Allow slashes in the dns_check_names_pattern default value.
[exim.git] / doc / doc-txt / ChangeLog
CommitLineData
230205fc 1$Cambridge: exim/doc/doc-txt/ChangeLog,v 1.303 2006/02/14 16:18:14 ph10 Exp $
495ae4b0
PH
2
3Change log file for Exim from version 4.21
4-------------------------------------------
5
214e2000
PH
6Exim version 4.61
7-----------------
8
9PH/01 The code for finding all the local interface addresses on a FreeBSD
10 system running IPv6 was broken. This may well have applied to all BSD
11 systems, as well as to others that have similar system calls. The broken
12 code found IPv4 interfaces correctly, but gave incorrect values for the
13 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
14 that it would not match correctly against @[] and not recognize the IPv6
15 addresses as local.
16
f9daeae0
PH
17PH/02 The ipliteral router was not recognizing addresses of the form user@
18 [ipv6:....] because it didn't know about the "ipv6:" prefix.
19
7e66e54d
PH
20PH/03 Added disable_ipv6.
21
c8ea1597
PH
22PH/04 Changed $reply_address to use the raw form of the headers instead of the
23 decoded form, because it is most often used to construct To: headers
24 lines in autoreplies, and the decoded form may well be syntactically
25 invalid. However, $reply_address has leading white space removed, and all
26 newlines turned into spaces so that the autoreply transport does not
27 grumble.
28
911f6fde
PH
29PH/05 If group was specified without a user on a router, and no group or user
30 was specified on a transport, the group from the router was ignored.
31
47ca6d6c
PH
32PH/06 Increased the number of ACL variables to 20 of each type, and arranged
33 for visible compile-time settings that can be used to change these
34 numbers, for those that want even more. Backwards compatibility with old
35 spool files has been maintained. However, going back to a previous Exim
36 release will lost any variables that are in spool files.
37
ed0e9820
PH
38PH/07 Two small changes when running in the test harness: increase delay when
39 passing a TCP/IP connection to a new process, in case the original
40 process has to generate a bounce, and remove special handling of
41 127.0.0.2 (sic), which is no longer necessary.
42
eff37e47
PH
43PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
44 be the same on different OS.
45
1921d2ea
PH
46PH/09 Moved a debug statement in filter processing to avoid a race problem when
47 testing.
48
b3f69ca8
JJ
49JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
50 whether --show-vars was specified or not
51
52JJ/02 exipick: Added support for new ACL variable spool format introduced
53 in 4.61-PH/06
54
424a1c63
PH
55PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
56 syntactically invalid From: or Reply-to: line, and a filter used this to
57 generate an autoreply, and therefore failed to obtain an address for the
58 autoreply, Exim could try to deliver to a non-existent relative file
59 name, causing unrelated and misleading errors. What now happens is that
60 it logs this as a hard delivery error, but does not attempt to create a
61 bounce message.
62
7a100415
PH
63PH/11 The exinext utility has a -C option for testing purposes, but although
64 the given file was scanned by exinext itself; it wasn't being passed on
65 when Exim was called.
66
19b9dc85
PH
67PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
68 an end-of-file indication when reading a command response.
69
309bd837
PH
70PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
71 compiled. In many other places in Exim, IPv6 addresses are always
72 recognized, so I have changed this. It also means that IPv4 domain
73 literals of the form [IPV4:n.n.n.n] are now always recognized.
74
59e82a2a
PH
75PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
76 used if the router is not running as root, for example, when verifying at
77 ACL time, or when using -bh. The debugging output from this situation was
78 non-existent - all you got was a failure to exec. I have made two
79 changes:
80
81 (a) Failures to set uid/gid, the current directory, or a process leader
82 in a subprocess such as that created by queryprogram now generate
83 suitable debugging ouput when -d is set.
84
85 (b) The queryprogram router detects when it is not running as root,
86 outputs suitable debugging information if -d is set, and then runs
87 the subprocess without attempting to change uid/gid.
88
9edc04ce
PH
89PH/15 Minor change to Makefile for building test_host (undocumented testing
90 feature).
91
1349e1e5
PH
92PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
93 additional section of a DNS packet that returns MX or SRV records.
94 Instead, it always explicitly searches for A/AAAA records. This avoids
95 major problems that occur when a DNS server includes only records of one
96 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
97 fixed another bug: if SRV records were looked up and the corresponding
98 address records were *not* found in the additional section, the port
99 values from the SRV records were lost.
100
ea49d0e1
PH
101PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
102 using the correct key (the original address) when searching the retry
103 rules in order to find which one to use for generating the retry hint.
104
064a94c9
PH
105PH/18 If quota_warn_message contains a From: header, Exim now refrains from
106 adding the default one. Similarly, if it contains a Reply-To: header, the
107 errors_reply_to option, if set, is not used.
108
727071f8
PH
109PH/19 When calculating a retry time, Exim used to measure the "time since
110 failure" by looking at the "first failed" field in the retry record. Now
111 it does not use this if it is later than than the arrival time of the
112 message. Instead it uses the arrival time. This makes for better
113 behaviour in cases where some deliveries succeed, thus re-setting the
114 "first failed" field. An example is a quota failure for a huge message
115 when small messages continue to be delivered. Without this change, the
116 "time since failure" will always be short, possible causing more frequent
117 delivery attempts for the huge message than are intended.
118
f78eb7c6
PH
119PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
120 $1, $2, $3) because the numerical variables can be reset during some
121 expansion items (e.g. "match"), thereby losing the authentication data.
122
21c28500
PH
123PH/21 Make -bV show the size of off_t variables so that the test suite can
124 decide whether to run tests for quotas > 2G.
125
126PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
127 mailbox_size, and mailbox_filecount in the appendfile transport. If a
128 filecount value is greater than 2G or if a quota value is greater than 2G
129 on a system where the size of off_t is not greater than 4, a panic error
130 is given.
131
1688f43b
PH
132PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
133 never match. The debug and -bh output now contains an explicit error
134 message indicating a malformed IPv4 address or mask.
135
136PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
137 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
138 PH/23 above applies.
139
9675b384
PH
140PH/25 Do not write to syslog when running in the test harness. The only
141 occasion when this arises is a failure to open the main or panic logs
142 (for which there is an explicit test).
143
6a3f1455
PH
144PH/26 Added the /no_tell option to "control=freeze".
145
dac79d3e
PH
146PH/27 If a host name lookup failed very early in a connection, for example, if
147 the IP address matched host_lookup and the reverse lookup yielded a name
148 that did not have a forward lookup, an error message of the form "no IP
149 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
150 could be logged. Now it outputs the IP address instead of "NULL".
1349e1e5 151
5977a0b3
PH
152PH/28 An enabling patch from MH: add new function child_open_exim2() which
153 allows the sender and the authenticated sender to be set when
154 submitting a message from within Exim. Since child_open_exim() is
155 documented for local_scan(), the new function should be too.
156
c91535f3
PH
157PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
158 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
159 results in an empty string is now treated as unset.
160
0d46a8c8
PH
161PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
162
278c6e6c
PH
163PH/31 Added sender_verify_fail logging option.
164
2cbb4081
PH
165PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
166 needed by RFC 822 but not by RFC 2822 was commented out. I have now
167 tidied the source and removed it altogether.
168
3eef829e
PH
169PH/33 When a queue run was abandoned because the load average was too high, a
170 log line was always written; now it is written only if the queue_run log
171 selector is set. In addition, the log line for abandonment now contains
172 information about the queue run such as the pid. This is always present
173 in "start" and "stop" lines but was omitted from the "abandon" line.
174
1ab95fa6
PH
175PH/34 Omit spaces between a header name and the colon in the error message that
176 is given when verify = headers_syntax fails (if there are lots of them,
177 the message gets confusing).
178
230205fc
PH
179PH/35 Change the default for dns_check_names_pattern to allow slashes within
180 names, as there are now some PTR records that contain slashes. This check
181 is only to protect against broken name servers that fall over on strange
182 characters, so the fact that it applies to all lookups doesn't matter.
183
214e2000 184
5de37277
PH
185Exim version 4.60
186-----------------
187
cc38ddbf
PH
188PH/01 Two changes to the default runtime configuration:
189
190 (1) Move the checks for relay_from_hosts and authenticated clients from
191 after to before the (commented out) DNS black list checks.
192
193 (2) Add control=submission to the relay_from_hosts and authenticated
194 clients checks, on the grounds that messages accepted by these
195 statements are most likely to be submissions.
5de37277 196
72fdd6ae
PH
197PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
198
199 (1) Generate an error if the third argument for the ${prvs expansion is
200 not a single digit.
201
202 (2) Treat a missing third argument of ${prvscheck as if it were an empty
203 string.
204
205 (3) Reset the variables that are obtained from the first argument of
206 ${prvscheck and used in the second argument before leaving the code,
207 because their memory is reclaimed, so using them afterwards may do
208 silly things.
209
210 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
211 one (it's much easier than Tom thought :-).
212
213 (5) Because of (4), we can now allow for the use of $prvscheck_result
214 inside the third argument.
cb9328de 215
cb741023
PH
216PH/03 For some reason, the default setting of PATH when running a command from
217 a pipe transport was just "/usr/bin". I have changed it to
218 "/bin:/usr/bin".
219
f174f16e
PH
220PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
221 anything to be listed in the output from -bV.
b2f5a032 222
c25242d7
PH
223PH/05 When a filter generated an autoreply, the entire To: header line was
224 quoted in the delivery log line, like this:
225
226 => >A.N.Other <ano@some.domain> <original@ddress> ...
227
228 This has been changed so that it extracts the operative address. There
229 may be more than one such address. If so, they are comma-separated, like
230 this:
231
232 => >ano@some.domain,ona@other.domain <original@ddress> ...
233
82c19f95
PH
234PH/06 When a client host used a correct literal IP address in a HELO or EHLO
235 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
236 not being looked up in the rDNS to get a host name, Exim was showing the
237 IP address twice in Received: lines, even though the IP addresses were
238 identical. For example:
239
240 Received: from [1.2.3.4] (helo=[1.2.3.4])
241
242 However, if the real host name was known, it was omitting the HELO data
243 if it matched the actual IP address. This has been tidied up so that it
244 doesn't show the same IP address twice.
245
d7ffbc12
PH
246PH/07 When both +timestamp and +memory debugging was on, the value given by
247 $tod_xxx expansions could be wrong, because the tod_stamp() function was
248 called by the debug printing, thereby overwriting the timestamp buffer.
249 Debugging no longer uses the tod_stamp() function when +timestamp is set.
250
9f526266
PH
251PH/08 When the original message was included in an autoreply transport, it
252 always said "this is a copy of the message, including all the headers",
253 even if body_only or headers_only was set. It now gives an appropriate
254 message.
255
87fcc8b9
PH
256PH/09 Applied a patch from the Sieve maintainer which:
257
258 o fixes some comments
259 o adds the (disabled) notify extension core
260 o adds some debug output for the result of if/elsif tests
261 o points to the current vacation draft in the documentation
262 and documents the missing references header update
263
264 and most important:
265
266 o fixes a bug in processing the envelope test (when testing
267 multiple envelope elements, the last element determinted the
268 result)
269
456682f5
PH
270PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
271 Electronic Mail") by including:
272
273 Auto-submitted: auto-generated
274
275 in the messages that it generates (bounce messages and others, such as
276 warnings). In the case of bounce messages for non-SMTP mesages, there was
277 also a typo: it was using "Auto_submitted" (underscore instead of
278 hyphen). Since every message generated by Exim is necessarily in response
279 to another message, thes have all been changed to:
280
281 Auto-Submitted: auto-replied
282
283 in accordance with these statements in the RFC:
284
285 The auto-replied keyword:
286
287 - SHOULD be used on messages sent in direct response to another
288 message by an automatic process,
289
290 - MUST NOT be used on manually-generated messages,
291
292 - MAY be used on Delivery Status Notifications (DSNs) and Message
293 Disposition Notifications (MDNs),
294
295 - MUST NOT be used on messages generated by automatic or periodic
296 processes, except for messages which are automatic responses to
297 other messages.
298
3e46c1aa
PH
299PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
300 to the default Received: header definition.
456682f5 301
49826d12
PH
302PH/12 Added log selector acl_warn_skipped (default on).
303
eba0c039
PH
304PH/13 After a successful wildlsearch lookup, discard the values of numeric
305 variables because (a) they are in the wrong storage pool and (b) even if
306 they were copied, it wouldn't work properly because of the caching.
307
a0d6ba8a
PH
308PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
309 checking when decoding. Apparently there are clients that generate
310 overlong encoded strings. Why am I not surprised?
311
f0917727
PH
312PH/15 If the first argument of "${if match_address" was not empty, but did not
313 contain an "@" character, Exim crashed. Now it writes a panic log message
314 and treats the condition as false.
315
096fee00
PH
316PH/16 In autoreply, treat an empty string for "once" the same as unset.
317
024bd3c2
PH
318PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
319 extension "envelope-auth". The code is finished and in agreement with
320 other implementations, but there is no documentation so far and in fact,
321 nobody wrote the draft yet. This extension is currently #undef'ed, thus
322 not changing the active code.
323
324 Print executed "if" and "elsif" statements when debugging is used. This
325 helps a great deal to understand what a filter does.
326
327 Document more things not specified clearly in RFC3028. I had all this
328 sorted out, when out of a sudden new issues came to my mind. Oops."
329
df199fec
PH
330PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
331 (Bugzilla #53).
332
d27f1df3
PH
333PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
334 canonical form (as documented). However, after a host name lookup from
335 the IP address, check_host() was doing a simple string comparison with
336 addresses acquired from the DNS when checking that the found name did
337 have the original IP as one of its addresses. Since any found IPv6
338 addresses are likely to be in abbreviated form, the comparison could
339 fail. Luckily, there already exists a function for doing the comparison
340 by converting both addresses to binary, so now that is used instead of
341 the text comparison.
342
96776534
PH
343PH/20 There was another similar case to PH/19, when a complete host name was
344 given in a host list; looking up its IP address could give an abbreviated
345 form, whereas the current host's name might or might not be abbreviated.
346 The same fix has been applied.
347
5de37277 348
9a799bc0
PH
349Exim version 4.54
350-----------------
351
352PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
353 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
354 It now does.
355
99a4b039
PH
356PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
357 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
358
359PH/03 Typo: missing ".o" in src/pcre/Makefile.
360
4b233853
PH
361PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
362 header line, restrict the check to what is listed in RFCs 2369 and 2929.
363 Also, for "Auto-Submitted", treat anything other than "no" as
364 non-personal, in accordance with RFC 3834. (Previously it treated
365 anything starting "auto-" as non-personal.)
366
8857ccfd
PH
367TF/01 The control=submission/name=... option had a problem with syntax
368 errors if the name included a slash character. The /name= option
369 now slurps the rest of the string, so it can include any characters
370 but it must come last in the list of options (after /sender_retain
371 or /domain=).
372
433a2980
PH
373PH/05 Some modifications to the interface to the fake nameserver for the new
374 testing suite.
375
3e46c1aa 376
9a799bc0 377
e3a311ba
TK
378Exim version 4.53
379-----------------
380
381TK/01 Added the "success_on_redirect" address verification option. See
382 NewStuff for rationale and an example.
383
13b685f9
PH
384PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
385
395ff96d
PH
386PH/02 Patch to exigrep to allow it to work on syslog lines.
387
5b68f6e4
PH
388PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
389 fread() to skip over the body file's header line, because in Cygwin the
390 header line is locked and is inaccessible.
391
1ab52c69
PH
392PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
393 co-exist for some time) to make it clear that it is the Exim ID that is
394 referenced, not the Message-ID: header line.
395
b07e6aa3
PH
396PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
397 string_format() function, because snprintf() does not exist on all
398 operating systems.
399
254e032f
PH
400PH/06 The use of forbid_filter_existstest now also locks out the use of the
401 ${stat: expansion item.
402
3af76a81
PH
403PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
404 protocol synchronization error", to keep the pedants happy.
405
2548ba04
PH
406PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
407 well as for IRIX systems, when gcc is being used. See the host.c source
408 file for comments.
409
b6c6011d
PH
410PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
411
cf39cf57
PH
412PH/10 Named domain lists were not working if used in a queue_smtp_domains
413 setting.
414
f1513293
PH
415PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
416 transport and to the smtp transport in LMTP mode.
417
727549a4
PH
418TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
419
af46795e
PH
420PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
421 run a filter in a subprocess. This could lead to confusion in subsequent
422 lookups in the parent process. There should also be a search_tidyup() at
423 the end of the subprocess.
424
d7b47fd0
PH
425PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
426 only if the host matched helo_try_verify_hosts, which caused the
427 verification to occur when the EHLO/HELO command was issued. The ACL just
428 tested the remembered result. Now, if a previous verification attempt has
429 not happened, "verify = helo" does it there and then.
430
ee744174
JJ
431JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
432
b582ab87
PH
433TK/03 Fix log output including CR from clamd.
434
41a13e0a
PH
435PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
436 exist provoked a memory error which could cause a segfault.
437
f625cc5a
PH
438PH/15 Installed PCRE 6.2
439
440PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
441
21f7af35
PH
442PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
443 of the problem. Specifically, suggested +O2 rather than +O1 for the
444 HP-UX compiler.
445
31480e42
PH
446PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
447
2d280592
PH
448PH/20 If a delivery was routed to a non-standard port by means of an SRV
449 record, the port was not correctly logged when the outgoing_port log
450 selector was set (it logged the transort's default port).
451
7cd1141b
PH
452PH/21 Added support for host-specific ports to manualroute, queryprogram,
453 fallback_hosts, and "hosts" in the smtp transport.
454
455PH/22 If the log selector "outgoing_port" is set, the port is now also given on
456 host errors such as "Connection refused".
457
750af86e
PH
458PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
459 authentication with radiusclient 0.4.9:
460
461 - Error returned from rc_read_config was caught wrongly
462 - Username/password not passed on to radius server due to wrong length.
463
464 The presumption is that some radiusclient API changes for 4.51/PH/17
465 were not taken care of correctly. The code is still untested by me (my
466 Linux distribution still has 0.3.2 of radiusclient), but it was
467 contributed by a Radius user.
468
469PH/24 When doing a callout, the value of $domain wasn't set correctly when
470 expanding the "port" option of the smtp transport.
471
4304270b
TK
472TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
473 while reading a MIME header. Thanks to Tom Hughes for a patch.
474
750af86e
PH
475PH/24 Include config.h inside local_scan.h so that configuration settings are
476 available.
477
64ffc24f
PH
478PH/25 Make $smtp_command_argument available after all SMTP commands. This means
479 that in an ACL for RCPT (for example), you can examine exactly what was
480 received.
481
5dd9625b
PH
482PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
483 commands, but it was not correctly comparing the address with the actual
484 client host address. Thus, it would show the EHLO address in Received:
485 header lines when this was not necessary.
486
5591031b
PH
487PH/27 Added the % operator to ${eval:}.
488
ba18e66a
PH
489PH/28 Exim tries to create and chdir to its spool directory when it starts;
490 it should be ignoring failures (because with -C, for example, it has lost
491 privilege). It wasn't ignoring creation failures other than "already
492 exists".
493
9cec981f
PH
494PH/29 Added "crypteq" to the list of supported features that Exim outputs when
495 -bV or -d is used.
496
aa2b5c79
PH
497PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
498 because an input line was too long, either on its own, or by virtue of
1509d3a8
PH
499 too many continuations, the temporary file was not being removed, and the
500 return code was incorrect.
aa2b5c79 501
48a53b7f
PH
502PH/31 Missing "BOOL" in function definition in filtertest.c.
503
1c59d63b
PH
504PH/32 Applied Sieve patches from the maintainer.
505
671012da
TK
506TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
507
1509d3a8
PH
508PH/33 Added "verify = not_blind".
509
510PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
511 Local/Makefile (with some defaults set). These are used in built scripts
512 such as exicyclog, but they have never been used in the exim_install
513 script (though there are many overriding facilities there). I have
514 arranged that the exim_install script now takes note of these two
515 settings.
516
517PH/35 Installed configuration files for Dragonfly.
518
2fe1a124
PH
519PH/36 When a locally submitted message by a trusted user did not contain a
520 From: header, and the sender address was obtained from -f or from an SMTP
521 MAIL command, and the trusted user did not use -F to supply a sender
522 name, $originator_name was incorrectly used when constructing a From:
523 header. Furthermore, $originator_name was used for submission mode
524 messages from external hosts without From: headers in a similar way,
525 which is clearly wrong.
526
8800895a
PH
527PH/37 Added control=suppress_local_fixups.
528
ccfdb010
PH
529PH/38 When log_selector = +received_sender was set, and the addition of the
530 sender made the log line's construction buffer exactly full, or one byte
531 less than full, an overflow happened when the terminating "\n" was
532 subsequently added.
533
1130bfb0
PH
534PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
535 when the result of a list match is failure because a DNS lookup failed.
536
ebcb507f
PH
537PH/40 RM_COMMAND is now used in the building process.
538
c35e155c
PH
539PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
540 the "build-* directories that it finds.
541
95d1f782
PH
542PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
543 address in a domain literal was a prefix of an interface address.
544
fd6de02e
PH
545PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
546 when verifying a sender address, unless rewrite_headers is false.
547
58de37c5
PH
548PH/44 Wrote a long comment about why errors_to addresses are verified as
549 recipients, not senders.
550
261cf466
TF
551TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
552 the ratelimit ACL was added.
553
3ee512ff
PH
554PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
555
e08c430f
PH
556PH/46 Added extra information about PostgreSQL errors to the error string.
557
bef5a11f
PH
558PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
559 avoiding the need to install special zones in a real server. This is
560 backwards compatible; if it can't find the fake resolver, it drops back.
561 Thus, both old and new test suites can be run.
562
7546de58
TF
563TF/02 Added util/ratelimit.pl
564
e5d5a95f
TF
565TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
566 clock is set back in time.
567
2e88a017
TF
568TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
569 Candler <B.Candler@pobox.com>.
570
a5f65aa4
TF
571TF/05 The fix for PH/43 was not completely correct; widen_domains is always
572 OK for addresses that are the result of redirections.
573
e7726cbf
PH
574PH/48 A number of further additions for the benefit of the new test suite,
575 including a fake gethostbyname() that interfaces to the fake DNS resolver
576 (see PH/47 above).
577
a7fdad5b
TF
578TF/06 The fix for widen_domains has also been applied to qualify_single and
579 search_parents which are the other dnslookup options that can cause
580 header rewrites.
581
6af56900
PH
582PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
583 type ("H").
584
0925ede6
PH
585PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
586
66afa403
TF
587TF/07 Exim produced the error message "an SRV record indicated no SMTP
588 service" if it encountered an MX record with an empty target hostname.
589 The message is now "an MX or SRV record indicated no SMTP service".
590
0154e85a
TF
591TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
592 if the DNS of the sending site is misconfigured. This is quite a
593 common situation. This change restores the behaviour of treating a
594 helo verification defer as a failure.
595
16f12c76
PH
596PH/51 If self=fail was set on a router, the bounce message did not include the
597 actual error message.
598
bbe902f0 599
e5a9dba6
PH
600Exim version 4.52
601-----------------
602
603TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
604
22c3b60b
PH
605PH/01 When a transport filter timed out in a pipe delivery, and the pipe
606 command itself ended in error, the underlying message about the transport
607 filter timeout was being overwritten with the pipe command error. Now the
608 underlying error message should be appended to the second error message.
609
06a9b4b5
PH
610TK/01 Fix poll() being unavailable on Mac OSX 10.2.
611
c1ac6996
PH
612PH/02 Reduce the amount of output that "make" produces by default. Full output
613 can still be requested.
614
9c7a242c
PH
615PH/03 The warning log line about a condition test deferring for a "warn" verb
616 was being output only once per connection, rather than after each
617 occurrence (because it was using the same function as for successful
618 "warn" verbs). This seems wrong, so I have changed it.
619
87ba3f5f
PH
620TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
621 it should not have, which might have caused a crash in the right
622 circumstances, but probably never did.
623
624PH/04 Installed a modified version of Tony Finch's patch to make submission
625 mode fix the return path as well as the Sender: header line, and to
626 add a /name= option so that you can make the user's friendly name appear
627 in the header line.
628
29aba418
TF
629TF/03 Added the control = fakedefer ACL modifier.
630
fe0dab11
TF
631TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
632 Mark Lowes for thorough testing.
870f6ba8 633
11d337a4
TK
634TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
635
636TK/03 Merged latest SRS patch from Miles Wilton.
637
415c8f3b
PH
638PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
639 with the definition in sysexits.h (which is #included earlier).
640 Fortunately, Exim does not actually use EX_OK. The code used to try to
641 preserve the sysexits.h value, by assumimg that macro definitions were
642 scanned for macro replacements. I have been disabused of this notion,
643 so now the code just undefines EX_OK before #including unistd.h.
11d337a4 644
958541e9
PH
645PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
646 in the smtp transport. When a block could not be written in a single
647 write() function, the timeout was being re-applied to each part-write.
648 This seems wrong - if the receiver was accepting one byte at a time it
649 would take for ever. The timeout is now adjusted when this happens. It
650 doesn't have to be particularly precise.
651
c206415f
TK
652TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
653 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
654
2a4be8f9
PH
655PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
656 without a domain if the check to <postmaster@domain> fails.
657
1cba11c5
SC
658SC/01 Eximstats: added -xls and the ability to specify output files
659 (patch written by Frank Heydlauf).
660
661SC/02 Eximstats: use FileHandles for outputing results.
662
663SC/03 Eximstats: allow any combination of xls, txt, and html output.
664
665SC/04 Eximstats: fixed display of large numbers with -nvr option
666
667SC/05 Eximstats: fixed merging of reports with empty tables.
668
669SC/06 Eximstats: added the -include_original_destination flag
670
671SC/07 Eximstats: removed tabs and trailing whitespace.
672
1005d00e
TK
673TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
674
675TK/06 MBOX spool code: Add real "From " MBOX separator line
676 so the .eml file is really in mbox format (even though
677 most programs do not really care). Patch from Alex Miller.
678
679TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
680 The latter is generated from $received_to and is only set if the
681 message has one envelope recipient. SA can use these headers,
682 obviously out-of-the-box. Patch from Alex Miller.
683
9b4768fa
PH
684PH/08 The ${def test on a variable was returning false if the variable's
685 value was "0", contrary to what the specification has always said!
686 The result should be true unless the variable is empty.
687
688PH/09 The syntax error of a character other than { following "${if
689 def:variable_name" (after optional whitespace) was not being diagnosed.
690 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
691 accidental colon was present, for example, could give incorrect results.
692
0d7eb84a
PH
693PH/10 Tidied the code in a number of places where the st_size field of a stat()
694 result is used (not including appendfile, where other changes are about
695 to be made).
696
697PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
698 This involved changing a lot of size variables from int to off_t. It
699 should work with maildirs and everything.
700
40727bee
TK
701TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
702 spamd dying while we are connected to it.
703
554d2369
TF
704TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
705 <jgh@wizmail.org>
706
1f922db1
PH
707PH/12 Applied Alex Kiernan's patch for the API change for the error callback
708 function for BDB 4.3.
709
ef213c3b
PH
710PH/13 Changed auto_thaw such that it does not apply to bounce messages.
711
8ac170f3
PH
712PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
713 the sources for PCRE have been re-arranged and more files are now
714 involved.
715
b1c749bb
PH
716PH/15 The code I had for printing potentially long long variables in PH/11
717 above was not the best (it lost precision). The length of off_t variables
718 is now inspected at build time, and an appropriate printing format (%ld
c6c2dc1d
PH
719 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
720 to be "long long int" or "long int". This is needed for the internal
721 formatting function string_vformat().
b1c749bb 722
4aac9b49
PH
723PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
724 the configuration file to be ":syslog", then the script "guesses" where
725 the logs files are, rather than using the compiled in default. In our
726 case the guess is not the same as the compiled default, so the script
727 suddenly stopped working when I started to use syslog. The patch checks
728 to see if log_file_path is "". If so, it attempts to read it from exim
729 with no configuration file to get the compiled in version, before it
730 falls back to the previous guessing code."
731
294520c8
TK
732TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
733 implementing BATV in an Exim configuration. See NewStuff for the gory
734 details.
735
5bd022fe
PH
736PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
737 Makefile that are specific to HP-UX.
738
90e9ce59
PH
739PH/18 If the "use_postmaster" option was set for a recipient callout together
740 with the "random" option, the postmaster address was used as the MAIL
741 FROM address for the random test, but not for the subsequent recipient
742 test. It is now used for both.
743
5ea81592
PH
744PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
745 patch removes a few documentation additions to RFC 3028, because the
746 latest draft now contains them. It adds the new en;ascii-case comparator
747 and a new error check for 8bit text in MIME parts. Comparator and
748 require names are now matched exactly. I enabled the subaddress
749 extension, but it is not well tested yet (read: it works for me)."
750
c6c2dc1d
PH
751PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
752 rework some of the code of TK/09 above to avoid the hardwired use of
753 "%lld" and "long long". Replaced the call to snprintf() with a call to
754 string_vformat().
755
fffffe4c
PH
756PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
757 records point to non-existent hosts", "retry timeout exceeded", and
758 "retry time not reached for any host after a long failure period".
ca02eafb 759
9a26b6b2
PH
760PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
761 experimental DomainKeys support:
762
763 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
764 (2) On an error such as an illegally used "control", the wrong name for
765 the control was given.
766
767 These problems did NOT occur unless DomainKeys support was compiled.
768
4aee0225
PH
769PH/23 Added daemon_startup_retries and daemon_startup_sleep.
770
32d668a5
PH
771PH/24 Added ${if match_ip condition.
772
8187c3f3
PH
773PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
774 databases so that it will be absolutely obvious if a crash occurs in the
775 DB library. This is a regular occurrence (often caused by mis-matched
776 db.h files).
777
ff790e47 778PH/26 Insert a lot of missing (void) casts for functions such as chown(),
f1e894f3
PH
779 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
780 picked up on a user's system that detects such things. There doesn't seem
781 to be a gcc warning option for this - only an attribute that has to be
782 put on the function's prototype. It seems that in Fedora Core 4 they have
783 set this on a number of new functions. No doubt there will be more in due
784 course.
ff790e47 785
5417f6d1
PH
786PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
787 specify a transport. However, if an address that was verified by such a
788 router was the subject of a callout, Exim crashed because it tried to
789 read the rcpt_include_affixes from the non-existent transport. Now it
790 just assumes that the setting of that option is false. This bug was
791 introduced by 4.51/PH/31.
792
59cf8544
PH
793PH/28 Changed -d+all to exclude +memory, because that information is very
794 rarely of interest, but it makes the output a lot bigger. People tend to
795 do -d+all out of habit.
796
e7ad8a65
PH
797PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
798 code in os-type was giving problems when libc.so lives in lib64, like on
799 x86_64 Fedora Core.
800
ade42478
PH
801PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
802 aren't the modern standard, and it seems that some systems' include files
803 don't always have them. Exim was already checking for some of the newer
804 ones like T_AAAA, and defining it itself. I've added checks for all the
805 record types that Exim uses.
806
182ad5cf
PH
807PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
808 not automatically generating a new one, as it is supposed to. This
809 prevented TLS from working. If the file did exist, but contained invalid
810 data, a new version was generated, as expected. It was only the case of a
811 non-existent file that was broken.
812
b0d9fc80
TK
813TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
814 with a change in libdomainkeys > 0.64.
815
816TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
817 from DNS. If the selector record carries the flag, it now has
818 precedence over the domain-wide flag.
819
820TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
821
47c7a64a
PH
822PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
823 the use of a port name that isn't defined in /etc/services) occurred, the
824 message was deferred as in a normal delivery, and thus remained on the
825 spool, instead of being failed because of the mua_wrapper setting. This
826 is now fixed, and I tidied up some of the mua_wrapper messages at the
827 same time.
828
a388bce4
SC
829SC/08 Eximstats: whilst parsing the mainlog(s), store information about
830 the messages in a hash of arrays rather than using individual hashes.
831 This is a bit cleaner and results in dramatic memory savings, albeit
832 at a slight CPU cost.
833
834SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
835 as requested by Marc Sherman.
836
837SC/10 Eximstats: added histograms for user specified patterns as requested
838 by Marc Sherman.
839
0793e4ed
SC
840SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
841
c58b88df
PH
842PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
843 fopen() in the content-scanning modules that did not already have it.
844
e7ad8a65 845
7982096b
PH
846Exim version 4.51
847-----------------
848
1a46a8c5
PH
849TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
850 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
851
2f079f46 852TK/02 Fix ACL "control" statement not being available in MIME ACL.
1a46a8c5
PH
853
854TK/03 Fix ACL "regex" condition not being available in MIME ACL.
855
856PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
857 to test Sieve filters that use "vacation".
858
859PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
860 that changes the way the GnuTLS parameters are stored in the cache file.
861 The new format can be generated externally. For backward compatibility,
862 if the data in the cache doesn't make sense, Exim assumes it has read an
863 old-format file, and it generates new data and writes a new file. This
864 means that you can't go back to an older release without removing the
865 file.
866
867PH/03 A redirect router that has both "unseen" and "one_time" set does not
868 work if there are any delivery delays because "one_time" forces the
869 parent to be marked "delivered", so its unseen clone is never tried
870 again. For this reason, Exim now forbids the simultaneous setting of
871 these two options.
872
873PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
874 redirected to themselves ("homonym" addresses). Read the long ChangeLog
875 entry if you want to know the details. The fix, however, neglected to
876 consider the case when local delivery batching is involved. The test for
877 "previously delivered" was not happening when checking to see if an
878 address could be batched with a previous (undelivered) one; under
879 certain circumstances this could lead to multiple deliveries to the same
c2c19e9d 880 address.
1a46a8c5
PH
881
882PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
883 in its include files, and this causes problems building Exim.
884
885PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
886 header_syntax) but Exim was just ignoring anything given after a slash.
887 In particular, this caused confusion with an attempt to use "verify =
888 reverse_host_lookup/defer_ok". An error is now given when options are
889 supplied for verify items that do not have them. (Maybe reverse_host_
890 lookup should have a defer_ok option, but that's a different point.)
891
892PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
893 defined by RFC 821) to 2048, because there were problems with some AUTH
894 commands, and RFC 1869 says the size should be increased for extended
895 SMTP commands that take arguments.
896
897PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
898 Finch).
899
900PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
901 "unknown" error; now it says that the functionality isn't in the binary.
8d67ada3 902
49c2d5ea
PH
903PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
904 an address' error message when a string expansion fails (syntax or
f331f3b6
PH
905 whatever). Otherwise the password may appear in the log. Following change
906 PH/42 below, there is no longer a chance of it appearing in a bounce
907 message.
49c2d5ea 908
bf759a8b
PH
909PH/11 Installed exipick version 20050225.0 from John Jetmore.
910
83364d30
PH
911PH/12 If the last host in a fallback_hosts list was multihomed, only the first
912 of its addresses was ever tried. (Bugzilla bug #2.)
913
7999bbd7
PH
914PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
915 the result incorrectly in the debug output. (It correctly added a newline
916 to what was transported.)
917
7dbf77c9
PH
918TF/01 Added $received_time.
919
74e0617f
PH
920PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
921 commented out examples of how to interface to a virus scanner and to
922 SpamAssassin. Also added commented examples of av_scanner and
923 spamd_address settings.
924
2f079f46
PH
925PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
926 and controls are allowed in which ACLs. There were a couple of minor
927 errors. Some of the entries in the conditions table (which is a table of
928 where they are NOT allowed) were getting very unwieldy; rewrote them as a
929 negation of where the condition IS allowed.
930
8c841523
PH
931PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
932
7766a4f0
PH
933PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
934 header file does not have a version number, so I've had to invent a new
935 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
936 API. The code is untested by me (my Linux distribution still has 0.3.2 of
937 radiusclient), but it was contributed by a Radius user.
938
8b417f2c
PH
939PH/18 Installed Lars Mainka's patch for the support of CRL collections in
940 files or directories, for OpenSSL.
941
901f42cb
PH
942PH/19 When an Exim process that is running as root has to create an Exim log
943 file, it does so in a subprocess that runs as exim:exim so as to get the
944 ownership right at creation (otherwise, other Exim processes might see
945 the file with the wrong ownership). There was no test for failure of this
946 fork() call, which would lead to the process getting stuck as it waited
947 for a non-existent subprocess. Forks do occasionally fail when resources
948 run out. I reviewed all the other calls to fork(); they all seem to check
949 for failure.
950
f9b9210e
PH
951PH/20 When checking for unexpected SMTP input at connect time (before writing
952 the banner), Exim was not dealing correctly with a non-positive return
953 from the read() function. If the client had disconnected by this time,
954 the result was a log entry for a synchronization error with an empty
955 string after "input=" when read() returned zero. If read() returned -1
956 (an event I could not check), uninitialized data bytes were printed.
957 There were reports of junk text (parts of files, etc) appearing after
958 "input=".
959
54cdb463
PH
960PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
961
cf00dad6
PH
962PH/22 Added support for macro redefinition, and (re)definition in between
963 driver and ACL definitions.
964
acb1b346
PH
965PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
966 forgetting to use the resulting value; it was using the unexpanded value.
967
c5ddb310
PH
968PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
969 hadn't been configured. The fix is from Juergen Kreileder, who
970 understands it better than I do:
971
972 "Here's what I see happening with three configured cyrus_sasl
973 authenticators configured (plain, login, cram-md5):
974
975 On startup auth_cyrus_sasl_init() gets called for each of these.
976 This means three calls to sasl_listmech() without a specified mech_list.
977 => SASL tests which mechs of all available mechs actually work
978 => three warnings about OTP not working
979 => the returned list contains: plain, login, cram-md5, digest-md5, ...
980
981 With the patch, sasl_listmech() also gets called three times. But now
982 SASL's mech_list option is set to the server_mech specified in the the
983 authenticator. Or in other words, the answer from sasl_listmech()
984 gets limited to just the mech you're testing for (which is different
985 for each call.)
986 => the return list contains just 'plain' or 'login', 'cram-md5' or
987 nothing depending on the value of ob->server_mech.
988
989 I've just tested the patch: Authentication still works fine,
990 unavailable mechs specified in the exim configuration are still
991 caught, and the auth.log warnings about OTP are gone."
992
31619da6
PH
993PH/25 When debugging is enabled, the contents of the command line are added
994 to the debugging output, even when log_selector=+arguments is not
995 specified.
996
bebaf0fc
PH
997PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
998 answer is "GNU", and only if the return is "GNU/something" is the answer
999 "Linux".
1000
475fe28a
PH
1001PH/27 $acl_verify_message is now set immediately after the failure of a
1002 verification in an ACL, and so is available in subsequent modifiers. In
1003 particular, the message can be preserved by coding like this:
1004
1005 warn !verify = sender
1006 set acl_m0 = $acl_verify_message
1007
1008 Previously, $acl_verify_message was set only while expanding "message"
1009 and "log_message" when a very denied access.
1010
7e8bec7a
PH
1011PH/28 Modified OS/os.c-Linux with
1012
1013 -#ifndef OS_LOAD_AVERAGE
1014 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
1015
1016 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
1017 nomenclature these days.)
1018
e4a89c47
PH
1019PH/29 Installed patch from the Sieve maintainer that adds the options
1020 sieve_useraddress and sieve_subaddress to the redirect router.
1021
5ca2a9a1
PH
1022PH/30 In these circumstances:
1023 . Two addresses routed to the same list of hosts;
1024 . First host does not offer TLS;
1025 . First host accepts first address;
1026 . First host gives temporary error to second address;
1027 . Second host offers TLS and a TLS session is established;
1028 . Second host accepts second address.
1029 Exim incorrectly logged both deliveries with the TLS parameters (cipher
1030 and peerdn, if requested) that were in fact used only for the second
1031 address.
7e8bec7a 1032
c688b954
PH
1033PH/31 When doing a callout as part of verifying an address, Exim was not paying
1034 attention to any local part prefix or suffix that was matched by the
1035 router that accepted the address. It now behaves in the same way as it
1036 does for delivery: the affixes are removed from the local part unless
1037 rcpt_include_affixes is set on the transport.
1038
fed77020
PH
1039PH/32 Add the sender address, as F=<...>, to the log line when logging a
1040 timeout during the DATA phase of an incoming message.
1041
7fe1560f
PH
1042PH/33 Sieve envelope tests were broken for match types other than :is. I have
1043 applied a patch sanctioned by the Sieve maintainer.
c688b954 1044
ebb6e6d5
PH
1045PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
1046 the uid or gid is negative. A case of a negative gid caused this to be
1047 noticed. The fix allows for either to be negative.
1048
9c4e8f60
PH
1049PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
1050 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
1051 overlooked.
1052
1053PH/36 The change PH/12 above was broken. Fixed it.
1054
d7174846
PH
1055PH/37 Exim used to check for duplicate addresses in the middle of routing, on
1056 the grounds that routing the same address twice would always produce the
1057 same answer. This might have been true once, but it is certainly no
1058 longer true now. Routing a child address may depend on the previous
1059 routing that produced that child. Some complicated redirection strategies
1060 went wrong when messages had multiple recipients, and made Exim's
1061 behaviour dependent on the order in which the addresses were given.
1062
1063 I have moved the duplicate checking until after the routing is complete.
1064 Exim scans the addresses that are assigned to local and remote
1065 transports, and removes any duplicates. This means that more work will be
1066 done, as duplicates will always all be routed, but duplicates are
1067 presumably rare, so I don't expect this is of any significance.
1068
1069 For deliveries to pipes, files, and autoreplies, the duplicate checking
1070 still happens during the routing process, since they are not going to be
1071 routed further.
1072
cfe75fc3
PH
1073PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
1074 It corrects a timeout issue with spamd. This is Ian's comment: "The
1075 background is that sometimes spamd either never reads data from a
1076 connection it has accepted, or it never writes response data. The exiscan
1077 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
1078 blindly assumes that writes won't block so it may never time out."
1079
be22d70e
PH
1080PH/39 Allow G after quota size as well as K and M.
1081
0612b098
PH
1082PH/40 The value set for $authenticated_id in an authenticator may not contain
1083 binary zeroes or newlines because the value is written to log lines and
1084 to spool files. There was no check on this. Now the value is run through
1085 the string_printing() function so that such characters are converted to
1086 printable escape sequences.
1087
2e0c1448
PH
1088PH/41 $message_linecount is a new variable that contains the total number of
1089 lines in the message. Compare $body_linecount, which is the count for the
1090 body only.
1091
447d236c
PH
1092PH/42 Exim no longer gives details of delivery errors for specific addresses in
1093 bounce and delay warning messages, except in certain special cases, which
1094 are as follows:
1095
1096 (a) An SMTP error message from a remote host;
1097 (b) A message specified in a :fail: redirection;
1098 (c) A message specified in a "fail" command in a system filter;
1099 (d) A message specified in a FAIL return from the queryprogram router;
1100 (e) A message specified by the cannot_route_message router option.
1101
1102 In these cases only, Exim does include the error details in bounce and
1103 warning messages. There are also a few cases where bland messages such
1104 as "unrouteable address" or "local delivery error" are given.
1105
d20976dc
PH
1106PH/43 $value is now also set for the "else" part of a ${run expansion.
1107
f656d135
PH
1108PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
1109 being worked on, but at least Exim now implements the latest version to
1110 play with."
1111
2e2a30b4
PH
1112PH/45 In a pipe transport, although a timeout while waiting for the pipe
1113 process to complete was treated as a delivery failure, a timeout while
1114 writing the message to the pipe was logged, but erroneously treated as a
1115 successful delivery. Such timeouts include transport filter timeouts. For
1116 consistency with the overall process timeout, these timeouts are now
1117 treated as errors, giving rise to delivery failures by default. However,
1118 there is now a new Boolean option for the pipe transport called
1119 timeout_defer, which, if set TRUE, converts the failures into defers for
1120 both kinds of timeout. A transport filter timeout is now identified in
1121 the log output.
1122
9176e9f0
PH
1123PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
1124 systems where "make" and "gmake" are different, calling "gmake" at top
1125 level broke things. I've arranged for the value of $(MAKE) to be passed
1126 from the Makefile to this script so that it can call the same version of
1127 "make".
1128
7982096b 1129
bbe902f0
PH
1130A note about Exim versions 4.44 and 4.50
1131----------------------------------------
1132
1133Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
1134changes of various kinds. As a consequence, a big documentation update was
1135needed. This delayed the release for rather longer than seemed good, especially
1136in the light of a couple of (minor) security issues. Therefore, the changes
1137that fixed bugs were backported into 4.43, to create a 4.44 maintenance
1138release. So 4.44 and 4.50 are in effect two different branches that both start
1139from 4.43.
1140
1141I have left the 4.50 change log unchanged; it contains all the changes since
11424.43. The change log for 4.44 is below; many of its items are identical to
1143those for 4.50. This seems to be the most sensible way to preserve the
1144historical information.
1145
1146
f7b63901 1147Exim version 4.50
495ae4b0
PH
1148-----------------
1149
5fe762f6
PH
1150 1. Minor wording change to the doc/README.SIEVE file.
1151
139059f6 1152 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
5fe762f6 1153 computation of the current number of files was incorrect.
495ae4b0 1154
7086e875
PH
1155 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
1156 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
1157 place.
1158
35af9f61
PH
1159 4. Give more explanation in the error message when the command for a transport
1160 filter fails to execute.
1161
b668c215
PH
1162 5. There are several places where Exim runs a non-Exim command in a
1163 subprocess. The SIGUSR1 signal should be disabled for these processes. This
1164 was being done only for the command run by the queryprogram router. It is
1165 now done for all such subprocesses. The other cases are: ${run, transport
1166 filters, and the commands run by the lmtp and pipe transports.
1167
a494b1e1
PH
1168 6. Added CONFIGURE_GROUP build-time option.
1169
1170 7. Some older OS have a limit of 256 on the maximum number of file
1171 descriptors. Exim was using setrlimit() to set 1000 as a large value
1172 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
1173 systems. I've change it so that if it can't get 1000, it tries for 256.
35edf2ff 1174
c5fcb476
PH
1175 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
1176 was an oversight, and furthermore, ever since the addition of extra
1177 controls (e.g. 4.43/32), the checks on when to allow different forms of
1178 "control" were broken. There should now be diagnostics for all cases when a
1179 control that does not make sense is encountered.
1180
69358f02
PH
1181 9. Added the /retain_sender option to "control=submission".
1182
5be20824
PH
118310. $recipients is now available in the predata ACL (oversight).
1184
eb2c0248
PH
118511. Tidy the search cache before the fork to do a delivery from a message
1186 received from the command line. Otherwise the child will trigger a lookup
1187 failure and thereby defer the delivery if it tries to use (for example) a
1188 cached ldap connection that the parent has called unbind on.
1189
2a3eea10
PH
119012. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
1191 of $address_data from the recipient verification was clobbered by the
1192 sender verification.
1193
119413. The value of address_data from a sender verification is now available in
1195 $sender_address_data in subsequent conditions in the ACL statement.
1196
23c7ff99
PH
119714. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
1198
4deaf07d
PH
119915. Added a new option "connect=<time>" to callout options, to set a different
1200 connection timeout.
1201
926e1192
PH
120216. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
1203 was its contents. (It was OK if the option was not defined at all.)
1204
650edc6f
PH
120517. A "Completed" log line is now written for messages that are removed from
1206 the spool by the -Mrm option.
1207
2c7db3f5
PH
120818. New variables $sender_verify_failure and $recipient_verify_failure contain
1209 information about exactly what failed.
1210
3d235903
PH
121119. Added -dd to debug only the daemon process.
1212
7c7ad977
PH
121320. Incorporated Michael Haardt's patch to ldap.c for improving the way it
1214 handles timeouts, both on the server side and network timeouts. Renamed the
1215 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
1216
981756db
PH
121721. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
1218 It is now set to "smtps".
1219
d4eb88df
PH
122022. $host_address is now set to the target address during the checking of
1221 ignore_target_hosts.
1222
122323. When checking ignore_target_hosts for an ipliteral router, no host name was
1224 being passed; this would have caused $sender_host_name to have been used if
1225 matching the list had actually called for a host name (not very likely,
1226 since this list is usually IP addresses). A host name is now passed as
1227 "[x.x.x.x]".
1228
7d468ab8
PH
122924. Changed the calls that set up the SIGCHLD handler in the daemon to use the
1230 code that specifies a non-restarting handler (typically sigaction() in
1231 modern systems) in an attempt to fix a rare and obscure crash bug.
1232
123325. Narrowed the window for a race in the daemon that could cause it to ignore
1234 SIGCHLD signals. This is not a major problem, because they are used only to
1235 wake it up if nothing else does.
1236
62c0818f
PH
123726. A malformed maildirsize file could cause Exim to calculate negative values
1238 for the mailbox size or file count. Odd effects could occur as a result.
1239 The maildirsize information is now recalculated if the size or filecount
1240 end up negative.
1241
26034054
PH
124227. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
1243 support for a long time. Removed HAVE_SYS_VFS_H.
1244
af66f652
PH
124528. Installed the latest version of exipick from John Jetmore.
1246
90af77f4
PH
124729. In an address list, if the pattern was not a regular expression, an empty
1248 subject address (from a bounce message) matched only if the pattern was an
1249 empty string. Non-empty patterns were not even tested. This was the wrong
1250 because it is perfectly reasonable to use an empty address as part of a
1251 database query. An empty address is now tested by patterns that are
1252 lookups. However, all the other forms of pattern expect the subject to
1253 contain a local part and a domain, and therefore, for them, an empty
1254 address still always fails if the pattern is not itself empty.
1255
d8ef3577
PH
125630. Exim went into a mad DNS loop when attempting to do a callout where the
1257 host was specified on an smtp transport, and looking it up yielded more
1258 than one IP address.
1259
5cb8cbc6
PH
126031. Re-factored the code for checking spool and log partition space into a
1261 function that finds that data and another that does the check. The former
1262 is then used to implement four new variables: $spool_space, $log_space,
1263 $spool_inodes, and $log_inodes.
1264
14702f5b
PH
126532. The RFC2047 encoding function was originally intended for short strings
1266 such as real names; it was not keeping to the 75-character limit for
1267 encoded words that the RFC imposes. It now respects the limit, and
1268 generates multiple encoded words if necessary. To be on the safe side, I
1269 have increased the buffer size for the ${rfc2047: expansion operator from
1270 1024 to 2048 bytes.
1271
063b1e99
PH
127233. It is now permitted to omit both strings after an "if" condition; if the
1273 condition is true, the result is "true". As before, when the second string
1274 is omitted, a false condition yields an empty string. This makes it less
1275 cumbersome to write custom ACL and router conditions.
1276
652e1b65
PH
127734. Failure to deliver a bounce message always caused it to be frozen, even if
1278 there was an errors_to setting on the router. The errors_to setting is now
1279 respected.
1280
6f0c9a4f
PH
128135. If an IPv6 address is given for -bh or -bhc, it is now converted to the
1282 canonical form (fully expanded) before being placed in
1283 $sender_host_address.
1284
33397d19
PH
128536. The table in the code that translates DNS record types into text (T_A to
1286 "A" for instance) was missing entries for NS and CNAME. It is just possible
1287 that this could have caused confusion if both these types were looked up
1288 for the same domain, because the text type is used as part of Exim's
1289 per-process caching. But the chance of anyone hitting this buglet seems
1290 very small.
1291
7bb56e1f
PH
129237. The dnsdb lookup has been extended in a number of ways.
1293
1294 (1) There is a new type, "zns", which walks up the domain tree until it
1295 finds some nameserver records. It should be used with care.
1296
ea3bc19b
PH
1297 (2) There is a new type, "mxh", which is like "mx" except that it returns
1298 just the host names, not the priorities.
1299
1300 (3) It is now possible to give a list of domains (or IP addresses) to be
ff4dbb19
PH
1301 looked up. The behaviour when one of the lookups defers can be
1302 controlled by a keyword.
7bb56e1f 1303
ea3bc19b 1304 (4) It is now possible to specify the separator character for use when
7bb56e1f 1305 multiple records are returned.
33397d19 1306
0bcb2a0e
PH
130738. The dnslists ACL condition has been extended: it is now possible to supply
1308 a list of IP addresses and/or domains to be looked up in a particular DNS
1309 domain.
1310
2ac0e484
PH
131139. Added log_selector=+queue_time_overall.
1312
4e1fde53
PH
131340. When running the queue in the test harness, wait just a tad after forking a
1314 delivery process, to get repeatability of debugging output.
1315
de365ded
PH
131641. Include certificate and key file names in error message when GnuTLS fails
1317 to set them up, because the GnuTLS error message doesn't include the name
1318 of the failing file when there is a problem reading it.
1319
f05da2e8
PH
132042. Allow both -bf and -bF in the same test run.
1321
d6453af2
PH
132243. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
1323
f7b63901
PH
132444. The "Exiscan patch" is now merged into the mainline Exim source.
1325
132645. Sometimes the final signoff response after QUIT could fail to get
1327 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
1328 < 0 before doing a fflush(). This bug looks as though it goes back to the
1329 introduction of TLS in release 3.20, but "sometimes" must have been rare
1330 because the tests only now provoked it.
1331
a444213a
PH
133246. Reset the locale to "C" after calling embedded Perl, in case it was changed
1333 (this can affect the format of dates).
1334
0ec020ea
PH
133547. exim_tidydb, when checking for the continued existence of a message for
1336 which it has found a message-specific retry record, was not finding
1337 messages that were in split spool directories. Consequently, it was
1338 deleting retry records that should have stayed in existence.
1339
b1206957
PH
134048. Steve fixed some bugs in eximstats.
1341
134249. The SPA authentication driver was not abandoning authentication and moving
1343 on to the next authenticator when an expansion was forced to fail,
1344 contradicting the general specification for all authenticators. Instead it
1345 was generating a temporary error. It now behaves as specified.
1346
26dd5a95
PH
134750. The default ordering of permitted cipher suites for GnuTLS was pessimal
1348 (the order specifies the preference for clients). The order is now AES256,
1349 AES128, 3DES, ARCFOUR128.
1350
343b2385
PH
135151. Small patch to Sieve code - explicitly set From: when generating an
1352 autoreply.
1353
1c5466b9
PH
135452. Exim crashed if a remote delivery caused a very long error message to be
1355 recorded - for instance if somebody sent an entire SpamAssassin report back
1356 as a large number of 550 error lines. This bug was coincidentally fixed by
1357 increasing the size of one of Exim's internal buffers (big_buffer) that
1358 happened as part of the Exiscan merge. However, to be on the safe side, I
1359 have made the code more robust (and fixed the comments that describe what
1360 is going on).
1361
55ee9ee3
PH
136253. Now that there can be additional text after "Completed" in log lines (if
1363 the queue_time_overall log selector is set), a one-byte patch to exigrep
1364 was needed to allow it to recognize "Completed" as not the last thing in
1365 the line.
1366
d38f8232
PH
136754. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
1368 patch that reportedly fixes this has been added. I am not expert enough to
1369 create a test for it. This is what the patch creator wrote:
1370
1371 "I found a little strange behaviour of ldap code when working with
1372 Windows 2003 AD Domain, where users was placed in more than one
1373 Organization Units. When I tried to give exim partial DN, the exit code
1374 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
1375 But simultaneously result of request was absolutely normal ldap result,
1376 so I produce this patch..."
1377
3295e65b
PH
1378 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
1379 REFERENCE, so I have modified the code to exclude the patch when that macro
1380 is not defined.
1381
7102e136
PH
138255. Some experimental protocols are using DNS PTR records for new purposes. The
1383 keys for these records are domain names, not reversed IP addresses. The
b975ba52
PH
1384 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
1385 leaves it alone. Component reversal etc. now happens only for IP addresses.
ea3a6f44 1386 CAN-2005-0021
7102e136 1387
3ca0ba97
PH
138856. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
1389
c2bcbe20
PH
139057. Double the size of the debug message buffer (to 2048) so that more of very
1391 long debug lines gets shown.
1392
18ce445d
PH
139358. The exicyclog utility now does better if the number of log files to keep
1394 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
1395
1f5b4c3d
PH
139659. Two changes related to the smtp_active_hostname option:
1397
1398 (1) $smtp_active_hostname is now available as a variable.
1399 (2) The default for smtp_banner uses $smtp_active_hostname instead
1400 of $primary_hostname.
1401
b975ba52
PH
140260. The host_aton() function is supposed to be passed a string that is known
1403 to be a valid IP address. However, in the case of IPv6 addresses, it was
1404 not checking this. This is a hostage to fortune. Exim now panics and dies
1405 if the condition is not met. A case was found where this could be provoked
85b87bc2
PH
1406 from a dnsdb PTR lookup with an IPv6 address that had more than 8
1407 components; fortuitously, this particular loophole had already been fixed
1408 by change 4.50/55 above.
1409
1410 If there are any other similar loopholes, the new check in host_aton()
1411 itself should stop them being exploited. The report I received stated that
1412 data on the command line could provoke the exploit when Exim was running as
1413 exim, but did not say which command line option was involved. All I could
1414 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
1415 running as the user.
ea3a6f44 1416 CAN-2005-0021
85b87bc2
PH
1417
141861. There was a buffer overflow vulnerability in the SPA authentication code
1419 (which came originally from the Samba project). I have added a test to the
1420 spa_base64_to_bits() function which I hope fixes it.
ea3a6f44 1421 CAN-2005-0022
b975ba52 1422
17ffcae7
PH
142362. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
1424 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
1425
d95f9fdb
PH
142663. The daemon start-up calls getloadavg() while still root for those OS that
1427 need the first call to be done as root, but it missed one case: when
1428 deliver_queue_load_max is set with deliver_drop_privilege. This is
1429 necessary for the benefit of the queue runner, because there is no re-exec
1430 when deliver_drop_privilege is set.
1431
86b8287f
PH
143264. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
1433 This has been fixed.
1434
60dc5e56
PH
143565. Caching of lookup data for "hosts =" ACL conditions, when a named host list
1436 was in use, was not putting the data itself into the right store pool;
1437 consequently, it could be overwritten for a subsequent message in the same
1438 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
1439 the caching.)
1440
533244af
PH
144166. Added hosts_max_try_hardlimit to the smtp transport, default 50.
1442
a5a28604
PH
144367. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
1444 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
1445 the function were treating the return as a boolean value, which happened to
1446 work because 0=false and not-0=true, but is not correct code.
1447
7e634d24
PH
144868. The host_aton() function was not handling scoped IPv6 addresses (those
1449 with, for example, "%eth0" on the end) correctly.
1450
3e11c26b
PH
145169. Fixed some compiler warnings in acl.c for the bitmaps specified with
1452 negated items (that is, ~something) in unsigned ints. Some compilers
1453 apparently mutter when there is no cast.
1454
6729cf78
PH
145570. If an address verification called from an ACL failed, and did not produce a
1456 user-specific message (i.e. there was only a "system" message), nothing was
1457 put in $acl_verify_message. In this situation, it now puts the system
1458 message there.
1459
00f00ca5
PH
146071. Change 4.23/11 added synchronization checking at the start of an SMTP
1461 session; change 4.31/43 added the unwanted input to the log line - except
1462 that it did not do this in the start of session case. It now does.
1463
c9bdd01c
PH
146472. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
1465 This is wrong and can cause the other end to generate a synchronization
1466 error if it is another Exim or anything else that does the synchronization
1467 check. A QUIT command is no longer sent after a timeout.
1468
d43194df
PH
146973. $host_lookup_deferred has been added, to make it easier to detect DEFERs
1470 during host lookups.
1471
fe5b5d0b
PH
147274. The defer_ok option of callout verification was not working if it was used
1473 when verifying addresses in header lines, that is, for this case:
1474
1475 verify = header_sender/callout=defer_ok
1476
76a2d7ba
PH
147775. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
1478 those file descriptors could be used for SMTP connections. If anything
1479 wrote to stderr (the example that came up was "warn" in embedded Perl), it
1480 could be sent to the SMTP client, causing chaos. The daemon now opens
1481 stdin, stdout, and stderr to /dev/null when it puts itself into the
1482 background.
1483
148476. Arrange for output from Perl's "warn" command to be written to Exim's main
1485 log by default. The user can override this with suitable Perl magic.
1486
04f7d5b9
PH
148777. The use of log_message on a "discard" ACL verb, which is supposed to add to
1488 the log message when discard triggers, was not working for the DATA ACL or
1489 for the non-SMTP ACL.
1490
bc60667e
PH
149178. Error message wording change in sieve.c.
1492
bb6e88ff
PH
149379. If smtp_accept_max_per_host was set, the number of connections could be
1494 restricted to fewer than expected, because the daemon was trying to set up
1495 a new connection before checking whether the processes handling previous
1496 connections had finished. The check for completed processes is now done
1497 earlier. On busy systems, this bug wouldn't be noticed because something
1498 else would have woken the daemon, and it would have reaped the completed
1499 process earlier.
1500
1e70f85b
PH
150180. If a message was submitted locally by a user whose login name contained one
1502 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
1503 It caused a spool format error. I have fixed the spool reading code. A
1504 related problem was that the "from" clause in the Received: line became
1505 illegal because of the space(s). It is now covered by ${quote_local_part.
1506
150781. Included the latest eximstats from Steve (adds average sizes to HTML Top
1508 tables).
1509
4e01f9d6
PH
151082. Updated OS/Makefile-AIX as per message from Mike Meredith.
1511
1ee1cef2
PH
151283. Patch from Sieve maintainer to fix unterminated string problem in
1513 "vacation" handling.
1514
6e2b4ccc
PH
151584. Some minor changes to the Linux configuration files to help with other
1516 OS variants using glibc.
1517
8e669ac1
PH
151885. One more patch for Sieve to update vacation handling to latest spec.
1519
495ae4b0 1520
bbe902f0
PH
1521----------------------------------------------------
1522See the note above about the 4.44 and 4.50 releases.
1523----------------------------------------------------
1524
1525
1526Exim version 4.44
1527-----------------
1528
1529 1. Change 4.43/35 introduced a bug that caused file counts to be
1530 incorrectly computed when quota_filecount was set in an appendfile
1531 transport
1532
1533 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
1534 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
1535 place.
1536
1537 3. Give more explanation in the error message when the command for a transport
1538 filter fails to execute.
1539
1540 4. There are several places where Exim runs a non-Exim command in a
1541 subprocess. The SIGUSR1 signal should be disabled for these processes. This
1542 was being done only for the command run by the queryprogram router. It is
1543 now done for all such subprocesses. The other cases are: ${run, transport
1544 filters, and the commands run by the lmtp and pipe transports.
1545
1546 5. Some older OS have a limit of 256 on the maximum number of file
1547 descriptors. Exim was using setrlimit() to set 1000 as a large value
1548 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
1549 systems. I've change it so that if it can't get 1000, it tries for 256.
1550
1551 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
1552 was an oversight, and furthermore, ever since the addition of extra
1553 controls (e.g. 4.43/32), the checks on when to allow different forms of
1554 "control" were broken. There should now be diagnostics for all cases when a
1555 control that does not make sense is encountered.
1556
1557 7. $recipients is now available in the predata ACL (oversight).
1558
1559 8. Tidy the search cache before the fork to do a delivery from a message
1560 received from the command line. Otherwise the child will trigger a lookup
1561 failure and thereby defer the delivery if it tries to use (for example) a
1562 cached ldap connection that the parent has called unbind on.
1563
1564 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
1565 of $address_data from the recipient verification was clobbered by the
1566 sender verification.
1567
156810. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
1569 was its contents. (It was OK if the option was not defined at all.)
1570
157111. A "Completed" log line is now written for messages that are removed from
1572 the spool by the -Mrm option.
1573
157412. $host_address is now set to the target address during the checking of
1575 ignore_target_hosts.
1576
157713. When checking ignore_target_hosts for an ipliteral router, no host name was
1578 being passed; this would have caused $sender_host_name to have been used if
1579 matching the list had actually called for a host name (not very likely,
1580 since this list is usually IP addresses). A host name is now passed as
1581 "[x.x.x.x]".
1582
158314. Changed the calls that set up the SIGCHLD handler in the daemon to use the
1584 code that specifies a non-restarting handler (typically sigaction() in
1585 modern systems) in an attempt to fix a rare and obscure crash bug.
1586
158715. Narrowed the window for a race in the daemon that could cause it to ignore
1588 SIGCHLD signals. This is not a major problem, because they are used only to
1589 wake it up if nothing else does.
1590
159116. A malformed maildirsize file could cause Exim to calculate negative values
1592 for the mailbox size or file count. Odd effects could occur as a result.
1593 The maildirsize information is now recalculated if the size or filecount
1594 end up negative.
1595
159617. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
1597 support for a long time. Removed HAVE_SYS_VFS_H.
1598
ea3a6f44 159918. Updated exipick to current release from John Jetmore.
bbe902f0
PH
1600
160119. Allow an empty sender to be matched against a lookup in an address list.
1602 Previously the only cases considered were a regular expression, or an
1603 empty pattern.
1604
160520. Exim went into a mad DNS lookup loop when doing a callout where the
1606 host was specified on the transport, if the DNS lookup yielded more than
1607 one IP address.
1608
ea3a6f44
NM
160921. The RFC2047 encoding function was originally intended for short strings
1610 such as real names; it was not keeping to the 75-character limit for
1611 encoded words that the RFC imposes. It now respects the limit, and
1612 generates multiple encoded words if necessary. To be on the safe side, I
1613 have increased the buffer size for the ${rfc2047: expansion operator from
1614 1024 to 2048 bytes.
bbe902f0 1615
ea3a6f44
NM
161622. Failure to deliver a bounce message always caused it to be frozen, even if
1617 there was an errors_to setting on the router. The errors_to setting is now
1618 respected.
bbe902f0
PH
1619
162023. If an IPv6 address is given for -bh or -bhc, it is now converted to the
1621 canonical form (fully expanded) before being placed in
1622 $sender_host_address.
1623
162424. Updated eximstats to version 1.33
1625
ea3a6f44
NM
162625. Include certificate and key file names in error message when GnuTLS fails
1627 to set them up, because the GnuTLS error message doesn't include the name
1628 of the failing file when there is a problem reading it.
bbe902f0
PH
1629
163026. Expand error message when OpenSSL has problems setting up cert/key files.
ea3a6f44 1631 As per change 25.
bbe902f0 1632
ea3a6f44
NM
163327. Reset the locale to "C" after calling embedded Perl, in case it was changed
1634 (this can affect the format of dates).
bbe902f0 1635
ea3a6f44
NM
163628. exim_tidydb, when checking for the continued existence of a message for
1637 which it has found a message-specific retry record, was not finding
1638 messages that were in split spool directories. Consequently, it was
1639 deleting retry records that should have stayed in existence.
bbe902f0
PH
1640
164129. eximstats updated to version 1.35
1642 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
1643 1.35 - bugfix such that pie charts by volume are generated correctly
1644
ea3a6f44
NM
164530. The SPA authentication driver was not abandoning authentication and moving
1646 on to the next authenticator when an expansion was forced to fail,
1647 contradicting the general specification for all authenticators. Instead it
1648 was generating a temporary error. It now behaves as specified.
bbe902f0 1649
ea3a6f44
NM
165031. The default ordering of permitted cipher suites for GnuTLS was pessimal
1651 (the order specifies the preference for clients). The order is now AES256,
1652 AES128, 3DES, ARCFOUR128.
bbe902f0 1653
ea3a6f44
NM
165431. Small patch to Sieve code - explicitly set From: when generating an
1655 autoreply.
bbe902f0 1656
ea3a6f44
NM
165732. Exim crashed if a remote delivery caused a very long error message to be
1658 recorded - for instance if somebody sent an entire SpamAssassin report back
1659 as a large number of 550 error lines. This bug was coincidentally fixed by
1660 increasing the size of one of Exim's internal buffers (big_buffer) that
1661 happened as part of the Exiscan merge. However, to be on the safe side, I
1662 have made the code more robust (and fixed the comments that describe what
1663 is going on).
bbe902f0 1664
ea3a6f44
NM
166533. Some experimental protocols are using DNS PTR records for new purposes. The
1666 keys for these records are domain names, not reversed IP addresses. The
1667 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
1668 leaves it alone. Component reversal etc. now happens only for IP addresses.
bbe902f0
PH
1669 CAN-2005-0021
1670
ea3a6f44
NM
167134. The host_aton() function is supposed to be passed a string that is known
1672 to be a valid IP address. However, in the case of IPv6 addresses, it was
1673 not checking this. This is a hostage to fortune. Exim now panics and dies
1674 if the condition is not met. A case was found where this could be provoked
1675 from a dnsdb PTR lookup with an IPv6 address that had more than 8
1676 components; fortuitously, this particular loophole had already been fixed
1677 by change 4.50/55 or 4.44/33 above.
1678
1679 If there are any other similar loopholes, the new check in host_aton()
1680 itself should stop them being exploited. The report I received stated that
1681 data on the command line could provoke the exploit when Exim was running as
1682 exim, but did not say which command line option was involved. All I could
1683 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
1684 running as the user.
bbe902f0
PH
1685 CAN-2005-0021
1686
ea3a6f44
NM
168735. There was a buffer overflow vulnerability in the SPA authentication code
1688 (which came originally from the Samba project). I have added a test to the
1689 spa_base64_to_bits() function which I hope fixes it.
bbe902f0
PH
1690 CAN-2005-0022
1691
ea3a6f44
NM
169236. The daemon start-up calls getloadavg() while still root for those OS that
1693 need the first call to be done as root, but it missed one case: when
1694 deliver_queue_load_max is set with deliver_drop_privilege. This is
1695 necessary for the benefit of the queue runner, because there is no re-exec
1696 when deliver_drop_privilege is set.
bbe902f0 1697
ea3a6f44
NM
169837. Caching of lookup data for "hosts =" ACL conditions, when a named host list
1699 was in use, was not putting the data itself into the right store pool;
1700 consequently, it could be overwritten for a subsequent message in the same
1701 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
1702 the caching.)
bbe902f0 1703
ea3a6f44
NM
170438. Sometimes the final signoff response after QUIT could fail to get
1705 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
1706 < 0 before doing a fflush(). This bug looks as though it goes back to the
1707 introduction of TLS in release 3.20, but "sometimes" must have been rare
1708 because the tests only now provoked it.
bbe902f0
PH
1709
1710
495ae4b0
PH
1711Exim version 4.43
1712-----------------
1713
1714 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
1715 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
1716 Now it returns an integer. A number of calls were still expecting a T/F
1717 return. Fortuitously, in all cases, the tests worked in OK situations,
1718 which is the norm. However, things would have gone wrong on any write
1719 failures on the smtp file descriptor. This function is used when sending
1720 messages over SMTP and also when doing verify callouts.
1721
1722 2. When Exim is called to do synchronous delivery of a locally submitted
1723 message (the -odf or -odi options), it no longer closes stderr before doing
1724 the delivery.
1725
1726 3. Implemented the mua_wrapper option.
1727
1728 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
1729
1730 5. Implemented the functions header_remove(), header_testname(),
1731 header_add_at_position(), and receive_remove_recipient(), and exported them
1732 to local_scan().
1733
1734 6. If an ACL "warn" statement specified the addition of headers, Exim already
1735 inserted X-ACL-Warn: at the start if there was no header name. However, it
1736 was not making this test for the second and subsequent header lines if
1737 there were newlines in the string. This meant that an invalid header could
1738 be inserted if Exim was badly configured.
1739
1740 7. Allow an ACL "warn" statement to add header lines at the start or after all
1741 the Received: headers, as well as at the end.
1742
1743 8. Added the rcpt_4xx retry error code.
1744
1745 9. Added postmaster_mailfrom=xxx to callout verification option.
1746
174710. Added mailfrom=xxxx to the callout verification option, for verify=
1748 header_sender only.
1749
175011. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
1751 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
1752
175312. Inserted some casts to stop certain compilers warning when using pointer
1754 differences as field lengths or precisions in printf-type calls (mostly
1755 affecting debugging statements).
1756
175713. Added optional readline() support for -be (dynamically loaded).
1758
175914. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
1760 same clock tick as a message's arrival, so that its received time was the
1761 same as the "first fail" time on the retry record, and that message
1762 remained on the queue past the ultimate address timeout, every queue runner
1763 would try a delivery (because it was past the ultimate address timeout) but
1764 after another failure, the ultimate address timeout, which should have then
1765 bounced the address, did not kick in. This was a "< instead of <=" error;
1766 in most cases the first failure would have been in the next clock tick
1767 after the received time, and all would be well.
1768
176915. The special items beginning with @ in domain lists (e.g. @mx_any) were not
1770 being recognized when the domain list was tested by the match_domain
1771 condition in an expansion string.
1772
177316. Added the ${str2b64: operator.
1774
177517. Exim was always calling setrlimit() to set a large limit for the number of
1776 processes, without checking whether the existing limit was already
1777 adequate. (It did check for the limit on file descriptors.) Furthermore,
1778 errors from getrlimit() and setrlimit() were being ignored. Now they are
1779 logged to the main and panic logs, but Exim does carry on, to try to do its
1780 job under whatever limits there are.
1781
178218. Imported PCRE 5.0.
1783
178419. Trivial typo in log message " temporarily refused connection" (the leading
1785 space).
1786
178720. If the log selector return_path_on_delivery was set and an address was
1788 redirected to /dev/null, the delivery process crashed because it assumed
1789 that a return path would always be set for a "successful" delivery. In this
1790 case, the whole delivery is bypassed as an optimization, and therefore no
1791 return path is set.
1792
179321. Internal re-arrangement: the function for sending a challenge and reading
1794 a response while authentication was assuming a zero-terminated challenge
1795 string. It's now changed to take a pointer and a length, to allow for
1796 binary data in such strings.
1797
179822. Added the cyrus_sasl authenticator (code supplied by MBM).
1799
180023. Exim was not respecting finduser_retries when seeking the login of the
1801 uid under which it was called; it was always trying 10 times. (The default
1802 setting of finduser_retries is zero.) Also, it was sleeping after the final
1803 failure, which is pointless.
1804
180524. Implemented tls_on_connect_ports.
1806
180725. Implemented acl_smtp_predata.
1808
180926. If the domain in control=submission is set empty, Exim assumes that the
1810 authenticated id is a complete email address when it generates From: or
1811 Sender: header lines.
1812
181327. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
1814 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
1815 chown and chgrp in /bin and hostname in /usr/bin.
1816
181728. Exim was keeping the "process log" file open after each use, just as it
1818 does for the main log. This opens the possibility of it remaining open for
1819 long periods when the USR1 signal hits a daemon. Occasional processlog
1820 errors were reported, that could have been caused by this. Anyway, it seems
1821 much more sensible not to leave this file open at all, so that is what now
1822 happens.
1823
182429. The long-running daemon process does not normally write to the log once it
1825 has entered its main loop, and it closes the log before doing so. This is
1826 so that log files can straightforwardly be renamed and moved. However,
1827 there are a couple of unusual error situations where the daemon does write
1828 log entries, and I had neglected to close the log afterwards.
1829
183030. The text of an SMTP error response that was received during a remote
1831 delivery was being truncated at 512 bytes. This is too short for some of
1832 the long messages that one sometimes sees. I've increased the limit to
1833 1024.
1834
183531. It is now possible to make retry rules that apply only when a message has a
1836 specific sender, in particular, an empty sender.
1837
183832. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
1839 it possible to be selective about when SMTP synchronization is enforced.
1840
184133. Added "control = caseful_local_part" and "control = "caselower_local_part".
1842
184332. Implemented hosts_connection_nolog.
1844
184533. Added an ACL for QUIT.
1846
184734. Setting "delay_warning=" to disable warnings was not working; it gave a
1848 syntax error.
1849
185035. Added mailbox_size and mailbox_filecount to appendfile.
1851
185236. Added control = no_multiline_responses to ACLs.
1853
185437. There was a bug in the logic of the code that waits for the clock to tick
1855 in the case where the clock went backwards by a substantial amount such
1856 that the microsecond fraction of "now" was more than the microsecond
1857 fraction of "then" (but the whole seconds number was less).
1858
185938. Added support for the libradius Radius client library this is found on
1860 FreeBSD (previously only the radiusclient library was supported).
1861
1862
1863Exim version 4.42
1864-----------------
1865
1866 1. When certain lookups returned multiple values in the form name=value, the
1867 quoting of the values was not always being done properly. Specifically:
1868 (a) If the value started with a double quote, but contained no whitespace,
1869 it was not quoted.
1870 (b) If the value contained whitespace other than a space character (i.e.
1871 tabs or newlines or carriage returns) it was not quoted.
1872 This fix has been applied to the mysql and pgsql lookups by writing a
1873 separate quoting function and calling it from the lookup code. The fix
1874 should probably also be applied to nisplus, ibase and oracle lookups, but
1875 since I cannot test any of those, I have not disturbed their existing code.
1876
1877 2. A hit in the callout cache for a specific address caused a log line with no
1878 reason for rejecting RCPT. Now it says "Previous (cached) callout
1879 verification failure".
1880
1881 3. There was an off-by-one bug in the queryprogram router. An over-long
1882 return line was truncated at 256 instead of 255 characters, thereby
1883 overflowing its buffer with the terminating zero. As well as fixing this, I
1884 have increased the buffer size to 1024 (and made a note to document this).
1885
1886 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
1887 when Exim is waiting for an SMTP response from a remote server, Exim
1888 restarts its select() call on the socket, thereby resetting its timeout.
1889 This is not a problem when such interrupts are rare. Somebody set up a cron
1890 job to run exiwhat every 2 minutes, which is less than the normal select()
1891 timeout (5 or 10 minutes). This meant that the select() timeout never
1892 kicked in because it was always reset. I have fixed this by comparing the
1893 time when an interrupt arrives with the time at the start of the first call
1894 to select(). If more time than the timeout has elapsed, the interrupt is
1895 treated as a timeout.
1896
1897 5. Some internal re-factoring in preparation for the addition of Sieve
1898 extensions (by MH). In particular, the "personal" test is moved to a
1899 separate function, and given an option for scanning Cc: and Bcc: (which is
1900 not set for Exim filters).
1901
1902 6. When Exim created an email address using the login of the caller as the
1903 local part (e.g. when creating a From: or Sender: header line), it was not
1904 quoting the local part when it contained special characters such as @.
1905
1906 7. Installed new OpenBSD configuration files.
1907
1908 8. Reworded some messages for syntax errors in "and" and "or" conditions to
1909 try to make them clearer.
1910
1911 9. Callout options, other than the timeout value, were being ignored when
1912 verifying sender addresses in header lines. For example, when using
1913
1914 verify = header_sender/callout=no_cache
1915
1916 the cache was (incorrectly) being used.
1917
191810. Added a missing instance of ${EXE} to the exim_install script; this affects
1919 only the Cygwin environment.
1920
192111. When return_path_on_delivery was set as a log selector, if different remote
1922 addresses in the same message used different return paths and parallel
1923 remote delivery occurred, the wrong values would sometimes be logged.
1924 (Whenever a remote delivery process finished, the return path value from
1925 the most recently started remote delivery process was logged.)
1926
192712. RFC 3848 specifies standard names for the "with" phrase in Received: header
1928 lines when AUTH and/or TLS are in use. This is the "received protocol"
1929 field. Exim used to use "asmtp" for authenticated SMTP, without any
1930 indication (in the protocol name) for TLS use. Now it follows the RFC and
1931 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
1932 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
1933 names appear in log lines as well as in Received: header lines.
1934
193513. Installed MH's patches for Sieve to add the "copy" and "vacation"
1936 extensions, and comparison tests, and to fix some bugs.
1937
193814. Changes to the "personal" filter test:
1939
1940 (1) The test was buggy in that it was just doing the equivalent of
1941 "contains" tests on header lines. For example, if a user's address was
1942 anne@some.where, the "personal" test would incorrectly be true for
1943
1944 To: susanne@some.where
1945
1946 This test is now done by extracting each address from the header in turn,
1947 and checking the entire address. Other tests that are part of "personal"
1948 are now done using regular expressions (for example, to check local parts
1949 of addresses in From: header lines).
1950
1951 (2) The list of non-personal local parts in From: addresses has been
1952 extended to include "listserv", "majordomo", "*-request", and "owner-*",
1953 taken from the Sieve specification recommendations.
1954
1955 (3) If the message contains any header line starting with "List-" it is
1956 treated as non-personal.
1957
1958 (4) The test for "circular" in the Subject: header line has been removed
1959 because it now seems ill-conceived.
1960
196115. Minor typos in src/EDITME comments corrected.
1962
196316. Installed latest exipick from John Jetmore.
1964
196517. If headers_add on a router specified a text string that was too long for
1966 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
1967 of string_sprintf() is now avoided.
1968
196918. $message_body_size was not set (it was always zero) when running the DATA
1970 ACL and the local_scan() function.
1971
197219. For the "mail" command in an Exim filter, no default was being set for
1973 the once_repeat time, causing a random time value to be used if "once" was
1974 specified. (If the value happened to be <= 0, no repeat happened.) The
1975 default is now 0s, meaning "never repeat". The "vacation" command was OK
1976 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
1977 (I found it when inspecting the code).
1978
197920. There is now an overall timeout for performing a callout verification. It
1980 defaults to 4 times the callout timeout, which applies to individual SMTP
1981 commands during the callout. The overall timeout applies when there is more
1982 than one host that can be tried. The timeout is checked before trying the
1983 next host. This prevents very long delays if there are a large number of
1984 hosts and all are timing out (e.g. when the network connections are timing
1985 out). The value of the overall timeout can be changed by specifying an
1986 additional sub-option for "callout", called "maxwait". For example:
1987
1988 verify = sender/callout=5s,maxwait=20s
1989
199021. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
1991 to the end before writing, but this should make it even safer).
1992
199322. Exim was forgetting that it had advertised PIPELINING for the second and
1994 subsequent messages on an SMTP connection. It was also not resetting its
1995 memory on STARTTLS and an internal HELO.
1996
199723. When Exim logs an SMTP synchronization error within a session, it now
1998 records whether PIPELINING has been advertised or not.
1999
200024. Added 3 instances of "(long int)" casts to time_t variables that were being
2001 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
2002 rather than long int.
2003
200425. Installed the latest Cygwin configuration files from the Cygwin maintainer.
2005
200626. Added the never_mail option to autoreply.
2007
2008
2009Exim version 4.41
2010-----------------
2011
2012 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
2013 crash if the getsockname() call failed; this can happen if a connection is
2014 closed very soon after it is established. The problem was simply in the
2015 order in which certain operations were done, causing Exim to try to write
2016 to the SMTP stream before it had set up the file descriptor. The bug has
2017 been fixed by making things happen in the correct order.
2018
2019
2020Exim version 4.40
2021-----------------
2022
2023 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
2024 before the connection was closed, thus losing the rejection response.
2025
2026 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
2027 some early Solaris releases, but causes trouble in current releases where
2028 socklen_t is defined.
2029
2030 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
2031 always exist.
2032
2033 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
2034 configured.
2035
2036 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
2037 that releasing the top of it at the end releases what was used for sub-
2038 expansions (unless the block got too big). However, discard this block if
2039 the first thing is a variable or header, so that we can use its block when
2040 it is dynamic (useful for very large $message_headers, for example).
2041
2042 6. Lookups now cache *every* query, not just the most recent. A new, separate
2043 store pool is used for this. It can be recovered when all lookup caches are
2044 flushed. Lookups now release memory at the end of their result strings.
2045 This has involved some general refactoring of the lookup sources.
2046
2047 7. Some code has been added to the store_xxx() functions to reduce the amount
2048 of flapping under certain conditions.
2049
2050 8. log_incoming_interface used to affect only the <= reception log lines. Now
2051 it causes the local interface and port to be added to several more SMTP log
2052 lines, for example "SMTP connection from", and rejection lines.
2053
2054 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
2055
205610. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
2057
205811. If $host_data was set by virtue of a hosts lookup in an ACL, its value
2059 could be overwritten at the end of the current message (or the start of a
2060 new message if it was set in a HELO ACL). The value is now preserved for
2061 the duration of the SMTP connection.
2062
206312. If a transport had a headers_rewrite setting, and a matching header line
2064 contained an unqualified address, that address was qualified, even if it
2065 did not match any rewriting rules. The underlying bug was that the values
2066 of the flags that permit the existence of unqualified sender and recipient
2067 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
2068 non-local messages, and by -bnq for local messages) were not being
2069 preserved with the message after it was received.
2070
207113. When Exim was logging an SMTP synchronization error, it could sometimes log
2072 "next input=" as part of the text comprising the host identity instead of
2073 the correct text. The code was using the same buffer for two different
2074 strings. However, depending on which order the printing function evaluated
2075 its arguments, the bug did not always show up. Under Linux, for example, my
2076 test suite worked just fine.
2077
207814. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
2079 doesn't work with some older versions of Perl. It has been changed to "my",
2080 which in any case is probably the better facility to use.
2081
208215. A really picky compiler found some instances of statements for creating
2083 error messages that either had too many or two few arguments for the format
2084 string.
2085
208616. The size of the buffer for calls to the DNS resolver has been increased
2087 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
2088 for addresses that have a lot of PTR records. This alleviates a problem; it
2089 does not fully solve it.
2090
209117. A dnsdb lookup for PTR records that receives more data than will fit in the
2092 buffer now truncates the list and logs the incident, which is the same
2093 action as happens when Exim is looking up a host name and its aliases.
2094 Previously in this situation something unpredictable would happen;
2095 sometimes it was "internal error: store_reset failed".
2096
209718. If a server dropped the connection unexpectedly when an Exim client was
2098 using GnuTLS and trying to read a response, the client delivery process
2099 crashed while trying to generate an error log message.
2100
210119. If a "warn" verb in an ACL added multiple headers to a message in a single
2102 string, for example:
2103
2104 warn message = H1: something\nH2: something
2105
2106 the text was added as a single header line from Exim's point of view
2107 though it ended up OK in the delivered message. However, searching for the
2108 second and subsequent header lines using $h_h2: did not work. This has been
2109 fixed. Similarly, if a system filter added multiple headers in this way,
2110 the routers could not see them.
2111
211220. Expanded the error message when iplsearch is called with an invalid key to
2113 suggest using net-iplsearch in a host list.
2114
211521. When running tests using -bh, any delays imposed by "delay" modifiers in
2116 ACLs are no longer actually imposed (and a message to that effect is
2117 output).
2118
211922. If a "gecos" field in a passwd entry contained escaped characters, in
2120 particular, if it contained a \" sequence, Exim got it wrong when building
2121 a From: or a Sender: header from that name. A second bug also caused
2122 incorrect handling when an unquoted " was present following a character
2123 that needed quoting.
2124
212523. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
2126 was not being matched caselessly.
2127
212824. Arranged for all hyphens in the exim.8 source to be escaped with
2129 backslashes.
2130
213125. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
2132 properly. Recipient callout cache records were still being keyed to include
2133 the sender, even when use_sender was set false. This led to far more
2134 callouts that were necessary. The sender is no longer included in the key
2135 when use_sender is false.
2136
213726. Added "control = submission" modifier to ACLs.
2138
213927. Added the ${base62d: operator to decode base 62 numbers.
2140
214128. dnsdb lookups can now access SRV records.
2142
214329. CONFIGURE_OWNER can be set at build time to define an alternative owner for
2144 the configuration file.
2145
214630. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
2147 (-v) mode. This makes the output for a verbose queue run more intelligible.
2148
214931. Added a use_postmaster feature to recipient callouts.
2150
215132. Added the $body_zerocount variable, containing the number of binary zero
2152 bytes in the message body.
2153
215433. The time of last modification of the "new" subdirectory is now used as the
2155 "mailbox time last read" when there is a quota error for a maildir
2156 delivery.
2157
215834. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
2159
216035. Added +ignore_unknown as a special item in host lists.
2161
216236. Code for decoding IPv6 addresses in host lists is now included, even if
2163 IPv6 support is not being compiled. This fixes a bug in which an IPv6
2164 address was recognized as an IP address, but was then not correctly decoded
2165 into binary, causing unexpected and incorrect effects when compared with
2166 another IP address.
2167
2168
2169Exim version 4.34
2170-----------------
2171
2172 1. Very minor rewording of debugging text in manualroute to say "list of
2173 hosts" instead of "hostlist".
2174
2175 2. If verify=header_syntax was set, and a header line with an unqualified
2176 address (no domain) and a large number of spaces between the end of the
2177 name and the colon was received, the reception process suffered a buffer
2178 overflow, and (when I tested it) crashed. This was caused by some obsolete
2179 code that should have been removed. The fix is to remove it!
2180
2181 3. When running in the test harness, delay a bit after writing a bounce
2182 message to get a bit more predictability in the log output.
2183
2184 4. Added a call to search_tidyup() just before forking a reception process. In
2185 theory, someone could use a lookup in the expansion of smtp_accept_max_
2186 per_host which, without the tidyup, could leave open a database connection.
2187
2188 5. Added the variables $recipient_data and $sender_data which get set from a
2189 lookup success in an ACL "recipients" or "senders" condition, or a router
2190 "senders" option, similar to $domain_data and $local_part_data.
2191
2192 6. Moved the writing of debug_print from before to after the "senders" test
2193 for routers.
2194
2195 7. Change 4.31/66 (moving the time when the Received: is generated) caused
2196 problems for message scanning, either using a data ACL, or using
2197 local_scan() because the Received: header was not generated till after they
2198 were called (in order to set the time as the time of reception completion).
2199 I have revised the way this works. The header is now generated after the
2200 body is received, but before the ACL or local_scan() are called. After they
2201 are run, the timestamp in the header is updated.
2202
2203
2204Exim version 4.33
2205-----------------
2206
2207 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
2208 before starting a queue runner without re-exec. This happened only when
2209 deliver_drop_privilege was set or when the Exim user was set to root. The
2210 effect of the bug was that timeouts during subsequent deliveries caused
2211 crashes instead of being properly handled. The handler is now left at its
2212 default (and expected) setting.
2213
2214 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
2215 message, again when deliver_drop_privilege is set or Exim is run as root.
2216 The bug described in (1) was not present in this case, but the tidying up
2217 of the other signals was missing. I have made the two cases consistent.
2218
2219 3. The ignore_target_hosts setting on a manualroute router was being ignored
2220 for hosts that were looked up using the /MX notation.
2221
2222 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
2223 in domain lists.
2224
2225 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
2226 operated on the sender address. After changing the $sender_address to <>
2227 for the sender address verify, Exim was re-instated it as the original
2228 (before rewriting) address, but remembering that it had rewritten it, so it
2229 wasn't rewriting it again. This bug also had the effect of breaking the
2230 sender address verification caching when the sender address was rewritten.
2231
2232 6. The ignore_target_hosts option was being ignored by the ipliteral router.
2233 This has been changed so that if the ip literal address matches
2234 ignore_target_hosts, the router declines.
2235
2236 7. Added expansion conditions match_domain, match_address, and match_local_
2237 part (NOT match_host).
2238
2239 8. The placeholder for the Received: header didn't have a length field set.
2240
2241 9. Added code to Exim itself and to exim_lock to test for a specific race
2242 condition that could lead to file corruption when using MBX delivery. The
2243 issue is with the lockfile that is created in /tmp. If this file is removed
2244 after a process has opened it but before that process has acquired a lock,
2245 there is the potential for a second process to recreate the file and also
2246 acquire a lock. This could lead to two Exim processes writing to the file
2247 at the same time. The added code performs the same test as UW imapd; it
2248 checks after acquiring the lock that its file descriptor still refers to
2249 the same named file.
2250
225110. The buffer for building added header lines was of fixed size, 8192 bytes.
2252 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
2253 when Exim is built.
2254
225511. Added the smtp_active_hostname option. If used, this will typically be made
2256 to depend on the incoming interface address. Because $interface_address is
2257 not set up until the daemon has forked a reception process, error responses
2258 that can happen earlier (such as "too many connections") no longer contain
2259 a host name.
2260
226112. If an expansion in a condition on a "warn" statement fails because a lookup
2262 defers, the "warn" statement is abandoned, and the next ACL statement is
2263 processed. Previously this caused the whole ACL to be aborted.
2264
226513. Added the iplsearch lookup type.
2266
226714. Added ident_timeout as a log selector.
2268
226915. Added tls_certificate_verified as a log selector.
2270
227116. Added a global option tls_require_ciphers (compare the smtp transport
2272 option of the same name). This controls incoming TLS connections.
2273
227417. I finally figured out how to make tls_require_ciphers do a similar thing
2275 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
2276 before starting the TLS session.
2277
227818. Tabs are now shown as \t in -bP output.
2279
228019. If the log selector return_path_on_delivery was set, Exim crashed when
2281 bouncing a message because it had too many Received: header lines.
2282
228320. If two routers both had headers_remove settings, and the first one included
2284 a superfluous trailing colon, the final name in the first list and the
2285 first name in the second list were incorrectly joined into one item (with a
2286 colon in the middle).
2287
2288
2289Exim version 4.32
2290-----------------
2291
2292 1. Added -C and -D options to the exinext utility, mainly to make it easier
2293 to include in the automated testing, but these could be helpful when
2294 multiple configurations are in use.
2295
2296 2. The exinext utility was not formatting the output nicely when there was
2297 an alternate port involved in the retry record key, nor when there was a
2298 message id as well (for retries that were specific to a specific message
2299 and a specific host). It was also confused by IPv6 addresses, because of
2300 the additional colons they contain. I have fixed the IPv4 problem, and
2301 patched it up to do a reasonable job for IPv6.
2302
2303 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
2304 delivery, the log line now contains "pipelined" if PIPELINING was used.
2305
2306 4. An SMTP transport process used to panic and die if the bind() call to set
2307 an explicit outgoing interface failed. This has been changed; it is now
2308 treated in the same way as a connect() failure.
2309
2310 5. A reference to $sender_host_name in the part of a conditional expansion
2311 that was being skipped was still causing a DNS lookup. This no longer
2312 occurs.
2313
2314 6. The def: expansion condition was not recognizing references to header lines
2315 that used bh_ and bheader_.
2316
2317 7. Added the _cache feature to named lists.
2318
2319 8. The code for checking quota_filecount in the appendfile transport was
2320 allowing one more file than it should have been.
2321
2322 9. For compatibility with Sendmail, the command line option
2323
2324 -prval:sval
2325
2326 is equivalent to
2327
2328 -oMr rval -oMs sval
2329
2330 and sets the incoming protocol and host name (for trusted callers). The
2331 host name and its colon can be omitted when only the protocol is to be set.
2332 Note the Exim already has two private options, -pd and -ps, that refer to
2333 embedded Perl. It is therefore impossible to set a protocol value of "d" or
2334 "s", but I don't think that's a major issue.
2335
233610. A number of refactoring changes to the code, none of which should affect
2337 Exim's behaviour:
2338
2339 (a) The number of logging options was getting close to filling up the
2340 32-bit word that was used as a bit map. I have split them into two classes:
2341 those that are passed in the argument to log_write(), and those that are
2342 only ever tested independently outside of that function. These are now in
2343 separate 32-bit words, so there is plenty of room for expansion again.
2344 There is no change in the user interface or the logging behaviour.
2345
2346 (b) When building, for example, log lines, the code previously used a
2347 macro that called string_cat() twice, in order to add two strings. This is
2348 not really sufficiently general. Furthermore, there was one instance where
2349 it was actually wrong because one of the argument was used twice, and in
2350 one call a function was used. (As it happened, calling the function twice
2351 did not affect the overall behaviour.) The macro has been replaced by a
2352 function that can join an arbitrary number of extra strings onto a growing
2353 string.
2354
2355 (c) The code for expansion conditions now uses a table and a binary chop
2356 instead of a serial search (which was left over from when there were very
2357 few conditions). Also, it now recognizes conditions like "pam" even when
2358 the relevant support is not compiled in: a suitably worded error message is
2359 given if an attempt is made to use such a condition.
2360
236111. Added ${time_interval:xxxxx}.
2362
236312. A bug was causing one of the ddress fields not to be passed back correctly
2364 from remote delivery subprocesses. The field in question was not being
2365 subsequently used, so this caused to problems in practice.
2366
236713. Added new log selectors queue_time and deliver_time.
2368
236914. Might have fixed a bug in maildirsizefile handling that threw up
2370 "unexpected character" debug warnings, and recalculated the data
2371 unnecessarily. In any case, I expanded the warning message to give more
2372 information.
2373
237415. Added the message "Restricted characters in address" to the statements in
2375 the default ACL that block characters like @ and % in local parts.
2376
237716. Change 71 for release 4.31 proved to be much less benign that I imagined.
2378 Three changes have been made:
2379
2380 (a) There was a serious bug; a negative response to MAIL caused the whole
2381 recipient domain to be cached as invalid, thereby blocking all messages
2382 to all local parts at the same domain, from all senders. This bug has
2383 been fixed. The domain is no longer cached after a negative response to
2384 MAIL if the sender used is not empty.
2385
2386 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
2387 been restored.
2388
2389 (c) A new callout option, "use_sender" has been added for people who want
2390 the modified behaviour.
2391
2392
2393Exim version 4.31
2394-----------------
2395
2396 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
2397 Larry Rosenman.
2398
2399 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
2400 indeed breaks things for older releases.
2401
2402 3. Added additional logging to the case where there is a problem reading data
2403 from a filter that is running in a subprocess using a pipe, in order to
2404 try to track down a specific problem.
2405
2406 4. Testing facility fudge: when running in the test harness and attempting
2407 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
2408 getting "No route to host". Convert this to a timeout.
2409
2410 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
2411 warning.
2412
2413 6. Some OS don't have socklen_t but use size_t instead. This affects the
2414 fifth argument of getsockopt() amongst other things. This is now
2415 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
2416 can be set for individual OS. I have set it for SunOS5, OSF1, and
2417 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
2418 some earlier ones do not.
2419
2420 7. Change 4.30/15 was not doing the test caselessly.
2421
2422 8. The standard form for an IPv6 address literal was being rejected by address
2423 parsing in, for example, MAIL and RCPT commands. An example of this kind of
2424 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
2425 this, as well as the form without the "IPv6" on the front (but only when
2426 address literals are enabled, of course).
2427
2428 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
2429
243010. Exim crashed if a message with an empty sender address specified by -f
2431 encountered a router with an errors_to setting. This could be provoked only
2432 by a command such as
2433
2434 exim -f "" ...
2435
2436 where an empty string was supplied; "<>" did not hit this bug.
2437
243811. Installed PCRE release 4.5.
2439
244012. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
2441 remained set. It is now erased.
2442
244313. exiqgrep wasn't working on MacOS X because it didn't correctly compute
2444 times from message ids (which are base 36 rather than the normal 62).
2445
244614. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
2447 were being counted as actual protocol errors, and logged if the log
2448 selector +smtp_protocol_error was set. One cannot be perfect in this test,
2449 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
2450 and DATA following a set of rejected RCPTs do not count as protocol errors.
2451 In other words, Exim assumes they were pipelined, though this may not
2452 actually be the case. Of course, in all cases the client gets an
2453 appropriate error code.
2454
245515. If a lookup fails in an ACL condition, a message about the failure may
2456 be available; it is used if testing the ACL cannot continue, because most
2457 such messages specify what the cause of the deferral is. However, some
2458 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
2459 that caused an old message to be retained and used if a later statement
2460 caused a defer, replacing the real cause of the deferral.
2461
246216. If an IP address had so many PTR records that the DNS lookup buffer
2463 was not large enough to hold them, Exim could crash while trying to process
2464 the truncated data. It now detects and logs this case.
2465
246617. Further to 4.21/58, another change has been made: if (and only if) the
2467 first line of a message (the first header line) ends with CRLF, a bare LF
2468 in a subsequent header line has a space inserted after it, so as not to
2469 terminate the header.
2470
247118. Refactoring: tidied an ugly bit of code in appendfile that copied data
2472 unnecessarily, used atoi() instead of strtol(), and didn't check the
2473 termination when getting file sizes from file names by regex.
2474
247519. Completely re-implemented the support for maildirsize files, in the light
2476 of a number of problems with the previous contributed implementation
2477 (4.30/29). In particular:
2478
2479 . If the quota is zero, the maildirsize file is maintained, but no quota is
2480 imposed.
2481
2482 . If the maildir directory does not exist, it is created before any attempt
2483 to write a maildirsize file.
2484
2485 . The quota value in the file is just a cache; if the quota is changed in
2486 the transport, the new value overrides.
2487
2488 . A regular expression is available for excluding directories from the
2489 count.
2490
249120. The autoreply transport checks the characters in options that define the
2492 message's headers; it allows continued headers, but it was checking with
2493 isspace() after an embedded newline instead of explicitly looking for a
2494 space or a tab.
2495
249621. If all the "regular" hosts to which an address was routed had passed their
2497 expiry times, and had not reached their retry times, the address was
2498 bounced, even if fallback hosts were defined. Now Exim should go on to try
2499 the fallback hosts.
2500
250122. Increased buffer sizes in the callout code from 1024 to 4096 to match the
2502 equivalent code in the SMTP transport. Some hosts send humungous responses
2503 to HELO/EHLO, more than 1024 it seems.
2504
250523. Refactoring: code in filter.c used (void *) for "any old type" but this
2506 gives compiler warnings in some environments. I've now done it "properly",
2507 using a union.
2508
250924. The replacement for inet_ntoa() that is used with gcc on IRIX systems
2510 (because of problems with the built-in one) was declared to return uschar *
2511 instead of char *, causing compiler failure.
2512
251325. Fixed a file descriptor leak when processing alias/forward files.
2514
251526. Fixed a minor format string issue in dbfn.c.
2516
251727. Typo in exim.c: ("dmbnz" for "dbmnz").
2518
251928. If a filter file refered to $h_xxx or $message_headers, and the headers
2520 contained RFC 2047 "words", Exim's memory could, under certain conditions,
2521 become corrupted.
2522
252329. When a sender address is verified, it is cached, to save repeating the test
2524 when there is more than one recipient in a message. However, when the
2525 verification involves a callout, it is possible for different callout
2526 options to be set for different recipients. It is too complicated to keep
2527 track of this in the cache, so now Exim always runs a verification when a
2528 callout is required, relying on the callout cache for the optimization.
2529 The overhead is duplication of the address routing, but this should not be
2530 too great.
2531
253230. Fixed a bug in callout caching. If a RCPT command caused the sender address
2533 to be verified with callout=postmaster, and the main callout worked but the
2534 postmaster check failed, the verification correctly failed. However, if a
2535 subsequent RCPT command asked for sender verification *without* the
2536 postmaster check, incorrect caching caused this verification also to fail,
2537 incorrectly.
2538
253931. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
2540 it was not caching the DNS options (qualify_single, search_parents) that
2541 were used when the lookup failed. A subsequent lookup with different
2542 options therefore always gave the same answer, though there were cases
2543 where it should not have. (Example: a "domains = !$mx_any" option on a
2544 dnslookup router: the "domains" option is always processed without any
2545 widening, but the router might have qualify_single set.) Now Exim uses the
2546 cached value only when the same options are set.
2547
254832. Added John Jetmore's "exipick" utility to the distribution.
2549
255033. GnuTLS: When an attempt to start a TLS session fails for any reason other
2551 than a timeout (e.g. a certificate is required, and is not provided), an
2552 Exim server now closes the connection immediately. Previously it waited for
2553 the client to close - but if the client is SSL, it seems that they each
2554 wait for each other, leading to a delay before one of them times out.
2555
255634: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
2557 maintained 0.8.x compatibility because I don't think many are using it, and
2558 it is clearly obsolete.
2559
256035. Added TLS support for CRLs: a tls_crl global option and one for the smtp
2561 transport.
2562
256336. OpenSSL: $tls_certificate_verified was being set to 1 even if the
2564 client certificate was expired. A simple patch fixes this, though I don't
2565 understand the full logic of why the verify callback is called multiple
2566 times.
2567
256837. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
2569 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
2570 which causes problems with some clients (such as the Certicom SSL Plus
2571 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
2572 disables the coutermeasure allowing Eudora to connect."
2573
257438. Exim was not checking that a write() to a log file succeeded. This could
2575 lead to Bad Things if a log got too big, in particular if it hit a file
2576 size limit. Exim now panics and dies if it cannot write to a log file, just
2577 as it does if it cannot open a log file.
2578
257939. Modified OS/Makefile-Linux so that it now contains
2580
2581 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
2582
2583 The two -D definitions ensure that Exim is compiled with large file
2584 support, which makes it possible to handle log files that are bigger than
2585 2^31.
2586
258740. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
2588 instance) a domain was checked against a named list that involved a lookup,
2589 causing $domain_data to be set, then another domain was checked against the
2590 same list, then the first domain was re-checked, the value of $domain_data
2591 after the final check could be wrong. In particular, if the second check
2592 failed, it could be set empty. This bug probably also applied to
2593 $localpart_data.
2594
259541. The strip_trailing_dot option was not being applied to the address given
2596 with the -f command-line option.
2597
259842. The code for reading a message's header from the spool was incrementing
2599 $received_count, but never initializing it. This meant that the value was
2600 incorrect (doubled) while delivering a message in the same process in which
2601 it was received. In the most common configuration of Exim, this never
2602 happens - a fresh exec is done - but it can happen when
2603 deliver_drop_privilege is set.
2604
260543. When Exim logs an SMTP synchronization error - client data sent too soon -
2606 it now includes up to 150 characters of the unexpected data in the log
2607 line.
2608
260944. The exim_dbmbuild utility uses fixed size buffers for reading input lines
2610 and building data strings. The size of both of these buffers was 10 000
2611 bytes - far larger than anybody would *ever* want, thought I. Needless to
2612 say, somebody hit the limit. I have increased the maximum line length to
2613 20 000 and the maximum data length of concatenated lines to 100 000. I have
2614 also fixed two bugs, because there was no checking on these buffers. Tsk,
2615 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
2616 buffer is too small.
2617
261845. The exim_dbmbuild utility did not support quoted keys, as Exim does in
2619 lsearch lookups. Now it does.
2620
262146. When parsing a route_list item in a manualroute router, a fixed-length
2622 buffer was used for the list of hosts. I made this 1024 bytes long,
2623 thinking that nobody would ever have a list of hosts that long. Wrong.
2624 Somebody had a whole pile of complicated expansion conditions, and the
2625 string was silently truncated, leading to an expansion error. It turns out
2626 that it is easier to change to an unlimited length (owing to other changes
2627 that have happened since this code was originally written) than to build
2628 structure for giving a limitation error. The length of the item that
2629 expands into the list of hosts is now unlimited.
2630
263147. The lsearch lookup could not handle data where the length of text line was
2632 more than 4095 characters. Such lines were truncated, leading to shortened
2633 data being returned. It should now handle lines of any length.
2634
263548. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
2636 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
2637 ACL").
2638
263949. Cosmetic tidy to scripts like exicyclog that are generated by globally
2640 replacing strings such as BIN_DIRECTORY in a source file: the replacement
2641 no longer happens in comment lines. A list of replacements is now placed
2642 at the head of all of the source files, except those whose only change is
2643 to replace PERL_COMMAND in the very first #! line.
2644
264550. Replaced the slow insertion sort in queue.c, for sorting the list of
2646 messages on the queue, with a bottom-up merge sort, using code contributed
2647 by Michael Haardt. This should make operations like -bp somewhat faster on
2648 large queues. It won't affect queue runners, except when queue_run_in_order
2649 is set.
2650
265151. Installed eximstats 1.31 in the distribution.
2652
265352. Added support for SRV lookups to the dnslookup router.
2654
265553. If an ACL referred to $message_body or $message_body_end, the value was not
2656 reset for any messages that followed in the same SMTP session.
2657
265854. The store-handling optimization for building very long strings was not
2659 differentiating between the different store pools. I don't think this
2660 actually made any difference in practice, but I've tidied it.
2661
266255. While running the routers to verify a sender address, $sender_address
2663 was still set to the sender address. This is wrong, because when routing to
2664 send a bounce to the sender, it would be empty. Therefore, I have changed
2665 it so that, while verifying a sender address, $sender_address is set to <>.
2666 (There is no change to what happens when verifying a recipient address.)
2667
266856. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
2669 target A or AAAA records (if not already returned) without resetting the
2670 qualify_single or search_parents options of the DNS resolver. These are
2671 inappropriate in this case because the targets of MX and SRV records must
2672 be FQDNs. A broken DNS record could cause trouble if it happened to have a
2673 target that, when qualified, matched something in the local domain. These
2674 two options are now turned off when doing these lookups.
2675
267657. It seems that at least some releases of Reiserfs (which does not have the
2677 concept of a fixed number of inodes) returns zero and not -1 for the
2678 number of available inodes. This interacted badly with check_spool_inodes,
2679 which assumed that -1 was the "no such thing" setting. What I have done is
2680 to check that the total number of inodes is greater than zero before doing
2681 the test of how many are available.
2682
268358. When a "warn" ACL statement has a log_message modifier, the message is
2684 remembered, and not repeated. This is to avoid a lot of repetition when a
2685 message has many recipients that cause the same warning to be written.
2686 Howewer, Exim was preserving the list of already written lines for an
2687 entire SMTP session, which doesn't seem right. The memory is now reset if a
2688 new message is started.
2689
269059. The "rewrite" debugging flag was not showing the result of rewriting in the
2691 debugging output unless log_rewrite was also set.
2692
269360. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
2694 of (int)(handle) when we know that handle contains (void *)(-1).
2695
269661. The Exim daemon panic-logs an error return when it closes the incoming
2697 connection. However "connection reset by peer" seems to be common, and
2698 isn't really an error worthy of noting specially, so that particular error
2699 is no long logged.
2700
270162. When Exim is trying to find all the local interfaces, it used to panic and
2702 die if the ioctl to get the interface flags failed. However, it seems that
2703 on at least one OS (Solaris 9) it is possible to have an interface that is
2704 included in the list of interfaces, but for which you get a failure error
2705 for this call. This happens when the interface is not "plumbed" into a
2706 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
2707 failure of the "get flags" call assumes that the interface is down.
2708
270963. Added a ${eval10: operator, which assumes all numbers are decimal. This
2710 makes life easier for people who are doing arithmetic on fields extracted
2711 from dates, where you often get leading zeros that should not be
2712 interpreted as octal.
2713
271464. Added qualify_domain to the redirect router, to override the global
2715 setting.
2716
271765. If a pathologically long header line contained very many addresses (the
2718 report of this problem mentioned 10 000) and each of them was rewritten,
2719 Exim could use up a very large amount of memory. (It kept on making new
2720 copies of the header line as it rewrote, and never released the old ones.)
2721 At the expense of a bit more processing, the header rewriting function has
2722 been changed so that it no longer eats memory in this way.
2723
272466. The generation of the Received: header has been moved from the time that a
2725 message starts to be received, to the time that it finishes. The timestamp
2726 in the Received: header should now be very close to that of the <= log
2727 line. There are two side-effects of this change:
2728
2729 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
2730 logged header lines no longer include the local Received: line, because
2731 it has not yet been created. The same applies to a copy of the message
2732 that is returned to a non-SMTP sender when a message is rejected.
2733
2734 (b) When a filter file is tested using -bf, no additional Received: header
2735 is added to the test message. After some thought, I decided that this
2736 is a bug fix.
2737
2738 This change does not affect the value of $received_for. It is still set
2739 after address rewriting, but before local_scan() is called.
2740
274167. Installed the latest Cygwin-specific files from the Cygwin maintainer.
2742
274368. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
2744 gave an unhelpful panic error message, and a defer error. I have managed to
2745 change this behaviour so that it now rejects any supplied certificate,
2746 which seems right, as the list of acceptable certificates is empty.
2747
274869. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
2749 gave an unhelpful defer error. I have not managed to make this reject any
2750 supplied certificates, but the error message it gives is "no certificate
2751 supplied", which is not helpful.
2752
275370. exigrep's output now also includes lines that are not associated with any
2754 message, but which match the given pattern. Implemented by a patch from
2755 Martin Sluka, which also tidied up the Perl a bit.
2756
275771. Recipient callout verification, like sender verification, was using <> in
2758 the MAIL FROM command. This isn't really the right thing, since the actual
2759 sender may affect whether the remote host accepts the recipient or not. I
2760 have changed it to use the actual sender in the callout; this means that
2761 the cache record is now keyed on a recipient/sender pair, not just the
2762 recipient address. There doesn't seem to be a real danger of callout loops,
2763 since a callout by the remote host to check the sender would use <>.
2764 [SEE ABOVE: changed after hitting problems.]
2765
276672. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
2767 temporary errors. However, in the case of such a code being given after
2768 the end of a data transmission (i.e. after ".") Exim was failing to write
2769 a retry record for the message. (Yes, there was some broken host that was
2770 actually sending 8xx at this point.)
2771
277273. An unknown lookup type in a host list could cause Exim to panic-die when
2773 the list was checked. (An example that provoked this was putting <; in the
2774 middle of a list instead of at the start.) If this happened during a DATA
2775 ACL check, a -D file could be left lying around. This kind of configuration
2776 error no longer causes Exim to die; instead it causes a defer errror. The
2777 incident is still logged to the main and panic logs.
2778
277974. Buglet left over from Exim 3 conversion. The message "too many messages
2780 in one connection" was written to the rejectlog but not the mainlog, except
2781 when address rewriting (yes!) was being logged.
2782
278375. Added write_rejectlog option.
2784
278576. When a system filter was run not as root (that is, when system_filter_user
2786 was set), the values of the $n variables were not being returned to the
2787 main process; thus, they were not subsequently available in the $sn
2788 variables.
2789
279077. Added +return_path_on_delivery log selector.
2791
279278. A connection timeout was being treated differently from recipients deferred
2793 when testing hosts_max_try with a message that was older than the host's
2794 retry timeout. (The host should not be counted, thus allowing all hosts to
2795 be tried at least once before bouncing.) This may have been the cause of an
2796 occasionally reported bug whereby a message would remain on the queue
2797 longer than the retry timeout, but would be bounced if a delivery was
2798 forced. I say "may" because I never totally pinned down the problem;
2799 setting up timeout/retry tests is difficult. See also the next item.
2800
280179. The ultimate address timeout was not being applied to errors that involved
2802 a combination of host plus message (for example, a timeout on a MAIL
2803 command). When an address resolved to a number of possible hosts, and they
2804 were not all tried for each delivery (e.g. because of hosts_max_try), a
2805 message could remain on the queue longer than the retry timeout.
2806
280780. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
2808 Haardt.
2809
281081. Fixed an obscure SMTP outgoing bug which required at least the following
2811 conditions: (a) there was another message waiting for the same server;
2812 (b) the server returned 5xx to all RCPT commands in the first message so
2813 that the message was not completed; (c) the server dropped the connection
2814 or gave a negative response to the RSET that Exim sends to abort the
2815 transaction. The observed case was a dropped connection after DATA that had
2816 been sent in pipelining mode. That is, the server had advertised PIPELINING
2817 but was not implementing it correctly. The effect of the bug was incorrect
2818 behaviour, such as trying another host, and this could lead to a crash.
2819
2820
2821Exim version 4.30
2822-----------------
2823
2824 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
2825 and daemon.c were passed as pointers to ints; they should have been
2826 pointers to socklen_t variables (which are typically unsigned ints).
2827
2828 2. Some signed/unsigned type warnings in the os.c file for Linux have been
2829 fixed.
2830
2831 3. Fixed a really odd bug that affected only the testing scheme; patching a
2832 certain fixed string in the binary changed the value of another string that
2833 happened to be identical to the end of the original first string.
2834
2835 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
2836 name", it returns that address as the IP address. On some operating
2837 systems (e.g. Solaris), it also passes back the IP address string as the
2838 "host name". However, on others (e.g. Linux), it passes back an empty
2839 string. Exim wasn't checking for this, and was changing the host name to an
2840 empty string, assuming it had been canonicized.
2841
2842 5. Although rare, it is permitted to have more than one PTR record for a given
2843 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
2844 all the names associated with an address, because they do in Solaris.
2845 However, it seems that they do not in Linux for data that comes from the
2846 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
2847 I found this out when I moved to a new Linux workstation and tried to run
2848 the Exim test suite.
2849
2850 To get round this problem I have changed the code so that it now does its
2851 own call to the DNS to look up PTR records when searching for a host name.
2852 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
2853 addresses that are only in /etc/hosts are still found.
2854
2855 This behaviour is, however, controlled by an option called host_lookup_
2856 order, which defaults to "bydns:byaddr". If people want to use the other
2857 order, or indeed, just use one or the other means of lookup, they can
2858 specify it in this variable.
2859
2860 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
2861 some operating systems, this comes back from gethostbyaddr() as an empty
2862 string, and this is what Exim used to test for. However, it seems that in
2863 other systems, "." is yielded. Exim now tests for this case too.
2864
2865 7. The values of check_spool_space and check_log_space are now held internally
2866 as a number of kilobytes instead of an absolute number of bytes. If a
2867 numbers is specified without 'K' or 'M', it is rounded up to the nearest
2868 kilobyte. This means that much larger values can be stored.
2869
2870 8. Exim monitor: an attempt to get the action menu when not actually pointing
2871 at a message produces an empty menu entitled "No message selected". This
2872 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
2873 no entries in it ("Shell widget menu has zero width and/or height"). So I
2874 have added a single, blank menu entry in this case.
2875
2876 9. Added ${quote_local_part.
2877
287810. MIME decoding is now applied to the contents of Subject: header lines when
2879 they are logged.
2880
288111. Now that a reference to $sender_host_address automatically causes a reverse
2882 lookup to occur if necessary (4.13/18), there is no need to arrange for a
2883 host lookup before query-style lookups in lists that might use this
2884 variable. This has therefore been abolished, and the "net-" prefix is no
2885 longer necessary for query-style lookups.
2886
288712. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
2888 have been a typo for LFLAGS, so it has been changed.
2889
289013. The install script calls Exim with "-C /dev/null" in order to find the
2891 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
2892 to be output. Howeve, since Exim outputs its version number before the
2893 error, it didn't break the script. It just looked ugly. I fixed this by
2894 always allowing "-C /dev/null" if the caller is root.
2895
289614. Ignore overlarge ACL variable number when reading spool file - insurance
2897 against a later release with more variables having written the file.
2898
289915. The standard form for an IPv6 address literal was being rejected by EHLO.
2900 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
2901 this, as well as the form without the "IPv6" on the front.
2902
290316. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
2904 OS/Makefile-Darwin file.
2905
290617. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
2907 only to LDAP libraries that do not have LDAP_OPT_DEREF.
2908
290918. After change 4.21/52, "%ld" was used to format the contents of the $inode
2910 variable. However, some OS use ints for inodes. I've added cast to long int
2911 to get rid of the compiler warning.
2912
291319. I had forgotten to lock out "/../" in configuration file names when
2914 ALT_CONFIG_PREFIX was set.
2915
291620. Routers used for verification do not need to specify transports. However,
2917 if such a router generated a host list, and callout was configured, Exim
2918 crashed, because it could not find a port number from the (non-existent)
2919 transport. It now assumes port 25 in this circumstance.
2920
292121. Added the -t option to exigrep.
2922
292322. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
2924 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
2925 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
2926 them is set, LOOKUP_LSEARCH is forced.
2927
292823. "exim -bV" now outputs a list of lookups that are included in the binary.
2929
293024. Added sender and host information to the "rejected by local_scan()" log
2931 line; previously there was no indication of these.
2932
293325. Added .include_if_exists.
2934
293526. Change 3.952/11 added an explicit directory sync on top of a file sync for
2936 Linux. It turns out that not all file systems support this. Apparently some
2937 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
2938 do it.) To cope with this, the error EINVAL, which means that sync-ing is
2939 not supported on the file descriptor, is now ignored when Exim is trying to
2940 sync a directory. This applies only to Linux.
2941
294227. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
2943
294428. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
2945 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
2946 this.
2947
294829. Added support for maildirsize files from supplied patch (modified a bit).
2949
295030. The use of :fail: followed by an empty string could lead Exim to respond to
2951 sender verification failures with (e.g.):
2952
2953 550 Verification failed for <xxx>
2954 550 Sender verify failed
2955
2956 where the first response line was missing the '-' that indicates it is not
2957 the final line of the response.
2958
295931. The loop for finding the name of the user that called Exim had a hardwired
2960 limit of 10; it now uses the value of finduser_retries, which is used for
2961 all other user lookups.
2962
296332. Added $received_count variable, available in data and not_smtp ACLs, and at
2964 delivery time.
2965
296633. Exim was neglecting to zero errno before one call of strtol() when
2967 expanding a string and expecting an integer value. On some systems this
2968 resulted in spurious "integer overflow" errors. Also, it was casting the
2969 result into an int without checking.
2970
297134. Testing for a connection timeout using "timeout_connect" in the retry rules
2972 did not work. The code looks as if it has *never* worked, though it appears
2973 to have been documented since at least releast 1.62. I have made it work.
2974
297535. The "timeout_DNS" error in retry rules, also documented since at least
2976 1.62, also never worked. As it isn't clear exactly what this means, and
2977 clearly it isn't a major issue, I have abolished the feature by treating it
2978 as "timeout", and writing a warning to the main and panic logs.
2979
298036. The display of retry rules for -brt wasn't always showing the error code
2981 correctly.
2982
298337. Added new error conditions to retry rules: timeout_A, timeout_MX,
2984 timeout_connect_A, timeout_connect_MX.
2985
298638. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
2987 to the empty sender.
2988
298939. The daemon was not analysing the content of -oX till after it had closed
2990 stderr and disconnected from the controlling terminal. This meant that any
2991 syntax errors were only noted on the panic log, and the return code from
2992 the command was 0. By re-arranging the code a little, I've made the
2993 decoding happen first, so such errors now appear on stderr, and the return
2994 code is 1. However, the actual setting up of the sockets still happens in
2995 the disconnected process, so errors there are still only recorded on the
2996 panic log.
2997
299840. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
2999 connections (as happens on some IP stacks) was logged at start up time as
3000 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
3001 it from "IPv6 and IPv4", which means that two separate sockets are being
3002 used.
3003
300441. The debug output for gethostbyname2() or getipnodebyname() failures now
3005 says whether AF_INET or AF_INET6 was passed as an argument.
3006
300742. Exiwhat output was messed up when time zones were included in log
3008 timestamps.
3009
301043. Exiwhat now gives more information about the daemon's listening ports,
3011 and whether -tls-on-connect was used.
3012
301344. The "port" option of the smtp transport is now expanded.
3014
301545. A "message" modifier in a "warn" statement in a non-message ACL was being
3016 silently ignored. Now an error message is written to the main and panic
3017 logs.
3018
301946. There's a new ACL modifier called "logwrite" which writes to a log file
3020 as soon as it is encountered.
3021
302247. Added $local_user_uid and $local_user_gid at routing time.
3023
302448. Exim crashed when trying to verify a sender address that was being
3025 rewritten to "<>".
3026
302749. Exim was recognizing only a space character after ".include". It now also
3028 recognizes a tab character.
3029
303050. Fixed several bugs in the Perl script that creates the exim.8 man page by
3031 extracting the relevant information from the specification. The man page no
3032 longer contains scrambled data for the -d option, and I've added a section
3033 at the front about calling Exim under different names.
3034
303551. Added "extra_headers" argument to the "mail" command in filter files.
3036
303752. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
3038 crash.
3039
304053. Installed eximstats 1.29.
3041
304254. Added transport_filter_timeout as a generic transport option.
3043
304455. Exim no longer adds an empty Bcc: header to messages that have no To: or
3045 Cc: header lines. This was required by RFC 822, but it not required by RFC
3046 2822.
3047
304856. Exim used to add From:, Date:, and Message-Id: header lines to any
3049 incoming messages that did not have them. Now it does so only if the
3050 message originates locally, that is, if there is no associated remote host
3051 address. When Resent- header lines are present, this applies to the Resent-
3052 lines rather than the non-Resent- lines.
3053
305457. Drop incoming SMTP connection after too many syntax or protocol errors. The
3055 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
3056
305758. Messages for configuration errors now include the name of the main
3058 configuration file - useful now that there may be more than one file in a
3059 list (.included file names were always shown).
3060
306159. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
3062 for those rare installations that do not start the daemon as root or run it
3063 setuid root. I've cut out the call to initgroups() if the daemon is not
3064 root at that time.
3065
306660. The Exim user and group can now be bound into the binary as text strings
3067 that are looked up at the start of Exim's processing.
3068
306961. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
3070
307162. Added $mailstore_basename variable.
3072
307363. Installed patch to sieve.c from Michael Haardt.
3074
307564. When Exim failed to open the panic log after failing to open the main log,
3076 the original message it was trying to log was written to stderr and debug
3077 output, but if they were not available (the usual case in production), it
3078 was lost. Now it is written to syslog before the two lines that record the
3079 failures to open the logs.
3080
308165. Users' Exim filters run in subprocesses under the user's uid. It is
3082 possible for a "deliver" command or an alias in a "personal" command to
3083 provoke an address rewrite. If logging of address rewriting is configured,
3084 this fails because the process is not running as root or exim. There may be
3085 a better way of dealing with this, but for the moment (because 4.30 needs
3086 to be released), I have disabled address rewrite logging when running a
3087 filter in a non-root, non-exim process.
3088
3089
3090Exim version 4.24
3091-----------------
3092
3093 1. The buildconfig auxiliary program wasn't quoting the value set for
3094 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
3095 not defined. This bug was masked in 4.22 by the effect that was fixed in
3096 change 4.23/1.
3097
3098 2. Some messages that were rejected after a message id was allocated were
3099 shown as "incomplete" by exigrep. It no longer does this for messages that
3100 are rejected by local_scan() or the DATA or non-SMTP ACLs.
3101
3102 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
3103 have allow_domain_literals set, the ID did not get logged in the <= line.
3104 Domain literals are now always recognized in Message-ID: header lines.
3105
3106 4. The first argument for a ${extract expansion item is the key name or field
3107 number. Leading and trailing spaces in this item were not being ignored,
3108 causing some misleading effects.
3109
3110 5. When deliver_drop_privilege was set, single queue runner processes started
3111 manually (i.e. by the command "exim -q") or by the daemon (which uses the
3112 same command in the process it spins off) were not dropping privilege.
3113
3114 6. When the daemon running as "exim" started a queue runner, it always
3115 re-executed Exim in the spun-off process. This is a waste of effort when
3116 deliver_drop_privilege is set. The new process now just calls the
3117 queue-runner function directly.
3118
3119
3120Exim version 4.23
3121-----------------
3122
3123 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
3124 HEADERS_CHARSET.
3125
3126 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
3127 ignored. Though the use of -oP was forcing the writing of a pid file, it
3128 was always written to the default place.
3129
3130 3. If the message "no IP address found for host xxxx" is generated during
3131 incoming verification, it is now followed by identification of the incoming
3132 connection (so you can more easily find what provoked it).
3133
3134 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
3135
3136 5. Added some features to "harden" Exim a bit more against certain attacks:
3137
3138 (a) There is now a build-time option called FIXED_NEVER_USERS that can
3139 be put in Local/Makefile. This is like the never_users runtime option,
3140 but it cannot be overridden. The default setting is "root".
3141
3142 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
3143 prefix string with which any file named in a -C command line option
3144 must start.
3145
3146 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
3147 is retained for -C and -D only if the caller of Exim is root. Without
3148 it, the exim user may also use -C and -D and retain privilege.
3149
3150 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
3151 command line option is disabled.
3152
3153 6. Macro names set by the -D option must start with an upper case letter, just
3154 like macro names defined in the configuration file.
3155
3156 7. Added "dereference=" facility to LDAP.
3157
3158 8. Two instances of the typo "uknown" in the source files are fixed.
3159
3160 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
3161 the Configure-Makefile script screwed up while processing it.
3162
316310. Incorporated PCRE 4.4.
3164
316511. The SMTP synchronization check was not operating right at the start of an
3166 SMTP session. For example, it could not catch a HELO sent before the client
3167 waited for the greeting. There is now a check for outstanding input at the
3168 point when the greeting is written. Because of the duplex, asynchronous
3169 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
3170 way, but not yet received, when the check is performed.
3171
317212. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
3173 on TCP/IP sockets, because this apparently causes some broken clients to
3174 timeout.
3175
317613. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
3177 unchanged) from the Cygwin maintainer.
3178
317914. The code for -bV that shows what is in the binary showed "mbx" when maildir
3180 was supported instead of testing for mbx. Effectively a typo.
3181
318215. The spa authenticator server code was not checking that the input it
3183 received was valid base64.
3184
318516. The debug output line for the "set" modifier in ACLs was not showing the
3186 name of the variable that was being set.
3187
318817. Code tidy: the variable type "vtype_string" was never used. Removed it.
3189
319018. Previously, a reference to $sender_host_name did not cause a DNS reverse
3191 lookup on its own. Something else was needed to trigger the lookup. For
3192 example, a match in host_lookup or the need for a host name in a host list.
3193 Now, if $sender_host_name is referenced and the host name has not yet been
3194 looked up, a lookup is performed. If the lookup fails, the variable remains
3195 empty, and $host_lookup_failed is set to "1".
3196
319719. Added "eqi" as a case-independent comparison operator.
3198
319920. The saslauthd authentication condition could segfault if neither service
3200 nor realm was specified.
3201
320221. If an overflowing value such as "2048M" was set for message_size_limit, the
3203 error message that was logged was misleading, and incoming SMTP
3204 connections were dropped. The message is now more accurate, and temporary
3205 errors are given to SMTP connections.
3206
320722. In some error situations (such as 21 above) Exim rejects all SMTP commands
3208 (except RSET) with a 421 error, until QUIT is received. However, it was
3209 failing to send a response to QUIT.
3210
321123. The HELO ACL was being run before the code for helo_try_verify_hosts,
3212 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
3213 ACL is now run after the helo_try_verify_hosts code.
3214
321524. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
3216 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
3217 case-independent, so other case variants are also recognized). Apparently
3218 some systems use these upper case variants.
3219
322025. If more than two messages were waiting for the same host, and a transport
3221 filter was specified for the transport, Exim sent two messages over the
3222 same TCP/IP connection, and then failed with "socket operation on non-
3223 socket" when it tried to send the third.
3224
322526. Added Exim::debug_write and Exim::log_write for embedded Perl use.
3226
322727. The extern definition of crypt16() in expand.c was not being excluded when
3228 the OS had its own crypt16() function.
3229
323028. Added bounce_return_body as a new option, and bounce_return_size_limit
3231 as a preferred synonym for return_size_limit, both as an option and as an
3232 expansion variable.
3233
323429. Added LIBS=-liconv to OS/Makefile-OSF1.
3235
323630. Changed the default configuration ACL to relax the local part checking rule
3237 for addresses that are not in any local domains. For these addresses,
3238 slashes and pipe symbols are allowed within local parts, but the sequence
3239 /../ is explicitly forbidden.
3240
324131. SPA server authentication was not clearing the challenge buffer before
3242 using it.
3243
324432. log_message in a "warn" ACL statement was writing to the reject log as
3245 well as to the main log, which contradicts the documentation and doesn't
3246 seem right (because no rejection is happening). So I have stopped it.
3247
324833. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
3249 However, I am unable to do any testing of this.
3250
325134. Fixed an infelicity in the appendfile transport. When checking directories
3252 for a mailbox, to see if any needed to be created, it was accidentally
3253 using path names with one or more superfluous leading slashes; tracing
3254 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
3255
325635. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
3257 contents are added to the log line that is written for every discarded
3258 recipient. (Previously a log_message setting was ignored.)
3259
326036. The ${quote: operator now quotes the string if it is empty.
3261
326237. The install script runs exim in order to find its version number. If for
3263 some reason other than non-existence or emptiness, which it checks, it
3264 could not run './exim', it was installing it with an empty version number,
3265 i.e. as "exim-". This error state is now caught, and the installation is
3266 aborted.
3267
326838. An argument was missing from the function that creates an error message
3269 when Exim fails to connect to the socket for saslauthd authentication.
3270 This could cause Exim to crash, or give a corrupted message.
3271
327239. Added isip, isip4, and isip6 to ${if conditions.
3273
327440. The ACL variables $acl_xx are now saved with the message, and can be
3275 accessed later in routers, transports, and filters.
3276
327741. The new lookup type nwildlsearch is like wildlsearch, except that the key
3278 strings in the file are not string-expanded.
3279
328042. If a MAIL command specified a SIZE value that was too large to fit into an
3281 int variable, the check against message_size_limit failed. Such values are
3282 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
3283 day this will have to be increased, but I don't think I want to be around
3284 when emails are that large.
3285
3286
3287
3288Exim version 4.22
3289-----------------
3290
3291 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
3292 iconv() is not standard in FreeBSD.
3293
3294 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
3295 IPv6 enabled. The observed symptom was a segmentation fault on return from
3296 the function os_common_find_running_interfaces() in src/os.c.
3297
3298 3. In the check_special_case() function in daemon.c I had used "errno" as an
3299 argument name, which causes warnings on some systems. This was basically a
3300 typo, since it was named "eno" in the comments!
3301
3302 4. The code that waits for the clock to tick (at a resolution of some fraction
3303 of a second) so as to ensure message-id uniqueness was always waiting for
3304 at least one whole tick, when it could have waited for less. [This is
3305 almost certainly not relevant at current processor speeds, where it is
3306 unlikely to ever wait at all. But we try to future-proof.]
3307
3308 5. The function that sleeps for a time interval that includes fractions of a
3309 second contained a race. It did not block SIGALRM between setting the
3310 timer, and suspending (a couple of lines later). If the interval was short
3311 and the sigsuspend() was delayed until after it had expired, the suspension
3312 never ended. On busy systems this could lead to processes getting stuck for
3313 ever.
3314
3315 6. Some uncommon configurations may cause a lookup to happen in a queue runner
3316 process, before it forks any delivery processes. The open lookup caching
3317 mechanism meant that the open file or database connection was passed into
3318 the delivery process. The problem was that delivery processes always tidy
3319 up cached lookup data. This could cause a problem for the next delivery
3320 process started by the queue runner, because the external queue runner
3321 process does not know about the closure. So the next delivery process
3322 still has data in the lookup cache. In the case of a file lookup, there was
3323 no problem because closing a file descriptor in a subprocess doesn't affect
3324 the parent. However, if the lookup was caching a connection to a database,
3325 the connection was closed, and the second delivery process was likely to
3326 see errors such as "PGSQL: query failed: server closed the connection
3327 unexpectedly". The problem has been fixed by closing all cached lookups
3328 in a queue runner before running a delivery process.
3329
3330 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
3331 seem to have the "const" qualifier which it has on other OS. I've
3332 parameterised it.
3333
3334 8. Change 4.21/2 was too strict. It is only if there are two authenticators
3335 *of the same type* (client or server) with the same public name that an
3336 error should be diagnosed.
3337
3338 9. When Exim looked up a host name for an IP address, but failed to find the
3339 original IP address when looking up the host name (a safety check), it
3340 output the message "<ip address> does not match any IP for NULL", which was
3341 confusing, to say the least. The bug was that the host name should have
3342 appeared instead of "NULL".
3343
334410. Since release 3.03, if Exim is called by a uid other than root or the Exim
3345 user that is built into the binary, and the -C or -D options is used, root
3346 privilege is dropped before the configuration file is read. In addition,
3347 logging is switched to stderr instead of the normal log files. If the
3348 configuration then re-defines the Exim user, the unprivileged environment
3349 is probably not what is expected, so Exim logs a panic warning message (but
3350 proceeds).
3351
3352 However, if deliver_drop_privilege is set, the unprivileged state may well
3353 be exactly what is intended, so the warning has been cut out in that case,
3354 and Exim is allowed to try to write to its normal log files.
3355
3356
3357Exim version 4.21
3358-----------------
3359
3360 1. smtp_return_error_details was not giving details for temporary sender
3361 or receiver verification errors.
3362
3363 2. Diagnose a configuration error if two authenticators have the same public
3364 name.
3365
3366 3. Exim used not to create the message log file for a message until the first
3367 delivery attempt. This could be confusing when incoming messages were held
3368 for policy or load reasons. The message log file is now created at the time
3369 the message is received, and an initial "Received" line is written to it.
3370
3371 4. The automatically generated man page for command line options had a minor
3372 bug that caused no ill effects; however, a more serious problem was that
3373 the procedure for building the man page automatically didn't always
3374 operate. Consequently, release 4.20 contains an out-of-date version. This
3375 shouldn't happen again.
3376
3377 5. When building Exim with embedded Perl support, the script that builds the
3378 Makefile was calling 'perl' to find its compile-time parameters, ignoring
3379 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
3380
3381 6. The freeze_tell option was not being used for messages that were frozen on
3382 arrival, either by an ACL or by local_scan().
3383
3384 7. Added the smtp_incomplete_transaction log selector.
3385
3386 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
3387 was accepting AUTH without a new EHLO.
3388
3389 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
3390 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
3391 RFC.
3392
339310. Logging of TCP/IP connections (when configured) now happens in the main
3394 daemon process instead of the child process, so that the TCP/IP connection
3395 count is more accurate (but it can never be perfect).
3396
339711. The use of "drop" in a nested ACL was not being handled correctly in the
3398 outer ACL. Now, if condition failure induced by the nested "drop" causes
3399 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
3400 or "require"), the connection is dropped.
3401
340212. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
3403 that yield "discard" can now be used with an "accept" or a "discard" verb,
3404 but an error is generated for any others (because I can't see a useful way
3405 to define what should happen).
3406
340713. When an ACL is read dynamically from a file (or anywhere else), the lines
3408 are now processed in the same way as lines in the Exim configuration file.
3409 In particular, continuation lines are supported.
3410
341114. Added the "dnslists = a.b.c!=n.n.n.n" feature.
3412
341315. Added -ti meaning -t -i.
3414
341516. Check for letters, digits, hyphens, and dots in the names of dnslist
3416 domains, and warn by logging if others are found.
3417
341817. At least on BSD, alignment is not guarenteed for the array of ifreq's
3419 returned from GIFCONF when Exim is trying to find the list of interfaces on
3420 a host. The code in os.c has been modified to copy each ifreq to an aligned
3421 structure in all cases.
3422
3423 Also, in some cases, the returned ifreq's were being copied to a 'struct
3424 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
3425 means the last couple of bytes of an IPv6 address could be chopped if the
3426 ifreq contained only a normal sockaddr (14 bytes storage).
3427
342818. Named domain lists were not supported in the hosts_treat_as_local option.
3429 An entry such as +xxxx was not recognized, and was treated as a literal
3430 domain name.
3431
343219. Ensure that header lines added by a DATA ACL are included in the reject log
3433 if the ACL subsequently rejects the message.
3434
343520. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
3436 MD5 (which is deprecated).
3437
343821. When testing a filter file using -bf, Exim was writing a message when it
3439 took the sender from a "From " line in the message, but it was not doing so
3440 when it took $return_path from a Return-Path: header line. It now does.
3441
344222. If the contents of a "message" modifier for a "warn" ACL verb do not begin
3443 with a valid header line field name (a series of printing characters
3444 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
3445
344623. Changed "disc" in the source to "disk" to conform to the documentation and
3447 the book and for uniformity.
3448
344924. Ignore Sendmail's -Ooption=value command line item.
3450
345125. When execve() failed while trying to run a command in a pipe transport,
3452 Exim was returning EX_UNAVAILBLE (69) from the subprocess. However, this
3453 could be confused with a return value of 69 from the command itself. This
3454 has been changed to 127, the value the shell returns if it is asked to run
3455 a non-existent command. The wording for the related log line suggests a
3456 non-existent command as the problem.
3457
345826. If received_header_text expands to an empty string, do not add a Received:
3459 header line to the message. (Well, it adds a token one on the spool, but
3460 marks it "old" so that it doesn't get used or transmitted.)
3461
346227. Installed eximstats 1.28 (addition of -nt option).
3463
346428. There was no check for failure on the call to getsockname() in the daemon
3465 code. This can fail if there is a shortage of resources on the system, with
3466 ENOMEM, for example. A temporary error is now given on failure.
3467
346829. Contrary to the C standard, it seems that in some environments, the
3469 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
3470 program. Exim now does this explicitly; it affects the formatting of
3471 timestamps using strftime().
3472
347330. If exiqsumm was given junk data, it threw up some uninitialized variable
3474 complaints. I've now initialized all the variables, to avoid this.
3475
347632. Header lines added by a system filter were not being "seen" during
3477 transport-time rewrites.
3478
347933. The info_callback() function passed to OpenSSL is set up with type void
3480 (*)(SSL *, int, int), as described somewhere. However, when calling the
3481 function (actually a macro) that sets it up, the type void(*)() is
3482 expected. I've put in a cast to prevent warnings from picky compilers.
3483
348434. If a DNS black list lookup found a CNAME record, but there were no A
3485 records associated with the domain it pointed at, Exim crashed.
3486
348735. If a DNS black list lookup returned more than one A record, Exim ignored
3488 all but the first. It now scans all returned addresses if a particular IP
3489 value is being sought. In this situation, the contents of the
3490 $dnslist_value variable are a list of all the addresses, separated by a
3491 comma and a space.
3492
349336. Tightened up the rules for host name lookups using reverse DNS. Exim used
3494 to accept a host name and all its aliases if the forward lookup for any of
3495 them yielded the IP address of the incoming connection. Now it accepts only
3496 those names whose forward lookup yields the correct IP address. Any other
3497 names are discarded. This closes a loophole whereby a rogue DNS
3498 administrator could create reverse DNS records to break through a
3499 wildcarded host restriction in an ACL.
3500
350137. If a user filter or a system filter that ran in a subprocess used any of
3502 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
3503 the wrong values were passed to the pipe command ($thisaddress had the
3504 value of $0, $0 had the value of $1, etc). This bug was introduced by
3505 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
3506
350738. Improved the line breaking for long SMTP error messages from ACLs.
3508 Previously, if there was no break point between 40 and 75 characters, Exim
3509 left the rest of the message alone. Two changes have been made: (a) I've
3510 reduced the minimum length to 35 characters; (b) if it can't find a break
3511 point between 35 and 75 characters, it looks ahead and uses the first one
3512 that it finds. This may give the occasional overlong line, but at least the
3513 remaining text gets split now.
3514
351539. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
3516 file descriptors might be low, and that setting 1000 would always raise it.
3517 It turns out that in some environments, the limit is already over 1000 and
3518 that lowering it causes trouble. So now Exim takes care not to decrease it.
3519
352040. When delivering a message, the value of $return_path is set to $sender_
3521 address at the start of routing (routers may change the value). By an
3522 oversight, this default was not being set up when an address was tested by
3523 -bt or -bv, which affected the outcome if any router or filter referred to
3524 $return_path.
3525
352641. The idea of the "warn" ACL verb is that it adds a header or writes to the
3527 log only when "message" or "log_message" are set. However, if one of the
3528 conditions was an address verification, or a call to a nested ACL, the
3529 messages generated by the underlying test were being passed through. This
3530 no longer happens. The underlying message is available in $acl_verify_
3531 message for both "message" and "log_message" expansions, so it can be
3532 passed through if needed.
3533
353442. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
3535 new expansion $bh_ to give the encoded byte string without charset
3536 translation. Translation happens only if iconv() is available; HAVE_ICONV
3537 indicates this at build time. HEADERS_CHARSET gives the charset to
3538 translate to; headers_charset can change it in the configuration, and
3539 "headers charset" can change it in an individual filter file.
3540
354143. Now that we have a default RFC 2047 charset (see above), the code in Exim
3542 that creates RFC 2047 encoded "words" labels them as that charset instead
3543 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
3544 expansion operator; (ii) when Exim creates a From: line for a local
3545 message; (iii) when a header line is rewritten to include a "phrase" part.
3546
354744. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
3548 buggy, causing it to skip the first lines of messages whose message ID
3549 ended in 'D'. This would not have bitten before Exim release 4.14, because
3550 message IDs were unlikely to end in 'D' before then. The effect was to have
3551 incorrect size information for certain domains.
3552
355345. #include "config.h" was missing at the start of the crypt16.c module. This
3554 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
3555 noticed.
3556
355746. If there was a timeout during a "random" callout check, Exim treated it as
3558 a failure of the random address, and carried on sending RSET and the real
3559 address. If the delay was just some slowness somewhere, the response to the
3560 original RCPT would be taken as a response to RSET and so on, causing
3561 mayhem of various kinds.
3562
356347. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
3564 when I implemented it. It didn't allow for the fact that some option values
3565 may legitimatetly be negative (e.g. size_addition), and it didn't even do
3566 the right test for positive values.
3567
356848. Domain names in DNS records are case-independent. Exim always looks them up
3569 in lower case. Some resolvers return domain names in exactly the case they
3570 appear in the zone file, that is, they may contain uppercase letters. Not
3571 all resolvers do this - some return always lower case. Exim was treating a
3572 change of case by a resolver as a change of domain, similar to a widening
3573 of a domain abbreviation. This triggered its re-routing code and so it was
3574 trying to route what was effectively the same domain again. This normally
3575 caused routing to fail (because the router wouldn't handle the domain
3576 twice). Now Exim checks for this case specially, and just changes the
3577 casing of the domain that it ultimately uses when it transmits the message
3578 envelope.
3579
358049. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
3581 module.
3582
358350. If a filter generated a file delivery with a non-absolute name (possible if
3584 no home directory exists for the router), the forbid_file option was not
3585 forbidding it.
3586
358751. Added '&' feature to dnslists, to provide bit mask matching in addition to
3588 the existing equality matching.
3589
359052. Exim was using ints instead of ino_t variables in some places where it was
3591 dealing with inode numbers.
3592
359353. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
3594 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
3595 called TMPDIR, and if it finds it is different, it changes its value.
3596
359754. The smtp_printf() function is now made available to local_scan() so
3598 additional output lines can be written before returning. There is also an
3599 smtp_fflush() function to enable the detection of a dropped connection.
3600 The variables smtp_input and smtp_batched_input are exported to
3601 local_scan().
3602
360355. Changed the default runtime configuration: the message "Unknown user"
3604 has been removed from the ACL, and instead placed on the localuser router,
3605 using the cannot_route_message feature. This means that any verification
3606 failures that generate their own messages won't get overridden. Similarly,
3607 the "Unrouteable address" message that was in the ACL for unverifiable
3608 relay addresses has also been removed.
3609
361056. Added hosts_avoid_esmtp to the smtp transport.
3611
361257. The exicyclog script was not checking for the esoteric option
3613 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
3614 will work only if exicyclog is run under the appropriate euid.
3615
361658. Following a discussion on the list, the rules by which Exim recognises line
3617 endings on incoming messages have been changed. The -dropcr and drop_cr
3618 options are now no-ops, retained only for backwards compatibility. The
3619 following line terminators are recognized: LF CRLF CR. However, special
3620 processing applies to CR:
3621
3622 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
3623 nor a local message in the state where . is a terminator.
3624
3625 (ii) If a bare CR is encountered in a header line, an extra space is added
3626 after the line terminator so as not to end the header. The reasoning
3627 behind this is that bare CRs in header lines are most likely either
3628 to be mistakes, or people trying to play silly games.
3629
363059. The size of a message, as listed by "-bp" or in the Exim monitor window,
3631 was being incorrectly given as 18 bytes larger than it should have been.
3632 This is a VOB (very old bug).
3633
363460. This may never have affected anything current, but just in case it has:
3635 When the local host is found other than at the start of a list of hosts,
3636 the local host, those with the same MX, and any that follow, are discarded.
3637 When the list in question was part of a longer list of hosts, the following
3638 hosts (not currently being processed) were also being discarded. This no
3639 longer happens. I'm not sure if this situation could ever has previously
3640 arisen.
3641
364261. Added the "/MX" feature to lists of hosts in the manualroute and query
3643 program routers.
3644
364562. Whenever Exim generates a new message, it now adds an Auto-Submitted:
3646 header. This is something that is recommended in a new Internet Draft, and
3647 is something that is documented as being done by Sendmail. There are two
3648 possible values. For messages generated by the autoreply transport, Exim
3649 adds:
3650
3651 Auto-Submitted: auto-replied
3652
3653 whereas for all other generated messages (e.g. bounces) it adds
3654
3655 Auto-Submitted: auto-generated
3656
365763. The "personal" condition in filters now includes a test for the
3658 Auto-Submitted: header. If it contains the string "auto-" the message it
3659 not considered personal.
3660
366164. Added rcpt_include_affixes as a generic transport option.
3662
366365. Added queue_only_override (default true).
3664
366566. Added the syslog_duplication option.
3666
366767. If what should have been the first header line of a message consisted of
3668 a space followed by a colon, Exim was mis-interpreting it as a header line.
3669 It isn't of course - it is syntactically invalid and should therefore be
3670 treated as the start of the message body. The misbehaviour could have
3671 caused a number of strange effects, including loss of data in subsequent
3672 header lines, and spool format errors.
3673
367468. Formerly, the AUTH parameter on a MAIL command was trusted only if the
3675 client host had authenticated. This control can now be exercised by an ACL
3676 for more flexibility.
3677
367869. By default, callouts do not happen when testing with -bh. There is now a
3679 variant, -bhc, which does actually run the callout code, including
3680 consulting and updating the callout ca