Harmonised TLS library version reporting.
[exim.git] / src / README.UPDATING
CommitLineData
2eda48c6 1$Cambridge: exim/src/README.UPDATING,v 1.17 2009/10/16 07:35:42 tom Exp $
0f4f2a88
PH
2
3This document contains detailed information about incompatibilities that might
4be encountered when upgrading from one release of Exim to another. The
5information is in reverse order of release numbers. Mostly these are relatively
6small points, and the configuration file is normally upwards compatible, but
7there have been two big upheavals...
8
9
10**************************************************************************
11* There was a big reworking of the way mail routing works for release *
12* 4.00. Previously used "directors" were abolished, and all routing is *
13* now done by routers. Policy controls for incoming mail are now done by *
14* Access Control Lists instead of separate options. All this means that *
15* pre-4.00 configuration files have to be massively converted. If you *
16* are coming from a 3.xx release, please read the document in the file *
17* doc/Exim4.upgrade, and allow some time to complete the upgrade. *
18* *
19* There was a big reworking of the way domain/host/net/address lists are *
20* handled at release 3.00. If you are coming from a pre-3.00 release, it *
21* might be easier to start again from a default configuration. Otherwise *
22* you need to read doc/Exim3.upgrade and do a double conversion of your *
23* configuration file. *
24**************************************************************************
25
26
27The rest of this document contains information about changes in 4.xx releases
28that might affect a running system.
2eda48c6 29
e6acf885
PP
30
31Exim version 4.74
32-----------------
33
4050a044
PP
34 * The integrated support for dynamically loadable lookup modules has an ABI
35 change from the modules supported by some OS vendors through an unofficial
36 patch. Don't try to mix & match.
37
38 * Some parts of the build system are now beginning to assume that the host
39 environment is POSIX. If you're building on a system where POSIX tools are
40 not the default, you might have an easier time if you switch to the POSIX
41 tools. Feel free to report non-POSIX issues as a request for a feature
42 enhancement, but if the POSIX variants are available then the fix will
43 probably just involve some coercion. See the README instructions for
44 building on such hosts.
e6acf885
PP
45
46
47Exim version 4.73
48-----------------
49
50 * The Exim run-time user can no longer be root; this was always
51 strongly discouraged, but is now prohibited both at build and
52 run-time. If you need Exim to run routinely as root, you'll need to
53 patch the source and accept the risk. Here be dragons.
54
55 * Exim will no longer accept a configuration file owned by the Exim
56 run-time user, unless that account is explicitly the value in
57 CONFIGURE_OWNER, which we discourage. Exim now checks to ensure that
58 files are not writable by other accounts.
59
60 * The ALT_CONFIG_ROOT_ONLY build option is no longer optional and is forced
61 on; the Exim user can, by default, no longer use -C/-D and retain privilege.
62 Two new build options mitigate this.
63
64 * TRUSTED_CONFIG_LIST defines a file containing a whitelist of config
65 files that are trusted to be selected by the Exim user; one per line.
66 This is the recommended approach going forward.
67
68 * WHITELIST_D_MACROS defines a colon-separated list of macro names which
69 the Exim run-time user may safely pass without dropping privileges.
70 Because changes to this involve a recompile, this is not the recommended
71 approach but may ease transition. The values of the macros, when
72 overriden, are constrained to match this regex: ^[A-Za-z0-9_/.-]*$
73
74 * The system_filter_user option now defaults to the Exim run-time user,
75 rather than root. You can still set it explicitly to root and this
76 can be done with prior versions too, letting you roll versions
77 without needing to change this configuration option.
78
79 * ClamAV must be at least version 0.95 unless WITH_OLD_CLAMAV_STREAM is
80 defined at build time.
81
82
2eda48c6
TK
83Exim version 4.70
84-----------------
85
861. Experimental Yahoo! Domainkeys support has been dropped in this release.
87It has been superceded by a native implementation of its successor DKIM.
88
892. Up to version 4.69, Exim came with an embedded version of the PCRE library.
90As of 4.70, this is no longer the case. To compile Exim, you will need PCRE
91installed. Most OS distributions have ready-made library and develoment
92packages.
0f4f2a88
PH
93
94
3c223874
PH
95Exim version 4.68
96-----------------
97
981. The internal implementation of the database keys that are used for ACL
99ratelimiting has been tidied up. This means that an update to 4.68 might cause
100Exim to "forget" previous rates that it had calculated, and reset them to zero.
101
102
26da7e20
PH
103Exim version 4.64
104-----------------
105
1061. Callouts were setting the name used for EHLO/HELO from $smtp_active_
107hostname. This is wrong, because it relates to the incoming message (and
108probably the interface on which it is arriving) and not to the outgoing
109callout (which could be using a different interface). This has been
110changed to use the value of the helo_data option from the smtp transport
111instead - this is what is used when a message is actually being sent. If
112there is no remote transport (possible with a router that sets up host
113addresses), $smtp_active_hostname is used. This change is mentioned here in
114case somebody is relying on the use of $smtp_active_hostname.
115
ebc9d865
PH
1162. A bug has been fixed that might just possibly be something that is relied on
117in some configurations. In expansion items such as ${if >{xxx}{yyy}...} an
118empty string (that is {}) was being interpreted as if it was {0} and therefore
119treated as the number zero. From release 4.64, such strings cause an error
120because a decimal number, possibly followed by K or M, is required (as has
121always been documented).
122
b6c27e28
PH
1233. There has been a change to the GnuTLS support (ChangeLog/PH/20) to improve
124Exim's performance. Unfortunately, this has the side effect of being slightly
125non-upwards compatible for versions 4.50 and earlier. If you are upgrading from
126one of these earlier versions and you use GnuTLS, you must remove the file
127called gnutls-params in Exim's spool directory. If you don't do this, you will
128see this error:
129
130 TLS error on connection from ... (DH params import): Base64 decoding error.
131
132Removing the file causes Exim to recompute the relevant encryption parameters
133and cache them in the new format that was introduced for release 4.51 (May
1342005). If you are upgrading from release 4.51 or later, there should be no
135problem.
136
26da7e20 137
a5bd321b
PH
138Exim version 4.63
139-----------------
140
141When an SMTP error message is specified in a "message" modifier in an ACL, or
142in a :fail: or :defer: message in a redirect router, Exim now checks the start
143of the message for an SMTP error code. This consists of three digits followed
144by a space, optionally followed by an extended code of the form n.n.n, also
145followed by a space. If this is the case and the very first digit is the same
146as the default error code, the code from the message is used instead. If the
147very first digit is incorrect, a panic error is logged, and the default code is
148used. This is an incompatible change, but it is not expected to affect many (if
149any) configurations. It is possible to suppress the use of the supplied code in
150a redirect router by setting the smtp_error_code option false. In this case,
151any SMTP code is quietly ignored.
152
153
47ca6d6c
PH
154Exim version 4.61
155-----------------
156
8def5aaf
PH
1571. The default number of ACL variables of each type has been increased to 20,
158and it's possible to compile Exim with more. You can safely upgrade to this
159release if you already have messages on the queue with saved ACL variable
160values. However, if you downgrade from this release with messages on the queue,
161any saved ACL values they may have will be lost.
162
1632. The default value for rfc1413_query_timeout has been changed from 30s to 5s.
47ca6d6c
PH
164
165
8857ccfd
PH
166Exim version 4.54
167-----------------
168
169There was a problem with 4.52/TF/02 in that a "name=" option on control=
170submission terminated at the next slash, thereby not allowing for slashes in
171the name. This has been changed so that "name=" takes the rest of the string as
172its data. It must therefore be the last option.
173
174
f3f0a1f0
PH
175Version 4.53
176------------
177
178If you are using the experimental Domain Keys support, you must upgrade to
179at least libdomainkeys 0.67 in order to run this release of Exim.
180
181
b5aea5e1
PH
182Version 4.51
183------------
184
c688b954
PH
1851. The format in which GnuTLS parameters are cached (in the file gnutls-params
186in the spool directory) has been changed. The new format can also be generated
b5aea5e1
PH
187externally, so it is now possible to update the values from outside Exim. This
188has been implemented in an upwards, BUT NOT downwards, compatible manner.
189Upgrading should be seamless: when Exim finds that it cannot understand an
190existing cache file, it generates new parameters and writes them to the cache
191in the new format. If, however, you downgrade from 4.51 to a previous release,
192you MUST delete the gnutls-params file in the spool directory, because the
193older Exim will not recognize the new format.
194
c688b954
PH
1952. When doing a callout as part of verifying an address, Exim was not paying
196attention to any local part prefix or suffix that was matched by the router
197that accepted the address. It now behaves in the same way as it does for
198delivery: the affixes are removed from the local part unless
199rcpt_include_affixes is set on the transport. If you have a configuration that
200uses prefixes or suffixes on addresses that could be used for callouts, and you
201want the affixes to be retained, you must make sure that rcpt_include_affixes
202is set on the transport.
203
447d236c
PH
2043. Bounce and delay warning messages no longer contain details of delivery
205errors, except for explicit messages (e.g. generated by :fail:) and SMTP
206responses from remote hosts.
207
b5aea5e1 208
8b5af54d
PH
209Version 4.50
210------------
211
4964e932
PH
212The exicyclog script has been updated to use three-digit numbers in rotated log
213files if the maximum number to keep is greater than 99. If you are already
214keeping more than 99, there will be an incompatible change when you upgrade.
215You will probably want to rename your old log files to the new form before
c3af992c 216running the new exicyclog.
8b5af54d
PH
217
218
0f4f2a88
PH
219Version 4.42
220------------
221
222RFC 3848 specifies standard names for the "with" phrase in Received: header
223lines when AUTH and/or TLS are in use. This is the "received protocol"
224field. Exim used to use "asmtp" for authenticated SMTP, without any
225indication (in the protocol name) for TLS use. Now it follows the RFC and
226uses "esmtpa" if the connection is authenticated, "esmtps" if it is
227encrypted, and "esmtpsa" if it is both encrypted and authenticated. These names
228appear in log lines as well as in Received: header lines.
229
230
231Version 4.34
232------------
233
234Change 4.31/2 gave problems to data ACLs and local_scan() functions that
235expected to see a Received: header. I have changed to yet another scheme. The
236Received: header is now generated after the body is received, but before the
237ACL or local_scan() is called. After they have run, the timestamp in the
238Received: header is updated.
239
240Thus, change (a) of 4.31/2 has been reversed, but change (b) is still true,
241which is lucky, since I decided it was a bug fix.
242
243
244Version 4.33
245------------
246
247If an expansion in a condition on a "warn" statement fails because a lookup
248defers, the "warn" statement is abandoned, and the next ACL statement is
249processed. Previously this caused the whole ACL to be aborted.
250
251
252Version 4.32
253------------
254
255Change 4.31/2 has been reversed, as it proved contentious. Recipient callout
256verification now uses <> in the MAIL command by default, as it did before. A
257new callout option, "use_sender", has been added to request the other
258behaviour.
259
260
261Version 4.31
262------------
263
2641. If you compile Exim to use GnuTLS, it now requires the use of release 1.0.0
265 or greater. The interface to the obsolete 0.8.x releases is no longer
266 supported. There is one externally visible change: the format for the
267 display of Distinguished Names now uses commas as a separator rather than a
268 slash. This is to comply with RFC 2253.
269
2702. When a message is received, the Received: header line is now generated when
271 reception is complete, instead of at the start of reception. For messages
272 that take a long time to come in, this changes the meaning of the timestamp.
273 There are several side-effects of this change:
274
275 (a) If a message is rejected by a DATA or non-SMTP ACL, or by local_scan(),
276 the logged header lines no longer include the local Received: line,
277 because it has not yet been created. If the message is a non-SMTP one,
278 and the error is processed by sending a message to the sender, the copy
279 of the original message that is returned does not have an added
280 Received: line.
281
282 (b) When a filter file is tested using -bf, no additional Received: header
283 is added to the test message. After some thought, I decided that this
284 is a bug fix.
285
286 The contents of $received_for are not affected by this change. This
287 variable still contains the single recipient of a message, copied after
288 addresses have been rewritten, but before local_scan() is run.
289
2902. Recipient callout verification, like sender verification, was using <> in
291 the MAIL FROM command. This isn't really the right thing, since the actual
292 sender may affect whether the remote host accepts the recipient or not. I
293 have changed it to use the actual sender in the callout; this means that
294 the cache record is now keyed on a recipient/sender pair, not just the
295 recipient address. There doesn't seem to be a real danger of callout loops,
296 since a callout by the remote host to check the sender would use <>.
297
298
299Version 4.30
300------------
301
3021. I have abolished timeout_DNS as an error that can be detected in retry
303 rules, because it has never worked. Despite the fact that it has been
304 documented since at least release 1.62, there was no code to support it.
305 If you have used it in your retry rules, you will now get a warning message
306 to the log and panic log. It is now treated as plain "timeout".
307
3082. After discussion on the mailing list, Exim no longer adds From:, Date:, or
309 Message-Id: header lines to messages that do not originate locally, that is,
310 messages that have an associated sending host address.
311
3123. When looking up a host name from an IP address, Exim now tries the DNS
313 first, and only if that fails does it use gethostbyaddr() (or equivalent).
314 This change was made because on some OS, not all the names are given for
315 addresses with multiple PTR records via the gethostbyaddr() interface. The
316 order of lookup can be changed by setting host_lookup_order.
317
318
319Version 4.23
320------------
321
3221. The new FIXED_NEVER_USERS build-time option creates a list of "never users"
323 that cannot be overridden. The default in the distributed EDITME is "root".
324 If for some reason you were (against advice) running deliveries as root, you
325 will have to ensure that FIXED_NEVER_USERS is not set in your
326 Local/Makefile.
327
3282. The ${quote: operator now quotes an empty string, which it did not before.
329
3303. Version 4.23 saves the contents of the ACL variables with the message, so
331 that they can be used later. If one of these variables contains a newline,
332 there will be a newline character in the spool that will not be interpreted
333 correctely by a previous version of Exim. (Exim ignores keyed spool file
334 items that it doesn't understand - precisely for this kind of problem - but
335 it expects them all to be on one line.)
336
337 So the bottom line is: if you have newlines in your ACL variables, you
338 cannot retreat from 4.23.
339
340
341Version 4.21
342------------
343
3441. The idea of the "warn" ACL verb is that it adds a header or writes to the
345 log only when "message" or "log_message" are set. However, if one of the
346 conditions was an address verification, or a call to a nested ACL, the
347 messages generated by the underlying test were being passed through. This
348 no longer happens. The underlying message is available in $acl_verify_
349 message for both "message" and "log_message" expansions, so it can be
350 passed through if needed.
351
3522. The way that the $h_ (and $header_) expansions work has been changed by the
353 addition of RFC 2047 decoding. See the main documentation (the NewStuff file
354 until release 4.30, then the manual) for full details. Briefly, there are
355 now three forms:
356
357 $rh_xxx: and $rheader_xxx: give the original content of the header
358 line(s), with no processing at all.
359
360 $bh_xxx: and $bheader_xxx: remove leading and trailing white space, and
361 then decode base64 or quoted-printable "words" within the header text,
362 but do not do charset translation.
363
364 $h_xxx: and $header_xxx: attempt to translate the $bh_ string to a
365 standard character set.
366
367 If you have previously been using $h_ expansions to access the raw
368 characters, you should change to $rh_ instead.
369
3703. When Exim creates an RFC 2047 encoded word in a header line, it labels it
371 with the default character set from the headers_charset option instead of
372 always using iso-8859-1.
373
3744. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
375 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
376 called TMPDIR, and if it finds it is different, it changes its value.
377
3785. Following a discussion on the list, the rules by which Exim recognises line
379 endings on incoming messages have been changed. The -dropcr and drop_cr
380 options are now no-ops, retained only for backwards compatibility. The
381 following line terminators are recognized: LF CRLF CR. However, special
382 processing applies to CR:
383
384 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
385 nor a local message in the state where . is a terminator.
386
387 (ii) If a bare CR is encountered in a header line, an extra space is added
388 after the line terminator so as not to end the header. The reasoning
389 behind this is that bare CRs in header lines are most likely either
390 to be mistakes, or people trying to play silly games.
391
3926. The code for using daemon_smtp_port, local_interfaces, and the -oX options
393 has been reorganized. It is supposed to be backwards compatible, but it is
394 mentioned here just in case I've screwed up.
395
396
397
398Version 4.20
399------------
400
4011. I have tidied and re-organized the code that uses alarm() for imposing time
402 limits on various things. It shouldn't affect anything, but if you notice
403 processes getting stuck, it may be that I've broken something.
404
4052. The "arguments" log selector now also logs the current working directory
406 when Exim is called.
407
4083. An incompatible change has been made to the appendfile transport. This
409 affects the case when it is used for file deliveries that are set up by
410 .forward and filter files. Previously, any settings of the "file" or
411 "directory" options were ignored. It is hoped that, like the address_file
412 transport in the default configuration, these options were never in fact set
413 on such transports, because they were of no use.
414
415 Now, if either of these options is set, it is used. The path that is passed
416 by the router is in $address_file (this is not new), so it can be used as
417 part of a longer path, or modified in any other way that expansion permits.
418
419 If neither "file" nor "directory" is set, the behaviour is unchanged.
420
4214. Related to the above: in a filter, if a "save" command specifies a non-
422 absolute path, the value of $home/ is pre-pended. This no longer happens if
423 $home is unset or is set to an empty string.
424
4255. Multiple file deliveries from a filter or .forward file can never be
426 batched; the value of batch_max on the transport is ignored for file
427 deliveries. I'm assuming that nobody ever actually set batch_max on the
428 address_file transport - it would have had odd effects previously.
429
4306. DESTDIR is the more common variable that ROOT for use when installing
431 software under a different root filing system. The Exim install script now
432 recognizes DESTDIR first; if it is not set, ROOT is used.
433
4347. If DESTDIR is set when installing Exim, it no longer prepends its value to
435 the path of the system aliases file that appears in the default
436 configuration (when a default configuration is installed). If an aliases
437 file is actually created, its name *does* use the prefix.
438
439
440Version 4.14
441------------
442
4431. The default for the maximum number of unknown SMTP commands that Exim will
444accept before dropping a connection has been reduced from 5 to 3. However, you
445can now change the value by setting smtp_max_unknown_commands.
446
4472. The ${quote: operator has been changed so that it turns newline and carriage
448return characters into \n and \r, respectively.
449
4503. The file names used for maildir messages now include the microsecond time
451fraction as well as the time in seconds, to cope with systems where the process
452id can be re-used within the same second. The format is now
453
454 <time>.H<microsec>P<pid>.<host>
455
456This should be a compatible change, but is noted here just in case.
457
4584. The rules for creating message ids have changed, to cope with systems where
459the process id can be re-used within the same second. The format, however, is
460unchanged, so this should not cause any problems, except as noted in the next
461item.
462
4635. The maximum value for localhost_number has been reduced from 255 to 16, in
464order to implement the new message id rules. For operating systems that have
465case-insensitive file systems (Cygwin and Darwin), the limit is 10.
466
4676. verify = header_syntax was allowing unqualified addresses in all cases. Now
468it allows them only for locally generated messages and from hosts that match
469sender_unqualified_hosts or recipient_unqualified_hosts, respectively.
470
4717. For reasons lost in the mists of time, when a pipe transport was run, the
472environment variable MESSAGE_ID was set to the message ID preceded by 'E' (the
473form used in Message-ID: header lines). The 'E' has been removed.
474
475
476Version 4.11
477------------
478
4791. The handling of lines in the configuration file has changed. Previously,
480macro expansion was applied to logical lines, after continuations had been
481joined on. This meant that it could not be used in .include lines, which are
482handled as physical rather than logical lines. Macro expansion is now done on
483physical lines rather than logical lines. This means there are two
484incompatibilities:
485
486 (a) A macro that expands to # to turn a line into a comment now applies only
487 to the physical line where it appears. Previously, it would have caused
488 any following continuations also to be ignored.
489
490 (b) A macro name can no longer be split over the boundary between a line and
491 its continuation. Actually, this is more of a bug fix. :-)
492
4932. The -D command line option must now all be within one command line item.
494This makes it possible to use -D to set a macro to the empty string by commands
495such as
496
497 exim -DABC ...
498 exim -DABC= ...
499
500Previously, these items would have moved on to the next item on the command
501line. To include spaces in a macro definition item, quotes must be used, in
502which case you can also have spaces after -D and surrounding the equals. For
503example:
504
505 exim '-D ABC = something' ...
506
5073. The way that addresses that redirect to themselves are handled has been
508changed, in order to fix an obscure bug. This should not cause any problems
509except in the case of wanting to go back from a 4.11 (or later) release to an
510earlier release. If there are undelivered messages on the spool that contain
511addresses which redirect to themselves, and the redirected addresses have
512already been delivered, you might get a duplicate delivery if you revert to an
513earlier Exim.
514
5154. The default way of looking up IP addresses for hosts in the manualroute and
516queryprogram routers has been changed. If "byname" or "bydns" is explicitly
517specified, there is no change, but if no method is specified, Exim now behaves
518as follows:
519
520 First, a DNS lookup is done. If this yields anything other than
521 HOST_NOT_FOUND, that result is used. Otherwise, Exim goes on to try a call to
522 getipnodebyname() (or gethostbyname() on older systems) and the result of the
523 lookup is the result of that call.
524
525This change has been made because it has been discovered that on some systems,
526if a DNS lookup called via getipnodebyname() times out, HOST_NOT_FOUND is
527returned instead of TRY_AGAIN. Thus, it is safest to try a DNS lookup directly
528first, and only if that gives a definite "no such host" to try the local
529function.
530
5315. In fixing the minor security problem with pid_file_path, I have removed some
532backwards-compatible (undocumented) code which was present to ease conversion
533from Exim 3. In Exim 4, pid_file_path is a literal; in Exim 3 it was allowed to
534contain "%s", which was replaced by the port number for daemons listening on
535non-standard ports. In Exim 4, such daemons do not write a pid file. The
536backwards compatibility feature was to replace "%s" by nothing if it occurred
537in an Exim 4 setting of pid_file_path. The bug was in this code. I have solved
538the problem by removing the backwards compatibility feature. Thus, if you still
539have "%s" somewhere in a setting of pid_file_path, you should remove it.
540
5416. There has been an extension to lsearch files. The keys in these files may
542now be quoted in order to allow for whitespace and colons in them. This means
543that if you were previously using keys that began with a doublequote, you will
544now have to wrap them with extra quotes and escape the internal quotes. The
545possibility that anybody is actually doing this seems extremely remote, but it
546is documented just in case.
547
548
549Version 4.10
550------------
551
552The build-time parameter EXIWHAT_KILL_ARG has been renamed EXIWHAT_KILL_SIGNAL
553to better reflect its function. The OS-specific files have been updated. Only
554if you have explicitly set this in your Makefile (highly unlikely) do you need
555to change anything.
556
557****