Fix GnuTLS privatekey forced fail bug; in both TLS's treat an empty
[exim.git] / doc / doc-txt / ChangeLog
CommitLineData
c91535f3 1$Cambridge: exim/doc/doc-txt/ChangeLog,v 1.297 2006/02/14 14:12:06 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
214e2000 161
5de37277
PH
162Exim version 4.60
163-----------------
164
cc38ddbf
PH
165PH/01 Two changes to the default runtime configuration:
166
167 (1) Move the checks for relay_from_hosts and authenticated clients from
168 after to before the (commented out) DNS black list checks.
169
170 (2) Add control=submission to the relay_from_hosts and authenticated
171 clients checks, on the grounds that messages accepted by these
172 statements are most likely to be submissions.
5de37277 173
72fdd6ae
PH
174PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
175
176 (1) Generate an error if the third argument for the ${prvs expansion is
177 not a single digit.
178
179 (2) Treat a missing third argument of ${prvscheck as if it were an empty
180 string.
181
182 (3) Reset the variables that are obtained from the first argument of
183 ${prvscheck and used in the second argument before leaving the code,
184 because their memory is reclaimed, so using them afterwards may do
185 silly things.
186
187 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
188 one (it's much easier than Tom thought :-).
189
190 (5) Because of (4), we can now allow for the use of $prvscheck_result
191 inside the third argument.
cb9328de 192
cb741023
PH
193PH/03 For some reason, the default setting of PATH when running a command from
194 a pipe transport was just "/usr/bin". I have changed it to
195 "/bin:/usr/bin".
196
f174f16e
PH
197PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
198 anything to be listed in the output from -bV.
b2f5a032 199
c25242d7
PH
200PH/05 When a filter generated an autoreply, the entire To: header line was
201 quoted in the delivery log line, like this:
202
203 => >A.N.Other <ano@some.domain> <original@ddress> ...
204
205 This has been changed so that it extracts the operative address. There
206 may be more than one such address. If so, they are comma-separated, like
207 this:
208
209 => >ano@some.domain,ona@other.domain <original@ddress> ...
210
82c19f95
PH
211PH/06 When a client host used a correct literal IP address in a HELO or EHLO
212 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
213 not being looked up in the rDNS to get a host name, Exim was showing the
214 IP address twice in Received: lines, even though the IP addresses were
215 identical. For example:
216
217 Received: from [1.2.3.4] (helo=[1.2.3.4])
218
219 However, if the real host name was known, it was omitting the HELO data
220 if it matched the actual IP address. This has been tidied up so that it
221 doesn't show the same IP address twice.
222
d7ffbc12
PH
223PH/07 When both +timestamp and +memory debugging was on, the value given by
224 $tod_xxx expansions could be wrong, because the tod_stamp() function was
225 called by the debug printing, thereby overwriting the timestamp buffer.
226 Debugging no longer uses the tod_stamp() function when +timestamp is set.
227
9f526266
PH
228PH/08 When the original message was included in an autoreply transport, it
229 always said "this is a copy of the message, including all the headers",
230 even if body_only or headers_only was set. It now gives an appropriate
231 message.
232
87fcc8b9
PH
233PH/09 Applied a patch from the Sieve maintainer which:
234
235 o fixes some comments
236 o adds the (disabled) notify extension core
237 o adds some debug output for the result of if/elsif tests
238 o points to the current vacation draft in the documentation
239 and documents the missing references header update
240
241 and most important:
242
243 o fixes a bug in processing the envelope test (when testing
244 multiple envelope elements, the last element determinted the
245 result)
246
456682f5
PH
247PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
248 Electronic Mail") by including:
249
250 Auto-submitted: auto-generated
251
252 in the messages that it generates (bounce messages and others, such as
253 warnings). In the case of bounce messages for non-SMTP mesages, there was
254 also a typo: it was using "Auto_submitted" (underscore instead of
255 hyphen). Since every message generated by Exim is necessarily in response
256 to another message, thes have all been changed to:
257
258 Auto-Submitted: auto-replied
259
260 in accordance with these statements in the RFC:
261
262 The auto-replied keyword:
263
264 - SHOULD be used on messages sent in direct response to another
265 message by an automatic process,
266
267 - MUST NOT be used on manually-generated messages,
268
269 - MAY be used on Delivery Status Notifications (DSNs) and Message
270 Disposition Notifications (MDNs),
271
272 - MUST NOT be used on messages generated by automatic or periodic
273 processes, except for messages which are automatic responses to
274 other messages.
275
3e46c1aa
PH
276PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
277 to the default Received: header definition.
456682f5 278
49826d12
PH
279PH/12 Added log selector acl_warn_skipped (default on).
280
eba0c039
PH
281PH/13 After a successful wildlsearch lookup, discard the values of numeric
282 variables because (a) they are in the wrong storage pool and (b) even if
283 they were copied, it wouldn't work properly because of the caching.
284
a0d6ba8a
PH
285PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
286 checking when decoding. Apparently there are clients that generate
287 overlong encoded strings. Why am I not surprised?
288
f0917727
PH
289PH/15 If the first argument of "${if match_address" was not empty, but did not
290 contain an "@" character, Exim crashed. Now it writes a panic log message
291 and treats the condition as false.
292
096fee00
PH
293PH/16 In autoreply, treat an empty string for "once" the same as unset.
294
024bd3c2
PH
295PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
296 extension "envelope-auth". The code is finished and in agreement with
297 other implementations, but there is no documentation so far and in fact,
298 nobody wrote the draft yet. This extension is currently #undef'ed, thus
299 not changing the active code.
300
301 Print executed "if" and "elsif" statements when debugging is used. This
302 helps a great deal to understand what a filter does.
303
304 Document more things not specified clearly in RFC3028. I had all this
305 sorted out, when out of a sudden new issues came to my mind. Oops."
306
df199fec
PH
307PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
308 (Bugzilla #53).
309
d27f1df3
PH
310PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
311 canonical form (as documented). However, after a host name lookup from
312 the IP address, check_host() was doing a simple string comparison with
313 addresses acquired from the DNS when checking that the found name did
314 have the original IP as one of its addresses. Since any found IPv6
315 addresses are likely to be in abbreviated form, the comparison could
316 fail. Luckily, there already exists a function for doing the comparison
317 by converting both addresses to binary, so now that is used instead of
318 the text comparison.
319
96776534
PH
320PH/20 There was another similar case to PH/19, when a complete host name was
321 given in a host list; looking up its IP address could give an abbreviated
322 form, whereas the current host's name might or might not be abbreviated.
323 The same fix has been applied.
324
5de37277 325
9a799bc0
PH
326Exim version 4.54
327-----------------
328
329PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
330 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
331 It now does.
332
99a4b039
PH
333PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
334 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
335
336PH/03 Typo: missing ".o" in src/pcre/Makefile.
337
4b233853
PH
338PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
339 header line, restrict the check to what is listed in RFCs 2369 and 2929.
340 Also, for "Auto-Submitted", treat anything other than "no" as
341 non-personal, in accordance with RFC 3834. (Previously it treated
342 anything starting "auto-" as non-personal.)
343
8857ccfd
PH
344TF/01 The control=submission/name=... option had a problem with syntax
345 errors if the name included a slash character. The /name= option
346 now slurps the rest of the string, so it can include any characters
347 but it must come last in the list of options (after /sender_retain
348 or /domain=).
349
433a2980
PH
350PH/05 Some modifications to the interface to the fake nameserver for the new
351 testing suite.
352
3e46c1aa 353
9a799bc0 354
e3a311ba
TK
355Exim version 4.53
356-----------------
357
358TK/01 Added the "success_on_redirect" address verification option. See
359 NewStuff for rationale and an example.
360
13b685f9
PH
361PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
362
395ff96d
PH
363PH/02 Patch to exigrep to allow it to work on syslog lines.
364
5b68f6e4
PH
365PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
366 fread() to skip over the body file's header line, because in Cygwin the
367 header line is locked and is inaccessible.
368
1ab52c69
PH
369PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
370 co-exist for some time) to make it clear that it is the Exim ID that is
371 referenced, not the Message-ID: header line.
372
b07e6aa3
PH
373PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
374 string_format() function, because snprintf() does not exist on all
375 operating systems.
376
254e032f
PH
377PH/06 The use of forbid_filter_existstest now also locks out the use of the
378 ${stat: expansion item.
379
3af76a81
PH
380PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
381 protocol synchronization error", to keep the pedants happy.
382
2548ba04
PH
383PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
384 well as for IRIX systems, when gcc is being used. See the host.c source
385 file for comments.
386
b6c6011d
PH
387PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
388
cf39cf57
PH
389PH/10 Named domain lists were not working if used in a queue_smtp_domains
390 setting.
391
f1513293
PH
392PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
393 transport and to the smtp transport in LMTP mode.
394
727549a4
PH
395TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
396
af46795e
PH
397PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
398 run a filter in a subprocess. This could lead to confusion in subsequent
399 lookups in the parent process. There should also be a search_tidyup() at
400 the end of the subprocess.
401
d7b47fd0
PH
402PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
403 only if the host matched helo_try_verify_hosts, which caused the
404 verification to occur when the EHLO/HELO command was issued. The ACL just
405 tested the remembered result. Now, if a previous verification attempt has
406 not happened, "verify = helo" does it there and then.
407
ee744174
JJ
408JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
409
b582ab87
PH
410TK/03 Fix log output including CR from clamd.
411
41a13e0a
PH
412PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
413 exist provoked a memory error which could cause a segfault.
414
f625cc5a
PH
415PH/15 Installed PCRE 6.2
416
417PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
418
21f7af35
PH
419PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
420 of the problem. Specifically, suggested +O2 rather than +O1 for the
421 HP-UX compiler.
422
31480e42
PH
423PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
424
2d280592
PH
425PH/20 If a delivery was routed to a non-standard port by means of an SRV
426 record, the port was not correctly logged when the outgoing_port log
427 selector was set (it logged the transort's default port).
428
7cd1141b
PH
429PH/21 Added support for host-specific ports to manualroute, queryprogram,
430 fallback_hosts, and "hosts" in the smtp transport.
431
432PH/22 If the log selector "outgoing_port" is set, the port is now also given on
433 host errors such as "Connection refused".
434
750af86e
PH
435PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
436 authentication with radiusclient 0.4.9:
437
438 - Error returned from rc_read_config was caught wrongly
439 - Username/password not passed on to radius server due to wrong length.
440
441 The presumption is that some radiusclient API changes for 4.51/PH/17
442 were not taken care of correctly. The code is still untested by me (my
443 Linux distribution still has 0.3.2 of radiusclient), but it was
444 contributed by a Radius user.
445
446PH/24 When doing a callout, the value of $domain wasn't set correctly when
447 expanding the "port" option of the smtp transport.
448
4304270b
TK
449TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
450 while reading a MIME header. Thanks to Tom Hughes for a patch.
451
750af86e
PH
452PH/24 Include config.h inside local_scan.h so that configuration settings are
453 available.
454
64ffc24f
PH
455PH/25 Make $smtp_command_argument available after all SMTP commands. This means
456 that in an ACL for RCPT (for example), you can examine exactly what was
457 received.
458
5dd9625b
PH
459PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
460 commands, but it was not correctly comparing the address with the actual
461 client host address. Thus, it would show the EHLO address in Received:
462 header lines when this was not necessary.
463
5591031b
PH
464PH/27 Added the % operator to ${eval:}.
465
ba18e66a
PH
466PH/28 Exim tries to create and chdir to its spool directory when it starts;
467 it should be ignoring failures (because with -C, for example, it has lost
468 privilege). It wasn't ignoring creation failures other than "already
469 exists".
470
9cec981f
PH
471PH/29 Added "crypteq" to the list of supported features that Exim outputs when
472 -bV or -d is used.
473
aa2b5c79
PH
474PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
475 because an input line was too long, either on its own, or by virtue of
1509d3a8
PH
476 too many continuations, the temporary file was not being removed, and the
477 return code was incorrect.
aa2b5c79 478
48a53b7f
PH
479PH/31 Missing "BOOL" in function definition in filtertest.c.
480
1c59d63b
PH
481PH/32 Applied Sieve patches from the maintainer.
482
671012da
TK
483TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
484
1509d3a8
PH
485PH/33 Added "verify = not_blind".
486
487PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
488 Local/Makefile (with some defaults set). These are used in built scripts
489 such as exicyclog, but they have never been used in the exim_install
490 script (though there are many overriding facilities there). I have
491 arranged that the exim_install script now takes note of these two
492 settings.
493
494PH/35 Installed configuration files for Dragonfly.
495
2fe1a124
PH
496PH/36 When a locally submitted message by a trusted user did not contain a
497 From: header, and the sender address was obtained from -f or from an SMTP
498 MAIL command, and the trusted user did not use -F to supply a sender
499 name, $originator_name was incorrectly used when constructing a From:
500 header. Furthermore, $originator_name was used for submission mode
501 messages from external hosts without From: headers in a similar way,
502 which is clearly wrong.
503
8800895a
PH
504PH/37 Added control=suppress_local_fixups.
505
ccfdb010
PH
506PH/38 When log_selector = +received_sender was set, and the addition of the
507 sender made the log line's construction buffer exactly full, or one byte
508 less than full, an overflow happened when the terminating "\n" was
509 subsequently added.
510
1130bfb0
PH
511PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
512 when the result of a list match is failure because a DNS lookup failed.
513
ebcb507f
PH
514PH/40 RM_COMMAND is now used in the building process.
515
c35e155c
PH
516PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
517 the "build-* directories that it finds.
518
95d1f782
PH
519PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
520 address in a domain literal was a prefix of an interface address.
521
fd6de02e
PH
522PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
523 when verifying a sender address, unless rewrite_headers is false.
524
58de37c5
PH
525PH/44 Wrote a long comment about why errors_to addresses are verified as
526 recipients, not senders.
527
261cf466
TF
528TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
529 the ratelimit ACL was added.
530
3ee512ff
PH
531PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
532
e08c430f
PH
533PH/46 Added extra information about PostgreSQL errors to the error string.
534
bef5a11f
PH
535PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
536 avoiding the need to install special zones in a real server. This is
537 backwards compatible; if it can't find the fake resolver, it drops back.
538 Thus, both old and new test suites can be run.
539
7546de58
TF
540TF/02 Added util/ratelimit.pl
541
e5d5a95f
TF
542TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
543 clock is set back in time.
544
2e88a017
TF
545TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
546 Candler <B.Candler@pobox.com>.
547
a5f65aa4
TF
548TF/05 The fix for PH/43 was not completely correct; widen_domains is always
549 OK for addresses that are the result of redirections.
550
e7726cbf
PH
551PH/48 A number of further additions for the benefit of the new test suite,
552 including a fake gethostbyname() that interfaces to the fake DNS resolver
553 (see PH/47 above).
554
a7fdad5b
TF
555TF/06 The fix for widen_domains has also been applied to qualify_single and
556 search_parents which are the other dnslookup options that can cause
557 header rewrites.
558
6af56900
PH
559PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
560 type ("H").
561
0925ede6
PH
562PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
563
66afa403
TF
564TF/07 Exim produced the error message "an SRV record indicated no SMTP
565 service" if it encountered an MX record with an empty target hostname.
566 The message is now "an MX or SRV record indicated no SMTP service".
567
0154e85a
TF
568TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
569 if the DNS of the sending site is misconfigured. This is quite a
570 common situation. This change restores the behaviour of treating a
571 helo verification defer as a failure.
572
16f12c76
PH
573PH/51 If self=fail was set on a router, the bounce message did not include the
574 actual error message.
575
bbe902f0 576
e5a9dba6
PH
577Exim version 4.52
578-----------------
579
580TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
581
22c3b60b
PH
582PH/01 When a transport filter timed out in a pipe delivery, and the pipe
583 command itself ended in error, the underlying message about the transport
584 filter timeout was being overwritten with the pipe command error. Now the
585 underlying error message should be appended to the second error message.
586
06a9b4b5
PH
587TK/01 Fix poll() being unavailable on Mac OSX 10.2.
588
c1ac6996
PH
589PH/02 Reduce the amount of output that "make" produces by default. Full output
590 can still be requested.
591
9c7a242c
PH
592PH/03 The warning log line about a condition test deferring for a "warn" verb
593 was being output only once per connection, rather than after each
594 occurrence (because it was using the same function as for successful
595 "warn" verbs). This seems wrong, so I have changed it.
596
87ba3f5f
PH
597TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
598 it should not have, which might have caused a crash in the right
599 circumstances, but probably never did.
600
601PH/04 Installed a modified version of Tony Finch's patch to make submission
602 mode fix the return path as well as the Sender: header line, and to
603 add a /name= option so that you can make the user's friendly name appear
604 in the header line.
605
29aba418
TF
606TF/03 Added the control = fakedefer ACL modifier.
607
fe0dab11
TF
608TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
609 Mark Lowes for thorough testing.
870f6ba8 610
11d337a4
TK
611TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
612
613TK/03 Merged latest SRS patch from Miles Wilton.
614
415c8f3b
PH
615PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
616 with the definition in sysexits.h (which is #included earlier).
617 Fortunately, Exim does not actually use EX_OK. The code used to try to
618 preserve the sysexits.h value, by assumimg that macro definitions were
619 scanned for macro replacements. I have been disabused of this notion,
620 so now the code just undefines EX_OK before #including unistd.h.
11d337a4 621
958541e9
PH
622PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
623 in the smtp transport. When a block could not be written in a single
624 write() function, the timeout was being re-applied to each part-write.
625 This seems wrong - if the receiver was accepting one byte at a time it
626 would take for ever. The timeout is now adjusted when this happens. It
627 doesn't have to be particularly precise.
628
c206415f
TK
629TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
630 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
631
2a4be8f9
PH
632PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
633 without a domain if the check to <postmaster@domain> fails.
634
1cba11c5
SC
635SC/01 Eximstats: added -xls and the ability to specify output files
636 (patch written by Frank Heydlauf).
637
638SC/02 Eximstats: use FileHandles for outputing results.
639
640SC/03 Eximstats: allow any combination of xls, txt, and html output.
641
642SC/04 Eximstats: fixed display of large numbers with -nvr option
643
644SC/05 Eximstats: fixed merging of reports with empty tables.
645
646SC/06 Eximstats: added the -include_original_destination flag
647
648SC/07 Eximstats: removed tabs and trailing whitespace.
649
1005d00e
TK
650TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
651
652TK/06 MBOX spool code: Add real "From " MBOX separator line
653 so the .eml file is really in mbox format (even though
654 most programs do not really care). Patch from Alex Miller.
655
656TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
657 The latter is generated from $received_to and is only set if the
658 message has one envelope recipient. SA can use these headers,
659 obviously out-of-the-box. Patch from Alex Miller.
660
9b4768fa
PH
661PH/08 The ${def test on a variable was returning false if the variable's
662 value was "0", contrary to what the specification has always said!
663 The result should be true unless the variable is empty.
664
665PH/09 The syntax error of a character other than { following "${if
666 def:variable_name" (after optional whitespace) was not being diagnosed.
667 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
668 accidental colon was present, for example, could give incorrect results.
669
0d7eb84a
PH
670PH/10 Tidied the code in a number of places where the st_size field of a stat()
671 result is used (not including appendfile, where other changes are about
672 to be made).
673
674PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
675 This involved changing a lot of size variables from int to off_t. It
676 should work with maildirs and everything.
677
40727bee
TK
678TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
679 spamd dying while we are connected to it.
680
554d2369
TF
681TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
682 <jgh@wizmail.org>
683
1f922db1
PH
684PH/12 Applied Alex Kiernan's patch for the API change for the error callback
685 function for BDB 4.3.
686
ef213c3b
PH
687PH/13 Changed auto_thaw such that it does not apply to bounce messages.
688
8ac170f3
PH
689PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
690 the sources for PCRE have been re-arranged and more files are now
691 involved.
692
b1c749bb
PH
693PH/15 The code I had for printing potentially long long variables in PH/11
694 above was not the best (it lost precision). The length of off_t variables
695 is now inspected at build time, and an appropriate printing format (%ld
c6c2dc1d
PH
696 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
697 to be "long long int" or "long int". This is needed for the internal
698 formatting function string_vformat().
b1c749bb 699
4aac9b49
PH
700PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
701 the configuration file to be ":syslog", then the script "guesses" where
702 the logs files are, rather than using the compiled in default. In our
703 case the guess is not the same as the compiled default, so the script
704 suddenly stopped working when I started to use syslog. The patch checks
705 to see if log_file_path is "". If so, it attempts to read it from exim
706 with no configuration file to get the compiled in version, before it
707 falls back to the previous guessing code."
708
294520c8
TK
709TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
710 implementing BATV in an Exim configuration. See NewStuff for the gory
711 details.
712
5bd022fe
PH
713PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
714 Makefile that are specific to HP-UX.
715
90e9ce59
PH
716PH/18 If the "use_postmaster" option was set for a recipient callout together
717 with the "random" option, the postmaster address was used as the MAIL
718 FROM address for the random test, but not for the subsequent recipient
719 test. It is now used for both.
720
5ea81592
PH
721PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
722 patch removes a few documentation additions to RFC 3028, because the
723 latest draft now contains them. It adds the new en;ascii-case comparator
724 and a new error check for 8bit text in MIME parts. Comparator and
725 require names are now matched exactly. I enabled the subaddress
726 extension, but it is not well tested yet (read: it works for me)."
727
c6c2dc1d
PH
728PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
729 rework some of the code of TK/09 above to avoid the hardwired use of
730 "%lld" and "long long". Replaced the call to snprintf() with a call to
731 string_vformat().
732
fffffe4c
PH
733PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
734 records point to non-existent hosts", "retry timeout exceeded", and
735 "retry time not reached for any host after a long failure period".
ca02eafb 736
9a26b6b2
PH
737PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
738 experimental DomainKeys support:
739
740 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
741 (2) On an error such as an illegally used "control", the wrong name for
742 the control was given.
743
744 These problems did NOT occur unless DomainKeys support was compiled.
745
4aee0225
PH
746PH/23 Added daemon_startup_retries and daemon_startup_sleep.
747
32d668a5
PH
748PH/24 Added ${if match_ip condition.
749
8187c3f3
PH
750PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
751 databases so that it will be absolutely obvious if a crash occurs in the
752 DB library. This is a regular occurrence (often caused by mis-matched
753 db.h files).
754
ff790e47 755PH/26 Insert a lot of missing (void) casts for functions such as chown(),
f1e894f3
PH
756 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
757 picked up on a user's system that detects such things. There doesn't seem
758 to be a gcc warning option for this - only an attribute that has to be
759 put on the function's prototype. It seems that in Fedora Core 4 they have
760 set this on a number of new functions. No doubt there will be more in due
761 course.
ff790e47 762
5417f6d1
PH
763PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
764 specify a transport. However, if an address that was verified by such a
765 router was the subject of a callout, Exim crashed because it tried to
766 read the rcpt_include_affixes from the non-existent transport. Now it
767 just assumes that the setting of that option is false. This bug was
768 introduced by 4.51/PH/31.
769
59cf8544
PH
770PH/28 Changed -d+all to exclude +memory, because that information is very
771 rarely of interest, but it makes the output a lot bigger. People tend to
772 do -d+all out of habit.
773
e7ad8a65
PH
774PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
775 code in os-type was giving problems when libc.so lives in lib64, like on
776 x86_64 Fedora Core.
777
ade42478
PH
778PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
779 aren't the modern standard, and it seems that some systems' include files
780 don't always have them. Exim was already checking for some of the newer
781 ones like T_AAAA, and defining it itself. I've added checks for all the
782 record types that Exim uses.
783
182ad5cf
PH
784PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
785 not automatically generating a new one, as it is supposed to. This
786 prevented TLS from working. If the file did exist, but contained invalid
787 data, a new version was generated, as expected. It was only the case of a
788 non-existent file that was broken.
789
b0d9fc80
TK
790TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
791 with a change in libdomainkeys > 0.64.
792
793TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
794 from DNS. If the selector record carries the flag, it now has
795 precedence over the domain-wide flag.
796
797TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
798
47c7a64a
PH
799PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
800 the use of a port name that isn't defined in /etc/services) occurred, the
801 message was deferred as in a normal delivery, and thus remained on the
802 spool, instead of being failed because of the mua_wrapper setting. This
803 is now fixed, and I tidied up some of the mua_wrapper messages at the
804 same time.
805
a388bce4
SC
806SC/08 Eximstats: whilst parsing the mainlog(s), store information about
807 the messages in a hash of arrays rather than using individual hashes.
808 This is a bit cleaner and results in dramatic memory savings, albeit
809 at a slight CPU cost.
810
811SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
812 as requested by Marc Sherman.
813
814SC/10 Eximstats: added histograms for user specified patterns as requested
815 by Marc Sherman.
816
0793e4ed
SC
817SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
818
c58b88df
PH
819PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
820 fopen() in the content-scanning modules that did not already have it.
821
e7ad8a65 822
7982096b
PH
823Exim version 4.51
824-----------------
825
1a46a8c5
PH
826TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
827 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
828
2f079f46 829TK/02 Fix ACL "control" statement not being available in MIME ACL.
1a46a8c5
PH
830
831TK/03 Fix ACL "regex" condition not being available in MIME ACL.
832
833PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
834 to test Sieve filters that use "vacation".
835
836PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
837 that changes the way the GnuTLS parameters are stored in the cache file.
838 The new format can be generated externally. For backward compatibility,
839 if the data in the cache doesn't make sense, Exim assumes it has read an
840 old-format file, and it generates new data and writes a new file. This
841 means that you can't go back to an older release without removing the
842 file.
843
844PH/03 A redirect router that has both "unseen" and "one_time" set does not
845 work if there are any delivery delays because "one_time" forces the
846 parent to be marked "delivered", so its unseen clone is never tried
847 again. For this reason, Exim now forbids the simultaneous setting of
848 these two options.
849
850PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
851 redirected to themselves ("homonym" addresses). Read the long ChangeLog
852 entry if you want to know the details. The fix, however, neglected to
853 consider the case when local delivery batching is involved. The test for
854 "previously delivered" was not happening when checking to see if an
855 address could be batched with a previous (undelivered) one; under
856 certain circumstances this could lead to multiple deliveries to the same
c2c19e9d 857 address.
1a46a8c5
PH
858
859PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
860 in its include files, and this causes problems building Exim.
861
862PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
863 header_syntax) but Exim was just ignoring anything given after a slash.
864 In particular, this caused confusion with an attempt to use "verify =
865 reverse_host_lookup/defer_ok". An error is now given when options are
866 supplied for verify items that do not have them. (Maybe reverse_host_
867 lookup should have a defer_ok option, but that's a different point.)
868
869PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
870 defined by RFC 821) to 2048, because there were problems with some AUTH
871 commands, and RFC 1869 says the size should be increased for extended
872 SMTP commands that take arguments.
873
874PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
875 Finch).
876
877PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
878 "unknown" error; now it says that the functionality isn't in the binary.
8d67ada3 879
49c2d5ea
PH
880PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
881 an address' error message when a string expansion fails (syntax or
f331f3b6
PH
882 whatever). Otherwise the password may appear in the log. Following change
883 PH/42 below, there is no longer a chance of it appearing in a bounce
884 message.
49c2d5ea 885
bf759a8b
PH
886PH/11 Installed exipick version 20050225.0 from John Jetmore.
887
83364d30
PH
888PH/12 If the last host in a fallback_hosts list was multihomed, only the first
889 of its addresses was ever tried. (Bugzilla bug #2.)
890
7999bbd7
PH
891PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
892 the result incorrectly in the debug output. (It correctly added a newline
893 to what was transported.)
894
7dbf77c9
PH
895TF/01 Added $received_time.
896
74e0617f
PH
897PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
898 commented out examples of how to interface to a virus scanner and to
899 SpamAssassin. Also added commented examples of av_scanner and
900 spamd_address settings.
901
2f079f46
PH
902PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
903 and controls are allowed in which ACLs. There were a couple of minor
904 errors. Some of the entries in the conditions table (which is a table of
905 where they are NOT allowed) were getting very unwieldy; rewrote them as a
906 negation of where the condition IS allowed.
907
8c841523
PH
908PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
909
7766a4f0
PH
910PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
911 header file does not have a version number, so I've had to invent a new
912 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
913 API. The code is untested by me (my Linux distribution still has 0.3.2 of
914 radiusclient), but it was contributed by a Radius user.
915
8b417f2c
PH
916PH/18 Installed Lars Mainka's patch for the support of CRL collections in
917 files or directories, for OpenSSL.
918
901f42cb
PH
919PH/19 When an Exim process that is running as root has to create an Exim log
920 file, it does so in a subprocess that runs as exim:exim so as to get the
921 ownership right at creation (otherwise, other Exim processes might see
922 the file with the wrong ownership). There was no test for failure of this
923 fork() call, which would lead to the process getting stuck as it waited
924 for a non-existent subprocess. Forks do occasionally fail when resources
925 run out. I reviewed all the other calls to fork(); they all seem to check
926 for failure.
927
f9b9210e
PH
928PH/20 When checking for unexpected SMTP input at connect time (before writing
929 the banner), Exim was not dealing correctly with a non-positive return
930 from the read() function. If the client had disconnected by this time,
931 the result was a log entry for a synchronization error with an empty
932 string after "input=" when read() returned zero. If read() returned -1
933 (an event I could not check), uninitialized data bytes were printed.
934 There were reports of junk text (parts of files, etc) appearing after
935 "input=".
936
54cdb463
PH
937PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
938
cf00dad6
PH
939PH/22 Added support for macro redefinition, and (re)definition in between
940 driver and ACL definitions.
941
acb1b346
PH
942PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
943 forgetting to use the resulting value; it was using the unexpanded value.
944
c5ddb310
PH
945PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
946 hadn't been configured. The fix is from Juergen Kreileder, who
947 understands it better than I do:
948
949 "Here's what I see happening with three configured cyrus_sasl
950 authenticators configured (plain, login, cram-md5):
951
952 On startup auth_cyrus_sasl_init() gets called for each of these.
953 This means three calls to sasl_listmech() without a specified mech_list.
954 => SASL tests which mechs of all available mechs actually work
955 => three warnings about OTP not working
956 => the returned list contains: plain, login, cram-md5, digest-md5, ...
957
958 With the patch, sasl_listmech() also gets called three times. But now
959 SASL's mech_list option is set to the server_mech specified in the the
960 authenticator. Or in other words, the answer from sasl_listmech()
961 gets limited to just the mech you're testing for (which is different
962 for each call.)
963 => the return list contains just 'plain' or 'login', 'cram-md5' or
964 nothing depending on the value of ob->server_mech.
965
966 I've just tested the patch: Authentication still works fine,
967 unavailable mechs specified in the exim configuration are still
968 caught, and the auth.log warnings about OTP are gone."
969
31619da6
PH
970PH/25 When debugging is enabled, the contents of the command line are added
971 to the debugging output, even when log_selector=+arguments is not
972 specified.
973
bebaf0fc
PH
974PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
975 answer is "GNU", and only if the return is "GNU/something" is the answer
976 "Linux".
977
475fe28a
PH
978PH/27 $acl_verify_message is now set immediately after the failure of a
979 verification in an ACL, and so is available in subsequent modifiers. In
980 particular, the message can be preserved by coding like this:
981
982 warn !verify = sender
983 set acl_m0 = $acl_verify_message
984
985 Previously, $acl_verify_message was set only while expanding "message"
986 and "log_message" when a very denied access.
987
7e8bec7a
PH
988PH/28 Modified OS/os.c-Linux with
989
990 -#ifndef OS_LOAD_AVERAGE
991 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
992
993 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
994 nomenclature these days.)
995
e4a89c47
PH
996PH/29 Installed patch from the Sieve maintainer that adds the options
997 sieve_useraddress and sieve_subaddress to the redirect router.
998
5ca2a9a1
PH
999PH/30 In these circumstances:
1000 . Two addresses routed to the same list of hosts;
1001 . First host does not offer TLS;
1002 . First host accepts first address;
1003 . First host gives temporary error to second address;
1004 . Second host offers TLS and a TLS session is established;
1005 . Second host accepts second address.
1006 Exim incorrectly logged both deliveries with the TLS parameters (cipher
1007 and peerdn, if requested) that were in fact used only for the second
1008 address.
7e8bec7a 1009
c688b954
PH
1010PH/31 When doing a callout as part of verifying an address, Exim was not paying
1011 attention to any local part prefix or suffix that was matched by the
1012 router that accepted the address. It now behaves in the same way as it
1013 does for delivery: the affixes are removed from the local part unless
1014 rcpt_include_affixes is set on the transport.
1015
fed77020
PH
1016PH/32 Add the sender address, as F=<...>, to the log line when logging a
1017 timeout during the DATA phase of an incoming message.
1018
7fe1560f
PH
1019PH/33 Sieve envelope tests were broken for match types other than :is. I have
1020 applied a patch sanctioned by the Sieve maintainer.
c688b954 1021
ebb6e6d5
PH
1022PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
1023 the uid or gid is negative. A case of a negative gid caused this to be
1024 noticed. The fix allows for either to be negative.
1025
9c4e8f60
PH
1026PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
1027 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
1028 overlooked.
1029
1030PH/36 The change PH/12 above was broken. Fixed it.
1031
d7174846
PH
1032PH/37 Exim used to check for duplicate addresses in the middle of routing, on
1033 the grounds that routing the same address twice would always produce the
1034 same answer. This might have been true once, but it is certainly no
1035 longer true now. Routing a child address may depend on the previous
1036 routing that produced that child. Some complicated redirection strategies
1037 went wrong when messages had multiple recipients, and made Exim's
1038 behaviour dependent on the order in which the addresses were given.
1039
1040 I have moved the duplicate checking until after the routing is complete.
1041 Exim scans the addresses that are assigned to local and remote
1042 transports, and removes any duplicates. This means that more work will be
1043 done, as duplicates will always all be routed, but duplicates are
1044 presumably rare, so I don't expect this is of any significance.
1045
1046 For deliveries to pipes, files, and autoreplies, the duplicate checking
1047 still happens during the routing process, since they are not going to be
1048 routed further.
1049
cfe75fc3
PH
1050PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
1051 It corrects a timeout issue with spamd. This is Ian's comment: "The
1052 background is that sometimes spamd either never reads data from a
1053 connection it has accepted, or it never writes response data. The exiscan
1054 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
1055 blindly assumes that writes won't block so it may never time out."
1056
be22d70e
PH
1057PH/39 Allow G after quota size as well as K and M.
1058
0612b098
PH
1059PH/40 The value set for $authenticated_id in an authenticator may not contain
1060 binary zeroes or newlines because the value is written to log lines and
1061 to spool files. There was no check on this. Now the value is run through
1062 the string_printing() function so that such characters are converted to
1063 printable escape sequences.
1064
2e0c1448
PH
1065PH/41 $message_linecount is a new variable that contains the total number of
1066 lines in the message. Compare $body_linecount, which is the count for the
1067 body only.
1068
447d236c
PH
1069PH/42 Exim no longer gives details of delivery errors for specific addresses in
1070 bounce and delay warning messages, except in certain special cases, which
1071 are as follows:
1072
1073 (a) An SMTP error message from a remote host;
1074 (b) A message specified in a :fail: redirection;
1075 (c) A message specified in a "fail" command in a system filter;
1076 (d) A message specified in a FAIL return from the queryprogram router;
1077 (e) A message specified by the cannot_route_message router option.
1078
1079 In these cases only, Exim does include the error details in bounce and
1080 warning messages. There are also a few cases where bland messages such
1081 as "unrouteable address" or "local delivery error" are given.
1082
d20976dc
PH
1083PH/43 $value is now also set for the "else" part of a ${run expansion.
1084
f656d135
PH
1085PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
1086 being worked on, but at least Exim now implements the latest version to
1087 play with."
1088
2e2a30b4
PH
1089PH/45 In a pipe transport, although a timeout while waiting for the pipe
1090 process to complete was treated as a delivery failure, a timeout while
1091 writing the message to the pipe was logged, but erroneously treated as a
1092 successful delivery. Such timeouts include transport filter timeouts. For
1093 consistency with the overall process timeout, these timeouts are now
1094 treated as errors, giving rise to delivery failures by default. However,
1095 there is now a new Boolean option for the pipe transport called
1096 timeout_defer, which, if set TRUE, converts the failures into defers for
1097 both kinds of timeout. A transport filter timeout is now identified in
1098 the log output.
1099
9176e9f0
PH
1100PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
1101 systems where "make" and "gmake" are different, calling "gmake" at top
1102 level broke things. I've arranged for the value of $(MAKE) to be passed
1103 from the Makefile to this script so that it can call the same version of
1104 "make".
1105
7982096b 1106
bbe902f0
PH
1107A note about Exim versions 4.44 and 4.50
1108----------------------------------------
1109
1110Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
1111changes of various kinds. As a consequence, a big documentation update was
1112needed. This delayed the release for rather longer than seemed good, especially
1113in the light of a couple of (minor) security issues. Therefore, the changes
1114that fixed bugs were backported into 4.43, to create a 4.44 maintenance
1115release. So 4.44 and 4.50 are in effect two different branches that both start
1116from 4.43.
1117
1118I have left the 4.50 change log unchanged; it contains all the changes since
11194.43. The change log for 4.44 is below; many of its items are identical to
1120those for 4.50. This seems to be the most sensible way to preserve the
1121historical information.
1122
1123
f7b63901 1124Exim version 4.50
495ae4b0
PH
1125-----------------
1126
5fe762f6
PH
1127 1. Minor wording change to the doc/README.SIEVE file.
1128
139059f6 1129 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
5fe762f6 1130 computation of the current number of files was incorrect.
495ae4b0 1131
7086e875
PH
1132 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
1133 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
1134 place.
1135
35af9f61
PH
1136 4. Give more explanation in the error message when the command for a transport
1137 filter fails to execute.
1138
b668c215
PH
1139 5. There are several places where Exim runs a non-Exim command in a
1140 subprocess. The SIGUSR1 signal should be disabled for these processes. This
1141 was being done only for the command run by the queryprogram router. It is
1142 now done for all such subprocesses. The other cases are: ${run, transport
1143 filters, and the commands run by the lmtp and pipe transports.
1144
a494b1e1
PH
1145 6. Added CONFIGURE_GROUP build-time option.
1146
1147 7. Some older OS have a limit of 256 on the maximum number of file
1148 descriptors. Exim was using setrlimit() to set 1000 as a large value
1149 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
1150 systems. I've change it so that if it can't get 1000, it tries for 256.
35edf2ff 1151
c5fcb476
PH
1152 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
1153 was an oversight, and furthermore, ever since the addition of extra
1154 controls (e.g. 4.43/32), the checks on when to allow different forms of
1155 "control" were broken. There should now be diagnostics for all cases when a
1156 control that does not make sense is encountered.
1157
69358f02
PH
1158 9. Added the /retain_sender option to "control=submission".
1159
5be20824
PH
116010. $recipients is now available in the predata ACL (oversight).
1161
eb2c0248
PH
116211. Tidy the search cache before the fork to do a delivery from a message
1163 received from the command line. Otherwise the child will trigger a lookup
1164 failure and thereby defer the delivery if it tries to use (for example) a
1165 cached ldap connection that the parent has called unbind on.
1166
2a3eea10
PH
116712. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
1168 of $address_data from the recipient verification was clobbered by the
1169 sender verification.
1170
117113. The value of address_data from a sender verification is now available in
1172 $sender_address_data in subsequent conditions in the ACL statement.
1173
23c7ff99
PH
117414. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
1175
4deaf07d
PH
117615. Added a new option "connect=<time>" to callout options, to set a different
1177 connection timeout.
1178
926e1192
PH
117916. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
1180 was its contents. (It was OK if the option was not defined at all.)
1181
650edc6f
PH
118217. A "Completed" log line is now written for messages that are removed from
1183 the spool by the -Mrm option.
1184
2c7db3f5
PH
118518. New variables $sender_verify_failure and $recipient_verify_failure contain
1186 information about exactly what failed.
1187
3d235903
PH
118819. Added -dd to debug only the daemon process.
1189
7c7ad977
PH
119020. Incorporated Michael Haardt's patch to ldap.c for improving the way it
1191 handles timeouts, both on the server side and network timeouts. Renamed the
1192 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
1193
981756db
PH
119421. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
1195 It is now set to "smtps".
1196
d4eb88df
PH
119722. $host_address is now set to the target address during the checking of
1198 ignore_target_hosts.
1199
120023. When checking ignore_target_hosts for an ipliteral router, no host name was
1201 being passed; this would have caused $sender_host_name to have been used if
1202 matching the list had actually called for a host name (not very likely,
1203 since this list is usually IP addresses). A host name is now passed as
1204 "[x.x.x.x]".
1205
7d468ab8
PH
120624. Changed the calls that set up the SIGCHLD handler in the daemon to use the
1207 code that specifies a non-restarting handler (typically sigaction() in
1208 modern systems) in an attempt to fix a rare and obscure crash bug.
1209
121025. Narrowed the window for a race in the daemon that could cause it to ignore
1211 SIGCHLD signals. This is not a major problem, because they are used only to
1212 wake it up if nothing else does.
1213
62c0818f
PH
121426. A malformed maildirsize file could cause Exim to calculate negative values
1215 for the mailbox size or file count. Odd effects could occur as a result.
1216 The maildirsize information is now recalculated if the size or filecount
1217 end up negative.
1218
26034054
PH
121927. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
1220 support for a long time. Removed HAVE_SYS_VFS_H.
1221
af66f652
PH
122228. Installed the latest version of exipick from John Jetmore.
1223
90af77f4
PH
122429. In an address list, if the pattern was not a regular expression, an empty
1225 subject address (from a bounce message) matched only if the pattern was an
1226 empty string. Non-empty patterns were not even tested. This was the wrong
1227 because it is perfectly reasonable to use an empty address as part of a
1228 database query. An empty address is now tested by patterns that are
1229 lookups. However, all the other forms of pattern expect the subject to
1230 contain a local part and a domain, and therefore, for them, an empty
1231 address still always fails if the pattern is not itself empty.
1232
d8ef3577
PH
123330. Exim went into a mad DNS loop when attempting to do a callout where the
1234 host was specified on an smtp transport, and looking it up yielded more
1235 than one IP address.
1236
5cb8cbc6
PH
123731. Re-factored the code for checking spool and log partition space into a
1238 function that finds that data and another that does the check. The former
1239 is then used to implement four new variables: $spool_space, $log_space,
1240 $spool_inodes, and $log_inodes.
1241
14702f5b
PH
124232. The RFC2047 encoding function was originally intended for short strings
1243 such as real names; it was not keeping to the 75-character limit for
1244 encoded words that the RFC imposes. It now respects the limit, and
1245 generates multiple encoded words if necessary. To be on the safe side, I
1246 have increased the buffer size for the ${rfc2047: expansion operator from
1247 1024 to 2048 bytes.
1248
063b1e99
PH
124933. It is now permitted to omit both strings after an "if" condition; if the
1250 condition is true, the result is "true". As before, when the second string
1251 is omitted, a false condition yields an empty string. This makes it less
1252 cumbersome to write custom ACL and router conditions.
1253
652e1b65
PH
125434. Failure to deliver a bounce message always caused it to be frozen, even if
1255 there was an errors_to setting on the router. The errors_to setting is now
1256 respected.
1257
6f0c9a4f
PH
125835. If an IPv6 address is given for -bh or -bhc, it is now converted to the
1259 canonical form (fully expanded) before being placed in
1260 $sender_host_address.
1261
33397d19
PH
126236. The table in the code that translates DNS record types into text (T_A to
1263 "A" for instance) was missing entries for NS and CNAME. It is just possible
1264 that this could have caused confusion if both these types were looked up
1265 for the same domain, because the text type is used as part of Exim's
1266 per-process caching. But the chance of anyone hitting this buglet seems
1267 very small.
1268
7bb56e1f
PH
126937. The dnsdb lookup has been extended in a number of ways.
1270
1271 (1) There is a new type, "zns", which walks up the domain tree until it
1272 finds some nameserver records. It should be used with care.
1273
ea3bc19b
PH
1274 (2) There is a new type, "mxh", which is like "mx" except that it returns
1275 just the host names, not the priorities.
1276
1277 (3) It is now possible to give a list of domains (or IP addresses) to be
ff4dbb19
PH
1278 looked up. The behaviour when one of the lookups defers can be
1279 controlled by a keyword.
7bb56e1f 1280
ea3bc19b 1281 (4) It is now possible to specify the separator character for use when
7bb56e1f 1282 multiple records are returned.
33397d19 1283
0bcb2a0e
PH
128438. The dnslists ACL condition has been extended: it is now possible to supply
1285 a list of IP addresses and/or domains to be looked up in a particular DNS
1286 domain.
1287
2ac0e484
PH
128839. Added log_selector=+queue_time_overall.
1289
4e1fde53
PH
129040. When running the queue in the test harness, wait just a tad after forking a
1291 delivery process, to get repeatability of debugging output.
1292
de365ded
PH
129341. Include certificate and key file names in error message when GnuTLS fails
1294 to set them up, because the GnuTLS error message doesn't include the name
1295 of the failing file when there is a problem reading it.
1296
f05da2e8
PH
129742. Allow both -bf and -bF in the same test run.
1298
d6453af2
PH
129943. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
1300
f7b63901
PH
130144. The "Exiscan patch" is now merged into the mainline Exim source.
1302
130345. Sometimes the final signoff response after QUIT could fail to get
1304 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
1305 < 0 before doing a fflush(). This bug looks as though it goes back to the
1306 introduction of TLS in release 3.20, but "sometimes" must have been rare
1307 because the tests only now provoked it.
1308
a444213a
PH
130946. Reset the locale to "C" after calling embedded Perl, in case it was changed
1310 (this can affect the format of dates).
1311
0ec020ea
PH
131247. exim_tidydb, when checking for the continued existence of a message for
1313 which it has found a message-specific retry record, was not finding
1314 messages that were in split spool directories. Consequently, it was
1315 deleting retry records that should have stayed in existence.
1316
b1206957
PH
131748. Steve fixed some bugs in eximstats.
1318
131949. The SPA authentication driver was not abandoning authentication and moving
1320 on to the next authenticator when an expansion was forced to fail,
1321 contradicting the general specification for all authenticators. Instead it
1322 was generating a temporary error. It now behaves as specified.
1323
26dd5a95
PH
132450. The default ordering of permitted cipher suites for GnuTLS was pessimal
1325 (the order specifies the preference for clients). The order is now AES256,
1326 AES128, 3DES, ARCFOUR128.
1327
343b2385
PH
132851. Small patch to Sieve code - explicitly set From: when generating an
1329 autoreply.
1330
1c5466b9
PH
133152. Exim crashed if a remote delivery caused a very long error message to be
1332 recorded - for instance if somebody sent an entire SpamAssassin report back
1333 as a large number of 550 error lines. This bug was coincidentally fixed by
1334 increasing the size of one of Exim's internal buffers (big_buffer) that
1335 happened as part of the Exiscan merge. However, to be on the safe side, I
1336 have made the code more robust (and fixed the comments that describe what
1337 is going on).
1338
55ee9ee3
PH
133953. Now that there can be additional text after "Completed" in log lines (if
1340 the queue_time_overall log selector is set), a one-byte patch to exigrep
1341 was needed to allow it to recognize "Completed" as not the last thing in
1342 the line.
1343
d38f8232
PH
134454. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
1345 patch that reportedly fixes this has been added. I am not expert enough to
1346 create a test for it. This is what the patch creator wrote:
1347
1348 "I found a little strange behaviour of ldap code when working with
1349 Windows 2003 AD Domain, where users was placed in more than one
1350 Organization Units. When I tried to give exim partial DN, the exit code
1351 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
1352 But simultaneously result of request was absolutely normal ldap result,
1353 so I produce this patch..."
1354
3295e65b
PH
1355 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
1356 REFERENCE, so I have modified the code to exclude the patch when that macro
1357 is not defined.
1358
7102e136
PH
135955. Some experimental protocols are using DNS PTR records for new purposes. The
1360 keys for these records are domain names, not reversed IP addresses. The
b975ba52
PH
1361 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
1362 leaves it alone. Component reversal etc. now happens only for IP addresses.
ea3a6f44 1363 CAN-2005-0021
7102e136 1364
3ca0ba97
PH
136556. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
1366
c2bcbe20
PH
136757. Double the size of the debug message buffer (to 2048) so that more of very
1368 long debug lines gets shown.
1369
18ce445d
PH
137058. The exicyclog utility now does better if the number of log files to keep
1371 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
1372
1f5b4c3d
PH
137359. Two changes related to the smtp_active_hostname option:
1374
1375 (1) $smtp_active_hostname is now available as a variable.
1376 (2) The default for smtp_banner uses $smtp_active_hostname instead
1377 of $primary_hostname.
1378
b975ba52
PH
137960. The host_aton() function is supposed to be passed a string that is known
1380 to be a valid IP address. However, in the case of IPv6 addresses, it was
1381 not checking this. This is a hostage to fortune. Exim now panics and dies
1382 if the condition is not met. A case was found where this could be provoked
85b87bc2
PH
1383 from a dnsdb PTR lookup with an IPv6 address that had more than 8
1384 components; fortuitously, this particular loophole had already been fixed
1385 by change 4.50/55 above.
1386
1387 If there are any other similar loopholes, the new check in host_aton()
1388 itself should stop them being exploited. The report I received stated that
1389 data on the command line could provoke the exploit when Exim was running as
1390 exim, but did not say which command line option was involved. All I could
1391 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
1392 running as the user.
ea3a6f44 1393 CAN-2005-0021
85b87bc2
PH
1394
139561. There was a buffer overflow vulnerability in the SPA authentication code
1396 (which came originally from the Samba project). I have added a test to the
1397 spa_base64_to_bits() function which I hope fixes it.
ea3a6f44 1398 CAN-2005-0022
b975ba52 1399
17ffcae7
PH
140062. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
1401 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
1402
d95f9fdb
PH
140363. The daemon start-up calls getloadavg() while still root for those OS that
1404 need the first call to be done as root, but it missed one case: when
1405 deliver_queue_load_max is set with deliver_drop_privilege. This is
1406 necessary for the benefit of the queue runner, because there is no re-exec
1407 when deliver_drop_privilege is set.
1408
86b8287f
PH
140964. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
1410 This has been fixed.
1411
60dc5e56
PH
141265. Caching of lookup data for "hosts =" ACL conditions, when a named host list
1413 was in use, was not putting the data itself into the right store pool;
1414 consequently, it could be overwritten for a subsequent message in the same
1415 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
1416 the caching.)
1417
533244af
PH
141866. Added hosts_max_try_hardlimit to the smtp transport, default 50.
1419
a5a28604
PH
142067. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
1421 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
1422 the function were treating the return as a boolean value, which happened to
1423 work because 0=false and not-0=true, but is not correct code.
1424
7e634d24
PH
142568. The host_aton() function was not handling scoped IPv6 addresses (those
1426 with, for example, "%eth0" on the end) correctly.
1427
3e11c26b
PH
142869. Fixed some compiler warnings in acl.c for the bitmaps specified with
1429 negated items (that is, ~something) in unsigned ints. Some compilers
1430 apparently mutter when there is no cast.
1431
6729cf78
PH
143270. If an address verification called from an ACL failed, and did not produce a
1433 user-specific message (i.e. there was only a "system" message), nothing was
1434 put in $acl_verify_message. In this situation, it now puts the system
1435 message there.
1436
00f00ca5
PH
143771. Change 4.23/11 added synchronization checking at the start of an SMTP
1438 session; change 4.31/43 added the unwanted input to the log line - except
1439 that it did not do this in the start of session case. It now does.
1440
c9bdd01c
PH
144172. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
1442 This is wrong and can cause the other end to generate a synchronization
1443 error if it is another Exim or anything else that does the synchronization
1444 check. A QUIT command is no longer sent after a timeout.
1445
d43194df
PH
144673. $host_lookup_deferred has been added, to make it easier to detect DEFERs
1447 during host lookups.
1448
fe5b5d0b
PH
144974. The defer_ok option of callout verification was not working if it was used
1450 when verifying addresses in header lines, that is, for this case:
1451
1452 verify = header_sender/callout=defer_ok
1453
76a2d7ba
PH
145475. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
1455 those file descriptors could be used for SMTP connections. If anything
1456 wrote to stderr (the example that came up was "warn" in embedded Perl), it
1457 could be sent to the SMTP client, causing chaos. The daemon now opens
1458 stdin, stdout, and stderr to /dev/null when it puts itself into the
1459 background.
1460
146176. Arrange for output from Perl's "warn" command to be written to Exim's main
1462 log by default. The user can override this with suitable Perl magic.
1463
04f7d5b9
PH
146477. The use of log_message on a "discard" ACL verb, which is supposed to add to
1465 the log message when discard triggers, was not working for the DATA ACL or
1466 for the non-SMTP ACL.
1467
bc60667e
PH
146878. Error message wording change in sieve.c.
1469
bb6e88ff
PH
147079. If smtp_accept_max_per_host was set, the number of connections could be
1471 restricted to fewer than expected, because the daemon was trying to set up
1472 a new connection before checking whether the processes handling previous
1473 connections had finished. The check for completed processes is now done
1474 earlier. On busy systems, this bug wouldn't be noticed because something
1475 else would have woken the daemon, and it would have reaped the completed
1476 process earlier.
1477
1e70f85b
PH
147880. If a message was submitted locally by a user whose login name contained one
1479 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
1480 It caused a spool format error. I have fixed the spool reading code. A
1481 related problem was that the "from" clause in the Received: line became
1482 illegal because of the space(s). It is now covered by ${quote_local_part.
1483
148481. Included the latest eximstats from Steve (adds average sizes to HTML Top
1485 tables).
1486
4e01f9d6
PH
148782. Updated OS/Makefile-AIX as per message from Mike Meredith.
1488
1ee1cef2
PH
148983. Patch from Sieve maintainer to fix unterminated string problem in
1490 "vacation" handling.
1491
6e2b4ccc
PH
149284. Some minor changes to the Linux configuration files to help with other
1493 OS variants using glibc.
1494
8e669ac1
PH
149585. One more patch for Sieve to update vacation handling to latest spec.
1496
495ae4b0 1497
bbe902f0
PH
1498----------------------------------------------------
1499See the note above about the 4.44 and 4.50 releases.
1500----------------------------------------------------
1501
1502
1503Exim version 4.44
1504-----------------
1505
1506 1. Change 4.43/35 introduced a bug that caused file counts to be
1507 incorrectly computed when quota_filecount was set in an appendfile
1508 transport
1509
1510 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
1511 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
1512 place.
1513
1514 3. Give more explanation in the error message when the command for a transport
1515 filter fails to execute.
1516
1517 4. There are several places where Exim runs a non-Exim command in a
1518 subprocess. The SIGUSR1 signal should be disabled for these processes. This
1519 was being done only for the command run by the queryprogram router. It is
1520 now done for all such subprocesses. The other cases are: ${run, transport
1521 filters, and the commands run by the lmtp and pipe transports.
1522
1523 5. Some older OS have a limit of 256 on the maximum number of file
1524 descriptors. Exim was using setrlimit() to set 1000 as a large value
1525 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
1526 systems. I've change it so that if it can't get 1000, it tries for 256.
1527
1528 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
1529 was an oversight, and furthermore, ever since the addition of extra
1530 controls (e.g. 4.43/32), the checks on when to allow different forms of
1531 "control" were broken. There should now be diagnostics for all cases when a
1532 control that does not make sense is encountered.
1533
1534 7. $recipients is now available in the predata ACL (oversight).
1535
1536 8. Tidy the search cache before the fork to do a delivery from a message
1537 received from the command line. Otherwise the child will trigger a lookup
1538 failure and thereby defer the delivery if it tries to use (for example) a
1539 cached ldap connection that the parent has called unbind on.
1540
1541 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
1542 of $address_data from the recipient verification was clobbered by the
1543 sender verification.
1544
154510. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
1546 was its contents. (It was OK if the option was not defined at all.)
1547
154811. A "Completed" log line is now written for messages that are removed from
1549 the spool by the -Mrm option.
1550
155112. $host_address is now set to the target address during the checking of
1552 ignore_target_hosts.
1553
155413. When checking ignore_target_hosts for an ipliteral router, no host name was
1555 being passed; this would have caused $sender_host_name to have been used if
1556 matching the list had actually called for a host name (not very likely,
1557 since this list is usually IP addresses). A host name is now passed as
1558 "[x.x.x.x]".
1559
156014. Changed the calls that set up the SIGCHLD handler in the daemon to use the
1561 code that specifies a non-restarting handler (typically sigaction() in
1562 modern systems) in an attempt to fix a rare and obscure crash bug.
1563
156415. Narrowed the window for a race in the daemon that could cause it to ignore
1565 SIGCHLD signals. This is not a major problem, because they are used only to
1566 wake it up if nothing else does.
1567
156816. A malformed maildirsize file could cause Exim to calculate negative values
1569 for the mailbox size or file count. Odd effects could occur as a result.
1570 The maildirsize information is now recalculated if the size or filecount
1571 end up negative.
1572
157317. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
1574 support for a long time. Removed HAVE_SYS_VFS_H.
1575
ea3a6f44 157618. Updated exipick to current release from John Jetmore.
bbe902f0
PH
1577
157819. Allow an empty sender to be matched against a lookup in an address list.
1579 Previously the only cases considered were a regular expression, or an
1580 empty pattern.
1581
158220. Exim went into a mad DNS lookup loop when doing a callout where the
1583 host was specified on the transport, if the DNS lookup yielded more than
1584 one IP address.
1585
ea3a6f44
NM
158621. The RFC2047 encoding function was originally intended for short strings
1587 such as real names; it was not keeping to the 75-character limit for
1588 encoded words that the RFC imposes. It now respects the limit, and
1589 generates multiple encoded words if necessary. To be on the safe side, I
1590 have increased the buffer size for the ${rfc2047: expansion operator from
1591 1024 to 2048 bytes.
bbe902f0 1592
ea3a6f44
NM
159322. Failure to deliver a bounce message always caused it to be frozen, even if
1594 there was an errors_to setting on the router. The errors_to setting is now
1595 respected.
bbe902f0
PH
1596
159723. If an IPv6 address is given for -bh or -bhc, it is now converted to the
1598 canonical form (fully expanded) before being placed in
1599 $sender_host_address.
1600
160124. Updated eximstats to version 1.33
1602
ea3a6f44
NM
160325. Include certificate and key file names in error message when GnuTLS fails
1604 to set them up, because the GnuTLS error message doesn't include the name
1605 of the failing file when there is a problem reading it.
bbe902f0
PH
1606
160726. Expand error message when OpenSSL has problems setting up cert/key files.
ea3a6f44 1608 As per change 25.
bbe902f0 1609
ea3a6f44
NM
161027. Reset the locale to "C" after calling embedded Perl, in case it was changed
1611 (this can affect the format of dates).
bbe902f0 1612
ea3a6f44
NM
161328. exim_tidydb, when checking for the continued existence of a message for
1614 which it has found a message-specific retry record, was not finding
1615 messages that were in split spool directories. Consequently, it was
1616 deleting retry records that should have stayed in existence.
bbe902f0
PH
1617
161829. eximstats updated to version 1.35
1619 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
1620 1.35 - bugfix such that pie charts by volume are generated correctly
1621
ea3a6f44
NM
162230. The SPA authentication driver was not abandoning authentication and moving
1623 on to the next authenticator when an expansion was forced to fail,
1624 contradicting the general specification for all authenticators. Instead it
1625 was generating a temporary error. It now behaves as specified.
bbe902f0 1626
ea3a6f44
NM
162731. The default ordering of permitted cipher suites for GnuTLS was pessimal
1628 (the order specifies the preference for clients). The order is now AES256,
1629 AES128, 3DES, ARCFOUR128.
bbe902f0 1630
ea3a6f44
NM
163131. Small patch to Sieve code - explicitly set From: when generating an
1632 autoreply.
bbe902f0 1633
ea3a6f44
NM
163432. Exim crashed if a remote delivery caused a very long error message to be
1635 recorded - for instance if somebody sent an entire SpamAssassin report back
1636 as a large number of 550 error lines. This bug was coincidentally fixed by
1637 increasing the size of one of Exim's internal buffers (big_buffer) that
1638 happened as part of the Exiscan merge. However, to be on the safe side, I
1639 have made the code more robust (and fixed the comments that describe what
1640 is going on).
bbe902f0 1641
ea3a6f44
NM
164233. Some experimental protocols are using DNS PTR records for new purposes. The
1643 keys for these records are domain names, not reversed IP addresses. The
1644 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
1645 leaves it alone. Component reversal etc. now happens only for IP addresses.
bbe902f0
PH
1646 CAN-2005-0021
1647
ea3a6f44
NM
164834. The host_aton() function is supposed to be passed a string that is known
1649 to be a valid IP address. However, in the case of IPv6 addresses, it was
1650 not checking this. This is a hostage to fortune. Exim now panics and dies
1651 if the condition is not met. A case was found where this could be provoked
1652 from a dnsdb PTR lookup with an IPv6 address that had more than 8
1653 components; fortuitously, this particular loophole had already been fixed
1654 by change 4.50/55 or 4.44/33 above.
1655
1656 If there are any other similar loopholes, the new check in host_aton()
1657 itself should stop them being exploited. The report I received stated that
1658 data on the command line could provoke the exploit when Exim was running as
1659 exim, but did not say which command line option was involved. All I could
1660 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
1661 running as the user.
bbe902f0
PH
1662 CAN-2005-0021
1663
ea3a6f44
NM
166435. There was a buffer overflow vulnerability in the SPA authentication code
1665 (which came originally from the Samba project). I have added a test to the
1666 spa_base64_to_bits() function which I hope fixes it.
bbe902f0
PH
1667 CAN-2005-0022
1668
ea3a6f44
NM
166936. The daemon start-up calls getloadavg() while still root for those OS that
1670 need the first call to be done as root, but it missed one case: when
1671 deliver_queue_load_max is set with deliver_drop_privilege. This is
1672 necessary for the benefit of the queue runner, because there is no re-exec
1673 when deliver_drop_privilege is set.
bbe902f0 1674
ea3a6f44
NM
167537. Caching of lookup data for "hosts =" ACL conditions, when a named host list
1676 was in use, was not putting the data itself into the right store pool;
1677 consequently, it could be overwritten for a subsequent message in the same
1678 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
1679 the caching.)
bbe902f0 1680
ea3a6f44
NM
168138. Sometimes the final signoff response after QUIT could fail to get
1682 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
1683 < 0 before doing a fflush(). This bug looks as though it goes back to the
1684 introduction of TLS in release 3.20, but "sometimes" must have been rare
1685 because the tests only now provoked it.
bbe902f0
PH
1686
1687
495ae4b0
PH
1688Exim version 4.43
1689-----------------
1690
1691 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
1692 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
1693 Now it returns an integer. A number of calls were still expecting a T/F
1694 return. Fortuitously, in all cases, the tests worked in OK situations,
1695 which is the norm. However, things would have gone wrong on any write
1696 failures on the smtp file descriptor. This function is used when sending
1697 messages over SMTP and also when doing verify callouts.
1698
1699 2. When Exim is called to do synchronous delivery of a locally submitted
1700 message (the -odf or -odi options), it no longer closes stderr before doing
1701 the delivery.
1702
1703 3. Implemented the mua_wrapper option.
1704
1705 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
1706
1707 5. Implemented the functions header_remove(), header_testname(),
1708 header_add_at_position(), and receive_remove_recipient(), and exported them
1709 to local_scan().
1710
1711 6. If an ACL "warn" statement specified the addition of headers, Exim already
1712 inserted X-ACL-Warn: at the start if there was no header name. However, it
1713 was not making this test for the second and subsequent header lines if
1714 there were newlines in the string. This meant that an invalid header could
1715 be inserted if Exim was badly configured.
1716
1717 7. Allow an ACL "warn" statement to add header lines at the start or after all
1718 the Received: headers, as well as at the end.
1719
1720 8. Added the rcpt_4xx retry error code.
1721
1722 9. Added postmaster_mailfrom=xxx to callout verification option.
1723
172410. Added mailfrom=xxxx to the callout verification option, for verify=
1725 header_sender only.
1726
172711. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
1728 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
1729
173012. Inserted some casts to stop certain compilers warning when using pointer
1731 differences as field lengths or precisions in printf-type calls (mostly
1732 affecting debugging statements).
1733
173413. Added optional readline() support for -be (dynamically loaded).
1735
173614. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
1737 same clock tick as a message's arrival, so that its received time was the
1738 same as the "first fail" time on the retry record, and that message
1739 remained on the queue past the ultimate address timeout, every queue runner
1740 would try a delivery (because it was past the ultimate address timeout) but
1741 after another failure, the ultimate address timeout, which should have then
1742 bounced the address, did not kick in. This was a "< instead of <=" error;
1743 in most cases the first failure would have been in the next clock tick
1744 after the received time, and all would be well.
1745
174615. The special items beginning with @ in domain lists (e.g. @mx_any) were not
1747 being recognized when the domain list was tested by the match_domain
1748 condition in an expansion string.
1749
175016. Added the ${str2b64: operator.
1751
175217. Exim was always calling setrlimit() to set a large limit for the number of
1753 processes, without checking whether the existing limit was already
1754 adequate. (It did check for the limit on file descriptors.) Furthermore,
1755 errors from getrlimit() and setrlimit() were being ignored. Now they are
1756 logged to the main and panic logs, but Exim does carry on, to try to do its
1757 job under whatever limits there are.
1758
175918. Imported PCRE 5.0.
1760
176119. Trivial typo in log message " temporarily refused connection" (the leading
1762 space).
1763
176420. If the log selector return_path_on_delivery was set and an address was
1765 redirected to /dev/null, the delivery process crashed because it assumed
1766 that a return path would always be set for a "successful" delivery. In this
1767 case, the whole delivery is bypassed as an optimization, and therefore no
1768 return path is set.
1769
177021. Internal re-arrangement: the function for sending a challenge and reading
1771 a response while authentication was assuming a zero-terminated challenge
1772 string. It's now changed to take a pointer and a length, to allow for
1773 binary data in such strings.
1774
177522. Added the cyrus_sasl authenticator (code supplied by MBM).
1776
177723. Exim was not respecting finduser_retries when seeking the login of the
1778 uid under which it was called; it was always trying 10 times. (The default
1779 setting of finduser_retries is zero.) Also, it was sleeping after the final
1780 failure, which is pointless.
1781
178224. Implemented tls_on_connect_ports.
1783
178425. Implemented acl_smtp_predata.
1785
178626. If the domain in control=submission is set empty, Exim assumes that the
1787 authenticated id is a complete email address when it generates From: or
1788 Sender: header lines.
1789
179027. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
1791 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
1792 chown and chgrp in /bin and hostname in /usr/bin.
1793
179428. Exim was keeping the "process log" file open after each use, just as it
1795 does for the main log. This opens the possibility of it remaining open for
1796 long periods when the USR1 signal hits a daemon. Occasional processlog
1797 errors were reported, that could have been caused by this. Anyway, it seems
1798 much more sensible not to leave this file open at all, so that is what now
1799 happens.
1800
180129. The long-running daemon process does not normally write to the log once it
1802 has entered its main loop, and it closes the log before doing so. This is
1803 so that log files can straightforwardly be renamed and moved. However,
1804 there are a couple of unusual error situations where the daemon does write
1805 log entries, and I had neglected to close the log afterwards.
1806
180730. The text of an SMTP error response that was received during a remote
1808 delivery was being truncated at 512 bytes. This is too short for some of
1809 the long messages that one sometimes sees. I've increased the limit to
1810 1024.
1811
181231. It is now possible to make retry rules that apply only when a message has a
1813 specific sender, in particular, an empty sender.
1814
181532. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
1816 it possible to be selective about when SMTP synchronization is enforced.
1817
181833. Added "control = caseful_local_part" and "control = "caselower_local_part".
1819
182032. Implemented hosts_connection_nolog.
1821
182233. Added an ACL for QUIT.
1823
182434. Setting "delay_warning=" to disable warnings was not working; it gave a
1825 syntax error.
1826
182735. Added mailbox_size and mailbox_filecount to appendfile.
1828
182936. Added control = no_multiline_responses to ACLs.
1830
183137. There was a bug in the logic of the code that waits for the clock to tick
1832 in the case where the clock went backwards by a substantial amount such
1833 that the microsecond fraction of "now" was more than the microsecond
1834 fraction of "then" (but the whole seconds number was less).
1835
183638. Added support for the libradius Radius client library this is found on
1837 FreeBSD (previously only the radiusclient library was supported).
1838
1839
1840Exim version 4.42
1841-----------------
1842
1843 1. When certain lookups returned multiple values in the form name=value, the
1844 quoting of the values was not always being done properly. Specifically:
1845 (a) If the value started with a double quote, but contained no whitespace,
1846 it was not quoted.
1847 (b) If the value contained whitespace other than a space character (i.e.
1848 tabs or newlines or carriage returns) it was not quoted.
1849 This fix has been applied to the mysql and pgsql lookups by writing a
1850 separate quoting function and calling it from the lookup code. The fix
1851 should probably also be applied to nisplus, ibase and oracle lookups, but
1852 since I cannot test any of those, I have not disturbed their existing code.
1853
1854 2. A hit in the callout cache for a specific address caused a log line with no
1855 reason for rejecting RCPT. Now it says "Previous (cached) callout
1856 verification failure".
1857
1858 3. There was an off-by-one bug in the queryprogram router. An over-long
1859 return line was truncated at 256 instead of 255 characters, thereby
1860 overflowing its buffer with the terminating zero. As well as fixing this, I
1861 have increased the buffer size to 1024 (and made a note to document this).
1862
1863 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
1864 when Exim is waiting for an SMTP response from a remote server, Exim
1865 restarts its select() call on the socket, thereby resetting its timeout.
1866 This is not a problem when such interrupts are rare. Somebody set up a cron
1867 job to run exiwhat every 2 minutes, which is less than the normal select()
1868 timeout (5 or 10 minutes). This meant that the select() timeout never
1869 kicked in because it was always reset. I have fixed this by comparing the
1870 time when an interrupt arrives with the time at the start of the first call
1871 to select(). If more time than the timeout has elapsed, the interrupt is
1872 treated as a timeout.
1873
1874 5. Some internal re-factoring in preparation for the addition of Sieve
1875 extensions (by MH). In particular, the "personal" test is moved to a
1876 separate function, and given an option for scanning Cc: and Bcc: (which is
1877 not set for Exim filters).
1878
1879 6. When Exim created an email address using the login of the caller as the
1880 local part (e.g. when creating a From: or Sender: header line), it was not
1881 quoting the local part when it contained special characters such as @.
1882
1883 7. Installed new OpenBSD configuration files.
1884
1885 8. Reworded some messages for syntax errors in "and" and "or" conditions to
1886 try to make them clearer.
1887
1888 9. Callout options, other than the timeout value, were being ignored when
1889 verifying sender addresses in header lines. For example, when using
1890
1891 verify = header_sender/callout=no_cache
1892
1893 the cache was (incorrectly) being used.
1894
189510. Added a missing instance of ${EXE} to the exim_install script; this affects
1896 only the Cygwin environment.
1897
189811. When return_path_on_delivery was set as a log selector, if different remote
1899 addresses in the same message used different return paths and parallel
1900 remote delivery occurred, the wrong values would sometimes be logged.
1901 (Whenever a remote delivery process finished, the return path value from
1902 the most recently started remote delivery process was logged.)
1903
190412. RFC 3848 specifies standard names for the "with" phrase in Received: header
1905 lines when AUTH and/or TLS are in use. This is the "received protocol"
1906 field. Exim used to use "asmtp" for authenticated SMTP, without any
1907 indication (in the protocol name) for TLS use. Now it follows the RFC and
1908 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
1909 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
1910 names appear in log lines as well as in Received: header lines.
1911
191213. Installed MH's patches for Sieve to add the "copy" and "vacation"
1913 extensions, and comparison tests, and to fix some bugs.
1914
191514. Changes to the "personal" filter test:
1916
1917 (1) The test was buggy in that it was just doing the equivalent of
1918 "contains" tests on header lines. For example, if a user's address was
1919 anne@some.where, the "personal" test would incorrectly be true for
1920
1921 To: susanne@some.where
1922
1923 This test is now done by extracting each address from the header in turn,
1924 and checking the entire address. Other tests that are part of "personal"
1925 are now done using regular expressions (for example, to check local parts
1926 of addresses in From: header lines).
1927
1928 (2) The list of non-personal local parts in From: addresses has been
1929 extended to include "listserv", "majordomo", "*-request", and "owner-*",
1930 taken from the Sieve specification recommendations.
1931
1932 (3) If the message contains any header line starting with "List-" it is
1933 treated as non-personal.
1934
1935 (4) The test for "circular" in the Subject: header line has been removed
1936 because it now seems ill-conceived.
1937
193815. Minor typos in src/EDITME comments corrected.
1939
194016. Installed latest exipick from John Jetmore.
1941
194217. If headers_add on a router specified a text string that was too long for
1943 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
1944 of string_sprintf() is now avoided.
1945
194618. $message_body_size was not set (it was always zero) when running the DATA
1947 ACL and the local_scan() function.
1948
194919. For the "mail" command in an Exim filter, no default was being set for
1950 the once_repeat time, causing a random time value to be used if "once" was
1951 specified. (If the value happened to be <= 0, no repeat happened.) The
1952 default is now 0s, meaning "never repeat". The "vacation" command was OK
1953 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
1954 (I found it when inspecting the code).
1955
195620. There is now an overall timeout for performing a callout verification. It
1957 defaults to 4 times the callout timeout, which applies to individual SMTP
1958 commands during the callout. The overall timeout applies when there is more
1959 than one host that can be tried. The timeout is checked before trying the
1960 next host. This prevents very long delays if there are a large number of
1961 hosts and all are timing out (e.g. when the network connections are timing
1962 out). The value of the overall timeout can be changed by specifying an
1963 additional sub-option for "callout", called "maxwait". For example:
1964
1965 verify = sender/callout=5s,maxwait=20s
1966
196721. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
1968 to the end before writing, but this should make it even safer).
1969
197022. Exim was forgetting that it had advertised PIPELINING for the second and
1971 subsequent messages on an SMTP connection. It was also not resetting its
1972 memory on STARTTLS and an internal HELO.
1973
197423. When Exim logs an SMTP synchronization error within a session, it now
1975 records whether PIPELINING has been advertised or not.
1976
197724. Added 3 instances of "(long int)" casts to time_t variables that were being
1978 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
1979 rather than long int.
1980
198125. Installed the latest Cygwin configuration files from the Cygwin maintainer.
1982
198326. Added the never_mail option to autoreply.
1984
1985
1986Exim version 4.41
1987-----------------
1988
1989 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
1990 crash if the getsockname() call failed; this can happen if a connection is
1991 closed very soon after it is established. The problem was simply in the
1992 order in which certain operations were done, causing Exim to try to write
1993 to the SMTP stream before it had set up the file descriptor. The bug has
1994 been fixed by making things happen in the correct order.
1995
1996
1997Exim version 4.40
1998-----------------
1999
2000 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
2001 before the connection was closed, thus losing the rejection response.
2002
2003 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
2004 some early Solaris releases, but causes trouble in current releases where
2005 socklen_t is defined.
2006
2007 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
2008 always exist.
2009
2010 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
2011 configured.
2012
2013 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
2014 that releasing the top of it at the end releases what was used for sub-
2015 expansions (unless the block got too big). However, discard this block if
2016 the first thing is a variable or header, so that we can use its block when
2017 it is dynamic (useful for very large $message_headers, for example).
2018
2019 6. Lookups now cache *every* query, not just the most recent. A new, separate
2020 store pool is used for this. It can be recovered when all lookup caches are
2021 flushed. Lookups now release memory at the end of their result strings.
2022 This has involved some general refactoring of the lookup sources.
2023
2024 7. Some code has been added to the store_xxx() functions to reduce the amount
2025 of flapping under certain conditions.
2026
2027 8. log_incoming_interface used to affect only the <= reception log lines. Now
2028 it causes the local interface and port to be added to several more SMTP log
2029 lines, for example "SMTP connection from", and rejection lines.
2030
2031 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
2032
203310. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
2034
203511. If $host_data was set by virtue of a hosts lookup in an ACL, its value
2036 could be overwritten at the end of the current message (or the start of a
2037 new message if it was set in a HELO ACL). The value is now preserved for
2038 the duration of the SMTP connection.
2039
204012. If a transport had a headers_rewrite setting, and a matching header line
2041 contained an unqualified address, that address was qualified, even if it
2042 did not match any rewriting rules. The underlying bug was that the values
2043 of the flags that permit the existence of unqualified sender and recipient
2044 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
2045 non-local messages, and by -bnq for local messages) were not being
2046 preserved with the message after it was received.
2047
204813. When Exim was logging an SMTP synchronization error, it could sometimes log
2049 "next input=" as part of the text comprising the host identity instead of
2050 the correct text. The code was using the same buffer for two different
2051 strings. However, depending on which order the printing function evaluated
2052 its arguments, the bug did not always show up. Under Linux, for example, my
2053 test suite worked just fine.
2054
205514. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
2056 doesn't work with some older versions of Perl. It has been changed to "my",
2057 which in any case is probably the better facility to use.
2058
205915. A really picky compiler found some instances of statements for creating
2060 error messages that either had too many or two few arguments for the format
2061 string.
2062
206316. The size of the buffer for calls to the DNS resolver has been increased
2064 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
2065 for addresses that have a lot of PTR records. This alleviates a problem; it
2066 does not fully solve it.
2067
206817. A dnsdb lookup for PTR records that receives more data than will fit in the
2069 buffer now truncates the list and logs the incident, which is the same
2070 action as happens when Exim is looking up a host name and its aliases.
2071 Previously in this situation something unpredictable would happen;
2072 sometimes it was "internal error: store_reset failed".
2073
207418. If a server dropped the connection unexpectedly when an Exim client was
2075 using GnuTLS and trying to read a response, the client delivery process
2076 crashed while trying to generate an error log message.
2077
207819. If a "warn" verb in an ACL added multiple headers to a message in a single
2079 string, for example:
2080
2081 warn message = H1: something\nH2: something
2082
2083 the text was added as a single header line from Exim's point of view
2084 though it ended up OK in the delivered message. However, searching for the
2085 second and subsequent header lines using $h_h2: did not work. This has been
2086 fixed. Similarly, if a system filter added multiple headers in this way,
2087 the routers could not see them.
2088
208920. Expanded the error message when iplsearch is called with an invalid key to
2090 suggest using net-iplsearch in a host list.
2091
209221. When running tests using -bh, any delays imposed by "delay" modifiers in
2093 ACLs are no longer actually imposed (and a message to that effect is
2094 output).
2095
209622. If a "gecos" field in a passwd entry contained escaped characters, in
2097 particular, if it contained a \" sequence, Exim got it wrong when building
2098 a From: or a Sender: header from that name. A second bug also caused
2099 incorrect handling when an unquoted " was present following a character
2100 that needed quoting.
2101
210223. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
2103 was not being matched caselessly.
2104
210524. Arranged for all hyphens in the exim.8 source to be escaped with
2106 backslashes.
2107
210825. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
2109 properly. Recipient callout cache records were still being keyed to include
2110 the sender, even when use_sender was set false. This led to far more
2111 callouts that were necessary. The sender is no longer included in the key
2112 when use_sender is false.
2113
211426. Added "control = submission" modifier to ACLs.
2115
211627. Added the ${base62d: operator to decode base 62 numbers.
2117
211828. dnsdb lookups can now access SRV records.
2119
212029. CONFIGURE_OWNER can be set at build time to define an alternative owner for
2121 the configuration file.
2122
212330. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
2124 (-v) mode. This makes the output for a verbose queue run more intelligible.
2125
212631. Added a use_postmaster feature to recipient callouts.
2127
212832. Added the $body_zerocount variable, containing the number of binary zero
2129 bytes in the message body.
2130
213133. The time of last modification of the "new" subdirectory is now used as the
2132 "mailbox time last read" when there is a quota error for a maildir
2133 delivery.
2134
213534. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
2136
213735. Added +ignore_unknown as a special item in host lists.
2138
213936. Code for decoding IPv6 addresses in host lists is now included, even if
2140 IPv6 support is not being compiled. This fixes a bug in which an IPv6
2141 address was recognized as an IP address, but was then not correctly decoded
2142 into binary, causing unexpected and incorrect effects when compared with
2143 another IP address.
2144
2145
2146Exim version 4.34
2147-----------------
2148
2149 1. Very minor rewording of debugging text in manualroute to say "list of
2150 hosts" instead of "hostlist".
2151
2152 2. If verify=header_syntax was set, and a header line with an unqualified
2153 address (no domain) and a large number of spaces between the end of the
2154 name and the colon was received, the reception process suffered a buffer
2155 overflow, and (when I tested it) crashed. This was caused by some obsolete
2156 code that should have been removed. The fix is to remove it!
2157
2158 3. When running in the test harness, delay a bit after writing a bounce
2159 message to get a bit more predictability in the log output.
2160
2161 4. Added a call to search_tidyup() just before forking a reception process. In
2162 theory, someone could use a lookup in the expansion of smtp_accept_max_
2163 per_host which, without the tidyup, could leave open a database connection.
2164
2165 5. Added the variables $recipient_data and $sender_data which get set from a
2166 lookup success in an ACL "recipients" or "senders" condition, or a router
2167 "senders" option, similar to $domain_data and $local_part_data.
2168
2169 6. Moved the writing of debug_print from before to after the "senders" test
2170 for routers.
2171
2172 7. Change 4.31/66 (moving the time when the Received: is generated) caused
2173 problems for message scanning, either using a data ACL, or using
2174 local_scan() because the Received: header was not generated till after they
2175 were called (in order to set the time as the time of reception completion).
2176 I have revised the way this works. The header is now generated after the
2177 body is received, but before the ACL or local_scan() are called. After they
2178 are run, the timestamp in the header is updated.
2179
2180
2181Exim version 4.33
2182-----------------
2183
2184 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
2185 before starting a queue runner without re-exec. This happened only when
2186 deliver_drop_privilege was set or when the Exim user was set to root. The
2187 effect of the bug was that timeouts during subsequent deliveries caused
2188 crashes instead of being properly handled. The handler is now left at its
2189 default (and expected) setting.
2190
2191 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
2192 message, again when deliver_drop_privilege is set or Exim is run as root.
2193 The bug described in (1) was not present in this case, but the tidying up
2194 of the other signals was missing. I have made the two cases consistent.
2195
2196 3. The ignore_target_hosts setting on a manualroute router was being ignored
2197 for hosts that were looked up using the /MX notation.
2198
2199 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
2200 in domain lists.
2201
2202 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
2203 operated on the sender address. After changing the $sender_address to <>
2204 for the sender address verify, Exim was re-instated it as the original
2205 (before rewriting) address, but remembering that it had rewritten it, so it
2206 wasn't rewriting it again. This bug also had the effect of breaking the
2207 sender address verification caching when the sender address was rewritten.
2208
2209 6. The ignore_target_hosts option was being ignored by the ipliteral router.
2210 This has been changed so that if the ip literal address matches
2211 ignore_target_hosts, the router declines.
2212
2213 7. Added expansion conditions match_domain, match_address, and match_local_
2214 part (NOT match_host).
2215
2216 8. The placeholder for the Received: header didn't have a length field set.
2217
2218 9. Added code to Exim itself and to exim_lock to test for a specific race
2219 condition that could lead to file corruption when using MBX delivery. The
2220 issue is with the lockfile that is created in /tmp. If this file is removed
2221 after a process has opened it but before that process has acquired a lock,
2222 there is the potential for a second process to recreate the file and also
2223 acquire a lock. This could lead to two Exim processes writing to the file
2224 at the same time. The added code performs the same test as UW imapd; it
2225 checks after acquiring the lock that its file descriptor still refers to
2226 the same named file.
2227
222810. The buffer for building added header lines was of fixed size, 8192 bytes.
2229 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
2230 when Exim is built.
2231
223211. Added the smtp_active_hostname option. If used, this will typically be made
2233 to depend on the incoming interface address. Because $interface_address is
2234 not set up until the daemon has forked a reception process, error responses
2235 that can happen earlier (such as "too many connections") no longer contain
2236 a host name.
2237
223812. If an expansion in a condition on a "warn" statement fails because a lookup
2239 defers, the "warn" statement is abandoned, and the next ACL statement is
2240 processed. Previously this caused the whole ACL to be aborted.
2241
224213. Added the iplsearch lookup type.
2243
224414. Added ident_timeout as a log selector.
2245
224615. Added tls_certificate_verified as a log selector.
2247
224816. Added a global option tls_require_ciphers (compare the smtp transport
2249 option of the same name). This controls incoming TLS connections.
2250
225117. I finally figured out how to make tls_require_ciphers do a similar thing
2252 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
2253 before starting the TLS session.
2254
225518. Tabs are now shown as \t in -bP output.
2256
225719. If the log selector return_path_on_delivery was set, Exim crashed when
2258 bouncing a message because it had too many Received: header lines.
2259
226020. If two routers both had headers_remove settings, and the first one included
2261 a superfluous trailing colon, the final name in the first list and the
2262 first name in the second list were incorrectly joined into one item (with a
2263 colon in the middle).
2264
2265
2266Exim version 4.32
2267-----------------
2268
2269 1. Added -C and -D options to the exinext utility, mainly to make it easier
2270 to include in the automated testing, but these could be helpful when
2271 multiple configurations are in use.
2272
2273 2. The exinext utility was not formatting the output nicely when there was
2274 an alternate port involved in the retry record key, nor when there was a
2275 message id as well (for retries that were specific to a specific message
2276 and a specific host). It was also confused by IPv6 addresses, because of
2277 the additional colons they contain. I have fixed the IPv4 problem, and
2278 patched it up to do a reasonable job for IPv6.
2279
2280 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
2281 delivery, the log line now contains "pipelined" if PIPELINING was used.
2282
2283 4. An SMTP transport process used to panic and die if the bind() call to set
2284 an explicit outgoing interface failed. This has been changed; it is now
2285 treated in the same way as a connect() failure.
2286
2287 5. A reference to $sender_host_name in the part of a conditional expansion
2288 that was being skipped was still causing a DNS lookup. This no longer
2289 occurs.
2290
2291 6. The def: expansion condition was not recognizing references to header lines
2292 that used bh_ and bheader_.
2293
2294 7. Added the _cache feature to named lists.
2295
2296 8. The code for checking quota_filecount in the appendfile transport was
2297 allowing one more file than it should have been.
2298
2299 9. For compatibility with Sendmail, the command line option
2300
2301 -prval:sval
2302
2303 is equivalent to
2304
2305 -oMr rval -oMs sval
2306
2307 and sets the incoming protocol and host name (for trusted callers). The
2308 host name and its colon can be omitted when only the protocol is to be set.
2309 Note the Exim already has two private options, -pd and -ps, that refer to
2310 embedded Perl. It is therefore impossible to set a protocol value of "d" or
2311 "s", but I don't think that's a major issue.
2312
231310. A number of refactoring changes to the code, none of which should affect
2314 Exim's behaviour:
2315
2316 (a) The number of logging options was getting close to filling up the
2317 32-bit word that was used as a bit map. I have split them into two classes:
2318 those that are passed in the argument to log_write(), and those that are
2319 only ever tested independently outside of that function. These are now in
2320 separate 32-bit words, so there is plenty of room for expansion again.
2321 There is no change in the user interface or the logging behaviour.
2322
2323 (b) When building, for example, log lines, the code previously used a
2324 macro that called string_cat() twice, in order to add two strings. This is
2325 not really sufficiently general. Furthermore, there was one instance where
2326 it was actually wrong because one of the argument was used twice, and in
2327 one call a function was used. (As it happened, calling the function twice
2328 did not affect the overall behaviour.) The macro has been replaced by a
2329 function that can join an arbitrary number of extra strings onto a growing
2330 string.
2331
2332 (c) The code for expansion conditions now uses a table and a binary chop
2333 instead of a serial search (which was left over from when there were very
2334 few conditions). Also, it now recognizes conditions like "pam" even when
2335 the relevant support is not compiled in: a suitably worded error message is
2336 given if an attempt is made to use such a condition.
2337
233811. Added ${time_interval:xxxxx}.
2339
234012. A bug was causing one of the ddress fields not to be passed back correctly
2341 from remote delivery subprocesses. The field in question was not being
2342 subsequently used, so this caused to problems in practice.
2343
234413. Added new log selectors queue_time and deliver_time.
2345
234614. Might have fixed a bug in maildirsizefile handling that threw up
2347 "unexpected character" debug warnings, and recalculated the data
2348 unnecessarily. In any case, I expanded the warning message to give more
2349 information.
2350
235115. Added the message "Restricted characters in address" to the statements in
2352 the default ACL that block characters like @ and % in local parts.
2353
235416. Change 71 for release 4.31 proved to be much less benign that I imagined.
2355 Three changes have been made:
2356
2357 (a) There was a serious bug; a negative response to MAIL caused the whole
2358 recipient domain to be cached as invalid, thereby blocking all messages
2359 to all local parts at the same domain, from all senders. This bug has
2360 been fixed. The domain is no longer cached after a negative response to
2361 MAIL if the sender used is not empty.
2362
2363 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
2364 been restored.
2365
2366 (c) A new callout option, "use_sender" has been added for people who want
2367 the modified behaviour.
2368
2369
2370Exim version 4.31
2371-----------------
2372
2373 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
2374 Larry Rosenman.
2375
2376 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
2377 indeed breaks things for older releases.
2378
2379 3. Added additional logging to the case where there is a problem reading data
2380 from a filter that is running in a subprocess using a pipe, in order to
2381 try to track down a specific problem.
2382
2383 4. Testing facility fudge: when running in the test harness and attempting
2384 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
2385 getting "No route to host". Convert this to a timeout.
2386
2387 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
2388 warning.
2389
2390 6. Some OS don't have socklen_t but use size_t instead. This affects the
2391 fifth argument of getsockopt() amongst other things. This is now
2392 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
2393 can be set for individual OS. I have set it for SunOS5, OSF1, and
2394 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
2395 some earlier ones do not.
2396
2397 7. Change 4.30/15 was not doing the test caselessly.
2398
2399 8. The standard form for an IPv6 address literal was being rejected by address
2400 parsing in, for example, MAIL and RCPT commands. An example of this kind of
2401 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
2402 this, as well as the form without the "IPv6" on the front (but only when
2403 address literals are enabled, of course).
2404
2405 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
2406
240710. Exim crashed if a message with an empty sender address specified by -f
2408 encountered a router with an errors_to setting. This could be provoked only
2409 by a command such as
2410
2411 exim -f "" ...
2412
2413 where an empty string was supplied; "<>" did not hit this bug.
2414
241511. Installed PCRE release 4.5.
2416
241712. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
2418 remained set. It is now erased.
2419
242013. exiqgrep wasn't working on MacOS X because it didn't correctly compute
2421 times from message ids (which are base 36 rather than the normal 62).
2422
242314. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
2424 were being counted as actual protocol errors, and logged if the log
2425 selector +smtp_protocol_error was set. One cannot be perfect in this test,
2426 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
2427 and DATA following a set of rejected RCPTs do not count as protocol errors.
2428 In other words, Exim assumes they were pipelined, though this may not
2429 actually be the case. Of course, in all cases the client gets an
2430 appropriate error code.
2431
243215. If a lookup fails in an ACL condition, a message about the failure may
2433 be available; it is used if testing the ACL cannot continue, because most
2434 such messages specify what the cause of the deferral is. However, some
2435 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
2436 that caused an old message to be retained and used if a later statement
2437 caused a defer, replacing the real cause of the deferral.
2438
243916. If an IP address had so many PTR records that the DNS lookup buffer
2440 was not large enough to hold them, Exim could crash while trying to process
2441 the truncated data. It now detects and logs this case.
2442
244317. Further to 4.21/58, another change has been made: if (and only if) the
2444 first line of a message (the first header line) ends with CRLF, a bare LF
2445 in a subsequent header line has a space inserted after it, so as not to
2446 terminate the header.
2447
244818. Refactoring: tidied an ugly bit of code in appendfile that copied data
2449 unnecessarily, used atoi() instead of strtol(), and didn't check the
2450 termination when getting file sizes from file names by regex.
2451
245219. Completely re-implemented the support for maildirsize files, in the light
2453 of a number of problems with the previous contributed implementation
2454 (4.30/29). In particular:
2455
2456 . If the quota is zero, the maildirsize file is maintained, but no quota is
2457 imposed.
2458
2459 . If the maildir directory does not exist, it is created before any attempt
2460 to write a maildirsize file.
2461
2462 . The quota value in the file is just a cache; if the quota is changed in
2463 the transport, the new value overrides.
2464
2465 . A regular expression is available for excluding directories from the
2466 count.
2467
246820. The autoreply transport checks the characters in options that define the
2469 message's headers; it allows continued headers, but it was checking with
2470 isspace() after an embedded newline instead of explicitly looking for a
2471 space or a tab.
2472
247321. If all the "regular" hosts to which an address was routed had passed their
2474 expiry times, and had not reached their retry times, the address was
2475 bounced, even if fallback hosts were defined. Now Exim should go on to try
2476 the fallback hosts.
2477
247822. Increased buffer sizes in the callout code from 1024 to 4096 to match the
2479 equivalent code in the SMTP transport. Some hosts send humungous responses
2480 to HELO/EHLO, more than 1024 it seems.
2481
248223. Refactoring: code in filter.c used (void *) for "any old type" but this
2483 gives compiler warnings in some environments. I've now done it "properly",
2484 using a union.
2485
248624. The replacement for inet_ntoa() that is used with gcc on IRIX systems
2487 (because of problems with the built-in one) was declared to return uschar *
2488 instead of char *, causing compiler failure.
2489
249025. Fixed a file descriptor leak when processing alias/forward files.
2491
249226. Fixed a minor format string issue in dbfn.c.
2493
249427. Typo in exim.c: ("dmbnz" for "dbmnz").
2495
249628. If a filter file refered to $h_xxx or $message_headers, and the headers
2497 contained RFC 2047 "words", Exim's memory could, under certain conditions,
2498 become corrupted.
2499
250029. When a sender address is verified, it is cached, to save repeating the test
2501 when there is more than one recipient in a message. However, when the
2502 verification involves a callout, it is possible for different callout
2503 options to be set for different recipients. It is too complicated to keep
2504 track of this in the cache, so now Exim always runs a verification when a
2505 callout is required, relying on the callout cache for the optimization.
2506 The overhead is duplication of the address routing, but this should not be
2507 too great.
2508
250930. Fixed a bug in callout caching. If a RCPT command caused the sender address
2510 to be verified with callout=postmaster, and the main callout worked but the
2511 postmaster check failed, the verification correctly failed. However, if a
2512 subsequent RCPT command asked for sender verification *without* the
2513 postmaster check, incorrect caching caused this verification also to fail,
2514 incorrectly.
2515
251631. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
2517 it was not caching the DNS options (qualify_single, search_parents) that
2518 were used when the lookup failed. A subsequent lookup with different
2519 options therefore always gave the same answer, though there were cases
2520 where it should not have. (Example: a "domains = !$mx_any" option on a
2521 dnslookup router: the "domains" option is always processed without any
2522 widening, but the router might have qualify_single set.) Now Exim uses the
2523 cached value only when the same options are set.
2524
252532. Added John Jetmore's "exipick" utility to the distribution.
2526
252733. GnuTLS: When an attempt to start a TLS session fails for any reason other
2528 than a timeout (e.g. a certificate is required, and is not provided), an
2529 Exim server now closes the connection immediately. Previously it waited for
2530 the client to close - but if the client is SSL, it seems that they each
2531 wait for each other, leading to a delay before one of them times out.
2532
253334: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
2534 maintained 0.8.x compatibility because I don't think many are using it, and
2535 it is clearly obsolete.
2536
253735. Added TLS support for CRLs: a tls_crl global option and one for the smtp
2538 transport.
2539
254036. OpenSSL: $tls_certificate_verified was being set to 1 even if the
2541 client certificate was expired. A simple patch fixes this, though I don't
2542 understand the full logic of why the verify callback is called multiple
2543 times.
2544
254537. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
2546 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
2547 which causes problems with some clients (such as the Certicom SSL Plus
2548 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
2549 disables the coutermeasure allowing Eudora to connect."
2550
255138. Exim was not checking that a write() to a log file succeeded. This could
2552 lead to Bad Things if a log got too big, in particular if it hit a file
2553 size limit. Exim now panics and dies if it cannot write to a log file, just
2554 as it does if it cannot open a log file.
2555
255639. Modified OS/Makefile-Linux so that it now contains
2557
2558 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
2559
2560 The two -D definitions ensure that Exim is compiled with large file
2561 support, which makes it possible to handle log files that are bigger than
2562 2^31.
2563
256440. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
2565 instance) a domain was checked against a named list that involved a lookup,
2566 causing $domain_data to be set, then another domain was checked against the
2567 same list, then the first domain was re-checked, the value of $domain_data
2568 after the final check could be wrong. In particular, if the second check
2569 failed, it could be set empty. This bug probably also applied to
2570 $localpart_data.
2571
257241. The strip_trailing_dot option was not being applied to the address given
2573 with the -f command-line option.
2574
257542. The code for reading a message's header from the spool was incrementing
2576 $received_count, but never initializing it. This meant that the value was
2577 incorrect (doubled) while delivering a message in the same process in which
2578 it was received. In the most common configuration of Exim, this never
2579 happens - a fresh exec is done - but it can happen when
2580 deliver_drop_privilege is set.
2581
258243. When Exim logs an SMTP synchronization error - client data sent too soon -
2583 it now includes up to 150 characters of the unexpected data in the log
2584 line.
2585
258644. The exim_dbmbuild utility uses fixed size buffers for reading input lines
2587 and building data strings. The size of both of these buffers was 10 000
2588 bytes - far larger than anybody would *ever* want, thought I. Needless to
2589 say, somebody hit the limit. I have increased the maximum line length to
2590 20 000 and the maximum data length of concatenated lines to 100 000. I have
2591 also fixed two bugs, because there was no checking on these buffers. Tsk,
2592 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
2593 buffer is too small.
2594
259545. The exim_dbmbuild utility did not support quoted keys, as Exim does in
2596 lsearch lookups. Now it does.
2597
259846. When parsing a route_list item in a manualroute router, a fixed-length
2599 buffer was used for the list of hosts. I made this 1024 bytes long,
2600 thinking that nobody would ever have a list of hosts that long. Wrong.
2601 Somebody had a whole pile of complicated expansion conditions, and the
2602 string was silently truncated, leading to an expansion error. It turns out
2603 that it is easier to change to an unlimited length (owing to other changes
2604 that have happened since this code was originally written) than to build
2605 structure for giving a limitation error. The length of the item that
2606 expands into the list of hosts is now unlimited.
2607
260847. The lsearch lookup could not handle data where the length of text line was
2609 more than 4095 characters. Such lines were truncated, leading to shortened
2610 data being returned. It should now handle lines of any length.
2611
261248. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
2613 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
2614 ACL").
2615
261649. Cosmetic tidy to scripts like exicyclog that are generated by globally
2617 replacing strings such as BIN_DIRECTORY in a source file: the replacement
2618 no longer happens in comment lines. A list of replacements is now placed
2619 at the head of all of the source files, except those whose only change is
2620 to replace PERL_COMMAND in the very first #! line.
2621
262250. Replaced the slow insertion sort in queue.c, for sorting the list of
2623 messages on the queue, with a bottom-up merge sort, using code contributed
2624 by Michael Haardt. This should make operations like -bp somewhat faster on
2625 large queues. It won't affect queue runners, except when queue_run_in_order
2626 is set.
2627
262851. Installed eximstats 1.31 in the distribution.
2629
263052. Added support for SRV lookups to the dnslookup router.
2631
263253. If an ACL referred to $message_body or $message_body_end, the value was not
2633 reset for any messages that followed in the same SMTP session.
2634
263554. The store-handling optimization for building very long strings was not
2636 differentiating between the different store pools. I don't think this
2637 actually made any difference in practice, but I've tidied it.
2638
263955. While running the routers to verify a sender address, $sender_address
2640 was still set to the sender address. This is wrong, because when routing to
2641 send a bounce to the sender, it would be empty. Therefore, I have changed
2642 it so that, while verifying a sender address, $sender_address is set to <>.
2643 (There is no change to what happens when verifying a recipient address.)
2644
264556. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
2646 target A or AAAA records (if not already returned) without resetting the
2647 qualify_single or search_parents options of the DNS resolver. These are
2648 inappropriate in this case because the targets of MX and SRV records must
2649 be FQDNs. A broken DNS record could cause trouble if it happened to have a
2650 target that, when qualified, matched something in the local domain. These
2651 two options are now turned off when doing these lookups.
2652
265357. It seems that at least some releases of Reiserfs (which does not have the
2654 concept of a fixed number of inodes) returns zero and not -1 for the
2655 number of available inodes. This interacted badly with check_spool_inodes,
2656 which assumed that -1 was the "no such thing" setting. What I have done is
2657 to check that the total number of inodes is greater than zero before doing
2658 the test of how many are available.
2659
266058. When a "warn" ACL statement has a log_message modifier, the message is
2661 remembered, and not repeated. This is to avoid a lot of repetition when a
2662 message has many recipients that cause the same warning to be written.
2663 Howewer, Exim was preserving the list of already written lines for an
2664 entire SMTP session, which doesn't seem right. The memory is now reset if a
2665 new message is started.
2666
266759. The "rewrite" debugging flag was not showing the result of rewriting in the
2668 debugging output unless log_rewrite was also set.
2669
267060. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
2671 of (int)(handle) when we know that handle contains (void *)(-1).
2672
267361. The Exim daemon panic-logs an error return when it closes the incoming
2674 connection. However "connection reset by peer" seems to be common, and
2675 isn't really an error worthy of noting specially, so that particular error
2676 is no long logged.
2677
267862. When Exim is trying to find all the local interfaces, it used to panic and
2679 die if the ioctl to get the interface flags failed. However, it seems that
2680 on at least one OS (Solaris 9) it is possible to have an interface that is
2681 included in the list of interfaces, but for which you get a failure error
2682 for this call. This happens when the interface is not "plumbed" into a
2683 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
2684 failure of the "get flags" call assumes that the interface is down.
2685
268663. Added a ${eval10: operator, which assumes all numbers are decimal. This
2687 makes life easier for people who are doing arithmetic on fields extracted
2688 from dates, where you often get leading zeros that should not be
2689 interpreted as octal.
2690
269164. Added qualify_domain to the redirect router, to override the global
2692 setting.
2693
269465. If a pathologically long header line contained very many addresses (the
2695 report of this problem mentioned 10 000) and each of them was rewritten,
2696 Exim could use up a very large amount of memory. (It kept on making new
2697 copies of the header line as it rewrote, and never released the old ones.)
2698 At the expense of a bit more processing, the header rewriting function has
2699 been changed so that it no longer eats memory in this way.
2700
270166. The generation of the Received: header has been moved from the time that a
2702 message starts to be received, to the time that it finishes. The timestamp
2703 in the Received: header should now be very close to that of the <= log
2704 line. There are two side-effects of this change:
2705
2706 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
2707 logged header lines no longer include the local Received: line, because
2708 it has not yet been created. The same applies to a copy of the message
2709 that is returned to a non-SMTP sender when a message is rejected.
2710
2711 (b) When a filter file is tested using -bf, no additional Received: header
2712 is added to the test message. After some thought, I decided that this
2713 is a bug fix.
2714
2715 This change does not affect the value of $received_for. It is still set
2716 after address rewriting, but before local_scan() is called.
2717
271867. Installed the latest Cygwin-specific files from the Cygwin maintainer.
2719
272068. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
2721 gave an unhelpful panic error message, and a defer error. I have managed to
2722 change this behaviour so that it now rejects any supplied certificate,
2723 which seems right, as the list of acceptable certificates is empty.
2724
272569. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
2726 gave an unhelpful defer error. I have not managed to make this reject any
2727 supplied certificates, but the error message it gives is "no certificate
2728 supplied", which is not helpful.
2729
273070. exigrep's output now also includes lines that are not associated with any
2731 message, but which match the given pattern. Implemented by a patch from
2732 Martin Sluka, which also tidied up the Perl a bit.
2733
273471. Recipient callout verification, like sender verification, was using <> in
2735 the MAIL FROM command. This isn't really the right thing, since the actual
2736 sender may affect whether the remote host accepts the recipient or not. I
2737 have changed it to use the actual sender in the callout; this means that
2738 the cache record is now keyed on a recipient/sender pair, not just the
2739 recipient address. There doesn't seem to be a real danger of callout loops,
2740 since a callout by the remote host to check the sender would use <>.
2741 [SEE ABOVE: changed after hitting problems.]
2742
274372. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
2744 temporary errors. However, in the case of such a code being given after
2745 the end of a data transmission (i.e. after ".") Exim was failing to write
2746 a retry record for the message. (Yes, there was some broken host that was
2747 actually sending 8xx at this point.)
2748
274973. An unknown lookup type in a host list could cause Exim to panic-die when
2750 the list was checked. (An example that provoked this was putting <; in the
2751 middle of a list instead of at the start.) If this happened during a DATA
2752 ACL check, a -D file could be left lying around. This kind of configuration
2753 error no longer causes Exim to die; instead it causes a defer errror. The
2754 incident is still logged to the main and panic logs.
2755
275674. Buglet left over from Exim 3 conversion. The message "too many messages
2757 in one connection" was written to the rejectlog but not the mainlog, except
2758 when address rewriting (yes!) was being logged.
2759
276075. Added write_rejectlog option.
2761
276276. When a system filter was run not as root (that is, when system_filter_user
2763 was set), the values of the $n variables were not being returned to the
2764 main process; thus, they were not subsequently available in the $sn
2765 variables.
2766
276777. Added +return_path_on_delivery log selector.
2768
276978. A connection timeout was being treated differently from recipients deferred
2770 when testing hosts_max_try with a message that was older than the host's
2771 retry timeout. (The host should not be counted, thus allowing all hosts to
2772 be tried at least once before bouncing.) This may have been the cause of an
2773 occasionally reported bug whereby a message would remain on the queue
2774 longer than the retry timeout, but would be bounced if a delivery was
2775 forced. I say "may" because I never totally pinned down the problem;
2776 setting up timeout/retry tests is difficult. See also the next item.
2777
277879. The ultimate address timeout was not being applied to errors that involved
2779 a combination of host plus message (for example, a timeout on a MAIL
2780 command). When an address resolved to a number of possible hosts, and they
2781 were not all tried for each delivery (e.g. because of hosts_max_try), a
2782 message could remain on the queue longer than the retry timeout.
2783
278480. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
2785 Haardt.
2786
278781. Fixed an obscure SMTP outgoing bug which required at least the following
2788 conditions: (a) there was another message waiting for the same server;
2789 (b) the server returned 5xx to all RCPT commands in the first message so
2790 that the message was not completed; (c) the server dropped the connection
2791 or gave a negative response to the RSET that Exim sends to abort the
2792 transaction. The observed case was a dropped connection after DATA that had
2793 been sent in pipelining mode. That is, the server had advertised PIPELINING
2794 but was not implementing it correctly. The effect of the bug was incorrect
2795 behaviour, such as trying another host, and this could lead to a crash.
2796
2797
2798Exim version 4.30
2799-----------------
2800
2801 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
2802 and daemon.c were passed as pointers to ints; they should have been
2803 pointers to socklen_t variables (which are typically unsigned ints).
2804
2805 2. Some signed/unsigned type warnings in the os.c file for Linux have been
2806 fixed.
2807
2808 3. Fixed a really odd bug that affected only the testing scheme; patching a
2809 certain fixed string in the binary changed the value of another string that
2810 happened to be identical to the end of the original first string.
2811
2812 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
2813 name", it returns that address as the IP address. On some operating
2814 systems (e.g. Solaris), it also passes back the IP address string as the
2815 "host name". However, on others (e.g. Linux), it passes back an empty
2816 string. Exim wasn't checking for this, and was changing the host name to an
2817 empty string, assuming it had been canonicized.
2818
2819 5. Although rare, it is permitted to have more than one PTR record for a given
2820 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
2821 all the names associated with an address, because they do in Solaris.
2822 However, it seems that they do not in Linux for data that comes from the
2823 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
2824 I found this out when I moved to a new Linux workstation and tried to run
2825 the Exim test suite.
2826
2827 To get round this problem I have changed the code so that it now does its
2828 own call to the DNS to look up PTR records when searching for a host name.
2829 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
2830 addresses that are only in /etc/hosts are still found.
2831
2832 This behaviour is, however, controlled by an option called host_lookup_
2833 order, which defaults to "bydns:byaddr". If people want to use the other
2834 order, or indeed, just use one or the other means of lookup, they can
2835 specify it in this variable.
2836
2837 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
2838 some operating systems, this comes back from gethostbyaddr() as an empty
2839 string, and this is what Exim used to test for. However, it seems that in
2840 other systems, "." is yielded. Exim now tests for this case too.
2841
2842 7. The values of check_spool_space and check_log_space are now held internally
2843 as a number of kilobytes instead of an absolute number of bytes. If a
2844 numbers is specified without 'K' or 'M', it is rounded up to the nearest
2845 kilobyte. This means that much larger values can be stored.
2846
2847 8. Exim monitor: an attempt to get the action menu when not actually pointing
2848 at a message produces an empty menu entitled "No message selected". This
2849 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
2850 no entries in it ("Shell widget menu has zero width and/or height"). So I
2851 have added a single, blank menu entry in this case.
2852
2853 9. Added ${quote_local_part.
2854
285510. MIME decoding is now applied to the contents of Subject: header lines when
2856 they are logged.
2857
285811. Now that a reference to $sender_host_address automatically causes a reverse
2859 lookup to occur if necessary (4.13/18), there is no need to arrange for a
2860 host lookup before query-style lookups in lists that might use this
2861 variable. This has therefore been abolished, and the "net-" prefix is no
2862 longer necessary for query-style lookups.
2863
286412. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
2865 have been a typo for LFLAGS, so it has been changed.
2866
286713. The install script calls Exim with "-C /dev/null" in order to find the
2868 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
2869 to be output. Howeve, since Exim outputs its version number before the
2870 error, it didn't break the script. It just looked ugly. I fixed this by
2871 always allowing "-C /dev/null" if the caller is root.
2872
287314. Ignore overlarge ACL variable number when reading spool file - insurance
2874 against a later release with more variables having written the file.
2875
287615. The standard form for an IPv6 address literal was being rejected by EHLO.
2877 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
2878 this, as well as the form without the "IPv6" on the front.
2879
288016. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
2881 OS/Makefile-Darwin file.
2882
288317. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
2884 only to LDAP libraries that do not have LDAP_OPT_DEREF.
2885
288618. After change 4.21/52, "%ld" was used to format the contents of the $inode
2887 variable. However, some OS use ints for inodes. I've added cast to long int
2888 to get rid of the compiler warning.
2889
289019. I had forgotten to lock out "/../" in configuration file names when
2891 ALT_CONFIG_PREFIX was set.
2892
289320. Routers used for verification do not need to specify transports. However,
2894 if such a router generated a host list, and callout was configured, Exim
2895 crashed, because it could not find a port number from the (non-existent)
2896 transport. It now assumes port 25 in this circumstance.
2897
289821. Added the -t option to exigrep.
2899
290022. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
2901 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
2902 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
2903 them is set, LOOKUP_LSEARCH is forced.
2904
290523. "exim -bV" now outputs a list of lookups that are included in the binary.
2906
290724. Added sender and host information to the "rejected by local_scan()" log
2908 line; previously there was no indication of these.
2909
291025. Added .include_if_exists.
2911
291226. Change 3.952/11 added an explicit directory sync on top of a file sync for
2913 Linux. It turns out that not all file systems support this. Apparently some
2914 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
2915 do it.) To cope with this, the error EINVAL, which means that sync-ing is
2916 not supported on the file descriptor, is now ignored when Exim is trying to
2917 sync a directory. This applies only to Linux.
2918
291927. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
2920
292128. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
2922 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
2923 this.
2924
292529. Added support for maildirsize files from supplied patch (modified a bit).
2926
292730. The use of :fail: followed by an empty string could lead Exim to respond to
2928 sender verification failures with (e.g.):
2929
2930 550 Verification failed for <xxx>
2931 550 Sender verify failed
2932
2933 where the first response line was missing the '-' that indicates it is not
2934 the final line of the response.
2935
293631. The loop for finding the name of the user that called Exim had a hardwired
2937 limit of 10; it now uses the value of finduser_retries, which is used for
2938 all other user lookups.
2939
294032. Added $received_count variable, available in data and not_smtp ACLs, and at
2941 delivery time.
2942
294333. Exim was neglecting to zero errno before one call of strtol() when
2944 expanding a string and expecting an integer value. On some systems this
2945 resulted in spurious "integer overflow" errors. Also, it was casting the
2946 result into an int without checking.
2947
294834. Testing for a connection timeout using "timeout_connect" in the retry rules
2949 did not work. The code looks as if it has *never* worked, though it appears
2950 to have been documented since at least releast 1.62. I have made it work.
2951
295235. The "timeout_DNS" error in retry rules, also documented since at least
2953 1.62, also never worked. As it isn't clear exactly what this means, and
2954 clearly it isn't a major issue, I have abolished the feature by treating it
2955 as "timeout", and writing a warning to the main and panic logs.
2956
295736. The display of retry rules for -brt wasn't always showing the error code
2958 correctly.
2959
296037. Added new error conditions to retry rules: timeout_A, timeout_MX,
2961 timeout_connect_A, timeout_connect_MX.
2962
296338. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
2964 to the empty sender.
2965
296639. The daemon was not analysing the content of -oX till after it had closed
2967 stderr and disconnected from the controlling terminal. This meant that any
2968 syntax errors were only noted on the panic log, and the return code from
2969 the command was 0. By re-arranging the code a little, I've made the
2970 decoding happen first, so such errors now appear on stderr, and the return
2971 code is 1. However, the actual setting up of the sockets still happens in
2972 the disconnected process, so errors there are still only recorded on the
2973 panic log.
2974
297540. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
2976 connections (as happens on some IP stacks) was logged at start up time as
2977 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
2978 it from "IPv6 and IPv4", which means that two separate sockets are being
2979 used.
2980
298141. The debug output for gethostbyname2() or getipnodebyname() failures now
2982 says whether AF_INET or AF_INET6 was passed as an argument.
2983
298442. Exiwhat output was messed up when time zones were included in log
2985 timestamps.
2986
298743. Exiwhat now gives more information about the daemon's listening ports,
2988 and whether -tls-on-connect was used.
2989
299044. The "port" option of the smtp transport is now expanded.
2991
299245. A "message" modifier in a "warn" statement in a non-message ACL was being
2993 silently ignored. Now an error message is written to the main and panic
2994 logs.
2995
299646. There's a new ACL modifier called "logwrite" which writes to a log file
2997 as soon as it is encountered.
2998
299947. Added $local_user_uid and $local_user_gid at routing time.
3000
300148. Exim crashed when trying to verify a sender address that was being
3002 rewritten to "<>".
3003
300449. Exim was recognizing only a space character after ".include". It now also
3005 recognizes a tab character.
3006
300750. Fixed several bugs in the Perl script that creates the exim.8 man page by
3008 extracting the relevant information from the specification. The man page no
3009 longer contains scrambled data for the -d option, and I've added a section
3010 at the front about calling Exim under different names.
3011
301251. Added "extra_headers" argument to the "mail" command in filter files.
3013
301452. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
3015 crash.
3016
301753. Installed eximstats 1.29.
3018
301954. Added transport_filter_timeout as a generic transport option.
3020
302155. Exim no longer adds an empty Bcc: header to messages that have no To: or
3022 Cc: header lines. This was required by RFC 822, but it not required by RFC
3023 2822.
3024
302556. Exim used to add From:, Date:, and Message-Id: header lines to any
3026 incoming messages that did not have them. Now it does so only if the
3027 message originates locally, that is, if there is no associated remote host
3028 address. When Resent- header lines are present, this applies to the Resent-
3029 lines rather than the non-Resent- lines.
3030
303157. Drop incoming SMTP connection after too many syntax or protocol errors. The
3032 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
3033
303458. Messages for configuration errors now include the name of the main
3035 configuration file - useful now that there may be more than one file in a
3036 list (.included file names were always shown).
3037
303859. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
3039 for those rare installations that do not start the daemon as root or run it
3040 setuid root. I've cut out the call to initgroups() if the daemon is not
3041 root at that time.
3042
304360. The Exim user and group can now be bound into the binary as text strings
3044 that are looked up at the start of Exim's processing.
3045
304661. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
3047
304862. Added $mailstore_basename variable.
3049
305063. Installed patch to sieve.c from Michael Haardt.
3051
305264. When Exim failed to open the panic log after failing to open the main log,
3053 the original message it was trying to log was written to stderr and debug
3054 output, but if they were not available (the usual case in production), it
3055 was lost. Now it is written to syslog before the two lines that record the
3056 failures to open the logs.
3057
305865. Users' Exim filters run in subprocesses under the user's uid. It is
3059 possible for a "deliver" command or an alias in a "personal" command to
3060 provoke an address rewrite. If logging of address rewriting is configured,
3061 this fails because the process is not running as root or exim. There may be
3062 a better way of dealing with this, but for the moment (because 4.30 needs
3063 to be released), I have disabled address rewrite logging when running a
3064 filter in a non-root, non-exim process.
3065
3066
3067Exim version 4.24
3068-----------------
3069
3070 1. The buildconfig auxiliary program wasn't quoting the value set for
3071 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
3072 not defined. This bug was masked in 4.22 by the effect that was fixed in
3073 change 4.23/1.
3074
3075 2. Some messages that were rejected after a message id was allocated were
3076 shown as "incomplete" by exigrep. It no longer does this for messages that
3077 are rejected by local_scan() or the DATA or non-SMTP ACLs.
3078
3079 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
3080 have allow_domain_literals set, the ID did not get logged in the <= line.
3081 Domain literals are now always recognized in Message-ID: header lines.
3082
3083 4. The first argument for a ${extract expansion item is the key name or field
3084 number. Leading and trailing spaces in this item were not being ignored,
3085 causing some misleading effects.
3086
3087 5. When deliver_drop_privilege was set, single queue runner processes started
3088 manually (i.e. by the command "exim -q") or by the daemon (which uses the
3089 same command in the process it spins off) were not dropping privilege.
3090
3091 6. When the daemon running as "exim" started a queue runner, it always
3092 re-executed Exim in the spun-off process. This is a waste of effort when
3093 deliver_drop_privilege is set. The new process now just calls the
3094 queue-runner function directly.
3095
3096
3097Exim version 4.23
3098-----------------
3099
3100 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
3101 HEADERS_CHARSET.
3102
3103 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
3104 ignored. Though the use of -oP was forcing the writing of a pid file, it
3105 was always written to the default place.
3106
3107 3. If the message "no IP address found for host xxxx" is generated during
3108 incoming verification, it is now followed by identification of the incoming
3109 connection (so you can more easily find what provoked it).
3110
3111 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
3112
3113 5. Added some features to "harden" Exim a bit more against certain attacks:
3114
3115 (a) There is now a build-time option called FIXED_NEVER_USERS that can
3116 be put in Local/Makefile. This is like the never_users runtime option,
3117 but it cannot be overridden. The default setting is "root".
3118
3119 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
3120 prefix string with which any file named in a -C command line option
3121 must start.
3122
3123 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
3124 is retained for -C and -D only if the caller of Exim is root. Without
3125 it, the exim user may also use -C and -D and retain privilege.
3126
3127 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
3128 command line option is disabled.
3129
3130 6. Macro names set by the -D option must start with an upper case letter, just
3131 like macro names defined in the configuration file.
3132
3133 7. Added "dereference=" facility to LDAP.
3134
3135 8. Two instances of the typo "uknown" in the source files are fixed.
3136
3137 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
3138 the Configure-Makefile script screwed up while processing it.
3139
314010. Incorporated PCRE 4.4.
3141
314211. The SMTP synchronization check was not operating right at the start of an
3143 SMTP session. For example, it could not catch a HELO sent before the client
3144 waited for the greeting. There is now a check for outstanding input at the
3145 point when the greeting is written. Because of the duplex, asynchronous
3146 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
3147 way, but not yet received, when the check is performed.
3148
314912. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
3150 on TCP/IP sockets, because this apparently causes some broken clients to
3151 timeout.
3152
315313. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
3154 unchanged) from the Cygwin maintainer.
3155
315614. The code for -bV that shows what is in the binary showed "mbx" when maildir
3157 was supported instead of testing for mbx. Effectively a typo.
3158
315915. The spa authenticator server code was not checking that the input it
3160 received was valid base64.
3161
316216. The debug output line for the "set" modifier in ACLs was not showing the
3163 name of the variable that was being set.
3164
316517. Code tidy: the variable type "vtype_string" was never used. Removed it.
3166
316718. Previously, a reference to $sender_host_name did not cause a DNS reverse
3168 lookup on its own. Something else was needed to trigger the lookup. For
3169 example, a match in host_lookup or the need for a host name in a host list.
3170 Now, if $sender_host_name is referenced and the host name has not yet been
3171 looked up, a lookup is performed. If the lookup fails, the variable remains
3172 empty, and $host_lookup_failed is set to "1".
3173
317419. Added "eqi" as a case-independent comparison operator.
3175
317620. The saslauthd authentication condition could segfault if neither service
3177 nor realm was specified.
3178
317921. If an overflowing value such as "2048M" was set for message_size_limit, the
3180 error message that was logged was misleading, and incoming SMTP
3181 connections were dropped. The message is now more accurate, and temporary
3182 errors are given to SMTP connections.
3183
318422. In some error situations (such as 21 above) Exim rejects all SMTP commands
3185 (except RSET) with a 421 error, until QUIT is received. However, it was
3186 failing to send a response to QUIT.
3187
318823. The HELO ACL was being run before the code for helo_try_verify_hosts,
3189 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
3190 ACL is now run after the helo_try_verify_hosts code.
3191
319224. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
3193 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
3194 case-independent, so other case variants are also recognized). Apparently
3195 some systems use these upper case variants.
3196
319725. If more than two messages were waiting for the same host, and a transport
3198 filter was specified for the transport, Exim sent two messages over the
3199 same TCP/IP connection, and then failed with "socket operation on non-
3200 socket" when it tried to send the third.
3201
320226. Added Exim::debug_write and Exim::log_write for embedded Perl use.
3203
320427. The extern definition of crypt16() in expand.c was not being excluded when
3205 the OS had its own crypt16() function.
3206
320728. Added bounce_return_body as a new option, and bounce_return_size_limit
3208 as a preferred synonym for return_size_limit, both as an option and as an
3209 expansion variable.
3210
321129. Added LIBS=-liconv to OS/Makefile-OSF1.
3212
321330. Changed the default configuration ACL to relax the local part checking rule
3214 for addresses that are not in any local domains. For these addresses,
3215 slashes and pipe symbols are allowed within local parts, but the sequence
3216 /../ is explicitly forbidden.
3217
321831. SPA server authentication was not clearing the challenge buffer before
3219 using it.
3220
322132. log_message in a "warn" ACL statement was writing to the reject log as
3222 well as to the main log, which contradicts the documentation and doesn't
3223 seem right (because no rejection is happening). So I have stopped it.
3224
322533. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
3226 However, I am unable to do any testing of this.
3227
322834. Fixed an infelicity in the appendfile transport. When checking directories
3229 for a mailbox, to see if any needed to be created, it was accidentally
3230 using path names with one or more superfluous leading slashes; tracing
3231 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
3232
323335. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
3234 contents are added to the log line that is written for every discarded
3235 recipient. (Previously a log_message setting was ignored.)
3236
323736. The ${quote: operator now quotes the string if it is empty.
3238
323937. The install script runs exim in order to find its version number. If for
3240 some reason other than non-existence or emptiness, which it checks, it
3241 could not run './exim', it was installing it with an empty version number,
3242 i.e. as "exim-". This error state is now caught, and the installation is
3243 aborted.
3244
324538. An argument was missing from the function that creates an error message
3246 when Exim fails to connect to the socket for saslauthd authentication.
3247 This could cause Exim to crash, or give a corrupted message.
3248
324939. Added isip, isip4, and isip6 to ${if conditions.
3250
325140. The ACL variables $acl_xx are now saved with the message, and can be
3252 accessed later in routers, transports, and filters.
3253
325441. The new lookup type nwildlsearch is like wildlsearch, except that the key
3255 strings in the file are not string-expanded.
3256
325742. If a MAIL command specified a SIZE value that was too large to fit into an
3258 int variable, the check against message_size_limit failed. Such values are
3259 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
3260 day this will have to be increased, but I don't think I want to be around
3261 when emails are that large.
3262
3263
3264
3265Exim version 4.22
3266-----------------
3267
3268 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
3269 iconv() is not standard in FreeBSD.
3270
3271 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
3272 IPv6 enabled. The observed symptom was a segmentation fault on return from
3273 the function os_common_find_running_interfaces() in src/os.c.
3274
3275 3. In the check_special_case() function in daemon.c I had used "errno" as an
3276 argument name, which causes warnings on some systems. This was basically a
3277 typo, since it was named "eno" in the comments!
3278
3279 4. The code that waits for the clock to tick (at a resolution of some fraction
3280 of a second) so as to ensure message-id uniqueness was always waiting for
3281 at least one whole tick, when it could have waited for less. [This is
3282 almost certainly not relevant at current processor speeds, where it is
3283 unlikely to ever wait at all. But we try to future-proof.]
3284
3285 5. The function that sleeps for a time interval that includes fractions of a
3286 second contained a race. It did not block SIGALRM between setting the
3287 timer, and suspending (a couple of lines later). If the interval was short
3288 and the sigsuspend() was delayed until after it had expired, the suspension
3289 never ended. On busy systems this could lead to processes getting stuck for
3290 ever.
3291
3292 6. Some uncommon configurations may cause a lookup to happen in a queue runner
3293 process, before it forks any delivery processes. The open lookup caching
3294 mechanism meant that the open file or database connection was passed into
3295 the delivery process. The problem was that delivery processes always tidy
3296 up cached lookup data. This could cause a problem for the next delivery
3297 process started by the queue runner, because the external queue runner
3298 process does not know about the closure. So the next delivery process
3299 still has data in the lookup cache. In the case of a file lookup, there was
3300 no problem because closing a file descriptor in a subprocess doesn't affect
3301 the parent. However, if the lookup was caching a connection to a database,
3302 the connection was closed, and the second delivery process was likely to
3303 see errors such as "PGSQL: query failed: server closed the connection
3304 unexpectedly". The problem has been fixed by closing all cached lookups
3305 in a queue runner before running a delivery process.
3306
3307 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
3308 seem to have the "const" qualifier which it has on other OS. I've
3309 parameterised it.
3310
3311 8. Change 4.21/2 was too strict. It is only if there are two authenticators
3312 *of the same type* (client or server) with the same public name that an
3313 error should be diagnosed.
3314
3315 9. When Exim looked up a host name for an IP address, but failed to find the
3316 original IP address when looking up the host name (a safety check), it
3317 output the message "<ip address> does not match any IP for NULL", which was
3318 confusing, to say the least. The bug was that the host name should have
3319 appeared instead of "NULL".
3320
332110. Since release 3.03, if Exim is called by a uid other than root or the Exim
3322 user that is built into the binary, and the -C or -D options is used, root
3323 privilege is dropped before the configuration file is read. In addition,
3324 logging is switched to stderr instead of the normal log files. If the
3325 configuration then re-defines the Exim user, the unprivileged environment
3326 is probably not what is expected, so Exim logs a panic warning message (but
3327 proceeds).
3328
3329 However, if deliver_drop_privilege is set, the unprivileged state may well
3330 be exactly what is intended, so the warning has been cut out in that case,
3331 and Exim is allowed to try to write to its normal log files.
3332
3333
3334Exim version 4.21
3335-----------------
3336
3337 1. smtp_return_error_details was not giving details for temporary sender
3338 or receiver verification errors.
3339
3340 2. Diagnose a configuration error if two authenticators have the same public
3341 name.
3342
3343 3. Exim used not to create the message log file for a message until the first
3344 delivery attempt. This could be confusing when incoming messages were held
3345 for policy or load reasons. The message log file is now created at the time
3346 the message is received, and an initial "Received" line is written to it.
3347
3348 4. The automatically generated man page for command line options had a minor
3349 bug that caused no ill effects; however, a more serious problem was that
3350 the procedure for building the man page automatically didn't always
3351 operate. Consequently, release 4.20 contains an out-of-date version. This
3352 shouldn't happen again.
3353
3354 5. When building Exim with embedded Perl support, the script that builds the
3355 Makefile was calling 'perl' to find its compile-time parameters, ignoring
3356 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
3357
3358 6. The freeze_tell option was not being used for messages that were frozen on
3359 arrival, either by an ACL or by local_scan().
3360
3361 7. Added the smtp_incomplete_transaction log selector.
3362
3363 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
3364 was accepting AUTH without a new EHLO.
3365
3366 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
3367 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
3368 RFC.
3369
337010. Logging of TCP/IP connections (when configured) now happens in the main
3371 daemon process instead of the child process, so that the TCP/IP connection
3372 count is more accurate (but it can never be perfect).
3373
337411. The use of "drop" in a nested ACL was not being handled correctly in the
3375 outer ACL. Now, if condition failure induced by the nested "drop" causes
3376 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
3377 or "require"), the connection is dropped.
3378
337912. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
3380 that yield "discard" can now be used with an "accept" or a "discard" verb,
3381 but an error is generated for any others (because I can't see a useful way
3382 to define what should happen).
3383
338413. When an ACL is read dynamically from a file (or anywhere else), the lines
3385 are now processed in the same way as lines in the Exim configuration file.
3386 In particular, continuation lines are supported.
3387
338814. Added the "dnslists = a.b.c!=n.n.n.n" feature.
3389
339015. Added -ti meaning -t -i.
3391
339216. Check for letters, digits, hyphens, and dots in the names of dnslist
3393 domains, and warn by logging if others are found.
3394
339517. At least on BSD, alignment is not guarenteed for the array of ifreq's
3396 returned from GIFCONF when Exim is trying to find the list of interfaces on
3397 a host. The code in os.c has been modified to copy each ifreq to an aligned
3398 structure in all cases.
3399
3400 Also, in some cases, the returned ifreq's were being copied to a 'struct
3401 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
3402 means the last couple of bytes of an IPv6 address could be chopped if the
3403 ifreq contained only a normal sockaddr (14 bytes storage).
3404
340518. Named domain lists were not supported in the hosts_treat_as_local option.
3406 An entry such as +xxxx was not recognized, and was treated as a literal
3407 domain name.
3408
340919. Ensure that header lines added by a DATA ACL are included in the reject log
3410 if the ACL subsequently rejects the message.
3411
341220. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
3413 MD5 (which is deprecated).
3414
341521. When testing a filter file using -bf, Exim was writing a message when it
3416 took the sender from a "From " line in the message, but it was not doing so
3417 when it took $return_path from a Return-Path: header line. It now does.
3418
341922. If the contents of a "message" modifier for a "warn" ACL verb do not begin
3420 with a valid header line field name (a series of printing characters
3421 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
3422
342323. Changed "disc" in the source to "disk" to conform to the documentation and
3424 the book and for uniformity.
3425
342624. Ignore Sendmail's -Ooption=value command line item.
3427
342825. When execve() failed while trying to run a command in a pipe transport,
3429 Exim was returning EX_UNAVAILBLE (69) from the subprocess. However, this
3430 could be confused with a return value of 69 from the command itself. This
3431 has been changed to 127, the value the shell returns if it is asked to run
3432 a non-existent command. The wording for the related log line suggests a
3433 non-existent command as the problem.
3434
343526. If received_header_text expands to an empty string, do not add a Received:
3436 header line to the message. (Well, it adds a token one on the spool, but
3437 marks it "old" so that it doesn't get used or transmitted.)
3438
343927. Installed eximstats 1.28 (addition of -nt option).
3440
344128. There was no check for failure on the call to getsockname() in the daemon
3442 code. This can fail if there is a shortage of resources on the system, with
3443 ENOMEM, for example. A temporary error is now given on failure.
3444
344529. Contrary to the C standard, it seems that in some environments, the
3446 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
3447 program. Exim now does this explicitly; it affects the formatting of
3448 timestamps using strftime().
3449
345030. If exiqsumm was given junk data, it threw up some uninitialized variable
3451 complaints. I've now initialized all the variables, to avoid this.
3452
345332. Header lines added by a system filter were not being "seen" during
3454 transport-time rewrites.
3455
345633. The info_callback() function passed to OpenSSL is set up with type void
3457 (*)(SSL *, int, int), as described somewhere. However, when calling the
3458 function (actually a macro) that sets it up, the type void(*)() is
3459 expected. I've put in a cast to prevent warnings from picky compilers.
3460
346134. If a DNS black list lookup found a CNAME record, but there were no A
3462 records associated with the domain it pointed at, Exim crashed.
3463
346435. If a DNS black list lookup returned more than one A record, Exim ignored
3465 all but the first. It now scans all returned addresses if a particular IP
3466 value is being sought. In this situation, the contents of the
3467 $dnslist_value variable are a list of all the addresses, separated by a
3468 comma and a space.
3469
347036. Tightened up the rules for host name lookups using reverse DNS. Exim used
3471 to accept a host name and all its aliases if the forward lookup for any of
3472 them yielded the IP address of the incoming connection. Now it accepts only
3473 those names whose forward lookup yields the correct IP address. Any other
3474 names are discarded. This closes a loophole whereby a rogue DNS
3475 administrator could create reverse DNS records to break through a
3476 wildcarded host restriction in an ACL.
3477
347837. If a user filter or a system filter that ran in a subprocess used any of
3479 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
3480 the wrong values were passed to the pipe command ($thisaddress had the
3481 value of $0, $0 had the value of $1, etc). This bug was introduced by
3482 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
3483
348438. Improved the line breaking for long SMTP error messages from ACLs.
3485 Previously, if there was no break point between 40 and 75 characters, Exim
3486 left the rest of the message alone. Two changes have been made: (a) I've
3487 reduced the minimum length to 35 characters; (b) if it can't find a break
3488 point between 35 and 75 characters, it looks ahead and uses the first one
3489 that it finds. This may give the occasional overlong line, but at least the
3490 remaining text gets split now.
3491
349239. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
3493 file descriptors might be low, and that setting 1000 would always raise it.
3494 It turns out that in some environments, the limit is already over 1000 and
3495 that lowering it causes trouble. So now Exim takes care not to decrease it.
3496
349740. When delivering a message, the value of $return_path is set to $sender_
3498 address at the start of routing (routers may change the value). By an
3499 oversight, this default was not being set up when an address was tested by
3500 -bt or -bv, which affected the outcome if any router or filter referred to
3501 $return_path.
3502
350341. The idea of the "warn" ACL verb is that it adds a header or writes to the
3504 log only when "message" or "log_message" are set. However, if one of the
3505 conditions was an address verification, or a call to a nested ACL, the
3506 messages generated by the underlying test were being passed through. This
3507 no longer happens. The underlying message is available in $acl_verify_
3508 message for both "message" and "log_message" expansions, so it can be
3509 passed through if needed.
3510
351142. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
3512 new expansion $bh_ to give the encoded byte string without charset
3513 translation. Translation happens only if iconv() is available; HAVE_ICONV
3514 indicates this at build time. HEADERS_CHARSET gives the charset to
3515 translate to; headers_charset can change it in the configuration, and
3516 "headers charset" can change it in an individual filter file.
3517
351843. Now that we have a default RFC 2047 charset (see above), the code in Exim
3519 that creates RFC 2047 encoded "words" labels them as that charset instead
3520 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
3521 expansion operator; (ii) when Exim creates a From: line for a local
3522 message; (iii) when a header line is rewritten to include a "phrase" part.
3523
352444. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
3525 buggy, causing it to skip the first lines of messages whose message ID
3526 ended in 'D'. This would not have bitten before Exim release 4.14, because
3527 message IDs were unlikely to end in 'D' before then. The effect was to have
3528 incorrect size information for certain domains.
3529
353045. #include "config.h" was missing at the start of the crypt16.c module. This
3531 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
3532 noticed.
3533
353446. If there was a timeout during a "random" callout check, Exim treated it as
3535 a failure of the random address, and carried on sending RSET and the real
3536 address. If the delay was just some slowness somewhere, the response to the
3537 original RCPT would be taken as a response to RSET and so on, causing
3538 mayhem of various kinds.
3539
354047. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
3541 when I implemented it. It didn't allow for the fact that some option values
3542 may legitimatetly be negative (e.g. size_addition), and it didn't even do
3543 the right test for positive values.
3544
354548. Domain names in DNS records are case-independent. Exim always looks them up
3546 in lower case. Some resolvers return domain names in exactly the case they
3547 appear in the zone file, that is, they may contain uppercase letters. Not
3548 all resolvers do this - some return always lower case. Exim was treating a
3549 change of case by a resolver as a change of domain, similar to a widening
3550 of a domain abbreviation. This triggered its re-routing code and so it was
3551 trying to route what was effectively the same domain again. This normally
3552 caused routing to fail (because the router wouldn't handle the domain
3553 twice). Now Exim checks for this case specially, and just changes the
3554 casing of the domain that it ultimately uses when it transmits the message
3555 envelope.
3556
355749. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
3558 module.
3559
356050. If a filter generated a file delivery with a non-absolute name (possible if
3561 no home directory exists for the router), the forbid_file option was not
3562 forbidding it.
3563
356451. Added '&' feature to dnslists, to provide bit mask matching in addition to
3565 the existing equality matching.
3566
356752. Exim was using ints instead of ino_t variables in some places where it was
3568 dealing with inode numbers.
3569
357053. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
3571 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
3572 called TMPDIR, and if it finds it is different, it changes its value.
3573
357454. The smtp_printf() function is now made available to local_scan() so
3575 additional output lines can be written before returning. There is also an
3576 smtp_fflush() function to enable the detection of a dropped connection.
3577 The variables smtp_input and smtp_batched_input are exported to
3578 local_scan().
3579
358055. Changed the default runtime configuration: the message "Unknown user"
3581 has been removed from the ACL, and instead placed on the localuser router,
3582 using the cannot_route_message feature. This means that any verification
3583 failures that generate their own messages won't get overridden. Similarly,
3584 the "Unrouteable address" message that was in the ACL for unverifiable
3585 relay addresses has also been removed.
3586
358756. Added hosts_avoid_esmtp to the smtp transport.
3588
358957. The exicyclog script was not checking for the esoteric option
3590 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
3591 will work only if exicyclog is run under the appropriate euid.
3592
359358. Following a discussion on the list, the rules by which Exim recognises line
3594 endings on incoming messages have been changed. The -dropcr and drop_cr
3595 options are now no-ops, retained only for backwards compatibility. The
3596 following line terminators are recognized: LF CRLF CR. However, special
3597 processing applies to CR:
3598
3599 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
3600 nor a local message in the state where . is a terminator.
3601
3602 (ii) If a bare CR is encountered in a header line, an extra space is added
3603 after the line terminator so as not to end the header. The reasoning
3604 behind this is that bare CRs in header lines are most likely either
3605 to be mistakes, or people trying to play silly games.
3606
360759. The size of a message, as listed by "-bp" or in the Exim monitor window,
3608 was being incorrectly given as 18 bytes larger than it should have been.
3609 This is a VOB (very old bug).
3610
361160. This may never have affected anything current, but just in case it has:
3612 When the local host is found other than at the start of a list of hosts,
3613 the local host, those with the same MX, and any that follow, are discarded.
3614 When the list in question was part of a longer list of hosts, the following
3615 hosts (not currently being processed) were also being discarded. This no
3616 longer happens. I'm not sure if this situation could ever has previously
3617 arisen.
3618
361961. Added the "/MX" feature to lists of hosts in the manualroute and query
3620 program routers.
3621
362262. Whenever Exim generates a new message, it now adds an Auto-Submitted:
3623 header. This is something that is recommended in a new Internet Draft, and
3624 is something that is documented as being done by Sendmail. There are two
3625 possible values. For messages generated by the autoreply transport, Exim
3626 adds:
3627
3628 Auto-Submitted: auto-replied
3629
3630 whereas for all other generated messages (e.g. bounces) it adds
3631
3632 Auto-Submitted: auto-generated
3633
363463. The "personal" condition in filters now includes a test for the
3635 Auto-Submitted: header. If it contains the string "auto-" the message it
3636 not considered personal.
3637
363864. Added rcpt_include_affixes as a generic transport option.
3639
364065. Added queue_only_override (default true).
3641
364266. Added the syslog_duplication option.
3643
364467. If what should have been the first header line of a message consisted of
3645 a space followed by a colon, Exim was mis-interpreting it as a header line.
3646 It isn't of course - it is syntactically invalid and should therefore be
3647 treated as the start of the message body. The misbehaviour could have
3648 caused a number of strange effects, including loss of data in subsequent
3649 header lines, and spool format errors.
3650
365168. Formerly, the AUTH parameter on a MAIL command was trusted only if the
3652 client host had authenticated. This control can now be exercised by an ACL
3653 for more flexibility.
3654
365569. By default, callouts do not happen when testing with -bh. There is now a
3656 variant, -bhc, which does actually run the callout code, including
3657 consulting and updating the callout cache.
3658
365970. Added support for saslauthd authentication, courtesy of Alexander
3660 Sabourenkov.
3661
366271. If statvfs() failed on the spool or log directories while checking their
3663 size for availability, Exim confusingly gave the error "space shortage".
3664 Furthermore, in debugging mode it crashed with a floating point exception.
3665 These checks are done if check_{spool,log}_{space,inodes} are set, and when
3666 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
3667 serious problem, Exim now writes to the main and panic logs when this
3668 happens, with details of the failure. It then refuses to accept the
3669 incoming message, giving the message "spool directory problem" or "log
3670 directory problem" with a 421 code for SMTP messages.
3671
367272. When Exim is about to re-exec itself, it ensures that the file descriptors
3673 0, 1, and 2 exist, because some OS complain for execs without them (see
3674 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
3675 descriptors. However, the code omitted to check that the open succeeded,
3676 causing mysterious errors if for some reason the permissions on /dev/null
3677 got screwed. Now Exim writes a message to the main and panic logs, and
3678 bombs out if it can't open /dev/null.
3679
368073. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
3681 interact so that it is all more flexible. It is supposed to remain
3682 backwards compatible. Also added extra_local_interfaces.