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 * GnuTLS will now attempt to use TLS 1.2 and TLS 1.1 before TLS 1.0 and SSL3,
33 if supported by your GnuTLS library. Use the existing
34 "gnutls_require_protocols" option to downgrade this if that will be a
35 problem. Prior to this release, supported values were "TLS1" and "SSL3",
36 so you should be able to update configuration prior to update.
38 * The match_<type>{string1}{string2} expansion conditions no longer subject
39 string2 to string expansion, unless Exim was built with the new
40 "EXPAND_LISTMATCH_RHS" option. Too many people have inadvertently created
41 insecure configurations that way. If you need the functionality and turn on
42 that build option, please let the developers know, and know why, so we can
43 try to provide a safer mechanism for you.
49 * The integrated support for dynamically loadable lookup modules has an ABI
50 change from the modules supported by some OS vendors through an unofficial
51 patch. Don't try to mix & match.
53 * Some parts of the build system are now beginning to assume that the host
54 environment is POSIX. If you're building on a system where POSIX tools are
55 not the default, you might have an easier time if you switch to the POSIX
56 tools. Feel free to report non-POSIX issues as a request for a feature
57 enhancement, but if the POSIX variants are available then the fix will
58 probably just involve some coercion. See the README instructions for
59 building on such hosts.
65 * The Exim run-time user can no longer be root; this was always
66 strongly discouraged, but is now prohibited both at build and
67 run-time. If you need Exim to run routinely as root, you'll need to
68 patch the source and accept the risk. Here be dragons.
70 * Exim will no longer accept a configuration file owned by the Exim
71 run-time user, unless that account is explicitly the value in
72 CONFIGURE_OWNER, which we discourage. Exim now checks to ensure that
73 files are not writable by other accounts.
75 * The ALT_CONFIG_ROOT_ONLY build option is no longer optional and is forced
76 on; the Exim user can, by default, no longer use -C/-D and retain privilege.
77 Two new build options mitigate this.
79 * TRUSTED_CONFIG_LIST defines a file containing a whitelist of config
80 files that are trusted to be selected by the Exim user; one per line.
81 This is the recommended approach going forward.
83 * WHITELIST_D_MACROS defines a colon-separated list of macro names which
84 the Exim run-time user may safely pass without dropping privileges.
85 Because changes to this involve a recompile, this is not the recommended
86 approach but may ease transition. The values of the macros, when
87 overridden, are constrained to match this regex: ^[A-Za-z0-9_/.-]*$
89 * The system_filter_user option now defaults to the Exim run-time user,
90 rather than root. You can still set it explicitly to root and this
91 can be done with prior versions too, letting you roll versions
92 without needing to change this configuration option.
94 * ClamAV must be at least version 0.95 unless WITH_OLD_CLAMAV_STREAM is
95 defined at build time.
101 1. Experimental Yahoo! Domainkeys support has been dropped in this release.
102 It has been superceded by a native implementation of its successor DKIM.
104 2. Up to version 4.69, Exim came with an embedded version of the PCRE library.
105 As of 4.70, this is no longer the case. To compile Exim, you will need PCRE
106 installed. Most OS distributions have ready-made library and develoment
113 1. The internal implementation of the database keys that are used for ACL
114 ratelimiting has been tidied up. This means that an update to 4.68 might cause
115 Exim to "forget" previous rates that it had calculated, and reset them to zero.
121 1. Callouts were setting the name used for EHLO/HELO from $smtp_active_
122 hostname. This is wrong, because it relates to the incoming message (and
123 probably the interface on which it is arriving) and not to the outgoing
124 callout (which could be using a different interface). This has been
125 changed to use the value of the helo_data option from the smtp transport
126 instead - this is what is used when a message is actually being sent. If
127 there is no remote transport (possible with a router that sets up host
128 addresses), $smtp_active_hostname is used. This change is mentioned here in
129 case somebody is relying on the use of $smtp_active_hostname.
131 2. A bug has been fixed that might just possibly be something that is relied on
132 in some configurations. In expansion items such as ${if >{xxx}{yyy}...} an
133 empty string (that is {}) was being interpreted as if it was {0} and therefore
134 treated as the number zero. From release 4.64, such strings cause an error
135 because a decimal number, possibly followed by K or M, is required (as has
136 always been documented).
138 3. There has been a change to the GnuTLS support (ChangeLog/PH/20) to improve
139 Exim's performance. Unfortunately, this has the side effect of being slightly
140 non-upwards compatible for versions 4.50 and earlier. If you are upgrading from
141 one of these earlier versions and you use GnuTLS, you must remove the file
142 called gnutls-params in Exim's spool directory. If you don't do this, you will
145 TLS error on connection from ... (DH params import): Base64 decoding error.
147 Removing the file causes Exim to recompute the relevant encryption parameters
148 and cache them in the new format that was introduced for release 4.51 (May
149 2005). If you are upgrading from release 4.51 or later, there should be no
156 When an SMTP error message is specified in a "message" modifier in an ACL, or
157 in a :fail: or :defer: message in a redirect router, Exim now checks the start
158 of the message for an SMTP error code. This consists of three digits followed
159 by a space, optionally followed by an extended code of the form n.n.n, also
160 followed by a space. If this is the case and the very first digit is the same
161 as the default error code, the code from the message is used instead. If the
162 very first digit is incorrect, a panic error is logged, and the default code is
163 used. This is an incompatible change, but it is not expected to affect many (if
164 any) configurations. It is possible to suppress the use of the supplied code in
165 a redirect router by setting the smtp_error_code option false. In this case,
166 any SMTP code is quietly ignored.
172 1. The default number of ACL variables of each type has been increased to 20,
173 and it's possible to compile Exim with more. You can safely upgrade to this
174 release if you already have messages on the queue with saved ACL variable
175 values. However, if you downgrade from this release with messages on the queue,
176 any saved ACL values they may have will be lost.
178 2. The default value for rfc1413_query_timeout has been changed from 30s to 5s.
184 There was a problem with 4.52/TF/02 in that a "name=" option on control=
185 submission terminated at the next slash, thereby not allowing for slashes in
186 the name. This has been changed so that "name=" takes the rest of the string as
187 its data. It must therefore be the last option.
193 If you are using the experimental Domain Keys support, you must upgrade to
194 at least libdomainkeys 0.67 in order to run this release of Exim.
200 1. The format in which GnuTLS parameters are cached (in the file gnutls-params
201 in the spool directory) has been changed. The new format can also be generated
202 externally, so it is now possible to update the values from outside Exim. This
203 has been implemented in an upwards, BUT NOT downwards, compatible manner.
204 Upgrading should be seamless: when Exim finds that it cannot understand an
205 existing cache file, it generates new parameters and writes them to the cache
206 in the new format. If, however, you downgrade from 4.51 to a previous release,
207 you MUST delete the gnutls-params file in the spool directory, because the
208 older Exim will not recognize the new format.
210 2. When doing a callout as part of verifying an address, Exim was not paying
211 attention to any local part prefix or suffix that was matched by the router
212 that accepted the address. It now behaves in the same way as it does for
213 delivery: the affixes are removed from the local part unless
214 rcpt_include_affixes is set on the transport. If you have a configuration that
215 uses prefixes or suffixes on addresses that could be used for callouts, and you
216 want the affixes to be retained, you must make sure that rcpt_include_affixes
217 is set on the transport.
219 3. Bounce and delay warning messages no longer contain details of delivery
220 errors, except for explicit messages (e.g. generated by :fail:) and SMTP
221 responses from remote hosts.
227 The exicyclog script has been updated to use three-digit numbers in rotated log
228 files if the maximum number to keep is greater than 99. If you are already
229 keeping more than 99, there will be an incompatible change when you upgrade.
230 You will probably want to rename your old log files to the new form before
231 running the new exicyclog.
237 RFC 3848 specifies standard names for the "with" phrase in Received: header
238 lines when AUTH and/or TLS are in use. This is the "received protocol"
239 field. Exim used to use "asmtp" for authenticated SMTP, without any
240 indication (in the protocol name) for TLS use. Now it follows the RFC and
241 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
242 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These names
243 appear in log lines as well as in Received: header lines.
249 Change 4.31/2 gave problems to data ACLs and local_scan() functions that
250 expected to see a Received: header. I have changed to yet another scheme. The
251 Received: header is now generated after the body is received, but before the
252 ACL or local_scan() is called. After they have run, the timestamp in the
253 Received: header is updated.
255 Thus, change (a) of 4.31/2 has been reversed, but change (b) is still true,
256 which is lucky, since I decided it was a bug fix.
262 If an expansion in a condition on a "warn" statement fails because a lookup
263 defers, the "warn" statement is abandoned, and the next ACL statement is
264 processed. Previously this caused the whole ACL to be aborted.
270 Change 4.31/2 has been reversed, as it proved contentious. Recipient callout
271 verification now uses <> in the MAIL command by default, as it did before. A
272 new callout option, "use_sender", has been added to request the other
279 1. If you compile Exim to use GnuTLS, it now requires the use of release 1.0.0
280 or greater. The interface to the obsolete 0.8.x releases is no longer
281 supported. There is one externally visible change: the format for the
282 display of Distinguished Names now uses commas as a separator rather than a
283 slash. This is to comply with RFC 2253.
285 2. When a message is received, the Received: header line is now generated when
286 reception is complete, instead of at the start of reception. For messages
287 that take a long time to come in, this changes the meaning of the timestamp.
288 There are several side-effects of this change:
290 (a) If a message is rejected by a DATA or non-SMTP ACL, or by local_scan(),
291 the logged header lines no longer include the local Received: line,
292 because it has not yet been created. If the message is a non-SMTP one,
293 and the error is processed by sending a message to the sender, the copy
294 of the original message that is returned does not have an added
297 (b) When a filter file is tested using -bf, no additional Received: header
298 is added to the test message. After some thought, I decided that this
301 The contents of $received_for are not affected by this change. This
302 variable still contains the single recipient of a message, copied after
303 addresses have been rewritten, but before local_scan() is run.
305 2. Recipient callout verification, like sender verification, was using <> in
306 the MAIL FROM command. This isn't really the right thing, since the actual
307 sender may affect whether the remote host accepts the recipient or not. I
308 have changed it to use the actual sender in the callout; this means that
309 the cache record is now keyed on a recipient/sender pair, not just the
310 recipient address. There doesn't seem to be a real danger of callout loops,
311 since a callout by the remote host to check the sender would use <>.
317 1. I have abolished timeout_DNS as an error that can be detected in retry
318 rules, because it has never worked. Despite the fact that it has been
319 documented since at least release 1.62, there was no code to support it.
320 If you have used it in your retry rules, you will now get a warning message
321 to the log and panic log. It is now treated as plain "timeout".
323 2. After discussion on the mailing list, Exim no longer adds From:, Date:, or
324 Message-Id: header lines to messages that do not originate locally, that is,
325 messages that have an associated sending host address.
327 3. When looking up a host name from an IP address, Exim now tries the DNS
328 first, and only if that fails does it use gethostbyaddr() (or equivalent).
329 This change was made because on some OS, not all the names are given for
330 addresses with multiple PTR records via the gethostbyaddr() interface. The
331 order of lookup can be changed by setting host_lookup_order.
337 1. The new FIXED_NEVER_USERS build-time option creates a list of "never users"
338 that cannot be overridden. The default in the distributed EDITME is "root".
339 If for some reason you were (against advice) running deliveries as root, you
340 will have to ensure that FIXED_NEVER_USERS is not set in your
343 2. The ${quote: operator now quotes an empty string, which it did not before.
345 3. Version 4.23 saves the contents of the ACL variables with the message, so
346 that they can be used later. If one of these variables contains a newline,
347 there will be a newline character in the spool that will not be interpreted
348 correctely by a previous version of Exim. (Exim ignores keyed spool file
349 items that it doesn't understand - precisely for this kind of problem - but
350 it expects them all to be on one line.)
352 So the bottom line is: if you have newlines in your ACL variables, you
353 cannot retreat from 4.23.
359 1. The idea of the "warn" ACL verb is that it adds a header or writes to the
360 log only when "message" or "log_message" are set. However, if one of the
361 conditions was an address verification, or a call to a nested ACL, the
362 messages generated by the underlying test were being passed through. This
363 no longer happens. The underlying message is available in $acl_verify_
364 message for both "message" and "log_message" expansions, so it can be
365 passed through if needed.
367 2. The way that the $h_ (and $header_) expansions work has been changed by the
368 addition of RFC 2047 decoding. See the main documentation (the NewStuff file
369 until release 4.30, then the manual) for full details. Briefly, there are
372 $rh_xxx: and $rheader_xxx: give the original content of the header
373 line(s), with no processing at all.
375 $bh_xxx: and $bheader_xxx: remove leading and trailing white space, and
376 then decode base64 or quoted-printable "words" within the header text,
377 but do not do charset translation.
379 $h_xxx: and $header_xxx: attempt to translate the $bh_ string to a
380 standard character set.
382 If you have previously been using $h_ expansions to access the raw
383 characters, you should change to $rh_ instead.
385 3. When Exim creates an RFC 2047 encoded word in a header line, it labels it
386 with the default character set from the headers_charset option instead of
387 always using iso-8859-1.
389 4. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
390 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
391 called TMPDIR, and if it finds it is different, it changes its value.
393 5. Following a discussion on the list, the rules by which Exim recognises line
394 endings on incoming messages have been changed. The -dropcr and drop_cr
395 options are now no-ops, retained only for backwards compatibility. The
396 following line terminators are recognized: LF CRLF CR. However, special
397 processing applies to CR:
399 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
400 nor a local message in the state where . is a terminator.
402 (ii) If a bare CR is encountered in a header line, an extra space is added
403 after the line terminator so as not to end the header. The reasoning
404 behind this is that bare CRs in header lines are most likely either
405 to be mistakes, or people trying to play silly games.
407 6. The code for using daemon_smtp_port, local_interfaces, and the -oX options
408 has been reorganized. It is supposed to be backwards compatible, but it is
409 mentioned here just in case I've screwed up.
416 1. I have tidied and re-organized the code that uses alarm() for imposing time
417 limits on various things. It shouldn't affect anything, but if you notice
418 processes getting stuck, it may be that I've broken something.
420 2. The "arguments" log selector now also logs the current working directory
423 3. An incompatible change has been made to the appendfile transport. This
424 affects the case when it is used for file deliveries that are set up by
425 .forward and filter files. Previously, any settings of the "file" or
426 "directory" options were ignored. It is hoped that, like the address_file
427 transport in the default configuration, these options were never in fact set
428 on such transports, because they were of no use.
430 Now, if either of these options is set, it is used. The path that is passed
431 by the router is in $address_file (this is not new), so it can be used as
432 part of a longer path, or modified in any other way that expansion permits.
434 If neither "file" nor "directory" is set, the behaviour is unchanged.
436 4. Related to the above: in a filter, if a "save" command specifies a non-
437 absolute path, the value of $home/ is pre-pended. This no longer happens if
438 $home is unset or is set to an empty string.
440 5. Multiple file deliveries from a filter or .forward file can never be
441 batched; the value of batch_max on the transport is ignored for file
442 deliveries. I'm assuming that nobody ever actually set batch_max on the
443 address_file transport - it would have had odd effects previously.
445 6. DESTDIR is the more common variable that ROOT for use when installing
446 software under a different root filing system. The Exim install script now
447 recognizes DESTDIR first; if it is not set, ROOT is used.
449 7. If DESTDIR is set when installing Exim, it no longer prepends its value to
450 the path of the system aliases file that appears in the default
451 configuration (when a default configuration is installed). If an aliases
452 file is actually created, its name *does* use the prefix.
458 1. The default for the maximum number of unknown SMTP commands that Exim will
459 accept before dropping a connection has been reduced from 5 to 3. However, you
460 can now change the value by setting smtp_max_unknown_commands.
462 2. The ${quote: operator has been changed so that it turns newline and carriage
463 return characters into \n and \r, respectively.
465 3. The file names used for maildir messages now include the microsecond time
466 fraction as well as the time in seconds, to cope with systems where the process
467 id can be re-used within the same second. The format is now
469 <time>.H<microsec>P<pid>.<host>
471 This should be a compatible change, but is noted here just in case.
473 4. The rules for creating message ids have changed, to cope with systems where
474 the process id can be re-used within the same second. The format, however, is
475 unchanged, so this should not cause any problems, except as noted in the next
478 5. The maximum value for localhost_number has been reduced from 255 to 16, in
479 order to implement the new message id rules. For operating systems that have
480 case-insensitive file systems (Cygwin and Darwin), the limit is 10.
482 6. verify = header_syntax was allowing unqualified addresses in all cases. Now
483 it allows them only for locally generated messages and from hosts that match
484 sender_unqualified_hosts or recipient_unqualified_hosts, respectively.
486 7. For reasons lost in the mists of time, when a pipe transport was run, the
487 environment variable MESSAGE_ID was set to the message ID preceded by 'E' (the
488 form used in Message-ID: header lines). The 'E' has been removed.
494 1. The handling of lines in the configuration file has changed. Previously,
495 macro expansion was applied to logical lines, after continuations had been
496 joined on. This meant that it could not be used in .include lines, which are
497 handled as physical rather than logical lines. Macro expansion is now done on
498 physical lines rather than logical lines. This means there are two
501 (a) A macro that expands to # to turn a line into a comment now applies only
502 to the physical line where it appears. Previously, it would have caused
503 any following continuations also to be ignored.
505 (b) A macro name can no longer be split over the boundary between a line and
506 its continuation. Actually, this is more of a bug fix. :-)
508 2. The -D command line option must now all be within one command line item.
509 This makes it possible to use -D to set a macro to the empty string by commands
515 Previously, these items would have moved on to the next item on the command
516 line. To include spaces in a macro definition item, quotes must be used, in
517 which case you can also have spaces after -D and surrounding the equals. For
520 exim '-D ABC = something' ...
522 3. The way that addresses that redirect to themselves are handled has been
523 changed, in order to fix an obscure bug. This should not cause any problems
524 except in the case of wanting to go back from a 4.11 (or later) release to an
525 earlier release. If there are undelivered messages on the spool that contain
526 addresses which redirect to themselves, and the redirected addresses have
527 already been delivered, you might get a duplicate delivery if you revert to an
530 4. The default way of looking up IP addresses for hosts in the manualroute and
531 queryprogram routers has been changed. If "byname" or "bydns" is explicitly
532 specified, there is no change, but if no method is specified, Exim now behaves
535 First, a DNS lookup is done. If this yields anything other than
536 HOST_NOT_FOUND, that result is used. Otherwise, Exim goes on to try a call to
537 getipnodebyname() (or gethostbyname() on older systems) and the result of the
538 lookup is the result of that call.
540 This change has been made because it has been discovered that on some systems,
541 if a DNS lookup called via getipnodebyname() times out, HOST_NOT_FOUND is
542 returned instead of TRY_AGAIN. Thus, it is safest to try a DNS lookup directly
543 first, and only if that gives a definite "no such host" to try the local
546 5. In fixing the minor security problem with pid_file_path, I have removed some
547 backwards-compatible (undocumented) code which was present to ease conversion
548 from Exim 3. In Exim 4, pid_file_path is a literal; in Exim 3 it was allowed to
549 contain "%s", which was replaced by the port number for daemons listening on
550 non-standard ports. In Exim 4, such daemons do not write a pid file. The
551 backwards compatibility feature was to replace "%s" by nothing if it occurred
552 in an Exim 4 setting of pid_file_path. The bug was in this code. I have solved
553 the problem by removing the backwards compatibility feature. Thus, if you still
554 have "%s" somewhere in a setting of pid_file_path, you should remove it.
556 6. There has been an extension to lsearch files. The keys in these files may
557 now be quoted in order to allow for whitespace and colons in them. This means
558 that if you were previously using keys that began with a doublequote, you will
559 now have to wrap them with extra quotes and escape the internal quotes. The
560 possibility that anybody is actually doing this seems extremely remote, but it
561 is documented just in case.
567 The build-time parameter EXIWHAT_KILL_ARG has been renamed EXIWHAT_KILL_SIGNAL
568 to better reflect its function. The OS-specific files have been updated. Only
569 if you have explicitly set this in your Makefile (highly unlikely) do you need
git://git.exim.org / exim.git / blob