Patch from Sieve maintainer.
[exim.git] / doc / doc-txt / ChangeLog
</
CommitLineData
50c99ba6 1$Cambridge: exim/doc/doc-txt/ChangeLog,v 1.317 2006/03/01 10:40:03 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
214e2000 248
5de37277
PH
249Exim version 4.60
250-----------------
251
cc38ddbf
PH
252PH/01 Two changes to the default runtime configuration:
253
254 (1) Move the checks for relay_from_hosts and authenticated clients from
255 after to before the (commented out) DNS black list checks.
256
257 (2) Add control=submission to the relay_from_hosts and authenticated
258 clients checks, on the grounds that messages accepted by these
259 statements are most likely to be submissions.
5de37277 260
72fdd6ae
PH
261PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
262
263 (1) Generate an error if the third argument for the ${prvs expansion is
264 not a single digit.
265
266 (2) Treat a missing third argument of ${prvscheck as if it were an empty
267 string.
268
269 (3) Reset the variables that are obtained from the first argument of
270 ${prvscheck and used in the second argument before leaving the code,
271 because their memory is reclaimed, so using them afterwards may do
272 silly things.
273
274 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
275 one (it's much easier than Tom thought :-).
276
277 (5) Because of (4), we can now allow for the use of $prvscheck_result
278 inside the third argument.
cb9328de 279
cb741023
PH
280PH/03 For some reason, the default setting of PATH when running a command from
281 a pipe transport was just "/usr/bin". I have changed it to
282 "/bin:/usr/bin".
283
f174f16e
PH
284PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
285 anything to be listed in the output from -bV.
b2f5a032 286
c25242d7
PH
287PH/05 When a filter generated an autoreply, the entire To: header line was
288 quoted in the delivery log line, like this:
289
290 => >A.N.Other <ano@some.domain> <original@ddress> ...
291
292 This has been changed so that it extracts the operative address. There
293 may be more than one such address. If so, they are comma-separated, like
294 this:
295
296 => >ano@some.domain,ona@other.domain <original@ddress> ...
297
82c19f95
PH
298PH/06 When a client host used a correct literal IP address in a HELO or EHLO
299 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
300 not being looked up in the rDNS to get a host name, Exim was showing the
301 IP address twice in Received: lines, even though the IP addresses were
302 identical. For example:
303
304 Received: from [1.2.3.4] (helo=[1.2.3.4])
305
306 However, if the real host name was known, it was omitting the HELO data
307 if it matched the actual IP address. This has been tidied up so that it
308 doesn't show the same IP address twice.
309
d7ffbc12
PH
310PH/07 When both +timestamp and +memory debugging was on, the value given by
311 $tod_xxx expansions could be wrong, because the tod_stamp() function was
312 called by the debug printing, thereby overwriting the timestamp buffer.
313 Debugging no longer uses the tod_stamp() function when +timestamp is set.
314
9f526266
PH
315PH/08 When the original message was included in an autoreply transport, it
316 always said "this is a copy of the message, including all the headers",
317 even if body_only or headers_only was set. It now gives an appropriate
318 message.
319
87fcc8b9
PH
320PH/09 Applied a patch from the Sieve maintainer which:
321
322 o fixes some comments
323 o adds the (disabled) notify extension core
324 o adds some debug output for the result of if/elsif tests
325 o points to the current vacation draft in the documentation
326 and documents the missing references header update
327
328 and most important:
329
330 o fixes a bug in processing the envelope test (when testing
331 multiple envelope elements, the last element determinted the
332 result)
333
456682f5
PH
334PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
335 Electronic Mail") by including:
336
337 Auto-submitted: auto-generated
338
339 in the messages that it generates (bounce messages and others, such as
340 warnings). In the case of bounce messages for non-SMTP mesages, there was
341 also a typo: it was using "Auto_submitted" (underscore instead of
342 hyphen). Since every message generated by Exim is necessarily in response
343 to another message, thes have all been changed to:
344
345 Auto-Submitted: auto-replied
346
347 in accordance with these statements in the RFC:
348
349 The auto-replied keyword:
350
351 - SHOULD be used on messages sent in direct response to another
352 message by an automatic process,
353
354 - MUST NOT be used on manually-generated messages,
355
356 - MAY be used on Delivery Status Notifications (DSNs) and Message
357 Disposition Notifications (MDNs),
358
359 - MUST NOT be used on messages generated by automatic or periodic
360 processes, except for messages which are automatic responses to
361 other messages.
362
3e46c1aa
PH
363PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
364 to the default Received: header definition.
456682f5 365
49826d12
PH
366PH/12 Added log selector acl_warn_skipped (default on).
367
eba0c039
PH
368PH/13 After a successful wildlsearch lookup, discard the values of numeric
369 variables because (a) they are in the wrong storage pool and (b) even if
370 they were copied, it wouldn't work properly because of the caching.
371
a0d6ba8a
PH
372PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
373 checking when decoding. Apparently there are clients that generate
374 overlong encoded strings. Why am I not surprised?
375
f0917727
PH
376PH/15 If the first argument of "${if match_address" was not empty, but did not
377 contain an "@" character, Exim crashed. Now it writes a panic log message
378 and treats the condition as false.
379
096fee00
PH
380PH/16 In autoreply, treat an empty string for "once" the same as unset.
381
024bd3c2
PH
382PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
383 extension "envelope-auth". The code is finished and in agreement with
384 other implementations, but there is no documentation so far and in fact,
385 nobody wrote the draft yet. This extension is currently #undef'ed, thus
386 not changing the active code.
387
388 Print executed "if" and "elsif" statements when debugging is used. This
389 helps a great deal to understand what a filter does.
390
391 Document more things not specified clearly in RFC3028. I had all this
392 sorted out, when out of a sudden new issues came to my mind. Oops."
393
df199fec
PH
394PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
395 (Bugzilla #53).
396
d27f1df3
PH
397PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
398 canonical form (as documented). However, after a host name lookup from
399 the IP address, check_host() was doing a simple string comparison with
400 addresses acquired from the DNS when checking that the found name did
401 have the original IP as one of its addresses. Since any found IPv6
402 addresses are likely to be in abbreviated form, the comparison could
403 fail. Luckily, there already exists a function for doing the comparison
404 by converting both addresses to binary, so now that is used instead of
405 the text comparison.
406
96776534
PH
407PH/20 There was another similar case to PH/19, when a complete host name was
408 given in a host list; looking up its IP address could give an abbreviated
409 form, whereas the current host's name might or might not be abbreviated.
410 The same fix has been applied.
411
5de37277 412
9a799bc0
PH
413Exim version 4.54
414-----------------
415
416PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
417 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
418 It now does.
419
99a4b039
PH
420PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
421 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
422
423PH/03 Typo: missing ".o" in src/pcre/Makefile.
424
4b233853
PH
425PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
426 header line, restrict the check to what is listed in RFCs 2369 and 2929.
427 Also, for "Auto-Submitted", treat anything other than "no" as
428 non-personal, in accordance with RFC 3834. (Previously it treated
429 anything starting "auto-" as non-personal.)
430
8857ccfd
PH
431TF/01 The control=submission/name=... option had a problem with syntax
432 errors if the name included a slash character. The /name= option
433 now slurps the rest of the string, so it can include any characters
434 but it must come last in the list of options (after /sender_retain
435 or /domain=).
436
433a2980
PH
437PH/05 Some modifications to the interface to the fake nameserver for the new
438 testing suite.
439
3e46c1aa 440
9a799bc0 441
e3a311ba
TK
442Exim version 4.53
443-----------------
444
445TK/01 Added the "success_on_redirect" address verification option. See
446 NewStuff for rationale and an example.
447
13b685f9
PH
448PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
449
395ff96d
PH
450PH/02 Patch to exigrep to allow it to work on syslog lines.
451
5b68f6e4
PH
452PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
453 fread() to skip over the body file's header line, because in Cygwin the
454 header line is locked and is inaccessible.
455
1ab52c69
PH
456PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
457 co-exist for some time) to make it clear that it is the Exim ID that is
458 referenced, not the Message-ID: header line.
459
b07e6aa3
PH
460PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
461 string_format() function, because snprintf() does not exist on all
462 operating systems.
463
254e032f
PH
464PH/06 The use of forbid_filter_existstest now also locks out the use of the
465 ${stat: expansion item.
466
3af76a81
PH
467PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
468 protocol synchronization error", to keep the pedants happy.
469
2548ba04
PH
470PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
471 well as for IRIX systems, when gcc is being used. See the host.c source
472 file for comments.
473
b6c6011d
PH
474PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
475
cf39cf57
PH
476PH/10 Named domain lists were not working if used in a queue_smtp_domains
477 setting.
478
f1513293
PH
479PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
480 transport and to the smtp transport in LMTP mode.
481
727549a4
PH
482TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
483
af46795e
PH
484PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
485 run a filter in a subprocess. This could lead to confusion in subsequent
486 lookups in the parent process. There should also be a search_tidyup() at
487 the end of the subprocess.
488
d7b47fd0
PH
489PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
490 only if the host matched helo_try_verify_hosts, which caused the
491 verification to occur when the EHLO/HELO command was issued. The ACL just
492 tested the remembered result. Now, if a previous verification attempt has
493 not happened, "verify = helo" does it there and then.
494
ee744174
JJ
495JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
496
b582ab87
PH
497TK/03 Fix log output including CR from clamd.
498
41a13e0a
PH
499PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
500 exist provoked a memory error which could cause a segfault.
501
f625cc5a
PH
502PH/15 Installed PCRE 6.2
503
504PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
505
21f7af35
PH
506PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
507 of the problem. Specifically, suggested +O2 rather than +O1 for the
508 HP-UX compiler.
509
31480e42
PH
510PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
511
2d280592
PH
512PH/20 If a delivery was routed to a non-standard port by means of an SRV
513 record, the port was not correctly logged when the outgoing_port log
514 selector was set (it logged the transort's default port).
515
7cd1141b
PH
516PH/21 Added support for host-specific ports to manualroute, queryprogram,
517 fallback_hosts, and "hosts" in the smtp transport.
518
519PH/22 If the log selector "outgoing_port" is set, the port is now also given on
520 host errors such as "Connection refused".
521
750af86e
PH
522PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
523 authentication with radiusclient 0.4.9:
524
525 - Error returned from rc_read_config was caught wrongly
526 - Username/password not passed on to radius server due to wrong length.
527
528 The presumption is that some radiusclient API changes for 4.51/PH/17
529 were not taken care of correctly. The code is still untested by me (my
530 Linux distribution still has 0.3.2 of radiusclient), but it was
531 contributed by a Radius user.
532
533PH/24 When doing a callout, the value of $domain wasn't set correctly when
534 expanding the "port" option of the smtp transport.
535
4304270b
TK
536TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
537 while reading a MIME header. Thanks to Tom Hughes for a patch.
538
750af86e
PH
539PH/24 Include config.h inside local_scan.h so that configuration settings are
540 available.
541
64ffc24f
PH
542PH/25 Make $smtp_command_argument available after all SMTP commands. This means
543 that in an ACL for RCPT (for example), you can examine exactly what was
544 received.
545
5dd9625b
PH
546PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
547 commands, but it was not correctly comparing the address with the actual
548 client host address. Thus, it would show the EHLO address in Received:
549 header lines when this was not necessary.
550
5591031b
PH
551PH/27 Added the % operator to ${eval:}.
552
ba18e66a
PH
553PH/28 Exim tries to create and chdir to its spool directory when it starts;
554 it should be ignoring failures (because with -C, for example, it has lost
555 privilege). It wasn't ignoring creation failures other than "already
556 exists".
557
9cec981f
PH
558PH/29 Added "crypteq" to the list of supported features that Exim outputs when
559 -bV or -d is used.
560
aa2b5c79
PH
561PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
562 because an input line was too long, either on its own, or by virtue of
1509d3a8
PH
563 too many continuations, the temporary file was not being removed, and the
564 return code was incorrect.
aa2b5c79 565
48a53b7f
PH
566PH/31 Missing "BOOL" in function definition in filtertest.c.
567
1c59d63b
PH
568PH/32 Applied Sieve patches from the maintainer.
569
671012da
TK
570TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
571
1509d3a8
PH
572PH/33 Added "verify = not_blind".
573
574PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
575 Local/Makefile (with some defaults set). These are used in built scripts
576 such as exicyclog, but they have never been used in the exim_install
577 script (though there are many overriding facilities there). I have
578 arranged that the exim_install script now takes note of these two
579 settings.
580
581PH/35 Installed configuration files for Dragonfly.
582
2fe1a124
PH
583PH/36 When a locally submitted message by a trusted user did not contain a
584 From: header, and the sender address was obtained from -f or from an SMTP
585 MAIL command, and the trusted user did not use -F to supply a sender
586 name, $originator_name was incorrectly used when constructing a From:
587 header. Furthermore, $originator_name was used for submission mode
588 messages from external hosts without From: headers in a similar way,
589 which is clearly wrong.
590
8800895a
PH
591PH/37 Added control=suppress_local_fixups.
592
ccfdb010
PH
593PH/38 When log_selector = +received_sender was set, and the addition of the
594 sender made the log line's construction buffer exactly full, or one byte
595 less than full, an overflow happened when the terminating "\n" was
596 subsequently added.
597
1130bfb0
PH
598PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
599 when the result of a list match is failure because a DNS lookup failed.
600
ebcb507f
PH
601PH/40 RM_COMMAND is now used in the building process.
602
c35e155c
PH
603PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
604 the "build-* directories that it finds.
605
95d1f782
PH
606PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
607 address in a domain literal was a prefix of an interface address.
608
fd6de02e
PH
609PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
610 when verifying a sender address, unless rewrite_headers is false.
611
58de37c5
PH
612PH/44 Wrote a long comment about why errors_to addresses are verified as
613 recipients, not senders.
614
261cf466
TF
615TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
616 the ratelimit ACL was added.
617
3ee512ff
PH
618PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
619
e08c430f
PH
620PH/46 Added extra information about PostgreSQL errors to the error string.
621
bef5a11f
PH
622PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
623 avoiding the need to install special zones in a real server. This is
624 backwards compatible; if it can't find the fake resolver, it drops back.
625 Thus, both old and new test suites can be run.
626
7546de58
TF
627TF/02 Added util/ratelimit.pl
628
e5d5a95f
TF
629TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
630 clock is set back in time.
631
2e88a017
TF
632TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
633 Candler <B.Candler@pobox.com>.
634
a5f65aa4
TF
635TF/05 The fix for PH/43 was not completely correct; widen_domains is always
636 OK for addresses that are the result of redirections.
637
e7726cbf
PH
638PH/48 A number of further additions for the benefit of the new test suite,
639 including a fake gethostbyname() that interfaces to the fake DNS resolver
640 (see PH/47 above).
641
a7fdad5b
TF
642TF/06 The fix for widen_domains has also been applied to qualify_single and
643 search_parents which are the other dnslookup options that can cause
644 header rewrites.
645
6af56900
PH
646PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
647 type ("H").
648
0925ede6
PH
649PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
650
66afa403
TF
651TF/07 Exim produced the error message "an SRV record indicated no SMTP
652 service" if it encountered an MX record with an empty target hostname.
653 The message is now "an MX or SRV record indicated no SMTP service".
654
0154e85a
TF
655TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
656 if the DNS of the sending site is misconfigured. This is quite a
657 common situation. This change restores the behaviour of treating a
658 helo verification defer as a failure.
659
16f12c76
PH
660PH/51 If self=fail was set on a router, the bounce message did not include the
661 actual error message.
662
bbe902f0 663
e5a9dba6
PH
664Exim version 4.52
665-----------------
666
667TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
668
22c3b60b
PH
669PH/01 When a transport filter timed out in a pipe delivery, and the pipe
670 command itself ended in error, the underlying message about the transport
671 filter timeout was being overwritten with the pipe command error. Now the
672 underlying error message should be appended to the second error message.
673
06a9b4b5
PH
674TK/01 Fix poll() being unavailable on Mac OSX 10.2.
675
c1ac6996
PH
676PH/02 Reduce the amount of output that "make" produces by default. Full output
677 can still be requested.
678
9c7a242c
PH
679PH/03 The warning log line about a condition test deferring for a "warn" verb
680 was being output only once per connection, rather than after each
681 occurrence (because it was using the same function as for successful
682 "warn" verbs). This seems wrong, so I have changed it.
683
87ba3f5f
PH
684TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
685 it should not have, which might have caused a crash in the right
686 circumstances, but probably never did.
687
688PH/04 Installed a modified version of Tony Finch's patch to make submission
689 mode fix the return path as well as the Sender: header line, and to
690 add a /name= option so that you can make the user's friendly name appear
691 in the header line.
692
29aba418
TF
693TF/03 Added the control = fakedefer ACL modifier.
694
fe0dab11
TF
695TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
696 Mark Lowes for thorough testing.
870f6ba8 697
11d337a4
TK
698TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
699
700TK/03 Merged latest SRS patch from Miles Wilton.
701
415c8f3b
PH
702PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
703 with the definition in sysexits.h (which is #included earlier).
704 Fortunately, Exim does not actually use EX_OK. The code used to try to
705 preserve the sysexits.h value, by assumimg that macro definitions were
706 scanned for macro replacements. I have been disabused of this notion,
707 so now the code just undefines EX_OK before #including unistd.h.
11d337a4 708
958541e9
PH
709PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
710 in the smtp transport. When a block could not be written in a single
711 write() function, the timeout was being re-applied to each part-write.
712 This seems wrong - if the receiver was accepting one byte at a time it
713 would take for ever. The timeout is now adjusted when this happens. It
714 doesn't have to be particularly precise.
715
c206415f
TK
716TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
717 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
718
2a4be8f9
PH
719PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
720 without a domain if the check to <postmaster@domain> fails.
721
1cba11c5
SC
722SC/01 Eximstats: added -xls and the ability to specify output files
723 (patch written by Frank Heydlauf).
724
725SC/02 Eximstats: use FileHandles for outputing results.
726
727SC/03 Eximstats: allow any combination of xls, txt, and html output.
728
729SC/04 Eximstats: fixed display of large numbers with -nvr option
730
731SC/05 Eximstats: fixed merging of reports with empty tables.
732
733SC/06 Eximstats: added the -include_original_destination flag
734
735SC/07 Eximstats: removed tabs and trailing whitespace.
736
1005d00e
TK
737TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
738
739TK/06 MBOX spool code: Add real "From " MBOX separator line
740 so the .eml file is really in mbox format (even though
741 most programs do not really care). Patch from Alex Miller.
742
743TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
744 The latter is generated from $received_to and is only set if the
745 message has one envelope recipient. SA can use these headers,
746 obviously out-of-the-box. Patch from Alex Miller.
747
9b4768fa
PH
748PH/08 The ${def test on a variable was returning false if the variable's
749 value was "0", contrary to what the specification has always said!
750 The result should be true unless the variable is empty.
751
752PH/09 The syntax error of a character other than { following "${if
753 def:variable_name" (after optional whitespace) was not being diagnosed.
754 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
755 accidental colon was present, for example, could give incorrect results.
756
0d7eb84a
PH
757PH/10 Tidied the code in a number of places where the st_size field of a stat()
758 result is used (not including appendfile, where other changes are about
759 to be made).
760
761PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
762 This involved changing a lot of size variables from int to off_t. It
763 should work with maildirs and everything.
764
40727bee
TK
765TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
766 spamd dying while we are connected to it.
767
554d2369
TF
768TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
769 <jgh@wizmail.org>
770
1f922db1
PH
771PH/12 Applied Alex Kiernan's patch for the API change for the error callback
772 function for BDB 4.3.
773
ef213c3b
PH
774PH/13 Changed auto_thaw such that it does not apply to bounce messages.
775
8ac170f3
PH
776PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
777 the sources for PCRE have been re-arranged and more files are now
778 involved.
779
b1c749bb
PH
780PH/15 The code I had for printing potentially long long variables in PH/11
781 above was not the best (it lost precision). The length of off_t variables
782 is now inspected at build time, and an appropriate printing format (%ld
c6c2dc1d
PH
783 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
784 to be "long long int" or "long int". This is needed for the internal
785 formatting function string_vformat().
b1c749bb 786
4aac9b49
PH
787PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
788 the configuration file to be ":syslog", then the script "guesses" where
789 the logs files are, rather than using the compiled in default. In our
790 case the guess is not the same as the compiled default, so the script
791 suddenly stopped working when I started to use syslog. The patch checks
792 to see if log_file_path is "". If so, it attempts to read it from exim
793 with no configuration file to get the compiled in version, before it
794 falls back to the previous guessing code."
795
294520c8
TK
796TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
797 implementing BATV in an Exim configuration. See NewStuff for the gory
798 details.
799
5bd022fe
PH
800PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
801 Makefile that are specific to HP-UX.
802
90e9ce59
PH
803PH/18 If the "use_postmaster" option was set for a recipient callout together
804 with the "random" option, the postmaster address was used as the MAIL
805 FROM address for the random test, but not for the subsequent recipient
806 test. It is now used for both.
807
5ea81592
PH
808PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
809 patch removes a few documentation additions to RFC 3028, because the
810 latest draft now contains them. It adds the new en;ascii-case comparator
811 and a new error check for 8bit text in MIME parts. Comparator and
812 require names are now matched exactly. I enabled the subaddress
813 extension, but it is not well tested yet (read: it works for me)."
814
c6c2dc1d
PH
815PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
816 rework some of the code of TK/09 above to avoid the hardwired use of
817 "%lld" and "long long". Replaced the call to snprintf() with a call to
818 string_vformat().
819
fffffe4c
PH
820PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
821 records point to non-existent hosts", "retry timeout exceeded", and
822 "retry time not reached for any host after a long failure period".
ca02eafb 823
9a26b6b2
PH
824PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
825 experimental DomainKeys support:
826
827 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
828 (2) On an error such as an illegally used "control", the wrong name for
829 the control was given.
830
831 These problems did NOT occur unless DomainKeys support was compiled.
832
4aee0225
PH
833PH/23 Added daemon_startup_retries and daemon_startup_sleep.
834
32d668a5
PH
835PH/24 Added ${if match_ip condition.
836
8187c3f3
PH
837PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
838 databases so that it will be absolutely obvious if a crash occurs in the
839 DB library. This is a regular occurrence (often caused by mis-matched
840 db.h files).
841
ff790e47 842PH/26 Insert a lot of missing (void) casts for functions such as chown(),
f1e894f3
PH
843 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
844 picked up on a user's system that detects such things. There doesn't seem
845 to be a gcc warning option for this - only an attribute that has to be
846 put on the function's prototype. It seems that in Fedora Core 4 they have
847 set this on a number of new functions. No doubt there will be more in due
848 course.
ff790e47 849
5417f6d1
PH
850PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
851 specify a transport. However, if an address that was verified by such a
852 router was the subject of a callout, Exim crashed because it tried to
853 read the rcpt_include_affixes from the non-existent transport. Now it
854 just assumes that the setting of that option is false. This bug was
855 introduced by 4.51/PH/31.
856
59cf8544
PH
857PH/28 Changed -d+all to exclude +memory, because that information is very
858 rarely of interest, but it makes the output a lot bigger. People tend to
859 do -d+all out of habit.
860
e7ad8a65
PH
861PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
862 code in os-type was giving problems when libc.so lives in lib64, like on
863 x86_64 Fedora Core.
864
ade42478
PH
865PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
866 aren't the modern standard, and it seems that some systems' include files
867 don't always have them. Exim was already checking for some of the newer
868 ones like T_AAAA, and defining it itself. I've added checks for all the
869 record types that Exim uses.
870
182ad5cf
PH
871PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
872 not automatically generating a new one, as it is supposed to. This
873 prevented TLS from working. If the file did exist, but contained invalid
874 data, a new version was generated, as expected. It was only the case of a
875 non-existent file that was broken.
876
b0d9fc80
TK
877TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
878 with a change in libdomainkeys > 0.64.
879
880TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
881 from DNS. If the selector record carries the flag, it now has
882 precedence over the domain-wide flag.
883
884TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
885
47c7a64a
PH
886PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
887 the use of a port name that isn't defined in /etc/services) occurred, the
888 message was deferred as in a normal delivery, and thus remained on the
889 spool, instead of being failed because of the mua_wrapper setting. This
890 is now fixed, and I tidied up some of the mua_wrapper messages at the
891 same time.
892
a388bce4
SC
893SC/08 Eximstats: whilst parsing the mainlog(s), store information about
894 the messages in a hash of arrays rather than using individual hashes.
895 This is a bit cleaner and results in dramatic memory savings, albeit
896 at a slight CPU cost.
897
898SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
899 as requested by Marc Sherman.
900
901SC/10 Eximstats: added histograms for user specified patterns as requested
902 by Marc Sherman.
903
0793e4ed
SC
904SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
905
c58b88df
PH
906PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
907 fopen() in the content-scanning modules that did not already have it.
908
e7ad8a65 909
7982096b
PH
910Exim version 4.51
911-----------------
912
1a46a8c5
PH
913TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
914 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
915
2f079f46 916TK/02 Fix ACL "control" statement not being available in MIME ACL.
1a46a8c5
PH
917
918TK/03 Fix ACL "regex" condition not being available in MIME ACL.
919
920PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
921 to test Sieve filters that use "vacation".
922
923PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
924 that changes the way the GnuTLS parameters are stored in the cache file.
925 The new format can be generated externally. For backward compatibility,
926 if the data in the cache doesn't make sense, Exim assumes it has read an
927 old-format file, and it generates new data and writes a new file. This
928 means that you can't go back to an older release without removing the
929 file.
930
931PH/03 A redirect router that has both "unseen" and "one_time" set does not
932 work if there are any delivery delays because "one_time" forces the
933 parent to be marked "delivered", so its unseen clone is never tried
934 again. For this reason, Exim now forbids the simultaneous setting of
935 these two options.
936
937PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
938 redirected to themselves ("homonym" addresses). Read the long ChangeLog
939 entry if you want to know the details. The fix, however, neglected to
940 consider the case when local delivery batching is involved. The test for
941 "previously delivered" was not happening when checking to see if an
942 address could be batched with a previous (undelivered) one; under
943 certain circumstances this could lead to multiple deliveries to the same
c2c19e9d 944 address.
1a46a8c5
PH
945
946PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
947 in its include files, and this causes problems building Exim.
948
949PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
950 header_syntax) but Exim was just ignoring anything given after a slash.
951 In particular, this caused confusion with an attempt to use "verify =
952 reverse_host_lookup/defer_ok". An error is now given when options are
953 supplied for verify items that do not have them. (Maybe reverse_host_
954 lookup should have a defer_ok option, but that's a different point.)
955
956PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
957 defined by RFC 821) to 2048, because there were problems with some AUTH
958 commands, and RFC 1869 says the size should be increased for extended
959 SMTP commands that take arguments.
960
961PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
962 Finch).
963
964PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
965 "unknown" error; now it says that the functionality isn't in the binary.
8d67ada3 966
49c2d5ea
PH
967PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
968 an address' error message when a string expansion fails (syntax or
f331f3b6
PH
969 whatever). Otherwise the password may appear in the log. Following change
970 PH/42 below, there is no longer a chance of it appearing in a bounce
971 message.
49c2d5ea 972
bf759a8b
PH
973PH/11 Installed exipick version 20050225.0 from John Jetmore.
974
83364d30
PH
975PH/12 If the last host in a fallback_hosts list was multihomed, only the first
976 of its addresses was ever tried. (Bugzilla bug #2.)
977
7999bbd7
PH
978PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
979 the result incorrectly in the debug output. (It correctly added a newline
980 to what was transported.)
981
7dbf77c9
PH
982TF/01 Added $received_time.
983
74e0617f
PH
984PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
985 commented out examples of how to interface to a virus scanner and to
986 SpamAssassin. Also added commented examples of av_scanner and
987 spamd_address settings.
988
2f079f46
PH
989PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
990 and controls are allowed in which ACLs. There were a couple of minor
991 errors. Some of the entries in the conditions table (which is a table of
992 where they are NOT allowed) were getting very unwieldy; rewrote them as a
993 negation of where the condition IS allowed.
994
8c841523
PH
995PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
996
7766a4f0
PH
997PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
998 header file does not have a version number, so I've had to invent a new
999 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
1000 API. The code is untested by me (my Linux distribution still has 0.3.2 of
1001 radiusclient), but it was contributed by a Radius user.
1002
8b417f2c
PH
1003PH/18 Installed Lars Mainka's patch for the support of CRL collections in
1004 files or directories, for OpenSSL.
1005
901f42cb
PH
1006PH/19 When an Exim process that is running as root has to create an Exim log
1007 file, it does so in a subprocess that runs as exim:exim so as to get the
1008 ownership right at creation (otherwise, other Exim processes might see
1009 the file with the wrong ownership). There was no test for failure of this
1010 fork() call, which would lead to the process getting stuck as it waited
1011 for a non-existent subprocess. Forks do occasionally fail when resources
1012 run out. I reviewed all the other calls to fork(); they all seem to check
1013 for failure.
1014
f9b9210e
PH
1015PH/20 When checking for unexpected SMTP input at connect time (before writing
1016 the banner), Exim was not dealing correctly with a non-positive return
1017 from the read() function. If the client had disconnected by this time,
1018 the result was a log entry for a synchronization error with an empty
1019 string after "input=" when read() returned zero. If read() returned -1
1020 (an event I could not check), uninitialized data bytes were printed.
1021 There were reports of junk text (parts of files, etc) appearing after
1022 "input=".
1023
54cdb463
PH
1024PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
1025
cf00dad6
PH
1026PH/22 Added support for macro redefinition, and (re)definition in between
1027 driver and ACL definitions.
1028
acb1b346
PH
1029PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
1030 forgetting to use the resulting value; it was using the unexpanded value.
1031
c5ddb310
PH
1032PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
1033 hadn't been configured. The fix is from Juergen Kreileder, who
1034 understands it better than I do:
1035
1036 "Here's what I see happening with three configured cyrus_sasl
1037 authenticators configured (plain, login, cram-md5):
1038
1039 On startup auth_cyrus_sasl_init() gets called for each of these.
1040 This means three calls to sasl_listmech() without a specified mech_list.
1041 => SASL tests which mechs of all available mechs actually work
1042 => three warnings about OTP not working
1043 => the returned list contains: plain, login, cram-md5, digest-md5, ...
1044
1045 With the patch, sasl_listmech() also gets called three times. But now
1046 SASL's mech_list option is set to the server_mech specified in the the
1047 authenticator. Or in other words, the answer from sasl_listmech()
1048 gets limited to just the mech you're testing for (which is different
1049 for each call.)
1050 => the return list contains just 'plain' or 'login', 'cram-md5' or
1051 nothing depending on the value of ob->server_mech.
1052
1053 I've just tested the patch: Authentication still works fine,
1054 unavailable mechs specified in the exim configuration are still
1055 caught, and the auth.log warnings about OTP are gone."
1056
31619da6
PH
1057PH/25 When debugging is enabled, the contents of the command line are added
1058 to the debugging output, even when log_selector=+arguments is not
1059 specified.
1060
bebaf0fc
PH
1061PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
1062 answer is "GNU", and only if the return is "GNU/something" is the answer
1063 "Linux".
1064
475fe28a
PH
1065PH/27 $acl_verify_message is now set immediately after the failure of a
1066 verification in an ACL, and so is available in subsequent modifiers. In
1067 particular, the message can be preserved by coding like this:
1068
1069 warn !verify = sender
1070 set acl_m0 = $acl_verify_message
1071
1072 Previously, $acl_verify_message was set only while expanding "message"
1073 and "log_message" when a very denied access.
1074
7e8bec7a
PH
1075PH/28 Modified OS/os.c-Linux with
1076
1077 -#ifndef OS_LOAD_AVERAGE
1078 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
1079
1080 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
1081 nomenclature these days.)
1082
e4a89c47
PH
1083PH/29 Installed patch from the Sieve maintainer that adds the options
1084 sieve_useraddress and sieve_subaddress to the redirect router.
1085
5ca2a9a1
PH
1086PH/30 In these circumstances:
1087 . Two addresses routed to the same list of hosts;
1088 . First host does not offer TLS;
1089 . First host accepts first address;
1090 . First host gives temporary error to second address;
1091 . Second host offers TLS and a TLS session is established;
1092 . Second host accepts second address.
1093 Exim incorrectly logged both deliveries with the TLS parameters (cipher
1094 and peerdn, if requested) that were in fact used only for the second
1095 address.
7e8bec7a 1096
c688b954
PH
1097PH/31 When doing a callout as part of verifying an address, Exim was not paying
1098 attention to any local part prefix or suffix that was matched by the
1099 router that accepted the address. It now behaves in the same way as it
1100 does for delivery: the affixes are removed from the local part unless
1101 rcpt_include_affixes is set on the transport.
1102
fed77020
PH
1103PH/32 Add the sender address, as F=<...>, to the log line when logging a
1104 timeout during the DATA phase of an incoming message.
1105
7fe1560f
PH
1106PH/33 Sieve envelope tests were broken for match types other than :is. I have
1107 applied a patch sanctioned by the Sieve maintainer.
c688b954 1108
ebb6e6d5
PH
1109PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
1110 the uid or gid is negative. A case of a negative gid caused this to be
1111 noticed. The fix allows for either to be negative.
1112
9c4e8f60
PH
1113PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
1114 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
1115 overlooked.
1116
1117PH/36 The change PH/12 above was broken. Fixed it.
1118
d7174846
PH
1119PH/37 Exim used to check for duplicate addresses in the middle of routing, on
1120 the grounds that routing the same address twice would always produce the
1121 same answer. This might have been true once, but it is certainly no
1122 longer true now. Routing a child address may depend on the previous
1123 routing that produced that child. Some complicated redirection strategies
1124 went wrong when messages had multiple recipients, and made Exim's
1125 behaviour dependent on the order in which the addresses were given.
1126
1127 I have moved the duplicate checking until after the routing is complete.
1128 Exim scans the addresses that are assigned to local and remote
1129 transports, and removes any duplicates. This means that more work will be
1130 done, as duplicates will always all be routed, but duplicates are
1131 presumably rare, so I don't expect this is of any significance.
1132
1133 For deliveries to pipes, files, and autoreplies, the duplicate checking
1134 still happens during the routing process, since they are not going to be
1135 routed further.
1136
cfe75fc3
PH
1137PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
1138 It corrects a timeout issue with spamd. This is Ian's comment: "The
1139 background is that sometimes spamd either never reads data from a
1140 connection it has accepted, or it never writes response data. The exiscan
1141 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
1142 blindly assumes that writes won't block so it may never time out."
1143
be22d70e
PH
1144PH/39 Allow G after quota size as well as K and M.
1145
0612b098
PH
1146PH/40 The value set for $authenticated_id in an authenticator may not contain
1147 binary zeroes or newlines because the value is written to log lines and
1148 to spool files. There was no check on this. Now the value is run through
1149 the string_printing() function so that such characters are converted to
1150 printable escape sequences.
1151
2e0c1448
PH
1152PH/41 $message_linecount is a new variable that contains the total number of
1153 lines in the message. Compare $body_linecount, which is the count for the
1154 body only.
1155
447d236c
PH
1156PH/42 Exim no longer gives details of delivery errors for specific addresses in
1157 bounce and delay warning messages, except in certain special cases, which
1158 are as follows:
1159
1160 (a) An SMTP error message from a remote host;
1161 (b) A message specified in a :fail: redirection;
1162 (c) A message specified in a "fail" command in a system filter;
1163 (d) A message specified in a FAIL return from the queryprogram router;
1164 (e) A message specified by the cannot_route_message router option.
1165
1166 In these cases only, Exim does include the error details in bounce and
1167 warning messages. There are also a few cases where bland messages such
1168 as "unrouteable address" or "local delivery error" are given.
1169
d20976dc
PH
1170PH/43 $value is now also set for the "else" part of a ${run expansion.
1171
f656d135
PH
1172PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
1173 being worked on, but at least Exim now implements the latest version to
1174 play with."
1175
2e2a30b4
PH
1176PH/45 In a pipe transport, although a timeout while waiting for the pipe
1177 process to complete was treated as a delivery failure, a timeout while
1178 writing the message to the pipe was logged, but erroneously treated as a
1179 successful delivery. Such timeouts include transport filter timeouts. For
1180 consistency with the overall process timeout, these timeouts are now
1181 treated as errors, giving rise to delivery failures by default. However,
1182 there is now a new Boolean option for the pipe transport called
1183 timeout_defer, which, if set TRUE, converts the failures into defers for
1184 both kinds of timeout. A transport filter timeout is now identified in
1185 the log output.
1186
9176e9f0
PH
1187PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
1188 systems where "make" and "gmake" are different, calling "gmake" at top
1189 level broke things. I've arranged for the value of $(MAKE) to be passed
1190 from the Makefile to this script so that it can call the same version of
1191 "make".
1192
7982096b 1193
bbe902f0
PH
1194A note about Exim versions 4.44 and 4.50
1195----------------------------------------
1196
1197Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
1198changes of various kinds. As a consequence, a big documentation update was
1199needed. This delayed the release for rather longer than seemed good, especially
1200in the light of a couple of (minor) security issues. Therefore, the changes
1201that fixed bugs were backported into 4.43, to create a 4.44 maintenance
1202release. So 4.44 and 4.50 are in effect two different branches that both start
1203from 4.43.
1204
1205I have left the 4.50 change log unchanged; it contains all the changes since
12064.43. The change log for 4.44 is below; many of its items are identical to
1207those for 4.50. This seems to be the most sensible way to preserve the
1208historical information.
1209
1210
f7b63901 1211Exim version 4.50
495ae4b0
PH
1212-----------------
1213
5fe762f6
PH
1214 1. Minor wording change to the doc/README.SIEVE file.
1215
139059f6 1216 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
5fe762f6 1217 computation of the current number of files was incorrect.
495ae4b0 1218
7086e875
PH
1219 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
1220 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
1221 place.
1222
35af9f61
PH
1223 4. Give more explanation in the error message when the command for a transport
1224 filter fails to execute.
1225
b668c215
PH
1226 5. There are several places where Exim runs a non-Exim command in a
1227 subprocess. The SIGUSR1 signal should be disabled for these processes. This
1228 was being done only for the command run by the queryprogram router. It is
1229 now done for all such subprocesses. The other cases are: ${run, transport
1230 filters, and the commands run by the lmtp and pipe transports.
1231
a494b1e1
PH
1232 6. Added CONFIGURE_GROUP build-time option.
1233
1234 7. Some older OS have a limit of 256 on the maximum number of file
1235 descriptors. Exim was using setrlimit() to set 1000 as a large value
1236 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
1237 systems. I've change it so that if it can't get 1000, it tries for 256.
35edf2ff 1238
c5fcb476
PH
1239 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
1240 was an oversight, and furthermore, ever since the addition of extra
1241 controls (e.g. 4.43/32), the checks on when to allow different forms of
1242 "control" were broken. There should now be diagnostics for all cases when a
1243 control that does not make sense is encountered.
1244
69358f02
PH
1245 9. Added the /retain_sender option to "control=submission".
1246
5be20824
PH
124710. $recipients is now available in the predata ACL (oversight).
1248
eb2c0248
PH
124911. Tidy the search cache before the fork to do a delivery from a message
1250 received from the command line. Otherwise the child will trigger a lookup
1251 failure and thereby defer the delivery if it tries to use (for example) a
1252 cached ldap connection that the parent has called unbind on.
1253
2a3eea10
PH
125412. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
1255 of $address_data from the recipient verification was clobbered by the
1256 sender verification.
1257
125813. The value of address_data from a sender verification is now available in
1259 $sender_address_data in subsequent conditions in the ACL statement.
1260
23c7ff99
PH
126114. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
1262
4deaf07d
PH
126315. Added a new option "connect=<time>" to callout options, to set a different
1264 connection timeout.
1265
926e1192
PH
126616. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
1267 was its contents. (It was OK if the option was not defined at all.)
1268
650edc6f
PH
126917. A "Completed" log line is now written for messages that are removed from
1270 the spool by the -Mrm option.
1271
2c7db3f5
PH
127218. New variables $sender_verify_failure and $recipient_verify_failure contain
1273 information about exactly what failed.
1274
3d235903
PH
127519. Added -dd to debug only the daemon process.
1276
7c7ad977
PH
127720. Incorporated Michael Haardt's patch to ldap.c for improving the way it
1278 handles timeouts, both on the server side and network timeouts. Renamed the
1279 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
1280
981756db
PH
128121. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
1282 It is now set to "smtps".
1283
d4eb88df
PH
128422. $host_address is now set to the target address during the checking of
1285 ignore_target_hosts.
1286
128723. When checking ignore_target_hosts for an ipliteral router, no host name was
1288 being passed; this would have caused $sender_host_name to have been used if
1289 matching the list had actually called for a host name (not very likely,
1290 since this list is usually IP addresses). A host name is now passed as
1291 "[x.x.x.x]".
1292
7d468ab8
PH
129324. Changed the calls that set up the SIGCHLD handler in the daemon to use the
1294 code that specifies a non-restarting handler (typically sigaction() in
1295 modern systems) in an attempt to fix a rare and obscure crash bug.
1296
129725. Narrowed the window for a race in the daemon that could cause it to ignore
1298 SIGCHLD signals. This is not a major problem, because they are used only to
1299 wake it up if nothing else does.
1300
62c0818f
PH
130126. A malformed maildirsize file could cause Exim to calculate negative values
1302 for the mailbox size or file count. Odd effects could occur as a result.
1303 The maildirsize information is now recalculated if the size or filecount
1304 end up negative.
1305
26034054
PH
130627. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
1307 support for a long time. Removed HAVE_SYS_VFS_H.
1308
af66f652
PH
130928. Installed the latest version of exipick from John Jetmore.
1310
90af77f4
PH
131129. In an address list, if the pattern was not a regular expression, an empty
1312 subject address (from a bounce message) matched only if the pattern was an
1313 empty string. Non-empty patterns were not even tested. This was the wrong
1314 because it is perfectly reasonable to use an empty address as part of a
1315 database query. An empty address is now tested by patterns that are
1316 lookups. However, all the other forms of pattern expect the subject to
1317 contain a local part and a domain, and therefore, for them, an empty
1318 address still always fails if the pattern is not itself empty.
1319
d8ef3577
PH
132030. Exim went into a mad DNS loop when attempting to do a callout where the
1321 host was specified on an smtp transport, and looking it up yielded more
1322 than one IP address.
1323
5cb8cbc6
PH
132431. Re-factored the code for checking spool and log partition space into a
1325 function that finds that data and another that does the check. The former
1326 is then used to implement four new variables: $spool_space, $log_space,
1327 $spool_inodes, and $log_inodes.
1328
14702f5b
PH
132932. The RFC2047 encoding function was originally intended for short strings
1330 such as real names; it was not keeping to the 75-character limit for
1331 encoded words that the RFC imposes. It now respects the limit, and
1332 generates multiple encoded words if necessary. To be on the safe side, I
1333 have increased the buffer size for the ${rfc2047: expansion operator from
1334 1024 to 2048 bytes.
1335
063b1e99
PH
133633. It is now permitted to omit both strings after an "if" condition; if the
1337 condition is true, the result is "true". As before, when the second string
1338 is omitted, a false condition yields an empty string. This makes it less
1339 cumbersome to write custom ACL and router conditions.
1340
652e1b65
PH
134134. Failure to deliver a bounce message always caused it to be frozen, even if
1342 there was an errors_to setting on the router. The errors_to setting is now
1343 respected.
1344
6f0c9a4f
PH
134535. If an IPv6 address is given for -bh or -bhc, it is now converted to the
1346 canonical form (fully expanded) before being placed in
1347 $sender_host_address.
1348
33397d19
PH
134936. The table in the code that translates DNS record types into text (T_A to
1350 "A" for instance) was missing entries for NS and CNAME. It is just possible
1351 that this could have caused confusion if both these types were looked up
1352 for the same domain, because the text type is used as part of Exim's
1353 per-process caching. But the chance of anyone hitting this buglet seems
1354 very small.
1355
7bb56e1f
PH
135637. The dnsdb lookup has been extended in a number of ways.
1357
1358 (1) There is a new type, "zns", which walks up the domain tree until it
1359 finds some nameserver records. It should be used with care.
1360
ea3bc19b
PH
1361 (2) There is a new type, "mxh", which is like "mx" except that it returns
1362 just the host names, not the priorities.
1363
1364 (3) It is now possible to give a list of domains (or IP addresses) to be
ff4dbb19
PH
1365 looked up. The behaviour when one of the lookups defers can be
1366 controlled by a keyword.
7bb56e1f 1367
ea3bc19b 1368 (4) It is now possible to specify the separator character for use when
7bb56e1f 1369 multiple records are returned.
33397d19 1370
0bcb2a0e
PH
137138. The dnslists ACL condition has been extended: it is now possible to supply
1372 a list of IP addresses and/or domains to be looked up in a particular DNS
1373 domain.
1374
2ac0e484
PH
137539. Added log_selector=+queue_time_overall.
1376
4e1fde53
PH
137740. When running the queue in the test harness, wait just a tad after forking a
1378 delivery process, to get repeatability of debugging output.
1379
de365ded
PH
138041. Include certificate and key file names in error message when GnuTLS fails
1381 to set them up, because the GnuTLS error message doesn't include the name
1382 of the failing file when there is a problem reading it.
1383
f05da2e8
PH
138442. Allow both -bf and -bF in the same test run.
1385
d6453af2
PH
138643. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
1387
f7b63901
PH
138844. The "Exiscan patch" is now merged into the mainline Exim source.
1389
139045. Sometimes the final signoff response after QUIT could fail to get
1391 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
1392 < 0 before doing a fflush(). This bug looks as though it goes back to the
1393 introduction of TLS in release 3.20, but "sometimes" must have been rare
1394 because the tests only now provoked it.
1395
a444213a
PH
139646. Reset the locale to "C" after calling embedded Perl, in case it was changed
1397 (this can affect the format of dates).
1398
0ec020ea
PH
139947. exim_tidydb, when checking for the continued existence of a message for
1400 which it has found a message-specific retry record, was not finding
1401 messages that were in split spool directories. Consequently, it was
1402 deleting retry records that should have stayed in existence.
1403
b1206957
PH
140448. Steve fixed some bugs in eximstats.
1405
140649. The SPA authentication driver was not abandoning authentication and moving
1407 on to the next authenticator when an expansion was forced to fail,
1408 contradicting the general specification for all authenticators. Instead it
1409 was generating a temporary error. It now behaves as specified.
1410
26dd5a95
PH
141150. The default ordering of permitted cipher suites for GnuTLS was pessimal
1412 (the order specifies the preference for clients). The order is now AES256,
1413 AES128, 3DES, ARCFOUR128.
1414
343b2385
PH
141551. Small patch to Sieve code - explicitly set From: when generating an
1416 autoreply.
1417
1c5466b9
PH
141852. Exim crashed if a remote delivery caused a very long error message to be
1419 recorded - for instance if somebody sent an entire SpamAssassin report back
1420 as a large number of 550 error lines. This bug was coincidentally fixed by
1421 increasing the size of one of Exim's internal buffers (big_buffer) that
1422 happened as part of the Exiscan merge. However, to be on the safe side, I
1423 have made the code more robust (and fixed the comments that describe what
1424 is going on).
1425
55ee9ee3
PH
142653. Now that there can be additional text after "Completed" in log lines (if
1427 the queue_time_overall log selector is set), a one-byte patch to exigrep
1428 was needed to allow it to recognize "Completed" as not the last thing in
1429 the line.
1430
d38f8232
PH
143154. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
1432 patch that reportedly fixes this has been added. I am not expert enough to
1433 create a test for it. This is what the patch creator wrote:
1434
1435 "I found a little strange behaviour of ldap code when working with
1436 Windows 2003 AD Domain, where users was placed in more than one
1437 Organization Units. When I tried to give exim partial DN, the exit code
1438 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
1439 But simultaneously result of request was absolutely normal ldap result,
1440 so I produce this patch..."
1441
3295e65b
PH
1442 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
1443 REFERENCE, so I have modified the code to exclude the patch when that macro
1444 is not defined.
1445
7102e136
PH
144655. Some experimental protocols are using DNS PTR records for new purposes. The
1447 keys for these records are domain names, not reversed IP addresses. The
b975ba52
PH
1448 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
1449 leaves it alone. Component reversal etc. now happens only for IP addresses.
ea3a6f44 1450 CAN-2005-0021
7102e136 1451
3ca0ba97
PH
145256. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
1453
c2bcbe20
PH
145457. Double the size of the debug message buffer (to 2048) so that more of very
1455 long debug lines gets shown.
1456
18ce445d
PH
145758. The exicyclog utility now does better if the number of log files to keep
1458 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
1459
1f5b4c3d
PH
146059. Two changes related to the smtp_active_hostname option:
1461
1462 (1) $smtp_active_hostname is now available as a variable.
1463 (2) The default for smtp_banner uses $smtp_active_hostname instead
1464 of $primary_hostname.
1465
b975ba52
PH
146660. The host_aton() function is supposed to be passed a string that is known
1467 to be a valid IP address. However, in the case of IPv6 addresses, it was
1468 not checking this. This is a hostage to fortune. Exim now panics and dies
1469 if the condition is not met. A case was found where this could be provoked
85b87bc2
PH
1470 from a dnsdb PTR lookup with an IPv6 address that had more than 8
1471 components; fortuitously, this particular loophole had already been fixed
1472 by change 4.50/55 above.
1473
1474 If there are any other similar loopholes, the new check in host_aton()
1475 itself should stop them being exploited. The report I received stated that
1476 data on the command line could provoke the exploit when Exim was running as
1477 exim, but did not say which command line option was involved. All I could
1478 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
1479 running as the user.
ea3a6f44 1480 CAN-2005-0021
85b87bc2
PH
1481
148261. There was a buffer overflow vulnerability in the SPA authentication code
1483 (which came originally from the Samba project). I have added a test to the
1484 spa_base64_to_bits() function which I hope fixes it.
ea3a6f44 1485 CAN-2005-0022
b975ba52 1486
17ffcae7
PH
148762. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
1488 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
1489
d95f9fdb
PH
149063. The daemon start-up calls getloadavg() while still root for those OS that
1491 need the first call to be done as root, but it missed one case: when
1492 deliver_queue_load_max is set with deliver_drop_privilege. This is
1493 necessary for the benefit of the queue runner, because there is no re-exec
1494 when deliver_drop_privilege is set.
1495
86b8287f
PH
149664. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
1497 This has been fixed.
1498
60dc5e56
PH
149965. Caching of lookup data for "hosts =" ACL conditions, when a named host list
1500 was in use, was not putting the data itself into the right store pool;
1501 consequently, it could be overwritten for a subsequent message in the same
1502 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
1503 the caching.)
1504
533244af
PH
150566. Added hosts_max_try_hardlimit to the smtp transport, default 50.
1506
a5a28604
PH
150767. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
1508 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
1509 the function were treating the return as a boolean value, which happened to
1510 work because 0=false and not-0=true, but is not correct code.
1511
7e634d24
PH
151268. The host_aton() function was not handling scoped IPv6 addresses (those
1513 with, for example, "%eth0" on the end) correctly.
1514
3e11c26b
PH
151569. Fixed some compiler warnings in acl.c for the bitmaps specified with
1516 negated items (that is, ~something) in unsigned ints. Some compilers
1517 apparently mutter when there is no cast.
1518
6729cf78
PH
151970. If an address verification called from an ACL failed, and did not produce a
1520 user-specific message (i.e. there was only a "system" message), nothing was
1521 put in $acl_verify_message. In this situation, it now puts the system
1522 message there.
1523
00f00ca5
PH
152471. Change 4.23/11 added synchronization checking at the start of an SMTP
1525 session; change 4.31/43 added the unwanted input to the log line - except
1526 that it did not do this in the start of session case. It now does.
1527
c9bdd01c
PH
152872. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
1529 This is wrong and can cause the other end to generate a synchronization
1530 error if it is another Exim or anything else that does the synchronization
1531 check. A QUIT command is no longer sent after a timeout.
1532
d43194df
PH
153373. $host_lookup_deferred has been added, to make it easier to detect DEFERs
1534 during host lookups.
1535
fe5b5d0b
PH
153674. The defer_ok option of callout verification was not working if it was used
1537 when verifying addresses in header lines, that is, for this case:
1538
1539 verify = header_sender/callout=defer_ok
1540
76a2d7ba
PH
154175. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
1542 those file descriptors could be used for SMTP connections. If anything
1543 wrote to stderr (the example that came up was "warn" in embedded Perl), it
1544 could be sent to the SMTP client, causing chaos. The daemon now opens
1545 stdin, stdout, and stderr to /dev/null when it puts itself into the
1546 background.
1547
154876. Arrange for output from Perl's "warn" command to be written to Exim's main
1549 log by default. The user can override this with suitable Perl magic.
1550
04f7d5b9
PH
155177. The use of log_message on a "discard" ACL verb, which is supposed to add to
1552 the log message when discard triggers, was not working for the DATA ACL or
1553 for the non-SMTP ACL.
1554
bc60667e
PH
155578. Error message wording change in sieve.c.
1556
bb6e88ff
PH
155779. If smtp_accept_max_per_host was set, the number of connections could be
1558 restricted to fewer than expected, because the daemon was trying to set up
1559 a new connection before checking whether the processes handling previous
1560 connections had finished. The check for completed processes is now done
1561 earlier. On busy systems, this bug wouldn't be noticed because something
1562 else would have woken the daemon, and it would have reaped the completed
1563 process earlier.
1564
1e70f85b
PH
156580. If a message was submitted locally by a user whose login name contained one
1566 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
1567 It caused a spool format error. I have fixed the spool reading code. A
1568 related problem was that the "from" clause in the Received: line became
1569 illegal because of the space(s). It is now covered by ${quote_local_part.
1570
157181. Included the latest eximstats from Steve (adds average sizes to HTML Top
1572 tables).
1573
4e01f9d6
PH
157482. Updated OS/Makefile-AIX as per message from Mike Meredith.
1575
1ee1cef2
PH
157683. Patch from Sieve maintainer to fix unterminated string problem in
1577 "vacation" handling.
1578
6e2b4ccc
PH
157984. Some minor changes to the Linux configuration files to help with other
1580 OS variants using glibc.
1581
8e669ac1
PH
158285. One more patch for Sieve to update vacation handling to latest spec.
1583
495ae4b0 1584
bbe902f0
PH
1585----------------------------------------------------
1586See the note above about the 4.44 and 4.50 releases.
1587----------------------------------------------------
1588
1589
1590Exim version 4.44
1591-----------------
1592
1593 1. Change 4.43/35 introduced a bug that caused file counts to be
1594 incorrectly computed when quota_filecount was set in an appendfile
1595 transport
1596
1597 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
1598 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
1599 place.
1600
1601 3. Give more explanation in the error message when the command for a transport
1602 filter fails to execute.
1603
1604 4. There are several places where Exim runs a non-Exim command in a
1605 subprocess. The SIGUSR1 signal should be disabled for these processes. This
1606 was being done only for the command run by the queryprogram router. It is
1607 now done for all such subprocesses. The other cases are: ${run, transport
1608 filters, and the commands run by the lmtp and pipe transports.
1609
1610 5. Some older OS have a limit of 256 on the maximum number of file
1611 descriptors. Exim was using setrlimit() to set 1000 as a large value
1612 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
1613 systems. I've change it so that if it can't get 1000, it tries for 256.
1614
1615 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
1616 was an oversight, and furthermore, ever since the addition of extra
1617 controls (e.g. 4.43/32), the checks on when to allow different forms of
1618 "control" were broken. There should now be diagnostics for all cases when a
1619 control that does not make sense is encountered.
1620
1621 7. $recipients is now available in the predata ACL (oversight).
1622
1623 8. Tidy the search cache before the fork to do a delivery from a message
1624 received from the command line. Otherwise the child will trigger a lookup
1625 failure and thereby defer the delivery if it tries to use (for example) a
1626 cached ldap connection that the parent has called unbind on.
1627
1628 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
1629 of $address_data from the recipient verification was clobbered by the
1630 sender verification.
1631
163210. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
1633 was its contents. (It was OK if the option was not defined at all.)
1634
163511. A "Completed" log line is now written for messages that are removed from
1636 the spool by the -Mrm option.
1637
163812. $host_address is now set to the target address during the checking of
1639 ignore_target_hosts.
1640
164113. When checking ignore_target_hosts for an ipliteral router, no host name was
1642 being passed; this would have caused $sender_host_name to have been used if
1643 matching the list had actually called for a host name (not very likely,
1644 since this list is usually IP addresses). A host name is now passed as
1645 "[x.x.x.x]".
1646
164714. Changed the calls that set up the SIGCHLD handler in the daemon to use the
1648 code that specifies a non-restarting handler (typically sigaction() in
1649 modern systems) in an attempt to fix a rare and obscure crash bug.
1650
165115. Narrowed the window for a race in the daemon that could cause it to ignore
1652 SIGCHLD signals. This is not a major problem, because they are used only to
1653 wake it up if nothing else does.
1654
165516. A malformed maildirsize file could cause Exim to calculate negative values
1656 for the mailbox size or file count. Odd effects could occur as a result.
1657 The maildirsize information is now recalculated if the size or filecount
1658 end up negative.
1659
166017. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
1661 support for a long time. Removed HAVE_SYS_VFS_H.
1662
ea3a6f44 166318. Updated exipick to current release from John Jetmore.
bbe902f0
PH
1664
166519. Allow an empty sender to be matched against a lookup in an address list.
1666 Previously the only cases considered were a regular expression, or an
1667 empty pattern.
1668
166920. Exim went into a mad DNS lookup loop when doing a callout where the
1670 host was specified on the transport, if the DNS lookup yielded more than
1671 one IP address.
1672
ea3a6f44
NM
167321. The RFC2047 encoding function was originally intended for short strings
1674 such as real names; it was not keeping to the 75-character limit for
1675 encoded words that the RFC imposes. It now respects the limit, and
1676 generates multiple encoded words if necessary. To be on the safe side, I
1677 have increased the buffer size for the ${rfc2047: expansion operator from
1678 1024 to 2048 bytes.
bbe902f0 1679
ea3a6f44
NM
168022. Failure to deliver a bounce message always caused it to be frozen, even if
1681 there was an errors_to setting on the router. The errors_to setting is now
1682 respected.
bbe902f0
PH
1683
168423. If an IPv6 address is given for -bh or -bhc, it is now converted to the
1685 canonical form (fully expanded) before being placed in
1686 $sender_host_address.
1687
168824. Updated eximstats to version 1.33
1689
ea3a6f44
NM
169025. Include certificate and key file names in error message when GnuTLS fails
1691 to set them up, because the GnuTLS error message doesn't include the name
1692 of the failing file when there is a problem reading it.
bbe902f0
PH
1693
169426. Expand error message when OpenSSL has problems setting up cert/key files.
ea3a6f44 1695 As per change 25.
bbe902f0 1696
ea3a6f44
NM
169727. Reset the locale to "C" after calling embedded Perl, in case it was changed
1698 (this can affect the format of dates).
bbe902f0 1699
ea3a6f44
NM
170028. exim_tidydb, when checking for the continued existence of a message for
1701 which it has found a message-specific retry record, was not finding
1702 messages that were in split spool directories. Consequently, it was
1703 deleting retry records that should have stayed in existence.
bbe902f0
PH
1704
170529. eximstats updated to version 1.35
1706 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
1707 1.35 - bugfix such that pie charts by volume are generated correctly
1708
ea3a6f44
NM
170930. The SPA authentication driver was not abandoning authentication and moving
1710 on to the next authenticator when an expansion was forced to fail,
1711 contradicting the general specification for all authenticators. Instead it
1712 was generating a temporary error. It now behaves as specified.
bbe902f0 1713
ea3a6f44
NM
171431. The default ordering of permitted cipher suites for GnuTLS was pessimal
1715 (the order specifies the preference for clients). The order is now AES256,
1716 AES128, 3DES, ARCFOUR128.
bbe902f0 1717
ea3a6f44
NM
171831. Small patch to Sieve code - explicitly set From: when generating an
1719 autoreply.
bbe902f0 1720
ea3a6f44
NM
172132. Exim crashed if a remote delivery caused a very long error message to be
1722 recorded - for instance if somebody sent an entire SpamAssassin report back
1723 as a large number of 550 error lines. This bug was coincidentally fixed by
1724 increasing the size of one of Exim's internal buffers (big_buffer) that
1725 happened as part of the Exiscan merge. However, to be on the safe side, I
1726 have made the code more robust (and fixed the comments that describe what
1727 is going on).
bbe902f0 1728
ea3a6f44
NM
172933. Some experimental protocols are using DNS PTR records for new purposes. The
1730 keys for these records are domain names, not reversed IP addresses. The
1731 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
1732 leaves it alone. Component reversal etc. now happens only for IP addresses.
bbe902f0
PH
1733 CAN-2005-0021
1734
ea3a6f44
NM
173534. The host_aton() function is supposed to be passed a string that is known
1736 to be a valid IP address. However, in the case of IPv6 addresses, it was
1737 not checking this. This is a hostage to fortune. Exim now panics and dies
1738 if the condition is not met. A case was found where this could be provoked
1739 from a dnsdb PTR lookup with an IPv6 address that had more than 8
1740 components; fortuitously, this particular loophole had already been fixed
1741 by change 4.50/55 or 4.44/33 above.
1742
1743 If there are any other similar loopholes, the new check in host_aton()
1744 itself should stop them being exploited. The report I received stated that
1745 data on the command line could provoke the exploit when Exim was running as
1746 exim, but did not say which command line option was involved. All I could
1747 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
1748 running as the user.
bbe902f0
PH
1749 CAN-2005-0021
1750
ea3a6f44
NM
175135. There was a buffer overflow vulnerability in the SPA authentication code
1752 (which came originally from the Samba project). I have added a test to the
1753 spa_base64_to_bits() function which I hope fixes it.
bbe902f0
PH
1754 CAN-2005-0022
1755
ea3a6f44
NM
175636. The daemon start-up calls getloadavg() while still root for those OS that
1757 need the first call to be done as root, but it missed one case: when
1758 deliver_queue_load_max is set with deliver_drop_privilege. This is
1759 necessary for the benefit of the queue runner, because there is no re-exec
1760 when deliver_drop_privilege is set.
bbe902f0 1761
ea3a6f44
NM
176237. Caching of lookup data for "hosts =" ACL conditions, when a named host list
1763 was in use, was not putting the data itself into the right store pool;
1764 consequently, it could be overwritten for a subsequent message in the same
1765 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
1766 the caching.)
bbe902f0 1767
ea3a6f44
NM
176838. Sometimes the final signoff response after QUIT could fail to get
1769 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
1770 < 0 before doing a fflush(). This bug looks as though it goes back to the
1771 introduction of TLS in release 3.20, but "sometimes" must have been rare
1772 because the tests only now provoked it.
bbe902f0
PH
1773
1774
495ae4b0
PH
1775Exim version 4.43
1776-----------------
1777
1778 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
1779 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
1780 Now it returns an integer. A number of calls were still expecting a T/F
1781 return. Fortuitously, in all cases, the tests worked in OK situations,
1782 which is the norm. However, things would have gone wrong on any write
1783 failures on the smtp file descriptor. This function is used when sending
1784 messages over SMTP and also when doing verify callouts.
1785
1786 2. When Exim is called to do synchronous delivery of a locally submitted
1787 message (the -odf or -odi options), it no longer closes stderr before doing
1788 the delivery.
1789
1790 3. Implemented the mua_wrapper option.
1791
1792 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
1793
1794 5. Implemented the functions header_remove(), header_testname(),
1795 header_add_at_position(), and receive_remove_recipient(), and exported them
1796 to local_scan().
1797
1798 6. If an ACL "warn" statement specified the addition of headers, Exim already
1799 inserted X-ACL-Warn: at the start if there was no header name. However, it
1800 was not making this test for the second and subsequent header lines if
1801 there were newlines in the string. This meant that an invalid header could
1802 be inserted if Exim was badly configured.
1803
1804 7. Allow an ACL "warn" statement to add header lines at the start or after all
1805 the Received: headers, as well as at the end.
1806
1807 8. Added the rcpt_4xx retry error code.
1808
1809 9. Added postmaster_mailfrom=xxx to callout verification option.
1810
181110. Added mailfrom=xxxx to the callout verification option, for verify=
1812 header_sender only.
1813
181411. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
1815 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
1816
181712. Inserted some casts to stop certain compilers warning when using pointer
1818 differences as field lengths or precisions in printf-type calls (mostly
1819 affecting debugging statements).
1820
182113. Added optional readline() support for -be (dynamically loaded).
1822
182314. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
1824 same clock tick as a message's arrival, so that its received time was the
1825 same as the "first fail" time on the retry record, and that message
1826 remained on the queue past the ultimate address timeout, every queue runner
1827 would try a delivery (because it was past the ultimate address timeout) but
1828 after another failure, the ultimate address timeout, which should have then
1829 bounced the address, did not kick in. This was a "< instead of <=" error;
1830 in most cases the first failure would have been in the next clock tick
1831 after the received time, and all would be well.
1832
183315. The special items beginning with @ in domain lists (e.g. @mx_any) were not
1834 being recognized when the domain list was tested by the match_domain
1835 condition in an expansion string.
1836
183716. Added the ${str2b64: operator.
1838
183917. Exim was always calling setrlimit() to set a large limit for the number of
1840 processes, without checking whether the existing limit was already
1841 adequate. (It did check for the limit on file descriptors.) Furthermore,
1842 errors from getrlimit() and setrlimit() were being ignored. Now they are
1843 logged to the main and panic logs, but Exim does carry on, to try to do its
1844 job under whatever limits there are.
1845
184618. Imported PCRE 5.0.
1847
184819. Trivial typo in log message " temporarily refused connection" (the leading
1849 space).
1850
185120. If the log selector return_path_on_delivery was set and an address was
1852 redirected to /dev/null, the delivery process crashed because it assumed
1853 that a return path would always be set for a "successful" delivery. In this
1854 case, the whole delivery is bypassed as an optimization, and therefore no
1855 return path is set.
1856
185721. Internal re-arrangement: the function for sending a challenge and reading
1858 a response while authentication was assuming a zero-terminated challenge
1859 string. It's now changed to take a pointer and a length, to allow for
1860 binary data in such strings.
1861
186222. Added the cyrus_sasl authenticator (code supplied by MBM).
1863
186423. Exim was not respecting finduser_retries when seeking the login of the
1865 uid under which it was called; it was always trying 10 times. (The default
1866 setting of finduser_retries is zero.) Also, it was sleeping after the final
1867 failure, which is pointless.
1868
186924. Implemented tls_on_connect_ports.
1870
187125. Implemented acl_smtp_predata.
1872
187326. If the domain in control=submission is set empty, Exim assumes that the
1874 authenticated id is a complete email address when it generates From: or
1875 Sender: header lines.
1876
187727. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
1878 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
1879 chown and chgrp in /bin and hostname in /usr/bin.
1880
188128. Exim was keeping the "process log" file open after each use, just as it
1882 does for the main log. This opens the possibility of it remaining open for
1883 long periods when the USR1 signal hits a daemon. Occasional processlog
1884 errors were reported, that could have been caused by this. Anyway, it seems
1885 much more sensible not to leave this file open at all, so that is what now
1886 happens.
1887
188829. The long-running daemon process does not normally write to the log once it
1889 has entered its main loop, and it closes the log before doing so. This is
1890 so that log files can straightforwardly be renamed and moved. However,
1891 there are a couple of unusual error situations where the daemon does write
1892 log entries, and I had neglected to close the log afterwards.
1893
189430. The text of an SMTP error response that was received during a remote
1895 delivery was being truncated at 512 bytes. This is too short for some of
1896 the long messages that one sometimes sees. I've increased the limit to
1897 1024.
1898
189931. It is now possible to make retry rules that apply only when a message has a
1900 specific sender, in particular, an empty sender.
1901
190232. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
1903 it possible to be selective about when SMTP synchronization is enforced.
1904
190533. Added "control = caseful_local_part" and "control = "caselower_local_part".
1906
190732. Implemented hosts_connection_nolog.
1908
190933. Added an ACL for QUIT.
1910
191134. Setting "delay_warning=" to disable warnings was not working; it gave a
1912 syntax error.
1913
191435. Added mailbox_size and mailbox_filecount to appendfile.
1915
191636. Added control = no_multiline_responses to ACLs.
1917
191837. There was a bug in the logic of the code that waits for the clock to tick
1919 in the case where the clock went backwards by a substantial amount such
1920 that the microsecond fraction of "now" was more than the microsecond
1921 fraction of "then" (but the whole seconds number was less).
1922
192338. Added support for the libradius Radius client library this is found on
1924 FreeBSD (previously only the radiusclient library was supported).
1925
1926
1927Exim version 4.42
1928-----------------
1929
1930 1. When certain lookups returned multiple values in the form name=value, the
1931 quoting of the values was not always being done properly. Specifically:
1932 (a) If the value started with a double quote, but contained no whitespace,
1933 it was not quoted.
1934 (b) If the value contained whitespace other than a space character (i.e.
1935 tabs or newlines or carriage returns) it was not quoted.
1936 This fix has been applied to the mysql and pgsql lookups by writing a
1937 separate quoting function and calling it from the lookup code. The fix
1938 should probably also be applied to nisplus, ibase and oracle lookups, but
1939 since I cannot test any of those, I have not disturbed their existing code.
1940
1941 2. A hit in the callout cache for a specific address caused a log line with no
1942 reason for rejecting RCPT. Now it says "Previous (cached) callout
1943 verification failure".
1944
1945 3. There was an off-by-one bug in the queryprogram router. An over-long
1946 return line was truncated at 256 instead of 255 characters, thereby
1947 overflowing its buffer with the terminating zero. As well as fixing this, I
1948 have increased the buffer size to 1024 (and made a note to document this).
1949
1950 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
1951 when Exim is waiting for an SMTP response from a remote server, Exim
1952 restarts its select() call on the socket, thereby resetting its timeout.
1953 This is not a problem when such interrupts are rare. Somebody set up a cron
1954 job to run exiwhat every 2 minutes, which is less than the normal select()
1955 timeout (5 or 10 minutes). This meant that the select() timeout never
1956 kicked in because it was always reset. I have fixed this by comparing the
1957 time when an interrupt arrives with the time at the start of the first call
1958 to select(). If more time than the timeout has elapsed, the interrupt is
1959 treated as a timeout.
1960
1961 5. Some internal re-factoring in preparation for the addition of Sieve
1962 extensions (by MH). In particular, the "personal" test is moved to a
1963 separate function, and given an option for scanning Cc: and Bcc: (which is
1964 not set for Exim filters).
1965
1966 6. When Exim created an email address using the login of the caller as the
1967 local part (e.g. when creating a From: or Sender: header line), it was not
1968 quoting the local part when it contained special characters such as @.
1969
1970 7. Installed new OpenBSD configuration files.
1971
1972 8. Reworded some messages for syntax errors in "and" and "or" conditions to
1973 try to make them clearer.
1974
1975 9. Callout options, other than the timeout value, were being ignored when
1976 verifying sender addresses in header lines. For example, when using
1977
1978 verify = header_sender/callout=no_cache
1979
1980 the cache was (incorrectly) being used.
1981
198210. Added a missing instance of ${EXE} to the exim_install script; this affects
1983 only the Cygwin environment.
1984
198511. When return_path_on_delivery was set as a log selector, if different remote
1986 addresses in the same message used different return paths and parallel
1987 remote delivery occurred, the wrong values would sometimes be logged.
1988 (Whenever a remote delivery process finished, the return path value from
1989 the most recently started remote delivery process was logged.)
1990
199112. RFC 3848 specifies standard names for the "with" phrase in Received: header
1992 lines when AUTH and/or TLS are in use. This is the "received protocol"
1993 field. Exim used to use "asmtp" for authenticated SMTP, without any
1994 indication (in the protocol name) for TLS use. Now it follows the RFC and
1995 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
1996 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
1997 names appear in log lines as well as in Received: header lines.
1998
199913. Installed MH's patches for Sieve to add the "copy" and "vacation"
2000 extensions, and comparison tests, and to fix some bugs.
2001
200214. Changes to the "personal" filter test:
2003
2004 (1) The test was buggy in that it was just doing the equivalent of
2005 "contains" tests on header lines. For example, if a user's address was
2006 anne@some.where, the "personal" test would incorrectly be true for
2007
2008 To: susanne@some.where
2009
2010 This test is now done by extracting each address from the header in turn,
2011 and checking the entire address. Other tests that are part of "personal"
2012 are now done using regular expressions (for example, to check local parts
2013 of addresses in From: header lines).
2014
2015 (2) The list of non-personal local parts in From: addresses has been
2016 extended to include "listserv", "majordomo", "*-request", and "owner-*",
2017 taken from the Sieve specification recommendations.
2018
2019 (3) If the message contains any header line starting with "List-" it is
2020 treated as non-personal.
2021
2022 (4) The test for "circular" in the Subject: header line has been removed
2023 because it now seems ill-conceived.
2024
202515. Minor typos in src/EDITME comments corrected.
2026
202716. Installed latest exipick from John Jetmore.
2028
202917. If headers_add on a router specified a text string that was too long for
2030 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
2031 of string_sprintf() is now avoided.
2032
203318. $message_body_size was not set (it was always zero) when running the DATA
2034 ACL and the local_scan() function.
2035
203619. For the "mail" command in an Exim filter, no default was being set for
2037 the once_repeat time, causing a random time value to be used if "once" was
2038 specified. (If the value happened to be <= 0, no repeat happened.) The
2039 default is now 0s, meaning "never repeat". The "vacation" command was OK
2040 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
2041 (I found it when inspecting the code).
2042
204320. There is now an overall timeout for performing a callout verification. It
2044 defaults to 4 times the callout timeout, which applies to individual SMTP
2045 commands during the callout. The overall timeout applies when there is more
2046 than one host that can be tried. The timeout is checked before trying the
2047 next host. This prevents very long delays if there are a large number of
2048 hosts and all are timing out (e.g. when the network connections are timing
2049 out). The value of the overall timeout can be changed by specifying an
2050 additional sub-option for "callout", called "maxwait". For example:
2051
2052 verify = sender/callout=5s,maxwait=20s
2053
205421. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
2055 to the end before writing, but this should make it even safer).
2056
205722. Exim was forgetting that it had advertised PIPELINING for the second and
2058 subsequent messages on an SMTP connection. It was also not resetting its
2059 memory on STARTTLS and an internal HELO.
2060
206123. When Exim logs an SMTP synchronization error within a session, it now
2062 records whether PIPELINING has been advertised or not.
2063
206424. Added 3 instances of "(long int)" casts to time_t variables that were being
2065 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
2066 rather than long int.
2067
206825. Installed the latest Cygwin configuration files from the Cygwin maintainer.
2069
207026. Added the never_mail option to autoreply.
2071
2072
2073Exim version 4.41
2074-----------------
2075
2076 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
2077 crash if the getsockname() call failed; this can happen if a connection is
2078 closed very soon after it is established. The problem was simply in the
2079 order in which certain operations were done, causing Exim to try to write
2080 to the SMTP stream before it had set up the file descriptor. The bug has
2081 been fixed by making things happen in the correct order.
2082
2083
2084Exim version 4.40
2085-----------------
2086
2087 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
2088 before the connection was closed, thus losing the rejection response.
2089
2090 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
2091 some early Solaris releases, but causes trouble in current releases where
2092 socklen_t is defined.
2093
2094 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
2095 always exist.
2096
2097 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
2098 configured.
2099
2100 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
2101 that releasing the top of it at the end releases what was used for sub-
2102 expansions (unless the block got too big). However, discard this block if
2103 the first thing is a variable or header, so that we can use its block when
2104 it is dynamic (useful for very large $message_headers, for example).
2105
2106 6. Lookups now cache *every* query, not just the most recent. A new, separate
2107 store pool is used for this. It can be recovered when all lookup caches are
2108 flushed. Lookups now release memory at the end of their result strings.
2109 This has involved some general refactoring of the lookup sources.
2110
2111 7. Some code has been added to the store_xxx() functions to reduce the amount
2112 of flapping under certain conditions.
2113
2114 8. log_incoming_interface used to affect only the <= reception log lines. Now
2115 it causes the local interface and port to be added to several more SMTP log
2116 lines, for example "SMTP connection from", and rejection lines.
2117
2118 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
2119
212010. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
2121
212211. If $host_data was set by virtue of a hosts lookup in an ACL, its value
2123 could be overwritten at the end of the current message (or the start of a
2124 new message if it was set in a HELO ACL). The value is now preserved for
2125 the duration of the SMTP connection.
2126
212712. If a transport had a headers_rewrite setting, and a matching header line
2128 contained an unqualified address, that address was qualified, even if it
2129 did not match any rewriting rules. The underlying bug was that the values
2130 of the flags that permit the existence of unqualified sender and recipient
2131 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
2132 non-local messages, and by -bnq for local messages) were not being
2133 preserved with the message after it was received.
2134
213513. When Exim was logging an SMTP synchronization error, it could sometimes log
2136 "next input=" as part of the text comprising the host identity instead of
2137 the correct text. The code was using the same buffer for two different
2138 strings. However, depending on which order the printing function evaluated
2139 its arguments, the bug did not always show up. Under Linux, for example, my
2140 test suite worked just fine.
2141
214214. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
2143 doesn't work with some older versions of Perl. It has been changed to "my",
2144 which in any case is probably the better facility to use.
2145
214615. A really picky compiler found some instances of statements for creating
2147 error messages that either had too many or two few arguments for the format
2148 string.
2149
215016. The size of the buffer for calls to the DNS resolver has been increased
2151 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
2152 for addresses that have a lot of PTR records. This alleviates a problem; it
2153 does not fully solve it.
2154
215517. A dnsdb lookup for PTR records that receives more data than will fit in the
2156 buffer now truncates the list and logs the incident, which is the same
2157 action as happens when Exim is looking up a host name and its aliases.
2158 Previously in this situation something unpredictable would happen;
2159 sometimes it was "internal error: store_reset failed".
2160
216118. If a server dropped the connection unexpectedly when an Exim client was
2162 using GnuTLS and trying to read a response, the client delivery process
2163 crashed while trying to generate an error log message.
2164
216519. If a "warn" verb in an ACL added multiple headers to a message in a single
2166 string, for example:
2167
2168 warn message = H1: something\nH2: something
2169
2170 the text was added as a single header line from Exim's point of view
2171 though it ended up OK in the delivered message. However, searching for the
2172 second and subsequent header lines using $h_h2: did not work. This has been
2173 fixed. Similarly, if a system filter added multiple headers in this way,
2174 the routers could not see them.
2175
217620. Expanded the error message when iplsearch is called with an invalid key to
2177 suggest using net-iplsearch in a host list.
2178
217921. When running tests using -bh, any delays imposed by "delay" modifiers in
2180 ACLs are no longer actually imposed (and a message to that effect is
2181 output).
2182
218322. If a "gecos" field in a passwd entry contained escaped characters, in
2184 particular, if it contained a \" sequence, Exim got it wrong when building
2185 a From: or a Sender: header from that name. A second bug also caused
2186 incorrect handling when an unquoted " was present following a character
2187 that needed quoting.
2188
218923. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
2190 was not being matched caselessly.
2191
219224. Arranged for all hyphens in the exim.8 source to be escaped with
2193 backslashes.
2194
219525. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
2196 properly. Recipient callout cache records were still being keyed to include
2197 the sender, even when use_sender was set false. This led to far more
2198 callouts that were necessary. The sender is no longer included in the key
2199 when use_sender is false.
2200
220126. Added "control = submission" modifier to ACLs.
2202
220327. Added the ${base62d: operator to decode base 62 numbers.
2204
220528. dnsdb lookups can now access SRV records.
2206
220729. CONFIGURE_OWNER can be set at build time to define an alternative owner for
2208 the configuration file.
2209
221030. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
2211 (-v) mode. This makes the output for a verbose queue run more intelligible.
2212
221331. Added a use_postmaster feature to recipient callouts.
2214
221532. Added the $body_zerocount variable, containing the number of binary zero
2216 bytes in the message body.
2217
221833. The time of last modification of the "new" subdirectory is now used as the
2219 "mailbox time last read" when there is a quota error for a maildir
2220 delivery.
2221
222234. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
2223
222435. Added +ignore_unknown as a special item in host lists.
2225
222636. Code for decoding IPv6 addresses in host lists is now included, even if
2227 IPv6 support is not being compiled. This fixes a bug in which an IPv6
2228 address was recognized as an IP address, but was then not correctly decoded
2229 into binary, causing unexpected and incorrect effects when compared with
2230 another IP address.
2231
2232
2233Exim version 4.34
2234-----------------
2235
2236 1. Very minor rewording of debugging text in manualroute to say "list of
2237 hosts" instead of "hostlist".
2238
2239 2. If verify=header_syntax was set, and a header line with an unqualified
2240 address (no domain) and a large number of spaces between the end of the
2241 name and the colon was received, the reception process suffered a buffer
2242 overflow, and (when I tested it) crashed. This was caused by some obsolete
2243 code that should have been removed. The fix is to remove it!
2244
2245 3. When running in the test harness, delay a bit after writing a bounce
2246 message to get a bit more predictability in the log output.
2247
2248 4. Added a call to search_tidyup() just before forking a reception process. In
2249 theory, someone could use a lookup in the expansion of smtp_accept_max_
2250 per_host which, without the tidyup, could leave open a database connection.
2251
2252 5. Added the variables $recipient_data and $sender_data which get set from a
2253 lookup success in an ACL "recipients" or "senders" condition, or a router
2254 "senders" option, similar to $domain_data and $local_part_data.
2255
2256 6. Moved the writing of debug_print from before to after the "senders" test
2257 for routers.
2258
2259 7. Change 4.31/66 (moving the time when the Received: is generated) caused
2260 problems for message scanning, either using a data ACL, or using
2261 local_scan() because the Received: header was not generated till after they
2262 were called (in order to set the time as the time of reception completion).
2263 I have revised the way this works. The header is now generated after the
2264 body is received, but before the ACL or local_scan() are called. After they
2265 are run, the timestamp in the header is updated.
2266
2267
2268Exim version 4.33
2269-----------------
2270
2271 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
2272 before starting a queue runner without re-exec. This happened only when
2273 deliver_drop_privilege was set or when the Exim user was set to root. The
2274 effect of the bug was that timeouts during subsequent deliveries caused
2275 crashes instead of being properly handled. The handler is now left at its
2276 default (and expected) setting.
2277
2278 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
2279 message, again when deliver_drop_privilege is set or Exim is run as root.
2280 The bug described in (1) was not present in this case, but the tidying up
2281 of the other signals was missing. I have made the two cases consistent.
2282
2283 3. The ignore_target_hosts setting on a manualroute router was being ignored
2284 for hosts that were looked up using the /MX notation.
2285
2286 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
2287 in domain lists.
2288
2289 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
2290 operated on the sender address. After changing the $sender_address to <>
2291 for the sender address verify, Exim was re-instated it as the original
2292 (before rewriting) address, but remembering that it had rewritten it, so it
2293 wasn't rewriting it again. This bug also had the effect of breaking the
2294 sender address verification caching when the sender address was rewritten.
2295
2296 6. The ignore_target_hosts option was being ignored by the ipliteral router.
2297 This has been changed so that if the ip literal address matches
2298 ignore_target_hosts, the router declines.
2299
2300 7. Added expansion conditions match_domain, match_address, and match_local_
2301 part (NOT match_host).
2302
2303 8. The placeholder for the Received: header didn't have a length field set.
2304
2305 9. Added code to Exim itself and to exim_lock to test for a specific race
2306 condition that could lead to file corruption when using MBX delivery. The
2307 issue is with the lockfile that is created in /tmp. If this file is removed
2308 after a process has opened it but before that process has acquired a lock,
2309 there is the potential for a second process to recreate the file and also
2310 acquire a lock. This could lead to two Exim processes writing to the file
2311 at the same time. The added code performs the same test as UW imapd; it
2312 checks after acquiring the lock that its file descriptor still refers to
2313 the same named file.
2314
231510. The buffer for building added header lines was of fixed size, 8192 bytes.
2316 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
2317 when Exim is built.
2318
231911. Added the smtp_active_hostname option. If used, this will typically be made
2320 to depend on the incoming interface address. Because $interface_address is
2321 not set up until the daemon has forked a reception process, error responses
2322 that can happen earlier (such as "too many connections") no longer contain
2323 a host name.
2324
232512. If an expansion in a condition on a "warn" statement fails because a lookup
2326 defers, the "warn" statement is abandoned, and the next ACL statement is
2327 processed. Previously this caused the whole ACL to be aborted.
2328
232913. Added the iplsearch lookup type.
2330
233114. Added ident_timeout as a log selector.
2332
233315. Added tls_certificate_verified as a log selector.
2334
233516. Added a global option tls_require_ciphers (compare the smtp transport
2336 option of the same name). This controls incoming TLS connections.
2337
233817. I finally figured out how to make tls_require_ciphers do a similar thing
2339 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
2340 before starting the TLS session.
2341
234218. Tabs are now shown as \t in -bP output.
2343
234419. If the log selector return_path_on_delivery was set, Exim crashed when
2345 bouncing a message because it had too many Received: header lines.
2346
234720. If two routers both had headers_remove settings, and the first one included
2348 a superfluous trailing colon, the final name in the first list and the
2349 first name in the second list were incorrectly joined into one item (with a
2350 colon in the middle).
2351
2352
2353Exim version 4.32
2354-----------------
2355
2356 1. Added -C and -D options to the exinext utility, mainly to make it easier
2357 to include in the automated testing, but these could be helpful when
2358 multiple configurations are in use.
2359
2360 2. The exinext utility was not formatting the output nicely when there was
2361 an alternate port involved in the retry record key, nor when there was a
2362 message id as well (for retries that were specific to a specific message
2363 and a specific host). It was also confused by IPv6 addresses, because of
2364 the additional colons they contain. I have fixed the IPv4 problem, and
2365 patched it up to do a reasonable job for IPv6.
2366
2367 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
2368 delivery, the log line now contains "pipelined" if PIPELINING was used.
2369
2370 4. An SMTP transport process used to panic and die if the bind() call to set
2371 an explicit outgoing interface failed. This has been changed; it is now
2372 treated in the same way as a connect() failure.
2373
2374 5. A reference to $sender_host_name in the part of a conditional expansion
2375 that was being skipped was still causing a DNS lookup. This no longer
2376 occurs.
2377
2378 6. The def: expansion condition was not recognizing references to header lines
2379 that used bh_ and bheader_.
2380
2381 7. Added the _cache feature to named lists.
2382
2383 8. The code for checking quota_filecount in the appendfile transport was
2384 allowing one more file than it should have been.
2385
2386 9. For compatibility with Sendmail, the command line option
2387
2388 -prval:sval
2389
2390 is equivalent to
2391
2392 -oMr rval -oMs sval
2393
2394 and sets the incoming protocol and host name (for trusted callers). The
2395 host name and its colon can be omitted when only the protocol is to be set.
2396 Note the Exim already has two private options, -pd and -ps, that refer to
2397 embedded Perl. It is therefore impossible to set a protocol value of "d" or
2398 "s", but I don't think that's a major issue.
2399
240010. A number of refactoring changes to the code, none of which should affect
2401 Exim's behaviour:
2402
2403 (a) The number of logging options was getting close to filling up the
2404 32-bit word that was used as a bit map. I have split them into two classes:
2405 those that are passed in the argument to log_write(), and those that are
2406 only ever tested independently outside of that function. These are now in
2407 separate 32-bit words, so there is plenty of room for expansion again.
2408 There is no change in the user interface or the logging behaviour.
2409
2410 (b) When building, for example, log lines, the code previously used a
2411 macro that called string_cat() twice, in order to add two strings. This is
2412 not really sufficiently general. Furthermore, there was one instance where
2413 it was actually wrong because one of the argument was used twice, and in
2414 one call a function was used. (As it happened, calling the function twice
2415 did not affect the overall behaviour.) The macro has been replaced by a
2416 function that can join an arbitrary number of extra strings onto a growing
2417 string.
2418
2419 (c) The code for expansion conditions now uses a table and a binary chop
2420 instead of a serial search (which was left over from when there were very
2421 few conditions). Also, it now recognizes conditions like "pam" even when
2422 the relevant support is not compiled in: a suitably worded error message is
2423 given if an attempt is made to use such a condition.
2424
242511. Added ${time_interval:xxxxx}.
2426
242712. A bug was causing one of the ddress fields not to be passed back correctly
2428 from remote delivery subprocesses. The field in question was not being
2429 subsequently used, so this caused to problems in practice.
2430
243113. Added new log selectors queue_time and deliver_time.
2432
243314. Might have fixed a bug in maildirsizefile handling that threw up
2434 "unexpected character" debug warnings, and recalculated the data
2435 unnecessarily. In any case, I expanded the warning message to give more
2436 information.
2437
243815. Added the message "Restricted characters in address" to the statements in
2439 the default ACL that block characters like @ and % in local parts.
2440
244116. Change 71 for release 4.31 proved to be much less benign that I imagined.
2442 Three changes have been made:
2443
2444 (a) There was a serious bug; a negative response to MAIL caused the whole
2445 recipient domain to be cached as invalid, thereby blocking all messages
2446 to all local parts at the same domain, from all senders. This bug has
2447 been fixed. The domain is no longer cached after a negative response to
2448 MAIL if the sender used is not empty.
2449
2450 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
2451 been restored.
2452
2453 (c) A new callout option, "use_sender" has been added for people who want
2454 the modified behaviour.
2455
2456
2457Exim version 4.31
2458-----------------
2459
2460 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
2461 Larry Rosenman.
2462
2463 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
2464 indeed breaks things for older releases.
2465
2466 3. Added additional logging to the case where there is a problem reading data
2467 from a filter that is running in a subprocess using a pipe, in order to
2468 try to track down a specific problem.
2469
2470 4. Testing facility fudge: when running in the test harness and attempting
2471 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
2472 getting "No route to host". Convert this to a timeout.
2473
2474 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
2475 warning.
2476
2477 6. Some OS don't have socklen_t but use size_t instead. This affects the
2478 fifth argument of getsockopt() amongst other things. This is now
2479 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
2480 can be set for individual OS. I have set it for SunOS5, OSF1, and
2481 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
2482 some earlier ones do not.
2483
2484 7. Change 4.30/15 was not doing the test caselessly.
2485
2486 8. The standard form for an IPv6 address literal was being rejected by address
2487 parsing in, for example, MAIL and RCPT commands. An example of this kind of
2488 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
2489 this, as well as the form without the "IPv6" on the front (but only when
2490 address literals are enabled, of course).
2491
2492 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
2493
249410. Exim crashed if a message with an empty sender address specified by -f
2495 encountered a router with an errors_to setting. This could be provoked only
2496 by a command such as
2497
2498 exim -f "" ...
2499
2500 where an empty string was supplied; "<>" did not hit this bug.
2501
250211. Installed PCRE release 4.5.
2503
250412. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
2505 remained set. It is now erased.
2506
250713. exiqgrep wasn't working on MacOS X because it didn't correctly compute
2508 times from message ids (which are base 36 rather than the normal 62).
2509
251014. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
2511 were being counted as actual protocol errors, and logged if the log
2512 selector +smtp_protocol_error was set. One cannot be perfect in this test,
2513 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
2514 and DATA following a set of rejected RCPTs do not count as protocol errors.
2515 In other words, Exim assumes they were pipelined, though this may not
2516 actually be the case. Of course, in all cases the client gets an
2517 appropriate error code.
2518
251915. If a lookup fails in an ACL condition, a message about the failure may
2520 be available; it is used if testing the ACL cannot continue, because most
2521 such messages specify what the cause of the deferral is. However, some
2522 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
2523 that caused an old message to be retained and used if a later statement
2524 caused a defer, replacing the real cause of the deferral.
2525
252616. If an IP address had so many PTR records that the DNS lookup buffer
2527 was not large enough to hold them, Exim could crash while trying to process
2528 the truncated data. It now detects and logs this case.
2529
253017. Further to 4.21/58, another change has been made: if (and only if) the
2531 first line of a message (the first header line) ends with CRLF, a bare LF
2532 in a subsequent header line has a space inserted after it, so as not to
2533 terminate the header.
2534
253518. Refactoring: tidied an ugly bit of code in appendfile that copied data
2536 unnecessarily, used atoi() instead of strtol(), and didn't check the
2537 termination when getting file sizes from file names by regex.
2538
253919. Completely re-implemented the support for maildirsize files, in the light
2540 of a number of problems with the previous contributed implementation
2541 (4.30/29). In particular:
2542
2543 . If the quota is zero, the maildirsize file is maintained, but no quota is
2544 imposed.
2545
2546 . If the maildir directory does not exist, it is created before any attempt
2547 to write a maildirsize file.
2548
2549 . The quota value in the file is just a cache; if the quota is changed in
2550 the transport, the new value overrides.
2551
2552 . A regular expression is available for excluding directories from the
2553 count.
2554
255520. The autoreply transport checks the characters in options that define the
2556 message's headers; it allows continued headers, but it was checking with
2557 isspace() after an embedded newline instead of explicitly looking for a
2558 space or a tab.
2559
256021. If all the "regular" hosts to which an address was routed had passed their
2561 expiry times, and had not reached their retry times, the address was
2562 bounced, even if fallback hosts were defined. Now Exim should go on to try
2563 the fallback hosts.
2564
256522. Increased buffer sizes in the callout code from 1024 to 4096 to match the
2566 equivalent code in the SMTP transport. Some hosts send humungous responses
2567 to HELO/EHLO, more than 1024 it seems.
2568
256923. Refactoring: code in filter.c used (void *) for "any old type" but this
2570 gives compiler warnings in some environments. I've now done it "properly",
2571 using a union.
2572
257324. The replacement for inet_ntoa() that is used with gcc on IRIX systems
2574 (because of problems with the built-in one) was declared to return uschar *
2575 instead of char *, causing compiler failure.
2576
257725. Fixed a file descriptor leak when processing alias/forward files.
2578
257926. Fixed a minor format string issue in dbfn.c.
2580
258127. Typo in exim.c: ("dmbnz" for "dbmnz").
2582
258328. If a filter file refered to $h_xxx or $message_headers, and the headers
2584 contained RFC 2047 "words", Exim's memory could, under certain conditions,
2585 become corrupted.
2586
258729. When a sender address is verified, it is cached, to save repeating the test
2588 when there is more than one recipient in a message. However, when the
2589 verification involves a callout, it is possible for different callout
2590 options to be set for different recipients. It is too complicated to keep
2591 track of this in the cache, so now Exim always runs a verification when a
2592 callout is required, relying on the callout cache for the optimization.
2593 The overhead is duplication of the address routing, but this should not be
2594 too great.
2595
259630. Fixed a bug in callout caching. If a RCPT command caused the sender address
2597 to be verified with callout=postmaster, and the main callout worked but the
2598 postmaster check failed, the verification correctly failed. However, if a
2599 subsequent RCPT command asked for sender verification *without* the
2600 postmaster check, incorrect caching caused this verification also to fail,
2601 incorrectly.
2602
260331. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
2604 it was not caching the DNS options (qualify_single, search_parents) that
2605 were used when the lookup failed. A subsequent lookup with different
2606 options therefore always gave the same answer, though there were cases
2607 where it should not have. (Example: a "domains = !$mx_any" option on a
2608 dnslookup router: the "domains" option is always processed without any
2609 widening, but the router might have qualify_single set.) Now Exim uses the
2610 cached value only when the same options are set.
2611
261232. Added John Jetmore's "exipick" utility to the distribution.
2613
261433. GnuTLS: When an attempt to start a TLS session fails for any reason other
2615 than a timeout (e.g. a certificate is required, and is not provided), an
2616 Exim server now closes the connection immediately. Previously it waited for
2617 the client to close - but if the client is SSL, it seems that they each
2618 wait for each other, leading to a delay before one of them times out.
2619
262034: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
2621 maintained 0.8.x compatibility because I don't think many are using it, and
2622 it is clearly obsolete.
2623
262435. Added TLS support for CRLs: a tls_crl global option and one for the smtp
2625 transport.
2626
262736. OpenSSL: $tls_certificate_verified was being set to 1 even if the
2628 client certificate was expired. A simple patch fixes this, though I don't
2629 understand the full logic of why the verify callback is called multiple
2630 times.
2631
263237. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
2633 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
2634 which causes problems with some clients (such as the Certicom SSL Plus
2635 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
2636 disables the coutermeasure allowing Eudora to connect."
2637
263838. Exim was not checking that a write() to a log file succeeded. This could
2639 lead to Bad Things if a log got too big, in particular if it hit a file
2640 size limit. Exim now panics and dies if it cannot write to a log file, just
2641 as it does if it cannot open a log file.
2642
264339. Modified OS/Makefile-Linux so that it now contains
2644
2645 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
2646
2647 The two -D definitions ensure that Exim is compiled with large file
2648 support, which makes it possible to handle log files that are bigger than
2649 2^31.
2650
265140. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
2652 instance) a domain was checked against a named list that involved a lookup,
2653 causing $domain_data to be set, then another domain was checked against the
2654 same list, then the first domain was re-checked, the value of $domain_data
2655 after the final check could be wrong. In particular, if the second check
2656 failed, it could be set empty. This bug probably also applied to
2657 $localpart_data.
2658
265941. The strip_trailing_dot option was not being applied to the address given
2660 with the -f command-line option.
2661
266242. The code for reading a message's header from the spool was incrementing
2663 $received_count, but never initializing it. This meant that the value was
2664 incorrect (doubled) while delivering a message in the same process in which
2665 it was received. In the most common configuration of Exim, this never
2666 happens - a fresh exec is done - but it can happen when
2667 deliver_drop_privilege is set.
2668
266943. When Exim logs an SMTP synchronization error - client data sent too soon -
2670 it now includes up to 150 characters of the unexpected data in the log
2671 line.
2672
267344. The exim_dbmbuild utility uses fixed size buffers for reading input lines
2674 and building data strings. The size of both of these buffers was 10 000
2675 bytes - far larger than anybody would *ever* want, thought I. Needless to
2676 say, somebody hit the limit. I have increased the maximum line length to
2677 20 000 and the maximum data length of concatenated lines to 100 000. I have
2678 also fixed two bugs, because there was no checking on these buffers. Tsk,
2679 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
2680 buffer is too small.
2681
268245. The exim_dbmbuild utility did not support quoted keys, as Exim does in
2683 lsearch lookups. Now it does.
2684
268546. When parsing a route_list item in a manualroute router, a fixed-length
2686 buffer was used for the list of hosts. I made this 1024 bytes long,
2687 thinking that nobody would ever have a list of hosts that long. Wrong.
2688 Somebody had a whole pile of complicated expansion conditions, and the
2689 string was silently truncated, leading to an expansion error. It turns out
2690 that it is easier to change to an unlimited length (owing to other changes
2691 that have happened since this code was originally written) than to build
2692 structure for giving a limitation error. The length of the item that
2693 expands into the list of hosts is now unlimited.
2694
269547. The lsearch lookup could not handle data where the length of text line was
2696 more than 4095 characters. Such lines were truncated, leading to shortened
2697 data being returned. It should now handle lines of any length.
2698
269948. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
2700 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
2701 ACL").
2702
270349. Cosmetic tidy to scripts like exicyclog that are generated by globally
2704 replacing strings such as BIN_DIRECTORY in a source file: the replacement
2705 no longer happens in comment lines. A list of replacements is now placed
2706 at the head of all of the source files, except those whose only change is
2707 to replace PERL_COMMAND in the very first #! line.
2708
270950. Replaced the slow insertion sort in queue.c, for sorting the list of
2710 messages on the queue, with a bottom-up merge sort, using code contributed
2711 by Michael Haardt. This should make operations like -bp somewhat faster on
2712 large queues. It won't affect queue runners, except when queue_run_in_order
2713 is set.
2714
271551. Installed eximstats 1.31 in the distribution.
2716
271752. Added support for SRV lookups to the dnslookup router.
2718
271953. If an ACL referred to $message_body or $message_body_end, the value was not
2720 reset for any messages that followed in the same SMTP session.
2721
272254. The store-handling optimization for building very long strings was not
2723 differentiating between the different store pools. I don't think this
2724 actually made any difference in practice, but I've tidied it.
2725
272655. While running the routers to verify a sender address, $sender_address
2727 was still set to the sender address. This is wrong, because when routing to
2728 send a bounce to the sender, it would be empty. Therefore, I have changed
2729 it so that, while verifying a sender address, $sender_address is set to <>.
2730 (There is no change to what happens when verifying a recipient address.)
2731
273256. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
2733 target A or AAAA records (if not already returned) without resetting the
2734 qualify_single or search_parents options of the DNS resolver. These are
2735 inappropriate in this case because the targets of MX and SRV records must
2736 be FQDNs. A broken DNS record could cause trouble if it happened to have a
2737 target that, when qualified, matched something in the local domain. These
2738 two options are now turned off when doing these lookups.
2739
274057. It seems that at least some releases of Reiserfs (which does not have the
2741 concept of a fixed number of inodes) returns zero and not -1 for the
2742 number of available inodes. This interacted badly with check_spool_inodes,
2743 which assumed that -1 was the "no such thing" setting. What I have done is
2744 to check that the total number of inodes is greater than zero before doing
2745 the test of how many are available.
2746
274758. When a "warn" ACL statement has a log_message modifier, the message is
2748 remembered, and not repeated. This is to avoid a lot of repetition when a
2749 message has many recipients that cause the same warning to be written.
2750 Howewer, Exim was preserving the list of already written lines for an
2751 entire SMTP session, which doesn't seem right. The memory is now reset if a
2752 new message is started.
2753
275459. The "rewrite" debugging flag was not showing the result of rewriting in the
2755 debugging output unless log_rewrite was also set.
2756
275760. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
2758 of (int)(handle) when we know that handle contains (void *)(-1).
2759
276061. The Exim daemon panic-logs an error return when it closes the incoming
2761 connection. However "connection reset by peer" seems to be common, and
2762 isn't really an error worthy of noting specially, so that particular error
2763 is no long logged.
2764
276562. When Exim is trying to find all the local interfaces, it used to panic and
2766 die if the ioctl to get the interface flags failed. However, it seems that
2767 on at least one OS (Solaris 9) it is possible to have an interface that is
2768 included in the list of interfaces, but for which you get a failure error
2769 for this call. This happens when the interface is not "plumbed" into a
2770 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
2771 failure of the "get flags" call assumes that the interface is down.
2772
277363. Added a ${eval10: operator, which assumes all numbers are decimal. This
2774 makes life easier for people who are doing arithmetic on fields extracted
2775 from dates, where you often get leading zeros that should not be
2776 interpreted as octal.
2777
277864. Added qualify_domain to the redirect router, to override the global
2779 setting.
2780
278165. If a pathologically long header line contained very many addresses (the
2782 report of this problem mentioned 10 000) and each of them was rewritten,
2783 Exim could use up a very large amount of memory. (It kept on making new
2784 copies of the header line as it rewrote, and never released the old ones.)
2785 At the expense of a bit more processing, the header rewriting function has
2786 been changed so that it no longer eats memory in this way.
2787
278866. The generation of the Received: header has been moved from the time that a
2789 message starts to be received, to the time that it finishes. The timestamp
2790 in the Received: header should now be very close to that of the <= log
2791 line. There are two side-effects of this change:
2792
2793 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
2794 logged header lines no longer include the local Received: line, because
2795 it has not yet been created. The same applies to a copy of the message
2796 that is returned to a non-SMTP sender when a message is rejected.
2797
2798 (b) When a filter file is tested using -bf, no additional Received: header
2799 is added to the test message. After some thought, I decided that this
2800 is a bug fix.
2801
2802 This change does not affect the value of $received_for. It is still set
2803 after address rewriting, but before local_scan() is called.
2804
280567. Installed the latest Cygwin-specific files from the Cygwin maintainer.
2806
280768. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
2808 gave an unhelpful panic error message, and a defer error. I have managed to
2809 change this behaviour so that it now rejects any supplied certificate,
2810 which seems right, as the list of acceptable certificates is empty.
2811
281269. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
2813 gave an unhelpful defer error. I have not managed to make this reject any
2814 supplied certificates, but the error message it gives is "no certificate
2815 supplied", which is not helpful.
2816
281770. exigrep's output now also includes lines that are not associated with any
2818 message, but which match the given pattern. Implemented by a patch from
2819 Martin Sluka, which also tidied up the Perl a bit.
2820
282171. Recipient callout verification, like sender verification, was using <> in
2822 the MAIL FROM command. This isn't really the right thing, since the actual
2823 sender may affect whether the remote host accepts the recipient or not. I
2824 have changed it to use the actual sender in the callout; this means that
2825 the cache record is now keyed on a recipient/sender pair, not just the
2826 recipient address. There doesn't seem to be a real danger of callout loops,
2827 since a callout by the remote host to check the sender would use <>.
2828 [SEE ABOVE: changed after hitting problems.]
2829
283072. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
2831 temporary errors. However, in the case of such a code being given after
2832 the end of a data transmission (i.e. after ".") Exim was failing to write
2833 a retry record for the message. (Yes, there was some broken host that was
2834 actually sending 8xx at this point.)
2835
283673. An unknown lookup type in a host list could cause Exim to panic-die when
2837 the list was checked. (An example that provoked this was putting <; in the
2838 middle of a list instead of at the start.) If this happened during a DATA
2839 ACL check, a -D file could be left lying around. This kind of configuration
2840 error no longer causes Exim to die; instead it causes a defer errror. The
2841 incident is still logged to the main and panic logs.
2842
284374. Buglet left over from Exim 3 conversion. The message "too many messages
2844 in one connection" was written to the rejectlog but not the mainlog, except
2845 when address rewriting (yes!) was being logged.
2846
284775. Added write_rejectlog option.
2848
284976. When a system filter was run not as root (that is, when system_filter_user
2850 was set), the values of the $n variables were not being returned to the
2851 main process; thus, they were not subsequently available in the $sn
2852 variables.
2853
285477. Added +return_path_on_delivery log selector.
2855
285678. A connection timeout was being treated differently from recipients deferred
2857 when testing hosts_max_try with a message that was older than the host's
2858 retry timeout. (The host should not be counted, thus allowing all hosts to
2859 be tried at least once before bouncing.) This may have been the cause of an
2860 occasionally reported bug whereby a message would remain on the queue
2861 longer than the retry timeout, but would be bounced if a delivery was
2862 forced. I say "may" because I never totally pinned down the problem;
2863 setting up timeout/retry tests is difficult. See also the next item.
2864
286579. The ultimate address timeout was not being applied to errors that involved
2866 a combination of host plus message (for example, a timeout on a MAIL
2867 command). When an address resolved to a number of possible hosts, and they
2868 were not all tried for each delivery (e.g. because of hosts_max_try), a
2869 message could remain on the queue longer than the retry timeout.
2870
287180. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
2872 Haardt.
2873
287481. Fixed an obscure SMTP outgoing bug which required at least the following
2875 conditions: (a) there was another message waiting for the same server;
2876 (b) the server returned 5xx to all RCPT commands in the first message so
2877 that the message was not completed; (c) the server dropped the connection
2878 or gave a negative response to the RSET that Exim sends to abort the
2879 transaction. The observed case was a dropped connection after DATA that had
2880 been sent in pipelining mode. That is, the server had advertised PIPELINING
2881 but was not implementing it correctly. The effect of the bug was incorrect
2882 behaviour, such as trying another host, and this could lead to a crash.
2883
2884
2885Exim version 4.30
2886-----------------
2887
2888 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
2889 and daemon.c were passed as pointers to ints; they should have been
2890 pointers to socklen_t variables (which are typically unsigned ints).
2891
2892 2. Some signed/unsigned type warnings in the os.c file for Linux have been
2893 fixed.
2894
2895 3. Fixed a really odd bug that affected only the testing scheme; patching a
2896 certain fixed string in the binary changed the value of another string that
2897 happened to be identical to the end of the original first string.
2898
2899 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
2900 name", it returns that address as the IP address. On some operating
2901 systems (e.g. Solaris), it also passes back the IP address string as the
2902 "host name". However, on others (e.g. Linux), it passes back an empty
2903 string. Exim wasn't checking for this, and was changing the host name to an
2904 empty string, assuming it had been canonicized.
2905
2906 5. Although rare, it is permitted to have more than one PTR record for a given
2907 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
2908 all the names associated with an address, because they do in Solaris.
2909 However, it seems that they do not in Linux for data that comes from the
2910 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
2911 I found this out when I moved to a new Linux workstation and tried to run
2912 the Exim test suite.
2913
2914 To get round this problem I have changed the code so that it now does its
2915 own call to the DNS to look up PTR records when searching for a host name.
2916 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
2917 addresses that are only in /etc/hosts are still found.
2918
2919 This behaviour is, however, controlled by an option called host_lookup_
2920 order, which defaults to "bydns:byaddr". If people want to use the other
2921 order, or indeed, just use one or the other means of lookup, they can
2922 specify it in this variable.
2923
2924 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
2925 some operating systems, this comes back from gethostbyaddr() as an empty
2926 string, and this is what Exim used to test for. However, it seems that in
2927 other systems, "." is yielded. Exim now tests for this case too.
2928
2929 7. The values of check_spool_space and check_log_space are now held internally
2930 as a number of kilobytes instead of an absolute number of bytes. If a
2931 numbers is specified without 'K' or 'M', it is rounded up to the nearest
2932 kilobyte. This means that much larger values can be stored.
2933
2934 8. Exim monitor: an attempt to get the action menu when not actually pointing
2935 at a message produces an empty menu entitled "No message selected". This
2936 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
2937 no entries in it ("Shell widget menu has zero width and/or height"). So I
2938 have added a single, blank menu entry in this case.
2939
2940 9. Added ${quote_local_part.
2941
294210. MIME decoding is now applied to the contents of Subject: header lines when
2943 they are logged.
2944
294511. Now that a reference to $sender_host_address automatically causes a reverse
2946 lookup to occur if necessary (4.13/18), there is no need to arrange for a
2947 host lookup before query-style lookups in lists that might use this
2948 variable. This has therefore been abolished, and the "net-" prefix is no
2949 longer necessary for query-style lookups.
2950
295112. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
2952 have been a typo for LFLAGS, so it has been changed.
2953
295413. The install script calls Exim with "-C /dev/null" in order to find the
2955 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
2956 to be output. Howeve, since Exim outputs its version number before the
2957 error, it didn't break the script. It just looked ugly. I fixed this by
2958 always allowing "-C /dev/null" if the caller is root.
2959
296014. Ignore overlarge ACL variable number when reading spool file - insurance
2961 against a later release with more variables having written the file.
2962
296315. The standard form for an IPv6 address literal was being rejected by EHLO.
2964 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
2965 this, as well as the form without the "IPv6" on the front.
2966
296716. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
2968 OS/Makefile-Darwin file.
2969
297017. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
2971 only to LDAP libraries that do not have LDAP_OPT_DEREF.
2972
297318. After change 4.21/52, "%ld" was used to format the contents of the $inode
2974 variable. However, some OS use ints for inodes. I've added cast to long int
2975 to get rid of the compiler warning.
2976
297719. I had forgotten to lock out "/../" in configuration file names when
2978 ALT_CONFIG_PREFIX was set.
2979
298020. Routers used for verification do not need to specify transports. However,
2981 if such a router generated a host list, and callout was configured, Exim
2982 crashed, because it could not find a port number from the (non-existent)
2983 transport. It now assumes port 25 in this circumstance.
2984
298521. Added the -t option to exigrep.
2986
298722. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
2988 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
2989 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
2990 them is set, LOOKUP_LSEARCH is forced.
2991
299223. "exim -bV" now outputs a list of lookups that are included in the binary.
2993
299424. Added sender and host information to the "rejected by local_scan()" log
2995 line; previously there was no indication of these.
2996
299725. Added .include_if_exists.
2998
299926. Change 3.952/11 added an explicit directory sync on top of a file sync for
3000 Linux. It turns out that not all file systems support this. Apparently some
3001 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
3002 do it.) To cope with this, the error EINVAL, which means that sync-ing is
3003 not supported on the file descriptor, is now ignored when Exim is trying to
3004 sync a directory. This applies only to Linux.
3005
300627. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
3007
300828. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
3009 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
3010 this.
3011
301229. Added support for maildirsize files from supplied patch (modified a bit).
3013
301430. The use of :fail: followed by an empty string could lead Exim to respond to
3015 sender verification failures with (e.g.):
3016
3017 550 Verification failed for <xxx>
3018 550 Sender verify failed
3019
3020 where the first response line was missing the '-' that indicates it is not
3021 the final line of the response.
3022
302331. The loop for finding the name of the user that called Exim had a hardwired
3024 limit of 10; it now uses the value of finduser_retries, which is used for
3025 all other user lookups.
3026
302732. Added $received_count variable, available in data and not_smtp ACLs, and at
3028 delivery time.
3029
303033. Exim was neglecting to zero errno before one call of strtol() when
3031 expanding a string and expecting an integer value. On some systems this
3032 resulted in spurious "integer overflow" errors. Also, it was casting the
3033 result into an int without checking.
3034
303534. Testing for a connection timeout using "timeout_connect" in the retry rules
3036 did not work. The code looks as if it has *never* worked, though it appears
3037 to have been documented since at least releast 1.62. I have made it work.
3038
303935. The "timeout_DNS" error in retry rules, also documented since at least
3040 1.62, also never worked. As it isn't clear exactly what this means, and
3041 clearly it isn't a major issue, I have abolished the feature by treating it
3042 as "timeout", and writing a warning to the main and panic logs.
3043
304436. The display of retry rules for -brt wasn't always showing the error code
3045 correctly.
3046
304737. Added new error conditions to retry rules: timeout_A, timeout_MX,
3048 timeout_connect_A, timeout_connect_MX.
3049
305038. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
3051 to the empty sender.
3052
305339. The daemon was not analysing the content of -oX till after it had closed
3054 stderr and disconnected from the controlling terminal. This meant that any
3055 syntax errors were only noted on the panic log, and the return code from
3056 the command was 0. By re-arranging the code a little, I've made the
3057 decoding happen first, so such errors now appear on stderr, and the return
3058 code is 1. However, the actual setting up of the sockets still happens in
3059 the disconnected process, so errors there are still only recorded on the
3060 panic log.
3061
306240. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
3063 connections (as happens on some IP stacks) was logged at start up time as
3064 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
3065 it from "IPv6 and IPv4", which means that two separate sockets are being
3066 used.
3067
306841. The debug output for gethostbyname2() or getipnodebyname() failures now
3069 says whether AF_INET or AF_INET6 was passed as an argument.
3070
307142. Exiwhat output was messed up when time zones were included in log
3072 timestamps.
3073
307443. Exiwhat now gives more information about the daemon's listening ports,
3075 and whether -tls-on-connect was used.
3076
307744. The "port" option of the smtp transport is now expanded.
3078
307945. A "message" modifier in a "warn" statement in a non-message ACL was being
3080 silently ignored. Now an error message is written to the main and panic
3081 logs.
3082
308346. There's a new ACL modifier called "logwrite" which writes to a log file
3084 as soon as it is encountered.
3085
308647. Added $local_user_uid and $local_user_gid at routing time.
3087
308848. Exim crashed when trying to verify a sender address that was being
3089 rewritten to "<>".
3090
309149. Exim was recognizing only a space character after ".include". It now also
3092 recognizes a tab character.
3093
309450. Fixed several bugs in the Perl script that creates the exim.8 man page by
3095 extracting the relevant information from the specification. The man page no
3096 longer contains scrambled data for the -d option, and I've added a section
3097 at the front about calling Exim under different names.
3098
309951. Added "extra_headers" argument to the "mail" command in filter files.
3100
310152. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
3102 crash.
3103
310453. Installed eximstats 1.29.
3105
310654. Added transport_filter_timeout as a generic transport option.
3107
310855. Exim no longer adds an empty Bcc: header to messages that have no To: or
3109 Cc: header lines. This was required by RFC 822, but it not required by RFC
3110 2822.
3111
311256. Exim used to add From:, Date:, and Message-Id: header lines to any
3113 incoming messages that did not have them. Now it does so only if the
3114 message originates locally, that is, if there is no associated remote host
3115 address. When Resent- header lines are present, this applies to the Resent-
3116 lines rather than the non-Resent- lines.
3117
311857. Drop incoming SMTP connection after too many syntax or protocol errors. The
3119 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
3120
312158. Messages for configuration errors now include the name of the main
3122 configuration file - useful now that there may be more than one file in a
3123 list (.included file names were always shown).
3124
312559. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
3126 for those rare installations that do not start the daemon as root or run it
3127 setuid root. I've cut out the call to initgroups() if the daemon is not
3128 root at that time.
3129
313060. The Exim user and group can now be bound into the binary as text strings
3131 that are looked up at the start of Exim's processing.
3132
313361. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
3134
313562. Added $mailstore_basename variable.
3136
313763. Installed patch to sieve.c from Michael Haardt.
3138
313964. When Exim failed to open the panic log after failing to open the main log,
3140 the original message it was trying to log was written to stderr and debug
3141 output, but if they were not available (the usual case in production), it
3142 was lost. Now it is written to syslog before the two lines that record the
3143 failures to open the logs.
3144
314565. Users' Exim filters run in subprocesses under the user's uid. It is
3146 possible for a "deliver" command or an alias in a "personal" command to
3147 provoke an address rewrite. If logging of address rewriting is configured,
3148 this fails because the process is not running as root or exim. There may be
3149 a better way of dealing with this, but for the moment (because 4.30 needs
3150 to be released), I have disabled address rewrite logging when running a
3151 filter in a non-root, non-exim process.
3152
3153
3154Exim version 4.24
3155-----------------
3156
3157 1. The buildconfig auxiliary program wasn't quoting the value set for
3158 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
3159 not defined. This bug was masked in 4.22 by the effect that was fixed in
3160 change 4.23/1.
3161
3162 2. Some messages that were rejected after a message id was allocated were
3163 shown as "incomplete" by exigrep. It no longer does this for messages that
3164 are rejected by local_scan() or the DATA or non-SMTP ACLs.
3165
3166 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
3167 have allow_domain_literals set, the ID did not get logged in the <= line.
3168 Domain literals are now always recognized in Message-ID: header lines.
3169
3170 4. The first argument for a ${extract expansion item is the key name or field
3171 number. Leading and trailing spaces in this item were not being ignored,
3172 causing some misleading effects.
3173
3174 5. When deliver_drop_privilege was set, single queue runner processes started
3175 manually (i.e. by the command "exim -q") or by the daemon (which uses the
3176 same command in the process it spins off) were not dropping privilege.
3177
3178 6. When the daemon running as "exim" started a queue runner, it always
3179 re-executed Exim in the spun-off process. This is a waste of effort when
3180 deliver_drop_privilege is set. The new process now just calls the
3181 queue-runner function directly.
3182
3183
3184Exim version 4.23
3185-----------------
3186
3187 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
3188 HEADERS_CHARSET.
3189
3190 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
3191 ignored. Though the use of -oP was forcing the writing of a pid file, it
3192 was always written to the default place.
3193
3194 3. If the message "no IP address found for host xxxx" is generated during
3195 incoming verification, it is now followed by identification of the incoming
3196 connection (so you can more easily find what provoked it).
3197
3198 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
3199
3200 5. Added some features to "harden" Exim a bit more against certain attacks:
3201
3202 (a) There is now a build-time option called FIXED_NEVER_USERS that can
3203 be put in Local/Makefile. This is like the never_users runtime option,
3204 but it cannot be overridden. The default setting is "root".
3205
3206 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
3207 prefix string with which any file named in a -C command line option
3208 must start.
3209
3210 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
3211 is retained for -C and -D only if the caller of Exim is root. Without
3212 it, the exim user may also use -C and -D and retain privilege.
3213
3214 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
3215 command line option is disabled.
3216
3217 6. Macro names set by the -D option must start with an upper case letter, just
3218 like macro names defined in the configuration file.
3219
3220 7. Added "dereference=" facility to LDAP.
3221
3222 8. Two instances of the typo "uknown" in the source files are fixed.
3223
3224 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
3225 the Configure-Makefile script screwed up while processing it.
3226
322710. Incorporated PCRE 4.4.
3228
322911. The SMTP synchronization check was not operating right at the start of an
3230 SMTP session. For example, it could not catch a HELO sent before the client
3231 waited for the greeting. There is now a check for outstanding input at the
3232 point when the greeting is written. Because of the duplex, asynchronous
3233 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
3234 way, but not yet received, when the check is performed.
3235
323612. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
3237 on TCP/IP sockets, because this apparently causes some broken clients to
3238 timeout.
3239
324013. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
3241 unchanged) from the Cygwin maintainer.
3242
324314. The code for -bV that shows what is in the binary showed "mbx" when maildir
3244 was supported instead of testing for mbx. Effectively a typo.
3245
324615. The spa authenticator server code was not checking that the input it
3247 received was valid base64.
3248
324916. The debug output line for the "set" modifier in ACLs was not showing the
3250 name of the variable that was being set.
3251
325217. Code tidy: the variable type "vtype_string" was never used. Removed it.
3253
325418. Previously, a reference to $sender_host_name did not cause a DNS reverse
3255 lookup on its own. Something else was needed to trigger the lookup. For
3256 example, a match in host_lookup or the need for a host name in a host list.
3257 Now, if $sender_host_name is referenced and the host name has not yet been
3258 looked up, a lookup is performed. If the lookup fails, the variable remains
3259 empty, and $host_lookup_failed is set to "1".
3260
326119. Added "eqi" as a case-independent comparison operator.
3262
326320. The saslauthd authentication condition could segfault if neither service
3264 nor realm was specified.
3265
326621. If an overflowing value such as "2048M" was set for message_size_limit, the
3267 error message that was logged was misleading, and incoming SMTP
3268 connections were dropped. The message is now more accurate, and temporary
3269 errors are given to SMTP connections.
3270
327122. In some error situations (such as 21 above) Exim rejects all SMTP commands
3272 (except RSET) with a 421 error, until QUIT is received. However, it was
3273 failing to send a response to QUIT.
3274
327523. The HELO ACL was being run before the code for helo_try_verify_hosts,
3276 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
3277 ACL is now run after the helo_try_verify_hosts code.
3278
327924. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
3280 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
3281 case-independent, so other case variants are also recognized). Apparently
3282 some systems use these upper case variants.
3283
328425. If more than two messages were waiting for the same host, and a transport
3285 filter was specified for the transport, Exim sent two messages over the
3286 same TCP/IP connection, and then failed with "socket operation on non-
3287 socket" when it tried to send the third.
3288
328926. Added Exim::debug_write and Exim::log_write for embedded Perl use.
3290
329127. The extern definition of crypt16() in expand.c was not being excluded when
3292 the OS had its own crypt16() function.
3293
329428. Added bounce_return_body as a new option, and bounce_return_size_limit
3295 as a preferred synonym for return_size_limit, both as an option and as an
3296 expansion variable.
3297
329829. Added LIBS=-liconv to OS/Makefile-OSF1.
3299
330030. Changed the default configuration ACL to relax the local part checking rule
3301 for addresses that are not in any local domains. For these addresses,
3302 slashes and pipe symbols are allowed within local parts, but the sequence
3303 /../ is explicitly forbidden.
3304
330531. SPA server authentication was not clearing the challenge buffer before
3306 using it.
3307
330832. log_message in a "warn" ACL statement was writing to the reject log as
3309 well as to the main log, which contradicts the documentation and doesn't
3310 seem right (because no rejection is happening). So I have stopped it.
3311
331233. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
3313 However, I am unable to do any testing of this.
3314
331534. Fixed an infelicity in the appendfile transport. When checking directories
3316 for a mailbox, to see if any needed to be created, it was accidentally
3317 using path names with one or more superfluous leading slashes; tracing
3318 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
3319
332035. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
3321 contents are added to the log line that is written for every discarded
3322 recipient. (Previously a log_message setting was ignored.)
3323
332436. The ${quote: operator now quotes the string if it is empty.
3325
332637. The install script runs exim in order to find its version number. If for
3327 some reason other than non-existence or emptiness, which it checks, it
3328 could not run './exim', it was installing it with an empty version number,
3329 i.e. as "exim-". This error state is now caught, and the installation is
3330 aborted.
3331
333238. An argument was missing from the function that creates an error message
3333 when Exim fails to connect to the socket for saslauthd authentication.
3334 This could cause Exim to crash, or give a corrupted message.
3335
333639. Added isip, isip4, and isip6 to ${if conditions.
3337
333840. The ACL variables $acl_xx are now saved with the message, and can be
3339 accessed later in routers, transports, and filters.
3340
334141. The new lookup type nwildlsearch is like wildlsearch, except that the key
3342 strings in the file are not string-expanded.
3343
334442. If a MAIL command specified a SIZE value that was too large to fit into an
3345 int variable, the check against message_size_limit failed. Such values are
3346 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
3347 day this will have to be increased, but I don't think I want to be around
3348 when emails are that large.
3349
3350
3351
3352Exim version 4.22
3353-----------------
3354
3355 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
3356 iconv() is not standard in FreeBSD.
3357
3358 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
3359 IPv6 enabled. The observed symptom was a segmentation fault on return from
3360 the function os_common_find_running_interfaces() in src/os.c.
3361
3362 3. In the check_special_case() function in daemon.c I had used "errno" as an
3363 argument name, which causes warnings on some systems. This was basically a
3364 typo, since it was named "eno" in the comments!
3365
3366 4. The code that waits for the clock to tick (at a resolution of some fraction
3367 of a second) so as to ensure message-id uniqueness was always waiting for
3368 at least one whole tick, when it could have waited for less. [This is
3369 almost certainly not relevant at current processor speeds, where it is
3370 unlikely to ever wait at all. But we try to future-proof.]
3371
3372 5. The function that sleeps for a time interval that includes fractions of a
3373 second contained a race. It did not block SIGALRM between setting the
3374 timer, and suspending (a couple of lines later). If the interval was short
3375 and the sigsuspend() was delayed until after it had expired, the suspension
3376 never ended. On busy systems this could lead to processes getting stuck for
3377 ever.
3378
3379 6. Some uncommon configurations may cause a lookup to happen in a queue runner
3380 process, before it forks any delivery processes. The open lookup caching
3381 mechanism meant that the open file or database connection was passed into
3382 the delivery process. The problem was that delivery processes always tidy
3383 up cached lookup data. This could cause a problem for the next delivery
3384 process started by the queue runner, because the external queue runner
3385 process does not know about the closure. So the next delivery process
3386 still has data in the lookup cache. In the case of a file lookup, there was
3387 no problem because closing a file descriptor in a subprocess doesn't affect
3388 the parent. However, if the lookup was caching a connection to a database,
3389 the connection was closed, and the second delivery process was likely to
3390 see errors such as "PGSQL: query failed: server closed the connection
3391 unexpectedly". The problem has been fixed by closing all cached lookups
3392 in a queue runner before running a delivery process.
3393
3394 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
3395 seem to have the "const" qualifier which it has on other OS. I've
3396 parameterised it.
3397
3398 8. Change 4.21/2 was too strict. It is only if there are two authenticators
3399 *of the same type* (client or server) with the same public name that an
3400 error should be diagnosed.
3401
3402 9. When Exim looked up a host name for an IP address, but failed to find the
3403 original IP address when looking up the host name (a safety check), it
3404 output the message "<ip address> does not match any IP for NULL", which was
3405 confusing, to say the least. The bug was that the host name should have
3406 appeared instead of "NULL".
3407
340810. Since release 3.03, if Exim is called by a uid other than root or the Exim
3409 user that is built into the binary, and the -C or -D options is used, root
3410 privilege is dropped before the configuration file is read. In addition,
3411 logging is switched to stderr instead of the normal log files. If the
3412 configuration then re-defines the Exim user, the unprivileged environment
3413 is probably not what is expected, so Exim logs a panic warning message (but
3414 proceeds).
3415
3416 However, if deliver_drop_privilege is set, the unprivileged state may well
3417 be exactly what is intended, so the warning has been cut out in that case,
3418 and Exim is allowed to try to write to its normal log files.
3419
3420
3421Exim version 4.21
3422-----------------
3423
3424 1. smtp_return_error_details was not giving details for temporary sender
3425 or receiver verification errors.
3426
3427 2. Diagnose a configuration error if two authenticators have the same public
3428 name.
3429
3430 3. Exim used not to create the message log file for a message until the first
3431 delivery attempt. This could be confusing when incoming messages were held
3432 for policy or load reasons. The message log file is now created at the time
3433 the message is received, and an initial "Received" line is written to it.
3434
3435 4. The automatically generated man page for command line options had a minor
3436 bug that caused no ill effects; however, a more serious problem was that
3437 the procedure for building the man page automatically didn't always
3438 operate. Consequently, release 4.20 contains an out-of-date version. This
3439 shouldn't happen again.
3440
3441 5. When building Exim with embedded Perl support, the script that builds the
3442 Makefile was calling 'perl' to find its compile-time parameters, ignoring
3443 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
3444
3445 6. The freeze_tell option was not being used for messages that were frozen on
3446 arrival, either by an ACL or by local_scan().
3447
3448 7. Added the smtp_incomplete_transaction log selector.
3449
3450 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
3451 was accepting AUTH without a new EHLO.
3452
3453 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
3454 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
3455 RFC.
3456
345710. Logging of TCP/IP connections (when configured) now happens in the main
3458 daemon process instead of the child process, so that the TCP/IP connection
3459 count is more accurate (but it can never be perfect).
3460
346111. The use of "drop" in a nested ACL was not being handled correctly in the
3462 outer ACL. Now, if condition failure induced by the nested "drop" causes
3463 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
3464 or "require"), the connection is dropped.
3465
346612. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
3467 that yield "discard" can now be used with an "accept" or a "discard" verb,
3468 but an error is generated for any others (because I can't see a useful way
3469 to define what should happen).
3470
347113. When an ACL is read dynamically from a file (or anywhere else), the lines
3472 are now processed in the same way as lines in the Exim configuration file.
3473 In particular, continuation lines are supported.
3474
347514. Added the "dnslists = a.b.c!=n.n.n.n" feature.
3476
347715. Added -ti meaning -t -i.
3478
347916. Check for letters, digits, hyphens, and dots in the names of dnslist
3480 domains, and warn by logging if others are found.
3481
348217. At least on BSD, alignment is not guarenteed for the array of ifreq's
3483 returned from GIFCONF when Exim is trying to find the list of interfaces on
3484 a host. The code in os.c has been modified to copy each ifreq to an aligned
3485 structure in all cases.
3486
3487 Also, in some cases, the returned ifreq's were being copied to a 'struct
3488 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
3489 means the last couple of bytes of an IPv6 address could be chopped if the
3490 ifreq contained only a normal sockaddr (14 bytes storage).
3491
349218. Named domain lists were not supported in the hosts_treat_as_local option.
3493 An entry such as +xxxx was not recognized, and was treated as a literal
3494 domain name.
3495
349619. Ensure that header lines added by a DATA ACL are included in the reject log
3497 if the ACL subsequently rejects the message.
3498
349920. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
3500 MD5 (which is deprecated).
3501
350221. When testing a filter file using -bf, Exim was writing a message when it
3503 took the sender from a "From " line in the message, but it was not doing so
3504 when it took $return_path from a Return-Path: header line. It now does.
3505
350622. If the contents of a "message" modifier for a "warn" ACL verb do not begin
3507 with a valid header line field name (a series of printing characters
3508 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
3509
351023. Changed "disc" in the source to "disk" to conform to the documentation and
3511 the book and for uniformity.
3512
351324. Ignore Sendmail's -Ooption=value command line item.
3514
351525. When execve() failed while trying to run a command in a pipe transport,
3516 Exim was returning EX_UNAVAILBLE (69) from the subprocess. However, this
3517 could be confused with a return value of 69 from the command itself. This
3518 has been changed to 127, the value the shell returns if it is asked to run
3519 a non-existent command. The wording for the related log line suggests a
3520 non-existent command as the problem.
3521
352226. If received_header_text expands to an empty string, do not add a Received:
3523 header line to the message. (Well, it adds a token one on the spool, but
3524 marks it "old" so that it doesn't get used or transmitted.)
3525
352627. Installed eximstats 1.28 (addition of -nt option).
3527
352828. There was no check for failure on the call to getsockname() in the daemon
3529 code. This can fail if there is a shortage of resources on the system, with
3530 ENOMEM, for example. A temporary error is now given on failure.
3531
353229. Contrary to the C standard, it seems that in some environments, the
3533 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
3534 program. Exim now does this explicitly; it affects the formatting of
3535 timestamps using strftime().
3536
353730. If exiqsumm was given junk data, it threw up some uninitialized variable
3538 complaints. I've now initialized all the variables, to avoid this.
3539
354032. Header lines added by a system filter were not being "seen" during
3541 transport-time rewrites.
3542
354333. The info_callback() function passed to OpenSSL is set up with type void
3544 (*)(SSL *, int, int), as described somewhere. However, when calling the
3545 function (actually a macro) that sets it up, the type void(*)() is
3546 expected. I've put in a cast to prevent warnings from picky compilers.
3547
354834. If a DNS black list lookup found a CNAME record, but there were no A
3549 records associated with the domain it pointed at, Exim crashed.
3550
355135. If a DNS black list lookup returned more than one A record, Exim ignored
3552 all but the first. It now scans all returned addresses if a particular IP
3553 value is being sought. In this situation, the contents of the
3554 $dnslist_value variable are a list of all the addresses, separated by a
3555 comma and a space.
3556
355736. Tightened up the rules for host name lookups using reverse DNS. Exim used
3558 to accept a host name and all its aliases if the forward lookup for any of
3559 them yielded the IP address of the incoming connection. Now it accepts only
3560 those names whose forward lookup yields the correct IP address. Any other
3561 names are discarded. This closes a loophole whereby a rogue DNS
3562 administrator could create reverse DNS records to break through a
3563 wildcarded host restriction in an ACL.
3564
356537. If a user filter or a system filter that ran in a subprocess used any of
3566 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
3567 the wrong values were passed to the pipe command ($thisaddress had the
3568 value of $0, $0 had the value of $1, etc). This bug was introduced by
3569 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
3570
357138. Improved the line breaking for long SMTP error messages from ACLs.
3572 Previously, if there was no break point between 40 and 75 characters, Exim
3573 left the rest of the message alone. Two changes have been made: (a) I've
3574 reduced the minimum length to 35 characters; (b) if it can't find a break
3575 point between 35 and 75 characters, it looks ahead and uses the first one
3576 that it finds. This may give the occasional overlong line, but at least the
3577 remaining text gets split now.
3578
357939. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
3580 file descriptors might be low, and that setting 1000 would always raise it.
3581 It turns out that in some environments, the limit is already over 1000 and
3582 that lowering it causes trouble. So now Exim takes care not to decrease it.
3583
358440. When delivering a message, the value of $return_path is set to $sender_
3585 address at the start of routing (routers may change the value). By an
3586 oversight, this default was not being set up when an address was tested by
3587 -bt or -bv, which affected the outcome if any router or filter referred to
3588 $return_path.
3589
359041. The idea of the "warn" ACL verb is that it adds a header or writes to the
3591 log only when "message" or "log_message" are set. However, if one of the
3592 conditions was an address verification, or a call to a nested ACL, the
3593 messages generated by the underlying test were being passed through. This
3594 no longer happens. The underlying message is available in $acl_verify_
3595 message for both "message" and "log_message" expansions, so it can be
3596 passed through if needed.
3597
359842. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
3599 new expansion $bh_ to give the encoded byte string without charset
3600 translation. Translation happens only if iconv() is available; HAVE_ICONV
3601 indicates this at build time. HEADERS_CHARSET gives the charset to
3602 translate to; headers_charset can change it in the configuration, and
3603 "headers charset" can change it in an individual filter file.
3604
360543. Now that we have a default RFC 2047 charset (see above), the code in Exim
3606 that creates RFC 2047 encoded "words" labels them as that charset instead
3607 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
3608 expansion operator; (ii) when Exim creates a From: line for a local
3609 message; (iii) when a header line is rewritten to include a "phrase" part.
3610
361144. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
3612 buggy, causing it to skip the first lines of messages whose message ID
3613 ended in 'D'. This would not have bitten before Exim release 4.14, because
3614 message IDs were unlikely to end in 'D' before then. The effect was to have
3615 incorrect size information for certain domains.
3616
361745. #include "config.h" was missing at the start of the crypt16.c module. This
3618 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
3619 noticed.
3620
362146. If there was a timeout during a "random" callout check, Exim treated it as
3622 a failure of the random address, and carried on sending RSET and the real
3623 address. If the delay was just some slowness somewhere, the response to the
3624 original RCPT would be taken as a response to RSET and so on, causing
3625 mayhem of various kinds.
3626
362747. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
3628 when I implemented it. It didn't allow for the fact that some option values
3629 may legitimatetly be negative (e.g. size_addition), and it didn't even do
3630 the right test for positive values.
3631
363248. Domain names in DNS records are case-independent. Exim always looks them up
3633 in lower case. Some resolvers return domain names in exactly the case they
3634 appear in the zone file, that is, they may contain uppercase letters. Not
3635 all resolvers do this - some return always lower case. Exim was treating a
3636 change of case by a resolver as a change of domain, similar to a widening
3637 of a domain abbreviation. This triggered its re-routing code and so it was
3638 trying to route what was effectively the same domain again. This normally
3639 caused routing to fail (because the router wouldn't handle the domain
3640 twice). Now Exim checks for this case specially, and just changes the
3641 casing of the domain that it ultimately uses when it transmits the message
3642 envelope.
3643
364449. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
3645 module.
3646
364750. If a filter generated a file delivery with a non-absolute name (possible if
3648 no home directory exists for the router), the forbid_file option was not
3649 forbidding it.
3650
365151. Added '&' feature to dnslists, to provide bit mask matching in addition to
3652 the existing equality matching.
3653
365452. Exim was using ints instead of ino_t variables in some places where it was
3655 dealing with inode numbers.
3656
365753. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
3658 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
3659 called TMPDIR, and if it finds it is different, it changes its value.
3660
366154. The smtp_printf() function is now made available to local_scan() so
3662 additional output lines can be written before returning. There is also an
3663 smtp_fflush() function to enable the detection of a dropped connection.
3664 The variables smtp_input and smtp_batched_input are exported to
3665 local_scan().
3666
366755. Changed the default runtime configuration: the message "Unknown user"
3668 has been removed from the ACL, and instead placed on the localuser router,
3669 using the cannot_route_message feature. This means that any verification
3670 failures that generate their own messages won't get overridden. Similarly,
3671 the "Unrouteable address" message that was in the ACL for unverifiable
3672 relay addresses has also been removed.