Reduce rfc1413_query_timeout default from 30s to 5s.
[exim.git] / doc / doc-txt / ChangeLog
CommitLineData
8def5aaf 1$Cambridge: exim/doc/doc-txt/ChangeLog,v 1.308 2006/02/20 16:31:48 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
214e2000 210
5de37277
PH
211Exim version 4.60
212-----------------
213
cc38ddbf
PH
214PH/01 Two changes to the default runtime configuration:
215
216 (1) Move the checks for relay_from_hosts and authenticated clients from
217 after to before the (commented out) DNS black list checks.
218
219 (2) Add control=submission to the relay_from_hosts and authenticated
220 clients checks, on the grounds that messages accepted by these
221 statements are most likely to be submissions.
5de37277 222
72fdd6ae
PH
223PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
224
225 (1) Generate an error if the third argument for the ${prvs expansion is
226 not a single digit.
227
228 (2) Treat a missing third argument of ${prvscheck as if it were an empty
229 string.
230
231 (3) Reset the variables that are obtained from the first argument of
232 ${prvscheck and used in the second argument before leaving the code,
233 because their memory is reclaimed, so using them afterwards may do
234 silly things.
235
236 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
237 one (it's much easier than Tom thought :-).
238
239 (5) Because of (4), we can now allow for the use of $prvscheck_result
240 inside the third argument.
cb9328de 241
cb741023
PH
242PH/03 For some reason, the default setting of PATH when running a command from
243 a pipe transport was just "/usr/bin". I have changed it to
244 "/bin:/usr/bin".
245
f174f16e
PH
246PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
247 anything to be listed in the output from -bV.
b2f5a032 248
c25242d7
PH
249PH/05 When a filter generated an autoreply, the entire To: header line was
250 quoted in the delivery log line, like this:
251
252 => >A.N.Other <ano@some.domain> <original@ddress> ...
253
254 This has been changed so that it extracts the operative address. There
255 may be more than one such address. If so, they are comma-separated, like
256 this:
257
258 => >ano@some.domain,ona@other.domain <original@ddress> ...
259
82c19f95
PH
260PH/06 When a client host used a correct literal IP address in a HELO or EHLO
261 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
262 not being looked up in the rDNS to get a host name, Exim was showing the
263 IP address twice in Received: lines, even though the IP addresses were
264 identical. For example:
265
266 Received: from [1.2.3.4] (helo=[1.2.3.4])
267
268 However, if the real host name was known, it was omitting the HELO data
269 if it matched the actual IP address. This has been tidied up so that it
270 doesn't show the same IP address twice.
271
d7ffbc12
PH
272PH/07 When both +timestamp and +memory debugging was on, the value given by
273 $tod_xxx expansions could be wrong, because the tod_stamp() function was
274 called by the debug printing, thereby overwriting the timestamp buffer.
275 Debugging no longer uses the tod_stamp() function when +timestamp is set.
276
9f526266
PH
277PH/08 When the original message was included in an autoreply transport, it
278 always said "this is a copy of the message, including all the headers",
279 even if body_only or headers_only was set. It now gives an appropriate
280 message.
281
87fcc8b9
PH
282PH/09 Applied a patch from the Sieve maintainer which:
283
284 o fixes some comments
285 o adds the (disabled) notify extension core
286 o adds some debug output for the result of if/elsif tests
287 o points to the current vacation draft in the documentation
288 and documents the missing references header update
289
290 and most important:
291
292 o fixes a bug in processing the envelope test (when testing
293 multiple envelope elements, the last element determinted the
294 result)
295
456682f5
PH
296PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
297 Electronic Mail") by including:
298
299 Auto-submitted: auto-generated
300
301 in the messages that it generates (bounce messages and others, such as
302 warnings). In the case of bounce messages for non-SMTP mesages, there was
303 also a typo: it was using "Auto_submitted" (underscore instead of
304 hyphen). Since every message generated by Exim is necessarily in response
305 to another message, thes have all been changed to:
306
307 Auto-Submitted: auto-replied
308
309 in accordance with these statements in the RFC:
310
311 The auto-replied keyword:
312
313 - SHOULD be used on messages sent in direct response to another
314 message by an automatic process,
315
316 - MUST NOT be used on manually-generated messages,
317
318 - MAY be used on Delivery Status Notifications (DSNs) and Message
319 Disposition Notifications (MDNs),
320
321 - MUST NOT be used on messages generated by automatic or periodic
322 processes, except for messages which are automatic responses to
323 other messages.
324
3e46c1aa
PH
325PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
326 to the default Received: header definition.
456682f5 327
49826d12
PH
328PH/12 Added log selector acl_warn_skipped (default on).
329
eba0c039
PH
330PH/13 After a successful wildlsearch lookup, discard the values of numeric
331 variables because (a) they are in the wrong storage pool and (b) even if
332 they were copied, it wouldn't work properly because of the caching.
333
a0d6ba8a
PH
334PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
335 checking when decoding. Apparently there are clients that generate
336 overlong encoded strings. Why am I not surprised?
337
f0917727
PH
338PH/15 If the first argument of "${if match_address" was not empty, but did not
339 contain an "@" character, Exim crashed. Now it writes a panic log message
340 and treats the condition as false.
341
096fee00
PH
342PH/16 In autoreply, treat an empty string for "once" the same as unset.
343
024bd3c2
PH
344PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
345 extension "envelope-auth". The code is finished and in agreement with
346 other implementations, but there is no documentation so far and in fact,
347 nobody wrote the draft yet. This extension is currently #undef'ed, thus
348 not changing the active code.
349
350 Print executed "if" and "elsif" statements when debugging is used. This
351 helps a great deal to understand what a filter does.
352
353 Document more things not specified clearly in RFC3028. I had all this
354 sorted out, when out of a sudden new issues came to my mind. Oops."
355
df199fec
PH
356PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
357 (Bugzilla #53).
358
d27f1df3
PH
359PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
360 canonical form (as documented). However, after a host name lookup from
361 the IP address, check_host() was doing a simple string comparison with
362 addresses acquired from the DNS when checking that the found name did
363 have the original IP as one of its addresses. Since any found IPv6
364 addresses are likely to be in abbreviated form, the comparison could
365 fail. Luckily, there already exists a function for doing the comparison
366 by converting both addresses to binary, so now that is used instead of
367 the text comparison.
368
96776534
PH
369PH/20 There was another similar case to PH/19, when a complete host name was
370 given in a host list; looking up its IP address could give an abbreviated
371 form, whereas the current host's name might or might not be abbreviated.
372 The same fix has been applied.
373
5de37277 374
9a799bc0
PH
375Exim version 4.54
376-----------------
377
378PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
379 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
380 It now does.
381
99a4b039
PH
382PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
383 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
384
385PH/03 Typo: missing ".o" in src/pcre/Makefile.
386
4b233853
PH
387PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
388 header line, restrict the check to what is listed in RFCs 2369 and 2929.
389 Also, for "Auto-Submitted", treat anything other than "no" as
390 non-personal, in accordance with RFC 3834. (Previously it treated
391 anything starting "auto-" as non-personal.)
392
8857ccfd
PH
393TF/01 The control=submission/name=... option had a problem with syntax
394 errors if the name included a slash character. The /name= option
395 now slurps the rest of the string, so it can include any characters
396 but it must come last in the list of options (after /sender_retain
397 or /domain=).
398
433a2980
PH
399PH/05 Some modifications to the interface to the fake nameserver for the new
400 testing suite.
401
3e46c1aa 402
9a799bc0 403
e3a311ba
TK
404Exim version 4.53
405-----------------
406
407TK/01 Added the "success_on_redirect" address verification option. See
408 NewStuff for rationale and an example.
409
13b685f9
PH
410PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
411
395ff96d
PH
412PH/02 Patch to exigrep to allow it to work on syslog lines.
413
5b68f6e4
PH
414PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
415 fread() to skip over the body file's header line, because in Cygwin the
416 header line is locked and is inaccessible.
417
1ab52c69
PH
418PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
419 co-exist for some time) to make it clear that it is the Exim ID that is
420 referenced, not the Message-ID: header line.
421
b07e6aa3
PH
422PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
423 string_format() function, because snprintf() does not exist on all
424 operating systems.
425
254e032f
PH
426PH/06 The use of forbid_filter_existstest now also locks out the use of the
427 ${stat: expansion item.
428
3af76a81
PH
429PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
430 protocol synchronization error", to keep the pedants happy.
431
2548ba04
PH
432PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
433 well as for IRIX systems, when gcc is being used. See the host.c source
434 file for comments.
435
b6c6011d
PH
436PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
437
cf39cf57
PH
438PH/10 Named domain lists were not working if used in a queue_smtp_domains
439 setting.
440
f1513293
PH
441PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
442 transport and to the smtp transport in LMTP mode.
443
727549a4
PH
444TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
445
af46795e
PH
446PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
447 run a filter in a subprocess. This could lead to confusion in subsequent
448 lookups in the parent process. There should also be a search_tidyup() at
449 the end of the subprocess.
450
d7b47fd0
PH
451PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
452 only if the host matched helo_try_verify_hosts, which caused the
453 verification to occur when the EHLO/HELO command was issued. The ACL just
454 tested the remembered result. Now, if a previous verification attempt has
455 not happened, "verify = helo" does it there and then.
456
ee744174
JJ
457JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
458
b582ab87
PH
459TK/03 Fix log output including CR from clamd.
460
41a13e0a
PH
461PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
462 exist provoked a memory error which could cause a segfault.
463
f625cc5a
PH
464PH/15 Installed PCRE 6.2
465
466PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
467
21f7af35
PH
468PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
469 of the problem. Specifically, suggested +O2 rather than +O1 for the
470 HP-UX compiler.
471
31480e42
PH
472PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
473
2d280592
PH
474PH/20 If a delivery was routed to a non-standard port by means of an SRV
475 record, the port was not correctly logged when the outgoing_port log
476 selector was set (it logged the transort's default port).
477
7cd1141b
PH
478PH/21 Added support for host-specific ports to manualroute, queryprogram,
479 fallback_hosts, and "hosts" in the smtp transport.
480
481PH/22 If the log selector "outgoing_port" is set, the port is now also given on
482 host errors such as "Connection refused".
483
750af86e
PH
484PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
485 authentication with radiusclient 0.4.9:
486
487 - Error returned from rc_read_config was caught wrongly
488 - Username/password not passed on to radius server due to wrong length.
489
490 The presumption is that some radiusclient API changes for 4.51/PH/17
491 were not taken care of correctly. The code is still untested by me (my
492 Linux distribution still has 0.3.2 of radiusclient), but it was
493 contributed by a Radius user.
494
495PH/24 When doing a callout, the value of $domain wasn't set correctly when
496 expanding the "port" option of the smtp transport.
497
4304270b
TK
498TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
499 while reading a MIME header. Thanks to Tom Hughes for a patch.
500
750af86e
PH
501PH/24 Include config.h inside local_scan.h so that configuration settings are
502 available.
503
64ffc24f
PH
504PH/25 Make $smtp_command_argument available after all SMTP commands. This means
505 that in an ACL for RCPT (for example), you can examine exactly what was
506 received.
507
5dd9625b
PH
508PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
509 commands, but it was not correctly comparing the address with the actual
510 client host address. Thus, it would show the EHLO address in Received:
511 header lines when this was not necessary.
512
5591031b
PH
513PH/27 Added the % operator to ${eval:}.
514
ba18e66a
PH
515PH/28 Exim tries to create and chdir to its spool directory when it starts;
516 it should be ignoring failures (because with -C, for example, it has lost
517 privilege). It wasn't ignoring creation failures other than "already
518 exists".
519
9cec981f
PH
520PH/29 Added "crypteq" to the list of supported features that Exim outputs when
521 -bV or -d is used.
522
aa2b5c79
PH
523PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
524 because an input line was too long, either on its own, or by virtue of
1509d3a8
PH
525 too many continuations, the temporary file was not being removed, and the
526 return code was incorrect.
aa2b5c79 527
48a53b7f
PH
528PH/31 Missing "BOOL" in function definition in filtertest.c.
529
1c59d63b
PH
530PH/32 Applied Sieve patches from the maintainer.
531
671012da
TK
532TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
533
1509d3a8
PH
534PH/33 Added "verify = not_blind".
535
536PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
537 Local/Makefile (with some defaults set). These are used in built scripts
538 such as exicyclog, but they have never been used in the exim_install
539 script (though there are many overriding facilities there). I have
540 arranged that the exim_install script now takes note of these two
541 settings.
542
543PH/35 Installed configuration files for Dragonfly.
544
2fe1a124
PH
545PH/36 When a locally submitted message by a trusted user did not contain a
546 From: header, and the sender address was obtained from -f or from an SMTP
547 MAIL command, and the trusted user did not use -F to supply a sender
548 name, $originator_name was incorrectly used when constructing a From:
549 header. Furthermore, $originator_name was used for submission mode
550 messages from external hosts without From: headers in a similar way,
551 which is clearly wrong.
552
8800895a
PH
553PH/37 Added control=suppress_local_fixups.
554
ccfdb010
PH
555PH/38 When log_selector = +received_sender was set, and the addition of the
556 sender made the log line's construction buffer exactly full, or one byte
557 less than full, an overflow happened when the terminating "\n" was
558 subsequently added.
559
1130bfb0
PH
560PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
561 when the result of a list match is failure because a DNS lookup failed.
562
ebcb507f
PH
563PH/40 RM_COMMAND is now used in the building process.
564
c35e155c
PH
565PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
566 the "build-* directories that it finds.
567
95d1f782
PH
568PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
569 address in a domain literal was a prefix of an interface address.
570
fd6de02e
PH
571PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
572 when verifying a sender address, unless rewrite_headers is false.
573
58de37c5
PH
574PH/44 Wrote a long comment about why errors_to addresses are verified as
575 recipients, not senders.
576
261cf466
TF
577TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
578 the ratelimit ACL was added.
579
3ee512ff
PH
580PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
581
e08c430f
PH
582PH/46 Added extra information about PostgreSQL errors to the error string.
583
bef5a11f
PH
584PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
585 avoiding the need to install special zones in a real server. This is
586 backwards compatible; if it can't find the fake resolver, it drops back.
587 Thus, both old and new test suites can be run.
588
7546de58
TF
589TF/02 Added util/ratelimit.pl
590
e5d5a95f
TF
591TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
592 clock is set back in time.
593
2e88a017
TF
594TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
595 Candler <B.Candler@pobox.com>.
596
a5f65aa4
TF
597TF/05 The fix for PH/43 was not completely correct; widen_domains is always
598 OK for addresses that are the result of redirections.
599
e7726cbf
PH
600PH/48 A number of further additions for the benefit of the new test suite,
601 including a fake gethostbyname() that interfaces to the fake DNS resolver
602 (see PH/47 above).
603
a7fdad5b
TF
604TF/06 The fix for widen_domains has also been applied to qualify_single and
605 search_parents which are the other dnslookup options that can cause
606 header rewrites.
607
6af56900
PH
608PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
609 type ("H").
610
0925ede6
PH
611PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
612
66afa403
TF
613TF/07 Exim produced the error message "an SRV record indicated no SMTP
614 service" if it encountered an MX record with an empty target hostname.
615 The message is now "an MX or SRV record indicated no SMTP service".
616
0154e85a
TF
617TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
618 if the DNS of the sending site is misconfigured. This is quite a
619 common situation. This change restores the behaviour of treating a
620 helo verification defer as a failure.
621
16f12c76
PH
622PH/51 If self=fail was set on a router, the bounce message did not include the
623 actual error message.
624
bbe902f0 625
e5a9dba6
PH
626Exim version 4.52
627-----------------
628
629TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
630
22c3b60b
PH
631PH/01 When a transport filter timed out in a pipe delivery, and the pipe
632 command itself ended in error, the underlying message about the transport
633 filter timeout was being overwritten with the pipe command error. Now the
634 underlying error message should be appended to the second error message.
635
06a9b4b5
PH
636TK/01 Fix poll() being unavailable on Mac OSX 10.2.
637
c1ac6996
PH
638PH/02 Reduce the amount of output that "make" produces by default. Full output
639 can still be requested.
640
9c7a242c
PH
641PH/03 The warning log line about a condition test deferring for a "warn" verb
642 was being output only once per connection, rather than after each
643 occurrence (because it was using the same function as for successful
644 "warn" verbs). This seems wrong, so I have changed it.
645
87ba3f5f
PH
646TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
647 it should not have, which might have caused a crash in the right
648 circumstances, but probably never did.
649
650PH/04 Installed a modified version of Tony Finch's patch to make submission
651 mode fix the return path as well as the Sender: header line, and to
652 add a /name= option so that you can make the user's friendly name appear
653 in the header line.
654
29aba418
TF
655TF/03 Added the control = fakedefer ACL modifier.
656
fe0dab11
TF
657TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
658 Mark Lowes for thorough testing.
870f6ba8 659
11d337a4
TK
660TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
661
662TK/03 Merged latest SRS patch from Miles Wilton.
663
415c8f3b
PH
664PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
665 with the definition in sysexits.h (which is #included earlier).
666 Fortunately, Exim does not actually use EX_OK. The code used to try to
667 preserve the sysexits.h value, by assumimg that macro definitions were
668 scanned for macro replacements. I have been disabused of this notion,
669 so now the code just undefines EX_OK before #including unistd.h.
11d337a4 670
958541e9
PH
671PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
672 in the smtp transport. When a block could not be written in a single
673 write() function, the timeout was being re-applied to each part-write.
674 This seems wrong - if the receiver was accepting one byte at a time it
675 would take for ever. The timeout is now adjusted when this happens. It
676 doesn't have to be particularly precise.
677
c206415f
TK
678TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
679 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
680
2a4be8f9
PH
681PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
682 without a domain if the check to <postmaster@domain> fails.
683
1cba11c5
SC
684SC/01 Eximstats: added -xls and the ability to specify output files
685 (patch written by Frank Heydlauf).
686
687SC/02 Eximstats: use FileHandles for outputing results.
688
689SC/03 Eximstats: allow any combination of xls, txt, and html output.
690
691SC/04 Eximstats: fixed display of large numbers with -nvr option
692
693SC/05 Eximstats: fixed merging of reports with empty tables.
694
695SC/06 Eximstats: added the -include_original_destination flag
696
697SC/07 Eximstats: removed tabs and trailing whitespace.
698
1005d00e
TK
699TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
700
701TK/06 MBOX spool code: Add real "From " MBOX separator line
702 so the .eml file is really in mbox format (even though
703 most programs do not really care). Patch from Alex Miller.
704
705TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
706 The latter is generated from $received_to and is only set if the
707 message has one envelope recipient. SA can use these headers,
708 obviously out-of-the-box. Patch from Alex Miller.
709
9b4768fa
PH
710PH/08 The ${def test on a variable was returning false if the variable's
711 value was "0", contrary to what the specification has always said!
712 The result should be true unless the variable is empty.
713
714PH/09 The syntax error of a character other than { following "${if
715 def:variable_name" (after optional whitespace) was not being diagnosed.
716 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
717 accidental colon was present, for example, could give incorrect results.
718
0d7eb84a
PH
719PH/10 Tidied the code in a number of places where the st_size field of a stat()
720 result is used (not including appendfile, where other changes are about
721 to be made).
722
723PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
724 This involved changing a lot of size variables from int to off_t. It
725 should work with maildirs and everything.
726
40727bee
TK
727TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
728 spamd dying while we are connected to it.
729
554d2369
TF
730TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
731 <jgh@wizmail.org>
732
1f922db1
PH
733PH/12 Applied Alex Kiernan's patch for the API change for the error callback
734 function for BDB 4.3.
735
ef213c3b
PH
736PH/13 Changed auto_thaw such that it does not apply to bounce messages.
737
8ac170f3
PH
738PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
739 the sources for PCRE have been re-arranged and more files are now
740 involved.
741
b1c749bb
PH
742PH/15 The code I had for printing potentially long long variables in PH/11
743 above was not the best (it lost precision). The length of off_t variables
744 is now inspected at build time, and an appropriate printing format (%ld
c6c2dc1d
PH
745 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
746 to be "long long int" or "long int". This is needed for the internal
747 formatting function string_vformat().
b1c749bb 748
4aac9b49
PH
749PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
750 the configuration file to be ":syslog", then the script "guesses" where
751 the logs files are, rather than using the compiled in default. In our
752 case the guess is not the same as the compiled default, so the script
753 suddenly stopped working when I started to use syslog. The patch checks
754 to see if log_file_path is "". If so, it attempts to read it from exim
755 with no configuration file to get the compiled in version, before it
756 falls back to the previous guessing code."
757
294520c8
TK
758TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
759 implementing BATV in an Exim configuration. See NewStuff for the gory
760 details.
761
5bd022fe
PH
762PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
763 Makefile that are specific to HP-UX.
764
90e9ce59
PH
765PH/18 If the "use_postmaster" option was set for a recipient callout together
766 with the "random" option, the postmaster address was used as the MAIL
767 FROM address for the random test, but not for the subsequent recipient
768 test. It is now used for both.
769
5ea81592
PH
770PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
771 patch removes a few documentation additions to RFC 3028, because the
772 latest draft now contains them. It adds the new en;ascii-case comparator
773 and a new error check for 8bit text in MIME parts. Comparator and
774 require names are now matched exactly. I enabled the subaddress
775 extension, but it is not well tested yet (read: it works for me)."
776
c6c2dc1d
PH
777PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
778 rework some of the code of TK/09 above to avoid the hardwired use of
779 "%lld" and "long long". Replaced the call to snprintf() with a call to
780 string_vformat().
781
fffffe4c
PH
782PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
783 records point to non-existent hosts", "retry timeout exceeded", and
784 "retry time not reached for any host after a long failure period".
ca02eafb 785
9a26b6b2
PH
786PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
787 experimental DomainKeys support:
788
789 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
790 (2) On an error such as an illegally used "control", the wrong name for
791 the control was given.
792
793 These problems did NOT occur unless DomainKeys support was compiled.
794
4aee0225
PH
795PH/23 Added daemon_startup_retries and daemon_startup_sleep.
796
32d668a5
PH
797PH/24 Added ${if match_ip condition.
798
8187c3f3
PH
799PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
800 databases so that it will be absolutely obvious if a crash occurs in the
801 DB library. This is a regular occurrence (often caused by mis-matched
802 db.h files).
803
ff790e47 804PH/26 Insert a lot of missing (void) casts for functions such as chown(),
f1e894f3
PH
805 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
806 picked up on a user's system that detects such things. There doesn't seem
807 to be a gcc warning option for this - only an attribute that has to be
808 put on the function's prototype. It seems that in Fedora Core 4 they have
809 set this on a number of new functions. No doubt there will be more in due
810 course.
ff790e47 811
5417f6d1
PH
812PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
813 specify a transport. However, if an address that was verified by such a
814 router was the subject of a callout, Exim crashed because it tried to
815 read the rcpt_include_affixes from the non-existent transport. Now it
816 just assumes that the setting of that option is false. This bug was
817 introduced by 4.51/PH/31.
818
59cf8544
PH
819PH/28 Changed -d+all to exclude +memory, because that information is very
820 rarely of interest, but it makes the output a lot bigger. People tend to
821 do -d+all out of habit.
822
e7ad8a65
PH
823PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
824 code in os-type was giving problems when libc.so lives in lib64, like on
825 x86_64 Fedora Core.
826
ade42478
PH
827PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
828 aren't the modern standard, and it seems that some systems' include files
829 don't always have them. Exim was already checking for some of the newer
830 ones like T_AAAA, and defining it itself. I've added checks for all the
831 record types that Exim uses.
832
182ad5cf
PH
833PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
834 not automatically generating a new one, as it is supposed to. This
835 prevented TLS from working. If the file did exist, but contained invalid
836 data, a new version was generated, as expected. It was only the case of a
837 non-existent file that was broken.
838
b0d9fc80
TK
839TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
840 with a change in libdomainkeys > 0.64.
841
842TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
843 from DNS. If the selector record carries the flag, it now has
844 precedence over the domain-wide flag.
845
846TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
847
47c7a64a
PH
848PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
849 the use of a port name that isn't defined in /etc/services) occurred, the
850 message was deferred as in a normal delivery, and thus remained on the
851 spool, instead of being failed because of the mua_wrapper setting. This
852 is now fixed, and I tidied up some of the mua_wrapper messages at the
853 same time.
854
a388bce4
SC
855SC/08 Eximstats: whilst parsing the mainlog(s), store information about
856 the messages in a hash of arrays rather than using individual hashes.
857 This is a bit cleaner and results in dramatic memory savings, albeit
858 at a slight CPU cost.
859
860SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
861 as requested by Marc Sherman.
862
863SC/10 Eximstats: added histograms for user specified patterns as requested
864 by Marc Sherman.
865
0793e4ed
SC
866SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
867
c58b88df
PH
868PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
869 fopen() in the content-scanning modules that did not already have it.
870
e7ad8a65 871
7982096b
PH
872Exim version 4.51
873-----------------
874
1a46a8c5
PH
875TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
876 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
877
2f079f46 878TK/02 Fix ACL "control" statement not being available in MIME ACL.
1a46a8c5
PH
879
880TK/03 Fix ACL "regex" condition not being available in MIME ACL.
881
882PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
883 to test Sieve filters that use "vacation".
884
885PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
886 that changes the way the GnuTLS parameters are stored in the cache file.
887 The new format can be generated externally. For backward compatibility,
888 if the data in the cache doesn't make sense, Exim assumes it has read an
889 old-format file, and it generates new data and writes a new file. This
890 means that you can't go back to an older release without removing the
891 file.
892
893PH/03 A redirect router that has both "unseen" and "one_time" set does not
894 work if there are any delivery delays because "one_time" forces the
895 parent to be marked "delivered", so its unseen clone is never tried
896 again. For this reason, Exim now forbids the simultaneous setting of
897 these two options.
898
899PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
900 redirected to themselves ("homonym" addresses). Read the long ChangeLog
901 entry if you want to know the details. The fix, however, neglected to
902 consider the case when local delivery batching is involved. The test for
903 "previously delivered" was not happening when checking to see if an
904 address could be batched with a previous (undelivered) one; under
905 certain circumstances this could lead to multiple deliveries to the same
c2c19e9d 906 address.
1a46a8c5
PH
907
908PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
909 in its include files, and this causes problems building Exim.
910
911PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
912 header_syntax) but Exim was just ignoring anything given after a slash.
913 In particular, this caused confusion with an attempt to use "verify =
914 reverse_host_lookup/defer_ok". An error is now given when options are
915 supplied for verify items that do not have them. (Maybe reverse_host_
916 lookup should have a defer_ok option, but that's a different point.)
917
918PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
919 defined by RFC 821) to 2048, because there were problems with some AUTH
920 commands, and RFC 1869 says the size should be increased for extended
921 SMTP commands that take arguments.
922
923PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
924 Finch).
925
926PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
927 "unknown" error; now it says that the functionality isn't in the binary.
8d67ada3 928
49c2d5ea
PH
929PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
930 an address' error message when a string expansion fails (syntax or
f331f3b6
PH
931 whatever). Otherwise the password may appear in the log. Following change
932 PH/42 below, there is no longer a chance of it appearing in a bounce
933 message.
49c2d5ea 934
bf759a8b
PH
935PH/11 Installed exipick version 20050225.0 from John Jetmore.
936
83364d30
PH
937PH/12 If the last host in a fallback_hosts list was multihomed, only the first
938 of its addresses was ever tried. (Bugzilla bug #2.)
939
7999bbd7
PH
940PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
941 the result incorrectly in the debug output. (It correctly added a newline
942 to what was transported.)
943
7dbf77c9
PH
944TF/01 Added $received_time.
945
74e0617f
PH
946PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
947 commented out examples of how to interface to a virus scanner and to
948 SpamAssassin. Also added commented examples of av_scanner and
949 spamd_address settings.
950
2f079f46
PH
951PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
952 and controls are allowed in which ACLs. There were a couple of minor
953 errors. Some of the entries in the conditions table (which is a table of
954 where they are NOT allowed) were getting very unwieldy; rewrote them as a
955 negation of where the condition IS allowed.
956
8c841523
PH
957PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
958
7766a4f0
PH
959PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
960 header file does not have a version number, so I've had to invent a new
961 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
962 API. The code is untested by me (my Linux distribution still has 0.3.2 of
963 radiusclient), but it was contributed by a Radius user.
964
8b417f2c
PH
965PH/18 Installed Lars Mainka's patch for the support of CRL collections in
966 files or directories, for OpenSSL.
967
901f42cb
PH
968PH/19 When an Exim process that is running as root has to create an Exim log
969 file, it does so in a subprocess that runs as exim:exim so as to get the
970 ownership right at creation (otherwise, other Exim processes might see
971 the file with the wrong ownership). There was no test for failure of this
972 fork() call, which would lead to the process getting stuck as it waited
973 for a non-existent subprocess. Forks do occasionally fail when resources
974 run out. I reviewed all the other calls to fork(); they all seem to check
975 for failure.
976
f9b9210e
PH
977PH/20 When checking for unexpected SMTP input at connect time (before writing
978 the banner), Exim was not dealing correctly with a non-positive return
979 from the read() function. If the client had disconnected by this time,
980 the result was a log entry for a synchronization error with an empty
981 string after "input=" when read() returned zero. If read() returned -1
982 (an event I could not check), uninitialized data bytes were printed.
983 There were reports of junk text (parts of files, etc) appearing after
984 "input=".
985
54cdb463
PH
986PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
987
cf00dad6
PH
988PH/22 Added support for macro redefinition, and (re)definition in between
989 driver and ACL definitions.
990
acb1b346
PH
991PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
992 forgetting to use the resulting value; it was using the unexpanded value.
993
c5ddb310
PH
994PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
995 hadn't been configured. The fix is from Juergen Kreileder, who
996 understands it better than I do:
997
998 "Here's what I see happening with three configured cyrus_sasl
999 authenticators configured (plain, login, cram-md5):
1000
1001 On startup auth_cyrus_sasl_init() gets called for each of these.
1002 This means three calls to sasl_listmech() without a specified mech_list.
1003 => SASL tests which mechs of all available mechs actually work
1004 => three warnings about OTP not working
1005 => the returned list contains: plain, login, cram-md5, digest-md5, ...
1006
1007 With the patch, sasl_listmech() also gets called three times. But now
1008 SASL's mech_list option is set to the server_mech specified in the the
1009 authenticator. Or in other words, the answer from sasl_listmech()
1010 gets limited to just the mech you're testing for (which is different
1011 for each call.)
1012 => the return list contains just 'plain' or 'login', 'cram-md5' or
1013 nothing depending on the value of ob->server_mech.
1014
1015 I've just tested the patch: Authentication still works fine,
1016 unavailable mechs specified in the exim configuration are still
1017 caught, and the auth.log warnings about OTP are gone."
1018
31619da6
PH
1019PH/25 When debugging is enabled, the contents of the command line are added
1020 to the debugging output, even when log_selector=+arguments is not
1021 specified.
1022
bebaf0fc
PH
1023PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
1024 answer is "GNU", and only if the return is "GNU/something" is the answer
1025 "Linux".
1026
475fe28a
PH
1027PH/27 $acl_verify_message is now set immediately after the failure of a
1028 verification in an ACL, and so is available in subsequent modifiers. In
1029 particular, the message can be preserved by coding like this:
1030
1031 warn !verify = sender
1032 set acl_m0 = $acl_verify_message
1033
1034 Previously, $acl_verify_message was set only while expanding "message"
1035 and "log_message" when a very denied access.
1036
7e8bec7a
PH
1037PH/28 Modified OS/os.c-Linux with
1038
1039 -#ifndef OS_LOAD_AVERAGE
1040 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
1041
1042 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
1043 nomenclature these days.)
1044
e4a89c47
PH
1045PH/29 Installed patch from the Sieve maintainer that adds the options
1046 sieve_useraddress and sieve_subaddress to the redirect router.
1047
5ca2a9a1
PH
1048PH/30 In these circumstances:
1049 . Two addresses routed to the same list of hosts;
1050 . First host does not offer TLS;
1051 . First host accepts first address;
1052 . First host gives temporary error to second address;
1053 . Second host offers TLS and a TLS session is established;
1054 . Second host accepts second address.
1055 Exim incorrectly logged both deliveries with the TLS parameters (cipher
1056 and peerdn, if requested) that were in fact used only for the second
1057 address.
7e8bec7a 1058
c688b954
PH
1059PH/31 When doing a callout as part of verifying an address, Exim was not paying
1060 attention to any local part prefix or suffix that was matched by the
1061 router that accepted the address. It now behaves in the same way as it
1062 does for delivery: the affixes are removed from the local part unless
1063 rcpt_include_affixes is set on the transport.
1064
fed77020
PH
1065PH/32 Add the sender address, as F=<...>, to the log line when logging a
1066 timeout during the DATA phase of an incoming message.
1067
7fe1560f
PH
1068PH/33 Sieve envelope tests were broken for match types other than :is. I have
1069 applied a patch sanctioned by the Sieve maintainer.
c688b954 1070
ebb6e6d5
PH
1071PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
1072 the uid or gid is negative. A case of a negative gid caused this to be
1073 noticed. The fix allows for either to be negative.
1074
9c4e8f60
PH
1075PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
1076 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
1077 overlooked.
1078
1079PH/36 The change PH/12 above was broken. Fixed it.
1080
d7174846
PH
1081PH/37 Exim used to check for duplicate addresses in the middle of routing, on
1082 the grounds that routing the same address twice would always produce the
1083 same answer. This might have been true once, but it is certainly no
1084 longer true now. Routing a child address may depend on the previous
1085 routing that produced that child. Some complicated redirection strategies
1086 went wrong when messages had multiple recipients, and made Exim's
1087 behaviour dependent on the order in which the addresses were given.
1088
1089 I have moved the duplicate checking until after the routing is complete.
1090 Exim scans the addresses that are assigned to local and remote
1091 transports, and removes any duplicates. This means that more work will be
1092 done, as duplicates will always all be routed, but duplicates are
1093 presumably rare, so I don't expect this is of any significance.
1094
1095 For deliveries to pipes, files, and autoreplies, the duplicate checking
1096 still happens during the routing process, since they are not going to be
1097 routed further.
1098
cfe75fc3
PH
1099PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
1100 It corrects a timeout issue with spamd. This is Ian's comment: "The
1101 background is that sometimes spamd either never reads data from a
1102 connection it has accepted, or it never writes response data. The exiscan
1103 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
1104 blindly assumes that writes won't block so it may never time out."
1105
be22d70e
PH
1106PH/39 Allow G after quota size as well as K and M.
1107
0612b098
PH
1108PH/40 The value set for $authenticated_id in an authenticator may not contain
1109 binary zeroes or newlines because the value is written to log lines and
1110 to spool files. There was no check on this. Now the value is run through
1111 the string_printing() function so that such characters are converted to
1112 printable escape sequences.
1113
2e0c1448
PH
1114PH/41 $message_linecount is a new variable that contains the total number of
1115 lines in the message. Compare $body_linecount, which is the count for the
1116 body only.
1117
447d236c
PH
1118PH/42 Exim no longer gives details of delivery errors for specific addresses in
1119 bounce and delay warning messages, except in certain special cases, which
1120 are as follows:
1121
1122 (a) An SMTP error message from a remote host;
1123 (b) A message specified in a :fail: redirection;
1124 (c) A message specified in a "fail" command in a system filter;
1125 (d) A message specified in a FAIL return from the queryprogram router;
1126 (e) A message specified by the cannot_route_message router option.
1127
1128 In these cases only, Exim does include the error details in bounce and
1129 warning messages. There are also a few cases where bland messages such
1130 as "unrouteable address" or "local delivery error" are given.
1131
d20976dc
PH
1132PH/43 $value is now also set for the "else" part of a ${run expansion.
1133
f656d135
PH
1134PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
1135 being worked on, but at least Exim now implements the latest version to
1136 play with."
1137
2e2a30b4
PH
1138PH/45 In a pipe transport, although a timeout while waiting for the pipe
1139 process to complete was treated as a delivery failure, a timeout while
1140 writing the message to the pipe was logged, but erroneously treated as a
1141 successful delivery. Such timeouts include transport filter timeouts. For
1142 consistency with the overall process timeout, these timeouts are now
1143 treated as errors, giving rise to delivery failures by default. However,
1144 there is now a new Boolean option for the pipe transport called
1145 timeout_defer, which, if set TRUE, converts the failures into defers for
1146 both kinds of timeout. A transport filter timeout is now identified in
1147 the log output.
1148
9176e9f0
PH
1149PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
1150 systems where "make" and "gmake" are different, calling "gmake" at top
1151 level broke things. I've arranged for the value of $(MAKE) to be passed
1152 from the Makefile to this script so that it can call the same version of
1153 "make".
1154
7982096b 1155
bbe902f0
PH
1156A note about Exim versions 4.44 and 4.50
1157----------------------------------------
1158
1159Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
1160changes of various kinds. As a consequence, a big documentation update was
1161needed. This delayed the release for rather longer than seemed good, especially
1162in the light of a couple of (minor) security issues. Therefore, the changes
1163that fixed bugs were backported into 4.43, to create a 4.44 maintenance
1164release. So 4.44 and 4.50 are in effect two different branches that both start
1165from 4.43.
1166
1167I have left the 4.50 change log unchanged; it contains all the changes since
11684.43. The change log for 4.44 is below; many of its items are identical to
1169those for 4.50. This seems to be the most sensible way to preserve the
1170historical information.
1171
1172
f7b63901 1173Exim version 4.50
495ae4b0
PH
1174-----------------
1175
5fe762f6
PH
1176 1. Minor wording change to the doc/README.SIEVE file.
1177
139059f6 1178 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
5fe762f6 1179 computation of the current number of files was incorrect.
495ae4b0 1180
7086e875
PH
1181 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
1182 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
1183 place.
1184
35af9f61
PH
1185 4. Give more explanation in the error message when the command for a transport
1186 filter fails to execute.
1187
b668c215
PH
1188 5. There are several places where Exim runs a non-Exim command in a
1189 subprocess. The SIGUSR1 signal should be disabled for these processes. This
1190 was being done only for the command run by the queryprogram router. It is
1191 now done for all such subprocesses. The other cases are: ${run, transport
1192 filters, and the commands run by the lmtp and pipe transports.
1193
a494b1e1
PH
1194 6. Added CONFIGURE_GROUP build-time option.
1195
1196 7. Some older OS have a limit of 256 on the maximum number of file
1197 descriptors. Exim was using setrlimit() to set 1000 as a large value
1198 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
1199 systems. I've change it so that if it can't get 1000, it tries for 256.
35edf2ff 1200
c5fcb476
PH
1201 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
1202 was an oversight, and furthermore, ever since the addition of extra
1203 controls (e.g. 4.43/32), the checks on when to allow different forms of
1204 "control" were broken. There should now be diagnostics for all cases when a
1205 control that does not make sense is encountered.
1206
69358f02
PH
1207 9. Added the /retain_sender option to "control=submission".
1208
5be20824
PH
120910. $recipients is now available in the predata ACL (oversight).
1210
eb2c0248
PH
121111. Tidy the search cache before the fork to do a delivery from a message
1212 received from the command line. Otherwise the child will trigger a lookup
1213 failure and thereby defer the delivery if it tries to use (for example) a
1214 cached ldap connection that the parent has called unbind on.
1215
2a3eea10
PH
121612. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
1217 of $address_data from the recipient verification was clobbered by the
1218 sender verification.
1219
122013. The value of address_data from a sender verification is now available in
1221 $sender_address_data in subsequent conditions in the ACL statement.
1222
23c7ff99
PH
122314. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
1224
4deaf07d
PH
122515. Added a new option "connect=<time>" to callout options, to set a different
1226 connection timeout.
1227
926e1192
PH
122816. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
1229 was its contents. (It was OK if the option was not defined at all.)
1230
650edc6f
PH
123117. A "Completed" log line is now written for messages that are removed from
1232 the spool by the -Mrm option.
1233
2c7db3f5
PH
123418. New variables $sender_verify_failure and $recipient_verify_failure contain
1235 information about exactly what failed.
1236
3d235903
PH
123719. Added -dd to debug only the daemon process.
1238
7c7ad977
PH
123920. Incorporated Michael Haardt's patch to ldap.c for improving the way it
1240 handles timeouts, both on the server side and network timeouts. Renamed the
1241 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
1242
981756db
PH
124321. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
1244 It is now set to "smtps".
1245
d4eb88df
PH
124622. $host_address is now set to the target address during the checking of
1247 ignore_target_hosts.
1248
124923. When checking ignore_target_hosts for an ipliteral router, no host name was
1250 being passed; this would have caused $sender_host_name to have been used if
1251 matching the list had actually called for a host name (not very likely,
1252 since this list is usually IP addresses). A host name is now passed as
1253 "[x.x.x.x]".
1254
7d468ab8
PH
125524. Changed the calls that set up the SIGCHLD handler in the daemon to use the
1256 code that specifies a non-restarting handler (typically sigaction() in
1257 modern systems) in an attempt to fix a rare and obscure crash bug.
1258
125925. Narrowed the window for a race in the daemon that could cause it to ignore
1260 SIGCHLD signals. This is not a major problem, because they are used only to
1261 wake it up if nothing else does.
1262
62c0818f
PH
126326. A malformed maildirsize file could cause Exim to calculate negative values
1264 for the mailbox size or file count. Odd effects could occur as a result.
1265 The maildirsize information is now recalculated if the size or filecount
1266 end up negative.
1267
26034054
PH
126827. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
1269 support for a long time. Removed HAVE_SYS_VFS_H.
1270
af66f652
PH
127128. Installed the latest version of exipick from John Jetmore.
1272
90af77f4
PH
127329. In an address list, if the pattern was not a regular expression, an empty
1274 subject address (from a bounce message) matched only if the pattern was an
1275 empty string. Non-empty patterns were not even tested. This was the wrong
1276 because it is perfectly reasonable to use an empty address as part of a
1277 database query. An empty address is now tested by patterns that are
1278 lookups. However, all the other forms of pattern expect the subject to
1279 contain a local part and a domain, and therefore, for them, an empty
1280 address still always fails if the pattern is not itself empty.
1281
d8ef3577
PH
128230. Exim went into a mad DNS loop when attempting to do a callout where the
1283 host was specified on an smtp transport, and looking it up yielded more
1284 than one IP address.
1285
5cb8cbc6
PH
128631. Re-factored the code for checking spool and log partition space into a
1287 function that finds that data and another that does the check. The former
1288 is then used to implement four new variables: $spool_space, $log_space,
1289 $spool_inodes, and $log_inodes.
1290
14702f5b
PH
129132. The RFC2047 encoding function was originally intended for short strings
1292 such as real names; it was not keeping to the 75-character limit for
1293 encoded words that the RFC imposes. It now respects the limit, and
1294 generates multiple encoded words if necessary. To be on the safe side, I
1295 have increased the buffer size for the ${rfc2047: expansion operator from
1296 1024 to 2048 bytes.
1297
063b1e99
PH
129833. It is now permitted to omit both strings after an "if" condition; if the
1299 condition is true, the result is "true". As before, when the second string
1300 is omitted, a false condition yields an empty string. This makes it less
1301 cumbersome to write custom ACL and router conditions.
1302
652e1b65
PH
130334. Failure to deliver a bounce message always caused it to be frozen, even if
1304 there was an errors_to setting on the router. The errors_to setting is now
1305 respected.
1306
6f0c9a4f
PH
130735. If an IPv6 address is given for -bh or -bhc, it is now converted to the
1308 canonical form (fully expanded) before being placed in
1309 $sender_host_address.
1310
33397d19
PH
131136. The table in the code that translates DNS record types into text (T_A to
1312 "A" for instance) was missing entries for NS and CNAME. It is just possible
1313 that this could have caused confusion if both these types were looked up
1314 for the same domain, because the text type is used as part of Exim's
1315 per-process caching. But the chance of anyone hitting this buglet seems
1316 very small.
1317
7bb56e1f
PH
131837. The dnsdb lookup has been extended in a number of ways.
1319
1320 (1) There is a new type, "zns", which walks up the domain tree until it
1321 finds some nameserver records. It should be used with care.
1322
ea3bc19b
PH
1323 (2) There is a new type, "mxh", which is like "mx" except that it returns
1324 just the host names, not the priorities.
1325
1326 (3) It is now possible to give a list of domains (or IP addresses) to be
ff4dbb19
PH
1327 looked up. The behaviour when one of the lookups defers can be
1328 controlled by a keyword.
7bb56e1f 1329
ea3bc19b 1330 (4) It is now possible to specify the separator character for use when
7bb56e1f 1331 multiple records are returned.
33397d19 1332
0bcb2a0e
PH
133338. The dnslists ACL condition has been extended: it is now possible to supply
1334 a list of IP addresses and/or domains to be looked up in a particular DNS
1335 domain.
1336
2ac0e484
PH
133739. Added log_selector=+queue_time_overall.
1338
4e1fde53
PH
133940. When running the queue in the test harness, wait just a tad after forking a
1340 delivery process, to get repeatability of debugging output.
1341
de365ded
PH
134241. Include certificate and key file names in error message when GnuTLS fails
1343 to set them up, because the GnuTLS error message doesn't include the name
1344 of the failing file when there is a problem reading it.
1345
f05da2e8
PH
134642. Allow both -bf and -bF in the same test run.
1347
d6453af2
PH
134843. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
1349
f7b63901
PH
135044. The "Exiscan patch" is now merged into the mainline Exim source.
1351
135245. Sometimes the final signoff response after QUIT could fail to get
1353 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
1354 < 0 before doing a fflush(). This bug looks as though it goes back to the
1355 introduction of TLS in release 3.20, but "sometimes" must have been rare
1356 because the tests only now provoked it.
1357
a444213a
PH
135846. Reset the locale to "C" after calling embedded Perl, in case it was changed
1359 (this can affect the format of dates).
1360
0ec020ea
PH
136147. exim_tidydb, when checking for the continued existence of a message for
1362 which it has found a message-specific retry record, was not finding
1363 messages that were in split spool directories. Consequently, it was
1364 deleting retry records that should have stayed in existence.
1365
b1206957
PH
136648. Steve fixed some bugs in eximstats.
1367
136849. The SPA authentication driver was not abandoning authentication and moving
1369 on to the next authenticator when an expansion was forced to fail,
1370 contradicting the general specification for all authenticators. Instead it
1371 was generating a temporary error. It now behaves as specified.
1372
26dd5a95
PH
137350. The default ordering of permitted cipher suites for GnuTLS was pessimal
1374 (the order specifies the preference for clients). The order is now AES256,
1375 AES128, 3DES, ARCFOUR128.
1376
343b2385
PH
137751. Small patch to Sieve code - explicitly set From: when generating an
1378 autoreply.
1379
1c5466b9
PH
138052. Exim crashed if a remote delivery caused a very long error message to be
1381 recorded - for instance if somebody sent an entire SpamAssassin report back
1382 as a large number of 550 error lines. This bug was coincidentally fixed by
1383 increasing the size of one of Exim's internal buffers (big_buffer) that
1384 happened as part of the Exiscan merge. However, to be on the safe side, I
1385 have made the code more robust (and fixed the comments that describe what
1386 is going on).
1387
55ee9ee3
PH
138853. Now that there can be additional text after "Completed" in log lines (if
1389 the queue_time_overall log selector is set), a one-byte patch to exigrep
1390 was needed to allow it to recognize "Completed" as not the last thing in
1391 the line.
1392
d38f8232
PH
139354. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
1394 patch that reportedly fixes this has been added. I am not expert enough to
1395 create a test for it. This is what the patch creator wrote:
1396
1397 "I found a little strange behaviour of ldap code when working with
1398 Windows 2003 AD Domain, where users was placed in more than one
1399 Organization Units. When I tried to give exim partial DN, the exit code
1400 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
1401 But simultaneously result of request was absolutely normal ldap result,
1402 so I produce this patch..."
1403
3295e65b
PH
1404 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
1405 REFERENCE, so I have modified the code to exclude the patch when that macro
1406 is not defined.
1407
7102e136
PH
140855. Some experimental protocols are using DNS PTR records for new purposes. The
1409 keys for these records are domain names, not reversed IP addresses. The
b975ba52
PH
1410 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
1411 leaves it alone. Component reversal etc. now happens only for IP addresses.
ea3a6f44 1412 CAN-2005-0021
7102e136 1413
3ca0ba97
PH
141456. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
1415
c2bcbe20
PH
141657. Double the size of the debug message buffer (to 2048) so that more of very
1417 long debug lines gets shown.
1418
18ce445d
PH
141958. The exicyclog utility now does better if the number of log files to keep
1420 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
1421
1f5b4c3d
PH
142259. Two changes related to the smtp_active_hostname option:
1423
1424 (1) $smtp_active_hostname is now available as a variable.
1425 (2) The default for smtp_banner uses $smtp_active_hostname instead
1426 of $primary_hostname.
1427
b975ba52
PH
142860. The host_aton() function is supposed to be passed a string that is known
1429 to be a valid IP address. However, in the case of IPv6 addresses, it was
1430 not checking this. This is a hostage to fortune. Exim now panics and dies
1431 if the condition is not met. A case was found where this could be provoked
85b87bc2
PH
1432 from a dnsdb PTR lookup with an IPv6 address that had more than 8
1433 components; fortuitously, this particular loophole had already been fixed
1434 by change 4.50/55 above.
1435
1436 If there are any other similar loopholes, the new check in host_aton()
1437 itself should stop them being exploited. The report I received stated that
1438 data on the command line could provoke the exploit when Exim was running as
1439 exim, but did not say which command line option was involved. All I could
1440 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
1441 running as the user.
ea3a6f44 1442 CAN-2005-0021
85b87bc2
PH
1443
144461. There was a buffer overflow vulnerability in the SPA authentication code
1445 (which came originally from the Samba project). I have added a test to the
1446 spa_base64_to_bits() function which I hope fixes it.
ea3a6f44 1447 CAN-2005-0022
b975ba52 1448
17ffcae7
PH
144962. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
1450 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
1451
d95f9fdb
PH
145263. The daemon start-up calls getloadavg() while still root for those OS that
1453 need the first call to be done as root, but it missed one case: when
1454 deliver_queue_load_max is set with deliver_drop_privilege. This is
1455 necessary for the benefit of the queue runner, because there is no re-exec
1456 when deliver_drop_privilege is set.
1457
86b8287f
PH
145864. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
1459 This has been fixed.
1460
60dc5e56
PH
146165. Caching of lookup data for "hosts =" ACL conditions, when a named host list
1462 was in use, was not putting the data itself into the right store pool;
1463 consequently, it could be overwritten for a subsequent message in the same
1464 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
1465 the caching.)
1466
533244af
PH
146766. Added hosts_max_try_hardlimit to the smtp transport, default 50.
1468
a5a28604
PH
146967. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
1470 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
1471 the function were treating the return as a boolean value, which happened to
1472 work because 0=false and not-0=true, but is not correct code.
1473
7e634d24
PH
147468. The host_aton() function was not handling scoped IPv6 addresses (those
1475 with, for example, "%eth0" on the end) correctly.
1476
3e11c26b
PH
147769. Fixed some compiler warnings in acl.c for the bitmaps specified with
1478 negated items (that is, ~something) in unsigned ints. Some compilers
1479 apparently mutter when there is no cast.
1480
6729cf78
PH
148170. If an address verification called from an ACL failed, and did not produce a
1482 user-specific message (i.e. there was only a "system" message), nothing was
1483 put in $acl_verify_message. In this situation, it now puts the system
1484 message there.
1485
00f00ca5
PH
148671. Change 4.23/11 added synchronization checking at the start of an SMTP
1487 session; change 4.31/43 added the unwanted input to the log line - except
1488 that it did not do this in the start of session case. It now does.
1489
c9bdd01c
PH
149072. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
1491 This is wrong and can cause the other end to generate a synchronization
1492 error if it is another Exim or anything else that does the synchronization
1493 check. A QUIT command is no longer sent after a timeout.
1494
d43194df
PH
149573. $host_lookup_deferred has been added, to make it easier to detect DEFERs
1496 during host lookups.
1497
fe5b5d0b
PH
149874. The defer_ok option of callout verification was not working if it was used
1499 when verifying addresses in header lines, that is, for this case:
1500
1501 verify = header_sender/callout=defer_ok
1502
76a2d7ba
PH
150375. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
1504 those file descriptors could be used for SMTP connections. If anything
1505 wrote to stderr (the example that came up was "warn" in embedded Perl), it
1506 could be sent to the SMTP client, causing chaos. The daemon now opens
1507 stdin, stdout, and stderr to /dev/null when it puts itself into the
1508 background.
1509
151076. Arrange for output from Perl's "warn" command to be written to Exim's main
1511 log by default. The user can override this with suitable Perl magic.
1512
04f7d5b9
PH
151377. The use of log_message on a "discard" ACL verb, which is supposed to add to
1514 the log message when discard triggers, was not working for the DATA ACL or
1515 for the non-SMTP ACL.
1516
bc60667e
PH
151778. Error message wording change in sieve.c.
1518
bb6e88ff
PH
151979. If smtp_accept_max_per_host was set, the number of connections could be
1520 restricted to fewer than expected, because the daemon was trying to set up
1521 a new connection before checking whether the processes handling previous
1522 connections had finished. The check for completed processes is now done
1523 earlier. On busy systems, this bug wouldn't be noticed because something
1524 else would have woken the daemon, and it would have reaped the completed
1525 process earlier.
1526
1e70f85b
PH
152780. If a message was submitted locally by a user whose login name contained one
1528 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
1529 It caused a spool format error. I have fixed the spool reading code. A
1530 related problem was that the "from" clause in the Received: line became
1531 illegal because of the space(s). It is now covered by ${quote_local_part.
1532
153381. Included the latest eximstats from Steve (adds average sizes to HTML Top
1534 tables).
1535
4e01f9d6
PH
153682. Updated OS/Makefile-AIX as per message from Mike Meredith.
1537
1ee1cef2
PH
153883. Patch from Sieve maintainer to fix unterminated string problem in
1539 "vacation" handling.
1540
6e2b4ccc
PH
154184. Some minor changes to the Linux configuration files to help with other
1542 OS variants using glibc.
1543
8e669ac1
PH
154485. One more patch for Sieve to update vacation handling to latest spec.
1545
495ae4b0 1546
bbe902f0
PH
1547----------------------------------------------------
1548See the note above about the 4.44 and 4.50 releases.
1549----------------------------------------------------
1550
1551
1552Exim version 4.44
1553-----------------
1554
1555 1. Change 4.43/35 introduced a bug that caused file counts to be
1556 incorrectly computed when quota_filecount was set in an appendfile
1557 transport
1558
1559 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
1560 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
1561 place.
1562
1563 3. Give more explanation in the error message when the command for a transport
1564 filter fails to execute.
1565
1566 4. There are several places where Exim runs a non-Exim command in a
1567 subprocess. The SIGUSR1 signal should be disabled for these processes. This
1568 was being done only for the command run by the queryprogram router. It is
1569 now done for all such subprocesses. The other cases are: ${run, transport
1570 filters, and the commands run by the lmtp and pipe transports.
1571
1572 5. Some older OS have a limit of 256 on the maximum number of file
1573 descriptors. Exim was using setrlimit() to set 1000 as a large value
1574 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
1575 systems. I've change it so that if it can't get 1000, it tries for 256.
1576
1577 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
1578 was an oversight, and furthermore, ever since the addition of extra
1579 controls (e.g. 4.43/32), the checks on when to allow different forms of
1580 "control" were broken. There should now be diagnostics for all cases when a
1581 control that does not make sense is encountered.
1582
1583 7. $recipients is now available in the predata ACL (oversight).
1584
1585 8. Tidy the search cache before the fork to do a delivery from a message
1586 received from the command line. Otherwise the child will trigger a lookup
1587 failure and thereby defer the delivery if it tries to use (for example) a
1588 cached ldap connection that the parent has called unbind on.
1589
1590 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
1591 of $address_data from the recipient verification was clobbered by the
1592 sender verification.
1593
159410. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
1595 was its contents. (It was OK if the option was not defined at all.)
1596
159711. A "Completed" log line is now written for messages that are removed from
1598 the spool by the -Mrm option.
1599
160012. $host_address is now set to the target address during the checking of
1601 ignore_target_hosts.
1602
160313. When checking ignore_target_hosts for an ipliteral router, no host name was
1604 being passed; this would have caused $sender_host_name to have been used if
1605 matching the list had actually called for a host name (not very likely,
1606 since this list is usually IP addresses). A host name is now passed as
1607 "[x.x.x.x]".
1608
160914. Changed the calls that set up the SIGCHLD handler in the daemon to use the
1610 code that specifies a non-restarting handler (typically sigaction() in
1611 modern systems) in an attempt to fix a rare and obscure crash bug.
1612
161315. Narrowed the window for a race in the daemon that could cause it to ignore
1614 SIGCHLD signals. This is not a major problem, because they are used only to
1615 wake it up if nothing else does.
1616
161716. A malformed maildirsize file could cause Exim to calculate negative values
1618 for the mailbox size or file count. Odd effects could occur as a result.
1619 The maildirsize information is now recalculated if the size or filecount
1620 end up negative.
1621
162217. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
1623 support for a long time. Removed HAVE_SYS_VFS_H.
1624
ea3a6f44 162518. Updated exipick to current release from John Jetmore.
bbe902f0
PH
1626
162719. Allow an empty sender to be matched against a lookup in an address list.
1628 Previously the only cases considered were a regular expression, or an
1629 empty pattern.
1630
163120. Exim went into a mad DNS lookup loop when doing a callout where the
1632 host was specified on the transport, if the DNS lookup yielded more than
1633 one IP address.
1634
ea3a6f44
NM
163521. The RFC2047 encoding function was originally intended for short strings
1636 such as real names; it was not keeping to the 75-character limit for
1637 encoded words that the RFC imposes. It now respects the limit, and
1638 generates multiple encoded words if necessary. To be on the safe side, I
1639 have increased the buffer size for the ${rfc2047: expansion operator from
1640 1024 to 2048 bytes.
bbe902f0 1641
ea3a6f44
NM
164222. Failure to deliver a bounce message always caused it to be frozen, even if
1643 there was an errors_to setting on the router. The errors_to setting is now
1644 respected.
bbe902f0
PH
1645
164623. If an IPv6 address is given for -bh or -bhc, it is now converted to the
1647 canonical form (fully expanded) before being placed in
1648 $sender_host_address.
1649
165024. Updated eximstats to version 1.33
1651
ea3a6f44
NM
165225. Include certificate and key file names in error message when GnuTLS fails
1653 to set them up, because the GnuTLS error message doesn't include the name
1654 of the failing file when there is a problem reading it.
bbe902f0
PH
1655
165626. Expand error message when OpenSSL has problems setting up cert/key files.
ea3a6f44 1657 As per change 25.
bbe902f0 1658
ea3a6f44
NM
165927. Reset the locale to "C" after calling embedded Perl, in case it was changed
1660 (this can affect the format of dates).
bbe902f0 1661
ea3a6f44
NM
166228. exim_tidydb, when checking for the continued existence of a message for
1663 which it has found a message-specific retry record, was not finding
1664 messages that were in split spool directories. Consequently, it was
1665 deleting retry records that should have stayed in existence.
bbe902f0
PH
1666
166729. eximstats updated to version 1.35
1668 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
1669 1.35 - bugfix such that pie charts by volume are generated correctly
1670
ea3a6f44
NM
167130. The SPA authentication driver was not abandoning authentication and moving
1672 on to the next authenticator when an expansion was forced to fail,
1673 contradicting the general specification for all authenticators. Instead it
1674 was generating a temporary error. It now behaves as specified.
bbe902f0 1675
ea3a6f44
NM
167631. The default ordering of permitted cipher suites for GnuTLS was pessimal
1677 (the order specifies the preference for clients). The order is now AES256,
1678 AES128, 3DES, ARCFOUR128.
bbe902f0 1679
ea3a6f44
NM
168031. Small patch to Sieve code - explicitly set From: when generating an
1681 autoreply.
bbe902f0 1682
ea3a6f44
NM
168332. Exim crashed if a remote delivery caused a very long error message to be
1684 recorded - for instance if somebody sent an entire SpamAssassin report back
1685 as a large number of 550 error lines. This bug was coincidentally fixed by
1686 increasing the size of one of Exim's internal buffers (big_buffer) that
1687 happened as part of the Exiscan merge. However, to be on the safe side, I
1688 have made the code more robust (and fixed the comments that describe what
1689 is going on).
bbe902f0 1690
ea3a6f44
NM
169133. Some experimental protocols are using DNS PTR records for new purposes. The
1692 keys for these records are domain names, not reversed IP addresses. The
1693 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
1694 leaves it alone. Component reversal etc. now happens only for IP addresses.
bbe902f0
PH
1695 CAN-2005-0021
1696
ea3a6f44
NM
169734. The host_aton() function is supposed to be passed a string that is known
1698 to be a valid IP address. However, in the case of IPv6 addresses, it was
1699 not checking this. This is a hostage to fortune. Exim now panics and dies
1700 if the condition is not met. A case was found where this could be provoked
1701 from a dnsdb PTR lookup with an IPv6 address that had more than 8
1702 components; fortuitously, this particular loophole had already been fixed
1703 by change 4.50/55 or 4.44/33 above.
1704
1705 If there are any other similar loopholes, the new check in host_aton()
1706 itself should stop them being exploited. The report I received stated that
1707 data on the command line could provoke the exploit when Exim was running as
1708 exim, but did not say which command line option was involved. All I could
1709 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
1710 running as the user.
bbe902f0
PH
1711 CAN-2005-0021
1712
ea3a6f44
NM
171335. There was a buffer overflow vulnerability in the SPA authentication code
1714 (which came originally from the Samba project). I have added a test to the
1715 spa_base64_to_bits() function which I hope fixes it.
bbe902f0
PH
1716 CAN-2005-0022
1717
ea3a6f44
NM
171836. The daemon start-up calls getloadavg() while still root for those OS that
1719 need the first call to be done as root, but it missed one case: when
1720 deliver_queue_load_max is set with deliver_drop_privilege. This is
1721 necessary for the benefit of the queue runner, because there is no re-exec
1722 when deliver_drop_privilege is set.
bbe902f0 1723
ea3a6f44
NM
172437. Caching of lookup data for "hosts =" ACL conditions, when a named host list
1725 was in use, was not putting the data itself into the right store pool;
1726 consequently, it could be overwritten for a subsequent message in the same
1727 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
1728 the caching.)
bbe902f0 1729
ea3a6f44
NM
173038. Sometimes the final signoff response after QUIT could fail to get
1731 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
1732 < 0 before doing a fflush(). This bug looks as though it goes back to the
1733 introduction of TLS in release 3.20, but "sometimes" must have been rare
1734 because the tests only now provoked it.
bbe902f0
PH
1735
1736
495ae4b0
PH
1737Exim version 4.43
1738-----------------
1739
1740 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
1741 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
1742 Now it returns an integer. A number of calls were still expecting a T/F
1743 return. Fortuitously, in all cases, the tests worked in OK situations,
1744 which is the norm. However, things would have gone wrong on any write
1745 failures on the smtp file descriptor. This function is used when sending
1746 messages over SMTP and also when doing verify callouts.
1747
1748 2. When Exim is called to do synchronous delivery of a locally submitted
1749 message (the -odf or -odi options), it no longer closes stderr before doing
1750 the delivery.
1751
1752 3. Implemented the mua_wrapper option.
1753
1754 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
1755
1756 5. Implemented the functions header_remove(), header_testname(),
1757 header_add_at_position(), and receive_remove_recipient(), and exported them
1758 to local_scan().
1759
1760 6. If an ACL "warn" statement specified the addition of headers, Exim already
1761 inserted X-ACL-Warn: at the start if there was no header name. However, it
1762 was not making this test for the second and subsequent header lines if
1763 there were newlines in the string. This meant that an invalid header could
1764 be inserted if Exim was badly configured.
1765
1766 7. Allow an ACL "warn" statement to add header lines at the start or after all
1767 the Received: headers, as well as at the end.
1768
1769 8. Added the rcpt_4xx retry error code.
1770
1771 9. Added postmaster_mailfrom=xxx to callout verification option.
1772
177310. Added mailfrom=xxxx to the callout verification option, for verify=
1774 header_sender only.
1775
177611. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
1777 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
1778
177912. Inserted some casts to stop certain compilers warning when using pointer
1780 differences as field lengths or precisions in printf-type calls (mostly
1781 affecting debugging statements).
1782
178313. Added optional readline() support for -be (dynamically loaded).
1784
178514. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
1786 same clock tick as a message's arrival, so that its received time was the
1787 same as the "first fail" time on the retry record, and that message
1788 remained on the queue past the ultimate address timeout, every queue runner
1789 would try a delivery (because it was past the ultimate address timeout) but
1790 after another failure, the ultimate address timeout, which should have then
1791 bounced the address, did not kick in. This was a "< instead of <=" error;
1792 in most cases the first failure would have been in the next clock tick
1793 after the received time, and all would be well.
1794
179515. The special items beginning with @ in domain lists (e.g. @mx_any) were not
1796 being recognized when the domain list was tested by the match_domain
1797 condition in an expansion string.
1798
179916. Added the ${str2b64: operator.
1800
180117. Exim was always calling setrlimit() to set a large limit for the number of
1802 processes, without checking whether the existing limit was already
1803 adequate. (It did check for the limit on file descriptors.) Furthermore,
1804 errors from getrlimit() and setrlimit() were being ignored. Now they are
1805 logged to the main and panic logs, but Exim does carry on, to try to do its
1806 job under whatever limits there are.
1807
180818. Imported PCRE 5.0.
1809
181019. Trivial typo in log message " temporarily refused connection" (the leading
1811 space).
1812
181320. If the log selector return_path_on_delivery was set and an address was
1814 redirected to /dev/null, the delivery process crashed because it assumed
1815 that a return path would always be set for a "successful" delivery. In this
1816 case, the whole delivery is bypassed as an optimization, and therefore no
1817 return path is set.
1818
181921. Internal re-arrangement: the function for sending a challenge and reading
1820 a response while authentication was assuming a zero-terminated challenge
1821 string. It's now changed to take a pointer and a length, to allow for
1822 binary data in such strings.
1823
182422. Added the cyrus_sasl authenticator (code supplied by MBM).
1825
182623. Exim was not respecting finduser_retries when seeking the login of the
1827 uid under which it was called; it was always trying 10 times. (The default
1828 setting of finduser_retries is zero.) Also, it was sleeping after the final
1829 failure, which is pointless.
1830
183124. Implemented tls_on_connect_ports.
1832
183325. Implemented acl_smtp_predata.
1834
183526. If the domain in control=submission is set empty, Exim assumes that the
1836 authenticated id is a complete email address when it generates From: or
1837 Sender: header lines.
1838
183927. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
1840 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
1841 chown and chgrp in /bin and hostname in /usr/bin.
1842
184328. Exim was keeping the "process log" file open after each use, just as it
1844 does for the main log. This opens the possibility of it remaining open for
1845 long periods when the USR1 signal hits a daemon. Occasional processlog
1846 errors were reported, that could have been caused by this. Anyway, it seems
1847 much more sensible not to leave this file open at all, so that is what now
1848 happens.
1849
185029. The long-running daemon process does not normally write to the log once it
1851 has entered its main loop, and it closes the log before doing so. This is
1852 so that log files can straightforwardly be renamed and moved. However,
1853 there are a couple of unusual error situations where the daemon does write
1854 log entries, and I had neglected to close the log afterwards.
1855
185630. The text of an SMTP error response that was received during a remote
1857 delivery was being truncated at 512 bytes. This is too short for some of
1858 the long messages that one sometimes sees. I've increased the limit to
1859 1024.
1860
186131. It is now possible to make retry rules that apply only when a message has a
1862 specific sender, in particular, an empty sender.
1863
186432. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
1865 it possible to be selective about when SMTP synchronization is enforced.
1866
186733. Added "control = caseful_local_part" and "control = "caselower_local_part".
1868
186932. Implemented hosts_connection_nolog.
1870
187133. Added an ACL for QUIT.
1872
187334. Setting "delay_warning=" to disable warnings was not working; it gave a
1874 syntax error.
1875
187635. Added mailbox_size and mailbox_filecount to appendfile.
1877
187836. Added control = no_multiline_responses to ACLs.
1879
188037. There was a bug in the logic of the code that waits for the clock to tick
1881 in the case where the clock went backwards by a substantial amount such
1882 that the microsecond fraction of "now" was more than the microsecond
1883 fraction of "then" (but the whole seconds number was less).
1884
188538. Added support for the libradius Radius client library this is found on
1886 FreeBSD (previously only the radiusclient library was supported).
1887
1888
1889Exim version 4.42
1890-----------------
1891
1892 1. When certain lookups returned multiple values in the form name=value, the
1893 quoting of the values was not always being done properly. Specifically:
1894 (a) If the value started with a double quote, but contained no whitespace,
1895 it was not quoted.
1896 (b) If the value contained whitespace other than a space character (i.e.
1897 tabs or newlines or carriage returns) it was not quoted.
1898 This fix has been applied to the mysql and pgsql lookups by writing a
1899 separate quoting function and calling it from the lookup code. The fix
1900 should probably also be applied to nisplus, ibase and oracle lookups, but
1901 since I cannot test any of those, I have not disturbed their existing code.
1902
1903 2. A hit in the callout cache for a specific address caused a log line with no
1904 reason for rejecting RCPT. Now it says "Previous (cached) callout
1905 verification failure".
1906
1907 3. There was an off-by-one bug in the queryprogram router. An over-long
1908 return line was truncated at 256 instead of 255 characters, thereby
1909 overflowing its buffer with the terminating zero. As well as fixing this, I
1910 have increased the buffer size to 1024 (and made a note to document this).
1911
1912 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
1913 when Exim is waiting for an SMTP response from a remote server, Exim
1914 restarts its select() call on the socket, thereby resetting its timeout.
1915 This is not a problem when such interrupts are rare. Somebody set up a cron
1916 job to run exiwhat every 2 minutes, which is less than the normal select()
1917 timeout (5 or 10 minutes). This meant that the select() timeout never
1918 kicked in because it was always reset. I have fixed this by comparing the
1919 time when an interrupt arrives with the time at the start of the first call
1920 to select(). If more time than the timeout has elapsed, the interrupt is
1921 treated as a timeout.
1922
1923 5. Some internal re-factoring in preparation for the addition of Sieve
1924 extensions (by MH). In particular, the "personal" test is moved to a
1925 separate function, and given an option for scanning Cc: and Bcc: (which is
1926 not set for Exim filters).
1927
1928 6. When Exim created an email address using the login of the caller as the
1929 local part (e.g. when creating a From: or Sender: header line), it was not
1930 quoting the local part when it contained special characters such as @.
1931
1932 7. Installed new OpenBSD configuration files.
1933
1934 8. Reworded some messages for syntax errors in "and" and "or" conditions to
1935 try to make them clearer.
1936
1937 9. Callout options, other than the timeout value, were being ignored when
1938 verifying sender addresses in header lines. For example, when using
1939
1940 verify = header_sender/callout=no_cache
1941
1942 the cache was (incorrectly) being used.
1943
194410. Added a missing instance of ${EXE} to the exim_install script; this affects
1945 only the Cygwin environment.
1946
194711. When return_path_on_delivery was set as a log selector, if different remote
1948 addresses in the same message used different return paths and parallel
1949 remote delivery occurred, the wrong values would sometimes be logged.
1950 (Whenever a remote delivery process finished, the return path value from
1951 the most recently started remote delivery process was logged.)
1952
195312. RFC 3848 specifies standard names for the "with" phrase in Received: header
1954 lines when AUTH and/or TLS are in use. This is the "received protocol"
1955 field. Exim used to use "asmtp" for authenticated SMTP, without any
1956 indication (in the protocol name) for TLS use. Now it follows the RFC and
1957 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
1958 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
1959 names appear in log lines as well as in Received: header lines.
1960
196113. Installed MH's patches for Sieve to add the "copy" and "vacation"
1962 extensions, and comparison tests, and to fix some bugs.
1963
196414. Changes to the "personal" filter test:
1965
1966 (1) The test was buggy in that it was just doing the equivalent of
1967 "contains" tests on header lines. For example, if a user's address was
1968 anne@some.where, the "personal" test would incorrectly be true for
1969
1970 To: susanne@some.where
1971
1972 This test is now done by extracting each address from the header in turn,
1973 and checking the entire address. Other tests that are part of "personal"
1974 are now done using regular expressions (for example, to check local parts
1975 of addresses in From: header lines).
1976
1977 (2) The list of non-personal local parts in From: addresses has been
1978 extended to include "listserv", "majordomo", "*-request", and "owner-*",
1979 taken from the Sieve specification recommendations.
1980
1981 (3) If the message contains any header line starting with "List-" it is
1982 treated as non-personal.
1983
1984 (4) The test for "circular" in the Subject: header line has been removed
1985 because it now seems ill-conceived.
1986
198715. Minor typos in src/EDITME comments corrected.
1988
198916. Installed latest exipick from John Jetmore.
1990
199117. If headers_add on a router specified a text string that was too long for
1992 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
1993 of string_sprintf() is now avoided.
1994
199518. $message_body_size was not set (it was always zero) when running the DATA
1996 ACL and the local_scan() function.
1997
199819. For the "mail" command in an Exim filter, no default was being set for
1999 the once_repeat time, causing a random time value to be used if "once" was
2000 specified. (If the value happened to be <= 0, no repeat happened.) The
2001 default is now 0s, meaning "never repeat". The "vacation" command was OK
2002 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
2003 (I found it when inspecting the code).
2004
200520. There is now an overall timeout for performing a callout verification. It
2006 defaults to 4 times the callout timeout, which applies to individual SMTP
2007 commands during the callout. The overall timeout applies when there is more
2008 than one host that can be tried. The timeout is checked before trying the
2009 next host. This prevents very long delays if there are a large number of
2010 hosts and all are timing out (e.g. when the network connections are timing
2011 out). The value of the overall timeout can be changed by specifying an
2012 additional sub-option for "callout", called "maxwait". For example:
2013
2014 verify = sender/callout=5s,maxwait=20s
2015
201621. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
2017 to the end before writing, but this should make it even safer).
2018
201922. Exim was forgetting that it had advertised PIPELINING for the second and
2020 subsequent messages on an SMTP connection. It was also not resetting its
2021 memory on STARTTLS and an internal HELO.
2022
202323. When Exim logs an SMTP synchronization error within a session, it now
2024 records whether PIPELINING has been advertised or not.
2025
202624. Added 3 instances of "(long int)" casts to time_t variables that were being
2027 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
2028 rather than long int.
2029
203025. Installed the latest Cygwin configuration files from the Cygwin maintainer.
2031
203226. Added the never_mail option to autoreply.
2033
2034
2035Exim version 4.41
2036-----------------
2037
2038 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
2039 crash if the getsockname() call failed; this can happen if a connection is
2040 closed very soon after it is established. The problem was simply in the
2041 order in which certain operations were done, causing Exim to try to write
2042 to the SMTP stream before it had set up the file descriptor. The bug has
2043 been fixed by making things happen in the correct order.
2044
2045
2046Exim version 4.40
2047-----------------
2048
2049 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
2050 before the connection was closed, thus losing the rejection response.
2051
2052 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
2053 some early Solaris releases, but causes trouble in current releases where
2054 socklen_t is defined.
2055
2056 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
2057 always exist.
2058
2059 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
2060 configured.
2061
2062 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
2063 that releasing the top of it at the end releases what was used for sub-
2064 expansions (unless the block got too big). However, discard this block if
2065 the first thing is a variable or header, so that we can use its block when
2066 it is dynamic (useful for very large $message_headers, for example).
2067
2068 6. Lookups now cache *every* query, not just the most recent. A new, separate
2069 store pool is used for this. It can be recovered when all lookup caches are
2070 flushed. Lookups now release memory at the end of their result strings.
2071 This has involved some general refactoring of the lookup sources.
2072
2073 7. Some code has been added to the store_xxx() functions to reduce the amount
2074 of flapping under certain conditions.
2075
2076 8. log_incoming_interface used to affect only the <= reception log lines. Now
2077 it causes the local interface and port to be added to several more SMTP log
2078 lines, for example "SMTP connection from", and rejection lines.
2079
2080 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
2081
208210. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
2083
208411. If $host_data was set by virtue of a hosts lookup in an ACL, its value
2085 could be overwritten at the end of the current message (or the start of a
2086 new message if it was set in a HELO ACL). The value is now preserved for
2087 the duration of the SMTP connection.
2088
208912. If a transport had a headers_rewrite setting, and a matching header line
2090 contained an unqualified address, that address was qualified, even if it
2091 did not match any rewriting rules. The underlying bug was that the values
2092 of the flags that permit the existence of unqualified sender and recipient
2093 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
2094 non-local messages, and by -bnq for local messages) were not being
2095 preserved with the message after it was received.
2096
209713. When Exim was logging an SMTP synchronization error, it could sometimes log
2098 "next input=" as part of the text comprising the host identity instead of
2099 the correct text. The code was using the same buffer for two different
2100 strings. However, depending on which order the printing function evaluated
2101 its arguments, the bug did not always show up. Under Linux, for example, my
2102 test suite worked just fine.
2103
210414. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
2105 doesn't work with some older versions of Perl. It has been changed to "my",
2106 which in any case is probably the better facility to use.
2107
210815. A really picky compiler found some instances of statements for creating
2109 error messages that either had too many or two few arguments for the format
2110 string.
2111
211216. The size of the buffer for calls to the DNS resolver has been increased
2113 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
2114 for addresses that have a lot of PTR records. This alleviates a problem; it
2115 does not fully solve it.
2116
211717. A dnsdb lookup for PTR records that receives more data than will fit in the
2118 buffer now truncates the list and logs the incident, which is the same
2119 action as happens when Exim is looking up a host name and its aliases.
2120 Previously in this situation something unpredictable would happen;
2121 sometimes it was "internal error: store_reset failed".
2122
212318. If a server dropped the connection unexpectedly when an Exim client was
2124 using GnuTLS and trying to read a response, the client delivery process
2125 crashed while trying to generate an error log message.
2126
212719. If a "warn" verb in an ACL added multiple headers to a message in a single
2128 string, for example:
2129
2130 warn message = H1: something\nH2: something
2131
2132 the text was added as a single header line from Exim's point of view
2133 though it ended up OK in the delivered message. However, searching for the
2134 second and subsequent header lines using $h_h2: did not work. This has been
2135 fixed. Similarly, if a system filter added multiple headers in this way,
2136 the routers could not see them.
2137
213820. Expanded the error message when iplsearch is called with an invalid key to
2139 suggest using net-iplsearch in a host list.
2140
214121. When running tests using -bh, any delays imposed by "delay" modifiers in
2142 ACLs are no longer actually imposed (and a message to that effect is
2143 output).
2144
214522. If a "gecos" field in a passwd entry contained escaped characters, in
2146 particular, if it contained a \" sequence, Exim got it wrong when building
2147 a From: or a Sender: header from that name. A second bug also caused
2148 incorrect handling when an unquoted " was present following a character
2149 that needed quoting.
2150
215123. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
2152 was not being matched caselessly.
2153
215424. Arranged for all hyphens in the exim.8 source to be escaped with
2155 backslashes.
2156
215725. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
2158 properly. Recipient callout cache records were still being keyed to include
2159 the sender, even when use_sender was set false. This led to far more
2160 callouts that were necessary. The sender is no longer included in the key
2161 when use_sender is false.
2162
216326. Added "control = submission" modifier to ACLs.
2164
216527. Added the ${base62d: operator to decode base 62 numbers.
2166
216728. dnsdb lookups can now access SRV records.
2168
216929. CONFIGURE_OWNER can be set at build time to define an alternative owner for
2170 the configuration file.
2171
217230. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
2173 (-v) mode. This makes the output for a verbose queue run more intelligible.
2174
217531. Added a use_postmaster feature to recipient callouts.
2176
217732. Added the $body_zerocount variable, containing the number of binary zero
2178 bytes in the message body.
2179
218033. The time of last modification of the "new" subdirectory is now used as the
2181 "mailbox time last read" when there is a quota error for a maildir
2182 delivery.
2183
218434. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
2185
218635. Added +ignore_unknown as a special item in host lists.
2187
218836. Code for decoding IPv6 addresses in host lists is now included, even if
2189 IPv6 support is not being compiled. This fixes a bug in which an IPv6
2190 address was recognized as an IP address, but was then not correctly decoded
2191 into binary, causing unexpected and incorrect effects when compared with
2192 another IP address.
2193
2194
2195Exim version 4.34
2196-----------------
2197
2198 1. Very minor rewording of debugging text in manualroute to say "list of
2199 hosts" instead of "hostlist".
2200
2201 2. If verify=header_syntax was set, and a header line with an unqualified
2202 address (no domain) and a large number of spaces between the end of the
2203 name and the colon was received, the reception process suffered a buffer
2204 overflow, and (when I tested it) crashed. This was caused by some obsolete
2205 code that should have been removed. The fix is to remove it!
2206
2207 3. When running in the test harness, delay a bit after writing a bounce
2208 message to get a bit more predictability in the log output.
2209
2210 4. Added a call to search_tidyup() just before forking a reception process. In
2211 theory, someone could use a lookup in the expansion of smtp_accept_max_
2212 per_host which, without the tidyup, could leave open a database connection.
2213
2214 5. Added the variables $recipient_data and $sender_data which get set from a
2215 lookup success in an ACL "recipients" or "senders" condition, or a router
2216 "senders" option, similar to $domain_data and $local_part_data.
2217
2218 6. Moved the writing of debug_print from before to after the "senders" test
2219 for routers.
2220
2221 7. Change 4.31/66 (moving the time when the Received: is generated) caused
2222 problems for message scanning, either using a data ACL, or using
2223 local_scan() because the Received: header was not generated till after they
2224 were called (in order to set the time as the time of reception completion).
2225 I have revised the way this works. The header is now generated after the
2226 body is received, but before the ACL or local_scan() are called. After they
2227 are run, the timestamp in the header is updated.
2228
2229
2230Exim version 4.33
2231-----------------
2232
2233 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
2234 before starting a queue runner without re-exec. This happened only when
2235 deliver_drop_privilege was set or when the Exim user was set to root. The
2236 effect of the bug was that timeouts during subsequent deliveries caused
2237 crashes instead of being properly handled. The handler is now left at its
2238 default (and expected) setting.
2239
2240 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
2241 message, again when deliver_drop_privilege is set or Exim is run as root.
2242 The bug described in (1) was not present in this case, but the tidying up
2243 of the other signals was missing. I have made the two cases consistent.
2244
2245 3. The ignore_target_hosts setting on a manualroute router was being ignored
2246 for hosts that were looked up using the /MX notation.
2247
2248 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
2249 in domain lists.
2250
2251 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
2252 operated on the sender address. After changing the $sender_address to <>
2253 for the sender address verify, Exim was re-instated it as the original
2254 (before rewriting) address, but remembering that it had rewritten it, so it
2255 wasn't rewriting it again. This bug also had the effect of breaking the
2256 sender address verification caching when the sender address was rewritten.
2257
2258 6. The ignore_target_hosts option was being ignored by the ipliteral router.
2259 This has been changed so that if the ip literal address matches
2260 ignore_target_hosts, the router declines.
2261
2262 7. Added expansion conditions match_domain, match_address, and match_local_
2263 part (NOT match_host).
2264
2265 8. The placeholder for the Received: header didn't have a length field set.
2266
2267 9. Added code to Exim itself and to exim_lock to test for a specific race
2268 condition that could lead to file corruption when using MBX delivery. The
2269 issue is with the lockfile that is created in /tmp. If this file is removed
2270 after a process has opened it but before that process has acquired a lock,
2271 there is the potential for a second process to recreate the file and also
2272 acquire a lock. This could lead to two Exim processes writing to the file
2273 at the same time. The added code performs the same test as UW imapd; it
2274 checks after acquiring the lock that its file descriptor still refers to
2275 the same named file.
2276
227710. The buffer for building added header lines was of fixed size, 8192 bytes.
2278 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
2279 when Exim is built.
2280
228111. Added the smtp_active_hostname option. If used, this will typically be made
2282 to depend on the incoming interface address. Because $interface_address is
2283 not set up until the daemon has forked a reception process, error responses
2284 that can happen earlier (such as "too many connections") no longer contain
2285 a host name.
2286
228712. If an expansion in a condition on a "warn" statement fails because a lookup
2288 defers, the "warn" statement is abandoned, and the next ACL statement is
2289 processed. Previously this caused the whole ACL to be aborted.
2290
229113. Added the iplsearch lookup type.
2292
229314. Added ident_timeout as a log selector.
2294
229515. Added tls_certificate_verified as a log selector.
2296
229716. Added a global option tls_require_ciphers (compare the smtp transport
2298 option of the same name). This controls incoming TLS connections.
2299
230017. I finally figured out how to make tls_require_ciphers do a similar thing
2301 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
2302 before starting the TLS session.
2303
230418. Tabs are now shown as \t in -bP output.
2305
230619. If the log selector return_path_on_delivery was set, Exim crashed when
2307 bouncing a message because it had too many Received: header lines.
2308
230920. If two routers both had headers_remove settings, and the first one included
2310 a superfluous trailing colon, the final name in the first list and the
2311 first name in the second list were incorrectly joined into one item (with a
2312 colon in the middle).
2313
2314
2315Exim version 4.32
2316-----------------
2317
2318 1. Added -C and -D options to the exinext utility, mainly to make it easier
2319 to include in the automated testing, but these could be helpful when
2320 multiple configurations are in use.
2321
2322 2. The exinext utility was not formatting the output nicely when there was
2323 an alternate port involved in the retry record key, nor when there was a
2324 message id as well (for retries that were specific to a specific message
2325 and a specific host). It was also confused by IPv6 addresses, because of
2326 the additional colons they contain. I have fixed the IPv4 problem, and
2327 patched it up to do a reasonable job for IPv6.
2328
2329 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
2330 delivery, the log line now contains "pipelined" if PIPELINING was used.
2331
2332 4. An SMTP transport process used to panic and die if the bind() call to set
2333 an explicit outgoing interface failed. This has been changed; it is now
2334 treated in the same way as a connect() failure.
2335
2336 5. A reference to $sender_host_name in the part of a conditional expansion
2337 that was being skipped was still causing a DNS lookup. This no longer
2338 occurs.
2339
2340 6. The def: expansion condition was not recognizing references to header lines
2341 that used bh_ and bheader_.
2342
2343 7. Added the _cache feature to named lists.
2344
2345 8. The code for checking quota_filecount in the appendfile transport was
2346 allowing one more file than it should have been.
2347
2348 9. For compatibility with Sendmail, the command line option
2349
2350 -prval:sval
2351
2352 is equivalent to
2353
2354 -oMr rval -oMs sval
2355
2356 and sets the incoming protocol and host name (for trusted callers). The
2357 host name and its colon can be omitted when only the protocol is to be set.
2358 Note the Exim already has two private options, -pd and -ps, that refer to
2359 embedded Perl. It is therefore impossible to set a protocol value of "d" or
2360 "s", but I don't think that's a major issue.
2361
236210. A number of refactoring changes to the code, none of which should affect
2363 Exim's behaviour:
2364
2365 (a) The number of logging options was getting close to filling up the
2366 32-bit word that was used as a bit map. I have split them into two classes:
2367 those that are passed in the argument to log_write(), and those that are
2368 only ever tested independently outside of that function. These are now in
2369 separate 32-bit words, so there is plenty of room for expansion again.
2370 There is no change in the user interface or the logging behaviour.
2371
2372 (b) When building, for example, log lines, the code previously used a
2373 macro that called string_cat() twice, in order to add two strings. This is
2374 not really sufficiently general. Furthermore, there was one instance where
2375 it was actually wrong because one of the argument was used twice, and in
2376 one call a function was used. (As it happened, calling the function twice
2377 did not affect the overall behaviour.) The macro has been replaced by a
2378 function that can join an arbitrary number of extra strings onto a growing
2379 string.
2380
2381 (c) The code for expansion conditions now uses a table and a binary chop
2382 instead of a serial search (which was left over from when there were very
2383 few conditions). Also, it now recognizes conditions like "pam" even when
2384 the relevant support is not compiled in: a suitably worded error message is
2385 given if an attempt is made to use such a condition.
2386
238711. Added ${time_interval:xxxxx}.
2388
238912. A bug was causing one of the ddress fields not to be passed back correctly
2390 from remote delivery subprocesses. The field in question was not being
2391 subsequently used, so this caused to problems in practice.
2392
239313. Added new log selectors queue_time and deliver_time.
2394
239514. Might have fixed a bug in maildirsizefile handling that threw up
2396 "unexpected character" debug warnings, and recalculated the data
2397 unnecessarily. In any case, I expanded the warning message to give more
2398 information.
2399
240015. Added the message "Restricted characters in address" to the statements in
2401 the default ACL that block characters like @ and % in local parts.
2402
240316. Change 71 for release 4.31 proved to be much less benign that I imagined.
2404 Three changes have been made:
2405
2406 (a) There was a serious bug; a negative response to MAIL caused the whole
2407 recipient domain to be cached as invalid, thereby blocking all messages
2408 to all local parts at the same domain, from all senders. This bug has
2409 been fixed. The domain is no longer cached after a negative response to
2410 MAIL if the sender used is not empty.
2411
2412 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
2413 been restored.
2414
2415 (c) A new callout option, "use_sender" has been added for people who want
2416 the modified behaviour.
2417
2418
2419Exim version 4.31
2420-----------------
2421
2422 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
2423 Larry Rosenman.
2424
2425 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
2426 indeed breaks things for older releases.
2427
2428 3. Added additional logging to the case where there is a problem reading data
2429 from a filter that is running in a subprocess using a pipe, in order to
2430 try to track down a specific problem.
2431
2432 4. Testing facility fudge: when running in the test harness and attempting
2433 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
2434 getting "No route to host". Convert this to a timeout.
2435
2436 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
2437 warning.
2438
2439 6. Some OS don't have socklen_t but use size_t instead. This affects the
2440 fifth argument of getsockopt() amongst other things. This is now
2441 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
2442 can be set for individual OS. I have set it for SunOS5, OSF1, and
2443 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
2444 some earlier ones do not.
2445
2446 7. Change 4.30/15 was not doing the test caselessly.
2447
2448 8. The standard form for an IPv6 address literal was being rejected by address
2449 parsing in, for example, MAIL and RCPT commands. An example of this kind of
2450 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
2451 this, as well as the form without the "IPv6" on the front (but only when
2452 address literals are enabled, of course).
2453
2454 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
2455
245610. Exim crashed if a message with an empty sender address specified by -f
2457 encountered a router with an errors_to setting. This could be provoked only
2458 by a command such as
2459
2460 exim -f "" ...
2461
2462 where an empty string was supplied; "<>" did not hit this bug.
2463
246411. Installed PCRE release 4.5.
2465
246612. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
2467 remained set. It is now erased.
2468
246913. exiqgrep wasn't working on MacOS X because it didn't correctly compute
2470 times from message ids (which are base 36 rather than the normal 62).
2471
247214. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
2473 were being counted as actual protocol errors, and logged if the log
2474 selector +smtp_protocol_error was set. One cannot be perfect in this test,
2475 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
2476 and DATA following a set of rejected RCPTs do not count as protocol errors.
2477 In other words, Exim assumes they were pipelined, though this may not
2478 actually be the case. Of course, in all cases the client gets an
2479 appropriate error code.
2480
248115. If a lookup fails in an ACL condition, a message about the failure may
2482 be available; it is used if testing the ACL cannot continue, because most
2483 such messages specify what the cause of the deferral is. However, some
2484 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
2485 that caused an old message to be retained and used if a later statement
2486 caused a defer, replacing the real cause of the deferral.
2487
248816. If an IP address had so many PTR records that the DNS lookup buffer
2489 was not large enough to hold them, Exim could crash while trying to process
2490 the truncated data. It now detects and logs this case.
2491
249217. Further to 4.21/58, another change has been made: if (and only if) the
2493 first line of a message (the first header line) ends with CRLF, a bare LF
2494 in a subsequent header line has a space inserted after it, so as not to
2495 terminate the header.
2496
249718. Refactoring: tidied an ugly bit of code in appendfile that copied data
2498 unnecessarily, used atoi() instead of strtol(), and didn't check the
2499 termination when getting file sizes from file names by regex.
2500
250119. Completely re-implemented the support for maildirsize files, in the light
2502 of a number of problems with the previous contributed implementation
2503 (4.30/29). In particular:
2504
2505 . If the quota is zero, the maildirsize file is maintained, but no quota is
2506 imposed.
2507
2508 . If the maildir directory does not exist, it is created before any attempt
2509 to write a maildirsize file.
2510
2511 . The quota value in the file is just a cache; if the quota is changed in
2512 the transport, the new value overrides.
2513
2514 . A regular expression is available for excluding directories from the
2515 count.
2516
251720. The autoreply transport checks the characters in options that define the
2518 message's headers; it allows continued headers, but it was checking with
2519 isspace() after an embedded newline instead of explicitly looking for a
2520 space or a tab.
2521
252221. If all the "regular" hosts to which an address was routed had passed their
2523 expiry times, and had not reached their retry times, the address was
2524 bounced, even if fallback hosts were defined. Now Exim should go on to try
2525 the fallback hosts.
2526
252722. Increased buffer sizes in the callout code from 1024 to 4096 to match the
2528 equivalent code in the SMTP transport. Some hosts send humungous responses
2529 to HELO/EHLO, more than 1024 it seems.
2530
253123. Refactoring: code in filter.c used (void *) for "any old type" but this
2532 gives compiler warnings in some environments. I've now done it "properly",
2533 using a union.
2534
253524. The replacement for inet_ntoa() that is used with gcc on IRIX systems
2536 (because of problems with the built-in one) was declared to return uschar *
2537 instead of char *, causing compiler failure.
2538
253925. Fixed a file descriptor leak when processing alias/forward files.
2540
254126. Fixed a minor format string issue in dbfn.c.
2542
254327. Typo in exim.c: ("dmbnz" for "dbmnz").
2544
254528. If a filter file refered to $h_xxx or $message_headers, and the headers
2546 contained RFC 2047 "words", Exim's memory could, under certain conditions,
2547 become corrupted.
2548
254929. When a sender address is verified, it is cached, to save repeating the test
2550 when there is more than one recipient in a message. However, when the
2551 verification involves a callout, it is possible for different callout
2552 options to be set for different recipients. It is too complicated to keep
2553 track of this in the cache, so now Exim always runs a verification when a
2554 callout is required, relying on the callout cache for the optimization.
2555 The overhead is duplication of the address routing, but this should not be
2556 too great.
2557
255830. Fixed a bug in callout caching. If a RCPT command caused the sender address
2559 to be verified with callout=postmaster, and the main callout worked but the
2560 postmaster check failed, the verification correctly failed. However, if a
2561 subsequent RCPT command asked for sender verification *without* the
2562 postmaster check, incorrect caching caused this verification also to fail,
2563 incorrectly.
2564
256531. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
2566 it was not caching the DNS options (qualify_single, search_parents) that
2567 were used when the lookup failed. A subsequent lookup with different
2568 options therefore always gave the same answer, though there were cases
2569 where it should not have. (Example: a "domains = !$mx_any" option on a
2570 dnslookup router: the "domains" option is always processed without any
2571 widening, but the router might have qualify_single set.) Now Exim uses the
2572 cached value only when the same options are set.
2573
257432. Added John Jetmore's "exipick" utility to the distribution.
2575
257633. GnuTLS: When an attempt to start a TLS session fails for any reason other
2577 than a timeout (e.g. a certificate is required, and is not provided), an
2578 Exim server now closes the connection immediately. Previously it waited for
2579 the client to close - but if the client is SSL, it seems that they each
2580 wait for each other, leading to a delay before one of them times out.
2581
258234: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
2583 maintained 0.8.x compatibility because I don't think many are using it, and
2584 it is clearly obsolete.
2585
258635. Added TLS support for CRLs: a tls_crl global option and one for the smtp
2587 transport.
2588
258936. OpenSSL: $tls_certificate_verified was being set to 1 even if the
2590 client certificate was expired. A simple patch fixes this, though I don't
2591 understand the full logic of why the verify callback is called multiple
2592 times.
2593
259437. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
2595 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
2596 which causes problems with some clients (such as the Certicom SSL Plus
2597 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
2598 disables the coutermeasure allowing Eudora to connect."
2599
260038. Exim was not checking that a write() to a log file succeeded. This could
2601 lead to Bad Things if a log got too big, in particular if it hit a file
2602 size limit. Exim now panics and dies if it cannot write to a log file, just
2603 as it does if it cannot open a log file.
2604
260539. Modified OS/Makefile-Linux so that it now contains
2606
2607 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
2608
2609 The two -D definitions ensure that Exim is compiled with large file
2610 support, which makes it possible to handle log files that are bigger than
2611 2^31.
2612
261340. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
2614 instance) a domain was checked against a named list that involved a lookup,
2615 causing $domain_data to be set, then another domain was checked against the
2616 same list, then the first domain was re-checked, the value of $domain_data
2617 after the final check could be wrong. In particular, if the second check
2618 failed, it could be set empty. This bug probably also applied to
2619 $localpart_data.
2620
262141. The strip_trailing_dot option was not being applied to the address given
2622 with the -f command-line option.
2623
262442. The code for reading a message's header from the spool was incrementing
2625 $received_count, but never initializing it. This meant that the value was
2626 incorrect (doubled) while delivering a message in the same process in which
2627 it was received. In the most common configuration of Exim, this never
2628 happens - a fresh exec is done - but it can happen when
2629 deliver_drop_privilege is set.
2630
263143. When Exim logs an SMTP synchronization error - client data sent too soon -
2632 it now includes up to 150 characters of the unexpected data in the log
2633 line.
2634
263544. The exim_dbmbuild utility uses fixed size buffers for reading input lines
2636 and building data strings. The size of both of these buffers was 10 000
2637 bytes - far larger than anybody would *ever* want, thought I. Needless to
2638 say, somebody hit the limit. I have increased the maximum line length to
2639 20 000 and the maximum data length of concatenated lines to 100 000. I have
2640 also fixed two bugs, because there was no checking on these buffers. Tsk,
2641 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
2642 buffer is too small.
2643
264445. The exim_dbmbuild utility did not support quoted keys, as Exim does in
2645 lsearch lookups. Now it does.
2646
264746. When parsing a route_list item in a manualroute router, a fixed-length
2648 buffer was used for the list of hosts. I made this 1024 bytes long,
2649 thinking that nobody would ever have a list of hosts that long. Wrong.
2650 Somebody had a whole pile of complicated expansion conditions, and the
2651 string was silently truncated, leading to an expansion error. It turns out
2652 that it is easier to change to an unlimited length (owing to other changes
2653 that have happened since this code was originally written) than to build
2654 structure for giving a limitation error. The length of the item that
2655 expands into the list of hosts is now unlimited.
2656
265747. The lsearch lookup could not handle data where the length of text line was
2658 more than 4095 characters. Such lines were truncated, leading to shortened
2659 data being returned. It should now handle lines of any length.
2660
266148. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
2662 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
2663 ACL").
2664
266549. Cosmetic tidy to scripts like exicyclog that are generated by globally
2666 replacing strings such as BIN_DIRECTORY in a source file: the replacement
2667 no longer happens in comment lines. A list of replacements is now placed
2668 at the head of all of the source files, except those whose only change is
2669 to replace PERL_COMMAND in the very first #! line.
2670
267150. Replaced the slow insertion sort in queue.c, for sorting the list of
2672 messages on the queue, with a bottom-up merge sort, using code contributed
2673 by Michael Haardt. This should make operations like -bp somewhat faster on
2674 large queues. It won't affect queue runners, except when queue_run_in_order
2675 is set.
2676
267751. Installed eximstats 1.31 in the distribution.
2678
267952. Added support for SRV lookups to the dnslookup router.
2680
268153. If an ACL referred to $message_body or $message_body_end, the value was not
2682 reset for any messages that followed in the same SMTP session.
2683
268454. The store-handling optimization for building very long strings was not
2685 differentiating between the different store pools. I don't think this
2686 actually made any difference in practice, but I've tidied it.
2687
268855. While running the routers to verify a sender address, $sender_address
2689 was still set to the sender address. This is wrong, because when routing to
2690 send a bounce to the sender, it would be empty. Therefore, I have changed
2691 it so that, while verifying a sender address, $sender_address is set to <>.
2692 (There is no change to what happens when verifying a recipient address.)
2693
269456. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
2695 target A or AAAA records (if not already returned) without resetting the
2696 qualify_single or search_parents options of the DNS resolver. These are
2697 inappropriate in this case because the targets of MX and SRV records must
2698 be FQDNs. A broken DNS record could cause trouble if it happened to have a
2699 target that, when qualified, matched something in the local domain. These
2700 two options are now turned off when doing these lookups.
2701
270257. It seems that at least some releases of Reiserfs (which does not have the
2703 concept of a fixed number of inodes) returns zero and not -1 for the
2704 number of available inodes. This interacted badly with check_spool_inodes,
2705 which assumed that -1 was the "no such thing" setting. What I have done is
2706 to check that the total number of inodes is greater than zero before doing
2707 the test of how many are available.
2708
270958. When a "warn" ACL statement has a log_message modifier, the message is
2710 remembered, and not repeated. This is to avoid a lot of repetition when a
2711 message has many recipients that cause the same warning to be written.
2712 Howewer, Exim was preserving the list of already written lines for an
2713 entire SMTP session, which doesn't seem right. The memory is now reset if a
2714 new message is started.
2715
271659. The "rewrite" debugging flag was not showing the result of rewriting in the
2717 debugging output unless log_rewrite was also set.
2718
271960. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
2720 of (int)(handle) when we know that handle contains (void *)(-1).
2721
272261. The Exim daemon panic-logs an error return when it closes the incoming
2723 connection. However "connection reset by peer" seems to be common, and
2724 isn't really an error worthy of noting specially, so that particular error
2725 is no long logged.
2726
272762. When Exim is trying to find all the local interfaces, it used to panic and
2728 die if the ioctl to get the interface flags failed. However, it seems that
2729 on at least one OS (Solaris 9) it is possible to have an interface that is
2730 included in the list of interfaces, but for which you get a failure error
2731 for this call. This happens when the interface is not "plumbed" into a
2732 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
2733 failure of the "get flags" call assumes that the interface is down.
2734
273563. Added a ${eval10: operator, which assumes all numbers are decimal. This
2736 makes life easier for people who are doing arithmetic on fields extracted
2737 from dates, where you often get leading zeros that should not be
2738 interpreted as octal.
2739
274064. Added qualify_domain to the redirect router, to override the global
2741 setting.
2742
274365. If a pathologically long header line contained very many addresses (the
2744 report of this problem mentioned 10 000) and each of them was rewritten,
2745 Exim could use up a very large amount of memory. (It kept on making new
2746 copies of the header line as it rewrote, and never released the old ones.)
2747 At the expense of a bit more processing, the header rewriting function has
2748 been changed so that it no longer eats memory in this way.
2749
275066. The generation of the Received: header has been moved from the time that a
2751 message starts to be received, to the time that it finishes. The timestamp
2752 in the Received: header should now be very close to that of the <= log
2753 line. There are two side-effects of this change:
2754
2755 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
2756 logged header lines no longer include the local Received: line, because
2757 it has not yet been created. The same applies to a copy of the message
2758 that is returned to a non-SMTP sender when a message is rejected.
2759
2760 (b) When a filter file is tested using -bf, no additional Received: header
2761 is added to the test message. After some thought, I decided that this
2762 is a bug fix.
2763
2764 This change does not affect the value of $received_for. It is still set
2765 after address rewriting, but before local_scan() is called.
2766
276767. Installed the latest Cygwin-specific files from the Cygwin maintainer.
2768
276968. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
2770 gave an unhelpful panic error message, and a defer error. I have managed to
2771 change this behaviour so that it now rejects any supplied certificate,
2772 which seems right, as the list of acceptable certificates is empty.
2773
277469. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
2775 gave an unhelpful defer error. I have not managed to make this reject any
2776 supplied certificates, but the error message it gives is "no certificate
2777 supplied", which is not helpful.
2778
277970. exigrep's output now also includes lines that are not associated with any
2780 message, but which match the given pattern. Implemented by a patch from
2781 Martin Sluka, which also tidied up the Perl a bit.
2782
278371. Recipient callout verification, like sender verification, was using <> in
2784 the MAIL FROM command. This isn't really the right thing, since the actual
2785 sender may affect whether the remote host accepts the recipient or not. I
2786 have changed it to use the actual sender in the callout; this means that
2787 the cache record is now keyed on a recipient/sender pair, not just the
2788 recipient address. There doesn't seem to be a real danger of callout loops,
2789 since a callout by the remote host to check the sender would use <>.
2790 [SEE ABOVE: changed after hitting problems.]
2791
279272. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
2793 temporary errors. However, in the case of such a code being given after
2794 the end of a data transmission (i.e. after ".") Exim was failing to write
2795 a retry record for the message. (Yes, there was some broken host that was
2796 actually sending 8xx at this point.)
2797
279873. An unknown lookup type in a host list could cause Exim to panic-die when
2799 the list was checked. (An example that provoked this was putting <; in the
2800 middle of a list instead of at the start.) If this happened during a DATA
2801 ACL check, a -D file could be left lying around. This kind of configuration
2802 error no longer causes Exim to die; instead it causes a defer errror. The
2803 incident is still logged to the main and panic logs.
2804
280574. Buglet left over from Exim 3 conversion. The message "too many messages
2806 in one connection" was written to the rejectlog but not the mainlog, except
2807 when address rewriting (yes!) was being logged.
2808
280975. Added write_rejectlog option.
2810
281176. When a system filter was run not as root (that is, when system_filter_user
2812 was set), the values of the $n variables were not being returned to the
2813 main process; thus, they were not subsequently available in the $sn
2814 variables.
2815
281677. Added +return_path_on_delivery log selector.
2817
281878. A connection timeout was being treated differently from recipients deferred
2819 when testing hosts_max_try with a message that was older than the host's
2820 retry timeout. (The host should not be counted, thus allowing all hosts to
2821 be tried at least once before bouncing.) This may have been the cause of an
2822 occasionally reported bug whereby a message would remain on the queue
2823 longer than the retry timeout, but would be bounced if a delivery was
2824 forced. I say "may" because I never totally pinned down the problem;
2825 setting up timeout/retry tests is difficult. See also the next item.
2826
282779. The ultimate address timeout was not being applied to errors that involved
2828 a combination of host plus message (for example, a timeout on a MAIL
2829 command). When an address resolved to a number of possible hosts, and they
2830 were not all tried for each delivery (e.g. because of hosts_max_try), a
2831 message could remain on the queue longer than the retry timeout.
2832
283380. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
2834 Haardt.
2835
283681. Fixed an obscure SMTP outgoing bug which required at least the following
2837 conditions: (a) there was another message waiting for the same server;
2838 (b) the server returned 5xx to all RCPT commands in the first message so
2839 that the message was not completed; (c) the server dropped the connection
2840 or gave a negative response to the RSET that Exim sends to abort the
2841 transaction. The observed case was a dropped connection after DATA that had
2842 been sent in pipelining mode. That is, the server had advertised PIPELINING
2843 but was not implementing it correctly. The effect of the bug was incorrect
2844 behaviour, such as trying another host, and this could lead to a crash.
2845
2846
2847Exim version 4.30
2848-----------------
2849
2850 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
2851 and daemon.c were passed as pointers to ints; they should have been
2852 pointers to socklen_t variables (which are typically unsigned ints).
2853
2854 2. Some signed/unsigned type warnings in the os.c file for Linux have been
2855 fixed.
2856
2857 3. Fixed a really odd bug that affected only the testing scheme; patching a
2858 certain fixed string in the binary changed the value of another string that
2859 happened to be identical to the end of the original first string.
2860
2861 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
2862 name", it returns that address as the IP address. On some operating
2863 systems (e.g. Solaris), it also passes back the IP address string as the
2864 "host name". However, on others (e.g. Linux), it passes back an empty
2865 string. Exim wasn't checking for this, and was changing the host name to an
2866 empty string, assuming it had been canonicized.
2867
2868 5. Although rare, it is permitted to have more than one PTR record for a given
2869 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
2870 all the names associated with an address, because they do in Solaris.
2871 However, it seems that they do not in Linux for data that comes from the
2872 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
2873 I found this out when I moved to a new Linux workstation and tried to run
2874 the Exim test suite.
2875
2876 To get round this problem I have changed the code so that it now does its
2877 own call to the DNS to look up PTR records when searching for a host name.
2878 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
2879 addresses that are only in /etc/hosts are still found.
2880
2881 This behaviour is, however, controlled by an option called host_lookup_
2882 order, which defaults to "bydns:byaddr". If people want to use the other
2883 order, or indeed, just use one or the other means of lookup, they can
2884 specify it in this variable.
2885
2886 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
2887 some operating systems, this comes back from gethostbyaddr() as an empty
2888 string, and this is what Exim used to test for. However, it seems that in
2889 other systems, "." is yielded. Exim now tests for this case too.
2890
2891 7. The values of check_spool_space and check_log_space are now held internally
2892 as a number of kilobytes instead of an absolute number of bytes. If a
2893 numbers is specified without 'K' or 'M', it is rounded up to the nearest
2894 kilobyte. This means that much larger values can be stored.
2895
2896 8. Exim monitor: an attempt to get the action menu when not actually pointing
2897 at a message produces an empty menu entitled "No message selected". This
2898 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
2899 no entries in it ("Shell widget menu has zero width and/or height"). So I
2900 have added a single, blank menu entry in this case.
2901
2902 9. Added ${quote_local_part.
2903
290410. MIME decoding is now applied to the contents of Subject: header lines when
2905 they are logged.
2906
290711. Now that a reference to $sender_host_address automatically causes a reverse
2908 lookup to occur if necessary (4.13/18), there is no need to arrange for a
2909 host lookup before query-style lookups in lists that might use this
2910 variable. This has therefore been abolished, and the "net-" prefix is no
2911 longer necessary for query-style lookups.
2912
291312. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
2914 have been a typo for LFLAGS, so it has been changed.
2915
291613. The install script calls Exim with "-C /dev/null" in order to find the
2917 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
2918 to be output. Howeve, since Exim outputs its version number before the
2919 error, it didn't break the script. It just looked ugly. I fixed this by
2920 always allowing "-C /dev/null" if the caller is root.
2921
292214. Ignore overlarge ACL variable number when reading spool file - insurance
2923 against a later release with more variables having written the file.
2924
292515. The standard form for an IPv6 address literal was being rejected by EHLO.
2926 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
2927 this, as well as the form without the "IPv6" on the front.
2928
292916. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
2930 OS/Makefile-Darwin file.
2931
293217. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
2933 only to LDAP libraries that do not have LDAP_OPT_DEREF.
2934
293518. After change 4.21/52, "%ld" was used to format the contents of the $inode
2936 variable. However, some OS use ints for inodes. I've added cast to long int
2937 to get rid of the compiler warning.
2938
293919. I had forgotten to lock out "/../" in configuration file names when
2940 ALT_CONFIG_PREFIX was set.
2941
294220. Routers used for verification do not need to specify transports. However,
2943 if such a router generated a host list, and callout was configured, Exim
2944 crashed, because it could not find a port number from the (non-existent)
2945 transport. It now assumes port 25 in this circumstance.
2946
294721. Added the -t option to exigrep.
2948
294922. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
2950 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
2951 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
2952 them is set, LOOKUP_LSEARCH is forced.
2953
295423. "exim -bV" now outputs a list of lookups that are included in the binary.
2955
295624. Added sender and host information to the "rejected by local_scan()" log
2957 line; previously there was no indication of these.
2958
295925. Added .include_if_exists.
2960
296126. Change 3.952/11 added an explicit directory sync on top of a file sync for
2962 Linux. It turns out that not all file systems support this. Apparently some
2963 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
2964 do it.) To cope with this, the error EINVAL, which means that sync-ing is
2965 not supported on the file descriptor, is now ignored when Exim is trying to
2966 sync a directory. This applies only to Linux.
2967
296827. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
2969
297028. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
2971 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
2972 this.
2973
297429. Added support for maildirsize files from supplied patch (modified a bit).
2975
297630. The use of :fail: followed by an empty string could lead Exim to respond to
2977 sender verification failures with (e.g.):
2978
2979 550 Verification failed for <xxx>
2980 550 Sender verify failed
2981
2982 where the first response line was missing the '-' that indicates it is not
2983 the final line of the response.
2984
298531. The loop for finding the name of the user that called Exim had a hardwired
2986 limit of 10; it now uses the value of finduser_retries, which is used for
2987 all other user lookups.
2988
298932. Added $received_count variable, available in data and not_smtp ACLs, and at
2990 delivery time.
2991
299233. Exim was neglecting to zero errno before one call of strtol() when
2993 expanding a string and expecting an integer value. On some systems this
2994 resulted in spurious "integer overflow" errors. Also, it was casting the
2995 result into an int without checking.
2996
299734. Testing for a connection timeout using "timeout_connect" in the retry rules
2998 did not work. The code looks as if it has *never* worked, though it appears
2999 to have been documented since at least releast 1.62. I have made it work.
3000
300135. The "timeout_DNS" error in retry rules, also documented since at least
3002 1.62, also never worked. As it isn't clear exactly what this means, and
3003 clearly it isn't a major issue, I have abolished the feature by treating it
3004 as "timeout", and writing a warning to the main and panic logs.
3005
300636. The display of retry rules for -brt wasn't always showing the error code
3007 correctly.
3008
300937. Added new error conditions to retry rules: timeout_A, timeout_MX,
3010 timeout_connect_A, timeout_connect_MX.
3011
301238. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
3013 to the empty sender.
3014
301539. The daemon was not analysing the content of -oX till after it had closed
3016 stderr and disconnected from the controlling terminal. This meant that any
3017 syntax errors were only noted on the panic log, and the return code from
3018 the command was 0. By re-arranging the code a little, I've made the
3019 decoding happen first, so such errors now appear on stderr, and the return
3020 code is 1. However, the actual setting up of the sockets still happens in
3021 the disconnected process, so errors there are still only recorded on the
3022 panic log.
3023
302440. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
3025 connections (as happens on some IP stacks) was logged at start up time as
3026 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
3027 it from "IPv6 and IPv4", which means that two separate sockets are being
3028 used.
3029
303041. The debug output for gethostbyname2() or getipnodebyname() failures now
3031 says whether AF_INET or AF_INET6 was passed as an argument.
3032
303342. Exiwhat output was messed up when time zones were included in log
3034 timestamps.
3035
303643. Exiwhat now gives more information about the daemon's listening ports,
3037 and whether -tls-on-connect was used.
3038
303944. The "port" option of the smtp transport is now expanded.
3040
304145. A "message" modifier in a "warn" statement in a non-message ACL was being
3042 silently ignored. Now an error message is written to the main and panic
3043 logs.
3044
304546. There's a new ACL modifier called "logwrite" which writes to a log file
3046 as soon as it is encountered.
3047
304847. Added $local_user_uid and $local_user_gid at routing time.
3049
305048. Exim crashed when trying to verify a sender address that was being
3051 rewritten to "<>".
3052
305349. Exim was recognizing only a space character after ".include". It now also
3054 recognizes a tab character.
3055
305650. Fixed several bugs in the Perl script that creates the exim.8 man page by
3057 extracting the relevant information from the specification. The man page no
3058 longer contains scrambled data for the -d option, and I've added a section
3059 at the front about calling Exim under different names.
3060
306151. Added "extra_headers" argument to the "mail" command in filter files.
3062
306352. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
3064 crash.
3065
306653. Installed eximstats 1.29.
3067
306854. Added transport_filter_timeout as a generic transport option.
3069
307055. Exim no longer adds an empty Bcc: header to messages that have no To: or
3071 Cc: header lines. This was required by RFC 822, but it not required by RFC
3072 2822.
3073
307456. Exim used to add From:, Date:, and Message-Id: header lines to any
3075 incoming messages that did not have them. Now it does so only if the
3076 message originates locally, that is, if there is no associated remote host
3077 address. When Resent- header lines are present, this applies to the Resent-
3078 lines rather than the non-Resent- lines.
3079
308057. Drop incoming SMTP connection after too many syntax or protocol errors. The
3081 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
3082
308358. Messages for configuration errors now include the name of the main
3084 configuration file - useful now that there may be more than one file in a
3085 list (.included file names were always shown).
3086
308759. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
3088 for those rare installations that do not start the daemon as root or run it
3089 setuid root. I've cut out the call to initgroups() if the daemon is not
3090 root at that time.
3091
309260. The Exim user and group can now be bound into the binary as text strings
3093 that are looked up at the start of Exim's processing.
3094
309561. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
3096
309762. Added $mailstore_basename variable.
3098
309963. Installed patch to sieve.c from Michael Haardt.
3100
310164. When Exim failed to open the panic log after failing to open the main log,
3102 the original message it was trying to log was written to stderr and debug
3103 output, but if they were not available (the usual case in production), it
3104 was lost. Now it is written to syslog before the two lines that record the
3105 failures to open the logs.
3106
310765. Users' Exim filters run in subprocesses under the user's uid. It is
3108 possible for a "deliver" command or an alias in a "personal" command to
3109 provoke an address rewrite. If logging of address rewriting is configured,
3110 this fails because the process is not running as root or exim. There may be
3111 a better way of dealing with this, but for the moment (because 4.30 needs
3112 to be released), I have disabled address rewrite logging when running a
3113 filter in a non-root, non-exim process.
3114
3115
3116Exim version 4.24
3117-----------------
3118
3119 1. The buildconfig auxiliary program wasn't quoting the value set for
3120 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
3121 not defined. This bug was masked in 4.22 by the effect that was fixed in
3122 change 4.23/1.
3123
3124 2. Some messages that were rejected after a message id was allocated were
3125 shown as "incomplete" by exigrep. It no longer does this for messages that
3126 are rejected by local_scan() or the DATA or non-SMTP ACLs.
3127
3128 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
3129 have allow_domain_literals set, the ID did not get logged in the <= line.
3130 Domain literals are now always recognized in Message-ID: header lines.
3131
3132 4. The first argument for a ${extract expansion item is the key name or field
3133 number. Leading and trailing spaces in this item were not being ignored,
3134 causing some misleading effects.
3135
3136 5. When deliver_drop_privilege was set, single queue runner processes started
3137 manually (i.e. by the command "exim -q") or by the daemon (which uses the
3138 same command in the process it spins off) were not dropping privilege.
3139
3140 6. When the daemon running as "exim" started a queue runner, it always
3141 re-executed Exim in the spun-off process. This is a waste of effort when
3142 deliver_drop_privilege is set. The new process now just calls the
3143 queue-runner function directly.
3144
3145
3146Exim version 4.23
3147-----------------
3148
3149 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
3150 HEADERS_CHARSET.
3151
3152 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
3153 ignored. Though the use of -oP was forcing the writing of a pid file, it
3154 was always written to the default place.
3155
3156 3. If the message "no IP address found for host xxxx" is generated during
3157 incoming verification, it is now followed by identification of the incoming
3158 connection (so you can more easily find what provoked it).
3159
3160 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
3161
3162 5. Added some features to "harden" Exim a bit more against certain attacks:
3163
3164 (a) There is now a build-time option called FIXED_NEVER_USERS that can
3165 be put in Local/Makefile. This is like the never_users runtime option,
3166 but it cannot be overridden. The default setting is "root".
3167
3168 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
3169 prefix string with which any file named in a -C command line option
3170 must start.
3171
3172 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
3173 is retained for -C and -D only if the caller of Exim is root. Without
3174 it, the exim user may also use -C and -D and retain privilege.
3175
3176 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
3177 command line option is disabled.
3178
3179 6. Macro names set by the -D option must start with an upper case letter, just
3180 like macro names defined in the configuration file.
3181
3182 7. Added "dereference=" facility to LDAP.
3183
3184 8. Two instances of the typo "uknown" in the source files are fixed.
3185
3186 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
3187 the Configure-Makefile script screwed up while processing it.
3188
318910. Incorporated PCRE 4.4.
3190
319111. The SMTP synchronization check was not operating right at the start of an
3192 SMTP session. For example, it could not catch a HELO sent before the client
3193 waited for the greeting. There is now a check for outstanding input at the
3194 point when the greeting is written. Because of the duplex, asynchronous
3195 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
3196 way, but not yet received, when the check is performed.
3197
319812. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
3199 on TCP/IP sockets, because this apparently causes some broken clients to
3200 timeout.
3201
320213. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
3203 unchanged) from the Cygwin maintainer.
3204
320514. The code for -bV that shows what is in the binary showed "mbx" when maildir
3206 was supported instead of testing for mbx. Effectively a typo.
3207
320815. The spa authenticator server code was not checking that the input it
3209 received was valid base64.
3210
321116. The debug output line for the "set" modifier in ACLs was not showing the
3212 name of the variable that was being set.
3213
321417. Code tidy: the variable type "vtype_string" was never used. Removed it.
3215
321618. Previously, a reference to $sender_host_name did not cause a DNS reverse
3217 lookup on its own. Something else was needed to trigger the lookup. For
3218 example, a match in host_lookup or the need for a host name in a host list.
3219 Now, if $sender_host_name is referenced and the host name has not yet been
3220 looked up, a lookup is performed. If the lookup fails, the variable remains
3221 empty, and $host_lookup_failed is set to "1".
3222
322319. Added "eqi" as a case-independent comparison operator.
3224
322520. The saslauthd authentication condition could segfault if neither service
3226 nor realm was specified.
3227
322821. If an overflowing value such as "2048M" was set for message_size_limit, the
3229 error message that was logged was misleading, and incoming SMTP
3230 connections were dropped. The message is now more accurate, and temporary
3231 errors are given to SMTP connections.
3232
323322. In some error situations (such as 21 above) Exim rejects all SMTP commands
3234 (except RSET) with a 421 error, until QUIT is received. However, it was
3235 failing to send a response to QUIT.
3236
323723. The HELO ACL was being run before the code for helo_try_verify_hosts,
3238 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
3239 ACL is now run after the helo_try_verify_hosts code.
3240
324124. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
3242 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
3243 case-independent, so other case variants are also recognized). Apparently
3244 some systems use these upper case variants.
3245
324625. If more than two messages were waiting for the same host, and a transport
3247 filter was specified for the transport, Exim sent two messages over the
3248 same TCP/IP connection, and then failed with "socket operation on non-
3249 socket" when it tried to send the third.
3250
325126. Added Exim::debug_write and Exim::log_write for embedded Perl use.
3252
325327. The extern definition of crypt16() in expand.c was not being excluded when
3254 the OS had its own crypt16() function.
3255
325628. Added bounce_return_body as a new option, and bounce_return_size_limit
3257 as a preferred synonym for return_size_limit, both as an option and as an
3258 expansion variable.
3259
326029. Added LIBS=-liconv to OS/Makefile-OSF1.
3261
326230. Changed the default configuration ACL to relax the local part checking rule
3263 for addresses that are not in any local domains. For these addresses,
3264 slashes and pipe symbols are allowed within local parts, but the sequence
3265 /../ is explicitly forbidden.
3266
326731. SPA server authentication was not clearing the challenge buffer before
3268 using it.
3269
327032. log_message in a "warn" ACL statement was writing to the reject log as
3271 well as to the main log, which contradicts the documentation and doesn't
3272 seem right (because no rejection is happening). So I have stopped it.
3273
327433. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
3275 However, I am unable to do any testing of this.
3276
327734. Fixed an infelicity in the appendfile transport. When checking directories
3278 for a mailbox, to see if any needed to be created, it was accidentally
3279 using path names with one or more superfluous leading slashes; tracing
3280 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
3281
328235. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
3283 contents are added to the log line that is written for every discarded
3284 recipient. (Previously a log_message setting was ignored.)
3285
328636. The ${quote: operator now quotes the string if it is empty.
3287
328837. The install script runs exim in order to find its version number. If for
3289 some reason other than non-existence or emptiness, which it checks, it
3290 could not run './exim', it was installing it with an empty version number,
3291 i.e. as "exim-". This error state is now caught, and the installation is
3292 aborted.
3293
329438. An argument was missing from the function that creates an error message
3295 when Exim fails to connect to the socket for saslauthd authentication.
3296 This could cause Exim to crash, or give a corrupted message.
3297
329839. Added isip, isip4, and isip6 to ${if conditions.
3299
330040. The ACL variables $acl_xx are now saved with the message, and can be
3301 accessed later in routers, transports, and filters.
3302
330341. The new lookup type nwildlsearch is like wildlsearch, except that the key
3304 strings in the file are not string-expanded.
3305
330642. If a MAIL command specified a SIZE value that was too large to fit into an
3307 int variable, the check against message_size_limit failed. Such values are
3308 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
3309 day this will have to be increased, but I don't think I want to be around
3310 when emails are that large.
3311
3312
3313
3314Exim version 4.22
3315-----------------
3316
3317 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
3318 iconv() is not standard in FreeBSD.
3319
3320 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
3321 IPv6 enabled. The observed symptom was a segmentation fault on return from
3322 the function os_common_find_running_interfaces() in src/os.c.
3323
3324 3. In the check_special_case() function in daemon.c I had used "errno" as an
3325 argument name, which causes warnings on some systems. This was basically a
3326 typo, since it was named "eno" in the comments!
3327
3328 4. The code that waits for the clock to tick (at a resolution of some fraction
3329 of a second) so as to ensure message-id uniqueness was always waiting for
3330 at least one whole tick, when it could have waited for less. [This is
3331 almost certainly not relevant at current processor speeds, where it is
3332 unlikely to ever wait at all. But we try to future-proof.]
3333
3334 5. The function that sleeps for a time interval that includes fractions of a
3335 second contained a race. It did not block SIGALRM between setting the
3336 timer, and suspending (a couple of lines later). If the interval was short
3337 and the sigsuspend() was delayed until after it had expired, the suspension
3338 never ended. On busy systems this could lead to processes getting stuck for
3339 ever.
3340
3341 6. Some uncommon configurations may cause a lookup to happen in a queue runner
3342 process, before it forks any delivery processes. The open lookup caching
3343 mechanism meant that the open file or database connection was passed into
3344 the delivery process. The problem was that delivery processes always tidy
3345 up cached lookup data. This could cause a problem for the next delivery
3346 process started by the queue runner, because the external queue runner
3347 process does not know about the closure. So the next delivery process
3348 still has data in the lookup cache. In the case of a file lookup, there was
3349 no problem because closing a file descriptor in a subprocess doesn't affect
3350 the parent. However, if the lookup was caching a connection to a database,
3351 the connection was closed, and the second delivery process was likely to
3352 see errors such as "PGSQL: query failed: server closed the connection
3353 unexpectedly". The problem has been fixed by closing all cached lookups
3354 in a queue runner before running a delivery process.
3355
3356 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
3357 seem to have the "const" qualifier which it has on other OS. I've
3358 parameterised it.
3359
3360 8. Change 4.21/2 was too strict. It is only if there are two authenticators
3361 *of the same type* (client or server) with the same public name that an
3362 error should be diagnosed.
3363
3364 9. When Exim looked up a host name for an IP address, but failed to find the
3365 original IP address when looking up the host name (a safety check), it
3366 output the message "<ip address> does not match any IP for NULL", which was
3367 confusing, to say the least. The bug was that the host name should have
3368 appeared instead of "NULL".
3369
337010. Since release 3.03, if Exim is called by a uid other than root or the Exim
3371 user that is built into the binary, and the -C or -D options is used, root
3372 privilege is dropped before the configuration file is read. In addition,
3373 logging is switched to stderr instead of the normal log files. If the
3374 configuration then re-defines the Exim user, the unprivileged environment
3375 is probably not what is expected, so Exim logs a panic warning message (but
3376 proceeds).
3377
3378 However, if deliver_drop_privilege is set, the unprivileged state may well
3379 be exactly what is intended, so the warning has been cut out in that case,
3380 and Exim is allowed to try to write to its normal log files.
3381
3382
3383Exim version 4.21
3384-----------------
3385
3386 1. smtp_return_error_details was not giving details for temporary sender
3387 or receiver verification errors.
3388
3389 2. Diagnose a configuration error if two authenticators have the same public
3390 name.
3391
3392 3. Exim used not to create the message log file for a message until the first
3393 delivery attempt. This could be confusing when incoming messages were held
3394 for policy or load reasons. The message log file is now created at the time
3395 the message is received, and an initial "Received" line is written to it.
3396
3397 4. The automatically generated man page for command line options had a minor
3398 bug that caused no ill effects; however, a more serious problem was that
3399 the procedure for building the man page automatically didn't always
3400 operate. Consequently, release 4.20 contains an out-of-date version. This
3401 shouldn't happen again.
3402
3403 5. When building Exim with embedded Perl support, the script that builds the
3404 Makefile was calling 'perl' to find its compile-time parameters, ignoring
3405 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
3406
3407 6. The freeze_tell option was not being used for messages that were frozen on
3408 arrival, either by an ACL or by local_scan().
3409
3410 7. Added the smtp_incomplete_transaction log selector.
3411
3412 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
3413 was accepting AUTH without a new EHLO.
3414
3415 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
3416 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
3417 RFC.
3418
341910. Logging of TCP/IP connections (when configured) now happens in the main
3420 daemon process instead of the child process, so that the TCP/IP connection
3421 count is more accurate (but it can never be perfect).
3422
342311. The use of "drop" in a nested ACL was not being handled correctly in the
3424 outer ACL. Now, if condition failure induced by the nested "drop" causes
3425 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
3426 or "require"), the connection is dropped.
3427
342812. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
3429 that yield "discard" can now be used with an "accept" or a "discard" verb,
3430 but an error is generated for any others (because I can't see a useful way
3431 to define what should happen).
3432
343313. When an ACL is read dynamically from a file (or anywhere else), the lines
3434 are now processed in the same way as lines in the Exim configuration file.
3435 In particular, continuation lines are supported.
3436
343714. Added the "dnslists = a.b.c!=n.n.n.n" feature.
3438
343915. Added -ti meaning -t -i.
3440
344116. Check for letters, digits, hyphens, and dots in the names of dnslist
3442 domains, and warn by logging if others are found.
3443
344417. At least on BSD, alignment is not guarenteed for the array of ifreq's
3445 returned from GIFCONF when Exim is trying to find the list of interfaces on
3446 a host. The code in os.c has been modified to copy each ifreq to an aligned
3447 structure in all cases.
3448
3449 Also, in some cases, the returned ifreq's were being copied to a 'struct
3450 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
3451 means the last couple of bytes of an IPv6 address could be chopped if the
3452 ifreq contained only a normal sockaddr (14 bytes storage).
3453
345418. Named domain lists were not supported in the hosts_treat_as_local option.
3455 An entry such as +xxxx was not recognized, and was treated as a literal
3456 domain name.
3457
345819. Ensure that header lines added by a DATA ACL are included in the reject log
3459 if the ACL subsequently rejects the message.
3460
346120. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
3462 MD5 (which is deprecated).
3463
346421. When testing a filter file using -bf, Exim was writing a message when it
3465 took the sender from a "From " line in the message, but it was not doing so
3466 when it took $return_path from a Return-Path: header line. It now does.
3467
346822. If the contents of a "message" modifier for a "warn" ACL verb do not begin
3469 with a valid header line field name (a series of printing characters
3470 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
3471
347223. Changed "disc" in the source to "disk" to conform to the documentation and
3473 the book and for uniformity.
3474
347524. Ignore Sendmail's -Ooption=value command line item.
3476
347725. When execve() failed while trying to run a command in a pipe transport,
3478 Exim was returning EX_UNAVAILBLE (69) from the subprocess. However, this
3479 could be confused with a return value of 69 from the command itself. This
3480 has been changed to 127, the value the shell returns if it is asked to run
3481 a non-existent command. The wording for the related log line suggests a
3482 non-existent command as the problem.
3483
348426. If received_header_text expands to an empty string, do not add a Received:
3485 header line to the message. (Well, it adds a token one on the spool, but
3486 marks it "old" so that it doesn't get used or transmitted.)
3487
348827. Installed eximstats 1.28 (addition of -nt option).
3489
349028. There was no check for failure on the call to getsockname() in the daemon
3491 code. This can fail if there is a shortage of resources on the system, with
3492 ENOMEM, for example. A temporary error is now given on failure.
3493
349429. Contrary to the C standard, it seems that in some environments, the
3495 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
3496 program. Exim now does this explicitly; it affects the formatting of
3497 timestamps using strftime().
3498
349930. If exiqsumm was given junk data, it threw up some uninitialized variable
3500 complaints. I've now initialized all the variables, to avoid this.
3501
350232. Header lines added by a system filter were not being "seen" during
3503 transport-time rewrites.
3504
350533. The info_callback() function passed to OpenSSL is set up with type void
3506 (*)(SSL *, int, int), as described somewhere. However, when calling the
3507 function (actually a macro) that sets it up, the type void(*)() is
3508 expected. I've put in a cast to prevent warnings from picky compilers.
3509
351034. If a DNS black list lookup found a CNAME record, but there were no A
3511 records associated with the domain it pointed at, Exim crashed.
3512
351335. If a DNS black list lookup returned more than one A record, Exim ignored
3514 all but the first. It now scans all returned addresses if a particular IP
3515 value is being sought. In this situation, the contents of the
3516 $dnslist_value variable are a list of all the addresses, separated by a
3517 comma and a space.
3518
351936. Tightened up the rules for host name lookups using reverse DNS. Exim used
3520 to accept a host name and all its aliases if the forward lookup for any of
3521 them yielded the IP address of the incoming connection. Now it accepts only
3522 those names whose forward lookup yields the correct IP address. Any other
3523 names are discarded. This closes a loophole whereby a rogue DNS
3524 administrator could create reverse DNS records to break through a
3525 wildcarded host restriction in an ACL.
3526
352737. If a user filter or a system filter that ran in a subprocess used any of
3528 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
3529 the wrong values were passed to the pipe command ($thisaddress had the
3530 value of $0, $0 had the value of $1, etc). This bug was introduced by
3531 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
3532
353338. Improved the line breaking for long SMTP error messages from ACLs.
3534 Previously, if there was no break point between 40 and 75 characters, Exim
3535 left the rest of the message alone. Two changes have been made: (a) I've
3536 reduced the minimum length to 35 characters; (b) if it can't find a break
3537 point between 35 and 75 characters, it looks ahead and uses the first one
3538 that it finds. This may give the occasional overlong line, but at least the
3539 remaining text gets split now.
3540
354139. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
3542 file descriptors might be low, and that setting 1000 would always raise it.
3543 It turns out that in some environments, the limit is already over 1000 and
3544 that lowering it causes trouble. So now Exim takes care not to decrease it.
3545
354640. When delivering a message, the value of $return_path is set to $sender_
3547 address at the start of routing (routers may change the value). By an
3548 oversight, this default was not being set up when an address was tested by
3549 -bt or -bv, which affected the outcome if any router or filter referred to
3550 $return_path.
3551
355241. The idea of the "warn" ACL verb is that it adds a header or writes to the
3553 log only when "message" or "log_message" are set. However, if one of the
3554 conditions was an address verification, or a call to a nested ACL, the
3555 messages generated by the underlying test were being passed through. This
3556 no longer happens. The underlying message is available in $acl_verify_
3557 message for both "message" and "log_message" expansions, so it can be
3558 passed through if needed.
3559
356042. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
3561 new expansion $bh_ to give the encoded byte string without charset
3562 translation. Translation happens only if iconv() is available; HAVE_ICONV
3563 indicates this at build time. HEADERS_CHARSET gives the charset to
3564 translate to; headers_charset can change it in the configuration, and
3565 "headers charset" can change it in an individual filter file.
3566
356743. Now that we have a default RFC 2047 charset (see above), the code in Exim
3568 that creates RFC 2047 encoded "words" labels them as that charset instead
3569 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
3570 expansion operator; (ii) when Exim creates a From: line for a local
3571 message; (iii) when a header line is rewritten to include a "phrase" part.
3572
357344. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
3574 buggy, causing it to skip the first lines of messages whose message ID
3575 ended in 'D'. This would not have bitten before Exim release 4.14, because
3576 message IDs were unlikely to end in 'D' before then. The effect was to have
3577 incorrect size information for certain domains.
3578
357945. #include "config.h" was missing at the start of the crypt16.c module. This
3580 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
3581 noticed.
3582
358346. If there was a timeout during a "random" callout check, Exim treated it as
3584 a failure of the random address, and carried on sending RSET and the real
3585 address. If the delay was just some slowness somewhere, the response to the
3586 original RCPT would be taken as a response to RSET and so on, causing
3587 mayhem of various kinds.
3588
358947. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
3590 when I implemented it. It didn't allow for the fact that some option values
3591 may legitimatetly be negative (e.g. size_addition), and it didn't even do
3592 the right test for positive values.
3593
359448. Domain names in DNS records are case-independent. Exim always looks them up
3595 in lower case. Some resolvers return domain names in exactly the case they
3596 appear in the zone file, that is, they may contain uppercase letters. Not
3597 all resolvers do this - some return always lower case. Exim was treating a
3598 change of case by a resolver as a change of domain, similar to a widening
3599 of a domain abbreviation. This triggered its re-routing code and so it was
3600 trying to route what was effectively the same domain again. This normally
3601 caused routing to fail (because the router wouldn't handle the domain
3602 twice). Now Exim checks for this case specially, and just changes the
3603 casing of the domain that it ultimately uses when it transmits the message
3604 envelope.
3605
360649. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
3607 module.
3608
360950. If a filter generated a file delivery with a non-absolute name (possible if
3610 no home directory exists for the router), the forbid_file option was not
3611 forbidding it.
3612
361351. Added '&' feature to dnslists, to provide bit mask matching in addition to
3614 the existing equality matching.
3615
361652. Exim was using ints instead of ino_t variables in some places where it was
3617 dealing with inode numbers.
3618
361953. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
3620 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
3621 called TMPDIR, and if it finds it is different, it changes its value.
3622
362354. The smtp_printf() function is now made available to local_scan() so
3624 additional output lines can be written before returning. There is also an
3625 smtp_fflush() function to enable the detection of a dropped connection.
3626 The variables smtp_input and smtp_batched_input are exported to
3627 local_scan().
3628
362955. Changed the default runtime configuration: the message "Unknown user"
3630 has been removed from the ACL, and instead placed on the localuser router,
3631 using the cannot_route_message feature. This means that any verification
3632 failures that generate their own messages won't get overridden. Similarly,
3633 the "Unrouteable address" message that was in the ACL for unverifiable
3634 relay addresses has also been removed.
3635
363656. Added hosts_avoid_esmtp to the smtp transport.
3637
363857. The exicyclog script was not checking for the esoteric option
3639 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
3640 will work only if exicyclog is run under the appropriate euid.
3641
364258. Following a discussion on the list, the rules by which Exim recognises line
3643 endings on incoming messages have been changed. The -dropcr and drop_cr
3644 options are now no-ops, retained only for backwards compatibility. The
3645 following line terminators are recognized: LF CRLF CR. However, special
3646 processing applies to CR:
3647
3648 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
3649 nor a local message in the state where . is a terminator.
3650
3651 (ii) If a bare CR is encountered in a header line, an extra space is added
3652 after the line terminator so as not to end the header. The reasoning
3653 behind this is that bare CRs in header lines are most likely either
3654 to be mistakes, or people trying to play silly games.
3655
365659. The size of a message, as listed by "-bp" or in the Exim monitor window,
3657 was being incorrectly given as 18 bytes larger than it should have been.
3658 This is a VOB (very old bug).
3659
366060. This may never have affected anything current, but just in case it has:
3661 When the local host is found other than at the start of a list of hosts,
3662 the local host, those with the same MX, and any that follow, are discarded.
3663 When the list in question was part of a longer list of hosts, the following
3664 hosts (not currently being processed) were also being discarded. This no
3665 longer happens. I'm not sure if this situation could ever has previously
3666 arisen.
3667
366861. Added the "/MX" feature to lists of hosts in the manualroute and query
3669 program routers.
3670
367162. Whenever Exim generates a new message, it now adds an Auto-Submitted:
3672 header. This is something that is recommended in a new Internet Draft, and
3673 is something that is documented as being done by Sendmail. There are two
3674 possible values. For messages generated by the autoreply transport, Exim
3675 adds:
3676
3677 Auto-Submitted: auto-replied
3678
3679 whereas for all other generated messages (e.g. bounces) it adds
3680
3681 Auto-Submitted: auto-generated
3682
368363. The "personal" condition in filters now includes a test for the
3684 Auto-Submitted: header. If it contains the string "auto-" the message it
3685