GnuTLS website moves
[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
2c422e6f 29Exim version 4.82
a5f239e4
PP
30-----------------
31
32 * New option gnutls_enable_pkcs11 defaults false; if you have GnuTLS 2.12.0
33 or later and do want PKCS11 modules to be autoloaded, then set this option.
34
8b260705
PP
35 * A per-transport wait-<name> database is no longer updated if the transport
36 sets "connection_max_messages" to 1, as it can not be used and causes
37 unnecessary serialisation and load. External tools tracking the state of
38 Exim by the hints databases may need modification to take this into account.
39
a5f239e4 40
83e2f8a2 41Exim version 4.80
dec5017e
PP
42-----------------
43
83e2f8a2
PP
44 * BEWARE backwards-incompatible changes in SSL libraries, thus the version
45 bump. See points below for details.
c2b063d4 46 Also an LDAP data returned format change.
83e2f8a2 47
dec5017e
PP
48 * The value of $tls_peerdn is now print-escaped when written to the spool file
49 in a -tls_peerdn line, and unescaped when read back in. We received reports
50 of values with embedded newlines, which caused spool file corruption.
51
52 If you have a corrupt spool file and you wish to recover the contents after
53 upgrading, then lock the message, replace the new-lines that should be part
54 of the -tls_peerdn line with the two-character sequence \n and then unlock
55 the message. No tool has been provided as we believe this is a rare
56 occurence.
57
f0f5a555
PP
58 * For OpenSSL, SSLv2 is now disabled by default. (GnuTLS does not support
59 SSLv2). RFC 6176 prohibits SSLv2 and some informal surveys suggest no
60 actual usage. You can re-enable with the "openssl_options" Exim option,
61 in the main configuration section. Note that supporting SSLv2 exposes
62 you to ciphersuite downgrade attacks.
63
c80c5570
PP
64 * With OpenSSL 1.0.1+, Exim now supports TLS 1.1 and TLS 1.2. If built
65 against 1.0.1a then you will get a warning message and the
66 "openssl_options" value will not parse "no_tlsv1_1": the value changes
67 incompatibly between 1.0.1a and 1.0.1b, because the value chosen for 1.0.1a
68 is infelicitous. We advise avoiding 1.0.1a.
69
70 "openssl_options" gains "no_tlsv1_1", "no_tlsv1_2" and "no_compression".
71
da3ad30d 72 COMPATIBILITY WARNING: The default value of "openssl_options" is no longer
f0f5a555
PP
73 "+dont_insert_empty_fragments". We default to "+no_sslv2".
74 That old default was grandfathered in from before openssl_options became a
75 configuration option.
da3ad30d
PP
76 Empty fragments are inserted by default through TLS1.0, to partially defend
77 against certain attacks; TLS1.1+ change the protocol so that this is not
78 needed. The DIEF SSL option was required for some old releases of mail
79 clients which did not gracefully handle the empty fragments, and was
80 initially set in Exim release 4.31 (see ChangeLog, item 37).
81
82 If you still have affected mail-clients, and you see SSL protocol failures
83 with this release of Exim, set:
84 openssl_options = +dont_insert_empty_fragments
85 in the main section of your Exim configuration file. You're trading off
86 security for compatibility. Exim is now defaulting to higher security and
87 rewarding more modern clients.
88
3375e053
PP
89 If the option tls_dhparams is set and the parameters loaded from the file
90 have a bit-count greater than the new option tls_dh_max_bits, then the file
91 will now be ignored. If this affects you, raise the tls_dh_max_bits limit.
92 We suspect that most folks are using dated defaults and will not be affected.
93
5407bfff
JH
94 * Ldap lookups returning multi-valued attributes now separate the attributes
95 with only a comma, not a comma-space sequence. Also, an actual comma within
96 a returned attribute is doubled. This makes it possible to parse the
97 attribute as a comma-separated list. Note the distinction from multiple
98 attributes being returned, where each one is a name=value pair.
c80c5570 99
c2b063d4
PP
100 If you are currently splitting the results from LDAP upon a comma, then you
101 should check carefully to see if adjustments are needed.
102
103 This change lets cautious folks distinguish "comma used as separator for
104 joining values" from "comma inside the data".
105
9cbad13b
PP
106 * accept_8bitmime now defaults on, which is not RFC compliant but is better
107 suited to today's Internet. See http://cr.yp.to/smtp/8bitmime.html for a
108 sane rationale. Those who wish to be strictly RFC compliant, or know that
109 they need to talk to servers that are not 8-bit-clean, now need to take
110 explicit configuration action to default this option off. This is not a
111 new option, you can safely force it off before upgrading, to decouple
112 configuration changes from the binary upgrade while remaining RFC compliant.
113
f90a9fd1
PP
114 * The GnuTLS support has been mostly rewritten, to use APIs which don't cause
115 deprecation warnings in GnuTLS 2.12.x. As part of this, these three options
116 are no longer supported:
17c76198
PP
117
118 gnutls_require_kx
119 gnutls_require_mac
120 gnutls_require_protocols
121
f90a9fd1
PP
122 Their functionality is entirely subsumed into tls_require_ciphers. In turn,
123 tls_require_ciphers is no longer an Exim list and is not parsed by Exim, but
124 is instead given to gnutls_priority_init(3), which expects a priority string;
125 this behaviour is much closer to the OpenSSL behaviour. See:
17c76198 126
42bfef1e 127 http://www.gnutls.org/manual/html_node/Priority-Strings.html
17c76198
PP
128
129 for fuller documentation of the strings parsed. The three gnutls_require_*
130 options are still parsed by Exim and, for this release, silently ignored.
131 A future release will add warnings, before a later still release removes
132 parsing entirely and the presence of the options will be a configuration
133 error.
134
83e2f8a2
PP
135 Note that by default, GnuTLS will not accept RSA-MD5 signatures in chains.
136 A tls_require_ciphers value of NORMAL:%VERIFY_ALLOW_SIGN_RSA_MD5 may
137 re-enable support, but this is not supported by the Exim maintainers.
138 Our test suite no longer includes MD5-based certificates.
139
17c76198
PP
140 This rewrite means that Exim will continue to build against GnuTLS in the
141 future, brings Exim closer to other GnuTLS applications and lets us add
142 support for SNI and other features more readily. We regret that it wasn't
143 feasible to retain the three dropped options.
144
3375e053
PP
145 * If built with TLS support, then Exim will now validate the value of
146 the main section tls_require_ciphers option at start-up. Before, this
147 would cause a STARTTLS 4xx failure, now it causes a failure to start.
148 Running with a broken configuration which causes failures that may only
149 be left in the logs has been traded off for something more visible. This
150 change makes an existing problem more prominent, but we do not believe
151 anyone would deliberately be running with an invalid tls_require_ciphers
152 option.
153
154 This also means that library linkage issues caused by conflicts of some
155 kind might take out the main daemon, not just the delivery or receiving
156 process. Conceivably some folks might prefer to continue delivering
157 mail plaintext when their binary is broken in this way, if there is a
158 server that is a candidate to receive such mails that does not advertise
159 STARTTLS. Note that Exim is typically a setuid root binary and given
160 broken linkage problems that cause segfaults, we feel it is safer to
161 fail completely. (The check is not done as root, to ensure that problems
162 here are not made worse by the check).
163
a799883d
PP
164 * The "tls_dhparam" option has been updated, so that it can now specify a
165 path or an identifier for a standard DH prime from one of a few RFCs.
166 The default for OpenSSL is no longer to not use DH but instead to use
167 one of these standard primes. The default for GnuTLS is no longer to use
168 a file in the spool directory, but to use that same standard prime.
169 The option is now used by GnuTLS too. If it points to a path, then
170 GnuTLS will use that path, instead of a file in the spool directory;
171 GnuTLS will attempt to create it if it does not exist.
172
173 To preserve the previous behaviour of generating files in the spool
174 directory, set "tls_dhparam = historic". Since prior releases of Exim
175 ignored tls_dhparam when using GnuTLS, this can safely be done before
176 the upgrade.
177
178
dec5017e 179
c566dd90
PP
180Exim version 4.77
181-----------------
182
183 * GnuTLS will now attempt to use TLS 1.2 and TLS 1.1 before TLS 1.0 and SSL3,
184 if supported by your GnuTLS library. Use the existing
185 "gnutls_require_protocols" option to downgrade this if that will be a
186 problem. Prior to this release, supported values were "TLS1" and "SSL3",
187 so you should be able to update configuration prior to update.
188
0215ec07
PP
189 [nb: gnutls_require_protocols removed in Exim 4.80, instead use
190 tls_require_ciphers to provide a priority string; see notes above]
191
39257585
PP
192 * The match_<type>{string1}{string2} expansion conditions no longer subject
193 string2 to string expansion, unless Exim was built with the new
194 "EXPAND_LISTMATCH_RHS" option. Too many people have inadvertently created
195 insecure configurations that way. If you need the functionality and turn on
196 that build option, please let the developers know, and know why, so we can
197 try to provide a safer mechanism for you.
198
a7ee5955
PP
199 The match{}{} expansion condition (for regular expressions) is NOT affected.
200 For match_<type>{s1}{s2}, all list functionality is unchanged. The only
201 change is that a '$' appearing in s2 will not trigger expansion, but instead
202 will be treated as a literal $ sign; the effect is very similar to having
203 wrapped s2 with \N...\N. If s2 contains a named list and the list definition
204 uses $expansions then those _will_ be processed as normal. It is only the
205 point at which s2 is read where expansion is inhibited.
206
207 If you are trying to test if two email addresses are equal, use eqi{s1}{s2}.
208 If you are testing if the address in s1 occurs in the list of items given
209 in s2, either use the new inlisti{s1}{s2} condition (added in 4.77) or use
210 the pre-existing forany{s2}{eqi{$item}{s1}} condition.
211
c566dd90 212
e6acf885
PP
213Exim version 4.74
214-----------------
215
4050a044
PP
216 * The integrated support for dynamically loadable lookup modules has an ABI
217 change from the modules supported by some OS vendors through an unofficial
218 patch. Don't try to mix & match.
219
220 * Some parts of the build system are now beginning to assume that the host
221 environment is POSIX. If you're building on a system where POSIX tools are
222 not the default, you might have an easier time if you switch to the POSIX
223 tools. Feel free to report non-POSIX issues as a request for a feature
224 enhancement, but if the POSIX variants are available then the fix will
225 probably just involve some coercion. See the README instructions for
226 building on such hosts.
e6acf885
PP
227
228
229Exim version 4.73
230-----------------
231
232 * The Exim run-time user can no longer be root; this was always
233 strongly discouraged, but is now prohibited both at build and
234 run-time. If you need Exim to run routinely as root, you'll need to
235 patch the source and accept the risk. Here be dragons.
236
237 * Exim will no longer accept a configuration file owned by the Exim
238 run-time user, unless that account is explicitly the value in
239 CONFIGURE_OWNER, which we discourage. Exim now checks to ensure that
a7ee5955 240 files are not writeable by other accounts.
e6acf885
PP
241
242 * The ALT_CONFIG_ROOT_ONLY build option is no longer optional and is forced
243 on; the Exim user can, by default, no longer use -C/-D and retain privilege.
244 Two new build options mitigate this.
245
246 * TRUSTED_CONFIG_LIST defines a file containing a whitelist of config
247 files that are trusted to be selected by the Exim user; one per line.
248 This is the recommended approach going forward.
249
250 * WHITELIST_D_MACROS defines a colon-separated list of macro names which
251 the Exim run-time user may safely pass without dropping privileges.
252 Because changes to this involve a recompile, this is not the recommended
253 approach but may ease transition. The values of the macros, when
43236f35 254 overridden, are constrained to match this regex: ^[A-Za-z0-9_/.-]*$
e6acf885
PP
255
256 * The system_filter_user option now defaults to the Exim run-time user,
257 rather than root. You can still set it explicitly to root and this
258 can be done with prior versions too, letting you roll versions
259 without needing to change this configuration option.
260
261 * ClamAV must be at least version 0.95 unless WITH_OLD_CLAMAV_STREAM is
262 defined at build time.
263
264
2eda48c6
TK
265Exim version 4.70
266-----------------
267
2681. Experimental Yahoo! Domainkeys support has been dropped in this release.
269It has been superceded by a native implementation of its successor DKIM.
270
2712. Up to version 4.69, Exim came with an embedded version of the PCRE library.
272As of 4.70, this is no longer the case. To compile Exim, you will need PCRE
a7ee5955 273installed. Most OS distributions have ready-made library and development
2eda48c6 274packages.
0f4f2a88
PH
275
276
3c223874
PH
277Exim version 4.68
278-----------------
279
2801. The internal implementation of the database keys that are used for ACL
281ratelimiting has been tidied up. This means that an update to 4.68 might cause
282Exim to "forget" previous rates that it had calculated, and reset them to zero.
283
284
26da7e20
PH
285Exim version 4.64
286-----------------
287
2881. Callouts were setting the name used for EHLO/HELO from $smtp_active_
289hostname. This is wrong, because it relates to the incoming message (and
290probably the interface on which it is arriving) and not to the outgoing
291callout (which could be using a different interface). This has been
292changed to use the value of the helo_data option from the smtp transport
293instead - this is what is used when a message is actually being sent. If
294there is no remote transport (possible with a router that sets up host
295addresses), $smtp_active_hostname is used. This change is mentioned here in
296case somebody is relying on the use of $smtp_active_hostname.
297
ebc9d865
PH
2982. A bug has been fixed that might just possibly be something that is relied on
299in some configurations. In expansion items such as ${if >{xxx}{yyy}...} an
300empty string (that is {}) was being interpreted as if it was {0} and therefore
301treated as the number zero. From release 4.64, such strings cause an error
302because a decimal number, possibly followed by K or M, is required (as has
303always been documented).
304
b6c27e28
PH
3053. There has been a change to the GnuTLS support (ChangeLog/PH/20) to improve
306Exim's performance. Unfortunately, this has the side effect of being slightly
307non-upwards compatible for versions 4.50 and earlier. If you are upgrading from
308one of these earlier versions and you use GnuTLS, you must remove the file
309called gnutls-params in Exim's spool directory. If you don't do this, you will
310see this error:
311
312 TLS error on connection from ... (DH params import): Base64 decoding error.
313
314Removing the file causes Exim to recompute the relevant encryption parameters
315and cache them in the new format that was introduced for release 4.51 (May
3162005). If you are upgrading from release 4.51 or later, there should be no
317problem.
318
26da7e20 319
a5bd321b
PH
320Exim version 4.63
321-----------------
322
323When an SMTP error message is specified in a "message" modifier in an ACL, or
324in a :fail: or :defer: message in a redirect router, Exim now checks the start
325of the message for an SMTP error code. This consists of three digits followed
326by a space, optionally followed by an extended code of the form n.n.n, also
327followed by a space. If this is the case and the very first digit is the same
328as the default error code, the code from the message is used instead. If the
329very first digit is incorrect, a panic error is logged, and the default code is
330used. This is an incompatible change, but it is not expected to affect many (if
331any) configurations. It is possible to suppress the use of the supplied code in
332a redirect router by setting the smtp_error_code option false. In this case,
333any SMTP code is quietly ignored.
334
335
47ca6d6c
PH
336Exim version 4.61
337-----------------
338
8def5aaf
PH
3391. The default number of ACL variables of each type has been increased to 20,
340and it's possible to compile Exim with more. You can safely upgrade to this
341release if you already have messages on the queue with saved ACL variable
342values. However, if you downgrade from this release with messages on the queue,
343any saved ACL values they may have will be lost.
344
3452. The default value for rfc1413_query_timeout has been changed from 30s to 5s.
47ca6d6c
PH
346
347
8857ccfd
PH
348Exim version 4.54
349-----------------
350
351There was a problem with 4.52/TF/02 in that a "name=" option on control=
352submission terminated at the next slash, thereby not allowing for slashes in
353the name. This has been changed so that "name=" takes the rest of the string as
354its data. It must therefore be the last option.
355
356
f3f0a1f0
PH
357Version 4.53
358------------
359
360If you are using the experimental Domain Keys support, you must upgrade to
361at least libdomainkeys 0.67 in order to run this release of Exim.
362
363
b5aea5e1
PH
364Version 4.51
365------------
366
c688b954
PH
3671. The format in which GnuTLS parameters are cached (in the file gnutls-params
368in the spool directory) has been changed. The new format can also be generated
b5aea5e1
PH
369externally, so it is now possible to update the values from outside Exim. This
370has been implemented in an upwards, BUT NOT downwards, compatible manner.
371Upgrading should be seamless: when Exim finds that it cannot understand an
372existing cache file, it generates new parameters and writes them to the cache
373in the new format. If, however, you downgrade from 4.51 to a previous release,
374you MUST delete the gnutls-params file in the spool directory, because the
375older Exim will not recognize the new format.
376
c688b954
PH
3772. When doing a callout as part of verifying an address, Exim was not paying
378attention to any local part prefix or suffix that was matched by the router
379that accepted the address. It now behaves in the same way as it does for
380delivery: the affixes are removed from the local part unless
381rcpt_include_affixes is set on the transport. If you have a configuration that
382uses prefixes or suffixes on addresses that could be used for callouts, and you
383want the affixes to be retained, you must make sure that rcpt_include_affixes
384is set on the transport.
385
447d236c
PH
3863. Bounce and delay warning messages no longer contain details of delivery
387errors, except for explicit messages (e.g. generated by :fail:) and SMTP
388responses from remote hosts.
389
b5aea5e1 390
8b5af54d
PH
391Version 4.50
392------------
393
4964e932
PH
394The exicyclog script has been updated to use three-digit numbers in rotated log
395files if the maximum number to keep is greater than 99. If you are already
396keeping more than 99, there will be an incompatible change when you upgrade.
397You will probably want to rename your old log files to the new form before
c3af992c 398running the new exicyclog.
8b5af54d
PH
399
400
0f4f2a88
PH
401Version 4.42
402------------
403
404RFC 3848 specifies standard names for the "with" phrase in Received: header
405lines when AUTH and/or TLS are in use. This is the "received protocol"
406field. Exim used to use "asmtp" for authenticated SMTP, without any
407indication (in the protocol name) for TLS use. Now it follows the RFC and
408uses "esmtpa" if the connection is authenticated, "esmtps" if it is
409encrypted, and "esmtpsa" if it is both encrypted and authenticated. These names
410appear in log lines as well as in Received: header lines.
411
412
413Version 4.34
414------------
415
416Change 4.31/2 gave problems to data ACLs and local_scan() functions that
417expected to see a Received: header. I have changed to yet another scheme. The
418Received: header is now generated after the body is received, but before the
419ACL or local_scan() is called. After they have run, the timestamp in the
420Received: header is updated.
421
422Thus, change (a) of 4.31/2 has been reversed, but change (b) is still true,
423which is lucky, since I decided it was a bug fix.
424
425
426Version 4.33
427------------
428
429If an expansion in a condition on a "warn" statement fails because a lookup
430defers, the "warn" statement is abandoned, and the next ACL statement is
431processed. Previously this caused the whole ACL to be aborted.
432
433
434Version 4.32
435------------
436
437Change 4.31/2 has been reversed, as it proved contentious. Recipient callout
438verification now uses <> in the MAIL command by default, as it did before. A
439new callout option, "use_sender", has been added to request the other
440behaviour.
441
442
443Version 4.31
444------------
445
4461. If you compile Exim to use GnuTLS, it now requires the use of release 1.0.0
447 or greater. The interface to the obsolete 0.8.x releases is no longer
448 supported. There is one externally visible change: the format for the
449 display of Distinguished Names now uses commas as a separator rather than a
450 slash. This is to comply with RFC 2253.
451
4522. When a message is received, the Received: header line is now generated when
453 reception is complete, instead of at the start of reception. For messages
454 that take a long time to come in, this changes the meaning of the timestamp.
455 There are several side-effects of this change:
456
457 (a) If a message is rejected by a DATA or non-SMTP ACL, or by local_scan(),
458 the logged header lines no longer include the local Received: line,
459 because it has not yet been created. If the message is a non-SMTP one,
460 and the error is processed by sending a message to the sender, the copy
461 of the original message that is returned does not have an added
462 Received: line.
463
464 (b) When a filter file is tested using -bf, no additional Received: header
465 is added to the test message. After some thought, I decided that this
466 is a bug fix.
467
468 The contents of $received_for are not affected by this change. This
469 variable still contains the single recipient of a message, copied after
470 addresses have been rewritten, but before local_scan() is run.
471
4722. Recipient callout verification, like sender verification, was using <> in
473 the MAIL FROM command. This isn't really the right thing, since the actual
474 sender may affect whether the remote host accepts the recipient or not. I
475 have changed it to use the actual sender in the callout; this means that
476 the cache record is now keyed on a recipient/sender pair, not just the
477 recipient address. There doesn't seem to be a real danger of callout loops,
478 since a callout by the remote host to check the sender would use <>.
479
480
481Version 4.30
482------------
483
4841. I have abolished timeout_DNS as an error that can be detected in retry
485 rules, because it has never worked. Despite the fact that it has been
486 documented since at least release 1.62, there was no code to support it.
487 If you have used it in your retry rules, you will now get a warning message
488 to the log and panic log. It is now treated as plain "timeout".
489
4902. After discussion on the mailing list, Exim no longer adds From:, Date:, or
491 Message-Id: header lines to messages that do not originate locally, that is,
492 messages that have an associated sending host address.
493
4943. When looking up a host name from an IP address, Exim now tries the DNS
495 first, and only if that fails does it use gethostbyaddr() (or equivalent).
496 This change was made because on some OS, not all the names are given for
497 addresses with multiple PTR records via the gethostbyaddr() interface. The
498 order of lookup can be changed by setting host_lookup_order.
499
500
501Version 4.23
502------------
503
5041. The new FIXED_NEVER_USERS build-time option creates a list of "never users"
505 that cannot be overridden. The default in the distributed EDITME is "root".
506 If for some reason you were (against advice) running deliveries as root, you
507 will have to ensure that FIXED_NEVER_USERS is not set in your
508 Local/Makefile.
509
5102. The ${quote: operator now quotes an empty string, which it did not before.
511
5123. Version 4.23 saves the contents of the ACL variables with the message, so
513 that they can be used later. If one of these variables contains a newline,
514 there will be a newline character in the spool that will not be interpreted
515 correctely by a previous version of Exim. (Exim ignores keyed spool file
516 items that it doesn't understand - precisely for this kind of problem - but
517 it expects them all to be on one line.)
518
519 So the bottom line is: if you have newlines in your ACL variables, you
520 cannot retreat from 4.23.
521
522
523Version 4.21
524------------
525
5261. The idea of the "warn" ACL verb is that it adds a header or writes to the
527 log only when "message" or "log_message" are set. However, if one of the
528 conditions was an address verification, or a call to a nested ACL, the
529 messages generated by the underlying test were being passed through. This
530 no longer happens. The underlying message is available in $acl_verify_
531 message for both "message" and "log_message" expansions, so it can be
532 passed through if needed.
533
5342. The way that the $h_ (and $header_) expansions work has been changed by the
535 addition of RFC 2047 decoding. See the main documentation (the NewStuff file
536 until release 4.30, then the manual) for full details. Briefly, there are
537 now three forms:
538
539 $rh_xxx: and $rheader_xxx: give the original content of the header
540 line(s), with no processing at all.
541
542 $bh_xxx: and $bheader_xxx: remove leading and trailing white space, and
543 then decode base64 or quoted-printable "words" within the header text,
544 but do not do charset translation.
545
546 $h_xxx: and $header_xxx: attempt to translate the $bh_ string to a
547 standard character set.
548
549 If you have previously been using $h_ expansions to access the raw
550 characters, you should change to $rh_ instead.
551
5523. When Exim creates an RFC 2047 encoded word in a header line, it labels it
553 with the default character set from the headers_charset option instead of
554 always using iso-8859-1.
555
5564. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
557 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
558 called TMPDIR, and if it finds it is different, it changes its value.
559
5605. Following a discussion on the list, the rules by which Exim recognises line
561 endings on incoming messages have been changed. The -dropcr and drop_cr
562 options are now no-ops, retained only for backwards compatibility. The
563 following line terminators are recognized: LF CRLF CR. However, special
564 processing applies to CR:
565
566 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
567 nor a local message in the state where . is a terminator.
568
569 (ii) If a bare CR is encountered in a header line, an extra space is added
570 after the line terminator so as not to end the header. The reasoning
571 behind this is that bare CRs in header lines are most likely either
572 to be mistakes, or people trying to play silly games.
573
5746. The code for using daemon_smtp_port, local_interfaces, and the -oX options
575 has been reorganized. It is supposed to be backwards compatible, but it is
576 mentioned here just in case I've screwed up.
577
578
579
580Version 4.20
581------------
582
5831. I have tidied and re-organized the code that uses alarm() for imposing time
584 limits on various things. It shouldn't affect anything, but if you notice
585 processes getting stuck, it may be that I've broken something.
586
5872. The "arguments" log selector now also logs the current working directory
588 when Exim is called.
589
5903. An incompatible change has been made to the appendfile transport. This
591 affects the case when it is used for file deliveries that are set up by
592 .forward and filter files. Previously, any settings of the "file" or
593 "directory" options were ignored. It is hoped that, like the address_file
594 transport in the default configuration, these options were never in fact set
595 on such transports, because they were of no use.
596
597 Now, if either of these options is set, it is used. The path that is passed
598 by the router is in $address_file (this is not new), so it can be used as
599 part of a longer path, or modified in any other way that expansion permits.
600
601 If neither "file" nor "directory" is set, the behaviour is unchanged.
602
6034. Related to the above: in a filter, if a "save" command specifies a non-
604 absolute path, the value of $home/ is pre-pended. This no longer happens if
605 $home is unset or is set to an empty string.
606
6075. Multiple file deliveries from a filter or .forward file can never be
608 batched; the value of batch_max on the transport is ignored for file
609 deliveries. I'm assuming that nobody ever actually set batch_max on the
610 address_file transport - it would have had odd effects previously.
611
6126. DESTDIR is the more common variable that ROOT for use when installing
613 software under a different root filing system. The Exim install script now
614 recognizes DESTDIR first; if it is not set, ROOT is used.
615
6167. If DESTDIR is set when installing Exim, it no longer prepends its value to
617 the path of the system aliases file that appears in the default
618 configuration (when a default configuration is installed). If an aliases
619 file is actually created, its name *does* use the prefix.
620
621
622Version 4.14
623------------
624
6251. The default for the maximum number of unknown SMTP commands that Exim will
626accept before dropping a connection has been reduced from 5 to 3. However, you
627can now change the value by setting smtp_max_unknown_commands.
628
6292. The ${quote: operator has been changed so that it turns newline and carriage
630return characters into \n and \r, respectively.
631
6323. The file names used for maildir messages now include the microsecond time
633fraction as well as the time in seconds, to cope with systems where the process
634id can be re-used within the same second. The format is now
635
636 <time>.H<microsec>P<pid>.<host>
637
638This should be a compatible change, but is noted here just in case.
639
6404. The rules for creating message ids have changed, to cope with systems where
641the process id can be re-used within the same second. The format, however, is
642unchanged, so this should not cause any problems, except as noted in the next
643item.
644
6455. The maximum value for localhost_number has been reduced from 255 to 16, in
646order to implement the new message id rules. For operating systems that have
647case-insensitive file systems (Cygwin and Darwin), the limit is 10.
648
6496. verify = header_syntax was allowing unqualified addresses in all cases. Now
650it allows them only for locally generated messages and from hosts that match
651sender_unqualified_hosts or recipient_unqualified_hosts, respectively.
652
6537. For reasons lost in the mists of time, when a pipe transport was run, the
654environment variable MESSAGE_ID was set to the message ID preceded by 'E' (the
655form used in Message-ID: header lines). The 'E' has been removed.
656
657
658Version 4.11
659------------
660
6611. The handling of lines in the configuration file has changed. Previously,
662macro expansion was applied to logical lines, after continuations had been
663joined on. This meant that it could not be used in .include lines, which are
664handled as physical rather than logical lines. Macro expansion is now done on
665physical lines rather than logical lines. This means there are two
666incompatibilities:
667
668 (a) A macro that expands to # to turn a line into a comment now applies only
669 to the physical line where it appears. Previously, it would have caused
670 any following continuations also to be ignored.
671
672 (b) A macro name can no longer be split over the boundary between a line and
673 its continuation. Actually, this is more of a bug fix. :-)
674
6752. The -D command line option must now all be within one command line item.
676This makes it possible to use -D to set a macro to the empty string by commands
677such as
678
679 exim -DABC ...
680 exim -DABC= ...
681
682Previously, these items would have moved on to the next item on the command
683line. To include spaces in a macro definition item, quotes must be used, in
684which case you can also have spaces after -D and surrounding the equals. For
685example:
686
687 exim '-D ABC = something' ...
688
6893. The way that addresses that redirect to themselves are handled has been
690changed, in order to fix an obscure bug. This should not cause any problems
691except in the case of wanting to go back from a 4.11 (or later) release to an
692earlier release. If there are undelivered messages on the spool that contain
693addresses which redirect to themselves, and the redirected addresses have
694already been delivered, you might get a duplicate delivery if you revert to an
695earlier Exim.
696
6974. The default way of looking up IP addresses for hosts in the manualroute and
698queryprogram routers has been changed. If "byname" or "bydns" is explicitly
699specified, there is no change, but if no method is specified, Exim now behaves
700as follows:
701
702 First, a DNS lookup is done. If this yields anything other than
703 HOST_NOT_FOUND, that result is used. Otherwise, Exim goes on to try a call to
704 getipnodebyname() (or gethostbyname() on older systems) and the result of the
705 lookup is the result of that call.
706
707This change has been made because it has been discovered that on some systems,
708if a DNS lookup called via getipnodebyname() times out, HOST_NOT_FOUND is
709returned instead of TRY_AGAIN. Thus, it is safest to try a DNS lookup directly
710first, and only if that gives a definite "no such host" to try the local
711function.
712
7135. In fixing the minor security problem with pid_file_path, I have removed some
714backwards-compatible (undocumented) code which was present to ease conversion
715from Exim 3. In Exim 4, pid_file_path is a literal; in Exim 3 it was allowed to
716contain "%s", which was replaced by the port number for daemons listening on
717non-standard ports. In Exim 4, such daemons do not write a pid file. The
718backwards compatibility feature was to replace "%s" by nothing if it occurred
719in an Exim 4 setting of pid_file_path. The bug was in this code. I have solved
720the problem by removing the backwards compatibility feature. Thus, if you still
721have "%s" somewhere in a setting of pid_file_path, you should remove it.
722
7236. There has been an extension to lsearch files. The keys in these files may
724now be quoted in order to allow for whitespace and colons in them. This means
725that if you were previously using keys that began with a doublequote, you will
726now have to wrap them with extra quotes and escape the internal quotes. The
727possibility that anybody is actually doing this seems extremely remote, but it
728is documented just in case.
729
730
731Version 4.10
732------------
733
734The build-time parameter EXIWHAT_KILL_ARG has been renamed EXIWHAT_KILL_SIGNAL
735to better reflect its function. The OS-specific files have been updated. Only
736if you have explicitly set this in your Makefile (highly unlikely) do you need
737to change anything.
738
739****