1 $Cambridge: exim/src/README.UPDATING,v 1.16 2007/06/20 14:19:23 ph10 Exp $
3 This document contains detailed information about incompatibilities that might
4 be encountered when upgrading from one release of Exim to another. The
5 information is in reverse order of release numbers. Mostly these are relatively
6 small points, and the configuration file is normally upwards compatible, but
7 there have been two big upheavals...
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. *
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 **************************************************************************
27 The rest of this document contains information about changes in 4.xx releases
28 that might affect a running system.
34 1. The internal implementation of the database keys that are used for ACL
35 ratelimiting has been tidied up. This means that an update to 4.68 might cause
36 Exim to "forget" previous rates that it had calculated, and reset them to zero.
42 1. Callouts were setting the name used for EHLO/HELO from $smtp_active_
43 hostname. This is wrong, because it relates to the incoming message (and
44 probably the interface on which it is arriving) and not to the outgoing
45 callout (which could be using a different interface). This has been
46 changed to use the value of the helo_data option from the smtp transport
47 instead - this is what is used when a message is actually being sent. If
48 there is no remote transport (possible with a router that sets up host
49 addresses), $smtp_active_hostname is used. This change is mentioned here in
50 case somebody is relying on the use of $smtp_active_hostname.
52 2. A bug has been fixed that might just possibly be something that is relied on
53 in some configurations. In expansion items such as ${if >{xxx}{yyy}...} an
54 empty string (that is {}) was being interpreted as if it was {0} and therefore
55 treated as the number zero. From release 4.64, such strings cause an error
56 because a decimal number, possibly followed by K or M, is required (as has
57 always been documented).
59 3. There has been a change to the GnuTLS support (ChangeLog/PH/20) to improve
60 Exim's performance. Unfortunately, this has the side effect of being slightly
61 non-upwards compatible for versions 4.50 and earlier. If you are upgrading from
62 one of these earlier versions and you use GnuTLS, you must remove the file
63 called gnutls-params in Exim's spool directory. If you don't do this, you will
66 TLS error on connection from ... (DH params import): Base64 decoding error.
68 Removing the file causes Exim to recompute the relevant encryption parameters
69 and cache them in the new format that was introduced for release 4.51 (May
70 2005). If you are upgrading from release 4.51 or later, there should be no
77 When an SMTP error message is specified in a "message" modifier in an ACL, or
78 in a :fail: or :defer: message in a redirect router, Exim now checks the start
79 of the message for an SMTP error code. This consists of three digits followed
80 by a space, optionally followed by an extended code of the form n.n.n, also
81 followed by a space. If this is the case and the very first digit is the same
82 as the default error code, the code from the message is used instead. If the
83 very first digit is incorrect, a panic error is logged, and the default code is
84 used. This is an incompatible change, but it is not expected to affect many (if
85 any) configurations. It is possible to suppress the use of the supplied code in
86 a redirect router by setting the smtp_error_code option false. In this case,
87 any SMTP code is quietly ignored.
93 1. The default number of ACL variables of each type has been increased to 20,
94 and it's possible to compile Exim with more. You can safely upgrade to this
95 release if you already have messages on the queue with saved ACL variable
96 values. However, if you downgrade from this release with messages on the queue,
97 any saved ACL values they may have will be lost.
99 2. The default value for rfc1413_query_timeout has been changed from 30s to 5s.
105 There was a problem with 4.52/TF/02 in that a "name=" option on control=
106 submission terminated at the next slash, thereby not allowing for slashes in
107 the name. This has been changed so that "name=" takes the rest of the string as
108 its data. It must therefore be the last option.
114 If you are using the experimental Domain Keys support, you must upgrade to
115 at least libdomainkeys 0.67 in order to run this release of Exim.
121 1. The format in which GnuTLS parameters are cached (in the file gnutls-params
122 in the spool directory) has been changed. The new format can also be generated
123 externally, so it is now possible to update the values from outside Exim. This
124 has been implemented in an upwards, BUT NOT downwards, compatible manner.
125 Upgrading should be seamless: when Exim finds that it cannot understand an
126 existing cache file, it generates new parameters and writes them to the cache
127 in the new format. If, however, you downgrade from 4.51 to a previous release,
128 you MUST delete the gnutls-params file in the spool directory, because the
129 older Exim will not recognize the new format.
131 2. When doing a callout as part of verifying an address, Exim was not paying
132 attention to any local part prefix or suffix that was matched by the router
133 that accepted the address. It now behaves in the same way as it does for
134 delivery: the affixes are removed from the local part unless
135 rcpt_include_affixes is set on the transport. If you have a configuration that
136 uses prefixes or suffixes on addresses that could be used for callouts, and you
137 want the affixes to be retained, you must make sure that rcpt_include_affixes
138 is set on the transport.
140 3. Bounce and delay warning messages no longer contain details of delivery
141 errors, except for explicit messages (e.g. generated by :fail:) and SMTP
142 responses from remote hosts.
148 The exicyclog script has been updated to use three-digit numbers in rotated log
149 files if the maximum number to keep is greater than 99. If you are already
150 keeping more than 99, there will be an incompatible change when you upgrade.
151 You will probably want to rename your old log files to the new form before
152 running the new exicyclog.
158 RFC 3848 specifies standard names for the "with" phrase in Received: header
159 lines when AUTH and/or TLS are in use. This is the "received protocol"
160 field. Exim used to use "asmtp" for authenticated SMTP, without any
161 indication (in the protocol name) for TLS use. Now it follows the RFC and
162 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
163 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These names
164 appear in log lines as well as in Received: header lines.
170 Change 4.31/2 gave problems to data ACLs and local_scan() functions that
171 expected to see a Received: header. I have changed to yet another scheme. The
172 Received: header is now generated after the body is received, but before the
173 ACL or local_scan() is called. After they have run, the timestamp in the
174 Received: header is updated.
176 Thus, change (a) of 4.31/2 has been reversed, but change (b) is still true,
177 which is lucky, since I decided it was a bug fix.
183 If an expansion in a condition on a "warn" statement fails because a lookup
184 defers, the "warn" statement is abandoned, and the next ACL statement is
185 processed. Previously this caused the whole ACL to be aborted.
191 Change 4.31/2 has been reversed, as it proved contentious. Recipient callout
192 verification now uses <> in the MAIL command by default, as it did before. A
193 new callout option, "use_sender", has been added to request the other
200 1. If you compile Exim to use GnuTLS, it now requires the use of release 1.0.0
201 or greater. The interface to the obsolete 0.8.x releases is no longer
202 supported. There is one externally visible change: the format for the
203 display of Distinguished Names now uses commas as a separator rather than a
204 slash. This is to comply with RFC 2253.
206 2. When a message is received, the Received: header line is now generated when
207 reception is complete, instead of at the start of reception. For messages
208 that take a long time to come in, this changes the meaning of the timestamp.
209 There are several side-effects of this change:
211 (a) If a message is rejected by a DATA or non-SMTP ACL, or by local_scan(),
212 the logged header lines no longer include the local Received: line,
213 because it has not yet been created. If the message is a non-SMTP one,
214 and the error is processed by sending a message to the sender, the copy
215 of the original message that is returned does not have an added
218 (b) When a filter file is tested using -bf, no additional Received: header
219 is added to the test message. After some thought, I decided that this
222 The contents of $received_for are not affected by this change. This
223 variable still contains the single recipient of a message, copied after
224 addresses have been rewritten, but before local_scan() is run.
226 2. Recipient callout verification, like sender verification, was using <> in
227 the MAIL FROM command. This isn't really the right thing, since the actual
228 sender may affect whether the remote host accepts the recipient or not. I
229 have changed it to use the actual sender in the callout; this means that
230 the cache record is now keyed on a recipient/sender pair, not just the
231 recipient address. There doesn't seem to be a real danger of callout loops,
232 since a callout by the remote host to check the sender would use <>.
238 1. I have abolished timeout_DNS as an error that can be detected in retry
239 rules, because it has never worked. Despite the fact that it has been
240 documented since at least release 1.62, there was no code to support it.
241 If you have used it in your retry rules, you will now get a warning message
242 to the log and panic log. It is now treated as plain "timeout".
244 2. After discussion on the mailing list, Exim no longer adds From:, Date:, or
245 Message-Id: header lines to messages that do not originate locally, that is,
246 messages that have an associated sending host address.
248 3. When looking up a host name from an IP address, Exim now tries the DNS
249 first, and only if that fails does it use gethostbyaddr() (or equivalent).
250 This change was made because on some OS, not all the names are given for
251 addresses with multiple PTR records via the gethostbyaddr() interface. The
252 order of lookup can be changed by setting host_lookup_order.
258 1. The new FIXED_NEVER_USERS build-time option creates a list of "never users"
259 that cannot be overridden. The default in the distributed EDITME is "root".
260 If for some reason you were (against advice) running deliveries as root, you
261 will have to ensure that FIXED_NEVER_USERS is not set in your
264 2. The ${quote: operator now quotes an empty string, which it did not before.
266 3. Version 4.23 saves the contents of the ACL variables with the message, so
267 that they can be used later. If one of these variables contains a newline,
268 there will be a newline character in the spool that will not be interpreted
269 correctely by a previous version of Exim. (Exim ignores keyed spool file
270 items that it doesn't understand - precisely for this kind of problem - but
271 it expects them all to be on one line.)
273 So the bottom line is: if you have newlines in your ACL variables, you
274 cannot retreat from 4.23.
280 1. The idea of the "warn" ACL verb is that it adds a header or writes to the
281 log only when "message" or "log_message" are set. However, if one of the
282 conditions was an address verification, or a call to a nested ACL, the
283 messages generated by the underlying test were being passed through. This
284 no longer happens. The underlying message is available in $acl_verify_
285 message for both "message" and "log_message" expansions, so it can be
286 passed through if needed.
288 2. The way that the $h_ (and $header_) expansions work has been changed by the
289 addition of RFC 2047 decoding. See the main documentation (the NewStuff file
290 until release 4.30, then the manual) for full details. Briefly, there are
293 $rh_xxx: and $rheader_xxx: give the original content of the header
294 line(s), with no processing at all.
296 $bh_xxx: and $bheader_xxx: remove leading and trailing white space, and
297 then decode base64 or quoted-printable "words" within the header text,
298 but do not do charset translation.
300 $h_xxx: and $header_xxx: attempt to translate the $bh_ string to a
301 standard character set.
303 If you have previously been using $h_ expansions to access the raw
304 characters, you should change to $rh_ instead.
306 3. When Exim creates an RFC 2047 encoded word in a header line, it labels it
307 with the default character set from the headers_charset option instead of
308 always using iso-8859-1.
310 4. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
311 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
312 called TMPDIR, and if it finds it is different, it changes its value.
314 5. Following a discussion on the list, the rules by which Exim recognises line
315 endings on incoming messages have been changed. The -dropcr and drop_cr
316 options are now no-ops, retained only for backwards compatibility. The
317 following line terminators are recognized: LF CRLF CR. However, special
318 processing applies to CR:
320 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
321 nor a local message in the state where . is a terminator.
323 (ii) If a bare CR is encountered in a header line, an extra space is added
324 after the line terminator so as not to end the header. The reasoning
325 behind this is that bare CRs in header lines are most likely either
326 to be mistakes, or people trying to play silly games.
328 6. The code for using daemon_smtp_port, local_interfaces, and the -oX options
329 has been reorganized. It is supposed to be backwards compatible, but it is
330 mentioned here just in case I've screwed up.
337 1. I have tidied and re-organized the code that uses alarm() for imposing time
338 limits on various things. It shouldn't affect anything, but if you notice
339 processes getting stuck, it may be that I've broken something.
341 2. The "arguments" log selector now also logs the current working directory
344 3. An incompatible change has been made to the appendfile transport. This
345 affects the case when it is used for file deliveries that are set up by
346 .forward and filter files. Previously, any settings of the "file" or
347 "directory" options were ignored. It is hoped that, like the address_file
348 transport in the default configuration, these options were never in fact set
349 on such transports, because they were of no use.
351 Now, if either of these options is set, it is used. The path that is passed
352 by the router is in $address_file (this is not new), so it can be used as
353 part of a longer path, or modified in any other way that expansion permits.
355 If neither "file" nor "directory" is set, the behaviour is unchanged.
357 4. Related to the above: in a filter, if a "save" command specifies a non-
358 absolute path, the value of $home/ is pre-pended. This no longer happens if
359 $home is unset or is set to an empty string.
361 5. Multiple file deliveries from a filter or .forward file can never be
362 batched; the value of batch_max on the transport is ignored for file
363 deliveries. I'm assuming that nobody ever actually set batch_max on the
364 address_file transport - it would have had odd effects previously.
366 6. DESTDIR is the more common variable that ROOT for use when installing
367 software under a different root filing system. The Exim install script now
368 recognizes DESTDIR first; if it is not set, ROOT is used.
370 7. If DESTDIR is set when installing Exim, it no longer prepends its value to
371 the path of the system aliases file that appears in the default
372 configuration (when a default configuration is installed). If an aliases
373 file is actually created, its name *does* use the prefix.
379 1. The default for the maximum number of unknown SMTP commands that Exim will
380 accept before dropping a connection has been reduced from 5 to 3. However, you
381 can now change the value by setting smtp_max_unknown_commands.
383 2. The ${quote: operator has been changed so that it turns newline and carriage
384 return characters into \n and \r, respectively.
386 3. The file names used for maildir messages now include the microsecond time
387 fraction as well as the time in seconds, to cope with systems where the process
388 id can be re-used within the same second. The format is now
390 <time>.H<microsec>P<pid>.<host>
392 This should be a compatible change, but is noted here just in case.
394 4. The rules for creating message ids have changed, to cope with systems where
395 the process id can be re-used within the same second. The format, however, is
396 unchanged, so this should not cause any problems, except as noted in the next
399 5. The maximum value for localhost_number has been reduced from 255 to 16, in
400 order to implement the new message id rules. For operating systems that have
401 case-insensitive file systems (Cygwin and Darwin), the limit is 10.
403 6. verify = header_syntax was allowing unqualified addresses in all cases. Now
404 it allows them only for locally generated messages and from hosts that match
405 sender_unqualified_hosts or recipient_unqualified_hosts, respectively.
407 7. For reasons lost in the mists of time, when a pipe transport was run, the
408 environment variable MESSAGE_ID was set to the message ID preceded by 'E' (the
409 form used in Message-ID: header lines). The 'E' has been removed.
415 1. The handling of lines in the configuration file has changed. Previously,
416 macro expansion was applied to logical lines, after continuations had been
417 joined on. This meant that it could not be used in .include lines, which are
418 handled as physical rather than logical lines. Macro expansion is now done on
419 physical lines rather than logical lines. This means there are two
422 (a) A macro that expands to # to turn a line into a comment now applies only
423 to the physical line where it appears. Previously, it would have caused
424 any following continuations also to be ignored.
426 (b) A macro name can no longer be split over the boundary between a line and
427 its continuation. Actually, this is more of a bug fix. :-)
429 2. The -D command line option must now all be within one command line item.
430 This makes it possible to use -D to set a macro to the empty string by commands
436 Previously, these items would have moved on to the next item on the command
437 line. To include spaces in a macro definition item, quotes must be used, in
438 which case you can also have spaces after -D and surrounding the equals. For
441 exim '-D ABC = something' ...
443 3. The way that addresses that redirect to themselves are handled has been
444 changed, in order to fix an obscure bug. This should not cause any problems
445 except in the case of wanting to go back from a 4.11 (or later) release to an
446 earlier release. If there are undelivered messages on the spool that contain
447 addresses which redirect to themselves, and the redirected addresses have
448 already been delivered, you might get a duplicate delivery if you revert to an
451 4. The default way of looking up IP addresses for hosts in the manualroute and
452 queryprogram routers has been changed. If "byname" or "bydns" is explicitly
453 specified, there is no change, but if no method is specified, Exim now behaves
456 First, a DNS lookup is done. If this yields anything other than
457 HOST_NOT_FOUND, that result is used. Otherwise, Exim goes on to try a call to
458 getipnodebyname() (or gethostbyname() on older systems) and the result of the
459 lookup is the result of that call.
461 This change has been made because it has been discovered that on some systems,
462 if a DNS lookup called via getipnodebyname() times out, HOST_NOT_FOUND is
463 returned instead of TRY_AGAIN. Thus, it is safest to try a DNS lookup directly
464 first, and only if that gives a definite "no such host" to try the local
467 5. In fixing the minor security problem with pid_file_path, I have removed some
468 backwards-compatible (undocumented) code which was present to ease conversion
469 from Exim 3. In Exim 4, pid_file_path is a literal; in Exim 3 it was allowed to
470 contain "%s", which was replaced by the port number for daemons listening on
471 non-standard ports. In Exim 4, such daemons do not write a pid file. The
472 backwards compatibility feature was to replace "%s" by nothing if it occurred
473 in an Exim 4 setting of pid_file_path. The bug was in this code. I have solved
474 the problem by removing the backwards compatibility feature. Thus, if you still
475 have "%s" somewhere in a setting of pid_file_path, you should remove it.
477 6. There has been an extension to lsearch files. The keys in these files may
478 now be quoted in order to allow for whitespace and colons in them. This means
479 that if you were previously using keys that began with a doublequote, you will
480 now have to wrap them with extra quotes and escape the internal quotes. The
481 possibility that anybody is actually doing this seems extremely remote, but it
482 is documented just in case.
488 The build-time parameter EXIWHAT_KILL_ARG has been renamed EXIWHAT_KILL_SIGNAL
489 to better reflect its function. The OS-specific files have been updated. Only
490 if you have explicitly set this in your Makefile (highly unlikely) do you need
git://git.exim.org / exim.git / blob