Fix too-often retry bug after 4xx with more hosts than hosts_max_retry.
[exim.git] / src / README.UPDATING
CommitLineData
b6c27e28 1$Cambridge: exim/src/README.UPDATING,v 1.15 2006/10/23 09:14:50 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
26da7e20
PH
31Exim version 4.64
32-----------------
33
341. Callouts were setting the name used for EHLO/HELO from $smtp_active_
35hostname. This is wrong, because it relates to the incoming message (and
36probably the interface on which it is arriving) and not to the outgoing
37callout (which could be using a different interface). This has been
38changed to use the value of the helo_data option from the smtp transport
39instead - this is what is used when a message is actually being sent. If
40there is no remote transport (possible with a router that sets up host
41addresses), $smtp_active_hostname is used. This change is mentioned here in
42case somebody is relying on the use of $smtp_active_hostname.
43
ebc9d865
PH
442. A bug has been fixed that might just possibly be something that is relied on
45in some configurations. In expansion items such as ${if >{xxx}{yyy}...} an
46empty string (that is {}) was being interpreted as if it was {0} and therefore
47treated as the number zero. From release 4.64, such strings cause an error
48because a decimal number, possibly followed by K or M, is required (as has
49always been documented).
50
b6c27e28
PH
513. There has been a change to the GnuTLS support (ChangeLog/PH/20) to improve
52Exim's performance. Unfortunately, this has the side effect of being slightly
53non-upwards compatible for versions 4.50 and earlier. If you are upgrading from
54one of these earlier versions and you use GnuTLS, you must remove the file
55called gnutls-params in Exim's spool directory. If you don't do this, you will
56see this error:
57
58 TLS error on connection from ... (DH params import): Base64 decoding error.
59
60Removing the file causes Exim to recompute the relevant encryption parameters
61and cache them in the new format that was introduced for release 4.51 (May
622005). If you are upgrading from release 4.51 or later, there should be no
63problem.
64
26da7e20 65
a5bd321b
PH
66Exim version 4.63
67-----------------
68
69When an SMTP error message is specified in a "message" modifier in an ACL, or
70in a :fail: or :defer: message in a redirect router, Exim now checks the start
71of the message for an SMTP error code. This consists of three digits followed
72by a space, optionally followed by an extended code of the form n.n.n, also
73followed by a space. If this is the case and the very first digit is the same
74as the default error code, the code from the message is used instead. If the
75very first digit is incorrect, a panic error is logged, and the default code is
76used. This is an incompatible change, but it is not expected to affect many (if
77any) configurations. It is possible to suppress the use of the supplied code in
78a redirect router by setting the smtp_error_code option false. In this case,
79any SMTP code is quietly ignored.
80
81
47ca6d6c
PH
82Exim version 4.61
83-----------------
84
8def5aaf
PH
851. The default number of ACL variables of each type has been increased to 20,
86and it's possible to compile Exim with more. You can safely upgrade to this
87release if you already have messages on the queue with saved ACL variable
88values. However, if you downgrade from this release with messages on the queue,
89any saved ACL values they may have will be lost.
90
912. The default value for rfc1413_query_timeout has been changed from 30s to 5s.
47ca6d6c
PH
92
93
8857ccfd
PH
94Exim version 4.54
95-----------------
96
97There was a problem with 4.52/TF/02 in that a "name=" option on control=
98submission terminated at the next slash, thereby not allowing for slashes in
99the name. This has been changed so that "name=" takes the rest of the string as
100its data. It must therefore be the last option.
101
102
f3f0a1f0
PH
103Version 4.53
104------------
105
106If you are using the experimental Domain Keys support, you must upgrade to
107at least libdomainkeys 0.67 in order to run this release of Exim.
108
109
b5aea5e1
PH
110Version 4.51
111------------
112
c688b954
PH
1131. The format in which GnuTLS parameters are cached (in the file gnutls-params
114in the spool directory) has been changed. The new format can also be generated
b5aea5e1
PH
115externally, so it is now possible to update the values from outside Exim. This
116has been implemented in an upwards, BUT NOT downwards, compatible manner.
117Upgrading should be seamless: when Exim finds that it cannot understand an
118existing cache file, it generates new parameters and writes them to the cache
119in the new format. If, however, you downgrade from 4.51 to a previous release,
120you MUST delete the gnutls-params file in the spool directory, because the
121older Exim will not recognize the new format.
122
c688b954
PH
1232. When doing a callout as part of verifying an address, Exim was not paying
124attention to any local part prefix or suffix that was matched by the router
125that accepted the address. It now behaves in the same way as it does for
126delivery: the affixes are removed from the local part unless
127rcpt_include_affixes is set on the transport. If you have a configuration that
128uses prefixes or suffixes on addresses that could be used for callouts, and you
129want the affixes to be retained, you must make sure that rcpt_include_affixes
130is set on the transport.
131
447d236c
PH
1323. Bounce and delay warning messages no longer contain details of delivery
133errors, except for explicit messages (e.g. generated by :fail:) and SMTP
134responses from remote hosts.
135
b5aea5e1 136
8b5af54d
PH
137Version 4.50
138------------
139
4964e932
PH
140The exicyclog script has been updated to use three-digit numbers in rotated log
141files if the maximum number to keep is greater than 99. If you are already
142keeping more than 99, there will be an incompatible change when you upgrade.
143You will probably want to rename your old log files to the new form before
c3af992c 144running the new exicyclog.
8b5af54d
PH
145
146
0f4f2a88
PH
147Version 4.42
148------------
149
150RFC 3848 specifies standard names for the "with" phrase in Received: header
151lines when AUTH and/or TLS are in use. This is the "received protocol"
152field. Exim used to use "asmtp" for authenticated SMTP, without any
153indication (in the protocol name) for TLS use. Now it follows the RFC and
154uses "esmtpa" if the connection is authenticated, "esmtps" if it is
155encrypted, and "esmtpsa" if it is both encrypted and authenticated. These names
156appear in log lines as well as in Received: header lines.
157
158
159Version 4.34
160------------
161
162Change 4.31/2 gave problems to data ACLs and local_scan() functions that
163expected to see a Received: header. I have changed to yet another scheme. The
164Received: header is now generated after the body is received, but before the
165ACL or local_scan() is called. After they have run, the timestamp in the
166Received: header is updated.
167
168Thus, change (a) of 4.31/2 has been reversed, but change (b) is still true,
169which is lucky, since I decided it was a bug fix.
170
171
172Version 4.33
173------------
174
175If an expansion in a condition on a "warn" statement fails because a lookup
176defers, the "warn" statement is abandoned, and the next ACL statement is
177processed. Previously this caused the whole ACL to be aborted.
178
179
180Version 4.32
181------------
182
183Change 4.31/2 has been reversed, as it proved contentious. Recipient callout
184verification now uses <> in the MAIL command by default, as it did before. A
185new callout option, "use_sender", has been added to request the other
186behaviour.
187
188
189Version 4.31
190------------
191
1921. If you compile Exim to use GnuTLS, it now requires the use of release 1.0.0
193 or greater. The interface to the obsolete 0.8.x releases is no longer
194 supported. There is one externally visible change: the format for the
195 display of Distinguished Names now uses commas as a separator rather than a
196 slash. This is to comply with RFC 2253.
197
1982. When a message is received, the Received: header line is now generated when
199 reception is complete, instead of at the start of reception. For messages
200 that take a long time to come in, this changes the meaning of the timestamp.
201 There are several side-effects of this change:
202
203 (a) If a message is rejected by a DATA or non-SMTP ACL, or by local_scan(),
204 the logged header lines no longer include the local Received: line,
205 because it has not yet been created. If the message is a non-SMTP one,
206 and the error is processed by sending a message to the sender, the copy
207 of the original message that is returned does not have an added
208 Received: line.
209
210 (b) When a filter file is tested using -bf, no additional Received: header
211 is added to the test message. After some thought, I decided that this
212 is a bug fix.
213
214 The contents of $received_for are not affected by this change. This
215 variable still contains the single recipient of a message, copied after
216 addresses have been rewritten, but before local_scan() is run.
217
2182. Recipient callout verification, like sender verification, was using <> in
219 the MAIL FROM command. This isn't really the right thing, since the actual
220 sender may affect whether the remote host accepts the recipient or not. I
221 have changed it to use the actual sender in the callout; this means that
222 the cache record is now keyed on a recipient/sender pair, not just the
223 recipient address. There doesn't seem to be a real danger of callout loops,
224 since a callout by the remote host to check the sender would use <>.
225
226
227Version 4.30
228------------
229
2301. I have abolished timeout_DNS as an error that can be detected in retry
231 rules, because it has never worked. Despite the fact that it has been
232 documented since at least release 1.62, there was no code to support it.
233 If you have used it in your retry rules, you will now get a warning message
234 to the log and panic log. It is now treated as plain "timeout".
235
2362. After discussion on the mailing list, Exim no longer adds From:, Date:, or
237 Message-Id: header lines to messages that do not originate locally, that is,
238 messages that have an associated sending host address.
239
2403. When looking up a host name from an IP address, Exim now tries the DNS
241 first, and only if that fails does it use gethostbyaddr() (or equivalent).
242 This change was made because on some OS, not all the names are given for
243 addresses with multiple PTR records via the gethostbyaddr() interface. The
244 order of lookup can be changed by setting host_lookup_order.
245
246
247Version 4.23
248------------
249
2501. The new FIXED_NEVER_USERS build-time option creates a list of "never users"
251 that cannot be overridden. The default in the distributed EDITME is "root".
252 If for some reason you were (against advice) running deliveries as root, you
253 will have to ensure that FIXED_NEVER_USERS is not set in your
254 Local/Makefile.
255
2562. The ${quote: operator now quotes an empty string, which it did not before.
257
2583. Version 4.23 saves the contents of the ACL variables with the message, so
259 that they can be used later. If one of these variables contains a newline,
260 there will be a newline character in the spool that will not be interpreted
261 correctely by a previous version of Exim. (Exim ignores keyed spool file
262 items that it doesn't understand - precisely for this kind of problem - but
263 it expects them all to be on one line.)
264
265 So the bottom line is: if you have newlines in your ACL variables, you
266 cannot retreat from 4.23.
267
268
269Version 4.21
270------------
271
2721. The idea of the "warn" ACL verb is that it adds a header or writes to the
273 log only when "message" or "log_message" are set. However, if one of the
274 conditions was an address verification, or a call to a nested ACL, the
275 messages generated by the underlying test were being passed through. This
276 no longer happens. The underlying message is available in $acl_verify_
277 message for both "message" and "log_message" expansions, so it can be
278 passed through if needed.
279
2802. The way that the $h_ (and $header_) expansions work has been changed by the
281 addition of RFC 2047 decoding. See the main documentation (the NewStuff file
282 until release 4.30, then the manual) for full details. Briefly, there are
283 now three forms:
284
285 $rh_xxx: and $rheader_xxx: give the original content of the header
286 line(s), with no processing at all.
287
288 $bh_xxx: and $bheader_xxx: remove leading and trailing white space, and
289 then decode base64 or quoted-printable "words" within the header text,
290 but do not do charset translation.
291
292 $h_xxx: and $header_xxx: attempt to translate the $bh_ string to a
293 standard character set.
294
295 If you have previously been using $h_ expansions to access the raw
296 characters, you should change to $rh_ instead.
297
2983. When Exim creates an RFC 2047 encoded word in a header line, it labels it
299 with the default character set from the headers_charset option instead of
300 always using iso-8859-1.
301
3024. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
303 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
304 called TMPDIR, and if it finds it is different, it changes its value.
305
3065. Following a discussion on the list, the rules by which Exim recognises line
307 endings on incoming messages have been changed. The -dropcr and drop_cr
308 options are now no-ops, retained only for backwards compatibility. The
309 following line terminators are recognized: LF CRLF CR. However, special
310 processing applies to CR:
311
312 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
313 nor a local message in the state where . is a terminator.
314
315 (ii) If a bare CR is encountered in a header line, an extra space is added
316 after the line terminator so as not to end the header. The reasoning
317 behind this is that bare CRs in header lines are most likely either
318 to be mistakes, or people trying to play silly games.
319
3206. The code for using daemon_smtp_port, local_interfaces, and the -oX options
321 has been reorganized. It is supposed to be backwards compatible, but it is
322 mentioned here just in case I've screwed up.
323
324
325
326Version 4.20
327------------
328
3291. I have tidied and re-organized the code that uses alarm() for imposing time
330 limits on various things. It shouldn't affect anything, but if you notice
331 processes getting stuck, it may be that I've broken something.
332
3332. The "arguments" log selector now also logs the current working directory
334 when Exim is called.
335
3363. An incompatible change has been made to the appendfile transport. This
337 affects the case when it is used for file deliveries that are set up by
338 .forward and filter files. Previously, any settings of the "file" or
339 "directory" options were ignored. It is hoped that, like the address_file
340 transport in the default configuration, these options were never in fact set
341 on such transports, because they were of no use.
342
343 Now, if either of these options is set, it is used. The path that is passed
344 by the router is in $address_file (this is not new), so it can be used as
345 part of a longer path, or modified in any other way that expansion permits.
346
347 If neither "file" nor "directory" is set, the behaviour is unchanged.
348
3494. Related to the above: in a filter, if a "save" command specifies a non-
350 absolute path, the value of $home/ is pre-pended. This no longer happens if
351 $home is unset or is set to an empty string.
352
3535. Multiple file deliveries from a filter or .forward file can never be
354 batched; the value of batch_max on the transport is ignored for file
355 deliveries. I'm assuming that nobody ever actually set batch_max on the
356 address_file transport - it would have had odd effects previously.
357
3586. DESTDIR is the more common variable that ROOT for use when installing
359 software under a different root filing system. The Exim install script now
360 recognizes DESTDIR first; if it is not set, ROOT is used.
361
3627. If DESTDIR is set when installing Exim, it no longer prepends its value to
363 the path of the system aliases file that appears in the default
364 configuration (when a default configuration is installed). If an aliases
365 file is actually created, its name *does* use the prefix.
366
367
368Version 4.14
369------------
370
3711. The default for the maximum number of unknown SMTP commands that Exim will
372accept before dropping a connection has been reduced from 5 to 3. However, you
373can now change the value by setting smtp_max_unknown_commands.
374
3752. The ${quote: operator has been changed so that it turns newline and carriage
376return characters into \n and \r, respectively.
377
3783. The file names used for maildir messages now include the microsecond time
379fraction as well as the time in seconds, to cope with systems where the process
380id can be re-used within the same second. The format is now
381
382 <time>.H<microsec>P<pid>.<host>
383
384This should be a compatible change, but is noted here just in case.
385
3864. The rules for creating message ids have changed, to cope with systems where
387the process id can be re-used within the same second. The format, however, is
388unchanged, so this should not cause any problems, except as noted in the next
389item.
390
3915. The maximum value for localhost_number has been reduced from 255 to 16, in
392order to implement the new message id rules. For operating systems that have
393case-insensitive file systems (Cygwin and Darwin), the limit is 10.
394
3956. verify = header_syntax was allowing unqualified addresses in all cases. Now
396it allows them only for locally generated messages and from hosts that match
397sender_unqualified_hosts or recipient_unqualified_hosts, respectively.
398
3997. For reasons lost in the mists of time, when a pipe transport was run, the
400environment variable MESSAGE_ID was set to the message ID preceded by 'E' (the
401form used in Message-ID: header lines). The 'E' has been removed.
402
403
404Version 4.11
405------------
406
4071. The handling of lines in the configuration file has changed. Previously,
408macro expansion was applied to logical lines, after continuations had been
409joined on. This meant that it could not be used in .include lines, which are
410handled as physical rather than logical lines. Macro expansion is now done on
411physical lines rather than logical lines. This means there are two
412incompatibilities:
413
414 (a) A macro that expands to # to turn a line into a comment now applies only
415 to the physical line where it appears. Previously, it would have caused
416 any following continuations also to be ignored.
417
418 (b) A macro name can no longer be split over the boundary between a line and
419 its continuation. Actually, this is more of a bug fix. :-)
420
4212. The -D command line option must now all be within one command line item.
422This makes it possible to use -D to set a macro to the empty string by commands
423such as
424
425 exim -DABC ...
426 exim -DABC= ...
427
428Previously, these items would have moved on to the next item on the command
429line. To include spaces in a macro definition item, quotes must be used, in
430which case you can also have spaces after -D and surrounding the equals. For
431example:
432
433 exim '-D ABC = something' ...
434
4353. The way that addresses that redirect to themselves are handled has been
436changed, in order to fix an obscure bug. This should not cause any problems
437except in the case of wanting to go back from a 4.11 (or later) release to an
438earlier release. If there are undelivered messages on the spool that contain
439addresses which redirect to themselves, and the redirected addresses have
440already been delivered, you might get a duplicate delivery if you revert to an
441earlier Exim.
442
4434. The default way of looking up IP addresses for hosts in the manualroute and
444queryprogram routers has been changed. If "byname" or "bydns" is explicitly
445specified, there is no change, but if no method is specified, Exim now behaves
446as follows:
447
448 First, a DNS lookup is done. If this yields anything other than
449 HOST_NOT_FOUND, that result is used. Otherwise, Exim goes on to try a call to
450 getipnodebyname() (or gethostbyname() on older systems) and the result of the
451 lookup is the result of that call.
452
453This change has been made because it has been discovered that on some systems,
454if a DNS lookup called via getipnodebyname() times out, HOST_NOT_FOUND is
455returned instead of TRY_AGAIN. Thus, it is safest to try a DNS lookup directly
456first, and only if that gives a definite "no such host" to try the local
457function.
458
4595. In fixing the minor security problem with pid_file_path, I have removed some
460backwards-compatible (undocumented) code which was present to ease conversion
461from Exim 3. In Exim 4, pid_file_path is a literal; in Exim 3 it was allowed to
462contain "%s", which was replaced by the port number for daemons listening on
463non-standard ports. In Exim 4, such daemons do not write a pid file. The
464backwards compatibility feature was to replace "%s" by nothing if it occurred
465in an Exim 4 setting of pid_file_path. The bug was in this code. I have solved
466the problem by removing the backwards compatibility feature. Thus, if you still
467have "%s" somewhere in a setting of pid_file_path, you should remove it.
468
4696. There has been an extension to lsearch files. The keys in these files may
470now be quoted in order to allow for whitespace and colons in them. This means
471that if you were previously using keys that began with a doublequote, you will
472now have to wrap them with extra quotes and escape the internal quotes. The
473possibility that anybody is actually doing this seems extremely remote, but it
474is documented just in case.
475
476
477Version 4.10
478------------
479
480The build-time parameter EXIWHAT_KILL_ARG has been renamed EXIWHAT_KILL_SIGNAL
481to better reflect its function. The OS-specific files have been updated. Only
482if you have explicitly set this in your Makefile (highly unlikely) do you need
483to change anything.
484
485****