Add support for setclassresources() in the pipe transport on FreeBSD,
[exim.git] / doc / doc-txt / ChangeLog
CommitLineData
929ba01c 1$Cambridge: exim/doc/doc-txt/ChangeLog,v 1.309 2006/02/21 16:24:19 ph10 Exp $
495ae4b0
PH
2
3Change log file for Exim from version 4.21
4-------------------------------------------
5
214e2000
PH
6Exim version 4.61
7-----------------
8
9PH/01 The code for finding all the local interface addresses on a FreeBSD
10 system running IPv6 was broken. This may well have applied to all BSD
11 systems, as well as to others that have similar system calls. The broken
12 code found IPv4 interfaces correctly, but gave incorrect values for the
13 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
14 that it would not match correctly against @[] and not recognize the IPv6
15 addresses as local.
16
f9daeae0
PH
17PH/02 The ipliteral router was not recognizing addresses of the form user@
18 [ipv6:....] because it didn't know about the "ipv6:" prefix.
19
7e66e54d
PH
20PH/03 Added disable_ipv6.
21
c8ea1597
PH
22PH/04 Changed $reply_address to use the raw form of the headers instead of the
23 decoded form, because it is most often used to construct To: headers
24 lines in autoreplies, and the decoded form may well be syntactically
25 invalid. However, $reply_address has leading white space removed, and all
26 newlines turned into spaces so that the autoreply transport does not
27 grumble.
28
911f6fde
PH
29PH/05 If group was specified without a user on a router, and no group or user
30 was specified on a transport, the group from the router was ignored.
31
47ca6d6c
PH
32PH/06 Increased the number of ACL variables to 20 of each type, and arranged
33 for visible compile-time settings that can be used to change these
34 numbers, for those that want even more. Backwards compatibility with old
35 spool files has been maintained. However, going back to a previous Exim
36 release will lost any variables that are in spool files.
37
ed0e9820
PH
38PH/07 Two small changes when running in the test harness: increase delay when
39 passing a TCP/IP connection to a new process, in case the original
40 process has to generate a bounce, and remove special handling of
41 127.0.0.2 (sic), which is no longer necessary.
42
eff37e47
PH
43PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
44 be the same on different OS.
45
1921d2ea
PH
46PH/09 Moved a debug statement in filter processing to avoid a race problem when
47 testing.
48
b3f69ca8
JJ
49JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
50 whether --show-vars was specified or not
51
52JJ/02 exipick: Added support for new ACL variable spool format introduced
53 in 4.61-PH/06
54
424a1c63
PH
55PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
56 syntactically invalid From: or Reply-to: line, and a filter used this to
57 generate an autoreply, and therefore failed to obtain an address for the
58 autoreply, Exim could try to deliver to a non-existent relative file
59 name, causing unrelated and misleading errors. What now happens is that
60 it logs this as a hard delivery error, but does not attempt to create a
61 bounce message.
62
7a100415
PH
63PH/11 The exinext utility has a -C option for testing purposes, but although
64 the given file was scanned by exinext itself; it wasn't being passed on
65 when Exim was called.
66
19b9dc85
PH
67PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
68 an end-of-file indication when reading a command response.
69
309bd837
PH
70PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
71 compiled. In many other places in Exim, IPv6 addresses are always
72 recognized, so I have changed this. It also means that IPv4 domain
73 literals of the form [IPV4:n.n.n.n] are now always recognized.
74
59e82a2a
PH
75PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
76 used if the router is not running as root, for example, when verifying at
77 ACL time, or when using -bh. The debugging output from this situation was
78 non-existent - all you got was a failure to exec. I have made two
79 changes:
80
81 (a) Failures to set uid/gid, the current directory, or a process leader
82 in a subprocess such as that created by queryprogram now generate
83 suitable debugging ouput when -d is set.
84
85 (b) The queryprogram router detects when it is not running as root,
86 outputs suitable debugging information if -d is set, and then runs
87 the subprocess without attempting to change uid/gid.
88
9edc04ce
PH
89PH/15 Minor change to Makefile for building test_host (undocumented testing
90 feature).
91
1349e1e5
PH
92PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
93 additional section of a DNS packet that returns MX or SRV records.
94 Instead, it always explicitly searches for A/AAAA records. This avoids
95 major problems that occur when a DNS server includes only records of one
96 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
97 fixed another bug: if SRV records were looked up and the corresponding
98 address records were *not* found in the additional section, the port
99 values from the SRV records were lost.
100
ea49d0e1
PH
101PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
102 using the correct key (the original address) when searching the retry
103 rules in order to find which one to use for generating the retry hint.
104
064a94c9
PH
105PH/18 If quota_warn_message contains a From: header, Exim now refrains from
106 adding the default one. Similarly, if it contains a Reply-To: header, the
107 errors_reply_to option, if set, is not used.
108
727071f8
PH
109PH/19 When calculating a retry time, Exim used to measure the "time since
110 failure" by looking at the "first failed" field in the retry record. Now
111 it does not use this if it is later than than the arrival time of the
112 message. Instead it uses the arrival time. This makes for better
113 behaviour in cases where some deliveries succeed, thus re-setting the
114 "first failed" field. An example is a quota failure for a huge message
115 when small messages continue to be delivered. Without this change, the
116 "time since failure" will always be short, possible causing more frequent
117 delivery attempts for the huge message than are intended.
118
f78eb7c6
PH
119PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
120 $1, $2, $3) because the numerical variables can be reset during some
121 expansion items (e.g. "match"), thereby losing the authentication data.
122
21c28500
PH
123PH/21 Make -bV show the size of off_t variables so that the test suite can
124 decide whether to run tests for quotas > 2G.
125
126PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
127 mailbox_size, and mailbox_filecount in the appendfile transport. If a
128 filecount value is greater than 2G or if a quota value is greater than 2G
129 on a system where the size of off_t is not greater than 4, a panic error
130 is given.
131
1688f43b
PH
132PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
133 never match. The debug and -bh output now contains an explicit error
134 message indicating a malformed IPv4 address or mask.
135
136PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
137 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
138 PH/23 above applies.
139
9675b384
PH
140PH/25 Do not write to syslog when running in the test harness. The only
141 occasion when this arises is a failure to open the main or panic logs
142 (for which there is an explicit test).
143
6a3f1455
PH
144PH/26 Added the /no_tell option to "control=freeze".
145
dac79d3e
PH
146PH/27 If a host name lookup failed very early in a connection, for example, if
147 the IP address matched host_lookup and the reverse lookup yielded a name
148 that did not have a forward lookup, an error message of the form "no IP
149 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
150 could be logged. Now it outputs the IP address instead of "NULL".
1349e1e5 151
5977a0b3
PH
152PH/28 An enabling patch from MH: add new function child_open_exim2() which
153 allows the sender and the authenticated sender to be set when
154 submitting a message from within Exim. Since child_open_exim() is
155 documented for local_scan(), the new function should be too.
156
c91535f3
PH
157PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
158 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
159 results in an empty string is now treated as unset.
160
0d46a8c8
PH
161PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
162
278c6e6c
PH
163PH/31 Added sender_verify_fail logging option.
164
2cbb4081
PH
165PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
166 needed by RFC 822 but not by RFC 2822 was commented out. I have now
167 tidied the source and removed it altogether.
168
3eef829e
PH
169PH/33 When a queue run was abandoned because the load average was too high, a
170 log line was always written; now it is written only if the queue_run log
171 selector is set. In addition, the log line for abandonment now contains
172 information about the queue run such as the pid. This is always present
173 in "start" and "stop" lines but was omitted from the "abandon" line.
174
1ab95fa6
PH
175PH/34 Omit spaces between a header name and the colon in the error message that
176 is given when verify = headers_syntax fails (if there are lots of them,
177 the message gets confusing).
178
230205fc
PH
179PH/35 Change the default for dns_check_names_pattern to allow slashes within
180 names, as there are now some PTR records that contain slashes. This check
181 is only to protect against broken name servers that fall over on strange
182 characters, so the fact that it applies to all lookups doesn't matter.
183
75e0e026
PH
184PH/36 Now that the new test suite is complete, we can remove some of the
185 special code in Exim that was needed for the old test suite. For example,
186 sorting DNS records because real resolvers return them in an arbitrary
187 order. The new test suite's fake resolver always returns records in the
188 same order.
189
190PH/37 When running in the test harness, use -odi for submitted messages (e.g.
191 bounces) except when queue_only is set, to avoid logging races between
192 the different processes.
193
145396a6
PH
194PH/38 Panic-die if .include specifies a non-absolute path.
195
3cd34f13
PH
196PH/39 A tweak to the "H" retry rule from its user.
197
11121d3d
JJ
198JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
199 a label. They prevented compilation on older perls.
200
201JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
202 a warning to be raised on newish perls.
203
204JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
205 on queue. Changes to match documented behaviour of showing count of
206 messages matching specified criteria.
207
8def5aaf
PH
208PH/40 Changed the default ident timeout from 30s to 5s.
209
929ba01c
PH
210PH/41 Added support for the use of login_cap features, on those BSD systems
211 that have them, for controlling the resources used by pipe deliveries.
212
214e2000 213
5de37277
PH
214Exim version 4.60
215-----------------
216
cc38ddbf
PH
217PH/01 Two changes to the default runtime configuration:
218
219 (1) Move the checks for relay_from_hosts and authenticated clients from
220 after to before the (commented out) DNS black list checks.
221
222 (2) Add control=submission to the relay_from_hosts and authenticated
223 clients checks, on the grounds that messages accepted by these
224 statements are most likely to be submissions.
5de37277 225
72fdd6ae
PH
226PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
227
228 (1) Generate an error if the third argument for the ${prvs expansion is
229 not a single digit.
230
231 (2) Treat a missing third argument of ${prvscheck as if it were an empty
232 string.
233
234 (3) Reset the variables that are obtained from the first argument of
235 ${prvscheck and used in the second argument before leaving the code,
236 because their memory is reclaimed, so using them afterwards may do
237 silly things.
238
239 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
240 one (it's much easier than Tom thought :-).
241
242 (5) Because of (4), we can now allow for the use of $prvscheck_result
243 inside the third argument.
cb9328de 244
cb741023
PH
245PH/03 For some reason, the default setting of PATH when running a command from
246 a pipe transport was just "/usr/bin". I have changed it to
247 "/bin:/usr/bin".
248
f174f16e
PH
249PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
250 anything to be listed in the output from -bV.
b2f5a032 251
c25242d7
PH
252PH/05 When a filter generated an autoreply, the entire To: header line was
253 quoted in the delivery log line, like this:
254
255 => >A.N.Other <ano@some.domain> <original@ddress> ...
256
257 This has been changed so that it extracts the operative address. There
258 may be more than one such address. If so, they are comma-separated, like
259 this:
260
261 => >ano@some.domain,ona@other.domain <original@ddress> ...
262
82c19f95
PH
263PH/06 When a client host used a correct literal IP address in a HELO or EHLO
264 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
265 not being looked up in the rDNS to get a host name, Exim was showing the
266 IP address twice in Received: lines, even though the IP addresses were
267 identical. For example:
268
269 Received: from [1.2.3.4] (helo=[1.2.3.4])
270
271 However, if the real host name was known, it was omitting the HELO data
272 if it matched the actual IP address. This has been tidied up so that it
273 doesn't show the same IP address twice.
274
d7ffbc12
PH
275PH/07 When both +timestamp and +memory debugging was on, the value given by
276 $tod_xxx expansions could be wrong, because the tod_stamp() function was
277 called by the debug printing, thereby overwriting the timestamp buffer.
278 Debugging no longer uses the tod_stamp() function when +timestamp is set.
279
9f526266
PH
280PH/08 When the original message was included in an autoreply transport, it
281 always said "this is a copy of the message, including all the headers",
282 even if body_only or headers_only was set. It now gives an appropriate
283 message.
284
87fcc8b9
PH
285PH/09 Applied a patch from the Sieve maintainer which:
286
287 o fixes some comments
288 o adds the (disabled) notify extension core
289 o adds some debug output for the result of if/elsif tests
290 o points to the current vacation draft in the documentation
291 and documents the missing references header update
292
293 and most important:
294
295 o fixes a bug in processing the envelope test (when testing
296 multiple envelope elements, the last element determinted the
297 result)
298
456682f5
PH
299PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
300 Electronic Mail") by including:
301
302 Auto-submitted: auto-generated
303
304 in the messages that it generates (bounce messages and others, such as
305 warnings). In the case of bounce messages for non-SMTP mesages, there was
306 also a typo: it was using "Auto_submitted" (underscore instead of
307 hyphen). Since every message generated by Exim is necessarily in response
308 to another message, thes have all been changed to:
309
310 Auto-Submitted: auto-replied
311
312 in accordance with these statements in the RFC:
313
314 The auto-replied keyword:
315
316 - SHOULD be used on messages sent in direct response to another
317 message by an automatic process,
318
319 - MUST NOT be used on manually-generated messages,
320
321 - MAY be used on Delivery Status Notifications (DSNs) and Message
322 Disposition Notifications (MDNs),
323
324 - MUST NOT be used on messages generated by automatic or periodic
325 processes, except for messages which are automatic responses to
326 other messages.
327
3e46c1aa
PH
328PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
329 to the default Received: header definition.
456682f5 330
49826d12
PH
331PH/12 Added log selector acl_warn_skipped (default on).
332
eba0c039
PH
333PH/13 After a successful wildlsearch lookup, discard the values of numeric
334 variables because (a) they are in the wrong storage pool and (b) even if
335 they were copied, it wouldn't work properly because of the caching.
336
a0d6ba8a
PH
337PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
338 checking when decoding. Apparently there are clients that generate
339 overlong encoded strings. Why am I not surprised?
340
f0917727
PH
341PH/15 If the first argument of "${if match_address" was not empty, but did not
342 contain an "@" character, Exim crashed. Now it writes a panic log message
343 and treats the condition as false.
344
096fee00
PH
345PH/16 In autoreply, treat an empty string for "once" the same as unset.
346
024bd3c2
PH
347PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
348 extension "envelope-auth". The code is finished and in agreement with
349 other implementations, but there is no documentation so far and in fact,
350 nobody wrote the draft yet. This extension is currently #undef'ed, thus
351 not changing the active code.
352
353 Print executed "if" and "elsif" statements when debugging is used. This
354 helps a great deal to understand what a filter does.
355
356 Document more things not specified clearly in RFC3028. I had all this
357 sorted out, when out of a sudden new issues came to my mind. Oops."
358
df199fec
PH
359PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
360 (Bugzilla #53).
361
d27f1df3
PH
362PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
363 canonical form (as documented). However, after a host name lookup from
364 the IP address, check_host() was doing a simple string comparison with
365 addresses acquired from the DNS when checking that the found name did
366 have the original IP as one of its addresses. Since any found IPv6
367 addresses are likely to be in abbreviated form, the comparison could
368 fail. Luckily, there already exists a function for doing the comparison
369 by converting both addresses to binary, so now that is used instead of
370 the text comparison.
371
96776534
PH
372PH/20 There was another similar case to PH/19, when a complete host name was
373 given in a host list; looking up its IP address could give an abbreviated
374 form, whereas the current host's name might or might not be abbreviated.
375 The same fix has been applied.
376
5de37277 377
9a799bc0
PH
378Exim version 4.54
379-----------------
380
381PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
382 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
383 It now does.
384
99a4b039
PH
385PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
386 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
387
388PH/03 Typo: missing ".o" in src/pcre/Makefile.
389
4b233853
PH
390PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
391 header line, restrict the check to what is listed in RFCs 2369 and 2929.
392 Also, for "Auto-Submitted", treat anything other than "no" as
393 non-personal, in accordance with RFC 3834. (Previously it treated
394 anything starting "auto-" as non-personal.)
395
8857ccfd
PH
396TF/01 The control=submission/name=... option had a problem with syntax
397 errors if the name included a slash character. The /name= option
398 now slurps the rest of the string, so it can include any characters
399 but it must come last in the list of options (after /sender_retain
400 or /domain=).
401
433a2980
PH
402PH/05 Some modifications to the interface to the fake nameserver for the new
403 testing suite.
404
3e46c1aa 405
9a799bc0 406
e3a311ba
TK
407Exim version 4.53
408-----------------
409
410TK/01 Added the "success_on_redirect" address verification option. See
411 NewStuff for rationale and an example.
412
13b685f9
PH
413PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
414
395ff96d
PH
415PH/02 Patch to exigrep to allow it to work on syslog lines.
416
5b68f6e4
PH
417PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
418 fread() to skip over the body file's header line, because in Cygwin the
419 header line is locked and is inaccessible.
420
1ab52c69
PH
421PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
422 co-exist for some time) to make it clear that it is the Exim ID that is
423 referenced, not the Message-ID: header line.
424
b07e6aa3
PH
425PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
426 string_format() function, because snprintf() does not exist on all
427 operating systems.
428
254e032f
PH
429PH/06 The use of forbid_filter_existstest now also locks out the use of the
430 ${stat: expansion item.
431
3af76a81
PH
432PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
433 protocol synchronization error", to keep the pedants happy.
434
2548ba04
PH
435PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
436 well as for IRIX systems, when gcc is being used. See the host.c source
437 file for comments.
438
b6c6011d
PH
439PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
440
cf39cf57
PH
441PH/10 Named domain lists were not working if used in a queue_smtp_domains
442 setting.
443
f1513293
PH
444PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
445 transport and to the smtp transport in LMTP mode.
446
727549a4
PH
447TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
448
af46795e
PH
449PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
450 run a filter in a subprocess. This could lead to confusion in subsequent
451 lookups in the parent process. There should also be a search_tidyup() at
452 the end of the subprocess.
453
d7b47fd0
PH
454PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
455 only if the host matched helo_try_verify_hosts, which caused the
456 verification to occur when the EHLO/HELO command was issued. The ACL just
457 tested the remembered result. Now, if a previous verification attempt has
458 not happened, "verify = helo" does it there and then.
459
ee744174
JJ
460JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
461
b582ab87
PH
462TK/03 Fix log output including CR from clamd.
463
41a13e0a
PH
464PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
465 exist provoked a memory error which could cause a segfault.
466
f625cc5a
PH
467PH/15 Installed PCRE 6.2
468
469PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
470
21f7af35
PH
471PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
472 of the problem. Specifically, suggested +O2 rather than +O1 for the
473 HP-UX compiler.
474
31480e42
PH
475PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
476
2d280592
PH
477PH/20 If a delivery was routed to a non-standard port by means of an SRV
478 record, the port was not correctly logged when the outgoing_port log
479 selector was set (it logged the transort's default port).
480
7cd1141b
PH
481PH/21 Added support for host-specific ports to manualroute, queryprogram,
482 fallback_hosts, and "hosts" in the smtp transport.
483
484PH/22 If the log selector "outgoing_port" is set, the port is now also given on
485 host errors such as "Connection refused".
486
750af86e
PH
487PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
488 authentication with radiusclient 0.4.9:
489
490 - Error returned from rc_read_config was caught wrongly
491 - Username/password not passed on to radius server due to wrong length.
492
493 The presumption is that some radiusclient API changes for 4.51/PH/17
494 were not taken care of correctly. The code is still untested by me (my
495 Linux distribution still has 0.3.2 of radiusclient), but it was
496 contributed by a Radius user.
497
498PH/24 When doing a callout, the value of $domain wasn't set correctly when
499 expanding the "port" option of the smtp transport.
500
4304270b
TK
501TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
502 while reading a MIME header. Thanks to Tom Hughes for a patch.
503
750af86e
PH
504PH/24 Include config.h inside local_scan.h so that configuration settings are
505 available.
506
64ffc24f
PH
507PH/25 Make $smtp_command_argument available after all SMTP commands. This means
508 that in an ACL for RCPT (for example), you can examine exactly what was
509 received.
510
5dd9625b
PH
511PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
512 commands, but it was not correctly comparing the address with the actual
513 client host address. Thus, it would show the EHLO address in Received:
514 header lines when this was not necessary.
515
5591031b
PH
516PH/27 Added the % operator to ${eval:}.
517
ba18e66a
PH
518PH/28 Exim tries to create and chdir to its spool directory when it starts;
519 it should be ignoring failures (because with -C, for example, it has lost
520 privilege). It wasn't ignoring creation failures other than "already
521 exists".
522
9cec981f
PH
523PH/29 Added "crypteq" to the list of supported features that Exim outputs when
524 -bV or -d is used.
525
aa2b5c79
PH
526PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
527 because an input line was too long, either on its own, or by virtue of
1509d3a8
PH
528 too many continuations, the temporary file was not being removed, and the
529 return code was incorrect.
aa2b5c79 530
48a53b7f
PH
531PH/31 Missing "BOOL" in function definition in filtertest.c.
532
1c59d63b
PH
533PH/32 Applied Sieve patches from the maintainer.
534
671012da
TK
535TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
536
1509d3a8
PH
537PH/33 Added "verify = not_blind".
538
539PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
540 Local/Makefile (with some defaults set). These are used in built scripts
541 such as exicyclog, but they have never been used in the exim_install
542 script (though there are many overriding facilities there). I have
543 arranged that the exim_install script now takes note of these two
544 settings.
545
546PH/35 Installed configuration files for Dragonfly.
547
2fe1a124
PH
548PH/36 When a locally submitted message by a trusted user did not contain a
549 From: header, and the sender address was obtained from -f or from an SMTP
550 MAIL command, and the trusted user did not use -F to supply a sender
551 name, $originator_name was incorrectly used when constructing a From:
552 header. Furthermore, $originator_name was used for submission mode
553 messages from external hosts without From: headers in a similar way,
554 which is clearly wrong.
555
8800895a
PH
556PH/37 Added control=suppress_local_fixups.
557
ccfdb010
PH
558PH/38 When log_selector = +received_sender was set, and the addition of the
559 sender made the log line's construction buffer exactly full, or one byte
560 less than full, an overflow happened when the terminating "\n" was
561 subsequently added.
562
1130bfb0
PH
563PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
564 when the result of a list match is failure because a DNS lookup failed.
565
ebcb507f
PH
566PH/40 RM_COMMAND is now used in the building process.
567
c35e155c
PH
568PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
569 the "build-* directories that it finds.
570
95d1f782
PH
571PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
572 address in a domain literal was a prefix of an interface address.
573
fd6de02e
PH
574PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
575 when verifying a sender address, unless rewrite_headers is false.
576
58de37c5
PH
577PH/44 Wrote a long comment about why errors_to addresses are verified as
578 recipients, not senders.
579
261cf466
TF
580TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
581 the ratelimit ACL was added.
582
3ee512ff
PH
583PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
584
e08c430f
PH
585PH/46 Added extra information about PostgreSQL errors to the error string.
586
bef5a11f
PH
587PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
588 avoiding the need to install special zones in a real server. This is
589 backwards compatible; if it can't find the fake resolver, it drops back.
590 Thus, both old and new test suites can be run.
591
7546de58
TF
592TF/02 Added util/ratelimit.pl
593
e5d5a95f
TF
594TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
595 clock is set back in time.
596
2e88a017
TF
597TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
598 Candler <B.Candler@pobox.com>.
599
a5f65aa4
TF
600TF/05 The fix for PH/43 was not completely correct; widen_domains is always
601 OK for addresses that are the result of redirections.
602
e7726cbf
PH
603PH/48 A number of further additions for the benefit of the new test suite,
604 including a fake gethostbyname() that interfaces to the fake DNS resolver
605 (see PH/47 above).
606
a7fdad5b
TF
607TF/06 The fix for widen_domains has also been applied to qualify_single and
608 search_parents which are the other dnslookup options that can cause
609 header rewrites.
610
6af56900
PH
611PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
612 type ("H").
613
0925ede6
PH
614PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
615
66afa403
TF
616TF/07 Exim produced the error message "an SRV record indicated no SMTP
617 service" if it encountered an MX record with an empty target hostname.
618 The message is now "an MX or SRV record indicated no SMTP service".
619
0154e85a
TF
620TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
621 if the DNS of the sending site is misconfigured. This is quite a
622 common situation. This change restores the behaviour of treating a
623 helo verification defer as a failure.
624
16f12c76
PH
625PH/51 If self=fail was set on a router, the bounce message did not include the
626 actual error message.
627
bbe902f0 628
e5a9dba6
PH
629Exim version 4.52
630-----------------
631
632TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
633
22c3b60b
PH
634PH/01 When a transport filter timed out in a pipe delivery, and the pipe
635 command itself ended in error, the underlying message about the transport
636 filter timeout was being overwritten with the pipe command error. Now the
637 underlying error message should be appended to the second error message.
638
06a9b4b5
PH
639TK/01 Fix poll() being unavailable on Mac OSX 10.2.
640
c1ac6996
PH
641PH/02 Reduce the amount of output that "make" produces by default. Full output
642 can still be requested.
643
9c7a242c
PH
644PH/03 The warning log line about a condition test deferring for a "warn" verb
645 was being output only once per connection, rather than after each
646 occurrence (because it was using the same function as for successful
647 "warn" verbs). This seems wrong, so I have changed it.
648
87ba3f5f
PH
649TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
650 it should not have, which might have caused a crash in the right
651 circumstances, but probably never did.
652
653PH/04 Installed a modified version of Tony Finch's patch to make submission
654 mode fix the return path as well as the Sender: header line, and to
655 add a /name= option so that you can make the user's friendly name appear
656 in the header line.
657
29aba418
TF
658TF/03 Added the control = fakedefer ACL modifier.
659
fe0dab11
TF
660TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
661 Mark Lowes for thorough testing.
870f6ba8 662
11d337a4
TK
663TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
664
665TK/03 Merged latest SRS patch from Miles Wilton.
666
415c8f3b
PH
667PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
668 with the definition in sysexits.h (which is #included earlier).
669 Fortunately, Exim does not actually use EX_OK. The code used to try to
670 preserve the sysexits.h value, by assumimg that macro definitions were
671 scanned for macro replacements. I have been disabused of this notion,
672 so now the code just undefines EX_OK before #including unistd.h.
11d337a4 673
958541e9
PH
674PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
675 in the smtp transport. When a block could not be written in a single
676 write() function, the timeout was being re-applied to each part-write.
677 This seems wrong - if the receiver was accepting one byte at a time it
678 would take for ever. The timeout is now adjusted when this happens. It
679 doesn't have to be particularly precise.
680
c206415f
TK
681TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
682 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
683
2a4be8f9
PH
684PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
685 without a domain if the check to <postmaster@domain> fails.
686
1cba11c5
SC
687SC/01 Eximstats: added -xls and the ability to specify output files
688 (patch written by Frank Heydlauf).
689
690SC/02 Eximstats: use FileHandles for outputing results.
691
692SC/03 Eximstats: allow any combination of xls, txt, and html output.
693
694SC/04 Eximstats: fixed display of large numbers with -nvr option
695
696SC/05 Eximstats: fixed merging of reports with empty tables.
697
698SC/06 Eximstats: added the -include_original_destination flag
699
700SC/07 Eximstats: removed tabs and trailing whitespace.
701
1005d00e
TK
702TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
703
704TK/06 MBOX spool code: Add real "From " MBOX separator line
705 so the .eml file is really in mbox format (even though
706 most programs do not really care). Patch from Alex Miller.
707
708TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
709 The latter is generated from $received_to and is only set if the
710 message has one envelope recipient. SA can use these headers,
711 obviously out-of-the-box. Patch from Alex Miller.
712
9b4768fa
PH
713PH/08 The ${def test on a variable was returning false if the variable's
714 value was "0", contrary to what the specification has always said!
715 The result should be true unless the variable is empty.
716
717PH/09 The syntax error of a character other than { following "${if
718 def:variable_name" (after optional whitespace) was not being diagnosed.
719 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
720 accidental colon was present, for example, could give incorrect results.
721
0d7eb84a
PH
722PH/10 Tidied the code in a number of places where the st_size field of a stat()
723 result is used (not including appendfile, where other changes are about
724 to be made).
725
726PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
727 This involved changing a lot of size variables from int to off_t. It
728 should work with maildirs and everything.
729
40727bee
TK
730TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
731 spamd dying while we are connected to it.
732
554d2369
TF
733TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
734 <jgh@wizmail.org>
735
1f922db1
PH
736PH/12 Applied Alex Kiernan's patch for the API change for the error callback
737 function for BDB 4.3.
738
ef213c3b
PH
739PH/13 Changed auto_thaw such that it does not apply to bounce messages.
740
8ac170f3
PH
741PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
742 the sources for PCRE have been re-arranged and more files are now
743 involved.
744
b1c749bb
PH
745PH/15 The code I had for printing potentially long long variables in PH/11
746 above was not the best (it lost precision). The length of off_t variables
747 is now inspected at build time, and an appropriate printing format (%ld
c6c2dc1d
PH
748 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
749 to be "long long int" or "long int". This is needed for the internal
750 formatting function string_vformat().
b1c749bb 751
4aac9b49
PH
752PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
753 the configuration file to be ":syslog", then the script "guesses" where
754 the logs files are, rather than using the compiled in default. In our
755 case the guess is not the same as the compiled default, so the script
756 suddenly stopped working when I started to use syslog. The patch checks
757 to see if log_file_path is "". If so, it attempts to read it from exim
758 with no configuration file to get the compiled in version, before it
759 falls back to the previous guessing code."
760
294520c8
TK
761TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
762 implementing BATV in an Exim configuration. See NewStuff for the gory
763 details.
764
5bd022fe
PH
765PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
766 Makefile that are specific to HP-UX.
767
90e9ce59
PH
768PH/18 If the "use_postmaster" option was set for a recipient callout together
769 with the "random" option, the postmaster address was used as the MAIL
770 FROM address for the random test, but not for the subsequent recipient
771 test. It is now used for both.
772
5ea81592
PH
773PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
774 patch removes a few documentation additions to RFC 3028, because the
775 latest draft now contains them. It adds the new en;ascii-case comparator
776 and a new error check for 8bit text in MIME parts. Comparator and
777 require names are now matched exactly. I enabled the subaddress
778 extension, but it is not well tested yet (read: it works for me)."
779
c6c2dc1d
PH
780PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
781 rework some of the code of TK/09 above to avoid the hardwired use of
782 "%lld" and "long long". Replaced the call to snprintf() with a call to
783 string_vformat().
784
fffffe4c
PH
785PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
786 records point to non-existent hosts", "retry timeout exceeded", and
787 "retry time not reached for any host after a long failure period".
ca02eafb 788
9a26b6b2
PH
789PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
790 experimental DomainKeys support:
791
792 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
793 (2) On an error such as an illegally used "control", the wrong name for
794 the control was given.
795
796 These problems did NOT occur unless DomainKeys support was compiled.
797
4aee0225
PH
798PH/23 Added daemon_startup_retries and daemon_startup_sleep.
799
32d668a5
PH
800PH/24 Added ${if match_ip condition.
801
8187c3f3
PH
802PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
803 databases so that it will be absolutely obvious if a crash occurs in the
804 DB library. This is a regular occurrence (often caused by mis-matched
805 db.h files).
806
ff790e47 807PH/26 Insert a lot of missing (void) casts for functions such as chown(),
f1e894f3
PH
808 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
809 picked up on a user's system that detects such things. There doesn't seem
810 to be a gcc warning option for this - only an attribute that has to be
811 put on the function's prototype. It seems that in Fedora Core 4 they have
812 set this on a number of new functions. No doubt there will be more in due
813 course.
ff790e47 814
5417f6d1
PH
815PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
816 specify a transport. However, if an address that was verified by such a
817 router was the subject of a callout, Exim crashed because it tried to
818 read the rcpt_include_affixes from the non-existent transport. Now it
819 just assumes that the setting of that option is false. This bug was
820 introduced by 4.51/PH/31.
821
59cf8544
PH
822PH/28 Changed -d+all to exclude +memory, because that information is very
823 rarely of interest, but it makes the output a lot bigger. People tend to
824 do -d+all out of habit.
825
e7ad8a65
PH
826PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
827 code in os-type was giving problems when libc.so lives in lib64, like on
828 x86_64 Fedora Core.
829
ade42478
PH
830PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
831 aren't the modern standard, and it seems that some systems' include files
832 don't always have them. Exim was already checking for some of the newer
833 ones like T_AAAA, and defining it itself. I've added checks for all the
834 record types that Exim uses.
835
182ad5cf
PH
836PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
837 not automatically generating a new one, as it is supposed to. This
838 prevented TLS from working. If the file did exist, but contained invalid
839 data, a new version was generated, as expected. It was only the case of a
840 non-existent file that was broken.
841
b0d9fc80
TK
842TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
843 with a change in libdomainkeys > 0.64.
844
845TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
846 from DNS. If the selector record carries the flag, it now has
847 precedence over the domain-wide flag.
848
849TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
850
47c7a64a
PH
851PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
852 the use of a port name that isn't defined in /etc/services) occurred, the
853 message was deferred as in a normal delivery, and thus remained on the
854 spool, instead of being failed because of the mua_wrapper setting. This
855 is now fixed, and I tidied up some of the mua_wrapper messages at the
856 same time.
857
a388bce4
SC
858SC/08 Eximstats: whilst parsing the mainlog(s), store information about
859 the messages in a hash of arrays rather than using individual hashes.
860 This is a bit cleaner and results in dramatic memory savings, albeit
861 at a slight CPU cost.
862
863SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
864 as requested by Marc Sherman.
865
866SC/10 Eximstats: added histograms for user specified patterns as requested
867 by Marc Sherman.
868
0793e4ed
SC
869SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
870
c58b88df
PH
871PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
872 fopen() in the content-scanning modules that did not already have it.
873
e7ad8a65 874
7982096b
PH
875Exim version 4.51
876-----------------
877
1a46a8c5
PH
878TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
879 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
880
2f079f46 881TK/02 Fix ACL "control" statement not being available in MIME ACL.
1a46a8c5
PH
882
883TK/03 Fix ACL "regex" condition not being available in MIME ACL.
884
885PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
886 to test Sieve filters that use "vacation".
887
888PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
889 that changes the way the GnuTLS parameters are stored in the cache file.
890 The new format can be generated externally. For backward compatibility,
891 if the data in the cache doesn't make sense, Exim assumes it has read an
892 old-format file, and it generates new data and writes a new file. This
893 means that you can't go back to an older release without removing the
894 file.
895
896PH/03 A redirect router that has both "unseen" and "one_time" set does not
897 work if there are any delivery delays because "one_time" forces the
898 parent to be marked "delivered", so its unseen clone is never tried
899 again. For this reason, Exim now forbids the simultaneous setting of
900 these two options.
901
902PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
903 redirected to themselves ("homonym" addresses). Read the long ChangeLog
904 entry if you want to know the details. The fix, however, neglected to
905 consider the case when local delivery batching is involved. The test for
906 "previously delivered" was not happening when checking to see if an
907 address could be batched with a previous (undelivered) one; under
908 certain circumstances this could lead to multiple deliveries to the same
c2c19e9d 909 address.
1a46a8c5
PH
910
911PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
912 in its include files, and this causes problems building Exim.
913
914PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
915 header_syntax) but Exim was just ignoring anything given after a slash.
916 In particular, this caused confusion with an attempt to use "verify =
917 reverse_host_lookup/defer_ok". An error is now given when options are
918 supplied for verify items that do not have them. (Maybe reverse_host_
919 lookup should have a defer_ok option, but that's a different point.)
920
921PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
922 defined by RFC 821) to 2048, because there were problems with some AUTH
923 commands, and RFC 1869 says the size should be increased for extended
924 SMTP commands that take arguments.
925
926PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
927 Finch).
928
929PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
930 "unknown" error; now it says that the functionality isn't in the binary.
8d67ada3 931
49c2d5ea
PH
932PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
933 an address' error message when a string expansion fails (syntax or
f331f3b6
PH
934 whatever). Otherwise the password may appear in the log. Following change
935 PH/42 below, there is no longer a chance of it appearing in a bounce
936 message.
49c2d5ea 937
bf759a8b
PH
938PH/11 Installed exipick version 20050225.0 from John Jetmore.
939
83364d30
PH
940PH/12 If the last host in a fallback_hosts list was multihomed, only the first
941 of its addresses was ever tried. (Bugzilla bug #2.)
942
7999bbd7
PH
943PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
944 the result incorrectly in the debug output. (It correctly added a newline
945 to what was transported.)
946
7dbf77c9
PH
947TF/01 Added $received_time.
948
74e0617f
PH
949PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
950 commented out examples of how to interface to a virus scanner and to
951 SpamAssassin. Also added commented examples of av_scanner and
952 spamd_address settings.
953
2f079f46
PH
954PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
955 and controls are allowed in which ACLs. There were a couple of minor
956 errors. Some of the entries in the conditions table (which is a table of
957 where they are NOT allowed) were getting very unwieldy; rewrote them as a
958 negation of where the condition IS allowed.
959
8c841523
PH
960PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
961
7766a4f0
PH
962PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
963 header file does not have a version number, so I've had to invent a new
964 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
965 API. The code is untested by me (my Linux distribution still has 0.3.2 of
966 radiusclient), but it was contributed by a Radius user.
967
8b417f2c
PH
968PH/18 Installed Lars Mainka's patch for the support of CRL collections in
969 files or directories, for OpenSSL.
970
901f42cb
PH
971PH/19 When an Exim process that is running as root has to create an Exim log
972 file, it does so in a subprocess that runs as exim:exim so as to get the
973 ownership right at creation (otherwise, other Exim processes might see
974 the file with the wrong ownership). There was no test for failure of this
975 fork() call, which would lead to the process getting stuck as it waited
976 for a non-existent subprocess. Forks do occasionally fail when resources
977 run out. I reviewed all the other calls to fork(); they all seem to check
978 for failure.
979
f9b9210e
PH
980PH/20 When checking for unexpected SMTP input at connect time (before writing
981 the banner), Exim was not dealing correctly with a non-positive return
982 from the read() function. If the client had disconnected by this time,
983 the result was a log entry for a synchronization error with an empty
984 string after "input=" when read() returned zero. If read() returned -1
985 (an event I could not check), uninitialized data bytes were printed.
986 There were reports of junk text (parts of files, etc) appearing after
987 "input=".
988
54cdb463
PH
989PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
990
cf00dad6
PH
991PH/22 Added support for macro redefinition, and (re)definition in between
992 driver and ACL definitions.
993
acb1b346
PH
994PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
995 forgetting to use the resulting value; it was using the unexpanded value.
996
c5ddb310
PH
997PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
998 hadn't been configured. The fix is from Juergen Kreileder, who
999 understands it better than I do:
1000
1001 "Here's what I see happening with three configured cyrus_sasl
1002 authenticators configured (plain, login, cram-md5):
1003
1004 On startup auth_cyrus_sasl_init() gets called for each of these.
1005 This means three calls to sasl_listmech() without a specified mech_list.
1006 => SASL tests which mechs of all available mechs actually work
1007 => three warnings about OTP not working
1008 => the returned list contains: plain, login, cram-md5, digest-md5, ...
1009
1010 With the patch, sasl_listmech() also gets called three times. But now
1011 SASL's mech_list option is set to the server_mech specified in the the
1012 authenticator. Or in other words, the answer from sasl_listmech()
1013 gets limited to just the mech you're testing for (which is different
1014 for each call.)
1015 => the return list contains just 'plain' or 'login', 'cram-md5' or
1016 nothing depending on the value of ob->server_mech.
1017
1018 I've just tested the patch: Authentication still works fine,
1019 unavailable mechs specified in the exim configuration are still
1020 caught, and the auth.log warnings about OTP are gone."
1021
31619da6
PH
1022PH/25 When debugging is enabled, the contents of the command line are added
1023 to the debugging output, even when log_selector=+arguments is not
1024 specified.
1025
bebaf0fc
PH
1026PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
1027 answer is "GNU", and only if the return is "GNU/something" is the answer
1028 "Linux".
1029
475fe28a
PH
1030PH/27 $acl_verify_message is now set immediately after the failure of a
1031 verification in an ACL, and so is available in subsequent modifiers. In
1032 particular, the message can be preserved by coding like this:
1033
1034 warn !verify = sender
1035 set acl_m0 = $acl_verify_message
1036
1037 Previously, $acl_verify_message was set only while expanding "message"
1038 and "log_message" when a very denied access.
1039
7e8bec7a
PH
1040PH/28 Modified OS/os.c-Linux with
1041
1042 -#ifndef OS_LOAD_AVERAGE
1043 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
1044
1045 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
1046 nomenclature these days.)
1047
e4a89c47
PH
1048PH/29 Installed patch from the Sieve maintainer that adds the options
1049 sieve_useraddress and sieve_subaddress to the redirect router.
1050
5ca2a9a1
PH
1051PH/30 In these circumstances:
1052 . Two addresses routed to the same list of hosts;
1053 . First host does not offer TLS;
1054 . First host accepts first address;
1055 . First host gives temporary error to second address;
1056 . Second host offers TLS and a TLS session is established;
1057 . Second host accepts second address.
1058 Exim incorrectly logged both deliveries with the TLS parameters (cipher
1059 and peerdn, if requested) that were in fact used only for the second
1060 address.
7e8bec7a 1061
c688b954
PH
1062PH/31 When doing a callout as part of verifying an address, Exim was not paying
1063 attention to any local part prefix or suffix that was matched by the
1064 router that accepted the address. It now behaves in the same way as it
1065 does for delivery: the affixes are removed from the local part unless
1066 rcpt_include_affixes is set on the transport.
1067
fed77020
PH
1068PH/32 Add the sender address, as F=<...>, to the log line when logging a
1069 timeout during the DATA phase of an incoming message.
1070
7fe1560f
PH
1071PH/33 Sieve envelope tests were broken for match types other than :is. I have
1072 applied a patch sanctioned by the Sieve maintainer.
c688b954 1073
ebb6e6d5
PH
1074PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
1075 the uid or gid is negative. A case of a negative gid caused this to be
1076 noticed. The fix allows for either to be negative.
1077
9c4e8f60
PH
1078PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
1079 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
1080 overlooked.
1081
1082PH/36 The change PH/12 above was broken. Fixed it.
1083
d7174846
PH
1084PH/37 Exim used to check for duplicate addresses in the middle of routing, on
1085 the grounds that routing the same address twice would always produce the
1086 same answer. This might have been true once, but it is certainly no
1087 longer true now. Routing a child address may depend on the previous
1088 routing that produced that child. Some complicated redirection strategies
1089 went wrong when messages had multiple recipients, and made Exim's
1090 behaviour dependent on the order in which the addresses were given.
1091
1092 I have moved the duplicate checking until after the routing is complete.
1093 Exim scans the addresses that are assigned to local and remote
1094 transports, and removes any duplicates. This means that more work will be
1095 done, as duplicates will always all be routed, but duplicates are
1096 presumably rare, so I don't expect this is of any significance.
1097
1098 For deliveries to pipes, files, and autoreplies, the duplicate checking
1099 still happens during the routing process, since they are not going to be
1100 routed further.
1101
cfe75fc3
PH
1102PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
1103 It corrects a timeout issue with spamd. This is Ian's comment: "The
1104 background is that sometimes spamd either never reads data from a
1105 connection it has accepted, or it never writes response data. The exiscan
1106 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
1107 blindly assumes that writes won't block so it may never time out."
1108
be22d70e
PH
1109PH/39 Allow G after quota size as well as K and M.
1110
0612b098
PH
1111PH/40 The value set for $authenticated_id in an authenticator may not contain
1112 binary zeroes or newlines because the value is written to log lines and
1113 to spool files. There was no check on this. Now the value is run through
1114 the string_printing() function so that such characters are converted to
1115 printable escape sequences.
1116
2e0c1448
PH
1117PH/41 $message_linecount is a new variable that contains the total number of
1118 lines in the message. Compare $body_linecount, which is the count for the
1119 body only.
1120
447d236c
PH
1121PH/42 Exim no longer gives details of delivery errors for specific addresses in
1122 bounce and delay warning messages, except in certain special cases, which
1123 are as follows:
1124
1125 (a) An SMTP error message from a remote host;
1126 (b) A message specified in a :fail: redirection;
1127 (c) A message specified in a "fail" command in a system filter;
1128 (d) A message specified in a FAIL return from the queryprogram router;
1129 (e) A message specified by the cannot_route_message router option.
1130
1131 In these cases only, Exim does include the error details in bounce and
1132 warning messages. There are also a few cases where bland messages such
1133 as "unrouteable address" or "local delivery error" are given.
1134
d20976dc
PH
1135PH/43 $value is now also set for the "else" part of a ${run expansion.
1136
f656d135
PH
1137PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
1138 being worked on, but at least Exim now implements the latest version to
1139 play with."
1140
2e2a30b4
PH
1141PH/45 In a pipe transport, although a timeout while waiting for the pipe
1142 process to complete was treated as a delivery failure, a timeout while
1143 writing the message to the pipe was logged, but erroneously treated as a
1144 successful delivery. Such timeouts include transport filter timeouts. For
1145 consistency with the overall process timeout, these timeouts are now
1146 treated as errors, giving rise to delivery failures by default. However,
1147 there is now a new Boolean option for the pipe transport called
1148 timeout_defer, which, if set TRUE, converts the failures into defers for
1149 both kinds of timeout. A transport filter timeout is now identified in
1150 the log output.
1151
9176e9f0
PH
1152PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
1153 systems where "make" and "gmake" are different, calling "gmake" at top
1154 level broke things. I've arranged for the value of $(MAKE) to be passed
1155 from the Makefile to this script so that it can call the same version of
1156 "make".
1157
7982096b 1158
bbe902f0
PH
1159A note about Exim versions 4.44 and 4.50
1160----------------------------------------
1161
1162Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
1163changes of various kinds. As a consequence, a big documentation update was
1164needed. This delayed the release for rather longer than seemed good, especially
1165in the light of a couple of (minor) security issues. Therefore, the changes
1166that fixed bugs were backported into 4.43, to create a 4.44 maintenance
1167release. So 4.44 and 4.50 are in effect two different branches that both start
1168from 4.43.
1169
1170I have left the 4.50 change log unchanged; it contains all the changes since
11714.43. The change log for 4.44 is below; many of its items are identical to
1172those for 4.50. This seems to be the most sensible way to preserve the
1173historical information.
1174
1175
f7b63901 1176Exim version 4.50
495ae4b0
PH
1177-----------------
1178
5fe762f6
PH
1179 1. Minor wording change to the doc/README.SIEVE file.
1180
139059f6 1181 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
5fe762f6 1182 computation of the current number of files was incorrect.
495ae4b0 1183
7086e875
PH
1184 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
1185 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
1186 place.
1187
35af9f61
PH
1188 4. Give more explanation in the error message when the command for a transport
1189 filter fails to execute.
1190
b668c215
PH
1191 5. There are several places where Exim runs a non-Exim command in a
1192 subprocess. The SIGUSR1 signal should be disabled for these processes. This
1193 was being done only for the command run by the queryprogram router. It is
1194 now done for all such subprocesses. The other cases are: ${run, transport
1195 filters, and the commands run by the lmtp and pipe transports.
1196
a494b1e1
PH
1197 6. Added CONFIGURE_GROUP build-time option.
1198
1199 7. Some older OS have a limit of 256 on the maximum number of file
1200 descriptors. Exim was using setrlimit() to set 1000 as a large value
1201 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
1202 systems. I've change it so that if it can't get 1000, it tries for 256.
35edf2ff 1203
c5fcb476
PH
1204 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
1205 was an oversight, and furthermore, ever since the addition of extra
1206 controls (e.g. 4.43/32), the checks on when to allow different forms of
1207 "control" were broken. There should now be diagnostics for all cases when a
1208 control that does not make sense is encountered.
1209
69358f02
PH
1210 9. Added the /retain_sender option to "control=submission".
1211
5be20824
PH
121210. $recipients is now available in the predata ACL (oversight).
1213
eb2c0248
PH
121411. Tidy the search cache before the fork to do a delivery from a message
1215 received from the command line. Otherwise the child will trigger a lookup
1216 failure and thereby defer the delivery if it tries to use (for example) a
1217 cached ldap connection that the parent has called unbind on.
1218
2a3eea10
PH
121912. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
1220 of $address_data from the recipient verification was clobbered by the
1221 sender verification.
1222
122313. The value of address_data from a sender verification is now available in
1224 $sender_address_data in subsequent conditions in the ACL statement.
1225
23c7ff99
PH
122614. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
1227
4deaf07d
PH
122815. Added a new option "connect=<time>" to callout options, to set a different
1229 connection timeout.
1230
926e1192
PH
123116. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
1232 was its contents. (It was OK if the option was not defined at all.)
1233
650edc6f
PH
123417. A "Completed" log line is now written for messages that are removed from
1235 the spool by the -Mrm option.
1236
2c7db3f5
PH
123718. New variables $sender_verify_failure and $recipient_verify_failure contain
1238 information about exactly what failed.
1239
3d235903
PH
124019. Added -dd to debug only the daemon process.
1241
7c7ad977
PH
124220. Incorporated Michael Haardt's patch to ldap.c for improving the way it
1243 handles timeouts, both on the server side and network timeouts. Renamed the
1244 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
1245
981756db
PH
124621. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
1247 It is now set to "smtps".
1248
d4eb88df
PH
124922. $host_address is now set to the target address during the checking of
1250 ignore_target_hosts.
1251
125223. When checking ignore_target_hosts for an ipliteral router, no host name was
1253 being passed; this would have caused $sender_host_name to have been used if
1254 matching the list had actually called for a host name (not very likely,
1255 since this list is usually IP addresses). A host name is now passed as
1256 "[x.x.x.x]".
1257
7d468ab8
PH
125824. Changed the calls that set up the SIGCHLD handler in the daemon to use the
1259 code that specifies a non-restarting handler (typically sigaction() in
1260 modern systems) in an attempt to fix a rare and obscure crash bug.
1261
126225. Narrowed the window for a race in the daemon that could cause it to ignore
1263 SIGCHLD signals. This is not a major problem, because they are used only to
1264 wake it up if nothing else does.
1265
62c0818f
PH
126626. A malformed maildirsize file could cause Exim to calculate negative values
1267 for the mailbox size or file count. Odd effects could occur as a result.
1268 The maildirsize information is now recalculated if the size or filecount
1269 end up negative.
1270
26034054
PH
127127. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
1272 support for a long time. Removed HAVE_SYS_VFS_H.
1273
af66f652
PH
127428. Installed the latest version of exipick from John Jetmore.
1275
90af77f4
PH
127629. In an address list, if the pattern was not a regular expression, an empty
1277 subject address (from a bounce message) matched only if the pattern was an
1278 empty string. Non-empty patterns were not even tested. This was the wrong
1279 because it is perfectly reasonable to use an empty address as part of a
1280 database query. An empty address is now tested by patterns that are
1281 lookups. However, all the other forms of pattern expect the subject to
1282 contain a local part and a domain, and therefore, for them, an empty
1283 address still always fails if the pattern is not itself empty.
1284
d8ef3577
PH
128530. Exim went into a mad DNS loop when attempting to do a callout where the
1286 host was specified on an smtp transport, and looking it up yielded more
1287 than one IP address.
1288
5cb8cbc6
PH
128931. Re-factored the code for checking spool and log partition space into a
1290 function that finds that data and another that does the check. The former
1291 is then used to implement four new variables: $spool_space, $log_space,
1292 $spool_inodes, and $log_inodes.
1293
14702f5b
PH
129432. The RFC2047 encoding function was originally intended for short strings
1295 such as real names; it was not keeping to the 75-character limit for
1296 encoded words that the RFC imposes. It now respects the limit, and
1297 generates multiple encoded words if necessary. To be on the safe side, I
1298 have increased the buffer size for the ${rfc2047: expansion operator from
1299 1024 to 2048 bytes.
1300
063b1e99
PH
130133. It is now permitted to omit both strings after an "if" condition; if the
1302 condition is true, the result is "true". As before, when the second string
1303 is omitted, a false condition yields an empty string. This makes it less
1304 cumbersome to write custom ACL and router conditions.
1305
652e1b65
PH
130634. Failure to deliver a bounce message always caused it to be frozen, even if
1307 there was an errors_to setting on the router. The errors_to setting is now
1308 respected.
1309
6f0c9a4f
PH
131035. If an IPv6 address is given for -bh or -bhc, it is now converted to the
1311 canonical form (fully expanded) before being placed in
1312 $sender_host_address.
1313
33397d19
PH
131436. The table in the code that translates DNS record types into text (T_A to
1315 "A" for instance) was missing entries for NS and CNAME. It is just possible
1316 that this could have caused confusion if both these types were looked up
1317 for the same domain, because the text type is used as part of Exim's
1318 per-process caching. But the chance of anyone hitting this buglet seems
1319 very small.
1320
7bb56e1f
PH
132137. The dnsdb lookup has been extended in a number of ways.
1322
1323 (1) There is a new type, "zns", which walks up the domain tree until it
1324 finds some nameserver records. It should be used with care.
1325
ea3bc19b
PH
1326 (2) There is a new type, "mxh", which is like "mx" except that it returns
1327 just the host names, not the priorities.
1328
1329 (3) It is now possible to give a list of domains (or IP addresses) to be
ff4dbb19
PH
1330 looked up. The behaviour when one of the lookups defers can be
1331 controlled by a keyword.
7bb56e1f 1332
ea3bc19b 1333 (4) It is now possible to specify the separator character for use when
7bb56e1f 1334 multiple records are returned.
33397d19 1335
0bcb2a0e
PH
133638. The dnslists ACL condition has been extended: it is now possible to supply
1337 a list of IP addresses and/or domains to be looked up in a particular DNS
1338 domain.
1339
2ac0e484
PH
134039. Added log_selector=+queue_time_overall.
1341
4e1fde53
PH
134240. When running the queue in the test harness, wait just a tad after forking a
1343 delivery process, to get repeatability of debugging output.
1344
de365ded
PH
134541. Include certificate and key file names in error message when GnuTLS fails
1346 to set them up, because the GnuTLS error message doesn't include the name
1347 of the failing file when there is a problem reading it.
1348
f05da2e8
PH
134942. Allow both -bf and -bF in the same test run.
1350
d6453af2
PH
135143. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
1352
f7b63901
PH
135344. The "Exiscan patch" is now merged into the mainline Exim source.
1354
135545. Sometimes the final signoff response after QUIT could fail to get
1356 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
1357 < 0 before doing a fflush(). This bug looks as though it goes back to the
1358 introduction of TLS in release 3.20, but "sometimes" must have been rare
1359 because the tests only now provoked it.
1360
a444213a
PH
136146. Reset the locale to "C" after calling embedded Perl, in case it was changed
1362 (this can affect the format of dates).
1363
0ec020ea
PH
136447. exim_tidydb, when checking for the continued existence of a message for
1365 which it has found a message-specific retry record, was not finding
1366 messages that were in split spool directories. Consequently, it was
1367 deleting retry records that should have stayed in existence.
1368
b1206957
PH
136948. Steve fixed some bugs in eximstats.
1370
137149. The SPA authentication driver was not abandoning authentication and moving
1372 on to the next authenticator when an expansion was forced to fail,
1373 contradicting the general specification for all authenticators. Instead it
1374 was generating a temporary error. It now behaves as specified.
1375
26dd5a95
PH
137650. The default ordering of permitted cipher suites for GnuTLS was pessimal
1377 (the order specifies the preference for clients). The order is now AES256,
1378 AES128, 3DES, ARCFOUR128.
1379
343b2385
PH
138051. Small patch to Sieve code - explicitly set From: when generating an
1381 autoreply.
1382
1c5466b9
PH
138352. Exim crashed if a remote delivery caused a very long error message to be
1384 recorded - for instance if somebody sent an entire SpamAssassin report back
1385 as a large number of 550 error lines. This bug was coincidentally fixed by
1386 increasing the size of one of Exim's internal buffers (big_buffer) that
1387 happened as part of the Exiscan merge. However, to be on the safe side, I
1388 have made the code more robust (and fixed the comments that describe what
1389 is going on).
1390
55ee9ee3
PH
139153. Now that there can be additional text after "Completed" in log lines (if
1392 the queue_time_overall log selector is set), a one-byte patch to exigrep
1393 was needed to allow it to recognize "Completed" as not the last thing in
1394 the line.
1395
d38f8232
PH
139654. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
1397 patch that reportedly fixes this has been added. I am not expert enough to
1398 create a test for it. This is what the patch creator wrote:
1399
1400 "I found a little strange behaviour of ldap code when working with
1401 Windows 2003 AD Domain, where users was placed in more than one
1402 Organization Units. When I tried to give exim partial DN, the exit code
1403 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
1404 But simultaneously result of request was absolutely normal ldap result,
1405 so I produce this patch..."
1406
3295e65b
PH
1407 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
1408 REFERENCE, so I have modified the code to exclude the patch when that macro
1409 is not defined.
1410
7102e136
PH
141155. Some experimental protocols are using DNS PTR records for new purposes. The
1412 keys for these records are domain names, not reversed IP addresses. The
b975ba52
PH
1413 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
1414 leaves it alone. Component reversal etc. now happens only for IP addresses.
ea3a6f44 1415 CAN-2005-0021
7102e136 1416
3ca0ba97
PH
141756. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
1418
c2bcbe20
PH
141957. Double the size of the debug message buffer (to 2048) so that more of very
1420 long debug lines gets shown.
1421
18ce445d
PH
142258. The exicyclog utility now does better if the number of log files to keep
1423 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
1424
1f5b4c3d
PH
142559. Two changes related to the smtp_active_hostname option:
1426
1427 (1) $smtp_active_hostname is now available as a variable.
1428 (2) The default for smtp_banner uses $smtp_active_hostname instead
1429 of $primary_hostname.
1430
b975ba52
PH
143160. The host_aton() function is supposed to be passed a string that is known
1432 to be a valid IP address. However, in the case of IPv6 addresses, it was
1433 not checking this. This is a hostage to fortune. Exim now panics and dies
1434 if the condition is not met. A case was found where this could be provoked
85b87bc2
PH
1435 from a dnsdb PTR lookup with an IPv6 address that had more than 8
1436 components; fortuitously, this particular loophole had already been fixed
1437 by change 4.50/55 above.
1438
1439 If there are any other similar loopholes, the new check in host_aton()
1440 itself should stop them being exploited. The report I received stated that
1441 data on the command line could provoke the exploit when Exim was running as
1442 exim, but did not say which command line option was involved. All I could
1443 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
1444 running as the user.
ea3a6f44 1445 CAN-2005-0021
85b87bc2
PH
1446
144761. There was a buffer overflow vulnerability in the SPA authentication code
1448 (which came originally from the Samba project). I have added a test to the
1449 spa_base64_to_bits() function which I hope fixes it.
ea3a6f44 1450 CAN-2005-0022
b975ba52 1451
17ffcae7
PH
145262. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
1453 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
1454
d95f9fdb
PH
145563. The daemon start-up calls getloadavg() while still root for those OS that
1456 need the first call to be done as root, but it missed one case: when
1457 deliver_queue_load_max is set with deliver_drop_privilege. This is
1458 necessary for the benefit of the queue runner, because there is no re-exec
1459 when deliver_drop_privilege is set.
1460
86b8287f
PH
146164. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
1462 This has been fixed.
1463
60dc5e56
PH
146465. Caching of lookup data for "hosts =" ACL conditions, when a named host list
1465 was in use, was not putting the data itself into the right store pool;
1466 consequently, it could be overwritten for a subsequent message in the same
1467 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
1468 the caching.)
1469
533244af
PH
147066. Added hosts_max_try_hardlimit to the smtp transport, default 50.
1471
a5a28604
PH
147267. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
1473 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
1474 the function were treating the return as a boolean value, which happened to
1475 work because 0=false and not-0=true, but is not correct code.
1476
7e634d24
PH
147768. The host_aton() function was not handling scoped IPv6 addresses (those
1478 with, for example, "%eth0" on the end) correctly.
1479
3e11c26b
PH
148069. Fixed some compiler warnings in acl.c for the bitmaps specified with
1481 negated items (that is, ~something) in unsigned ints. Some compilers
1482 apparently mutter when there is no cast.
1483
6729cf78
PH
148470. If an address verification called from an ACL failed, and did not produce a
1485 user-specific message (i.e. there was only a "system" message), nothing was
1486 put in $acl_verify_message. In this situation, it now puts the system
1487 message there.
1488
00f00ca5
PH
148971. Change 4.23/11 added synchronization checking at the start of an SMTP
1490 session; change 4.31/43 added the unwanted input to the log line - except
1491 that it did not do this in the start of session case. It now does.
1492
c9bdd01c
PH
149372. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
1494 This is wrong and can cause the other end to generate a synchronization
1495 error if it is another Exim or anything else that does the synchronization
1496 check. A QUIT command is no longer sent after a timeout.
1497
d43194df
PH
149873. $host_lookup_deferred has been added, to make it easier to detect DEFERs
1499 during host lookups.
1500
fe5b5d0b
PH
150174. The defer_ok option of callout verification was not working if it was used
1502 when verifying addresses in header lines, that is, for this case:
1503
1504 verify = header_sender/callout=defer_ok
1505
76a2d7ba
PH
150675. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
1507 those file descriptors could be used for SMTP connections. If anything
1508 wrote to stderr (the example that came up was "warn" in embedded Perl), it
1509 could be sent to the SMTP client, causing chaos. The daemon now opens
1510 stdin, stdout, and stderr to /dev/null when it puts itself into the
1511 background.
1512
151376. Arrange for output from Perl's "warn" command to be written to Exim's main
1514 log by default. The user can override this with suitable Perl magic.
1515
04f7d5b9
PH
151677. The use of log_message on a "discard" ACL verb, which is supposed to add to
1517 the log message when discard triggers, was not working for the DATA ACL or
1518 for the non-SMTP ACL.
1519
bc60667e
PH
152078. Error message wording change in sieve.c.
1521
bb6e88ff
PH
152279. If smtp_accept_max_per_host was set, the number of connections could be
1523 restricted to fewer than expected, because the daemon was trying to set up
1524 a new connection before checking whether the processes handling previous
1525 connections had finished. The check for completed processes is now done
1526 earlier. On busy systems, this bug wouldn't be noticed because something
1527 else would have woken the daemon, and it would have reaped the completed
1528 process earlier.
1529
1e70f85b
PH
153080. If a message was submitted locally by a user whose login name contained one
1531 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
1532 It caused a spool format error. I have fixed the spool reading code. A
1533 related problem was that the "from" clause in the Received: line became
1534 illegal because of the space(s). It is now covered by ${quote_local_part.
1535
153681. Included the latest eximstats from Steve (adds average sizes to HTML Top
1537 tables).
1538
4e01f9d6
PH
153982. Updated OS/Makefile-AIX as per message from Mike Meredith.
1540
1ee1cef2
PH
154183. Patch from Sieve maintainer to fix unterminated string problem in
1542 "vacation" handling.
1543
6e2b4ccc
PH
154484. Some minor changes to the Linux configuration files to help with other
1545 OS variants using glibc.
1546
8e669ac1
PH
154785. One more patch for Sieve to update vacation handling to latest spec.
1548
495ae4b0 1549
bbe902f0
PH
1550----------------------------------------------------
1551See the note above about the 4.44 and 4.50 releases.
1552----------------------------------------------------
1553
1554
1555Exim version 4.44
1556-----------------
1557
1558 1. Change 4.43/35 introduced a bug that caused file counts to be
1559 incorrectly computed when quota_filecount was set in an appendfile
1560 transport
1561
1562 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
1563 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
1564 place.
1565
1566 3. Give more explanation in the error message when the command for a transport
1567 filter fails to execute.
1568
1569 4. There are several places where Exim runs a non-Exim command in a
1570 subprocess. The SIGUSR1 signal should be disabled for these processes. This
1571 was being done only for the command run by the queryprogram router. It is
1572 now done for all such subprocesses. The other cases are: ${run, transport
1573 filters, and the commands run by the lmtp and pipe transports.
1574
1575 5. Some older OS have a limit of 256 on the maximum number of file
1576 descriptors. Exim was using setrlimit() to set 1000 as a large value
1577 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
1578 systems. I've change it so that if it can't get 1000, it tries for 256.
1579
1580 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
1581 was an oversight, and furthermore, ever since the addition of extra
1582 controls (e.g. 4.43/32), the checks on when to allow different forms of
1583 "control" were broken. There should now be diagnostics for all cases when a
1584 control that does not make sense is encountered.
1585
1586 7. $recipients is now available in the predata ACL (oversight).
1587
1588 8. Tidy the search cache before the fork to do a delivery from a message
1589 received from the command line. Otherwise the child will trigger a lookup
1590 failure and thereby defer the delivery if it tries to use (for example) a
1591 cached ldap connection that the parent has called unbind on.
1592
1593 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
1594 of $address_data from the recipient verification was clobbered by the
1595 sender verification.
1596
159710. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
1598 was its contents. (It was OK if the option was not defined at all.)
1599
160011. A "Completed" log line is now written for messages that are removed from
1601 the spool by the -Mrm option.
1602
160312. $host_address is now set to the target address during the checking of
1604 ignore_target_hosts.
1605
160613. When checking ignore_target_hosts for an ipliteral router, no host name was
1607 being passed; this would have caused $sender_host_name to have been used if
1608 matching the list had actually called for a host name (not very likely,
1609 since this list is usually IP addresses). A host name is now passed as
1610 "[x.x.x.x]".
1611
161214. Changed the calls that set up the SIGCHLD handler in the daemon to use the
1613 code that specifies a non-restarting handler (typically sigaction() in
1614 modern systems) in an attempt to fix a rare and obscure crash bug.
1615
161615. Narrowed the window for a race in the daemon that could cause it to ignore
1617 SIGCHLD signals. This is not a major problem, because they are used only to
1618 wake it up if nothing else does.
1619
162016. A malformed maildirsize file could cause Exim to calculate negative values
1621 for the mailbox size or file count. Odd effects could occur as a result.
1622 The maildirsize information is now recalculated if the size or filecount
1623 end up negative.
1624
162517. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
1626 support for a long time. Removed HAVE_SYS_VFS_H.
1627
ea3a6f44 162818. Updated exipick to current release from John Jetmore.
bbe902f0
PH
1629
163019. Allow an empty sender to be matched against a lookup in an address list.
1631 Previously the only cases considered were a regular expression, or an
1632 empty pattern.
1633
163420. Exim went into a mad DNS lookup loop when doing a callout where the
1635 host was specified on the transport, if the DNS lookup yielded more than
1636 one IP address.
1637
ea3a6f44
NM
163821. The RFC2047 encoding function was originally intended for short strings
1639 such as real names; it was not keeping to the 75-character limit for
1640 encoded words that the RFC imposes. It now respects the limit, and
1641 generates multiple encoded words if necessary. To be on the safe side, I
1642 have increased the buffer size for the ${rfc2047: expansion operator from
1643 1024 to 2048 bytes.
bbe902f0 1644
ea3a6f44
NM
164522. Failure to deliver a bounce message always caused it to be frozen, even if
1646 there was an errors_to setting on the router. The errors_to setting is now
1647 respected.
bbe902f0
PH
1648
164923. If an IPv6 address is given for -bh or -bhc, it is now converted to the
1650 canonical form (fully expanded) before being placed in
1651 $sender_host_address.
1652
165324. Updated eximstats to version 1.33
1654
ea3a6f44
NM
165525. Include certificate and key file names in error message when GnuTLS fails
1656 to set them up, because the GnuTLS error message doesn't include the name
1657 of the failing file when there is a problem reading it.
bbe902f0
PH
1658
165926. Expand error message when OpenSSL has problems setting up cert/key files.
ea3a6f44 1660 As per change 25.
bbe902f0 1661
ea3a6f44
NM
166227. Reset the locale to "C" after calling embedded Perl, in case it was changed
1663 (this can affect the format of dates).
bbe902f0 1664
ea3a6f44
NM
166528. exim_tidydb, when checking for the continued existence of a message for
1666 which it has found a message-specific retry record, was not finding
1667 messages that were in split spool directories. Consequently, it was
1668 deleting retry records that should have stayed in existence.
bbe902f0
PH
1669
167029. eximstats updated to version 1.35
1671 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
1672 1.35 - bugfix such that pie charts by volume are generated correctly
1673
ea3a6f44
NM
167430. The SPA authentication driver was not abandoning authentication and moving
1675 on to the next authenticator when an expansion was forced to fail,
1676 contradicting the general specification for all authenticators. Instead it
1677 was generating a temporary error. It now behaves as specified.
bbe902f0 1678
ea3a6f44
NM
167931. The default ordering of permitted cipher suites for GnuTLS was pessimal
1680 (the order specifies the preference for clients). The order is now AES256,
1681 AES128, 3DES, ARCFOUR128.
bbe902f0 1682
ea3a6f44
NM
168331. Small patch to Sieve code - explicitly set From: when generating an
1684 autoreply.
bbe902f0 1685
ea3a6f44
NM
168632. Exim crashed if a remote delivery caused a very long error message to be
1687 recorded - for instance if somebody sent an entire SpamAssassin report back
1688 as a large number of 550 error lines. This bug was coincidentally fixed by
1689 increasing the size of one of Exim's internal buffers (big_buffer) that
1690 happened as part of the Exiscan merge. However, to be on the safe side, I
1691 have made the code more robust (and fixed the comments that describe what
1692 is going on).
bbe902f0 1693
ea3a6f44
NM
169433. Some experimental protocols are using DNS PTR records for new purposes. The
1695 keys for these records are domain names, not reversed IP addresses. The
1696 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
1697 leaves it alone. Component reversal etc. now happens only for IP addresses.
bbe902f0
PH
1698 CAN-2005-0021
1699
ea3a6f44
NM
170034. The host_aton() function is supposed to be passed a string that is known
1701 to be a valid IP address. However, in the case of IPv6 addresses, it was
1702 not checking this. This is a hostage to fortune. Exim now panics and dies
1703 if the condition is not met. A case was found where this could be provoked
1704 from a dnsdb PTR lookup with an IPv6 address that had more than 8
1705 components; fortuitously, this particular loophole had already been fixed
1706 by change 4.50/55 or 4.44/33 above.
1707
1708 If there are any other similar loopholes, the new check in host_aton()
1709 itself should stop them being exploited. The report I received stated that
1710 data on the command line could provoke the exploit when Exim was running as
1711 exim, but did not say which command line option was involved. All I could
1712 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
1713 running as the user.
bbe902f0
PH
1714 CAN-2005-0021
1715
ea3a6f44
NM
171635. There was a buffer overflow vulnerability in the SPA authentication code
1717 (which came originally from the Samba project). I have added a test to the
1718 spa_base64_to_bits() function which I hope fixes it.
bbe902f0
PH
1719 CAN-2005-0022
1720
ea3a6f44
NM
172136. The daemon start-up calls getloadavg() while still root for those OS that
1722 need the first call to be done as root, but it missed one case: when
1723 deliver_queue_load_max is set with deliver_drop_privilege. This is
1724 necessary for the benefit of the queue runner, because there is no re-exec
1725 when deliver_drop_privilege is set.
bbe902f0 1726
ea3a6f44
NM
172737. Caching of lookup data for "hosts =" ACL conditions, when a named host list
1728 was in use, was not putting the data itself into the right store pool;
1729 consequently, it could be overwritten for a subsequent message in the same
1730 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
1731 the caching.)
bbe902f0 1732
ea3a6f44
NM
173338. Sometimes the final signoff response after QUIT could fail to get
1734 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
1735 < 0 before doing a fflush(). This bug looks as though it goes back to the
1736 introduction of TLS in release 3.20, but "sometimes" must have been rare
1737 because the tests only now provoked it.
bbe902f0
PH
1738
1739
495ae4b0
PH
1740Exim version 4.43
1741-----------------
1742
1743 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
1744 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
1745 Now it returns an integer. A number of calls were still expecting a T/F
1746 return. Fortuitously, in all cases, the tests worked in OK situations,
1747 which is the norm. However, things would have gone wrong on any write
1748 failures on the smtp file descriptor. This function is used when sending
1749 messages over SMTP and also when doing verify callouts.
1750
1751 2. When Exim is called to do synchronous delivery of a locally submitted
1752 message (the -odf or -odi options), it no longer closes stderr before doing
1753 the delivery.
1754
1755 3. Implemented the mua_wrapper option.
1756
1757 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
1758
1759 5. Implemented the functions header_remove(), header_testname(),
1760 header_add_at_position(), and receive_remove_recipient(), and exported them
1761 to local_scan().
1762
1763 6. If an ACL "warn" statement specified the addition of headers, Exim already
1764 inserted X-ACL-Warn: at the start if there was no header name. However, it
1765 was not making this test for the second and subsequent header lines if
1766 there were newlines in the string. This meant that an invalid header could
1767 be inserted if Exim was badly configured.
1768
1769 7. Allow an ACL "warn" statement to add header lines at the start or after all
1770 the Received: headers, as well as at the end.
1771
1772 8. Added the rcpt_4xx retry error code.
1773
1774 9. Added postmaster_mailfrom=xxx to callout verification option.
1775
177610. Added mailfrom=xxxx to the callout verification option, for verify=
1777 header_sender only.
1778
177911. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
1780 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
1781
178212. Inserted some casts to stop certain compilers warning when using pointer
1783 differences as field lengths or precisions in printf-type calls (mostly
1784 affecting debugging statements).
1785
178613. Added optional readline() support for -be (dynamically loaded).
1787
178814. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
1789 same clock tick as a message's arrival, so that its received time was the
1790 same as the "first fail" time on the retry record, and that message
1791 remained on the queue past the ultimate address timeout, every queue runner
1792 would try a delivery (because it was past the ultimate address timeout) but
1793 after another failure, the ultimate address timeout, which should have then
1794 bounced the address, did not kick in. This was a "< instead of <=" error;
1795 in most cases the first failure would have been in the next clock tick
1796 after the received time, and all would be well.
1797
179815. The special items beginning with @ in domain lists (e.g. @mx_any) were not
1799 being recognized when the domain list was tested by the match_domain
1800 condition in an expansion string.
1801
180216. Added the ${str2b64: operator.
1803
180417. Exim was always calling setrlimit() to set a large limit for the number of
1805 processes, without checking whether the existing limit was already
1806 adequate. (It did check for the limit on file descriptors.) Furthermore,
1807 errors from getrlimit() and setrlimit() were being ignored. Now they are
1808 logged to the main and panic logs, but Exim does carry on, to try to do its
1809 job under whatever limits there are.
1810
181118. Imported PCRE 5.0.
1812
181319. Trivial typo in log message " temporarily refused connection" (the leading
1814 space).
1815
181620. If the log selector return_path_on_delivery was set and an address was
1817 redirected to /dev/null, the delivery process crashed because it assumed
1818 that a return path would always be set for a "successful" delivery. In this
1819 case, the whole delivery is bypassed as an optimization, and therefore no
1820 return path is set.
1821
182221. Internal re-arrangement: the function for sending a challenge and reading
1823 a response while authentication was assuming a zero-terminated challenge
1824 string. It's now changed to take a pointer and a length, to allow for
1825 binary data in such strings.
1826
182722. Added the cyrus_sasl authenticator (code supplied by MBM).
1828
182923. Exim was not respecting finduser_retries when seeking the login of the
1830 uid under which it was called; it was always trying 10 times. (The default
1831 setting of finduser_retries is zero.) Also, it was sleeping after the final
1832 failure, which is pointless.
1833
183424. Implemented tls_on_connect_ports.
1835
183625. Implemented acl_smtp_predata.
1837
183826. If the domain in control=submission is set empty, Exim assumes that the
1839 authenticated id is a complete email address when it generates From: or
1840 Sender: header lines.
1841
184227. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
1843 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
1844 chown and chgrp in /bin and hostname in /usr/bin.
1845
184628. Exim was keeping the "process log" file open after each use, just as it
1847 does for the main log. This opens the possibility of it remaining open for
1848 long periods when the USR1 signal hits a daemon. Occasional processlog
1849 errors were reported, that could have been caused by this. Anyway, it seems
1850 much more sensible not to leave this file open at all, so that is what now
1851 happens.
1852
185329. The long-running daemon process does not normally write to the log once it
1854 has entered its main loop, and it closes the log before doing so. This is
1855 so that log files can straightforwardly be renamed and moved. However,
1856 there are a couple of unusual error situations where the daemon does write
1857 log entries, and I had neglected to close the log afterwards.
1858
185930. The text of an SMTP error response that was received during a remote
1860 delivery was being truncated at 512 bytes. This is too short for some of
1861 the long messages that one sometimes sees. I've increased the limit to
1862 1024.
1863
186431. It is now possible to make retry rules that apply only when a message has a
1865 specific sender, in particular, an empty sender.
1866
186732. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
1868 it possible to be selective about when SMTP synchronization is enforced.
1869
187033. Added "control = caseful_local_part" and "control = "caselower_local_part".
1871
187232. Implemented hosts_connection_nolog.
1873
187433. Added an ACL for QUIT.
1875
187634. Setting "delay_warning=" to disable warnings was not working; it gave a
1877 syntax error.
1878
187935. Added mailbox_size and mailbox_filecount to appendfile.
1880
188136. Added control = no_multiline_responses to ACLs.
1882
188337. There was a bug in the logic of the code that waits for the clock to tick
1884 in the case where the clock went backwards by a substantial amount such
1885 that the microsecond fraction of "now" was more than the microsecond
1886 fraction of "then" (but the whole seconds number was less).
1887
188838. Added support for the libradius Radius client library this is found on
1889 FreeBSD (previously only the radiusclient library was supported).
1890
1891
1892Exim version 4.42
1893-----------------
1894
1895 1. When certain lookups returned multiple values in the form name=value, the
1896 quoting of the values was not always being done properly. Specifically:
1897 (a) If the value started with a double quote, but contained no whitespace,
1898 it was not quoted.
1899 (b) If the value contained whitespace other than a space character (i.e.
1900 tabs or newlines or carriage returns) it was not quoted.
1901 This fix has been applied to the mysql and pgsql lookups by writing a
1902 separate quoting function and calling it from the lookup code. The fix
1903 should probably also be applied to nisplus, ibase and oracle lookups, but
1904 since I cannot test any of those, I have not disturbed their existing code.
1905
1906 2. A hit in the callout cache for a specific address caused a log line with no
1907 reason for rejecting RCPT. Now it says "Previous (cached) callout
1908 verification failure".
1909
1910 3. There was an off-by-one bug in the queryprogram router. An over-long
1911 return line was truncated at 256 instead of 255 characters, thereby
1912 overflowing its buffer with the terminating zero. As well as fixing this, I
1913 have increased the buffer size to 1024 (and made a note to document this).
1914
1915 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
1916 when Exim is waiting for an SMTP response from a remote server, Exim
1917 restarts its select() call on the socket, thereby resetting its timeout.
1918 This is not a problem when such interrupts are rare. Somebody set up a cron
1919 job to run exiwhat every 2 minutes, which is less than the normal select()
1920 timeout (5 or 10 minutes). This meant that the select() timeout never
1921 kicked in because it was always reset. I have fixed this by comparing the
1922 time when an interrupt arrives with the time at the start of the first call
1923 to select(). If more time than the timeout has elapsed, the interrupt is
1924 treated as a timeout.
1925
1926 5. Some internal re-factoring in preparation for the addition of Sieve
1927 extensions (by MH). In particular, the "personal" test is moved to a
1928 separate function, and given an option for scanning Cc: and Bcc: (which is
1929 not set for Exim filters).
1930
1931 6. When Exim created an email address using the login of the caller as the
1932 local part (e.g. when creating a From: or Sender: header line), it was not
1933 quoting the local part when it contained special characters such as @.
1934
1935 7. Installed new OpenBSD configuration files.
1936
1937 8. Reworded some messages for syntax errors in "and" and "or" conditions to
1938 try to make them clearer.
1939
1940 9. Callout options, other than the timeout value, were being ignored when
1941 verifying sender addresses in header lines. For example, when using
1942
1943 verify = header_sender/callout=no_cache
1944
1945 the cache was (incorrectly) being used.
1946
194710. Added a missing instance of ${EXE} to the exim_install script; this affects
1948 only the Cygwin environment.
1949
195011. When return_path_on_delivery was set as a log selector, if different remote
1951 addresses in the same message used different return paths and parallel
1952 remote delivery occurred, the wrong values would sometimes be logged.
1953 (Whenever a remote delivery process finished, the return path value from
1954 the most recently started remote delivery process was logged.)
1955
195612. RFC 3848 specifies standard names for the "with" phrase in Received: header
1957 lines when AUTH and/or TLS are in use. This is the "received protocol"
1958 field. Exim used to use "asmtp" for authenticated SMTP, without any
1959 indication (in the protocol name) for TLS use. Now it follows the RFC and
1960 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
1961 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
1962 names appear in log lines as well as in Received: header lines.
1963
196413. Installed MH's patches for Sieve to add the "copy" and "vacation"
1965 extensions, and comparison tests, and to fix some bugs.
1966
196714. Changes to the "personal" filter test:
1968
1969 (1) The test was buggy in that it was just doing the equivalent of
1970 "contains" tests on header lines. For example, if a user's address was
1971 anne@some.where, the "personal" test would incorrectly be true for
1972
1973 To: susanne@some.where
1974
1975 This test is now done by extracting each address from the header in turn,
1976 and checking the entire address. Other tests that are part of "personal"
1977 are now done using regular expressions (for example, to check local parts
1978 of addresses in From: header lines).
1979
1980 (2) The list of non-personal local parts in From: addresses has been
1981 extended to include "listserv", "majordomo", "*-request", and "owner-*",
1982 taken from the Sieve specification recommendations.
1983
1984 (3) If the message contains any header line starting with "List-" it is
1985 treated as non-personal.
1986
1987 (4) The test for "circular" in the Subject: header line has been removed
1988 because it now seems ill-conceived.
1989
199015. Minor typos in src/EDITME comments corrected.
1991
199216. Installed latest exipick from John Jetmore.
1993
199417. If headers_add on a router specified a text string that was too long for
1995 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
1996 of string_sprintf() is now avoided.
1997
199818. $message_body_size was not set (it was always zero) when running the DATA
1999 ACL and the local_scan() function.
2000
200119. For the "mail" command in an Exim filter, no default was being set for
2002 the once_repeat time, causing a random time value to be used if "once" was
2003 specified. (If the value happened to be <= 0, no repeat happened.) The
2004 default is now 0s, meaning "never repeat". The "vacation" command was OK
2005 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
2006 (I found it when inspecting the code).
2007
200820. There is now an overall timeout for performing a callout verification. It
2009 defaults to 4 times the callout timeout, which applies to individual SMTP
2010 commands during the callout. The overall timeout applies when there is more
2011 than one host that can be tried. The timeout is checked before trying the
2012 next host. This prevents very long delays if there are a large number of
2013 hosts and all are timing out (e.g. when the network connections are timing
2014 out). The value of the overall timeout can be changed by specifying an
2015 additional sub-option for "callout", called "maxwait". For example:
2016
2017 verify = sender/callout=5s,maxwait=20s
2018
201921. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
2020 to the end before writing, but this should make it even safer).
2021
202222. Exim was forgetting that it had advertised PIPELINING for the second and
2023 subsequent messages on an SMTP connection. It was also not resetting its
2024 memory on STARTTLS and an internal HELO.
2025
202623. When Exim logs an SMTP synchronization error within a session, it now
2027 records whether PIPELINING has been advertised or not.
2028
202924. Added 3 instances of "(long int)" casts to time_t variables that were being
2030 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
2031 rather than long int.
2032
203325. Installed the latest Cygwin configuration files from the Cygwin maintainer.
2034
203526. Added the never_mail option to autoreply.
2036
2037
2038Exim version 4.41
2039-----------------
2040
2041 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
2042 crash if the getsockname() call failed; this can happen if a connection is
2043 closed very soon after it is established. The problem was simply in the
2044 order in which certain operations were done, causing Exim to try to write
2045 to the SMTP stream before it had set up the file descriptor. The bug has
2046 been fixed by making things happen in the correct order.
2047
2048
2049Exim version 4.40
2050-----------------
2051
2052 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
2053 before the connection was closed, thus losing the rejection response.
2054
2055 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
2056 some early Solaris releases, but causes trouble in current releases where
2057 socklen_t is defined.
2058
2059 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
2060 always exist.
2061
2062 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
2063 configured.
2064
2065 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
2066 that releasing the top of it at the end releases what was used for sub-
2067 expansions (unless the block got too big). However, discard this block if
2068 the first thing is a variable or header, so that we can use its block when
2069 it is dynamic (useful for very large $message_headers, for example).
2070
2071 6. Lookups now cache *every* query, not just the most recent. A new, separate
2072 store pool is used for this. It can be recovered when all lookup caches are
2073 flushed. Lookups now release memory at the end of their result strings.
2074 This has involved some general refactoring of the lookup sources.
2075
2076 7. Some code has been added to the store_xxx() functions to reduce the amount
2077 of flapping under certain conditions.
2078
2079 8. log_incoming_interface used to affect only the <= reception log lines. Now
2080 it causes the local interface and port to be added to several more SMTP log
2081 lines, for example "SMTP connection from", and rejection lines.
2082
2083 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
2084
208510. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
2086
208711. If $host_data was set by virtue of a hosts lookup in an ACL, its value
2088 could be overwritten at the end of the current message (or the start of a
2089 new message if it was set in a HELO ACL). The value is now preserved for
2090 the duration of the SMTP connection.
2091
209212. If a transport had a headers_rewrite setting, and a matching header line
2093 contained an unqualified address, that address was qualified, even if it
2094 did not match any rewriting rules. The underlying bug was that the values
2095 of the flags that permit the existence of unqualified sender and recipient
2096 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
2097 non-local messages, and by -bnq for local messages) were not being
2098 preserved with the message after it was received.
2099
210013. When Exim was logging an SMTP synchronization error, it could sometimes log
2101 "next input=" as part of the text comprising the host identity instead of
2102 the correct text. The code was using the same buffer for two different
2103 strings. However, depending on which order the printing function evaluated
2104 its arguments, the bug did not always show up. Under Linux, for example, my
2105 test suite worked just fine.
2106
210714. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
2108 doesn't work with some older versions of Perl. It has been changed to "my",
2109 which in any case is probably the better facility to use.
2110
211115. A really picky compiler found some instances of statements for creating
2112 error messages that either had too many or two few arguments for the format
2113 string.
2114
211516. The size of the buffer for calls to the DNS resolver has been increased
2116 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
2117 for addresses that have a lot of PTR records. This alleviates a problem; it
2118 does not fully solve it.
2119
212017. A dnsdb lookup for PTR records that receives more data than will fit in the
2121 buffer now truncates the list and logs the incident, which is the same
2122 action as happens when Exim is looking up a host name and its aliases.
2123 Previously in this situation something unpredictable would happen;
2124 sometimes it was "internal error: store_reset failed".
2125
212618. If a server dropped the connection unexpectedly when an Exim client was
2127 using GnuTLS and trying to read a response, the client delivery process
2128 crashed while trying to generate an error log message.
2129
213019. If a "warn" verb in an ACL added multiple headers to a message in a single
2131 string, for example:
2132
2133 warn message = H1: something\nH2: something
2134
2135 the text was added as a single header line from Exim's point of view
2136 though it ended up OK in the delivered message. However, searching for the
2137 second and subsequent header lines using $h_h2: did not work. This has been
2138 fixed. Similarly, if a system filter added multiple headers in this way,
2139 the routers could not see them.
2140
214120. Expanded the error message when iplsearch is called with an invalid key to
2142 suggest using net-iplsearch in a host list.
2143
214421. When running tests using -bh, any delays imposed by "delay" modifiers in
2145 ACLs are no longer actually imposed (and a message to that effect is
2146 output).
2147
214822. If a "gecos" field in a passwd entry contained escaped characters, in
2149 particular, if it contained a \" sequence, Exim got it wrong when building
2150 a From: or a Sender: header from that name. A second bug also caused
2151 incorrect handling when an unquoted " was present following a character
2152 that needed quoting.
2153
215423. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
2155 was not being matched caselessly.
2156
215724. Arranged for all hyphens in the exim.8 source to be escaped with
2158 backslashes.
2159
216025. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
2161 properly. Recipient callout cache records were still being keyed to include
2162 the sender, even when use_sender was set false. This led to far more
2163 callouts that were necessary. The sender is no longer included in the key
2164 when use_sender is false.
2165
216626. Added "control = submission" modifier to ACLs.
2167
216827. Added the ${base62d: operator to decode base 62 numbers.
2169
217028. dnsdb lookups can now access SRV records.
2171
217229. CONFIGURE_OWNER can be set at build time to define an alternative owner for
2173 the configuration file.
2174
217530. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
2176 (-v) mode. This makes the output for a verbose queue run more intelligible.
2177
217831. Added a use_postmaster feature to recipient callouts.
2179
218032. Added the $body_zerocount variable, containing the number of binary zero
2181 bytes in the message body.
2182
218333. The time of last modification of the "new" subdirectory is now used as the
2184 "mailbox time last read" when there is a quota error for a maildir
2185 delivery.
2186
218734. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
2188
218935. Added +ignore_unknown as a special item in host lists.
2190
219136. Code for decoding IPv6 addresses in host lists is now included, even if
2192 IPv6 support is not being compiled. This fixes a bug in which an IPv6
2193 address was recognized as an IP address, but was then not correctly decoded
2194 into binary, causing unexpected and incorrect effects when compared with
2195 another IP address.
2196
2197
2198Exim version 4.34
2199-----------------
2200
2201 1. Very minor rewording of debugging text in manualroute to say "list of
2202 hosts" instead of "hostlist".
2203
2204 2. If verify=header_syntax was set, and a header line with an unqualified
2205 address (no domain) and a large number of spaces between the end of the
2206 name and the colon was received, the reception process suffered a buffer
2207 overflow, and (when I tested it) crashed. This was caused by some obsolete
2208 code that should have been removed. The fix is to remove it!
2209
2210 3. When running in the test harness, delay a bit after writing a bounce
2211 message to get a bit more predictability in the log output.
2212
2213 4. Added a call to search_tidyup() just before forking a reception process. In
2214 theory, someone could use a lookup in the expansion of smtp_accept_max_
2215 per_host which, without the tidyup, could leave open a database connection.
2216
2217 5. Added the variables $recipient_data and $sender_data which get set from a
2218 lookup success in an ACL "recipients" or "senders" condition, or a router
2219 "senders" option, similar to $domain_data and $local_part_data.
2220
2221 6. Moved the writing of debug_print from before to after the "senders" test
2222 for routers.
2223
2224 7. Change 4.31/66 (moving the time when the Received: is generated) caused
2225 problems for message scanning, either using a data ACL, or using
2226 local_scan() because the Received: header was not generated till after they
2227 were called (in order to set the time as the time of reception completion).
2228 I have revised the way this works. The header is now generated after the
2229 body is received, but before the ACL or local_scan() are called. After they
2230 are run, the timestamp in the header is updated.
2231
2232
2233Exim version 4.33
2234-----------------
2235
2236 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
2237 before starting a queue runner without re-exec. This happened only when
2238 deliver_drop_privilege was set or when the Exim user was set to root. The
2239 effect of the bug was that timeouts during subsequent deliveries caused
2240 crashes instead of being properly handled. The handler is now left at its
2241 default (and expected) setting.
2242
2243 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
2244 message, again when deliver_drop_privilege is set or Exim is run as root.
2245 The bug described in (1) was not present in this case, but the tidying up
2246 of the other signals was missing. I have made the two cases consistent.
2247
2248 3. The ignore_target_hosts setting on a manualroute router was being ignored
2249 for hosts that were looked up using the /MX notation.
2250
2251 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
2252 in domain lists.
2253
2254 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
2255 operated on the sender address. After changing the $sender_address to <>
2256 for the sender address verify, Exim was re-instated it as the original
2257 (before rewriting) address, but remembering that it had rewritten it, so it
2258 wasn't rewriting it again. This bug also had the effect of breaking the
2259 sender address verification caching when the sender address was rewritten.
2260
2261 6. The ignore_target_hosts option was being ignored by the ipliteral router.
2262 This has been changed so that if the ip literal address matches
2263 ignore_target_hosts, the router declines.
2264
2265 7. Added expansion conditions match_domain, match_address, and match_local_
2266 part (NOT match_host).
2267
2268 8. The placeholder for the Received: header didn't have a length field set.
2269
2270 9. Added code to Exim itself and to exim_lock to test for a specific race
2271 condition that could lead to file corruption when using MBX delivery. The
2272 issue is with the lockfile that is created in /tmp. If this file is removed
2273 after a process has opened it but before that process has acquired a lock,
2274 there is the potential for a second process to recreate the file and also
2275 acquire a lock. This could lead to two Exim processes writing to the file
2276 at the same time. The added code performs the same test as UW imapd; it
2277 checks after acquiring the lock that its file descriptor still refers to
2278 the same named file.
2279
228010. The buffer for building added header lines was of fixed size, 8192 bytes.
2281 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
2282 when Exim is built.
2283
228411. Added the smtp_active_hostname option. If used, this will typically be made
2285 to depend on the incoming interface address. Because $interface_address is
2286 not set up until the daemon has forked a reception process, error responses
2287 that can happen earlier (such as "too many connections") no longer contain
2288 a host name.
2289
229012. If an expansion in a condition on a "warn" statement fails because a lookup
2291 defers, the "warn" statement is abandoned, and the next ACL statement is
2292 processed. Previously this caused the whole ACL to be aborted.
2293
229413. Added the iplsearch lookup type.
2295
229614. Added ident_timeout as a log selector.
2297
229815. Added tls_certificate_verified as a log selector.
2299
230016. Added a global option tls_require_ciphers (compare the smtp transport
2301 option of the same name). This controls incoming TLS connections.
2302
230317. I finally figured out how to make tls_require_ciphers do a similar thing
2304 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
2305 before starting the TLS session.
2306
230718. Tabs are now shown as \t in -bP output.
2308
230919. If the log selector return_path_on_delivery was set, Exim crashed when
2310 bouncing a message because it had too many Received: header lines.
2311
231220. If two routers both had headers_remove settings, and the first one included
2313 a superfluous trailing colon, the final name in the first list and the
2314 first name in the second list were incorrectly joined into one item (with a
2315 colon in the middle).
2316
2317
2318Exim version 4.32
2319-----------------
2320
2321 1. Added -C and -D options to the exinext utility, mainly to make it easier
2322 to include in the automated testing, but these could be helpful when
2323 multiple configurations are in use.
2324
2325 2. The exinext utility was not formatting the output nicely when there was
2326 an alternate port involved in the retry record key, nor when there was a
2327 message id as well (for retries that were specific to a specific message
2328 and a specific host). It was also confused by IPv6 addresses, because of
2329 the additional colons they contain. I have fixed the IPv4 problem, and
2330 patched it up to do a reasonable job for IPv6.
2331
2332 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
2333 delivery, the log line now contains "pipelined" if PIPELINING was used.
2334
2335 4. An SMTP transport process used to panic and die if the bind() call to set
2336 an explicit outgoing interface failed. This has been changed; it is now
2337 treated in the same way as a connect() failure.
2338
2339 5. A reference to $sender_host_name in the part of a conditional expansion
2340 that was being skipped was still causing a DNS lookup. This no longer
2341 occurs.
2342
2343 6. The def: expansion condition was not recognizing references to header lines
2344 that used bh_ and bheader_.
2345
2346 7. Added the _cache feature to named lists.
2347
2348 8. The code for checking quota_filecount in the appendfile transport was
2349 allowing one more file than it should have been.
2350
2351 9. For compatibility with Sendmail, the command line option
2352
2353 -prval:sval
2354
2355 is equivalent to
2356
2357 -oMr rval -oMs sval
2358
2359 and sets the incoming protocol and host name (for trusted callers). The
2360 host name and its colon can be omitted when only the protocol is to be set.
2361 Note the Exim already has two private options, -pd and -ps, that refer to
2362 embedded Perl. It is therefore impossible to set a protocol value of "d" or
2363 "s", but I don't think that's a major issue.
2364
236510. A number of refactoring changes to the code, none of which should affect
2366 Exim's behaviour:
2367
2368 (a) The number of logging options was getting close to filling up the
2369 32-bit word that was used as a bit map. I have split them into two classes:
2370 those that are passed in the argument to log_write(), and those that are
2371 only ever tested independently outside of that function. These are now in
2372 separate 32-bit words, so there is plenty of room for expansion again.
2373 There is no change in the user interface or the logging behaviour.
2374
2375 (b) When building, for example, log lines, the code previously used a
2376 macro that called string_cat() twice, in order to add two strings. This is
2377 not really sufficiently general. Furthermore, there was one instance where
2378 it was actually wrong because one of the argument was used twice, and in
2379 one call a function was used. (As it happened, calling the function twice
2380 did not affect the overall behaviour.) The macro has been replaced by a
2381 function that can join an arbitrary number of extra strings onto a growing
2382 string.
2383
2384 (c) The code for expansion conditions now uses a table and a binary chop
2385 instead of a serial search (which was left over from when there were very
2386 few conditions). Also, it now recognizes conditions like "pam" even when
2387 the relevant support is not compiled in: a suitably worded error message is
2388 given if an attempt is made to use such a condition.
2389
239011. Added ${time_interval:xxxxx}.
2391
239212. A bug was causing one of the ddress fields not to be passed back correctly
2393 from remote delivery subprocesses. The field in question was not being
2394 subsequently used, so this caused to problems in practice.
2395
239613. Added new log selectors queue_time and deliver_time.
2397
239814. Might have fixed a bug in maildirsizefile handling that threw up
2399 "unexpected character" debug warnings, and recalculated the data
2400 unnecessarily. In any case, I expanded the warning message to give more
2401 information.
2402
240315. Added the message "Restricted characters in address" to the statements in
2404 the default ACL that block characters like @ and % in local parts.
2405
240616. Change 71 for release 4.31 proved to be much less benign that I imagined.
2407 Three changes have been made:
2408
2409 (a) There was a serious bug; a negative response to MAIL caused the whole
2410 recipient domain to be cached as invalid, thereby blocking all messages
2411 to all local parts at the same domain, from all senders. This bug has
2412 been fixed. The domain is no longer cached after a negative response to
2413 MAIL if the sender used is not empty.
2414
2415 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
2416 been restored.
2417
2418 (c) A new callout option, "use_sender" has been added for people who want
2419 the modified behaviour.
2420
2421
2422Exim version 4.31
2423-----------------
2424
2425 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
2426 Larry Rosenman.
2427
2428 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
2429 indeed breaks things for older releases.
2430
2431 3. Added additional logging to the case where there is a problem reading data
2432 from a filter that is running in a subprocess using a pipe, in order to
2433 try to track down a specific problem.
2434
2435 4. Testing facility fudge: when running in the test harness and attempting
2436 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
2437 getting "No route to host". Convert this to a timeout.
2438
2439 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
2440 warning.
2441
2442 6. Some OS don't have socklen_t but use size_t instead. This affects the
2443 fifth argument of getsockopt() amongst other things. This is now
2444 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
2445 can be set for individual OS. I have set it for SunOS5, OSF1, and
2446 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
2447 some earlier ones do not.
2448
2449 7. Change 4.30/15 was not doing the test caselessly.
2450
2451 8. The standard form for an IPv6 address literal was being rejected by address
2452 parsing in, for example, MAIL and RCPT commands. An example of this kind of
2453 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
2454 this, as well as the form without the "IPv6" on the front (but only when
2455 address literals are enabled, of course).
2456
2457 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
2458
245910. Exim crashed if a message with an empty sender address specified by -f
2460 encountered a router with an errors_to setting. This could be provoked only
2461 by a command such as
2462
2463 exim -f "" ...
2464
2465 where an empty string was supplied; "<>" did not hit this bug.
2466
246711. Installed PCRE release 4.5.
2468
246912. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
2470 remained set. It is now erased.
2471
247213. exiqgrep wasn't working on MacOS X because it didn't correctly compute
2473 times from message ids (which are base 36 rather than the normal 62).
2474
247514. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
2476 were being counted as actual protocol errors, and logged if the log
2477 selector +smtp_protocol_error was set. One cannot be perfect in this test,
2478 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
2479 and DATA following a set of rejected RCPTs do not count as protocol errors.
2480 In other words, Exim assumes they were pipelined, though this may not
2481 actually be the case. Of course, in all cases the client gets an
2482 appropriate error code.
2483
248415. If a lookup fails in an ACL condition, a message about the failure may
2485 be available; it is used if testing the ACL cannot continue, because most
2486 such messages specify what the cause of the deferral is. However, some
2487 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
2488 that caused an old message to be retained and used if a later statement
2489 caused a defer, replacing the real cause of the deferral.
2490
249116. If an IP address had so many PTR records that the DNS lookup buffer
2492 was not large enough to hold them, Exim could crash while trying to process
2493 the truncated data. It now detects and logs this case.
2494
249517. Further to 4.21/58, another change has been made: if (and only if) the
2496 first line of a message (the first header line) ends with CRLF, a bare LF
2497 in a subsequent header line has a space inserted after it, so as not to
2498 terminate the header.
2499
250018. Refactoring: tidied an ugly bit of code in appendfile that copied data
2501 unnecessarily, used atoi() instead of strtol(), and didn't check the
2502 termination when getting file sizes from file names by regex.
2503
250419. Completely re-implemented the support for maildirsize files, in the light
2505 of a number of problems with the previous contributed implementation
2506 (4.30/29). In particular:
2507
2508 . If the quota is zero, the maildirsize file is maintained, but no quota is
2509 imposed.
2510
2511 . If the maildir directory does not exist, it is created before any attempt
2512 to write a maildirsize file.
2513
2514 . The quota value in the file is just a cache; if the quota is changed in
2515 the transport, the new value overrides.
2516
2517 . A regular expression is available for excluding directories from the
2518 count.
2519
252020. The autoreply transport checks the characters in options that define the
2521 message's headers; it allows continued headers, but it was checking with
2522 isspace() after an embedded newline instead of explicitly looking for a
2523 space or a tab.
2524
252521. If all the "regular" hosts to which an address was routed had passed their
2526 expiry times, and had not reached their retry times, the address was
2527 bounced, even if fallback hosts were defined. Now Exim should go on to try
2528 the fallback hosts.
2529
253022. Increased buffer sizes in the callout code from 1024 to 4096 to match the
2531 equivalent code in the SMTP transport. Some hosts send humungous responses
2532 to HELO/EHLO, more than 1024 it seems.
2533
253423. Refactoring: code in filter.c used (void *) for "any old type" but this
2535 gives compiler warnings in some environments. I've now done it "properly",
2536 using a union.
2537
253824. The replacement for inet_ntoa() that is used with gcc on IRIX systems
2539 (because of problems with the built-in one) was declared to return uschar *
2540 instead of char *, causing compiler failure.
2541
254225. Fixed a file descriptor leak when processing alias/forward files.
2543
254426. Fixed a minor format string issue in dbfn.c.
2545
254627. Typo in exim.c: ("dmbnz" for "dbmnz").
2547
254828. If a filter file refered to $h_xxx or $message_headers, and the headers
2549 contained RFC 2047 "words", Exim's memory could, under certain conditions,
2550 become corrupted.
2551
255229. When a sender address is verified, it is cached, to save repeating the test
2553 when there is more than one recipient in a message. However, when the
2554 verification involves a callout, it is possible for different callout
2555 options to be set for different recipients. It is too complicated to keep
2556 track of this in the cache, so now Exim always runs a verification when a
2557 callout is required, relying on the callout cache for the optimization.
2558 The overhead is duplication of the address routing, but this should not be
2559 too great.
2560
256130. Fixed a bug in callout caching. If a RCPT command caused the sender address
2562 to be verified with callout=postmaster, and the main callout worked but the
2563 postmaster check failed, the verification correctly failed. However, if a
2564 subsequent RCPT command asked for sender verification *without* the
2565 postmaster check, incorrect caching caused this verification also to fail,
2566 incorrectly.
2567
256831. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
2569 it was not caching the DNS options (qualify_single, search_parents) that
2570 were used when the lookup failed. A subsequent lookup with different
2571 options therefore always gave the same answer, though there were cases
2572 where it should not have. (Example: a "domains = !$mx_any" option on a
2573 dnslookup router: the "domains" option is always processed without any
2574 widening, but the router might have qualify_single set.) Now Exim uses the
2575 cached value only when the same options are set.
2576
257732. Added John Jetmore's "exipick" utility to the distribution.
2578
257933. GnuTLS: When an attempt to start a TLS session fails for any reason other
2580 than a timeout (e.g. a certificate is required, and is not provided), an
2581 Exim server now closes the connection immediately. Previously it waited for
2582 the client to close - but if the client is SSL, it seems that they each
2583 wait for each other, leading to a delay before one of them times out.
2584
258534: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
2586 maintained 0.8.x compatibility because I don't think many are using it, and
2587 it is clearly obsolete.
2588
258935. Added TLS support for CRLs: a tls_crl global option and one for the smtp
2590 transport.
2591
259236. OpenSSL: $tls_certificate_verified was being set to 1 even if the
2593 client certificate was expired. A simple patch fixes this, though I don't
2594 understand the full logic of why the verify callback is called multiple
2595 times.
2596
259737. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
2598 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
2599 which causes problems with some clients (such as the Certicom SSL Plus
2600 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
2601 disables the coutermeasure allowing Eudora to connect."
2602
260338. Exim was not checking that a write() to a log file succeeded. This could
2604 lead to Bad Things if a log got too big, in particular if it hit a file
2605 size limit. Exim now panics and dies if it cannot write to a log file, just
2606 as it does if it cannot open a log file.
2607
260839. Modified OS/Makefile-Linux so that it now contains
2609
2610 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
2611
2612 The two -D definitions ensure that Exim is compiled with large file
2613 support, which makes it possible to handle log files that are bigger than
2614 2^31.
2615
261640. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
2617 instance) a domain was checked against a named list that involved a lookup,
2618 causing $domain_data to be set, then another domain was checked against the
2619 same list, then the first domain was re-checked, the value of $domain_data
2620 after the final check could be wrong. In particular, if the second check
2621 failed, it could be set empty. This bug probably also applied to
2622 $localpart_data.
2623
262441. The strip_trailing_dot option was not being applied to the address given
2625 with the -f command-line option.
2626
262742. The code for reading a message's header from the spool was incrementing
2628 $received_count, but never initializing it. This meant that the value was
2629 incorrect (doubled) while delivering a message in the same process in which
2630 it was received. In the most common configuration of Exim, this never
2631 happens - a fresh exec is done - but it can happen when
2632 deliver_drop_privilege is set.
2633
263443. When Exim logs an SMTP synchronization error - client data sent too soon -
2635 it now includes up to 150 characters of the unexpected data in the log
2636 line.
2637
263844. The exim_dbmbuild utility uses fixed size buffers for reading input lines
2639 and building data strings. The size of both of these buffers was 10 000
2640 bytes - far larger than anybody would *ever* want, thought I. Needless to
2641 say, somebody hit the limit. I have increased the maximum line length to
2642 20 000 and the maximum data length of concatenated lines to 100 000. I have
2643 also fixed two bugs, because there was no checking on these buffers. Tsk,
2644 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
2645 buffer is too small.
2646
264745. The exim_dbmbuild utility did not support quoted keys, as Exim does in
2648 lsearch lookups. Now it does.
2649
265046. When parsing a route_list item in a manualroute router, a fixed-length
2651 buffer was used for the list of hosts. I made this 1024 bytes long,
2652 thinking that nobody would ever have a list of hosts that long. Wrong.
2653 Somebody had a whole pile of complicated expansion conditions, and the
2654 string was silently truncated, leading to an expansion error. It turns out
2655 that it is easier to change to an unlimited length (owing to other changes
2656 that have happened since this code was originally written) than to build
2657 structure for giving a limitation error. The length of the item that
2658 expands into the list of hosts is now unlimited.
2659
266047. The lsearch lookup could not handle data where the length of text line was
2661 more than 4095 characters. Such lines were truncated, leading to shortened
2662 data being returned. It should now handle lines of any length.
2663
266448. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
2665 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
2666 ACL").
2667
266849. Cosmetic tidy to scripts like exicyclog that are generated by globally
2669 replacing strings such as BIN_DIRECTORY in a source file: the replacement
2670 no longer happens in comment lines. A list of replacements is now placed
2671 at the head of all of the source files, except those whose only change is
2672 to replace PERL_COMMAND in the very first #! line.
2673
267450. Replaced the slow insertion sort in queue.c, for sorting the list of
2675 messages on the queue, with a bottom-up merge sort, using code contributed
2676 by Michael Haardt. This should make operations like -bp somewhat faster on
2677 large queues. It won't affect queue runners, except when queue_run_in_order
2678 is set.
2679
268051. Installed eximstats 1.31 in the distribution.
2681
268252. Added support for SRV lookups to the dnslookup router.
2683
268453. If an ACL referred to $message_body or $message_body_end, the value was not
2685 reset for any messages that followed in the same SMTP session.
2686
268754. The store-handling optimization for building very long strings was not
2688 differentiating between the different store pools. I don't think this
2689 actually made any difference in practice, but I've tidied it.
2690
269155. While running the routers to verify a sender address, $sender_address
2692 was still set to the sender address. This is wrong, because when routing to
2693 send a bounce to the sender, it would be empty. Therefore, I have changed
2694 it so that, while verifying a sender address, $sender_address is set to <>.
2695 (There is no change to what happens when verifying a recipient address.)
2696
269756. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
2698 target A or AAAA records (if not already returned) without resetting the
2699 qualify_single or search_parents options of the DNS resolver. These are
2700 inappropriate in this case because the targets of MX and SRV records must
2701 be FQDNs. A broken DNS record could cause trouble if it happened to have a
2702 target that, when qualified, matched something in the local domain. These
2703 two options are now turned off when doing these lookups.
2704
270557. It seems that at least some releases of Reiserfs (which does not have the
2706 concept of a fixed number of inodes) returns zero and not -1 for the
2707 number of available inodes. This interacted badly with check_spool_inodes,
2708 which assumed that -1 was the "no such thing" setting. What I have done is
2709 to check that the total number of inodes is greater than zero before doing
2710 the test of how many are available.
2711
271258. When a "warn" ACL statement has a log_message modifier, the message is
2713 remembered, and not repeated. This is to avoid a lot of repetition when a
2714 message has many recipients that cause the same warning to be written.
2715 Howewer, Exim was preserving the list of already written lines for an
2716 entire SMTP session, which doesn't seem right. The memory is now reset if a
2717 new message is started.
2718
271959. The "rewrite" debugging flag was not showing the result of rewriting in the
2720 debugging output unless log_rewrite was also set.
2721
272260. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
2723 of (int)(handle) when we know that handle contains (void *)(-1).
2724
272561. The Exim daemon panic-logs an error return when it closes the incoming
2726 connection. However "connection reset by peer" seems to be common, and
2727 isn't really an error worthy of noting specially, so that particular error
2728 is no long logged.
2729
273062. When Exim is trying to find all the local interfaces, it used to panic and
2731 die if the ioctl to get the interface flags failed. However, it seems that
2732 on at least one OS (Solaris 9) it is possible to have an interface that is
2733 included in the list of interfaces, but for which you get a failure error
2734 for this call. This happens when the interface is not "plumbed" into a
2735 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
2736 failure of the "get flags" call assumes that the interface is down.
2737
273863. Added a ${eval10: operator, which assumes all numbers are decimal. This
2739 makes life easier for people who are doing arithmetic on fields extracted
2740 from dates, where you often get leading zeros that should not be
2741 interpreted as octal.
2742
274364. Added qualify_domain to the redirect router, to override the global
2744 setting.
2745
274665. If a pathologically long header line contained very many addresses (the
2747 report of this problem mentioned 10 000) and each of them was rewritten,
2748 Exim could use up a very large amount of memory. (It kept on making new
2749 copies of the header line as it rewrote, and never released the old ones.)
2750 At the expense of a bit more processing, the header rewriting function has
2751 been changed so that it no longer eats memory in this way.
2752
275366. The generation of the Received: header has been moved from the time that a
2754 message starts to be received, to the time that it finishes. The timestamp
2755 in the Received: header should now be very close to that of the <= log
2756 line. There are two side-effects of this change:
2757
2758 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
2759 logged header lines no longer include the local Received: line, because
2760 it has not yet been created. The same applies to a copy of the message
2761 that is returned to a non-SMTP sender when a message is rejected.
2762
2763 (b) When a filter file is tested using -bf, no additional Received: header
2764 is added to the test message. After some thought, I decided that this
2765 is a bug fix.
2766
2767 This change does not affect the value of $received_for. It is still set
2768 after address rewriting, but before local_scan() is called.
2769
277067. Installed the latest Cygwin-specific files from the Cygwin maintainer.
2771
277268. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
2773 gave an unhelpful panic error message, and a defer error. I have managed to
2774 change this behaviour so that it now rejects any supplied certificate,
2775 which seems right, as the list of acceptable certificates is empty.
2776
277769. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
2778 gave an unhelpful defer error. I have not managed to make this reject any
2779 supplied certificates, but the error message it gives is "no certificate
2780 supplied", which is not helpful.
2781
278270. exigrep's output now also includes lines that are not associated with any
2783 message, but which match the given pattern. Implemented by a patch from
2784 Martin Sluka, which also tidied up the Perl a bit.
2785
278671. Recipient callout verification, like sender verification, was using <> in
2787 the MAIL FROM command. This isn't really the right thing, since the actual
2788 sender may affect whether the remote host accepts the recipient or not. I
2789 have changed it to use the actual sender in the callout; this means that
2790 the cache record is now keyed on a recipient/sender pair, not just the
2791 recipient address. There doesn't seem to be a real danger of callout loops,
2792 since a callout by the remote host to check the sender would use <>.
2793 [SEE ABOVE: changed after hitting problems.]
2794
279572. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
2796 temporary errors. However, in the case of such a code being given after
2797 the end of a data transmission (i.e. after ".") Exim was failing to write
2798 a retry record for the message. (Yes, there was some broken host that was
2799 actually sending 8xx at this point.)
2800
280173. An unknown lookup type in a host list could cause Exim to panic-die when
2802 the list was checked. (An example that provoked this was putting <; in the
2803 middle of a list instead of at the start.) If this happened during a DATA
2804 ACL check, a -D file could be left lying around. This kind of configuration
2805 error no longer causes Exim to die; instead it causes a defer errror. The
2806 incident is still logged to the main and panic logs.
2807
280874. Buglet left over from Exim 3 conversion. The message "too many messages
2809 in one connection" was written to the rejectlog but not the mainlog, except
2810 when address rewriting (yes!) was being logged.
2811
281275. Added write_rejectlog option.
2813
281476. When a system filter was run not as root (that is, when system_filter_user
2815 was set), the values of the $n variables were not being returned to the
2816 main process; thus, they were not subsequently available in the $sn
2817 variables.
2818
281977. Added +return_path_on_delivery log selector.
2820
282178. A connection timeout was being treated differently from recipients deferred
2822 when testing hosts_max_try with a message that was older than the host's
2823 retry timeout. (The host should not be counted, thus allowing all hosts to
2824 be tried at least once before bouncing.) This may have been the cause of an
2825 occasionally reported bug whereby a message would remain on the queue
2826 longer than the retry timeout, but would be bounced if a delivery was
2827 forced. I say "may" because I never totally pinned down the problem;
2828 setting up timeout/retry tests is difficult. See also the next item.
2829
283079. The ultimate address timeout was not being applied to errors that involved
2831 a combination of host plus message (for example, a timeout on a MAIL
2832 command). When an address resolved to a number of possible hosts, and they
2833 were not all tried for each delivery (e.g. because of hosts_max_try), a
2834 message could remain on the queue longer than the retry timeout.
2835
283680. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
2837 Haardt.
2838
283981. Fixed an obscure SMTP outgoing bug which required at least the following
2840 conditions: (a) there was another message waiting for the same server;
2841 (b) the server returned 5xx to all RCPT commands in the first message so
2842 that the message was not completed; (c) the server dropped the connection
2843 or gave a negative response to the RSET that Exim sends to abort the
2844 transaction. The observed case was a dropped connection after DATA that had
2845 been sent in pipelining mode. That is, the server had advertised PIPELINING
2846 but was not implementing it correctly. The effect of the bug was incorrect
2847 behaviour, such as trying another host, and this could lead to a crash.
2848
2849
2850Exim version 4.30
2851-----------------
2852
2853 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
2854 and daemon.c were passed as pointers to ints; they should have been
2855 pointers to socklen_t variables (which are typically unsigned ints).
2856
2857 2. Some signed/unsigned type warnings in the os.c file for Linux have been
2858 fixed.
2859
2860 3. Fixed a really odd bug that affected only the testing scheme; patching a
2861 certain fixed string in the binary changed the value of another string that
2862 happened to be identical to the end of the original first string.
2863
2864 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
2865 name", it returns that address as the IP address. On some operating
2866 systems (e.g. Solaris), it also passes back the IP address string as the
2867 "host name". However, on others (e.g. Linux), it passes back an empty
2868 string. Exim wasn't checking for this, and was changing the host name to an
2869 empty string, assuming it had been canonicized.
2870
2871 5. Although rare, it is permitted to have more than one PTR record for a given
2872 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
2873 all the names associated with an address, because they do in Solaris.
2874 However, it seems that they do not in Linux for data that comes from the
2875 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
2876 I found this out when I moved to a new Linux workstation and tried to run
2877 the Exim test suite.
2878
2879 To get round this problem I have changed the code so that it now does its
2880 own call to the DNS to look up PTR records when searching for a host name.
2881 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
2882 addresses that are only in /etc/hosts are still found.
2883
2884 This behaviour is, however, controlled by an option called host_lookup_
2885 order, which defaults to "bydns:byaddr". If people want to use the other
2886 order, or indeed, just use one or the other means of lookup, they can
2887 specify it in this variable.
2888
2889 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
2890 some operating systems, this comes back from gethostbyaddr() as an empty
2891 string, and this is what Exim used to test for. However, it seems that in
2892 other systems, "." is yielded. Exim now tests for this case too.
2893
2894 7. The values of check_spool_space and check_log_space are now held internally
2895 as a number of kilobytes instead of an absolute number of bytes. If a
2896 numbers is specified without 'K' or 'M', it is rounded up to the nearest
2897 kilobyte. This means that much larger values can be stored.
2898
2899 8. Exim monitor: an attempt to get the action menu when not actually pointing
2900 at a message produces an empty menu entitled "No message selected". This
2901 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
2902 no entries in it ("Shell widget menu has zero width and/or height"). So I
2903 have added a single, blank menu entry in this case.
2904
2905 9. Added ${quote_local_part.
2906
290710. MIME decoding is now applied to the contents of Subject: header lines when
2908 they are logged.
2909
291011. Now that a reference to $sender_host_address automatically causes a reverse
2911 lookup to occur if necessary (4.13/18), there is no need to arrange for a
2912 host lookup before query-style lookups in lists that might use this
2913 variable. This has therefore been abolished, and the "net-" prefix is no
2914 longer necessary for query-style lookups.
2915
291612. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
2917 have been a typo for LFLAGS, so it has been changed.
2918
291913. The install script calls Exim with "-C /dev/null" in order to find the
2920 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
2921 to be output. Howeve, since Exim outputs its version number before the
2922 error, it didn't break the script. It just looked ugly. I fixed this by
2923 always allowing "-C /dev/null" if the caller is root.
2924
292514. Ignore overlarge ACL variable number when reading spool file - insurance
2926 against a later release with more variables having written the file.
2927
292815. The standard form for an IPv6 address literal was being rejected by EHLO.
2929 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
2930 this, as well as the form without the "IPv6" on the front.
2931
293216. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
2933 OS/Makefile-Darwin file.
2934
293517. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
2936 only to LDAP libraries that do not have LDAP_OPT_DEREF.
2937
293818. After change 4.21/52, "%ld" was used to format the contents of the $inode
2939 variable. However, some OS use ints for inodes. I've added cast to long int
2940 to get rid of the compiler warning.
2941
294219. I had forgotten to lock out "/../" in configuration file names when
2943 ALT_CONFIG_PREFIX was set.
2944
294520. Routers used for verification do not need to specify transports. However,
2946 if such a router generated a host list, and callout was configured, Exim
2947 crashed, because it could not find a port number from the (non-existent)
2948 transport. It now assumes port 25 in this circumstance.
2949
295021. Added the -t option to exigrep.
2951
295222. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
2953 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
2954 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
2955 them is set, LOOKUP_LSEARCH is forced.
2956
295723. "exim -bV" now outputs a list of lookups that are included in the binary.
2958
295924. Added sender and host information to the "rejected by local_scan()" log
2960 line; previously there was no indication of these.
2961
296225. Added .include_if_exists.
2963
296426. Change 3.952/11 added an explicit directory sync on top of a file sync for
2965 Linux. It turns out that not all file systems support this. Apparently some
2966 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
2967 do it.) To cope with this, the error EINVAL, which means that sync-ing is
2968 not supported on the file descriptor, is now ignored when Exim is trying to
2969 sync a directory. This applies only to Linux.
2970
297127. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
2972
297328. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
2974 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
2975 this.
2976
297729. Added support for maildirsize files from supplied patch (modified a bit).
2978
297930. The use of :fail: followed by an empty string could lead Exim to respond to
2980 sender verification failures with (e.g.):
2981
2982 550 Verification failed for <xxx>
2983 550 Sender verify failed
2984
2985 where the first response line was missing the '-' that indicates it is not
2986 the final line of the response.
2987
298831. The loop for finding the name of the user that called Exim had a hardwired
2989 limit of 10; it now uses the value of finduser_retries, which is used for
2990 all other user lookups.
2991
299232. Added $received_count variable, available in data and not_smtp ACLs, and at
2993 delivery time.
2994
299533. Exim was neglecting to zero errno before one call of strtol() when
2996 expanding a string and expecting an integer value. On some systems this
2997 resulted in spurious "integer overflow" errors. Also, it was casting the
2998 result into an int without checking.
2999
300034. Testing for a connection timeout using "timeout_connect" in the retry rules
3001 did not work. The code looks as if it has *never* worked, though it appears
3002 to have been documented since at least releast 1.62. I have made it work.
3003
300435. The "timeout_DNS" error in retry rules, also documented since at least
3005 1.62, also never worked. As it isn't clear exactly what this means, and
3006 clearly it isn't a major issue, I have abolished the feature by treating it
3007 as "timeout", and writing a warning to the main and panic logs.
3008
300936. The display of retry rules for -brt wasn't always showing the error code
3010 correctly.
3011
301237. Added new error conditions to retry rules: timeout_A, timeout_MX,
3013 timeout_connect_A, timeout_connect_MX.
3014
301538. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
3016 to the empty sender.
3017
301839. The daemon was not analysing the content of -oX till after it had closed
3019 stderr and disconnected from the controlling terminal. This meant that any
3020 syntax errors were only noted on the panic log, and the return code from
3021 the command was 0. By re-arranging the code a little, I've made the
3022 decoding happen first, so such errors now appear on stderr, and the return
3023 code is 1. However, the actual setting up of the sockets still happens in
3024 the disconnected process, so errors there are still only recorded on the
3025 panic log.
3026
302740. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
3028 connections (as happens on some IP stacks) was logged at start up time as
3029 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
3030 it from "IPv6 and IPv4", which means that two separate sockets are being
3031 used.
3032
303341. The debug output for gethostbyname2() or getipnodebyname() failures now
3034 says whether AF_INET or AF_INET6 was passed as an argument.
3035
303642. Exiwhat output was messed up when time zones were included in log
3037 timestamps.
3038
303943. Exiwhat now gives more information about the daemon's listening ports,
3040 and whether -tls-on-connect was used.
3041
304244. The "port" option of the smtp transport is now expanded.
3043
304445. A "message" modifier in a "warn" statement in a non-message ACL was being
3045 silently ignored. Now an error message is written to the main and panic
3046 logs.
3047
304846. There's a new ACL modifier called "logwrite" which writes to a log file
3049 as soon as it is encountered.
3050
305147. Added $local_user_uid and $local_user_gid at routing time.
3052
305348. Exim crashed when trying to verify a sender address that was being
3054 rewritten to "<>".
3055
305649. Exim was recognizing only a space character after ".include". It now also
3057 recognizes a tab character.
3058
305950. Fixed several bugs in the Perl script that creates the exim.8 man page by
3060 extracting the relevant information from the specification. The man page no
3061 longer contains scrambled data for the -d option, and I've added a section
3062 at the front about calling Exim under different names.
3063
306451. Added "extra_headers" argument to the "mail" command in filter files.
3065
306652. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
3067 crash.
3068
306953. Installed eximstats 1.29.
3070
307154. Added transport_filter_timeout as a generic transport option.
3072
307355. Exim no longer adds an empty Bcc: header to messages that have no To: or
3074 Cc: header lines. This was required by RFC 822, but it not required by RFC
3075 2822.
3076
307756. Exim used to add From:, Date:, and Message-Id: header lines to any
3078 incoming messages that did not have them. Now it does so only if the
3079 message originates locally, that is, if there is no associated remote host
3080 address. When Resent- header lines are present, this applies to the Resent-
3081 lines rather than the non-Resent- lines.
3082
308357. Drop incoming SMTP connection after too many syntax or protocol errors. The
3084 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
3085
308658. Messages for configuration errors now include the name of the main
3087 configuration file - useful now that there may be more than one file in a
3088 list (.included file names were always shown).
3089
309059. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
3091 for those rare installations that do not start the daemon as root or run it
3092 setuid root. I've cut out the call to initgroups() if the daemon is not
3093 root at that time.
3094
309560. The Exim user and group can now be bound into the binary as text strings
3096 that are looked up at the start of Exim's processing.
3097
309861. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
3099
310062. Added $mailstore_basename variable.
3101
310263. Installed patch to sieve.c from Michael Haardt.
3103
310464. When Exim failed to open the panic log after failing to open the main log,
3105 the original message it was trying to log was written to stderr and debug
3106 output, but if they were not available (the usual case in production), it
3107 was lost. Now it is written to syslog before the two lines that record the
3108 failures to open the logs.
3109
311065. Users' Exim filters run in subprocesses under the user's uid. It is
3111 possible for a "deliver" command or an alias in a "personal" command to
3112 provoke an address rewrite. If logging of address rewriting is configured,
3113 this fails because the process is not running as root or exim. There may be
3114 a better way of dealing with this, but for the moment (because 4.30 needs
3115 to be released), I have disabled address rewrite logging when running a
3116 filter in a non-root, non-exim process.
3117
3118
3119Exim version 4.24
3120-----------------
3121
3122 1. The buildconfig auxiliary program wasn't quoting the value set for
3123 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
3124 not defined. This bug was masked in 4.22 by the effect that was fixed in
3125 change 4.23/1.
3126
3127 2. Some messages that were rejected after a message id was allocated were
3128 shown as "incomplete" by exigrep. It no longer does this for messages that
3129 are rejected by local_scan() or the DATA or non-SMTP ACLs.
3130
3131 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
3132 have allow_domain_literals set, the ID did not get logged in the <= line.
3133 Domain literals are now always recognized in Message-ID: header lines.
3134
3135 4. The first argument for a ${extract expansion item is the key name or field
3136 number. Leading and trailing spaces in this item were not being ignored,
3137 causing some misleading effects.
3138
3139 5. When deliver_drop_privilege was set, single queue runner processes started
3140 manually (i.e. by the command "exim -q") or by the daemon (which uses the
3141 same command in the process it spins off) were not dropping privilege.
3142
3143 6. When the daemon running as "exim" started a queue runner, it always
3144 re-executed Exim in the spun-off process. This is a waste of effort when
3145 deliver_drop_privilege is set. The new process now just calls the
3146 queue-runner function directly.
3147
3148
3149Exim version 4.23
3150-----------------
3151
3152 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
3153 HEADERS_CHARSET.
3154
3155 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
3156 ignored. Though the use of -oP was forcing the writing of a pid file, it
3157 was always written to the default place.
3158
3159 3. If the message "no IP address found for host xxxx" is generated during
3160 incoming verification, it is now followed by identification of the incoming
3161 connection (so you can more easily find what provoked it).
3162
3163 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
3164
3165 5. Added some features to "harden" Exim a bit more against certain attacks:
3166
3167 (a) There is now a build-time option called FIXED_NEVER_USERS that can
3168 be put in Local/Makefile. This is like the never_users runtime option,
3169 but it cannot be overridden. The default setting is "root".
3170
3171 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
3172 prefix string with which any file named in a -C command line option
3173 must start.
3174
3175 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
3176 is retained for -C and -D only if the caller of Exim is root. Without
3177 it, the exim user may also use -C and -D and retain privilege.
3178
3179 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
3180 command line option is disabled.
3181
3182 6. Macro names set by the -D option must start with an upper case letter, just
3183 like macro names defined in the configuration file.
3184
3185 7. Added "dereference=" facility to LDAP.
3186
3187 8. Two instances of the typo "uknown" in the source files are fixed.
3188
3189 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
3190 the Configure-Makefile script screwed up while processing it.
3191
319210. Incorporated PCRE 4.4.
3193
319411. The SMTP synchronization check was not operating right at the start of an
3195 SMTP session. For example, it could not catch a HELO sent before the client
3196 waited for the greeting. There is now a check for outstanding input at the
3197 point when the greeting is written. Because of the duplex, asynchronous
3198 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
3199 way, but not yet received, when the check is performed.
3200
320112. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
3202 on TCP/IP sockets, because this apparently causes some broken clients to
3203 timeout.
3204
320513. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
3206 unchanged) from the Cygwin maintainer.
3207
320814. The code for -bV that shows what is in the binary showed "mbx" when maildir
3209 was supported instead of testing for mbx. Effectively a typo.
3210
321115. The spa authenticator server code was not checking that the input it
3212 received was valid base64.
3213
321416. The debug output line for the "set" modifier in ACLs was not showing the
3215 name of the variable that was being set.
3216
321717. Code tidy: the variable type "vtype_string" was never used. Removed it.
3218
321918. Previously, a reference to $sender_host_name did not cause a DNS reverse
3220 lookup on its own. Something else was needed to trigger the lookup. For
3221 example, a match in host_lookup or the need for a host name in a host list.
3222 Now, if $sender_host_name is referenced and the host name has not yet been
3223 looked up, a lookup is performed. If the lookup fails, the variable remains
3224 empty, and $host_lookup_failed is set to "1".
3225
322619. Added "eqi" as a case-independent comparison operator.
3227
322820. The saslauthd authentication condition could segfault if neither service
3229 nor realm was specified.
3230
323121. If an overflowing value such as "2048M" was set for message_size_limit, the
3232 error message that was logged was misleading, and incoming SMTP
3233 connections were dropped. The message is now more accurate, and temporary
3234 errors are given to SMTP connections.
3235
323622. In some error situations (such as 21 above) Exim rejects all SMTP commands
3237 (except RSET) with a 421 error, until QUIT is received. However, it was
3238 failing to send a response to QUIT.
3239
324023. The HELO ACL was being run before the code for helo_try_verify_hosts,
3241 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
3242 ACL is now run after the helo_try_verify_hosts code.
3243
324424. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
3245 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
3246 case-independent, so other case variants are also recognized). Apparently
3247 some systems use these upper case variants.
3248
324925. If more than two messages were waiting for the same host, and a transport
3250 filter was specified for the transport, Exim sent two messages over the
3251 same TCP/IP connection, and then failed with "socket operation on non-
3252 socket" when it tried to send the third.
3253
325426. Added Exim::debug_write and Exim::log_write for embedded Perl use.
3255
325627. The extern definition of crypt16() in expand.c was not being excluded when
3257 the OS had its own crypt16() function.
3258
325928. Added bounce_return_body as a new option, and bounce_return_size_limit
3260 as a preferred synonym for return_size_limit, both as an option and as an
3261 expansion variable.
3262
326329. Added LIBS=-liconv to OS/Makefile-OSF1.
3264
326530. Changed the default configuration ACL to relax the local part checking rule
3266 for addresses that are not in any local domains. For these addresses,
3267 slashes and pipe symbols are allowed within local parts, but the sequence
3268 /../ is explicitly forbidden.
3269
327031. SPA server authentication was not clearing the challenge buffer before
3271 using it.
3272
327332. log_message in a "warn" ACL statement was writing to the reject log as
3274 well as to the main log, which contradicts the documentation and doesn't
3275 seem right (because no rejection is happening). So I have stopped it.
3276
327733. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
3278 However, I am unable to do any testing of this.
3279
328034. Fixed an infelicity in the appendfile transport. When checking directories
3281 for a mailbox, to see if any needed to be created, it was accidentally
3282 using path names with one or more superfluous leading slashes; tracing
3283 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
3284
328535. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
3286 contents are added to the log line that is written for every discarded
3287 recipient. (Previously a log_message setting was ignored.)
3288
328936. The ${quote: operator now quotes the string if it is empty.
3290
329137. The install script runs exim in order to find its version number. If for
3292 some reason other than non-existence or emptiness, which it checks, it
3293 could not run './exim', it was installing it with an empty version number,
3294 i.e. as "exim-". This error state is now caught, and the installation is
3295 aborted.
3296
329738. An argument was missing from the function that creates an error message
3298 when Exim fails to connect to the socket for saslauthd authentication.
3299 This could cause Exim to crash, or give a corrupted message.
3300
330139. Added isip, isip4, and isip6 to ${if conditions.
3302
330340. The ACL variables $acl_xx are now saved with the message, and can be
3304 accessed later in routers, transports, and filters.
3305
330641. The new lookup type nwildlsearch is like wildlsearch, except that the key
3307 strings in the file are not string-expanded.
3308
330942. If a MAIL command specified a SIZE value that was too large to fit into an
3310 int variable, the check against message_size_limit failed. Such values are
3311 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
3312 day this will have to be increased, but I don't think I want to be around
3313 when emails are that large.
3314
3315
3316
3317Exim version 4.22
3318-----------------
3319
3320 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
3321 iconv() is not standard in FreeBSD.
3322
3323 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
3324 IPv6 enabled. The observed symptom was a segmentation fault on return from
3325 the function os_common_find_running_interfaces() in src/os.c.
3326
3327 3. In the check_special_case() function in daemon.c I had used "errno" as an
3328 argument name, which causes warnings on some systems. This was basically a
3329 typo, since it was named "eno" in the comments!
3330
3331 4. The code that waits for the clock to tick (at a resolution of some fraction
3332 of a second) so as to ensure message-id uniqueness was always waiting for
3333 at least one whole tick, when it could have waited for less. [This is
3334 almost certainly not relevant at current processor speeds, where it is
3335 unlikely to ever wait at all. But we try to future-proof.]
3336
3337 5. The function that sleeps for a time interval that includes fractions of a
3338 second contained a race. It did not block SIGALRM between setting the
3339 timer, and suspending (a couple of lines later). If the interval was short
3340 and the sigsuspend() was delayed until after it had expired, the suspension
3341 never ended. On busy systems this could lead to processes getting stuck for
3342 ever.
3343
3344 6. Some uncommon configurations may cause a lookup to happen in a queue runner
3345 process, before it forks any delivery processes. The open lookup caching
3346 mechanism meant that the open file or database connection was passed into
3347 the delivery process. The problem was that delivery processes always tidy
3348 up cached lookup data. This could cause a problem for the next delivery
3349 process started by the queue runner, because the external queue runner
3350 process does not know about the closure. So the next delivery process
3351 still has data in the lookup cache. In the case of a file lookup, there was
3352 no problem because closing a file descriptor in a subprocess doesn't affect
3353 the parent. However, if the lookup was caching a connection to a database,
3354 the connection was closed, and the second delivery process was likely to
3355 see errors such as "PGSQL: query failed: server closed the connection
3356 unexpectedly". The problem has been fixed by closing all cached lookups
3357 in a queue runner before running a delivery process.
3358
3359 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
3360 seem to have the "const" qualifier which it has on other OS. I've
3361 parameterised it.
3362
3363 8. Change 4.21/2 was too strict. It is only if there are two authenticators
3364 *of the same type* (client or server) with the same public name that an
3365 error should be diagnosed.
3366
3367 9. When Exim looked up a host name for an IP address, but failed to find the
3368 original IP address when looking up the host name (a safety check), it
3369 output the message "<ip address> does not match any IP for NULL", which was
3370 confusing, to say the least. The bug was that the host name should have
3371 appeared instead of "NULL".
3372
337310. Since release 3.03, if Exim is called by a uid other than root or the Exim
3374 user that is built into the binary, and the -C or -D options is used, root
3375 privilege is dropped before the configuration file is read. In addition,
3376 logging is switched to stderr instead of the normal log files. If the
3377 configuration then re-defines the Exim user, the unprivileged environment
3378 is probably not what is expected, so Exim logs a panic warning message (but
3379 proceeds).
3380
3381 However, if deliver_drop_privilege is set, the unprivileged state may well
3382 be exactly what is intended, so the warning has been cut out in that case,
3383 and Exim is allowed to try to write to its normal log files.
3384
3385
3386Exim version 4.21
3387-----------------
3388
3389 1. smtp_return_error_details was not giving details for temporary sender
3390 or receiver verification errors.
3391
3392 2. Diagnose a configuration error if two authenticators have the same public
3393 name.
3394
3395 3. Exim used not to create the message log file for a message until the first
3396 delivery attempt. This could be confusing when incoming messages were held
3397 for policy or load reasons. The message log file is now created at the time
3398 the message is received, and an initial "Received" line is written to it.
3399
3400 4. The automatically generated man page for command line options had a minor
3401 bug that caused no ill effects; however, a more serious problem was that
3402 the procedure for building the man page automatically didn't always
3403 operate. Consequently, release 4.20 contains an out-of-date version. This
3404 shouldn't happen again.
3405
3406 5. When building Exim with embedded Perl support, the script that builds the
3407 Makefile was calling 'perl' to find its compile-time parameters, ignoring
3408 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
3409
3410 6. The freeze_tell option was not being used for messages that were frozen on
3411 arrival, either by an ACL or by local_scan().
3412
3413 7. Added the smtp_incomplete_transaction log selector.
3414
3415 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
3416 was accepting AUTH without a new EHLO.
3417
3418 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
3419 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
3420 RFC.
3421
342210. Logging of TCP/IP connections (when configured) now happens in the main
3423 daemon process instead of the child process, so that the TCP/IP connection
3424 count is more accurate (but it can never be perfect).
3425
342611. The use of "drop" in a nested ACL was not being handled correctly in the
3427 outer ACL. Now, if condition failure induced by the nested "drop" causes
3428 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
3429 or "require"), the connection is dropped.
3430
343112. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
3432 that yield "discard" can now be used with an "accept" or a "discard" verb,
3433 but an error is generated for any others (because I can't see a useful way
3434 to define what should happen).
3435
343613. When an ACL is read dynamically from a file (or anywhere else), the lines
3437 are now processed in the same way as lines in the Exim configuration file.
3438 In particular, continuation lines are supported.
3439
344014. Added the "dnslists = a.b.c!=n.n.n.n" feature.
3441
344215. Added -ti meaning -t -i.
3443
344416. Check for letters, digits, hyphens, and dots in the names of dnslist
3445 domains, and warn by logging if others are found.
3446
344717. At least on BSD, alignment is not guarenteed for the array of ifreq's
3448 returned from GIFCONF when Exim is trying to find the list of interfaces on
3449 a host. The code in os.c has been modified to copy each ifreq to an aligned
3450 structure in all cases.
3451
3452 Also, in some cases, the returned ifreq's were being copied to a 'struct
3453 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
3454 means the last couple of bytes of an IPv6 address could be chopped if the
3455 ifreq contained only a normal sockaddr (14 bytes storage).
3456
345718. Named domain lists were not supported in the hosts_treat_as_local option.
3458 An entry such as +xxxx was not recognized, and was treated as a literal
3459 domain name.
3460
346119. Ensure that header lines added by a DATA ACL are included in the reject log
3462 if the ACL subsequently rejects the message.
3463
346420. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
3465 MD5 (which is deprecated).
3466
346721. When testing a filter file using -bf, Exim was writing a message when it
3468 took the sender from a "From " line in the message, but it was not doing so
3469 when it took $return_path from a Return-Path: header line. It now does.
3470
347122. If the contents of a "message" modifier for a "warn" ACL verb do not begin
3472 with a valid header line field name (a series of printing characters
3473 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
3474
347523. Changed "disc" in the source to "disk" to conform to the documentation and
3476 the book and for uniformity.
3477
347824. Ignore Sendmail's -Ooption=value command line item.
3479
348025. When execve() failed while trying to run a command in a pipe transport,
3481 Exim was returning EX_UNAVAILBLE (69) from the subprocess. However, this
3482 could be confused with a return value of 69 from the command itself. This
3483 has been changed to 127, the value the shell returns if it is asked to run
3484 a non-existent command. The wording for the related log line suggests a
3485 non-existent command as the problem.
3486
348726. If received_header_text expands to an empty string, do not add a Received:
3488 header line to the message. (Well, it adds a token one on the spool, but
3489 marks it "old" so that it doesn't get used or transmitted.)
3490
349127. Installed eximstats 1.28 (addition of -nt option).
3492
349328. There was no check for failure on the call to getsockname() in the daemon
3494 code. This can fail if there is a shortage of resources on the system, with
3495 ENOMEM, for example. A temporary error is now given on failure.
3496
349729. Contrary to the C standard, it seems that in some environments, the
3498 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
3499 program. Exim now does this explicitly; it affects the formatting of
3500 timestamps using strftime().
3501
350230. If exiqsumm was given junk data, it threw up some uninitialized variable
3503 complaints. I've now initialized all the variables, to avoid this.
3504
350532. Header lines added by a system filter were not being "seen" during
3506 transport-time rewrites.
3507
350833. The info_callback() function passed to OpenSSL is set up with type void
3509 (*)(SSL *, int, int), as described somewhere. However, when calling the
3510 function (actually a macro) that sets it up, the type void(*)() is
3511 expected. I've put in a cast to prevent warnings from picky compilers.
3512
351334. If a DNS black list lookup found a CNAME record, but there were no A
3514 records associated with the domain it pointed at, Exim crashed.
3515
351635. If a DNS black list lookup returned more than one A record, Exim ignored
3517 all but the first. It now scans all returned addresses if a particular IP
3518 value is being sought. In this situation, the contents of the
3519 $dnslist_value variable are a list of all the addresses, separated by a
3520 comma and a space.
3521
352236. Tightened up the rules for host name lookups using reverse DNS. Exim used
3523 to accept a host name and all its aliases if the forward lookup for any of
3524 them yielded the IP address of the incoming connection. Now it accepts only
3525 those names whose forward lookup yields the correct IP address. Any other
3526 names are discarded. This closes a loophole whereby a rogue DNS
3527 administrator could create reverse DNS records to break through a
3528 wildcarded host restriction in an ACL.
3529
353037. If a user filter or a system filter that ran in a subprocess used any of
3531 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
3532 the wrong values were passed to the pipe command ($thisaddress had the
3533 value of $0, $0 had the value of $1, etc). This bug was introduced by
3534 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
3535
353638. Improved the line breaking for long SMTP error messages from ACLs.
3537 Previously, if there was no break point between 40 and 75 characters, Exim
3538 left the rest of the message alone. Two changes have been made: (a) I've
3539 reduced the minimum length to 35 characters; (b) if it can't find a break
3540 point between 35 and 75 characters, it looks ahead and uses the first one
3541 that it finds. This may give the occasional overlong line, but at least the
3542 remaining text gets split now.
3543
354439. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
3545 file descriptors might be low, and that setting 1000 would always raise it.
3546 It turns out that in some environments, the limit is already over 1000 and
3547 that lowering it causes trouble. So now Exim takes care not to decrease it.
3548
354940. When delivering a message, the value of $return_path is set to $sender_
3550 address at the start of routing (routers may change the value). By an
3551 oversight, this default was not being set up when an address was tested by
3552 -bt or -bv, which affected the outcome if any router or filter referred to
3553 $return_path.
3554
355541. The idea of the "warn" ACL verb is that it adds a header or writes to the
3556 log only when "message" or "log_message" are set. However, if one of the
3557 conditions was an address verification, or a call to a nested ACL, the
3558 messages generated by the underlying test were being passed through. This
3559 no longer happens. The underlying message is available in $acl_verify_
3560 message for both "message" and "log_message" expansions, so it can be
3561 passed through if needed.
3562
356342. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
3564 new expansion $bh_ to give the encoded byte string without charset
3565 translation. Translation happens only if iconv() is available; HAVE_ICONV
3566 indicates this at build time. HEADERS_CHARSET gives the charset to
3567 translate to; headers_charset can change it in the configuration, and
3568 "headers charset" can change it in an individual filter file.
3569
357043. Now that we have a default RFC 2047 charset (see above), the code in Exim
3571 that creates RFC 2047 encoded "words" labels them as that charset instead
3572 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
3573 expansion operator; (ii) when Exim creates a From: line for a local
3574 message; (iii) when a header line is rewritten to include a "phrase" part.
3575
357644. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
3577 buggy, causing it to skip the first lines of messages whose message ID
3578 ended in 'D'. This would not have bitten before Exim release 4.14, because
3579 message IDs were unlikely to end in 'D' before then. The effect was to have
3580 incorrect size information for certain domains.
3581
358245. #include "config.h" was missing at the start of the crypt16.c module. This
3583 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
3584 noticed.
3585
358646. If there was a timeout during a "random" callout check, Exim treated it as
3587 a failure of the random address, and carried on sending RSET and the real
3588 address. If the delay was just some slowness somewhere, the response to the
3589 original RCPT would be taken as a response to RSET and so on, causing
3590 mayhem of various kinds.
3591
359247. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
3593 when I implemented it. It didn't allow for the fact that some option values
3594 may legitimatetly be negative (e.g. size_addition), and it didn't even do
3595 the right test for positive values.
3596
359748. Domain names in DNS records are case-independent. Exim always looks them up
3598 in lower case. Some resolvers return domain names in exactly the case they
3599 appear in the zone file, that is, they may contain uppercase letters. Not
3600 all resolvers do this - some return always lower case. Exim was treating a
3601 change of case by a resolver as a change of domain, similar to a widening
3602 of a domain abbreviation. This triggered its re-routing code and so it was
3603 trying to route what was effectively the same domain again. This normally
3604 caused routing to fail (because the router wouldn't handle the domain
3605 twice). Now Exim checks for this case specially, and just changes the
3606 casing of the domain that it ultimately uses when it transmits the message
3607 envelope.
3608
360949. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
3610 module.
3611
361250. If a filter generated a file delivery with a non-absolute name (possible if
3613 no home directory exists for the router), the forbid_file option was not
3614 forbidding it.
3615
361651. Added '&' feature to dnslists, to provide bit mask matching in addition to
3617 the existing equality matching.
3618
361952. Exim was using ints instead of ino_t variables in some places where it was
3620 dealing with inode numbers.
3621
362253. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
3623 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
3624 called TMPDIR, and if it finds it is different, it changes its value.
3625
362654. The smtp_printf() function is now made available to local_scan() so
3627 additional output lines can be written before returning. There is also an
3628 smtp_fflush() function to enable the detection of a dropped connection.
3629 The variables smtp_input and smtp_batched_input are exported to
3630 local_scan().
3631
363255. Changed the default runtime configuration: the message "Unknown user"
3633 has been removed from the ACL, and instead placed on the localuser router,
3634 using the cannot_route_message feature. This means that any verification
3635 failures that generate their own messages won't get overridden. Similarly,
3636 the "Unrouteable address" message that was in the ACL for unverifiable
3637 relay addresses has also been removed.
3638
363956. Added hosts_avoid_esmtp to the smtp transport.
3640
364157. The exicyclog script was not checking for the esoteric option
3642 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
3643 will work only if exicyclog is run under the appropriate euid.
3644
364558. Following a discussion on the list, the rules by which Exim recognises line
3646 endings on incoming messages have been changed. The -dropcr and drop_cr
3647 options are now no-ops, retained only for backwards compatibility. The
3648 following line terminators are recognized: LF CRLF CR. However, special
3649 processing applies to CR:
3650
3651 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
3652 nor a local message in the state where . is a terminator.
3653
3654 (ii) If a bare CR is encountered in a header line, an extra space is added
3655 after the line terminator so as not to end the header. The reasoning
3656 behind this is that bare CRs in header lines are most likely either
3657 to be mistakes, or people trying to play silly games.
3658
365959. The size of a message, as listed by "-bp" or in the Exim monitor window,
3660 was being incorrectly given as 18 bytes larger than it should have been.
3661 This is a VOB (very old bug).
3662
366360. This may never have affected anything current, but just in case it has:
3664 When the local host is found other than at the start of a list of hosts,
3665 the local host, those with the same MX, and any that follow, are discarded.
3666 When the list in question was part of a longer list of hosts, the following
3667 hosts (not currently being processed) were also being discarded. This no
3668 longer happens. I'm not sure if this situation could ever has previously
3669 arisen.
3670
367161. Added the "/MX" feature to lists of hosts in the manualroute and query
3672 program routers.
3673
367462. Whenever Exim generates a new message, it now adds an Auto-Submitted:
3675 header. This is something that is recommended in a new Internet Draft, and
3676 is something that is documented as being done by Sendmail. There are two