sender verification, was using <> in
515 the MAIL FROM command. This isn't really the right thing, since the actual
516 sender may affect whether the remote host accepts the recipient or not. I
517 have changed it to use the actual sender in the callout; this means that
518 the cache record is now keyed on a recipient/sender pair, not just the
519 recipient address. There doesn't seem to be a real danger of callout loops,
520 since a callout by the remote host to check the sender would use <>.
521
522
523Version 4.30
524------------
525
5261. I have abolished timeout_DNS as an error that can be detected in retry
527 rules, because it has never worked. Despite the fact that it has been
528 documented since at least release 1.62, there was no code to support it.
529 If you have used it in your retry rules, you will now get a warning message
530 to the log and panic log. It is now treated as plain "timeout".
531
5322. After discussion on the mailing list, Exim no longer adds From:, Date:, or
533 Message-Id: header lines to messages that do not originate locally, that is,
534 messages that have an associated sending host address.
535
5363. When looking up a host name from an IP address, Exim now tries the DNS
537 first, and only if that fails does it use gethostbyaddr() (or equivalent).
538 This change was made because on some OS, not all the names are given for
539 addresses with multiple PTR records via the gethostbyaddr() interface. The
540 order of lookup can be changed by setting host_lookup_order.
541
542
543Version 4.23
544------------
545
5461. The new FIXED_NEVER_USERS build-time option creates a list of "never users"
547 that cannot be overridden. The default in the distributed EDITME is "root".
548 If for some reason you were (against advice) running deliveries as root, you
549 will have to ensure that FIXED_NEVER_USERS is not set in your
550 Local/Makefile.
551
5522. The ${quote: operator now quotes an empty string, which it did not before.
553
5543. Version 4.23 saves the contents of the ACL variables with the message, so
555 that they can be used later. If one of these variables contains a newline,
556 there will be a newline character in the spool that will not be interpreted
4c04137d 557 correctly by a previous version of Exim. (Exim ignores keyed spool file
0f4f2a88
PH
558 items that it doesn't understand - precisely for this kind of problem - but
559 it expects them all to be on one line.)
560
561 So the bottom line is: if you have newlines in your ACL variables, you
562 cannot retreat from 4.23.
563
564
565Version 4.21
566------------
567
5681. The idea of the "warn" ACL verb is that it adds a header or writes to the
569 log only when "message" or "log_message" are set. However, if one of the
570 conditions was an address verification, or a call to a nested ACL, the
571 messages generated by the underlying test were being passed through. This
572 no longer happens. The underlying message is available in $acl_verify_
573 message for both "message" and "log_message" expansions, so it can be
574 passed through if needed.
575
5762. The way that the $h_ (and $header_) expansions work has been changed by the
577 addition of RFC 2047 decoding. See the main documentation (the NewStuff file
578 until release 4.30, then the manual) for full details. Briefly, there are
579 now three forms:
580
581 $rh_xxx: and $rheader_xxx: give the original content of the header
582 line(s), with no processing at all.
583
584 $bh_xxx: and $bheader_xxx: remove leading and trailing white space, and
585 then decode base64 or quoted-printable "words" within the header text,
586 but do not do charset translation.
587
588 $h_xxx: and $header_xxx: attempt to translate the $bh_ string to a
589 standard character set.
590
591 If you have previously been using $h_ expansions to access the raw
592 characters, you should change to $rh_ instead.
593
5943. When Exim creates an RFC 2047 encoded word in a header line, it labels it
595 with the default character set from the headers_charset option instead of
596 always using iso-8859-1.
597
5984. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
599 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
600 called TMPDIR, and if it finds it is different, it changes its value.
601
6025. Following a discussion on the list, the rules by which Exim recognises line
603 endings on incoming messages have been changed. The -dropcr and drop_cr
604 options are now no-ops, retained only for backwards compatibility. The
605 following line terminators are recognized: LF CRLF CR. However, special
606 processing applies to CR:
607
608 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
609 nor a local message in the state where . is a terminator.
610
611 (ii) If a bare CR is encountered in a header line, an extra space is added
612 after the line terminator so as not to end the header. The reasoning
613 behind this is that bare CRs in header lines are most likely either
614 to be mistakes, or people trying to play silly games.
615
6166. The code for using daemon_smtp_port, local_interfaces, and the -oX options
617 has been reorganized. It is supposed to be backwards compatible, but it is
618 mentioned here just in case I've screwed up.
619
620
621
622Version 4.20
623------------
624
6251. I have tidied and re-organized the code that uses alarm() for imposing time
626 limits on various things. It shouldn't affect anything, but if you notice
627 processes getting stuck, it may be that I've broken something.
628
6292. The "arguments" log selector now also logs the current working directory
630 when Exim is called.
631
6323. An incompatible change has been made to the appendfile transport. This
633 affects the case when it is used for file deliveries that are set up by
634 .forward and filter files. Previously, any settings of the "file" or
635 "directory" options were ignored. It is hoped that, like the address_file
636 transport in the default configuration, these options were never in fact set
637 on such transports, because they were of no use.
638
639 Now, if either of these options is set, it is used. The path that is passed
640 by the router is in $address_file (this is not new), so it can be used as
641 part of a longer path, or modified in any other way that expansion permits.
642
643 If neither "file" nor "directory" is set, the behaviour is unchanged.
644
6454. Related to the above: in a filter, if a "save" command specifies a non-
646 absolute path, the value of $home/ is pre-pended. This no longer happens if
647 $home is unset or is set to an empty string.
648
6495. Multiple file deliveries from a filter or .forward file can never be
650 batched; the value of batch_max on the transport is ignored for file
651 deliveries. I'm assuming that nobody ever actually set batch_max on the
652 address_file transport - it would have had odd effects previously.
653
6546. DESTDIR is the more common variable that ROOT for use when installing
655 software under a different root filing system. The Exim install script now
656 recognizes DESTDIR first; if it is not set, ROOT is used.
657
6587. If DESTDIR is set when installing Exim, it no longer prepends its value to
659 the path of the system aliases file that appears in the default
660 configuration (when a default configuration is installed). If an aliases
661 file is actually created, its name *does* use the prefix.
662
663
664Version 4.14
665------------
666
6671. The default for the maximum number of unknown SMTP commands that Exim will
668accept before dropping a connection has been reduced from 5 to 3. However, you
669can now change the value by setting smtp_max_unknown_commands.
670
6712. The ${quote: operator has been changed so that it turns newline and carriage
672return characters into \n and \r, respectively.
673
6743. The file names used for maildir messages now include the microsecond time
675fraction as well as the time in seconds, to cope with systems where the process
676id can be re-used within the same second. The format is now
677
678 <time>.H<microsec>P<pid>.<host>
679
680This should be a compatible change, but is noted here just in case.
681
6824. The rules for creating message ids have changed, to cope with systems where
683the process id can be re-used within the same second. The format, however, is
684unchanged, so this should not cause any problems, except as noted in the next
685item.
686
6875. The maximum value for localhost_number has been reduced from 255 to 16, in
688order to implement the new message id rules. For operating systems that have
689case-insensitive file systems (Cygwin and Darwin), the limit is 10.
690
6916. verify = header_syntax was allowing unqualified addresses in all cases. Now
692it allows them only for locally generated messages and from hosts that match
693sender_unqualified_hosts or recipient_unqualified_hosts, respectively.
694
6957. For reasons lost in the mists of time, when a pipe transport was run, the
696environment variable MESSAGE_ID was set to the message ID preceded by 'E' (the
697form used in Message-ID: header lines). The 'E' has been removed.
698
699
700Version 4.11
701------------
702
7031. The handling of lines in the configuration file has changed. Previously,
704macro expansion was applied to logical lines, after continuations had been
705joined on. This meant that it could not be used in .include lines, which are
706handled as physical rather than logical lines. Macro expansion is now done on
707physical lines rather than logical lines. This means there are two
708incompatibilities:
709
710 (a) A macro that expands to # to turn a line into a comment now applies only
711 to the physical line where it appears. Previously, it would have caused
712 any following continuations also to be ignored.
713
714 (b) A macro name can no longer be split over the boundary between a line and
715 its continuation. Actually, this is more of a bug fix. :-)
716
7172. The -D command line option must now all be within one command line item.
718This makes it possible to use -D to set a macro to the empty string by commands
719such as
720
721 exim -DABC ...
722 exim -DABC= ...
723
724Previously, these items would have moved on to the next item on the command
725line. To include spaces in a macro definition item, quotes must be used, in
726which case you can also have spaces after -D and surrounding the equals. For
727example:
728
729 exim '-D ABC = something' ...
730
7313. The way that addresses that redirect to themselves are handled has been
732changed, in order to fix an obscure bug. This should not cause any problems
733except in the case of wanting to go back from a 4.11 (or later) release to an
734earlier release. If there are undelivered messages on the spool that contain
735addresses which redirect to themselves, and the redirected addresses have
736already been delivered, you might get a duplicate delivery if you revert to an
737earlier Exim.
738
7394. The default way of looking up IP addresses for hosts in the manualroute and
740queryprogram routers has been changed. If "byname" or "bydns" is explicitly
741specified, there is no change, but if no method is specified, Exim now behaves
742as follows:
743
744 First, a DNS lookup is done. If this yields anything other than
745 HOST_NOT_FOUND, that result is used. Otherwise, Exim goes on to try a call to
746 getipnodebyname() (or gethostbyname() on older systems) and the result of the
747 lookup is the result of that call.
748
749This change has been made because it has been discovered that on some systems,
750if a DNS lookup called via getipnodebyname() times out, HOST_NOT_FOUND is
751returned instead of TRY_AGAIN. Thus, it is safest to try a DNS lookup directly
752first, and only if that gives a definite "no such host" to try the local
753function.
754
7555. In fixing the minor security problem with pid_file_path, I have removed some
756backwards-compatible (undocumented) code which was present to ease conversion
757from Exim 3. In Exim 4, pid_file_path is a literal; in Exim 3 it was allowed to
758contain "%s", which was replaced by the port number for daemons listening on
759non-standard ports. In Exim 4, such daemons do not write a pid file. The
760backwards compatibility feature was to replace "%s" by nothing if it occurred
761in an Exim 4 setting of pid_file_path. The bug was in this code. I have solved
762the problem by removing the backwards compatibility feature. Thus, if you still
763have "%s" somewhere in a setting of pid_file_path, you should remove it.
764
7656. There has been an extension to lsearch files. The keys in these files may
766now be quoted in order to allow for whitespace and colons in them. This means
767that if you were previously using keys that began with a doublequote, you will
768now have to wrap them with extra quotes and escape the internal quotes. The
769possibility that anybody is actually doing this seems extremely remote, but it
770is documented just in case.
771
772
773Version 4.10
774------------
775
776The build-time parameter EXIWHAT_KILL_ARG has been renamed EXIWHAT_KILL_SIGNAL
777to better reflect its function. The OS-specific files have been updated. Only
778if you have explicitly set this in your Makefile (highly unlikely) do you need
779to change anything.
780
781****