Add new errors mail_4xx, data_4xx, lost_connection, tls_required.
[exim.git] / doc / doc-txt / ChangeLog
CommitLineData
e97957bc 1$Cambridge: exim/doc/doc-txt/ChangeLog,v 1.327 2006/03/09 15:10:16 ph10 Exp $
495ae4b0
PH
2
3Change log file for Exim from version 4.21
4-------------------------------------------
5
214e2000
PH
6Exim version 4.61
7-----------------
8
9PH/01 The code for finding all the local interface addresses on a FreeBSD
10 system running IPv6 was broken. This may well have applied to all BSD
11 systems, as well as to others that have similar system calls. The broken
12 code found IPv4 interfaces correctly, but gave incorrect values for the
13 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
14 that it would not match correctly against @[] and not recognize the IPv6
15 addresses as local.
16
f9daeae0
PH
17PH/02 The ipliteral router was not recognizing addresses of the form user@
18 [ipv6:....] because it didn't know about the "ipv6:" prefix.
19
7e66e54d
PH
20PH/03 Added disable_ipv6.
21
c8ea1597
PH
22PH/04 Changed $reply_address to use the raw form of the headers instead of the
23 decoded form, because it is most often used to construct To: headers
24 lines in autoreplies, and the decoded form may well be syntactically
25 invalid. However, $reply_address has leading white space removed, and all
26 newlines turned into spaces so that the autoreply transport does not
27 grumble.
28
911f6fde
PH
29PH/05 If group was specified without a user on a router, and no group or user
30 was specified on a transport, the group from the router was ignored.
31
47ca6d6c
PH
32PH/06 Increased the number of ACL variables to 20 of each type, and arranged
33 for visible compile-time settings that can be used to change these
34 numbers, for those that want even more. Backwards compatibility with old
35 spool files has been maintained. However, going back to a previous Exim
36 release will lost any variables that are in spool files.
37
ed0e9820
PH
38PH/07 Two small changes when running in the test harness: increase delay when
39 passing a TCP/IP connection to a new process, in case the original
40 process has to generate a bounce, and remove special handling of
41 127.0.0.2 (sic), which is no longer necessary.
42
eff37e47
PH
43PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
44 be the same on different OS.
45
1921d2ea
PH
46PH/09 Moved a debug statement in filter processing to avoid a race problem when
47 testing.
48
b3f69ca8
JJ
49JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
50 whether --show-vars was specified or not
51
52JJ/02 exipick: Added support for new ACL variable spool format introduced
53 in 4.61-PH/06
54
424a1c63
PH
55PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
56 syntactically invalid From: or Reply-to: line, and a filter used this to
57 generate an autoreply, and therefore failed to obtain an address for the
58 autoreply, Exim could try to deliver to a non-existent relative file
59 name, causing unrelated and misleading errors. What now happens is that
60 it logs this as a hard delivery error, but does not attempt to create a
61 bounce message.
62
7a100415
PH
63PH/11 The exinext utility has a -C option for testing purposes, but although
64 the given file was scanned by exinext itself; it wasn't being passed on
65 when Exim was called.
66
19b9dc85
PH
67PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
68 an end-of-file indication when reading a command response.
69
309bd837
PH
70PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
71 compiled. In many other places in Exim, IPv6 addresses are always
72 recognized, so I have changed this. It also means that IPv4 domain
73 literals of the form [IPV4:n.n.n.n] are now always recognized.
74
59e82a2a
PH
75PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
76 used if the router is not running as root, for example, when verifying at
77 ACL time, or when using -bh. The debugging output from this situation was
78 non-existent - all you got was a failure to exec. I have made two
79 changes:
80
81 (a) Failures to set uid/gid, the current directory, or a process leader
82 in a subprocess such as that created by queryprogram now generate
83 suitable debugging ouput when -d is set.
84
85 (b) The queryprogram router detects when it is not running as root,
86 outputs suitable debugging information if -d is set, and then runs
87 the subprocess without attempting to change uid/gid.
88
9edc04ce
PH
89PH/15 Minor change to Makefile for building test_host (undocumented testing
90 feature).
91
1349e1e5
PH
92PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
93 additional section of a DNS packet that returns MX or SRV records.
94 Instead, it always explicitly searches for A/AAAA records. This avoids
95 major problems that occur when a DNS server includes only records of one
96 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
97 fixed another bug: if SRV records were looked up and the corresponding
98 address records were *not* found in the additional section, the port
99 values from the SRV records were lost.
100
ea49d0e1
PH
101PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
102 using the correct key (the original address) when searching the retry
103 rules in order to find which one to use for generating the retry hint.
104
064a94c9
PH
105PH/18 If quota_warn_message contains a From: header, Exim now refrains from
106 adding the default one. Similarly, if it contains a Reply-To: header, the
107 errors_reply_to option, if set, is not used.
108
727071f8
PH
109PH/19 When calculating a retry time, Exim used to measure the "time since
110 failure" by looking at the "first failed" field in the retry record. Now
111 it does not use this if it is later than than the arrival time of the
112 message. Instead it uses the arrival time. This makes for better
113 behaviour in cases where some deliveries succeed, thus re-setting the
114 "first failed" field. An example is a quota failure for a huge message
115 when small messages continue to be delivered. Without this change, the
116 "time since failure" will always be short, possible causing more frequent
117 delivery attempts for the huge message than are intended.
118
f78eb7c6
PH
119PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
120 $1, $2, $3) because the numerical variables can be reset during some
121 expansion items (e.g. "match"), thereby losing the authentication data.
122
21c28500
PH
123PH/21 Make -bV show the size of off_t variables so that the test suite can
124 decide whether to run tests for quotas > 2G.
125
126PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
127 mailbox_size, and mailbox_filecount in the appendfile transport. If a
128 filecount value is greater than 2G or if a quota value is greater than 2G
129 on a system where the size of off_t is not greater than 4, a panic error
130 is given.
131
1688f43b
PH
132PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
133 never match. The debug and -bh output now contains an explicit error
134 message indicating a malformed IPv4 address or mask.
135
136PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
137 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
138 PH/23 above applies.
139
9675b384
PH
140PH/25 Do not write to syslog when running in the test harness. The only
141 occasion when this arises is a failure to open the main or panic logs
142 (for which there is an explicit test).
143
6a3f1455
PH
144PH/26 Added the /no_tell option to "control=freeze".
145
dac79d3e
PH
146PH/27 If a host name lookup failed very early in a connection, for example, if
147 the IP address matched host_lookup and the reverse lookup yielded a name
148 that did not have a forward lookup, an error message of the form "no IP
149 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
150 could be logged. Now it outputs the IP address instead of "NULL".
1349e1e5 151
5977a0b3
PH
152PH/28 An enabling patch from MH: add new function child_open_exim2() which
153 allows the sender and the authenticated sender to be set when
154 submitting a message from within Exim. Since child_open_exim() is
155 documented for local_scan(), the new function should be too.
156
c91535f3
PH
157PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
158 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
159 results in an empty string is now treated as unset.
160
0d46a8c8
PH
161PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
162
278c6e6c
PH
163PH/31 Added sender_verify_fail logging option.
164
2cbb4081
PH
165PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
166 needed by RFC 822 but not by RFC 2822 was commented out. I have now
167 tidied the source and removed it altogether.
168
3eef829e
PH
169PH/33 When a queue run was abandoned because the load average was too high, a
170 log line was always written; now it is written only if the queue_run log
171 selector is set. In addition, the log line for abandonment now contains
172 information about the queue run such as the pid. This is always present
173 in "start" and "stop" lines but was omitted from the "abandon" line.
174
1ab95fa6
PH
175PH/34 Omit spaces between a header name and the colon in the error message that
176 is given when verify = headers_syntax fails (if there are lots of them,
177 the message gets confusing).
178
230205fc
PH
179PH/35 Change the default for dns_check_names_pattern to allow slashes within
180 names, as there are now some PTR records that contain slashes. This check
181 is only to protect against broken name servers that fall over on strange
182 characters, so the fact that it applies to all lookups doesn't matter.
183
75e0e026
PH
184PH/36 Now that the new test suite is complete, we can remove some of the
185 special code in Exim that was needed for the old test suite. For example,
186 sorting DNS records because real resolvers return them in an arbitrary
187 order. The new test suite's fake resolver always returns records in the
188 same order.
189
190PH/37 When running in the test harness, use -odi for submitted messages (e.g.
191 bounces) except when queue_only is set, to avoid logging races between
192 the different processes.
193
145396a6
PH
194PH/38 Panic-die if .include specifies a non-absolute path.
195
3cd34f13
PH
196PH/39 A tweak to the "H" retry rule from its user.
197
11121d3d
JJ
198JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
199 a label. They prevented compilation on older perls.
200
201JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
202 a warning to be raised on newish perls.
203
204JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
205 on queue. Changes to match documented behaviour of showing count of
206 messages matching specified criteria.
207
8def5aaf
PH
208PH/40 Changed the default ident timeout from 30s to 5s.
209
929ba01c
PH
210PH/41 Added support for the use of login_cap features, on those BSD systems
211 that have them, for controlling the resources used by pipe deliveries.
212
2632889e
PH
213PH/42 The content-scanning code uses fopen() to create files in which to put
214 message data. Previously it was not paying any attention to the mode of
215 the files. Exim runs with umask(0) because the rest of the code creates
216 files with open(), and sets the required mode explicitly. Thus, these
217 files were ending up world-writeable. This was not a big issue, because,
218 being within the spool directory, they were not world-accessible. I have
219 created a function called modefopen, which takes an additional mode
220 argument. It sets umask(777), creates the file, chmods it to the required
221 mode, then resets the umask. All the relevant calls to fopen() in the
222 content scanning code have been changed to use this function.
223
944a9c55
PH
224PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
225 to 24 hours. This avoids potential overflow problems when processing G
226 and H retry rules. I suspect nobody ever tinkers with this value.
227
4a23603b
PH
228PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
229
4730f942
PH
230PH/45 When the plaintext authenticator is running as a client, the server's
231 challenges are checked to ensure they are valid base64 strings. By
232 default, the authentication attempt is cancelled if an invalid string is
233 received. Setting client_ignore_invalid_base64 true ignores these errors.
234 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
235 they are received. Thus, the responses can be made to depend on the
236 challenges. If an invalid string is ignored, an empty string is placed in
237 the variable.
238
30dba1e6
PH
239PH/46 Messages that are created by the autoreply transport now contains a
240 References: header, in accordance with RFCs 2822 and 3834.
241
382afc6b
PH
242PH/47 Added authenticated_sender_force to the smtp transport.
243
a86229cf
PH
244PH/48 The ${prvs expansion was broken on systems where time_t was long long.
245
50c99ba6
PH
246PH/49 Installed latest patch from the Sieve maintainer.
247
d35e429d
PH
248PH/50 When an Exim quota was set without a file count quota, and mailbox_size
249 was also set, the appendfile transport was unnecessarily scanning a
250 directory of message files (e.g. for maildir delivery) to find the count
251 of files (along with the size), even though it did not need this
252 information. It now does the scan only if it needs to find either the
253 size of the count of files.
254
f90d018c
PH
255PH/51 Added ${time_eval: to convert Exim time strings into seconds.
256
75def545
PH
257PH/52 Two bugs concerned with error handling when the smtp transport is
258 used in LMTP mode:
259
260 (i) Exim was not creating retry information for temporary errors given
261 for individual recipients after the DATA command when the smtp transport
262 was used in LMTP mode. This meant that they could be retried too
263 frequently, and not timed out correctly.
264
265 (ii) Exim was setting the flag that allows error details to be returned
266 for LMTP errors on RCPT commands, but not for LMTP errors for individual
267 recipients that were returned after the DATA command.
268
269PH/53 This is related to PH/52, but is more general: for any failing address,
270 when detailed error information was permitted to be returned to the
271 sender, but the error was temporary, then after the final timeout, only
272 "retry timeout exceeded" was returned. Now it returns the full error as
273 well as "retry timeout exceeded".
274
c46782ef
PH
275PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
276 do this, and (what is worse) MTAs that accept it.
277
71fafd95
PH
278PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
279 will now be deprecated.
280
2c5db4fd
PH
281PH/56 New os.c-cygwin from the Cygwin maintainer.
282
9cf6b11a
JJ
283JJ/06 exipick: added --unsorted option to allow unsorted output in all output
284 formats (previously only available in exim formats via -bpr, -bpru,
285 and -bpra. Now also available in native and exiqgrep formats)
286
287JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
288 with very large, slow to parse queues
289
290JJ/08 exipick: added ! as generic prefix to negate any criteria format
291
292JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
293
898d150f
PH
294PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
295 responses to authentication challenges, though it was showing the
296 challenges; (ii) I've removed the CR characters from the debug output for
297 SMTP output lines.
298
46218253
PH
299PH/58 Allow for the insertion of a newline as well as a space when a string
300 is turned into more than one encoded-word during RFC 2047 encoding. The
301 Sieve code now uses this.
302
e97957bc
PH
303PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
304 data_4xx, lost_connection, tls_required.
305
214e2000 306
5de37277
PH
307Exim version 4.60
308-----------------
309
cc38ddbf
PH
310PH/01 Two changes to the default runtime configuration:
311
312 (1) Move the checks for relay_from_hosts and authenticated clients from
313 after to before the (commented out) DNS black list checks.
314
315 (2) Add control=submission to the relay_from_hosts and authenticated
316 clients checks, on the grounds that messages accepted by these
317 statements are most likely to be submissions.
5de37277 318
72fdd6ae
PH
319PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
320
321 (1) Generate an error if the third argument for the ${prvs expansion is
322 not a single digit.
323
324 (2) Treat a missing third argument of ${prvscheck as if it were an empty
325 string.
326
327 (3) Reset the variables that are obtained from the first argument of
328 ${prvscheck and used in the second argument before leaving the code,
329 because their memory is reclaimed, so using them afterwards may do
330 silly things.
331
332 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
333 one (it's much easier than Tom thought :-).
334
335 (5) Because of (4), we can now allow for the use of $prvscheck_result
336 inside the third argument.
cb9328de 337
cb741023
PH
338PH/03 For some reason, the default setting of PATH when running a command from
339 a pipe transport was just "/usr/bin". I have changed it to
340 "/bin:/usr/bin".
341
f174f16e
PH
342PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
343 anything to be listed in the output from -bV.
b2f5a032 344
c25242d7
PH
345PH/05 When a filter generated an autoreply, the entire To: header line was
346 quoted in the delivery log line, like this:
347
348 => >A.N.Other <ano@some.domain> <original@ddress> ...
349
350 This has been changed so that it extracts the operative address. There
351 may be more than one such address. If so, they are comma-separated, like
352 this:
353
354 => >ano@some.domain,ona@other.domain <original@ddress> ...
355
82c19f95
PH
356PH/06 When a client host used a correct literal IP address in a HELO or EHLO
357 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
358 not being looked up in the rDNS to get a host name, Exim was showing the
359 IP address twice in Received: lines, even though the IP addresses were
360 identical. For example:
361
362 Received: from [1.2.3.4] (helo=[1.2.3.4])
363
364 However, if the real host name was known, it was omitting the HELO data
365 if it matched the actual IP address. This has been tidied up so that it
366 doesn't show the same IP address twice.
367
d7ffbc12
PH
368PH/07 When both +timestamp and +memory debugging was on, the value given by
369 $tod_xxx expansions could be wrong, because the tod_stamp() function was
370 called by the debug printing, thereby overwriting the timestamp buffer.
371 Debugging no longer uses the tod_stamp() function when +timestamp is set.
372
9f526266
PH
373PH/08 When the original message was included in an autoreply transport, it
374 always said "this is a copy of the message, including all the headers",
375 even if body_only or headers_only was set. It now gives an appropriate
376 message.
377
87fcc8b9
PH
378PH/09 Applied a patch from the Sieve maintainer which:
379
380 o fixes some comments
381 o adds the (disabled) notify extension core
382 o adds some debug output for the result of if/elsif tests
383 o points to the current vacation draft in the documentation
384 and documents the missing references header update
385
386 and most important:
387
388 o fixes a bug in processing the envelope test (when testing
389 multiple envelope elements, the last element determinted the
390 result)
391
456682f5
PH
392PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
393 Electronic Mail") by including:
394
395 Auto-submitted: auto-generated
396
397 in the messages that it generates (bounce messages and others, such as
398 warnings). In the case of bounce messages for non-SMTP mesages, there was
399 also a typo: it was using "Auto_submitted" (underscore instead of
400 hyphen). Since every message generated by Exim is necessarily in response
401 to another message, thes have all been changed to:
402
403 Auto-Submitted: auto-replied
404
405 in accordance with these statements in the RFC:
406
407 The auto-replied keyword:
408
409 - SHOULD be used on messages sent in direct response to another
410 message by an automatic process,
411
412 - MUST NOT be used on manually-generated messages,
413
414 - MAY be used on Delivery Status Notifications (DSNs) and Message
415 Disposition Notifications (MDNs),
416
417 - MUST NOT be used on messages generated by automatic or periodic
418 processes, except for messages which are automatic responses to
419 other messages.
420
3e46c1aa
PH
421PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
422 to the default Received: header definition.
456682f5 423
49826d12
PH
424PH/12 Added log selector acl_warn_skipped (default on).
425
eba0c039
PH
426PH/13 After a successful wildlsearch lookup, discard the values of numeric
427 variables because (a) they are in the wrong storage pool and (b) even if
428 they were copied, it wouldn't work properly because of the caching.
429
a0d6ba8a
PH
430PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
431 checking when decoding. Apparently there are clients that generate
432 overlong encoded strings. Why am I not surprised?
433
f0917727
PH
434PH/15 If the first argument of "${if match_address" was not empty, but did not
435 contain an "@" character, Exim crashed. Now it writes a panic log message
436 and treats the condition as false.
437
096fee00
PH
438PH/16 In autoreply, treat an empty string for "once" the same as unset.
439
024bd3c2
PH
440PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
441 extension "envelope-auth". The code is finished and in agreement with
442 other implementations, but there is no documentation so far and in fact,
443 nobody wrote the draft yet. This extension is currently #undef'ed, thus
444 not changing the active code.
445
446 Print executed "if" and "elsif" statements when debugging is used. This
447 helps a great deal to understand what a filter does.
448
449 Document more things not specified clearly in RFC3028. I had all this
450 sorted out, when out of a sudden new issues came to my mind. Oops."
451
df199fec
PH
452PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
453 (Bugzilla #53).
454
d27f1df3
PH
455PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
456 canonical form (as documented). However, after a host name lookup from
457 the IP address, check_host() was doing a simple string comparison with
458 addresses acquired from the DNS when checking that the found name did
459 have the original IP as one of its addresses. Since any found IPv6
460 addresses are likely to be in abbreviated form, the comparison could
461 fail. Luckily, there already exists a function for doing the comparison
462 by converting both addresses to binary, so now that is used instead of
463 the text comparison.
464
96776534
PH
465PH/20 There was another similar case to PH/19, when a complete host name was
466 given in a host list; looking up its IP address could give an abbreviated
467 form, whereas the current host's name might or might not be abbreviated.
468 The same fix has been applied.
469
5de37277 470
9a799bc0
PH
471Exim version 4.54
472-----------------
473
474PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
475 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
476 It now does.
477
99a4b039
PH
478PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
479 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
480
481PH/03 Typo: missing ".o" in src/pcre/Makefile.
482
4b233853
PH
483PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
484 header line, restrict the check to what is listed in RFCs 2369 and 2929.
485 Also, for "Auto-Submitted", treat anything other than "no" as
486 non-personal, in accordance with RFC 3834. (Previously it treated
487 anything starting "auto-" as non-personal.)
488
8857ccfd
PH
489TF/01 The control=submission/name=... option had a problem with syntax
490 errors if the name included a slash character. The /name= option
491 now slurps the rest of the string, so it can include any characters
492 but it must come last in the list of options (after /sender_retain
493 or /domain=).
494
433a2980
PH
495PH/05 Some modifications to the interface to the fake nameserver for the new
496 testing suite.
497
3e46c1aa 498
9a799bc0 499
e3a311ba
TK
500Exim version 4.53
501-----------------
502
503TK/01 Added the "success_on_redirect" address verification option. See
504 NewStuff for rationale and an example.
505
13b685f9
PH
506PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
507
395ff96d
PH
508PH/02 Patch to exigrep to allow it to work on syslog lines.
509
5b68f6e4
PH
510PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
511 fread() to skip over the body file's header line, because in Cygwin the
512 header line is locked and is inaccessible.
513
1ab52c69
PH
514PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
515 co-exist for some time) to make it clear that it is the Exim ID that is
516 referenced, not the Message-ID: header line.
517
b07e6aa3
PH
518PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
519 string_format() function, because snprintf() does not exist on all
520 operating systems.
521
254e032f
PH
522PH/06 The use of forbid_filter_existstest now also locks out the use of the
523 ${stat: expansion item.
524
3af76a81
PH
525PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
526 protocol synchronization error", to keep the pedants happy.
527
2548ba04
PH
528PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
529 well as for IRIX systems, when gcc is being used. See the host.c source
530 file for comments.
531
b6c6011d
PH
532PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
533
cf39cf57
PH
534PH/10 Named domain lists were not working if used in a queue_smtp_domains
535 setting.
536
f1513293
PH
537PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
538 transport and to the smtp transport in LMTP mode.
539
727549a4
PH
540TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
541
af46795e
PH
542PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
543 run a filter in a subprocess. This could lead to confusion in subsequent
544 lookups in the parent process. There should also be a search_tidyup() at
545 the end of the subprocess.
546
d7b47fd0
PH
547PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
548 only if the host matched helo_try_verify_hosts, which caused the
549 verification to occur when the EHLO/HELO command was issued. The ACL just
550 tested the remembered result. Now, if a previous verification attempt has
551 not happened, "verify = helo" does it there and then.
552
ee744174
JJ
553JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
554
b582ab87
PH
555TK/03 Fix log output including CR from clamd.
556
41a13e0a
PH
557PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
558 exist provoked a memory error which could cause a segfault.
559
f625cc5a
PH
560PH/15 Installed PCRE 6.2
561
562PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
563
21f7af35
PH
564PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
565 of the problem. Specifically, suggested +O2 rather than +O1 for the
566 HP-UX compiler.
567
31480e42
PH
568PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
569
2d280592
PH
570PH/20 If a delivery was routed to a non-standard port by means of an SRV
571 record, the port was not correctly logged when the outgoing_port log
572 selector was set (it logged the transort's default port).
573
7cd1141b
PH
574PH/21 Added support for host-specific ports to manualroute, queryprogram,
575 fallback_hosts, and "hosts" in the smtp transport.
576
577PH/22 If the log selector "outgoing_port" is set, the port is now also given on
578 host errors such as "Connection refused".
579
750af86e
PH
580PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
581 authentication with radiusclient 0.4.9:
582
583 - Error returned from rc_read_config was caught wrongly
584 - Username/password not passed on to radius server due to wrong length.
585
586 The presumption is that some radiusclient API changes for 4.51/PH/17
587 were not taken care of correctly. The code is still untested by me (my
588 Linux distribution still has 0.3.2 of radiusclient), but it was
589 contributed by a Radius user.
590
591PH/24 When doing a callout, the value of $domain wasn't set correctly when
592 expanding the "port" option of the smtp transport.
593
4304270b
TK
594TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
595 while reading a MIME header. Thanks to Tom Hughes for a patch.
596
750af86e
PH
597PH/24 Include config.h inside local_scan.h so that configuration settings are
598 available.
599
64ffc24f
PH
600PH/25 Make $smtp_command_argument available after all SMTP commands. This means
601 that in an ACL for RCPT (for example), you can examine exactly what was
602 received.
603
5dd9625b
PH
604PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
605 commands, but it was not correctly comparing the address with the actual
606 client host address. Thus, it would show the EHLO address in Received:
607 header lines when this was not necessary.
608
5591031b
PH
609PH/27 Added the % operator to ${eval:}.
610
ba18e66a
PH
611PH/28 Exim tries to create and chdir to its spool directory when it starts;
612 it should be ignoring failures (because with -C, for example, it has lost
613 privilege). It wasn't ignoring creation failures other than "already
614 exists".
615
9cec981f
PH
616PH/29 Added "crypteq" to the list of supported features that Exim outputs when
617 -bV or -d is used.
618
aa2b5c79
PH
619PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
620 because an input line was too long, either on its own, or by virtue of
1509d3a8
PH
621 too many continuations, the temporary file was not being removed, and the
622 return code was incorrect.
aa2b5c79 623
48a53b7f
PH
624PH/31 Missing "BOOL" in function definition in filtertest.c.
625
1c59d63b
PH
626PH/32 Applied Sieve patches from the maintainer.
627
671012da
TK
628TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
629
1509d3a8
PH
630PH/33 Added "verify = not_blind".
631
632PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
633 Local/Makefile (with some defaults set). These are used in built scripts
634 such as exicyclog, but they have never been used in the exim_install
635 script (though there are many overriding facilities there). I have
636 arranged that the exim_install script now takes note of these two
637 settings.
638
639PH/35 Installed configuration files for Dragonfly.
640
2fe1a124
PH
641PH/36 When a locally submitted message by a trusted user did not contain a
642 From: header, and the sender address was obtained from -f or from an SMTP
643 MAIL command, and the trusted user did not use -F to supply a sender
644 name, $originator_name was incorrectly used when constructing a From:
645 header. Furthermore, $originator_name was used for submission mode
646 messages from external hosts without From: headers in a similar way,
647 which is clearly wrong.
648
8800895a
PH
649PH/37 Added control=suppress_local_fixups.
650
ccfdb010
PH
651PH/38 When log_selector = +received_sender was set, and the addition of the
652 sender made the log line's construction buffer exactly full, or one byte
653 less than full, an overflow happened when the terminating "\n" was
654 subsequently added.
655
1130bfb0
PH
656PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
657 when the result of a list match is failure because a DNS lookup failed.
658
ebcb507f
PH
659PH/40 RM_COMMAND is now used in the building process.
660
c35e155c
PH
661PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
662 the "build-* directories that it finds.
663
95d1f782
PH
664PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
665 address in a domain literal was a prefix of an interface address.
666
fd6de02e
PH
667PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
668 when verifying a sender address, unless rewrite_headers is false.
669
58de37c5
PH
670PH/44 Wrote a long comment about why errors_to addresses are verified as
671 recipients, not senders.
672
261cf466
TF
673TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
674 the ratelimit ACL was added.
675
3ee512ff
PH
676PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
677
e08c430f
PH
678PH/46 Added extra information about PostgreSQL errors to the error string.
679
bef5a11f
PH
680PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
681 avoiding the need to install special zones in a real server. This is
682 backwards compatible; if it can't find the fake resolver, it drops back.
683 Thus, both old and new test suites can be run.
684
7546de58
TF
685TF/02 Added util/ratelimit.pl
686
e5d5a95f
TF
687TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
688 clock is set back in time.
689
2e88a017
TF
690TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
691 Candler <B.Candler@pobox.com>.
692
a5f65aa4
TF
693TF/05 The fix for PH/43 was not completely correct; widen_domains is always
694 OK for addresses that are the result of redirections.
695
e7726cbf
PH
696PH/48 A number of further additions for the benefit of the new test suite,
697 including a fake gethostbyname() that interfaces to the fake DNS resolver
698 (see PH/47 above).
699
a7fdad5b
TF
700TF/06 The fix for widen_domains has also been applied to qualify_single and
701 search_parents which are the other dnslookup options that can cause
702 header rewrites.
703
6af56900
PH
704PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
705 type ("H").
706
0925ede6
PH
707PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
708
66afa403
TF
709TF/07 Exim produced the error message "an SRV record indicated no SMTP
710 service" if it encountered an MX record with an empty target hostname.
711 The message is now "an MX or SRV record indicated no SMTP service".
712
0154e85a
TF
713TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
714 if the DNS of the sending site is misconfigured. This is quite a
715 common situation. This change restores the behaviour of treating a
716 helo verification defer as a failure.
717
16f12c76
PH
718PH/51 If self=fail was set on a router, the bounce message did not include the
719 actual error message.
720
bbe902f0 721
e5a9dba6
PH
722Exim version 4.52
723-----------------
724
725TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
726
22c3b60b
PH
727PH/01 When a transport filter timed out in a pipe delivery, and the pipe
728 command itself ended in error, the underlying message about the transport
729 filter timeout was being overwritten with the pipe command error. Now the
730 underlying error message should be appended to the second error message.
731
06a9b4b5
PH
732TK/01 Fix poll() being unavailable on Mac OSX 10.2.
733
c1ac6996
PH
734PH/02 Reduce the amount of output that "make" produces by default. Full output
735 can still be requested.
736
9c7a242c
PH
737PH/03 The warning log line about a condition test deferring for a "warn" verb
738 was being output only once per connection, rather than after each
739 occurrence (because it was using the same function as for successful
740 "warn" verbs). This seems wrong, so I have changed it.
741
87ba3f5f
PH
742TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
743 it should not have, which might have caused a crash in the right
744 circumstances, but probably never did.
745
746PH/04 Installed a modified version of Tony Finch's patch to make submission
747 mode fix the return path as well as the Sender: header line, and to
748 add a /name= option so that you can make the user's friendly name appear
749 in the header line.
750
29aba418
TF
751TF/03 Added the control = fakedefer ACL modifier.
752
fe0dab11
TF
753TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
754 Mark Lowes for thorough testing.
870f6ba8 755
11d337a4
TK
756TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
757
758TK/03 Merged latest SRS patch from Miles Wilton.
759
415c8f3b
PH
760PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
761 with the definition in sysexits.h (which is #included earlier).
762 Fortunately, Exim does not actually use EX_OK. The code used to try to
763 preserve the sysexits.h value, by assumimg that macro definitions were
764 scanned for macro replacements. I have been disabused of this notion,
765 so now the code just undefines EX_OK before #including unistd.h.
11d337a4 766
958541e9
PH
767PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
768 in the smtp transport. When a block could not be written in a single
769 write() function, the timeout was being re-applied to each part-write.
770 This seems wrong - if the receiver was accepting one byte at a time it
771 would take for ever. The timeout is now adjusted when this happens. It
772 doesn't have to be particularly precise.
773
c206415f
TK
774TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
775 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
776
2a4be8f9
PH
777PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
778 without a domain if the check to <postmaster@domain> fails.
779
1cba11c5
SC
780SC/01 Eximstats: added -xls and the ability to specify output files
781 (patch written by Frank Heydlauf).
782
783SC/02 Eximstats: use FileHandles for outputing results.
784
785SC/03 Eximstats: allow any combination of xls, txt, and html output.
786
787SC/04 Eximstats: fixed display of large numbers with -nvr option
788
789SC/05 Eximstats: fixed merging of reports with empty tables.
790
791SC/06 Eximstats: added the -include_original_destination flag
792
793SC/07 Eximstats: removed tabs and trailing whitespace.
794
1005d00e
TK
795TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
796
797TK/06 MBOX spool code: Add real "From " MBOX separator line
798 so the .eml file is really in mbox format (even though
799 most programs do not really care). Patch from Alex Miller.
800
801TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
802 The latter is generated from $received_to and is only set if the
803 message has one envelope recipient. SA can use these headers,
804 obviously out-of-the-box. Patch from Alex Miller.
805
9b4768fa
PH
806PH/08 The ${def test on a variable was returning false if the variable's
807 value was "0", contrary to what the specification has always said!
808 The result should be true unless the variable is empty.
809
810PH/09 The syntax error of a character other than { following "${if
811 def:variable_name" (after optional whitespace) was not being diagnosed.
812 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
813 accidental colon was present, for example, could give incorrect results.
814
0d7eb84a
PH
815PH/10 Tidied the code in a number of places where the st_size field of a stat()
816 result is used (not including appendfile, where other changes are about
817 to be made).
818
819PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
820 This involved changing a lot of size variables from int to off_t. It
821 should work with maildirs and everything.
822
40727bee
TK
823TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
824 spamd dying while we are connected to it.
825
554d2369
TF
826TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
827 <jgh@wizmail.org>
828
1f922db1
PH
829PH/12 Applied Alex Kiernan's patch for the API change for the error callback
830 function for BDB 4.3.
831
ef213c3b
PH
832PH/13 Changed auto_thaw such that it does not apply to bounce messages.
833
8ac170f3
PH
834PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
835 the sources for PCRE have been re-arranged and more files are now
836 involved.
837
b1c749bb
PH
838PH/15 The code I had for printing potentially long long variables in PH/11
839 above was not the best (it lost precision). The length of off_t variables
840 is now inspected at build time, and an appropriate printing format (%ld
c6c2dc1d
PH
841 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
842 to be "long long int" or "long int". This is needed for the internal
843 formatting function string_vformat().
b1c749bb 844
4aac9b49
PH
845PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
846 the configuration file to be ":syslog", then the script "guesses" where
847 the logs files are, rather than using the compiled in default. In our
848 case the guess is not the same as the compiled default, so the script
849 suddenly stopped working when I started to use syslog. The patch checks
850 to see if log_file_path is "". If so, it attempts to read it from exim
851 with no configuration file to get the compiled in version, before it
852 falls back to the previous guessing code."
853
294520c8
TK
854TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
855 implementing BATV in an Exim configuration. See NewStuff for the gory
856 details.
857
5bd022fe
PH
858PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
859 Makefile that are specific to HP-UX.
860
90e9ce59
PH
861PH/18 If the "use_postmaster" option was set for a recipient callout together
862 with the "random" option, the postmaster address was used as the MAIL
863 FROM address for the random test, but not for the subsequent recipient
864 test. It is now used for both.
865
5ea81592
PH
866PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
867 patch removes a few documentation additions to RFC 3028, because the
868 latest draft now contains them. It adds the new en;ascii-case comparator
869 and a new error check for 8bit text in MIME parts. Comparator and
870 require names are now matched exactly. I enabled the subaddress
871 extension, but it is not well tested yet (read: it works for me)."
872
c6c2dc1d
PH
873PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
874 rework some of the code of TK/09 above to avoid the hardwired use of
875 "%lld" and "long long". Replaced the call to snprintf() with a call to
876 string_vformat().
877
fffffe4c
PH
878PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
879 records point to non-existent hosts", "retry timeout exceeded", and
880 "retry time not reached for any host after a long failure period".
ca02eafb 881
9a26b6b2
PH
882PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
883 experimental DomainKeys support:
884
885 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
886 (2) On an error such as an illegally used "control", the wrong name for
887 the control was given.
888
889 These problems did NOT occur unless DomainKeys support was compiled.
890
4aee0225
PH
891PH/23 Added daemon_startup_retries and daemon_startup_sleep.
892
32d668a5
PH
893PH/24 Added ${if match_ip condition.
894
8187c3f3
PH
895PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
896 databases so that it will be absolutely obvious if a crash occurs in the
897 DB library. This is a regular occurrence (often caused by mis-matched
898 db.h files).
899
ff790e47 900PH/26 Insert a lot of missing (void) casts for functions such as chown(),
f1e894f3
PH
901 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
902 picked up on a user's system that detects such things. There doesn't seem
903 to be a gcc warning option for this - only an attribute that has to be
904 put on the function's prototype. It seems that in Fedora Core 4 they have
905 set this on a number of new functions. No doubt there will be more in due
906 course.
ff790e47 907
5417f6d1
PH
908PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
909 specify a transport. However, if an address that was verified by such a
910 router was the subject of a callout, Exim crashed because it tried to
911 read the rcpt_include_affixes from the non-existent transport. Now it
912 just assumes that the setting of that option is false. This bug was
913 introduced by 4.51/PH/31.
914
59cf8544
PH
915PH/28 Changed -d+all to exclude +memory, because that information is very
916 rarely of interest, but it makes the output a lot bigger. People tend to
917 do -d+all out of habit.
918
e7ad8a65
PH
919PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
920 code in os-type was giving problems when libc.so lives in lib64, like on
921 x86_64 Fedora Core.
922
ade42478
PH
923PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
924 aren't the modern standard, and it seems that some systems' include files
925 don't always have them. Exim was already checking for some of the newer
926 ones like T_AAAA, and defining it itself. I've added checks for all the
927 record types that Exim uses.
928
182ad5cf
PH
929PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
930 not automatically generating a new one, as it is supposed to. This
931 prevented TLS from working. If the file did exist, but contained invalid
932 data, a new version was generated, as expected. It was only the case of a
933 non-existent file that was broken.
934
b0d9fc80
TK
935TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
936 with a change in libdomainkeys > 0.64.
937
938TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
939 from DNS. If the selector record carries the flag, it now has
940 precedence over the domain-wide flag.
941
942TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
943
47c7a64a
PH
944PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
945 the use of a port name that isn't defined in /etc/services) occurred, the
946 message was deferred as in a normal delivery, and thus remained on the
947 spool, instead of being failed because of the mua_wrapper setting. This
948 is now fixed, and I tidied up some of the mua_wrapper messages at the
949 same time.
950
a388bce4
SC
951SC/08 Eximstats: whilst parsing the mainlog(s), store information about
952 the messages in a hash of arrays rather than using individual hashes.
953 This is a bit cleaner and results in dramatic memory savings, albeit
954 at a slight CPU cost.
955
956SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
957 as requested by Marc Sherman.
958
959SC/10 Eximstats: added histograms for user specified patterns as requested
960 by Marc Sherman.
961
0793e4ed
SC
962SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
963
c58b88df
PH
964PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
965 fopen() in the content-scanning modules that did not already have it.
966
e7ad8a65 967
7982096b
PH
968Exim version 4.51
969-----------------
970
1a46a8c5
PH
971TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
972 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
973
2f079f46 974TK/02 Fix ACL "control" statement not being available in MIME ACL.
1a46a8c5
PH
975
976TK/03 Fix ACL "regex" condition not being available in MIME ACL.
977
978PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
979 to test Sieve filters that use "vacation".
980
981PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
982 that changes the way the GnuTLS parameters are stored in the cache file.
983 The new format can be generated externally. For backward compatibility,
984 if the data in the cache doesn't make sense, Exim assumes it has read an
985 old-format file, and it generates new data and writes a new file. This
986 means that you can't go back to an older release without removing the
987 file.
988
989PH/03 A redirect router that has both "unseen" and "one_time" set does not
990 work if there are any delivery delays because "one_time" forces the
991 parent to be marked "delivered", so its unseen clone is never tried
992 again. For this reason, Exim now forbids the simultaneous setting of
993 these two options.
994
995PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
996 redirected to themselves ("homonym" addresses). Read the long ChangeLog
997 entry if you want to know the details. The fix, however, neglected to
998 consider the case when local delivery batching is involved. The test for
999 "previously delivered" was not happening when checking to see if an
1000 address could be batched with a previous (undelivered) one; under
1001 certain circumstances this could lead to multiple deliveries to the same
c2c19e9d 1002 address.
1a46a8c5
PH
1003
1004PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
1005 in its include files, and this causes problems building Exim.
1006
1007PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
1008 header_syntax) but Exim was just ignoring anything given after a slash.
1009 In particular, this caused confusion with an attempt to use "verify =
1010 reverse_host_lookup/defer_ok". An error is now given when options are
1011 supplied for verify items that do not have them. (Maybe reverse_host_
1012 lookup should have a defer_ok option, but that's a different point.)
1013
1014PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
1015 defined by RFC 821) to 2048, because there were problems with some AUTH
1016 commands, and RFC 1869 says the size should be increased for extended
1017 SMTP commands that take arguments.
1018
1019PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
1020 Finch).
1021
1022PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
1023 "unknown" error; now it says that the functionality isn't in the binary.
8d67ada3 1024
49c2d5ea
PH
1025PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
1026 an address' error message when a string expansion fails (syntax or
f331f3b6
PH
1027 whatever). Otherwise the password may appear in the log. Following change
1028 PH/42 below, there is no longer a chance of it appearing in a bounce
1029 message.
49c2d5ea 1030
bf759a8b
PH
1031PH/11 Installed exipick version 20050225.0 from John Jetmore.
1032
83364d30
PH
1033PH/12 If the last host in a fallback_hosts list was multihomed, only the first
1034 of its addresses was ever tried. (Bugzilla bug #2.)
1035
7999bbd7
PH
1036PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
1037 the result incorrectly in the debug output. (It correctly added a newline
1038 to what was transported.)
1039
7dbf77c9
PH
1040TF/01 Added $received_time.
1041
74e0617f
PH
1042PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
1043 commented out examples of how to interface to a virus scanner and to
1044 SpamAssassin. Also added commented examples of av_scanner and
1045 spamd_address settings.
1046
2f079f46
PH
1047PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
1048 and controls are allowed in which ACLs. There were a couple of minor
1049 errors. Some of the entries in the conditions table (which is a table of
1050 where they are NOT allowed) were getting very unwieldy; rewrote them as a
1051 negation of where the condition IS allowed.
1052
8c841523
PH
1053PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
1054
7766a4f0
PH
1055PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
1056 header file does not have a version number, so I've had to invent a new
1057 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
1058 API. The code is untested by me (my Linux distribution still has 0.3.2 of
1059 radiusclient), but it was contributed by a Radius user.
1060
8b417f2c
PH
1061PH/18 Installed Lars Mainka's patch for the support of CRL collections in
1062 files or directories, for OpenSSL.
1063
901f42cb
PH
1064PH/19 When an Exim process that is running as root has to create an Exim log
1065 file, it does so in a subprocess that runs as exim:exim so as to get the
1066 ownership right at creation (otherwise, other Exim processes might see
1067 the file with the wrong ownership). There was no test for failure of this
1068 fork() call, which would lead to the process getting stuck as it waited
1069 for a non-existent subprocess. Forks do occasionally fail when resources
1070 run out. I reviewed all the other calls to fork(); they all seem to check
1071 for failure.
1072
f9b9210e
PH
1073PH/20 When checking for unexpected SMTP input at connect time (before writing
1074 the banner), Exim was not dealing correctly with a non-positive return
1075 from the read() function. If the client had disconnected by this time,
1076 the result was a log entry for a synchronization error with an empty
1077 string after "input=" when read() returned zero. If read() returned -1
1078 (an event I could not check), uninitialized data bytes were printed.
1079 There were reports of junk text (parts of files, etc) appearing after
1080 "input=".
1081
54cdb463
PH
1082PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
1083
cf00dad6
PH
1084PH/22 Added support for macro redefinition, and (re)definition in between
1085 driver and ACL definitions.
1086
acb1b346
PH
1087PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
1088 forgetting to use the resulting value; it was using the unexpanded value.
1089
c5ddb310
PH
1090PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
1091 hadn't been configured. The fix is from Juergen Kreileder, who
1092 understands it better than I do:
1093
1094 "Here's what I see happening with three configured cyrus_sasl
1095 authenticators configured (plain, login, cram-md5):
1096
1097 On startup auth_cyrus_sasl_init() gets called for each of these.
1098 This means three calls to sasl_listmech() without a specified mech_list.
1099 => SASL tests which mechs of all available mechs actually work
1100 => three warnings about OTP not working
1101 => the returned list contains: plain, login, cram-md5, digest-md5, ...
1102
1103 With the patch, sasl_listmech() also gets called three times. But now
1104 SASL's mech_list option is set to the server_mech specified in the the
1105 authenticator. Or in other words, the answer from sasl_listmech()
1106 gets limited to just the mech you're testing for (which is different
1107 for each call.)
1108 => the return list contains just 'plain' or 'login', 'cram-md5' or
1109 nothing depending on the value of ob->server_mech.
1110
1111 I've just tested the patch: Authentication still works fine,
1112 unavailable mechs specified in the exim configuration are still
1113 caught, and the auth.log warnings about OTP are gone."
1114
31619da6
PH
1115PH/25 When debugging is enabled, the contents of the command line are added
1116 to the debugging output, even when log_selector=+arguments is not
1117 specified.
1118
bebaf0fc
PH
1119PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
1120 answer is "GNU", and only if the return is "GNU/something" is the answer
1121 "Linux".
1122
475fe28a
PH
1123PH/27 $acl_verify_message is now set immediately after the failure of a
1124 verification in an ACL, and so is available in subsequent modifiers. In
1125 particular, the message can be preserved by coding like this:
1126
1127 warn !verify = sender
1128 set acl_m0 = $acl_verify_message
1129
1130 Previously, $acl_verify_message was set only while expanding "message"
1131 and "log_message" when a very denied access.
1132
7e8bec7a
PH
1133PH/28 Modified OS/os.c-Linux with
1134
1135 -#ifndef OS_LOAD_AVERAGE
1136 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
1137
1138 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
1139 nomenclature these days.)
1140
e4a89c47
PH
1141PH/29 Installed patch from the Sieve maintainer that adds the options
1142 sieve_useraddress and sieve_subaddress to the redirect router.
1143
5ca2a9a1
PH
1144PH/30 In these circumstances:
1145 . Two addresses routed to the same list of hosts;
1146 . First host does not offer TLS;
1147 . First host accepts first address;
1148 . First host gives temporary error to second address;
1149 . Second host offers TLS and a TLS session is established;
1150 . Second host accepts second address.
1151 Exim incorrectly logged both deliveries with the TLS parameters (cipher
1152 and peerdn, if requested) that were in fact used only for the second
1153 address.
7e8bec7a 1154
c688b954
PH
1155PH/31 When doing a callout as part of verifying an address, Exim was not paying
1156 attention to any local part prefix or suffix that was matched by the
1157 router that accepted the address. It now behaves in the same way as it
1158 does for delivery: the affixes are removed from the local part unless
1159 rcpt_include_affixes is set on the transport.
1160
fed77020
PH
1161PH/32 Add the sender address, as F=<...>, to the log line when logging a
1162 timeout during the DATA phase of an incoming message.
1163
7fe1560f
PH
1164PH/33 Sieve envelope tests were broken for match types other than :is. I have
1165 applied a patch sanctioned by the Sieve maintainer.
c688b954 1166
ebb6e6d5
PH
1167PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
1168 the uid or gid is negative. A case of a negative gid caused this to be
1169 noticed. The fix allows for either to be negative.
1170
9c4e8f60
PH
1171PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
1172 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
1173 overlooked.
1174
1175PH/36 The change PH/12 above was broken. Fixed it.
1176
d7174846
PH
1177PH/37 Exim used to check for duplicate addresses in the middle of routing, on
1178 the grounds that routing the same address twice would always produce the
1179 same answer. This might have been true once, but it is certainly no
1180 longer true now. Routing a child address may depend on the previous
1181 routing that produced that child. Some complicated redirection strategies
1182 went wrong when messages had multiple recipients, and made Exim's
1183 behaviour dependent on the order in which the addresses were given.
1184
1185 I have moved the duplicate checking until after the routing is complete.
1186 Exim scans the addresses that are assigned to local and remote
1187 transports, and removes any duplicates. This means that more work will be
1188 done, as duplicates will always all be routed, but duplicates are
1189 presumably rare, so I don't expect this is of any significance.
1190
1191 For deliveries to pipes, files, and autoreplies, the duplicate checking
1192 still happens during the routing process, since they are not going to be
1193 routed further.
1194
cfe75fc3
PH
1195PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
1196 It corrects a timeout issue with spamd. This is Ian's comment: "The
1197 background is that sometimes spamd either never reads data from a
1198 connection it has accepted, or it never writes response data. The exiscan
1199 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
1200 blindly assumes that writes won't block so it may never time out."
1201
be22d70e
PH
1202PH/39 Allow G after quota size as well as K and M.
1203
0612b098
PH
1204PH/40 The value set for $authenticated_id in an authenticator may not contain
1205 binary zeroes or newlines because the value is written to log lines and
1206 to spool files. There was no check on this. Now the value is run through
1207 the string_printing() function so that such characters are converted to
1208 printable escape sequences.
1209
2e0c1448
PH
1210PH/41 $message_linecount is a new variable that contains the total number of
1211 lines in the message. Compare $body_linecount, which is the count for the
1212 body only.
1213
447d236c
PH
1214PH/42 Exim no longer gives details of delivery errors for specific addresses in
1215 bounce and delay warning messages, except in certain special cases, which
1216 are as follows:
1217
1218 (a) An SMTP error message from a remote host;
1219 (b) A message specified in a :fail: redirection;
1220 (c) A message specified in a "fail" command in a system filter;
1221 (d) A message specified in a FAIL return from the queryprogram router;
1222 (e) A message specified by the cannot_route_message router option.
1223
1224 In these cases only, Exim does include the error details in bounce and
1225 warning messages. There are also a few cases where bland messages such
1226 as "unrouteable address" or "local delivery error" are given.
1227
d20976dc
PH
1228PH/43 $value is now also set for the "else" part of a ${run expansion.
1229
f656d135
PH
1230PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
1231 being worked on, but at least Exim now implements the latest version to
1232 play with."
1233
2e2a30b4
PH
1234PH/45 In a pipe transport, although a timeout while waiting for the pipe
1235 process to complete was treated as a delivery failure, a timeout while
1236 writing the message to the pipe was logged, but erroneously treated as a
1237 successful delivery. Such timeouts include transport filter timeouts. For
1238 consistency with the overall process timeout, these timeouts are now
1239 treated as errors, giving rise to delivery failures by default. However,
1240 there is now a new Boolean option for the pipe transport called
1241 timeout_defer, which, if set TRUE, converts the failures into defers for
1242 both kinds of timeout. A transport filter timeout is now identified in
1243 the log output.
1244
9176e9f0
PH
1245PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
1246 systems where "make" and "gmake" are different, calling "gmake" at top
1247 level broke things. I've arranged for the value of $(MAKE) to be passed
1248 from the Makefile to this script so that it can call the same version of
1249 "make".
1250
7982096b 1251
bbe902f0
PH
1252A note about Exim versions 4.44 and 4.50
1253----------------------------------------
1254
1255Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
1256changes of various kinds. As a consequence, a big documentation update was
1257needed. This delayed the release for rather longer than seemed good, especially
1258in the light of a couple of (minor) security issues. Therefore, the changes
1259that fixed bugs were backported into 4.43, to create a 4.44 maintenance
1260release. So 4.44 and 4.50 are in effect two different branches that both start
1261from 4.43.
1262
1263I have left the 4.50 change log unchanged; it contains all the changes since
12644.43. The change log for 4.44 is below; many of its items are identical to
1265those for 4.50. This seems to be the most sensible way to preserve the
1266historical information.
1267
1268
f7b63901 1269Exim version 4.50
495ae4b0
PH
1270-----------------
1271
5fe762f6
PH
1272 1. Minor wording change to the doc/README.SIEVE file.
1273
139059f6 1274 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
5fe762f6 1275 computation of the current number of files was incorrect.
495ae4b0 1276
7086e875
PH
1277 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
1278 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
1279 place.
1280
35af9f61
PH
1281 4. Give more explanation in the error message when the command for a transport
1282 filter fails to execute.
1283
b668c215
PH
1284 5. There are several places where Exim runs a non-Exim command in a
1285 subprocess. The SIGUSR1 signal should be disabled for these processes. This
1286 was being done only for the command run by the queryprogram router. It is
1287 now done for all such subprocesses. The other cases are: ${run, transport
1288 filters, and the commands run by the lmtp and pipe transports.
1289
a494b1e1
PH
1290 6. Added CONFIGURE_GROUP build-time option.
1291
1292 7. Some older OS have a limit of 256 on the maximum number of file
1293 descriptors. Exim was using setrlimit() to set 1000 as a large value
1294 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
1295 systems. I've change it so that if it can't get 1000, it tries for 256.
35edf2ff 1296
c5fcb476
PH
1297 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
1298 was an oversight, and furthermore, ever since the addition of extra
1299 controls (e.g. 4.43/32), the checks on when to allow different forms of
1300 "control" were broken. There should now be diagnostics for all cases when a
1301 control that does not make sense is encountered.
1302
69358f02
PH
1303 9. Added the /retain_sender option to "control=submission".
1304
5be20824
PH
130510. $recipients is now available in the predata ACL (oversight).
1306
eb2c0248
PH
130711. Tidy the search cache before the fork to do a delivery from a message
1308 received from the command line. Otherwise the child will trigger a lookup
1309 failure and thereby defer the delivery if it tries to use (for example) a
1310 cached ldap connection that the parent has called unbind on.
1311
2a3eea10
PH
131212. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
1313 of $address_data from the recipient verification was clobbered by the
1314 sender verification.
1315
131613. The value of address_data from a sender verification is now available in
1317 $sender_address_data in subsequent conditions in the ACL statement.
1318
23c7ff99
PH
131914. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
1320
4deaf07d
PH
132115. Added a new option "connect=<time>" to callout options, to set a different
1322 connection timeout.
1323
926e1192
PH
132416. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
1325 was its contents. (It was OK if the option was not defined at all.)
1326
650edc6f
PH
132717. A "Completed" log line is now written for messages that are removed from
1328 the spool by the -Mrm option.
1329
2c7db3f5
PH
133018. New variables $sender_verify_failure and $recipient_verify_failure contain
1331 information about exactly what failed.
1332
3d235903
PH
133319. Added -dd to debug only the daemon process.
1334
7c7ad977
PH
133520. Incorporated Michael Haardt's patch to ldap.c for improving the way it
1336 handles timeouts, both on the server side and network timeouts. Renamed the
1337 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
1338
981756db
PH
133921. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
1340 It is now set to "smtps".
1341
d4eb88df
PH
134222. $host_address is now set to the target address during the checking of
1343 ignore_target_hosts.
1344
134523. When checking ignore_target_hosts for an ipliteral router, no host name was
1346 being passed; this would have caused $sender_host_name to have been used if
1347 matching the list had actually called for a host name (not very likely,
1348 since this list is usually IP addresses). A host name is now passed as
1349 "[x.x.x.x]".
1350
7d468ab8
PH
135124. Changed the calls that set up the SIGCHLD handler in the daemon to use the
1352 code that specifies a non-restarting handler (typically sigaction() in
1353 modern systems) in an attempt to fix a rare and obscure crash bug.
1354
135525. Narrowed the window for a race in the daemon that could cause it to ignore
1356 SIGCHLD signals. This is not a major problem, because they are used only to
1357 wake it up if nothing else does.
1358
62c0818f
PH
135926. A malformed maildirsize file could cause Exim to calculate negative values
1360 for the mailbox size or file count. Odd effects could occur as a result.
1361 The maildirsize information is now recalculated if the size or filecount
1362 end up negative.
1363
26034054
PH
136427. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
1365 support for a long time. Removed HAVE_SYS_VFS_H.
1366
af66f652
PH
136728. Installed the latest version of exipick from John Jetmore.
1368
90af77f4
PH
136929. In an address list, if the pattern was not a regular expression, an empty
1370 subject address (from a bounce message) matched only if the pattern was an
1371 empty string. Non-empty patterns were not even tested. This was the wrong
1372 because it is perfectly reasonable to use an empty address as part of a
1373 database query. An empty address is now tested by patterns that are
1374 lookups. However, all the other forms of pattern expect the subject to
1375 contain a local part and a domain, and therefore, for them, an empty
1376 address still always fails if the pattern is not itself empty.
1377
d8ef3577
PH
137830. Exim went into a mad DNS loop when attempting to do a callout where the
1379 host was specified on an smtp transport, and looking it up yielded more
1380 than one IP address.
1381
5cb8cbc6
PH
138231. Re-factored the code for checking spool and log partition space into a
1383 function that finds that data and another that does the check. The former
1384 is then used to implement four new variables: $spool_space, $log_space,
1385 $spool_inodes, and $log_inodes.
1386
14702f5b
PH
138732. The RFC2047 encoding function was originally intended for short strings
1388 such as real names; it was not keeping to the 75-character limit for
1389 encoded words that the RFC imposes. It now respects the limit, and
1390 generates multiple encoded words if necessary. To be on the safe side, I
1391 have increased the buffer size for the ${rfc2047: expansion operator from
1392 1024 to 2048 bytes.
1393
063b1e99
PH
139433. It is now permitted to omit both strings after an "if" condition; if the
1395 condition is true, the result is "true". As before, when the second string
1396 is omitted, a false condition yields an empty string. This makes it less
1397 cumbersome to write custom ACL and router conditions.
1398
652e1b65
PH
139934. Failure to deliver a bounce message always caused it to be frozen, even if
1400 there was an errors_to setting on the router. The errors_to setting is now
1401 respected.
1402
6f0c9a4f
PH
140335. If an IPv6 address is given for -bh or -bhc, it is now converted to the
1404 canonical form (fully expanded) before being placed in
1405 $sender_host_address.
1406
33397d19
PH
140736. The table in the code that translates DNS record types into text (T_A to
1408 "A" for instance) was missing entries for NS and CNAME. It is just possible
1409 that this could have caused confusion if both these types were looked up
1410 for the same domain, because the text type is used as part of Exim's
1411 per-process caching. But the chance of anyone hitting this buglet seems
1412 very small.
1413
7bb56e1f
PH
141437. The dnsdb lookup has been extended in a number of ways.
1415
1416 (1) There is a new type, "zns", which walks up the domain tree until it
1417 finds some nameserver records. It should be used with care.
1418
ea3bc19b
PH
1419 (2) There is a new type, "mxh", which is like "mx" except that it returns
1420 just the host names, not the priorities.
1421
1422 (3) It is now possible to give a list of domains (or IP addresses) to be
ff4dbb19
PH
1423 looked up. The behaviour when one of the lookups defers can be
1424 controlled by a keyword.
7bb56e1f 1425
ea3bc19b 1426 (4) It is now possible to specify the separator character for use when
7bb56e1f 1427 multiple records are returned.
33397d19 1428
0bcb2a0e
PH
142938. The dnslists ACL condition has been extended: it is now possible to supply
1430 a list of IP addresses and/or domains to be looked up in a particular DNS
1431 domain.
1432
2ac0e484
PH
143339. Added log_selector=+queue_time_overall.
1434
4e1fde53
PH
143540. When running the queue in the test harness, wait just a tad after forking a
1436 delivery process, to get repeatability of debugging output.
1437
de365ded
PH
143841. Include certificate and key file names in error message when GnuTLS fails
1439 to set them up, because the GnuTLS error message doesn't include the name
1440 of the failing file when there is a problem reading it.
1441
f05da2e8
PH
144242. Allow both -bf and -bF in the same test run.
1443
d6453af2
PH
144443. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
1445
f7b63901
PH
144644. The "Exiscan patch" is now merged into the mainline Exim source.
1447
144845. Sometimes the final signoff response after QUIT could fail to get
1449 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
1450 < 0 before doing a fflush(). This bug looks as though it goes back to the
1451 introduction of TLS in release 3.20, but "sometimes" must have been rare
1452 because the tests only now provoked it.
1453
a444213a
PH
145446. Reset the locale to "C" after calling embedded Perl, in case it was changed
1455 (this can affect the format of dates).
1456
0ec020ea
PH
145747. exim_tidydb, when checking for the continued existence of a message for
1458 which it has found a message-specific retry record, was not finding
1459 messages that were in split spool directories. Consequently, it was
1460 deleting retry records that should have stayed in existence.
1461
b1206957
PH
146248. Steve fixed some bugs in eximstats.
1463
146449. The SPA authentication driver was not abandoning authentication and moving
1465 on to the next authenticator when an expansion was forced to fail,
1466 contradicting the general specification for all authenticators. Instead it
1467 was generating a temporary error. It now behaves as specified.
1468
26dd5a95
PH
146950. The default ordering of permitted cipher suites for GnuTLS was pessimal
1470 (the order specifies the preference for clients). The order is now AES256,
1471 AES128, 3DES, ARCFOUR128.
1472
343b2385
PH
147351. Small patch to Sieve code - explicitly set From: when generating an
1474 autoreply.
1475
1c5466b9
PH
147652. Exim crashed if a remote delivery caused a very long error message to be
1477 recorded - for instance if somebody sent an entire SpamAssassin report back
1478 as a large number of 550 error lines. This bug was coincidentally fixed by
1479 increasing the size of one of Exim's internal buffers (big_buffer) that
1480 happened as part of the Exiscan merge. However, to be on the safe side, I
1481 have made the code more robust (and fixed the comments that describe what
1482 is going on).
1483
55ee9ee3
PH
148453. Now that there can be additional text after "Completed" in log lines (if
1485 the queue_time_overall log selector is set), a one-byte patch to exigrep
1486 was needed to allow it to recognize "Completed" as not the last thing in
1487 the line.
1488
d38f8232
PH
148954. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
1490 patch that reportedly fixes this has been added. I am not expert enough to
1491 create a test for it. This is what the patch creator wrote:
1492
1493 "I found a little strange behaviour of ldap code when working with
1494 Windows 2003 AD Domain, where users was placed in more than one
1495 Organization Units. When I tried to give exim partial DN, the exit code
1496 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
1497 But simultaneously result of request was absolutely normal ldap result,
1498 so I produce this patch..."
1499
3295e65b
PH
1500 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
1501 REFERENCE, so I have modified the code to exclude the patch when that macro
1502 is not defined.
1503
7102e136
PH
150455. Some experimental protocols are using DNS PTR records for new purposes. The
1505 keys for these records are domain names, not reversed IP addresses. The
b975ba52
PH
1506 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
1507 leaves it alone. Component reversal etc. now happens only for IP addresses.
ea3a6f44 1508 CAN-2005-0021
7102e136 1509
3ca0ba97
PH
151056. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
1511
c2bcbe20
PH
151257. Double the size of the debug message buffer (to 2048) so that more of very
1513 long debug lines gets shown.
1514
18ce445d
PH
151558. The exicyclog utility now does better if the number of log files to keep
1516 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
1517
1f5b4c3d
PH
151859. Two changes related to the smtp_active_hostname option:
1519
1520 (1) $smtp_active_hostname is now available as a variable.
1521 (2) The default for smtp_banner uses $smtp_active_hostname instead
1522 of $primary_hostname.
1523
b975ba52
PH
152460. The host_aton() function is supposed to be passed a string that is known
1525 to be a valid IP address. However, in the case of IPv6 addresses, it was
1526 not checking this. This is a hostage to fortune. Exim now panics and dies
1527 if the condition is not met. A case was found where this could be provoked
85b87bc2
PH
1528 from a dnsdb PTR lookup with an IPv6 address that had more than 8
1529 components; fortuitously, this particular loophole had already been fixed
1530 by change 4.50/55 above.
1531
1532 If there are any other similar loopholes, the new check in host_aton()
1533 itself should stop them being exploited. The report I received stated that
1534 data on the command line could provoke the exploit when Exim was running as
1535 exim, but did not say which command line option was involved. All I could
1536 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
1537 running as the user.
ea3a6f44 1538 CAN-2005-0021
85b87bc2
PH
1539
154061. There was a buffer overflow vulnerability in the SPA authentication code
1541 (which came originally from the Samba project). I have added a test to the
1542 spa_base64_to_bits() function which I hope fixes it.
ea3a6f44 1543 CAN-2005-0022
b975ba52 1544
17ffcae7
PH
154562. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
1546 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
1547
d95f9fdb
PH
154863. The daemon start-up calls getloadavg() while still root for those OS that
1549 need the first call to be done as root, but it missed one case: when
1550 deliver_queue_load_max is set with deliver_drop_privilege. This is
1551 necessary for the benefit of the queue runner, because there is no re-exec
1552 when deliver_drop_privilege is set.
1553
86b8287f
PH
155464. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
1555 This has been fixed.
1556
60dc5e56
PH
155765. Caching of lookup data for "hosts =" ACL conditions, when a named host list
1558 was in use, was not putting the data itself into the right store pool;
1559 consequently, it could be overwritten for a subsequent message in the same
1560 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
1561 the caching.)
1562
533244af
PH
156366. Added hosts_max_try_hardlimit to the smtp transport, default 50.
1564
a5a28604
PH
156567. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
1566 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
1567 the function were treating the return as a boolean value, which happened to
1568 work because 0=false and not-0=true, but is not correct code.
1569
7e634d24
PH
157068. The host_aton() function was not handling scoped IPv6 addresses (those
1571 with, for example, "%eth0" on the end) correctly.
1572
3e11c26b
PH
157369. Fixed some compiler warnings in acl.c for the bitmaps specified with
1574 negated items (that is, ~something) in unsigned ints. Some compilers
1575 apparently mutter when there is no cast.
1576
6729cf78
PH
157770. If an address verification called from an ACL failed, and did not produce a
1578 user-specific message (i.e. there was only a "system" message), nothing was
1579 put in $acl_verify_message. In this situation, it now puts the system
1580 message there.
1581
00f00ca5
PH
158271. Change 4.23/11 added synchronization checking at the start of an SMTP
1583 session; change 4.31/43 added the unwanted input to the log line - except
1584 that it did not do this in the start of session case. It now does.
1585
c9bdd01c
PH
158672. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
1587 This is wrong and can cause the other end to generate a synchronization
1588 error if it is another Exim or anything else that does the synchronization
1589 check. A QUIT command is no longer sent after a timeout.
1590
d43194df
PH
159173. $host_lookup_deferred has been added, to make it easier to detect DEFERs
1592 during host lookups.
1593
fe5b5d0b
PH
159474. The defer_ok option of callout verification was not working if it was used
1595 when verifying addresses in header lines, that is, for this case:
1596
1597 verify = header_sender/callout=defer_ok
1598
76a2d7ba
PH
159975. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
1600 those file descriptors could be used for SMTP connections. If anything
1601 wrote to stderr (the example that came up was "warn" in embedded Perl), it
1602 could be sent to the SMTP client, causing chaos. The daemon now opens
1603 stdin, stdout, and stderr to /dev/null when it puts itself into the
1604 background.
1605
160676. Arrange for output from Perl's "warn" command to be written to Exim's main
1607 log by default. The user can override this with suitable Perl magic.
1608
04f7d5b9
PH
160977. The use of log_message on a "discard" ACL verb, which is supposed to add to
1610 the log message when discard triggers, was not working for the DATA ACL or
1611 for the non-SMTP ACL.
1612
bc60667e
PH
161378. Error message wording change in sieve.c.
1614
bb6e88ff
PH
161579. If smtp_accept_max_per_host was set, the number of connections could be
1616 restricted to fewer than expected, because the daemon was trying to set up
1617 a new connection before checking whether the processes handling previous
1618 connections had finished. The check for completed processes is now done
1619 earlier. On busy systems, this bug wouldn't be noticed because something
1620 else would have woken the daemon, and it would have reaped the completed
1621 process earlier.
1622
1e70f85b
PH
162380. If a message was submitted locally by a user whose login name contained one
1624 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
1625 It caused a spool format error. I have fixed the spool reading code. A
1626 related problem was that the "from" clause in the Received: line became
1627 illegal because of the space(s). It is now covered by ${quote_local_part.
1628
162981. Included the latest eximstats from Steve (adds average sizes to HTML Top
1630 tables).
1631
4e01f9d6
PH
163282. Updated OS/Makefile-AIX as per message from Mike Meredith.
1633
1ee1cef2
PH
163483. Patch from Sieve maintainer to fix unterminated string problem in
1635 "vacation" handling.
1636
6e2b4ccc
PH
163784. Some minor changes to the Linux configuration files to help with other
1638 OS variants using glibc.
1639
8e669ac1
PH
164085. One more patch for Sieve to update vacation handling to latest spec.
1641
495ae4b0 1642
bbe902f0
PH
1643----------------------------------------------------
1644See the note above about the 4.44 and 4.50 releases.
1645----------------------------------------------------
1646
1647
1648Exim version 4.44
1649-----------------
1650
1651 1. Change 4.43/35 introduced a bug that caused file counts to be
1652 incorrectly computed when quota_filecount was set in an appendfile
1653 transport
1654
1655 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
1656 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
1657 place.
1658
1659 3. Give more explanation in the error message when the command for a transport
1660 filter fails to execute.
1661
1662 4. There are several places where Exim runs a non-Exim command in a
1663 subprocess. The SIGUSR1 signal should be disabled for these processes. This
1664 was being done only for the command run by the queryprogram router. It is
1665 now done for all such subprocesses. The other cases are: ${run, transport
1666 filters, and the commands run by the lmtp and pipe transports.
1667
1668 5. Some older OS have a limit of 256 on the maximum number of file
1669 descriptors. Exim was using setrlimit() to set 1000 as a large value
1670 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
1671 systems. I've change it so that if it can't get 1000, it tries for 256.
1672
1673 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
1674 was an oversight, and furthermore, ever since the addition of extra
1675 controls (e.g. 4.43/32), the checks on when to allow different forms of
1676 "control" were broken. There should now be diagnostics for all cases when a
1677 control that does not make sense is encountered.
1678
1679 7. $recipients is now available in the predata ACL (oversight).
1680
1681 8. Tidy the search cache before the fork to do a delivery from a message
1682 received from the command line. Otherwise the child will trigger a lookup
1683 failure and thereby defer the delivery if it tries to use (for example) a
1684 cached ldap connection that the parent has called unbind on.
1685
1686 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
1687 of $address_data from the recipient verification was clobbered by the
1688 sender verification.
1689
169010. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
1691 was its contents. (It was OK if the option was not defined at all.)
1692
169311. A "Completed" log line is now written for messages that are removed from
1694 the spool by the -Mrm option.
1695
169612. $host_address is now set to the target address during the checking of
1697 ignore_target_hosts.
1698
169913. When checking ignore_target_hosts for an ipliteral router, no host name was
1700 being passed; this would have caused $sender_host_name to have been used if
1701 matching the list had actually called for a host name (not very likely,
1702 since this list is usually IP addresses). A host name is now passed as
1703 "[x.x.x.x]".
1704
170514. Changed the calls that set up the SIGCHLD handler in the daemon to use the
1706 code that specifies a non-restarting handler (typically sigaction() in
1707 modern systems) in an attempt to fix a rare and obscure crash bug.
1708
170915. Narrowed the window for a race in the daemon that could cause it to ignore
1710 SIGCHLD signals. This is not a major problem, because they are used only to
1711 wake it up if nothing else does.
1712
171316. A malformed maildirsize file could cause Exim to calculate negative values
1714 for the mailbox size or file count. Odd effects could occur as a result.
1715 The maildirsize information is now recalculated if the size or filecount
1716 end up negative.
1717
171817. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
1719 support for a long time. Removed HAVE_SYS_VFS_H.
1720
ea3a6f44 172118. Updated exipick to current release from John Jetmore.
bbe902f0
PH
1722
172319. Allow an empty sender to be matched against a lookup in an address list.
1724 Previously the only cases considered were a regular expression, or an
1725 empty pattern.
1726
172720. Exim went into a mad DNS lookup loop when doing a callout where the
1728 host was specified on the transport, if the DNS lookup yielded more than
1729 one IP address.
1730
ea3a6f44
NM
173121. The RFC2047 encoding function was originally intended for short strings
1732 such as real names; it was not keeping to the 75-character limit for
1733 encoded words that the RFC imposes. It now respects the limit, and
1734 generates multiple encoded words if necessary. To be on the safe side, I
1735 have increased the buffer size for the ${rfc2047: expansion operator from
1736 1024 to 2048 bytes.
bbe902f0 1737
ea3a6f44
NM
173822. Failure to deliver a bounce message always caused it to be frozen, even if
1739 there was an errors_to setting on the router. The errors_to setting is now
1740 respected.
bbe902f0
PH
1741
174223. If an IPv6 address is given for -bh or -bhc, it is now converted to the
1743 canonical form (fully expanded) before being placed in
1744 $sender_host_address.
1745
174624. Updated eximstats to version 1.33
1747
ea3a6f44
NM
174825. Include certificate and key file names in error message when GnuTLS fails
1749 to set them up, because the GnuTLS error message doesn't include the name
1750 of the failing file when there is a problem reading it.
bbe902f0
PH
1751
175226. Expand error message when OpenSSL has problems setting up cert/key files.
ea3a6f44 1753 As per change 25.
bbe902f0 1754
ea3a6f44
NM
175527. Reset the locale to "C" after calling embedded Perl, in case it was changed
1756 (this can affect the format of dates).
bbe902f0 1757
ea3a6f44
NM
175828. exim_tidydb, when checking for the continued existence of a message for
1759 which it has found a message-specific retry record, was not finding
1760 messages that were in split spool directories. Consequently, it was
1761 deleting retry records that should have stayed in existence.
bbe902f0
PH
1762
176329. eximstats updated to version 1.35
1764 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
1765 1.35 - bugfix such that pie charts by volume are generated correctly
1766
ea3a6f44
NM
176730. The SPA authentication driver was not abandoning authentication and moving
1768 on to the next authenticator when an expansion was forced to fail,
1769 contradicting the general specification for all authenticators. Instead it
1770 was generating a temporary error. It now behaves as specified.
bbe902f0 1771
ea3a6f44
NM
177231. The default ordering of permitted cipher suites for GnuTLS was pessimal
1773 (the order specifies the preference for clients). The order is now AES256,
1774 AES128, 3DES, ARCFOUR128.
bbe902f0 1775
ea3a6f44
NM
177631. Small patch to Sieve code - explicitly set From: when generating an
1777 autoreply.
bbe902f0 1778
ea3a6f44
NM
177932. Exim crashed if a remote delivery caused a very long error message to be
1780 recorded - for instance if somebody sent an entire SpamAssassin report back
1781 as a large number of 550 error lines. This bug was coincidentally fixed by
1782 increasing the size of one of Exim's internal buffers (big_buffer) that
1783 happened as part of the Exiscan merge. However, to be on the safe side, I
1784 have made the code more robust (and fixed the comments that describe what
1785 is going on).
bbe902f0 1786
ea3a6f44
NM
178733. Some experimental protocols are using DNS PTR records for new purposes. The
1788 keys for these records are domain names, not reversed IP addresses. The
1789 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
1790 leaves it alone. Component reversal etc. now happens only for IP addresses.
bbe902f0
PH
1791 CAN-2005-0021
1792
ea3a6f44
NM
179334. The host_aton() function is supposed to be passed a string that is known
1794 to be a valid IP address. However, in the case of IPv6 addresses, it was
1795 not checking this. This is a hostage to fortune. Exim now panics and dies
1796 if the condition is not met. A case was found where this could be provoked
1797 from a dnsdb PTR lookup with an IPv6 address that had more than 8
1798 components; fortuitously, this particular loophole had already been fixed
1799 by change 4.50/55 or 4.44/33 above.
1800
1801 If there are any other similar loopholes, the new check in host_aton()
1802 itself should stop them being exploited. The report I received stated that
1803 data on the command line could provoke the exploit when Exim was running as
1804 exim, but did not say which command line option was involved. All I could
1805 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
1806 running as the user.
bbe902f0
PH
1807 CAN-2005-0021
1808
ea3a6f44
NM
180935. There was a buffer overflow vulnerability in the SPA authentication code
1810 (which came originally from the Samba project). I have added a test to the
1811 spa_base64_to_bits() function which I hope fixes it.
bbe902f0
PH
1812 CAN-2005-0022
1813
ea3a6f44
NM
181436. The daemon start-up calls getloadavg() while still root for those OS that
1815 need the first call to be done as root, but it missed one case: when
1816 deliver_queue_load_max is set with deliver_drop_privilege. This is
1817 necessary for the benefit of the queue runner, because there is no re-exec
1818 when deliver_drop_privilege is set.
bbe902f0 1819
ea3a6f44
NM
182037. Caching of lookup data for "hosts =" ACL conditions, when a named host list
1821 was in use, was not putting the data itself into the right store pool;
1822 consequently, it could be overwritten for a subsequent message in the same
1823 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
1824 the caching.)
bbe902f0 1825
ea3a6f44
NM
182638. Sometimes the final signoff response after QUIT could fail to get
1827 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
1828 < 0 before doing a fflush(). This bug looks as though it goes back to the
1829 introduction of TLS in release 3.20, but "sometimes" must have been rare
1830 because the tests only now provoked it.
bbe902f0
PH
1831
1832
495ae4b0
PH
1833Exim version 4.43
1834-----------------
1835
1836 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
1837 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
1838 Now it returns an integer. A number of calls were still expecting a T/F
1839 return. Fortuitously, in all cases, the tests worked in OK situations,
1840 which is the norm. However, things would have gone wrong on any write
1841 failures on the smtp file descriptor. This function is used when sending
1842 messages over SMTP and also when doing verify callouts.
1843
1844 2. When Exim is called to do synchronous delivery of a locally submitted
1845 message (the -odf or -odi options), it no longer closes stderr before doing
1846 the delivery.
1847
1848 3. Implemented the mua_wrapper option.
1849
1850 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
1851
1852 5. Implemented the functions header_remove(), header_testname(),
1853 header_add_at_position(), and receive_remove_recipient(), and exported them
1854 to local_scan().
1855
1856 6. If an ACL "warn" statement specified the addition of headers, Exim already
1857 inserted X-ACL-Warn: at the start if there was no header name. However, it
1858 was not making this test for the second and subsequent header lines if
1859 there were newlines in the string. This meant that an invalid header could
1860 be inserted if Exim was badly configured.
1861
1862 7. Allow an ACL "warn" statement to add header lines at the start or after all
1863 the Received: headers, as well as at the end.
1864
1865 8. Added the rcpt_4xx retry error code.
1866
1867 9. Added postmaster_mailfrom=xxx to callout verification option.
1868
186910. Added mailfrom=xxxx to the callout verification option, for verify=
1870 header_sender only.
1871
187211. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
1873 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
1874
187512. Inserted some casts to stop certain compilers warning when using pointer
1876 differences as field lengths or precisions in printf-type calls (mostly
1877 affecting debugging statements).
1878
187913. Added optional readline() support for -be (dynamically loaded).
1880
188114. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
1882 same clock tick as a message's arrival, so that its received time was the
1883 same as the "first fail" time on the retry record, and that message
1884 remained on the queue past the ultimate address timeout, every queue runner
1885 would try a delivery (because it was past the ultimate address timeout) but
1886 after another failure, the ultimate address timeout, which should have then
1887 bounced the address, did not kick in. This was a "< instead of <=" error;
1888 in most cases the first failure would have been in the next clock tick
1889 after the received time, and all would be well.
1890
189115. The special items beginning with @ in domain lists (e.g. @mx_any) were not
1892 being recognized when the domain list was tested by the match_domain
1893 condition in an expansion string.
1894
189516. Added the ${str2b64: operator.
1896
189717. Exim was always calling setrlimit() to set a large limit for the number of
1898 processes, without checking whether the existing limit was already
1899 adequate. (It did check for the limit on file descriptors.) Furthermore,
1900 errors from getrlimit() and setrlimit() were being ignored. Now they are
1901 logged to the main and panic logs, but Exim does carry on, to try to do its
1902 job under whatever limits there are.
1903
190418. Imported PCRE 5.0.
1905
190619. Trivial typo in log message " temporarily refused connection" (the leading
1907 space).
1908
190920. If the log selector return_path_on_delivery was set and an address was
1910 redirected to /dev/null, the delivery process crashed because it assumed
1911 that a return path would always be set for a "successful" delivery. In this
1912 case, the whole delivery is bypassed as an optimization, and therefore no
1913 return path is set.
1914
191521. Internal re-arrangement: the function for sending a challenge and reading
1916 a response while authentication was assuming a zero-terminated challenge
1917 string. It's now changed to take a pointer and a length, to allow for
1918 binary data in such strings.
1919
192022. Added the cyrus_sasl authenticator (code supplied by MBM).
1921
192223. Exim was not respecting finduser_retries when seeking the login of the
1923 uid under which it was called; it was always trying 10 times. (The default
1924 setting of finduser_retries is zero.) Also, it was sleeping after the final
1925 failure, which is pointless.
1926
192724. Implemented tls_on_connect_ports.
1928
192925. Implemented acl_smtp_predata.
1930
193126. If the domain in control=submission is set empty, Exim assumes that the
1932 authenticated id is a complete email address when it generates From: or
1933 Sender: header lines.
1934
193527. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
1936 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
1937 chown and chgrp in /bin and hostname in /usr/bin.
1938
193928. Exim was keeping the "process log" file open after each use, just as it
1940 does for the main log. This opens the possibility of it remaining open for
1941 long periods when the USR1 signal hits a daemon. Occasional processlog
1942 errors were reported, that could have been caused by this. Anyway, it seems
1943 much more sensible not to leave this file open at all, so that is what now
1944 happens.
1945
194629. The long-running daemon process does not normally write to the log once it
1947 has entered its main loop, and it closes the log before doing so. This is
1948 so that log files can straightforwardly be renamed and moved. However,
1949 there are a couple of unusual error situations where the daemon does write
1950 log entries, and I had neglected to close the log afterwards.
1951
195230. The text of an SMTP error response that was received during a remote
1953 delivery was being truncated at 512 bytes. This is too short for some of
1954 the long messages that one sometimes sees. I've increased the limit to
1955 1024.
1956
195731. It is now possible to make retry rules that apply only when a message has a
1958 specific sender, in particular, an empty sender.
1959
196032. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
1961 it possible to be selective about when SMTP synchronization is enforced.
1962
196333. Added "control = caseful_local_part" and "control = "caselower_local_part".
1964
196532. Implemented hosts_connection_nolog.
1966
196733. Added an ACL for QUIT.
1968
196934. Setting "delay_warning=" to disable warnings was not working; it gave a
1970 syntax error.
1971
197235. Added mailbox_size and mailbox_filecount to appendfile.
1973
197436. Added control = no_multiline_responses to ACLs.
1975
197637. There was a bug in the logic of the code that waits for the clock to tick
1977 in the case where the clock went backwards by a substantial amount such
1978 that the microsecond fraction of "now" was more than the microsecond
1979 fraction of "then" (but the whole seconds number was less).
1980
198138. Added support for the libradius Radius client library this is found on
1982 FreeBSD (previously only the radiusclient library was supported).
1983
1984
1985Exim version 4.42
1986-----------------
1987
1988 1. When certain lookups returned multiple values in the form name=value, the
1989 quoting of the values was not always being done properly. Specifically:
1990 (a) If the value started with a double quote, but contained no whitespace,
1991 it was not quoted.
1992 (b) If the value contained whitespace other than a space character (i.e.
1993 tabs or newlines or carriage returns) it was not quoted.
1994 This fix has been applied to the mysql and pgsql lookups by writing a
1995 separate quoting function and calling it from the lookup code. The fix
1996 should probably also be applied to nisplus, ibase and oracle lookups, but
1997 since I cannot test any of those, I have not disturbed their existing code.
1998
1999 2. A hit in the callout cache for a specific address caused a log line with no
2000 reason for rejecting RCPT. Now it says "Previous (cached) callout
2001 verification failure".
2002
2003 3. There was an off-by-one bug in the queryprogram router. An over-long
2004 return line was truncated at 256 instead of 255 characters, thereby
2005 overflowing its buffer with the terminating zero. As well as fixing this, I
2006 have increased the buffer size to 1024 (and made a note to document this).
2007
2008 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
2009 when Exim is waiting for an SMTP response from a remote server, Exim
2010 restarts its select() call on the socket, thereby resetting its timeout.
2011 This is not a problem when such interrupts are rare. Somebody set up a cron
2012 job to run exiwhat every 2 minutes, which is less than the normal select()
2013 timeout (5 or 10 minutes). This meant that the select() timeout never
2014 kicked in because it was always reset. I have fixed this by comparing the
2015 time when an interrupt arrives with the time at the start of the first call
2016 to select(). If more time than the timeout has elapsed, the interrupt is
2017 treated as a timeout.
2018
2019 5. Some internal re-factoring in preparation for the addition of Sieve
2020 extensions (by MH). In particular, the "personal" test is moved to a
2021 separate function, and given an option for scanning Cc: and Bcc: (which is
2022 not set for Exim filters).
2023
2024 6. When Exim created an email address using the login of the caller as the
2025 local part (e.g. when creating a From: or Sender: header line), it was not
2026 quoting the local part when it contained special characters such as @.
2027
2028 7. Installed new OpenBSD configuration files.
2029
2030 8. Reworded some messages for syntax errors in "and" and "or" conditions to
2031 try to make them clearer.
2032
2033 9. Callout options, other than the timeout value, were being ignored when
2034 verifying sender addresses in header lines. For example, when using
2035
2036 verify = header_sender/callout=no_cache
2037
2038 the cache was (incorrectly) being used.
2039
204010. Added a missing instance of ${EXE} to the exim_install script; this affects
2041 only the Cygwin environment.
2042
204311. When return_path_on_delivery was set as a log selector, if different remote
2044 addresses in the same message used different return paths and parallel
2045 remote delivery occurred, the wrong values would sometimes be logged.
2046 (Whenever a remote delivery process finished, the return path value from
2047 the most recently started remote delivery process was logged.)
2048
204912. RFC 3848 specifies standard names for the "with" phrase in Received: header
2050 lines when AUTH and/or TLS are in use. This is the "received protocol"
2051 field. Exim used to use "asmtp" for authenticated SMTP, without any
2052 indication (in the protocol name) for TLS use. Now it follows the RFC and
2053 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
2054 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
2055 names appear in log lines as well as in Received: header lines.
2056
205713. Installed MH's patches for Sieve to add the "copy" and "vacation"
2058 extensions, and comparison tests, and to fix some bugs.
2059
206014. Changes to the "personal" filter test:
2061
2062 (1) The test was buggy in that it was just doing the equivalent of
2063 "contains" tests on header lines. For example, if a user's address was
2064 anne@some.where, the "personal" test would incorrectly be true for
2065
2066 To: susanne@some.where
2067
2068 This test is now done by extracting each address from the header in turn,
2069 and checking the entire address. Other tests that are part of "personal"
2070 are now done using regular expressions (for example, to check local parts
2071 of addresses in From: header lines).
2072
2073 (2) The list of non-personal local parts in From: addresses has been
2074 extended to include "listserv", "majordomo", "*-request", and "owner-*",
2075 taken from the Sieve specification recommendations.
2076
2077 (3) If the message contains any header line starting with "List-" it is
2078 treated as non-personal.
2079
2080 (4) The test for "circular" in the Subject: header line has been removed
2081 because it now seems ill-conceived.
2082
208315. Minor typos in src/EDITME comments corrected.
2084
208516. Installed latest exipick from John Jetmore.
2086
208717. If headers_add on a router specified a text string that was too long for
2088 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
2089 of string_sprintf() is now avoided.
2090
209118. $message_body_size was not set (it was always zero) when running the DATA
2092 ACL and the local_scan() function.
2093
209419. For the "mail" command in an Exim filter, no default was being set for
2095 the once_repeat time, causing a random time value to be used if "once" was
2096 specified. (If the value happened to be <= 0, no repeat happened.) The
2097 default is now 0s, meaning "never repeat". The "vacation" command was OK
2098 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
2099 (I found it when inspecting the code).
2100
210120. There is now an overall timeout for performing a callout verification. It
2102 defaults to 4 times the callout timeout, which applies to individual SMTP
2103 commands during the callout. The overall timeout applies when there is more
2104 than one host that can be tried. The timeout is checked before trying the
2105 next host. This prevents very long delays if there are a large number of
2106 hosts and all are timing out (e.g. when the network connections are timing
2107 out). The value of the overall timeout can be changed by specifying an
2108 additional sub-option for "callout", called "maxwait". For example:
2109
2110 verify = sender/callout=5s,maxwait=20s
2111
211221. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
2113 to the end before writing, but this should make it even safer).
2114
211522. Exim was forgetting that it had advertised PIPELINING for the second and
2116 subsequent messages on an SMTP connection. It was also not resetting its
2117 memory on STARTTLS and an internal HELO.
2118
211923. When Exim logs an SMTP synchronization error within a session, it now
2120 records whether PIPELINING has been advertised or not.
2121
212224. Added 3 instances of "(long int)" casts to time_t variables that were being
2123 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
2124 rather than long int.
2125
212625. Installed the latest Cygwin configuration files from the Cygwin maintainer.
2127
212826. Added the never_mail option to autoreply.
2129
2130
2131Exim version 4.41
2132-----------------
2133
2134 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
2135 crash if the getsockname() call failed; this can happen if a connection is
2136 closed very soon after it is established. The problem was simply in the
2137 order in which certain operations were done, causing Exim to try to write
2138 to the SMTP stream before it had set up the file descriptor. The bug has
2139 been fixed by making things happen in the correct order.
2140
2141
2142Exim version 4.40
2143-----------------
2144
2145 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
2146 before the connection was closed, thus losing the rejection response.
2147
2148 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
2149 some early Solaris releases, but causes trouble in current releases where
2150 socklen_t is defined.
2151
2152 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
2153 always exist.
2154
2155 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
2156 configured.
2157
2158 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
2159 that releasing the top of it at the end releases what was used for sub-
2160 expansions (unless the block got too big). However, discard this block if
2161 the first thing is a variable or header, so that we can use its block when
2162 it is dynamic (useful for very large $message_headers, for example).
2163
2164 6. Lookups now cache *every* query, not just the most recent. A new, separate
2165 store pool is used for this. It can be recovered when all lookup caches are
2166 flushed. Lookups now release memory at the end of their result strings.
2167 This has involved some general refactoring of the lookup sources.
2168
2169 7. Some code has been added to the store_xxx() functions to reduce the amount
2170 of flapping under certain conditions.
2171
2172 8. log_incoming_interface used to affect only the <= reception log lines. Now
2173 it causes the local interface and port to be added to several more SMTP log
2174 lines, for example "SMTP connection from", and rejection lines.
2175
2176 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
2177
217810. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
2179
218011. If $host_data was set by virtue of a hosts lookup in an ACL, its value
2181 could be overwritten at the end of the current message (or the start of a
2182 new message if it was set in a HELO ACL). The value is now preserved for
2183 the duration of the SMTP connection.
2184
218512. If a transport had a headers_rewrite setting, and a matching header line
2186 contained an unqualified address, that address was qualified, even if it
2187 did not match any rewriting rules. The underlying bug was that the values
2188 of the flags that permit the existence of unqualified sender and recipient
2189 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
2190 non-local messages, and by -bnq for local messages) were not being
2191 preserved with the message after it was received.
2192
219313. When Exim was logging an SMTP synchronization error, it could sometimes log
2194 "next input=" as part of the text comprising the host identity instead of
2195 the correct text. The code was using the same buffer for two different
2196 strings. However, depending on which order the printing function evaluated
2197 its arguments, the bug did not always show up. Under Linux, for example, my
2198 test suite worked just fine.
2199
220014. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
2201 doesn't work with some older versions of Perl. It has been changed to "my",
2202 which in any case is probably the better facility to use.
2203
220415. A really picky compiler found some instances of statements for creating
2205 error messages that either had too many or two few arguments for the format
2206 string.
2207
220816. The size of the buffer for calls to the DNS resolver has been increased
2209 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
2210 for addresses that have a lot of PTR records. This alleviates a problem; it
2211 does not fully solve it.
2212
221317. A dnsdb lookup for PTR records that receives more data than will fit in the
2214 buffer now truncates the list and logs the incident, which is the same
2215 action as happens when Exim is looking up a host name and its aliases.
2216 Previously in this situation something unpredictable would happen;
2217 sometimes it was "internal error: store_reset failed".
2218
221918. If a server dropped the connection unexpectedly when an Exim client was
2220 using GnuTLS and trying to read a response, the client delivery process
2221 crashed while trying to generate an error log message.
2222
222319. If a "warn" verb in an ACL added multiple headers to a message in a single
2224 string, for example:
2225
2226 warn message = H1: something\nH2: something
2227
2228 the text was added as a single header line from Exim's point of view
2229 though it ended up OK in the delivered message. However, searching for the
2230 second and subsequent header lines using $h_h2: did not work. This has been
2231 fixed. Similarly, if a system filter added multiple headers in this way,
2232 the routers could not see them.
2233
223420. Expanded the error message when iplsearch is called with an invalid key to
2235 suggest using net-iplsearch in a host list.
2236
223721. When running tests using -bh, any delays imposed by "delay" modifiers in
2238 ACLs are no longer actually imposed (and a message to that effect is
2239 output).
2240
224122. If a "gecos" field in a passwd entry contained escaped characters, in
2242 particular, if it contained a \" sequence, Exim got it wrong when building
2243 a From: or a Sender: header from that name. A second bug also caused
2244 incorrect handling when an unquoted " was present following a character
2245 that needed quoting.
2246
224723. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
2248 was not being matched caselessly.
2249
225024. Arranged for all hyphens in the exim.8 source to be escaped with
2251 backslashes.
2252
225325. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
2254 properly. Recipient callout cache records were still being keyed to include
2255 the sender, even when use_sender was set false. This led to far more
2256 callouts that were necessary. The sender is no longer included in the key
2257 when use_sender is false.
2258
225926. Added "control = submission" modifier to ACLs.
2260
226127. Added the ${base62d: operator to decode base 62 numbers.
2262
226328. dnsdb lookups can now access SRV records.
2264
226529. CONFIGURE_OWNER can be set at build time to define an alternative owner for
2266 the configuration file.
2267
226830. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
2269 (-v) mode. This makes the output for a verbose queue run more intelligible.
2270
227131. Added a use_postmaster feature to recipient callouts.
2272
227332. Added the $body_zerocount variable, containing the number of binary zero
2274 bytes in the message body.
2275
227633. The time of last modification of the "new" subdirectory is now used as the
2277 "mailbox time last read" when there is a quota error for a maildir
2278 delivery.
2279
228034. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
2281
228235. Added +ignore_unknown as a special item in host lists.
2283
228436. Code for decoding IPv6 addresses in host lists is now included, even if
2285 IPv6 support is not being compiled. This fixes a bug in which an IPv6
2286 address was recognized as an IP address, but was then not correctly decoded
2287 into binary, causing unexpected and incorrect effects when compared with
2288 another IP address.
2289
2290
2291Exim version 4.34
2292-----------------
2293
2294 1. Very minor rewording of debugging text in manualroute to say "list of
2295 hosts" instead of "hostlist".
2296
2297 2. If verify=header_syntax was set, and a header line with an unqualified
2298 address (no domain) and a large number of spaces between the end of the
2299 name and the colon was received, the reception process suffered a buffer
2300 overflow, and (when I tested it) crashed. This was caused by some obsolete
2301 code that should have been removed. The fix is to remove it!
2302
2303 3. When running in the test harness, delay a bit after writing a bounce
2304 message to get a bit more predictability in the log output.
2305
2306 4. Added a call to search_tidyup() just before forking a reception process. In
2307 theory, someone could use a lookup in the expansion of smtp_accept_max_
2308 per_host which, without the tidyup, could leave open a database connection.
2309
2310 5. Added the variables $recipient_data and $sender_data which get set from a
2311 lookup success in an ACL "recipients" or "senders" condition, or a router
2312 "senders" option, similar to $domain_data and $local_part_data.
2313
2314 6. Moved the writing of debug_print from before to after the "senders" test
2315 for routers.
2316
2317 7. Change 4.31/66 (moving the time when the Received: is generated) caused
2318 problems for message scanning, either using a data ACL, or using
2319 local_scan() because the Received: header was not generated till after they
2320 were called (in order to set the time as the time of reception completion).
2321 I have revised the way this works. The header is now generated after the
2322 body is received, but before the ACL or local_scan() are called. After they
2323 are run, the timestamp in the header is updated.
2324
2325
2326Exim version 4.33
2327-----------------
2328
2329 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
2330 before starting a queue runner without re-exec. This happened only when
2331 deliver_drop_privilege was set or when the Exim user was set to root. The
2332 effect of the bug was that timeouts during subsequent deliveries caused
2333 crashes instead of being properly handled. The handler is now left at its
2334 default (and expected) setting.
2335
2336 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
2337 message, again when deliver_drop_privilege is set or Exim is run as root.
2338 The bug described in (1) was not present in this case, but the tidying up
2339 of the other signals was missing. I have made the two cases consistent.
2340
2341 3. The ignore_target_hosts setting on a manualroute router was being ignored
2342 for hosts that were looked up using the /MX notation.
2343
2344 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
2345 in domain lists.
2346
2347 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
2348 operated on the sender address. After changing the $sender_address to <>
2349 for the sender address verify, Exim was re-instated it as the original
2350 (before rewriting) address, but remembering that it had rewritten it, so it
2351 wasn't rewriting it again. This bug also had the effect of breaking the
2352 sender address verification caching when the sender address was rewritten.
2353
2354 6. The ignore_target_hosts option was being ignored by the ipliteral router.
2355 This has been changed so that if the ip literal address matches
2356 ignore_target_hosts, the router declines.
2357
2358 7. Added expansion conditions match_domain, match_address, and match_local_
2359 part (NOT match_host).
2360
2361 8. The placeholder for the Received: header didn't have a length field set.
2362
2363 9. Added code to Exim itself and to exim_lock to test for a specific race
2364 condition that could lead to file corruption when using MBX delivery. The
2365 issue is with the lockfile that is created in /tmp. If this file is removed
2366 after a process has opened it but before that process has acquired a lock,
2367 there is the potential for a second process to recreate the file and also
2368 acquire a lock. This could lead to two Exim processes writing to the file
2369 at the same time. The added code performs the same test as UW imapd; it
2370 checks after acquiring the lock that its file descriptor still refers to
2371 the same named file.
2372
237310. The buffer for building added header lines was of fixed size, 8192 bytes.
2374 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
2375 when Exim is built.
2376
237711. Added the smtp_active_hostname option. If used, this will typically be made
2378 to depend on the incoming interface address. Because $interface_address is
2379 not set up until the daemon has forked a reception process, error responses
2380 that can happen earlier (such as "too many connections") no longer contain
2381 a host name.
2382
238312. If an expansion in a condition on a "warn" statement fails because a lookup
2384 defers, the "warn" statement is abandoned, and the next ACL statement is
2385 processed. Previously this caused the whole ACL to be aborted.
2386
238713. Added the iplsearch lookup type.
2388
238914. Added ident_timeout as a log selector.
2390
239115. Added tls_certificate_verified as a log selector.
2392
239316. Added a global option tls_require_ciphers (compare the smtp transport
2394 option of the same name). This controls incoming TLS connections.
2395
239617. I finally figured out how to make tls_require_ciphers do a similar thing
2397 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
2398 before starting the TLS session.
2399
240018. Tabs are now shown as \t in -bP output.
2401
240219. If the log selector return_path_on_delivery was set, Exim crashed when
2403 bouncing a message because it had too many Received: header lines.
2404
240520. If two routers both had headers_remove settings, and the first one included
2406 a superfluous trailing colon, the final name in the first list and the
2407 first name in the second list were incorrectly joined into one item (with a
2408 colon in the middle).
2409
2410
2411Exim version 4.32
2412-----------------
2413
2414 1. Added -C and -D options to the exinext utility, mainly to make it easier
2415 to include in the automated testing, but these could be helpful when
2416 multiple configurations are in use.
2417
2418 2. The exinext utility was not formatting the output nicely when there was
2419 an alternate port involved in the retry record key, nor when there was a
2420 message id as well (for retries that were specific to a specific message
2421 and a specific host). It was also confused by IPv6 addresses, because of
2422 the additional colons they contain. I have fixed the IPv4 problem, and
2423 patched it up to do a reasonable job for IPv6.
2424
2425 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
2426 delivery, the log line now contains "pipelined" if PIPELINING was used.
2427
2428 4. An SMTP transport process used to panic and die if the bind() call to set
2429 an explicit outgoing interface failed. This has been changed; it is now
2430 treated in the same way as a connect() failure.
2431
2432 5. A reference to $sender_host_name in the part of a conditional expansion
2433 that was being skipped was still causing a DNS lookup. This no longer
2434 occurs.
2435
2436 6. The def: expansion condition was not recognizing references to header lines
2437 that used bh_ and bheader_.
2438
2439 7. Added the _cache feature to named lists.
2440
2441 8. The code for checking quota_filecount in the appendfile transport was
2442 allowing one more file than it should have been.
2443
2444 9. For compatibility with Sendmail, the command line option
2445
2446 -prval:sval
2447
2448 is equivalent to
2449
2450 -oMr rval -oMs sval
2451
2452 and sets the incoming protocol and host name (for trusted callers). The
2453 host name and its colon can be omitted when only the protocol is to be set.
2454 Note the Exim already has two private options, -pd and -ps, that refer to
2455 embedded Perl. It is therefore impossible to set a protocol value of "d" or
2456 "s", but I don't think that's a major issue.
2457
245810. A number of refactoring changes to the code, none of which should affect
2459 Exim's behaviour:
2460
2461 (a) The number of logging options was getting close to filling up the
2462 32-bit word that was used as a bit map. I have split them into two classes:
2463 those that are passed in the argument to log_write(), and those that are
2464 only ever tested independently outside of that function. These are now in
2465 separate 32-bit words, so there is plenty of room for expansion again.
2466 There is no change in the user interface or the logging behaviour.
2467
2468 (b) When building, for example, log lines, the code previously used a
2469 macro that called string_cat() twice, in order to add two strings. This is
2470 not really sufficiently general. Furthermore, there was one instance where
2471 it was actually wrong because one of the argument was used twice, and in
2472 one call a function was used. (As it happened, calling the function twice
2473 did not affect the overall behaviour.) The macro has been replaced by a
2474 function that can join an arbitrary number of extra strings onto a growing
2475 string.
2476
2477 (c) The code for expansion conditions now uses a table and a binary chop
2478 instead of a serial search (which was left over from when there were very
2479 few conditions). Also, it now recognizes conditions like "pam" even when
2480 the relevant support is not compiled in: a suitably worded error message is
2481 given if an attempt is made to use such a condition.
2482
248311. Added ${time_interval:xxxxx}.
2484
248512. A bug was causing one of the ddress fields not to be passed back correctly
2486 from remote delivery subprocesses. The field in question was not being
2487 subsequently used, so this caused to problems in practice.
2488
248913. Added new log selectors queue_time and deliver_time.
2490
249114. Might have fixed a bug in maildirsizefile handling that threw up
2492 "unexpected character" debug warnings, and recalculated the data
2493 unnecessarily. In any case, I expanded the warning message to give more
2494 information.
2495
249615. Added the message "Restricted characters in address" to the statements in
2497 the default ACL that block characters like @ and % in local parts.
2498
249916. Change 71 for release 4.31 proved to be much less benign that I imagined.
2500 Three changes have been made:
2501
2502 (a) There was a serious bug; a negative response to MAIL caused the whole
2503 recipient domain to be cached as invalid, thereby blocking all messages
2504 to all local parts at the same domain, from all senders. This bug has
2505 been fixed. The domain is no longer cached after a negative response to
2506 MAIL if the sender used is not empty.
2507
2508 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
2509 been restored.
2510
2511 (c) A new callout option, "use_sender" has been added for people who want
2512 the modified behaviour.
2513
2514
2515Exim version 4.31
2516-----------------
2517
2518 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
2519 Larry Rosenman.
2520
2521 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
2522 indeed breaks things for older releases.
2523
2524 3. Added additional logging to the case where there is a problem reading data
2525 from a filter that is running in a subprocess using a pipe, in order to
2526 try to track down a specific problem.
2527
2528 4. Testing facility fudge: when running in the test harness and attempting
2529 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
2530 getting "No route to host". Convert this to a timeout.
2531
2532 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
2533 warning.
2534
2535 6. Some OS don't have socklen_t but use size_t instead. This affects the
2536 fifth argument of getsockopt() amongst other things. This is now
2537 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
2538 can be set for individual OS. I have set it for SunOS5, OSF1, and
2539 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
2540 some earlier ones do not.
2541
2542 7. Change 4.30/15 was not doing the test caselessly.
2543
2544 8. The standard form for an IPv6 address literal was being rejected by address
2545 parsing in, for example, MAIL and RCPT commands. An example of this kind of
2546 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
2547 this, as well as the form without the "IPv6" on the front (but only when
2548 address literals are enabled, of course).
2549
2550 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
2551
255210. Exim crashed if a message with an empty sender address specified by -f
2553 encountered a router with an errors_to setting. This could be provoked only
2554 by a command such as
2555
2556 exim -f "" ...
2557
2558 where an empty string was supplied; "<>" did not hit this bug.
2559
256011. Installed PCRE release 4.5.
2561
256212. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
2563 remained set. It is now erased.
2564
256513. exiqgrep wasn't working on MacOS X because it didn't correctly compute
2566 times from message ids (which are base 36 rather than the normal 62).
2567
256814. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
2569 were being counted as actual protocol errors, and logged if the log
2570 selector +smtp_protocol_error was set. One cannot be perfect in this test,
2571 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
2572 and DATA following a set of rejected RCPTs do not count as protocol errors.
2573 In other words, Exim assumes they were pipelined, though this may not
2574 actually be the case. Of course, in all cases the client gets an
2575 appropriate error code.
2576
257715. If a lookup fails in an ACL condition, a message about the failure may
2578 be available; it is used if testing the ACL cannot continue, because most
2579 such messages specify what the cause of the deferral is. However, some
2580 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
2581 that caused an old message to be retained and used if a later statement
2582 caused a defer, replacing the real cause of the deferral.
2583
258416. If an IP address had so many PTR records that the DNS lookup buffer
2585 was not large enough to hold them, Exim could crash while trying to process
2586 the truncated data. It now detects and logs this case.
2587
258817. Further to 4.21/58, another change has been made: if (and only if) the
2589 first line of a message (the first header line) ends with CRLF, a bare LF
2590 in a subsequent header line has a space inserted after it, so as not to
2591 terminate the header.
2592
259318. Refactoring: tidied an ugly bit of code in appendfile that copied data
2594 unnecessarily, used atoi() instead of strtol(), and didn't check the
2595 termination when getting file sizes from file names by regex.
2596
259719. Completely re-implemented the support for maildirsize files, in the light
2598 of a number of problems with the previous contributed implementation
2599 (4.30/29). In particular:
2600
2601 . If the quota is zero, the maildirsize file is maintained, but no quota is
2602 imposed.
2603
2604 . If the maildir directory does not exist, it is created before any attempt
2605 to write a maildirsize file.
2606
2607 . The quota value in the file is just a cache; if the quota is changed in
2608 the transport, the new value overrides.
2609
2610 . A regular expression is available for excluding directories from the
2611 count.
2612
261320. The autoreply transport checks the characters in options that define the
2614 message's headers; it allows continued headers, but it was checking with
2615 isspace() after an embedded newline instead of explicitly looking for a
2616 space or a tab.
2617
261821. If all the "regular" hosts to which an address was routed had passed their
2619 expiry times, and had not reached their retry times, the address was
2620 bounced, even if fallback hosts were defined. Now Exim should go on to try
2621 the fallback hosts.
2622
262322. Increased buffer sizes in the callout code from 1024 to 4096 to match the
2624 equivalent code in the SMTP transport. Some hosts send humungous responses
2625 to HELO/EHLO, more than 1024 it seems.
2626
262723. Refactoring: code in filter.c used (void *) for "any old type" but this
2628 gives compiler warnings in some environments. I've now done it "properly",
2629 using a union.
2630
263124. The replacement for inet_ntoa() that is used with gcc on IRIX systems
2632 (because of problems with the built-in one) was declared to return uschar *
2633 instead of char *, causing compiler failure.
2634
263525. Fixed a file descriptor leak when processing alias/forward files.
2636
263726. Fixed a minor format string issue in dbfn.c.
2638
263927. Typo in exim.c: ("dmbnz" for "dbmnz").
2640
264128. If a filter file refered to $h_xxx or $message_headers, and the headers
2642 contained RFC 2047 "words", Exim's memory could, under certain conditions,
2643 become corrupted.
2644
264529. When a sender address is verified, it is cached, to save repeating the test
2646 when there is more than one recipient in a message. However, when the
2647 verification involves a callout, it is possible for different callout
2648 options to be set for different recipients. It is too complicated to keep
2649 track of this in the cache, so now Exim always runs a verification when a
2650 callout is required, relying on the callout cache for the optimization.
2651 The overhead is duplication of the address routing, but this should not be
2652 too great.
2653
265430. Fixed a bug in callout caching. If a RCPT command caused the sender address
2655 to be verified with callout=postmaster, and the main callout worked but the
2656 postmaster check failed, the verification correctly failed. However, if a
2657 subsequent RCPT command asked for sender verification *without* the
2658 postmaster check, incorrect caching caused this verification also to fail,
2659 incorrectly.
2660
266131. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
2662 it was not caching the DNS options (qualify_single, search_parents) that
2663 were used when the lookup failed. A subsequent lookup with different
2664 options therefore always gave the same answer, though there were cases
2665 where it should not have. (Example: a "domains = !$mx_any" option on a
2666 dnslookup router: the "domains" option is always processed without any
2667 widening, but the router might have qualify_single set.) Now Exim uses the
2668 cached value only when the same options are set.
2669
267032. Added John Jetmore's "exipick" utility to the distribution.
2671
267233. GnuTLS: When an attempt to start a TLS session fails for any reason other
2673 than a timeout (e.g. a certificate is required, and is not provided), an
2674 Exim server now closes the connection immediately. Previously it waited for
2675 the client to close - but if the client is SSL, it seems that they each
2676 wait for each other, leading to a delay before one of them times out.
2677
267834: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
2679 maintained 0.8.x compatibility because I don't think many are using it, and
2680 it is clearly obsolete.
2681
268235. Added TLS support for CRLs: a tls_crl global option and one for the smtp
2683 transport.
2684
268536. OpenSSL: $tls_certificate_verified was being set to 1 even if the
2686 client certificate was expired. A simple patch fixes this, though I don't
2687 understand the full logic of why the verify callback is called multiple
2688 times.
2689
269037. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
2691 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
2692 which causes problems with some clients (such as the Certicom SSL Plus
2693 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
2694 disables the coutermeasure allowing Eudora to connect."
2695
269638. Exim was not checking that a write() to a log file succeeded. This could
2697 lead to Bad Things if a log got too big, in particular if it hit a file
2698 size limit. Exim now panics and dies if it cannot write to a log file, just
2699 as it does if it cannot open a log file.
2700
270139. Modified OS/Makefile-Linux so that it now contains
2702
2703 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
2704
2705 The two -D definitions ensure that Exim is compiled with large file
2706 support, which makes it possible to handle log files that are bigger than
2707 2^31.
2708
270940. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
2710 instance) a domain was checked against a named list that involved a lookup,
2711 causing $domain_data to be set, then another domain was checked against the
2712 same list, then the first domain was re-checked, the value of $domain_data
2713 after the final check could be wrong. In particular, if the second check
2714 failed, it could be set empty. This bug probably also applied to
2715 $localpart_data.
2716
271741. The strip_trailing_dot option was not being applied to the address given
2718 with the -f command-line option.
2719
272042. The code for reading a message's header from the spool was incrementing
2721 $received_count, but never initializing it. This meant that the value was
2722 incorrect (doubled) while delivering a message in the same process in which
2723 it was received. In the most common configuration of Exim, this never
2724 happens - a fresh exec is done - but it can happen when
2725 deliver_drop_privilege is set.
2726
272743. When Exim logs an SMTP synchronization error - client data sent too soon -
2728 it now includes up to 150 characters of the unexpected data in the log
2729 line.
2730
273144. The exim_dbmbuild utility uses fixed size buffers for reading input lines
2732 and building data strings. The size of both of these buffers was 10 000
2733 bytes - far larger than anybody would *ever* want, thought I. Needless to
2734 say, somebody hit the limit. I have increased the maximum line length to
2735 20 000 and the maximum data length of concatenated lines to 100 000. I have
2736 also fixed two bugs, because there was no checking on these buffers. Tsk,
2737 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
2738 buffer is too small.
2739
274045. The exim_dbmbuild utility did not support quoted keys, as Exim does in
2741 lsearch lookups. Now it does.
2742
274346. When parsing a route_list item in a manualroute router, a fixed-length
2744 buffer was used for the list of hosts. I made this 1024 bytes long,
2745 thinking that nobody would ever have a list of hosts that long. Wrong.
2746 Somebody had a whole pile of complicated expansion conditions, and the
2747 string was silently truncated, leading to an expansion error. It turns out
2748 that it is easier to change to an unlimited length (owing to other changes
2749 that have happened since this code was originally written) than to build
2750 structure for giving a limitation error. The length of the item that
2751 expands into the list of hosts is now unlimited.
2752
275347. The lsearch lookup could not handle data where the length of text line was
2754 more than 4095 characters. Such lines were truncated, leading to shortened
2755 data being returned. It should now handle lines of any length.
2756
275748. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
2758 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
2759 ACL").
2760
276149. Cosmetic tidy to scripts like exicyclog that are generated by globally
2762 replacing strings such as BIN_DIRECTORY in a source file: the replacement
2763 no longer happens in comment lines. A list of replacements is now placed
2764 at the head of all of the source files, except those whose only change is
2765 to replace PERL_COMMAND in the very first #! line.
2766
276750. Replaced the slow insertion sort in queue.c, for sorting the list of
2768 messages on the queue, with a bottom-up merge sort, using code contributed
2769 by Michael Haardt. This should make operations like -bp somewhat faster on
2770 large queues. It won't affect queue runners, except when queue_run_in_order
2771 is set.
2772
277351. Installed eximstats 1.31 in the distribution.
2774
277552. Added support for SRV lookups to the dnslookup router.
2776
277753. If an ACL referred to $message_body or $message_body_end, the value was not
2778 reset for any messages that followed in the same SMTP session.
2779
278054. The store-handling optimization for building very long strings was not
2781 differentiating between the different store pools. I don't think this
2782 actually made any difference in practice, but I've tidied it.
2783
278455. While running the routers to verify a sender address, $sender_address
2785 was still set to the sender address. This is wrong, because when routing to
2786 send a bounce to the sender, it would be empty. Therefore, I have changed
2787 it so that, while verifying a sender address, $sender_address is set to <>.
2788 (There is no change to what happens when verifying a recipient address.)
2789
279056. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
2791 target A or AAAA records (if not already returned) without resetting the
2792 qualify_single or search_parents options of the DNS resolver. These are
2793 inappropriate in this case because the targets of MX and SRV records must
2794 be FQDNs. A broken DNS record could cause trouble if it happened to have a
2795 target that, when qualified, matched something in the local domain. These
2796 two options are now turned off when doing these lookups.
2797
279857. It seems that at least some releases of Reiserfs (which does not have the
2799 concept of a fixed number of inodes) returns zero and not -1 for the
2800 number of available inodes. This interacted badly with check_spool_inodes,
2801 which assumed that -1 was the "no such thing" setting. What I have done is
2802 to check that the total number of inodes is greater than zero before doing
2803 the test of how many are available.
2804
280558. When a "warn" ACL statement has a log_message modifier, the message is
2806 remembered, and not repeated. This is to avoid a lot of repetition when a
2807 message has many recipients that cause the same warning to be written.
2808 Howewer, Exim was preserving the list of already written lines for an
2809 entire SMTP session, which doesn't seem right. The memory is now reset if a
2810 new message is started.
2811
281259. The "rewrite" debugging flag was not showing the result of rewriting in the
2813 debugging output unless log_rewrite was also set.
2814
281560. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
2816 of (int)(handle) when we know that handle contains (void *)(-1).
2817
281861. The Exim daemon panic-logs an error return when it closes the incoming
2819 connection. However "connection reset by peer" seems to be common, and
2820 isn't really an error worthy of noting specially, so that particular error
2821 is no long logged.
2822
282362. When Exim is trying to find all the local interfaces, it used to panic and
2824 die if the ioctl to get the interface flags failed. However, it seems that
2825 on at least one OS (Solaris 9) it is possible to have an interface that is
2826 included in the list of interfaces, but for which you get a failure error
2827 for this call. This happens when the interface is not "plumbed" into a
2828 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
2829 failure of the "get flags" call assumes that the interface is down.
2830
283163. Added a ${eval10: operator, which assumes all numbers are decimal. This
2832 makes life easier for people who are doing arithmetic on fields extracted
2833 from dates, where you often get leading zeros that should not be
2834 interpreted as octal.
2835
283664. Added qualify_domain to the redirect router, to override the global
2837 setting.
2838
283965. If a pathologically long header line contained very many addresses (the
2840 report of this problem mentioned 10 000) and each of them was rewritten,
2841 Exim could use up a very large amount of memory. (It kept on making new
2842 copies of the header line as it rewrote, and never released the old ones.)
2843 At the expense of a bit more processing, the header rewriting function has
2844 been changed so that it no longer eats memory in this way.
2845
284666. The generation of the Received: header has been moved from the time that a
2847 message starts to be received, to the time that it finishes. The timestamp
2848 in the Received: header should now be very close to that of the <= log
2849 line. There are two side-effects of this change:
2850
2851 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
2852 logged header lines no longer include the local Received: line, because
2853 it has not yet been created. The same applies to a copy of the message
2854 that is returned to a non-SMTP sender when a message is rejected.
2855
2856 (b) When a filter file is tested using -bf, no additional Received: header
2857 is added to the test message. After some thought, I decided that this
2858 is a bug fix.
2859
2860 This change does not affect the value of $received_for. It is still set
2861 after address rewriting, but before local_scan() is called.
2862
286367. Installed the latest Cygwin-specific files from the Cygwin maintainer.
2864
286568. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
2866 gave an unhelpful panic error message, and a defer error. I have managed to
2867 change this behaviour so that it now rejects any supplied certificate,
2868 which seems right, as the list of acceptable certificates is empty.
2869
287069. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
2871 gave an unhelpful defer error. I have not managed to make this reject any
2872 supplied certificates, but the error message it gives is "no certificate
2873 supplied", which is not helpful.
2874
287570. exigrep's output now also includes lines that are not associated with any
2876 message, but which match the given pattern. Implemented by a patch from
2877 Martin Sluka, which also tidied up the Perl a bit.
2878
287971. Recipient callout verification, like sender verification, was using <> in
2880 the MAIL FROM command. This isn't really the right thing, since the actual
2881 sender may affect whether the remote host accepts the recipient or not. I
2882 have changed it to use the actual sender in the callout; this means that
2883 the cache record is now keyed on a recipient/sender pair, not just the
2884 recipient address. There doesn't seem to be a real danger of callout loops,
2885 since a callout by the remote host to check the sender would use <>.
2886 [SEE ABOVE: changed after hitting problems.]
2887
288872. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
2889 temporary errors. However, in the case of such a code being given after
2890 the end of a data transmission (i.e. after ".") Exim was failing to write
2891 a retry record for the message. (Yes, there was some broken host that was
2892 actually sending 8xx at this point.)
2893
289473. An unknown lookup type in a host list could cause Exim to panic-die when
2895 the list was checked. (An example that provoked this was putting <; in the
2896 middle of a list instead of at the start.) If this happened during a DATA
2897 ACL check, a -D file could be left lying around. This kind of configuration
2898 error no longer causes Exim to die; instead it causes a defer errror. The
2899 incident is still logged to the main and panic logs.
2900
290174. Buglet left over from Exim 3 conversion. The message "too many messages
2902 in one connection" was written to the rejectlog but not the mainlog, except
2903 when address rewriting (yes!) was being logged.
2904
290575. Added write_rejectlog option.
2906
290776. When a system filter was run not as root (that is, when system_filter_user
2908 was set), the values of the $n variables were not being returned to the
2909 main process; thus, they were not subsequently available in the $sn
2910 variables.
2911
291277. Added +return_path_on_delivery log selector.
2913
291478. A connection timeout was being treated differently from recipients deferred
2915 when testing hosts_max_try with a message that was older than the host's
2916 retry timeout. (The host should not be counted, thus allowing all hosts to
2917 be tried at least once before bouncing.) This may have been the cause of an
2918 occasionally reported bug whereby a message would remain on the queue
2919 longer than the retry timeout, but would be bounced if a delivery was
2920 forced. I say "may" because I never totally pinned down the problem;
2921 setting up timeout/retry tests is difficult. See also the next item.
2922
292379. The ultimate address timeout was not being applied to errors that involved
2924 a combination of host plus message (for example, a timeout on a MAIL
2925 command). When an address resolved to a number of possible hosts, and they
2926 were not all tried for each delivery (e.g. because of hosts_max_try), a
2927 message could remain on the queue longer than the retry timeout.
2928
292980. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
2930 Haardt.
2931
293281. Fixed an obscure SMTP outgoing bug which required at least the following
2933 conditions: (a) there was another message waiting for the same server;
2934 (b) the server returned 5xx to all RCPT commands in the first message so
2935 that the message was not completed; (c) the server dropped the connection
2936 or gave a negative response to the RSET that Exim sends to abort the
2937 transaction. The observed case was a dropped connection after DATA that had
2938 been sent in pipelining mode. That is, the server had advertised PIPELINING
2939 but was not implementing it correctly. The effect of the bug was incorrect
2940 behaviour, such as trying another host, and this could lead to a crash.
2941
2942
2943Exim version 4.30
2944-----------------
2945
2946 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
2947 and daemon.c were passed as pointers to ints; they should have been
2948 pointers to socklen_t variables (which are typically unsigned ints).
2949
2950 2. Some signed/unsigned type warnings in the os.c file for Linux have been
2951 fixed.
2952
2953 3. Fixed a really odd bug that affected only the testing scheme; patching a
2954 certain fixed string in the binary changed the value of another string that
2955 happened to be identical to the end of the original first string.
2956
2957 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
2958 name", it returns that address as the IP address. On some operating
2959 systems (e.g. Solaris), it also passes back the IP address string as the
2960 "host name". However, on others (e.g. Linux), it passes back an empty
2961 string. Exim wasn't checking for this, and was changing the host name to an
2962 empty string, assuming it had been canonicized.
2963
2964 5. Although rare, it is permitted to have more than one PTR record for a given
2965 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
2966 all the names associated with an address, because they do in Solaris.
2967 However, it seems that they do not in Linux for data that comes from the
2968 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
2969 I found this out when I moved to a new Linux workstation and tried to run
2970 the Exim test suite.
2971
2972 To get round this problem I have changed the code so that it now does its
2973 own call to the DNS to look up PTR records when searching for a host name.
2974 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
2975 addresses that are only in /etc/hosts are still found.
2976
2977 This behaviour is, however, controlled by an option called host_lookup_
2978 order, which defaults to "bydns:byaddr". If people want to use the other
2979 order, or indeed, just use one or the other means of lookup, they can
2980 specify it in this variable.
2981
2982 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
2983 some operating systems, this comes back from gethostbyaddr() as an empty
2984 string, and this is what Exim used to test for. However, it seems that in
2985 other systems, "." is yielded. Exim now tests for this case too.
2986
2987 7. The values of check_spool_space and check_log_space are now held internally
2988 as a number of kilobytes instead of an absolute number of bytes. If a
2989 numbers is specified without 'K' or 'M', it is rounded up to the nearest
2990 kilobyte. This means that much larger values can be stored.
2991
2992 8. Exim monitor: an attempt to get the action menu when not actually pointing
2993 at a message produces an empty menu entitled "No message selected". This
2994 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
2995 no entries in it ("Shell widget menu has zero width and/or height"). So I
2996 have added a single, blank menu entry in this case.
2997
2998 9. Added ${quote_local_part.
2999
300010. MIME decoding is now applied to the contents of Subject: header lines when
3001 they are logged.
3002
300311. Now that a reference to $sender_host_address automatically causes a reverse
3004 lookup to occur if necessary (4.13/18), there is no need to arrange for a
3005 host lookup before query-style lookups in lists that might use this
3006 variable. This has therefore been abolished, and the "net-" prefix is no
3007 longer necessary for query-style lookups.
3008
300912. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
3010 have been a typo for LFLAGS, so it has been changed.
3011
301213. The install script calls Exim with "-C /dev/null" in order to find the
3013 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
3014 to be output. Howeve, since Exim outputs its version number before the
3015 error, it didn't break the script. It just looked ugly. I fixed this by
3016 always allowing "-C /dev/null" if the caller is root.
3017
301814. Ignore overlarge ACL variable number when reading spool file - insurance
3019 against a later release with more variables having written the file.
3020
302115. The standard form for an IPv6 address literal was being rejected by EHLO.
3022 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
3023 this, as well as the form without the "IPv6" on the front.
3024
302516. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
3026 OS/Makefile-Darwin file.
3027
302817. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
3029 only to LDAP libraries that do not have LDAP_OPT_DEREF.
3030
303118. After change 4.21/52, "%ld" was used to format the contents of the $inode
3032 variable. However, some OS use ints for inodes. I've added cast to long int
3033 to get rid of the compiler warning.
3034
303519. I had forgotten to lock out "/../" in configuration file names when
3036 ALT_CONFIG_PREFIX was set.
3037
303820. Routers used for verification do not need to specify transports. However,
3039 if such a router generated a host list, and callout was configured, Exim
3040 crashed, because it could not find a port number from the (non-existent)
3041 transport. It now assumes port 25 in this circumstance.
3042
304321. Added the -t option to exigrep.
3044
304522. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
3046 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
3047 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
3048 them is set, LOOKUP_LSEARCH is forced.
3049
305023. "exim -bV" now outputs a list of lookups that are included in the binary.
3051
305224. Added sender and host information to the "rejected by local_scan()" log
3053 line; previously there was no indication of these.
3054
305525. Added .include_if_exists.
3056
305726. Change 3.952/11 added an explicit directory sync on top of a file sync for
3058 Linux. It turns out that not all file systems support this. Apparently some
3059 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
3060 do it.) To cope with this, the error EINVAL, which means that sync-ing is
3061 not supported on the file descriptor, is now ignored when Exim is trying to
3062 sync a directory. This applies only to Linux.
3063
306427. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
3065
306628. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
3067 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
3068 this.
3069
307029. Added support for maildirsize files from supplied patch (modified a bit).
3071
307230. The use of :fail: followed by an empty string could lead Exim to respond to
3073 sender verification failures with (e.g.):
3074
3075 550 Verification failed for <xxx>
3076 550 Sender verify failed
3077
3078 where the first response line was missing the '-' that indicates it is not
3079 the final line of the response.
3080
308131. The loop for finding the name of the user that called Exim had a hardwired
3082 limit of 10; it now uses the value of finduser_retries, which is used for
3083 all other user lookups.
3084
308532. Added $received_count variable, available in data and not_smtp ACLs, and at
3086 delivery time.
3087
308833. Exim was neglecting to zero errno before one call of strtol() when
3089 expanding a string and expecting an integer value. On some systems this
3090 resulted in spurious "integer overflow" errors. Also, it was casting the
3091 result into an int without checking.
3092
309334. Testing for a connection timeout using "timeout_connect" in the retry rules
3094 did not work. The code looks as if it has *never* worked, though it appears
3095 to have been documented since at least releast 1.62. I have made it work.
3096
309735. The "timeout_DNS" error in retry rules, also documented since at least
3098 1.62, also never worked. As it isn't clear exactly what this means, and
3099 clearly it isn't a major issue, I have abolished the feature by treating it
3100 as "timeout", and writing a warning to the main and panic logs.
3101
310236. The display of retry rules for -brt wasn't always showing the error code
3103 correctly.
3104
310537. Added new error conditions to retry rules: timeout_A, timeout_MX,
3106 timeout_connect_A, timeout_connect_MX.
3107
310838. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
3109 to the empty sender.
3110
311139. The daemon was not analysing the content of -oX till after it had closed
3112 stderr and disconnected from the controlling terminal. This meant that any
3113 syntax errors were only noted on the panic log, and the return code from
3114 the command was 0. By re-arranging the code a little, I've made the
3115 decoding happen first, so such errors now appear on stderr, and the return
3116 code is 1. However, the actual setting up of the sockets still happens in
3117 the disconnected process, so errors there are still only recorded on the
3118 panic log.
3119
312040. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
3121 connections (as happens on some IP stacks) was logged at start up time as
3122 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
3123 it from "IPv6 and IPv4", which means that two separate sockets are being
3124 used.
3125
312641. The debug output for gethostbyname2() or getipnodebyname() failures now
3127 says whether AF_INET or AF_INET6 was passed as an argument.
3128
312942. Exiwhat output was messed up when time zones were included in log
3130 timestamps.
3131
313243. Exiwhat now gives more information about the daemon's listening ports,
3133 and whether -tls-on-connect was used.
3134
313544. The "port" option of the smtp transport is now expanded.
3136
313745. A "message" modifier in a "warn" statement in a non-message ACL was being
3138 silently ignored. Now an error message is written to the main and panic
3139 logs.
3140
314146. There's a new ACL modifier called "logwrite" which writes to a log file
3142 as soon as it is encountered.
3143
314447. Added $local_user_uid and $local_user_gid at routing time.
3145
314648. Exim crashed when trying to verify a sender address that was being
3147 rewritten to "<>".
3148
314949. Exim was recognizing only a space character after ".include". It now also
3150 recognizes a tab character.
3151
315250. Fixed several bugs in the Perl script that creates the exim.8 man page by
3153 extracting the relevant information from the specification. The man page no
3154 longer contains scrambled data for the -d option, and I've added a section
3155 at the front about calling Exim under different names.
3156
315751. Added "extra_headers" argument to the "mail" command in filter files.
3158
315952. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
3160 crash.
3161
316253. Installed eximstats 1.29.
3163
316454. Added transport_filter_timeout as a generic transport option.
3165
316655. Exim no longer adds an empty Bcc: header to messages that have no To: or
3167 Cc: header lines. This was required by RFC 822, but it not required by RFC
3168 2822.
3169
317056. Exim used to add From:, Date:, and Message-Id: header lines to any
3171 incoming messages that did not have them. Now it does so only if the
3172 message originates locally, that is, if there is no associated remote host
3173 address. When Resent- header lines are present, this applies to the Resent-
3174 lines rather than the non-Resent- lines.
3175
317657. Drop incoming SMTP connection after too many syntax or protocol errors. The
3177 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
3178
317958. Messages for configuration errors now include the name of the main
3180 configuration file - useful now that there may be more than one file in a
3181 list (.included file names were always shown).
3182
318359. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
3184 for those rare installations that do not start the daemon as root or run it
3185 setuid root. I've cut out the call to initgroups() if the daemon is not
3186 root at that time.
3187
318860. The Exim user and group can now be bound into the binary as text strings
3189 that are looked up at the start of Exim's processing.
3190
319161. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
3192
319362. Added $mailstore_basename variable.
3194
319563. Installed patch to sieve.c from Michael Haardt.
3196
319764. When Exim failed to open the panic log after failing to open the main log,
3198 the original message it was trying to log was written to stderr and debug
3199 output, but if they were not available (the usual case in production), it
3200 was lost. Now it is written to syslog before the two lines that record the
3201 failures to open the logs.
3202
320365. Users' Exim filters run in subprocesses under the user's uid. It is
3204 possible for a "deliver" command or an alias in a "personal" command to
3205 provoke an address rewrite. If logging of address rewriting is configured,
3206 this fails because the process is not running as root or exim. There may be
3207 a better way of dealing with this, but for the moment (because 4.30 needs
3208 to be released), I have disabled address rewrite logging when running a
3209 filter in a non-root, non-exim process.
3210
3211
3212Exim version 4.24
3213-----------------
3214
3215 1. The buildconfig auxiliary program wasn't quoting the value set for
3216 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
3217 not defined. This bug was masked in 4.22 by the effect that was fixed in
3218 change 4.23/1.
3219
3220 2. Some messages that were rejected after a message id was allocated were
3221 shown as "incomplete" by exigrep. It no longer does this for messages that
3222 are rejected by local_scan() or the DATA or non-SMTP ACLs.
3223
3224 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
3225 have allow_domain_literals set, the ID did not get logged in the <= line.
3226 Domain literals are now always recognized in Message-ID: header lines.
3227
3228 4. The first argument for a ${extract expansion item is the key name or field
3229 number. Leading and trailing spaces in this item were not being ignored,
3230 causing some misleading effects.
3231
3232 5. When deliver_drop_privilege was set, single queue runner processes started
3233 manually (i.e. by the command "exim -q") or by the daemon (which uses the
3234 same command in the process it spins off) were not dropping privilege.
3235
3236 6. When the daemon running as "exim" started a queue runner, it always
3237 re-executed Exim in the spun-off process. This is a waste of effort when
3238 deliver_drop_privilege is set. The new process now just calls the
3239 queue-runner function directly.
3240
3241
3242Exim version 4.23
3243-----------------
3244
3245 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
3246 HEADERS_CHARSET.
3247
3248 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
3249 ignored. Though the use of -oP was forcing the writing of a pid file, it
3250 was always written to the default place.
3251
3252 3. If the message "no IP address found for host xxxx" is generated during
3253 incoming verification, it is now followed by identification of the incoming
3254 connection (so you can more easily find what provoked it).
3255
3256 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
3257
3258 5. Added some features to "harden" Exim a bit more against certain attacks:
3259
3260 (a) There is now a build-time option called FIXED_NEVER_USERS that can
3261 be put in Local/Makefile. This is like the never_users runtime option,
3262 but it cannot be overridden. The default setting is "root".
3263
3264 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
3265 prefix string with which any file named in a -C command line option
3266 must start.
3267
3268 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
3269 is retained for -C and -D only if the caller of Exim is root. Without
3270 it, the exim user may also use -C and -D and retain privilege.
3271
3272 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
3273 command line option is disabled.
3274
3275 6. Macro names set by the -D option must start with an upper case letter, just
3276 like macro names defined in the configuration file.
3277
3278 7. Added "dereference=" facility to LDAP.
3279
3280 8. Two instances of the typo "uknown" in the source files are fixed.
3281
3282 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
3283 the Configure-Makefile script screwed up while processing it.
3284
328510. Incorporated PCRE 4.4.
3286
328711. The SMTP synchronization check was not operating right at the start of an
3288 SMTP session. For example, it could not catch a HELO sent before the client
3289 waited for the greeting. There is now a check for outstanding input at the
3290 point when the greeting is written. Because of the duplex, asynchronous
3291 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
3292 way, but not yet received, when the check is performed.
3293
329412. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
3295 on TCP/IP sockets, because this apparently causes some broken clients to
3296 timeout.
3297
329813. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
3299 unchanged) from the Cygwin maintainer.
3300
330114. The code for -bV that shows what is in the binary showed "mbx" when maildir
3302 was supported instead of testing for mbx. Effectively a typo.
3303
330415. The spa authenticator server code was not checking that the input it
3305 received was valid base64.
3306
330716. The debug output line for the "set" modifier in ACLs was not showing the
3308 name of the variable that was being set.
3309
331017. Code tidy: the variable type "vtype_string" was never used. Removed it.
3311
331218. Previously, a reference to $sender_host_name did not cause a DNS reverse
3313 lookup on its own. Something else was needed to trigger the lookup. For
3314 example, a match in host_lookup or the need for a host name in a host list.
3315 Now, if $sender_host_name is referenced and the host name has not yet been
3316 looked up, a lookup is performed. If the lookup fails, the variable remains
3317 empty, and $host_lookup_failed is set to "1".
3318
331919. Added "eqi" as a case-independent comparison operator.
3320
332120. The saslauthd authentication condition could segfault if neither service
3322 nor realm was specified.
3323
332421. If an overflowing value such as "2048M" was set for message_size_limit, the
3325 error message that was logged was misleading, and incoming SMTP
3326 connections were dropped. The message is now more accurate, and temporary
3327 errors are given to SMTP connections.
3328
332922. In some error situations (such as 21 above) Exim rejects all SMTP commands
3330 (except RSET) with a 421 error, until QUIT is received. However, it was
3331 failing to send a response to QUIT.
3332
333323. The HELO ACL was being run before the code for helo_try_verify_hosts,
3334 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
3335 ACL is now run after the helo_try_verify_hosts code.
3336
333724. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
3338 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
3339 case-independent, so other case variants are also recognized). Apparently
3340 some systems use these upper case variants.
3341
334225. If more than two messages were waiting for the same host, and a transport
3343 filter was specified for the transport, Exim sent two messages over the
3344 same TCP/IP connection, and then failed with "socket operation on non-
3345 socket" when it tried to send the third.
3346
334726. Added Exim::debug_write and Exim::log_write for embedded Perl use.
3348
334927. The extern definition of crypt16() in expand.c was not being excluded when
3350 the OS had its own crypt16() function.
3351
335228. Added bounce_return_body as a new option, and bounce_return_size_limit
3353 as a preferred synonym for return_size_limit, both as an option and as an
3354 expansion variable.
3355
335629. Added LIBS=-liconv to OS/Makefile-OSF1.
3357
335830. Changed the default configuration ACL to relax the local part checking rule
3359 for addresses that are not in any local domains. For these addresses,
3360 slashes and pipe symbols are allowed within local parts, but the sequence
3361 /../ is explicitly forbidden.
3362
336331. SPA server authentication was not clearing the challenge buffer before
3364 using it.
3365
336632. log_message in a "warn" ACL statement was writing to the reject log as
3367 well as to the main log, which contradicts the documentation and doesn't
3368 seem right (because no rejection is happening). So I have stopped it.
3369
337033. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
3371 However, I am unable to do any testing of this.
3372
337334. Fixed an infelicity in the appendfile transport. When checking directories
3374 for a mailbox, to see if any needed to be created, it was accidentally
3375 using path names with one or more superfluous leading slashes; tracing
3376 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
3377
337835. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
3379 contents are added to the log line that is written for every discarded
3380 recipient. (Previously a log_message setting was ignored.)
3381
338236. The ${quote: operator now quotes the string if it is empty.
3383
338437. The install script runs exim in order to find its version number. If for
3385 some reason other than non-existence or emptiness, which it checks, it
3386 could not run './exim', it was installing it with an empty version number,
3387 i.e. as "exim-". This error state is now caught, and the installation is
3388 aborted.
3389
339038. An argument was missing from the function that creates an error message
3391 when Exim fails to connect to the socket for saslauthd authentication.
3392 This could cause Exim to crash, or give a corrupted message.
3393
339439. Added isip, isip4, and isip6 to ${if conditions.
3395
339640. The ACL variables $acl_xx are now saved with the message, and can be
3397 accessed later in routers, transports, and filters.
3398
339941. The new lookup type nwildlsearch is like wildlsearch, except that the key
3400 strings in the file are not string-expanded.
3401
340242. If a MAIL command specified a SIZE value that was too large to fit into an
3403 int variable, the check against message_size_limit failed. Such values are
3404 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
3405 day this will have to be increased, but I don't think I want to be around
3406 when emails are that large.
3407
3408
3409
3410Exim version 4.22
3411-----------------
3412
3413 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
3414 iconv() is not standard in FreeBSD.
3415
3416 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
3417 IPv6 enabled. The observed symptom was a segmentation fault on return from
3418 the function os_common_find_running_interfaces() in src/os.c.
3419
3420 3. In the check_special_case() function in daemon.c I had used "errno" as an
3421 argument name, which causes warnings on some systems. This was basically a
3422 typo, since it was named "eno" in the comments!
3423
3424 4. The code that waits for the clock to tick (at a resolution of some fraction
3425 of a second) so as to ensure message-id uniqueness was always waiting for
3426 at least one whole tick, when it could have waited for less. [This is
3427 almost certainly not relevant at current processor speeds, where it is
3428 unlikely to ever wait at all. But we try to future-proof.]
3429
3430 5. The function that sleeps for a time interval that includes fractions of a
3431 second contained a race. It did not block SIGALRM between setting the
3432 timer, and suspending (a couple of lines later). If the interval was short
3433 and the sigsuspend() was delayed until after it had expired, the suspension
3434 never ended. On busy systems this could lead to processes getting stuck for
3435 ever.
3436
3437 6. Some uncommon configurations may cause a lookup to happen in a queue runner
3438 process, before it forks any delivery processes. The open lookup caching
3439 mechanism meant that the open file or database connection was passed into
3440 the delivery process. The problem was that delivery processes always tidy
3441 up cached lookup data. This could cause a problem for the next delivery
3442 process started by the queue runner, because the external queue runner
3443 process does not know about the closure. So the next delivery process
3444 still has data in the lookup cache. In the case of a file lookup, there was
3445 no problem because closing a file descriptor in a subprocess doesn't affect
3446 the parent. However, if the lookup was caching a connection to a database,
3447 the connection was closed, and the second delivery process was likely to
3448 see errors such as "PGSQL: query failed: server closed the connection
3449 unexpectedly". The problem has been fixed by closing all cached lookups
3450 in a queue runner before running a delivery process.
3451
3452 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
3453 seem to have the "const" qualifier which it has on other OS. I've
3454 parameterised it.
3455
3456 8. Change 4.21/2 was too strict. It is only if there are two authenticators
3457 *of the same type* (client or server) with the same public name that an
3458 error should be diagnosed.
3459
3460 9. When Exim looked up a host name for an IP address, but failed to find the
3461 original IP address when looking up the host name (a safety check), it
3462 output the message "<ip address> does not match any IP for NULL", which was
3463 confusing, to say the least. The bug was that the host name should have
3464 appeared instead of "NULL".
3465
346610. Since release 3.03, if Exim is called by a uid other than root or the Exim
3467 user that is built into the binary, and the -C or -D options is used, root
3468 privilege is dropped before the configuration file is read. In addition,
3469 logging is switched to stderr instead of the normal log files. If the
3470 configuration then re-defines the Exim user, the unprivileged environment
3471 is probably not what is expected, so Exim logs a panic warning message (but
3472 proceeds).
3473
3474 However, if deliver_drop_privilege is set, the unprivileged state may well
3475 be exactly what is intended, so the warning has been cut out in that case,
3476 and Exim is allowed to try to write to its normal log files.
3477
3478
3479Exim version 4.21
3480-----------------
3481
3482 1. smtp_return_error_details was not giving details for temporary sender
3483 or receiver verification errors.
3484
3485 2. Diagnose a configuration error if two authenticators have the same public
3486 name.
3487
3488 3. Exim used not to create the message log file for a message until the first
3489 delivery attempt. This could be confusing when incoming messages were held
3490 for policy or load reasons. The message log file is now created at the time
3491 the message is received, and an initial "Received" line is written to it.
3492
3493 4. The automatically generated man page for command line options had a minor
3494 bug that caused no ill effects; however, a more serious problem was that
3495 the procedure for building the man page automatically didn't always
3496 operate. Consequently, release 4.20 contains an out-of-date version. This
3497 shouldn't happen again.
3498
3499 5. When building Exim with embedded Perl support, the script that builds the
3500 Makefile was calling 'perl' to find its compile-time parameters, ignoring
3501 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
3502
3503 6. The freeze_tell option was not being used for messages that were frozen on
3504 arrival, either by an ACL or by local_scan().
3505
3506 7. Added the smtp_incomplete_transaction log selector.
3507
3508 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
3509 was accepting AUTH without a new EHLO.
3510
3511 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
3512 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
3513 RFC.
3514
351510. Logging of TCP/IP connections (when configured) now happens in the main
3516 daemon process instead of the child process, so that the TCP/IP connection
3517 count is more accurate (but it can never be perfect).
3518
351911. The use of "drop" in a nested ACL was not being handled correctly in the
3520 outer ACL. Now, if condition failure induced by the nested "drop" causes
3521 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
3522 or "require"), the connection is dropped.
3523
352412. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
3525 that yield "discard" can now be used with an "accept" or a "discard" verb,
3526 but an error is generated for any others (because I can't see a useful way
3527 to define what should happen).
3528
352913. When an ACL is read dynamically from a file (or anywhere else), the lines
3530 are now processed in the same way as lines in the Exim configuration file.
3531 In particular, continuation lines are supported.
3532
353314. Added the "dnslists = a.b.c!=n.n.n.n" feature.
3534
353515. Added -ti meaning -t -i.
3536
353716. Check for letters, digits, hyphens, and dots in the names of dnslist
3538 domains, and warn by logging if others are found.
3539
354017. At least on BSD, alignment is not guarenteed for the array of ifreq's
3541 returned from GIFCONF when Exim is trying to find the list of interfaces on
3542 a host. The code in os.c has been modified to copy each ifreq to an aligned
3543 structure in all cases.
3544
3545 Also, in some cases, the returned ifreq's were being copied to a 'struct
3546 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
3547 means the last couple of bytes of an IPv6 address could be chopped if the
3548 ifreq contained only a normal sockaddr (14 bytes storage).
3549
355018. Named domain lists were not supported in the hosts_treat_as_local option.
3551 An entry such as +xxxx was not recognized, and was treated as a literal
3552 domain name.
3553
355419. Ensure that header lines added by a DATA ACL are included in the reject log
3555 if the ACL subsequently rejects the message.
3556
355720. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
3558 MD5 (which is deprecated).
3559
356021. When testing a filter file using -bf, Exim was writing a message when it
3561 took the sender from a "From " line in the message, but it was not doing so
3562 when it took $return_path from a Return-Path: header line. It now does.
3563
356422. If the contents of a "message" modifier for a "warn" ACL verb do not begin
3565 with a valid header line field name (a series of printing characters
3566 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
3567
356823. Changed "disc" in the source to "disk" to conform to the documentation and
3569 the book and for uniformity.
3570
357124. Ignore Sendmail's -Ooption=value command line item.
3572
357325. When execve() failed while trying to run a command in a pipe transport,
3574 Exim was returning EX_UNAVAILBLE (69) from the subprocess. However, this
3575 could be confused with a return value of 69 from the command itself. This
3576 has been changed to 127, the value the shell returns if it is asked to run
3577 a non-existent command. The wording for the related log line suggests a
3578 non-existent command as the problem.
3579
358026. If received_header_text expands to an empty string, do not add a Received:
3581 header line to the message. (Well, it adds a token one on the spool, but
3582 marks it "old" so that it doesn't get used or transmitted.)
3583
358427. Installed eximstats 1.28 (addition of -nt option).
3585
358628. There was no check for failure on the call to getsockname() in the daemon
3587 code. This can fail if there is a shortage of resources on the system, with
3588 ENOMEM, for example. A temporary error is now given on failure.
3589
359029. Contrary to the C standard, it seems that in some environments, the
3591 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
3592 program. Exim now does this explicitly; it affects the formatting of
3593 timestamps using strftime().
3594
359530. If exiqsumm was given junk data, it threw up some uninitialized variable
3596 complaints. I've now initialized all the variables, to avoid this.
3597
359832. Header lines added by a system filter were not being "seen" during
3599 transport-time rewrites.
3600
360133. The info_callback() function passed to OpenSSL is set up with type void
3602 (*)(SSL *, int, int), as described somewhere. However, when calling the
3603 function (actually a macro) that sets it up, the type void(*)() is
3604 expected. I've put in a cast to prevent warnings from picky compilers.
3605
360634. If a DNS black list lookup found a CNAME record, but there were no A
3607 records associated with the domain it pointed at, Exim crashed.
3608
360935. If a DNS black list lookup returned more than one A record, Exim ignored
3610 all but the first. It now scans all returned addresses if a particular IP
3611 value is being sought. In this situation, the contents of the
3612 $dnslist_value variable are a list of all the addresses, separated by a
3613 comma and a space.
3614
361536. Tightened up the rules for host name lookups using reverse DNS. Exim used
3616 to accept a host name and all its aliases if the forward lookup for any of
3617 them yielded the IP address of the incoming connection. Now it accepts only
3618 those names whose forward lookup yields the correct IP address. Any other
3619 names are discarded. This closes a loophole whereby a rogue DNS
3620 administrator could create reverse DNS records to break through a
3621 wildcarded host restriction in an ACL.
3622
362337. If a user filter or a system filter that ran in a subprocess used any of
3624 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
3625 the wrong values were passed to the pipe command ($thisaddress had the
3626 value of $0, $0 had the value of $1, etc). This bug was introduced by
3627 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
3628
362938. Improved the line breaking for long SMTP error messages from ACLs.
3630 Previously, if there was no break point between 40 and 75 characters, Exim
3631 left the rest of the message alone. Two changes have been made: (a) I've
3632 reduced the minimum length to 35 characters; (b) if it can't find a break
3633 point between 35 and 75 characters, it looks ahead and uses the first one
3634 that it finds. This may give the occasional overlong line, but at least the
3635 remaining text gets split now.
3636
363739. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
3638 file descriptors might be low, and that setting 1000 would always raise it.
3639 It turns out that in some environments, the limit is already over 1000 and
3640 that lowering it causes trouble. So now Exim takes care not to decrease it.
3641
364240. When delivering a message, the value of $return_path is set to $sender_
3643 address at the start of routing (routers may change the value). By an
3644 oversight, this default was not being set up when an address was tested by
3645 -bt or -bv, which affected the outcome if any router or filter referred to
3646 $return_path.
3647
364841. The idea of the "warn" ACL verb is that it adds a header or writes to the
3649 log only when "message" or "log_message" are set. However, if one of the
3650 conditions was an address verification, or a call to a nested ACL, the
3651 messages generated by the underlying test were being passed through. This
3652 no longer happens. The underlying message is available in $acl_verify_
3653 message for both "message" and "log_message" expansions, so it can be
3654 passed through if needed.
3655
365642. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
3657 new expansion $bh_ to give the encoded byte string without charset
3658 translation. Translation happens only if iconv() is available; HAVE_ICONV
3659 indicates this at build time. HEADERS_CHARSET gives the charset to
3660 translate to; headers_charset can change it in the configuration, and
3661 "headers charset" can change it in an individual filter file.
3662
366343. Now that we have a default RFC 2047 charset (see above), the code in Exim
3664 that creates RFC 2047 encoded "words" labels them as that charset instead
3665 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
3666 expansion operator; (ii) when Exim creates a From: line for a local
3667 message; (iii) when a header line is rewritten to include a "phrase" part.
3668
366944. Nasty b