Change HAVE_LOGIN_CAP to HAVE_SETCLASSRESOURCES because the former isn't
[exim.git] / doc / doc-txt / ChangeLog
CommitLineData
79378e0f 1$Cambridge: exim/doc/doc-txt/ChangeLog,v 1.331 2006/03/16 12:25:24 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
214e2000 322
5de37277
PH
323Exim version 4.60
324-----------------
325
cc38ddbf
PH
326PH/01 Two changes to the default runtime configuration:
327
328 (1) Move the checks for relay_from_hosts and authenticated clients from
329 after to before the (commented out) DNS black list checks.
330
331 (2) Add control=submission to the relay_from_hosts and authenticated
332 clients checks, on the grounds that messages accepted by these
333 statements are most likely to be submissions.
5de37277 334
72fdd6ae
PH
335PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
336
337 (1) Generate an error if the third argument for the ${prvs expansion is
338 not a single digit.
339
340 (2) Treat a missing third argument of ${prvscheck as if it were an empty
341 string.
342
343 (3) Reset the variables that are obtained from the first argument of
344 ${prvscheck and used in the second argument before leaving the code,
345 because their memory is reclaimed, so using them afterwards may do
346 silly things.
347
348 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
349 one (it's much easier than Tom thought :-).
350
351 (5) Because of (4), we can now allow for the use of $prvscheck_result
352 inside the third argument.
cb9328de 353
cb741023
PH
354PH/03 For some reason, the default setting of PATH when running a command from
355 a pipe transport was just "/usr/bin". I have changed it to
356 "/bin:/usr/bin".
357
f174f16e
PH
358PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
359 anything to be listed in the output from -bV.
b2f5a032 360
c25242d7
PH
361PH/05 When a filter generated an autoreply, the entire To: header line was
362 quoted in the delivery log line, like this:
363
364 => >A.N.Other <ano@some.domain> <original@ddress> ...
365
366 This has been changed so that it extracts the operative address. There
367 may be more than one such address. If so, they are comma-separated, like
368 this:
369
370 => >ano@some.domain,ona@other.domain <original@ddress> ...
371
82c19f95
PH
372PH/06 When a client host used a correct literal IP address in a HELO or EHLO
373 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
374 not being looked up in the rDNS to get a host name, Exim was showing the
375 IP address twice in Received: lines, even though the IP addresses were
376 identical. For example:
377
378 Received: from [1.2.3.4] (helo=[1.2.3.4])
379
380 However, if the real host name was known, it was omitting the HELO data
381 if it matched the actual IP address. This has been tidied up so that it
382 doesn't show the same IP address twice.
383
d7ffbc12
PH
384PH/07 When both +timestamp and +memory debugging was on, the value given by
385 $tod_xxx expansions could be wrong, because the tod_stamp() function was
386 called by the debug printing, thereby overwriting the timestamp buffer.
387 Debugging no longer uses the tod_stamp() function when +timestamp is set.
388
9f526266
PH
389PH/08 When the original message was included in an autoreply transport, it
390 always said "this is a copy of the message, including all the headers",
391 even if body_only or headers_only was set. It now gives an appropriate
392 message.
393
87fcc8b9
PH
394PH/09 Applied a patch from the Sieve maintainer which:
395
396 o fixes some comments
397 o adds the (disabled) notify extension core
398 o adds some debug output for the result of if/elsif tests
399 o points to the current vacation draft in the documentation
400 and documents the missing references header update
401
402 and most important:
403
404 o fixes a bug in processing the envelope test (when testing
405 multiple envelope elements, the last element determinted the
406 result)
407
456682f5
PH
408PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
409 Electronic Mail") by including:
410
411 Auto-submitted: auto-generated
412
413 in the messages that it generates (bounce messages and others, such as
414 warnings). In the case of bounce messages for non-SMTP mesages, there was
415 also a typo: it was using "Auto_submitted" (underscore instead of
416 hyphen). Since every message generated by Exim is necessarily in response
417 to another message, thes have all been changed to:
418
419 Auto-Submitted: auto-replied
420
421 in accordance with these statements in the RFC:
422
423 The auto-replied keyword:
424
425 - SHOULD be used on messages sent in direct response to another
426 message by an automatic process,
427
428 - MUST NOT be used on manually-generated messages,
429
430 - MAY be used on Delivery Status Notifications (DSNs) and Message
431 Disposition Notifications (MDNs),
432
433 - MUST NOT be used on messages generated by automatic or periodic
434 processes, except for messages which are automatic responses to
435 other messages.
436
3e46c1aa
PH
437PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
438 to the default Received: header definition.
456682f5 439
49826d12
PH
440PH/12 Added log selector acl_warn_skipped (default on).
441
eba0c039
PH
442PH/13 After a successful wildlsearch lookup, discard the values of numeric
443 variables because (a) they are in the wrong storage pool and (b) even if
444 they were copied, it wouldn't work properly because of the caching.
445
a0d6ba8a
PH
446PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
447 checking when decoding. Apparently there are clients that generate
448 overlong encoded strings. Why am I not surprised?
449
f0917727
PH
450PH/15 If the first argument of "${if match_address" was not empty, but did not
451 contain an "@" character, Exim crashed. Now it writes a panic log message
452 and treats the condition as false.
453
096fee00
PH
454PH/16 In autoreply, treat an empty string for "once" the same as unset.
455
024bd3c2
PH
456PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
457 extension "envelope-auth". The code is finished and in agreement with
458 other implementations, but there is no documentation so far and in fact,
459 nobody wrote the draft yet. This extension is currently #undef'ed, thus
460 not changing the active code.
461
462 Print executed "if" and "elsif" statements when debugging is used. This
463 helps a great deal to understand what a filter does.
464
465 Document more things not specified clearly in RFC3028. I had all this
466 sorted out, when out of a sudden new issues came to my mind. Oops."
467
df199fec
PH
468PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
469 (Bugzilla #53).
470
d27f1df3
PH
471PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
472 canonical form (as documented). However, after a host name lookup from
473 the IP address, check_host() was doing a simple string comparison with
474 addresses acquired from the DNS when checking that the found name did
475 have the original IP as one of its addresses. Since any found IPv6
476 addresses are likely to be in abbreviated form, the comparison could
477 fail. Luckily, there already exists a function for doing the comparison
478 by converting both addresses to binary, so now that is used instead of
479 the text comparison.
480
96776534
PH
481PH/20 There was another similar case to PH/19, when a complete host name was
482 given in a host list; looking up its IP address could give an abbreviated
483 form, whereas the current host's name might or might not be abbreviated.
484 The same fix has been applied.
485
5de37277 486
9a799bc0
PH
487Exim version 4.54
488-----------------
489
490PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
491 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
492 It now does.
493
99a4b039
PH
494PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
495 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
496
497PH/03 Typo: missing ".o" in src/pcre/Makefile.
498
4b233853
PH
499PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
500 header line, restrict the check to what is listed in RFCs 2369 and 2929.
501 Also, for "Auto-Submitted", treat anything other than "no" as
502 non-personal, in accordance with RFC 3834. (Previously it treated
503 anything starting "auto-" as non-personal.)
504
8857ccfd
PH
505TF/01 The control=submission/name=... option had a problem with syntax
506 errors if the name included a slash character. The /name= option
507 now slurps the rest of the string, so it can include any characters
508 but it must come last in the list of options (after /sender_retain
509 or /domain=).
510
433a2980
PH
511PH/05 Some modifications to the interface to the fake nameserver for the new
512 testing suite.
513
3e46c1aa 514
9a799bc0 515
e3a311ba
TK
516Exim version 4.53
517-----------------
518
519TK/01 Added the "success_on_redirect" address verification option. See
520 NewStuff for rationale and an example.
521
13b685f9
PH
522PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
523
395ff96d
PH
524PH/02 Patch to exigrep to allow it to work on syslog lines.
525
5b68f6e4
PH
526PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
527 fread() to skip over the body file's header line, because in Cygwin the
528 header line is locked and is inaccessible.
529
1ab52c69
PH
530PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
531 co-exist for some time) to make it clear that it is the Exim ID that is
532 referenced, not the Message-ID: header line.
533
b07e6aa3
PH
534PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
535 string_format() function, because snprintf() does not exist on all
536 operating systems.
537
254e032f
PH
538PH/06 The use of forbid_filter_existstest now also locks out the use of the
539 ${stat: expansion item.
540
3af76a81
PH
541PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
542 protocol synchronization error", to keep the pedants happy.
543
2548ba04
PH
544PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
545 well as for IRIX systems, when gcc is being used. See the host.c source
546 file for comments.
547
b6c6011d
PH
548PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
549
cf39cf57
PH
550PH/10 Named domain lists were not working if used in a queue_smtp_domains
551 setting.
552
f1513293
PH
553PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
554 transport and to the smtp transport in LMTP mode.
555
727549a4
PH
556TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
557
af46795e
PH
558PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
559 run a filter in a subprocess. This could lead to confusion in subsequent
560 lookups in the parent process. There should also be a search_tidyup() at
561 the end of the subprocess.
562
d7b47fd0
PH
563PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
564 only if the host matched helo_try_verify_hosts, which caused the
565 verification to occur when the EHLO/HELO command was issued. The ACL just
566 tested the remembered result. Now, if a previous verification attempt has
567 not happened, "verify = helo" does it there and then.
568
ee744174
JJ
569JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
570
b582ab87
PH
571TK/03 Fix log output including CR from clamd.
572
41a13e0a
PH
573PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
574 exist provoked a memory error which could cause a segfault.
575
f625cc5a
PH
576PH/15 Installed PCRE 6.2
577
578PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
579
21f7af35
PH
580PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
581 of the problem. Specifically, suggested +O2 rather than +O1 for the
582 HP-UX compiler.
583
31480e42
PH
584PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
585
2d280592
PH
586PH/20 If a delivery was routed to a non-standard port by means of an SRV
587 record, the port was not correctly logged when the outgoing_port log
588 selector was set (it logged the transort's default port).
589
7cd1141b
PH
590PH/21 Added support for host-specific ports to manualroute, queryprogram,
591 fallback_hosts, and "hosts" in the smtp transport.
592
593PH/22 If the log selector "outgoing_port" is set, the port is now also given on
594 host errors such as "Connection refused".
595
750af86e
PH
596PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
597 authentication with radiusclient 0.4.9:
598
599 - Error returned from rc_read_config was caught wrongly
600 - Username/password not passed on to radius server due to wrong length.
601
602 The presumption is that some radiusclient API changes for 4.51/PH/17
603 were not taken care of correctly. The code is still untested by me (my
604 Linux distribution still has 0.3.2 of radiusclient), but it was
605 contributed by a Radius user.
606
607PH/24 When doing a callout, the value of $domain wasn't set correctly when
608 expanding the "port" option of the smtp transport.
609
4304270b
TK
610TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
611 while reading a MIME header. Thanks to Tom Hughes for a patch.
612
750af86e
PH
613PH/24 Include config.h inside local_scan.h so that configuration settings are
614 available.
615
64ffc24f
PH
616PH/25 Make $smtp_command_argument available after all SMTP commands. This means
617 that in an ACL for RCPT (for example), you can examine exactly what was
618 received.
619
5dd9625b
PH
620PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
621 commands, but it was not correctly comparing the address with the actual
622 client host address. Thus, it would show the EHLO address in Received:
623 header lines when this was not necessary.
624
5591031b
PH
625PH/27 Added the % operator to ${eval:}.
626
ba18e66a
PH
627PH/28 Exim tries to create and chdir to its spool directory when it starts;
628 it should be ignoring failures (because with -C, for example, it has lost
629 privilege). It wasn't ignoring creation failures other than "already
630 exists".
631
9cec981f
PH
632PH/29 Added "crypteq" to the list of supported features that Exim outputs when
633 -bV or -d is used.
634
aa2b5c79
PH
635PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
636 because an input line was too long, either on its own, or by virtue of
1509d3a8
PH
637 too many continuations, the temporary file was not being removed, and the
638 return code was incorrect.
aa2b5c79 639
48a53b7f
PH
640PH/31 Missing "BOOL" in function definition in filtertest.c.
641
1c59d63b
PH
642PH/32 Applied Sieve patches from the maintainer.
643
671012da
TK
644TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
645
1509d3a8
PH
646PH/33 Added "verify = not_blind".
647
648PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
649 Local/Makefile (with some defaults set). These are used in built scripts
650 such as exicyclog, but they have never been used in the exim_install
651 script (though there are many overriding facilities there). I have
652 arranged that the exim_install script now takes note of these two
653 settings.
654
655PH/35 Installed configuration files for Dragonfly.
656
2fe1a124
PH
657PH/36 When a locally submitted message by a trusted user did not contain a
658 From: header, and the sender address was obtained from -f or from an SMTP
659 MAIL command, and the trusted user did not use -F to supply a sender
660 name, $originator_name was incorrectly used when constructing a From:
661 header. Furthermore, $originator_name was used for submission mode
662 messages from external hosts without From: headers in a similar way,
663 which is clearly wrong.
664
8800895a
PH
665PH/37 Added control=suppress_local_fixups.
666
ccfdb010
PH
667PH/38 When log_selector = +received_sender was set, and the addition of the
668 sender made the log line's construction buffer exactly full, or one byte
669 less than full, an overflow happened when the terminating "\n" was
670 subsequently added.
671
1130bfb0
PH
672PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
673 when the result of a list match is failure because a DNS lookup failed.
674
ebcb507f
PH
675PH/40 RM_COMMAND is now used in the building process.
676
c35e155c
PH
677PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
678 the "build-* directories that it finds.
679
95d1f782
PH
680PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
681 address in a domain literal was a prefix of an interface address.
682
fd6de02e
PH
683PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
684 when verifying a sender address, unless rewrite_headers is false.
685
58de37c5
PH
686PH/44 Wrote a long comment about why errors_to addresses are verified as
687 recipients, not senders.
688
261cf466
TF
689TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
690 the ratelimit ACL was added.
691
3ee512ff
PH
692PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
693
e08c430f
PH
694PH/46 Added extra information about PostgreSQL errors to the error string.
695
bef5a11f
PH
696PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
697 avoiding the need to install special zones in a real server. This is
698 backwards compatible; if it can't find the fake resolver, it drops back.
699 Thus, both old and new test suites can be run.
700
7546de58
TF
701TF/02 Added util/ratelimit.pl
702
e5d5a95f
TF
703TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
704 clock is set back in time.
705
2e88a017
TF
706TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
707 Candler <B.Candler@pobox.com>.
708
a5f65aa4
TF
709TF/05 The fix for PH/43 was not completely correct; widen_domains is always
710 OK for addresses that are the result of redirections.
711
e7726cbf
PH
712PH/48 A number of further additions for the benefit of the new test suite,
713 including a fake gethostbyname() that interfaces to the fake DNS resolver
714 (see PH/47 above).
715
a7fdad5b
TF
716TF/06 The fix for widen_domains has also been applied to qualify_single and
717 search_parents which are the other dnslookup options that can cause
718 header rewrites.
719
6af56900
PH
720PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
721 type ("H").
722
0925ede6
PH
723PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
724
66afa403
TF
725TF/07 Exim produced the error message "an SRV record indicated no SMTP
726 service" if it encountered an MX record with an empty target hostname.
727 The message is now "an MX or SRV record indicated no SMTP service".
728
0154e85a
TF
729TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
730 if the DNS of the sending site is misconfigured. This is quite a
731 common situation. This change restores the behaviour of treating a
732 helo verification defer as a failure.
733
16f12c76
PH
734PH/51 If self=fail was set on a router, the bounce message did not include the
735 actual error message.
736
bbe902f0 737
e5a9dba6
PH
738Exim version 4.52
739-----------------
740
741TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
742
22c3b60b
PH
743PH/01 When a transport filter timed out in a pipe delivery, and the pipe
744 command itself ended in error, the underlying message about the transport
745 filter timeout was being overwritten with the pipe command error. Now the
746 underlying error message should be appended to the second error message.
747
06a9b4b5
PH
748TK/01 Fix poll() being unavailable on Mac OSX 10.2.
749
c1ac6996
PH
750PH/02 Reduce the amount of output that "make" produces by default. Full output
751 can still be requested.
752
9c7a242c
PH
753PH/03 The warning log line about a condition test deferring for a "warn" verb
754 was being output only once per connection, rather than after each
755 occurrence (because it was using the same function as for successful
756 "warn" verbs). This seems wrong, so I have changed it.
757
87ba3f5f
PH
758TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
759 it should not have, which might have caused a crash in the right
760 circumstances, but probably never did.
761
762PH/04 Installed a modified version of Tony Finch's patch to make submission
763 mode fix the return path as well as the Sender: header line, and to
764 add a /name= option so that you can make the user's friendly name appear
765 in the header line.
766
29aba418
TF
767TF/03 Added the control = fakedefer ACL modifier.
768
fe0dab11
TF
769TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
770 Mark Lowes for thorough testing.
870f6ba8 771
11d337a4
TK
772TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
773
774TK/03 Merged latest SRS patch from Miles Wilton.
775
415c8f3b
PH
776PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
777 with the definition in sysexits.h (which is #included earlier).
778 Fortunately, Exim does not actually use EX_OK. The code used to try to
779 preserve the sysexits.h value, by assumimg that macro definitions were
780 scanned for macro replacements. I have been disabused of this notion,
781 so now the code just undefines EX_OK before #including unistd.h.
11d337a4 782
958541e9
PH
783PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
784 in the smtp transport. When a block could not be written in a single
785 write() function, the timeout was being re-applied to each part-write.
786 This seems wrong - if the receiver was accepting one byte at a time it
787 would take for ever. The timeout is now adjusted when this happens. It
788 doesn't have to be particularly precise.
789
c206415f
TK
790TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
791 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
792
2a4be8f9
PH
793PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
794 without a domain if the check to <postmaster@domain> fails.
795
1cba11c5
SC
796SC/01 Eximstats: added -xls and the ability to specify output files
797 (patch written by Frank Heydlauf).
798
799SC/02 Eximstats: use FileHandles for outputing results.
800
801SC/03 Eximstats: allow any combination of xls, txt, and html output.
802
803SC/04 Eximstats: fixed display of large numbers with -nvr option
804
805SC/05 Eximstats: fixed merging of reports with empty tables.
806
807SC/06 Eximstats: added the -include_original_destination flag
808
809SC/07 Eximstats: removed tabs and trailing whitespace.
810
1005d00e
TK
811TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
812
813TK/06 MBOX spool code: Add real "From " MBOX separator line
814 so the .eml file is really in mbox format (even though
815 most programs do not really care). Patch from Alex Miller.
816
817TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
818 The latter is generated from $received_to and is only set if the
819 message has one envelope recipient. SA can use these headers,
820 obviously out-of-the-box. Patch from Alex Miller.
821
9b4768fa
PH
822PH/08 The ${def test on a variable was returning false if the variable's
823 value was "0", contrary to what the specification has always said!
824 The result should be true unless the variable is empty.
825
826PH/09 The syntax error of a character other than { following "${if
827 def:variable_name" (after optional whitespace) was not being diagnosed.
828 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
829 accidental colon was present, for example, could give incorrect results.
830
0d7eb84a
PH
831PH/10 Tidied the code in a number of places where the st_size field of a stat()
832 result is used (not including appendfile, where other changes are about
833 to be made).
834
835PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
836 This involved changing a lot of size variables from int to off_t. It
837 should work with maildirs and everything.
838
40727bee
TK
839TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
840 spamd dying while we are connected to it.
841
554d2369
TF
842TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
843 <jgh@wizmail.org>
844
1f922db1
PH
845PH/12 Applied Alex Kiernan's patch for the API change for the error callback
846 function for BDB 4.3.
847
ef213c3b
PH
848PH/13 Changed auto_thaw such that it does not apply to bounce messages.
849
8ac170f3
PH
850PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
851 the sources for PCRE have been re-arranged and more files are now
852 involved.
853
b1c749bb
PH
854PH/15 The code I had for printing potentially long long variables in PH/11
855 above was not the best (it lost precision). The length of off_t variables
856 is now inspected at build time, and an appropriate printing format (%ld
c6c2dc1d
PH
857 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
858 to be "long long int" or "long int". This is needed for the internal
859 formatting function string_vformat().
b1c749bb 860
4aac9b49
PH
861PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
862 the configuration file to be ":syslog", then the script "guesses" where
863 the logs files are, rather than using the compiled in default. In our
864 case the guess is not the same as the compiled default, so the script
865 suddenly stopped working when I started to use syslog. The patch checks
866 to see if log_file_path is "". If so, it attempts to read it from exim
867 with no configuration file to get the compiled in version, before it
868 falls back to the previous guessing code."
869
294520c8
TK
870TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
871 implementing BATV in an Exim configuration. See NewStuff for the gory
872 details.
873
5bd022fe
PH
874PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
875 Makefile that are specific to HP-UX.
876
90e9ce59
PH
877PH/18 If the "use_postmaster" option was set for a recipient callout together
878 with the "random" option, the postmaster address was used as the MAIL
879 FROM address for the random test, but not for the subsequent recipient
880 test. It is now used for both.
881
5ea81592
PH
882PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
883 patch removes a few documentation additions to RFC 3028, because the
884 latest draft now contains them. It adds the new en;ascii-case comparator
885 and a new error check for 8bit text in MIME parts. Comparator and
886 require names are now matched exactly. I enabled the subaddress
887 extension, but it is not well tested yet (read: it works for me)."
888
c6c2dc1d
PH
889PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
890 rework some of the code of TK/09 above to avoid the hardwired use of
891 "%lld" and "long long". Replaced the call to snprintf() with a call to
892 string_vformat().
893
fffffe4c
PH
894PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
895 records point to non-existent hosts", "retry timeout exceeded", and
896 "retry time not reached for any host after a long failure period".
ca02eafb 897
9a26b6b2
PH
898PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
899 experimental DomainKeys support:
900
901 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
902 (2) On an error such as an illegally used "control", the wrong name for
903 the control was given.
904
905 These problems did NOT occur unless DomainKeys support was compiled.
906
4aee0225
PH
907PH/23 Added daemon_startup_retries and daemon_startup_sleep.
908
32d668a5
PH
909PH/24 Added ${if match_ip condition.
910
8187c3f3
PH
911PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
912 databases so that it will be absolutely obvious if a crash occurs in the
913 DB library. This is a regular occurrence (often caused by mis-matched
914 db.h files).
915
ff790e47 916PH/26 Insert a lot of missing (void) casts for functions such as chown(),
f1e894f3
PH
917 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
918 picked up on a user's system that detects such things. There doesn't seem
919 to be a gcc warning option for this - only an attribute that has to be
920 put on the function's prototype. It seems that in Fedora Core 4 they have
921 set this on a number of new functions. No doubt there will be more in due
922 course.
ff790e47 923
5417f6d1
PH
924PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
925 specify a transport. However, if an address that was verified by such a
926 router was the subject of a callout, Exim crashed because it tried to
927 read the rcpt_include_affixes from the non-existent transport. Now it
928 just assumes that the setting of that option is false. This bug was
929 introduced by 4.51/PH/31.
930
59cf8544
PH
931PH/28 Changed -d+all to exclude +memory, because that information is very
932 rarely of interest, but it makes the output a lot bigger. People tend to
933 do -d+all out of habit.
934
e7ad8a65
PH
935PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
936 code in os-type was giving problems when libc.so lives in lib64, like on
937 x86_64 Fedora Core.
938
ade42478
PH
939PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
940 aren't the modern standard, and it seems that some systems' include files
941 don't always have them. Exim was already checking for some of the newer
942 ones like T_AAAA, and defining it itself. I've added checks for all the
943 record types that Exim uses.
944
182ad5cf
PH
945PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
946 not automatically generating a new one, as it is supposed to. This
947 prevented TLS from working. If the file did exist, but contained invalid
948 data, a new version was generated, as expected. It was only the case of a
949 non-existent file that was broken.
950
b0d9fc80
TK
951TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
952 with a change in libdomainkeys > 0.64.
953
954TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
955 from DNS. If the selector record carries the flag, it now has
956 precedence over the domain-wide flag.
957
958TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
959
47c7a64a
PH
960PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
961 the use of a port name that isn't defined in /etc/services) occurred, the
962 message was deferred as in a normal delivery, and thus remained on the
963 spool, instead of being failed because of the mua_wrapper setting. This
964 is now fixed, and I tidied up some of the mua_wrapper messages at the
965 same time.
966
a388bce4
SC
967SC/08 Eximstats: whilst parsing the mainlog(s), store information about
968 the messages in a hash of arrays rather than using individual hashes.
969 This is a bit cleaner and results in dramatic memory savings, albeit
970 at a slight CPU cost.
971
972SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
973 as requested by Marc Sherman.
974
975SC/10 Eximstats: added histograms for user specified patterns as requested
976 by Marc Sherman.
977
0793e4ed
SC
978SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
979
c58b88df
PH
980PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
981 fopen() in the content-scanning modules that did not already have it.
982
e7ad8a65 983
7982096b
PH
984Exim version 4.51
985-----------------
986
1a46a8c5
PH
987TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
988 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
989
2f079f46 990TK/02 Fix ACL "control" statement not being available in MIME ACL.
1a46a8c5
PH
991
992TK/03 Fix ACL "regex" condition not being available in MIME ACL.
993
994PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
995 to test Sieve filters that use "vacation".
996
997PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
998 that changes the way the GnuTLS parameters are stored in the cache file.
999 The new format can be generated externally. For backward compatibility,
1000 if the data in the cache doesn't make sense, Exim assumes it has read an
1001 old-format file, and it generates new data and writes a new file. This
1002 means that you can't go back to an older release without removing the
1003 file.
1004
1005PH/03 A redirect router that has both "unseen" and "one_time" set does not
1006 work if there are any delivery delays because "one_time" forces the
1007 parent to be marked "delivered", so its unseen clone is never tried
1008 again. For this reason, Exim now forbids the simultaneous setting of
1009 these two options.
1010
1011PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
1012 redirected to themselves ("homonym" addresses). Read the long ChangeLog
1013 entry if you want to know the details. The fix, however, neglected to
1014 consider the case when local delivery batching is involved. The test for
1015 "previously delivered" was not happening when checking to see if an
1016 address could be batched with a previous (undelivered) one; under
1017 certain circumstances this could lead to multiple deliveries to the same
c2c19e9d 1018 address.
1a46a8c5
PH
1019
1020PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
1021 in its include files, and this causes problems building Exim.
1022
1023PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
1024 header_syntax) but Exim was just ignoring anything given after a slash.
1025 In particular, this caused confusion with an attempt to use "verify =
1026 reverse_host_lookup/defer_ok". An error is now given when options are
1027 supplied for verify items that do not have them. (Maybe reverse_host_
1028 lookup should have a defer_ok option, but that's a different point.)
1029
1030PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
1031 defined by RFC 821) to 2048, because there were problems with some AUTH
1032 commands, and RFC 1869 says the size should be increased for extended
1033 SMTP commands that take arguments.
1034
1035PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
1036 Finch).
1037
1038PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
1039 "unknown" error; now it says that the functionality isn't in the binary.
8d67ada3 1040
49c2d5ea
PH
1041PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
1042 an address' error message when a string expansion fails (syntax or
f331f3b6
PH
1043 whatever). Otherwise the password may appear in the log. Following change
1044 PH/42 below, there is no longer a chance of it appearing in a bounce
1045 message.
49c2d5ea 1046
bf759a8b
PH
1047PH/11 Installed exipick version 20050225.0 from John Jetmore.
1048
83364d30
PH
1049PH/12 If the last host in a fallback_hosts list was multihomed, only the first
1050 of its addresses was ever tried. (Bugzilla bug #2.)
1051
7999bbd7
PH
1052PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
1053 the result incorrectly in the debug output. (It correctly added a newline
1054 to what was transported.)
1055
7dbf77c9
PH
1056TF/01 Added $received_time.
1057
74e0617f
PH
1058PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
1059 commented out examples of how to interface to a virus scanner and to
1060 SpamAssassin. Also added commented examples of av_scanner and
1061 spamd_address settings.
1062
2f079f46
PH
1063PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
1064 and controls are allowed in which ACLs. There were a couple of minor
1065 errors. Some of the entries in the conditions table (which is a table of
1066 where they are NOT allowed) were getting very unwieldy; rewrote them as a
1067 negation of where the condition IS allowed.
1068
8c841523
PH
1069PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
1070
7766a4f0
PH
1071PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
1072 header file does not have a version number, so I've had to invent a new
1073 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
1074 API. The code is untested by me (my Linux distribution still has 0.3.2 of
1075 radiusclient), but it was contributed by a Radius user.
1076
8b417f2c
PH
1077PH/18 Installed Lars Mainka's patch for the support of CRL collections in
1078 files or directories, for OpenSSL.
1079
901f42cb
PH
1080PH/19 When an Exim process that is running as root has to create an Exim log
1081 file, it does so in a subprocess that runs as exim:exim so as to get the
1082 ownership right at creation (otherwise, other Exim processes might see
1083 the file with the wrong ownership). There was no test for failure of this
1084 fork() call, which would lead to the process getting stuck as it waited
1085 for a non-existent subprocess. Forks do occasionally fail when resources
1086 run out. I reviewed all the other calls to fork(); they all seem to check
1087 for failure.
1088
f9b9210e
PH
1089PH/20 When checking for unexpected SMTP input at connect time (before writing
1090 the banner), Exim was not dealing correctly with a non-positive return
1091 from the read() function. If the client had disconnected by this time,
1092 the result was a log entry for a synchronization error with an empty
1093 string after "input=" when read() returned zero. If read() returned -1
1094 (an event I could not check), uninitialized data bytes were printed.
1095 There were reports of junk text (parts of files, etc) appearing after
1096 "input=".
1097
54cdb463
PH
1098PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
1099
cf00dad6
PH
1100PH/22 Added support for macro redefinition, and (re)definition in between
1101 driver and ACL definitions.
1102
acb1b346
PH
1103PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
1104 forgetting to use the resulting value; it was using the unexpanded value.
1105
c5ddb310
PH
1106PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
1107 hadn't been configured. The fix is from Juergen Kreileder, who
1108 understands it better than I do:
1109
1110 "Here's what I see happening with three configured cyrus_sasl
1111 authenticators configured (plain, login, cram-md5):
1112
1113 On startup auth_cyrus_sasl_init() gets called for each of these.
1114 This means three calls to sasl_listmech() without a specified mech_list.
1115 => SASL tests which mechs of all available mechs actually work
1116 => three warnings about OTP not working
1117 => the returned list contains: plain, login, cram-md5, digest-md5, ...
1118
1119 With the patch, sasl_listmech() also gets called three times. But now
1120 SASL's mech_list option is set to the server_mech specified in the the
1121 authenticator. Or in other words, the answer from sasl_listmech()
1122 gets limited to just the mech you're testing for (which is different
1123 for each call.)
1124 => the return list contains just 'plain' or 'login', 'cram-md5' or
1125 nothing depending on the value of ob->server_mech.
1126
1127 I've just tested the patch: Authentication still works fine,
1128 unavailable mechs specified in the exim configuration are still
1129 caught, and the auth.log warnings about OTP are gone."
1130
31619da6
PH
1131PH/25 When debugging is enabled, the contents of the command line are added
1132 to the debugging output, even when log_selector=+arguments is not
1133 specified.
1134
bebaf0fc
PH
1135PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
1136 answer is "GNU", and only if the return is "GNU/something" is the answer
1137 "Linux".
1138
475fe28a
PH
1139PH/27 $acl_verify_message is now set immediately after the failure of a
1140 verification in an ACL, and so is available in subsequent modifiers. In
1141 particular, the message can be preserved by coding like this:
1142
1143 warn !verify = sender
1144 set acl_m0 = $acl_verify_message
1145
1146 Previously, $acl_verify_message was set only while expanding "message"
1147 and "log_message" when a very denied access.
1148
7e8bec7a
PH
1149PH/28 Modified OS/os.c-Linux with
1150
1151 -#ifndef OS_LOAD_AVERAGE
1152 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
1153
1154 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
1155 nomenclature these days.)
1156
e4a89c47
PH
1157PH/29 Installed patch from the Sieve maintainer that adds the options
1158 sieve_useraddress and sieve_subaddress to the redirect router.
1159
5ca2a9a1
PH
1160PH/30 In these circumstances:
1161 . Two addresses routed to the same list of hosts;
1162 . First host does not offer TLS;
1163 . First host accepts first address;
1164 . First host gives temporary error to second address;
1165 . Second host offers TLS and a TLS session is established;
1166 . Second host accepts second address.
1167 Exim incorrectly logged both deliveries with the TLS parameters (cipher
1168 and peerdn, if requested) that were in fact used only for the second
1169 address.
7e8bec7a 1170
c688b954
PH
1171PH/31 When doing a callout as part of verifying an address, Exim was not paying
1172 attention to any local part prefix or suffix that was matched by the
1173 router that accepted the address. It now behaves in the same way as it
1174 does for delivery: the affixes are removed from the local part unless
1175 rcpt_include_affixes is set on the transport.
1176
fed77020
PH
1177PH/32 Add the sender address, as F=<...>, to the log line when logging a
1178 timeout during the DATA phase of an incoming message.
1179
7fe1560f
PH
1180PH/33 Sieve envelope tests were broken for match types other than :is. I have
1181 applied a patch sanctioned by the Sieve maintainer.
c688b954 1182
ebb6e6d5
PH
1183PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
1184 the uid or gid is negative. A case of a negative gid caused this to be
1185 noticed. The fix allows for either to be negative.
1186
9c4e8f60
PH
1187PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
1188 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
1189 overlooked.
1190
1191PH/36 The change PH/12 above was broken. Fixed it.
1192
d7174846
PH
1193PH/37 Exim used to check for duplicate addresses in the middle of routing, on
1194 the grounds that routing the same address twice would always produce the
1195 same answer. This might have been true once, but it is certainly no
1196 longer true now. Routing a child address may depend on the previous
1197 routing that produced that child. Some complicated redirection strategies
1198 went wrong when messages had multiple recipients, and made Exim's
1199 behaviour dependent on the order in which the addresses were given.
1200
1201 I have moved the duplicate checking until after the routing is complete.
1202 Exim scans the addresses that are assigned to local and remote
1203 transports, and removes any duplicates. This means that more work will be
1204 done, as duplicates will always all be routed, but duplicates are
1205 presumably rare, so I don't expect this is of any significance.
1206
1207 For deliveries to pipes, files, and autoreplies, the duplicate checking
1208 still happens during the routing process, since they are not going to be
1209 routed further.
1210
cfe75fc3
PH
1211PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
1212 It corrects a timeout issue with spamd. This is Ian's comment: "The
1213 background is that sometimes spamd either never reads data from a
1214 connection it has accepted, or it never writes response data. The exiscan
1215 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
1216 blindly assumes that writes won't block so it may never time out."
1217
be22d70e
PH
1218PH/39 Allow G after quota size as well as K and M.
1219
0612b098
PH
1220PH/40 The value set for $authenticated_id in an authenticator may not contain
1221 binary zeroes or newlines because the value is written to log lines and
1222 to spool files. There was no check on this. Now the value is run through
1223 the string_printing() function so that such characters are converted to
1224 printable escape sequences.
1225
2e0c1448
PH
1226PH/41 $message_linecount is a new variable that contains the total number of
1227 lines in the message. Compare $body_linecount, which is the count for the
1228 body only.
1229
447d236c
PH
1230PH/42 Exim no longer gives details of delivery errors for specific addresses in
1231 bounce and delay warning messages, except in certain special cases, which
1232 are as follows:
1233
1234 (a) An SMTP error message from a remote host;
1235 (b) A message specified in a :fail: redirection;
1236 (c) A message specified in a "fail" command in a system filter;
1237 (d) A message specified in a FAIL return from the queryprogram router;
1238 (e) A message specified by the cannot_route_message router option.
1239
1240 In these cases only, Exim does include the error details in bounce and
1241 warning messages. There are also a few cases where bland messages such
1242 as "unrouteable address" or "local delivery error" are given.
1243
d20976dc
PH
1244PH/43 $value is now also set for the "else" part of a ${run expansion.
1245
f656d135
PH
1246PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
1247 being worked on, but at least Exim now implements the latest version to
1248 play with."
1249
2e2a30b4
PH
1250PH/45 In a pipe transport, although a timeout while waiting for the pipe
1251 process to complete was treated as a delivery failure, a timeout while
1252 writing the message to the pipe was logged, but erroneously treated as a
1253 successful delivery. Such timeouts include transport filter timeouts. For
1254 consistency with the overall process timeout, these timeouts are now
1255 treated as errors, giving rise to delivery failures by default. However,
1256 there is now a new Boolean option for the pipe transport called
1257 timeout_defer, which, if set TRUE, converts the failures into defers for
1258 both kinds of timeout. A transport filter timeout is now identified in
1259 the log output.
1260
9176e9f0
PH
1261PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
1262 systems where "make" and "gmake" are different, calling "gmake" at top
1263 level broke things. I've arranged for the value of $(MAKE) to be passed
1264 from the Makefile to this script so that it can call the same version of
1265 "make".
1266
7982096b 1267
bbe902f0
PH
1268A note about Exim versions 4.44 and 4.50
1269----------------------------------------
1270
1271Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
1272changes of various kinds. As a consequence, a big documentation update was
1273needed. This delayed the release for rather longer than seemed good, especially
1274in the light of a couple of (minor) security issues. Therefore, the changes
1275that fixed bugs were backported into 4.43, to create a 4.44 maintenance
1276release. So 4.44 and 4.50 are in effect two different branches that both start
1277from 4.43.
1278
1279I have left the 4.50 change log unchanged; it contains all the changes since
12804.43. The change log for 4.44 is below; many of its items are identical to
1281those for 4.50. This seems to be the most sensible way to preserve the
1282historical information.
1283
1284
f7b63901 1285Exim version 4.50
495ae4b0
PH
1286-----------------
1287
5fe762f6
PH
1288 1. Minor wording change to the doc/README.SIEVE file.
1289
139059f6 1290 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
5fe762f6 1291 computation of the current number of files was incorrect.
495ae4b0 1292
7086e875
PH
1293 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
1294 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
1295 place.
1296
35af9f61
PH
1297 4. Give more explanation in the error message when the command for a transport
1298 filter fails to execute.
1299
b668c215
PH
1300 5. There are several places where Exim runs a non-Exim command in a
1301 subprocess. The SIGUSR1 signal should be disabled for these processes. This
1302 was being done only for the command run by the queryprogram router. It is
1303 now done for all such subprocesses. The other cases are: ${run, transport
1304 filters, and the commands run by the lmtp and pipe transports.
1305
a494b1e1
PH
1306 6. Added CONFIGURE_GROUP build-time option.
1307
1308 7. Some older OS have a limit of 256 on the maximum number of file
1309 descriptors. Exim was using setrlimit() to set 1000 as a large value
1310 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
1311 systems. I've change it so that if it can't get 1000, it tries for 256.
35edf2ff 1312
c5fcb476
PH
1313 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
1314 was an oversight, and furthermore, ever since the addition of extra
1315 controls (e.g. 4.43/32), the checks on when to allow different forms of
1316 "control" were broken. There should now be diagnostics for all cases when a
1317 control that does not make sense is encountered.
1318
69358f02
PH
1319 9. Added the /retain_sender option to "control=submission".
1320
5be20824
PH
132110. $recipients is now available in the predata ACL (oversight).
1322
eb2c0248
PH
132311. Tidy the search cache before the fork to do a delivery from a message
1324 received from the command line. Otherwise the child will trigger a lookup
1325 failure and thereby defer the delivery if it tries to use (for example) a
1326 cached ldap connection that the parent has called unbind on.
1327
2a3eea10
PH
132812. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
1329 of $address_data from the recipient verification was clobbered by the
1330 sender verification.
1331
133213. The value of address_data from a sender verification is now available in
1333 $sender_address_data in subsequent conditions in the ACL statement.
1334
23c7ff99
PH
133514. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
1336
4deaf07d
PH
133715. Added a new option "connect=<time>" to callout options, to set a different
1338 connection timeout.
1339
926e1192
PH
134016. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
1341 was its contents. (It was OK if the option was not defined at all.)
1342
650edc6f
PH
134317. A "Completed" log line is now written for messages that are removed from
1344 the spool by the -Mrm option.
1345
2c7db3f5
PH
134618. New variables $sender_verify_failure and $recipient_verify_failure contain
1347 information about exactly what failed.
1348
3d235903
PH
134919. Added -dd to debug only the daemon process.
1350
7c7ad977
PH
135120. Incorporated Michael Haardt's patch to ldap.c for improving the way it
1352 handles timeouts, both on the server side and network timeouts. Renamed the
1353 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
1354
981756db
PH
135521. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
1356 It is now set to "smtps".
1357
d4eb88df
PH
135822. $host_address is now set to the target address during the checking of
1359 ignore_target_hosts.
1360
136123. When checking ignore_target_hosts for an ipliteral router, no host name was
1362 being passed; this would have caused $sender_host_name to have been used if
1363 matching the list had actually called for a host name (not very likely,
1364 since this list is usually IP addresses). A host name is now passed as
1365 "[x.x.x.x]".
1366
7d468ab8
PH
136724. Changed the calls that set up the SIGCHLD handler in the daemon to use the
1368 code that specifies a non-restarting handler (typically sigaction() in
1369 modern systems) in an attempt to fix a rare and obscure crash bug.
1370
137125. Narrowed the window for a race in the daemon that could cause it to ignore
1372 SIGCHLD signals. This is not a major problem, because they are used only to
1373 wake it up if nothing else does.
1374
62c0818f
PH
137526. A malformed maildirsize file could cause Exim to calculate negative values
1376 for the mailbox size or file count. Odd effects could occur as a result.
1377 The maildirsize information is now recalculated if the size or filecount
1378 end up negative.
1379
26034054
PH
138027. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
1381 support for a long time. Removed HAVE_SYS_VFS_H.
1382
af66f652
PH
138328. Installed the latest version of exipick from John Jetmore.
1384
90af77f4
PH
138529. In an address list, if the pattern was not a regular expression, an empty
1386 subject address (from a bounce message) matched only if the pattern was an
1387 empty string. Non-empty patterns were not even tested. This was the wrong
1388 because it is perfectly reasonable to use an empty address as part of a
1389 database query. An empty address is now tested by patterns that are
1390 lookups. However, all the other forms of pattern expect the subject to
1391 contain a local part and a domain, and therefore, for them, an empty
1392 address still always fails if the pattern is not itself empty.
1393
d8ef3577
PH
139430. Exim went into a mad DNS loop when attempting to do a callout where the
1395 host was specified on an smtp transport, and looking it up yielded more
1396 than one IP address.
1397
5cb8cbc6
PH
139831. Re-factored the code for checking spool and log partition space into a
1399 function that finds that data and another that does the check. The former
1400 is then used to implement four new variables: $spool_space, $log_space,
1401 $spool_inodes, and $log_inodes.
1402
14702f5b
PH
140332. The RFC2047 encoding function was originally intended for short strings
1404 such as real names; it was not keeping to the 75-character limit for
1405 encoded words that the RFC imposes. It now respects the limit, and
1406 generates multiple encoded words if necessary. To be on the safe side, I
1407 have increased the buffer size for the ${rfc2047: expansion operator from
1408 1024 to 2048 bytes.
1409
063b1e99
PH
141033. It is now permitted to omit both strings after an "if" condition; if the
1411 condition is true, the result is "true". As before, when the second string
1412 is omitted, a false condition yields an empty string. This makes it less
1413 cumbersome to write custom ACL and router conditions.
1414
652e1b65
PH
141534. Failure to deliver a bounce message always caused it to be frozen, even if
1416 there was an errors_to setting on the router. The errors_to setting is now
1417 respected.
1418
6f0c9a4f
PH
141935. If an IPv6 address is given for -bh or -bhc, it is now converted to the
1420 canonical form (fully expanded) before being placed in
1421 $sender_host_address.
1422
33397d19
PH
142336. The table in the code that translates DNS record types into text (T_A to
1424 "A" for instance) was missing entries for NS and CNAME. It is just possible
1425 that this could have caused confusion if both these types were looked up
1426 for the same domain, because the text type is used as part of Exim's
1427 per-process caching. But the chance of anyone hitting this buglet seems
1428 very small.
1429
7bb56e1f
PH
143037. The dnsdb lookup has been extended in a number of ways.
1431
1432 (1) There is a new type, "zns", which walks up the domain tree until it
1433 finds some nameserver records. It should be used with care.
1434
ea3bc19b
PH
1435 (2) There is a new type, "mxh", which is like "mx" except that it returns
1436 just the host names, not the priorities.
1437
1438 (3) It is now possible to give a list of domains (or IP addresses) to be
ff4dbb19
PH
1439 looked up. The behaviour when one of the lookups defers can be
1440 controlled by a keyword.
7bb56e1f 1441
ea3bc19b 1442 (4) It is now possible to specify the separator character for use when
7bb56e1f 1443 multiple records are returned.
33397d19 1444
0bcb2a0e
PH
144538. The dnslists ACL condition has been extended: it is now possible to supply
1446 a list of IP addresses and/or domains to be looked up in a particular DNS
1447 domain.
1448
2ac0e484
PH
144939. Added log_selector=+queue_time_overall.
1450
4e1fde53
PH
145140. When running the queue in the test harness, wait just a tad after forking a
1452 delivery process, to get repeatability of debugging output.
1453
de365ded
PH
145441. Include certificate and key file names in error message when GnuTLS fails
1455 to set them up, because the GnuTLS error message doesn't include the name
1456 of the failing file when there is a problem reading it.
1457
f05da2e8
PH
145842. Allow both -bf and -bF in the same test run.
1459
d6453af2
PH
146043. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
1461
f7b63901
PH
146244. The "Exiscan patch" is now merged into the mainline Exim source.
1463
146445. Sometimes the final signoff response after QUIT could fail to get
1465 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
1466 < 0 before doing a fflush(). This bug looks as though it goes back to the
1467 introduction of TLS in release 3.20, but "sometimes" must have been rare
1468 because the tests only now provoked it.
1469
a444213a
PH
147046. Reset the locale to "C" after calling embedded Perl, in case it was changed
1471 (this can affect the format of dates).
1472
0ec020ea
PH
147347. exim_tidydb, when checking for the continued existence of a message for
1474 which it has found a message-specific retry record, was not finding
1475 messages that were in split spool directories. Consequently, it was
1476 deleting retry records that should have stayed in existence.
1477
b1206957
PH
147848. Steve fixed some bugs in eximstats.
1479
148049. The SPA authentication driver was not abandoning authentication and moving
1481 on to the next authenticator when an expansion was forced to fail,
1482 contradicting the general specification for all authenticators. Instead it
1483 was generating a temporary error. It now behaves as specified.
1484
26dd5a95
PH
148550. The default ordering of permitted cipher suites for GnuTLS was pessimal
1486 (the order specifies the preference for clients). The order is now AES256,
1487 AES128, 3DES, ARCFOUR128.
1488
343b2385
PH
148951. Small patch to Sieve code - explicitly set From: when generating an
1490 autoreply.
1491
1c5466b9
PH
149252. Exim crashed if a remote delivery caused a very long error message to be
1493 recorded - for instance if somebody sent an entire SpamAssassin report back
1494 as a large number of 550 error lines. This bug was coincidentally fixed by
1495 increasing the size of one of Exim's internal buffers (big_buffer) that
1496 happened as part of the Exiscan merge. However, to be on the safe side, I
1497 have made the code more robust (and fixed the comments that describe what
1498 is going on).
1499
55ee9ee3
PH
150053. Now that there can be additional text after "Completed" in log lines (if
1501 the queue_time_overall log selector is set), a one-byte patch to exigrep
1502 was needed to allow it to recognize "Completed" as not the last thing in
1503 the line.
1504
d38f8232
PH
150554. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
1506 patch that reportedly fixes this has been added. I am not expert enough to
1507 create a test for it. This is what the patch creator wrote:
1508
1509 "I found a little strange behaviour of ldap code when working with
1510 Windows 2003 AD Domain, where users was placed in more than one
1511 Organization Units. When I tried to give exim partial DN, the exit code
1512 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
1513 But simultaneously result of request was absolutely normal ldap result,
1514 so I produce this patch..."
1515
3295e65b
PH
1516 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
1517 REFERENCE, so I have modified the code to exclude the patch when that macro
1518 is not defined.
1519
7102e136
PH
152055. Some experimental protocols are using DNS PTR records for new purposes. The
1521 keys for these records are domain names, not reversed IP addresses. The
b975ba52
PH
1522 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
1523 leaves it alone. Component reversal etc. now happens only for IP addresses.
ea3a6f44 1524 CAN-2005-0021
7102e136 1525
3ca0ba97
PH
152656. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
1527
c2bcbe20
PH
152857. Double the size of the debug message buffer (to 2048) so that more of very
1529 long debug lines gets shown.
1530
18ce445d
PH
153158. The exicyclog utility now does better if the number of log files to keep
1532 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
1533
1f5b4c3d
PH
153459. Two changes related to the smtp_active_hostname option:
1535
1536 (1) $smtp_active_hostname is now available as a variable.
1537 (2) The default for smtp_banner uses $smtp_active_hostname instead
1538 of $primary_hostname.
1539
b975ba52
PH
154060. The host_aton() function is supposed to be passed a string that is known
1541 to be a valid IP address. However, in the case of IPv6 addresses, it was
1542 not checking this. This is a hostage to fortune. Exim now panics and dies
1543 if the condition is not met. A case was found where this could be provoked
85b87bc2
PH
1544 from a dnsdb PTR lookup with an IPv6 address that had more than 8
1545 components; fortuitously, this particular loophole had already been fixed
1546 by change 4.50/55 above.
1547
1548 If there are any other similar loopholes, the new check in host_aton()
1549 itself should stop them being exploited. The report I received stated that
1550 data on the command line could provoke the exploit when Exim was running as
1551 exim, but did not say which command line option was involved. All I could
1552 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
1553 running as the user.
ea3a6f44 1554 CAN-2005-0021
85b87bc2
PH
1555
155661. There was a buffer overflow vulnerability in the SPA authentication code
1557 (which came originally from the Samba project). I have added a test to the
1558 spa_base64_to_bits() function which I hope fixes it.
ea3a6f44 1559 CAN-2005-0022
b975ba52 1560
17ffcae7
PH
156162. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
1562 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
1563
d95f9fdb
PH
156463. The daemon start-up calls getloadavg() while still root for those OS that
1565 need the first call to be done as root, but it missed one case: when
1566 deliver_queue_load_max is set with deliver_drop_privilege. This is
1567 necessary for the benefit of the queue runner, because there is no re-exec
1568 when deliver_drop_privilege is set.
1569
86b8287f
PH
157064. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
1571 This has been fixed.
1572
60dc5e56
PH
157365. Caching of lookup data for "hosts =" ACL conditions, when a named host list
1574 was in use, was not putting the data itself into the right store pool;
1575 consequently, it could be overwritten for a subsequent message in the same
1576 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
1577 the caching.)
1578
533244af
PH
157966. Added hosts_max_try_hardlimit to the smtp transport, default 50.
1580
a5a28604
PH
158167. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
1582 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
1583 the function were treating the return as a boolean value, which happened to
1584 work because 0=false and not-0=true, but is not correct code.
1585
7e634d24
PH
158668. The host_aton() function was not handling scoped IPv6 addresses (those
1587 with, for example, "%eth0" on the end) correctly.
1588
3e11c26b
PH
158969. Fixed some compiler warnings in acl.c for the bitmaps specified with
1590 negated items (that is, ~something) in unsigned ints. Some compilers
1591 apparently mutter when there is no cast.
1592
6729cf78
PH
159370. If an address verification called from an ACL failed, and did not produce a
1594 user-specific message (i.e. there was only a "system" message), nothing was
1595 put in $acl_verify_message. In this situation, it now puts the system
1596 message there.
1597
00f00ca5
PH
159871. Change 4.23/11 added synchronization checking at the start of an SMTP
1599 session; change 4.31/43 added the unwanted input to the log line - except
1600 that it did not do this in the start of session case. It now does.
1601
c9bdd01c
PH
160272. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
1603 This is wrong and can cause the other end to generate a synchronization
1604 error if it is another Exim or anything else that does the synchronization
1605 check. A QUIT command is no longer sent after a timeout.
1606
d43194df
PH
160773. $host_lookup_deferred has been added, to make it easier to detect DEFERs
1608 during host lookups.
1609
fe5b5d0b
PH
161074. The defer_ok option of callout verification was not working if it was used
1611 when verifying addresses in header lines, that is, for this case:
1612
1613 verify = header_sender/callout=defer_ok
1614
76a2d7ba
PH
161575. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
1616 those file descriptors could be used for SMTP connections. If anything
1617 wrote to stderr (the example that came up was "warn" in embedded Perl), it
1618 could be sent to the SMTP client, causing chaos. The daemon now opens
1619 stdin, stdout, and stderr to /dev/null when it puts itself into the
1620 background.
1621
162276. Arrange for output from Perl's "warn" command to be written to Exim's main
1623 log by default. The user can override this with suitable Perl magic.
1624
04f7d5b9
PH
162577. The use of log_message on a "discard" ACL verb, which is supposed to add to
1626 the log message when discard triggers, was not working for the DATA ACL or
1627 for the non-SMTP ACL.
1628
bc60667e
PH
162978. Error message wording change in sieve.c.
1630
bb6e88ff
PH
163179. If smtp_accept_max_per_host was set, the number of connections could be
1632 restricted to fewer than expected, because the daemon was trying to set up
1633 a new connection before checking whether the processes handling previous
1634 connections had finished. The check for completed processes is now done
1635 earlier. On busy systems, this bug wouldn't be noticed because something
1636 else would have woken the daemon, and it would have reaped the completed
1637 process earlier.
1638
1e70f85b
PH
163980. If a message was submitted locally by a user whose login name contained one
1640 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
1641 It caused a spool format error. I have fixed the spool reading code. A
1642 related problem was that the "from" clause in the Received: line became
1643 illegal because of the space(s). It is now covered by ${quote_local_part.
1644
164581. Included the latest eximstats from Steve (adds average sizes to HTML Top
1646 tables).
1647
4e01f9d6
PH
164882. Updated OS/Makefile-AIX as per message from Mike Meredith.
1649
1ee1cef2
PH
165083. Patch from Sieve maintainer to fix unterminated string problem in
1651 "vacation" handling.
1652
6e2b4ccc
PH
165384. Some minor changes to the Linux configuration files to help with other
1654 OS variants using glibc.
1655
8e669ac1
PH
165685. One more patch for Sieve to update vacation handling to latest spec.
1657
495ae4b0 1658
bbe902f0
PH
1659----------------------------------------------------
1660See the note above about the 4.44 and 4.50 releases.
1661----------------------------------------------------
1662
1663
1664Exim version 4.44
1665-----------------
1666
1667 1. Change 4.43/35 introduced a bug that caused file counts to be
1668 incorrectly computed when quota_filecount was set in an appendfile
1669 transport
1670
1671 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
1672 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
1673 place.
1674
1675 3. Give more explanation in the error message when the command for a transport
1676 filter fails to execute.
1677
1678 4. There are several places where Exim runs a non-Exim command in a
1679 subprocess. The SIGUSR1 signal should be disabled for these processes. This
1680 was being done only for the command run by the queryprogram router. It is
1681 now done for all such subprocesses. The other cases are: ${run, transport
1682 filters, and the commands run by the lmtp and pipe transports.
1683
1684 5. Some older OS have a limit of 256 on the maximum number of file
1685 descriptors. Exim was using setrlimit() to set 1000 as a large value
1686 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
1687 systems. I've change it so that if it can't get 1000, it tries for 256.
1688
1689 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
1690 was an oversight, and furthermore, ever since the addition of extra
1691 controls (e.g. 4.43/32), the checks on when to allow different forms of
1692 "control" were broken. There should now be diagnostics for all cases when a
1693 control that does not make sense is encountered.
1694
1695 7. $recipients is now available in the predata ACL (oversight).
1696
1697 8. Tidy the search cache before the fork to do a delivery from a message
1698 received from the command line. Otherwise the child will trigger a lookup
1699 failure and thereby defer the delivery if it tries to use (for example) a
1700 cached ldap connection that the parent has called unbind on.
1701
1702 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
1703 of $address_data from the recipient verification was clobbered by the
1704 sender verification.
1705
170610. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
1707 was its contents. (It was OK if the option was not defined at all.)
1708
170911. A "Completed" log line is now written for messages that are removed from
1710 the spool by the -Mrm option.
1711
171212. $host_address is now set to the target address during the checking of
1713 ignore_target_hosts.
1714
171513. When checking ignore_target_hosts for an ipliteral router, no host name was
1716 being passed; this would have caused $sender_host_name to have been used if
1717 matching the list had actually called for a host name (not very likely,
1718 since this list is usually IP addresses). A host name is now passed as
1719 "[x.x.x.x]".
1720
172114. Changed the calls that set up the SIGCHLD handler in the daemon to use the
1722 code that specifies a non-restarting handler (typically sigaction() in
1723 modern systems) in an attempt to fix a rare and obscure crash bug.
1724
172515. Narrowed the window for a race in the daemon that could cause it to ignore
1726 SIGCHLD signals. This is not a major problem, because they are used only to
1727 wake it up if nothing else does.
1728
172916. A malformed maildirsize file could cause Exim to calculate negative values
1730 for the mailbox size or file count. Odd effects could occur as a result.
1731 The maildirsize information is now recalculated if the size or filecount
1732 end up negative.
1733
173417. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
1735 support for a long time. Removed HAVE_SYS_VFS_H.
1736
ea3a6f44 173718. Updated exipick to current release from John Jetmore.
bbe902f0
PH
1738
173919. Allow an empty sender to be matched against a lookup in an address list.
1740 Previously the only cases considered were a regular expression, or an
1741 empty pattern.
1742
174320. Exim went into a mad DNS lookup loop when doing a callout where the
1744 host was specified on the transport, if the DNS lookup yielded more than
1745 one IP address.
1746
ea3a6f44
NM
174721. The RFC2047 encoding function was originally intended for short strings
1748 such as real names; it was not keeping to the 75-character limit for
1749 encoded words that the RFC imposes. It now respects the limit, and
1750 generates multiple encoded words if necessary. To be on the safe side, I
1751 have increased the buffer size for the ${rfc2047: expansion operator from
1752 1024 to 2048 bytes.
bbe902f0 1753
ea3a6f44
NM
175422. Failure to deliver a bounce message always caused it to be frozen, even if
1755 there was an errors_to setting on the router. The errors_to setting is now
1756 respected.
bbe902f0
PH
1757
175823. If an IPv6 address is given for -bh or -bhc, it is now converted to the
1759 canonical form (fully expanded) before being placed in
1760 $sender_host_address.
1761
176224. Updated eximstats to version 1.33
1763
ea3a6f44
NM
176425. Include certificate and key file names in error message when GnuTLS fails
1765 to set them up, because the GnuTLS error message doesn't include the name
1766 of the failing file when there is a problem reading it.
bbe902f0
PH
1767
176826. Expand error message when OpenSSL has problems setting up cert/key files.
ea3a6f44 1769 As per change 25.
bbe902f0 1770
ea3a6f44
NM
177127. Reset the locale to "C" after calling embedded Perl, in case it was changed
1772 (this can affect the format of dates).
bbe902f0 1773
ea3a6f44
NM
177428. exim_tidydb, when checking for the continued existence of a message for
1775 which it has found a message-specific retry record, was not finding
1776 messages that were in split spool directories. Consequently, it was
1777 deleting retry records that should have stayed in existence.
bbe902f0
PH
1778
177929. eximstats updated to version 1.35
1780 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
1781 1.35 - bugfix such that pie charts by volume are generated correctly
1782
ea3a6f44
NM
178330. The SPA authentication driver was not abandoning authentication and moving
1784 on to the next authenticator when an expansion was forced to fail,
1785 contradicting the general specification for all authenticators. Instead it
1786 was generating a temporary error. It now behaves as specified.
bbe902f0 1787
ea3a6f44
NM
178831. The default ordering of permitted cipher suites for GnuTLS was pessimal
1789 (the order specifies the preference for clients). The order is now AES256,
1790 AES128, 3DES, ARCFOUR128.
bbe902f0 1791
ea3a6f44
NM
179231. Small patch to Sieve code - explicitly set From: when generating an
1793 autoreply.
bbe902f0 1794
ea3a6f44
NM
179532. Exim crashed if a remote delivery caused a very long error message to be
1796 recorded - for instance if somebody sent an entire SpamAssassin report back
1797 as a large number of 550 error lines. This bug was coincidentally fixed by
1798 increasing the size of one of Exim's internal buffers (big_buffer) that
1799 happened as part of the Exiscan merge. However, to be on the safe side, I
1800 have made the code more robust (and fixed the comments that describe what
1801 is going on).
bbe902f0 1802
ea3a6f44
NM
180333. Some experimental protocols are using DNS PTR records for new purposes. The
1804 keys for these records are domain names, not reversed IP addresses. The
1805 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
1806 leaves it alone. Component reversal etc. now happens only for IP addresses.
bbe902f0
PH
1807 CAN-2005-0021
1808
ea3a6f44
NM
180934. The host_aton() function is supposed to be passed a string that is known
1810 to be a valid IP address. However, in the case of IPv6 addresses, it was
1811 not checking this. This is a hostage to fortune. Exim now panics and dies
1812 if the condition is not met. A case was found where this could be provoked
1813 from a dnsdb PTR lookup with an IPv6 address that had more than 8
1814 components; fortuitously, this particular loophole had already been fixed
1815 by change 4.50/55 or 4.44/33 above.
1816
1817 If there are any other similar loopholes, the new check in host_aton()
1818 itself should stop them being exploited. The report I received stated that
1819 data on the command line could provoke the exploit when Exim was running as
1820 exim, but did not say which command line option was involved. All I could
1821 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
1822 running as the user.
bbe902f0
PH
1823 CAN-2005-0021
1824
ea3a6f44
NM
182535. There was a buffer overflow vulnerability in the SPA authentication code
1826 (which came originally from the Samba project). I have added a test to the
1827 spa_base64_to_bits() function which I hope fixes it.
bbe902f0
PH
1828 CAN-2005-0022
1829
ea3a6f44
NM
183036. The daemon start-up calls getloadavg() while still root for those OS that
1831 need the first call to be done as root, but it missed one case: when
1832 deliver_queue_load_max is set with deliver_drop_privilege. This is
1833 necessary for the benefit of the queue runner, because there is no re-exec
1834 when deliver_drop_privilege is set.
bbe902f0 1835
ea3a6f44
NM
183637. Caching of lookup data for "hosts =" ACL conditions, when a named host list
1837 was in use, was not putting the data itself into the right store pool;
1838 consequently, it could be overwritten for a subsequent message in the same
1839 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
1840 the caching.)
bbe902f0 1841
ea3a6f44
NM
184238. Sometimes the final signoff response after QUIT could fail to get
1843 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
1844 < 0 before doing a fflush(). This bug looks as though it goes back to the
1845 introduction of TLS in release 3.20, but "sometimes" must have been rare
1846 because the tests only now provoked it.
bbe902f0
PH
1847
1848
495ae4b0
PH
1849Exim version 4.43
1850-----------------
1851
1852 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
1853 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
1854 Now it returns an integer. A number of calls were still expecting a T/F
1855 return. Fortuitously, in all cases, the tests worked in OK situations,
1856 which is the norm. However, things would have gone wrong on any write
1857 failures on the smtp file descriptor. This function is used when sending
1858 messages over SMTP and also when doing verify callouts.
1859
1860 2. When Exim is called to do synchronous delivery of a locally submitted
1861 message (the -odf or -odi options), it no longer closes stderr before doing
1862 the delivery.
1863
1864 3. Implemented the mua_wrapper option.
1865
1866 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
1867
1868 5. Implemented the functions header_remove(), header_testname(),
1869 header_add_at_position(), and receive_remove_recipient(), and exported them
1870 to local_scan().
1871
1872 6. If an ACL "warn" statement specified the addition of headers, Exim already
1873 inserted X-ACL-Warn: at the start if there was no header name. However, it
1874 was not making this test for the second and subsequent header lines if
1875 there were newlines in the string. This meant that an invalid header could
1876 be inserted if Exim was badly configured.
1877
1878 7. Allow an ACL "warn" statement to add header lines at the start or after all
1879 the Received: headers, as well as at the end.
1880
1881 8. Added the rcpt_4xx retry error code.
1882
1883 9. Added postmaster_mailfrom=xxx to callout verification option.
1884
188510. Added mailfrom=xxxx to the callout verification option, for verify=
1886 header_sender only.
1887
188811. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
1889 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
1890
189112. Inserted some casts to stop certain compilers warning when using pointer
1892 differences as field lengths or precisions in printf-type calls (mostly
1893 affecting debugging statements).
1894
189513. Added optional readline() support for -be (dynamically loaded).
1896
189714. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
1898 same clock tick as a message's arrival, so that its received time was the
1899 same as the "first fail" time on the retry record, and that message
1900 remained on the queue past the ultimate address timeout, every queue runner
1901 would try a delivery (because it was past the ultimate address timeout) but
1902 after another failure, the ultimate address timeout, which should have then
1903 bounced the address, did not kick in. This was a "< instead of <=" error;
1904 in most cases the first failure would have been in the next clock tick
1905 after the received time, and all would be well.
1906
190715. The special items beginning with @ in domain lists (e.g. @mx_any) were not
1908 being recognized when the domain list was tested by the match_domain
1909 condition in an expansion string.
1910
191116. Added the ${str2b64: operator.
1912
191317. Exim was always calling setrlimit() to set a large limit for the number of
1914 processes, without checking whether the existing limit was already
1915 adequate. (It did check for the limit on file descriptors.) Furthermore,
1916 errors from getrlimit() and setrlimit() were being ignored. Now they are
1917 logged to the main and panic logs, but Exim does carry on, to try to do its
1918 job under whatever limits there are.
1919
192018. Imported PCRE 5.0.
1921
192219. Trivial typo in log message " temporarily refused connection" (the leading
1923 space).
1924
192520. If the log selector return_path_on_delivery was set and an address was
1926 redirected to /dev/null, the delivery process crashed because it assumed
1927 that a return path would always be set for a "successful" delivery. In this
1928 case, the whole delivery is bypassed as an optimization, and therefore no
1929 return path is set.
1930
193121. Internal re-arrangement: the function for sending a challenge and reading
1932 a response while authentication was assuming a zero-terminated challenge
1933 string. It's now changed to take a pointer and a length, to allow for
1934 binary data in such strings.
1935
193622. Added the cyrus_sasl authenticator (code supplied by MBM).
1937
193823. Exim was not respecting finduser_retries when seeking the login of the
1939 uid under which it was called; it was always trying 10 times. (The default
1940 setting of finduser_retries is zero.) Also, it was sleeping after the final
1941 failure, which is pointless.
1942
194324. Implemented tls_on_connect_ports.
1944
194525. Implemented acl_smtp_predata.
1946
194726. If the domain in control=submission is set empty, Exim assumes that the
1948 authenticated id is a complete email address when it generates From: or
1949 Sender: header lines.
1950
195127. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
1952 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
1953 chown and chgrp in /bin and hostname in /usr/bin.
1954
195528. Exim was keeping the "process log" file open after each use, just as it
1956 does for the main log. This opens the possibility of it remaining open for
1957 long periods when the USR1 signal hits a daemon. Occasional processlog
1958 errors were reported, that could have been caused by this. Anyway, it seems
1959 much more sensible not to leave this file open at all, so that is what now
1960 happens.
1961
196229. The long-running daemon process does not normally write to the log once it
1963 has entered its main loop, and it closes the log before doing so. This is
1964 so that log files can straightforwardly be renamed and moved. However,
1965 there are a couple of unusual error situations where the daemon does write
1966 log entries, and I had neglected to close the log afterwards.
1967
196830. The text of an SMTP error response that was received during a remote
1969 delivery was being truncated at 512 bytes. This is too short for some of
1970 the long messages that one sometimes sees. I've increased the limit to
1971 1024.
1972
197331. It is now possible to make retry rules that apply only when a message has a
1974 specific sender, in particular, an empty sender.
1975
197632. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
1977 it possible to be selective about when SMTP synchronization is enforced.
1978
197933. Added "control = caseful_local_part" and "control = "caselower_local_part".
1980
198132. Implemented hosts_connection_nolog.
1982
198333. Added an ACL for QUIT.
1984
198534. Setting "delay_warning=" to disable warnings was not working; it gave a
1986 syntax error.
1987
198835. Added mailbox_size and mailbox_filecount to appendfile.
1989
199036. Added control = no_multiline_responses to ACLs.
1991
199237. There was a bug in the logic of the code that waits for the clock to tick
1993 in the case where the clock went backwards by a substantial amount such
1994 that the microsecond fraction of "now" was more than the microsecond
1995 fraction of "then" (but the whole seconds number was less).
1996
199738. Added support for the libradius Radius client library this is found on
1998 FreeBSD (previously only the radiusclient library was supported).
1999
2000
2001Exim version 4.42
2002-----------------
2003
2004 1. When certain lookups returned multiple values in the form name=value, the
2005 quoting of the values was not always being done properly. Specifically:
2006 (a) If the value started with a double quote, but contained no whitespace,
2007 it was not quoted.
2008 (b) If the value contained whitespace other than a space character (i.e.
2009 tabs or newlines or carriage returns) it was not quoted.
2010 This fix has been applied to the mysql and pgsql lookups by writing a
2011 separate quoting function and calling it from the lookup code. The fix
2012 should probably also be applied to nisplus, ibase and oracle lookups, but
2013 since I cannot test any of those, I have not disturbed their existing code.
2014
2015 2. A hit in the callout cache for a specific address caused a log line with no
2016 reason for rejecting RCPT. Now it says "Previous (cached) callout
2017 verification failure".
2018
2019 3. There was an off-by-one bug in the queryprogram router. An over-long
2020 return line was truncated at 256 instead of 255 characters, thereby
2021 overflowing its buffer with the terminating zero. As well as fixing this, I
2022 have increased the buffer size to 1024 (and made a note to document this).
2023
2024 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
2025 when Exim is waiting for an SMTP response from a remote server, Exim
2026 restarts its select() call on the socket, thereby resetting its timeout.
2027 This is not a problem when such interrupts are rare. Somebody set up a cron
2028 job to run exiwhat every 2 minutes, which is less than the normal select()
2029 timeout (5 or 10 minutes). This meant that the select() timeout never
2030 kicked in because it was always reset. I have fixed this by comparing the
2031 time when an interrupt arrives with the time at the start of the first call
2032 to select(). If more time than the timeout has elapsed, the interrupt is
2033 treated as a timeout.
2034
2035 5. Some internal re-factoring in preparation for the addition of Sieve
2036 extensions (by MH). In particular, the "personal" test is moved to a
2037 separate function, and given an option for scanning Cc: and Bcc: (which is
2038 not set for Exim filters).
2039
2040 6. When Exim created an email address using the login of the caller as the
2041 local part (e.g. when creating a From: or Sender: header line), it was not
2042 quoting the local part when it contained special characters such as @.
2043
2044 7. Installed new OpenBSD configuration files.
2045
2046 8. Reworded some messages for syntax errors in "and" and "or" conditions to
2047 try to make them clearer.
2048
2049 9. Callout options, other than the timeout value, were being ignored when
2050 verifying sender addresses in header lines. For example, when using
2051
2052 verify = header_sender/callout=no_cache
2053
2054 the cache was (incorrectly) being used.
2055
205610. Added a missing instance of ${EXE} to the exim_install script; this affects
2057 only the Cygwin environment.
2058
205911. When return_path_on_delivery was set as a log selector, if different remote
2060 addresses in the same message used different return paths and parallel
2061 remote delivery occurred, the wrong values would sometimes be logged.
2062 (Whenever a remote delivery process finished, the return path value from
2063 the most recently started remote delivery process was logged.)
2064
206512. RFC 3848 specifies standard names for the "with" phrase in Received: header
2066 lines when AUTH and/or TLS are in use. This is the "received protocol"
2067 field. Exim used to use "asmtp" for authenticated SMTP, without any
2068 indication (in the protocol name) for TLS use. Now it follows the RFC and
2069 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
2070 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
2071 names appear in log lines as well as in Received: header lines.
2072
207313. Installed MH's patches for Sieve to add the "copy" and "vacation"
2074 extensions, and comparison tests, and to fix some bugs.
2075
207614. Changes to the "personal" filter test:
2077
2078 (1) The test was buggy in that it was just doing the equivalent of
2079 "contains" tests on header lines. For example, if a user's address was
2080 anne@some.where, the "personal" test would incorrectly be true for
2081
2082 To: susanne@some.where
2083
2084 This test is now done by extracting each address from the header in turn,
2085 and checking the entire address. Other tests that are part of "personal"
2086 are now done using regular expressions (for example, to check local parts
2087 of addresses in From: header lines).
2088
2089 (2) The list of non-personal local parts in From: addresses has been
2090 extended to include "listserv", "majordomo", "*-request", and "owner-*",
2091 taken from the Sieve specification recommendations.
2092
2093 (3) If the message contains any header line starting with "List-" it is
2094 treated as non-personal.
2095
2096 (4) The test for "circular" in the Subject: header line has been removed
2097 because it now seems ill-conceived.
2098
209915. Minor typos in src/EDITME comments corrected.
2100
210116. Installed latest exipick from John Jetmore.
2102
210317. If headers_add on a router specified a text string that was too long for
2104 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
2105 of string_sprintf() is now avoided.
2106
210718. $message_body_size was not set (it was always zero) when running the DATA
2108 ACL and the local_scan() function.
2109
211019. For the "mail" command in an Exim filter, no default was being set for
2111 the once_repeat time, causing a random time value to be used if "once" was
2112 specified. (If the value happened to be <= 0, no repeat happened.) The
2113 default is now 0s, meaning "never repeat". The "vacation" command was OK
2114 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
2115 (I found it when inspecting the code).
2116
211720. There is now an overall timeout for performing a callout verification. It
2118 defaults to 4 times the callout timeout, which applies to individual SMTP
2119 commands during the callout. The overall timeout applies when there is more
2120 than one host that can be tried. The timeout is checked before trying the
2121 next host. This prevents very long delays if there are a large number of
2122 hosts and all are timing out (e.g. when the network connections are timing
2123 out). The value of the overall timeout can be changed by specifying an
2124 additional sub-option for "callout", called "maxwait". For example:
2125
2126 verify = sender/callout=5s,maxwait=20s
2127
212821. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
2129 to the end before writing, but this should make it even safer).
2130
213122. Exim was forgetting that it had advertised PIPELINING for the second and
2132 subsequent messages on an SMTP connection. It was also not resetting its
2133 memory on STARTTLS and an internal HELO.
2134
213523. When Exim logs an SMTP synchronization error within a session, it now
2136 records whether PIPELINING has been advertised or not.
2137
213824. Added 3 instances of "(long int)" casts to time_t variables that were being
2139 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
2140 rather than long int.
2141
214225. Installed the latest Cygwin configuration files from the Cygwin maintainer.
2143
214426. Added the never_mail option to autoreply.
2145
2146
2147Exim version 4.41
2148-----------------
2149
2150 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
2151 crash if the getsockname() call failed; this can happen if a connection is
2152 closed very soon after it is established. The problem was simply in the
2153 order in which certain operations were done, causing Exim to try to write
2154 to the SMTP stream before it had set up the file descriptor. The bug has
2155 been fixed by making things happen in the correct order.
2156
2157
2158Exim version 4.40
2159-----------------
2160
2161 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
2162 before the connection was closed, thus losing the rejection response.
2163
2164 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
2165 some early Solaris releases, but causes trouble in current releases where
2166 socklen_t is defined.
2167
2168 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
2169 always exist.
2170
2171 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
2172 configured.
2173
2174 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
2175 that releasing the top of it at the end releases what was used for sub-
2176 expansions (unless the block got too big). However, discard this block if
2177 the first thing is a variable or header, so that we can use its block when
2178 it is dynamic (useful for very large $message_headers, for example).
2179
2180 6. Lookups now cache *every* query, not just the most recent. A new, separate
2181 store pool is used for this. It can be recovered when all lookup caches are
2182 flushed. Lookups now release memory at the end of their result strings.
2183 This has involved some general refactoring of the lookup sources.
2184
2185 7. Some code has been added to the store_xxx() functions to reduce the amount
2186 of flapping under certain conditions.
2187
2188 8. log_incoming_interface used to affect only the <= reception log lines. Now
2189 it causes the local interface and port to be added to several more SMTP log
2190 lines, for example "SMTP connection from", and rejection lines.
2191
2192 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
2193
219410. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
2195
219611. If $host_data was set by virtue of a hosts lookup in an ACL, its value
2197 could be overwritten at the end of the current message (or the start of a
2198 new message if it was set in a HELO ACL). The value is now preserved for
2199 the duration of the SMTP connection.
2200
220112. If a transport had a headers_rewrite setting, and a matching header line
2202 contained an unqualified address, that address was qualified, even if it
2203 did not match any rewriting rules. The underlying bug was that the values
2204 of the flags that permit the existence of unqualified sender and recipient
2205 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
2206 non-local messages, and by -bnq for local messages) were not being
2207 preserved with the message after it was received.
2208
220913. When Exim was logging an SMTP synchronization error, it could sometimes log
2210 "next input=" as part of the text comprising the host identity instead of
2211 the correct text. The code was using the same buffer for two different
2212 strings. However, depending on which order the printing function evaluated
2213 its arguments, the bug did not always show up. Under Linux, for example, my
2214 test suite worked just fine.
2215
221614. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
2217 doesn't work with some older versions of Perl. It has been changed to "my",
2218 which in any case is probably the better facility to use.
2219
222015. A really picky compiler found some instances of statements for creating
2221 error messages that either had too many or two few arguments for the format
2222 string.
2223
222416. The size of the buffer for calls to the DNS resolver has been increased
2225 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
2226 for addresses that have a lot of PTR records. This alleviates a problem; it
2227 does not fully solve it.
2228
222917. A dnsdb lookup for PTR records that receives more data than will fit in the
2230 buffer now truncates the list and logs the incident, which is the same
2231 action as happens when Exim is looking up a host name and its aliases.
2232 Previously in this situation something unpredictable would happen;
2233 sometimes it was "internal error: store_reset failed".
2234
223518. If a server dropped the connection unexpectedly when an Exim client was
2236 using GnuTLS and trying to read a response, the client delivery process
2237 crashed while trying to generate an error log message.
2238
223919. If a "warn" verb in an ACL added multiple headers to a message in a single
2240 string, for example:
2241
2242 warn message = H1: something\nH2: something
2243
2244 the text was added as a single header line from Exim's point of view
2245 though it ended up OK in the delivered message. However, searching for the
2246 second and subsequent header lines using $h_h2: did not work. This has been
2247 fixed. Similarly, if a system filter added multiple headers in this way,
2248 the routers could not see them.
2249
225020. Expanded the error message when iplsearch is called with an invalid key to
2251 suggest using net-iplsearch in a host list.
2252
225321. When running tests using -bh, any delays imposed by "delay" modifiers in
2254 ACLs are no longer actually imposed (and a message to that effect is
2255 output).
2256
225722. If a "gecos" field in a passwd entry contained escaped characters, in
2258 particular, if it contained a \" sequence, Exim got it wrong when building
2259 a From: or a Sender: header from that name. A second bug also caused
2260 incorrect handling when an unquoted " was present following a character
2261 that needed quoting.
2262
226323. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
2264 was not being matched caselessly.
2265
226624. Arranged for all hyphens in the exim.8 source to be escaped with
2267 backslashes.
2268
226925. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
2270 properly. Recipient callout cache records were still being keyed to include
2271 the sender, even when use_sender was set false. This led to far more
2272 callouts that were necessary. The sender is no longer included in the key
2273 when use_sender is false.
2274
227526. Added "control = submission" modifier to ACLs.
2276
227727. Added the ${base62d: operator to decode base 62 numbers.
2278
227928. dnsdb lookups can now access SRV records.
2280
228129. CONFIGURE_OWNER can be set at build time to define an alternative owner for
2282 the configuration file.
2283
228430. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
2285 (-v) mode. This makes the output for a verbose queue run more intelligible.
2286
228731. Added a use_postmaster feature to recipient callouts.
2288
228932. Added the $body_zerocount variable, containing the number of binary zero
2290 bytes in the message body.
2291
229233. The time of last modification of the "new" subdirectory is now used as the
2293 "mailbox time last read" when there is a quota error for a maildir
2294 delivery.
2295
229634. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
2297
229835. Added +ignore_unknown as a special item in host lists.
2299
230036. Code for decoding IPv6 addresses in host lists is now included, even if
2301 IPv6 support is not being compiled. This fixes a bug in which an IPv6
2302 address was recognized as an IP address, but was then not correctly decoded
2303 into binary, causing unexpected and incorrect effects when compared with
2304 another IP address.
2305
2306
2307Exim version 4.34
2308-----------------
2309
2310 1. Very minor rewording of debugging text in manualroute to say "list of
2311 hosts" instead of "hostlist".
2312
2313 2. If verify=header_syntax was set, and a header line with an unqualified
2314 address (no domain) and a large number of spaces between the end of the
2315 name and the colon was received, the reception process suffered a buffer
2316 overflow, and (when I tested it) crashed. This was caused by some obsolete
2317 code that should have been removed. The fix is to remove it!
2318
2319 3. When running in the test harness, delay a bit after writing a bounce
2320 message to get a bit more predictability in the log output.
2321
2322 4. Added a call to search_tidyup() just before forking a reception process. In
2323 theory, someone could use a lookup in the expansion of smtp_accept_max_
2324 per_host which, without the tidyup, could leave open a database connection.
2325
2326 5. Added the variables $recipient_data and $sender_data which get set from a
2327 lookup success in an ACL "recipients" or "senders" condition, or a router
2328 "senders" option, similar to $domain_data and $local_part_data.
2329
2330 6. Moved the writing of debug_print from before to after the "senders" test
2331 for routers.
2332
2333 7. Change 4.31/66 (moving the time when the Received: is generated) caused
2334 problems for message scanning, either using a data ACL, or using
2335 local_scan() because the Received: header was not generated till after they
2336 were called (in order to set the time as the time of reception completion).
2337 I have revised the way this works. The header is now generated after the
2338 body is received, but before the ACL or local_scan() are called. After they
2339 are run, the timestamp in the header is updated.
2340
2341
2342Exim version 4.33
2343-----------------
2344
2345 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
2346 before starting a queue runner without re-exec. This happened only when
2347 deliver_drop_privilege was set or when the Exim user was set to root. The
2348 effect of the bug was that timeouts during subsequent deliveries caused
2349 crashes instead of being properly handled. The handler is now left at its
2350 default (and expected) setting.
2351
2352 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
2353 message, again when deliver_drop_privilege is set or Exim is run as root.
2354 The bug described in (1) was not present in this case, but the tidying up
2355 of the other signals was missing. I have made the two cases consistent.
2356
2357 3. The ignore_target_hosts setting on a manualroute router was being ignored
2358 for hosts that were looked up using the /MX notation.
2359
2360 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
2361 in domain lists.
2362
2363 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
2364 operated on the sender address. After changing the $sender_address to <>
2365 for the sender address verify, Exim was re-instated it as the original
2366 (before rewriting) address, but remembering that it had rewritten it, so it
2367 wasn't rewriting it again. This bug also had the effect of breaking the
2368 sender address verification caching when the sender address was rewritten.
2369
2370 6. The ignore_target_hosts option was being ignored by the ipliteral router.
2371 This has been changed so that if the ip literal address matches
2372 ignore_target_hosts, the router declines.
2373
2374 7. Added expansion conditions match_domain, match_address, and match_local_
2375 part (NOT match_host).
2376
2377 8. The placeholder for the Received: header didn't have a length field set.
2378
2379 9. Added code to Exim itself and to exim_lock to test for a specific race
2380 condition that could lead to file corruption when using MBX delivery. The
2381 issue is with the lockfile that is created in /tmp. If this file is removed
2382 after a process has opened it but before that process has acquired a lock,
2383 there is the potential for a second process to recreate the file and also
2384 acquire a lock. This could lead to two Exim processes writing to the file
2385 at the same time. The added code performs the same test as UW imapd; it
2386 checks after acquiring the lock that its file descriptor still refers to
2387 the same named file.
2388
238910. The buffer for building added header lines was of fixed size, 8192 bytes.
2390 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
2391 when Exim is built.
2392
239311. Added the smtp_active_hostname option. If used, this will typically be made
2394 to depend on the incoming interface address. Because $interface_address is
2395 not set up until the daemon has forked a reception process, error responses
2396 that can happen earlier (such as "too many connections") no longer contain
2397 a host name.
2398
239912. If an expansion in a condition on a "warn" statement fails because a lookup
2400 defers, the "warn" statement is abandoned, and the next ACL statement is
2401 processed. Previously this caused the whole ACL to be aborted.
2402
240313. Added the iplsearch lookup type.
2404
240514. Added ident_timeout as a log selector.
2406
240715. Added tls_certificate_verified as a log selector.
2408
240916. Added a global option tls_require_ciphers (compare the smtp transport
2410 option of the same name). This controls incoming TLS connections.
2411
241217. I finally figured out how to make tls_require_ciphers do a similar thing
2413 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
2414 before starting the TLS session.
2415
241618. Tabs are now shown as \t in -bP output.
2417
241819. If the log selector return_path_on_delivery was set, Exim crashed when
2419 bouncing a message because it had too many Received: header lines.
2420
242120. If two routers both had headers_remove settings, and the first one included
2422 a superfluous trailing colon, the final name in the first list and the
2423 first name in the second list were incorrectly joined into one item (with a
2424 colon in the middle).
2425
2426
2427Exim version 4.32
2428-----------------
2429
2430 1. Added -C and -D options to the exinext utility, mainly to make it easier
2431 to include in the automated testing, but these could be helpful when
2432 multiple configurations are in use.
2433
2434 2. The exinext utility was not formatting the output nicely when there was
2435 an alternate port involved in the retry record key, nor when there was a
2436 message id as well (for retries that were specific to a specific message
2437 and a specific host). It was also confused by IPv6 addresses, because of
2438 the additional colons they contain. I have fixed the IPv4 problem, and
2439 patched it up to do a reasonable job for IPv6.
2440
2441 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
2442 delivery, the log line now contains "pipelined" if PIPELINING was used.
2443
2444 4. An SMTP transport process used to panic and die if the bind() call to set
2445 an explicit outgoing interface failed. This has been changed; it is now
2446 treated in the same way as a connect() failure.
2447
2448 5. A reference to $sender_host_name in the part of a conditional expansion
2449 that was being skipped was still causing a DNS lookup. This no longer
2450 occurs.
2451
2452 6. The def: expansion condition was not recognizing references to header lines
2453 that used bh_ and bheader_.
2454
2455 7. Added the _cache feature to named lists.
2456
2457 8. The code for checking quota_filecount in the appendfile transport was
2458 allowing one more file than it should have been.
2459
2460 9. For compatibility with Sendmail, the command line option
2461
2462 -prval:sval
2463
2464 is equivalent to
2465
2466 -oMr rval -oMs sval
2467
2468 and sets the incoming protocol and host name (for trusted callers). The
2469 host name and its colon can be omitted when only the protocol is to be set.
2470 Note the Exim already has two private options, -pd and -ps, that refer to
2471 embedded Perl. It is therefore impossible to set a protocol value of "d" or
2472 "s", but I don't think that's a major issue.
2473
247410. A number of refactoring changes to the code, none of which should affect
2475 Exim's behaviour:
2476
2477 (a) The number of logging options was getting close to filling up the
2478 32-bit word that was used as a bit map. I have split them into two classes:
2479 those that are passed in the argument to log_write(), and those that are
2480 only ever tested independently outside of that function. These are now in
2481 separate 32-bit words, so there is plenty of room for expansion again.
2482 There is no change in the user interface or the logging behaviour.
2483
2484 (b) When building, for example, log lines, the code previously used a
2485 macro that called string_cat() twice, in order to add two strings. This is
2486 not really sufficiently general. Furthermore, there was one instance where
2487 it was actually wrong because one of the argument was used twice, and in
2488 one call a function was used. (As it happened, calling the function twice
2489 did not affect the overall behaviour.) The macro has been replaced by a
2490 function that can join an arbitrary number of extra strings onto a growing
2491 string.
2492
2493 (c) The code for expansion conditions now uses a table and a binary chop
2494 instead of a serial search (which was left over from when there were very
2495 few conditions). Also, it now recognizes conditions like "pam" even when
2496 the relevant support is not compiled in: a suitably worded error message is
2497 given if an attempt is made to use such a condition.
2498
249911. Added ${time_interval:xxxxx}.
2500
250112. A bug was causing one of the ddress fields not to be passed back correctly
2502 from remote delivery subprocesses. The field in question was not being
2503 subsequently used, so this caused to problems in practice.
2504
250513. Added new log selectors queue_time and deliver_time.
2506
250714. Might have fixed a bug in maildirsizefile handling that threw up
2508 "unexpected character" debug warnings, and recalculated the data
2509 unnecessarily. In any case, I expanded the warning message to give more
2510 information.
2511
251215. Added the message "Restricted characters in address" to the statements in
2513 the default ACL that block characters like @ and % in local parts.
2514
251516. Change 71 for release 4.31 proved to be much less benign that I imagined.
2516 Three changes have been made:
2517
2518 (a) There was a serious bug; a negative response to MAIL caused the whole
2519 recipient domain to be cached as invalid, thereby blocking all messages
2520 to all local parts at the same domain, from all senders. This bug has
2521 been fixed. The domain is no longer cached after a negative response to
2522 MAIL if the sender used is not empty.
2523
2524 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
2525 been restored.
2526
2527 (c) A new callout option, "use_sender" has been added for people who want
2528 the modified behaviour.
2529
2530
2531Exim version 4.31
2532-----------------
2533
2534 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
2535 Larry Rosenman.
2536
2537 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
2538 indeed breaks things for older releases.
2539
2540 3. Added additional logging to the case where there is a problem reading data
2541 from a filter that is running in a subprocess using a pipe, in order to
2542 try to track down a specific problem.
2543
2544 4. Testing facility fudge: when running in the test harness and attempting
2545 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
2546 getting "No route to host". Convert this to a timeout.
2547
2548 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
2549 warning.
2550
2551 6. Some OS don't have socklen_t but use size_t instead. This affects the
2552 fifth argument of getsockopt() amongst other things. This is now
2553 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
2554 can be set for individual OS. I have set it for SunOS5, OSF1, and
2555 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
2556 some earlier ones do not.
2557
2558 7. Change 4.30/15 was not doing the test caselessly.
2559
2560 8. The standard form for an IPv6 address literal was being rejected by address
2561 parsing in, for example, MAIL and RCPT commands. An example of this kind of
2562 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
2563 this, as well as the form without the "IPv6" on the front (but only when
2564 address literals are enabled, of course).
2565
2566 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
2567
256810. Exim crashed if a message with an empty sender address specified by -f
2569 encountered a router with an errors_to setting. This could be provoked only
2570 by a command such as
2571
2572 exim -f "" ...
2573
2574 where an empty string was supplied; "<>" did not hit this bug.
2575
257611. Installed PCRE release 4.5.
2577
257812. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
2579 remained set. It is now erased.
2580
258113. exiqgrep wasn't working on MacOS X because it didn't correctly compute
2582 times from message ids (which are base 36 rather than the normal 62).
2583
258414. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
2585 were being counted as actual protocol errors, and logged if the log
2586 selector +smtp_protocol_error was set. One cannot be perfect in this test,
2587 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
2588 and DATA following a set of rejected RCPTs do not count as protocol errors.
2589 In other words, Exim assumes they were pipelined, though this may not
2590 actually be the case. Of course, in all cases the client gets an
2591 appropriate error code.
2592
259315. If a lookup fails in an ACL condition, a message about the failure may
2594 be available; it is used if testing the ACL cannot continue, because most
2595 such messages specify what the cause of the deferral is. However, some
2596 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
2597 that caused an old message to be retained and used if a later statement
2598 caused a defer, replacing the real cause of the deferral.
2599
260016. If an IP address had so many PTR records that the DNS lookup buffer
2601 was not large enough to hold them, Exim could crash while trying to process
2602 the truncated data. It now detects and logs this case.
2603
260417. Further to 4.21/58, another change has been made: if (and only if) the
2605 first line of a message (the first header line) ends with CRLF, a bare LF
2606 in a subsequent header line has a space inserted after it, so as not to
2607 terminate the header.
2608
260918. Refactoring: tidied an ugly bit of code in appendfile that copied data
2610 unnecessarily, used atoi() instead of strtol(), and didn't check the
2611 termination when getting file sizes from file names by regex.
2612
261319. Completely re-implemented the support for maildirsize files, in the light
2614 of a number of problems with the previous contributed implementation
2615 (4.30/29). In particular:
2616
2617 . If the quota is zero, the maildirsize file is maintained, but no quota is
2618 imposed.
2619
2620 . If the maildir directory does not exist, it is created before any attempt
2621 to write a maildirsize file.
2622
2623 . The quota value in the file is just a cache; if the quota is changed in
2624 the transport, the new value overrides.
2625
2626 . A regular expression is available for excluding directories from the
2627 count.
2628
262920. The autoreply transport checks the characters in options that define the
2630 message's headers; it allows continued headers, but it was checking with
2631 isspace() after an embedded newline instead of explicitly looking for a
2632 space or a tab.
2633
263421. If all the "regular" hosts to which an address was routed had passed their
2635 expiry times, and had not reached their retry times, the address was
2636 bounced, even if fallback hosts were defined. Now Exim should go on to try
2637 the fallback hosts.
2638
263922. Increased buffer sizes in the callout code from 1024 to 4096 to match the
2640 equivalent code in the SMTP transport. Some hosts send humungous responses
2641 to HELO/EHLO, more than 1024 it seems.
2642
264323. Refactoring: code in filter.c used (void *) for "any old type" but this
2644 gives compiler warnings in some environments. I've now done it "properly",
2645 using a union.
2646
264724. The replacement for inet_ntoa() that is used with gcc on IRIX systems
2648 (because of problems with the built-in one) was declared to return uschar *
2649 instead of char *, causing compiler failure.
2650
265125. Fixed a file descriptor leak when processing alias/forward files.
2652
265326. Fixed a minor format string issue in dbfn.c.
2654
265527. Typo in exim.c: ("dmbnz" for "dbmnz").
2656
265728. If a filter file refered to $h_xxx or $message_headers, and the headers
2658 contained RFC 2047 "words", Exim's memory could, under certain conditions,
2659 become corrupted.
2660
266129. When a sender address is verified, it is cached, to save repeating the test
2662 when there is more than one recipient in a message. However, when the
2663 verification involves a callout, it is possible for different callout
2664 options to be set for different recipients. It is too complicated to keep
2665 track of this in the cache, so now Exim always runs a verification when a
2666 callout is required, relying on the callout cache for the optimization.
2667 The overhead is duplication of the address routing, but this should not be
2668 too great.
2669
267030. Fixed a bug in callout caching. If a RCPT command caused the sender address
2671 to be verified with callout=postmaster, and the main callout worked but the
2672 postmaster check failed, the verification correctly failed. However, if a
2673 subsequent RCPT command asked for sender verification *without* the
2674 postmaster check, incorrect caching caused this verification also to fail,
2675 incorrectly.
2676
267731. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
2678 it was not caching the DNS options (qualify_single, search_parents) that
2679 were used when the lookup failed. A subsequent lookup with different
2680 options therefore always gave the same answer, though there were cases
2681 where it should not have. (Example: a "domains = !$mx_any" option on a
2682 dnslookup router: the "domains" option is always processed without any
2683 widening, but the router might have qualify_single set.) Now Exim uses the
2684 cached value only when the same options are set.
2685
268632. Added John Jetmore's "exipick" utility to the distribution.
2687
268833. GnuTLS: When an attempt to start a TLS session fails for any reason other
2689 than a timeout (e.g. a certificate is required, and is not provided), an
2690 Exim server now closes the connection immediately. Previously it waited for
2691 the client to close - but if the client is SSL, it seems that they each
2692 wait for each other, leading to a delay before one of them times out.
2693
269434: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
2695 maintained 0.8.x compatibility because I don't think many are using it, and
2696 it is clearly obsolete.
2697
269835. Added TLS support for CRLs: a tls_crl global option and one for the smtp
2699 transport.
2700
270136. OpenSSL: $tls_certificate_verified was being set to 1 even if the
2702 client certificate was expired. A simple patch fixes this, though I don't
2703 understand the full logic of why the verify callback is called multiple
2704 times.
2705
270637. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
2707 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
2708 which causes problems with some clients (such as the Certicom SSL Plus
2709 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
2710 disables the coutermeasure allowing Eudora to connect."
2711
271238. Exim was not checking that a write() to a log file succeeded. This could
2713 lead to Bad Things if a log got too big, in particular if it hit a file
2714 size limit. Exim now panics and dies if it cannot write to a log file, just
2715 as it does if it cannot open a log file.
2716
271739. Modified OS/Makefile-Linux so that it now contains
2718
2719 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
2720
2721 The two -D definitions ensure that Exim is compiled with large file
2722 support, which makes it possible to handle log files that are bigger than
2723 2^31.
2724
272540. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
2726 instance) a domain was checked against a named list that involved a lookup,
2727 causing $domain_data to be set, then another domain was checked against the
2728 same list, then the first domain was re-checked, the value of $domain_data
2729 after the final check could be wrong. In particular, if the second check
2730 failed, it could be set empty. This bug probably also applied to
2731 $localpart_data.
2732
273341. The strip_trailing_dot option was not being applied to the address given
2734 with the -f command-line option.
2735
273642. The code for reading a message's header from the spool was incrementing
2737 $received_count, but never initializing it. This meant that the value was
2738 incorrect (doubled) while delivering a message in the same process in which
2739 it was received. In the most common configuration of Exim, this never
2740 happens - a fresh exec is done - but it can happen when
2741 deliver_drop_privilege is set.
2742
274343. When Exim logs an SMTP synchronization error - client data sent too soon -
2744 it now includes up to 150 characters of the unexpected data in the log
2745 line.
2746
274744. The exim_dbmbuild utility uses fixed size buffers for reading input lines
2748 and building data strings. The size of both of these buffers was 10 000
2749 bytes - far larger than anybody would *ever* want, thought I. Needless to
2750 say, somebody hit the limit. I have increased the maximum line length to
2751 20 000 and the maximum data length of concatenated lines to 100 000. I have
2752 also fixed two bugs, because there was no checking on these buffers. Tsk,
2753 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
2754 buffer is too small.
2755
275645. The exim_dbmbuild utility did not support quoted keys, as Exim does in
2757 lsearch lookups. Now it does.
2758
275946. When parsing a route_list item in a manualroute router, a fixed-length
2760 buffer was used for the list of hosts. I made this 1024 bytes long,
2761 thinking that nobody would ever have a list of hosts that long. Wrong.
2762 Somebody had a whole pile of complicated expansion conditions, and the
2763 string was silently truncated, leading to an expansion error. It turns out
2764 that it is easier to change to an unlimited length (owing to other changes
2765 that have happened since this code was originally written) than to build
2766 structure for giving a limitation error. The length of the item that
2767 expands into the list of hosts is now unlimited.
2768
276947. The lsearch lookup could not handle data where the length of text line was
2770 more than 4095 characters. Such lines were truncated, leading to shortened
2771 data being returned. It should now handle lines of any length.
2772
277348. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
2774 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
2775 ACL").
2776
277749. Cosmetic tidy to scripts like exicyclog that are generated by globally
2778 replacing strings such as BIN_DIRECTORY in a source file: the replacement
2779 no longer happens in comment lines. A list of replacements is now placed
2780 at the head of all of the source files, except those whose only change is
2781 to replace PERL_COMMAND in the very first #! line.
2782
278350. Replaced the slow insertion sort in queue.c, for sorting the list of
2784 messages on the queue, with a bottom-up merge sort, using code contributed
2785 by Michael Haardt. This should make operations like -bp somewhat faster on
2786 large queues. It won't affect queue runners, except when queue_run_in_order
2787 is set.
2788
278951. Installed eximstats 1.31 in the distribution.
2790
279152. Added support for SRV lookups to the dnslookup router.
2792
279353. If an ACL referred to $message_body or $message_body_end, the value was not
2794 reset for any messages that followed in the same SMTP session.
2795
279654. The store-handling optimization for building very long strings was not
2797 differentiating between the different store pools. I don't think this
2798 actually made any difference in practice, but I've tidied it.
2799
280055. While running the routers to verify a sender address, $sender_address
2801 was still set to the sender address. This is wrong, because when routing to
2802 send a bounce to the sender, it would be empty. Therefore, I have changed
2803 it so that, while verifying a sender address, $sender_address is set to <>.
2804 (There is no change to what happens when verifying a recipient address.)
2805
280656. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
2807 target A or AAAA records (if not already returned) without resetting the
2808 qualify_single or search_parents options of the DNS resolver. These are
2809 inappropriate in this case because the targets of MX and SRV records must
2810 be FQDNs. A broken DNS record could cause trouble if it happened to have a
2811 target that, when qualified, matched something in the local domain. These
2812 two options are now turned off when doing these lookups.
2813
281457. It seems that at least some releases of Reiserfs (which does not have the
2815 concept of a fixed number of inodes) returns zero and not -1 for the
2816 number of available inodes. This interacted badly with check_spool_inodes,
2817 which assumed that -1 was the "no such thing" setting. What I have done is
2818 to check that the total number of inodes is greater than zero before doing
2819 the test of how many are available.
2820
282158. When a "warn" ACL statement has a log_message modifier, the message is
2822 remembered, and not repeated. This is to avoid a lot of repetition when a
2823 message has many recipients that cause the same warning to be written.
2824 Howewer, Exim was preserving the list of already written lines for an
2825 entire SMTP session, which doesn't seem right. The memory is now reset if a
2826 new message is started.
2827
282859. The "rewrite" debugging flag was not showing the result of rewriting in the
2829 debugging output unless log_rewrite was also set.
2830
283160. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
2832 of (int)(handle) when we know that handle contains (void *)(-1).
2833
283461. The Exim daemon panic-logs an error return when it closes the incoming
2835 connection. However "connection reset by peer" seems to be common, and
2836 isn't really an error worthy of noting specially, so that particular error
2837 is no long logged.
2838
283962. When Exim is trying to find all the local interfaces, it used to panic and
2840 die if the ioctl to get the interface flags failed. However, it seems that
2841 on at least one OS (Solaris 9) it is possible to have an interface that is
2842 included in the list of interfaces, but for which you get a failure error
2843 for this call. This happens when the interface is not "plumbed" into a
2844 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
2845 failure of the "get flags" call assumes that the interface is down.
2846
284763. Added a ${eval10: operator, which assumes all numbers are decimal. This
2848 makes life easier for people who are doing arithmetic on fields extracted
2849 from dates, where you often get leading zeros that should not be
2850 interpreted as octal.
2851
285264. Added qualify_domain to the redirect router, to override the global
2853 setting.
2854
285565. If a pathologically long header line contained very many addresses (the
2856 report of this problem mentioned 10 000) and each of them was rewritten,
2857 Exim could use up a very large amount of memory. (It kept on making new
2858 copies of the header line as it rewrote, and never released the old ones.)
2859 At the expense of a bit more processing, the header rewriting function has
2860 been changed so that it no longer eats memory in this way.
2861
286266. The generation of the Received: header has been moved from the time that a
2863 message starts to be received, to the time that it finishes. The timestamp
2864 in the Received: header should now be very close to that of the <= log
2865 line. There are two side-effects of this change:
2866
2867 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
2868 logged header lines no longer include the local Received: line, because
2869 it has not yet been created. The same applies to a copy of the message
2870 that is returned to a non-SMTP sender when a message is rejected.
2871
2872 (b) When a filter file is tested using -bf, no additional Received: header
2873 is added to the test message. After some thought, I decided that this
2874 is a bug fix.
2875
2876 This change does not affect the value of $received_for. It is still set
2877 after address rewriting, but before local_scan() is called.
2878
287967. Installed the latest Cygwin-specific files from the Cygwin maintainer.
2880
288168. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
2882 gave an unhelpful panic error message, and a defer error. I have managed to
2883 change this behaviour so that it now rejects any supplied certificate,
2884 which seems right, as the list of acceptable certificates is empty.
2885
288669. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
2887 gave an unhelpful defer error. I have not managed to make this reject any
2888 supplied certificates, but the error message it gives is "no certificate
2889 supplied", which is not helpful.
2890
289170. exigrep's output now also includes lines that are not associated with any
2892 message, but which match the given pattern. Implemented by a patch from
2893 Martin Sluka, which also tidied up the Perl a bit.
2894
289571. Recipient callout verification, like sender verification, was using <> in
2896 the MAIL FROM command. This isn't really the right thing, since the actual
2897 sender may affect whether the remote host accepts the recipient or not. I
2898 have changed it to use the actual sender in the callout; this means that
2899 the cache record is now keyed on a recipient/sender pair, not just the
2900 recipient address. There doesn't seem to be a real danger of callout loops,
2901 since a callout by the remote host to check the sender would use <>.
2902 [SEE ABOVE: changed after hitting problems.]
2903
290472. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
2905 temporary errors. However, in the case of such a code being given after
2906 the end of a data transmission (i.e. after ".") Exim was failing to write
2907 a retry record for the message. (Yes, there was some broken host that was
2908 actually sending 8xx at this point.)
2909
291073. An unknown lookup type in a host list could cause Exim to panic-die when
2911 the list was checked. (An example that provoked this was putting <; in the
2912 middle of a list instead of at the start.) If this happened during a DATA
2913 ACL check, a -D file could be left lying around. This kind of configuration
2914 error no longer causes Exim to die; instead it causes a defer errror. The
2915 incident is still logged to the main and panic logs.
2916
291774. Buglet left over from Exim 3 conversion. The message "too many messages
2918 in one connection" was written to the rejectlog but not the mainlog, except
2919 when address rewriting (yes!) was being logged.
2920
292175. Added write_rejectlog option.
2922
292376. When a system filter was run not as root (that is, when system_filter_user
2924 was set), the values of the $n variables were not being returned to the
2925 main process; thus, they were not subsequently available in the $sn
2926 variables.
2927
292877. Added +return_path_on_delivery log selector.
2929
293078. A connection timeout was being treated differently from recipients deferred
2931 when testing hosts_max_try with a message that was older than the host's
2932 retry timeout. (The host should not be counted, thus allowing all hosts to
2933 be tried at least once before bouncing.) This may have been the cause of an
2934 occasionally reported bug whereby a message would remain on the queue
2935 longer than the retry timeout, but would be bounced if a delivery was
2936 forced. I say "may" because I never totally pinned down the problem;
2937 setting up timeout/retry tests is difficult. See also the next item.
2938
293979. The ultimate address timeout was not being applied to errors that involved
2940 a combination of host plus message (for example, a timeout on a MAIL
2941 command). When an address resolved to a number of possible hosts, and they
2942 were not all tried for each delivery (e.g. because of hosts_max_try), a
2943 message could remain on the queue longer than the retry timeout.
2944
294580. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
2946 Haardt.
2947
294881. Fixed an obscure SMTP outgoing bug which required at least the following
2949 conditions: (a) there was another message waiting for the same server;
2950 (b) the server returned 5xx to all RCPT commands in the first message so
2951 that the message was not completed; (c) the server dropped the connection
2952 or gave a negative response to the RSET that Exim sends to abort the
2953 transaction. The observed case was a dropped connection after DATA that had
2954 been sent in pipelining mode. That is, the server had advertised PIPELINING
2955 but was not implementing it correctly. The effect of the bug was incorrect
2956 behaviour, such as trying another host, and this could lead to a crash.
2957
2958
2959Exim version 4.30
2960-----------------
2961
2962 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
2963 and daemon.c were passed as pointers to ints; they should have been
2964 pointers to socklen_t variables (which are typically unsigned ints).
2965
2966 2. Some signed/unsigned type warnings in the os.c file for Linux have been
2967 fixed.
2968
2969 3. Fixed a really odd bug that affected only the testing scheme; patching a
2970 certain fixed string in the binary changed the value of another string that
2971 happened to be identical to the end of the original first string.
2972
2973 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
2974 name", it returns that address as the IP address. On some operating
2975 systems (e.g. Solaris), it also passes back the IP address string as the
2976 "host name". However, on others (e.g. Linux), it passes back an empty
2977 string. Exim wasn't checking for this, and was changing the host name to an
2978 empty string, assuming it had been canonicized.
2979
2980 5. Although rare, it is permitted to have more than one PTR record for a given
2981 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
2982 all the names associated with an address, because they do in Solaris.
2983 However, it seems that they do not in Linux for data that comes from the
2984 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
2985 I found this out when I moved to a new Linux workstation and tried to run
2986 the Exim test suite.
2987
2988 To get round this problem I have changed the code so that it now does its
2989 own call to the DNS to look up PTR records when searching for a host name.
2990 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
2991 addresses that are only in /etc/hosts are still found.
2992
2993 This behaviour is, however, controlled by an option called host_lookup_
2994 order, which defaults to "bydns:byaddr". If people want to use the other
2995 order, or indeed, just use one or the other means of lookup, they can
2996 specify it in this variable.
2997
2998 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
2999 some operating systems, this comes back from gethostbyaddr() as an empty
3000 string, and this is what Exim used to test for. However, it seems that in
3001 other systems, "." is yielded. Exim now tests for this case too.
3002
3003 7. The values of check_spool_space and check_log_space are now held internally
3004 as a number of kilobytes instead of an absolute number of bytes. If a
3005 numbers is specified without 'K' or 'M', it is rounded up to the nearest
3006 kilobyte. This means that much larger values can be stored.
3007
3008 8. Exim monitor: an attempt to get the action menu when not actually pointing
3009 at a message produces an empty menu entitled "No message selected". This
3010 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
3011 no entries in it ("Shell widget menu has zero width and/or height"). So I
3012 have added a single, blank menu entry in this case.
3013
3014 9. Added ${quote_local_part.
3015
301610. MIME decoding is now applied to the contents of Subject: header lines when
3017 they are logged.
3018
301911. Now that a reference to $sender_host_address automatically causes a reverse
3020 lookup to occur if necessary (4.13/18), there is no need to arrange for a
3021 host lookup before query-style lookups in lists that might use this
3022 variable. This has therefore been abolished, and the "net-" prefix is no
3023 longer necessary for query-style lookups.
3024
302512. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
3026 have been a typo for LFLAGS, so it has been changed.
3027
302813. The install script calls Exim with "-C /dev/null" in order to find the
3029 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
3030 to be output. Howeve, since Exim outputs its version number before the
3031 error, it didn't break the script. It just looked ugly. I fixed this by
3032 always allowing "-C /dev/null" if the caller is root.
3033
303414. Ignore overlarge ACL variable number when reading spool file - insurance
3035 against a later release with more variables having written the file.
3036
303715. The standard form for an IPv6 address literal was being rejected by EHLO.
3038 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
3039 this, as well as the form without the "IPv6" on the front.
3040
304116. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
3042 OS/Makefile-Darwin file.
3043
304417. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
3045 only to LDAP libraries that do not have LDAP_OPT_DEREF.
3046
304718. After change 4.21/52, "%ld" was used to format the contents of the $inode
3048 variable. However, some OS use ints for inodes. I've added cast to long int
3049 to get rid of the compiler warning.
3050
305119. I had forgotten to lock out "/../" in configuration file names when
3052 ALT_CONFIG_PREFIX was set.
3053
305420. Routers used for verification do not need to specify transports. However,
3055 if such a router generated a host list, and callout was configured, Exim
3056 crashed, because it could not find a port number from the (non-existent)
3057 transport. It now assumes port 25 in this circumstance.
3058
305921. Added the -t option to exigrep.
3060
306122. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
3062 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
3063 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
3064 them is set, LOOKUP_LSEARCH is forced.
3065
306623. "exim -bV" now outputs a list of lookups that are included in the binary.
3067
306824. Added sender and host information to the "rejected by local_scan()" log
3069 line; previously there was no indication of these.
3070
307125. Added .include_if_exists.
3072
307326. Change 3.952/11 added an explicit directory sync on top of a file sync for
3074 Linux. It turns out that not all file systems support this. Apparently some
3075 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
3076 do it.) To cope with this, the error EINVAL, which means that sync-ing is
3077 not supported on the file descriptor, is now ignored when Exim is trying to
3078 sync a directory. This applies only to Linux.
3079
308027. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
3081
308228. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
3083 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
3084 this.
3085
308629. Added support for maildirsize files from supplied patch (modified a bit).
3087
308830. The use of :fail: followed by an empty string could lead Exim to respond to
3089 sender verification failures with (e.g.):
3090
3091 550 Verification failed for <xxx>
3092 550 Sender verify failed
3093
3094 where the first response line was missing the '-' that indicates it is not
3095 the final line of the response.
3096
309731. The loop for finding the name of the user that called Exim had a hardwired
3098 limit of 10; it now uses the value of finduser_retries, which is used for
3099 all other user lookups.
3100
310132. Added $received_count variable, available in data and not_smtp ACLs, and at
3102 delivery time.
3103
310433. Exim was neglecting to zero errno before one call of strtol() when
3105 expanding a string and expecting an integer value. On some systems this
3106 resulted in spurious "integer overflow" errors. Also, it was casting the
3107 result into an int without checking.
3108
310934. Testing for a connection timeout using "timeout_connect" in the retry rules
3110 did not work. The code looks as if it has *never* worked, though it appears
3111 to have been documented since at least releast 1.62. I have made it work.
3112
311335. The "timeout_DNS" error in retry rules, also documented since at least
3114 1.62, also never worked. As it isn't clear exactly what this means, and
3115 clearly it isn't a major issue, I have abolished the feature by treating it
3116 as "timeout", and writing a warning to the main and panic logs.
3117
311836. The display of retry rules for -brt wasn't always showing the error code
3119 correctly.
3120
312137. Added new error conditions to retry rules: timeout_A, timeout_MX,
3122 timeout_connect_A, timeout_connect_MX.
3123
312438. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
3125 to the empty sender.
3126
312739. The daemon was not analysing the content of -oX till after it had closed
3128 stderr and disconnected from the controlling terminal. This meant that any
3129 syntax errors were only noted on the panic log, and the return code from
3130 the command was 0. By re-arranging the code a little, I've made the
3131 decoding happen first, so such errors now appear on stderr, and the return
3132 code is 1. However, the actual setting up of the sockets still happens in
3133 the disconnected process, so errors there are still only recorded on the
3134 panic log.
3135
313640. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
3137 connections (as happens on some IP stacks) was logged at start up time as
3138 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
3139 it from "IPv6 and IPv4", which means that two separate sockets are being
3140 used.
3141
314241. The debug output for gethostbyname2() or getipnodebyname() failures now
3143 says whether AF_INET or AF_INET6 was passed as an argument.
3144
314542. Exiwhat output was messed up when time zones were included in log
3146 timestamps.
3147
314843. Exiwhat now gives more information about the daemon's listening ports,
3149 and whether -tls-on-connect was used.
3150
315144. The "port" option of the smtp transport is now expanded.
3152
315345. A "message" modifier in a "warn" statement in a non-message ACL was being
3154 silently ignored. Now an error message is written to the main and panic
3155 logs.
3156
315746. There's a new ACL modifier called "logwrite" which writes to a log file
3158 as soon as it is encountered.
3159
316047. Added $local_user_uid and $local_user_gid at routing time.
3161
316248. Exim crashed when trying to verify a sender address that was being
3163 rewritten to "<>".
3164
316549. Exim was recognizing only a space character after ".include". It now also
3166 recognizes a tab character.
3167
316850. Fixed several bugs in the Perl script that creates the exim.8 man page by
3169 extracting the relevant information from the specification. The man page no
3170 longer contains scrambled data for the -d option, and I've added a section
3171 at the front about calling Exim under different names.
3172
317351. Added "extra_headers" argument to the "mail" command in filter files.
3174
317552. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
3176 crash.
3177
317853. Installed eximstats 1.29.
3179
318054. Added transport_filter_timeout as a generic transport option.
3181
318255. Exim no longer adds an empty Bcc: header to messages that have no To: or
3183 Cc: header lines. This was required by RFC 822, but it not required by RFC
3184 2822.
3185
318656. Exim used to add From:, Date:, and Message-Id: header lines to any
3187 incoming messages that did not have them. Now it does so only if the
3188 message originates locally, that is, if there is no associated remote host
3189 address. When Resent- header lines are present, this applies to the Resent-
3190 lines rather than the non-Resent- lines.
3191
319257. Drop incoming SMTP connection after too many syntax or protocol errors. The
3193 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
3194
319558. Messages for configuration errors now include the name of the main
3196 configuration file - useful now that there may be more than one file in a
3197 list (.included file names were always shown).
3198
319959. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
3200 for those rare installations that do not start the daemon as root or run it
3201 setuid root. I've cut out the call to initgroups() if the daemon is not
3202 root at that time.
3203
320460. The Exim user and group can now be bound into the binary as text strings
3205 that are looked up at the start of Exim's processing.
3206
320761. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
3208
320962. Added $mailstore_basename variable.
3210
321163. Installed patch to sieve.c from Michael Haardt.
3212
321364. When Exim failed to open the panic log after failing to open the main log,
3214 the original message it was trying to log was written to stderr and debug
3215 output, but if they were not available (the usual case in production), it
3216 was lost. Now it is written to syslog before the two lines that record the
3217 failures to open the logs.
3218
321965. Users' Exim filters run in subprocesses under the user's uid. It is
3220 possible for a "deliver" command or an alias in a "personal" command to
3221 provoke an address rewrite. If logging of address rewriting is configured,
3222 this fails because the process is not running as root or exim. There may be
3223 a better way of dealing with this, but for the moment (because 4.30 needs
3224 to be released), I have disabled address rewrite logging when running a
3225 filter in a non-root, non-exim process.
3226
3227
3228Exim version 4.24
3229-----------------
3230
3231 1. The buildconfig auxiliary program wasn't quoting the value set for
3232 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
3233 not defined. This bug was masked in 4.22 by the effect that was fixed in
3234 change 4.23/1.
3235
3236 2. Some messages that were rejected after a message id was allocated were
3237 shown as "incomplete" by exigrep. It no longer does this for messages that
3238 are rejected by local_scan() or the DATA or non-SMTP ACLs.
3239
3240 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
3241 have allow_domain_literals set, the ID did not get logged in the <= line.
3242 Domain literals are now always recognized in Message-ID: header lines.
3243
3244 4. The first argument for a ${extract expansion item is the key name or field
3245 number. Leading and trailing spaces in this item were not being ignored,
3246 causing some misleading effects.
3247
3248 5. When deliver_drop_privilege was set, single queue runner processes started
3249 manually (i.e. by the command "exim -q") or by the daemon (which uses the
3250 same command in the process it spins off) were not dropping privilege.
3251
3252 6. When the daemon running as "exim" started a queue runner, it always
3253 re-executed Exim in the spun-off process. This is a waste of effort when
3254 deliver_drop_privilege is set. The new process now just calls the
3255 queue-runner function directly.
3256
3257
3258Exim version 4.23
3259-----------------
3260
3261 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
3262 HEADERS_CHARSET.
3263
3264 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
3265 ignored. Though the use of -oP was forcing the writing of a pid file, it
3266 was always written to the default place.
3267
3268 3. If the message "no IP address found for host xxxx" is generated during
3269 incoming verification, it is now followed by identification of the incoming
3270 connection (so you can more easily find what provoked it).
3271
3272 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
3273
3274 5. Added some features to "harden" Exim a bit more against certain attacks:
3275
3276 (a) There is now a build-time option called FIXED_NEVER_USERS that can
3277 be put in Local/Makefile. This is like the never_users runtime option,
3278 but it cannot be overridden. The default setting is "root".
3279
3280 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
3281 prefix string with which any file named in a -C command line option
3282 must start.
3283
3284 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
3285 is retained for -C and -D only if the caller of Exim is root. Without
3286 it, the exim user may also use -C and -D and retain privilege.
3287
3288 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
3289 command line option is disabled.
3290
3291 6. Macro names set by the -D option must start with an upper case letter, just
3292 like macro names defined in the configuration file.
3293
3294 7. Added "dereference=" facility to LDAP.
3295
3296 8. Two instances of the typo "uknown" in the source files are fixed.
3297
3298 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
3299 the Configure-Makefile script screwed up while processing it.
3300
330110. Incorporated PCRE 4.4.
3302
330311. The SMTP synchronization check was not operating right at the start of an
3304 SMTP session. For example, it could not catch a HELO sent before the client
3305 waited for the greeting. There is now a check for outstanding input at the
3306 point when the greeting is written. Because of the duplex, asynchronous
3307 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
3308 way, but not yet received, when the check is performed.
3309
331012. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
3311 on TCP/IP sockets, because this apparently causes some broken clients to
3312 timeout.
3313
331413. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
3315 unchanged) from the Cygwin maintainer.
3316
331714. The code for -bV that shows what is in the binary showed "mbx" when maildir
3318 was supported instead of testing for mbx. Effectively a typo.
3319
332015. The spa authenticator server code was not checking that the input it
3321 received was valid base64.
3322
332316. The debug output line for the "set" modifier in ACLs was not showing the
3324 name of the variable that was being set.
3325
332617. Code tidy: the variable type "vtype_string" was never used. Removed it.
3327
332818. Previously, a reference to $sender_host_name did not cause a DNS reverse
3329 lookup on its own. Something else was needed to trigger the lookup. For
3330 example, a match in host_lookup or the need for a host name in a host list.
3331 Now, if $sender_host_name is referenced and the host name has not yet been
3332 looked up, a lookup is performed. If the lookup fails, the variable remains
3333 empty, and $host_lookup_failed is set to "1".
3334
333519. Added "eqi" as a case-independent comparison operator.
3336
333720. The saslauthd authentication condition could segfault if neither service
3338 nor realm was specified.
3339
334021. If an overflowing value such as "2048M" was set for message_size_limit, the
3341 error message that was logged was misleading, and incoming SMTP
3342 connections were dropped. The message is now more accurate, and temporary
3343 errors are given to SMTP connections.
3344
334522. In some error situations (such as 21 above) Exim rejects all SMTP commands
3346 (except RSET) with a 421 error, until QUIT is received. However, it was
3347 failing to send a response to QUIT.
3348
334923. The HELO ACL was being run before the code for helo_try_verify_hosts,
3350 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
3351 ACL is now run after the helo_try_verify_hosts code.
3352
335324. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
3354 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
3355 case-independent, so other case variants are also recognized). Apparently
3356 some systems use these upper case variants.
3357
335825. If more than two messages were waiting for the same host, and a transport
3359 filter was specified for the transport, Exim sent two messages over the
3360 same TCP/IP connection, and then failed with "socket operation on non-
3361 socket" when it tried to send the third.
3362
336326. Added Exim::debug_write and Exim::log_write for embedded Perl use.
3364
336527. The extern definition of crypt16() in expand.c was not being excluded when
3366 the OS had its own crypt16() function.
3367
336828. Added bounce_return_body as a new option, and bounce_return_size_limit
3369 as a preferred synonym for return_size_limit, both as an option and as an
3370 expansion variable.
3371
337229. Added LIBS=-liconv to OS/Makefile-OSF1.
3373
337430. Changed the default configuration ACL to relax the local part checking rule
3375 for addresses that are not in any local domains. For these addresses,
3376 slashes and pipe symbols are allowed within local parts, but the sequence
3377 /../ is explicitly forbidden.
3378
337931. SPA server authentication was not clearing the challenge buffer before
3380 using it.
3381
338232. log_message in a "warn" ACL statement was writing to the reject log as
3383 well as to the main log, which contradicts the documentation and doesn't
3384 seem right (because no rejection is happening). So I have stopped it.
3385
338633. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
3387 However, I am unable to do any testing of this.
3388
338934. Fixed an infelicity in the appendfile transport. When checking directories
3390 for a mailbox, to see if any needed to be created, it was accidentally
3391 using path names with one or more superfluous leading slashes; tracing
3392 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
3393
339435. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
3395 contents are added to the log line that is written for every discarded
3396 recipient. (Previously a log_message setting was ignored.)
3397
339836. The ${quote: operator now quotes the string if it is empty.
3399
340037. The install script runs exim in order to find its version number. If for
3401 some reason other than non-existence or emptiness, which it checks, it
3402 could not run './exim', it was installing it with an empty version number,
3403 i.e. as "exim-". This error state is now caught, and the installation is
3404 aborted.
3405
340638. An argument was missing from the function that creates an error message
3407 when Exim fails to connect to the socket for saslauthd authentication.
3408 This could cause Exim to crash, or give a corrupted message.
3409
341039. Added isip, isip4, and isip6 to ${if conditions.
3411
341240. The ACL variables $acl_xx are now saved with the message, and can be
3413 accessed later in routers, transports, and filters.
3414
341541. The new lookup type nwildlsearch is like wildlsearch, except that the key
3416 strings in the file are not string-expanded.
3417
341842. If a MAIL command specified a SIZE value that was too large to fit into an
3419 int variable, the check against message_size_limit failed. Such values are
3420 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
3421 day this will have to be increased, but I don't think I want to be around
3422 when emails are that large.
3423
3424
3425
3426Exim version 4.22
3427-----------------
3428
3429 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
3430 iconv() is not standard in FreeBSD.
3431
3432 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
3433 IPv6 enabled. The observed symptom was a segmentation fault on return from
3434 the function os_common_find_running_interfaces() in src/os.c.
3435
3436 3. In the check_special_case() function in daemon.c I had used "errno" as an
3437 argument name, which causes warnings on some systems. This was basically a
3438 typo, since it was named "eno" in the comments!
3439
3440 4. The code that waits for the clock to tick (at a resolution of some fraction
3441 of a second) so as to ensure message-id uniqueness was always waiting for
3442 at least one whole tick, when it could have waited for less. [This is
3443 almost certainly not relevant at current processor speeds, where it is
3444 unlikely to ever wait at all. But we try to future-proof.]
3445
3446 5. The function that sleeps for a time interval that includes fractions of a
3447 second contained a race. It did not block SIGALRM between setting the
3448 timer, and suspending (a couple of lines later). If the interval was short
3449 and the sigsuspend() was delayed until after it had expired, the suspension
3450 never ended. On busy systems this could lead to processes getting stuck for
3451 ever.
3452
3453 6. Some uncommon configurations may cause a lookup to happen in a queue runner
3454 process, before it forks any delivery processes. The open lookup caching
3455 mechanism meant that the open file or database connection was passed into
3456 the delivery process. The problem was that delivery processes always tidy
3457 up cached lookup data. This could cause a problem for the next delivery
3458 process started by the queue runner, because the external queue runner
3459 process does not know about the closure. So the next delivery process
3460 still has data in the lookup cache. In the case of a file lookup, there was
3461 no problem because closing a file descriptor in a subprocess doesn't affect
3462 the parent. However, if the lookup was caching a connection to a database,
3463 the connection was closed, and the second delivery process was likely to
3464 see errors such as "PGSQL: query failed: server closed the connection
3465 unexpectedly". The problem has been fixed by closing all cached lookups
3466 in a queue runner before running a delivery process.
3467
3468 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
3469 seem to have the "const" qualifier which it has on other OS. I've
3470 parameterised it.
3471
3472 8. Change 4.21/2 was too strict. It is only if there are two authenticators
3473 *of the same type* (client or server) with the same public name that an
3474 error should be diagnosed.
3475
3476 9. When Exim looked up a host name for an IP address, but failed to find the
3477 original IP address when looking up the host name (a safety check), it
3478 output the message "<ip address> does not match any IP for NULL", which was
3479 confusing, to say the least. The bug was that the host name should have
3480 appeared instead of "NULL".
3481
348210. Since release 3.03, if Exim is called by a uid other than root or the Exim
3483 user that is built into the binary, and the -C or -D options is used, root
3484 privilege is dropped before the configuration file is read. In addition,
3485 logging is switched to stderr instead of the normal log files. If the
3486 configuration then re-defines the Exim user, the unprivileged environment
3487 is probably not what is expected, so Exim logs a panic warning message (but
3488 proceeds).
3489
3490 However, if deliver_drop_privilege is set, the unprivileged state may well
3491 be exactly what is intended, so the warning has been cut out in that case,
3492 and Exim is allowed to try to write to its normal log files.
3493
3494
3495Exim version 4.21
3496-----------------
3497
3498 1. smtp_return_error_details was not giving details for temporary sender
3499 or receiver verification errors.
3500
3501 2. Diagnose a configuration error if two authenticators have the same public
3502 name.
3503
3504 3. Exim used not to create the message log file for a message until the first
3505 delivery attempt. This could be confusing when incoming messages were held
3506 for policy or load reasons. The message log file is now created at the time
3507 the message is received, and an initial "Received" line is written to it.
3508
3509 4. The automatically generated man page for command line options had a minor
3510 bug that caused no ill effects; however, a more serious problem was that
3511 the procedure for building the man page automatically didn't always
3512 operate. Consequently, release 4.20 contains an out-of-date version. This
3513 shouldn't happen again.
3514
3515 5. When building Exim with embedded Perl support, the script that builds the
3516 Makefile was calling 'perl' to find its compile-time parameters, ignoring
3517 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
3518
3519 6. The freeze_tell option was not being used for messages that were frozen on
3520 arrival, either by an ACL or by local_scan().
3521
3522 7. Added the smtp_incomplete_transaction log selector.
3523
3524 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
3525 was accepting AUTH without a new EHLO.
3526
3527 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
3528 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
3529 RFC.
3530
353110. Logging of TCP/IP connections (when configured) now happens in the main
3532 daemon process instead of the child process, so that the TCP/IP connection
3533 count is more accurate (but it can never be perfect).
3534
353511. The use of "drop" in a nested ACL was not being handled correctly in the
3536 outer ACL. Now, if condition failure induced by the nested "drop" causes
3537 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
3538 or "require"), the connection is dropped.
3539
354012. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
3541 that yield "discard" can now be used with an "accept" or a "discard" verb,
3542 but an error is generated for any others (because I can't see a useful way
3543 to define what should happen).
3544
354513. When an ACL is read dynamically from a file (or anywhere else), the lines
3546 are now processed in the same way as lines in the Exim configuration file.
3547 In particular, continuation lines are supported.
3548
354914. Added the "dnslists = a.b.c!=n.n.n.n" feature.
3550
355115. Added -ti meaning -t -i.
3552
355316. Check for letters, digits, hyphens, and dots in the names of dnslist
3554 domains, and warn by logging if others are found.
3555
355617. At least on BSD, alignment is not guarenteed for the array of ifreq's
3557 returned from GIFCONF when Exim is trying to find the list of interfaces on
3558 a host. The code in os.c has been modified to copy each ifreq to an aligned
3559 structure in all cases.
3560
3561 Also, in some cases, the returned ifreq's were being copied to a 'struct
3562 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
3563 means the last couple of bytes of an IPv6 address could be chopped if the
3564 ifreq contained only a normal sockaddr (14 bytes storage).
3565
356618. Named domain lists were not supported in the hosts_treat_as_local option.
3567 An entry such as +xxxx was not recognized, and was treated as a literal
3568 domain name.
3569
357019. Ensure that header lines added by a DATA ACL are included in the reject log
3571 if the ACL subsequently rejects the message.
3572
357320. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
3574 MD5 (which is deprecated).
3575
357621. When testing a filter file using -bf, Exim was writing a message when it
3577 took the sender from a "From " line in the message, but it was not doing so
3578 when it took $return_path from a Return-Path: header line. It now does.
3579
358022. If the contents of a "message" modifier for a "warn" ACL verb do not begin
3581 with a valid header line field name (a series of printing characters
3582 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
3583
358423. Changed "disc" in the source to "disk" to conform to the documentation and
3585 the book and for uniformity.
3586
358724. Ignore Sendmail's -Ooption=value command line item.
3588
358925. When execve() failed while trying to run a command in a pipe transport,
3590 Exim was returning EX_UNAVAILBLE (69) from the subprocess. However, this
3591 could be confused with a return value of 69 from the command itself. This
3592 has been changed to 127, the value the shell returns if it is asked to run
3593 a non-existent command. The wording for the related log line suggests a
3594 non-existent command as the problem.
3595
359626. If received_header_text expands to an empty string, do not add a Received:
3597 header line to the message. (Well, it adds a token one on the spool, but
3598 marks it "old" so that it doesn't get used or transmitted.)
3599
360027. Installed eximstats 1.28 (addition of -nt option).
3601
360228. There was no check for failure on the call to getsockname() in the daemon
3603 code. This can fail if there is a shortage of resources on the system, with
3604 ENOMEM, for example. A temporary error is now given on failure.
3605
360629. Contrary to the C standard, it seems that in some environments, the
3607 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
3608 program. Exim now does this explicitly; it affects the formatting of
3609 timestamps using strftime().
3610
361130. If exiqsumm was given junk data, it threw up some uninitialized variable
3612 complaints. I've now initialized all the variables, to avoid this.
3613
361432. Header lines added by a system filter were not being "seen" during
3615 transport-time rewrites.
3616
361733. The info_callback() function passed to OpenSSL is set up with type void
3618 (*)(SSL *, int, int), as described somewhere. However, when calling the
3619 function (actually a macro) that sets it up, the type void(*)() is
3620 expected. I've put in a cast to prevent warnings from picky compilers.
3621
362234. If a DNS black list lookup found a CNAME record, but there were no A
3623 records associated with the domain it pointed at, Exim crashed.
3624
362535. If a DNS black list lookup returned more than one A record, Exim ignored
3626 all but the first. It now scans all returned addresses if a particular IP
3627 value is being sought. In this situation, the contents of the
3628 $dnslist_value variable are a list of all the addresses, separated by a
3629 comma and a space.
3630
363136. Tightened up the rules for host name lookups using reverse DNS. Exim used
3632 to accept a host name and all its aliases if the forward lookup for any of
3633 them yielded the IP address of the incoming connection. Now it accepts only
3634 those names whose forward lookup yields the correct IP address. Any other
3635 names are discarded. This closes a loophole whereby a rogue DNS
3636 administrator could create reverse DNS records to break through a
3637 wildcarded host restriction in an ACL.
3638
363937. If a user filter or a system filter that ran in a subprocess used any of
3640 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
3641 the wrong values were passed to the pipe command ($thisaddress had the
3642 value of $0, $0 had the value of $1, etc). This bug was introduced by
3643 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
3644
364538. Improved the line breaking for long SMTP error messages from ACLs.
3646 Previously, if there was no break point between 40 and 75 characters, Exim
3647 left the rest of the message alone. Two changes have been made: (a) I've
3648 reduced the minimum length to 35 characters; (b) if it can't find a break
3649 point between 35 and 75 characters, it looks ahead and uses the first one
3650 that it finds. This may give the occasional overlong line, but at least the
3651 remaining text gets split now.
3652
365339. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
3654 file descriptors might be low, and that setting 1000 would always raise it.
3655 It turns out that in some environments, the limit is already over 1000 and
3656 that lowering it causes trouble. So now Exim takes care not to decrease it.
3657
365840. When delivering a message, the value of $return_path is set to $sender_
3659 address at the start of routing (routers may change the value). By an
3660 oversight, this default was not being set up when an address was tested by
3661 -bt or -bv, which affected the outcome if any router or filter referred to
3662 $return_path.
3663
366441. The idea of the "warn" ACL verb is that it adds a header or writes to the
3665 log only when "message" or "log_message" are set. However, if one of the
3666 conditions was an address verification, or a call to a nested ACL, the
3667 messages generated by the underlying test were being passed through. This