Add log selector sender_verify_fail.
[exim.git] / doc / doc-txt / ChangeLog
CommitLineData
278c6e6c 1$Cambridge: exim/doc/doc-txt/ChangeLog,v 1.299 2006/02/14 14:55:37 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
214e2000 165
5de37277
PH
166Exim version 4.60
167-----------------
168
cc38ddbf
PH
169PH/01 Two changes to the default runtime configuration:
170
171 (1) Move the checks for relay_from_hosts and authenticated clients from
172 after to before the (commented out) DNS black list checks.
173
174 (2) Add control=submission to the relay_from_hosts and authenticated
175 clients checks, on the grounds that messages accepted by these
176 statements are most likely to be submissions.
5de37277 177
72fdd6ae
PH
178PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
179
180 (1) Generate an error if the third argument for the ${prvs expansion is
181 not a single digit.
182
183 (2) Treat a missing third argument of ${prvscheck as if it were an empty
184 string.
185
186 (3) Reset the variables that are obtained from the first argument of
187 ${prvscheck and used in the second argument before leaving the code,
188 because their memory is reclaimed, so using them afterwards may do
189 silly things.
190
191 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
192 one (it's much easier than Tom thought :-).
193
194 (5) Because of (4), we can now allow for the use of $prvscheck_result
195 inside the third argument.
cb9328de 196
cb741023
PH
197PH/03 For some reason, the default setting of PATH when running a command from
198 a pipe transport was just "/usr/bin". I have changed it to
199 "/bin:/usr/bin".
200
f174f16e
PH
201PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
202 anything to be listed in the output from -bV.
b2f5a032 203
c25242d7
PH
204PH/05 When a filter generated an autoreply, the entire To: header line was
205 quoted in the delivery log line, like this:
206
207 => >A.N.Other <ano@some.domain> <original@ddress> ...
208
209 This has been changed so that it extracts the operative address. There
210 may be more than one such address. If so, they are comma-separated, like
211 this:
212
213 => >ano@some.domain,ona@other.domain <original@ddress> ...
214
82c19f95
PH
215PH/06 When a client host used a correct literal IP address in a HELO or EHLO
216 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
217 not being looked up in the rDNS to get a host name, Exim was showing the
218 IP address twice in Received: lines, even though the IP addresses were
219 identical. For example:
220
221 Received: from [1.2.3.4] (helo=[1.2.3.4])
222
223 However, if the real host name was known, it was omitting the HELO data
224 if it matched the actual IP address. This has been tidied up so that it
225 doesn't show the same IP address twice.
226
d7ffbc12
PH
227PH/07 When both +timestamp and +memory debugging was on, the value given by
228 $tod_xxx expansions could be wrong, because the tod_stamp() function was
229 called by the debug printing, thereby overwriting the timestamp buffer.
230 Debugging no longer uses the tod_stamp() function when +timestamp is set.
231
9f526266
PH
232PH/08 When the original message was included in an autoreply transport, it
233 always said "this is a copy of the message, including all the headers",
234 even if body_only or headers_only was set. It now gives an appropriate
235 message.
236
87fcc8b9
PH
237PH/09 Applied a patch from the Sieve maintainer which:
238
239 o fixes some comments
240 o adds the (disabled) notify extension core
241 o adds some debug output for the result of if/elsif tests
242 o points to the current vacation draft in the documentation
243 and documents the missing references header update
244
245 and most important:
246
247 o fixes a bug in processing the envelope test (when testing
248 multiple envelope elements, the last element determinted the
249 result)
250
456682f5
PH
251PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
252 Electronic Mail") by including:
253
254 Auto-submitted: auto-generated
255
256 in the messages that it generates (bounce messages and others, such as
257 warnings). In the case of bounce messages for non-SMTP mesages, there was
258 also a typo: it was using "Auto_submitted" (underscore instead of
259 hyphen). Since every message generated by Exim is necessarily in response
260 to another message, thes have all been changed to:
261
262 Auto-Submitted: auto-replied
263
264 in accordance with these statements in the RFC:
265
266 The auto-replied keyword:
267
268 - SHOULD be used on messages sent in direct response to another
269 message by an automatic process,
270
271 - MUST NOT be used on manually-generated messages,
272
273 - MAY be used on Delivery Status Notifications (DSNs) and Message
274 Disposition Notifications (MDNs),
275
276 - MUST NOT be used on messages generated by automatic or periodic
277 processes, except for messages which are automatic responses to
278 other messages.
279
3e46c1aa
PH
280PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
281 to the default Received: header definition.
456682f5 282
49826d12
PH
283PH/12 Added log selector acl_warn_skipped (default on).
284
eba0c039
PH
285PH/13 After a successful wildlsearch lookup, discard the values of numeric
286 variables because (a) they are in the wrong storage pool and (b) even if
287 they were copied, it wouldn't work properly because of the caching.
288
a0d6ba8a
PH
289PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
290 checking when decoding. Apparently there are clients that generate
291 overlong encoded strings. Why am I not surprised?
292
f0917727
PH
293PH/15 If the first argument of "${if match_address" was not empty, but did not
294 contain an "@" character, Exim crashed. Now it writes a panic log message
295 and treats the condition as false.
296
096fee00
PH
297PH/16 In autoreply, treat an empty string for "once" the same as unset.
298
024bd3c2
PH
299PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
300 extension "envelope-auth". The code is finished and in agreement with
301 other implementations, but there is no documentation so far and in fact,
302 nobody wrote the draft yet. This extension is currently #undef'ed, thus
303 not changing the active code.
304
305 Print executed "if" and "elsif" statements when debugging is used. This
306 helps a great deal to understand what a filter does.
307
308 Document more things not specified clearly in RFC3028. I had all this
309 sorted out, when out of a sudden new issues came to my mind. Oops."
310
df199fec
PH
311PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
312 (Bugzilla #53).
313
d27f1df3
PH
314PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
315 canonical form (as documented). However, after a host name lookup from
316 the IP address, check_host() was doing a simple string comparison with
317 addresses acquired from the DNS when checking that the found name did
318 have the original IP as one of its addresses. Since any found IPv6
319 addresses are likely to be in abbreviated form, the comparison could
320 fail. Luckily, there already exists a function for doing the comparison
321 by converting both addresses to binary, so now that is used instead of
322 the text comparison.
323
96776534
PH
324PH/20 There was another similar case to PH/19, when a complete host name was
325 given in a host list; looking up its IP address could give an abbreviated
326 form, whereas the current host's name might or might not be abbreviated.
327 The same fix has been applied.
328
5de37277 329
9a799bc0
PH
330Exim version 4.54
331-----------------
332
333PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
334 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
335 It now does.
336
99a4b039
PH
337PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
338 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
339
340PH/03 Typo: missing ".o" in src/pcre/Makefile.
341
4b233853
PH
342PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
343 header line, restrict the check to what is listed in RFCs 2369 and 2929.
344 Also, for "Auto-Submitted", treat anything other than "no" as
345 non-personal, in accordance with RFC 3834. (Previously it treated
346 anything starting "auto-" as non-personal.)
347
8857ccfd
PH
348TF/01 The control=submission/name=... option had a problem with syntax
349 errors if the name included a slash character. The /name= option
350 now slurps the rest of the string, so it can include any characters
351 but it must come last in the list of options (after /sender_retain
352 or /domain=).
353
433a2980
PH
354PH/05 Some modifications to the interface to the fake nameserver for the new
355 testing suite.
356
3e46c1aa 357
9a799bc0 358
e3a311ba
TK
359Exim version 4.53
360-----------------
361
362TK/01 Added the "success_on_redirect" address verification option. See
363 NewStuff for rationale and an example.
364
13b685f9
PH
365PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
366
395ff96d
PH
367PH/02 Patch to exigrep to allow it to work on syslog lines.
368
5b68f6e4
PH
369PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
370 fread() to skip over the body file's header line, because in Cygwin the
371 header line is locked and is inaccessible.
372
1ab52c69
PH
373PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
374 co-exist for some time) to make it clear that it is the Exim ID that is
375 referenced, not the Message-ID: header line.
376
b07e6aa3
PH
377PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
378 string_format() function, because snprintf() does not exist on all
379 operating systems.
380
254e032f
PH
381PH/06 The use of forbid_filter_existstest now also locks out the use of the
382 ${stat: expansion item.
383
3af76a81
PH
384PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
385 protocol synchronization error", to keep the pedants happy.
386
2548ba04
PH
387PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
388 well as for IRIX systems, when gcc is being used. See the host.c source
389 file for comments.
390
b6c6011d
PH
391PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
392
cf39cf57
PH
393PH/10 Named domain lists were not working if used in a queue_smtp_domains
394 setting.
395
f1513293
PH
396PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
397 transport and to the smtp transport in LMTP mode.
398
727549a4
PH
399TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
400
af46795e
PH
401PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
402 run a filter in a subprocess. This could lead to confusion in subsequent
403 lookups in the parent process. There should also be a search_tidyup() at
404 the end of the subprocess.
405
d7b47fd0
PH
406PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
407 only if the host matched helo_try_verify_hosts, which caused the
408 verification to occur when the EHLO/HELO command was issued. The ACL just
409 tested the remembered result. Now, if a previous verification attempt has
410 not happened, "verify = helo" does it there and then.
411
ee744174
JJ
412JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
413
b582ab87
PH
414TK/03 Fix log output including CR from clamd.
415
41a13e0a
PH
416PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
417 exist provoked a memory error which could cause a segfault.
418
f625cc5a
PH
419PH/15 Installed PCRE 6.2
420
421PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
422
21f7af35
PH
423PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
424 of the problem. Specifically, suggested +O2 rather than +O1 for the
425 HP-UX compiler.
426
31480e42
PH
427PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
428
2d280592
PH
429PH/20 If a delivery was routed to a non-standard port by means of an SRV
430 record, the port was not correctly logged when the outgoing_port log
431 selector was set (it logged the transort's default port).
432
7cd1141b
PH
433PH/21 Added support for host-specific ports to manualroute, queryprogram,
434 fallback_hosts, and "hosts" in the smtp transport.
435
436PH/22 If the log selector "outgoing_port" is set, the port is now also given on
437 host errors such as "Connection refused".
438
750af86e
PH
439PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
440 authentication with radiusclient 0.4.9:
441
442 - Error returned from rc_read_config was caught wrongly
443 - Username/password not passed on to radius server due to wrong length.
444
445 The presumption is that some radiusclient API changes for 4.51/PH/17
446 were not taken care of correctly. The code is still untested by me (my
447 Linux distribution still has 0.3.2 of radiusclient), but it was
448 contributed by a Radius user.
449
450PH/24 When doing a callout, the value of $domain wasn't set correctly when
451 expanding the "port" option of the smtp transport.
452
4304270b
TK
453TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
454 while reading a MIME header. Thanks to Tom Hughes for a patch.
455
750af86e
PH
456PH/24 Include config.h inside local_scan.h so that configuration settings are
457 available.
458
64ffc24f
PH
459PH/25 Make $smtp_command_argument available after all SMTP commands. This means
460 that in an ACL for RCPT (for example), you can examine exactly what was
461 received.
462
5dd9625b
PH
463PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
464 commands, but it was not correctly comparing the address with the actual
465 client host address. Thus, it would show the EHLO address in Received:
466 header lines when this was not necessary.
467
5591031b
PH
468PH/27 Added the % operator to ${eval:}.
469
ba18e66a
PH
470PH/28 Exim tries to create and chdir to its spool directory when it starts;
471 it should be ignoring failures (because with -C, for example, it has lost
472 privilege). It wasn't ignoring creation failures other than "already
473 exists".
474
9cec981f
PH
475PH/29 Added "crypteq" to the list of supported features that Exim outputs when
476 -bV or -d is used.
477
aa2b5c79
PH
478PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
479 because an input line was too long, either on its own, or by virtue of
1509d3a8
PH
480 too many continuations, the temporary file was not being removed, and the
481 return code was incorrect.
aa2b5c79 482
48a53b7f
PH
483PH/31 Missing "BOOL" in function definition in filtertest.c.
484
1c59d63b
PH
485PH/32 Applied Sieve patches from the maintainer.
486
671012da
TK
487TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
488
1509d3a8
PH
489PH/33 Added "verify = not_blind".
490
491PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
492 Local/Makefile (with some defaults set). These are used in built scripts
493 such as exicyclog, but they have never been used in the exim_install
494 script (though there are many overriding facilities there). I have
495 arranged that the exim_install script now takes note of these two
496 settings.
497
498PH/35 Installed configuration files for Dragonfly.
499
2fe1a124
PH
500PH/36 When a locally submitted message by a trusted user did not contain a
501 From: header, and the sender address was obtained from -f or from an SMTP
502 MAIL command, and the trusted user did not use -F to supply a sender
503 name, $originator_name was incorrectly used when constructing a From:
504 header. Furthermore, $originator_name was used for submission mode
505 messages from external hosts without From: headers in a similar way,
506 which is clearly wrong.
507
8800895a
PH
508PH/37 Added control=suppress_local_fixups.
509
ccfdb010
PH
510PH/38 When log_selector = +received_sender was set, and the addition of the
511 sender made the log line's construction buffer exactly full, or one byte
512 less than full, an overflow happened when the terminating "\n" was
513 subsequently added.
514
1130bfb0
PH
515PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
516 when the result of a list match is failure because a DNS lookup failed.
517
ebcb507f
PH
518PH/40 RM_COMMAND is now used in the building process.
519
c35e155c
PH
520PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
521 the "build-* directories that it finds.
522
95d1f782
PH
523PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
524 address in a domain literal was a prefix of an interface address.
525
fd6de02e
PH
526PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
527 when verifying a sender address, unless rewrite_headers is false.
528
58de37c5
PH
529PH/44 Wrote a long comment about why errors_to addresses are verified as
530 recipients, not senders.
531
261cf466
TF
532TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
533 the ratelimit ACL was added.
534
3ee512ff
PH
535PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
536
e08c430f
PH
537PH/46 Added extra information about PostgreSQL errors to the error string.
538
bef5a11f
PH
539PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
540 avoiding the need to install special zones in a real server. This is
541 backwards compatible; if it can't find the fake resolver, it drops back.
542 Thus, both old and new test suites can be run.
543
7546de58
TF
544TF/02 Added util/ratelimit.pl
545
e5d5a95f
TF
546TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
547 clock is set back in time.
548
2e88a017
TF
549TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
550 Candler <B.Candler@pobox.com>.
551
a5f65aa4
TF
552TF/05 The fix for PH/43 was not completely correct; widen_domains is always
553 OK for addresses that are the result of redirections.
554
e7726cbf
PH
555PH/48 A number of further additions for the benefit of the new test suite,
556 including a fake gethostbyname() that interfaces to the fake DNS resolver
557 (see PH/47 above).
558
a7fdad5b
TF
559TF/06 The fix for widen_domains has also been applied to qualify_single and
560 search_parents which are the other dnslookup options that can cause
561 header rewrites.
562
6af56900
PH
563PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
564 type ("H").
565
0925ede6
PH
566PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
567
66afa403
TF
568TF/07 Exim produced the error message "an SRV record indicated no SMTP
569 service" if it encountered an MX record with an empty target hostname.
570 The message is now "an MX or SRV record indicated no SMTP service".
571
0154e85a
TF
572TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
573 if the DNS of the sending site is misconfigured. This is quite a
574 common situation. This change restores the behaviour of treating a
575 helo verification defer as a failure.
576
16f12c76
PH
577PH/51 If self=fail was set on a router, the bounce message did not include the
578 actual error message.
579
bbe902f0 580
e5a9dba6
PH
581Exim version 4.52
582-----------------
583
584TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
585
22c3b60b
PH
586PH/01 When a transport filter timed out in a pipe delivery, and the pipe
587 command itself ended in error, the underlying message about the transport
588 filter timeout was being overwritten with the pipe command error. Now the
589 underlying error message should be appended to the second error message.
590
06a9b4b5
PH
591TK/01 Fix poll() being unavailable on Mac OSX 10.2.
592
c1ac6996
PH
593PH/02 Reduce the amount of output that "make" produces by default. Full output
594 can still be requested.
595
9c7a242c
PH
596PH/03 The warning log line about a condition test deferring for a "warn" verb
597 was being output only once per connection, rather than after each
598 occurrence (because it was using the same function as for successful
599 "warn" verbs). This seems wrong, so I have changed it.
600
87ba3f5f
PH
601TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
602 it should not have, which might have caused a crash in the right
603 circumstances, but probably never did.
604
605PH/04 Installed a modified version of Tony Finch's patch to make submission
606 mode fix the return path as well as the Sender: header line, and to
607 add a /name= option so that you can make the user's friendly name appear
608 in the header line.
609
29aba418
TF
610TF/03 Added the control = fakedefer ACL modifier.
611
fe0dab11
TF
612TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
613 Mark Lowes for thorough testing.
870f6ba8 614
11d337a4
TK
615TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
616
617TK/03 Merged latest SRS patch from Miles Wilton.
618
415c8f3b
PH
619PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
620 with the definition in sysexits.h (which is #included earlier).
621 Fortunately, Exim does not actually use EX_OK. The code used to try to
622 preserve the sysexits.h value, by assumimg that macro definitions were
623 scanned for macro replacements. I have been disabused of this notion,
624 so now the code just undefines EX_OK before #including unistd.h.
11d337a4 625
958541e9
PH
626PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
627 in the smtp transport. When a block could not be written in a single
628 write() function, the timeout was being re-applied to each part-write.
629 This seems wrong - if the receiver was accepting one byte at a time it
630 would take for ever. The timeout is now adjusted when this happens. It
631 doesn't have to be particularly precise.
632
c206415f
TK
633TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
634 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
635
2a4be8f9
PH
636PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
637 without a domain if the check to <postmaster@domain> fails.
638
1cba11c5
SC
639SC/01 Eximstats: added -xls and the ability to specify output files
640 (patch written by Frank Heydlauf).
641
642SC/02 Eximstats: use FileHandles for outputing results.
643
644SC/03 Eximstats: allow any combination of xls, txt, and html output.
645
646SC/04 Eximstats: fixed display of large numbers with -nvr option
647
648SC/05 Eximstats: fixed merging of reports with empty tables.
649
650SC/06 Eximstats: added the -include_original_destination flag
651
652SC/07 Eximstats: removed tabs and trailing whitespace.
653
1005d00e
TK
654TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
655
656TK/06 MBOX spool code: Add real "From " MBOX separator line
657 so the .eml file is really in mbox format (even though
658 most programs do not really care). Patch from Alex Miller.
659
660TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
661 The latter is generated from $received_to and is only set if the
662 message has one envelope recipient. SA can use these headers,
663 obviously out-of-the-box. Patch from Alex Miller.
664
9b4768fa
PH
665PH/08 The ${def test on a variable was returning false if the variable's
666 value was "0", contrary to what the specification has always said!
667 The result should be true unless the variable is empty.
668
669PH/09 The syntax error of a character other than { following "${if
670 def:variable_name" (after optional whitespace) was not being diagnosed.
671 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
672 accidental colon was present, for example, could give incorrect results.
673
0d7eb84a
PH
674PH/10 Tidied the code in a number of places where the st_size field of a stat()
675 result is used (not including appendfile, where other changes are about
676 to be made).
677
678PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
679 This involved changing a lot of size variables from int to off_t. It
680 should work with maildirs and everything.
681
40727bee
TK
682TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
683 spamd dying while we are connected to it.
684
554d2369
TF
685TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
686 <jgh@wizmail.org>
687
1f922db1
PH
688PH/12 Applied Alex Kiernan's patch for the API change for the error callback
689 function for BDB 4.3.
690
ef213c3b
PH
691PH/13 Changed auto_thaw such that it does not apply to bounce messages.
692
8ac170f3
PH
693PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
694 the sources for PCRE have been re-arranged and more files are now
695 involved.
696
b1c749bb
PH
697PH/15 The code I had for printing potentially long long variables in PH/11
698 above was not the best (it lost precision). The length of off_t variables
699 is now inspected at build time, and an appropriate printing format (%ld
c6c2dc1d
PH
700 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
701 to be "long long int" or "long int". This is needed for the internal
702 formatting function string_vformat().
b1c749bb 703
4aac9b49
PH
704PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
705 the configuration file to be ":syslog", then the script "guesses" where
706 the logs files are, rather than using the compiled in default. In our
707 case the guess is not the same as the compiled default, so the script
708 suddenly stopped working when I started to use syslog. The patch checks
709 to see if log_file_path is "". If so, it attempts to read it from exim
710 with no configuration file to get the compiled in version, before it
711 falls back to the previous guessing code."
712
294520c8
TK
713TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
714 implementing BATV in an Exim configuration. See NewStuff for the gory
715 details.
716
5bd022fe
PH
717PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
718 Makefile that are specific to HP-UX.
719
90e9ce59
PH
720PH/18 If the "use_postmaster" option was set for a recipient callout together
721 with the "random" option, the postmaster address was used as the MAIL
722 FROM address for the random test, but not for the subsequent recipient
723 test. It is now used for both.
724
5ea81592
PH
725PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
726 patch removes a few documentation additions to RFC 3028, because the
727 latest draft now contains them. It adds the new en;ascii-case comparator
728 and a new error check for 8bit text in MIME parts. Comparator and
729 require names are now matched exactly. I enabled the subaddress
730 extension, but it is not well tested yet (read: it works for me)."
731
c6c2dc1d
PH
732PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
733 rework some of the code of TK/09 above to avoid the hardwired use of
734 "%lld" and "long long". Replaced the call to snprintf() with a call to
735 string_vformat().
736
fffffe4c
PH
737PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
738 records point to non-existent hosts", "retry timeout exceeded", and
739 "retry time not reached for any host after a long failure period".
ca02eafb 740
9a26b6b2
PH
741PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
742 experimental DomainKeys support:
743
744 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
745 (2) On an error such as an illegally used "control", the wrong name for
746 the control was given.
747
748 These problems did NOT occur unless DomainKeys support was compiled.
749
4aee0225
PH
750PH/23 Added daemon_startup_retries and daemon_startup_sleep.
751
32d668a5
PH
752PH/24 Added ${if match_ip condition.
753
8187c3f3
PH
754PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
755 databases so that it will be absolutely obvious if a crash occurs in the
756 DB library. This is a regular occurrence (often caused by mis-matched
757 db.h files).
758
ff790e47 759PH/26 Insert a lot of missing (void) casts for functions such as chown(),
f1e894f3
PH
760 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
761 picked up on a user's system that detects such things. There doesn't seem
762 to be a gcc warning option for this - only an attribute that has to be
763 put on the function's prototype. It seems that in Fedora Core 4 they have
764 set this on a number of new functions. No doubt there will be more in due
765 course.
ff790e47 766
5417f6d1
PH
767PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
768 specify a transport. However, if an address that was verified by such a
769 router was the subject of a callout, Exim crashed because it tried to
770 read the rcpt_include_affixes from the non-existent transport. Now it
771 just assumes that the setting of that option is false. This bug was
772 introduced by 4.51/PH/31.
773
59cf8544
PH
774PH/28 Changed -d+all to exclude +memory, because that information is very
775 rarely of interest, but it makes the output a lot bigger. People tend to
776 do -d+all out of habit.
777
e7ad8a65
PH
778PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
779 code in os-type was giving problems when libc.so lives in lib64, like on
780 x86_64 Fedora Core.
781
ade42478
PH
782PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
783 aren't the modern standard, and it seems that some systems' include files
784 don't always have them. Exim was already checking for some of the newer
785 ones like T_AAAA, and defining it itself. I've added checks for all the
786 record types that Exim uses.
787
182ad5cf
PH
788PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
789 not automatically generating a new one, as it is supposed to. This
790 prevented TLS from working. If the file did exist, but contained invalid
791 data, a new version was generated, as expected. It was only the case of a
792 non-existent file that was broken.
793
b0d9fc80
TK
794TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
795 with a change in libdomainkeys > 0.64.
796
797TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
798 from DNS. If the selector record carries the flag, it now has
799 precedence over the domain-wide flag.
800
801TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
802
47c7a64a
PH
803PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
804 the use of a port name that isn't defined in /etc/services) occurred, the
805 message was deferred as in a normal delivery, and thus remained on the
806 spool, instead of being failed because of the mua_wrapper setting. This
807 is now fixed, and I tidied up some of the mua_wrapper messages at the
808 same time.
809
a388bce4
SC
810SC/08 Eximstats: whilst parsing the mainlog(s), store information about
811 the messages in a hash of arrays rather than using individual hashes.
812 This is a bit cleaner and results in dramatic memory savings, albeit
813 at a slight CPU cost.
814
815SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
816 as requested by Marc Sherman.
817
818SC/10 Eximstats: added histograms for user specified patterns as requested
819 by Marc Sherman.
820
0793e4ed
SC
821SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
822
c58b88df
PH
823PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
824 fopen() in the content-scanning modules that did not already have it.
825
e7ad8a65 826
7982096b
PH
827Exim version 4.51
828-----------------
829
1a46a8c5
PH
830TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
831 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
832
2f079f46 833TK/02 Fix ACL "control" statement not being available in MIME ACL.
1a46a8c5
PH
834
835TK/03 Fix ACL "regex" condition not being available in MIME ACL.
836
837PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
838 to test Sieve filters that use "vacation".
839
840PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
841 that changes the way the GnuTLS parameters are stored in the cache file.
842 The new format can be generated externally. For backward compatibility,
843 if the data in the cache doesn't make sense, Exim assumes it has read an
844 old-format file, and it generates new data and writes a new file. This
845 means that you can't go back to an older release without removing the
846 file.
847
848PH/03 A redirect router that has both "unseen" and "one_time" set does not
849 work if there are any delivery delays because "one_time" forces the
850 parent to be marked "delivered", so its unseen clone is never tried
851 again. For this reason, Exim now forbids the simultaneous setting of
852 these two options.
853
854PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
855 redirected to themselves ("homonym" addresses). Read the long ChangeLog
856 entry if you want to know the details. The fix, however, neglected to
857 consider the case when local delivery batching is involved. The test for
858 "previously delivered" was not happening when checking to see if an
859 address could be batched with a previous (undelivered) one; under
860 certain circumstances this could lead to multiple deliveries to the same
c2c19e9d 861 address.
1a46a8c5
PH
862
863PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
864 in its include files, and this causes problems building Exim.
865
866PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
867 header_syntax) but Exim was just ignoring anything given after a slash.
868 In particular, this caused confusion with an attempt to use "verify =
869 reverse_host_lookup/defer_ok". An error is now given when options are
870 supplied for verify items that do not have them. (Maybe reverse_host_
871 lookup should have a defer_ok option, but that's a different point.)
872
873PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
874 defined by RFC 821) to 2048, because there were problems with some AUTH
875 commands, and RFC 1869 says the size should be increased for extended
876 SMTP commands that take arguments.
877
878PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
879 Finch).
880
881PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
882 "unknown" error; now it says that the functionality isn't in the binary.
8d67ada3 883
49c2d5ea
PH
884PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
885 an address' error message when a string expansion fails (syntax or
f331f3b6
PH
886 whatever). Otherwise the password may appear in the log. Following change
887 PH/42 below, there is no longer a chance of it appearing in a bounce
888 message.
49c2d5ea 889
bf759a8b
PH
890PH/11 Installed exipick version 20050225.0 from John Jetmore.
891
83364d30
PH
892PH/12 If the last host in a fallback_hosts list was multihomed, only the first
893 of its addresses was ever tried. (Bugzilla bug #2.)
894
7999bbd7
PH
895PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
896 the result incorrectly in the debug output. (It correctly added a newline
897 to what was transported.)
898
7dbf77c9
PH
899TF/01 Added $received_time.
900
74e0617f
PH
901PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
902 commented out examples of how to interface to a virus scanner and to
903 SpamAssassin. Also added commented examples of av_scanner and
904 spamd_address settings.
905
2f079f46
PH
906PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
907 and controls are allowed in which ACLs. There were a couple of minor
908 errors. Some of the entries in the conditions table (which is a table of
909 where they are NOT allowed) were getting very unwieldy; rewrote them as a
910 negation of where the condition IS allowed.
911
8c841523
PH
912PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
913
7766a4f0
PH
914PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
915 header file does not have a version number, so I've had to invent a new
916 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
917 API. The code is untested by me (my Linux distribution still has 0.3.2 of
918 radiusclient), but it was contributed by a Radius user.
919
8b417f2c
PH
920PH/18 Installed Lars Mainka's patch for the support of CRL collections in
921 files or directories, for OpenSSL.
922
901f42cb
PH
923PH/19 When an Exim process that is running as root has to create an Exim log
924 file, it does so in a subprocess that runs as exim:exim so as to get the
925 ownership right at creation (otherwise, other Exim processes might see
926 the file with the wrong ownership). There was no test for failure of this
927 fork() call, which would lead to the process getting stuck as it waited
928 for a non-existent subprocess. Forks do occasionally fail when resources
929 run out. I reviewed all the other calls to fork(); they all seem to check
930 for failure.
931
f9b9210e
PH
932PH/20 When checking for unexpected SMTP input at connect time (before writing
933 the banner), Exim was not dealing correctly with a non-positive return
934 from the read() function. If the client had disconnected by this time,
935 the result was a log entry for a synchronization error with an empty
936 string after "input=" when read() returned zero. If read() returned -1
937 (an event I could not check), uninitialized data bytes were printed.
938 There were reports of junk text (parts of files, etc) appearing after
939 "input=".
940
54cdb463
PH
941PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
942
cf00dad6
PH
943PH/22 Added support for macro redefinition, and (re)definition in between
944 driver and ACL definitions.
945
acb1b346
PH
946PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
947 forgetting to use the resulting value; it was using the unexpanded value.
948
c5ddb310
PH
949PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
950 hadn't been configured. The fix is from Juergen Kreileder, who
951 understands it better than I do:
952
953 "Here's what I see happening with three configured cyrus_sasl
954 authenticators configured (plain, login, cram-md5):
955
956 On startup auth_cyrus_sasl_init() gets called for each of these.
957 This means three calls to sasl_listmech() without a specified mech_list.
958 => SASL tests which mechs of all available mechs actually work
959 => three warnings about OTP not working
960 => the returned list contains: plain, login, cram-md5, digest-md5, ...
961
962 With the patch, sasl_listmech() also gets called three times. But now
963 SASL's mech_list option is set to the server_mech specified in the the
964 authenticator. Or in other words, the answer from sasl_listmech()
965 gets limited to just the mech you're testing for (which is different
966 for each call.)
967 => the return list contains just 'plain' or 'login', 'cram-md5' or
968 nothing depending on the value of ob->server_mech.
969
970 I've just tested the patch: Authentication still works fine,
971 unavailable mechs specified in the exim configuration are still
972 caught, and the auth.log warnings about OTP are gone."
973
31619da6
PH
974PH/25 When debugging is enabled, the contents of the command line are added
975 to the debugging output, even when log_selector=+arguments is not
976 specified.
977
bebaf0fc
PH
978PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
979 answer is "GNU", and only if the return is "GNU/something" is the answer
980 "Linux".
981
475fe28a
PH
982PH/27 $acl_verify_message is now set immediately after the failure of a
983 verification in an ACL, and so is available in subsequent modifiers. In
984 particular, the message can be preserved by coding like this:
985
986 warn !verify = sender
987 set acl_m0 = $acl_verify_message
988
989 Previously, $acl_verify_message was set only while expanding "message"
990 and "log_message" when a very denied access.
991
7e8bec7a
PH
992PH/28 Modified OS/os.c-Linux with
993
994 -#ifndef OS_LOAD_AVERAGE
995 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
996
997 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
998 nomenclature these days.)
999
e4a89c47
PH
1000PH/29 Installed patch from the Sieve maintainer that adds the options
1001 sieve_useraddress and sieve_subaddress to the redirect router.
1002
5ca2a9a1
PH
1003PH/30 In these circumstances:
1004 . Two addresses routed to the same list of hosts;
1005 . First host does not offer TLS;
1006 . First host accepts first address;
1007 . First host gives temporary error to second address;
1008 . Second host offers TLS and a TLS session is established;
1009 . Second host accepts second address.
1010 Exim incorrectly logged both deliveries with the TLS parameters (cipher
1011 and peerdn, if requested) that were in fact used only for the second
1012 address.
7e8bec7a 1013
c688b954
PH
1014PH/31 When doing a callout as part of verifying an address, Exim was not paying
1015 attention to any local part prefix or suffix that was matched by the
1016 router that accepted the address. It now behaves in the same way as it
1017 does for delivery: the affixes are removed from the local part unless
1018 rcpt_include_affixes is set on the transport.
1019
fed77020
PH
1020PH/32 Add the sender address, as F=<...>, to the log line when logging a
1021 timeout during the DATA phase of an incoming message.
1022
7fe1560f
PH
1023PH/33 Sieve envelope tests were broken for match types other than :is. I have
1024 applied a patch sanctioned by the Sieve maintainer.
c688b954 1025
ebb6e6d5
PH
1026PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
1027 the uid or gid is negative. A case of a negative gid caused this to be
1028 noticed. The fix allows for either to be negative.
1029
9c4e8f60
PH
1030PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
1031 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
1032 overlooked.
1033
1034PH/36 The change PH/12 above was broken. Fixed it.
1035
d7174846
PH
1036PH/37 Exim used to check for duplicate addresses in the middle of routing, on
1037 the grounds that routing the same address twice would always produce the
1038 same answer. This might have been true once, but it is certainly no
1039 longer true now. Routing a child address may depend on the previous
1040 routing that produced that child. Some complicated redirection strategies
1041 went wrong when messages had multiple recipients, and made Exim's
1042 behaviour dependent on the order in which the addresses were given.
1043
1044 I have moved the duplicate checking until after the routing is complete.
1045 Exim scans the addresses that are assigned to local and remote
1046 transports, and removes any duplicates. This means that more work will be
1047 done, as duplicates will always all be routed, but duplicates are
1048 presumably rare, so I don't expect this is of any significance.
1049
1050 For deliveries to pipes, files, and autoreplies, the duplicate checking
1051 still happens during the routing process, since they are not going to be
1052 routed further.
1053
cfe75fc3
PH
1054PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
1055 It corrects a timeout issue with spamd. This is Ian's comment: "The
1056 background is that sometimes spamd either never reads data from a
1057 connection it has accepted, or it never writes response data. The exiscan
1058 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
1059 blindly assumes that writes won't block so it may never time out."
1060
be22d70e
PH
1061PH/39 Allow G after quota size as well as K and M.
1062
0612b098
PH
1063PH/40 The value set for $authenticated_id in an authenticator may not contain
1064 binary zeroes or newlines because the value is written to log lines and
1065 to spool files. There was no check on this. Now the value is run through
1066 the string_printing() function so that such characters are converted to
1067 printable escape sequences.
1068
2e0c1448
PH
1069PH/41 $message_linecount is a new variable that contains the total number of
1070 lines in the message. Compare $body_linecount, which is the count for the
1071 body only.
1072
447d236c
PH
1073PH/42 Exim no longer gives details of delivery errors for specific addresses in
1074 bounce and delay warning messages, except in certain special cases, which
1075 are as follows:
1076
1077 (a) An SMTP error message from a remote host;
1078 (b) A message specified in a :fail: redirection;
1079 (c) A message specified in a "fail" command in a system filter;
1080 (d) A message specified in a FAIL return from the queryprogram router;
1081 (e) A message specified by the cannot_route_message router option.
1082
1083 In these cases only, Exim does include the error details in bounce and
1084 warning messages. There are also a few cases where bland messages such
1085 as "unrouteable address" or "local delivery error" are given.
1086
d20976dc
PH
1087PH/43 $value is now also set for the "else" part of a ${run expansion.
1088
f656d135
PH
1089PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
1090 being worked on, but at least Exim now implements the latest version to
1091 play with."
1092
2e2a30b4
PH
1093PH/45 In a pipe transport, although a timeout while waiting for the pipe
1094 process to complete was treated as a delivery failure, a timeout while
1095 writing the message to the pipe was logged, but erroneously treated as a
1096 successful delivery. Such timeouts include transport filter timeouts. For
1097 consistency with the overall process timeout, these timeouts are now
1098 treated as errors, giving rise to delivery failures by default. However,
1099 there is now a new Boolean option for the pipe transport called
1100 timeout_defer, which, if set TRUE, converts the failures into defers for
1101 both kinds of timeout. A transport filter timeout is now identified in
1102 the log output.
1103
9176e9f0
PH
1104PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
1105 systems where "make" and "gmake" are different, calling "gmake" at top
1106 level broke things. I've arranged for the value of $(MAKE) to be passed
1107 from the Makefile to this script so that it can call the same version of
1108 "make".
1109
7982096b 1110
bbe902f0
PH
1111A note about Exim versions 4.44 and 4.50
1112----------------------------------------
1113
1114Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
1115changes of various kinds. As a consequence, a big documentation update was
1116needed. This delayed the release for rather longer than seemed good, especially
1117in the light of a couple of (minor) security issues. Therefore, the changes
1118that fixed bugs were backported into 4.43, to create a 4.44 maintenance
1119release. So 4.44 and 4.50 are in effect two different branches that both start
1120from 4.43.
1121
1122I have left the 4.50 change log unchanged; it contains all the changes since
11234.43. The change log for 4.44 is below; many of its items are identical to
1124those for 4.50. This seems to be the most sensible way to preserve the
1125historical information.
1126
1127
f7b63901 1128Exim version 4.50
495ae4b0
PH
1129-----------------
1130
5fe762f6
PH
1131 1. Minor wording change to the doc/README.SIEVE file.
1132
139059f6 1133 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
5fe762f6 1134 computation of the current number of files was incorrect.
495ae4b0 1135
7086e875
PH
1136 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
1137 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
1138 place.
1139
35af9f61
PH
1140 4. Give more explanation in the error message when the command for a transport
1141 filter fails to execute.
1142
b668c215
PH
1143 5. There are several places where Exim runs a non-Exim command in a
1144 subprocess. The SIGUSR1 signal should be disabled for these processes. This
1145 was being done only for the command run by the queryprogram router. It is
1146 now done for all such subprocesses. The other cases are: ${run, transport
1147 filters, and the commands run by the lmtp and pipe transports.
1148
a494b1e1
PH
1149 6. Added CONFIGURE_GROUP build-time option.
1150
1151 7. Some older OS have a limit of 256 on the maximum number of file
1152 descriptors. Exim was using setrlimit() to set 1000 as a large value
1153 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
1154 systems. I've change it so that if it can't get 1000, it tries for 256.
35edf2ff 1155
c5fcb476
PH
1156 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
1157 was an oversight, and furthermore, ever since the addition of extra
1158 controls (e.g. 4.43/32), the checks on when to allow different forms of
1159 "control" were broken. There should now be diagnostics for all cases when a
1160 control that does not make sense is encountered.
1161
69358f02
PH
1162 9. Added the /retain_sender option to "control=submission".
1163
5be20824
PH
116410. $recipients is now available in the predata ACL (oversight).
1165
eb2c0248
PH
116611. Tidy the search cache before the fork to do a delivery from a message
1167 received from the command line. Otherwise the child will trigger a lookup
1168 failure and thereby defer the delivery if it tries to use (for example) a
1169 cached ldap connection that the parent has called unbind on.
1170
2a3eea10
PH
117112. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
1172 of $address_data from the recipient verification was clobbered by the
1173 sender verification.
1174
117513. The value of address_data from a sender verification is now available in
1176 $sender_address_data in subsequent conditions in the ACL statement.
1177
23c7ff99
PH
117814. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
1179
4deaf07d
PH
118015. Added a new option "connect=<time>" to callout options, to set a different
1181 connection timeout.
1182
926e1192
PH
118316. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
1184 was its contents. (It was OK if the option was not defined at all.)
1185
650edc6f
PH
118617. A "Completed" log line is now written for messages that are removed from
1187 the spool by the -Mrm option.
1188
2c7db3f5
PH
118918. New variables $sender_verify_failure and $recipient_verify_failure contain
1190 information about exactly what failed.
1191
3d235903
PH
119219. Added -dd to debug only the daemon process.
1193
7c7ad977
PH
119420. Incorporated Michael Haardt's patch to ldap.c for improving the way it
1195 handles timeouts, both on the server side and network timeouts. Renamed the
1196 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
1197
981756db
PH
119821. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
1199 It is now set to "smtps".
1200
d4eb88df
PH
120122. $host_address is now set to the target address during the checking of
1202 ignore_target_hosts.
1203
120423. When checking ignore_target_hosts for an ipliteral router, no host name was
1205 being passed; this would have caused $sender_host_name to have been used if
1206 matching the list had actually called for a host name (not very likely,
1207 since this list is usually IP addresses). A host name is now passed as
1208 "[x.x.x.x]".
1209
7d468ab8
PH
121024. Changed the calls that set up the SIGCHLD handler in the daemon to use the
1211 code that specifies a non-restarting handler (typically sigaction() in
1212 modern systems) in an attempt to fix a rare and obscure crash bug.
1213
121425. Narrowed the window for a race in the daemon that could cause it to ignore
1215 SIGCHLD signals. This is not a major problem, because they are used only to
1216 wake it up if nothing else does.
1217
62c0818f
PH
121826. A malformed maildirsize file could cause Exim to calculate negative values
1219 for the mailbox size or file count. Odd effects could occur as a result.
1220 The maildirsize information is now recalculated if the size or filecount
1221 end up negative.
1222
26034054
PH
122327. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
1224 support for a long time. Removed HAVE_SYS_VFS_H.
1225
af66f652
PH
122628. Installed the latest version of exipick from John Jetmore.
1227
90af77f4
PH
122829. In an address list, if the pattern was not a regular expression, an empty
1229 subject address (from a bounce message) matched only if the pattern was an
1230 empty string. Non-empty patterns were not even tested. This was the wrong
1231 because it is perfectly reasonable to use an empty address as part of a
1232 database query. An empty address is now tested by patterns that are
1233 lookups. However, all the other forms of pattern expect the subject to
1234 contain a local part and a domain, and therefore, for them, an empty
1235 address still always fails if the pattern is not itself empty.
1236
d8ef3577
PH
123730. Exim went into a mad DNS loop when attempting to do a callout where the
1238 host was specified on an smtp transport, and looking it up yielded more
1239 than one IP address.
1240
5cb8cbc6
PH
124131. Re-factored the code for checking spool and log partition space into a
1242 function that finds that data and another that does the check. The former
1243 is then used to implement four new variables: $spool_space, $log_space,
1244 $spool_inodes, and $log_inodes.
1245
14702f5b
PH
124632. The RFC2047 encoding function was originally intended for short strings
1247 such as real names; it was not keeping to the 75-character limit for
1248 encoded words that the RFC imposes. It now respects the limit, and
1249 generates multiple encoded words if necessary. To be on the safe side, I
1250 have increased the buffer size for the ${rfc2047: expansion operator from
1251 1024 to 2048 bytes.
1252
063b1e99
PH
125333. It is now permitted to omit both strings after an "if" condition; if the
1254 condition is true, the result is "true". As before, when the second string
1255 is omitted, a false condition yields an empty string. This makes it less
1256 cumbersome to write custom ACL and router conditions.
1257
652e1b65
PH
125834. Failure to deliver a bounce message always caused it to be frozen, even if
1259 there was an errors_to setting on the router. The errors_to setting is now
1260 respected.
1261
6f0c9a4f
PH
126235. If an IPv6 address is given for -bh or -bhc, it is now converted to the
1263 canonical form (fully expanded) before being placed in
1264 $sender_host_address.
1265
33397d19
PH
126636. The table in the code that translates DNS record types into text (T_A to
1267 "A" for instance) was missing entries for NS and CNAME. It is just possible
1268 that this could have caused confusion if both these types were looked up
1269 for the same domain, because the text type is used as part of Exim's
1270 per-process caching. But the chance of anyone hitting this buglet seems
1271 very small.
1272
7bb56e1f
PH
127337. The dnsdb lookup has been extended in a number of ways.
1274
1275 (1) There is a new type, "zns", which walks up the domain tree until it
1276 finds some nameserver records. It should be used with care.
1277
ea3bc19b
PH
1278 (2) There is a new type, "mxh", which is like "mx" except that it returns
1279 just the host names, not the priorities.
1280
1281 (3) It is now possible to give a list of domains (or IP addresses) to be
ff4dbb19
PH
1282 looked up. The behaviour when one of the lookups defers can be
1283 controlled by a keyword.
7bb56e1f 1284
ea3bc19b 1285 (4) It is now possible to specify the separator character for use when
7bb56e1f 1286 multiple records are returned.
33397d19 1287
0bcb2a0e
PH
128838. The dnslists ACL condition has been extended: it is now possible to supply
1289 a list of IP addresses and/or domains to be looked up in a particular DNS
1290 domain.
1291
2ac0e484
PH
129239. Added log_selector=+queue_time_overall.
1293
4e1fde53
PH
129440. When running the queue in the test harness, wait just a tad after forking a
1295 delivery process, to get repeatability of debugging output.
1296
de365ded
PH
129741. Include certificate and key file names in error message when GnuTLS fails
1298 to set them up, because the GnuTLS error message doesn't include the name
1299 of the failing file when there is a problem reading it.
1300
f05da2e8
PH
130142. Allow both -bf and -bF in the same test run.
1302
d6453af2
PH
130343. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
1304
f7b63901
PH
130544. The "Exiscan patch" is now merged into the mainline Exim source.
1306
130745. Sometimes the final signoff response after QUIT could fail to get
1308 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
1309 < 0 before doing a fflush(). This bug looks as though it goes back to the
1310 introduction of TLS in release 3.20, but "sometimes" must have been rare
1311 because the tests only now provoked it.
1312
a444213a
PH
131346. Reset the locale to "C" after calling embedded Perl, in case it was changed
1314 (this can affect the format of dates).
1315
0ec020ea
PH
131647. exim_tidydb, when checking for the continued existence of a message for
1317 which it has found a message-specific retry record, was not finding
1318 messages that were in split spool directories. Consequently, it was
1319 deleting retry records that should have stayed in existence.
1320
b1206957
PH
132148. Steve fixed some bugs in eximstats.
1322
132349. The SPA authentication driver was not abandoning authentication and moving
1324 on to the next authenticator when an expansion was forced to fail,
1325 contradicting the general specification for all authenticators. Instead it
1326 was generating a temporary error. It now behaves as specified.
1327
26dd5a95
PH
132850. The default ordering of permitted cipher suites for GnuTLS was pessimal
1329 (the order specifies the preference for clients). The order is now AES256,
1330 AES128, 3DES, ARCFOUR128.
1331
343b2385
PH
133251. Small patch to Sieve code - explicitly set From: when generating an
1333 autoreply.
1334
1c5466b9
PH
133552. Exim crashed if a remote delivery caused a very long error message to be
1336 recorded - for instance if somebody sent an entire SpamAssassin report back
1337 as a large number of 550 error lines. This bug was coincidentally fixed by
1338 increasing the size of one of Exim's internal buffers (big_buffer) that
1339 happened as part of the Exiscan merge. However, to be on the safe side, I
1340 have made the code more robust (and fixed the comments that describe what
1341 is going on).
1342
55ee9ee3
PH
134353. Now that there can be additional text after "Completed" in log lines (if
1344 the queue_time_overall log selector is set), a one-byte patch to exigrep
1345 was needed to allow it to recognize "Completed" as not the last thing in
1346 the line.
1347
d38f8232
PH
134854. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
1349 patch that reportedly fixes this has been added. I am not expert enough to
1350 create a test for it. This is what the patch creator wrote:
1351
1352 "I found a little strange behaviour of ldap code when working with
1353 Windows 2003 AD Domain, where users was placed in more than one
1354 Organization Units. When I tried to give exim partial DN, the exit code
1355 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
1356 But simultaneously result of request was absolutely normal ldap result,
1357 so I produce this patch..."
1358
3295e65b
PH
1359 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
1360 REFERENCE, so I have modified the code to exclude the patch when that macro
1361 is not defined.
1362
7102e136
PH
136355. Some experimental protocols are using DNS PTR records for new purposes. The
1364 keys for these records are domain names, not reversed IP addresses. The
b975ba52
PH
1365 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
1366 leaves it alone. Component reversal etc. now happens only for IP addresses.
ea3a6f44 1367 CAN-2005-0021
7102e136 1368
3ca0ba97
PH
136956. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
1370
c2bcbe20
PH
137157. Double the size of the debug message buffer (to 2048) so that more of very
1372 long debug lines gets shown.
1373
18ce445d
PH
137458. The exicyclog utility now does better if the number of log files to keep
1375 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
1376
1f5b4c3d
PH
137759. Two changes related to the smtp_active_hostname option:
1378
1379 (1) $smtp_active_hostname is now available as a variable.
1380 (2) The default for smtp_banner uses $smtp_active_hostname instead
1381 of $primary_hostname.
1382
b975ba52
PH
138360. The host_aton() function is supposed to be passed a string that is known
1384 to be a valid IP address. However, in the case of IPv6 addresses, it was
1385 not checking this. This is a hostage to fortune. Exim now panics and dies
1386 if the condition is not met. A case was found where this could be provoked
85b87bc2
PH
1387 from a dnsdb PTR lookup with an IPv6 address that had more than 8
1388 components; fortuitously, this particular loophole had already been fixed
1389 by change 4.50/55 above.
1390
1391 If there are any other similar loopholes, the new check in host_aton()
1392 itself should stop them being exploited. The report I received stated that
1393 data on the command line could provoke the exploit when Exim was running as
1394 exim, but did not say which command line option was involved. All I could
1395 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
1396 running as the user.
ea3a6f44 1397 CAN-2005-0021
85b87bc2
PH
1398
139961. There was a buffer overflow vulnerability in the SPA authentication code
1400 (which came originally from the Samba project). I have added a test to the
1401 spa_base64_to_bits() function which I hope fixes it.
ea3a6f44 1402 CAN-2005-0022
b975ba52 1403
17ffcae7
PH
140462. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
1405 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
1406
d95f9fdb
PH
140763. The daemon start-up calls getloadavg() while still root for those OS that
1408 need the first call to be done as root, but it missed one case: when
1409 deliver_queue_load_max is set with deliver_drop_privilege. This is
1410 necessary for the benefit of the queue runner, because there is no re-exec
1411 when deliver_drop_privilege is set.
1412
86b8287f
PH
141364. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
1414 This has been fixed.
1415
60dc5e56
PH
141665. Caching of lookup data for "hosts =" ACL conditions, when a named host list
1417 was in use, was not putting the data itself into the right store pool;
1418 consequently, it could be overwritten for a subsequent message in the same
1419 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
1420 the caching.)
1421
533244af
PH
142266. Added hosts_max_try_hardlimit to the smtp transport, default 50.
1423
a5a28604
PH
142467. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
1425 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
1426 the function were treating the return as a boolean value, which happened to
1427 work because 0=false and not-0=true, but is not correct code.
1428
7e634d24
PH
142968. The host_aton() function was not handling scoped IPv6 addresses (those
1430 with, for example, "%eth0" on the end) correctly.
1431
3e11c26b
PH
143269. Fixed some compiler warnings in acl.c for the bitmaps specified with
1433 negated items (that is, ~something) in unsigned ints. Some compilers
1434 apparently mutter when there is no cast.
1435
6729cf78
PH
143670. If an address verification called from an ACL failed, and did not produce a
1437 user-specific message (i.e. there was only a "system" message), nothing was
1438 put in $acl_verify_message. In this situation, it now puts the system
1439 message there.
1440
00f00ca5
PH
144171. Change 4.23/11 added synchronization checking at the start of an SMTP
1442 session; change 4.31/43 added the unwanted input to the log line - except
1443 that it did not do this in the start of session case. It now does.
1444
c9bdd01c
PH
144572. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
1446 This is wrong and can cause the other end to generate a synchronization
1447 error if it is another Exim or anything else that does the synchronization
1448 check. A QUIT command is no longer sent after a timeout.
1449
d43194df
PH
145073. $host_lookup_deferred has been added, to make it easier to detect DEFERs
1451 during host lookups.
1452
fe5b5d0b
PH
145374. The defer_ok option of callout verification was not working if it was used
1454 when verifying addresses in header lines, that is, for this case:
1455
1456 verify = header_sender/callout=defer_ok
1457
76a2d7ba
PH
145875. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
1459 those file descriptors could be used for SMTP connections. If anything
1460 wrote to stderr (the example that came up was "warn" in embedded Perl), it
1461 could be sent to the SMTP client, causing chaos. The daemon now opens
1462 stdin, stdout, and stderr to /dev/null when it puts itself into the
1463 background.
1464
146576. Arrange for output from Perl's "warn" command to be written to Exim's main
1466 log by default. The user can override this with suitable Perl magic.
1467
04f7d5b9
PH
146877. The use of log_message on a "discard" ACL verb, which is supposed to add to
1469 the log message when discard triggers, was not working for the DATA ACL or
1470 for the non-SMTP ACL.
1471
bc60667e
PH
147278. Error message wording change in sieve.c.
1473
bb6e88ff
PH
147479. If smtp_accept_max_per_host was set, the number of connections could be
1475 restricted to fewer than expected, because the daemon was trying to set up
1476 a new connection before checking whether the processes handling previous
1477 connections had finished. The check for completed processes is now done
1478 earlier. On busy systems, this bug wouldn't be noticed because something
1479 else would have woken the daemon, and it would have reaped the completed
1480 process earlier.
1481
1e70f85b
PH
148280. If a message was submitted locally by a user whose login name contained one
1483 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
1484 It caused a spool format error. I have fixed the spool reading code. A
1485 related problem was that the "from" clause in the Received: line became
1486 illegal because of the space(s). It is now covered by ${quote_local_part.
1487
148881. Included the latest eximstats from Steve (adds average sizes to HTML Top
1489 tables).
1490
4e01f9d6
PH
149182. Updated OS/Makefile-AIX as per message from Mike Meredith.
1492
1ee1cef2
PH
149383. Patch from Sieve maintainer to fix unterminated string problem in
1494 "vacation" handling.
1495
6e2b4ccc
PH
149684. Some minor changes to the Linux configuration files to help with other
1497 OS variants using glibc.
1498
8e669ac1
PH
149985. One more patch for Sieve to update vacation handling to latest spec.
1500
495ae4b0 1501
bbe902f0
PH
1502----------------------------------------------------
1503See the note above about the 4.44 and 4.50 releases.
1504----------------------------------------------------
1505
1506
1507Exim version 4.44
1508-----------------
1509
1510 1. Change 4.43/35 introduced a bug that caused file counts to be
1511 incorrectly computed when quota_filecount was set in an appendfile
1512 transport
1513
1514 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
1515 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
1516 place.
1517
1518 3. Give more explanation in the error message when the command for a transport
1519 filter fails to execute.
1520
1521 4. There are several places where Exim runs a non-Exim command in a
1522 subprocess. The SIGUSR1 signal should be disabled for these processes. This
1523 was being done only for the command run by the queryprogram router. It is
1524 now done for all such subprocesses. The other cases are: ${run, transport
1525 filters, and the commands run by the lmtp and pipe transports.
1526
1527 5. Some older OS have a limit of 256 on the maximum number of file
1528 descriptors. Exim was using setrlimit() to set 1000 as a large value
1529 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
1530 systems. I've change it so that if it can't get 1000, it tries for 256.
1531
1532 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
1533 was an oversight, and furthermore, ever since the addition of extra
1534 controls (e.g. 4.43/32), the checks on when to allow different forms of
1535 "control" were broken. There should now be diagnostics for all cases when a
1536 control that does not make sense is encountered.
1537
1538 7. $recipients is now available in the predata ACL (oversight).
1539
1540 8. Tidy the search cache before the fork to do a delivery from a message
1541 received from the command line. Otherwise the child will trigger a lookup
1542 failure and thereby defer the delivery if it tries to use (for example) a
1543 cached ldap connection that the parent has called unbind on.
1544
1545 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
1546 of $address_data from the recipient verification was clobbered by the
1547 sender verification.
1548
154910. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
1550 was its contents. (It was OK if the option was not defined at all.)
1551
155211. A "Completed" log line is now written for messages that are removed from
1553 the spool by the -Mrm option.
1554
155512. $host_address is now set to the target address during the checking of
1556 ignore_target_hosts.
1557
155813. When checking ignore_target_hosts for an ipliteral router, no host name was
1559 being passed; this would have caused $sender_host_name to have been used if
1560 matching the list had actually called for a host name (not very likely,
1561 since this list is usually IP addresses). A host name is now passed as
1562 "[x.x.x.x]".
1563
156414. Changed the calls that set up the SIGCHLD handler in the daemon to use the
1565 code that specifies a non-restarting handler (typically sigaction() in
1566 modern systems) in an attempt to fix a rare and obscure crash bug.
1567
156815. Narrowed the window for a race in the daemon that could cause it to ignore
1569 SIGCHLD signals. This is not a major problem, because they are used only to
1570 wake it up if nothing else does.
1571
157216. A malformed maildirsize file could cause Exim to calculate negative values
1573 for the mailbox size or file count. Odd effects could occur as a result.
1574 The maildirsize information is now recalculated if the size or filecount
1575 end up negative.
1576
157717. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
1578 support for a long time. Removed HAVE_SYS_VFS_H.
1579
ea3a6f44 158018. Updated exipick to current release from John Jetmore.
bbe902f0
PH
1581
158219. Allow an empty sender to be matched against a lookup in an address list.
1583 Previously the only cases considered were a regular expression, or an
1584 empty pattern.
1585
158620. Exim went into a mad DNS lookup loop when doing a callout where the
1587 host was specified on the transport, if the DNS lookup yielded more than
1588 one IP address.
1589
ea3a6f44
NM
159021. The RFC2047 encoding function was originally intended for short strings
1591 such as real names; it was not keeping to the 75-character limit for
1592 encoded words that the RFC imposes. It now respects the limit, and
1593 generates multiple encoded words if necessary. To be on the safe side, I
1594 have increased the buffer size for the ${rfc2047: expansion operator from
1595 1024 to 2048 bytes.
bbe902f0 1596
ea3a6f44
NM
159722. Failure to deliver a bounce message always caused it to be frozen, even if
1598 there was an errors_to setting on the router. The errors_to setting is now
1599 respected.
bbe902f0
PH
1600
160123. If an IPv6 address is given for -bh or -bhc, it is now converted to the
1602 canonical form (fully expanded) before being placed in
1603 $sender_host_address.
1604
160524. Updated eximstats to version 1.33
1606
ea3a6f44
NM
160725. Include certificate and key file names in error message when GnuTLS fails
1608 to set them up, because the GnuTLS error message doesn't include the name
1609 of the failing file when there is a problem reading it.
bbe902f0
PH
1610
161126. Expand error message when OpenSSL has problems setting up cert/key files.
ea3a6f44 1612 As per change 25.
bbe902f0 1613
ea3a6f44
NM
161427. Reset the locale to "C" after calling embedded Perl, in case it was changed
1615 (this can affect the format of dates).
bbe902f0 1616
ea3a6f44
NM
161728. exim_tidydb, when checking for the continued existence of a message for
1618 which it has found a message-specific retry record, was not finding
1619 messages that were in split spool directories. Consequently, it was
1620 deleting retry records that should have stayed in existence.
bbe902f0
PH
1621
162229. eximstats updated to version 1.35
1623 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
1624 1.35 - bugfix such that pie charts by volume are generated correctly
1625
ea3a6f44
NM
162630. The SPA authentication driver was not abandoning authentication and moving
1627 on to the next authenticator when an expansion was forced to fail,
1628 contradicting the general specification for all authenticators. Instead it
1629 was generating a temporary error. It now behaves as specified.
bbe902f0 1630
ea3a6f44
NM
163131. The default ordering of permitted cipher suites for GnuTLS was pessimal
1632 (the order specifies the preference for clients). The order is now AES256,
1633 AES128, 3DES, ARCFOUR128.
bbe902f0 1634
ea3a6f44
NM
163531. Small patch to Sieve code - explicitly set From: when generating an
1636 autoreply.
bbe902f0 1637
ea3a6f44
NM
163832. Exim crashed if a remote delivery caused a very long error message to be
1639 recorded - for instance if somebody sent an entire SpamAssassin report back
1640 as a large number of 550 error lines. This bug was coincidentally fixed by
1641 increasing the size of one of Exim's internal buffers (big_buffer) that
1642 happened as part of the Exiscan merge. However, to be on the safe side, I
1643 have made the code more robust (and fixed the comments that describe what
1644 is going on).
bbe902f0 1645
ea3a6f44
NM
164633. Some experimental protocols are using DNS PTR records for new purposes. The
1647 keys for these records are domain names, not reversed IP addresses. The
1648 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
1649 leaves it alone. Component reversal etc. now happens only for IP addresses.
bbe902f0
PH
1650 CAN-2005-0021
1651
ea3a6f44
NM
165234. The host_aton() function is supposed to be passed a string that is known
1653 to be a valid IP address. However, in the case of IPv6 addresses, it was
1654 not checking this. This is a hostage to fortune. Exim now panics and dies
1655 if the condition is not met. A case was found where this could be provoked
1656 from a dnsdb PTR lookup with an IPv6 address that had more than 8
1657 components; fortuitously, this particular loophole had already been fixed
1658 by change 4.50/55 or 4.44/33 above.
1659
1660 If there are any other similar loopholes, the new check in host_aton()
1661 itself should stop them being exploited. The report I received stated that
1662 data on the command line could provoke the exploit when Exim was running as
1663 exim, but did not say which command line option was involved. All I could
1664 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
1665 running as the user.
bbe902f0
PH
1666 CAN-2005-0021
1667
ea3a6f44
NM
166835. There was a buffer overflow vulnerability in the SPA authentication code
1669 (which came originally from the Samba project). I have added a test to the
1670 spa_base64_to_bits() function which I hope fixes it.
bbe902f0
PH
1671 CAN-2005-0022
1672
ea3a6f44
NM
167336. The daemon start-up calls getloadavg() while still root for those OS that
1674 need the first call to be done as root, but it missed one case: when
1675 deliver_queue_load_max is set with deliver_drop_privilege. This is
1676 necessary for the benefit of the queue runner, because there is no re-exec
1677 when deliver_drop_privilege is set.
bbe902f0 1678
ea3a6f44
NM
167937. Caching of lookup data for "hosts =" ACL conditions, when a named host list
1680 was in use, was not putting the data itself into the right store pool;
1681 consequently, it could be overwritten for a subsequent message in the same
1682 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
1683 the caching.)
bbe902f0 1684
ea3a6f44
NM
168538. Sometimes the final signoff response after QUIT could fail to get
1686 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
1687 < 0 before doing a fflush(). This bug looks as though it goes back to the
1688 introduction of TLS in release 3.20, but "sometimes" must have been rare
1689 because the tests only now provoked it.
bbe902f0
PH
1690
1691
495ae4b0
PH
1692Exim version 4.43
1693-----------------
1694
1695 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
1696 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
1697 Now it returns an integer. A number of calls were still expecting a T/F
1698 return. Fortuitously, in all cases, the tests worked in OK situations,
1699 which is the norm. However, things would have gone wrong on any write
1700 failures on the smtp file descriptor. This function is used when sending
1701 messages over SMTP and also when doing verify callouts.
1702
1703 2. When Exim is called to do synchronous delivery of a locally submitted
1704 message (the -odf or -odi options), it no longer closes stderr before doing
1705 the delivery.
1706
1707 3. Implemented the mua_wrapper option.
1708
1709 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
1710
1711 5. Implemented the functions header_remove(), header_testname(),
1712 header_add_at_position(), and receive_remove_recipient(), and exported them
1713 to local_scan().
1714
1715 6. If an ACL "warn" statement specified the addition of headers, Exim already
1716 inserted X-ACL-Warn: at the start if there was no header name. However, it
1717 was not making this test for the second and subsequent header lines if
1718 there were newlines in the string. This meant that an invalid header could
1719 be inserted if Exim was badly configured.
1720
1721 7. Allow an ACL "warn" statement to add header lines at the start or after all
1722 the Received: headers, as well as at the end.
1723
1724 8. Added the rcpt_4xx retry error code.
1725
1726 9. Added postmaster_mailfrom=xxx to callout verification option.
1727
172810. Added mailfrom=xxxx to the callout verification option, for verify=
1729 header_sender only.
1730
173111. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
1732 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
1733
173412. Inserted some casts to stop certain compilers warning when using pointer
1735 differences as field lengths or precisions in printf-type calls (mostly
1736 affecting debugging statements).
1737
173813. Added optional readline() support for -be (dynamically loaded).
1739
174014. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
1741 same clock tick as a message's arrival, so that its received time was the
1742 same as the "first fail" time on the retry record, and that message
1743 remained on the queue past the ultimate address timeout, every queue runner
1744 would try a delivery (because it was past the ultimate address timeout) but
1745 after another failure, the ultimate address timeout, which should have then
1746 bounced the address, did not kick in. This was a "< instead of <=" error;
1747 in most cases the first failure would have been in the next clock tick
1748 after the received time, and all would be well.
1749
175015. The special items beginning with @ in domain lists (e.g. @mx_any) were not
1751 being recognized when the domain list was tested by the match_domain
1752 condition in an expansion string.
1753
175416. Added the ${str2b64: operator.
1755
175617. Exim was always calling setrlimit() to set a large limit for the number of
1757 processes, without checking whether the existing limit was already
1758 adequate. (It did check for the limit on file descriptors.) Furthermore,
1759 errors from getrlimit() and setrlimit() were being ignored. Now they are
1760 logged to the main and panic logs, but Exim does carry on, to try to do its
1761 job under whatever limits there are.
1762
176318. Imported PCRE 5.0.
1764
176519. Trivial typo in log message " temporarily refused connection" (the leading
1766 space).
1767
176820. If the log selector return_path_on_delivery was set and an address was
1769 redirected to /dev/null, the delivery process crashed because it assumed
1770 that a return path would always be set for a "successful" delivery. In this
1771 case, the whole delivery is bypassed as an optimization, and therefore no
1772 return path is set.
1773
177421. Internal re-arrangement: the function for sending a challenge and reading
1775 a response while authentication was assuming a zero-terminated challenge
1776 string. It's now changed to take a pointer and a length, to allow for
1777 binary data in such strings.
1778
177922. Added the cyrus_sasl authenticator (code supplied by MBM).
1780
178123. Exim was not respecting finduser_retries when seeking the login of the
1782 uid under which it was called; it was always trying 10 times. (The default
1783 setting of finduser_retries is zero.) Also, it was sleeping after the final
1784 failure, which is pointless.
1785
178624. Implemented tls_on_connect_ports.
1787
178825. Implemented acl_smtp_predata.
1789
179026. If the domain in control=submission is set empty, Exim assumes that the
1791 authenticated id is a complete email address when it generates From: or
1792 Sender: header lines.
1793
179427. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
1795 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
1796 chown and chgrp in /bin and hostname in /usr/bin.
1797
179828. Exim was keeping the "process log" file open after each use, just as it
1799 does for the main log. This opens the possibility of it remaining open for
1800 long periods when the USR1 signal hits a daemon. Occasional processlog
1801 errors were reported, that could have been caused by this. Anyway, it seems
1802 much more sensible not to leave this file open at all, so that is what now
1803 happens.
1804
180529. The long-running daemon process does not normally write to the log once it
1806 has entered its main loop, and it closes the log before doing so. This is
1807 so that log files can straightforwardly be renamed and moved. However,
1808 there are a couple of unusual error situations where the daemon does write
1809 log entries, and I had neglected to close the log afterwards.
1810
181130. The text of an SMTP error response that was received during a remote
1812 delivery was being truncated at 512 bytes. This is too short for some of
1813 the long messages that one sometimes sees. I've increased the limit to
1814 1024.
1815
181631. It is now possible to make retry rules that apply only when a message has a
1817 specific sender, in particular, an empty sender.
1818
181932. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
1820 it possible to be selective about when SMTP synchronization is enforced.
1821
182233. Added "control = caseful_local_part" and "control = "caselower_local_part".
1823
182432. Implemented hosts_connection_nolog.
1825
182633. Added an ACL for QUIT.
1827
182834. Setting "delay_warning=" to disable warnings was not working; it gave a
1829 syntax error.
1830
183135. Added mailbox_size and mailbox_filecount to appendfile.
1832
183336. Added control = no_multiline_responses to ACLs.
1834
183537. There was a bug in the logic of the code that waits for the clock to tick
1836 in the case where the clock went backwards by a substantial amount such
1837 that the microsecond fraction of "now" was more than the microsecond
1838 fraction of "then" (but the whole seconds number was less).
1839
184038. Added support for the libradius Radius client library this is found on
1841 FreeBSD (previously only the radiusclient library was supported).
1842
1843
1844Exim version 4.42
1845-----------------
1846
1847 1. When certain lookups returned multiple values in the form name=value, the
1848 quoting of the values was not always being done properly. Specifically:
1849 (a) If the value started with a double quote, but contained no whitespace,
1850 it was not quoted.
1851 (b) If the value contained whitespace other than a space character (i.e.
1852 tabs or newlines or carriage returns) it was not quoted.
1853 This fix has been applied to the mysql and pgsql lookups by writing a
1854 separate quoting function and calling it from the lookup code. The fix
1855 should probably also be applied to nisplus, ibase and oracle lookups, but
1856 since I cannot test any of those, I have not disturbed their existing code.
1857
1858 2. A hit in the callout cache for a specific address caused a log line with no
1859 reason for rejecting RCPT. Now it says "Previous (cached) callout
1860 verification failure".
1861
1862 3. There was an off-by-one bug in the queryprogram router. An over-long
1863 return line was truncated at 256 instead of 255 characters, thereby
1864 overflowing its buffer with the terminating zero. As well as fixing this, I
1865 have increased the buffer size to 1024 (and made a note to document this).
1866
1867 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
1868 when Exim is waiting for an SMTP response from a remote server, Exim
1869 restarts its select() call on the socket, thereby resetting its timeout.
1870 This is not a problem when such interrupts are rare. Somebody set up a cron
1871 job to run exiwhat every 2 minutes, which is less than the normal select()
1872 timeout (5 or 10 minutes). This meant that the select() timeout never
1873 kicked in because it was always reset. I have fixed this by comparing the
1874 time when an interrupt arrives with the time at the start of the first call
1875 to select(). If more time than the timeout has elapsed, the interrupt is
1876 treated as a timeout.
1877
1878 5. Some internal re-factoring in preparation for the addition of Sieve
1879 extensions (by MH). In particular, the "personal" test is moved to a
1880 separate function, and given an option for scanning Cc: and Bcc: (which is
1881 not set for Exim filters).
1882
1883 6. When Exim created an email address using the login of the caller as the
1884 local part (e.g. when creating a From: or Sender: header line), it was not
1885 quoting the local part when it contained special characters such as @.
1886
1887 7. Installed new OpenBSD configuration files.
1888
1889 8. Reworded some messages for syntax errors in "and" and "or" conditions to
1890 try to make them clearer.
1891
1892 9. Callout options, other than the timeout value, were being ignored when
1893 verifying sender addresses in header lines. For example, when using
1894
1895 verify = header_sender/callout=no_cache
1896
1897 the cache was (incorrectly) being used.
1898
189910. Added a missing instance of ${EXE} to the exim_install script; this affects
1900 only the Cygwin environment.
1901
190211. When return_path_on_delivery was set as a log selector, if different remote
1903 addresses in the same message used different return paths and parallel
1904 remote delivery occurred, the wrong values would sometimes be logged.
1905 (Whenever a remote delivery process finished, the return path value from
1906 the most recently started remote delivery process was logged.)
1907
190812. RFC 3848 specifies standard names for the "with" phrase in Received: header
1909 lines when AUTH and/or TLS are in use. This is the "received protocol"
1910 field. Exim used to use "asmtp" for authenticated SMTP, without any
1911 indication (in the protocol name) for TLS use. Now it follows the RFC and
1912 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
1913 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
1914 names appear in log lines as well as in Received: header lines.
1915
191613. Installed MH's patches for Sieve to add the "copy" and "vacation"
1917 extensions, and comparison tests, and to fix some bugs.
1918
191914. Changes to the "personal" filter test:
1920
1921 (1) The test was buggy in that it was just doing the equivalent of
1922 "contains" tests on header lines. For example, if a user's address was
1923 anne@some.where, the "personal" test would incorrectly be true for
1924
1925 To: susanne@some.where
1926
1927 This test is now done by extracting each address from the header in turn,
1928 and checking the entire address. Other tests that are part of "personal"
1929 are now done using regular expressions (for example, to check local parts
1930 of addresses in From: header lines).
1931
1932 (2) The list of non-personal local parts in From: addresses has been
1933 extended to include "listserv", "majordomo", "*-request", and "owner-*",
1934 taken from the Sieve specification recommendations.
1935
1936 (3) If the message contains any header line starting with "List-" it is
1937 treated as non-personal.
1938
1939 (4) The test for "circular" in the Subject: header line has been removed
1940 because it now seems ill-conceived.
1941
194215. Minor typos in src/EDITME comments corrected.
1943
194416. Installed latest exipick from John Jetmore.
1945
194617. If headers_add on a router specified a text string that was too long for
1947 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
1948 of string_sprintf() is now avoided.
1949
195018. $message_body_size was not set (it was always zero) when running the DATA
1951 ACL and the local_scan() function.
1952
195319. For the "mail" command in an Exim filter, no default was being set for
1954 the once_repeat time, causing a random time value to be used if "once" was
1955 specified. (If the value happened to be <= 0, no repeat happened.) The
1956 default is now 0s, meaning "never repeat". The "vacation" command was OK
1957 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
1958 (I found it when inspecting the code).
1959
196020. There is now an overall timeout for performing a callout verification. It
1961 defaults to 4 times the callout timeout, which applies to individual SMTP
1962 commands during the callout. The overall timeout applies when there is more
1963 than one host that can be tried. The timeout is checked before trying the
1964 next host. This prevents very long delays if there are a large number of
1965 hosts and all are timing out (e.g. when the network connections are timing
1966 out). The value of the overall timeout can be changed by specifying an
1967 additional sub-option for "callout", called "maxwait". For example:
1968
1969 verify = sender/callout=5s,maxwait=20s
1970
197121. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
1972 to the end before writing, but this should make it even safer).
1973
197422. Exim was forgetting that it had advertised PIPELINING for the second and
1975 subsequent messages on an SMTP connection. It was also not resetting its
1976 memory on STARTTLS and an internal HELO.
1977
197823. When Exim logs an SMTP synchronization error within a session, it now
1979 records whether PIPELINING has been advertised or not.
1980
198124. Added 3 instances of "(long int)" casts to time_t variables that were being
1982 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
1983 rather than long int.
1984
198525. Installed the latest Cygwin configuration files from the Cygwin maintainer.
1986
198726. Added the never_mail option to autoreply.
1988
1989
1990Exim version 4.41
1991-----------------
1992
1993 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
1994 crash if the getsockname() call failed; this can happen if a connection is
1995 closed very soon after it is established. The problem was simply in the
1996 order in which certain operations were done, causing Exim to try to write
1997 to the SMTP stream before it had set up the file descriptor. The bug has
1998 been fixed by making things happen in the correct order.
1999
2000
2001Exim version 4.40
2002-----------------
2003
2004 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
2005 before the connection was closed, thus losing the rejection response.
2006
2007 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
2008 some early Solaris releases, but causes trouble in current releases where
2009 socklen_t is defined.
2010
2011 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
2012 always exist.
2013
2014 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
2015 configured.
2016
2017 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
2018 that releasing the top of it at the end releases what was used for sub-
2019 expansions (unless the block got too big). However, discard this block if
2020 the first thing is a variable or header, so that we can use its block when
2021 it is dynamic (useful for very large $message_headers, for example).
2022
2023 6. Lookups now cache *every* query, not just the most recent. A new, separate
2024 store pool is used for this. It can be recovered when all lookup caches are
2025 flushed. Lookups now release memory at the end of their result strings.
2026 This has involved some general refactoring of the lookup sources.
2027
2028 7. Some code has been added to the store_xxx() functions to reduce the amount
2029 of flapping under certain conditions.
2030
2031 8. log_incoming_interface used to affect only the <= reception log lines. Now
2032 it causes the local interface and port to be added to several more SMTP log
2033 lines, for example "SMTP connection from", and rejection lines.
2034
2035 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
2036
203710. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
2038
203911. If $host_data was set by virtue of a hosts lookup in an ACL, its value
2040 could be overwritten at the end of the current message (or the start of a
2041 new message if it was set in a HELO ACL). The value is now preserved for
2042 the duration of the SMTP connection.
2043
204412. If a transport had a headers_rewrite setting, and a matching header line
2045 contained an unqualified address, that address was qualified, even if it
2046 did not match any rewriting rules. The underlying bug was that the values
2047 of the flags that permit the existence of unqualified sender and recipient
2048 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
2049 non-local messages, and by -bnq for local messages) were not being
2050 preserved with the message after it was received.
2051
205213. When Exim was logging an SMTP synchronization error, it could sometimes log
2053 "next input=" as part of the text comprising the host identity instead of
2054 the correct text. The code was using the same buffer for two different
2055 strings. However, depending on which order the printing function evaluated
2056 its arguments, the bug did not always show up. Under Linux, for example, my
2057 test suite worked just fine.
2058
205914. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
2060 doesn't work with some older versions of Perl. It has been changed to "my",
2061 which in any case is probably the better facility to use.
2062
206315. A really picky compiler found some instances of statements for creating
2064 error messages that either had too many or two few arguments for the format
2065 string.
2066
206716. The size of the buffer for calls to the DNS resolver has been increased
2068 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
2069 for addresses that have a lot of PTR records. This alleviates a problem; it
2070 does not fully solve it.
2071
207217. A dnsdb lookup for PTR records that receives more data than will fit in the
2073 buffer now truncates the list and logs the incident, which is the same
2074 action as happens when Exim is looking up a host name and its aliases.
2075 Previously in this situation something unpredictable would happen;
2076 sometimes it was "internal error: store_reset failed".
2077
207818. If a server dropped the connection unexpectedly when an Exim client was
2079 using GnuTLS and trying to read a response, the client delivery process
2080 crashed while trying to generate an error log message.
2081
208219. If a "warn" verb in an ACL added multiple headers to a message in a single
2083 string, for example:
2084
2085 warn message = H1: something\nH2: something
2086
2087 the text was added as a single header line from Exim's point of view
2088 though it ended up OK in the delivered message. However, searching for the
2089 second and subsequent header lines using $h_h2: did not work. This has been
2090 fixed. Similarly, if a system filter added multiple headers in this way,
2091 the routers could not see them.
2092
209320. Expanded the error message when iplsearch is called with an invalid key to
2094 suggest using net-iplsearch in a host list.
2095
209621. When running tests using -bh, any delays imposed by "delay" modifiers in
2097 ACLs are no longer actually imposed (and a message to that effect is
2098 output).
2099
210022. If a "gecos" field in a passwd entry contained escaped characters, in
2101 particular, if it contained a \" sequence, Exim got it wrong when building
2102 a From: or a Sender: header from that name. A second bug also caused
2103 incorrect handling when an unquoted " was present following a character
2104 that needed quoting.
2105
210623. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
2107 was not being matched caselessly.
2108
210924. Arranged for all hyphens in the exim.8 source to be escaped with
2110 backslashes.
2111
211225. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
2113 properly. Recipient callout cache records were still being keyed to include
2114 the sender, even when use_sender was set false. This led to far more
2115 callouts that were necessary. The sender is no longer included in the key
2116 when use_sender is false.
2117
211826. Added "control = submission" modifier to ACLs.
2119
212027. Added the ${base62d: operator to decode base 62 numbers.
2121
212228. dnsdb lookups can now access SRV records.
2123
212429. CONFIGURE_OWNER can be set at build time to define an alternative owner for
2125 the configuration file.
2126
212730. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
2128 (-v) mode. This makes the output for a verbose queue run more intelligible.
2129
213031. Added a use_postmaster feature to recipient callouts.
2131
213232. Added the $body_zerocount variable, containing the number of binary zero
2133 bytes in the message body.
2134
213533. The time of last modification of the "new" subdirectory is now used as the
2136 "mailbox time last read" when there is a quota error for a maildir
2137 delivery.
2138
213934. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
2140
214135. Added +ignore_unknown as a special item in host lists.
2142
214336. Code for decoding IPv6 addresses in host lists is now included, even if
2144 IPv6 support is not being compiled. This fixes a bug in which an IPv6
2145 address was recognized as an IP address, but was then not correctly decoded
2146 into binary, causing unexpected and incorrect effects when compared with
2147 another IP address.
2148
2149
2150Exim version 4.34
2151-----------------
2152
2153 1. Very minor rewording of debugging text in manualroute to say "list of
2154 hosts" instead of "hostlist".
2155
2156 2. If verify=header_syntax was set, and a header line with an unqualified
2157 address (no domain) and a large number of spaces between the end of the
2158 name and the colon was received, the reception process suffered a buffer
2159 overflow, and (when I tested it) crashed. This was caused by some obsolete
2160 code that should have been removed. The fix is to remove it!
2161
2162 3. When running in the test harness, delay a bit after writing a bounce
2163 message to get a bit more predictability in the log output.
2164
2165 4. Added a call to search_tidyup() just before forking a reception process. In
2166 theory, someone could use a lookup in the expansion of smtp_accept_max_
2167 per_host which, without the tidyup, could leave open a database connection.
2168
2169 5. Added the variables $recipient_data and $sender_data which get set from a
2170 lookup success in an ACL "recipients" or "senders" condition, or a router
2171 "senders" option, similar to $domain_data and $local_part_data.
2172
2173 6. Moved the writing of debug_print from before to after the "senders" test
2174 for routers.
2175
2176 7. Change 4.31/66 (moving the time when the Received: is generated) caused
2177 problems for message scanning, either using a data ACL, or using
2178 local_scan() because the Received: header was not generated till after they
2179 were called (in order to set the time as the time of reception completion).
2180 I have revised the way this works. The header is now generated after the
2181 body is received, but before the ACL or local_scan() are called. After they
2182 are run, the timestamp in the header is updated.
2183
2184
2185Exim version 4.33
2186-----------------
2187
2188 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
2189 before starting a queue runner without re-exec. This happened only when
2190 deliver_drop_privilege was set or when the Exim user was set to root. The
2191 effect of the bug was that timeouts during subsequent deliveries caused
2192 crashes instead of being properly handled. The handler is now left at its
2193 default (and expected) setting.
2194
2195 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
2196 message, again when deliver_drop_privilege is set or Exim is run as root.
2197 The bug described in (1) was not present in this case, but the tidying up
2198 of the other signals was missing. I have made the two cases consistent.
2199
2200 3. The ignore_target_hosts setting on a manualroute router was being ignored
2201 for hosts that were looked up using the /MX notation.
2202
2203 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
2204 in domain lists.
2205
2206 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
2207 operated on the sender address. After changing the $sender_address to <>
2208 for the sender address verify, Exim was re-instated it as the original
2209 (before rewriting) address, but remembering that it had rewritten it, so it
2210 wasn't rewriting it again. This bug also had the effect of breaking the
2211 sender address verification caching when the sender address was rewritten.
2212
2213 6. The ignore_target_hosts option was being ignored by the ipliteral router.
2214 This has been changed so that if the ip literal address matches
2215 ignore_target_hosts, the router declines.
2216
2217 7. Added expansion conditions match_domain, match_address, and match_local_
2218 part (NOT match_host).
2219
2220 8. The placeholder for the Received: header didn't have a length field set.
2221
2222 9. Added code to Exim itself and to exim_lock to test for a specific race
2223 condition that could lead to file corruption when using MBX delivery. The
2224 issue is with the lockfile that is created in /tmp. If this file is removed
2225 after a process has opened it but before that process has acquired a lock,
2226 there is the potential for a second process to recreate the file and also
2227 acquire a lock. This could lead to two Exim processes writing to the file
2228 at the same time. The added code performs the same test as UW imapd; it
2229 checks after acquiring the lock that its file descriptor still refers to
2230 the same named file.
2231
223210. The buffer for building added header lines was of fixed size, 8192 bytes.
2233 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
2234 when Exim is built.
2235
223611. Added the smtp_active_hostname option. If used, this will typically be made
2237 to depend on the incoming interface address. Because $interface_address is
2238 not set up until the daemon has forked a reception process, error responses
2239 that can happen earlier (such as "too many connections") no longer contain
2240 a host name.
2241
224212. If an expansion in a condition on a "warn" statement fails because a lookup
2243 defers, the "warn" statement is abandoned, and the next ACL statement is
2244 processed. Previously this caused the whole ACL to be aborted.
2245
224613. Added the iplsearch lookup type.
2247
224814. Added ident_timeout as a log selector.
2249
225015. Added tls_certificate_verified as a log selector.
2251
225216. Added a global option tls_require_ciphers (compare the smtp transport
2253 option of the same name). This controls incoming TLS connections.
2254
225517. I finally figured out how to make tls_require_ciphers do a similar thing
2256 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
2257 before starting the TLS session.
2258
225918. Tabs are now shown as \t in -bP output.
2260
226119. If the log selector return_path_on_delivery was set, Exim crashed when
2262 bouncing a message because it had too many Received: header lines.
2263
226420. If two routers both had headers_remove settings, and the first one included
2265 a superfluous trailing colon, the final name in the first list and the
2266 first name in the second list were incorrectly joined into one item (with a
2267 colon in the middle).
2268
2269
2270Exim version 4.32
2271-----------------
2272
2273 1. Added -C and -D options to the exinext utility, mainly to make it easier
2274 to include in the automated testing, but these could be helpful when
2275 multiple configurations are in use.
2276
2277 2. The exinext utility was not formatting the output nicely when there was
2278 an alternate port involved in the retry record key, nor when there was a
2279 message id as well (for retries that were specific to a specific message
2280 and a specific host). It was also confused by IPv6 addresses, because of
2281 the additional colons they contain. I have fixed the IPv4 problem, and
2282 patched it up to do a reasonable job for IPv6.
2283
2284 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
2285 delivery, the log line now contains "pipelined" if PIPELINING was used.
2286
2287 4. An SMTP transport process used to panic and die if the bind() call to set
2288 an explicit outgoing interface failed. This has been changed; it is now
2289 treated in the same way as a connect() failure.
2290
2291 5. A reference to $sender_host_name in the part of a conditional expansion
2292 that was being skipped was still causing a DNS lookup. This no longer
2293 occurs.
2294
2295 6. The def: expansion condition was not recognizing references to header lines
2296 that used bh_ and bheader_.
2297
2298 7. Added the _cache feature to named lists.
2299
2300 8. The code for checking quota_filecount in the appendfile transport was
2301 allowing one more file than it should have been.
2302
2303 9. For compatibility with Sendmail, the command line option
2304
2305 -prval:sval
2306
2307 is equivalent to
2308
2309 -oMr rval -oMs sval
2310
2311 and sets the incoming protocol and host name (for trusted callers). The
2312 host name and its colon can be omitted when only the protocol is to be set.
2313 Note the Exim already has two private options, -pd and -ps, that refer to
2314 embedded Perl. It is therefore impossible to set a protocol value of "d" or
2315 "s", but I don't think that's a major issue.
2316
231710. A number of refactoring changes to the code, none of which should affect
2318 Exim's behaviour:
2319
2320 (a) The number of logging options was getting close to filling up the
2321 32-bit word that was used as a bit map. I have split them into two classes:
2322 those that are passed in the argument to log_write(), and those that are
2323 only ever tested independently outside of that function. These are now in
2324 separate 32-bit words, so there is plenty of room for expansion again.
2325 There is no change in the user interface or the logging behaviour.
2326
2327 (b) When building, for example, log lines, the code previously used a
2328 macro that called string_cat() twice, in order to add two strings. This is
2329 not really sufficiently general. Furthermore, there was one instance where
2330 it was actually wrong because one of the argument was used twice, and in
2331 one call a function was used. (As it happened, calling the function twice
2332 did not affect the overall behaviour.) The macro has been replaced by a
2333 function that can join an arbitrary number of extra strings onto a growing
2334 string.
2335
2336 (c) The code for expansion conditions now uses a table and a binary chop
2337 instead of a serial search (which was left over from when there were very
2338 few conditions). Also, it now recognizes conditions like "pam" even when
2339 the relevant support is not compiled in: a suitably worded error message is
2340 given if an attempt is made to use such a condition.
2341
234211. Added ${time_interval:xxxxx}.
2343
234412. A bug was causing one of the ddress fields not to be passed back correctly
2345 from remote delivery subprocesses. The field in question was not being
2346 subsequently used, so this caused to problems in practice.
2347
234813. Added new log selectors queue_time and deliver_time.
2349
235014. Might have fixed a bug in maildirsizefile handling that threw up
2351 "unexpected character" debug warnings, and recalculated the data
2352 unnecessarily. In any case, I expanded the warning message to give more
2353 information.
2354
235515. Added the message "Restricted characters in address" to the statements in
2356 the default ACL that block characters like @ and % in local parts.
2357
235816. Change 71 for release 4.31 proved to be much less benign that I imagined.
2359 Three changes have been made:
2360
2361 (a) There was a serious bug; a negative response to MAIL caused the whole
2362 recipient domain to be cached as invalid, thereby blocking all messages
2363 to all local parts at the same domain, from all senders. This bug has
2364 been fixed. The domain is no longer cached after a negative response to
2365 MAIL if the sender used is not empty.
2366
2367 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
2368 been restored.
2369
2370 (c) A new callout option, "use_sender" has been added for people who want
2371 the modified behaviour.
2372
2373
2374Exim version 4.31
2375-----------------
2376
2377 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
2378 Larry Rosenman.
2379
2380 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
2381 indeed breaks things for older releases.
2382
2383 3. Added additional logging to the case where there is a problem reading data
2384 from a filter that is running in a subprocess using a pipe, in order to
2385 try to track down a specific problem.
2386
2387 4. Testing facility fudge: when running in the test harness and attempting
2388 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
2389 getting "No route to host". Convert this to a timeout.
2390
2391 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
2392 warning.
2393
2394 6. Some OS don't have socklen_t but use size_t instead. This affects the
2395 fifth argument of getsockopt() amongst other things. This is now
2396 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
2397 can be set for individual OS. I have set it for SunOS5, OSF1, and
2398 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
2399 some earlier ones do not.
2400
2401 7. Change 4.30/15 was not doing the test caselessly.
2402
2403 8. The standard form for an IPv6 address literal was being rejected by address
2404 parsing in, for example, MAIL and RCPT commands. An example of this kind of
2405 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
2406 this, as well as the form without the "IPv6" on the front (but only when
2407 address literals are enabled, of course).
2408
2409 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
2410
241110. Exim crashed if a message with an empty sender address specified by -f
2412 encountered a router with an errors_to setting. This could be provoked only
2413 by a command such as
2414
2415 exim -f "" ...
2416
2417 where an empty string was supplied; "<>" did not hit this bug.
2418
241911. Installed PCRE release 4.5.
2420
242112. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
2422 remained set. It is now erased.
2423
242413. exiqgrep wasn't working on MacOS X because it didn't correctly compute
2425 times from message ids (which are base 36 rather than the normal 62).
2426
242714. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
2428 were being counted as actual protocol errors, and logged if the log
2429 selector +smtp_protocol_error was set. One cannot be perfect in this test,
2430 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
2431 and DATA following a set of rejected RCPTs do not count as protocol errors.
2432 In other words, Exim assumes they were pipelined, though this may not
2433 actually be the case. Of course, in all cases the client gets an
2434 appropriate error code.
2435
243615. If a lookup fails in an ACL condition, a message about the failure may
2437 be available; it is used if testing the ACL cannot continue, because most
2438 such messages specify what the cause of the deferral is. However, some
2439 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
2440 that caused an old message to be retained and used if a later statement
2441 caused a defer, replacing the real cause of the deferral.
2442
244316. If an IP address had so many PTR records that the DNS lookup buffer
2444 was not large enough to hold them, Exim could crash while trying to process
2445 the truncated data. It now detects and logs this case.
2446
244717. Further to 4.21/58, another change has been made: if (and only if) the
2448 first line of a message (the first header line) ends with CRLF, a bare LF
2449 in a subsequent header line has a space inserted after it, so as not to
2450 terminate the header.
2451
245218. Refactoring: tidied an ugly bit of code in appendfile that copied data
2453 unnecessarily, used atoi() instead of strtol(), and didn't check the
2454 termination when getting file sizes from file names by regex.
2455
245619. Completely re-implemented the support for maildirsize files, in the light
2457 of a number of problems with the previous contributed implementation
2458 (4.30/29). In particular:
2459
2460 . If the quota is zero, the maildirsize file is maintained, but no quota is
2461 imposed.
2462
2463 . If the maildir directory does not exist, it is created before any attempt
2464 to write a maildirsize file.
2465
2466 . The quota value in the file is just a cache; if the quota is changed in
2467 the transport, the new value overrides.
2468
2469 . A regular expression is available for excluding directories from the
2470 count.
2471
247220. The autoreply transport checks the characters in options that define the
2473 message's headers; it allows continued headers, but it was checking with
2474 isspace() after an embedded newline instead of explicitly looking for a
2475 space or a tab.
2476
247721. If all the "regular" hosts to which an address was routed had passed their
2478 expiry times, and had not reached their retry times, the address was
2479 bounced, even if fallback hosts were defined. Now Exim should go on to try
2480 the fallback hosts.
2481
248222. Increased buffer sizes in the callout code from 1024 to 4096 to match the
2483 equivalent code in the SMTP transport. Some hosts send humungous responses
2484 to HELO/EHLO, more than 1024 it seems.
2485
248623. Refactoring: code in filter.c used (void *) for "any old type" but this
2487 gives compiler warnings in some environments. I've now done it "properly",
2488 using a union.
2489
249024. The replacement for inet_ntoa() that is used with gcc on IRIX systems
2491 (because of problems with the built-in one) was declared to return uschar *
2492 instead of char *, causing compiler failure.
2493
249425. Fixed a file descriptor leak when processing alias/forward files.
2495
249626. Fixed a minor format string issue in dbfn.c.
2497
249827. Typo in exim.c: ("dmbnz" for "dbmnz").
2499
250028. If a filter file refered to $h_xxx or $message_headers, and the headers
2501 contained RFC 2047 "words", Exim's memory could, under certain conditions,
2502 become corrupted.
2503
250429. When a sender address is verified, it is cached, to save repeating the test
2505 when there is more than one recipient in a message. However, when the
2506 verification involves a callout, it is possible for different callout
2507 options to be set for different recipients. It is too complicated to keep
2508 track of this in the cache, so now Exim always runs a verification when a
2509 callout is required, relying on the callout cache for the optimization.
2510 The overhead is duplication of the address routing, but this should not be
2511 too great.
2512
251330. Fixed a bug in callout caching. If a RCPT command caused the sender address
2514 to be verified with callout=postmaster, and the main callout worked but the
2515 postmaster check failed, the verification correctly failed. However, if a
2516 subsequent RCPT command asked for sender verification *without* the
2517 postmaster check, incorrect caching caused this verification also to fail,
2518 incorrectly.
2519
252031. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
2521 it was not caching the DNS options (qualify_single, search_parents) that
2522 were used when the lookup failed. A subsequent lookup with different
2523 options therefore always gave the same answer, though there were cases
2524 where it should not have. (Example: a "domains = !$mx_any" option on a
2525 dnslookup router: the "domains" option is always processed without any
2526 widening, but the router might have qualify_single set.) Now Exim uses the
2527 cached value only when the same options are set.
2528
252932. Added John Jetmore's "exipick" utility to the distribution.
2530
253133. GnuTLS: When an attempt to start a TLS session fails for any reason other
2532 than a timeout (e.g. a certificate is required, and is not provided), an
2533 Exim server now closes the connection immediately. Previously it waited for
2534 the client to close - but if the client is SSL, it seems that they each
2535 wait for each other, leading to a delay before one of them times out.
2536
253734: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
2538 maintained 0.8.x compatibility because I don't think many are using it, and
2539 it is clearly obsolete.
2540
254135. Added TLS support for CRLs: a tls_crl global option and one for the smtp
2542 transport.
2543
254436. OpenSSL: $tls_certificate_verified was being set to 1 even if the
2545 client certificate was expired. A simple patch fixes this, though I don't
2546 understand the full logic of why the verify callback is called multiple
2547 times.
2548
254937. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
2550 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
2551 which causes problems with some clients (such as the Certicom SSL Plus
2552 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
2553 disables the coutermeasure allowing Eudora to connect."
2554
255538. Exim was not checking that a write() to a log file succeeded. This could
2556 lead to Bad Things if a log got too big, in particular if it hit a file
2557 size limit. Exim now panics and dies if it cannot write to a log file, just
2558 as it does if it cannot open a log file.
2559
256039. Modified OS/Makefile-Linux so that it now contains
2561
2562 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
2563
2564 The two -D definitions ensure that Exim is compiled with large file
2565 support, which makes it possible to handle log files that are bigger than
2566 2^31.
2567
256840. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
2569 instance) a domain was checked against a named list that involved a lookup,
2570 causing $domain_data to be set, then another domain was checked against the
2571 same list, then the first domain was re-checked, the value of $domain_data
2572 after the final check could be wrong. In particular, if the second check
2573 failed, it could be set empty. This bug probably also applied to
2574 $localpart_data.
2575
257641. The strip_trailing_dot option was not being applied to the address given
2577 with the -f command-line option.
2578
257942. The code for reading a message's header from the spool was incrementing
2580 $received_count, but never initializing it. This meant that the value was
2581 incorrect (doubled) while delivering a message in the same process in which
2582 it was received. In the most common configuration of Exim, this never
2583 happens - a fresh exec is done - but it can happen when
2584 deliver_drop_privilege is set.
2585
258643. When Exim logs an SMTP synchronization error - client data sent too soon -
2587 it now includes up to 150 characters of the unexpected data in the log
2588 line.
2589
259044. The exim_dbmbuild utility uses fixed size buffers for reading input lines
2591 and building data strings. The size of both of these buffers was 10 000
2592 bytes - far larger than anybody would *ever* want, thought I. Needless to
2593 say, somebody hit the limit. I have increased the maximum line length to
2594 20 000 and the maximum data length of concatenated lines to 100 000. I have
2595 also fixed two bugs, because there was no checking on these buffers. Tsk,
2596 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
2597 buffer is too small.
2598
259945. The exim_dbmbuild utility did not support quoted keys, as Exim does in
2600 lsearch lookups. Now it does.
2601
260246. When parsing a route_list item in a manualroute router, a fixed-length
2603 buffer was used for the list of hosts. I made this 1024 bytes long,
2604 thinking that nobody would ever have a list of hosts that long. Wrong.
2605 Somebody had a whole pile of complicated expansion conditions, and the
2606 string was silently truncated, leading to an expansion error. It turns out
2607 that it is easier to change to an unlimited length (owing to other changes
2608 that have happened since this code was originally written) than to build
2609 structure for giving a limitation error. The length of the item that
2610 expands into the list of hosts is now unlimited.
2611
261247. The lsearch lookup could not handle data where the length of text line was
2613 more than 4095 characters. Such lines were truncated, leading to shortened
2614 data being returned. It should now handle lines of any length.
2615
261648. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
2617 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
2618 ACL").
2619
262049. Cosmetic tidy to scripts like exicyclog that are generated by globally
2621 replacing strings such as BIN_DIRECTORY in a source file: the replacement
2622 no longer happens in comment lines. A list of replacements is now placed
2623 at the head of all of the source files, except those whose only change is
2624 to replace PERL_COMMAND in the very first #! line.
2625
262650. Replaced the slow insertion sort in queue.c, for sorting the list of
2627 messages on the queue, with a bottom-up merge sort, using code contributed
2628 by Michael Haardt. This should make operations like -bp somewhat faster on
2629 large queues. It won't affect queue runners, except when queue_run_in_order
2630 is set.
2631
263251. Installed eximstats 1.31 in the distribution.
2633
263452. Added support for SRV lookups to the dnslookup router.
2635
263653. If an ACL referred to $message_body or $message_body_end, the value was not
2637 reset for any messages that followed in the same SMTP session.
2638
263954. The store-handling optimization for building very long strings was not
2640 differentiating between the different store pools. I don't think this
2641 actually made any difference in practice, but I've tidied it.
2642
264355. While running the routers to verify a sender address, $sender_address
2644 was still set to the sender address. This is wrong, because when routing to
2645 send a bounce to the sender, it would be empty. Therefore, I have changed
2646 it so that, while verifying a sender address, $sender_address is set to <>.
2647 (There is no change to what happens when verifying a recipient address.)
2648
264956. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
2650 target A or AAAA records (if not already returned) without resetting the
2651 qualify_single or search_parents options of the DNS resolver. These are
2652 inappropriate in this case because the targets of MX and SRV records must
2653 be FQDNs. A broken DNS record could cause trouble if it happened to have a
2654 target that, when qualified, matched something in the local domain. These
2655 two options are now turned off when doing these lookups.
2656
265757. It seems that at least some releases of Reiserfs (which does not have the
2658 concept of a fixed number of inodes) returns zero and not -1 for the
2659 number of available inodes. This interacted badly with check_spool_inodes,
2660 which assumed that -1 was the "no such thing" setting. What I have done is
2661 to check that the total number of inodes is greater than zero before doing
2662 the test of how many are available.
2663
266458. When a "warn" ACL statement has a log_message modifier, the message is
2665 remembered, and not repeated. This is to avoid a lot of repetition when a
2666 message has many recipients that cause the same warning to be written.
2667 Howewer, Exim was preserving the list of already written lines for an
2668 entire SMTP session, which doesn't seem right. The memory is now reset if a
2669 new message is started.
2670
267159. The "rewrite" debugging flag was not showing the result of rewriting in the
2672 debugging output unless log_rewrite was also set.
2673
267460. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
2675 of (int)(handle) when we know that handle contains (void *)(-1).
2676
267761. The Exim daemon panic-logs an error return when it closes the incoming
2678 connection. However "connection reset by peer" seems to be common, and
2679 isn't really an error worthy of noting specially, so that particular error
2680 is no long logged.
2681
268262. When Exim is trying to find all the local interfaces, it used to panic and
2683 die if the ioctl to get the interface flags failed. However, it seems that
2684 on at least one OS (Solaris 9) it is possible to have an interface that is
2685 included in the list of interfaces, but for which you get a failure error
2686 for this call. This happens when the interface is not "plumbed" into a
2687 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
2688 failure of the "get flags" call assumes that the interface is down.
2689
269063. Added a ${eval10: operator, which assumes all numbers are decimal. This
2691 makes life easier for people who are doing arithmetic on fields extracted
2692 from dates, where you often get leading zeros that should not be
2693 interpreted as octal.
2694
269564. Added qualify_domain to the redirect router, to override the global
2696 setting.
2697
269865. If a pathologically long header line contained very many addresses (the
2699 report of this problem mentioned 10 000) and each of them was rewritten,
2700 Exim could use up a very large amount of memory. (It kept on making new
2701 copies of the header line as it rewrote, and never released the old ones.)
2702 At the expense of a bit more processing, the header rewriting function has
2703 been changed so that it no longer eats memory in this way.
2704
270566. The generation of the Received: header has been moved from the time that a
2706 message starts to be received, to the time that it finishes. The timestamp
2707 in the Received: header should now be very close to that of the <= log
2708 line. There are two side-effects of this change:
2709
2710 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
2711 logged header lines no longer include the local Received: line, because
2712 it has not yet been created. The same applies to a copy of the message
2713 that is returned to a non-SMTP sender when a message is rejected.
2714
2715 (b) When a filter file is tested using -bf, no additional Received: header
2716 is added to the test message. After some thought, I decided that this
2717 is a bug fix.
2718
2719 This change does not affect the value of $received_for. It is still set
2720 after address rewriting, but before local_scan() is called.
2721
272267. Installed the latest Cygwin-specific files from the Cygwin maintainer.
2723
272468. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
2725 gave an unhelpful panic error message, and a defer error. I have managed to
2726 change this behaviour so that it now rejects any supplied certificate,
2727 which seems right, as the list of acceptable certificates is empty.
2728
272969. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
2730 gave an unhelpful defer error. I have not managed to make this reject any
2731 supplied certificates, but the error message it gives is "no certificate
2732 supplied", which is not helpful.
2733
273470. exigrep's output now also includes lines that are not associated with any
2735 message, but which match the given pattern. Implemented by a patch from
2736 Martin Sluka, which also tidied up the Perl a bit.
2737
273871. Recipient callout verification, like sender verification, was using <> in
2739 the MAIL FROM command. This isn't really the right thing, since the actual
2740 sender may affect whether the remote host accepts the recipient or not. I
2741 have changed it to use the actual sender in the callout; this means that
2742 the cache record is now keyed on a recipient/sender pair, not just the
2743 recipient address. There doesn't seem to be a real danger of callout loops,
2744 since a callout by the remote host to check the sender would use <>.
2745 [SEE ABOVE: changed after hitting problems.]
2746
274772. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
2748 temporary errors. However, in the case of such a code being given after
2749 the end of a data transmission (i.e. after ".") Exim was failing to write
2750 a retry record for the message. (Yes, there was some broken host that was
2751 actually sending 8xx at this point.)
2752
275373. An unknown lookup type in a host list could cause Exim to panic-die when
2754 the list was checked. (An example that provoked this was putting <; in the
2755 middle of a list instead of at the start.) If this happened during a DATA
2756 ACL check, a -D file could be left lying around. This kind of configuration
2757 error no longer causes Exim to die; instead it causes a defer errror. The
2758 incident is still logged to the main and panic logs.
2759
276074. Buglet left over from Exim 3 conversion. The message "too many messages
2761 in one connection" was written to the rejectlog but not the mainlog, except
2762 when address rewriting (yes!) was being logged.
2763
276475. Added write_rejectlog option.
2765
276676. When a system filter was run not as root (that is, when system_filter_user
2767 was set), the values of the $n variables were not being returned to the
2768 main process; thus, they were not subsequently available in the $sn
2769 variables.
2770
277177. Added +return_path_on_delivery log selector.
2772
277378. A connection timeout was being treated differently from recipients deferred
2774 when testing hosts_max_try with a message that was older than the host's
2775 retry timeout. (The host should not be counted, thus allowing all hosts to
2776 be tried at least once before bouncing.) This may have been the cause of an
2777 occasionally reported bug whereby a message would remain on the queue
2778 longer than the retry timeout, but would be bounced if a delivery was
2779 forced. I say "may" because I never totally pinned down the problem;
2780 setting up timeout/retry tests is difficult. See also the next item.
2781
278279. The ultimate address timeout was not being applied to errors that involved
2783 a combination of host plus message (for example, a timeout on a MAIL
2784 command). When an address resolved to a number of possible hosts, and they
2785 were not all tried for each delivery (e.g. because of hosts_max_try), a
2786 message could remain on the queue longer than the retry timeout.
2787
278880. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
2789 Haardt.
2790
279181. Fixed an obscure SMTP outgoing bug which required at least the following
2792 conditions: (a) there was another message waiting for the same server;
2793 (b) the server returned 5xx to all RCPT commands in the first message so
2794 that the message was not completed; (c) the server dropped the connection
2795 or gave a negative response to the RSET that Exim sends to abort the
2796 transaction. The observed case was a dropped connection after DATA that had
2797 been sent in pipelining mode. That is, the server had advertised PIPELINING
2798 but was not implementing it correctly. The effect of the bug was incorrect
2799 behaviour, such as trying another host, and this could lead to a crash.
2800
2801
2802Exim version 4.30
2803-----------------
2804
2805 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
2806 and daemon.c were passed as pointers to ints; they should have been
2807 pointers to socklen_t variables (which are typically unsigned ints).
2808
2809 2. Some signed/unsigned type warnings in the os.c file for Linux have been
2810 fixed.
2811
2812 3. Fixed a really odd bug that affected only the testing scheme; patching a
2813 certain fixed string in the binary changed the value of another string that
2814 happened to be identical to the end of the original first string.
2815
2816 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
2817 name", it returns that address as the IP address. On some operating
2818 systems (e.g. Solaris), it also passes back the IP address string as the
2819 "host name". However, on others (e.g. Linux), it passes back an empty
2820 string. Exim wasn't checking for this, and was changing the host name to an
2821 empty string, assuming it had been canonicized.
2822
2823 5. Although rare, it is permitted to have more than one PTR record for a given
2824 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
2825 all the names associated with an address, because they do in Solaris.
2826 However, it seems that they do not in Linux for data that comes from the
2827 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
2828 I found this out when I moved to a new Linux workstation and tried to run
2829 the Exim test suite.
2830
2831 To get round this problem I have changed the code so that it now does its
2832 own call to the DNS to look up PTR records when searching for a host name.
2833 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
2834 addresses that are only in /etc/hosts are still found.
2835
2836 This behaviour is, however, controlled by an option called host_lookup_
2837 order, which defaults to "bydns:byaddr". If people want to use the other
2838 order, or indeed, just use one or the other means of lookup, they can
2839 specify it in this variable.
2840
2841 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
2842 some operating systems, this comes back from gethostbyaddr() as an empty
2843 string, and this is what Exim used to test for. However, it seems that in
2844 other systems, "." is yielded. Exim now tests for this case too.
2845
2846 7. The values of check_spool_space and check_log_space are now held internally
2847 as a number of kilobytes instead of an absolute number of bytes. If a
2848 numbers is specified without 'K' or 'M', it is rounded up to the nearest
2849 kilobyte. This means that much larger values can be stored.
2850
2851 8. Exim monitor: an attempt to get the action menu when not actually pointing
2852 at a message produces an empty menu entitled "No message selected". This
2853 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
2854 no entries in it ("Shell widget menu has zero width and/or height"). So I
2855 have added a single, blank menu entry in this case.
2856
2857 9. Added ${quote_local_part.
2858
285910. MIME decoding is now applied to the contents of Subject: header lines when
2860 they are logged.
2861
286211. Now that a reference to $sender_host_address automatically causes a reverse
2863 lookup to occur if necessary (4.13/18), there is no need to arrange for a
2864 host lookup before query-style lookups in lists that might use this
2865 variable. This has therefore been abolished, and the "net-" prefix is no
2866 longer necessary for query-style lookups.
2867
286812. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
2869 have been a typo for LFLAGS, so it has been changed.
2870
287113. The install script calls Exim with "-C /dev/null" in order to find the
2872 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
2873 to be output. Howeve, since Exim outputs its version number before the
2874 error, it didn't break the script. It just looked ugly. I fixed this by
2875 always allowing "-C /dev/null" if the caller is root.
2876
287714. Ignore overlarge ACL variable number when reading spool file - insurance
2878 against a later release with more variables having written the file.
2879
288015. The standard form for an IPv6 address literal was being rejected by EHLO.
2881 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
2882 this, as well as the form without the "IPv6" on the front.
2883
288416. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
2885 OS/Makefile-Darwin file.
2886
288717. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
2888 only to LDAP libraries that do not have LDAP_OPT_DEREF.
2889
289018. After change 4.21/52, "%ld" was used to format the contents of the $inode
2891 variable. However, some OS use ints for inodes. I've added cast to long int
2892 to get rid of the compiler warning.
2893
289419. I had forgotten to lock out "/../" in configuration file names when
2895 ALT_CONFIG_PREFIX was set.
2896
289720. Routers used for verification do not need to specify transports. However,
2898 if such a router generated a host list, and callout was configured, Exim
2899 crashed, because it could not find a port number from the (non-existent)
2900 transport. It now assumes port 25 in this circumstance.
2901
290221. Added the -t option to exigrep.
2903
290422. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
2905 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
2906 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
2907 them is set, LOOKUP_LSEARCH is forced.
2908
290923. "exim -bV" now outputs a list of lookups that are included in the binary.
2910
291124. Added sender and host information to the "rejected by local_scan()" log
2912 line; previously there was no indication of these.
2913
291425. Added .include_if_exists.
2915
291626. Change 3.952/11 added an explicit directory sync on top of a file sync for
2917 Linux. It turns out that not all file systems support this. Apparently some
2918 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
2919 do it.) To cope with this, the error EINVAL, which means that sync-ing is
2920 not supported on the file descriptor, is now ignored when Exim is trying to
2921 sync a directory. This applies only to Linux.
2922
292327. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
2924
292528. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
2926 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
2927 this.
2928
292929. Added support for maildirsize files from supplied patch (modified a bit).
2930
293130. The use of :fail: followed by an empty string could lead Exim to respond to
2932 sender verification failures with (e.g.):
2933
2934 550 Verification failed for <xxx>
2935 550 Sender verify failed
2936
2937 where the first response line was missing the '-' that indicates it is not
2938 the final line of the response.
2939
294031. The loop for finding the name of the user that called Exim had a hardwired
2941 limit of 10; it now uses the value of finduser_retries, which is used for
2942 all other user lookups.
2943
294432. Added $received_count variable, available in data and not_smtp ACLs, and at
2945 delivery time.
2946
294733. Exim was neglecting to zero errno before one call of strtol() when
2948 expanding a string and expecting an integer value. On some systems this
2949 resulted in spurious "integer overflow" errors. Also, it was casting the
2950 result into an int without checking.
2951
295234. Testing for a connection timeout using "timeout_connect" in the retry rules
2953 did not work. The code looks as if it has *never* worked, though it appears
2954 to have been documented since at least releast 1.62. I have made it work.
2955
295635. The "timeout_DNS" error in retry rules, also documented since at least
2957 1.62, also never worked. As it isn't clear exactly what this means, and
2958 clearly it isn't a major issue, I have abolished the feature by treating it
2959 as "timeout", and writing a warning to the main and panic logs.
2960
296136. The display of retry rules for -brt wasn't always showing the error code
2962 correctly.
2963
296437. Added new error conditions to retry rules: timeout_A, timeout_MX,
2965 timeout_connect_A, timeout_connect_MX.
2966
296738. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
2968 to the empty sender.
2969
297039. The daemon was not analysing the content of -oX till after it had closed
2971 stderr and disconnected from the controlling terminal. This meant that any
2972 syntax errors were only noted on the panic log, and the return code from
2973 the command was 0. By re-arranging the code a little, I've made the
2974 decoding happen first, so such errors now appear on stderr, and the return
2975 code is 1. However, the actual setting up of the sockets still happens in
2976 the disconnected process, so errors there are still only recorded on the
2977 panic log.
2978
297940. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
2980 connections (as happens on some IP stacks) was logged at start up time as
2981 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
2982 it from "IPv6 and IPv4", which means that two separate sockets are being
2983 used.
2984
298541. The debug output for gethostbyname2() or getipnodebyname() failures now
2986 says whether AF_INET or AF_INET6 was passed as an argument.
2987
298842. Exiwhat output was messed up when time zones were included in log
2989 timestamps.
2990
299143. Exiwhat now gives more information about the daemon's listening ports,
2992 and whether -tls-on-connect was used.
2993
299444. The "port" option of the smtp transport is now expanded.
2995
299645. A "message" modifier in a "warn" statement in a non-message ACL was being
2997 silently ignored. Now an error message is written to the main and panic
2998 logs.
2999
300046. There's a new ACL modifier called "logwrite" which writes to a log file
3001 as soon as it is encountered.
3002
300347. Added $local_user_uid and $local_user_gid at routing time.
3004
300548. Exim crashed when trying to verify a sender address that was being
3006 rewritten to "<>".
3007
300849. Exim was recognizing only a space character after ".include". It now also
3009 recognizes a tab character.
3010
301150. Fixed several bugs in the Perl script that creates the exim.8 man page by
3012 extracting the relevant information from the specification. The man page no
3013 longer contains scrambled data for the -d option, and I've added a section
3014 at the front about calling Exim under different names.
3015
301651. Added "extra_headers" argument to the "mail" command in filter files.
3017
301852. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
3019 crash.
3020
302153. Installed eximstats 1.29.
3022
302354. Added transport_filter_timeout as a generic transport option.
3024
302555. Exim no longer adds an empty Bcc: header to messages that have no To: or
3026 Cc: header lines. This was required by RFC 822, but it not required by RFC
3027 2822.
3028
302956. Exim used to add From:, Date:, and Message-Id: header lines to any
3030 incoming messages that did not have them. Now it does so only if the
3031 message originates locally, that is, if there is no associated remote host
3032 address. When Resent- header lines are present, this applies to the Resent-
3033 lines rather than the non-Resent- lines.
3034
303557. Drop incoming SMTP connection after too many syntax or protocol errors. The
3036 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
3037
303858. Messages for configuration errors now include the name of the main
3039 configuration file - useful now that there may be more than one file in a
3040 list (.included file names were always shown).
3041
304259. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
3043 for those rare installations that do not start the daemon as root or run it
3044 setuid root. I've cut out the call to initgroups() if the daemon is not
3045 root at that time.
3046
304760. The Exim user and group can now be bound into the binary as text strings
3048 that are looked up at the start of Exim's processing.
3049
305061. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
3051
305262. Added $mailstore_basename variable.
3053
305463. Installed patch to sieve.c from Michael Haardt.
3055
305664. When Exim failed to open the panic log after failing to open the main log,
3057 the original message it was trying to log was written to stderr and debug
3058 output, but if they were not available (the usual case in production), it
3059 was lost. Now it is written to syslog before the two lines that record the
3060 failures to open the logs.
3061
306265. Users' Exim filters run in subprocesses under the user's uid. It is
3063 possible for a "deliver" command or an alias in a "personal" command to
3064 provoke an address rewrite. If logging of address rewriting is configured,
3065 this fails because the process is not running as root or exim. There may be
3066 a better way of dealing with this, but for the moment (because 4.30 needs
3067 to be released), I have disabled address rewrite logging when running a
3068 filter in a non-root, non-exim process.
3069
3070
3071Exim version 4.24
3072-----------------
3073
3074 1. The buildconfig auxiliary program wasn't quoting the value set for
3075 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
3076 not defined. This bug was masked in 4.22 by the effect that was fixed in
3077 change 4.23/1.
3078
3079 2. Some messages that were rejected after a message id was allocated were
3080 shown as "incomplete" by exigrep. It no longer does this for messages that
3081 are rejected by local_scan() or the DATA or non-SMTP ACLs.
3082
3083 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
3084 have allow_domain_literals set, the ID did not get logged in the <= line.
3085 Domain literals are now always recognized in Message-ID: header lines.
3086
3087 4. The first argument for a ${extract expansion item is the key name or field
3088 number. Leading and trailing spaces in this item were not being ignored,
3089 causing some misleading effects.
3090
3091 5. When deliver_drop_privilege was set, single queue runner processes started
3092 manually (i.e. by the command "exim -q") or by the daemon (which uses the
3093 same command in the process it spins off) were not dropping privilege.
3094
3095 6. When the daemon running as "exim" started a queue runner, it always
3096 re-executed Exim in the spun-off process. This is a waste of effort when
3097 deliver_drop_privilege is set. The new process now just calls the
3098 queue-runner function directly.
3099
3100
3101Exim version 4.23
3102-----------------
3103
3104 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
3105 HEADERS_CHARSET.
3106
3107 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
3108 ignored. Though the use of -oP was forcing the writing of a pid file, it
3109 was always written to the default place.
3110
3111 3. If the message "no IP address found for host xxxx" is generated during
3112 incoming verification, it is now followed by identification of the incoming
3113 connection (so you can more easily find what provoked it).
3114
3115 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
3116
3117 5. Added some features to "harden" Exim a bit more against certain attacks:
3118
3119 (a) There is now a build-time option called FIXED_NEVER_USERS that can
3120 be put in Local/Makefile. This is like the never_users runtime option,
3121 but it cannot be overridden. The default setting is "root".
3122
3123 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
3124 prefix string with which any file named in a -C command line option
3125 must start.
3126
3127 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
3128 is retained for -C and -D only if the caller of Exim is root. Without
3129 it, the exim user may also use -C and -D and retain privilege.
3130
3131 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
3132 command line option is disabled.
3133
3134 6. Macro names set by the -D option must start with an upper case letter, just
3135 like macro names defined in the configuration file.
3136
3137 7. Added "dereference=" facility to LDAP.
3138
3139 8. Two instances of the typo "uknown" in the source files are fixed.
3140
3141 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
3142 the Configure-Makefile script screwed up while processing it.
3143
314410. Incorporated PCRE 4.4.
3145
314611. The SMTP synchronization check was not operating right at the start of an
3147 SMTP session. For example, it could not catch a HELO sent before the client
3148 waited for the greeting. There is now a check for outstanding input at the
3149 point when the greeting is written. Because of the duplex, asynchronous
3150 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
3151 way, but not yet received, when the check is performed.
3152
315312. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
3154 on TCP/IP sockets, because this apparently causes some broken clients to
3155 timeout.
3156
315713. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
3158 unchanged) from the Cygwin maintainer.
3159
316014. The code for -bV that shows what is in the binary showed "mbx" when maildir
3161 was supported instead of testing for mbx. Effectively a typo.
3162
316315. The spa authenticator server code was not checking that the input it
3164 received was valid base64.
3165
316616. The debug output line for the "set" modifier in ACLs was not showing the
3167 name of the variable that was being set.
3168
316917. Code tidy: the variable type "vtype_string" was never used. Removed it.
3170
317118. Previously, a reference to $sender_host_name did not cause a DNS reverse
3172 lookup on its own. Something else was needed to trigger the lookup. For
3173 example, a match in host_lookup or the need for a host name in a host list.
3174 Now, if $sender_host_name is referenced and the host name has not yet been
3175 looked up, a lookup is performed. If the lookup fails, the variable remains
3176 empty, and $host_lookup_failed is set to "1".
3177
317819. Added "eqi" as a case-independent comparison operator.
3179
318020. The saslauthd authentication condition could segfault if neither service
3181 nor realm was specified.
3182
318321. If an overflowing value such as "2048M" was set for message_size_limit, the
3184 error message that was logged was misleading, and incoming SMTP
3185 connections were dropped. The message is now more accurate, and temporary
3186 errors are given to SMTP connections.
3187
318822. In some error situations (such as 21 above) Exim rejects all SMTP commands
3189 (except RSET) with a 421 error, until QUIT is received. However, it was
3190 failing to send a response to QUIT.
3191
319223. The HELO ACL was being run before the code for helo_try_verify_hosts,
3193 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
3194 ACL is now run after the helo_try_verify_hosts code.
3195
319624. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
3197 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
3198 case-independent, so other case variants are also recognized). Apparently
3199 some systems use these upper case variants.
3200
320125. If more than two messages were waiting for the same host, and a transport
3202 filter was specified for the transport, Exim sent two messages over the
3203 same TCP/IP connection, and then failed with "socket operation on non-
3204 socket" when it tried to send the third.
3205
320626. Added Exim::debug_write and Exim::log_write for embedded Perl use.
3207
320827. The extern definition of crypt16() in expand.c was not being excluded when
3209 the OS had its own crypt16() function.
3210
321128. Added bounce_return_body as a new option, and bounce_return_size_limit
3212 as a preferred synonym for return_size_limit, both as an option and as an
3213 expansion variable.
3214
321529. Added LIBS=-liconv to OS/Makefile-OSF1.
3216
321730. Changed the default configuration ACL to relax the local part checking rule
3218 for addresses that are not in any local domains. For these addresses,
3219 slashes and pipe symbols are allowed within local parts, but the sequence
3220 /../ is explicitly forbidden.
3221
322231. SPA server authentication was not clearing the challenge buffer before
3223 using it.
3224
322532. log_message in a "warn" ACL statement was writing to the reject log as
3226 well as to the main log, which contradicts the documentation and doesn't
3227 seem right (because no rejection is happening). So I have stopped it.
3228
322933. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
3230 However, I am unable to do any testing of this.
3231
323234. Fixed an infelicity in the appendfile transport. When checking directories
3233 for a mailbox, to see if any needed to be created, it was accidentally
3234 using path names with one or more superfluous leading slashes; tracing
3235 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
3236
323735. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
3238 contents are added to the log line that is written for every discarded
3239 recipient. (Previously a log_message setting was ignored.)
3240
324136. The ${quote: operator now quotes the string if it is empty.
3242
324337. The install script runs exim in order to find its version number. If for
3244 some reason other than non-existence or emptiness, which it checks, it
3245 could not run './exim', it was installing it with an empty version number,
3246 i.e. as "exim-". This error state is now caught, and the installation is
3247 aborted.
3248
324938. An argument was missing from the function that creates an error message
3250 when Exim fails to connect to the socket for saslauthd authentication.
3251 This could cause Exim to crash, or give a corrupted message.
3252
325339. Added isip, isip4, and isip6 to ${if conditions.
3254
325540. The ACL variables $acl_xx are now saved with the message, and can be
3256 accessed later in routers, transports, and filters.
3257
325841. The new lookup type nwildlsearch is like wildlsearch, except that the key
3259 strings in the file are not string-expanded.
3260
326142. If a MAIL command specified a SIZE value that was too large to fit into an
3262 int variable, the check against message_size_limit failed. Such values are
3263 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
3264 day this will have to be increased, but I don't think I want to be around
3265 when emails are that large.
3266
3267
3268
3269Exim version 4.22
3270-----------------
3271
3272 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
3273 iconv() is not standard in FreeBSD.
3274
3275 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
3276 IPv6 enabled. The observed symptom was a segmentation fault on return from
3277 the function os_common_find_running_interfaces() in src/os.c.
3278
3279 3. In the check_special_case() function in daemon.c I had used "errno" as an
3280 argument name, which causes warnings on some systems. This was basically a
3281 typo, since it was named "eno" in the comments!
3282
3283 4. The code that waits for the clock to tick (at a resolution of some fraction
3284 of a second) so as to ensure message-id uniqueness was always waiting for
3285 at least one whole tick, when it could have waited for less. [This is
3286 almost certainly not relevant at current processor speeds, where it is
3287 unlikely to ever wait at all. But we try to future-proof.]
3288
3289 5. The function that sleeps for a time interval that includes fractions of a
3290 second contained a race. It did not block SIGALRM between setting the
3291 timer, and suspending (a couple of lines later). If the interval was short
3292 and the sigsuspend() was delayed until after it had expired, the suspension
3293 never ended. On busy systems this could lead to processes getting stuck for
3294 ever.
3295
3296 6. Some uncommon configurations may cause a lookup to happen in a queue runner
3297 process, before it forks any delivery processes. The open lookup caching
3298 mechanism meant that the open file or database connection was passed into
3299 the delivery process. The problem was that delivery processes always tidy
3300 up cached lookup data. This could cause a problem for the next delivery
3301 process started by the queue runner, because the external queue runner
3302 process does not know about the closure. So the next delivery process
3303 still has data in the lookup cache. In the case of a file lookup, there was
3304 no problem because closing a file descriptor in a subprocess doesn't affect
3305 the parent. However, if the lookup was caching a connection to a database,
3306 the connection was closed, and the second delivery process was likely to
3307 see errors such as "PGSQL: query failed: server closed the connection
3308 unexpectedly". The problem has been fixed by closing all cached lookups
3309 in a queue runner before running a delivery process.
3310
3311 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
3312 seem to have the "const" qualifier which it has on other OS. I've
3313 parameterised it.
3314
3315 8. Change 4.21/2 was too strict. It is only if there are two authenticators
3316 *of the same type* (client or server) with the same public name that an
3317 error should be diagnosed.
3318
3319 9. When Exim looked up a host name for an IP address, but failed to find the
3320 original IP address when looking up the host name (a safety check), it
3321 output the message "<ip address> does not match any IP for NULL", which was
3322 confusing, to say the least. The bug was that the host name should have
3323 appeared instead of "NULL".
3324
332510. Since release 3.03, if Exim is called by a uid other than root or the Exim
3326 user that is built into the binary, and the -C or -D options is used, root
3327 privilege is dropped before the configuration file is read. In addition,
3328 logging is switched to stderr instead of the normal log files. If the
3329 configuration then re-defines the Exim user, the unprivileged environment
3330 is probably not what is expected, so Exim logs a panic warning message (but
3331 proceeds).
3332
3333 However, if deliver_drop_privilege is set, the unprivileged state may well
3334 be exactly what is intended, so the warning has been cut out in that case,
3335 and Exim is allowed to try to write to its normal log files.
3336
3337
3338Exim version 4.21
3339-----------------
3340
3341 1. smtp_return_error_details was not giving details for temporary sender
3342 or receiver verification errors.
3343
3344 2. Diagnose a configuration error if two authenticators have the same public
3345 name.
3346
3347 3. Exim used not to create the message log file for a message until the first
3348 delivery attempt. This could be confusing when incoming messages were held
3349 for policy or load reasons. The message log file is now created at the time
3350 the message is received, and an initial "Received" line is written to it.
3351
3352 4. The automatically generated man page for command line options had a minor
3353 bug that caused no ill effects; however, a more serious problem was that
3354 the procedure for building the man page automatically didn't always
3355 operate. Consequently, release 4.20 contains an out-of-date version. This
3356 shouldn't happen again.
3357
3358 5. When building Exim with embedded Perl support, the script that builds the
3359 Makefile was calling 'perl' to find its compile-time parameters, ignoring
3360 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
3361
3362 6. The freeze_tell option was not being used for messages that were frozen on
3363 arrival, either by an ACL or by local_scan().
3364
3365 7. Added the smtp_incomplete_transaction log selector.
3366
3367 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
3368 was accepting AUTH without a new EHLO.
3369
3370 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
3371 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
3372 RFC.
3373
337410. Logging of TCP/IP connections (when configured) now happens in the main
3375 daemon process instead of the child process, so that the TCP/IP connection
3376 count is more accurate (but it can never be perfect).
3377
337811. The use of "drop" in a nested ACL was not being handled correctly in the
3379 outer ACL. Now, if condition failure induced by the nested "drop" causes
3380 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
3381 or "require"), the connection is dropped.
3382
338312. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
3384 that yield "discard" can now be used with an "accept" or a "discard" verb,
3385 but an error is generated for any others (because I can't see a useful way
3386 to define what should happen).
3387
338813. When an ACL is read dynamically from a file (or anywhere else), the lines
3389 are now processed in the same way as lines in the Exim configuration file.
3390 In particular, continuation lines are supported.
3391
339214. Added the "dnslists = a.b.c!=n.n.n.n" feature.
3393
339415. Added -ti meaning -t -i.
3395
339616. Check for letters, digits, hyphens, and dots in the names of dnslist
3397 domains, and warn by logging if others are found.
3398
339917. At least on BSD, alignment is not guarenteed for the array of ifreq's
3400 returned from GIFCONF when Exim is trying to find the list of interfaces on
3401 a host. The code in os.c has been modified to copy each ifreq to an aligned
3402 structure in all cases.
3403
3404 Also, in some cases, the returned ifreq's were being copied to a 'struct
3405 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
3406 means the last couple of bytes of an IPv6 address could be chopped if the
3407 ifreq contained only a normal sockaddr (14 bytes storage).
3408
340918. Named domain lists were not supported in the hosts_treat_as_local option.
3410 An entry such as +xxxx was not recognized, and was treated as a literal
3411 domain name.
3412
341319. Ensure that header lines added by a DATA ACL are included in the reject log
3414 if the ACL subsequently rejects the message.
3415
341620. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
3417 MD5 (which is deprecated).
3418
341921. When testing a filter file using -bf, Exim was writing a message when it
3420 took the sender from a "From " line in the message, but it was not doing so
3421 when it took $return_path from a Return-Path: header line. It now does.
3422
342322. If the contents of a "message" modifier for a "warn" ACL verb do not begin
3424 with a valid header line field name (a series of printing characters
3425 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
3426
342723. Changed "disc" in the source to "disk" to conform to the documentation and
3428 the book and for uniformity.
3429
343024. Ignore Sendmail's -Ooption=value command line item.
3431
343225. When execve() failed while trying to run a command in a pipe transport,
3433 Exim was returning EX_UNAVAILBLE (69) from the subprocess. However, this
3434 could be confused with a return value of 69 from the command itself. This
3435 has been changed to 127, the value the shell returns if it is asked to run
3436 a non-existent command. The wording for the related log line suggests a
3437 non-existent command as the problem.
3438
343926. If received_header_text expands to an empty string, do not add a Received:
3440 header line to the message. (Well, it adds a token one on the spool, but
3441 marks it "old" so that it doesn't get used or transmitted.)
3442
344327. Installed eximstats 1.28 (addition of -nt option).
3444
344528. There was no check for failure on the call to getsockname() in the daemon
3446 code. This can fail if there is a shortage of resources on the system, with
3447 ENOMEM, for example. A temporary error is now given on failure.
3448
344929. Contrary to the C standard, it seems that in some environments, the
3450 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
3451 program. Exim now does this explicitly; it affects the formatting of
3452 timestamps using strftime().
3453
345430. If exiqsumm was given junk data, it threw up some uninitialized variable
3455 complaints. I've now initialized all the variables, to avoid this.
3456
345732. Header lines added by a system filter were not being "seen" during
3458 transport-time rewrites.
3459
346033. The info_callback() function passed to OpenSSL is set up with type void
3461 (*)(SSL *, int, int), as described somewhere. However, when calling the
3462 function (actually a macro) that sets it up, the type void(*)() is
3463 expected. I've put in a cast to prevent warnings from picky compilers.
3464
346534. If a DNS black list lookup found a CNAME record, but there were no A
3466 records associated with the domain it pointed at, Exim crashed.
3467
346835. If a DNS black list lookup returned more than one A record, Exim ignored
3469 all but the first. It now scans all returned addresses if a particular IP
3470 value is being sought. In this situation, the contents of the
3471 $dnslist_value variable are a list of all the addresses, separated by a
3472 comma and a space.
3473
347436. Tightened up the rules for host name lookups using reverse DNS. Exim used
3475 to accept a host name and all its aliases if the forward lookup for any of
3476 them yielded the IP address of the incoming connection. Now it accepts only
3477 those names whose forward lookup yields the correct IP address. Any other
3478 names are discarded. This closes a loophole whereby a rogue DNS
3479 administrator could create reverse DNS records to break through a
3480 wildcarded host restriction in an ACL.
3481
348237. If a user filter or a system filter that ran in a subprocess used any of
3483 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
3484 the wrong values were passed to the pipe command ($thisaddress had the
3485 value of $0, $0 had the value of $1, etc). This bug was introduced by
3486 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
3487
348838. Improved the line breaking for long SMTP error messages from ACLs.
3489 Previously, if there was no break point between 40 and 75 characters, Exim
3490 left the rest of the message alone. Two changes have been made: (a) I've
3491 reduced the minimum length to 35 characters; (b) if it can't find a break
3492 point between 35 and 75 characters, it looks ahead and uses the first one
3493 that it finds. This may give the occasional overlong line, but at least the
3494 remaining text gets split now.
3495
349639. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
3497 file descriptors might be low, and that setting 1000 would always raise it.
3498 It turns out that in some environments, the limit is already over 1000 and
3499 that lowering it causes trouble. So now Exim takes care not to decrease it.
3500
350140. When delivering a message, the value of $return_path is set to $sender_
3502 address at the start of routing (routers may change the value). By an
3503 oversight, this default was not being set up when an address was tested by
3504 -bt or -bv, which affected the outcome if any router or filter referred to
3505 $return_path.
3506
350741. The idea of the "warn" ACL verb is that it adds a header or writes to the
3508 log only when "message" or "log_message" are set. However, if one of the
3509 conditions was an address verification, or a call to a nested ACL, the
3510 messages generated by the underlying test were being passed through. This
3511 no longer happens. The underlying message is available in $acl_verify_
3512 message for both "message" and "log_message" expansions, so it can be
3513 passed through if needed.
3514
351542. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
3516 new expansion $bh_ to give the encoded byte string without charset
3517 translation. Translation happens only if iconv() is available; HAVE_ICONV
3518 indicates this at build time. HEADERS_CHARSET gives the charset to
3519 translate to; headers_charset can change it in the configuration, and
3520 "headers charset" can change it in an individual filter file.
3521
352243. Now that we have a default RFC 2047 charset (see above), the code in Exim
3523 that creates RFC 2047 encoded "words" labels them as that charset instead
3524 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
3525 expansion operator; (ii) when Exim creates a From: line for a local
3526 message; (iii) when a header line is rewritten to include a "phrase" part.
3527
352844. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
3529 buggy, causing it to skip the first lines of messages whose message ID
3530 ended in 'D'. This would not have bitten before Exim release 4.14, because
3531 message IDs were unlikely to end in 'D' before then. The effect was to have
3532 incorrect size information for certain domains.
3533
353445. #include "config.h" was missing at the start of the crypt16.c module. This
3535 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
3536 noticed.
3537
353846. If there was a timeout during a "random" callout check, Exim treated it as
3539 a failure of the random address, and carried on sending RSET and the real
3540 address. If the delay was just some slowness somewhere, the response to the
3541 original RCPT would be taken as a response to RSET and so on, causing
3542 mayhem of various kinds.
3543
354447. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
3545 when I implemented it. It didn't allow for the fact that some option values
3546 may legitimatetly be negative (e.g. size_addition), and it didn't even do
3547 the right test for positive values.
3548
354948. Domain names in DNS records are case-independent. Exim always looks them up
3550 in lower case. Some resolvers return domain names in exactly the case they
3551 appear in the zone file, that is, they may contain uppercase letters. Not
3552 all resolvers do this - some return always lower case. Exim was treating a
3553 change of case by a resolver as a change of domain, similar to a widening
3554 of a domain abbreviation. This triggered its re-routing code and so it was
3555 trying to route what was effectively the same domain again. This normally
3556 caused routing to fail (because the router wouldn't handle the domain
3557 twice). Now Exim checks for this case specially, and just changes the
3558 casing of the domain that it ultimately uses when it transmits the message
3559 envelope.
3560
356149. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
3562 module.
3563
356450. If a filter generated a file delivery with a non-absolute name (possible if
3565 no home directory exists for the router), the forbid_file option was not
3566 forbidding it.
3567
356851. Added '&' feature to dnslists, to provide bit mask matching in addition to
3569 the existing equality matching.
3570
357152. Exim was using ints instead of ino_t variables in some places where it was
3572 dealing with inode numbers.
3573
357453. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
3575 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
3576 called TMPDIR, and if it finds it is different, it changes its value.
3577
357854. The smtp_printf() function is now made available to local_scan() so
3579 additional output lines can be written before returning. There is also an
3580 smtp_fflush() function to enable the detection of a dropped connection.
3581 The variables smtp_input and smtp_batched_input are exported to
3582 local_scan().
3583
358455. Changed the default runtime configuration: the message "Unknown user"
3585 has been removed from the ACL, and instead placed on the localuser router,
3586 using the cannot_route_message feature. This means that any verification
3587 failures that generate their own messages won't get overridden. Similarly,
3588 the "Unrouteable address" message that was in the ACL for unverifiable
3589 relay addresses has also been removed.
3590
359156. Added hosts_avoid_esmtp to the smtp transport.
3592
359357. The exicyclog script was not checking for the esoteric option
3594 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
3595 will work only if exicyclog is run under the appropriate euid.
3596
359758. Following a discussion on the list, the rules by which Exim recognises line
3598 endings on incoming messages have been changed. The -dropcr and drop_cr
3599 options are now no-ops, retained only for backwards compatibility. The
3600 following line terminators are recognized: LF CRLF CR. However, special
3601 processing applies to CR:
3602
3603 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
3604 nor a local message in the state where . is a terminator.
3605
3606 (ii) If a bare CR is encountered in a header line, an extra space is added
3607 after the line terminator so as not to end the header. The reasoning
3608 behind this is that bare CRs in header lines are most likely either
3609 to be mistakes, or people trying to play silly games.
3610
361159. The size of a message, as listed by "-bp" or in the Exim monitor window,
3612 was being incorrectly given as 18 bytes larger than it should have been.
3613 This is a VOB (very old bug).
3614
361560. This may never have affected anything current, but just in case it has:
3616 When the local host is found other than at the start of a list of hosts,
3617 the local host, those with the same MX, and any that follow, are discarded.
3618 When the list in question was part of a longer list of hosts, the following
3619 hosts (not currently being processed) were also being discarded. This no
3620 longer happens. I'm not sure if this situation could ever has previously
3621 arisen.
3622
362361. Added the "/MX" feature to lists of hosts in the manualroute and query
3624 program routers.
3625
362662. Whenever Exim generates a new message, it now adds an Auto-Submitted:
3627 header. This is something that is recommended in a new Internet Draft, and
3628 is something that is documented as being done by Sendmail. There are two
3629 possible values. For messages generated by the autoreply transport, Exim
3630 adds:
3631
3632 Auto-Submitted: auto-replied
3633
3634 whereas for all other generated messages (e.g. bounces) it adds
3635
3636 Auto-Submitted: auto-generated
3637
363863. The "personal" condition in filters now includes a test for the
3639 Auto-Submitted: header. If it contains the string "auto-" the message it
3640 not considered personal.
3641
364264. Added rcpt_include_affixes as a generic transport option.
3643
364465. Added queue_only_override (default true).
3645
364666. Added the syslog_duplication option.
3647
364867. If what should have been the first header line of a message consisted of
3649 a space followed by a colon, Exim was mis-interpreting it as a header line.
3650 It isn't of course - it is syntactically invalid and should therefore be
3651 treated as the start of the message body. The misbehaviour could have
3652 caused a number of strange effects, including loss of data in subsequent
3653 header lines, and spool format errors.
3654
365568. Formerly, the AUTH parameter on a MAIL command was trusted only if the
3656 client host had authenticated. This control can now be exercised by an ACL
3657 for more flexibility.
3658
365969. By default, callouts do not happen when testing with -bh. There is now a
3660 variant, -bhc, which does actually run the callout code, including
3661 consulting and updating the callout cache.
3662
366370. Added support for saslauthd authentication, courtesy of Alexander
3664 Sabourenkov.
3665
366671. If statvfs() failed on the spool or log directories while checking their
3667 size for availability, Exim confusingly gave the error "space shortage".
3668 Furthermore, in debugging mode it crashed with a floating point exception.
3669 These checks are done if check_{spool,log}_{space,inodes} are set, and when
3670 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
3671 serious problem, Exim now writes to the main and panic logs when this
3672 happens, with details of the failure. It then refuses to accept the
3673 incoming message, giving the message "spool directory problem" or "log
3674 directory problem" with a 421 code for SMTP messages.
3675
367672. When Exim is about to re-exec itself, it ensures that the file descriptors
3677 0, 1, and 2 exist, because some OS complain for execs without them (see
3678 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
3679 descriptors. However, the code omitted to check that the open succeeded,
3680 causing mysterious errors if for some reason the permissions on /dev/null
3681 got screwed. Now Exim writes a message to the main and panic logs, and
3682