Install Tony's patch for $message_linecount.
[exim.git] / src / README.UPDATING
CommitLineData
c688b954 1$Cambridge: exim/src/README.UPDATING,v 1.6 2005/04/06 16:26:42 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
b5aea5e1
PH
31Version 4.51
32------------
33
c688b954
PH
341. The format in which GnuTLS parameters are cached (in the file gnutls-params
35in the spool directory) has been changed. The new format can also be generated
b5aea5e1
PH
36externally, so it is now possible to update the values from outside Exim. This
37has been implemented in an upwards, BUT NOT downwards, compatible manner.
38Upgrading should be seamless: when Exim finds that it cannot understand an
39existing cache file, it generates new parameters and writes them to the cache
40in the new format. If, however, you downgrade from 4.51 to a previous release,
41you MUST delete the gnutls-params file in the spool directory, because the
42older Exim will not recognize the new format.
43
c688b954
PH
442. When doing a callout as part of verifying an address, Exim was not paying
45attention to any local part prefix or suffix that was matched by the router
46that accepted the address. It now behaves in the same way as it does for
47delivery: the affixes are removed from the local part unless
48rcpt_include_affixes is set on the transport. If you have a configuration that
49uses prefixes or suffixes on addresses that could be used for callouts, and you
50want the affixes to be retained, you must make sure that rcpt_include_affixes
51is set on the transport.
52
b5aea5e1 53
8b5af54d
PH
54Version 4.50
55------------
56
4964e932
PH
57The exicyclog script has been updated to use three-digit numbers in rotated log
58files if the maximum number to keep is greater than 99. If you are already
59keeping more than 99, there will be an incompatible change when you upgrade.
60You will probably want to rename your old log files to the new form before
c3af992c 61running the new exicyclog.
8b5af54d
PH
62
63
0f4f2a88
PH
64Version 4.42
65------------
66
67RFC 3848 specifies standard names for the "with" phrase in Received: header
68lines when AUTH and/or TLS are in use. This is the "received protocol"
69field. Exim used to use "asmtp" for authenticated SMTP, without any
70indication (in the protocol name) for TLS use. Now it follows the RFC and
71uses "esmtpa" if the connection is authenticated, "esmtps" if it is
72encrypted, and "esmtpsa" if it is both encrypted and authenticated. These names
73appear in log lines as well as in Received: header lines.
74
75
76Version 4.34
77------------
78
79Change 4.31/2 gave problems to data ACLs and local_scan() functions that
80expected to see a Received: header. I have changed to yet another scheme. The
81Received: header is now generated after the body is received, but before the
82ACL or local_scan() is called. After they have run, the timestamp in the
83Received: header is updated.
84
85Thus, change (a) of 4.31/2 has been reversed, but change (b) is still true,
86which is lucky, since I decided it was a bug fix.
87
88
89Version 4.33
90------------
91
92If an expansion in a condition on a "warn" statement fails because a lookup
93defers, the "warn" statement is abandoned, and the next ACL statement is
94processed. Previously this caused the whole ACL to be aborted.
95
96
97Version 4.32
98------------
99
100Change 4.31/2 has been reversed, as it proved contentious. Recipient callout
101verification now uses <> in the MAIL command by default, as it did before. A
102new callout option, "use_sender", has been added to request the other
103behaviour.
104
105
106Version 4.31
107------------
108
1091. If you compile Exim to use GnuTLS, it now requires the use of release 1.0.0
110 or greater. The interface to the obsolete 0.8.x releases is no longer
111 supported. There is one externally visible change: the format for the
112 display of Distinguished Names now uses commas as a separator rather than a
113 slash. This is to comply with RFC 2253.
114
1152. When a message is received, the Received: header line is now generated when
116 reception is complete, instead of at the start of reception. For messages
117 that take a long time to come in, this changes the meaning of the timestamp.
118 There are several side-effects of this change:
119
120 (a) If a message is rejected by a DATA or non-SMTP ACL, or by local_scan(),
121 the logged header lines no longer include the local Received: line,
122 because it has not yet been created. If the message is a non-SMTP one,
123 and the error is processed by sending a message to the sender, the copy
124 of the original message that is returned does not have an added
125 Received: line.
126
127 (b) When a filter file is tested using -bf, no additional Received: header
128 is added to the test message. After some thought, I decided that this
129 is a bug fix.
130
131 The contents of $received_for are not affected by this change. This
132 variable still contains the single recipient of a message, copied after
133 addresses have been rewritten, but before local_scan() is run.
134
1352. Recipient callout verification, like sender verification, was using <> in
136 the MAIL FROM command. This isn't really the right thing, since the actual
137 sender may affect whether the remote host accepts the recipient or not. I
138 have changed it to use the actual sender in the callout; this means that
139 the cache record is now keyed on a recipient/sender pair, not just the
140 recipient address. There doesn't seem to be a real danger of callout loops,
141 since a callout by the remote host to check the sender would use <>.
142
143
144Version 4.30
145------------
146
1471. I have abolished timeout_DNS as an error that can be detected in retry
148 rules, because it has never worked. Despite the fact that it has been
149 documented since at least release 1.62, there was no code to support it.
150 If you have used it in your retry rules, you will now get a warning message
151 to the log and panic log. It is now treated as plain "timeout".
152
1532. After discussion on the mailing list, Exim no longer adds From:, Date:, or
154 Message-Id: header lines to messages that do not originate locally, that is,
155 messages that have an associated sending host address.
156
1573. When looking up a host name from an IP address, Exim now tries the DNS
158 first, and only if that fails does it use gethostbyaddr() (or equivalent).
159 This change was made because on some OS, not all the names are given for
160 addresses with multiple PTR records via the gethostbyaddr() interface. The
161 order of lookup can be changed by setting host_lookup_order.
162
163
164Version 4.23
165------------
166
1671. The new FIXED_NEVER_USERS build-time option creates a list of "never users"
168 that cannot be overridden. The default in the distributed EDITME is "root".
169 If for some reason you were (against advice) running deliveries as root, you
170 will have to ensure that FIXED_NEVER_USERS is not set in your
171 Local/Makefile.
172
1732. The ${quote: operator now quotes an empty string, which it did not before.
174
1753. Version 4.23 saves the contents of the ACL variables with the message, so
176 that they can be used later. If one of these variables contains a newline,
177 there will be a newline character in the spool that will not be interpreted
178 correctely by a previous version of Exim. (Exim ignores keyed spool file
179 items that it doesn't understand - precisely for this kind of problem - but
180 it expects them all to be on one line.)
181
182 So the bottom line is: if you have newlines in your ACL variables, you
183 cannot retreat from 4.23.
184
185
186Version 4.21
187------------
188
1891. The idea of the "warn" ACL verb is that it adds a header or writes to the
190 log only when "message" or "log_message" are set. However, if one of the
191 conditions was an address verification, or a call to a nested ACL, the
192 messages generated by the underlying test were being passed through. This
193 no longer happens. The underlying message is available in $acl_verify_
194 message for both "message" and "log_message" expansions, so it can be
195 passed through if needed.
196
1972. The way that the $h_ (and $header_) expansions work has been changed by the
198 addition of RFC 2047 decoding. See the main documentation (the NewStuff file
199 until release 4.30, then the manual) for full details. Briefly, there are
200 now three forms:
201
202 $rh_xxx: and $rheader_xxx: give the original content of the header
203 line(s), with no processing at all.
204
205 $bh_xxx: and $bheader_xxx: remove leading and trailing white space, and
206 then decode base64 or quoted-printable "words" within the header text,
207 but do not do charset translation.
208
209 $h_xxx: and $header_xxx: attempt to translate the $bh_ string to a
210 standard character set.
211
212 If you have previously been using $h_ expansions to access the raw
213 characters, you should change to $rh_ instead.
214
2153. When Exim creates an RFC 2047 encoded word in a header line, it labels it
216 with the default character set from the headers_charset option instead of
217 always using iso-8859-1.
218
2194. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
220 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
221 called TMPDIR, and if it finds it is different, it changes its value.
222
2235. Following a discussion on the list, the rules by which Exim recognises line
224 endings on incoming messages have been changed. The -dropcr and drop_cr
225 options are now no-ops, retained only for backwards compatibility. The
226 following line terminators are recognized: LF CRLF CR. However, special
227 processing applies to CR:
228
229 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
230 nor a local message in the state where . is a terminator.
231
232 (ii) If a bare CR is encountered in a header line, an extra space is added
233 after the line terminator so as not to end the header. The reasoning
234 behind this is that bare CRs in header lines are most likely either
235 to be mistakes, or people trying to play silly games.
236
2376. The code for using daemon_smtp_port, local_interfaces, and the -oX options
238 has been reorganized. It is supposed to be backwards compatible, but it is
239 mentioned here just in case I've screwed up.
240
241
242
243Version 4.20
244------------
245
2461. I have tidied and re-organized the code that uses alarm() for imposing time
247 limits on various things. It shouldn't affect anything, but if you notice
248 processes getting stuck, it may be that I've broken something.
249
2502. The "arguments" log selector now also logs the current working directory
251 when Exim is called.
252
2533. An incompatible change has been made to the appendfile transport. This
254 affects the case when it is used for file deliveries that are set up by
255 .forward and filter files. Previously, any settings of the "file" or
256 "directory" options were ignored. It is hoped that, like the address_file
257 transport in the default configuration, these options were never in fact set
258 on such transports, because they were of no use.
259
260 Now, if either of these options is set, it is used. The path that is passed
261 by the router is in $address_file (this is not new), so it can be used as
262 part of a longer path, or modified in any other way that expansion permits.
263
264 If neither "file" nor "directory" is set, the behaviour is unchanged.
265
2664. Related to the above: in a filter, if a "save" command specifies a non-
267 absolute path, the value of $home/ is pre-pended. This no longer happens if
268 $home is unset or is set to an empty string.
269
2705. Multiple file deliveries from a filter or .forward file can never be
271 batched; the value of batch_max on the transport is ignored for file
272 deliveries. I'm assuming that nobody ever actually set batch_max on the
273 address_file transport - it would have had odd effects previously.
274
2756. DESTDIR is the more common variable that ROOT for use when installing
276 software under a different root filing system. The Exim install script now
277 recognizes DESTDIR first; if it is not set, ROOT is used.
278
2797. If DESTDIR is set when installing Exim, it no longer prepends its value to
280 the path of the system aliases file that appears in the default
281 configuration (when a default configuration is installed). If an aliases
282 file is actually created, its name *does* use the prefix.
283
284
285Version 4.14
286------------
287
2881. The default for the maximum number of unknown SMTP commands that Exim will
289accept before dropping a connection has been reduced from 5 to 3. However, you
290can now change the value by setting smtp_max_unknown_commands.
291
2922. The ${quote: operator has been changed so that it turns newline and carriage
293return characters into \n and \r, respectively.
294
2953. The file names used for maildir messages now include the microsecond time
296fraction as well as the time in seconds, to cope with systems where the process
297id can be re-used within the same second. The format is now
298
299 <time>.H<microsec>P<pid>.<host>
300
301This should be a compatible change, but is noted here just in case.
302
3034. The rules for creating message ids have changed, to cope with systems where
304the process id can be re-used within the same second. The format, however, is
305unchanged, so this should not cause any problems, except as noted in the next
306item.
307
3085. The maximum value for localhost_number has been reduced from 255 to 16, in
309order to implement the new message id rules. For operating systems that have
310case-insensitive file systems (Cygwin and Darwin), the limit is 10.
311
3126. verify = header_syntax was allowing unqualified addresses in all cases. Now
313it allows them only for locally generated messages and from hosts that match
314sender_unqualified_hosts or recipient_unqualified_hosts, respectively.
315
3167. For reasons lost in the mists of time, when a pipe transport was run, the
317environment variable MESSAGE_ID was set to the message ID preceded by 'E' (the
318form used in Message-ID: header lines). The 'E' has been removed.
319
320
321Version 4.11
322------------
323
3241. The handling of lines in the configuration file has changed. Previously,
325macro expansion was applied to logical lines, after continuations had been
326joined on. This meant that it could not be used in .include lines, which are
327handled as physical rather than logical lines. Macro expansion is now done on
328physical lines rather than logical lines. This means there are two
329incompatibilities:
330
331 (a) A macro that expands to # to turn a line into a comment now applies only
332 to the physical line where it appears. Previously, it would have caused
333 any following continuations also to be ignored.
334
335 (b) A macro name can no longer be split over the boundary between a line and
336 its continuation. Actually, this is more of a bug fix. :-)
337
3382. The -D command line option must now all be within one command line item.
339This makes it possible to use -D to set a macro to the empty string by commands
340such as
341
342 exim -DABC ...
343 exim -DABC= ...
344
345Previously, these items would have moved on to the next item on the command
346line. To include spaces in a macro definition item, quotes must be used, in
347which case you can also have spaces after -D and surrounding the equals. For
348example:
349
350 exim '-D ABC = something' ...
351
3523. The way that addresses that redirect to themselves are handled has been
353changed, in order to fix an obscure bug. This should not cause any problems
354except in the case of wanting to go back from a 4.11 (or later) release to an
355earlier release. If there are undelivered messages on the spool that contain
356addresses which redirect to themselves, and the redirected addresses have
357already been delivered, you might get a duplicate delivery if you revert to an
358earlier Exim.
359
3604. The default way of looking up IP addresses for hosts in the manualroute and
361queryprogram routers has been changed. If "byname" or "bydns" is explicitly
362specified, there is no change, but if no method is specified, Exim now behaves
363as follows:
364
365 First, a DNS lookup is done. If this yields anything other than
366 HOST_NOT_FOUND, that result is used. Otherwise, Exim goes on to try a call to
367 getipnodebyname() (or gethostbyname() on older systems) and the result of the
368 lookup is the result of that call.
369
370This change has been made because it has been discovered that on some systems,
371if a DNS lookup called via getipnodebyname() times out, HOST_NOT_FOUND is
372returned instead of TRY_AGAIN. Thus, it is safest to try a DNS lookup directly
373first, and only if that gives a definite "no such host" to try the local
374function.
375
3765. In fixing the minor security problem with pid_file_path, I have removed some
377backwards-compatible (undocumented) code which was present to ease conversion
378from Exim 3. In Exim 4, pid_file_path is a literal; in Exim 3 it was allowed to
379contain "%s", which was replaced by the port number for daemons listening on
380non-standard ports. In Exim 4, such daemons do not write a pid file. The
381backwards compatibility feature was to replace "%s" by nothing if it occurred
382in an Exim 4 setting of pid_file_path. The bug was in this code. I have solved
383the problem by removing the backwards compatibility feature. Thus, if you still
384have "%s" somewhere in a setting of pid_file_path, you should remove it.
385
3866. There has been an extension to lsearch files. The keys in these files may
387now be quoted in order to allow for whitespace and colons in them. This means
388that if you were previously using keys that began with a doublequote, you will
389now have to wrap them with extra quotes and escape the internal quotes. The
390possibility that anybody is actually doing this seems extremely remote, but it
391is documented just in case.
392
393
394Version 4.10
395------------
396
397The build-time parameter EXIWHAT_KILL_ARG has been renamed EXIWHAT_KILL_SIGNAL
398to better reflect its function. The OS-specific files have been updated. Only
399if you have explicitly set this in your Makefile (highly unlikely) do you need
400to change anything.
401
402****