Wish List update.
[exim.git] / doc / doc-txt / ChangeLog
CommitLineData
715ab376 1$Cambridge: exim/doc/doc-txt/ChangeLog,v 1.335 2006/03/20 10:55:21 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
81e509d7
PH
306PH/60 When a VRFY deferred or FAILED, the log message rather than the user
307 message was being sent as an SMTP response.
308
3d240ff7
PH
309PH/61 Add -l and -k options to exicyclog.
310
b37c4101
PH
311PH/62 When verifying, if an address was redirected to one new address, so that
312 verification continued, and the new address failed or deferred after
313 having set something in $address_data, the value of $address_data was not
314 passed back to the ACL. This was different to the case when no
315 redirection occurred. The value is now passed back in both cases.
316
79378e0f
PH
317PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
318 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
319 use login_cap.h, so on its own it isn't the distinguishing feature. The
320 new name refers directly to the setclassresources() function.
321
e49c7bb4
PH
322PH/65 Added configuration files for NetBSD3.
323
d114ec46
PH
324PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
325
f3d7df6c
PH
326PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
327 is preferred over IPv4.
328
715ab376
PH
329PH/68 The bounce_return_message and bounce_return_body options were not being
330 honoured for bounces generated during the reception of non-SMTP messages.
331 In particular, this applied to messages rejected by the ACL. This bug has
332 been fixed. However, if bounce_return_message is true and bounce_return_
333 body is false, the headers that are returned for a non-SMTP message
334 include only those that have been read before the error was detected.
335 (In the case of an ACL rejection, they have all been read.)
336
214e2000 337
5de37277
PH
338Exim version 4.60
339-----------------
340
cc38ddbf
PH
341PH/01 Two changes to the default runtime configuration:
342
343 (1) Move the checks for relay_from_hosts and authenticated clients from
344 after to before the (commented out) DNS black list checks.
345
346 (2) Add control=submission to the relay_from_hosts and authenticated
347 clients checks, on the grounds that messages accepted by these
348 statements are most likely to be submissions.
5de37277 349
72fdd6ae
PH
350PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
351
352 (1) Generate an error if the third argument for the ${prvs expansion is
353 not a single digit.
354
355 (2) Treat a missing third argument of ${prvscheck as if it were an empty
356 string.
357
358 (3) Reset the variables that are obtained from the first argument of
359 ${prvscheck and used in the second argument before leaving the code,
360 because their memory is reclaimed, so using them afterwards may do
361 silly things.
362
363 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
364 one (it's much easier than Tom thought :-).
365
366 (5) Because of (4), we can now allow for the use of $prvscheck_result
367 inside the third argument.
cb9328de 368
cb741023
PH
369PH/03 For some reason, the default setting of PATH when running a command from
370 a pipe transport was just "/usr/bin". I have changed it to
371 "/bin:/usr/bin".
372
f174f16e
PH
373PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
374 anything to be listed in the output from -bV.
b2f5a032 375
c25242d7
PH
376PH/05 When a filter generated an autoreply, the entire To: header line was
377 quoted in the delivery log line, like this:
378
379 => >A.N.Other <ano@some.domain> <original@ddress> ...
380
381 This has been changed so that it extracts the operative address. There
382 may be more than one such address. If so, they are comma-separated, like
383 this:
384
385 => >ano@some.domain,ona@other.domain <original@ddress> ...
386
82c19f95
PH
387PH/06 When a client host used a correct literal IP address in a HELO or EHLO
388 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
389 not being looked up in the rDNS to get a host name, Exim was showing the
390 IP address twice in Received: lines, even though the IP addresses were
391 identical. For example:
392
393 Received: from [1.2.3.4] (helo=[1.2.3.4])
394
395 However, if the real host name was known, it was omitting the HELO data
396 if it matched the actual IP address. This has been tidied up so that it
397 doesn't show the same IP address twice.
398
d7ffbc12
PH
399PH/07 When both +timestamp and +memory debugging was on, the value given by
400 $tod_xxx expansions could be wrong, because the tod_stamp() function was
401 called by the debug printing, thereby overwriting the timestamp buffer.
402 Debugging no longer uses the tod_stamp() function when +timestamp is set.
403
9f526266
PH
404PH/08 When the original message was included in an autoreply transport, it
405 always said "this is a copy of the message, including all the headers",
406 even if body_only or headers_only was set. It now gives an appropriate
407 message.
408
87fcc8b9
PH
409PH/09 Applied a patch from the Sieve maintainer which:
410
411 o fixes some comments
412 o adds the (disabled) notify extension core
413 o adds some debug output for the result of if/elsif tests
414 o points to the current vacation draft in the documentation
415 and documents the missing references header update
416
417 and most important:
418
419 o fixes a bug in processing the envelope test (when testing
420 multiple envelope elements, the last element determinted the
421 result)
422
456682f5
PH
423PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
424 Electronic Mail") by including:
425
426 Auto-submitted: auto-generated
427
428 in the messages that it generates (bounce messages and others, such as
429 warnings). In the case of bounce messages for non-SMTP mesages, there was
430 also a typo: it was using "Auto_submitted" (underscore instead of
431 hyphen). Since every message generated by Exim is necessarily in response
432 to another message, thes have all been changed to:
433
434 Auto-Submitted: auto-replied
435
436 in accordance with these statements in the RFC:
437
438 The auto-replied keyword:
439
440 - SHOULD be used on messages sent in direct response to another
441 message by an automatic process,
442
443 - MUST NOT be used on manually-generated messages,
444
445 - MAY be used on Delivery Status Notifications (DSNs) and Message
446 Disposition Notifications (MDNs),
447
448 - MUST NOT be used on messages generated by automatic or periodic
449 processes, except for messages which are automatic responses to
450 other messages.
451
3e46c1aa
PH
452PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
453 to the default Received: header definition.
456682f5 454
49826d12
PH
455PH/12 Added log selector acl_warn_skipped (default on).
456
eba0c039
PH
457PH/13 After a successful wildlsearch lookup, discard the values of numeric
458 variables because (a) they are in the wrong storage pool and (b) even if
459 they were copied, it wouldn't work properly because of the caching.
460
a0d6ba8a
PH
461PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
462 checking when decoding. Apparently there are clients that generate
463 overlong encoded strings. Why am I not surprised?
464
f0917727
PH
465PH/15 If the first argument of "${if match_address" was not empty, but did not
466 contain an "@" character, Exim crashed. Now it writes a panic log message
467 and treats the condition as false.
468
096fee00
PH
469PH/16 In autoreply, treat an empty string for "once" the same as unset.
470
024bd3c2
PH
471PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
472 extension "envelope-auth". The code is finished and in agreement with
473 other implementations, but there is no documentation so far and in fact,
474 nobody wrote the draft yet. This extension is currently #undef'ed, thus
475 not changing the active code.
476
477 Print executed "if" and "elsif" statements when debugging is used. This
478 helps a great deal to understand what a filter does.
479
480 Document more things not specified clearly in RFC3028. I had all this
481 sorted out, when out of a sudden new issues came to my mind. Oops."
482
df199fec
PH
483PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
484 (Bugzilla #53).
485
d27f1df3
PH
486PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
487 canonical form (as documented). However, after a host name lookup from
488 the IP address, check_host() was doing a simple string comparison with
489 addresses acquired from the DNS when checking that the found name did
490 have the original IP as one of its addresses. Since any found IPv6
491 addresses are likely to be in abbreviated form, the comparison could
492 fail. Luckily, there already exists a function for doing the comparison
493 by converting both addresses to binary, so now that is used instead of
494 the text comparison.
495
96776534
PH
496PH/20 There was another similar case to PH/19, when a complete host name was
497 given in a host list; looking up its IP address could give an abbreviated
498 form, whereas the current host's name might or might not be abbreviated.
499 The same fix has been applied.
500
5de37277 501
9a799bc0
PH
502Exim version 4.54
503-----------------
504
505PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
506 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
507 It now does.
508
99a4b039
PH
509PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
510 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
511
512PH/03 Typo: missing ".o" in src/pcre/Makefile.
513
4b233853
PH
514PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
515 header line, restrict the check to what is listed in RFCs 2369 and 2929.
516 Also, for "Auto-Submitted", treat anything other than "no" as
517 non-personal, in accordance with RFC 3834. (Previously it treated
518 anything starting "auto-" as non-personal.)
519
8857ccfd
PH
520TF/01 The control=submission/name=... option had a problem with syntax
521 errors if the name included a slash character. The /name= option
522 now slurps the rest of the string, so it can include any characters
523 but it must come last in the list of options (after /sender_retain
524 or /domain=).
525
433a2980
PH
526PH/05 Some modifications to the interface to the fake nameserver for the new
527 testing suite.
528
3e46c1aa 529
9a799bc0 530
e3a311ba
TK
531Exim version 4.53
532-----------------
533
534TK/01 Added the "success_on_redirect" address verification option. See
535 NewStuff for rationale and an example.
536
13b685f9
PH
537PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
538
395ff96d
PH
539PH/02 Patch to exigrep to allow it to work on syslog lines.
540
5b68f6e4
PH
541PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
542 fread() to skip over the body file's header line, because in Cygwin the
543 header line is locked and is inaccessible.
544
1ab52c69
PH
545PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
546 co-exist for some time) to make it clear that it is the Exim ID that is
547 referenced, not the Message-ID: header line.
548
b07e6aa3
PH
549PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
550 string_format() function, because snprintf() does not exist on all
551 operating systems.
552
254e032f
PH
553PH/06 The use of forbid_filter_existstest now also locks out the use of the
554 ${stat: expansion item.
555
3af76a81
PH
556PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
557 protocol synchronization error", to keep the pedants happy.
558
2548ba04
PH
559PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
560 well as for IRIX systems, when gcc is being used. See the host.c source
561 file for comments.
562
b6c6011d
PH
563PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
564
cf39cf57
PH
565PH/10 Named domain lists were not working if used in a queue_smtp_domains
566 setting.
567
f1513293
PH
568PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
569 transport and to the smtp transport in LMTP mode.
570
727549a4
PH
571TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
572
af46795e
PH
573PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
574 run a filter in a subprocess. This could lead to confusion in subsequent
575 lookups in the parent process. There should also be a search_tidyup() at
576 the end of the subprocess.
577
d7b47fd0
PH
578PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
579 only if the host matched helo_try_verify_hosts, which caused the
580 verification to occur when the EHLO/HELO command was issued. The ACL just
581 tested the remembered result. Now, if a previous verification attempt has
582 not happened, "verify = helo" does it there and then.
583
ee744174
JJ
584JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
585
b582ab87
PH
586TK/03 Fix log output including CR from clamd.
587
41a13e0a
PH
588PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
589 exist provoked a memory error which could cause a segfault.
590
f625cc5a
PH
591PH/15 Installed PCRE 6.2
592
593PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
594
21f7af35
PH
595PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
596 of the problem. Specifically, suggested +O2 rather than +O1 for the
597 HP-UX compiler.
598
31480e42
PH
599PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
600
2d280592
PH
601PH/20 If a delivery was routed to a non-standard port by means of an SRV
602 record, the port was not correctly logged when the outgoing_port log
603 selector was set (it logged the transort's default port).
604
7cd1141b
PH
605PH/21 Added support for host-specific ports to manualroute, queryprogram,
606 fallback_hosts, and "hosts" in the smtp transport.
607
608PH/22 If the log selector "outgoing_port" is set, the port is now also given on
609 host errors such as "Connection refused".
610
750af86e
PH
611PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
612 authentication with radiusclient 0.4.9:
613
614 - Error returned from rc_read_config was caught wrongly
615 - Username/password not passed on to radius server due to wrong length.
616
617 The presumption is that some radiusclient API changes for 4.51/PH/17
618 were not taken care of correctly. The code is still untested by me (my
619 Linux distribution still has 0.3.2 of radiusclient), but it was
620 contributed by a Radius user.
621
622PH/24 When doing a callout, the value of $domain wasn't set correctly when
623 expanding the "port" option of the smtp transport.
624
4304270b
TK
625TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
626 while reading a MIME header. Thanks to Tom Hughes for a patch.
627
750af86e
PH
628PH/24 Include config.h inside local_scan.h so that configuration settings are
629 available.
630
64ffc24f
PH
631PH/25 Make $smtp_command_argument available after all SMTP commands. This means
632 that in an ACL for RCPT (for example), you can examine exactly what was
633 received.
634
5dd9625b
PH
635PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
636 commands, but it was not correctly comparing the address with the actual
637 client host address. Thus, it would show the EHLO address in Received:
638 header lines when this was not necessary.
639
5591031b
PH
640PH/27 Added the % operator to ${eval:}.
641
ba18e66a
PH
642PH/28 Exim tries to create and chdir to its spool directory when it starts;
643 it should be ignoring failures (because with -C, for example, it has lost
644 privilege). It wasn't ignoring creation failures other than "already
645 exists".
646
9cec981f
PH
647PH/29 Added "crypteq" to the list of supported features that Exim outputs when
648 -bV or -d is used.
649
aa2b5c79
PH
650PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
651 because an input line was too long, either on its own, or by virtue of
1509d3a8
PH
652 too many continuations, the temporary file was not being removed, and the
653 return code was incorrect.
aa2b5c79 654
48a53b7f
PH
655PH/31 Missing "BOOL" in function definition in filtertest.c.
656
1c59d63b
PH
657PH/32 Applied Sieve patches from the maintainer.
658
671012da
TK
659TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
660
1509d3a8
PH
661PH/33 Added "verify = not_blind".
662
663PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
664 Local/Makefile (with some defaults set). These are used in built scripts
665 such as exicyclog, but they have never been used in the exim_install
666 script (though there are many overriding facilities there). I have
667 arranged that the exim_install script now takes note of these two
668 settings.
669
670PH/35 Installed configuration files for Dragonfly.
671
2fe1a124
PH
672PH/36 When a locally submitted message by a trusted user did not contain a
673 From: header, and the sender address was obtained from -f or from an SMTP
674 MAIL command, and the trusted user did not use -F to supply a sender
675 name, $originator_name was incorrectly used when constructing a From:
676 header. Furthermore, $originator_name was used for submission mode
677 messages from external hosts without From: headers in a similar way,
678 which is clearly wrong.
679
8800895a
PH
680PH/37 Added control=suppress_local_fixups.
681
ccfdb010
PH
682PH/38 When log_selector = +received_sender was set, and the addition of the
683 sender made the log line's construction buffer exactly full, or one byte
684 less than full, an overflow happened when the terminating "\n" was
685 subsequently added.
686
1130bfb0
PH
687PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
688 when the result of a list match is failure because a DNS lookup failed.
689
ebcb507f
PH
690PH/40 RM_COMMAND is now used in the building process.
691
c35e155c
PH
692PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
693 the "build-* directories that it finds.
694
95d1f782
PH
695PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
696 address in a domain literal was a prefix of an interface address.
697
fd6de02e
PH
698PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
699 when verifying a sender address, unless rewrite_headers is false.
700
58de37c5
PH
701PH/44 Wrote a long comment about why errors_to addresses are verified as
702 recipients, not senders.
703
261cf466
TF
704TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
705 the ratelimit ACL was added.
706
3ee512ff
PH
707PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
708
e08c430f
PH
709PH/46 Added extra information about PostgreSQL errors to the error string.
710
bef5a11f
PH
711PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
712 avoiding the need to install special zones in a real server. This is
713 backwards compatible; if it can't find the fake resolver, it drops back.
714 Thus, both old and new test suites can be run.
715
7546de58
TF
716TF/02 Added util/ratelimit.pl
717
e5d5a95f
TF
718TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
719 clock is set back in time.
720
2e88a017
TF
721TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
722 Candler <B.Candler@pobox.com>.
723
a5f65aa4
TF
724TF/05 The fix for PH/43 was not completely correct; widen_domains is always
725 OK for addresses that are the result of redirections.
726
e7726cbf
PH
727PH/48 A number of further additions for the benefit of the new test suite,
728 including a fake gethostbyname() that interfaces to the fake DNS resolver
729 (see PH/47 above).
730
a7fdad5b
TF
731TF/06 The fix for widen_domains has also been applied to qualify_single and
732 search_parents which are the other dnslookup options that can cause
733 header rewrites.
734
6af56900
PH
735PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
736 type ("H").
737
0925ede6
PH
738PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
739
66afa403
TF
740TF/07 Exim produced the error message "an SRV record indicated no SMTP
741 service" if it encountered an MX record with an empty target hostname.
742 The message is now "an MX or SRV record indicated no SMTP service".
743
0154e85a
TF
744TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
745 if the DNS of the sending site is misconfigured. This is quite a
746 common situation. This change restores the behaviour of treating a
747 helo verification defer as a failure.
748
16f12c76
PH
749PH/51 If self=fail was set on a router, the bounce message did not include the
750 actual error message.
751
bbe902f0 752
e5a9dba6
PH
753Exim version 4.52
754-----------------
755
756TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
757
22c3b60b
PH
758PH/01 When a transport filter timed out in a pipe delivery, and the pipe
759 command itself ended in error, the underlying message about the transport
760 filter timeout was being overwritten with the pipe command error. Now the
761 underlying error message should be appended to the second error message.
762
06a9b4b5
PH
763TK/01 Fix poll() being unavailable on Mac OSX 10.2.
764
c1ac6996
PH
765PH/02 Reduce the amount of output that "make" produces by default. Full output
766 can still be requested.
767
9c7a242c
PH
768PH/03 The warning log line about a condition test deferring for a "warn" verb
769 was being output only once per connection, rather than after each
770 occurrence (because it was using the same function as for successful
771 "warn" verbs). This seems wrong, so I have changed it.
772
87ba3f5f
PH
773TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
774 it should not have, which might have caused a crash in the right
775 circumstances, but probably never did.
776
777PH/04 Installed a modified version of Tony Finch's patch to make submission
778 mode fix the return path as well as the Sender: header line, and to
779 add a /name= option so that you can make the user's friendly name appear
780 in the header line.
781
29aba418
TF
782TF/03 Added the control = fakedefer ACL modifier.
783
fe0dab11
TF
784TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
785 Mark Lowes for thorough testing.
870f6ba8 786
11d337a4
TK
787TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
788
789TK/03 Merged latest SRS patch from Miles Wilton.
790
415c8f3b
PH
791PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
792 with the definition in sysexits.h (which is #included earlier).
793 Fortunately, Exim does not actually use EX_OK. The code used to try to
794 preserve the sysexits.h value, by assumimg that macro definitions were
795 scanned for macro replacements. I have been disabused of this notion,
796 so now the code just undefines EX_OK before #including unistd.h.
11d337a4 797
958541e9
PH
798PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
799 in the smtp transport. When a block could not be written in a single
800 write() function, the timeout was being re-applied to each part-write.
801 This seems wrong - if the receiver was accepting one byte at a time it
802 would take for ever. The timeout is now adjusted when this happens. It
803 doesn't have to be particularly precise.
804
c206415f
TK
805TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
806 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
807
2a4be8f9
PH
808PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
809 without a domain if the check to <postmaster@domain> fails.
810
1cba11c5
SC
811SC/01 Eximstats: added -xls and the ability to specify output files
812 (patch written by Frank Heydlauf).
813
814SC/02 Eximstats: use FileHandles for outputing results.
815
816SC/03 Eximstats: allow any combination of xls, txt, and html output.
817
818SC/04 Eximstats: fixed display of large numbers with -nvr option
819
820SC/05 Eximstats: fixed merging of reports with empty tables.
821
822SC/06 Eximstats: added the -include_original_destination flag
823
824SC/07 Eximstats: removed tabs and trailing whitespace.
825
1005d00e
TK
826TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
827
828TK/06 MBOX spool code: Add real "From " MBOX separator line
829 so the .eml file is really in mbox format (even though
830 most programs do not really care). Patch from Alex Miller.
831
832TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
833 The latter is generated from $received_to and is only set if the
834 message has one envelope recipient. SA can use these headers,
835 obviously out-of-the-box. Patch from Alex Miller.
836
9b4768fa
PH
837PH/08 The ${def test on a variable was returning false if the variable's
838 value was "0", contrary to what the specification has always said!
839 The result should be true unless the variable is empty.
840
841PH/09 The syntax error of a character other than { following "${if
842 def:variable_name" (after optional whitespace) was not being diagnosed.
843 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
844 accidental colon was present, for example, could give incorrect results.
845
0d7eb84a
PH
846PH/10 Tidied the code in a number of places where the st_size field of a stat()
847 result is used (not including appendfile, where other changes are about
848 to be made).
849
850PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
851 This involved changing a lot of size variables from int to off_t. It
852 should work with maildirs and everything.
853
40727bee
TK
854TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
855 spamd dying while we are connected to it.
856
554d2369
TF
857TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
858 <jgh@wizmail.org>
859
1f922db1
PH
860PH/12 Applied Alex Kiernan's patch for the API change for the error callback
861 function for BDB 4.3.
862
ef213c3b
PH
863PH/13 Changed auto_thaw such that it does not apply to bounce messages.
864
8ac170f3
PH
865PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
866 the sources for PCRE have been re-arranged and more files are now
867 involved.
868
b1c749bb
PH
869PH/15 The code I had for printing potentially long long variables in PH/11
870 above was not the best (it lost precision). The length of off_t variables
871 is now inspected at build time, and an appropriate printing format (%ld
c6c2dc1d
PH
872 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
873 to be "long long int" or "long int". This is needed for the internal
874 formatting function string_vformat().
b1c749bb 875
4aac9b49
PH
876PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
877 the configuration file to be ":syslog", then the script "guesses" where
878 the logs files are, rather than using the compiled in default. In our
879 case the guess is not the same as the compiled default, so the script
880 suddenly stopped working when I started to use syslog. The patch checks
881 to see if log_file_path is "". If so, it attempts to read it from exim
882 with no configuration file to get the compiled in version, before it
883 falls back to the previous guessing code."
884
294520c8
TK
885TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
886 implementing BATV in an Exim configuration. See NewStuff for the gory
887 details.
888
5bd022fe
PH
889PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
890 Makefile that are specific to HP-UX.
891
90e9ce59
PH
892PH/18 If the "use_postmaster" option was set for a recipient callout together
893 with the "random" option, the postmaster address was used as the MAIL
894 FROM address for the random test, but not for the subsequent recipient
895 test. It is now used for both.
896
5ea81592
PH
897PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
898 patch removes a few documentation additions to RFC 3028, because the
899 latest draft now contains them. It adds the new en;ascii-case comparator
900 and a new error check for 8bit text in MIME parts. Comparator and
901 require names are now matched exactly. I enabled the subaddress
902 extension, but it is not well tested yet (read: it works for me)."
903
c6c2dc1d
PH
904PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
905 rework some of the code of TK/09 above to avoid the hardwired use of
906 "%lld" and "long long". Replaced the call to snprintf() with a call to
907 string_vformat().
908
fffffe4c
PH
909PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
910 records point to non-existent hosts", "retry timeout exceeded", and
911 "retry time not reached for any host after a long failure period".
ca02eafb 912
9a26b6b2
PH
913PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
914 experimental DomainKeys support:
915
916 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
917 (2) On an error such as an illegally used "control", the wrong name for
918 the control was given.
919
920 These problems did NOT occur unless DomainKeys support was compiled.
921
4aee0225
PH
922PH/23 Added daemon_startup_retries and daemon_startup_sleep.
923
32d668a5
PH
924PH/24 Added ${if match_ip condition.
925
8187c3f3
PH
926PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
927 databases so that it will be absolutely obvious if a crash occurs in the
928 DB library. This is a regular occurrence (often caused by mis-matched
929 db.h files).
930
ff790e47 931PH/26 Insert a lot of missing (void) casts for functions such as chown(),
f1e894f3
PH
932 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
933 picked up on a user's system that detects such things. There doesn't seem
934 to be a gcc warning option for this - only an attribute that has to be
935 put on the function's prototype. It seems that in Fedora Core 4 they have
936 set this on a number of new functions. No doubt there will be more in due
937 course.
ff790e47 938
5417f6d1
PH
939PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
940 specify a transport. However, if an address that was verified by such a
941 router was the subject of a callout, Exim crashed because it tried to
942 read the rcpt_include_affixes from the non-existent transport. Now it
943 just assumes that the setting of that option is false. This bug was
944 introduced by 4.51/PH/31.
945
59cf8544
PH
946PH/28 Changed -d+all to exclude +memory, because that information is very
947 rarely of interest, but it makes the output a lot bigger. People tend to
948 do -d+all out of habit.
949
e7ad8a65
PH
950PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
951 code in os-type was giving problems when libc.so lives in lib64, like on
952 x86_64 Fedora Core.
953
ade42478
PH
954PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
955 aren't the modern standard, and it seems that some systems' include files
956 don't always have them. Exim was already checking for some of the newer
957 ones like T_AAAA, and defining it itself. I've added checks for all the
958 record types that Exim uses.
959
182ad5cf
PH
960PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
961 not automatically generating a new one, as it is supposed to. This
962 prevented TLS from working. If the file did exist, but contained invalid
963 data, a new version was generated, as expected. It was only the case of a
964 non-existent file that was broken.
965
b0d9fc80
TK
966TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
967 with a change in libdomainkeys > 0.64.
968
969TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
970 from DNS. If the selector record carries the flag, it now has
971 precedence over the domain-wide flag.
972
973TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
974
47c7a64a
PH
975PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
976 the use of a port name that isn't defined in /etc/services) occurred, the
977 message was deferred as in a normal delivery, and thus remained on the
978 spool, instead of being failed because of the mua_wrapper setting. This
979 is now fixed, and I tidied up some of the mua_wrapper messages at the
980 same time.
981
a388bce4
SC
982SC/08 Eximstats: whilst parsing the mainlog(s), store information about
983 the messages in a hash of arrays rather than using individual hashes.
984 This is a bit cleaner and results in dramatic memory savings, albeit
985 at a slight CPU cost.
986
987SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
988 as requested by Marc Sherman.
989
990SC/10 Eximstats: added histograms for user specified patterns as requested
991 by Marc Sherman.
992
0793e4ed
SC
993SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
994
c58b88df
PH
995PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
996 fopen() in the content-scanning modules that did not already have it.
997
e7ad8a65 998
7982096b
PH
999Exim version 4.51
1000-----------------
1001
1a46a8c5
PH
1002TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
1003 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
1004
2f079f46 1005TK/02 Fix ACL "control" statement not being available in MIME ACL.
1a46a8c5
PH
1006
1007TK/03 Fix ACL "regex" condition not being available in MIME ACL.
1008
1009PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
1010 to test Sieve filters that use "vacation".
1011
1012PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
1013 that changes the way the GnuTLS parameters are stored in the cache file.
1014 The new format can be generated externally. For backward compatibility,
1015 if the data in the cache doesn't make sense, Exim assumes it has read an
1016 old-format file, and it generates new data and writes a new file. This
1017 means that you can't go back to an older release without removing the
1018 file.
1019
1020PH/03 A redirect router that has both "unseen" and "one_time" set does not
1021 work if there are any delivery delays because "one_time" forces the
1022 parent to be marked "delivered", so its unseen clone is never tried
1023 again. For this reason, Exim now forbids the simultaneous setting of
1024 these two options.
1025
1026PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
1027 redirected to themselves ("homonym" addresses). Read the long ChangeLog
1028 entry if you want to know the details. The fix, however, neglected to
1029 consider the case when local delivery batching is involved. The test for
1030 "previously delivered" was not happening when checking to see if an
1031 address could be batched with a previous (undelivered) one; under
1032 certain circumstances this could lead to multiple deliveries to the same
c2c19e9d 1033 address.
1a46a8c5
PH
1034
1035PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
1036 in its include files, and this causes problems building Exim.
1037
1038PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
1039 header_syntax) but Exim was just ignoring anything given after a slash.
1040 In particular, this caused confusion with an attempt to use "verify =
1041 reverse_host_lookup/defer_ok". An error is now given when options are
1042 supplied for verify items that do not have them. (Maybe reverse_host_
1043 lookup should have a defer_ok option, but that's a different point.)
1044
1045PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
1046 defined by RFC 821) to 2048, because there were problems with some AUTH
1047 commands, and RFC 1869 says the size should be increased for extended
1048 SMTP commands that take arguments.
1049
1050PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
1051 Finch).
1052
1053PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
1054 "unknown" error; now it says that the functionality isn't in the binary.
8d67ada3 1055
49c2d5ea
PH
1056PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
1057 an address' error message when a string expansion fails (syntax or
f331f3b6
PH
1058 whatever). Otherwise the password may appear in the log. Following change
1059 PH/42 below, there is no longer a chance of it appearing in a bounce
1060 message.
49c2d5ea 1061
bf759a8b
PH
1062PH/11 Installed exipick version 20050225.0 from John Jetmore.
1063
83364d30
PH
1064PH/12 If the last host in a fallback_hosts list was multihomed, only the first
1065 of its addresses was ever tried. (Bugzilla bug #2.)
1066
7999bbd7
PH
1067PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
1068 the result incorrectly in the debug output. (It correctly added a newline
1069 to what was transported.)
1070
7dbf77c9
PH
1071TF/01 Added $received_time.
1072
74e0617f
PH
1073PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
1074 commented out examples of how to interface to a virus scanner and to
1075 SpamAssassin. Also added commented examples of av_scanner and
1076 spamd_address settings.
1077
2f079f46
PH
1078PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
1079 and controls are allowed in which ACLs. There were a couple of minor
1080 errors. Some of the entries in the conditions table (which is a table of
1081 where they are NOT allowed) were getting very unwieldy; rewrote them as a
1082 negation of where the condition IS allowed.
1083
8c841523
PH
1084PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
1085
7766a4f0
PH
1086PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
1087 header file does not have a version number, so I've had to invent a new
1088 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
1089 API. The code is untested by me (my Linux distribution still has 0.3.2 of
1090 radiusclient), but it was contributed by a Radius user.
1091
8b417f2c
PH
1092PH/18 Installed Lars Mainka's patch for the support of CRL collections in
1093 files or directories, for OpenSSL.
1094
901f42cb
PH
1095PH/19 When an Exim process that is running as root has to create an Exim log
1096 file, it does so in a subprocess that runs as exim:exim so as to get the
1097 ownership right at creation (otherwise, other Exim processes might see
1098 the file with the wrong ownership). There was no test for failure of this
1099 fork() call, which would lead to the process getting stuck as it waited
1100 for a non-existent subprocess. Forks do occasionally fail when resources
1101 run out. I reviewed all the other calls to fork(); they all seem to check
1102 for failure.
1103
f9b9210e
PH
1104PH/20 When checking for unexpected SMTP input at connect time (before writing
1105 the banner), Exim was not dealing correctly with a non-positive return
1106 from the read() function. If the client had disconnected by this time,
1107 the result was a log entry for a synchronization error with an empty
1108 string after "input=" when read() returned zero. If read() returned -1
1109 (an event I could not check), uninitialized data bytes were printed.
1110 There were reports of junk text (parts of files, etc) appearing after
1111 "input=".
1112
54cdb463
PH
1113PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
1114
cf00dad6
PH
1115PH/22 Added support for macro redefinition, and (re)definition in between
1116 driver and ACL definitions.
1117
acb1b346
PH
1118PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
1119 forgetting to use the resulting value; it was using the unexpanded value.
1120
c5ddb310
PH
1121PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
1122 hadn't been configured. The fix is from Juergen Kreileder, who
1123 understands it better than I do:
1124
1125 "Here's what I see happening with three configured cyrus_sasl
1126 authenticators configured (plain, login, cram-md5):
1127
1128 On startup auth_cyrus_sasl_init() gets called for each of these.
1129 This means three calls to sasl_listmech() without a specified mech_list.
1130 => SASL tests which mechs of all available mechs actually work
1131 => three warnings about OTP not working
1132 => the returned list contains: plain, login, cram-md5, digest-md5, ...
1133
1134 With the patch, sasl_listmech() also gets called three times. But now
1135 SASL's mech_list option is set to the server_mech specified in the the
1136 authenticator. Or in other words, the answer from sasl_listmech()
1137 gets limited to just the mech you're testing for (which is different
1138 for each call.)
1139 => the return list contains just 'plain' or 'login', 'cram-md5' or
1140 nothing depending on the value of ob->server_mech.
1141
1142 I've just tested the patch: Authentication still works fine,
1143 unavailable mechs specified in the exim configuration are still
1144 caught, and the auth.log warnings about OTP are gone."
1145
31619da6
PH
1146PH/25 When debugging is enabled, the contents of the command line are added
1147 to the debugging output, even when log_selector=+arguments is not
1148 specified.
1149
bebaf0fc
PH
1150PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
1151 answer is "GNU", and only if the return is "GNU/something" is the answer
1152 "Linux".
1153
475fe28a
PH
1154PH/27 $acl_verify_message is now set immediately after the failure of a
1155 verification in an ACL, and so is available in subsequent modifiers. In
1156 particular, the message can be preserved by coding like this:
1157
1158 warn !verify = sender
1159 set acl_m0 = $acl_verify_message
1160
1161 Previously, $acl_verify_message was set only while expanding "message"
1162 and "log_message" when a very denied access.
1163
7e8bec7a
PH
1164PH/28 Modified OS/os.c-Linux with
1165
1166 -#ifndef OS_LOAD_AVERAGE
1167 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
1168
1169 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
1170 nomenclature these days.)
1171
e4a89c47
PH
1172PH/29 Installed patch from the Sieve maintainer that adds the options
1173 sieve_useraddress and sieve_subaddress to the redirect router.
1174
5ca2a9a1
PH
1175PH/30 In these circumstances:
1176 . Two addresses routed to the same list of hosts;
1177 . First host does not offer TLS;
1178 . First host accepts first address;
1179 . First host gives temporary error to second address;
1180 . Second host offers TLS and a TLS session is established;
1181 . Second host accepts second address.
1182 Exim incorrectly logged both deliveries with the TLS parameters (cipher
1183 and peerdn, if requested) that were in fact used only for the second
1184 address.
7e8bec7a 1185
c688b954
PH
1186PH/31 When doing a callout as part of verifying an address, Exim was not paying
1187 attention to any local part prefix or suffix that was matched by the
1188 router that accepted the address. It now behaves in the same way as it
1189 does for delivery: the affixes are removed from the local part unless
1190 rcpt_include_affixes is set on the transport.
1191
fed77020
PH
1192PH/32 Add the sender address, as F=<...>, to the log line when logging a
1193 timeout during the DATA phase of an incoming message.
1194
7fe1560f
PH
1195PH/33 Sieve envelope tests were broken for match types other than :is. I have
1196 applied a patch sanctioned by the Sieve maintainer.
c688b954 1197
ebb6e6d5
PH
1198PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
1199 the uid or gid is negative. A case of a negative gid caused this to be
1200 noticed. The fix allows for either to be negative.
1201
9c4e8f60
PH
1202PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
1203 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
1204 overlooked.
1205
1206PH/36 The change PH/12 above was broken. Fixed it.
1207
d7174846
PH
1208PH/37 Exim used to check for duplicate addresses in the middle of routing, on
1209 the grounds that routing the same address twice would always produce the
1210 same answer. This might have been true once, but it is certainly no
1211 longer true now. Routing a child address may depend on the previous
1212 routing that produced that child. Some complicated redirection strategies
1213 went wrong when messages had multiple recipients, and made Exim's
1214 behaviour dependent on the order in which the addresses were given.
1215
1216 I have moved the duplicate checking until after the routing is complete.
1217 Exim scans the addresses that are assigned to local and remote
1218 transports, and removes any duplicates. This means that more work will be
1219 done, as duplicates will always all be routed, but duplicates are
1220 presumably rare, so I don't expect this is of any significance.
1221
1222 For deliveries to pipes, files, and autoreplies, the duplicate checking
1223 still happens during the routing process, since they are not going to be
1224 routed further.
1225
cfe75fc3
PH
1226PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
1227 It corrects a timeout issue with spamd. This is Ian's comment: "The
1228 background is that sometimes spamd either never reads data from a
1229 connection it has accepted, or it never writes response data. The exiscan
1230 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
1231 blindly assumes that writes won't block so it may never time out."
1232
be22d70e
PH
1233PH/39 Allow G after quota size as well as K and M.
1234
0612b098
PH
1235PH/40 The value set for $authenticated_id in an authenticator may not contain
1236 binary zeroes or newlines because the value is written to log lines and
1237 to spool files. There was no check on this. Now the value is run through
1238 the string_printing() function so that such characters are converted to
1239 printable escape sequences.
1240
2e0c1448
PH
1241PH/41 $message_linecount is a new variable that contains the total number of
1242 lines in the message. Compare $body_linecount, which is the count for the
1243 body only.
1244
447d236c
PH
1245PH/42 Exim no longer gives details of delivery errors for specific addresses in
1246 bounce and delay warning messages, except in certain special cases, which
1247 are as follows:
1248
1249 (a) An SMTP error message from a remote host;
1250 (b) A message specified in a :fail: redirection;
1251 (c) A message specified in a "fail" command in a system filter;
1252 (d) A message specified in a FAIL return from the queryprogram router;
1253 (e) A message specified by the cannot_route_message router option.
1254
1255 In these cases only, Exim does include the error details in bounce and
1256 warning messages. There are also a few cases where bland messages such
1257 as "unrouteable address" or "local delivery error" are given.
1258
d20976dc
PH
1259PH/43 $value is now also set for the "else" part of a ${run expansion.
1260
f656d135
PH
1261PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
1262 being worked on, but at least Exim now implements the latest version to
1263 play with."
1264
2e2a30b4
PH
1265PH/45 In a pipe transport, although a timeout while waiting for the pipe
1266 process to complete was treated as a delivery failure, a timeout while
1267 writing the message to the pipe was logged, but erroneously treated as a
1268 successful delivery. Such timeouts include transport filter timeouts. For
1269 consistency with the overall process timeout, these timeouts are now
1270 treated as errors, giving rise to delivery failures by default. However,
1271 there is now a new Boolean option for the pipe transport called
1272 timeout_defer, which, if set TRUE, converts the failures into defers for
1273 both kinds of timeout. A transport filter timeout is now identified in
1274 the log output.
1275
9176e9f0
PH
1276PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
1277 systems where "make" and "gmake" are different, calling "gmake" at top
1278 level broke things. I've arranged for the value of $(MAKE) to be passed
1279 from the Makefile to this script so that it can call the same version of
1280 "make".
1281
7982096b 1282
bbe902f0
PH
1283A note about Exim versions 4.44 and 4.50
1284----------------------------------------
1285
1286Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
1287changes of various kinds. As a consequence, a big documentation update was
1288needed. This delayed the release for rather longer than seemed good, especially
1289in the light of a couple of (minor) security issues. Therefore, the changes
1290that fixed bugs were backported into 4.43, to create a 4.44 maintenance
1291release. So 4.44 and 4.50 are in effect two different branches that both start
1292from 4.43.
1293
1294I have left the 4.50 change log unchanged; it contains all the changes since
12954.43. The change log for 4.44 is below; many of its items are identical to
1296those for 4.50. This seems to be the most sensible way to preserve the
1297historical information.
1298
1299
f7b63901 1300Exim version 4.50
495ae4b0
PH
1301-----------------
1302
5fe762f6
PH
1303 1. Minor wording change to the doc/README.SIEVE file.
1304
139059f6 1305 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
5fe762f6 1306 computation of the current number of files was incorrect.
495ae4b0 1307
7086e875
PH
1308 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
1309 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
1310 place.
1311
35af9f61
PH
1312 4. Give more explanation in the error message when the command for a transport
1313 filter fails to execute.
1314
b668c215
PH
1315 5. There are several places where Exim runs a non-Exim command in a
1316 subprocess. The SIGUSR1 signal should be disabled for these processes. This
1317 was being done only for the command run by the queryprogram router. It is
1318 now done for all such subprocesses. The other cases are: ${run, transport
1319 filters, and the commands run by the lmtp and pipe transports.
1320
a494b1e1
PH
1321 6. Added CONFIGURE_GROUP build-time option.
1322
1323 7. Some older OS have a limit of 256 on the maximum number of file
1324 descriptors. Exim was using setrlimit() to set 1000 as a large value
1325 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
1326 systems. I've change it so that if it can't get 1000, it tries for 256.
35edf2ff 1327
c5fcb476
PH
1328 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
1329 was an oversight, and furthermore, ever since the addition of extra
1330 controls (e.g. 4.43/32), the checks on when to allow different forms of
1331 "control" were broken. There should now be diagnostics for all cases when a
1332 control that does not make sense is encountered.
1333
69358f02
PH
1334 9. Added the /retain_sender option to "control=submission".
1335
5be20824
PH
133610. $recipients is now available in the predata ACL (oversight).
1337
eb2c0248
PH
133811. Tidy the search cache before the fork to do a delivery from a message
1339 received from the command line. Otherwise the child will trigger a lookup
1340 failure and thereby defer the delivery if it tries to use (for example) a
1341 cached ldap connection that the parent has called unbind on.
1342
2a3eea10
PH
134312. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
1344 of $address_data from the recipient verification was clobbered by the
1345 sender verification.
1346
134713. The value of address_data from a sender verification is now available in
1348 $sender_address_data in subsequent conditions in the ACL statement.
1349
23c7ff99
PH
135014. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
1351
4deaf07d
PH
135215. Added a new option "connect=<time>" to callout options, to set a different
1353 connection timeout.
1354
926e1192
PH
135516. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
1356 was its contents. (It was OK if the option was not defined at all.)
1357
650edc6f
PH
135817. A "Completed" log line is now written for messages that are removed from
1359 the spool by the -Mrm option.
1360
2c7db3f5
PH
136118. New variables $sender_verify_failure and $recipient_verify_failure contain
1362 information about exactly what failed.
1363
3d235903
PH
136419. Added -dd to debug only the daemon process.
1365
7c7ad977
PH
136620. Incorporated Michael Haardt's patch to ldap.c for improving the way it
1367 handles timeouts, both on the server side and network timeouts. Renamed the
1368 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
1369
981756db
PH
137021. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
1371 It is now set to "smtps".
1372
d4eb88df
PH
137322. $host_address is now set to the target address during the checking of
1374 ignore_target_hosts.
1375
137623. When checking ignore_target_hosts for an ipliteral router, no host name was
1377 being passed; this would have caused $sender_host_name to have been used if
1378 matching the list had actually called for a host name (not very likely,
1379 since this list is usually IP addresses). A host name is now passed as
1380 "[x.x.x.x]".
1381
7d468ab8
PH
138224. Changed the calls that set up the SIGCHLD handler in the daemon to use the
1383 code that specifies a non-restarting handler (typically sigaction() in
1384 modern systems) in an attempt to fix a rare and obscure crash bug.
1385
138625. Narrowed the window for a race in the daemon that could cause it to ignore
1387 SIGCHLD signals. This is not a major problem, because they are used only to
1388 wake it up if nothing else does.
1389
62c0818f
PH
139026. A malformed maildirsize file could cause Exim to calculate negative values
1391 for the mailbox size or file count. Odd effects could occur as a result.
1392 The maildirsize information is now recalculated if the size or filecount
1393 end up negative.
1394
26034054
PH
139527. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
1396 support for a long time. Removed HAVE_SYS_VFS_H.
1397
af66f652
PH
139828. Installed the latest version of exipick from John Jetmore.
1399
90af77f4
PH
140029. In an address list, if the pattern was not a regular expression, an empty
1401 subject address (from a bounce message) matched only if the pattern was an
1402 empty string. Non-empty patterns were not even tested. This was the wrong
1403 because it is perfectly reasonable to use an empty address as part of a
1404 database query. An empty address is now tested by patterns that are
1405 lookups. However, all the other forms of pattern expect the subject to
1406 contain a local part and a domain, and therefore, for them, an empty
1407 address still always fails if the pattern is not itself empty.
1408
d8ef3577
PH
140930. Exim went into a mad DNS loop when attempting to do a callout where the
1410 host was specified on an smtp transport, and looking it up yielded more
1411 than one IP address.
1412
5cb8cbc6
PH
141331. Re-factored the code for checking spool and log partition space into a
1414 function that finds that data and another that does the check. The former
1415 is then used to implement four new variables: $spool_space, $log_space,
1416 $spool_inodes, and $log_inodes.
1417
14702f5b
PH
141832. The RFC2047 encoding function was originally intended for short strings
1419 such as real names; it was not keeping to the 75-character limit for
1420 encoded words that the RFC imposes. It now respects the limit, and
1421 generates multiple encoded words if necessary. To be on the safe side, I
1422 have increased the buffer size for the ${rfc2047: expansion operator from
1423 1024 to 2048 bytes.
1424
063b1e99
PH
142533. It is now permitted to omit both strings after an "if" condition; if the
1426 condition is true, the result is "true". As before, when the second string
1427 is omitted, a false condition yields an empty string. This makes it less
1428 cumbersome to write custom ACL and router conditions.
1429
652e1b65
PH
143034. Failure to deliver a bounce message always caused it to be frozen, even if
1431 there was an errors_to setting on the router. The errors_to setting is now
1432 respected.
1433
6f0c9a4f
PH
143435. If an IPv6 address is given for -bh or -bhc, it is now converted to the
1435 canonical form (fully expanded) before being placed in
1436 $sender_host_address.
1437
33397d19
PH
143836. The table in the code that translates DNS record types into text (T_A to
1439 "A" for instance) was missing entries for NS and CNAME. It is just possible
1440 that this could have caused confusion if both these types were looked up
1441 for the same domain, because the text type is used as part of Exim's
1442 per-process caching. But the chance of anyone hitting this buglet seems
1443 very small.
1444
7bb56e1f
PH
144537. The dnsdb lookup has been extended in a number of ways.
1446
1447 (1) There is a new type, "zns", which walks up the domain tree until it
1448 finds some nameserver records. It should be used with care.
1449
ea3bc19b
PH
1450 (2) There is a new type, "mxh", which is like "mx" except that it returns
1451 just the host names, not the priorities.
1452
1453 (3) It is now possible to give a list of domains (or IP addresses) to be
ff4dbb19
PH
1454 looked up. The behaviour when one of the lookups defers can be
1455 controlled by a keyword.
7bb56e1f 1456
ea3bc19b 1457 (4) It is now possible to specify the separator character for use when
7bb56e1f 1458 multiple records are returned.
33397d19 1459
0bcb2a0e
PH
146038. The dnslists ACL condition has been extended: it is now possible to supply
1461 a list of IP addresses and/or domains to be looked up in a particular DNS
1462 domain.
1463
2ac0e484
PH
146439. Added log_selector=+queue_time_overall.
1465
4e1fde53
PH
146640. When running the queue in the test harness, wait just a tad after forking a
1467 delivery process, to get repeatability of debugging output.
1468
de365ded
PH
146941. Include certificate and key file names in error message when GnuTLS fails
1470 to set them up, because the GnuTLS error message doesn't include the name
1471 of the failing file when there is a problem reading it.
1472
f05da2e8
PH
147342. Allow both -bf and -bF in the same test run.
1474
d6453af2
PH
147543. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
1476
f7b63901
PH
147744. The "Exiscan patch" is now merged into the mainline Exim source.
1478
147945. Sometimes the final signoff response after QUIT could fail to get
1480 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
1481 < 0 before doing a fflush(). This bug looks as though it goes back to the
1482 introduction of TLS in release 3.20, but "sometimes" must have been rare
1483 because the tests only now provoked it.
1484
a444213a
PH
148546. Reset the locale to "C" after calling embedded Perl, in case it was changed
1486 (this can affect the format of dates).
1487
0ec020ea
PH
148847. exim_tidydb, when checking for the continued existence of a message for
1489 which it has found a message-specific retry record, was not finding
1490 messages that were in split spool directories. Consequently, it was
1491 deleting retry records that should have stayed in existence.
1492
b1206957
PH
149348. Steve fixed some bugs in eximstats.
1494
149549. The SPA authentication driver was not abandoning authentication and moving
1496 on to the next authenticator when an expansion was forced to fail,
1497 contradicting the general specification for all authenticators. Instead it
1498 was generating a temporary error. It now behaves as specified.
1499
26dd5a95
PH
150050. The default ordering of permitted cipher suites for GnuTLS was pessimal
1501 (the order specifies the preference for clients). The order is now AES256,
1502 AES128, 3DES, ARCFOUR128.
1503
343b2385
PH
150451. Small patch to Sieve code - explicitly set From: when generating an
1505 autoreply.
1506
1c5466b9
PH
150752. Exim crashed if a remote delivery caused a very long error message to be
1508 recorded - for instance if somebody sent an entire SpamAssassin report back
1509 as a large number of 550 error lines. This bug was coincidentally fixed by
1510 increasing the size of one of Exim's internal buffers (big_buffer) that
1511 happened as part of the Exiscan merge. However, to be on the safe side, I
1512 have made the code more robust (and fixed the comments that describe what
1513 is going on).
1514
55ee9ee3
PH
151553. Now that there can be additional text after "Completed" in log lines (if
1516 the queue_time_overall log selector is set), a one-byte patch to exigrep
1517 was needed to allow it to recognize "Completed" as not the last thing in
1518 the line.
1519
d38f8232
PH
152054. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
1521 patch that reportedly fixes this has been added. I am not expert enough to
1522 create a test for it. This is what the patch creator wrote:
1523
1524 "I found a little strange behaviour of ldap code when working with
1525 Windows 2003 AD Domain, where users was placed in more than one
1526 Organization Units. When I tried to give exim partial DN, the exit code
1527 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
1528 But simultaneously result of request was absolutely normal ldap result,
1529 so I produce this patch..."
1530
3295e65b
PH
1531 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
1532 REFERENCE, so I have modified the code to exclude the patch when that macro
1533 is not defined.
1534
7102e136
PH
153555. Some experimental protocols are using DNS PTR records for new purposes. The
1536 keys for these records are domain names, not reversed IP addresses. The
b975ba52
PH
1537 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
1538 leaves it alone. Component reversal etc. now happens only for IP addresses.
ea3a6f44 1539 CAN-2005-0021
7102e136 1540
3ca0ba97
PH
154156. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
1542
c2bcbe20
PH
154357. Double the size of the debug message buffer (to 2048) so that more of very
1544 long debug lines gets shown.
1545
18ce445d
PH
154658. The exicyclog utility now does better if the number of log files to keep
1547 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
1548
1f5b4c3d
PH
154959. Two changes related to the smtp_active_hostname option:
1550
1551 (1) $smtp_active_hostname is now available as a variable.
1552 (2) The default for smtp_banner uses $smtp_active_hostname instead
1553 of $primary_hostname.
1554
b975ba52
PH
155560. The host_aton() function is supposed to be passed a string that is known
1556 to be a valid IP address. However, in the case of IPv6 addresses, it was
1557 not checking this. This is a hostage to fortune. Exim now panics and dies
1558 if the condition is not met. A case was found where this could be provoked
85b87bc2
PH
1559 from a dnsdb PTR lookup with an IPv6 address that had more than 8
1560 components; fortuitously, this particular loophole had already been fixed
1561 by change 4.50/55 above.
1562
1563 If there are any other similar loopholes, the new check in host_aton()
1564 itself should stop them being exploited. The report I received stated that
1565 data on the command line could provoke the exploit when Exim was running as
1566 exim, but did not say which command line option was involved. All I could
1567 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
1568 running as the user.
ea3a6f44 1569 CAN-2005-0021
85b87bc2
PH
1570
157161. There was a buffer overflow vulnerability in the SPA authentication code
1572 (which came originally from the Samba project). I have added a test to the
1573 spa_base64_to_bits() function which I hope fixes it.
ea3a6f44 1574 CAN-2005-0022
b975ba52 1575
17ffcae7
PH
157662. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
1577 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
1578
d95f9fdb
PH
157963. The daemon start-up calls getloadavg() while still root for those OS that
1580 need the first call to be done as root, but it missed one case: when
1581 deliver_queue_load_max is set with deliver_drop_privilege. This is
1582 necessary for the benefit of the queue runner, because there is no re-exec
1583 when deliver_drop_privilege is set.
1584
86b8287f
PH
158564. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
1586 This has been fixed.
1587
60dc5e56
PH
158865. Caching of lookup data for "hosts =" ACL conditions, when a named host list
1589 was in use, was not putting the data itself into the right store pool;
1590 consequently, it could be overwritten for a subsequent message in the same
1591 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
1592 the caching.)
1593
533244af
PH
159466. Added hosts_max_try_hardlimit to the smtp transport, default 50.
1595
a5a28604
PH
159667. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
1597 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
1598 the function were treating the return as a boolean value, which happened to
1599 work because 0=false and not-0=true, but is not correct code.
1600
7e634d24
PH
160168. The host_aton() function was not handling scoped IPv6 addresses (those
1602 with, for example, "%eth0" on the end) correctly.
1603
3e11c26b
PH
160469. Fixed some compiler warnings in acl.c for the bitmaps specified with
1605 negated items (that is, ~something) in unsigned ints. Some compilers
1606 apparently mutter when there is no cast.
1607
6729cf78
PH
160870. If an address verification called from an ACL failed, and did not produce a
1609 user-specific message (i.e. there was only a "system" message), nothing was
1610 put in $acl_verify_message. In this situation, it now puts the system
1611 message there.
1612
00f00ca5
PH
161371. Change 4.23/11 added synchronization checking at the start of an SMTP
1614 session; change 4.31/43 added the unwanted input to the log line - except
1615 that it did not do this in the start of session case. It now does.
1616
c9bdd01c
PH
161772. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
1618 This is wrong and can cause the other end to generate a synchronization
1619 error if it is another Exim or anything else that does the synchronization
1620 check. A QUIT command is no longer sent after a timeout.
1621
d43194df
PH
162273. $host_lookup_deferred has been added, to make it easier to detect DEFERs
1623 during host lookups.
1624
fe5b5d0b
PH
162574. The defer_ok option of callout verification was not working if it was used
1626 when verifying addresses in header lines, that is, for this case:
1627
1628 verify = header_sender/callout=defer_ok
1629
76a2d7ba
PH
163075. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
1631 those file descriptors could be used for SMTP connections. If anything
1632 wrote to stderr (the example that came up was "warn" in embedded Perl), it
1633 could be sent to the SMTP client, causing chaos. The daemon now opens
1634 stdin, stdout, and stderr to /dev/null when it puts itself into the
1635 background.
1636
163776. Arrange for output from Perl's "warn" command to be written to Exim's main
1638 log by default. The user can override this with suitable Perl magic.
1639
04f7d5b9
PH
164077. The use of log_message on a "discard" ACL verb, which is supposed to add to
1641 the log message when discard triggers, was not working for the DATA ACL or
1642 for the non-SMTP ACL.
1643
bc60667e
PH
164478. Error message wording change in sieve.c.
1645
bb6e88ff
PH
164679. If smtp_accept_max_per_host was set, the number of connections could be
1647 restricted to fewer than expected, because the daemon was trying to set up
1648 a new connection before checking whether the processes handling previous
1649 connections had finished. The check for completed processes is now done
1650 earlier. On busy systems, this bug wouldn't be noticed because something
1651 else would have woken the daemon, and it would have reaped the completed
1652 process earlier.
1653
1e70f85b
PH
165480. If a message was submitted locally by a user whose login name contained one
1655 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
1656 It caused a spool format error. I have fixed the spool reading code. A
1657 related problem was that the "from" clause in the Received: line became
1658 illegal because of the space(s). It is now covered by ${quote_local_part.
1659
166081. Included the latest eximstats from Steve (adds average sizes to HTML Top
1661 tables).
1662
4e01f9d6
PH
166382. Updated OS/Makefile-AIX as per message from Mike Meredith.
1664
1ee1cef2
PH
166583. Patch from Sieve maintainer to fix unterminated string problem in
1666 "vacation" handling.
1667
6e2b4ccc
PH
166884. Some minor changes to the Linux configuration files to help with other
1669 OS variants using glibc.
1670
8e669ac1
PH
167185. One more patch for Sieve to update vacation handling to latest spec.
1672
495ae4b0 1673
bbe902f0
PH
1674----------------------------------------------------
1675See the note above about the 4.44 and 4.50 releases.
1676----------------------------------------------------
1677
1678
1679Exim version 4.44
1680-----------------
1681
1682 1. Change 4.43/35 introduced a bug that caused file counts to be
1683 incorrectly computed when quota_filecount was set in an appendfile
1684 transport
1685
1686 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
1687 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
1688 place.
1689
1690 3. Give more explanation in the error message when the command for a transport
1691 filter fails to execute.
1692
1693 4. There are several places where Exim runs a non-Exim command in a
1694 subprocess. The SIGUSR1 signal should be disabled for these processes. This
1695 was being done only for the command run by the queryprogram router. It is
1696 now done for all such subprocesses. The other cases are: ${run, transport
1697 filters, and the commands run by the lmtp and pipe transports.
1698
1699 5. Some older OS have a limit of 256 on the maximum number of file
1700 descriptors. Exim was using setrlimit() to set 1000 as a large value
1701 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
1702 systems. I've change it so that if it can't get 1000, it tries for 256.
1703
1704 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
1705 was an oversight, and furthermore, ever since the addition of extra
1706 controls (e.g. 4.43/32), the checks on when to allow different forms of
1707 "control" were broken. There should now be diagnostics for all cases when a
1708 control that does not make sense is encountered.
1709
1710 7. $recipients is now available in the predata ACL (oversight).
1711
1712 8. Tidy the search cache before the fork to do a delivery from a message
1713 received from the command line. Otherwise the child will trigger a lookup
1714 failure and thereby defer the delivery if it tries to use (for example) a
1715 cached ldap connection that the parent has called unbind on.
1716
1717 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
1718 of $address_data from the recipient verification was clobbered by the
1719 sender verification.
1720
172110. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
1722 was its contents. (It was OK if the option was not defined at all.)
1723
172411. A "Completed" log line is now written for messages that are removed from
1725 the spool by the -Mrm option.
1726
172712. $host_address is now set to the target address during the checking of
1728 ignore_target_hosts.
1729
173013. When checking ignore_target_hosts for an ipliteral router, no host name was
1731 being passed; this would have caused $sender_host_name to have been used if
1732 matching the list had actually called for a host name (not very likely,
1733 since this list is usually IP addresses). A host name is now passed as
1734 "[x.x.x.x]".
1735
173614. Changed the calls that set up the SIGCHLD handler in the daemon to use the
1737 code that specifies a non-restarting handler (typically sigaction() in
1738 modern systems) in an attempt to fix a rare and obscure crash bug.
1739
174015. Narrowed the window for a race in the daemon that could cause it to ignore
1741 SIGCHLD signals. This is not a major problem, because they are used only to
1742 wake it up if nothing else does.
1743
174416. A malformed maildirsize file could cause Exim to calculate negative values
1745 for the mailbox size or file count. Odd effects could occur as a result.
1746 The maildirsize information is now recalculated if the size or filecount
1747 end up negative.
1748
174917. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
1750 support for a long time. Removed HAVE_SYS_VFS_H.
1751
ea3a6f44 175218. Updated exipick to current release from John Jetmore.
bbe902f0
PH
1753
175419. Allow an empty sender to be matched against a lookup in an address list.
1755 Previously the only cases considered were a regular expression, or an
1756 empty pattern.
1757
175820. Exim went into a mad DNS lookup loop when doing a callout where the
1759 host was specified on the transport, if the DNS lookup yielded more than
1760 one IP address.
1761
ea3a6f44
NM
176221. The RFC2047 encoding function was originally intended for short strings
1763 such as real names; it was not keeping to the 75-character limit for
1764 encoded words that the RFC imposes. It now respects the limit, and
1765 generates multiple encoded words if necessary. To be on the safe side, I
1766 have increased the buffer size for the ${rfc2047: expansion operator from
1767 1024 to 2048 bytes.
bbe902f0 1768
ea3a6f44
NM
176922. Failure to deliver a bounce message always caused it to be frozen, even if
1770 there was an errors_to setting on the router. The errors_to setting is now
1771 respected.
bbe902f0
PH
1772
177323. If an IPv6 address is given for -bh or -bhc, it is now converted to the
1774 canonical form (fully expanded) before being placed in
1775 $sender_host_address.
1776
177724. Updated eximstats to version 1.33
1778
ea3a6f44
NM
177925. Include certificate and key file names in error message when GnuTLS fails
1780 to set them up, because the GnuTLS error message doesn't include the name
1781 of the failing file when there is a problem reading it.
bbe902f0
PH
1782
178326. Expand error message when OpenSSL has problems setting up cert/key files.
ea3a6f44 1784 As per change 25.
bbe902f0 1785
ea3a6f44
NM
178627. Reset the locale to "C" after calling embedded Perl, in case it was changed
1787 (this can affect the format of dates).
bbe902f0 1788
ea3a6f44
NM
178928. exim_tidydb, when checking for the continued existence of a message for
1790 which it has found a message-specific retry record, was not finding
1791 messages that were in split spool directories. Consequently, it was
1792 deleting retry records that should have stayed in existence.
bbe902f0
PH
1793
179429. eximstats updated to version 1.35
1795 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
1796 1.35 - bugfix such that pie charts by volume are generated correctly
1797
ea3a6f44
NM
179830. The SPA authentication driver was not abandoning authentication and moving
1799 on to the next authenticator when an expansion was forced to fail,
1800 contradicting the general specification for all authenticators. Instead it
1801 was generating a temporary error. It now behaves as specified.
bbe902f0 1802
ea3a6f44
NM
180331. The default ordering of permitted cipher suites for GnuTLS was pessimal
1804 (the order specifies the preference for clients). The order is now AES256,
1805 AES128, 3DES, ARCFOUR128.
bbe902f0 1806
ea3a6f44
NM
180731. Small patch to Sieve code - explicitly set From: when generating an
1808 autoreply.
bbe902f0 1809
ea3a6f44
NM
181032. Exim crashed if a remote delivery caused a very long error message to be
1811 recorded - for instance if somebody sent an entire SpamAssassin report back
1812 as a large number of 550 error lines. This bug was coincidentally fixed by
1813 increasing the size of one of Exim's internal buffers (big_buffer) that
1814 happened as part of the Exiscan merge. However, to be on the safe side, I
1815 have made the code more robust (and fixed the comments that describe what
1816 is going on).
bbe902f0 1817
ea3a6f44
NM
181833. Some experimental protocols are using DNS PTR records for new purposes. The
1819 keys for these records are domain names, not reversed IP addresses. The
1820 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
1821 leaves it alone. Component reversal etc. now happens only for IP addresses.
bbe902f0
PH
1822 CAN-2005-0021
1823
ea3a6f44
NM
182434. The host_aton() function is supposed to be passed a string that is known
1825 to be a valid IP address. However, in the case of IPv6 addresses, it was
1826 not checking this. This is a hostage to fortune. Exim now panics and dies
1827 if the condition is not met. A case was found where this could be provoked
1828 from a dnsdb PTR lookup with an IPv6 address that had more than 8
1829 components; fortuitously, this particular loophole had already been fixed
1830 by change 4.50/55 or 4.44/33 above.
1831
1832 If there are any other similar loopholes, the new check in host_aton()
1833 itself should stop them being exploited. The report I received stated that
1834 data on the command line could provoke the exploit when Exim was running as
1835 exim, but did not say which command line option was involved. All I could
1836 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
1837 running as the user.
bbe902f0
PH
1838 CAN-2005-0021
1839
ea3a6f44
NM
184035. There was a buffer overflow vulnerability in the SPA authentication code
1841 (which came originally from the Samba project). I have added a test to the
1842 spa_base64_to_bits() function which I hope fixes it.
bbe902f0
PH
1843 CAN-2005-0022
1844
ea3a6f44
NM
184536. The daemon start-up calls getloadavg() while still root for those OS that
1846 need the first call to be done as root, but it missed one case: when
1847 deliver_queue_load_max is set with deliver_drop_privilege. This is
1848 necessary for the benefit of the queue runner, because there is no re-exec
1849 when deliver_drop_privilege is set.
bbe902f0 1850
ea3a6f44
NM
185137. Caching of lookup data for "hosts =" ACL conditions, when a named host list
1852 was in use, was not putting the data itself into the right store pool;
1853 consequently, it could be overwritten for a subsequent message in the same
1854 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
1855 the caching.)
bbe902f0 1856
ea3a6f44
NM
185738. Sometimes the final signoff response after QUIT could fail to get
1858 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
1859 < 0 before doing a fflush(). This bug looks as though it goes back to the
1860 introduction of TLS in release 3.20, but "sometimes" must have been rare
1861 because the tests only now provoked it.
bbe902f0
PH
1862
1863
495ae4b0
PH
1864Exim version 4.43
1865-----------------
1866
1867 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
1868 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
1869 Now it returns an integer. A number of calls were still expecting a T/F
1870 return. Fortuitously, in all cases, the tests worked in OK situations,
1871 which is the norm. However, things would have gone wrong on any write
1872 failures on the smtp file descriptor. This function is used when sending
1873 messages over SMTP and also when doing verify callouts.
1874
1875 2. When Exim is called to do synchronous delivery of a locally submitted
1876 message (the -odf or -odi options), it no longer closes stderr before doing
1877 the delivery.
1878
1879 3. Implemented the mua_wrapper option.
1880
1881 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
1882
1883 5. Implemented the functions header_remove(), header_testname(),
1884 header_add_at_position(), and receive_remove_recipient(), and exported them
1885 to local_scan().
1886
1887 6. If an ACL "warn" statement specified the addition of headers, Exim already
1888 inserted X-ACL-Warn: at the start if there was no header name. However, it
1889 was not making this test for the second and subsequent header lines if
1890 there were newlines in the string. This meant that an invalid header could
1891 be inserted if Exim was badly configured.
1892
1893 7. Allow an ACL "warn" statement to add header lines at the start or after all
1894 the Received: headers, as well as at the end.
1895
1896 8. Added the rcpt_4xx retry error code.
1897
1898 9. Added postmaster_mailfrom=xxx to callout verification option.
1899
190010. Added mailfrom=xxxx to the callout verification option, for verify=
1901 header_sender only.
1902
190311. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
1904 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
1905
190612. Inserted some casts to stop certain compilers warning when using pointer
1907 differences as field lengths or precisions in printf-type calls (mostly
1908 affecting debugging statements).
1909
191013. Added optional readline() support for -be (dynamically loaded).
1911
191214. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
1913 same clock tick as a message's arrival, so that its received time was the
1914 same as the "first fail" time on the retry record, and that message
1915 remained on the queue past the ultimate address timeout, every queue runner
1916 would try a delivery (because it was past the ultimate address timeout) but
1917 after another failure, the ultimate address timeout, which should have then
1918 bounced the address, did not kick in. This was a "< instead of <=" error;
1919 in most cases the first failure would have been in the next clock tick
1920 after the received time, and all would be well.
1921
192215. The special items beginning with @ in domain lists (e.g. @mx_any) were not
1923 being recognized when the domain list was tested by the match_domain
1924 condition in an expansion string.
1925
192616. Added the ${str2b64: operator.
1927
192817. Exim was always calling setrlimit() to set a large limit for the number of
1929 processes, without checking whether the existing limit was already
1930 adequate. (It did check for the limit on file descriptors.) Furthermore,
1931 errors from getrlimit() and setrlimit() were being ignored. Now they are
1932 logged to the main and panic logs, but Exim does carry on, to try to do its
1933 job under whatever limits there are.
1934
193518. Imported PCRE 5.0.
1936
193719. Trivial typo in log message " temporarily refused connection" (the leading
1938 space).
1939
194020. If the log selector return_path_on_delivery was set and an address was
1941 redirected to /dev/null, the delivery process crashed because it assumed
1942 that a return path would always be set for a "successful" delivery. In this
1943 case, the whole delivery is bypassed as an optimization, and therefore no
1944 return path is set.
1945
194621. Internal re-arrangement: the function for sending a challenge and reading
1947 a response while authentication was assuming a zero-terminated challenge
1948 string. It's now changed to take a pointer and a length, to allow for
1949 binary data in such strings.
1950
195122. Added the cyrus_sasl authenticator (code supplied by MBM).
1952
195323. Exim was not respecting finduser_retries when seeking the login of the
1954 uid under which it was called; it was always trying 10 times. (The default
1955 setting of finduser_retries is zero.) Also, it was sleeping after the final
1956 failure, which is pointless.
1957
195824. Implemented tls_on_connect_ports.
1959
196025. Implemented acl_smtp_predata.
1961
196226. If the domain in control=submission is set empty, Exim assumes that the
1963 authenticated id is a complete email address when it generates From: or
1964 Sender: header lines.
1965
196627. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
1967 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
1968 chown and chgrp in /bin and hostname in /usr/bin.
1969
197028. Exim was keeping the "process log" file open after each use, just as it
1971 does for the main log. This opens the possibility of it remaining open for
1972 long periods when the USR1 signal hits a daemon. Occasional processlog
1973 errors were reported, that could have been caused by this. Anyway, it seems
1974 much more sensible not to leave this file open at all, so that is what now
1975 happens.
1976
197729. The long-running daemon process does not normally write to the log once it
1978 has entered its main loop, and it closes the log before doing so. This is
1979 so that log files can straightforwardly be renamed and moved. However,
1980 there are a couple of unusual error situations where the daemon does write
1981 log entries, and I had neglected to close the log afterwards.
1982
198330. The text of an SMTP error response that was received during a remote
1984 delivery was being truncated at 512 bytes. This is too short for some of
1985 the long messages that one sometimes sees. I've increased the limit to
1986 1024.
1987
198831. It is now possible to make retry rules that apply only when a message has a
1989 specific sender, in particular, an empty sender.
1990
199132. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
1992 it possible to be selective about when SMTP synchronization is enforced.
1993
199433. Added "control = caseful_local_part" and "control = "caselower_local_part".
1995
199632. Implemented hosts_connection_nolog.
1997
199833. Added an ACL for QUIT.
1999
200034. Setting "delay_warning=" to disable warnings was not working; it gave a
2001 syntax error.
2002
200335. Added mailbox_size and mailbox_filecount to appendfile.
2004
200536. Added control = no_multiline_responses to ACLs.
2006
200737. There was a bug in the logic of the code that waits for the clock to tick
2008 in the case where the clock went backwards by a substantial amount such
2009 that the microsecond fraction of "now" was more than the microsecond
2010 fraction of "then" (but the whole seconds number was less).
2011
201238. Added support for the libradius Radius client library this is found on
2013 FreeBSD (previously only the radiusclient library was supported).
2014
2015
2016Exim version 4.42
2017-----------------
2018
2019 1. When certain lookups returned multiple values in the form name=value, the
2020 quoting of the values was not always being done properly. Specifically:
2021 (a) If the value started with a double quote, but contained no whitespace,
2022 it was not quoted.
2023 (b) If the value contained whitespace other than a space character (i.e.
2024 tabs or newlines or carriage returns) it was not quoted.
2025 This fix has been applied to the mysql and pgsql lookups by writing a
2026 separate quoting function and calling it from the lookup code. The fix
2027 should probably also be applied to nisplus, ibase and oracle lookups, but
2028 since I cannot test any of those, I have not disturbed their existing code.
2029
2030 2. A hit in the callout cache for a specific address caused a log line with no
2031 reason for rejecting RCPT. Now it says "Previous (cached) callout
2032 verification failure".
2033
2034 3. There was an off-by-one bug in the queryprogram router. An over-long
2035 return line was truncated at 256 instead of 255 characters, thereby
2036 overflowing its buffer with the terminating zero. As well as fixing this, I
2037 have increased the buffer size to 1024 (and made a note to document this).
2038
2039 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
2040 when Exim is waiting for an SMTP response from a remote server, Exim
2041 restarts its select() call on the socket, thereby resetting its timeout.
2042 This is not a problem when such interrupts are rare. Somebody set up a cron
2043 job to run exiwhat every 2 minutes, which is less than the normal select()
2044 timeout (5 or 10 minutes). This meant that the select() timeout never
2045 kicked in because it was always reset. I have fixed this by comparing the
2046 time when an interrupt arrives with the time at the start of the first call
2047 to select(). If more time than the timeout has elapsed, the interrupt is
2048 treated as a timeout.
2049
2050 5. Some internal re-factoring in preparation for the addition of Sieve
2051 extensions (by MH). In particular, the "personal" test is moved to a
2052 separate function, and given an option for scanning Cc: and Bcc: (which is
2053 not set for Exim filters).
2054
2055 6. When Exim created an email address using the login of the caller as the
2056 local part (e.g. when creating a From: or Sender: header line), it was not
2057 quoting the local part when it contained special characters such as @.
2058
2059 7. Installed new OpenBSD configuration files.
2060
2061 8. Reworded some messages for syntax errors in "and" and "or" conditions to
2062 try to make them clearer.
2063
2064 9. Callout options, other than the timeout value, were being ignored when
2065 verifying sender addresses in header lines. For example, when using
2066
2067 verify = header_sender/callout=no_cache
2068
2069 the cache was (incorrectly) being used.
2070
207110. Added a missing instance of ${EXE} to the exim_install script; this affects
2072 only the Cygwin environment.
2073
207411. When return_path_on_delivery was set as a log selector, if different remote
2075 addresses in the same message used different return paths and parallel
2076 remote delivery occurred, the wrong values would sometimes be logged.
2077 (Whenever a remote delivery process finished, the return path value from
2078 the most recently started remote delivery process was logged.)
2079
208012. RFC 3848 specifies standard names for the "with" phrase in Received: header
2081 lines when AUTH and/or TLS are in use. This is the "received protocol"
2082 field. Exim used to use "asmtp" for authenticated SMTP, without any
2083 indication (in the protocol name) for TLS use. Now it follows the RFC and
2084 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
2085 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
2086 names appear in log lines as well as in Received: header lines.
2087
208813. Installed MH's patches for Sieve to add the "copy" and "vacation"
2089 extensions, and comparison tests, and to fix some bugs.
2090
209114. Changes to the "personal" filter test:
2092
2093 (1) The test was buggy in that it was just doing the equivalent of
2094 "contains" tests on header lines. For example, if a user's address was
2095 anne@some.where, the "personal" test would incorrectly be true for
2096
2097 To: susanne@some.where
2098
2099 This test is now done by extracting each address from the header in turn,
2100 and checking the entire address. Other tests that are part of "personal"
2101 are now done using regular expressions (for example, to check local parts
2102 of addresses in From: header lines).
2103
2104 (2) The list of non-personal local parts in From: addresses has been
2105 extended to include "listserv", "majordomo", "*-request", and "owner-*",
2106 taken from the Sieve specification recommendations.
2107
2108 (3) If the message contains any header line starting with "List-" it is
2109 treated as non-personal.
2110
2111 (4) The test for "circular" in the Subject: header line has been removed
2112 because it now seems ill-conceived.
2113
211415. Minor typos in src/EDITME comments corrected.
2115
211616. Installed latest exipick from John Jetmore.
2117
211817. If headers_add on a router specified a text string that was too long for
2119 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
2120 of string_sprintf() is now avoided.
2121
212218. $message_body_size was not set (it was always zero) when running the DATA
2123 ACL and the local_scan() function.
2124
212519. For the "mail" command in an Exim filter, no default was being set for
2126 the once_repeat time, causing a random time value to be used if "once" was
2127 specified. (If the value happened to be <= 0, no repeat happened.) The
2128 default is now 0s, meaning "never repeat". The "vacation" command was OK
2129 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
2130 (I found it when inspecting the code).
2131
213220. There is now an overall timeout for performing a callout verification. It
2133 defaults to 4 times the callout timeout, which applies to individual SMTP
2134 commands during the callout. The overall timeout applies when there is more
2135 than one host that can be tried. The timeout is checked before trying the
2136 next host. This prevents very long delays if there are a large number of
2137 hosts and all are timing out (e.g. when the network connections are timing
2138 out). The value of the overall timeout can be changed by specifying an
2139 additional sub-option for "callout", called "maxwait". For example:
2140
2141 verify = sender/callout=5s,maxwait=20s
2142
214321. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
2144 to the end before writing, but this should make it even safer).
2145
214622. Exim was forgetting that it had advertised PIPELINING for the second and
2147 subsequent messages on an SMTP connection. It was also not resetting its
2148 memory on STARTTLS and an internal HELO.
2149
215023. When Exim logs an SMTP synchronization error within a session, it now
2151 records whether PIPELINING has been advertised or not.
2152
215324. Added 3 instances of "(long int)" casts to time_t variables that were being
2154 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
2155 rather than long int.
2156
215725. Installed the latest Cygwin configuration files from the Cygwin maintainer.
2158
215926. Added the never_mail option to autoreply.
2160
2161
2162Exim version 4.41
2163-----------------
2164
2165 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
2166 crash if the getsockname() call failed; this can happen if a connection is
2167 closed very soon after it is established. The problem was simply in the
2168 order in which certain operations were done, causing Exim to try to write
2169 to the SMTP stream before it had set up the file descriptor. The bug has
2170 been fixed by making things happen in the correct order.
2171
2172
2173Exim version 4.40
2174-----------------
2175
2176 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
2177 before the connection was closed, thus losing the rejection response.
2178
2179 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
2180 some early Solaris releases, but causes trouble in current releases where
2181 socklen_t is defined.
2182
2183 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
2184 always exist.
2185
2186 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
2187 configured.
2188
2189 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
2190 that releasing the top of it at the end releases what was used for sub-
2191 expansions (unless the block got too big). However, discard this block if
2192 the first thing is a variable or header, so that we can use its block when
2193 it is dynamic (useful for very large $message_headers, for example).
2194
2195 6. Lookups now cache *every* query, not just the most recent. A new, separate
2196 store pool is used for this. It can be recovered when all lookup caches are
2197 flushed. Lookups now release memory at the end of their result strings.
2198 This has involved some general refactoring of the lookup sources.
2199
2200 7. Some code has been added to the store_xxx() functions to reduce the amount
2201 of flapping under certain conditions.
2202
2203 8. log_incoming_interface used to affect only the <= reception log lines. Now
2204 it causes the local interface and port to be added to several more SMTP log
2205 lines, for example "SMTP connection from", and rejection lines.
2206
2207 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
2208
220910. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
2210
221111. If $host_data was set by virtue of a hosts lookup in an ACL, its value
2212 could be overwritten at the end of the current message (or the start of a
2213 new message if it was set in a HELO ACL). The value is now preserved for
2214 the duration of the SMTP connection.
2215
221612. If a transport had a headers_rewrite setting, and a matching header line
2217 contained an unqualified address, that address was qualified, even if it
2218 did not match any rewriting rules. The underlying bug was that the values
2219 of the flags that permit the existence of unqualified sender and recipient
2220 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
2221 non-local messages, and by -bnq for local messages) were not being
2222 preserved with the message after it was received.
2223
222413. When Exim was logging an SMTP synchronization error, it could sometimes log
2225 "next input=" as part of the text comprising the host identity instead of
2226 the correct text. The code was using the same buffer for two different
2227 strings. However, depending on which order the printing function evaluated
2228 its arguments, the bug did not always show up. Under Linux, for example, my
2229 test suite worked just fine.
2230
223114. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
2232 doesn't work with some older versions of Perl. It has been changed to "my",
2233 which in any case is probably the better facility to use.
2234
223515. A really picky compiler found some instances of statements for creating
2236 error messages that either had too many or two few arguments for the format
2237 string.
2238
223916. The size of the buffer for calls to the DNS resolver has been increased
2240 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
2241 for addresses that have a lot of PTR records. This alleviates a problem; it
2242 does not fully solve it.
2243
224417. A dnsdb lookup for PTR records that receives more data than will fit in the
2245 buffer now truncates the list and logs the incident, which is the same
2246 action as happens when Exim is looking up a host name and its aliases.
2247 Previously in this situation something unpredictable would happen;
2248 sometimes it was "internal error: store_reset failed".
2249
225018. If a server dropped the connection unexpectedly when an Exim client was
2251 using GnuTLS and trying to read a response, the client delivery process
2252 crashed while trying to generate an error log message.
2253
225419. If a "warn" verb in an ACL added multiple headers to a message in a single
2255 string, for example:
2256
2257 warn message = H1: something\nH2: something
2258
2259 the text was added as a single header line from Exim's point of view
2260 though it ended up OK in the delivered message. However, searching for the
2261 second and subsequent header lines using $h_h2: did not work. This has been
2262 fixed. Similarly, if a system filter added multiple headers in this way,
2263 the routers could not see them.
2264
226520. Expanded the error message when iplsearch is called with an invalid key to
2266 suggest using net-iplsearch in a host list.
2267
226821. When running tests using -bh, any delays imposed by "delay" modifiers in
2269 ACLs are no longer actually imposed (and a message to that effect is
2270 output).
2271
227222. If a "gecos" field in a passwd entry contained escaped characters, in
2273 particular, if it contained a \" sequence, Exim got it wrong when building
2274 a From: or a Sender: header from that name. A second bug also caused
2275 incorrect handling when an unquoted " was present following a character
2276 that needed quoting.
2277
227823. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
2279 was not being matched caselessly.
2280
228124. Arranged for all hyphens in the exim.8 source to be escaped with
2282 backslashes.
2283
228425. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
2285 properly. Recipient callout cache records were still being keyed to include
2286 the sender, even when use_sender was set false. This led to far more
2287 callouts that were necessary. The sender is no longer included in the key
2288 when use_sender is false.
2289
229026. Added "control = submission" modifier to ACLs.
2291
229227. Added the ${base62d: operator to decode base 62 numbers.
2293
229428. dnsdb lookups can now access SRV records.
2295
229629. CONFIGURE_OWNER can be set at build time to define an alternative owner for
2297 the configuration file.
2298
229930. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
2300 (-v) mode. This makes the output for a verbose queue run more intelligible.
2301
230231. Added a use_postmaster feature to recipient callouts.
2303
230432. Added the $body_zerocount variable, containing the number of binary zero
2305 bytes in the message body.
2306
230733. The time of last modification of the "new" subdirectory is now used as the
2308 "mailbox time last read" when there is a quota error for a maildir
2309 delivery.
2310
231134. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
2312
231335. Added +ignore_unknown as a special item in host lists.
2314
231536. Code for decoding IPv6 addresses in host lists is now included, even if
2316 IPv6 support is not being compiled. This fixes a bug in which an IPv6
2317 address was recognized as an IP address, but was then not correctly decoded
2318 into binary, causing unexpected and incorrect effects when compared with
2319 another IP address.
2320
2321
2322Exim version 4.34
2323-----------------
2324
2325 1. Very minor rewording of debugging text in manualroute to say "list of
2326 hosts" instead of "hostlist".
2327
2328 2. If verify=header_syntax was set, and a header line with an unqualified
2329 address (no domain) and a large number of spaces between the end of the
2330 name and the colon was received, the reception process suffered a buffer
2331 overflow, and (when I tested it) crashed. This was caused by some obsolete
2332 code that should have been removed. The fix is to remove it!
2333
2334 3. When running in the test harness, delay a bit after writing a bounce
2335 message to get a bit more predictability in the log output.
2336
2337 4. Added a call to search_tidyup() just before forking a reception process. In
2338 theory, someone could use a lookup in the expansion of smtp_accept_max_
2339 per_host which, without the tidyup, could leave open a database connection.
2340
2341 5. Added the variables $recipient_data and $sender_data which get set from a
2342 lookup success in an ACL "recipients" or "senders" condition, or a router
2343 "senders" option, similar to $domain_data and $local_part_data.
2344
2345 6. Moved the writing of debug_print from before to after the "senders" test
2346 for routers.
2347
2348 7. Change 4.31/66 (moving the time when the Received: is generated) caused
2349 problems for message scanning, either using a data ACL, or using
2350 local_scan() because the Received: header was not generated till after they
2351 were called (in order to set the time as the time of reception completion).
2352 I have revised the way this works. The header is now generated after the
2353 body is received, but before the ACL or local_scan() are called. After they
2354 are run, the timestamp in the header is updated.
2355
2356
2357Exim version 4.33
2358-----------------
2359
2360 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
2361 before starting a queue runner without re-exec. This happened only when
2362 deliver_drop_privilege was set or when the Exim user was set to root. The
2363 effect of the bug was that timeouts during subsequent deliveries caused
2364 crashes instead of being properly handled. The handler is now left at its
2365 default (and expected) setting.
2366
2367 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
2368 message, again when deliver_drop_privilege is set or Exim is run as root.
2369 The bug described in (1) was not present in this case, but the tidying up
2370 of the other signals was missing. I have made the two cases consistent.
2371
2372 3. The ignore_target_hosts setting on a manualroute router was being ignored
2373 for hosts that were looked up using the /MX notation.
2374
2375 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
2376 in domain lists.
2377
2378 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
2379 operated on the sender address. After changing the $sender_address to <>
2380 for the sender address verify, Exim was re-instated it as the original
2381 (before rewriting) address, but remembering that it had rewritten it, so it
2382 wasn't rewriting it again. This bug also had the effect of breaking the
2383 sender address verification caching when the sender address was rewritten.
2384
2385 6. The ignore_target_hosts option was being ignored by the ipliteral router.
2386 This has been changed so that if the ip literal address matches
2387 ignore_target_hosts, the router declines.
2388
2389 7. Added expansion conditions match_domain, match_address, and match_local_
2390 part (NOT match_host).
2391
2392 8. The placeholder for the Received: header didn't have a length field set.
2393
2394 9. Added code to Exim itself and to exim_lock to test for a specific race
2395 condition that could lead to file corruption when using MBX delivery. The
2396 issue is with the lockfile that is created in /tmp. If this file is removed
2397 after a process has opened it but before that process has acquired a lock,
2398 there is the potential for a second process to recreate the file and also
2399 acquire a lock. This could lead to two Exim processes writing to the file
2400 at the same time. The added code performs the same test as UW imapd; it
2401 checks after acquiring the lock that its file descriptor still refers to
2402 the same named file.
2403
240410. The buffer for building added header lines was of fixed size, 8192 bytes.
2405 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
2406 when Exim is built.
2407
240811. Added the smtp_active_hostname option. If used, this will typically be made
2409 to depend on the incoming interface address. Because $interface_address is
2410 not set up until the daemon has forked a reception process, error responses
2411 that can happen earlier (such as "too many connections") no longer contain
2412 a host name.
2413
241412. If an expansion in a condition on a "warn" statement fails because a lookup
2415 defers, the "warn" statement is abandoned, and the next ACL statement is
2416 processed. Previously this caused the whole ACL to be aborted.
2417
241813. Added the iplsearch lookup type.
2419
242014. Added ident_timeout as a log selector.
2421
242215. Added tls_certificate_verified as a log selector.
2423
242416. Added a global option tls_require_ciphers (compare the smtp transport
2425 option of the same name). This controls incoming TLS connections.
2426
242717. I finally figured out how to make tls_require_ciphers do a similar thing
2428 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
2429 before starting the TLS session.
2430
243118. Tabs are now shown as \t in -bP output.
2432
243319. If the log selector return_path_on_delivery was set, Exim crashed when
2434 bouncing a message because it had too many Received: header lines.
2435
243620. If two routers both had headers_remove settings, and the first one included
2437 a superfluous trailing colon, the final name in the first list and the
2438 first name in the second list were incorrectly joined into one item (with a
2439 colon in the middle).
2440
2441
2442Exim version 4.32
2443-----------------
2444
2445 1. Added -C and -D options to the exinext utility, mainly to make it easier
2446 to include in the automated testing, but these could be helpful when
2447 multiple configurations are in use.
2448
2449 2. The exinext utility was not formatting the output nicely when there was
2450 an alternate port involved in the retry record key, nor when there was a
2451 message id as well (for retries that were specific to a specific message
2452 and a specific host). It was also confused by IPv6 addresses, because of
2453 the additional colons they contain. I have fixed the IPv4 problem, and
2454 patched it up to do a reasonable job for IPv6.
2455
2456 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
2457 delivery, the log line now contains "pipelined" if PIPELINING was used.
2458
2459 4. An SMTP transport process used to panic and die if the bind() call to set
2460 an explicit outgoing interface failed. This has been changed; it is now
2461 treated in the same way as a connect() failure.
2462
2463 5. A reference to $sender_host_name in the part of a conditional expansion
2464 that was being skipped was still causing a DNS lookup. This no longer
2465 occurs.
2466
2467 6. The def: expansion condition was not recognizing references to header lines
2468 that used bh_ and bheader_.
2469
2470 7. Added the _cache feature to named lists.
2471
2472 8. The code for checking quota_filecount in the appendfile transport was
2473 allowing one more file than it should have been.
2474
2475 9. For compatibility with Sendmail, the command line option
2476
2477 -prval:sval
2478
2479 is equivalent to
2480
2481 -oMr rval -oMs sval
2482
2483 and sets the incoming protocol and host name (for trusted callers). The
2484 host name and its colon can be omitted when only the protocol is to be set.
2485 Note the Exim already has two private options, -pd and -ps, that refer to
2486 embedded Perl. It is therefore impossible to set a protocol value of "d" or
2487 "s", but I don't think that's a major issue.
2488
248910. A number of refactoring changes to the code, none of which should affect
2490 Exim's behaviour:
2491
2492 (a) The number of logging options was getting close to filling up the
2493 32-bit word that was used as a bit map. I have split them into two classes:
2494 those that are passed in the argument to log_write(), and those that are
2495 only ever tested independently outside of that function. These are now in
2496 separate 32-bit words, so there is plenty of room for expansion again.
2497 There is no change in the user interface or the logging behaviour.
2498
2499 (b) When building, for example, log lines, the code previously used a
2500 macro that called string_cat() twice, in order to add two strings. This is
2501 not really sufficiently general. Furthermore, there was one instance where
2502 it was actually wrong because one of the argument was used twice, and in
2503 one call a function was used. (As it happened, calling the function twice
2504 did not affect the overall behaviour.) The macro has been replaced by a
2505 function that can join an arbitrary number of extra strings onto a growing
2506 string.
2507
2508 (c) The code for expansion conditions now uses a table and a binary chop
2509 instead of a serial search (which was left over from when there were very
2510 few conditions). Also, it now recognizes conditions like "pam" even when
2511 the relevant support is not compiled in: a suitably worded error message is
2512 given if an attempt is made to use such a condition.
2513
251411. Added ${time_interval:xxxxx}.
2515
251612. A bug was causing one of the ddress fields not to be passed back correctly
2517 from remote delivery subprocesses. The field in question was not being
2518 subsequently used, so this caused to problems in practice.
2519
252013. Added new log selectors queue_time and deliver_time.
2521
252214. Might have fixed a bug in maildirsizefile handling that threw up
2523 "unexpected character" debug warnings, and recalculated the data
2524 unnecessarily. In any case, I expanded the warning message to give more
2525 information.
2526
252715. Added the message "Restricted characters in address" to the statements in
2528 the default ACL that block characters like @ and % in local parts.
2529
253016. Change 71 for release 4.31 proved to be much less benign that I imagined.
2531 Three changes have been made:
2532
2533 (a) There was a serious bug; a negative response to MAIL caused the whole
2534 recipient domain to be cached as invalid, thereby blocking all messages
2535 to all local parts at the same domain, from all senders. This bug has
2536 been fixed. The domain is no longer cached after a negative response to
2537 MAIL if the sender used is not empty.
2538
2539 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
2540 been restored.
2541
2542 (c) A new callout option, "use_sender" has been added for people who want
2543 the modified behaviour.
2544
2545
2546Exim version 4.31
2547-----------------
2548
2549 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
2550 Larry Rosenman.
2551
2552 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
2553 indeed breaks things for older releases.
2554
2555 3. Added additional logging to the case where there is a problem reading data
2556 from a filter that is running in a subprocess using a pipe, in order to
2557 try to track down a specific problem.
2558
2559 4. Testing facility fudge: when running in the test harness and attempting
2560 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
2561 getting "No route to host". Convert this to a timeout.
2562
2563 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
2564 warning.
2565
2566 6. Some OS don't have socklen_t but use size_t instead. This affects the
2567 fifth argument of getsockopt() amongst other things. This is now
2568 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
2569 can be set for individual OS. I have set it for SunOS5, OSF1, and
2570 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
2571 some earlier ones do not.
2572
2573 7. Change 4.30/15 was not doing the test caselessly.
2574
2575 8. The standard form for an IPv6 address literal was being rejected by address
2576 parsing in, for example, MAIL and RCPT commands. An example of this kind of
2577 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
2578 this, as well as the form without the "IPv6" on the front (but only when
2579 address literals are enabled, of course).
2580
2581 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
2582
258310. Exim crashed if a message with an empty sender address specified by -f
2584 encountered a router with an errors_to setting. This could be provoked only
2585 by a command such as
2586
2587 exim -f "" ...
2588
2589 where an empty string was supplied; "<>" did not hit this bug.
2590
259111. Installed PCRE release 4.5.
2592
259312. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
2594 remained set. It is now erased.
2595
259613. exiqgrep wasn't working on MacOS X because it didn't correctly compute
2597 times from message ids (which are base 36 rather than the normal 62).
2598
259914. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
2600 were being counted as actual protocol errors, and logged if the log
2601 selector +smtp_protocol_error was set. One cannot be perfect in this test,
2602 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
2603 and DATA following a set of rejected RCPTs do not count as protocol errors.
2604 In other words, Exim assumes they were pipelined, though this may not
2605 actually be the case. Of course, in all cases the client gets an
2606 appropriate error code.
2607
260815. If a lookup fails in an ACL condition, a message about the failure may
2609 be available; it is used if testing the ACL cannot continue, because most
2610 such messages specify what the cause of the deferral is. However, some
2611 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
2612 that caused an old message to be retained and used if a later statement
2613 caused a defer, replacing the real cause of the deferral.
2614
261516. If an IP address had so many PTR records that the DNS lookup buffer
2616 was not large enough to hold them, Exim could crash while trying to process
2617 the truncated data. It now detects and logs this case.
2618
261917. Further to 4.21/58, another change has been made: if (and only if) the
2620 first line of a message (the first header line) ends with CRLF, a bare LF
2621 in a subsequent header line has a space inserted after it, so as not to
2622 terminate the header.
2623
262418. Refactoring: tidied an ugly bit of code in appendfile that copied data
2625 unnecessarily, used atoi() instead of strtol(), and didn't check the
2626 termination when getting file sizes from file names by regex.
2627
262819. Completely re-implemented the support for maildirsize files, in the light
2629 of a number of problems with the previous contributed implementation
2630 (4.30/29). In particular:
2631
2632 . If the quota is zero, the maildirsize file is maintained, but no quota is
2633 imposed.
2634
2635 . If the maildir directory does not exist, it is created before any attempt
2636 to write a maildirsize file.
2637
2638 . The quota value in the file is just a cache; if the quota is changed in
2639 the transport, the new value overrides.
2640
2641 . A regular expression is available for excluding directories from the
2642 count.
2643
264420. The autoreply transport checks the characters in options that define the
2645 message's headers; it allows continued headers, but it was checking with
2646 isspace() after an embedded newline instead of explicitly looking for a
2647 space or a tab.
2648
264921. If all the "regular" hosts to which an address was routed had passed their
2650 expiry times, and had not reached their retry times, the address was
2651 bounced, even if fallback hosts were defined. Now Exim should go on to try
2652 the fallback hosts.
2653
265422. Increased buffer sizes in the callout code from 1024 to 4096 to match the
2655 equivalent code in the SMTP transport. Some hosts send humungous responses
2656 to HELO/EHLO, more than 1024 it seems.
2657
265823. Refactoring: code in filter.c used (void *) for "any old type" but this
2659 gives compiler warnings in some environments. I've now done it "properly",
2660 using a union.
2661
266224. The replacement for inet_ntoa() that is used with gcc on IRIX systems
2663 (because of problems with the built-in one) was declared to return uschar *
2664 instead of char *, causing compiler failure.
2665
266625. Fixed a file descriptor leak when processing alias/forward files.
2667
266826. Fixed a minor format string issue in dbfn.c.
2669
267027. Typo in exim.c: ("dmbnz" for "dbmnz").
2671
267228. If a filter file refered to $h_xxx or $message_headers, and the headers
2673 contained RFC 2047 "words", Exim's memory could, under certain conditions,
2674 become corrupted.
2675
267629. When a sender address is verified, it is cached, to save repeating the test
2677 when there is more than one recipient in a message. However, when the
2678 verification involves a callout, it is possible for different callout
2679 options to be set for different recipients. It is too complicated to keep
2680 track of this in the cache, so now Exim always runs a verification when a
2681 callout is required, relying on the callout cache for the optimization.
2682 The overhead is duplication of the address routing, but this should not be
2683 too great.
2684
268530. Fixed a bug in callout caching. If a RCPT command caused the sender address
2686 to be verified with callout=postmaster, and the main callout worked but the
2687 postmaster check failed, the verification correctly failed. However, if a
2688 subsequent RCPT command asked for sender verification *without* the
2689 postmaster check, incorrect caching caused this verification also to fail,
2690 incorrectly.
2691
269231. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
2693 it was not caching the DNS options (qualify_single, search_parents) that
2694 were used when the lookup failed. A subsequent lookup with different
2695 options therefore always gave the same answer, though there were cases
2696 where it should not have. (Example: a "domains = !$mx_any" option on a
2697 dnslookup router: the "domains" option is always processed without any
2698 widening, but the router might have qualify_single set.) Now Exim uses the
2699 cached value only when the same options are set.
2700
270132. Added John Jetmore's "exipick" utility to the distribution.
2702
270333. GnuTLS: When an attempt to start a TLS session fails for any reason other
2704 than a timeout (e.g. a certificate is required, and is not provided), an
2705 Exim server now closes the connection immediately. Previously it waited for
2706 the client to close - but if the client is SSL, it seems that they each
2707 wait for each other, leading to a delay before one of them times out.
2708
270934: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
2710 maintained 0.8.x compatibility because I don't think many are using it, and
2711 it is clearly obsolete.
2712
271335. Added TLS support for CRLs: a tls_crl global option and one for the smtp
2714 transport.
2715
271636. OpenSSL: $tls_certificate_verified was being set to 1 even if the
2717 client certificate was expired. A simple patch fixes this, though I don't
2718 understand the full logic of why the verify callback is called multiple
2719 times.
2720
272137. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
2722 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
2723 which causes problems with some clients (such as the Certicom SSL Plus
2724 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
2725 disables the coutermeasure allowing Eudora to connect."
2726
272738. Exim was not checking that a write() to a log file succeeded. This could
2728 lead to Bad Things if a log got too big, in particular if it hit a file
2729 size limit. Exim now panics and dies if it cannot write to a log file, just
2730 as it does if it cannot open a log file.
2731
273239. Modified OS/Makefile-Linux so that it now contains
2733
2734 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
2735
2736 The two -D definitions ensure that Exim is compiled with large file
2737 support, which makes it possible to handle log files that are bigger than
2738 2^31.
2739
274040. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
2741 instance) a domain was checked against a named list that involved a lookup,
2742 causing $domain_data to be set, then another domain was checked against the
2743 same list, then the first domain was re-checked, the value of $domain_data
2744 after the final check could be wrong. In particular, if the second check
2745 failed, it could be set empty. This bug probably also applied to
2746 $localpart_data.
2747
274841. The strip_trailing_dot option was not being applied to the address given
2749 with the -f command-line option.
2750
275142. The code for reading a message's header from the spool was incrementing
2752 $received_count, but never initializing it. This meant that the value was
2753 incorrect (doubled) while delivering a message in the same process in which
2754 it was received. In the most common configuration of Exim, this never
2755 happens - a fresh exec is done - but it can happen when
2756 deliver_drop_privilege is set.
2757
275843. When Exim logs an SMTP synchronization error - client data sent too soon -
2759 it now includes up to 150 characters of the unexpected data in the log
2760 line.
2761
276244. The exim_dbmbuild utility uses fixed size buffers for reading input lines
2763 and building data strings. The size of both of these buffers was 10 000
2764 bytes - far larger than anybody would *ever* want, thought I. Needless to
2765 say, somebody hit the limit. I have increased the maximum line length to
2766 20 000 and the maximum data length of concatenated lines to 100 000. I have
2767 also fixed two bugs, because there was no checking on these buffers. Tsk,
2768 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
2769 buffer is too small.
2770
277145. The exim_dbmbuild utility did not support quoted keys, as Exim does in
2772 lsearch lookups. Now it does.
2773
277446. When parsing a route_list item in a manualroute router, a fixed-length
2775 buffer was used for the list of hosts. I made this 1024 bytes long,
2776 thinking that nobody would ever have a list of hosts that long. Wrong.
2777 Somebody had a whole pile of complicated expansion conditions, and the
2778 string was silently truncated, leading to an expansion error. It turns out
2779 that it is easier to change to an unlimited length (owing to other changes
2780 that have happened since this code was originally written) than to build
2781 structure for giving a limitation error. The length of the item that
2782 expands into the list of hosts is now unlimited.
2783
278447. The lsearch lookup could not handle data where the length of text line was
2785 more than 4095 characters. Such lines were truncated, leading to shortened
2786 data being returned. It should now handle lines of any length.
2787
278848. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
2789 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
2790 ACL").
2791
279249. Cosmetic tidy to scripts like exicyclog that are generated by globally
2793 replacing strings such as BIN_DIRECTORY in a source file: the replacement
2794 no longer happens in comment lines. A list of replacements is now placed
2795 at the head of all of the source files, except those whose only change is
2796 to replace PERL_COMMAND in the very first #! line.
2797
279850. Replaced the slow insertion sort in queue.c, for sorting the list of
2799 messages on the queue, with a bottom-up merge sort, using code contributed
2800 by Michael Haardt. This should make operations like -bp somewhat faster on
2801 large queues. It won't affect queue runners, except when queue_run_in_order
2802 is set.
2803
280451. Installed eximstats 1.31 in the distribution.
2805
280652. Added support for SRV lookups to the dnslookup router.
2807
280853. If an ACL referred to $message_body or $message_body_end, the value was not
2809 reset for any messages that followed in the same SMTP session.
2810
281154. The store-handling optimization for building very long strings was not
2812 differentiating between the different store pools. I don't think this
2813 actually made any difference in practice, but I've tidied it.
2814
281555. While running the routers to verify a sender address, $sender_address
2816 was still set to the sender address. This is wrong, because when routing to
2817 send a bounce to the sender, it would be empty. Therefore, I have changed
2818 it so that, while verifying a sender address, $sender_address is set to <>.
2819 (There is no change to what happens when verifying a recipient address.)
2820
282156. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
2822 target A or AAAA records (if not already returned) without resetting the
2823 qualify_single or search_parents options of the DNS resolver. These are
2824 inappropriate in this case because the targets of MX and SRV records must
2825 be FQDNs. A broken DNS record could cause trouble if it happened to have a
2826 target that, when qualified, matched something in the local domain. These
2827 two options are now turned off when doing these lookups.
2828
282957. It seems that at least some releases of Reiserfs (which does not have the
2830 concept of a fixed number of inodes) returns zero and not -1 for the
2831 number of available inodes. This interacted badly with check_spool_inodes,
2832 which assumed that -1 was the "no such thing" setting. What I have done is
2833 to check that the total number of inodes is greater than zero before doing
2834 the test of how many are available.
2835
283658. When a "warn" ACL statement has a log_message modifier, the message is
2837 remembered, and not repeated. This is to avoid a lot of repetition when a
2838 message has many recipients that cause the same warning to be written.
2839 Howewer, Exim was preserving the list of already written lines for an
2840 entire SMTP session, which doesn't seem right. The memory is now reset if a
2841 new message is started.
2842
284359. The "rewrite" debugging flag was not showing the result of rewriting in the
2844 debugging output unless log_rewrite was also set.
2845
284660. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
2847 of (int)(handle) when we know that handle contains (void *)(-1).
2848
284961. The Exim daemon panic-logs an error return when it closes the incoming
2850 connection. However "connection reset by peer" seems to be common, and
2851 isn't really an error worthy of noting specially, so that particular error
2852 is no long logged.
2853
285462. When Exim is trying to find all the local interfaces, it used to panic and
2855 die if the ioctl to get the interface flags failed. However, it seems that
2856 on at least one OS (Solaris 9) it is possible to have an interface that is
2857 included in the list of interfaces, but for which you get a failure error
2858 for this call. This happens when the interface is not "plumbed" into a
2859 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
2860 failure of the "get flags" call assumes that the interface is down.
2861
286263. Added a ${eval10: operator, which assumes all numbers are decimal. This
2863 makes life easier for people who are doing arithmetic on fields extracted
2864 from dates, where you often get leading zeros that should not be
2865 interpreted as octal.
2866
286764. Added qualify_domain to the redirect router, to override the global
2868 setting.
2869
287065. If a pathologically long header line contained very many addresses (the
2871 report of this problem mentioned 10 000) and each of them was rewritten,
2872 Exim could use up a very large amount of memory. (It kept on making new
2873 copies of the header line as it rewrote, and never released the old ones.)
2874 At the expense of a bit more processing, the header rewriting function has
2875 been changed so that it no longer eats memory in this way.
2876
287766. The generation of the Received: header has been moved from the time that a
2878 message starts to be received, to the time that it finishes. The timestamp
2879 in the Received: header should now be very close to that of the <= log
2880 line. There are two side-effects of this change:
2881
2882 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
2883 logged header lines no longer include the local Received: line, because
2884 it has not yet been created. The same applies to a copy of the message
2885 that is returned to a non-SMTP sender when a message is rejected.
2886
2887 (b) When a filter file is tested using -bf, no additional Received: header
2888 is added to the test message. After some thought, I decided that this
2889 is a bug fix.
2890
2891 This change does not affect the value of $received_for. It is still set
2892 after address rewriting, but before local_scan() is called.
2893
289467. Installed the latest Cygwin-specific files from the Cygwin maintainer.
2895
289668. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
2897 gave an unhelpful panic error message, and a defer error. I have managed to
2898 change this behaviour so that it now rejects any supplied certificate,
2899 which seems right, as the list of acceptable certificates is empty.
2900
290169. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
2902 gave an unhelpful defer error. I have not managed to make this reject any
2903 supplied certificates, but the error message it gives is "no certificate
2904 supplied", which is not helpful.
2905
290670. exigrep's output now also includes lines that are not associated with any
2907 message, but which match the given pattern. Implemented by a patch from
2908 Martin Sluka, which also tidied up the Perl a bit.
2909
291071. Recipient callout verification, like sender verification, was using <> in
2911 the MAIL FROM command. This isn't really the right thing, since the actual
2912 sender may affect whether the remote host accepts the recipient or not. I
2913 have changed it to use the actual sender in the callout; this means that
2914 the cache record is now keyed on a recipient/sender pair, not just the
2915 recipient address. There doesn't seem to be a real danger of callout loops,
2916 since a callout by the remote host to check the sender would use <>.
2917 [SEE ABOVE: changed after hitting problems.]
2918
291972. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
2920 temporary errors. However, in the case of such a code being given after
2921 the end of a data transmission (i.e. after ".") Exim was failing to write
2922 a retry record for the message. (Yes, there was some broken host that was
2923 actually sending 8xx at this point.)
2924
292573. An unknown lookup type in a host list could cause Exim to panic-die when
2926 the list was checked. (An example that provoked this was putting <; in the
2927 middle of a list instead of at the start.) If this happened during a DATA
2928 ACL check, a -D file could be left lying around. This kind of configuration
2929 error no longer causes Exim to die; instead it causes a defer errror. The
2930 incident is still logged to the main and panic logs.
2931
293274. Buglet left over from Exim 3 conversion. The message "too many messages
2933 in one connection" was written to the rejectlog but not the mainlog, except
2934 when address rewriting (yes!) was being logged.
2935
293675. Added write_rejectlog option.
2937
293876. When a system filter was run not as root (that is, when system_filter_user
2939 was set), the values of the $n variables were not being returned to the
2940 main process; thus, they were not subsequently available in the $sn
2941 variables.
2942
294377. Added +return_path_on_delivery log selector.
2944
294578. A connection timeout was being treated differently from recipients deferred
2946 when testing hosts_max_try with a message that was older than the host's
2947 retry timeout. (The host should not be counted, thus allowing all hosts to
2948 be tried at least once before bouncing.) This may have been the cause of an
2949 occasionally reported bug whereby a message would remain on the queue
2950 longer than the retry timeout, but would be bounced if a delivery was
2951 forced. I say "may" because I never totally pinned down the problem;
2952 setting up timeout/retry tests is difficult. See also the next item.
2953
295479. The ultimate address timeout was not being applied to errors that involved
2955 a combination of host plus message (for example, a timeout on a MAIL
2956 command). When an address resolved to a number of possible hosts, and they
2957 were not all tried for each delivery (e.g. because of hosts_max_try), a
2958 message could remain on the queue longer than the retry timeout.
2959
296080. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
2961 Haardt.
2962
296381. Fixed an obscure SMTP outgoing bug which required at least the following
2964 conditions: (a) there was another message waiting for the same server;
2965 (b) the server returned 5xx to all RCPT commands in the first message so
2966 that the message was not completed; (c) the server dropped the connection
2967 or gave a negative response to the RSET that Exim sends to abort the
2968 transaction. The observed case was a dropped connection after DATA that had
2969 been sent in pipelining mode. That is, the server had advertised PIPELINING
2970 but was not implementing it correctly. The effect of the bug was incorrect
2971 behaviour, such as trying another host, and this could lead to a crash.
2972
2973
2974Exim version 4.30
2975-----------------
2976
2977 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
2978 and daemon.c were passed as pointers to ints; they should have been
2979 pointers to socklen_t variables (which are typically unsigned ints).
2980
2981 2. Some signed/unsigned type warnings in the os.c file for Linux have been
2982 fixed.
2983
2984 3. Fixed a really odd bug that affected only the testing scheme; patching a
2985 certain fixed string in the binary changed the value of another string that
2986 happened to be identical to the end of the original first string.
2987
2988 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
2989 name", it returns that address as the IP address. On some operating
2990 systems (e.g. Solaris), it also passes back the IP address string as the
2991 "host name". However, on others (e.g. Linux), it passes back an empty
2992 string. Exim wasn't checking for this, and was changing the host name to an
2993 empty string, assuming it had been canonicized.
2994
2995 5. Although rare, it is permitted to have more than one PTR record for a given
2996 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
2997 all the names associated with an address, because they do in Solaris.
2998 However, it seems that they do not in Linux for data that comes from the
2999 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
3000 I found this out when I moved to a new Linux workstation and tried to run
3001 the Exim test suite.
3002
3003 To get round this problem I have changed the code so that it now does its
3004 own call to the DNS to look up PTR records when searching for a host name.
3005 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
3006 addresses that are only in /etc/hosts are still found.
3007
3008 This behaviour is, however, controlled by an option called host_lookup_
3009 order, which defaults to "bydns:byaddr". If people want to use the other
3010 order, or indeed, just use one or the other means of lookup, they can
3011 specify it in this variable.
3012
3013 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
3014 some operating systems, this comes back from gethostbyaddr() as an empty
3015 string, and this is what Exim used to test for. However, it seems that in
3016 other systems, "." is yielded. Exim now tests for this case too.
3017
3018 7. The values of check_spool_space and check_log_space are now held internally
3019 as a number of kilobytes instead of an absolute number of bytes. If a
3020 numbers is specified without 'K' or 'M', it is rounded up to the nearest
3021 kilobyte. This means that much larger values can be stored.
3022
3023 8. Exim monitor: an attempt to get the action menu when not actually pointing
3024 at a message produces an empty menu entitled "No message selected". This
3025 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
3026 no entries in it ("Shell widget menu has zero width and/or height"). So I
3027 have added a single, blank menu entry in this case.
3028
3029 9. Added ${quote_local_part.
3030
303110. MIME decoding is now applied to the contents of Subject: header lines when
3032 they are logged.
3033
303411. Now that a reference to $sender_host_address automatically causes a reverse
3035 lookup to occur if necessary (4.13/18), there is no need to arrange for a
3036 host lookup before query-style lookups in lists that might use this
3037 variable. This has therefore been abolished, and the "net-" prefix is no
3038 longer necessary for query-style lookups.
3039
304012. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
3041 have been a typo for LFLAGS, so it has been changed.
3042
304313. The install script calls Exim with "-C /dev/null" in order to find the
3044 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
3045 to be output. Howeve, since Exim outputs its version number before the
3046 error, it didn't break the script. It just looked ugly. I fixed this by
3047 always allowing "-C /dev/null" if the caller is root.
3048
304914. Ignore overlarge ACL variable number when reading spool file - insurance
3050 against a later release with more variables having written the file.
3051
305215. The standard form for an IPv6 address literal was being rejected by EHLO.
3053 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
3054 this, as well as the form without the "IPv6" on the front.
3055
305616. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
3057 OS/Makefile-Darwin file.
3058
305917. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
3060 only to LDAP libraries that do not have LDAP_OPT_DEREF.
3061
306218. After change 4.21/52, "%ld" was used to format the contents of the $inode
3063 variable. However, some OS use ints for inodes. I've added cast to long int
3064 to get rid of the compiler warning.
3065
306619. I had forgotten to lock out "/../" in configuration file names when
3067 ALT_CONFIG_PREFIX was set.
3068
306920. Routers used for verification do not need to specify transports. However,
3070 if such a router generated a host list, and callout was configured, Exim
3071 crashed, because it could not find a port number from the (non-existent)
3072 transport. It now assumes port 25 in this circumstance.
3073
307421. Added the -t option to exigrep.
3075
307622. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
3077 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
3078 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
3079 them is set, LOOKUP_LSEARCH is forced.
3080
308123. "exim -bV" now outputs a list of lookups that are included in the binary.
3082
308324. Added sender and host information to the "rejected by local_scan()" log
3084 line; previously there was no indication of these.
3085
308625. Added .include_if_exists.
3087
308826. Change 3.952/11 added an explicit directory sync on top of a file sync for
3089 Linux. It turns out that not all file systems support this. Apparently some
3090 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
3091 do it.) To cope with this, the error EINVAL, which means that sync-ing is
3092 not supported on the file descriptor, is now ignored when Exim is trying to
3093 sync a directory. This applies only to Linux.
3094
309527. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
3096
309728. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
3098 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
3099 this.
3100
310129. Added support for maildirsize files from supplied patch (modified a bit).
3102
310330. The use of :fail: followed by an empty string could lead Exim to respond to
3104 sender verification failures with (e.g.):
3105
3106 550 Verification failed for <xxx>
3107 550 Sender verify failed
3108
3109 where the first response line was missing the '-' that indicates it is not
3110 the final line of the response.
3111
311231. The loop for finding the name of the user that called Exim had a hardwired
3113 limit of 10; it now uses the value of finduser_retries, which is used for
3114 all other user lookups.
3115
311632. Added $received_count variable, available in data and not_smtp ACLs, and at
3117 delivery time.
3118
311933. Exim was neglecting to zero errno before one call of strtol() when
3120 expanding a string and expecting an integer value. On some systems this
3121 resulted in spurious "integer overflow" errors. Also, it was casting the
3122 result into an int without checking.
3123
312434. Testing for a connection timeout using "timeout_connect" in the retry rules
3125 did not work. The code looks as if it has *never* worked, though it appears
3126 to have been documented since at least releast 1.62. I have made it work.
3127
312835. The "timeout_DNS" error in retry rules, also documented since at least
3129 1.62, also never worked. As it isn't clear exactly what this means, and
3130 clearly it isn't a major issue, I have abolished the feature by treating it
3131 as "timeout", and writing a warning to the main and panic logs.
3132
313336. The display of retry rules for -brt wasn't always showing the error code
3134 correctly.
3135
313637. Added new error conditions to retry rules: timeout_A, timeout_MX,
3137 timeout_connect_A, timeout_connect_MX.
3138
313938. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
3140 to the empty sender.
3141
314239. The daemon was not analysing the content of -oX till after it had closed
3143 stderr and disconnected from the controlling terminal. This meant that any
3144 syntax errors were only noted on the panic log, and the return code from
3145 the command was 0. By re-arranging the code a little, I've made the
3146 decoding happen first, so such errors now appear on stderr, and the return
3147 code is 1. However, the actual setting up of the sockets still happens in
3148 the disconnected process, so errors there are still only recorded on the
3149 panic log.
3150
315140. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
3152 connections (as happens on some IP stacks) was logged at start up time as
3153 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
3154 it from "IPv6 and IPv4", which means that two separate sockets are being
3155 used.
3156
315741. The debug output for gethostbyname2() or getipnodebyname() failures now
3158 says whether AF_INET or AF_INET6 was passed as an argument.
3159
316042. Exiwhat output was messed up when time zones were included in log
3161 timestamps.
3162
316343. Exiwhat now gives more information about the daemon's listening ports,
3164 and whether -tls-on-connect was used.
3165
316644. The "port" option of the smtp transport is now expanded.
3167
316845. A "message" modifier in a "warn" statement in a non-message ACL was being
3169 silently ignored. Now an error message is written to the main and panic
3170 logs.
3171
317246. There's a new ACL modifier called "logwrite" which writes to a log file
3173 as soon as it is encountered.
3174
317547. Added $local_user_uid and $local_user_gid at routing time.
3176
317748. Exim crashed when trying to verify a sender address that was being
3178 rewritten to "<>".
3179
318049. Exim was recognizing only a space character after ".include". It now also
3181 recognizes a tab character.
3182
318350. Fixed several bugs in the Perl script that creates the exim.8 man page by
3184 extracting the relevant information from the specification. The man page no
3185 longer contains scrambled data for the -d option, and I've added a section
3186 at the front about calling Exim under different names.
3187
318851. Added "extra_headers" argument to the "mail" command in filter files.
3189
319052. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
3191 crash.
3192
319353. Installed eximstats 1.29.
3194
319554. Added transport_filter_timeout as a generic transport option.
3196
319755. Exim no longer adds an empty Bcc: header to messages that have no To: or
3198 Cc: header lines. This was required by RFC 822, but it not required by RFC
3199 2822.
3200
320156. Exim used to add From:, Date:, and Message-Id: header lines to any
3202 incoming messages that did not have them. Now it does so only if the
3203 message originates locally, that is, if there is no associated remote host
3204 address. When Resent- header lines are present, this applies to the Resent-
3205 lines rather than the non-Resent- lines.
3206
320757. Drop incoming SMTP connection after too many syntax or protocol errors. The
3208 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
3209
321058. Messages for configuration errors now include the name of the main
3211 configuration file - useful now that there may be more than one file in a
3212 list (.included file names were always shown).
3213
321459. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
3215 for those rare installations that do not start the daemon as root or run it
3216 setuid root. I've cut out the call to initgroups() if the daemon is not
3217 root at that time.
3218
321960. The Exim user and group can now be bound into the binary as text strings
3220 that are looked up at the start of Exim's processing.
3221
322261. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
3223
322462. Added $mailstore_basename variable.
3225
322663. Installed patch to sieve.c from Michael Haardt.
3227
322864. When Exim failed to open the panic log after failing to open the main log,
3229 the original message it was trying to log was written to stderr and debug
3230 output, but if they were not available (the usual case in production), it
3231 was lost. Now it is written to syslog before the two lines that record the
3232 failures to open the logs.
3233
323465. Users' Exim filters run in subprocesses under the user's uid. It is
3235 possible for a "deliver" command or an alias in a "personal" command to
3236 provoke an address rewrite. If logging of address rewriting is configured,
3237 this fails because the process is not running as root or exim. There may be
3238 a better way of dealing with this, but for the moment (because 4.30 needs
3239 to be released), I have disabled address rewrite logging when running a
3240 filter in a non-root, non-exim process.
3241
3242
3243Exim version 4.24
3244-----------------
3245
3246 1. The buildconfig auxiliary program wasn't quoting the value set for
3247 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
3248 not defined. This bug was masked in 4.22 by the effect that was fixed in
3249 change 4.23/1.
3250
3251 2. Some messages that were rejected after a message id was allocated were
3252 shown as "incomplete" by exigrep. It no longer does this for messages that
3253 are rejected by local_scan() or the DATA or non-SMTP ACLs.
3254
3255 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
3256 have allow_domain_literals set, the ID did not get logged in the <= line.
3257 Domain literals are now always recognized in Message-ID: header lines.
3258
3259 4. The first argument for a ${extract expansion item is the key name or field
3260 number. Leading and trailing spaces in this item were not being ignored,
3261 causing some misleading effects.
3262
3263 5. When deliver_drop_privilege was set, single queue runner processes started
3264 manually (i.e. by the command "exim -q") or by the daemon (which uses the
3265 same command in the process it spins off) were not dropping privilege.
3266
3267 6. When the daemon running as "exim" started a queue runner, it always
3268 re-executed Exim in the spun-off process. This is a waste of effort when
3269 deliver_drop_privilege is set. The new process now just calls the
3270 queue-runner function directly.
3271
3272
3273Exim version 4.23
3274-----------------
3275
3276 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
3277 HEADERS_CHARSET.
3278
3279 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
3280 ignored. Though the use of -oP was forcing the writing of a pid file, it
3281 was always written to the default place.
3282
3283 3. If the message "no IP address found for host xxxx" is generated during
3284 incoming verification, it is now followed by identification of the incoming
3285 connection (so you can more easily find what provoked it).
3286
3287 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
3288
3289 5. Added some features to "harden" Exim a bit more against certain attacks:
3290
3291 (a) There is now a build-time option called FIXED_NEVER_USERS that can
3292 be put in Local/Makefile. This is like the never_users runtime option,
3293 but it cannot be overridden. The default setting is "root".
3294
3295 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
3296 prefix string with which any file named in a -C command line option
3297 must start.
3298
3299 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
3300 is retained for -C and -D only if the caller of Exim is root. Without
3301 it, the exim user may also use -C and -D and retain privilege.
3302
3303 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
3304 command line option is disabled.
3305
3306 6. Macro names set by the -D option must start with an upper case letter, just
3307 like macro names defined in the configuration file.
3308
3309 7. Added "dereference=" facility to LDAP.
3310
3311 8. Two instances of the typo "uknown" in the source files are fixed.
3312
3313 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
3314 the Configure-Makefile script screwed up while processing it.
3315
331610. Incorporated PCRE 4.4.
3317
331811. The SMTP synchronization check was not operating right at the start of an
3319 SMTP session. For example, it could not catch a HELO sent before the client
3320 waited for the greeting. There is now a check for outstanding input at the
3321 point when the greeting is written. Because of the duplex, asynchronous
3322 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
3323 way, but not yet received, when the check is performed.
3324
332512. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
3326 on TCP/IP sockets, because this apparently causes some broken clients to
3327 timeout.
3328
332913. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
3330 unchanged) from the Cygwin maintainer.
3331
333214. The code for -bV that shows what is in the binary showed "mbx" when maildir
3333 was supported instead of testing for mbx. Effectively a typo.
3334
333515. The spa authenticator server code was not checking that the input it
3336 received was valid base64.
3337
333816. The debug output line for the "set" modifier in ACLs was not showing the
3339 name of the variable that was being set.
3340
334117. Code tidy: the variable type "vtype_string" was never used. Removed it.
3342
334318. Previously, a reference to $sender_host_name did not cause a DNS reverse
3344 lookup on its own. Something else was needed to trigger the lookup. For
3345 example, a match in host_lookup or the need for a host name in a host list.
3346 Now, if $sender_host_name is referenced and the host name has not yet been
3347 looked up, a lookup is performed. If the lookup fails, the variable remains
3348 empty, and $host_lookup_failed is set to "1".
3349
335019. Added "eqi" as a case-independent comparison operator.
3351
335220. The saslauthd authentication condition could segfault if neither service
3353 nor realm was specified.
3354
335521. If an overflowing value such as "2048M" was set for message_size_limit, the
3356 error message that was logged was misleading, and incoming SMTP
3357 connections were dropped. The message is now more accurate, and temporary
3358 errors are given to SMTP connections.
3359
336022. In some error situations (such as 21 above) Exim rejects all SMTP commands
3361 (except RSET) with a 421 error, until QUIT is received. However, it was
3362 failing to send a response to QUIT.
3363
336423. The HELO ACL was being run before the code for helo_try_verify_hosts,
3365 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
3366 ACL is now run after the helo_try_verify_hosts code.
3367
336824. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
3369 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
3370 case-independent, so other case variants are also recognized). Apparently
3371 some systems use these upper case variants.
3372
337325. If more than two messages were waiting for the same host, and a transport
3374 filter was specified for the transport, Exim sent two messages over the
3375 same TCP/IP connection, and then failed with "socket operation on non-
3376 socket" when it tried to send the third.
3377
337826. Added Exim::debug_write and Exim::log_write for embedded Perl use.
3379
338027. The extern definition of crypt16() in expand.c was not being excluded when
3381 the OS had its own crypt16() function.
3382
338328. Added bounce_return_body as a new option, and bounce_return_size_limit
3384 as a preferred synonym for return_size_limit, both as an option and as an
3385 expansion variable.
3386
338729. Added LIBS=-liconv to OS/Makefile-OSF1.
3388
338930. Changed the default configuration ACL to relax the local part checking rule
3390 for addresses that are not in any local domains. For these addresses,
3391 slashes and pipe symbols are allowed within local parts, but the sequence
3392 /../ is explicitly forbidden.
3393
339431. SPA server authentication was not clearing the challenge buffer before
3395 using it.
3396
339732. log_message in a "warn" ACL statement was writing to the reject log as
3398 well as to the main log, which contradicts the documentation and doesn't
3399 seem right (because no rejection is happening). So I have stopped it.
3400
340133. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
3402 However, I am unable to do any testing of this.
3403
340434. Fixed an infelicity in the appendfile transport. When checking directories
3405 for a mailbox, to see if any needed to be created, it was accidentally
3406 using path names with one or more superfluous leading slashes; tracing
3407 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
3408
340935. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
3410 contents are added to the log line that is written for every discarded
3411 recipient. (Previously a log_message setting was ignored.)
3412
341336. The ${quote: operator now quotes the string if it is empty.
3414
341537. The install script runs exim in order to find its version number. If for
3416 some reason other than non-existence or emptiness, which it checks, it
3417 could not run './exim', it was installing it with an empty version number,
3418 i.e. as "exim-". This error state is now caught, and the installation is
3419 aborted.
3420
342138. An argument was missing from the function that creates an error message
3422 when Exim fails to connect to the socket for saslauthd authentication.
3423 This could cause Exim to crash, or give a corrupted message.
3424
342539. Added isip, isip4, and isip6 to ${if conditions.
3426
342740. The ACL variables $acl_xx are now saved with the message, and can be
3428 accessed later in routers, transports, and filters.
3429
343041. The new lookup type nwildlsearch is like wildlsearch, except that the key
3431 strings in the file are not string-expanded.
3432
343342. If a MAIL command specified a SIZE value that was too large to fit into an
3434 int variable, the check against message_size_limit failed. Such values are
3435 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
3436 day this will have to be increased, but I don't think I want to be around
3437 when emails are that large.
3438
3439
3440
3441Exim version 4.22
3442-----------------
3443
3444 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
3445 iconv() is not standard in FreeBSD.
3446
3447 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
3448 IPv6 enabled. The observed symptom was a segmentation fault on return from
3449 the function os_common_find_running_interfaces() in src/os.c.
3450
3451 3. In the check_special_case() function in daemon.c I had used "errno" as an
3452 argument name, which causes warnings on some systems. This was basically a
3453 typo, since it was named "eno" in the comments!
3454
3455 4. The code that waits for the clock to tick (at a resolution of some fraction
3456 of a second) so as to ensure message-id uniqueness was always waiting for
3457 at least one whole tick, when it could have waited for less. [This is
3458 almost certainly not relevant at current processor speeds, where it is
3459 unlikely to ever wait at all. But we try to future-proof.]
3460
3461 5. The function that sleeps for a time interval that includes fractions of a
3462 second contained a race. It did not block SIGALRM between setting the
3463 timer, and suspending (a couple of lines later). If the interval was short
3464 and the sigsuspend() was delayed until after it had expired, the suspension
3465 never ended. On busy systems this could lead to processes getting stuck for
3466 ever.
3467
3468 6. Some uncommon configurations may cause a lookup to happen in a queue runner
3469 process, before it forks any delivery processes. The open lookup caching
3470 mechanism meant that the open file or database connection was passed into
3471 the delivery process. The problem was that delivery processes always tidy
3472 up cached lookup data. This could cause a problem for the next delivery
3473 process started by the queue runner, because the external queue runner
3474 process does not know about the closure. So the next delivery process
3475 still has data in the lookup cache. In the case of a file lookup, there was
3476 no problem because closing a file descriptor in a subprocess doesn't affect
3477 the parent. However, if the lookup was caching a connection to a database,
3478 the connection was closed, and the second delivery process was likely to
3479 see errors such as "PGSQL: query failed: server closed the connection
3480 unexpectedly". The problem has been fixed by closing all cached lookups
3481 in a queue runner before running a delivery process.
3482
3483 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
3484 seem to have the "const" qualifier which it has on other OS. I've
3485 parameterised it.
3486
3487 8. Change 4.21/2 was too strict. It is only if there are two authenticators
3488 *of the same type* (client or server) with the same public name that an
3489 error should be diagnosed.
3490
3491 9. When Exim looked up a host name for an IP address, but failed to find the
3492 original IP address when looking up the host name (a safety check), it
3493 output the message "<ip address> does not match any IP for NULL", which was
3494 confusing, to say the least. The bug was that the host name should have
3495 appeared instead of "NULL".
3496
349710. Since release 3.03, if Exim is called by a uid other than root or the Exim
3498 user that is built into the binary, and the -C or -D options is used, root
3499 privilege is dropped before the configuration file is read. In addition,
3500 logging is switched to stderr instead of the normal log files. If the
3501 configuration then re-defines the Exim user, the unprivileged environment
3502 is probably not what is expected, so Exim logs a panic warning message (but
3503 proceeds).
3504
3505 However, if deliver_drop_privilege is set, the unprivileged state may well
3506 be exactly what is intended, so the warning has been cut out in that case,
3507 and Exim is allowed to try to write to its normal log files.
3508
3509
3510Exim version 4.21
3511-----------------
3512
3513 1. smtp_return_error_details was not giving details for temporary sender
3514 or receiver verification errors.
3515
3516 2. Diagnose a configuration error if two authenticators have the same public
3517 name.
3518
3519 3. Exim used not to create the message log file for a message until the first
3520 delivery attempt. This could be confusing when incoming messages were held
3521 for policy or load reasons. The message log file is now created at the time
3522 the message is received, and an initial "Received" line is written to it.
3523
3524 4. The automatically generated man page for command line options had a minor
3525 bug that caused no ill effects; however, a more serious problem was that
3526 the procedure for building the man page automatically didn't always
3527 operate. Consequently, release 4.20 contains an out-of-date version. This
3528 shouldn't happen again.
3529
3530 5. When building Exim with embedded Perl support, the script that builds the
3531 Makefile was calling 'perl' to find its compile-time parameters, ignoring
3532 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
3533
3534 6. The freeze_tell option was not being used for messages that were frozen on
3535 arrival, either by an ACL or by local_scan().
3536
3537 7. Added the smtp_incomplete_transaction log selector.
3538
3539 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
3540 was accepting AUTH without a new EHLO.
3541
3542 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
3543 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
3544 RFC.
3545
354610. Logging of TCP/IP connections (when configured) now happens in the main
3547 daemon process instead of the child process, so that the TCP/IP connection
3548 count is more accurate (but it can never be perfect).
3549
355011. The use of "drop" in a nested ACL was not being handled correctly in the
3551 outer ACL. Now, if condition failure induced by the nested "drop" causes
3552 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
3553 or "require"), the connection is dropped.
3554
355512. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
3556 that yield "discard" can now be used with an "accept" or a "discard" verb,
3557 but an error is generated for any others (because I can't see a useful way
3558 to define what should happen).
3559
356013. When an ACL is read dynamically from a file (or anywhere else), the lines
3561 are now processed in the same way as lines in the Exim configuration file.
3562 In particular, continuation lines are supported.
3563
356414. Added the "dnslists = a.b.c!=n.n.n.n" feature.
3565
356615. Added -ti meaning -t -i.
3567
356816. Check for letters, digits, hyphens, and dots in the names of dnslist
3569 domains, and warn by logging if others are found.
3570
357117. At least on BSD, alignment is not guarenteed for the array of ifreq's
3572 returned from GIFCONF when Exim is trying to find the list of interfaces on
3573 a host. The code in os.c has been modified to copy each ifreq to an aligned
3574 structure in all cases.
3575
3576 Also, in some cases, the returned ifreq's were being copied to a 'struct
3577 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
3578 means the last couple of bytes of an IPv6 address could be chopped if the
3579 ifreq contained only a normal sockaddr (14 bytes storage).
3580
358118. Named domain lists were not supported in the hosts_treat_as_local option.
3582 An entry such as +xxxx was not recognized, and was treated as a literal
3583 domain name.
3584
358519. Ensure that header lines added by a DATA ACL are included in the reject log
3586 if the ACL subsequently rejects the message.
3587
358820. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
3589 MD5 (which is deprecated).
3590
359121. When testing a filter file using -bf, Exim was writing a message when it
3592 took the sender from a "From " line in the message, but it was not doing so
3593 when it took $return_path from a Return-Path: header line. It now does.
3594
359522. If the contents of a "message" modifier for a "warn" ACL verb do not begin
3596 with a valid header line field name (a series of printing characters
3597 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
3598
359923. Changed "disc" in the source to "disk" to conform to the documentation and
3600 the book and for uniformity.
3601
360224. Ignore Sendmail's -Ooption=value command line item.
3603
360425. When execve() failed while trying to run a command in a pipe transport,
3605 Exim was returning EX_UNAVAILBLE (69) from the subprocess. However, this
3606 could be confused with a return value of 69 from the command itself. This
3607 has been changed to 127, the value the shell returns if it is asked to run
3608 a non-existent command. The wording for the related log line suggests a
3609 non-existent command as the problem.
3610
361126. If received_header_text expands to an empty string, do not add a Received:
3612 header line to the message. (Well, it adds a token one on the spool, but
3613 marks it "old" so that it doesn't get used or transmitted.)
3614
361527. Installed eximstats 1.28 (addition of -nt option).
3616
361728. There was no check for failure on the call to getsockname() in the daemon
3618 code. This can fail if there is a shortage of resources on the system, with
3619 ENOMEM, for example. A temporary error is now given on failure.
3620
362129. Contrary to the C standard, it seems that in some environments, the
3622 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
3623 program. Exim now does this explicitly; it affects the formatting of
3624 timestamps using strftime().
3625
362630. If exiqsumm was given junk data, it threw up some uninitialized variable
3627 complaints. I've now initialized all the variables, to avoid this.
3628
362932. Header lines added by a system filter were not being "seen" during
3630 transport-time rewrites.
3631
363233. The info_callback() function passed to OpenSSL is set up with type void
3633 (*)(SSL *, int, int), as described somewhere. However, when calling the
3634 function (actually a macro) that sets it up, the type void(*)() is
3635 expected. I've put in a cast to prevent warnings from picky compilers.
3636
363734. If a DNS black list lookup found a CNAME record, but there were no A
3638 records associated with the domain it pointed at, Exim crashed.
3639
364035. If a DNS black list lookup returned more than one A record, Exim ignored
3641 all but the first. It now scans all returned addresses if a particular IP
3642 value is being sought. In this situation, the contents of the
3643 $dnslist_value variable are a list of all the addresses, separated by a
3644 comma and a space.
3645
364636. Tightened up the rules for host name lookups using reverse DNS. Exim used
3647 to accept a host name and all its aliases if the forward lookup for any of
3648 them yielded the IP address of the incoming connection. Now it accepts only
3649 those names whose forward lookup yields the correct IP address. Any other
3650 names are discarded. This closes a loophole whereby a rogue DNS
3651 administrator could create reverse DNS records to break through a
3652 wildcarded host restriction in an ACL.
3653
365437. If a user filter or a system filter that ran in a subprocess used any of
3655 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
3656 the wrong values were passed to the pipe command ($thisaddress had the
3657 value of $0, $0 had the value of $1, etc). This bug was introduced by
3658 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
3659
366038. Improved the line breaking for long SMTP error messages from ACLs.
3661 Previously, if there was no break point between 40 and 75 characters, Exim
3662 left the rest of the message alone. Two changes have been made: (a) I've
3663 reduced the minimum length to 35 characters; (b) if it can't find a break
3664 point between 35 and 75 characters, it looks ahead and uses the first one
3665 that it finds. This may give the occasional overlong line, but at least the
3666