ipliteral was not recognizing "ipv6" prefix.
[exim.git] / src / README.UPDATING
CommitLineData
8857ccfd 1$Cambridge: exim/src/README.UPDATING,v 1.9 2005/10/03 13:25:32 ph10 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.
29
30
8857ccfd
PH
31Exim version 4.54
32-----------------
33
34There was a problem with 4.52/TF/02 in that a "name=" option on control=
35submission terminated at the next slash, thereby not allowing for slashes in
36the name. This has been changed so that "name=" takes the rest of the string as
37its data. It must therefore be the last option.
38
39
f3f0a1f0
PH
40Version 4.53
41------------
42
43If you are using the experimental Domain Keys support, you must upgrade to
44at least libdomainkeys 0.67 in order to run this release of Exim.
45
46
b5aea5e1
PH
47Version 4.51
48------------
49
c688b954
PH
501. The format in which GnuTLS parameters are cached (in the file gnutls-params
51in the spool directory) has been changed. The new format can also be generated
b5aea5e1
PH
52externally, so it is now possible to update the values from outside Exim. This
53has been implemented in an upwards, BUT NOT downwards, compatible manner.
54Upgrading should be seamless: when Exim finds that it cannot understand an
55existing cache file, it generates new parameters and writes them to the cache
56in the new format. If, however, you downgrade from 4.51 to a previous release,
57you MUST delete the gnutls-params file in the spool directory, because the
58older Exim will not recognize the new format.
59
c688b954
PH
602. When doing a callout as part of verifying an address, Exim was not paying
61attention to any local part prefix or suffix that was matched by the router
62that accepted the address. It now behaves in the same way as it does for
63delivery: the affixes are removed from the local part unless
64rcpt_include_affixes is set on the transport. If you have a configuration that
65uses prefixes or suffixes on addresses that could be used for callouts, and you
66want the affixes to be retained, you must make sure that rcpt_include_affixes
67is set on the transport.
68
447d236c
PH
693. Bounce and delay warning messages no longer contain details of delivery
70errors, except for explicit messages (e.g. generated by :fail:) and SMTP
71responses from remote hosts.
72
b5aea5e1 73
8b5af54d
PH
74Version 4.50
75------------
76
4964e932
PH
77The exicyclog script has been updated to use three-digit numbers in rotated log
78files if the maximum number to keep is greater than 99. If you are already
79keeping more than 99, there will be an incompatible change when you upgrade.
80You will probably want to rename your old log files to the new form before
c3af992c 81running the new exicyclog.
8b5af54d
PH
82
83
0f4f2a88
PH
84Version 4.42
85------------
86
87RFC 3848 specifies standard names for the "with" phrase in Received: header
88lines when AUTH and/or TLS are in use. This is the "received protocol"
89field. Exim used to use "asmtp" for authenticated SMTP, without any
90indication (in the protocol name) for TLS use. Now it follows the RFC and
91uses "esmtpa" if the connection is authenticated, "esmtps" if it is
92encrypted, and "esmtpsa" if it is both encrypted and authenticated. These names
93appear in log lines as well as in Received: header lines.
94
95
96Version 4.34
97------------
98
99Change 4.31/2 gave problems to data ACLs and local_scan() functions that
100expected to see a Received: header. I have changed to yet another scheme. The
101Received: header is now generated after the body is received, but before the
102ACL or local_scan() is called. After they have run, the timestamp in the
103Received: header is updated.
104
105Thus, change (a) of 4.31/2 has been reversed, but change (b) is still true,
106which is lucky, since I decided it was a bug fix.
107
108
109Version 4.33
110------------
111
112If an expansion in a condition on a "warn" statement fails because a lookup
113defers, the "warn" statement is abandoned, and the next ACL statement is
114processed. Previously this caused the whole ACL to be aborted.
115
116
117Version 4.32
118------------
119
120Change 4.31/2 has been reversed, as it proved contentious. Recipient callout
121verification now uses <> in the MAIL command by default, as it did before. A
122new callout option, "use_sender", has been added to request the other
123behaviour.
124
125
126Version 4.31
127------------
128
1291. If you compile Exim to use GnuTLS, it now requires the use of release 1.0.0
130 or greater. The interface to the obsolete 0.8.x releases is no longer
131 supported. There is one externally visible change: the format for the
132 display of Distinguished Names now uses commas as a separator rather than a
133 slash. This is to comply with RFC 2253.
134
1352. When a message is received, the Received: header line is now generated when
136 reception is complete, instead of at the start of reception. For messages
137 that take a long time to come in, this changes the meaning of the timestamp.
138 There are several side-effects of this change:
139
140 (a) If a message is rejected by a DATA or non-SMTP ACL, or by local_scan(),
141 the logged header lines no longer include the local Received: line,
142 because it has not yet been created. If the message is a non-SMTP one,
143 and the error is processed by sending a message to the sender, the copy
144 of the original message that is returned does not have an added
145 Received: line.
146
147 (b) When a filter file is tested using -bf, no additional Received: header
148 is added to the test message. After some thought, I decided that this
149 is a bug fix.
150
151 The contents of $received_for are not affected by this change. This
152 variable still contains the single recipient of a message, copied after
153 addresses have been rewritten, but before local_scan() is run.
154
1552. Recipient callout verification, like sender verification, was using <> in
156 the MAIL FROM command. This isn't really the right thing, since the actual
157 sender may affect whether the remote host accepts the recipient or not. I
158 have changed it to use the actual sender in the callout; this means that
159 the cache record is now keyed on a recipient/sender pair, not just the
160 recipient address. There doesn't seem to be a real danger of callout loops,
161 since a callout by the remote host to check the sender would use <>.
162
163
164Version 4.30
165------------
166
1671. I have abolished timeout_DNS as an error that can be detected in retry
168 rules, because it has never worked. Despite the fact that it has been
169 documented since at least release 1.62, there was no code to support it.
170 If you have used it in your retry rules, you will now get a warning message
171 to the log and panic log. It is now treated as plain "timeout".
172
1732. After discussion on the mailing list, Exim no longer adds From:, Date:, or
174 Message-Id: header lines to messages that do not originate locally, that is,
175 messages that have an associated sending host address.
176
1773. When looking up a host name from an IP address, Exim now tries the DNS
178 first, and only if that fails does it use gethostbyaddr() (or equivalent).
179 This change was made because on some OS, not all the names are given for
180 addresses with multiple PTR records via the gethostbyaddr() interface. The
181 order of lookup can be changed by setting host_lookup_order.
182
183
184Version 4.23
185------------
186
1871. The new FIXED_NEVER_USERS build-time option creates a list of "never users"
188 that cannot be overridden. The default in the distributed EDITME is "root".
189 If for some reason you were (against advice) running deliveries as root, you
190 will have to ensure that FIXED_NEVER_USERS is not set in your
191 Local/Makefile.
192
1932. The ${quote: operator now quotes an empty string, which it did not before.
194
1953. Version 4.23 saves the contents of the ACL variables with the message, so
196 that they can be used later. If one of these variables contains a newline,
197 there will be a newline character in the spool that will not be interpreted
198 correctely by a previous version of Exim. (Exim ignores keyed spool file
199 items that it doesn't understand - precisely for this kind of problem - but
200 it expects them all to be on one line.)
201
202 So the bottom line is: if you have newlines in your ACL variables, you
203 cannot retreat from 4.23.
204
205
206Version 4.21
207------------
208
2091. The idea of the "warn" ACL verb is that it adds a header or writes to the
210 log only when "message" or "log_message" are set. However, if one of the
211 conditions was an address verification, or a call to a nested ACL, the
212 messages generated by the underlying test were being passed through. This
213 no longer happens. The underlying message is available in $acl_verify_
214 message for both "message" and "log_message" expansions, so it can be
215 passed through if needed.
216
2172. The way that the $h_ (and $header_) expansions work has been changed by the
218 addition of RFC 2047 decoding. See the main documentation (the NewStuff file
219 until release 4.30, then the manual) for full details. Briefly, there are
220 now three forms:
221
222 $rh_xxx: and $rheader_xxx: give the original content of the header
223 line(s), with no processing at all.
224
225 $bh_xxx: and $bheader_xxx: remove leading and trailing white space, and
226 then decode base64 or quoted-printable "words" within the header text,
227 but do not do charset translation.
228
229 $h_xxx: and $header_xxx: attempt to translate the $bh_ string to a
230 standard character set.
231
232 If you have previously been using $h_ expansions to access the raw
233 characters, you should change to $rh_ instead.
234
2353. When Exim creates an RFC 2047 encoded word in a header line, it labels it
236 with the default character set from the headers_charset option instead of
237 always using iso-8859-1.
238
2394. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
240 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
241 called TMPDIR, and if it finds it is different, it changes its value.
242
2435. Following a discussion on the list, the rules by which Exim recognises line
244 endings on incoming messages have been changed. The -dropcr and drop_cr
245 options are now no-ops, retained only for backwards compatibility. The
246 following line terminators are recognized: LF CRLF CR. However, special
247 processing applies to CR:
248
249 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
250 nor a local message in the state where . is a terminator.
251
252 (ii) If a bare CR is encountered in a header line, an extra space is added
253 after the line terminator so as not to end the header. The reasoning
254 behind this is that bare CRs in header lines are most likely either
255 to be mistakes, or people trying to play silly games.
256
2576. The code for using daemon_smtp_port, local_interfaces, and the -oX options
258 has been reorganized. It is supposed to be backwards compatible, but it is
259 mentioned here just in case I've screwed up.
260
261
262
263Version 4.20
264------------
265
2661. I have tidied and re-organized the code that uses alarm() for imposing time
267 limits on various things. It shouldn't affect anything, but if you notice
268 processes getting stuck, it may be that I've broken something.
269
2702. The "arguments" log selector now also logs the current working directory
271 when Exim is called.
272
2733. An incompatible change has been made to the appendfile transport. This
274 affects the case when it is used for file deliveries that are set up by
275 .forward and filter files. Previously, any settings of the "file" or
276 "directory" options were ignored. It is hoped that, like the address_file
277 transport in the default configuration, these options were never in fact set
278 on such transports, because they were of no use.
279
280 Now, if either of these options is set, it is used. The path that is passed
281 by the router is in $address_file (this is not new), so it can be used as
282 part of a longer path, or modified in any other way that expansion permits.
283
284 If neither "file" nor "directory" is set, the behaviour is unchanged.
285
2864. Related to the above: in a filter, if a "save" command specifies a non-
287 absolute path, the value of $home/ is pre-pended. This no longer happens if
288 $home is unset or is set to an empty string.
289
2905. Multiple file deliveries from a filter or .forward file can never be
291 batched; the value of batch_max on the transport is ignored for file
292 deliveries. I'm assuming that nobody ever actually set batch_max on the
293 address_file transport - it would have had odd effects previously.
294
2956. DESTDIR is the more common variable that ROOT for use when installing
296 software under a different root filing system. The Exim install script now
297 recognizes DESTDIR first; if it is not set, ROOT is used.
298
2997. If DESTDIR is set when installing Exim, it no longer prepends its value to
300 the path of the system aliases file that appears in the default
301 configuration (when a default configuration is installed). If an aliases
302 file is actually created, its name *does* use the prefix.
303
304
305Version 4.14
306------------
307
3081. The default for the maximum number of unknown SMTP commands that Exim will
309accept before dropping a connection has been reduced from 5 to 3. However, you
310can now change the value by setting smtp_max_unknown_commands.
311
3122. The ${quote: operator has been changed so that it turns newline and carriage
313return characters into \n and \r, respectively.
314
3153. The file names used for maildir messages now include the microsecond time
316fraction as well as the time in seconds, to cope with systems where the process
317id can be re-used within the same second. The format is now
318
319 <time>.H<microsec>P<pid>.<host>
320
321This should be a compatible change, but is noted here just in case.
322
3234. The rules for creating message ids have changed, to cope with systems where
324the process id can be re-used within the same second. The format, however, is
325unchanged, so this should not cause any problems, except as noted in the next
326item.
327
3285. The maximum value for localhost_number has been reduced from 255 to 16, in
329order to implement the new message id rules. For operating systems that have
330case-insensitive file systems (Cygwin and Darwin), the limit is 10.
331
3326. verify = header_syntax was allowing unqualified addresses in all cases. Now
333it allows them only for locally generated messages and from hosts that match
334sender_unqualified_hosts or recipient_unqualified_hosts, respectively.
335
3367. For reasons lost in the mists of time, when a pipe transport was run, the
337environment variable MESSAGE_ID was set to the message ID preceded by 'E' (the
338form used in Message-ID: header lines). The 'E' has been removed.
339
340
341Version 4.11
342------------
343
3441. The handling of lines in the configuration file has changed. Previously,
345macro expansion was applied to logical lines, after continuations had been
346joined on. This meant that it could not be used in .include lines, which are
347handled as physical rather than logical lines. Macro expansion is now done on
348physical lines rather than logical lines. This means there are two
349incompatibilities:
350
351 (a) A macro that expands to # to turn a line into a comment now applies only
352 to the physical line where it appears. Previously, it would have caused
353 any following continuations also to be ignored.
354
355 (b) A macro name can no longer be split over the boundary between a line and
356 its continuation. Actually, this is more of a bug fix. :-)
357
3582. The -D command line option must now all be within one command line item.
359This makes it possible to use -D to set a macro to the empty string by commands
360such as
361
362 exim -DABC ...
363 exim -DABC= ...
364
365Previously, these items would have moved on to the next item on the command
366line. To include spaces in a macro definition item, quotes must be used, in
367which case you can also have spaces after -D and surrounding the equals. For
368example:
369
370 exim '-D ABC = something' ...
371
3723. The way that addresses that redirect to themselves are handled has been
373changed, in order to fix an obscure bug. This should not cause any problems
374except in the case of wanting to go back from a 4.11 (or later) release to an
375earlier release. If there are undelivered messages on the spool that contain
376addresses which redirect to themselves, and the redirected addresses have
377already been delivered, you might get a duplicate delivery if you revert to an
378earlier Exim.
379
3804. The default way of looking up IP addresses for hosts in the manualroute and
381queryprogram routers has been changed. If "byname" or "bydns" is explicitly
382specified, there is no change, but if no method is specified, Exim now behaves
383as follows:
384
385 First, a DNS lookup is done. If this yields anything other than
386 HOST_NOT_FOUND, that result is used. Otherwise, Exim goes on to try a call to
387 getipnodebyname() (or gethostbyname() on older systems) and the result of the
388 lookup is the result of that call.
389
390This change has been made because it has been discovered that on some systems,
391if a DNS lookup called via getipnodebyname() times out, HOST_NOT_FOUND is
392returned instead of TRY_AGAIN. Thus, it is safest to try a DNS lookup directly
393first, and only if that gives a definite "no such host" to try the local
394function.
395
3965. In fixing the minor security problem with pid_file_path, I have removed some
397backwards-compatible (undocumented) code which was present to ease conversion
398from Exim 3. In Exim 4, pid_file_path is a literal; in Exim 3 it was allowed to
399contain "%s", which was replaced by the port number for daemons listening on
400non-standard ports. In Exim 4, such daemons do not write a pid file. The
401backwards compatibility feature was to replace "%s" by nothing if it occurred
402in an Exim 4 setting of pid_file_path. The bug was in this code. I have solved
403the problem by removing the backwards compatibility feature. Thus, if you still
404have "%s" somewhere in a setting of pid_file_path, you should remove it.
405
4066. There has been an extension to lsearch files. The keys in these files may
407now be quoted in order to allow for whitespace and colons in them. This means
408that if you were previously using keys that began with a doublequote, you will
409now have to wrap them with extra quotes and escape the internal quotes. The
410possibility that anybody is actually doing this seems extremely remote, but it
411is documented just in case.
412
413
414Version 4.10
415------------
416
417The build-time parameter EXIWHAT_KILL_ARG has been renamed EXIWHAT_KILL_SIGNAL
418to better reflect its function. The OS-specific files have been updated. Only
419if you have explicitly set this in your Makefile (highly unlikely) do you need
420to change anything.
421
422****