Include note about exicyclog and keeping more than 99 back copies.
[exim.git] / src / README.UPDATING
8b5af54d 1$Cambridge: exim/src/README.UPDATING,v 1.2 2005/01/11 15:15:33 ph10 Exp $
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...
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. *
27The rest of this document contains information about changes in 4.xx releases
28that might affect a running system.
31Version 4.50
34The exicyclog script has been updated to use three-digit numbers in rotated log
35files if the maximum number to keep is greater than 99. If you are already
36keeping more than 99, there will be an incompatible change when you upgrade.
37You will probably want to rename your old log files to the new form before
38running the new exicyclog. The following script, by Mick Swisher, can do this
39for you:
41# Begin conversion
42# Rename all the old files by adding additional leading zeros.
44while [ $count -gt 0 ]; do
45if [ $count -lt 100 ]; then newt=0$count; oldt=$count; fi
46if [ $count -lt 10 ]; then newt=00$count; oldt=0$count; fi
47if [ -f $mainlog.$oldt ]; then
48 $mv $mainlog.$oldt $mainlog.$newt
49elif [ -f $mainlog.$oldt.$suffix ]; then
50 $mv $mainlog.$oldt.$suffix $mainlog.$newt.$suffix
52if [ -f $rejectlog.$oldt ]; then
53 $mv $rejectlog.$oldt $rejectlog.$newt
54elif [ -f $rejectlog.$oldt.$suffix ]; then
55 $mv $rejectlog.$oldt.$suffix $rejectlog.$newt.$suffix
57count=`expr $count - 1`
61Version 4.42
64RFC 3848 specifies standard names for the "with" phrase in Received: header
65lines when AUTH and/or TLS are in use. This is the "received protocol"
66field. Exim used to use "asmtp" for authenticated SMTP, without any
67indication (in the protocol name) for TLS use. Now it follows the RFC and
68uses "esmtpa" if the connection is authenticated, "esmtps" if it is
69encrypted, and "esmtpsa" if it is both encrypted and authenticated. These names
70appear in log lines as well as in Received: header lines.
73Version 4.34
76Change 4.31/2 gave problems to data ACLs and local_scan() functions that
77expected to see a Received: header. I have changed to yet another scheme. The
78Received: header is now generated after the body is received, but before the
79ACL or local_scan() is called. After they have run, the timestamp in the
80Received: header is updated.
82Thus, change (a) of 4.31/2 has been reversed, but change (b) is still true,
83which is lucky, since I decided it was a bug fix.
86Version 4.33
89If an expansion in a condition on a "warn" statement fails because a lookup
90defers, the "warn" statement is abandoned, and the next ACL statement is
91processed. Previously this caused the whole ACL to be aborted.
94Version 4.32
97Change 4.31/2 has been reversed, as it proved contentious. Recipient callout
98verification now uses <> in the MAIL command by default, as it did before. A
99new callout option, "use_sender", has been added to request the other
103Version 4.31
1061. If you compile Exim to use GnuTLS, it now requires the use of release 1.0.0
107 or greater. The interface to the obsolete 0.8.x releases is no longer
108 supported. There is one externally visible change: the format for the
109 display of Distinguished Names now uses commas as a separator rather than a
110 slash. This is to comply with RFC 2253.
1122. When a message is received, the Received: header line is now generated when
113 reception is complete, instead of at the start of reception. For messages
114 that take a long time to come in, this changes the meaning of the timestamp.
115 There are several side-effects of this change:
117 (a) If a message is rejected by a DATA or non-SMTP ACL, or by local_scan(),
118 the logged header lines no longer include the local Received: line,
119 because it has not yet been created. If the message is a non-SMTP one,
120 and the error is processed by sending a message to the sender, the copy
121 of the original message that is returned does not have an added
122 Received: line.
124 (b) When a filter file is tested using -bf, no additional Received: header
125 is added to the test message. After some thought, I decided that this
126 is a bug fix.
128 The contents of $received_for are not affected by this change. This
129 variable still contains the single recipient of a message, copied after
130 addresses have been rewritten, but before local_scan() is run.
1322. Recipient callout verification, like sender verification, was using <> in
133 the MAIL FROM command. This isn't really the right thing, since the actual
134 sender may affect whether the remote host accepts the recipient or not. I
135 have changed it to use the actual sender in the callout; this means that
136 the cache record is now keyed on a recipient/sender pair, not just the
137 recipient address. There doesn't seem to be a real danger of callout loops,
138 since a callout by the remote host to check the sender would use <>.
141Version 4.30
1441. I have abolished timeout_DNS as an error that can be detected in retry
145 rules, because it has never worked. Despite the fact that it has been
146 documented since at least release 1.62, there was no code to support it.
147 If you have used it in your retry rules, you will now get a warning message
148 to the log and panic log. It is now treated as plain "timeout".
1502. After discussion on the mailing list, Exim no longer adds From:, Date:, or
151 Message-Id: header lines to messages that do not originate locally, that is,
152 messages that have an associated sending host address.
1543. When looking up a host name from an IP address, Exim now tries the DNS
155 first, and only if that fails does it use gethostbyaddr() (or equivalent).
156 This change was made because on some OS, not all the names are given for
157 addresses with multiple PTR records via the gethostbyaddr() interface. The
158 order of lookup can be changed by setting host_lookup_order.
161Version 4.23
1641. The new FIXED_NEVER_USERS build-time option creates a list of "never users"
165 that cannot be overridden. The default in the distributed EDITME is "root".
166 If for some reason you were (against advice) running deliveries as root, you
167 will have to ensure that FIXED_NEVER_USERS is not set in your
168 Local/Makefile.
1702. The ${quote: operator now quotes an empty string, which it did not before.
1723. Version 4.23 saves the contents of the ACL variables with the message, so
173 that they can be used later. If one of these variables contains a newline,
174 there will be a newline character in the spool that will not be interpreted
175 correctely by a previous version of Exim. (Exim ignores keyed spool file
176 items that it doesn't understand - precisely for this kind of problem - but
177 it expects them all to be on one line.)
179 So the bottom line is: if you have newlines in your ACL variables, you
180 cannot retreat from 4.23.
183Version 4.21
1861. The idea of the "warn" ACL verb is that it adds a header or writes to the
187 log only when "message" or "log_message" are set. However, if one of the
188 conditions was an address verification, or a call to a nested ACL, the
189 messages generated by the underlying test were being passed through. This
190 no longer happens. The underlying message is available in $acl_verify_
191 message for both "message" and "log_message" expansions, so it can be
192 passed through if needed.
1942. The way that the $h_ (and $header_) expansions work has been changed by the
195 addition of RFC 2047 decoding. See the main documentation (the NewStuff file
196 until release 4.30, then the manual) for full details. Briefly, there are
197 now three forms:
199 $rh_xxx: and $rheader_xxx: give the original content of the header
200 line(s), with no processing at all.
202 $bh_xxx: and $bheader_xxx: remove leading and trailing white space, and
203 then decode base64 or quoted-printable "words" within the header text,
204 but do not do charset translation.
206 $h_xxx: and $header_xxx: attempt to translate the $bh_ string to a
207 standard character set.
209 If you have previously been using $h_ expansions to access the raw
210 characters, you should change to $rh_ instead.
2123. When Exim creates an RFC 2047 encoded word in a header line, it labels it
213 with the default character set from the headers_charset option instead of
214 always using iso-8859-1.
2164. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
217 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
218 called TMPDIR, and if it finds it is different, it changes its value.
2205. Following a discussion on the list, the rules by which Exim recognises line
221 endings on incoming messages have been changed. The -dropcr and drop_cr
222 options are now no-ops, retained only for backwards compatibility. The
223 following line terminators are recognized: LF CRLF CR. However, special
224 processing applies to CR:
226 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
227 nor a local message in the state where . is a terminator.
229 (ii) If a bare CR is encountered in a header line, an extra space is added
230 after the line terminator so as not to end the header. The reasoning
231 behind this is that bare CRs in header lines are most likely either
232 to be mistakes, or people trying to play silly games.
2346. The code for using daemon_smtp_port, local_interfaces, and the -oX options
235 has been reorganized. It is supposed to be backwards compatible, but it is
236 mentioned here just in case I've screwed up.
240Version 4.20
2431. I have tidied and re-organized the code that uses alarm() for imposing time
244 limits on various things. It shouldn't affect anything, but if you notice
245 processes getting stuck, it may be that I've broken something.
2472. The "arguments" log selector now also logs the current working directory
248 when Exim is called.
2503. An incompatible change has been made to the appendfile transport. This
251 affects the case when it is used for file deliveries that are set up by
252 .forward and filter files. Previously, any settings of the "file" or
253 "directory" options were ignored. It is hoped that, like the address_file
254 transport in the default configuration, these options were never in fact set
255 on such transports, because they were of no use.
257 Now, if either of these options is set, it is used. The path that is passed
258 by the router is in $address_file (this is not new), so it can be used as
259 part of a longer path, or modified in any other way that expansion permits.
261 If neither "file" nor "directory" is set, the behaviour is unchanged.
2634. Related to the above: in a filter, if a "save" command specifies a non-
264 absolute path, the value of $home/ is pre-pended. This no longer happens if
265 $home is unset or is set to an empty string.
2675. Multiple file deliveries from a filter or .forward file can never be
268 batched; the value of batch_max on the transport is ignored for file
269 deliveries. I'm assuming that nobody ever actually set batch_max on the
270 address_file transport - it would have had odd effects previously.
2726. DESTDIR is the more common variable that ROOT for use when installing
273 software under a different root filing system. The Exim install script now
274 recognizes DESTDIR first; if it is not set, ROOT is used.
2767. If DESTDIR is set when installing Exim, it no longer prepends its value to
277 the path of the system aliases file that appears in the default
278 configuration (when a default configuration is installed). If an aliases
279 file is actually created, its name *does* use the prefix.
282Version 4.14
2851. The default for the maximum number of unknown SMTP commands that Exim will
286accept before dropping a connection has been reduced from 5 to 3. However, you
287can now change the value by setting smtp_max_unknown_commands.
2892. The ${quote: operator has been changed so that it turns newline and carriage
290return characters into \n and \r, respectively.
2923. The file names used for maildir messages now include the microsecond time
293fraction as well as the time in seconds, to cope with systems where the process
294id can be re-used within the same second. The format is now
296 <time>.H<microsec>P<pid>.<host>
298This should be a compatible change, but is noted here just in case.
3004. The rules for creating message ids have changed, to cope with systems where
301the process id can be re-used within the same second. The format, however, is
302unchanged, so this should not cause any problems, except as noted in the next
3055. The maximum value for localhost_number has been reduced from 255 to 16, in
306order to implement the new message id rules. For operating systems that have
307case-insensitive file systems (Cygwin and Darwin), the limit is 10.
3096. verify = header_syntax was allowing unqualified addresses in all cases. Now
310it allows them only for locally generated messages and from hosts that match
311sender_unqualified_hosts or recipient_unqualified_hosts, respectively.
3137. For reasons lost in the mists of time, when a pipe transport was run, the
314environment variable MESSAGE_ID was set to the message ID preceded by 'E' (the
315form used in Message-ID: header lines). The 'E' has been removed.
318Version 4.11
3211. The handling of lines in the configuration file has changed. Previously,
322macro expansion was applied to logical lines, after continuations had been
323joined on. This meant that it could not be used in .include lines, which are
324handled as physical rather than logical lines. Macro expansion is now done on
325physical lines rather than logical lines. This means there are two
328 (a) A macro that expands to # to turn a line into a comment now applies only
329 to the physical line where it appears. Previously, it would have caused
330 any following continuations also to be ignored.
332 (b) A macro name can no longer be split over the boundary between a line and
333 its continuation. Actually, this is more of a bug fix. :-)
3352. The -D command line option must now all be within one command line item.
336This makes it possible to use -D to set a macro to the empty string by commands
337such as
339 exim -DABC ...
340 exim -DABC= ...
342Previously, these items would have moved on to the next item on the command
343line. To include spaces in a macro definition item, quotes must be used, in
344which case you can also have spaces after -D and surrounding the equals. For
347 exim '-D ABC = something' ...
3493. The way that addresses that redirect to themselves are handled has been
350changed, in order to fix an obscure bug. This should not cause any problems
351except in the case of wanting to go back from a 4.11 (or later) release to an
352earlier release. If there are undelivered messages on the spool that contain
353addresses which redirect to themselves, and the redirected addresses have
354already been delivered, you might get a duplicate delivery if you revert to an
355earlier Exim.
3574. The default way of looking up IP addresses for hosts in the manualroute and
358queryprogram routers has been changed. If "byname" or "bydns" is explicitly
359specified, there is no change, but if no method is specified, Exim now behaves
360as follows:
362 First, a DNS lookup is done. If this yields anything other than
363 HOST_NOT_FOUND, that result is used. Otherwise, Exim goes on to try a call to
364 getipnodebyname() (or gethostbyname() on older systems) and the result of the
365 lookup is the result of that call.
367This change has been made because it has been discovered that on some systems,
368if a DNS lookup called via getipnodebyname() times out, HOST_NOT_FOUND is
369returned instead of TRY_AGAIN. Thus, it is safest to try a DNS lookup directly
370first, and only if that gives a definite "no such host" to try the local
3735. In fixing the minor security problem with pid_file_path, I have removed some
374backwards-compatible (undocumented) code which was present to ease conversion
375from Exim 3. In Exim 4, pid_file_path is a literal; in Exim 3 it was allowed to
376contain "%s", which was replaced by the port number for daemons listening on
377non-standard ports. In Exim 4, such daemons do not write a pid file. The
378backwards compatibility feature was to replace "%s" by nothing if it occurred
379in an Exim 4 setting of pid_file_path. The bug was in this code. I have solved
380the problem by removing the backwards compatibility feature. Thus, if you still
381have "%s" somewhere in a setting of pid_file_path, you should remove it.
3836. There has been an extension to lsearch files. The keys in these files may
384now be quoted in order to allow for whitespace and colons in them. This means
385that if you were previously using keys that began with a doublequote, you will
386now have to wrap them with extra quotes and escape the internal quotes. The
387possibility that anybody is actually doing this seems extremely remote, but it
388is documented just in case.
391Version 4.10
394The build-time parameter EXIWHAT_KILL_ARG has been renamed EXIWHAT_KILL_SIGNAL
395to better reflect its function. The OS-specific files have been updated. Only
396if you have explicitly set this in your Makefile (highly unlikely) do you need
397to change anything.