1 This document contains detailed information about incompatibilities that might
2 be encountered when upgrading from one release of Exim to another. The
3 information is in reverse order of release numbers. Mostly these are relatively
4 small points, and the configuration file is normally upwards compatible, but
5 there have been two big upheavals...
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. *
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 **************************************************************************
25 The rest of this document contains information about changes in 4.xx releases
26 that might affect a running system.
32 * Exim used to manually follow CNAME chains, to a limited depth. In this
33 day-and-age we expect the resolver to be doing this for us, so the loop
34 is limited to one retry unless the (new) config option dns_cname_loops
40 * DANE and SPF have been promoted from Experimental to Supported status, thus
41 the options to enable them in Local/Makefile have been renamed.
42 See current src/EDITME for full details, including changes in dependencies,
43 but loosely: replace EXPERIMENTAL_SPF with SUPPORT_SPF and replace
44 EXPERIMENTAL_DANE with SUPPORT_DANE.
46 * Ancient ClamAV stream support, long deprecated by ClamAV, has been removed;
47 if you were building with WITH_OLD_CLAMAV_STREAM enabled then your problems
48 have marginally increased.
50 * A number of logging changes; if relying upon the previous DKIM additional
51 log-line, explicit log_selector configuration is needed to keep it.
53 * Other incompatible changes in EXPERIMENTAL_* features, read NewStuff and
54 ChangeLog carefully if relying upon an experimental feature such as DMARC.
55 Note that this includes changes to SPF as it was promoted into Supported.
61 * SMTP CHUNKING in Exim 4.88 did not ensure that received mails had a final
62 newline; attempts to deliver such messages onwards to non-chunking hosts
63 would probably hang, as Exim does not insert the newline before a ".".
64 In 4.89, the newline is added upon receipt. For already-received messages
65 in your queue, try util/chunking_fixqueue_finalnewlines.pl
66 to walk the queue, fixing any affected messages. Note that because a
67 delivery attempt will be hanging, attempts to lock the messages for fixing
68 them will stall; stopping all queue-runners temporarily is recommended.
70 * OpenSSL: oldest supported release series is now 1.0.2, which is the oldest
71 supported by the OpenSSL project. If you can build Exim with an older
72 release series, congratulations. If you can't, then upgrade.
73 The file doc/openssl.txt contains instructions for installing a current
74 OpenSSL outside the system library paths and building Exim to use it.
76 * FreeBSD: we now always use the system iconv in libc, as all versions of
77 FreeBSD supported by the FreeBSD project provide this functionality.
83 * The "demime" ACL condition, deprecated for the past 10 years, has
86 * Old GnuTLS configuration options "gnutls_require_kx", "gnutls_require_mac",
87 and "gnutls_require_protocols" have now been removed. (Inoperative from
88 4.80, per below; logging warnings since 4.83, again per below).
94 * SPF condition results renamed "permerror" and "temperror". The old
95 names are still accepted for back-compatability, for this release.
97 * TLS details are now logged on rejects, subject to log selectors.
99 * Items in headers_remove lists must now have any embedded list-separators
102 * Attempted use of the deprecated options "gnutls_require_kx" et. al.
103 now result in logged warning.
109 * New option gnutls_allow_auto_pkcs11 defaults false; if you have GnuTLS 2.12.0
110 or later and do want PKCS11 modules to be autoloaded, then set this option.
112 * A per-transport wait-<name> database is no longer updated if the transport
113 sets "connection_max_messages" to 1, as it can not be used and causes
114 unnecessary serialisation and load. External tools tracking the state of
115 Exim by the hints databases may need modification to take this into account.
117 * The av_scanner option can now accept multiple clamd TCP targets, all other
118 setting limitations remain.
124 * BEWARE backwards-incompatible changes in SSL libraries, thus the version
125 bump. See points below for details.
126 Also an LDAP data returned format change.
128 * The value of $tls_peerdn is now print-escaped when written to the spool file
129 in a -tls_peerdn line, and unescaped when read back in. We received reports
130 of values with embedded newlines, which caused spool file corruption.
132 If you have a corrupt spool file and you wish to recover the contents after
133 upgrading, then lock the message, replace the new-lines that should be part
134 of the -tls_peerdn line with the two-character sequence \n and then unlock
135 the message. No tool has been provided as we believe this is a rare
138 * For OpenSSL, SSLv2 is now disabled by default. (GnuTLS does not support
139 SSLv2). RFC 6176 prohibits SSLv2 and some informal surveys suggest no
140 actual usage. You can re-enable with the "openssl_options" Exim option,
141 in the main configuration section. Note that supporting SSLv2 exposes
142 you to ciphersuite downgrade attacks.
144 * With OpenSSL 1.0.1+, Exim now supports TLS 1.1 and TLS 1.2. If built
145 against 1.0.1a then you will get a warning message and the
146 "openssl_options" value will not parse "no_tlsv1_1": the value changes
147 incompatibly between 1.0.1a and 1.0.1b, because the value chosen for 1.0.1a
148 is infelicitous. We advise avoiding 1.0.1a.
150 "openssl_options" gains "no_tlsv1_1", "no_tlsv1_2" and "no_compression".
152 COMPATIBILITY WARNING: The default value of "openssl_options" is no longer
153 "+dont_insert_empty_fragments". We default to "+no_sslv2".
154 That old default was grandfathered in from before openssl_options became a
155 configuration option.
156 Empty fragments are inserted by default through TLS1.0, to partially defend
157 against certain attacks; TLS1.1+ change the protocol so that this is not
158 needed. The DIEF SSL option was required for some old releases of mail
159 clients which did not gracefully handle the empty fragments, and was
160 initially set in Exim release 4.31 (see ChangeLog, item 37).
162 If you still have affected mail-clients, and you see SSL protocol failures
163 with this release of Exim, set:
164 openssl_options = +dont_insert_empty_fragments
165 in the main section of your Exim configuration file. You're trading off
166 security for compatibility. Exim is now defaulting to higher security and
167 rewarding more modern clients.
169 If the option tls_dhparams is set and the parameters loaded from the file
170 have a bit-count greater than the new option tls_dh_max_bits, then the file
171 will now be ignored. If this affects you, raise the tls_dh_max_bits limit.
172 We suspect that most folks are using dated defaults and will not be affected.
174 * Ldap lookups returning multi-valued attributes now separate the attributes
175 with only a comma, not a comma-space sequence. Also, an actual comma within
176 a returned attribute is doubled. This makes it possible to parse the
177 attribute as a comma-separated list. Note the distinction from multiple
178 attributes being returned, where each one is a name=value pair.
180 If you are currently splitting the results from LDAP upon a comma, then you
181 should check carefully to see if adjustments are needed.
183 This change lets cautious folks distinguish "comma used as separator for
184 joining values" from "comma inside the data".
186 * accept_8bitmime now defaults on, which is not RFC compliant but is better
187 suited to today's Internet. See http://cr.yp.to/smtp/8bitmime.html for a
188 sane rationale. Those who wish to be strictly RFC compliant, or know that
189 they need to talk to servers that are not 8-bit-clean, now need to take
190 explicit configuration action to default this option off. This is not a
191 new option, you can safely force it off before upgrading, to decouple
192 configuration changes from the binary upgrade while remaining RFC compliant.
194 * The GnuTLS support has been mostly rewritten, to use APIs which don't cause
195 deprecation warnings in GnuTLS 2.12.x. As part of this, these three options
196 are no longer supported:
200 gnutls_require_protocols
202 Their functionality is entirely subsumed into tls_require_ciphers. In turn,
203 tls_require_ciphers is no longer an Exim list and is not parsed by Exim, but
204 is instead given to gnutls_priority_init(3), which expects a priority string;
205 this behaviour is much closer to the OpenSSL behaviour. See:
207 http://www.gnutls.org/manual/html_node/Priority-Strings.html
209 for fuller documentation of the strings parsed. The three gnutls_require_*
210 options are still parsed by Exim and, for this release, silently ignored.
211 A future release will add warnings, before a later still release removes
212 parsing entirely and the presence of the options will be a configuration
215 Note that by default, GnuTLS will not accept RSA-MD5 signatures in chains.
216 A tls_require_ciphers value of NORMAL:%VERIFY_ALLOW_SIGN_RSA_MD5 may
217 re-enable support, but this is not supported by the Exim maintainers.
218 Our test suite no longer includes MD5-based certificates.
220 This rewrite means that Exim will continue to build against GnuTLS in the
221 future, brings Exim closer to other GnuTLS applications and lets us add
222 support for SNI and other features more readily. We regret that it wasn't
223 feasible to retain the three dropped options.
225 * If built with TLS support, then Exim will now validate the value of
226 the main section tls_require_ciphers option at start-up. Before, this
227 would cause a STARTTLS 4xx failure, now it causes a failure to start.
228 Running with a broken configuration which causes failures that may only
229 be left in the logs has been traded off for something more visible. This
230 change makes an existing problem more prominent, but we do not believe
231 anyone would deliberately be running with an invalid tls_require_ciphers
234 This also means that library linkage issues caused by conflicts of some
235 kind might take out the main daemon, not just the delivery or receiving
236 process. Conceivably some folks might prefer to continue delivering
237 mail plaintext when their binary is broken in this way, if there is a
238 server that is a candidate to receive such mails that does not advertise
239 STARTTLS. Note that Exim is typically a setuid root binary and given
240 broken linkage problems that cause segfaults, we feel it is safer to
241 fail completely. (The check is not done as root, to ensure that problems
242 here are not made worse by the check).
244 * The "tls_dhparam" option has been updated, so that it can now specify a
245 path or an identifier for a standard DH prime from one of a few RFCs.
246 The default for OpenSSL is no longer to not use DH but instead to use
247 one of these standard primes. The default for GnuTLS is no longer to use
248 a file in the spool directory, but to use that same standard prime.
249 The option is now used by GnuTLS too. If it points to a path, then
250 GnuTLS will use that path, instead of a file in the spool directory;
251 GnuTLS will attempt to create it if it does not exist.
253 To preserve the previous behaviour of generating files in the spool
254 directory, set "tls_dhparam = historic". Since prior releases of Exim
255 ignored tls_dhparam when using GnuTLS, this can safely be done before
263 * GnuTLS will now attempt to use TLS 1.2 and TLS 1.1 before TLS 1.0 and SSL3,
264 if supported by your GnuTLS library. Use the existing
265 "gnutls_require_protocols" option to downgrade this if that will be a
266 problem. Prior to this release, supported values were "TLS1" and "SSL3",
267 so you should be able to update configuration prior to update.
269 [nb: gnutls_require_protocols removed in Exim 4.80, instead use
270 tls_require_ciphers to provide a priority string; see notes above]
272 * The match_<type>{string1}{string2} expansion conditions no longer subject
273 string2 to string expansion, unless Exim was built with the new
274 "EXPAND_LISTMATCH_RHS" option. Too many people have inadvertently created
275 insecure configurations that way. If you need the functionality and turn on
276 that build option, please let the developers know, and know why, so we can
277 try to provide a safer mechanism for you.
279 The match{}{} expansion condition (for regular expressions) is NOT affected.
280 For match_<type>{s1}{s2}, all list functionality is unchanged. The only
281 change is that a '$' appearing in s2 will not trigger expansion, but instead
282 will be treated as a literal $ sign; the effect is very similar to having
283 wrapped s2 with \N...\N. If s2 contains a named list and the list definition
284 uses $expansions then those _will_ be processed as normal. It is only the
285 point at which s2 is read where expansion is inhibited.
287 If you are trying to test if two email addresses are equal, use eqi{s1}{s2}.
288 If you are testing if the address in s1 occurs in the list of items given
289 in s2, either use the new inlisti{s1}{s2} condition (added in 4.77) or use
290 the pre-existing forany{s2}{eqi{$item}{s1}} condition.
296 * The integrated support for dynamically loadable lookup modules has an ABI
297 change from the modules supported by some OS vendors through an unofficial
298 patch. Don't try to mix & match.
300 * Some parts of the build system are now beginning to assume that the host
301 environment is POSIX. If you're building on a system where POSIX tools are
302 not the default, you might have an easier time if you switch to the POSIX
303 tools. Feel free to report non-POSIX issues as a request for a feature
304 enhancement, but if the POSIX variants are available then the fix will
305 probably just involve some coercion. See the README instructions for
306 building on such hosts.
312 * The Exim run-time user can no longer be root; this was always
313 strongly discouraged, but is now prohibited both at build and
314 run-time. If you need Exim to run routinely as root, you'll need to
315 patch the source and accept the risk. Here be dragons.
317 * Exim will no longer accept a configuration file owned by the Exim
318 run-time user, unless that account is explicitly the value in
319 CONFIGURE_OWNER, which we discourage. Exim now checks to ensure that
320 files are not writeable by other accounts.
322 * The ALT_CONFIG_ROOT_ONLY build option is no longer optional and is forced
323 on; the Exim user can, by default, no longer use -C/-D and retain privilege.
324 Two new build options mitigate this.
326 * TRUSTED_CONFIG_LIST defines a file containing a whitelist of config
327 files that are trusted to be selected by the Exim user; one per line.
328 This is the recommended approach going forward.
330 * WHITELIST_D_MACROS defines a colon-separated list of macro names which
331 the Exim run-time user may safely pass without dropping privileges.
332 Because changes to this involve a recompile, this is not the recommended
333 approach but may ease transition. The values of the macros, when
334 overridden, are constrained to match this regex: ^[A-Za-z0-9_/.-]*$
336 * The system_filter_user option now defaults to the Exim run-time user,
337 rather than root. You can still set it explicitly to root and this
338 can be done with prior versions too, letting you roll versions
339 without needing to change this configuration option.
341 * ClamAV must be at least version 0.95 unless WITH_OLD_CLAMAV_STREAM is
342 defined at build time.
348 1. Experimental Yahoo! Domainkeys support has been dropped in this release.
349 It has been superceded by a native implementation of its successor DKIM.
351 2. Up to version 4.69, Exim came with an embedded version of the PCRE library.
352 As of 4.70, this is no longer the case. To compile Exim, you will need PCRE
353 installed. Most OS distributions have ready-made library and development
360 1. The internal implementation of the database keys that are used for ACL
361 ratelimiting has been tidied up. This means that an update to 4.68 might cause
362 Exim to "forget" previous rates that it had calculated, and reset them to zero.
368 1. Callouts were setting the name used for EHLO/HELO from $smtp_active_
369 hostname. This is wrong, because it relates to the incoming message (and
370 probably the interface on which it is arriving) and not to the outgoing
371 callout (which could be using a different interface). This has been
372 changed to use the value of the helo_data option from the smtp transport
373 instead - this is what is used when a message is actually being sent. If
374 there is no remote transport (possible with a router that sets up host
375 addresses), $smtp_active_hostname is used. This change is mentioned here in
376 case somebody is relying on the use of $smtp_active_hostname.
378 2. A bug has been fixed that might just possibly be something that is relied on
379 in some configurations. In expansion items such as ${if >{xxx}{yyy}...} an
380 empty string (that is {}) was being interpreted as if it was {0} and therefore
381 treated as the number zero. From release 4.64, such strings cause an error
382 because a decimal number, possibly followed by K or M, is required (as has
383 always been documented).
385 3. There has been a change to the GnuTLS support (ChangeLog/PH/20) to improve
386 Exim's performance. Unfortunately, this has the side effect of being slightly
387 non-upwards compatible for versions 4.50 and earlier. If you are upgrading from
388 one of these earlier versions and you use GnuTLS, you must remove the file
389 called gnutls-params in Exim's spool directory. If you don't do this, you will
392 TLS error on connection from ... (DH params import): Base64 decoding error.
394 Removing the file causes Exim to recompute the relevant encryption parameters
395 and cache them in the new format that was introduced for release 4.51 (May
396 2005). If you are upgrading from release 4.51 or later, there should be no
403 When an SMTP error message is specified in a "message" modifier in an ACL, or
404 in a :fail: or :defer: message in a redirect router, Exim now checks the start
405 of the message for an SMTP error code. This consists of three digits followed
406 by a space, optionally followed by an extended code of the form n.n.n, also
407 followed by a space. If this is the case and the very first digit is the same
408 as the default error code, the code from the message is used instead. If the
409 very first digit is incorrect, a panic error is logged, and the default code is
410 used. This is an incompatible change, but it is not expected to affect many (if
411 any) configurations. It is possible to suppress the use of the supplied code in
412 a redirect router by setting the smtp_error_code option false. In this case,
413 any SMTP code is quietly ignored.
419 1. The default number of ACL variables of each type has been increased to 20,
420 and it's possible to compile Exim with more. You can safely upgrade to this
421 release if you already have messages on the queue with saved ACL variable
422 values. However, if you downgrade from this release with messages on the queue,
423 any saved ACL values they may have will be lost.
425 2. The default value for rfc1413_query_timeout has been changed from 30s to 5s.
431 There was a problem with 4.52/TF/02 in that a "name=" option on control=
432 submission terminated at the next slash, thereby not allowing for slashes in
433 the name. This has been changed so that "name=" takes the rest of the string as
434 its data. It must therefore be the last option.
440 If you are using the experimental Domain Keys support, you must upgrade to
441 at least libdomainkeys 0.67 in order to run this release of Exim.
447 1. The format in which GnuTLS parameters are cached (in the file gnutls-params
448 in the spool directory) has been changed. The new format can also be generated
449 externally, so it is now possible to update the values from outside Exim. This
450 has been implemented in an upwards, BUT NOT downwards, compatible manner.
451 Upgrading should be seamless: when Exim finds that it cannot understand an
452 existing cache file, it generates new parameters and writes them to the cache
453 in the new format. If, however, you downgrade from 4.51 to a previous release,
454 you MUST delete the gnutls-params file in the spool directory, because the
455 older Exim will not recognize the new format.
457 2. When doing a callout as part of verifying an address, Exim was not paying
458 attention to any local part prefix or suffix that was matched by the router
459 that accepted the address. It now behaves in the same way as it does for
460 delivery: the affixes are removed from the local part unless
461 rcpt_include_affixes is set on the transport. If you have a configuration that
462 uses prefixes or suffixes on addresses that could be used for callouts, and you
463 want the affixes to be retained, you must make sure that rcpt_include_affixes
464 is set on the transport.
466 3. Bounce and delay warning messages no longer contain details of delivery
467 errors, except for explicit messages (e.g. generated by :fail:) and SMTP
468 responses from remote hosts.
474 The exicyclog script has been updated to use three-digit numbers in rotated log
475 files if the maximum number to keep is greater than 99. If you are already
476 keeping more than 99, there will be an incompatible change when you upgrade.
477 You will probably want to rename your old log files to the new form before
478 running the new exicyclog.
484 RFC 3848 specifies standard names for the "with" phrase in Received: header
485 lines when AUTH and/or TLS are in use. This is the "received protocol"
486 field. Exim used to use "asmtp" for authenticated SMTP, without any
487 indication (in the protocol name) for TLS use. Now it follows the RFC and
488 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
489 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These names
490 appear in log lines as well as in Received: header lines.
496 Change 4.31/2 gave problems to data ACLs and local_scan() functions that
497 expected to see a Received: header. I have changed to yet another scheme. The
498 Received: header is now generated after the body is received, but before the
499 ACL or local_scan() is called. After they have run, the timestamp in the
500 Received: header is updated.
502 Thus, change (a) of 4.31/2 has been reversed, but change (b) is still true,
503 which is lucky, since I decided it was a bug fix.
509 If an expansion in a condition on a "warn" statement fails because a lookup
510 defers, the "warn" statement is abandoned, and the next ACL statement is
511 processed. Previously this caused the whole ACL to be aborted.
517 Change 4.31/2 has been reversed, as it proved contentious. Recipient callout
518 verification now uses <> in the MAIL command by default, as it did before. A
519 new callout option, "use_sender", has been added to request the other
526 1. If you compile Exim to use GnuTLS, it now requires the use of release 1.0.0
527 or greater. The interface to the obsolete 0.8.x releases is no longer
528 supported. There is one externally visible change: the format for the
529 display of Distinguished Names now uses commas as a separator rather than a
530 slash. This is to comply with RFC 2253.
532 2. When a message is received, the Received: header line is now generated when
533 reception is complete, instead of at the start of reception. For messages
534 that take a long time to come in, this changes the meaning of the timestamp.
535 There are several side-effects of this change:
537 (a) If a message is rejected by a DATA or non-SMTP ACL, or by local_scan(),
538 the logged header lines no longer include the local Received: line,
539 because it has not yet been created. If the message is a non-SMTP one,
540 and the error is processed by sending a message to the sender, the copy
541 of the original message that is returned does not have an added
544 (b) When a filter file is tested using -bf, no additional Received: header
545 is added to the test message. After some thought, I decided that this
548 The contents of $received_for are not affected by this change. This
549 variable still contains the single recipient of a message, copied after
550 addresses have been rewritten, but before local_scan() is run.
552 2. Recipient callout verification, like sender verification, was using <> in
553 the MAIL FROM command. This isn't really the right thing, since the actual
554 sender may affect whether the remote host accepts the recipient or not. I
555 have changed it to use the actual sender in the callout; this means that
556 the cache record is now keyed on a recipient/sender pair, not just the
557 recipient address. There doesn't seem to be a real danger of callout loops,
558 since a callout by the remote host to check the sender would use <>.
564 1. I have abolished timeout_DNS as an error that can be detected in retry
565 rules, because it has never worked. Despite the fact that it has been
566 documented since at least release 1.62, there was no code to support it.
567 If you have used it in your retry rules, you will now get a warning message
568 to the log and panic log. It is now treated as plain "timeout".
570 2. After discussion on the mailing list, Exim no longer adds From:, Date:, or
571 Message-Id: header lines to messages that do not originate locally, that is,
572 messages that have an associated sending host address.
574 3. When looking up a host name from an IP address, Exim now tries the DNS
575 first, and only if that fails does it use gethostbyaddr() (or equivalent).
576 This change was made because on some OS, not all the names are given for
577 addresses with multiple PTR records via the gethostbyaddr() interface. The
578 order of lookup can be changed by setting host_lookup_order.
584 1. The new FIXED_NEVER_USERS build-time option creates a list of "never users"
585 that cannot be overridden. The default in the distributed EDITME is "root".
586 If for some reason you were (against advice) running deliveries as root, you
587 will have to ensure that FIXED_NEVER_USERS is not set in your
590 2. The ${quote: operator now quotes an empty string, which it did not before.
592 3. Version 4.23 saves the contents of the ACL variables with the message, so
593 that they can be used later. If one of these variables contains a newline,
594 there will be a newline character in the spool that will not be interpreted
595 correctly by a previous version of Exim. (Exim ignores keyed spool file
596 items that it doesn't understand - precisely for this kind of problem - but
597 it expects them all to be on one line.)
599 So the bottom line is: if you have newlines in your ACL variables, you
600 cannot retreat from 4.23.
606 1. The idea of the "warn" ACL verb is that it adds a header or writes to the
607 log only when "message" or "log_message" are set. However, if one of the
608 conditions was an address verification, or a call to a nested ACL, the
609 messages generated by the underlying test were being passed through. This
610 no longer happens. The underlying message is available in $acl_verify_
611 message for both "message" and "log_message" expansions, so it can be
612 passed through if needed.
614 2. The way that the $h_ (and $header_) expansions work has been changed by the
615 addition of RFC 2047 decoding. See the main documentation (the NewStuff file
616 until release 4.30, then the manual) for full details. Briefly, there are
619 $rh_xxx: and $rheader_xxx: give the original content of the header
620 line(s), with no processing at all.
622 $bh_xxx: and $bheader_xxx: remove leading and trailing white space, and
623 then decode base64 or quoted-printable "words" within the header text,
624 but do not do charset translation.
626 $h_xxx: and $header_xxx: attempt to translate the $bh_ string to a
627 standard character set.
629 If you have previously been using $h_ expansions to access the raw
630 characters, you should change to $rh_ instead.
632 3. When Exim creates an RFC 2047 encoded word in a header line, it labels it
633 with the default character set from the headers_charset option instead of
634 always using iso-8859-1.
636 4. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
637 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
638 called TMPDIR, and if it finds it is different, it changes its value.
640 5. Following a discussion on the list, the rules by which Exim recognises line
641 endings on incoming messages have been changed. The -dropcr and drop_cr
642 options are now no-ops, retained only for backwards compatibility. The
643 following line terminators are recognized: LF CRLF CR. However, special
644 processing applies to CR:
646 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
647 nor a local message in the state where . is a terminator.
649 (ii) If a bare CR is encountered in a header line, an extra space is added
650 after the line terminator so as not to end the header. The reasoning
651 behind this is that bare CRs in header lines are most likely either
652 to be mistakes, or people trying to play silly games.
654 6. The code for using daemon_smtp_port, local_interfaces, and the -oX options
655 has been reorganized. It is supposed to be backwards compatible, but it is
656 mentioned here just in case I've screwed up.
663 1. I have tidied and re-organized the code that uses alarm() for imposing time
664 limits on various things. It shouldn't affect anything, but if you notice
665 processes getting stuck, it may be that I've broken something.
667 2. The "arguments" log selector now also logs the current working directory
670 3. An incompatible change has been made to the appendfile transport. This
671 affects the case when it is used for file deliveries that are set up by
672 .forward and filter files. Previously, any settings of the "file" or
673 "directory" options were ignored. It is hoped that, like the address_file
674 transport in the default configuration, these options were never in fact set
675 on such transports, because they were of no use.
677 Now, if either of these options is set, it is used. The path that is passed
678 by the router is in $address_file (this is not new), so it can be used as
679 part of a longer path, or modified in any other way that expansion permits.
681 If neither "file" nor "directory" is set, the behaviour is unchanged.
683 4. Related to the above: in a filter, if a "save" command specifies a non-
684 absolute path, the value of $home/ is pre-pended. This no longer happens if
685 $home is unset or is set to an empty string.
687 5. Multiple file deliveries from a filter or .forward file can never be
688 batched; the value of batch_max on the transport is ignored for file
689 deliveries. I'm assuming that nobody ever actually set batch_max on the
690 address_file transport - it would have had odd effects previously.
692 6. DESTDIR is the more common variable that ROOT for use when installing
693 software under a different root filing system. The Exim install script now
694 recognizes DESTDIR first; if it is not set, ROOT is used.
696 7. If DESTDIR is set when installing Exim, it no longer prepends its value to
697 the path of the system aliases file that appears in the default
698 configuration (when a default configuration is installed). If an aliases
699 file is actually created, its name *does* use the prefix.
705 1. The default for the maximum number of unknown SMTP commands that Exim will
706 accept before dropping a connection has been reduced from 5 to 3. However, you
707 can now change the value by setting smtp_max_unknown_commands.
709 2. The ${quote: operator has been changed so that it turns newline and carriage
710 return characters into \n and \r, respectively.
712 3. The file names used for maildir messages now include the microsecond time
713 fraction as well as the time in seconds, to cope with systems where the process
714 id can be re-used within the same second. The format is now
716 <time>.H<microsec>P<pid>.<host>
718 This should be a compatible change, but is noted here just in case.
720 4. The rules for creating message ids have changed, to cope with systems where
721 the process id can be re-used within the same second. The format, however, is
722 unchanged, so this should not cause any problems, except as noted in the next
725 5. The maximum value for localhost_number has been reduced from 255 to 16, in
726 order to implement the new message id rules. For operating systems that have
727 case-insensitive file systems (Cygwin and Darwin), the limit is 10.
729 6. verify = header_syntax was allowing unqualified addresses in all cases. Now
730 it allows them only for locally generated messages and from hosts that match
731 sender_unqualified_hosts or recipient_unqualified_hosts, respectively.
733 7. For reasons lost in the mists of time, when a pipe transport was run, the
734 environment variable MESSAGE_ID was set to the message ID preceded by 'E' (the
735 form used in Message-ID: header lines). The 'E' has been removed.
741 1. The handling of lines in the configuration file has changed. Previously,
742 macro expansion was applied to logical lines, after continuations had been
743 joined on. This meant that it could not be used in .include lines, which are
744 handled as physical rather than logical lines. Macro expansion is now done on
745 physical lines rather than logical lines. This means there are two
748 (a) A macro that expands to # to turn a line into a comment now applies only
749 to the physical line where it appears. Previously, it would have caused
750 any following continuations also to be ignored.
752 (b) A macro name can no longer be split over the boundary between a line and
753 its continuation. Actually, this is more of a bug fix. :-)
755 2. The -D command line option must now all be within one command line item.
756 This makes it possible to use -D to set a macro to the empty string by commands
762 Previously, these items would have moved on to the next item on the command
763 line. To include spaces in a macro definition item, quotes must be used, in
764 which case you can also have spaces after -D and surrounding the equals. For
767 exim '-D ABC = something' ...
769 3. The way that addresses that redirect to themselves are handled has been
770 changed, in order to fix an obscure bug. This should not cause any problems
771 except in the case of wanting to go back from a 4.11 (or later) release to an
772 earlier release. If there are undelivered messages on the spool that contain
773 addresses which redirect to themselves, and the redirected addresses have
774 already been delivered, you might get a duplicate delivery if you revert to an
777 4. The default way of looking up IP addresses for hosts in the manualroute and
778 queryprogram routers has been changed. If "byname" or "bydns" is explicitly
779 specified, there is no change, but if no method is specified, Exim now behaves
782 First, a DNS lookup is done. If this yields anything other than
783 HOST_NOT_FOUND, that result is used. Otherwise, Exim goes on to try a call to
784 getipnodebyname() (or gethostbyname() on older systems) and the result of the
785 lookup is the result of that call.
787 This change has been made because it has been discovered that on some systems,
788 if a DNS lookup called via getipnodebyname() times out, HOST_NOT_FOUND is
789 returned instead of TRY_AGAIN. Thus, it is safest to try a DNS lookup directly
790 first, and only if that gives a definite "no such host" to try the local
793 5. In fixing the minor security problem with pid_file_path, I have removed some
794 backwards-compatible (undocumented) code which was present to ease conversion
795 from Exim 3. In Exim 4, pid_file_path is a literal; in Exim 3 it was allowed to
796 contain "%s", which was replaced by the port number for daemons listening on
797 non-standard ports. In Exim 4, such daemons do not write a pid file. The
798 backwards compatibility feature was to replace "%s" by nothing if it occurred
799 in an Exim 4 setting of pid_file_path. The bug was in this code. I have solved
800 the problem by removing the backwards compatibility feature. Thus, if you still
801 have "%s" somewhere in a setting of pid_file_path, you should remove it.
803 6. There has been an extension to lsearch files. The keys in these files may
804 now be quoted in order to allow for whitespace and colons in them. This means
805 that if you were previously using keys that began with a doublequote, you will
806 now have to wrap them with extra quotes and escape the internal quotes. The
807 possibility that anybody is actually doing this seems extremely remote, but it
808 is documented just in case.
814 The build-time parameter EXIWHAT_KILL_ARG has been renamed EXIWHAT_KILL_SIGNAL
815 to better reflect its function. The OS-specific files have been updated. Only
816 if you have explicitly set this in your Makefile (highly unlikely) do you need