Commit | Line | Data |
---|---|---|
0f4f2a88 PH |
1 | $Cambridge: exim/src/README.UPDATING,v 1.1 2004/10/06 14:12:40 ph10 Exp $ |
2 | ||
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... | |
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 | ||
27 | The rest of this document contains information about changes in 4.xx releases | |
28 | that might affect a running system. | |
29 | ||
30 | ||
31 | Version 4.42 | |
32 | ------------ | |
33 | ||
34 | RFC 3848 specifies standard names for the "with" phrase in Received: header | |
35 | lines when AUTH and/or TLS are in use. This is the "received protocol" | |
36 | field. Exim used to use "asmtp" for authenticated SMTP, without any | |
37 | indication (in the protocol name) for TLS use. Now it follows the RFC and | |
38 | uses "esmtpa" if the connection is authenticated, "esmtps" if it is | |
39 | encrypted, and "esmtpsa" if it is both encrypted and authenticated. These names | |
40 | appear in log lines as well as in Received: header lines. | |
41 | ||
42 | ||
43 | Version 4.34 | |
44 | ------------ | |
45 | ||
46 | Change 4.31/2 gave problems to data ACLs and local_scan() functions that | |
47 | expected to see a Received: header. I have changed to yet another scheme. The | |
48 | Received: header is now generated after the body is received, but before the | |
49 | ACL or local_scan() is called. After they have run, the timestamp in the | |
50 | Received: header is updated. | |
51 | ||
52 | Thus, change (a) of 4.31/2 has been reversed, but change (b) is still true, | |
53 | which is lucky, since I decided it was a bug fix. | |
54 | ||
55 | ||
56 | Version 4.33 | |
57 | ------------ | |
58 | ||
59 | If an expansion in a condition on a "warn" statement fails because a lookup | |
60 | defers, the "warn" statement is abandoned, and the next ACL statement is | |
61 | processed. Previously this caused the whole ACL to be aborted. | |
62 | ||
63 | ||
64 | Version 4.32 | |
65 | ------------ | |
66 | ||
67 | Change 4.31/2 has been reversed, as it proved contentious. Recipient callout | |
68 | verification now uses <> in the MAIL command by default, as it did before. A | |
69 | new callout option, "use_sender", has been added to request the other | |
70 | behaviour. | |
71 | ||
72 | ||
73 | Version 4.31 | |
74 | ------------ | |
75 | ||
76 | 1. If you compile Exim to use GnuTLS, it now requires the use of release 1.0.0 | |
77 | or greater. The interface to the obsolete 0.8.x releases is no longer | |
78 | supported. There is one externally visible change: the format for the | |
79 | display of Distinguished Names now uses commas as a separator rather than a | |
80 | slash. This is to comply with RFC 2253. | |
81 | ||
82 | 2. When a message is received, the Received: header line is now generated when | |
83 | reception is complete, instead of at the start of reception. For messages | |
84 | that take a long time to come in, this changes the meaning of the timestamp. | |
85 | There are several side-effects of this change: | |
86 | ||
87 | (a) If a message is rejected by a DATA or non-SMTP ACL, or by local_scan(), | |
88 | the logged header lines no longer include the local Received: line, | |
89 | because it has not yet been created. If the message is a non-SMTP one, | |
90 | and the error is processed by sending a message to the sender, the copy | |
91 | of the original message that is returned does not have an added | |
92 | Received: line. | |
93 | ||
94 | (b) When a filter file is tested using -bf, no additional Received: header | |
95 | is added to the test message. After some thought, I decided that this | |
96 | is a bug fix. | |
97 | ||
98 | The contents of $received_for are not affected by this change. This | |
99 | variable still contains the single recipient of a message, copied after | |
100 | addresses have been rewritten, but before local_scan() is run. | |
101 | ||
102 | 2. Recipient callout verification, like sender verification, was using <> in | |
103 | the MAIL FROM command. This isn't really the right thing, since the actual | |
104 | sender may affect whether the remote host accepts the recipient or not. I | |
105 | have changed it to use the actual sender in the callout; this means that | |
106 | the cache record is now keyed on a recipient/sender pair, not just the | |
107 | recipient address. There doesn't seem to be a real danger of callout loops, | |
108 | since a callout by the remote host to check the sender would use <>. | |
109 | ||
110 | ||
111 | Version 4.30 | |
112 | ------------ | |
113 | ||
114 | 1. I have abolished timeout_DNS as an error that can be detected in retry | |
115 | rules, because it has never worked. Despite the fact that it has been | |
116 | documented since at least release 1.62, there was no code to support it. | |
117 | If you have used it in your retry rules, you will now get a warning message | |
118 | to the log and panic log. It is now treated as plain "timeout". | |
119 | ||
120 | 2. After discussion on the mailing list, Exim no longer adds From:, Date:, or | |
121 | Message-Id: header lines to messages that do not originate locally, that is, | |
122 | messages that have an associated sending host address. | |
123 | ||
124 | 3. When looking up a host name from an IP address, Exim now tries the DNS | |
125 | first, and only if that fails does it use gethostbyaddr() (or equivalent). | |
126 | This change was made because on some OS, not all the names are given for | |
127 | addresses with multiple PTR records via the gethostbyaddr() interface. The | |
128 | order of lookup can be changed by setting host_lookup_order. | |
129 | ||
130 | ||
131 | Version 4.23 | |
132 | ------------ | |
133 | ||
134 | 1. The new FIXED_NEVER_USERS build-time option creates a list of "never users" | |
135 | that cannot be overridden. The default in the distributed EDITME is "root". | |
136 | If for some reason you were (against advice) running deliveries as root, you | |
137 | will have to ensure that FIXED_NEVER_USERS is not set in your | |
138 | Local/Makefile. | |
139 | ||
140 | 2. The ${quote: operator now quotes an empty string, which it did not before. | |
141 | ||
142 | 3. Version 4.23 saves the contents of the ACL variables with the message, so | |
143 | that they can be used later. If one of these variables contains a newline, | |
144 | there will be a newline character in the spool that will not be interpreted | |
145 | correctely by a previous version of Exim. (Exim ignores keyed spool file | |
146 | items that it doesn't understand - precisely for this kind of problem - but | |
147 | it expects them all to be on one line.) | |
148 | ||
149 | So the bottom line is: if you have newlines in your ACL variables, you | |
150 | cannot retreat from 4.23. | |
151 | ||
152 | ||
153 | Version 4.21 | |
154 | ------------ | |
155 | ||
156 | 1. The idea of the "warn" ACL verb is that it adds a header or writes to the | |
157 | log only when "message" or "log_message" are set. However, if one of the | |
158 | conditions was an address verification, or a call to a nested ACL, the | |
159 | messages generated by the underlying test were being passed through. This | |
160 | no longer happens. The underlying message is available in $acl_verify_ | |
161 | message for both "message" and "log_message" expansions, so it can be | |
162 | passed through if needed. | |
163 | ||
164 | 2. The way that the $h_ (and $header_) expansions work has been changed by the | |
165 | addition of RFC 2047 decoding. See the main documentation (the NewStuff file | |
166 | until release 4.30, then the manual) for full details. Briefly, there are | |
167 | now three forms: | |
168 | ||
169 | $rh_xxx: and $rheader_xxx: give the original content of the header | |
170 | line(s), with no processing at all. | |
171 | ||
172 | $bh_xxx: and $bheader_xxx: remove leading and trailing white space, and | |
173 | then decode base64 or quoted-printable "words" within the header text, | |
174 | but do not do charset translation. | |
175 | ||
176 | $h_xxx: and $header_xxx: attempt to translate the $bh_ string to a | |
177 | standard character set. | |
178 | ||
179 | If you have previously been using $h_ expansions to access the raw | |
180 | characters, you should change to $rh_ instead. | |
181 | ||
182 | 3. When Exim creates an RFC 2047 encoded word in a header line, it labels it | |
183 | with the default character set from the headers_charset option instead of | |
184 | always using iso-8859-1. | |
185 | ||
186 | 4. If TMPDIR is defined in Local/Makefile (default in src/EDITME is | |
187 | TMPDIR="/tmp"), Exim checks for the presence of an environment variable | |
188 | called TMPDIR, and if it finds it is different, it changes its value. | |
189 | ||
190 | 5. Following a discussion on the list, the rules by which Exim recognises line | |
191 | endings on incoming messages have been changed. The -dropcr and drop_cr | |
192 | options are now no-ops, retained only for backwards compatibility. The | |
193 | following line terminators are recognized: LF CRLF CR. However, special | |
194 | processing applies to CR: | |
195 | ||
196 | (i) The sequence CR . CR does *not* terminate an incoming SMTP message, | |
197 | nor a local message in the state where . is a terminator. | |
198 | ||
199 | (ii) If a bare CR is encountered in a header line, an extra space is added | |
200 | after the line terminator so as not to end the header. The reasoning | |
201 | behind this is that bare CRs in header lines are most likely either | |
202 | to be mistakes, or people trying to play silly games. | |
203 | ||
204 | 6. The code for using daemon_smtp_port, local_interfaces, and the -oX options | |
205 | has been reorganized. It is supposed to be backwards compatible, but it is | |
206 | mentioned here just in case I've screwed up. | |
207 | ||
208 | ||
209 | ||
210 | Version 4.20 | |
211 | ------------ | |
212 | ||
213 | 1. I have tidied and re-organized the code that uses alarm() for imposing time | |
214 | limits on various things. It shouldn't affect anything, but if you notice | |
215 | processes getting stuck, it may be that I've broken something. | |
216 | ||
217 | 2. The "arguments" log selector now also logs the current working directory | |
218 | when Exim is called. | |
219 | ||
220 | 3. An incompatible change has been made to the appendfile transport. This | |
221 | affects the case when it is used for file deliveries that are set up by | |
222 | .forward and filter files. Previously, any settings of the "file" or | |
223 | "directory" options were ignored. It is hoped that, like the address_file | |
224 | transport in the default configuration, these options were never in fact set | |
225 | on such transports, because they were of no use. | |
226 | ||
227 | Now, if either of these options is set, it is used. The path that is passed | |
228 | by the router is in $address_file (this is not new), so it can be used as | |
229 | part of a longer path, or modified in any other way that expansion permits. | |
230 | ||
231 | If neither "file" nor "directory" is set, the behaviour is unchanged. | |
232 | ||
233 | 4. Related to the above: in a filter, if a "save" command specifies a non- | |
234 | absolute path, the value of $home/ is pre-pended. This no longer happens if | |
235 | $home is unset or is set to an empty string. | |
236 | ||
237 | 5. Multiple file deliveries from a filter or .forward file can never be | |
238 | batched; the value of batch_max on the transport is ignored for file | |
239 | deliveries. I'm assuming that nobody ever actually set batch_max on the | |
240 | address_file transport - it would have had odd effects previously. | |
241 | ||
242 | 6. DESTDIR is the more common variable that ROOT for use when installing | |
243 | software under a different root filing system. The Exim install script now | |
244 | recognizes DESTDIR first; if it is not set, ROOT is used. | |
245 | ||
246 | 7. If DESTDIR is set when installing Exim, it no longer prepends its value to | |
247 | the path of the system aliases file that appears in the default | |
248 | configuration (when a default configuration is installed). If an aliases | |
249 | file is actually created, its name *does* use the prefix. | |
250 | ||
251 | ||
252 | Version 4.14 | |
253 | ------------ | |
254 | ||
255 | 1. The default for the maximum number of unknown SMTP commands that Exim will | |
256 | accept before dropping a connection has been reduced from 5 to 3. However, you | |
257 | can now change the value by setting smtp_max_unknown_commands. | |
258 | ||
259 | 2. The ${quote: operator has been changed so that it turns newline and carriage | |
260 | return characters into \n and \r, respectively. | |
261 | ||
262 | 3. The file names used for maildir messages now include the microsecond time | |
263 | fraction as well as the time in seconds, to cope with systems where the process | |
264 | id can be re-used within the same second. The format is now | |
265 | ||
266 | <time>.H<microsec>P<pid>.<host> | |
267 | ||
268 | This should be a compatible change, but is noted here just in case. | |
269 | ||
270 | 4. The rules for creating message ids have changed, to cope with systems where | |
271 | the process id can be re-used within the same second. The format, however, is | |
272 | unchanged, so this should not cause any problems, except as noted in the next | |
273 | item. | |
274 | ||
275 | 5. The maximum value for localhost_number has been reduced from 255 to 16, in | |
276 | order to implement the new message id rules. For operating systems that have | |
277 | case-insensitive file systems (Cygwin and Darwin), the limit is 10. | |
278 | ||
279 | 6. verify = header_syntax was allowing unqualified addresses in all cases. Now | |
280 | it allows them only for locally generated messages and from hosts that match | |
281 | sender_unqualified_hosts or recipient_unqualified_hosts, respectively. | |
282 | ||
283 | 7. For reasons lost in the mists of time, when a pipe transport was run, the | |
284 | environment variable MESSAGE_ID was set to the message ID preceded by 'E' (the | |
285 | form used in Message-ID: header lines). The 'E' has been removed. | |
286 | ||
287 | ||
288 | Version 4.11 | |
289 | ------------ | |
290 | ||
291 | 1. The handling of lines in the configuration file has changed. Previously, | |
292 | macro expansion was applied to logical lines, after continuations had been | |
293 | joined on. This meant that it could not be used in .include lines, which are | |
294 | handled as physical rather than logical lines. Macro expansion is now done on | |
295 | physical lines rather than logical lines. This means there are two | |
296 | incompatibilities: | |
297 | ||
298 | (a) A macro that expands to # to turn a line into a comment now applies only | |
299 | to the physical line where it appears. Previously, it would have caused | |
300 | any following continuations also to be ignored. | |
301 | ||
302 | (b) A macro name can no longer be split over the boundary between a line and | |
303 | its continuation. Actually, this is more of a bug fix. :-) | |
304 | ||
305 | 2. The -D command line option must now all be within one command line item. | |
306 | This makes it possible to use -D to set a macro to the empty string by commands | |
307 | such as | |
308 | ||
309 | exim -DABC ... | |
310 | exim -DABC= ... | |
311 | ||
312 | Previously, these items would have moved on to the next item on the command | |
313 | line. To include spaces in a macro definition item, quotes must be used, in | |
314 | which case you can also have spaces after -D and surrounding the equals. For | |
315 | example: | |
316 | ||
317 | exim '-D ABC = something' ... | |
318 | ||
319 | 3. The way that addresses that redirect to themselves are handled has been | |
320 | changed, in order to fix an obscure bug. This should not cause any problems | |
321 | except in the case of wanting to go back from a 4.11 (or later) release to an | |
322 | earlier release. If there are undelivered messages on the spool that contain | |
323 | addresses which redirect to themselves, and the redirected addresses have | |
324 | already been delivered, you might get a duplicate delivery if you revert to an | |
325 | earlier Exim. | |
326 | ||
327 | 4. The default way of looking up IP addresses for hosts in the manualroute and | |
328 | queryprogram routers has been changed. If "byname" or "bydns" is explicitly | |
329 | specified, there is no change, but if no method is specified, Exim now behaves | |
330 | as follows: | |
331 | ||
332 | First, a DNS lookup is done. If this yields anything other than | |
333 | HOST_NOT_FOUND, that result is used. Otherwise, Exim goes on to try a call to | |
334 | getipnodebyname() (or gethostbyname() on older systems) and the result of the | |
335 | lookup is the result of that call. | |
336 | ||
337 | This change has been made because it has been discovered that on some systems, | |
338 | if a DNS lookup called via getipnodebyname() times out, HOST_NOT_FOUND is | |
339 | returned instead of TRY_AGAIN. Thus, it is safest to try a DNS lookup directly | |
340 | first, and only if that gives a definite "no such host" to try the local | |
341 | function. | |
342 | ||
343 | 5. In fixing the minor security problem with pid_file_path, I have removed some | |
344 | backwards-compatible (undocumented) code which was present to ease conversion | |
345 | from Exim 3. In Exim 4, pid_file_path is a literal; in Exim 3 it was allowed to | |
346 | contain "%s", which was replaced by the port number for daemons listening on | |
347 | non-standard ports. In Exim 4, such daemons do not write a pid file. The | |
348 | backwards compatibility feature was to replace "%s" by nothing if it occurred | |
349 | in an Exim 4 setting of pid_file_path. The bug was in this code. I have solved | |
350 | the problem by removing the backwards compatibility feature. Thus, if you still | |
351 | have "%s" somewhere in a setting of pid_file_path, you should remove it. | |
352 | ||
353 | 6. There has been an extension to lsearch files. The keys in these files may | |
354 | now be quoted in order to allow for whitespace and colons in them. This means | |
355 | that if you were previously using keys that began with a doublequote, you will | |
356 | now have to wrap them with extra quotes and escape the internal quotes. The | |
357 | possibility that anybody is actually doing this seems extremely remote, but it | |
358 | is documented just in case. | |
359 | ||
360 | ||
361 | Version 4.10 | |
362 | ------------ | |
363 | ||
364 | The build-time parameter EXIWHAT_KILL_ARG has been renamed EXIWHAT_KILL_SIGNAL | |
365 | to better reflect its function. The OS-specific files have been updated. Only | |
366 | if you have explicitly set this in your Makefile (highly unlikely) do you need | |
367 | to change anything. | |
368 | ||
369 | **** |