Add the add_header modifier for use with all ACL verbs.
[exim.git] / doc / doc-txt / ChangeLog
CommitLineData
71fafd95 1$Cambridge: exim/doc/doc-txt/ChangeLog,v 1.322 2006/03/06 16:05:12 ph10 Exp $
495ae4b0
PH
2
3Change log file for Exim from version 4.21
4-------------------------------------------
5
214e2000
PH
6Exim version 4.61
7-----------------
8
9PH/01 The code for finding all the local interface addresses on a FreeBSD
10 system running IPv6 was broken. This may well have applied to all BSD
11 systems, as well as to others that have similar system calls. The broken
12 code found IPv4 interfaces correctly, but gave incorrect values for the
13 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
14 that it would not match correctly against @[] and not recognize the IPv6
15 addresses as local.
16
f9daeae0
PH
17PH/02 The ipliteral router was not recognizing addresses of the form user@
18 [ipv6:....] because it didn't know about the "ipv6:" prefix.
19
7e66e54d
PH
20PH/03 Added disable_ipv6.
21
c8ea1597
PH
22PH/04 Changed $reply_address to use the raw form of the headers instead of the
23 decoded form, because it is most often used to construct To: headers
24 lines in autoreplies, and the decoded form may well be syntactically
25 invalid. However, $reply_address has leading white space removed, and all
26 newlines turned into spaces so that the autoreply transport does not
27 grumble.
28
911f6fde
PH
29PH/05 If group was specified without a user on a router, and no group or user
30 was specified on a transport, the group from the router was ignored.
31
47ca6d6c
PH
32PH/06 Increased the number of ACL variables to 20 of each type, and arranged
33 for visible compile-time settings that can be used to change these
34 numbers, for those that want even more. Backwards compatibility with old
35 spool files has been maintained. However, going back to a previous Exim
36 release will lost any variables that are in spool files.
37
ed0e9820
PH
38PH/07 Two small changes when running in the test harness: increase delay when
39 passing a TCP/IP connection to a new process, in case the original
40 process has to generate a bounce, and remove special handling of
41 127.0.0.2 (sic), which is no longer necessary.
42
eff37e47
PH
43PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
44 be the same on different OS.
45
1921d2ea
PH
46PH/09 Moved a debug statement in filter processing to avoid a race problem when
47 testing.
48
b3f69ca8
JJ
49JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
50 whether --show-vars was specified or not
51
52JJ/02 exipick: Added support for new ACL variable spool format introduced
53 in 4.61-PH/06
54
424a1c63
PH
55PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
56 syntactically invalid From: or Reply-to: line, and a filter used this to
57 generate an autoreply, and therefore failed to obtain an address for the
58 autoreply, Exim could try to deliver to a non-existent relative file
59 name, causing unrelated and misleading errors. What now happens is that
60 it logs this as a hard delivery error, but does not attempt to create a
61 bounce message.
62
7a100415
PH
63PH/11 The exinext utility has a -C option for testing purposes, but although
64 the given file was scanned by exinext itself; it wasn't being passed on
65 when Exim was called.
66
19b9dc85
PH
67PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
68 an end-of-file indication when reading a command response.
69
309bd837
PH
70PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
71 compiled. In many other places in Exim, IPv6 addresses are always
72 recognized, so I have changed this. It also means that IPv4 domain
73 literals of the form [IPV4:n.n.n.n] are now always recognized.
74
59e82a2a
PH
75PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
76 used if the router is not running as root, for example, when verifying at
77 ACL time, or when using -bh. The debugging output from this situation was
78 non-existent - all you got was a failure to exec. I have made two
79 changes:
80
81 (a) Failures to set uid/gid, the current directory, or a process leader
82 in a subprocess such as that created by queryprogram now generate
83 suitable debugging ouput when -d is set.
84
85 (b) The queryprogram router detects when it is not running as root,
86 outputs suitable debugging information if -d is set, and then runs
87 the subprocess without attempting to change uid/gid.
88
9edc04ce
PH
89PH/15 Minor change to Makefile for building test_host (undocumented testing
90 feature).
91
1349e1e5
PH
92PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
93 additional section of a DNS packet that returns MX or SRV records.
94 Instead, it always explicitly searches for A/AAAA records. This avoids
95 major problems that occur when a DNS server includes only records of one
96 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
97 fixed another bug: if SRV records were looked up and the corresponding
98 address records were *not* found in the additional section, the port
99 values from the SRV records were lost.
100
ea49d0e1
PH
101PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
102 using the correct key (the original address) when searching the retry
103 rules in order to find which one to use for generating the retry hint.
104
064a94c9
PH
105PH/18 If quota_warn_message contains a From: header, Exim now refrains from
106 adding the default one. Similarly, if it contains a Reply-To: header, the
107 errors_reply_to option, if set, is not used.
108
727071f8
PH
109PH/19 When calculating a retry time, Exim used to measure the "time since
110 failure" by looking at the "first failed" field in the retry record. Now
111 it does not use this if it is later than than the arrival time of the
112 message. Instead it uses the arrival time. This makes for better
113 behaviour in cases where some deliveries succeed, thus re-setting the
114 "first failed" field. An example is a quota failure for a huge message
115 when small messages continue to be delivered. Without this change, the
116 "time since failure" will always be short, possible causing more frequent
117 delivery attempts for the huge message than are intended.
118
f78eb7c6
PH
119PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
120 $1, $2, $3) because the numerical variables can be reset during some
121 expansion items (e.g. "match"), thereby losing the authentication data.
122
21c28500
PH
123PH/21 Make -bV show the size of off_t variables so that the test suite can
124 decide whether to run tests for quotas > 2G.
125
126PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
127 mailbox_size, and mailbox_filecount in the appendfile transport. If a
128 filecount value is greater than 2G or if a quota value is greater than 2G
129 on a system where the size of off_t is not greater than 4, a panic error
130 is given.
131
1688f43b
PH
132PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
133 never match. The debug and -bh output now contains an explicit error
134 message indicating a malformed IPv4 address or mask.
135
136PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
137 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
138 PH/23 above applies.
139
9675b384
PH
140PH/25 Do not write to syslog when running in the test harness. The only
141 occasion when this arises is a failure to open the main or panic logs
142 (for which there is an explicit test).
143
6a3f1455
PH
144PH/26 Added the /no_tell option to "control=freeze".
145
dac79d3e
PH
146PH/27 If a host name lookup failed very early in a connection, for example, if
147 the IP address matched host_lookup and the reverse lookup yielded a name
148 that did not have a forward lookup, an error message of the form "no IP
149 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
150 could be logged. Now it outputs the IP address instead of "NULL".
1349e1e5 151
5977a0b3
PH
152PH/28 An enabling patch from MH: add new function child_open_exim2() which
153 allows the sender and the authenticated sender to be set when
154 submitting a message from within Exim. Since child_open_exim() is
155 documented for local_scan(), the new function should be too.
156
c91535f3
PH
157PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
158 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
159 results in an empty string is now treated as unset.
160
0d46a8c8
PH
161PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
162
278c6e6c
PH
163PH/31 Added sender_verify_fail logging option.
164
2cbb4081
PH
165PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
166 needed by RFC 822 but not by RFC 2822 was commented out. I have now
167 tidied the source and removed it altogether.
168
3eef829e
PH
169PH/33 When a queue run was abandoned because the load average was too high, a
170 log line was always written; now it is written only if the queue_run log
171 selector is set. In addition, the log line for abandonment now contains
172 information about the queue run such as the pid. This is always present
173 in "start" and "stop" lines but was omitted from the "abandon" line.
174
1ab95fa6
PH
175PH/34 Omit spaces between a header name and the colon in the error message that
176 is given when verify = headers_syntax fails (if there are lots of them,
177 the message gets confusing).
178
230205fc
PH
179PH/35 Change the default for dns_check_names_pattern to allow slashes within
180 names, as there are now some PTR records that contain slashes. This check
181 is only to protect against broken name servers that fall over on strange
182 characters, so the fact that it applies to all lookups doesn't matter.
183
75e0e026
PH
184PH/36 Now that the new test suite is complete, we can remove some of the
185 special code in Exim that was needed for the old test suite. For example,
186 sorting DNS records because real resolvers return them in an arbitrary
187 order. The new test suite's fake resolver always returns records in the
188 same order.
189
190PH/37 When running in the test harness, use -odi for submitted messages (e.g.
191 bounces) except when queue_only is set, to avoid logging races between
192 the different processes.
193
145396a6
PH
194PH/38 Panic-die if .include specifies a non-absolute path.
195
3cd34f13
PH
196PH/39 A tweak to the "H" retry rule from its user.
197
11121d3d
JJ
198JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
199 a label. They prevented compilation on older perls.
200
201JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
202 a warning to be raised on newish perls.
203
204JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
205 on queue. Changes to match documented behaviour of showing count of
206 messages matching specified criteria.
207
8def5aaf
PH
208PH/40 Changed the default ident timeout from 30s to 5s.
209
929ba01c
PH
210PH/41 Added support for the use of login_cap features, on those BSD systems
211 that have them, for controlling the resources used by pipe deliveries.
212
2632889e
PH
213PH/42 The content-scanning code uses fopen() to create files in which to put
214 message data. Previously it was not paying any attention to the mode of
215 the files. Exim runs with umask(0) because the rest of the code creates
216 files with open(), and sets the required mode explicitly. Thus, these
217 files were ending up world-writeable. This was not a big issue, because,
218 being within the spool directory, they were not world-accessible. I have
219 created a function called modefopen, which takes an additional mode
220 argument. It sets umask(777), creates the file, chmods it to the required
221 mode, then resets the umask. All the relevant calls to fopen() in the
222 content scanning code have been changed to use this function.
223
944a9c55
PH
224PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
225 to 24 hours. This avoids potential overflow problems when processing G
226 and H retry rules. I suspect nobody ever tinkers with this value.
227
4a23603b
PH
228PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
229
4730f942
PH
230PH/45 When the plaintext authenticator is running as a client, the server's
231 challenges are checked to ensure they are valid base64 strings. By
232 default, the authentication attempt is cancelled if an invalid string is
233 received. Setting client_ignore_invalid_base64 true ignores these errors.
234 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
235 they are received. Thus, the responses can be made to depend on the
236 challenges. If an invalid string is ignored, an empty string is placed in
237 the variable.
238
30dba1e6
PH
239PH/46 Messages that are created by the autoreply transport now contains a
240 References: header, in accordance with RFCs 2822 and 3834.
241
382afc6b
PH
242PH/47 Added authenticated_sender_force to the smtp transport.
243
a86229cf
PH
244PH/48 The ${prvs expansion was broken on systems where time_t was long long.
245
50c99ba6
PH
246PH/49 Installed latest patch from the Sieve maintainer.
247
d35e429d
PH
248PH/50 When an Exim quota was set without a file count quota, and mailbox_size
249 was also set, the appendfile transport was unnecessarily scanning a
250 directory of message files (e.g. for maildir delivery) to find the count
251 of files (along with the size), even though it did not need this
252 information. It now does the scan only if it needs to find either the
253 size of the count of files.
254
f90d018c
PH
255PH/51 Added ${time_eval: to convert Exim time strings into seconds.
256
75def545
PH
257PH/52 Two bugs concerned with error handling when the smtp transport is
258 used in LMTP mode:
259
260 (i) Exim was not creating retry information for temporary errors given
261 for individual recipients after the DATA command when the smtp transport
262 was used in LMTP mode. This meant that they could be retried too
263 frequently, and not timed out correctly.
264
265 (ii) Exim was setting the flag that allows error details to be returned
266 for LMTP errors on RCPT commands, but not for LMTP errors for individual
267 recipients that were returned after the DATA command.
268
269PH/53 This is related to PH/52, but is more general: for any failing address,
270 when detailed error information was permitted to be returned to the
271 sender, but the error was temporary, then after the final timeout, only
272 "retry timeout exceeded" was returned. Now it returns the full error as
273 well as "retry timeout exceeded".
274
c46782ef
PH
275PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
276 do this, and (what is worse) MTAs that accept it.
277
71fafd95
PH
278PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
279 will now be deprecated.
280
214e2000 281
5de37277
PH
282Exim version 4.60
283-----------------
284
cc38ddbf
PH
285PH/01 Two changes to the default runtime configuration:
286
287 (1) Move the checks for relay_from_hosts and authenticated clients from
288 after to before the (commented out) DNS black list checks.
289
290 (2) Add control=submission to the relay_from_hosts and authenticated
291 clients checks, on the grounds that messages accepted by these
292 statements are most likely to be submissions.
5de37277 293
72fdd6ae
PH
294PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
295
296 (1) Generate an error if the third argument for the ${prvs expansion is
297 not a single digit.
298
299 (2) Treat a missing third argument of ${prvscheck as if it were an empty
300 string.
301
302 (3) Reset the variables that are obtained from the first argument of
303 ${prvscheck and used in the second argument before leaving the code,
304 because their memory is reclaimed, so using them afterwards may do
305 silly things.
306
307 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
308 one (it's much easier than Tom thought :-).
309
310 (5) Because of (4), we can now allow for the use of $prvscheck_result
311 inside the third argument.
cb9328de 312
cb741023
PH
313PH/03 For some reason, the default setting of PATH when running a command from
314 a pipe transport was just "/usr/bin". I have changed it to
315 "/bin:/usr/bin".
316
f174f16e
PH
317PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
318 anything to be listed in the output from -bV.
b2f5a032 319
c25242d7
PH
320PH/05 When a filter generated an autoreply, the entire To: header line was
321 quoted in the delivery log line, like this:
322
323 => >A.N.Other <ano@some.domain> <original@ddress> ...
324
325 This has been changed so that it extracts the operative address. There
326 may be more than one such address. If so, they are comma-separated, like
327 this:
328
329 => >ano@some.domain,ona@other.domain <original@ddress> ...
330
82c19f95
PH
331PH/06 When a client host used a correct literal IP address in a HELO or EHLO
332 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
333 not being looked up in the rDNS to get a host name, Exim was showing the
334 IP address twice in Received: lines, even though the IP addresses were
335 identical. For example:
336
337 Received: from [1.2.3.4] (helo=[1.2.3.4])
338
339 However, if the real host name was known, it was omitting the HELO data
340 if it matched the actual IP address. This has been tidied up so that it
341 doesn't show the same IP address twice.
342
d7ffbc12
PH
343PH/07 When both +timestamp and +memory debugging was on, the value given by
344 $tod_xxx expansions could be wrong, because the tod_stamp() function was
345 called by the debug printing, thereby overwriting the timestamp buffer.
346 Debugging no longer uses the tod_stamp() function when +timestamp is set.
347
9f526266
PH
348PH/08 When the original message was included in an autoreply transport, it
349 always said "this is a copy of the message, including all the headers",
350 even if body_only or headers_only was set. It now gives an appropriate
351 message.
352
87fcc8b9
PH
353PH/09 Applied a patch from the Sieve maintainer which:
354
355 o fixes some comments
356 o adds the (disabled) notify extension core
357 o adds some debug output for the result of if/elsif tests
358 o points to the current vacation draft in the documentation
359 and documents the missing references header update
360
361 and most important:
362
363 o fixes a bug in processing the envelope test (when testing
364 multiple envelope elements, the last element determinted the
365 result)
366
456682f5
PH
367PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
368 Electronic Mail") by including:
369
370 Auto-submitted: auto-generated
371
372 in the messages that it generates (bounce messages and others, such as
373 warnings). In the case of bounce messages for non-SMTP mesages, there was
374 also a typo: it was using "Auto_submitted" (underscore instead of
375 hyphen). Since every message generated by Exim is necessarily in response
376 to another message, thes have all been changed to:
377
378 Auto-Submitted: auto-replied
379
380 in accordance with these statements in the RFC:
381
382 The auto-replied keyword:
383
384 - SHOULD be used on messages sent in direct response to another
385 message by an automatic process,
386
387 - MUST NOT be used on manually-generated messages,
388
389 - MAY be used on Delivery Status Notifications (DSNs) and Message
390 Disposition Notifications (MDNs),
391
392 - MUST NOT be used on messages generated by automatic or periodic
393 processes, except for messages which are automatic responses to
394 other messages.
395
3e46c1aa
PH
396PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
397 to the default Received: header definition.
456682f5 398
49826d12
PH
399PH/12 Added log selector acl_warn_skipped (default on).
400
eba0c039
PH
401PH/13 After a successful wildlsearch lookup, discard the values of numeric
402 variables because (a) they are in the wrong storage pool and (b) even if
403 they were copied, it wouldn't work properly because of the caching.
404
a0d6ba8a
PH
405PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
406 checking when decoding. Apparently there are clients that generate
407 overlong encoded strings. Why am I not surprised?
408
f0917727
PH
409PH/15 If the first argument of "${if match_address" was not empty, but did not
410 contain an "@" character, Exim crashed. Now it writes a panic log message
411 and treats the condition as false.
412
096fee00
PH
413PH/16 In autoreply, treat an empty string for "once" the same as unset.
414
024bd3c2
PH
415PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
416 extension "envelope-auth". The code is finished and in agreement with
417 other implementations, but there is no documentation so far and in fact,
418 nobody wrote the draft yet. This extension is currently #undef'ed, thus
419 not changing the active code.
420
421 Print executed "if" and "elsif" statements when debugging is used. This
422 helps a great deal to understand what a filter does.
423
424 Document more things not specified clearly in RFC3028. I had all this
425 sorted out, when out of a sudden new issues came to my mind. Oops."
426
df199fec
PH
427PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
428 (Bugzilla #53).
429
d27f1df3
PH
430PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
431 canonical form (as documented). However, after a host name lookup from
432 the IP address, check_host() was doing a simple string comparison with
433 addresses acquired from the DNS when checking that the found name did
434 have the original IP as one of its addresses. Since any found IPv6
435 addresses are likely to be in abbreviated form, the comparison could
436 fail. Luckily, there already exists a function for doing the comparison
437 by converting both addresses to binary, so now that is used instead of
438 the text comparison.
439
96776534
PH
440PH/20 There was another similar case to PH/19, when a complete host name was
441 given in a host list; looking up its IP address could give an abbreviated
442 form, whereas the current host's name might or might not be abbreviated.
443 The same fix has been applied.
444
5de37277 445
9a799bc0
PH
446Exim version 4.54
447-----------------
448
449PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
450 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
451 It now does.
452
99a4b039
PH
453PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
454 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
455
456PH/03 Typo: missing ".o" in src/pcre/Makefile.
457
4b233853
PH
458PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
459 header line, restrict the check to what is listed in RFCs 2369 and 2929.
460 Also, for "Auto-Submitted", treat anything other than "no" as
461 non-personal, in accordance with RFC 3834. (Previously it treated
462 anything starting "auto-" as non-personal.)
463
8857ccfd
PH
464TF/01 The control=submission/name=... option had a problem with syntax
465 errors if the name included a slash character. The /name= option
466 now slurps the rest of the string, so it can include any characters
467 but it must come last in the list of options (after /sender_retain
468 or /domain=).
469
433a2980
PH
470PH/05 Some modifications to the interface to the fake nameserver for the new
471 testing suite.
472
3e46c1aa 473
9a799bc0 474
e3a311ba
TK
475Exim version 4.53
476-----------------
477
478TK/01 Added the "success_on_redirect" address verification option. See
479 NewStuff for rationale and an example.
480
13b685f9
PH
481PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
482
395ff96d
PH
483PH/02 Patch to exigrep to allow it to work on syslog lines.
484
5b68f6e4
PH
485PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
486 fread() to skip over the body file's header line, because in Cygwin the
487 header line is locked and is inaccessible.
488
1ab52c69
PH
489PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
490 co-exist for some time) to make it clear that it is the Exim ID that is
491 referenced, not the Message-ID: header line.
492
b07e6aa3
PH
493PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
494 string_format() function, because snprintf() does not exist on all
495 operating systems.
496
254e032f
PH
497PH/06 The use of forbid_filter_existstest now also locks out the use of the
498 ${stat: expansion item.
499
3af76a81
PH
500PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
501 protocol synchronization error", to keep the pedants happy.
502
2548ba04
PH
503PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
504 well as for IRIX systems, when gcc is being used. See the host.c source
505 file for comments.
506
b6c6011d
PH
507PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
508
cf39cf57
PH
509PH/10 Named domain lists were not working if used in a queue_smtp_domains
510 setting.
511
f1513293
PH
512PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
513 transport and to the smtp transport in LMTP mode.
514
727549a4
PH
515TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
516
af46795e
PH
517PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
518 run a filter in a subprocess. This could lead to confusion in subsequent
519 lookups in the parent process. There should also be a search_tidyup() at
520 the end of the subprocess.
521
d7b47fd0
PH
522PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
523 only if the host matched helo_try_verify_hosts, which caused the
524 verification to occur when the EHLO/HELO command was issued. The ACL just
525 tested the remembered result. Now, if a previous verification attempt has
526 not happened, "verify = helo" does it there and then.
527
ee744174
JJ
528JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
529
b582ab87
PH
530TK/03 Fix log output including CR from clamd.
531
41a13e0a
PH
532PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
533 exist provoked a memory error which could cause a segfault.
534
f625cc5a
PH
535PH/15 Installed PCRE 6.2
536
537PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
538
21f7af35
PH
539PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
540 of the problem. Specifically, suggested +O2 rather than +O1 for the
541 HP-UX compiler.
542
31480e42
PH
543PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
544
2d280592
PH
545PH/20 If a delivery was routed to a non-standard port by means of an SRV
546 record, the port was not correctly logged when the outgoing_port log
547 selector was set (it logged the transort's default port).
548
7cd1141b
PH
549PH/21 Added support for host-specific ports to manualroute, queryprogram,
550 fallback_hosts, and "hosts" in the smtp transport.
551
552PH/22 If the log selector "outgoing_port" is set, the port is now also given on
553 host errors such as "Connection refused".
554
750af86e
PH
555PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
556 authentication with radiusclient 0.4.9:
557
558 - Error returned from rc_read_config was caught wrongly
559 - Username/password not passed on to radius server due to wrong length.
560
561 The presumption is that some radiusclient API changes for 4.51/PH/17
562 were not taken care of correctly. The code is still untested by me (my
563 Linux distribution still has 0.3.2 of radiusclient), but it was
564 contributed by a Radius user.
565
566PH/24 When doing a callout, the value of $domain wasn't set correctly when
567 expanding the "port" option of the smtp transport.
568
4304270b
TK
569TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
570 while reading a MIME header. Thanks to Tom Hughes for a patch.
571
750af86e
PH
572PH/24 Include config.h inside local_scan.h so that configuration settings are
573 available.
574
64ffc24f
PH
575PH/25 Make $smtp_command_argument available after all SMTP commands. This means
576 that in an ACL for RCPT (for example), you can examine exactly what was
577 received.
578
5dd9625b
PH
579PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
580 commands, but it was not correctly comparing the address with the actual
581 client host address. Thus, it would show the EHLO address in Received:
582 header lines when this was not necessary.
583
5591031b
PH
584PH/27 Added the % operator to ${eval:}.
585
ba18e66a
PH
586PH/28 Exim tries to create and chdir to its spool directory when it starts;
587 it should be ignoring failures (because with -C, for example, it has lost
588 privilege). It wasn't ignoring creation failures other than "already
589 exists".
590
9cec981f
PH
591PH/29 Added "crypteq" to the list of supported features that Exim outputs when
592 -bV or -d is used.
593
aa2b5c79
PH
594PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
595 because an input line was too long, either on its own, or by virtue of
1509d3a8
PH
596 too many continuations, the temporary file was not being removed, and the
597 return code was incorrect.
aa2b5c79 598
48a53b7f
PH
599PH/31 Missing "BOOL" in function definition in filtertest.c.
600
1c59d63b
PH
601PH/32 Applied Sieve patches from the maintainer.
602
671012da
TK
603TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
604
1509d3a8
PH
605PH/33 Added "verify = not_blind".
606
607PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
608 Local/Makefile (with some defaults set). These are used in built scripts
609 such as exicyclog, but they have never been used in the exim_install
610 script (though there are many overriding facilities there). I have
611 arranged that the exim_install script now takes note of these two
612 settings.
613
614PH/35 Installed configuration files for Dragonfly.
615
2fe1a124
PH
616PH/36 When a locally submitted message by a trusted user did not contain a
617 From: header, and the sender address was obtained from -f or from an SMTP
618 MAIL command, and the trusted user did not use -F to supply a sender
619 name, $originator_name was incorrectly used when constructing a From:
620 header. Furthermore, $originator_name was used for submission mode
621 messages from external hosts without From: headers in a similar way,
622 which is clearly wrong.
623
8800895a
PH
624PH/37 Added control=suppress_local_fixups.
625
ccfdb010
PH
626PH/38 When log_selector = +received_sender was set, and the addition of the
627 sender made the log line's construction buffer exactly full, or one byte
628 less than full, an overflow happened when the terminating "\n" was
629 subsequently added.
630
1130bfb0
PH
631PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
632 when the result of a list match is failure because a DNS lookup failed.
633
ebcb507f
PH
634PH/40 RM_COMMAND is now used in the building process.
635
c35e155c
PH
636PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
637 the "build-* directories that it finds.
638
95d1f782
PH
639PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
640 address in a domain literal was a prefix of an interface address.
641
fd6de02e
PH
642PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
643 when verifying a sender address, unless rewrite_headers is false.
644
58de37c5
PH
645PH/44 Wrote a long comment about why errors_to addresses are verified as
646 recipients, not senders.
647
261cf466
TF
648TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
649 the ratelimit ACL was added.
650
3ee512ff
PH
651PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
652
e08c430f
PH
653PH/46 Added extra information about PostgreSQL errors to the error string.
654
bef5a11f
PH
655PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
656 avoiding the need to install special zones in a real server. This is
657 backwards compatible; if it can't find the fake resolver, it drops back.
658 Thus, both old and new test suites can be run.
659
7546de58
TF
660TF/02 Added util/ratelimit.pl
661
e5d5a95f
TF
662TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
663 clock is set back in time.
664
2e88a017
TF
665TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
666 Candler <B.Candler@pobox.com>.
667
a5f65aa4
TF
668TF/05 The fix for PH/43 was not completely correct; widen_domains is always
669 OK for addresses that are the result of redirections.
670
e7726cbf
PH
671PH/48 A number of further additions for the benefit of the new test suite,
672 including a fake gethostbyname() that interfaces to the fake DNS resolver
673 (see PH/47 above).
674
a7fdad5b
TF
675TF/06 The fix for widen_domains has also been applied to qualify_single and
676 search_parents which are the other dnslookup options that can cause
677 header rewrites.
678
6af56900
PH
679PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
680 type ("H").
681
0925ede6
PH
682PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
683
66afa403
TF
684TF/07 Exim produced the error message "an SRV record indicated no SMTP
685 service" if it encountered an MX record with an empty target hostname.
686 The message is now "an MX or SRV record indicated no SMTP service".
687
0154e85a
TF
688TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
689 if the DNS of the sending site is misconfigured. This is quite a
690 common situation. This change restores the behaviour of treating a
691 helo verification defer as a failure.
692
16f12c76
PH
693PH/51 If self=fail was set on a router, the bounce message did not include the
694 actual error message.
695
bbe902f0 696
e5a9dba6
PH
697Exim version 4.52
698-----------------
699
700TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
701
22c3b60b
PH
702PH/01 When a transport filter timed out in a pipe delivery, and the pipe
703 command itself ended in error, the underlying message about the transport
704 filter timeout was being overwritten with the pipe command error. Now the
705 underlying error message should be appended to the second error message.
706
06a9b4b5
PH
707TK/01 Fix poll() being unavailable on Mac OSX 10.2.
708
c1ac6996
PH
709PH/02 Reduce the amount of output that "make" produces by default. Full output
710 can still be requested.
711
9c7a242c
PH
712PH/03 The warning log line about a condition test deferring for a "warn" verb
713 was being output only once per connection, rather than after each
714 occurrence (because it was using the same function as for successful
715 "warn" verbs). This seems wrong, so I have changed it.
716
87ba3f5f
PH
717TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
718 it should not have, which might have caused a crash in the right
719 circumstances, but probably never did.
720
721PH/04 Installed a modified version of Tony Finch's patch to make submission
722 mode fix the return path as well as the Sender: header line, and to
723 add a /name= option so that you can make the user's friendly name appear
724 in the header line.
725
29aba418
TF
726TF/03 Added the control = fakedefer ACL modifier.
727
fe0dab11
TF
728TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
729 Mark Lowes for thorough testing.
870f6ba8 730
11d337a4
TK
731TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
732
733TK/03 Merged latest SRS patch from Miles Wilton.
734
415c8f3b
PH
735PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
736 with the definition in sysexits.h (which is #included earlier).
737 Fortunately, Exim does not actually use EX_OK. The code used to try to
738 preserve the sysexits.h value, by assumimg that macro definitions were
739 scanned for macro replacements. I have been disabused of this notion,
740 so now the code just undefines EX_OK before #including unistd.h.
11d337a4 741
958541e9
PH
742PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
743 in the smtp transport. When a block could not be written in a single
744 write() function, the timeout was being re-applied to each part-write.
745 This seems wrong - if the receiver was accepting one byte at a time it
746 would take for ever. The timeout is now adjusted when this happens. It
747 doesn't have to be particularly precise.
748
c206415f
TK
749TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
750 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
751
2a4be8f9
PH
752PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
753 without a domain if the check to <postmaster@domain> fails.
754
1cba11c5
SC
755SC/01 Eximstats: added -xls and the ability to specify output files
756 (patch written by Frank Heydlauf).
757
758SC/02 Eximstats: use FileHandles for outputing results.
759
760SC/03 Eximstats: allow any combination of xls, txt, and html output.
761
762SC/04 Eximstats: fixed display of large numbers with -nvr option
763
764SC/05 Eximstats: fixed merging of reports with empty tables.
765
766SC/06 Eximstats: added the -include_original_destination flag
767
768SC/07 Eximstats: removed tabs and trailing whitespace.
769
1005d00e
TK
770TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
771
772TK/06 MBOX spool code: Add real "From " MBOX separator line
773 so the .eml file is really in mbox format (even though
774 most programs do not really care). Patch from Alex Miller.
775
776TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
777 The latter is generated from $received_to and is only set if the
778 message has one envelope recipient. SA can use these headers,
779 obviously out-of-the-box. Patch from Alex Miller.
780
9b4768fa
PH
781PH/08 The ${def test on a variable was returning false if the variable's
782 value was "0", contrary to what the specification has always said!
783 The result should be true unless the variable is empty.
784
785PH/09 The syntax error of a character other than { following "${if
786 def:variable_name" (after optional whitespace) was not being diagnosed.
787 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
788 accidental colon was present, for example, could give incorrect results.
789
0d7eb84a
PH
790PH/10 Tidied the code in a number of places where the st_size field of a stat()
791 result is used (not including appendfile, where other changes are about
792 to be made).
793
794PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
795 This involved changing a lot of size variables from int to off_t. It
796 should work with maildirs and everything.
797
40727bee
TK
798TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
799 spamd dying while we are connected to it.
800
554d2369
TF
801TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
802 <jgh@wizmail.org>
803
1f922db1
PH
804PH/12 Applied Alex Kiernan's patch for the API change for the error callback
805 function for BDB 4.3.
806
ef213c3b
PH
807PH/13 Changed auto_thaw such that it does not apply to bounce messages.
808
8ac170f3
PH
809PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
810 the sources for PCRE have been re-arranged and more files are now
811 involved.
812
b1c749bb
PH
813PH/15 The code I had for printing potentially long long variables in PH/11
814 above was not the best (it lost precision). The length of off_t variables
815 is now inspected at build time, and an appropriate printing format (%ld
c6c2dc1d
PH
816 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
817 to be "long long int" or "long int". This is needed for the internal
818 formatting function string_vformat().
b1c749bb 819
4aac9b49
PH
820PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
821 the configuration file to be ":syslog", then the script "guesses" where
822 the logs files are, rather than using the compiled in default. In our
823 case the guess is not the same as the compiled default, so the script
824 suddenly stopped working when I started to use syslog. The patch checks
825 to see if log_file_path is "". If so, it attempts to read it from exim
826 with no configuration file to get the compiled in version, before it
827 falls back to the previous guessing code."
828
294520c8
TK
829TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
830 implementing BATV in an Exim configuration. See NewStuff for the gory
831 details.
832
5bd022fe
PH
833PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
834 Makefile that are specific to HP-UX.
835
90e9ce59
PH
836PH/18 If the "use_postmaster" option was set for a recipient callout together
837 with the "random" option, the postmaster address was used as the MAIL
838 FROM address for the random test, but not for the subsequent recipient
839 test. It is now used for both.
840
5ea81592
PH
841PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
842 patch removes a few documentation additions to RFC 3028, because the
843 latest draft now contains them. It adds the new en;ascii-case comparator
844 and a new error check for 8bit text in MIME parts. Comparator and
845 require names are now matched exactly. I enabled the subaddress
846 extension, but it is not well tested yet (read: it works for me)."
847
c6c2dc1d
PH
848PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
849 rework some of the code of TK/09 above to avoid the hardwired use of
850 "%lld" and "long long". Replaced the call to snprintf() with a call to
851 string_vformat().
852
fffffe4c
PH
853PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
854 records point to non-existent hosts", "retry timeout exceeded", and
855 "retry time not reached for any host after a long failure period".
ca02eafb 856
9a26b6b2
PH
857PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
858 experimental DomainKeys support:
859
860 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
861 (2) On an error such as an illegally used "control", the wrong name for
862 the control was given.
863
864 These problems did NOT occur unless DomainKeys support was compiled.
865
4aee0225
PH
866PH/23 Added daemon_startup_retries and daemon_startup_sleep.
867
32d668a5
PH
868PH/24 Added ${if match_ip condition.
869
8187c3f3
PH
870PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
871 databases so that it will be absolutely obvious if a crash occurs in the
872 DB library. This is a regular occurrence (often caused by mis-matched
873 db.h files).
874
ff790e47 875PH/26 Insert a lot of missing (void) casts for functions such as chown(),
f1e894f3
PH
876 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
877 picked up on a user's system that detects such things. There doesn't seem
878 to be a gcc warning option for this - only an attribute that has to be
879 put on the function's prototype. It seems that in Fedora Core 4 they have
880 set this on a number of new functions. No doubt there will be more in due
881 course.
ff790e47 882
5417f6d1
PH
883PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
884 specify a transport. However, if an address that was verified by such a
885 router was the subject of a callout, Exim crashed because it tried to
886 read the rcpt_include_affixes from the non-existent transport. Now it
887 just assumes that the setting of that option is false. This bug was
888 introduced by 4.51/PH/31.
889
59cf8544
PH
890PH/28 Changed -d+all to exclude +memory, because that information is very
891 rarely of interest, but it makes the output a lot bigger. People tend to
892 do -d+all out of habit.
893
e7ad8a65
PH
894PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
895 code in os-type was giving problems when libc.so lives in lib64, like on
896 x86_64 Fedora Core.
897
ade42478
PH
898PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
899 aren't the modern standard, and it seems that some systems' include files
900 don't always have them. Exim was already checking for some of the newer
901 ones like T_AAAA, and defining it itself. I've added checks for all the
902 record types that Exim uses.
903
182ad5cf
PH
904PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
905 not automatically generating a new one, as it is supposed to. This
906 prevented TLS from working. If the file did exist, but contained invalid
907 data, a new version was generated, as expected. It was only the case of a
908 non-existent file that was broken.
909
b0d9fc80
TK
910TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
911 with a change in libdomainkeys > 0.64.
912
913TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
914 from DNS. If the selector record carries the flag, it now has
915 precedence over the domain-wide flag.
916
917TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
918
47c7a64a
PH
919PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
920 the use of a port name that isn't defined in /etc/services) occurred, the
921 message was deferred as in a normal delivery, and thus remained on the
922 spool, instead of being failed because of the mua_wrapper setting. This
923 is now fixed, and I tidied up some of the mua_wrapper messages at the
924 same time.
925
a388bce4
SC
926SC/08 Eximstats: whilst parsing the mainlog(s), store information about
927 the messages in a hash of arrays rather than using individual hashes.
928 This is a bit cleaner and results in dramatic memory savings, albeit
929 at a slight CPU cost.
930
931SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
932 as requested by Marc Sherman.
933
934SC/10 Eximstats: added histograms for user specified patterns as requested
935 by Marc Sherman.
936
0793e4ed
SC
937SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
938
c58b88df
PH
939PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
940 fopen() in the content-scanning modules that did not already have it.
941
e7ad8a65 942
7982096b
PH
943Exim version 4.51
944-----------------
945
1a46a8c5
PH
946TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
947 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
948
2f079f46 949TK/02 Fix ACL "control" statement not being available in MIME ACL.
1a46a8c5
PH
950
951TK/03 Fix ACL "regex" condition not being available in MIME ACL.
952
953PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
954 to test Sieve filters that use "vacation".
955
956PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
957 that changes the way the GnuTLS parameters are stored in the cache file.
958 The new format can be generated externally. For backward compatibility,
959 if the data in the cache doesn't make sense, Exim assumes it has read an
960 old-format file, and it generates new data and writes a new file. This
961 means that you can't go back to an older release without removing the
962 file.
963
964PH/03 A redirect router that has both "unseen" and "one_time" set does not
965 work if there are any delivery delays because "one_time" forces the
966 parent to be marked "delivered", so its unseen clone is never tried
967 again. For this reason, Exim now forbids the simultaneous setting of
968 these two options.
969
970PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
971 redirected to themselves ("homonym" addresses). Read the long ChangeLog
972 entry if you want to know the details. The fix, however, neglected to
973 consider the case when local delivery batching is involved. The test for
974 "previously delivered" was not happening when checking to see if an
975 address could be batched with a previous (undelivered) one; under
976 certain circumstances this could lead to multiple deliveries to the same
c2c19e9d 977 address.
1a46a8c5
PH
978
979PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
980 in its include files, and this causes problems building Exim.
981
982PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
983 header_syntax) but Exim was just ignoring anything given after a slash.
984 In particular, this caused confusion with an attempt to use "verify =
985 reverse_host_lookup/defer_ok". An error is now given when options are
986 supplied for verify items that do not have them. (Maybe reverse_host_
987 lookup should have a defer_ok option, but that's a different point.)
988
989PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
990 defined by RFC 821) to 2048, because there were problems with some AUTH
991 commands, and RFC 1869 says the size should be increased for extended
992 SMTP commands that take arguments.
993
994PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
995 Finch).
996
997PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
998 "unknown" error; now it says that the functionality isn't in the binary.
8d67ada3 999
49c2d5ea
PH
1000PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
1001 an address' error message when a string expansion fails (syntax or
f331f3b6
PH
1002 whatever). Otherwise the password may appear in the log. Following change
1003 PH/42 below, there is no longer a chance of it appearing in a bounce
1004 message.
49c2d5ea 1005
bf759a8b
PH
1006PH/11 Installed exipick version 20050225.0 from John Jetmore.
1007
83364d30
PH
1008PH/12 If the last host in a fallback_hosts list was multihomed, only the first
1009 of its addresses was ever tried. (Bugzilla bug #2.)
1010
7999bbd7
PH
1011PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
1012 the result incorrectly in the debug output. (It correctly added a newline
1013 to what was transported.)
1014
7dbf77c9
PH
1015TF/01 Added $received_time.
1016
74e0617f
PH
1017PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
1018 commented out examples of how to interface to a virus scanner and to
1019 SpamAssassin. Also added commented examples of av_scanner and
1020 spamd_address settings.
1021
2f079f46
PH
1022PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
1023 and controls are allowed in which ACLs. There were a couple of minor
1024 errors. Some of the entries in the conditions table (which is a table of
1025 where they are NOT allowed) were getting very unwieldy; rewrote them as a
1026 negation of where the condition IS allowed.
1027
8c841523
PH
1028PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
1029
7766a4f0
PH
1030PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
1031 header file does not have a version number, so I've had to invent a new
1032 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
1033 API. The code is untested by me (my Linux distribution still has 0.3.2 of
1034 radiusclient), but it was contributed by a Radius user.
1035
8b417f2c
PH
1036PH/18 Installed Lars Mainka's patch for the support of CRL collections in
1037 files or directories, for OpenSSL.
1038
901f42cb
PH
1039PH/19 When an Exim process that is running as root has to create an Exim log
1040 file, it does so in a subprocess that runs as exim:exim so as to get the
1041 ownership right at creation (otherwise, other Exim processes might see
1042 the file with the wrong ownership). There was no test for failure of this
1043 fork() call, which would lead to the process getting stuck as it waited
1044 for a non-existent subprocess. Forks do occasionally fail when resources
1045 run out. I reviewed all the other calls to fork(); they all seem to check
1046 for failure.
1047
f9b9210e
PH
1048PH/20 When checking for unexpected SMTP input at connect time (before writing
1049 the banner), Exim was not dealing correctly with a non-positive return
1050 from the read() function. If the client had disconnected by this time,
1051 the result was a log entry for a synchronization error with an empty
1052 string after "input=" when read() returned zero. If read() returned -1
1053 (an event I could not check), uninitialized data bytes were printed.
1054 There were reports of junk text (parts of files, etc) appearing after
1055 "input=".
1056
54cdb463
PH
1057PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
1058
cf00dad6
PH
1059PH/22 Added support for macro redefinition, and (re)definition in between
1060 driver and ACL definitions.
1061
acb1b346
PH
1062PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
1063 forgetting to use the resulting value; it was using the unexpanded value.
1064
c5ddb310
PH
1065PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
1066 hadn't been configured. The fix is from Juergen Kreileder, who
1067 understands it better than I do:
1068
1069 "Here's what I see happening with three configured cyrus_sasl
1070 authenticators configured (plain, login, cram-md5):
1071
1072 On startup auth_cyrus_sasl_init() gets called for each of these.
1073 This means three calls to sasl_listmech() without a specified mech_list.
1074 => SASL tests which mechs of all available mechs actually work
1075 => three warnings about OTP not working
1076 => the returned list contains: plain, login, cram-md5, digest-md5, ...
1077
1078 With the patch, sasl_listmech() also gets called three times. But now
1079 SASL's mech_list option is set to the server_mech specified in the the
1080 authenticator. Or in other words, the answer from sasl_listmech()
1081 gets limited to just the mech you're testing for (which is different
1082 for each call.)
1083 => the return list contains just 'plain' or 'login', 'cram-md5' or
1084 nothing depending on the value of ob->server_mech.
1085
1086 I've just tested the patch: Authentication still works fine,
1087 unavailable mechs specified in the exim configuration are still
1088 caught, and the auth.log warnings about OTP are gone."
1089
31619da6
PH
1090PH/25 When debugging is enabled, the contents of the command line are added
1091 to the debugging output, even when log_selector=+arguments is not
1092 specified.
1093
bebaf0fc
PH
1094PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
1095 answer is "GNU", and only if the return is "GNU/something" is the answer
1096 "Linux".
1097
475fe28a
PH
1098PH/27 $acl_verify_message is now set immediately after the failure of a
1099 verification in an ACL, and so is available in subsequent modifiers. In
1100 particular, the message can be preserved by coding like this:
1101
1102 warn !verify = sender
1103 set acl_m0 = $acl_verify_message
1104
1105 Previously, $acl_verify_message was set only while expanding "message"
1106 and "log_message" when a very denied access.
1107
7e8bec7a
PH
1108PH/28 Modified OS/os.c-Linux with
1109
1110 -#ifndef OS_LOAD_AVERAGE
1111 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
1112
1113 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
1114 nomenclature these days.)
1115
e4a89c47
PH
1116PH/29 Installed patch from the Sieve maintainer that adds the options
1117 sieve_useraddress and sieve_subaddress to the redirect router.
1118
5ca2a9a1
PH
1119PH/30 In these circumstances:
1120 . Two addresses routed to the same list of hosts;
1121 . First host does not offer TLS;
1122 . First host accepts first address;
1123 . First host gives temporary error to second address;
1124 . Second host offers TLS and a TLS session is established;
1125 . Second host accepts second address.
1126 Exim incorrectly logged both deliveries with the TLS parameters (cipher
1127 and peerdn, if requested) that were in fact used only for the second
1128 address.
7e8bec7a 1129
c688b954
PH
1130PH/31 When doing a callout as part of verifying an address, Exim was not paying
1131 attention to any local part prefix or suffix that was matched by the
1132 router that accepted the address. It now behaves in the same way as it
1133 does for delivery: the affixes are removed from the local part unless
1134 rcpt_include_affixes is set on the transport.
1135
fed77020
PH
1136PH/32 Add the sender address, as F=<...>, to the log line when logging a
1137 timeout during the DATA phase of an incoming message.
1138
7fe1560f
PH
1139PH/33 Sieve envelope tests were broken for match types other than :is. I have
1140 applied a patch sanctioned by the Sieve maintainer.
c688b954 1141
ebb6e6d5
PH
1142PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
1143 the uid or gid is negative. A case of a negative gid caused this to be
1144 noticed. The fix allows for either to be negative.
1145
9c4e8f60
PH
1146PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
1147 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
1148 overlooked.
1149
1150PH/36 The change PH/12 above was broken. Fixed it.
1151
d7174846
PH
1152PH/37 Exim used to check for duplicate addresses in the middle of routing, on
1153 the grounds that routing the same address twice would always produce the
1154 same answer. This might have been true once, but it is certainly no
1155 longer true now. Routing a child address may depend on the previous
1156 routing that produced that child. Some complicated redirection strategies
1157 went wrong when messages had multiple recipients, and made Exim's
1158 behaviour dependent on the order in which the addresses were given.
1159
1160 I have moved the duplicate checking until after the routing is complete.
1161 Exim scans the addresses that are assigned to local and remote
1162 transports, and removes any duplicates. This means that more work will be
1163 done, as duplicates will always all be routed, but duplicates are
1164 presumably rare, so I don't expect this is of any significance.
1165
1166 For deliveries to pipes, files, and autoreplies, the duplicate checking
1167 still happens during the routing process, since they are not going to be
1168 routed further.
1169
cfe75fc3
PH
1170PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
1171 It corrects a timeout issue with spamd. This is Ian's comment: "The
1172 background is that sometimes spamd either never reads data from a
1173 connection it has accepted, or it never writes response data. The exiscan
1174 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
1175 blindly assumes that writes won't block so it may never time out."
1176
be22d70e
PH
1177PH/39 Allow G after quota size as well as K and M.
1178
0612b098
PH
1179PH/40 The value set for $authenticated_id in an authenticator may not contain
1180 binary zeroes or newlines because the value is written to log lines and
1181 to spool files. There was no check on this. Now the value is run through
1182 the string_printing() function so that such characters are converted to
1183 printable escape sequences.
1184
2e0c1448
PH
1185PH/41 $message_linecount is a new variable that contains the total number of
1186 lines in the message. Compare $body_linecount, which is the count for the
1187 body only.
1188
447d236c
PH
1189PH/42 Exim no longer gives details of delivery errors for specific addresses in
1190 bounce and delay warning messages, except in certain special cases, which
1191 are as follows:
1192
1193 (a) An SMTP error message from a remote host;
1194 (b) A message specified in a :fail: redirection;
1195 (c) A message specified in a "fail" command in a system filter;
1196 (d) A message specified in a FAIL return from the queryprogram router;
1197 (e) A message specified by the cannot_route_message router option.
1198
1199 In these cases only, Exim does include the error details in bounce and
1200 warning messages. There are also a few cases where bland messages such
1201 as "unrouteable address" or "local delivery error" are given.
1202
d20976dc
PH
1203PH/43 $value is now also set for the "else" part of a ${run expansion.
1204
f656d135
PH
1205PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
1206 being worked on, but at least Exim now implements the latest version to
1207 play with."
1208
2e2a30b4
PH
1209PH/45 In a pipe transport, although a timeout while waiting for the pipe
1210 process to complete was treated as a delivery failure, a timeout while
1211 writing the message to the pipe was logged, but erroneously treated as a
1212 successful delivery. Such timeouts include transport filter timeouts. For
1213 consistency with the overall process timeout, these timeouts are now
1214 treated as errors, giving rise to delivery failures by default. However,
1215 there is now a new Boolean option for the pipe transport called
1216 timeout_defer, which, if set TRUE, converts the failures into defers for
1217 both kinds of timeout. A transport filter timeout is now identified in
1218 the log output.
1219
9176e9f0
PH
1220PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
1221 systems where "make" and "gmake" are different, calling "gmake" at top
1222 level broke things. I've arranged for the value of $(MAKE) to be passed
1223 from the Makefile to this script so that it can call the same version of
1224 "make".
1225
7982096b 1226
bbe902f0
PH
1227A note about Exim versions 4.44 and 4.50
1228----------------------------------------
1229
1230Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
1231changes of various kinds. As a consequence, a big documentation update was
1232needed. This delayed the release for rather longer than seemed good, especially
1233in the light of a couple of (minor) security issues. Therefore, the changes
1234that fixed bugs were backported into 4.43, to create a 4.44 maintenance
1235release. So 4.44 and 4.50 are in effect two different branches that both start
1236from 4.43.
1237
1238I have left the 4.50 change log unchanged; it contains all the changes since
12394.43. The change log for 4.44 is below; many of its items are identical to
1240those for 4.50. This seems to be the most sensible way to preserve the
1241historical information.
1242
1243
f7b63901 1244Exim version 4.50
495ae4b0
PH
1245-----------------
1246
5fe762f6
PH
1247 1. Minor wording change to the doc/README.SIEVE file.
1248
139059f6 1249 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
5fe762f6 1250 computation of the current number of files was incorrect.
495ae4b0 1251
7086e875
PH
1252 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
1253 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
1254 place.
1255
35af9f61
PH
1256 4. Give more explanation in the error message when the command for a transport
1257 filter fails to execute.
1258
b668c215
PH
1259 5. There are several places where Exim runs a non-Exim command in a
1260 subprocess. The SIGUSR1 signal should be disabled for these processes. This
1261 was being done only for the command run by the queryprogram router. It is
1262 now done for all such subprocesses. The other cases are: ${run, transport
1263 filters, and the commands run by the lmtp and pipe transports.
1264
a494b1e1
PH
1265 6. Added CONFIGURE_GROUP build-time option.
1266
1267 7. Some older OS have a limit of 256 on the maximum number of file
1268 descriptors. Exim was using setrlimit() to set 1000 as a large value
1269 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
1270 systems. I've change it so that if it can't get 1000, it tries for 256.
35edf2ff 1271
c5fcb476
PH
1272 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
1273 was an oversight, and furthermore, ever since the addition of extra
1274 controls (e.g. 4.43/32), the checks on when to allow different forms of
1275 "control" were broken. There should now be diagnostics for all cases when a
1276 control that does not make sense is encountered.
1277
69358f02
PH
1278 9. Added the /retain_sender option to "control=submission".
1279
5be20824
PH
128010. $recipients is now available in the predata ACL (oversight).
1281
eb2c0248
PH
128211. Tidy the search cache before the fork to do a delivery from a message
1283 received from the command line. Otherwise the child will trigger a lookup
1284 failure and thereby defer the delivery if it tries to use (for example) a
1285 cached ldap connection that the parent has called unbind on.
1286
2a3eea10
PH
128712. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
1288 of $address_data from the recipient verification was clobbered by the
1289 sender verification.
1290
129113. The value of address_data from a sender verification is now available in
1292 $sender_address_data in subsequent conditions in the ACL statement.
1293
23c7ff99
PH
129414. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
1295
4deaf07d
PH
129615. Added a new option "connect=<time>" to callout options, to set a different
1297 connection timeout.
1298
926e1192
PH
129916. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
1300 was its contents. (It was OK if the option was not defined at all.)
1301
650edc6f
PH
130217. A "Completed" log line is now written for messages that are removed from
1303 the spool by the -Mrm option.
1304
2c7db3f5
PH
130518. New variables $sender_verify_failure and $recipient_verify_failure contain
1306 information about exactly what failed.
1307
3d235903
PH
130819. Added -dd to debug only the daemon process.
1309
7c7ad977
PH
131020. Incorporated Michael Haardt's patch to ldap.c for improving the way it
1311 handles timeouts, both on the server side and network timeouts. Renamed the
1312 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
1313
981756db
PH
131421. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
1315 It is now set to "smtps".
1316
d4eb88df
PH
131722. $host_address is now set to the target address during the checking of
1318 ignore_target_hosts.
1319
132023. When checking ignore_target_hosts for an ipliteral router, no host name was
1321 being passed; this would have caused $sender_host_name to have been used if
1322 matching the list had actually called for a host name (not very likely,
1323 since this list is usually IP addresses). A host name is now passed as
1324 "[x.x.x.x]".
1325
7d468ab8
PH
132624. Changed the calls that set up the SIGCHLD handler in the daemon to use the
1327 code that specifies a non-restarting handler (typically sigaction() in
1328 modern systems) in an attempt to fix a rare and obscure crash bug.
1329
133025. Narrowed the window for a race in the daemon that could cause it to ignore
1331 SIGCHLD signals. This is not a major problem, because they are used only to
1332 wake it up if nothing else does.
1333
62c0818f
PH
133426. A malformed maildirsize file could cause Exim to calculate negative values
1335 for the mailbox size or file count. Odd effects could occur as a result.
1336 The maildirsize information is now recalculated if the size or filecount
1337 end up negative.
1338
26034054
PH
133927. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
1340 support for a long time. Removed HAVE_SYS_VFS_H.
1341
af66f652
PH
134228. Installed the latest version of exipick from John Jetmore.
1343
90af77f4
PH
134429. In an address list, if the pattern was not a regular expression, an empty
1345 subject address (from a bounce message) matched only if the pattern was an
1346 empty string. Non-empty patterns were not even tested. This was the wrong
1347 because it is perfectly reasonable to use an empty address as part of a
1348 database query. An empty address is now tested by patterns that are
1349 lookups. However, all the other forms of pattern expect the subject to
1350 contain a local part and a domain, and therefore, for them, an empty
1351 address still always fails if the pattern is not itself empty.
1352
d8ef3577
PH
135330. Exim went into a mad DNS loop when attempting to do a callout where the
1354 host was specified on an smtp transport, and looking it up yielded more
1355 than one IP address.
1356
5cb8cbc6
PH
135731. Re-factored the code for checking spool and log partition space into a
1358 function that finds that data and another that does the check. The former
1359 is then used to implement four new variables: $spool_space, $log_space,
1360 $spool_inodes, and $log_inodes.
1361
14702f5b
PH
136232. The RFC2047 encoding function was originally intended for short strings
1363 such as real names; it was not keeping to the 75-character limit for
1364 encoded words that the RFC imposes. It now respects the limit, and
1365 generates multiple encoded words if necessary. To be on the safe side, I
1366 have increased the buffer size for the ${rfc2047: expansion operator from
1367 1024 to 2048 bytes.
1368
063b1e99
PH
136933. It is now permitted to omit both strings after an "if" condition; if the
1370 condition is true, the result is "true". As before, when the second string
1371 is omitted, a false condition yields an empty string. This makes it less
1372 cumbersome to write custom ACL and router conditions.
1373
652e1b65
PH
137434. Failure to deliver a bounce message always caused it to be frozen, even if
1375 there was an errors_to setting on the router. The errors_to setting is now
1376 respected.
1377
6f0c9a4f
PH
137835. If an IPv6 address is given for -bh or -bhc, it is now converted to the
1379 canonical form (fully expanded) before being placed in
1380 $sender_host_address.
1381
33397d19
PH
138236. The table in the code that translates DNS record types into text (T_A to
1383 "A" for instance) was missing entries for NS and CNAME. It is just possible
1384 that this could have caused confusion if both these types were looked up
1385 for the same domain, because the text type is used as part of Exim's
1386 per-process caching. But the chance of anyone hitting this buglet seems
1387 very small.
1388
7bb56e1f
PH
138937. The dnsdb lookup has been extended in a number of ways.
1390
1391 (1) There is a new type, "zns", which walks up the domain tree until it
1392 finds some nameserver records. It should be used with care.
1393
ea3bc19b
PH
1394 (2) There is a new type, "mxh", which is like "mx" except that it returns
1395 just the host names, not the priorities.
1396
1397 (3) It is now possible to give a list of domains (or IP addresses) to be
ff4dbb19
PH
1398 looked up. The behaviour when one of the lookups defers can be
1399 controlled by a keyword.
7bb56e1f 1400
ea3bc19b 1401 (4) It is now possible to specify the separator character for use when
7bb56e1f 1402 multiple records are returned.
33397d19 1403
0bcb2a0e
PH
140438. The dnslists ACL condition has been extended: it is now possible to supply
1405 a list of IP addresses and/or domains to be looked up in a particular DNS
1406 domain.
1407
2ac0e484
PH
140839. Added log_selector=+queue_time_overall.
1409
4e1fde53
PH
141040. When running the queue in the test harness, wait just a tad after forking a
1411 delivery process, to get repeatability of debugging output.
1412
de365ded
PH
141341. Include certificate and key file names in error message when GnuTLS fails
1414 to set them up, because the GnuTLS error message doesn't include the name
1415 of the failing file when there is a problem reading it.
1416
f05da2e8
PH
141742. Allow both -bf and -bF in the same test run.
1418
d6453af2
PH
141943. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
1420
f7b63901
PH
142144. The "Exiscan patch" is now merged into the mainline Exim source.
1422
142345. Sometimes the final signoff response after QUIT could fail to get
1424 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
1425 < 0 before doing a fflush(). This bug looks as though it goes back to the
1426 introduction of TLS in release 3.20, but "sometimes" must have been rare
1427 because the tests only now provoked it.
1428
a444213a
PH
142946. Reset the locale to "C" after calling embedded Perl, in case it was changed
1430 (this can affect the format of dates).
1431
0ec020ea
PH
143247. exim_tidydb, when checking for the continued existence of a message for
1433 which it has found a message-specific retry record, was not finding
1434 messages that were in split spool directories. Consequently, it was
1435 deleting retry records that should have stayed in existence.
1436
b1206957
PH
143748. Steve fixed some bugs in eximstats.
1438
143949. The SPA authentication driver was not abandoning authentication and moving
1440 on to the next authenticator when an expansion was forced to fail,
1441 contradicting the general specification for all authenticators. Instead it
1442 was generating a temporary error. It now behaves as specified.
1443
26dd5a95
PH
144450. The default ordering of permitted cipher suites for GnuTLS was pessimal
1445 (the order specifies the preference for clients). The order is now AES256,
1446 AES128, 3DES, ARCFOUR128.
1447
343b2385
PH
144851. Small patch to Sieve code - explicitly set From: when generating an
1449 autoreply.
1450
1c5466b9
PH
145152. Exim crashed if a remote delivery caused a very long error message to be
1452 recorded - for instance if somebody sent an entire SpamAssassin report back
1453 as a large number of 550 error lines. This bug was coincidentally fixed by
1454 increasing the size of one of Exim's internal buffers (big_buffer) that
1455 happened as part of the Exiscan merge. However, to be on the safe side, I
1456 have made the code more robust (and fixed the comments that describe what
1457 is going on).
1458
55ee9ee3
PH
145953. Now that there can be additional text after "Completed" in log lines (if
1460 the queue_time_overall log selector is set), a one-byte patch to exigrep
1461 was needed to allow it to recognize "Completed" as not the last thing in
1462 the line.
1463
d38f8232
PH
146454. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
1465 patch that reportedly fixes this has been added. I am not expert enough to
1466 create a test for it. This is what the patch creator wrote:
1467
1468 "I found a little strange behaviour of ldap code when working with
1469 Windows 2003 AD Domain, where users was placed in more than one
1470 Organization Units. When I tried to give exim partial DN, the exit code
1471 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
1472 But simultaneously result of request was absolutely normal ldap result,
1473 so I produce this patch..."
1474
3295e65b
PH
1475 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
1476 REFERENCE, so I have modified the code to exclude the patch when that macro
1477 is not defined.
1478
7102e136
PH
147955. Some experimental protocols are using DNS PTR records for new purposes. The
1480 keys for these records are domain names, not reversed IP addresses. The
b975ba52
PH
1481 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
1482 leaves it alone. Component reversal etc. now happens only for IP addresses.
ea3a6f44 1483 CAN-2005-0021
7102e136 1484
3ca0ba97
PH
148556. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
1486
c2bcbe20
PH
148757. Double the size of the debug message buffer (to 2048) so that more of very
1488 long debug lines gets shown.
1489
18ce445d
PH
149058. The exicyclog utility now does better if the number of log files to keep
1491 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
1492
1f5b4c3d
PH
149359. Two changes related to the smtp_active_hostname option:
1494
1495 (1) $smtp_active_hostname is now available as a variable.
1496 (2) The default for smtp_banner uses $smtp_active_hostname instead
1497 of $primary_hostname.
1498
b975ba52
PH
149960. The host_aton() function is supposed to be passed a string that is known
1500 to be a valid IP address. However, in the case of IPv6 addresses, it was
1501 not checking this. This is a hostage to fortune. Exim now panics and dies
1502 if the condition is not met. A case was found where this could be provoked
85b87bc2
PH
1503 from a dnsdb PTR lookup with an IPv6 address that had more than 8
1504 components; fortuitously, this particular loophole had already been fixed
1505 by change 4.50/55 above.
1506
1507 If there are any other similar loopholes, the new check in host_aton()
1508 itself should stop them being exploited. The report I received stated that
1509 data on the command line could provoke the exploit when Exim was running as
1510 exim, but did not say which command line option was involved. All I could
1511 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
1512 running as the user.
ea3a6f44 1513 CAN-2005-0021
85b87bc2
PH
1514
151561. There was a buffer overflow vulnerability in the SPA authentication code
1516 (which came originally from the Samba project). I have added a test to the
1517 spa_base64_to_bits() function which I hope fixes it.
ea3a6f44 1518 CAN-2005-0022
b975ba52 1519
17ffcae7
PH
152062. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
1521 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
1522
d95f9fdb
PH
152363. The daemon start-up calls getloadavg() while still root for those OS that
1524 need the first call to be done as root, but it missed one case: when
1525 deliver_queue_load_max is set with deliver_drop_privilege. This is
1526 necessary for the benefit of the queue runner, because there is no re-exec
1527 when deliver_drop_privilege is set.
1528
86b8287f
PH
152964. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
1530 This has been fixed.
1531
60dc5e56
PH
153265. Caching of lookup data for "hosts =" ACL conditions, when a named host list
1533 was in use, was not putting the data itself into the right store pool;
1534 consequently, it could be overwritten for a subsequent message in the same
1535 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
1536 the caching.)
1537
533244af
PH
153866. Added hosts_max_try_hardlimit to the smtp transport, default 50.
1539
a5a28604
PH
154067. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
1541 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
1542 the function were treating the return as a boolean value, which happened to
1543 work because 0=false and not-0=true, but is not correct code.
1544
7e634d24
PH
154568. The host_aton() function was not handling scoped IPv6 addresses (those
1546 with, for example, "%eth0" on the end) correctly.
1547
3e11c26b
PH
154869. Fixed some compiler warnings in acl.c for the bitmaps specified with
1549 negated items (that is, ~something) in unsigned ints. Some compilers
1550 apparently mutter when there is no cast.
1551
6729cf78
PH
155270. If an address verification called from an ACL failed, and did not produce a
1553 user-specific message (i.e. there was only a "system" message), nothing was
1554 put in $acl_verify_message. In this situation, it now puts the system
1555 message there.
1556
00f00ca5
PH
155771. Change 4.23/11 added synchronization checking at the start of an SMTP
1558 session; change 4.31/43 added the unwanted input to the log line - except
1559 that it did not do this in the start of session case. It now does.
1560
c9bdd01c
PH
156172. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
1562 This is wrong and can cause the other end to generate a synchronization
1563 error if it is another Exim or anything else that does the synchronization
1564 check. A QUIT command is no longer sent after a timeout.
1565
d43194df
PH
156673. $host_lookup_deferred has been added, to make it easier to detect DEFERs
1567 during host lookups.
1568
fe5b5d0b
PH
156974. The defer_ok option of callout verification was not working if it was used
1570 when verifying addresses in header lines, that is, for this case:
1571
1572 verify = header_sender/callout=defer_ok
1573
76a2d7ba
PH
157475. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
1575 those file descriptors could be used for SMTP connections. If anything
1576 wrote to stderr (the example that came up was "warn" in embedded Perl), it
1577 could be sent to the SMTP client, causing chaos. The daemon now opens
1578 stdin, stdout, and stderr to /dev/null when it puts itself into the
1579 background.
1580
158176. Arrange for output from Perl's "warn" command to be written to Exim's main
1582 log by default. The user can override this with suitable Perl magic.
1583
04f7d5b9
PH
158477. The use of log_message on a "discard" ACL verb, which is supposed to add to
1585 the log message when discard triggers, was not working for the DATA ACL or
1586 for the non-SMTP ACL.
1587
bc60667e
PH
158878. Error message wording change in sieve.c.
1589
bb6e88ff
PH
159079. If smtp_accept_max_per_host was set, the number of connections could be
1591 restricted to fewer than expected, because the daemon was trying to set up
1592 a new connection before checking whether the processes handling previous
1593 connections had finished. The check for completed processes is now done
1594 earlier. On busy systems, this bug wouldn't be noticed because something
1595 else would have woken the daemon, and it would have reaped the completed
1596 process earlier.
1597
1e70f85b
PH
159880. If a message was submitted locally by a user whose login name contained one
1599 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
1600 It caused a spool format error. I have fixed the spool reading code. A
1601 related problem was that the "from" clause in the Received: line became
1602 illegal because of the space(s). It is now covered by ${quote_local_part.
1603
160481. Included the latest eximstats from Steve (adds average sizes to HTML Top
1605 tables).
1606
4e01f9d6
PH
160782. Updated OS/Makefile-AIX as per message from Mike Meredith.
1608
1ee1cef2
PH
160983. Patch from Sieve maintainer to fix unterminated string problem in
1610 "vacation" handling.
1611
6e2b4ccc
PH
161284. Some minor changes to the Linux configuration files to help with other
1613 OS variants using glibc.
1614
8e669ac1
PH
161585. One more patch for Sieve to update vacation handling to latest spec.
1616
495ae4b0 1617
bbe902f0
PH
1618----------------------------------------------------
1619See the note above about the 4.44 and 4.50 releases.
1620----------------------------------------------------
1621
1622
1623Exim version 4.44
1624-----------------
1625
1626 1. Change 4.43/35 introduced a bug that caused file counts to be
1627 incorrectly computed when quota_filecount was set in an appendfile
1628 transport
1629
1630 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
1631 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
1632 place.
1633
1634 3. Give more explanation in the error message when the command for a transport
1635 filter fails to execute.
1636
1637 4. There are several places where Exim runs a non-Exim command in a
1638 subprocess. The SIGUSR1 signal should be disabled for these processes. This
1639 was being done only for the command run by the queryprogram router. It is
1640 now done for all such subprocesses. The other cases are: ${run, transport
1641 filters, and the commands run by the lmtp and pipe transports.
1642
1643 5. Some older OS have a limit of 256 on the maximum number of file
1644 descriptors. Exim was using setrlimit() to set 1000 as a large value
1645 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
1646 systems. I've change it so that if it can't get 1000, it tries for 256.
1647
1648 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
1649 was an oversight, and furthermore, ever since the addition of extra
1650 controls (e.g. 4.43/32), the checks on when to allow different forms of
1651 "control" were broken. There should now be diagnostics for all cases when a
1652 control that does not make sense is encountered.
1653
1654 7. $recipients is now available in the predata ACL (oversight).
1655
1656 8. Tidy the search cache before the fork to do a delivery from a message
1657 received from the command line. Otherwise the child will trigger a lookup
1658 failure and thereby defer the delivery if it tries to use (for example) a
1659 cached ldap connection that the parent has called unbind on.
1660
1661 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
1662 of $address_data from the recipient verification was clobbered by the
1663 sender verification.
1664
166510. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
1666 was its contents. (It was OK if the option was not defined at all.)
1667
166811. A "Completed" log line is now written for messages that are removed from
1669 the spool by the -Mrm option.
1670
167112. $host_address is now set to the target address during the checking of
1672 ignore_target_hosts.
1673
167413. When checking ignore_target_hosts for an ipliteral router, no host name was
1675 being passed; this would have caused $sender_host_name to have been used if
1676 matching the list had actually called for a host name (not very likely,
1677 since this list is usually IP addresses). A host name is now passed as
1678 "[x.x.x.x]".
1679
168014. Changed the calls that set up the SIGCHLD handler in the daemon to use the
1681 code that specifies a non-restarting handler (typically sigaction() in
1682 modern systems) in an attempt to fix a rare and obscure crash bug.
1683
168415. Narrowed the window for a race in the daemon that could cause it to ignore
1685 SIGCHLD signals. This is not a major problem, because they are used only to
1686 wake it up if nothing else does.
1687
168816. A malformed maildirsize file could cause Exim to calculate negative values
1689 for the mailbox size or file count. Odd effects could occur as a result.
1690 The maildirsize information is now recalculated if the size or filecount
1691 end up negative.
1692
169317. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
1694 support for a long time. Removed HAVE_SYS_VFS_H.
1695
ea3a6f44 169618. Updated exipick to current release from John Jetmore.
bbe902f0
PH
1697
169819. Allow an empty sender to be matched against a lookup in an address list.
1699 Previously the only cases considered were a regular expression, or an
1700 empty pattern.
1701
170220. Exim went into a mad DNS lookup loop when doing a callout where the
1703 host was specified on the transport, if the DNS lookup yielded more than
1704 one IP address.
1705
ea3a6f44
NM
170621. The RFC2047 encoding function was originally intended for short strings
1707 such as real names; it was not keeping to the 75-character limit for
1708 encoded words that the RFC imposes. It now respects the limit, and
1709 generates multiple encoded words if necessary. To be on the safe side, I
1710 have increased the buffer size for the ${rfc2047: expansion operator from
1711 1024 to 2048 bytes.
bbe902f0 1712
ea3a6f44
NM
171322. Failure to deliver a bounce message always caused it to be frozen, even if
1714 there was an errors_to setting on the router. The errors_to setting is now
1715 respected.
bbe902f0
PH
1716
171723. If an IPv6 address is given for -bh or -bhc, it is now converted to the
1718 canonical form (fully expanded) before being placed in
1719 $sender_host_address.
1720
172124. Updated eximstats to version 1.33
1722
ea3a6f44
NM
172325. Include certificate and key file names in error message when GnuTLS fails
1724 to set them up, because the GnuTLS error message doesn't include the name
1725 of the failing file when there is a problem reading it.
bbe902f0
PH
1726
172726. Expand error message when OpenSSL has problems setting up cert/key files.
ea3a6f44 1728 As per change 25.
bbe902f0 1729
ea3a6f44
NM
173027. Reset the locale to "C" after calling embedded Perl, in case it was changed
1731 (this can affect the format of dates).
bbe902f0 1732
ea3a6f44
NM
173328. exim_tidydb, when checking for the continued existence of a message for
1734 which it has found a message-specific retry record, was not finding
1735 messages that were in split spool directories. Consequently, it was
1736 deleting retry records that should have stayed in existence.
bbe902f0
PH
1737
173829. eximstats updated to version 1.35
1739 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
1740 1.35 - bugfix such that pie charts by volume are generated correctly
1741
ea3a6f44
NM
174230. The SPA authentication driver was not abandoning authentication and moving
1743 on to the next authenticator when an expansion was forced to fail,
1744 contradicting the general specification for all authenticators. Instead it
1745 was generating a temporary error. It now behaves as specified.
bbe902f0 1746
ea3a6f44
NM
174731. The default ordering of permitted cipher suites for GnuTLS was pessimal
1748 (the order specifies the preference for clients). The order is now AES256,
1749 AES128, 3DES, ARCFOUR128.
bbe902f0 1750
ea3a6f44
NM
175131. Small patch to Sieve code - explicitly set From: when generating an
1752 autoreply.
bbe902f0 1753
ea3a6f44
NM
175432. Exim crashed if a remote delivery caused a very long error message to be
1755 recorded - for instance if somebody sent an entire SpamAssassin report back
1756 as a large number of 550 error lines. This bug was coincidentally fixed by
1757 increasing the size of one of Exim's internal buffers (big_buffer) that
1758 happened as part of the Exiscan merge. However, to be on the safe side, I
1759 have made the code more robust (and fixed the comments that describe what
1760 is going on).
bbe902f0 1761
ea3a6f44
NM
176233. Some experimental protocols are using DNS PTR records for new purposes. The
1763 keys for these records are domain names, not reversed IP addresses. The
1764 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
1765 leaves it alone. Component reversal etc. now happens only for IP addresses.
bbe902f0
PH
1766 CAN-2005-0021
1767
ea3a6f44
NM
176834. The host_aton() function is supposed to be passed a string that is known
1769 to be a valid IP address. However, in the case of IPv6 addresses, it was
1770 not checking this. This is a hostage to fortune. Exim now panics and dies
1771 if the condition is not met. A case was found where this could be provoked
1772 from a dnsdb PTR lookup with an IPv6 address that had more than 8
1773 components; fortuitously, this particular loophole had already been fixed
1774 by change 4.50/55 or 4.44/33 above.
1775
1776 If there are any other similar loopholes, the new check in host_aton()
1777 itself should stop them being exploited. The report I received stated that
1778 data on the command line could provoke the exploit when Exim was running as
1779 exim, but did not say which command line option was involved. All I could
1780 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
1781 running as the user.
bbe902f0
PH
1782 CAN-2005-0021
1783
ea3a6f44
NM
178435. There was a buffer overflow vulnerability in the SPA authentication code
1785 (which came originally from the Samba project). I have added a test to the
1786 spa_base64_to_bits() function which I hope fixes it.
bbe902f0
PH
1787 CAN-2005-0022
1788
ea3a6f44
NM
178936. The daemon start-up calls getloadavg() while still root for those OS that
1790 need the first call to be done as root, but it missed one case: when
1791 deliver_queue_load_max is set with deliver_drop_privilege. This is
1792 necessary for the benefit of the queue runner, because there is no re-exec
1793 when deliver_drop_privilege is set.
bbe902f0 1794
ea3a6f44
NM
179537. Caching of lookup data for "hosts =" ACL conditions, when a named host list
1796 was in use, was not putting the data itself into the right store pool;
1797 consequently, it could be overwritten for a subsequent message in the same
1798 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
1799 the caching.)
bbe902f0 1800
ea3a6f44
NM
180138. Sometimes the final signoff response after QUIT could fail to get
1802 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
1803 < 0 before doing a fflush(). This bug looks as though it goes back to the
1804 introduction of TLS in release 3.20, but "sometimes" must have been rare
1805 because the tests only now provoked it.
bbe902f0
PH
1806
1807
495ae4b0
PH
1808Exim version 4.43
1809-----------------
1810
1811 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
1812 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
1813 Now it returns an integer. A number of calls were still expecting a T/F
1814 return. Fortuitously, in all cases, the tests worked in OK situations,
1815 which is the norm. However, things would have gone wrong on any write
1816 failures on the smtp file descriptor. This function is used when sending
1817 messages over SMTP and also when doing verify callouts.
1818
1819 2. When Exim is called to do synchronous delivery of a locally submitted
1820 message (the -odf or -odi options), it no longer closes stderr before doing
1821 the delivery.
1822
1823 3. Implemented the mua_wrapper option.
1824
1825 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
1826
1827 5. Implemented the functions header_remove(), header_testname(),
1828 header_add_at_position(), and receive_remove_recipient(), and exported them
1829 to local_scan().
1830
1831 6. If an ACL "warn" statement specified the addition of headers, Exim already
1832 inserted X-ACL-Warn: at the start if there was no header name. However, it
1833 was not making this test for the second and subsequent header lines if
1834 there were newlines in the string. This meant that an invalid header could
1835 be inserted if Exim was badly configured.
1836
1837 7. Allow an ACL "warn" statement to add header lines at the start or after all
1838 the Received: headers, as well as at the end.
1839
1840 8. Added the rcpt_4xx retry error code.
1841
1842 9. Added postmaster_mailfrom=xxx to callout verification option.
1843
184410. Added mailfrom=xxxx to the callout verification option, for verify=
1845 header_sender only.
1846
184711. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
1848 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
1849
185012. Inserted some casts to stop certain compilers warning when using pointer
1851 differences as field lengths or precisions in printf-type calls (mostly
1852 affecting debugging statements).
1853
185413. Added optional readline() support for -be (dynamically loaded).
1855
185614. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
1857 same clock tick as a message's arrival, so that its received time was the
1858 same as the "first fail" time on the retry record, and that message
1859 remained on the queue past the ultimate address timeout, every queue runner
1860 would try a delivery (because it was past the ultimate address timeout) but
1861 after another failure, the ultimate address timeout, which should have then
1862 bounced the address, did not kick in. This was a "< instead of <=" error;
1863 in most cases the first failure would have been in the next clock tick
1864 after the received time, and all would be well.
1865
186615. The special items beginning with @ in domain lists (e.g. @mx_any) were not
1867 being recognized when the domain list was tested by the match_domain
1868 condition in an expansion string.
1869
187016. Added the ${str2b64: operator.
1871
187217. Exim was always calling setrlimit() to set a large limit for the number of
1873 processes, without checking whether the existing limit was already
1874 adequate. (It did check for the limit on file descriptors.) Furthermore,
1875 errors from getrlimit() and setrlimit() were being ignored. Now they are
1876 logged to the main and panic logs, but Exim does carry on, to try to do its
1877 job under whatever limits there are.
1878
187918. Imported PCRE 5.0.
1880
188119. Trivial typo in log message " temporarily refused connection" (the leading
1882 space).
1883
188420. If the log selector return_path_on_delivery was set and an address was
1885 redirected to /dev/null, the delivery process crashed because it assumed
1886 that a return path would always be set for a "successful" delivery. In this
1887 case, the whole delivery is bypassed as an optimization, and therefore no
1888 return path is set.
1889
189021. Internal re-arrangement: the function for sending a challenge and reading
1891 a response while authentication was assuming a zero-terminated challenge
1892 string. It's now changed to take a pointer and a length, to allow for
1893 binary data in such strings.
1894
189522. Added the cyrus_sasl authenticator (code supplied by MBM).
1896
189723. Exim was not respecting finduser_retries when seeking the login of the
1898 uid under which it was called; it was always trying 10 times. (The default
1899 setting of finduser_retries is zero.) Also, it was sleeping after the final
1900 failure, which is pointless.
1901
190224. Implemented tls_on_connect_ports.
1903
190425. Implemented acl_smtp_predata.
1905
190626. If the domain in control=submission is set empty, Exim assumes that the
1907 authenticated id is a complete email address when it generates From: or
1908 Sender: header lines.
1909
191027. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
1911 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
1912 chown and chgrp in /bin and hostname in /usr/bin.
1913
191428. Exim was keeping the "process log" file open after each use, just as it
1915 does for the main log. This opens the possibility of it remaining open for
1916 long periods when the USR1 signal hits a daemon. Occasional processlog
1917 errors were reported, that could have been caused by this. Anyway, it seems
1918 much more sensible not to leave this file open at all, so that is what now
1919 happens.
1920
192129. The long-running daemon process does not normally write to the log once it
1922 has entered its main loop, and it closes the log before doing so. This is
1923 so that log files can straightforwardly be renamed and moved. However,
1924 there are a couple of unusual error situations where the daemon does write
1925 log entries, and I had neglected to close the log afterwards.
1926
192730. The text of an SMTP error response that was received during a remote
1928 delivery was being truncated at 512 bytes. This is too short for some of
1929 the long messages that one sometimes sees. I've increased the limit to
1930 1024.
1931
193231. It is now possible to make retry rules that apply only when a message has a
1933 specific sender, in particular, an empty sender.
1934
193532. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
1936 it possible to be selective about when SMTP synchronization is enforced.
1937
193833. Added "control = caseful_local_part" and "control = "caselower_local_part".
1939
194032. Implemented hosts_connection_nolog.
1941
194233. Added an ACL for QUIT.
1943
194434. Setting "delay_warning=" to disable warnings was not working; it gave a
1945 syntax error.
1946
194735. Added mailbox_size and mailbox_filecount to appendfile.
1948
194936. Added control = no_multiline_responses to ACLs.
1950
195137. There was a bug in the logic of the code that waits for the clock to tick
1952 in the case where the clock went backwards by a substantial amount such
1953 that the microsecond fraction of "now" was more than the microsecond
1954 fraction of "then" (but the whole seconds number was less).
1955
195638. Added support for the libradius Radius client library this is found on
1957 FreeBSD (previously only the radiusclient library was supported).
1958
1959
1960Exim version 4.42
1961-----------------
1962
1963 1. When certain lookups returned multiple values in the form name=value, the
1964 quoting of the values was not always being done properly. Specifically:
1965 (a) If the value started with a double quote, but contained no whitespace,
1966 it was not quoted.
1967 (b) If the value contained whitespace other than a space character (i.e.
1968 tabs or newlines or carriage returns) it was not quoted.
1969 This fix has been applied to the mysql and pgsql lookups by writing a
1970 separate quoting function and calling it from the lookup code. The fix
1971 should probably also be applied to nisplus, ibase and oracle lookups, but
1972 since I cannot test any of those, I have not disturbed their existing code.
1973
1974 2. A hit in the callout cache for a specific address caused a log line with no
1975 reason for rejecting RCPT. Now it says "Previous (cached) callout
1976 verification failure".
1977
1978 3. There was an off-by-one bug in the queryprogram router. An over-long
1979 return line was truncated at 256 instead of 255 characters, thereby
1980 overflowing its buffer with the terminating zero. As well as fixing this, I
1981 have increased the buffer size to 1024 (and made a note to document this).
1982
1983 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
1984 when Exim is waiting for an SMTP response from a remote server, Exim
1985 restarts its select() call on the socket, thereby resetting its timeout.
1986 This is not a problem when such interrupts are rare. Somebody set up a cron
1987 job to run exiwhat every 2 minutes, which is less than the normal select()
1988 timeout (5 or 10 minutes). This meant that the select() timeout never
1989 kicked in because it was always reset. I have fixed this by comparing the
1990 time when an interrupt arrives with the time at the start of the first call
1991 to select(). If more time than the timeout has elapsed, the interrupt is
1992 treated as a timeout.
1993
1994 5. Some internal re-factoring in preparation for the addition of Sieve
1995 extensions (by MH). In particular, the "personal" test is moved to a
1996 separate function, and given an option for scanning Cc: and Bcc: (which is
1997 not set for Exim filters).
1998
1999 6. When Exim created an email address using the login of the caller as the
2000 local part (e.g. when creating a From: or Sender: header line), it was not
2001 quoting the local part when it contained special characters such as @.
2002
2003 7. Installed new OpenBSD configuration files.
2004
2005 8. Reworded some messages for syntax errors in "and" and "or" conditions to
2006 try to make them clearer.
2007
2008 9. Callout options, other than the timeout value, were being ignored when
2009 verifying sender addresses in header lines. For example, when using
2010
2011 verify = header_sender/callout=no_cache
2012
2013 the cache was (incorrectly) being used.
2014
201510. Added a missing instance of ${EXE} to the exim_install script; this affects
2016 only the Cygwin environment.
2017
201811. When return_path_on_delivery was set as a log selector, if different remote
2019 addresses in the same message used different return paths and parallel
2020 remote delivery occurred, the wrong values would sometimes be logged.
2021 (Whenever a remote delivery process finished, the return path value from
2022 the most recently started remote delivery process was logged.)
2023
202412. RFC 3848 specifies standard names for the "with" phrase in Received: header
2025 lines when AUTH and/or TLS are in use. This is the "received protocol"
2026 field. Exim used to use "asmtp" for authenticated SMTP, without any
2027 indication (in the protocol name) for TLS use. Now it follows the RFC and
2028 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
2029 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
2030 names appear in log lines as well as in Received: header lines.
2031
203213. Installed MH's patches for Sieve to add the "copy" and "vacation"
2033 extensions, and comparison tests, and to fix some bugs.
2034
203514. Changes to the "personal" filter test:
2036
2037 (1) The test was buggy in that it was just doing the equivalent of
2038 "contains" tests on header lines. For example, if a user's address was
2039 anne@some.where, the "personal" test would incorrectly be true for
2040
2041 To: susanne@some.where
2042
2043 This test is now done by extracting each address from the header in turn,
2044 and checking the entire address. Other tests that are part of "personal"
2045 are now done using regular expressions (for example, to check local parts
2046 of addresses in From: header lines).
2047
2048 (2) The list of non-personal local parts in From: addresses has been
2049 extended to include "listserv", "majordomo", "*-request", and "owner-*",
2050 taken from the Sieve specification recommendations.
2051
2052 (3) If the message contains any header line starting with "List-" it is
2053 treated as non-personal.
2054
2055 (4) The test for "circular" in the Subject: header line has been removed
2056 because it now seems ill-conceived.
2057
205815. Minor typos in src/EDITME comments corrected.
2059
206016. Installed latest exipick from John Jetmore.
2061
206217. If headers_add on a router specified a text string that was too long for
2063 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
2064 of string_sprintf() is now avoided.
2065
206618. $message_body_size was not set (it was always zero) when running the DATA
2067 ACL and the local_scan() function.
2068
206919. For the "mail" command in an Exim filter, no default was being set for
2070 the once_repeat time, causing a random time value to be used if "once" was
2071 specified. (If the value happened to be <= 0, no repeat happened.) The
2072 default is now 0s, meaning "never repeat". The "vacation" command was OK
2073 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
2074 (I found it when inspecting the code).
2075
207620. There is now an overall timeout for performing a callout verification. It
2077 defaults to 4 times the callout timeout, which applies to individual SMTP
2078 commands during the callout. The overall timeout applies when there is more
2079 than one host that can be tried. The timeout is checked before trying the
2080 next host. This prevents very long delays if there are a large number of
2081 hosts and all are timing out (e.g. when the network connections are timing
2082 out). The value of the overall timeout can be changed by specifying an
2083 additional sub-option for "callout", called "maxwait". For example:
2084
2085 verify = sender/callout=5s,maxwait=20s
2086
208721. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
2088 to the end before writing, but this should make it even safer).
2089
209022. Exim was forgetting that it had advertised PIPELINING for the second and
2091 subsequent messages on an SMTP connection. It was also not resetting its
2092 memory on STARTTLS and an internal HELO.
2093
209423. When Exim logs an SMTP synchronization error within a session, it now
2095 records whether PIPELINING has been advertised or not.
2096
209724. Added 3 instances of "(long int)" casts to time_t variables that were being
2098 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
2099 rather than long int.
2100
210125. Installed the latest Cygwin configuration files from the Cygwin maintainer.
2102
210326. Added the never_mail option to autoreply.
2104
2105
2106Exim version 4.41
2107-----------------
2108
2109 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
2110 crash if the getsockname() call failed; this can happen if a connection is
2111 closed very soon after it is established. The problem was simply in the
2112 order in which certain operations were done, causing Exim to try to write
2113 to the SMTP stream before it had set up the file descriptor. The bug has
2114 been fixed by making things happen in the correct order.
2115
2116
2117Exim version 4.40
2118-----------------
2119
2120 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
2121 before the connection was closed, thus losing the rejection response.
2122
2123 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
2124 some early Solaris releases, but causes trouble in current releases where
2125 socklen_t is defined.
2126
2127 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
2128 always exist.
2129
2130 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
2131 configured.
2132
2133 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
2134 that releasing the top of it at the end releases what was used for sub-
2135 expansions (unless the block got too big). However, discard this block if
2136 the first thing is a variable or header, so that we can use its block when
2137 it is dynamic (useful for very large $message_headers, for example).
2138
2139 6. Lookups now cache *every* query, not just the most recent. A new, separate
2140 store pool is used for this. It can be recovered when all lookup caches are
2141 flushed. Lookups now release memory at the end of their result strings.
2142 This has involved some general refactoring of the lookup sources.
2143
2144 7. Some code has been added to the store_xxx() functions to reduce the amount
2145 of flapping under certain conditions.
2146
2147 8. log_incoming_interface used to affect only the <= reception log lines. Now
2148 it causes the local interface and port to be added to several more SMTP log
2149 lines, for example "SMTP connection from", and rejection lines.
2150
2151 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
2152
215310. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
2154
215511. If $host_data was set by virtue of a hosts lookup in an ACL, its value
2156 could be overwritten at the end of the current message (or the start of a
2157 new message if it was set in a HELO ACL). The value is now preserved for
2158 the duration of the SMTP connection.
2159
216012. If a transport had a headers_rewrite setting, and a matching header line
2161 contained an unqualified address, that address was qualified, even if it
2162 did not match any rewriting rules. The underlying bug was that the values
2163 of the flags that permit the existence of unqualified sender and recipient
2164 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
2165 non-local messages, and by -bnq for local messages) were not being
2166 preserved with the message after it was received.
2167
216813. When Exim was logging an SMTP synchronization error, it could sometimes log
2169 "next input=" as part of the text comprising the host identity instead of
2170 the correct text. The code was using the same buffer for two different
2171 strings. However, depending on which order the printing function evaluated
2172 its arguments, the bug did not always show up. Under Linux, for example, my
2173 test suite worked just fine.
2174
217514. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
2176 doesn't work with some older versions of Perl. It has been changed to "my",
2177 which in any case is probably the better facility to use.
2178
217915. A really picky compiler found some instances of statements for creating
2180 error messages that either had too many or two few arguments for the format
2181 string.
2182
218316. The size of the buffer for calls to the DNS resolver has been increased
2184 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
2185 for addresses that have a lot of PTR records. This alleviates a problem; it
2186 does not fully solve it.
2187
218817. A dnsdb lookup for PTR records that receives more data than will fit in the
2189 buffer now truncates the list and logs the incident, which is the same
2190 action as happens when Exim is looking up a host name and its aliases.
2191 Previously in this situation something unpredictable would happen;
2192 sometimes it was "internal error: store_reset failed".
2193
219418. If a server dropped the connection unexpectedly when an Exim client was
2195 using GnuTLS and trying to read a response, the client delivery process
2196 crashed while trying to generate an error log message.
2197
219819. If a "warn" verb in an ACL added multiple headers to a message in a single
2199 string, for example:
2200
2201 warn message = H1: something\nH2: something
2202
2203 the text was added as a single header line from Exim's point of view
2204 though it ended up OK in the delivered message. However, searching for the
2205 second and subsequent header lines using $h_h2: did not work. This has been
2206 fixed. Similarly, if a system filter added multiple headers in this way,
2207 the routers could not see them.
2208
220920. Expanded the error message when iplsearch is called with an invalid key to
2210 suggest using net-iplsearch in a host list.
2211
221221. When running tests using -bh, any delays imposed by "delay" modifiers in
2213 ACLs are no longer actually imposed (and a message to that effect is
2214 output).
2215
221622. If a "gecos" field in a passwd entry contained escaped characters, in
2217 particular, if it contained a \" sequence, Exim got it wrong when building
2218 a From: or a Sender: header from that name. A second bug also caused
2219 incorrect handling when an unquoted " was present following a character
2220 that needed quoting.
2221
222223. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
2223 was not being matched caselessly.
2224
222524. Arranged for all hyphens in the exim.8 source to be escaped with
2226 backslashes.
2227
222825. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
2229 properly. Recipient callout cache records were still being keyed to include
2230 the sender, even when use_sender was set false. This led to far more
2231 callouts that were necessary. The sender is no longer included in the key
2232 when use_sender is false.
2233
223426. Added "control = submission" modifier to ACLs.
2235
223627. Added the ${base62d: operator to decode base 62 numbers.
2237
223828. dnsdb lookups can now access SRV records.
2239
224029. CONFIGURE_OWNER can be set at build time to define an alternative owner for
2241 the configuration file.
2242
224330. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
2244 (-v) mode. This makes the output for a verbose queue run more intelligible.
2245
224631. Added a use_postmaster feature to recipient callouts.
2247
224832. Added the $body_zerocount variable, containing the number of binary zero
2249 bytes in the message body.
2250
225133. The time of last modification of the "new" subdirectory is now used as the
2252 "mailbox time last read" when there is a quota error for a maildir
2253 delivery.
2254
225534. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
2256
225735. Added +ignore_unknown as a special item in host lists.
2258
225936. Code for decoding IPv6 addresses in host lists is now included, even if
2260 IPv6 support is not being compiled. This fixes a bug in which an IPv6
2261 address was recognized as an IP address, but was then not correctly decoded
2262 into binary, causing unexpected and incorrect effects when compared with
2263 another IP address.
2264
2265
2266Exim version 4.34
2267-----------------
2268
2269 1. Very minor rewording of debugging text in manualroute to say "list of
2270 hosts" instead of "hostlist".
2271
2272 2. If verify=header_syntax was set, and a header line with an unqualified
2273 address (no domain) and a large number of spaces between the end of the
2274 name and the colon was received, the reception process suffered a buffer
2275 overflow, and (when I tested it) crashed. This was caused by some obsolete
2276 code that should have been removed. The fix is to remove it!
2277
2278 3. When running in the test harness, delay a bit after writing a bounce
2279 message to get a bit more predictability in the log output.
2280
2281 4. Added a call to search_tidyup() just before forking a reception process. In
2282 theory, someone could use a lookup in the expansion of smtp_accept_max_
2283 per_host which, without the tidyup, could leave open a database connection.
2284
2285 5. Added the variables $recipient_data and $sender_data which get set from a
2286 lookup success in an ACL "recipients" or "senders" condition, or a router
2287 "senders" option, similar to $domain_data and $local_part_data.
2288
2289 6. Moved the writing of debug_print from before to after the "senders" test
2290 for routers.
2291
2292 7. Change 4.31/66 (moving the time when the Received: is generated) caused
2293 problems for message scanning, either using a data ACL, or using
2294 local_scan() because the Received: header was not generated till after they
2295 were called (in order to set the time as the time of reception completion).
2296 I have revised the way this works. The header is now generated after the
2297 body is received, but before the ACL or local_scan() are called. After they
2298 are run, the timestamp in the header is updated.
2299
2300
2301Exim version 4.33
2302-----------------
2303
2304 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
2305 before starting a queue runner without re-exec. This happened only when
2306 deliver_drop_privilege was set or when the Exim user was set to root. The
2307 effect of the bug was that timeouts during subsequent deliveries caused
2308 crashes instead of being properly handled. The handler is now left at its
2309 default (and expected) setting.
2310
2311 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
2312 message, again when deliver_drop_privilege is set or Exim is run as root.
2313 The bug described in (1) was not present in this case, but the tidying up
2314 of the other signals was missing. I have made the two cases consistent.
2315
2316 3. The ignore_target_hosts setting on a manualroute router was being ignored
2317 for hosts that were looked up using the /MX notation.
2318
2319 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
2320 in domain lists.
2321
2322 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
2323 operated on the sender address. After changing the $sender_address to <>
2324 for the sender address verify, Exim was re-instated it as the original
2325 (before rewriting) address, but remembering that it had rewritten it, so it
2326 wasn't rewriting it again. This bug also had the effect of breaking the
2327 sender address verification caching when the sender address was rewritten.
2328
2329 6. The ignore_target_hosts option was being ignored by the ipliteral router.
2330 This has been changed so that if the ip literal address matches
2331 ignore_target_hosts, the router declines.
2332
2333 7. Added expansion conditions match_domain, match_address, and match_local_
2334 part (NOT match_host).
2335
2336 8. The placeholder for the Received: header didn't have a length field set.
2337
2338 9. Added code to Exim itself and to exim_lock to test for a specific race
2339 condition that could lead to file corruption when using MBX delivery. The
2340 issue is with the lockfile that is created in /tmp. If this file is removed
2341 after a process has opened it but before that process has acquired a lock,
2342 there is the potential for a second process to recreate the file and also
2343 acquire a lock. This could lead to two Exim processes writing to the file
2344 at the same time. The added code performs the same test as UW imapd; it
2345 checks after acquiring the lock that its file descriptor still refers to
2346 the same named file.
2347
234810. The buffer for building added header lines was of fixed size, 8192 bytes.
2349 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
2350 when Exim is built.
2351
235211. Added the smtp_active_hostname option. If used, this will typically be made
2353 to depend on the incoming interface address. Because $interface_address is
2354 not set up until the daemon has forked a reception process, error responses
2355 that can happen earlier (such as "too many connections") no longer contain
2356 a host name.
2357
235812. If an expansion in a condition on a "warn" statement fails because a lookup
2359 defers, the "warn" statement is abandoned, and the next ACL statement is
2360 processed. Previously this caused the whole ACL to be aborted.
2361
236213. Added the iplsearch lookup type.
2363
236414. Added ident_timeout as a log selector.
2365
236615. Added tls_certificate_verified as a log selector.
2367
236816. Added a global option tls_require_ciphers (compare the smtp transport
2369 option of the same name). This controls incoming TLS connections.
2370
237117. I finally figured out how to make tls_require_ciphers do a similar thing
2372 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
2373 before starting the TLS session.
2374
237518. Tabs are now shown as \t in -bP output.
2376
237719. If the log selector return_path_on_delivery was set, Exim crashed when
2378 bouncing a message because it had too many Received: header lines.
2379
238020. If two routers both had headers_remove settings, and the first one included
2381 a superfluous trailing colon, the final name in the first list and the
2382 first name in the second list were incorrectly joined into one item (with a
2383 colon in the middle).
2384
2385
2386Exim version 4.32
2387-----------------
2388
2389 1. Added -C and -D options to the exinext utility, mainly to make it easier
2390 to include in the automated testing, but these could be helpful when
2391 multiple configurations are in use.
2392
2393 2. The exinext utility was not formatting the output nicely when there was
2394 an alternate port involved in the retry record key, nor when there was a
2395 message id as well (for retries that were specific to a specific message
2396 and a specific host). It was also confused by IPv6 addresses, because of
2397 the additional colons they contain. I have fixed the IPv4 problem, and
2398 patched it up to do a reasonable job for IPv6.
2399
2400 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
2401 delivery, the log line now contains "pipelined" if PIPELINING was used.
2402
2403 4. An SMTP transport process used to panic and die if the bind() call to set
2404 an explicit outgoing interface failed. This has been changed; it is now
2405 treated in the same way as a connect() failure.
2406
2407 5. A reference to $sender_host_name in the part of a conditional expansion
2408 that was being skipped was still causing a DNS lookup. This no longer
2409 occurs.
2410
2411 6. The def: expansion condition was not recognizing references to header lines
2412 that used bh_ and bheader_.
2413
2414 7. Added the _cache feature to named lists.
2415
2416 8. The code for checking quota_filecount in the appendfile transport was
2417 allowing one more file than it should have been.
2418
2419 9. For compatibility with Sendmail, the command line option
2420
2421 -prval:sval
2422
2423 is equivalent to
2424
2425 -oMr rval -oMs sval
2426
2427 and sets the incoming protocol and host name (for trusted callers). The
2428 host name and its colon can be omitted when only the protocol is to be set.
2429 Note the Exim already has two private options, -pd and -ps, that refer to
2430 embedded Perl. It is therefore impossible to set a protocol value of "d" or
2431 "s", but I don't think that's a major issue.
2432
243310. A number of refactoring changes to the code, none of which should affect
2434 Exim's behaviour:
2435
2436 (a) The number of logging options was getting close to filling up the
2437 32-bit word that was used as a bit map. I have split them into two classes:
2438 those that are passed in the argument to log_write(), and those that are
2439 only ever tested independently outside of that function. These are now in
2440 separate 32-bit words, so there is plenty of room for expansion again.
2441 There is no change in the user interface or the logging behaviour.
2442
2443 (b) When building, for example, log lines, the code previously used a
2444 macro that called string_cat() twice, in order to add two strings. This is
2445 not really sufficiently general. Furthermore, there was one instance where
2446 it was actually wrong because one of the argument was used twice, and in
2447 one call a function was used. (As it happened, calling the function twice
2448 did not affect the overall behaviour.) The macro has been replaced by a
2449 function that can join an arbitrary number of extra strings onto a growing
2450 string.
2451
2452 (c) The code for expansion conditions now uses a table and a binary chop
2453 instead of a serial search (which was left over from when there were very
2454 few conditions). Also, it now recognizes conditions like "pam" even when
2455 the relevant support is not compiled in: a suitably worded error message is
2456 given if an attempt is made to use such a condition.
2457
245811. Added ${time_interval:xxxxx}.
2459
246012. A bug was causing one of the ddress fields not to be passed back correctly
2461 from remote delivery subprocesses. The field in question was not being
2462 subsequently used, so this caused to problems in practice.
2463
246413. Added new log selectors queue_time and deliver_time.
2465
246614. Might have fixed a bug in maildirsizefile handling that threw up
2467 "unexpected character" debug warnings, and recalculated the data
2468 unnecessarily. In any case, I expanded the warning message to give more
2469 information.
2470
247115. Added the message "Restricted characters in address" to the statements in
2472 the default ACL that block characters like @ and % in local parts.
2473
247416. Change 71 for release 4.31 proved to be much less benign that I imagined.
2475 Three changes have been made:
2476
2477 (a) There was a serious bug; a negative response to MAIL caused the whole
2478 recipient domain to be cached as invalid, thereby blocking all messages
2479 to all local parts at the same domain, from all senders. This bug has
2480 been fixed. The domain is no longer cached after a negative response to
2481 MAIL if the sender used is not empty.
2482
2483 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
2484 been restored.
2485
2486 (c) A new callout option, "use_sender" has been added for people who want
2487 the modified behaviour.
2488
2489
2490Exim version 4.31
2491-----------------
2492
2493 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
2494 Larry Rosenman.
2495
2496 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
2497 indeed breaks things for older releases.
2498
2499 3. Added additional logging to the case where there is a problem reading data
2500 from a filter that is running in a subprocess using a pipe, in order to
2501 try to track down a specific problem.
2502
2503 4. Testing facility fudge: when running in the test harness and attempting
2504 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
2505 getting "No route to host". Convert this to a timeout.
2506
2507 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
2508 warning.
2509
2510 6. Some OS don't have socklen_t but use size_t instead. This affects the
2511 fifth argument of getsockopt() amongst other things. This is now
2512 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
2513 can be set for individual OS. I have set it for SunOS5, OSF1, and
2514 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
2515 some earlier ones do not.
2516
2517 7. Change 4.30/15 was not doing the test caselessly.
2518
2519 8. The standard form for an IPv6 address literal was being rejected by address
2520 parsing in, for example, MAIL and RCPT commands. An example of this kind of
2521 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
2522 this, as well as the form without the "IPv6" on the front (but only when
2523 address literals are enabled, of course).
2524
2525 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
2526
252710. Exim crashed if a message with an empty sender address specified by -f
2528 encountered a router with an errors_to setting. This could be provoked only
2529 by a command such as
2530
2531 exim -f "" ...
2532
2533 where an empty string was supplied; "<>" did not hit this bug.
2534
253511. Installed PCRE release 4.5.
2536
253712. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
2538 remained set. It is now erased.
2539
254013. exiqgrep wasn't working on MacOS X because it didn't correctly compute
2541 times from message ids (which are base 36 rather than the normal 62).
2542
254314. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
2544 were being counted as actual protocol errors, and logged if the log
2545 selector +smtp_protocol_error was set. One cannot be perfect in this test,
2546 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
2547 and DATA following a set of rejected RCPTs do not count as protocol errors.
2548 In other words, Exim assumes they were pipelined, though this may not
2549 actually be the case. Of course, in all cases the client gets an
2550 appropriate error code.
2551
255215. If a lookup fails in an ACL condition, a message about the failure may
2553 be available; it is used if testing the ACL cannot continue, because most
2554 such messages specify what the cause of the deferral is. However, some
2555 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
2556 that caused an old message to be retained and used if a later statement
2557 caused a defer, replacing the real cause of the deferral.
2558
255916. If an IP address had so many PTR records that the DNS lookup buffer
2560 was not large enough to hold them, Exim could crash while trying to process
2561 the truncated data. It now detects and logs this case.
2562
256317. Further to 4.21/58, another change has been made: if (and only if) the
2564 first line of a message (the first header line) ends with CRLF, a bare LF
2565 in a subsequent header line has a space inserted after it, so as not to
2566 terminate the header.
2567
256818. Refactoring: tidied an ugly bit of code in appendfile that copied data
2569 unnecessarily, used atoi() instead of strtol(), and didn't check the
2570 termination when getting file sizes from file names by regex.
2571
257219. Completely re-implemented the support for maildirsize files, in the light
2573 of a number of problems with the previous contributed implementation
2574 (4.30/29). In particular:
2575
2576 . If the quota is zero, the maildirsize file is maintained, but no quota is
2577 imposed.
2578
2579 . If the maildir directory does not exist, it is created before any attempt
2580 to write a maildirsize file.
2581
2582 . The quota value in the file is just a cache; if the quota is changed in
2583 the transport, the new value overrides.
2584
2585 . A regular expression is available for excluding directories from the
2586 count.
2587
258820. The autoreply transport checks the characters in options that define the
2589 message's headers; it allows continued headers, but it was checking with
2590 isspace() after an embedded newline instead of explicitly looking for a
2591 space or a tab.
2592
259321. If all the "regular" hosts to which an address was routed had passed their
2594 expiry times, and had not reached their retry times, the address was
2595 bounced, even if fallback hosts were defined. Now Exim should go on to try
2596 the fallback hosts.
2597
259822. Increased buffer sizes in the callout code from 1024 to 4096 to match the
2599 equivalent code in the SMTP transport. Some hosts send humungous responses
2600 to HELO/EHLO, more than 1024 it seems.
2601
260223. Refactoring: code in filter.c used (void *) for "any old type" but this
2603 gives compiler warnings in some environments. I've now done it "properly",
2604 using a union.
2605
260624. The replacement for inet_ntoa() that is used with gcc on IRIX systems
2607 (because of problems with the built-in one) was declared to return uschar *
2608 instead of char *, causing compiler failure.
2609
261025. Fixed a file descriptor leak when processing alias/forward files.
2611
261226. Fixed a minor format string issue in dbfn.c.
2613
261427. Typo in exim.c: ("dmbnz" for "dbmnz").
2615
261628. If a filter file refered to $h_xxx or $message_headers, and the headers
2617 contained RFC 2047 "words", Exim's memory could, under certain conditions,
2618 become corrupted.
2619
262029. When a sender address is verified, it is cached, to save repeating the test
2621 when there is more than one recipient in a message. However, when the
2622 verification involves a callout, it is possible for different callout
2623 options to be set for different recipients. It is too complicated to keep
2624 track of this in the cache, so now Exim always runs a verification when a
2625 callout is required, relying on the callout cache for the optimization.
2626 The overhead is duplication of the address routing, but this should not be
2627 too great.
2628
262930. Fixed a bug in callout caching. If a RCPT command caused the sender address
2630 to be verified with callout=postmaster, and the main callout worked but the
2631 postmaster check failed, the verification correctly failed. However, if a
2632 subsequent RCPT command asked for sender verification *without* the
2633 postmaster check, incorrect caching caused this verification also to fail,
2634 incorrectly.
2635
263631. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
2637 it was not caching the DNS options (qualify_single, search_parents) that
2638 were used when the lookup failed. A subsequent lookup with different
2639 options therefore always gave the same answer, though there were cases
2640 where it should not have. (Example: a "domains = !$mx_any" option on a
2641 dnslookup router: the "domains" option is always processed without any
2642 widening, but the router might have qualify_single set.) Now Exim uses the
2643 cached value only when the same options are set.
2644
264532. Added John Jetmore's "exipick" utility to the distribution.
2646
264733. GnuTLS: When an attempt to start a TLS session fails for any reason other
2648 than a timeout (e.g. a certificate is required, and is not provided), an
2649 Exim server now closes the connection immediately. Previously it waited for
2650 the client to close - but if the client is SSL, it seems that they each
2651 wait for each other, leading to a delay before one of them times out.
2652
265334: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
2654 maintained 0.8.x compatibility because I don't think many are using it, and
2655 it is clearly obsolete.
2656
265735. Added TLS support for CRLs: a tls_crl global option and one for the smtp
2658 transport.
2659
266036. OpenSSL: $tls_certificate_verified was being set to 1 even if the
2661 client certificate was expired. A simple patch fixes this, though I don't
2662 understand the full logic of why the verify callback is called multiple
2663 times.
2664
266537. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
2666 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
2667 which causes problems with some clients (such as the Certicom SSL Plus
2668 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
2669 disables the coutermeasure allowing Eudora to connect."
2670
267138. Exim was not checking that a write() to a log file succeeded. This could
2672 lead to Bad Things if a log got too big, in particular if it hit a file
2673 size limit. Exim now panics and dies if it cannot write to a log file, just
2674 as it does if it cannot open a log file.
2675
267639. Modified OS/Makefile-Linux so that it now contains
2677
2678 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
2679
2680 The two -D definitions ensure that Exim is compiled with large file
2681 support, which makes it possible to handle log files that are bigger than
2682 2^31.
2683
268440. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
2685 instance) a domain was checked against a named list that involved a lookup,
2686 causing $domain_data to be set, then another domain was checked against the
2687 same list, then the first domain was re-checked, the value of $domain_data
2688 after the final check could be wrong. In particular, if the second check
2689 failed, it could be set empty. This bug probably also applied to
2690 $localpart_data.
2691
269241. The strip_trailing_dot option was not being applied to the address given
2693 with the -f command-line option.
2694
269542. The code for reading a message's header from the spool was incrementing
2696 $received_count, but never initializing it. This meant that the value was
2697 incorrect (doubled) while delivering a message in the same process in which
2698 it was received. In the most common configuration of Exim, this never
2699 happens - a fresh exec is done - but it can happen when
2700 deliver_drop_privilege is set.
2701
270243. When Exim logs an SMTP synchronization error - client data sent too soon -
2703 it now includes up to 150 characters of the unexpected data in the log
2704 line.
2705
270644. The exim_dbmbuild utility uses fixed size buffers for reading input lines
2707 and building data strings. The size of both of these buffers was 10 000
2708 bytes - far larger than anybody would *ever* want, thought I. Needless to
2709 say, somebody hit the limit. I have increased the maximum line length to
2710 20 000 and the maximum data length of concatenated lines to 100 000. I have
2711 also fixed two bugs, because there was no checking on these buffers. Tsk,
2712 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
2713 buffer is too small.
2714
271545. The exim_dbmbuild utility did not support quoted keys, as Exim does in
2716 lsearch lookups. Now it does.
2717
271846. When parsing a route_list item in a manualroute router, a fixed-length
2719 buffer was used for the list of hosts. I made this 1024 bytes long,
2720 thinking that nobody would ever have a list of hosts that long. Wrong.
2721 Somebody had a whole pile of complicated expansion conditions, and the
2722 string was silently truncated, leading to an expansion error. It turns out
2723 that it is easier to change to an unlimited length (owing to other changes
2724 that have happened since this code was originally written) than to build
2725 structure for giving a limitation error. The length of the item that
2726 expands into the list of hosts is now unlimited.
2727
272847. The lsearch lookup could not handle data where the length of text line was
2729 more than 4095 characters. Such lines were truncated, leading to shortened
2730 data being returned. It should now handle lines of any length.
2731
273248. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
2733 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
2734 ACL").
2735
273649. Cosmetic tidy to scripts like exicyclog that are generated by globally
2737 replacing strings such as BIN_DIRECTORY in a source file: the replacement
2738 no longer happens in comment lines. A list of replacements is now placed
2739 at the head of all of the source files, except those whose only change is
2740 to replace PERL_COMMAND in the very first #! line.
2741
274250. Replaced the slow insertion sort in queue.c, for sorting the list of
2743 messages on the queue, with a bottom-up merge sort, using code contributed
2744 by Michael Haardt. This should make operations like -bp somewhat faster on
2745 large queues. It won't affect queue runners, except when queue_run_in_order
2746 is set.
2747
274851. Installed eximstats 1.31 in the distribution.
2749
275052. Added support for SRV lookups to the dnslookup router.
2751
275253. If an ACL referred to $message_body or $message_body_end, the value was not
2753 reset for any messages that followed in the same SMTP session.
2754
275554. The store-handling optimization for building very long strings was not
2756 differentiating between the different store pools. I don't think this
2757 actually made any difference in practice, but I've tidied it.
2758
275955. While running the routers to verify a sender address, $sender_address
2760 was still set to the sender address. This is wrong, because when routing to
2761 send a bounce to the sender, it would be empty. Therefore, I have changed
2762 it so that, while verifying a sender address, $sender_address is set to <>.
2763 (There is no change to what happens when verifying a recipient address.)
2764
276556. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
2766 target A or AAAA records (if not already returned) without resetting the
2767 qualify_single or search_parents options of the DNS resolver. These are
2768 inappropriate in this case because the targets of MX and SRV records must
2769 be FQDNs. A broken DNS record could cause trouble if it happened to have a
2770 target that, when qualified, matched something in the local domain. These
2771 two options are now turned off when doing these lookups.
2772
277357. It seems that at least some releases of Reiserfs (which does not have the
2774 concept of a fixed number of inodes) returns zero and not -1 for the
2775 number of available inodes. This interacted badly with check_spool_inodes,
2776 which assumed that -1 was the "no such thing" setting. What I have done is
2777 to check that the total number of inodes is greater than zero before doing
2778 the test of how many are available.
2779
278058. When a "warn" ACL statement has a log_message modifier, the message is
2781 remembered, and not repeated. This is to avoid a lot of repetition when a
2782 message has many recipients that cause the same warning to be written.
2783 Howewer, Exim was preserving the list of already written lines for an
2784 entire SMTP session, which doesn't seem right. The memory is now reset if a
2785 new message is started.
2786
278759. The "rewrite" debugging flag was not showing the result of rewriting in the
2788 debugging output unless log_rewrite was also set.
2789
279060. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
2791 of (int)(handle) when we know that handle contains (void *)(-1).
2792
279361. The Exim daemon panic-logs an error return when it closes the incoming
2794 connection. However "connection reset by peer" seems to be common, and
2795 isn't really an error worthy of noting specially, so that particular error
2796 is no long logged.
2797
279862. When Exim is trying to find all the local interfaces, it used to panic and
2799 die if the ioctl to get the interface flags failed. However, it seems that
2800 on at least one OS (Solaris 9) it is possible to have an interface that is
2801 included in the list of interfaces, but for which you get a failure error
2802 for this call. This happens when the interface is not "plumbed" into a
2803 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
2804 failure of the "get flags" call assumes that the interface is down.
2805
280663. Added a ${eval10: operator, which assumes all numbers are decimal. This
2807 makes life easier for people who are doing arithmetic on fields extracted
2808 from dates, where you often get leading zeros that should not be
2809 interpreted as octal.
2810
281164. Added qualify_domain to the redirect router, to override the global
2812 setting.
2813
281465. If a pathologically long header line contained very many addresses (the
2815 report of this problem mentioned 10 000) and each of them was rewritten,
2816 Exim could use up a very large amount of memory. (It kept on making new
2817 copies of the header line as it rewrote, and never released the old ones.)
2818 At the expense of a bit more processing, the header rewriting function has
2819 been changed so that it no longer eats memory in this way.
2820
282166. The generation of the Received: header has been moved from the time that a
2822 message starts to be received, to the time that it finishes. The timestamp
2823 in the Received: header should now be very close to that of the <= log
2824 line. There are two side-effects of this change:
2825
2826 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
2827 logged header lines no longer include the local Received: line, because
2828 it has not yet been created. The same applies to a copy of the message
2829 that is returned to a non-SMTP sender when a message is rejected.
2830
2831 (b) When a filter file is tested using -bf, no additional Received: header
2832 is added to the test message. After some thought, I decided that this
2833 is a bug fix.
2834
2835 This change does not affect the value of $received_for. It is still set
2836 after address rewriting, but before local_scan() is called.
2837
283867. Installed the latest Cygwin-specific files from the Cygwin maintainer.
2839
284068. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
2841 gave an unhelpful panic error message, and a defer error. I have managed to
2842 change this behaviour so that it now rejects any supplied certificate,
2843 which seems right, as the list of acceptable certificates is empty.
2844
284569. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
2846 gave an unhelpful defer error. I have not managed to make this reject any
2847 supplied certificates, but the error message it gives is "no certificate
2848 supplied", which is not helpful.
2849
285070. exigrep's output now also includes lines that are not associated with any
2851 message, but which match the given pattern. Implemented by a patch from
2852 Martin Sluka, which also tidied up the Perl a bit.
2853
285471. Recipient callout verification, like sender verification, was using <> in
2855 the MAIL FROM command. This isn't really the right thing, since the actual
2856 sender may affect whether the remote host accepts the recipient or not. I
2857 have changed it to use the actual sender in the callout; this means that
2858 the cache record is now keyed on a recipient/sender pair, not just the
2859 recipient address. There doesn't seem to be a real danger of callout loops,
2860 since a callout by the remote host to check the sender would use <>.
2861 [SEE ABOVE: changed after hitting problems.]
2862
286372. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
2864 temporary errors. However, in the case of such a code being given after
2865 the end of a data transmission (i.e. after ".") Exim was failing to write
2866 a retry record for the message. (Yes, there was some broken host that was
2867 actually sending 8xx at this point.)
2868
286973. An unknown lookup type in a host list could cause Exim to panic-die when
2870 the list was checked. (An example that provoked this was putting <; in the
2871 middle of a list instead of at the start.) If this happened during a DATA
2872 ACL check, a -D file could be left lying around. This kind of configuration
2873 error no longer causes Exim to die; instead it causes a defer errror. The
2874 incident is still logged to the main and panic logs.
2875
287674. Buglet left over from Exim 3 conversion. The message "too many messages
2877 in one connection" was written to the rejectlog but not the mainlog, except
2878 when address rewriting (yes!) was being logged.
2879
288075. Added write_rejectlog option.
2881
288276. When a system filter was run not as root (that is, when system_filter_user
2883 was set), the values of the $n variables were not being returned to the
2884 main process; thus, they were not subsequently available in the $sn
2885 variables.
2886
288777. Added +return_path_on_delivery log selector.
2888
288978. A connection timeout was being treated differently from recipients deferred
2890 when testing hosts_max_try with a message that was older than the host's
2891 retry timeout. (The host should not be counted, thus allowing all hosts to
2892 be tried at least once before bouncing.) This may have been the cause of an
2893 occasionally reported bug whereby a message would remain on the queue
2894 longer than the retry timeout, but would be bounced if a delivery was
2895 forced. I say "may" because I never totally pinned down the problem;
2896 setting up timeout/retry tests is difficult. See also the next item.
2897
289879. The ultimate address timeout was not being applied to errors that involved
2899 a combination of host plus message (for example, a timeout on a MAIL
2900 command). When an address resolved to a number of possible hosts, and they
2901 were not all tried for each delivery (e.g. because of hosts_max_try), a
2902 message could remain on the queue longer than the retry timeout.
2903
290480. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
2905 Haardt.
2906
290781. Fixed an obscure SMTP outgoing bug which required at least the following
2908 conditions: (a) there was another message waiting for the same server;
2909 (b) the server returned 5xx to all RCPT commands in the first message so
2910 that the message was not completed; (c) the server dropped the connection
2911 or gave a negative response to the RSET that Exim sends to abort the
2912 transaction. The observed case was a dropped connection after DATA that had
2913 been sent in pipelining mode. That is, the server had advertised PIPELINING
2914 but was not implementing it correctly. The effect of the bug was incorrect
2915 behaviour, such as trying another host, and this could lead to a crash.
2916
2917
2918Exim version 4.30
2919-----------------
2920
2921 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
2922 and daemon.c were passed as pointers to ints; they should have been
2923 pointers to socklen_t variables (which are typically unsigned ints).
2924
2925 2. Some signed/unsigned type warnings in the os.c file for Linux have been
2926 fixed.
2927
2928 3. Fixed a really odd bug that affected only the testing scheme; patching a
2929 certain fixed string in the binary changed the value of another string that
2930 happened to be identical to the end of the original first string.
2931
2932 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
2933 name", it returns that address as the IP address. On some operating
2934 systems (e.g. Solaris), it also passes back the IP address string as the
2935 "host name". However, on others (e.g. Linux), it passes back an empty
2936 string. Exim wasn't checking for this, and was changing the host name to an
2937 empty string, assuming it had been canonicized.
2938
2939 5. Although rare, it is permitted to have more than one PTR record for a given
2940 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
2941 all the names associated with an address, because they do in Solaris.
2942 However, it seems that they do not in Linux for data that comes from the
2943 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
2944 I found this out when I moved to a new Linux workstation and tried to run
2945 the Exim test suite.
2946
2947 To get round this problem I have changed the code so that it now does its
2948 own call to the DNS to look up PTR records when searching for a host name.
2949 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
2950 addresses that are only in /etc/hosts are still found.
2951
2952 This behaviour is, however, controlled by an option called host_lookup_
2953 order, which defaults to "bydns:byaddr". If people want to use the other
2954 order, or indeed, just use one or the other means of lookup, they can
2955 specify it in this variable.
2956
2957 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
2958 some operating systems, this comes back from gethostbyaddr() as an empty
2959 string, and this is what Exim used to test for. However, it seems that in
2960 other systems, "." is yielded. Exim now tests for this case too.
2961
2962 7. The values of check_spool_space and check_log_space are now held internally
2963 as a number of kilobytes instead of an absolute number of bytes. If a
2964 numbers is specified without 'K' or 'M', it is rounded up to the nearest
2965 kilobyte. This means that much larger values can be stored.
2966
2967 8. Exim monitor: an attempt to get the action menu when not actually pointing
2968 at a message produces an empty menu entitled "No message selected". This
2969 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
2970 no entries in it ("Shell widget menu has zero width and/or height"). So I
2971 have added a single, blank menu entry in this case.
2972
2973 9. Added ${quote_local_part.
2974
297510. MIME decoding is now applied to the contents of Subject: header lines when
2976 they are logged.
2977
297811. Now that a reference to $sender_host_address automatically causes a reverse
2979 lookup to occur if necessary (4.13/18), there is no need to arrange for a
2980 host lookup before query-style lookups in lists that might use this
2981 variable. This has therefore been abolished, and the "net-" prefix is no
2982 longer necessary for query-style lookups.
2983
298412. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
2985 have been a typo for LFLAGS, so it has been changed.
2986
298713. The install script calls Exim with "-C /dev/null" in order to find the
2988 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
2989 to be output. Howeve, since Exim outputs its version number before the
2990 error, it didn't break the script. It just looked ugly. I fixed this by
2991 always allowing "-C /dev/null" if the caller is root.
2992
299314. Ignore overlarge ACL variable number when reading spool file - insurance
2994 against a later release with more variables having written the file.
2995
299615. The standard form for an IPv6 address literal was being rejected by EHLO.
2997 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
2998 this, as well as the form without the "IPv6" on the front.
2999
300016. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
3001 OS/Makefile-Darwin file.
3002
300317. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
3004 only to LDAP libraries that do not have LDAP_OPT_DEREF.
3005
300618. After change 4.21/52, "%ld" was used to format the contents of the $inode
3007 variable. However, some OS use ints for inodes. I've added cast to long int
3008 to get rid of the compiler warning.
3009
301019. I had forgotten to lock out "/../" in configuration file names when
3011 ALT_CONFIG_PREFIX was set.
3012
301320. Routers used for verification do not need to specify transports. However,
3014 if such a router generated a host list, and callout was configured, Exim
3015 crashed, because it could not find a port number from the (non-existent)
3016 transport. It now assumes port 25 in this circumstance.
3017
301821. Added the -t option to exigrep.
3019
302022. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
3021 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
3022 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
3023 them is set, LOOKUP_LSEARCH is forced.
3024
302523. "exim -bV" now outputs a list of lookups that are included in the binary.
3026
302724. Added sender and host information to the "rejected by local_scan()" log
3028 line; previously there was no indication of these.
3029
303025. Added .include_if_exists.
3031
303226. Change 3.952/11 added an explicit directory sync on top of a file sync for
3033 Linux. It turns out that not all file systems support this. Apparently some
3034 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
3035 do it.) To cope with this, the error EINVAL, which means that sync-ing is
3036 not supported on the file descriptor, is now ignored when Exim is trying to
3037 sync a directory. This applies only to Linux.
3038
303927. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
3040
304128. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
3042 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
3043 this.
3044
304529. Added support for maildirsize files from supplied patch (modified a bit).
3046
304730. The use of :fail: followed by an empty string could lead Exim to respond to
3048 sender verification failures with (e.g.):
3049
3050 550 Verification failed for <xxx>
3051 550 Sender verify failed
3052
3053 where the first response line was missing the '-' that indicates it is not
3054 the final line of the response.
3055
305631. The loop for finding the name of the user that called Exim had a hardwired
3057 limit of 10; it now uses the value of finduser_retries, which is used for
3058 all other user lookups.
3059
306032. Added $received_count variable, available in data and not_smtp ACLs, and at
3061 delivery time.
3062
306333. Exim was neglecting to zero errno before one call of strtol() when
3064 expanding a string and expecting an integer value. On some systems this
3065 resulted in spurious "integer overflow" errors. Also, it was casting the
3066 result into an int without checking.
3067
306834. Testing for a connection timeout using "timeout_connect" in the retry rules
3069 did not work. The code looks as if it has *never* worked, though it appears
3070 to have been documented since at least releast 1.62. I have made it work.
3071
307235. The "timeout_DNS" error in retry rules, also documented since at least
3073 1.62, also never worked. As it isn't clear exactly what this means, and
3074 clearly it isn't a major issue, I have abolished the feature by treating it
3075 as "timeout", and writing a warning to the main and panic logs.
3076
307736. The display of retry rules for -brt wasn't always showing the error code
3078 correctly.
3079
308037. Added new error conditions to retry rules: timeout_A, timeout_MX,
3081 timeout_connect_A, timeout_connect_MX.
3082
308338. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
3084 to the empty sender.
3085
308639. The daemon was not analysing the content of -oX till after it had closed
3087 stderr and disconnected from the controlling terminal. This meant that any
3088 syntax errors were only noted on the panic log, and the return code from
3089 the command was 0. By re-arranging the code a little, I've made the
3090 decoding happen first, so such errors now appear on stderr, and the return
3091 code is 1. However, the actual setting up of the sockets still happens in
3092 the disconnected process, so errors there are still only recorded on the
3093 panic log.
3094
309540. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
3096 connections (as happens on some IP stacks) was logged at start up time as
3097 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
3098 it from "IPv6 and IPv4", which means that two separate sockets are being
3099 used.
3100
310141. The debug output for gethostbyname2() or getipnodebyname() failures now
3102 says whether AF_INET or AF_INET6 was passed as an argument.
3103
310442. Exiwhat output was messed up when time zones were included in log
3105 timestamps.
3106
310743. Exiwhat now gives more information about the daemon's listening ports,
3108 and whether -tls-on-connect was used.
3109
311044. The "port" option of the smtp transport is now expanded.
3111
311245. A "message" modifier in a "warn" statement in a non-message ACL was being
3113 silently ignored. Now an error message is written to the main and panic
3114 logs.
3115
311646. There's a new ACL modifier called "logwrite" which writes to a log file
3117 as soon as it is encountered.
3118
311947. Added $local_user_uid and $local_user_gid at routing time.
3120
312148. Exim crashed when trying to verify a sender address that was being
3122 rewritten to "<>".
3123
312449. Exim was recognizing only a space character after ".include". It now also
3125 recognizes a tab character.
3126
312750. Fixed several bugs in the Perl script that creates the exim.8 man page by
3128 extracting the relevant information from the specification. The man page no
3129 longer contains scrambled data for the -d option, and I've added a section
3130 at the front about calling Exim under different names.
3131
313251. Added "extra_headers" argument to the "mail" command in filter files.
3133
313452. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
3135 crash.
3136
313753. Installed eximstats 1.29.
3138
313954. Added transport_filter_timeout as a generic transport option.
3140
314155. Exim no longer adds an empty Bcc: header to messages that have no To: or
3142 Cc: header lines. This was required by RFC 822, but it not required by RFC
3143 2822.
3144
314556. Exim used to add From:, Date:, and Message-Id: header lines to any
3146 incoming messages that did not have them. Now it does so only if the
3147 message originates locally, that is, if there is no associated remote host
3148 address. When Resent- header lines are present, this applies to the Resent-
3149 lines rather than the non-Resent- lines.
3150
315157. Drop incoming SMTP connection after too many syntax or protocol errors. The
3152 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
3153
315458. Messages for configuration errors now include the name of the main
3155 configuration file - useful now that there may be more than one file in a
3156 list (.included file names were always shown).
3157
315859. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
3159 for those rare installations that do not start the daemon as root or run it
3160 setuid root. I've cut out the call to initgroups() if the daemon is not
3161 root at that time.
3162
316360. The Exim user and group can now be bound into the binary as text strings
3164 that are looked up at the start of Exim's processing.
3165
316661. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
3167
316862. Added $mailstore_basename variable.
3169
317063. Installed patch to sieve.c from Michael Haardt.
3171
317264. When Exim failed to open the panic log after failing to open the main log,
3173 the original message it was trying to log was written to stderr and debug
3174 output, but if they were not available (the usual case in production), it
3175 was lost. Now it is written to syslog before the two lines that record the
3176 failures to open the logs.
3177
317865. Users' Exim filters run in subprocesses under the user's uid. It is
3179 possible for a "deliver" command or an alias in a "personal" command to
3180 provoke an address rewrite. If logging of address rewriting is configured,
3181 this fails because the process is not running as root or exim. There may be
3182 a better way of dealing with this, but for the moment (because 4.30 needs
3183 to be released), I have disabled address rewrite logging when running a
3184 filter in a non-root, non-exim process.
3185
3186
3187Exim version 4.24
3188-----------------
3189
3190 1. The buildconfig auxiliary program wasn't quoting the value set for
3191 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
3192 not defined. This bug was masked in 4.22 by the effect that was fixed in
3193 change 4.23/1.
3194
3195 2. Some messages that were rejected after a message id was allocated were
3196 shown as "incomplete" by exigrep. It no longer does this for messages that
3197 are rejected by local_scan() or the DATA or non-SMTP ACLs.
3198
3199 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
3200 have allow_domain_literals set, the ID did not get logged in the <= line.
3201 Domain literals are now always recognized in Message-ID: header lines.
3202
3203 4. The first argument for a ${extract expansion item is the key name or field
3204 number. Leading and trailing spaces in this item were not being ignored,
3205 causing some misleading effects.
3206
3207 5. When deliver_drop_privilege was set, single queue runner processes started
3208 manually (i.e. by the command "exim -q") or by the daemon (which uses the
3209 same command in the process it spins off) were not dropping privilege.
3210
3211 6. When the daemon running as "exim" started a queue runner, it always
3212 re-executed Exim in the spun-off process. This is a waste of effort when
3213 deliver_drop_privilege is set. The new process now just calls the
3214 queue-runner function directly.
3215
3216
3217Exim version 4.23
3218-----------------
3219
3220 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
3221 HEADERS_CHARSET.
3222
3223 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
3224 ignored. Though the use of -oP was forcing the writing of a pid file, it
3225 was always written to the default place.
3226
3227 3. If the message "no IP address found for host xxxx" is generated during
3228 incoming verification, it is now followed by identification of the incoming
3229 connection (so you can more easily find what provoked it).
3230
3231 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
3232
3233 5. Added some features to "harden" Exim a bit more against certain attacks:
3234
3235 (a) There is now a build-time option called FIXED_NEVER_USERS that can
3236 be put in Local/Makefile. This is like the never_users runtime option,
3237 but it cannot be overridden. The default setting is "root".
3238
3239 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
3240 prefix string with which any file named in a -C command line option
3241 must start.
3242
3243 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
3244 is retained for -C and -D only if the caller of Exim is root. Without
3245 it, the exim user may also use -C and -D and retain privilege.
3246
3247 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
3248 command line option is disabled.
3249
3250 6. Macro names set by the -D option must start with an upper case letter, just
3251 like macro names defined in the configuration file.
3252
3253 7. Added "dereference=" facility to LDAP.
3254
3255 8. Two instances of the typo "uknown" in the source files are fixed.
3256
3257 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
3258 the Configure-Makefile script screwed up while processing it.
3259
326010. Incorporated PCRE 4.4.
3261
326211. The SMTP synchronization check was not operating right at the start of an
3263 SMTP session. For example, it could not catch a HELO sent before the client
3264 waited for the greeting. There is now a check for outstanding input at the
3265 point when the greeting is written. Because of the duplex, asynchronous
3266 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
3267 way, but not yet received, when the check is performed.
3268
326912. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
3270 on TCP/IP sockets, because this apparently causes some broken clients to
3271 timeout.
3272
327313. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
3274 unchanged) from the Cygwin maintainer.
3275
327614. The code for -bV that shows what is in the binary showed "mbx" when maildir
3277 was supported instead of testing for mbx. Effectively a typo.
3278
327915. The spa authenticator server code was not checking that the input it
3280 received was valid base64.
3281
328216. The debug output line for the "set" modifier in ACLs was not showing the
3283 name of the variable that was being set.
3284
328517. Code tidy: the variable type "vtype_string" was never used. Removed it.
3286
328718. Previously, a reference to $sender_host_name did not cause a DNS reverse
3288 lookup on its own. Something else was needed to trigger the lookup. For
3289 example, a match in host_lookup or the need for a host name in a host list.
3290 Now, if $sender_host_name is referenced and the host name has not yet been
3291 looked up, a lookup is performed. If the lookup fails, the variable remains
3292 empty, and $host_lookup_failed is set to "1".
3293
329419. Added "eqi" as a case-independent comparison operator.
3295
329620. The saslauthd authentication condition could segfault if neither service
3297 nor realm was specified.
3298
329921. If an overflowing value such as "2048M" was set for message_size_limit, the
3300 error message that was logged was misleading, and incoming SMTP
3301 connections were dropped. The message is now more accurate, and temporary
3302 errors are given to SMTP connections.
3303
330422. In some error situations (such as 21 above) Exim rejects all SMTP commands
3305 (except RSET) with a 421 error, until QUIT is received. However, it was
3306 failing to send a response to QUIT.
3307
330823. The HELO ACL was being run before the code for helo_try_verify_hosts,
3309 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
3310 ACL is now run after the helo_try_verify_hosts code.
3311
331224. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
3313 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
3314 case-independent, so other case variants are also recognized). Apparently
3315 some systems use these upper case variants.
3316
331725. If more than two messages were waiting for the same host, and a transport
3318 filter was specified for the transport, Exim sent two messages over the
3319 same TCP/IP connection, and then failed with "socket operation on non-
3320 socket" when it tried to send the third.
3321
332226. Added Exim::debug_write and Exim::log_write for embedded Perl use.
3323
332427. The extern definition of crypt16() in expand.c was not being excluded when
3325 the OS had its own crypt16() function.
3326
332728. Added bounce_return_body as a new option, and bounce_return_size_limit
3328 as a preferred synonym for return_size_limit, both as an option and as an
3329 expansion variable.
3330
333129. Added LIBS=-liconv to OS/Makefile-OSF1.
3332
333330. Changed the default configuration ACL to relax the local part checking rule
3334 for addresses that are not in any local domains. For these addresses,
3335 slashes and pipe symbols are allowed within local parts, but the sequence
3336 /../ is explicitly forbidden.
3337
333831. SPA server authentication was not clearing the challenge buffer before
3339 using it.
3340
334132. log_message in a "warn" ACL statement was writing to the reject log as
3342 well as to the main log, which contradicts the documentation and doesn't
3343 seem right (because no rejection is happening). So I have stopped it.
3344
334533. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
3346 However, I am unable to do any testing of this.
3347
334834. Fixed an infelicity in the appendfile transport. When checking directories
3349 for a mailbox, to see if any needed to be created, it was accidentally
3350 using path names with one or more superfluous leading slashes; tracing
3351 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
3352
335335. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
3354 contents are added to the log line that is written for every discarded
3355 recipient. (Previously a log_message setting was ignored.)
3356
335736. The ${quote: operator now quotes the string if it is empty.
3358
335937. The install script runs exim in order to find its version number. If for
3360 some reason other than non-existence or emptiness, which it checks, it
3361 could not run './exim', it was installing it with an empty version number,
3362 i.e. as "exim-". This error state is now caught, and the installation is
3363 aborted.
3364
336538. An argument was missing from the function that creates an error message
3366 when Exim fails to connect to the socket for saslauthd authentication.
3367 This could cause Exim to crash, or give a corrupted message.
3368
336939. Added isip, isip4, and isip6 to ${if conditions.
3370
337140. The ACL variables $acl_xx are now saved with the message, and can be
3372 accessed later in routers, transports, and filters.
3373
337441. The new lookup type nwildlsearch is like wildlsearch, except that the key
3375 strings in the file are not string-expanded.
3376
337742. If a MAIL command specified a SIZE value that was too large to fit into an
3378 int variable, the check against message_size_limit failed. Such values are
3379 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
3380 day this will have to be increased, but I don't think I want to be around
3381 when emails are that large.
3382
3383
3384
3385Exim version 4.22
3386-----------------
3387
3388 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
3389 iconv() is not standard in FreeBSD.
3390
3391 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
3392 IPv6 enabled. The observed symptom was a segmentation fault on return from
3393 the function os_common_find_running_interfaces() in src/os.c.
3394
3395 3. In the check_special_case() function in daemon.c I had used "errno" as an
3396 argument name, which causes warnings on some systems. This was basically a
3397 typo, since it was named "eno" in the comments!
3398
3399 4. The code that waits for the clock to tick (at a resolution of some fraction
3400 of a second) so as to ensure message-id uniqueness was always waiting for
3401 at least one whole tick, when it could have waited for less. [This is
3402 almost certainly not relevant at current processor speeds, where it is
3403 unlikely to ever wait at all. But we try to future-proof.]
3404
3405 5. The function that sleeps for a time interval that includes fractions of a
3406 second contained a race. It did not block SIGALRM between setting the
3407 timer, and suspending (a couple of lines later). If the interval was short
3408 and the sigsuspend() was delayed until after it had expired, the suspension
3409 never ended. On busy systems this could lead to processes getting stuck for
3410 ever.
3411
3412 6. Some uncommon configurations may cause a lookup to happen in a queue runner
3413 process, before it forks any delivery processes. The open lookup caching
3414 mechanism meant that the open file or database connection was passed into
3415 the delivery process. The problem was that delivery processes always tidy
3416 up cached lookup data. This could cause a problem for the next delivery
3417 process started by the queue runner, because the external queue runner
3418 process does not know about the closure. So the next delivery process
3419 still has data in the lookup cache. In the case of a file lookup, there was
3420 no problem because closing a file descriptor in a subprocess doesn't affect
3421 the parent. However, if the lookup was caching a connection to a database,
3422 the connection was closed, and the second delivery process was likely to
3423 see errors such as "PGSQL: query failed: server closed the connection
3424 unexpectedly". The problem has been fixed by closing all cached lookups
3425 in a queue runner before running a delivery process.
3426
3427 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
3428 seem to have the "const" qualifier which it has on other OS. I've
3429 parameterised it.
3430
3431 8. Change 4.21/2 was too strict. It is only if there are two authenticators
3432 *of the same type* (client or server) with the same public name that an
3433 error should be diagnosed.
3434
3435 9. When Exim looked up a host name for an IP address, but failed to find the
3436 original IP address when looking up the host name (a safety check), it
3437 output the message "<ip address> does not match any IP for NULL", which was
3438 confusing, to say the least. The bug was that the host name should have
3439 appeared instead of "NULL".
3440
344110. Since release 3.03, if Exim is called by a uid other than root or the Exim
3442 user that is built into the binary, and the -C or -D options is used, root
3443 privilege is dropped before the configuration file is read. In addition,
3444 logging is switched to stderr instead of the normal log files. If the
3445 configuration then re-defines the Exim user, the unprivileged environment
3446 is probably not what is expected, so Exim logs a panic warning message (but
3447 proceeds).
3448
3449 However, if deliver_drop_privilege is set, the unprivileged state may well
3450 be exactly what is intended, so the warning has been cut out in that case,
3451 and Exim is allowed to try to write to its normal log files.
3452
3453
3454Exim version 4.21
3455-----------------
3456
3457 1. smtp_return_error_details was not giving details for temporary sender
3458 or receiver verification errors.
3459
3460 2. Diagnose a configuration error if two authenticators have the same public
3461 name.
3462
3463 3. Exim used not to create the message log file for a message until the first
3464 delivery attempt. This could be confusing when incoming messages were held
3465 for policy or load reasons. The message log file is now created at the time
3466 the message is received, and an initial "Received" line is written to it.
3467
3468 4. The automatically generated man page for command line options had a minor
3469 bug that caused no ill effects; however, a more serious problem was that
3470 the procedure for building the man page automatically didn't always
3471 operate. Consequently, release 4.20 contains an out-of-date version. This
3472 shouldn't happen again.
3473
3474 5. When building Exim with embedded Perl support, the script that builds the
3475 Makefile was calling 'perl' to find its compile-time parameters, ignoring
3476 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
3477
3478 6. The freeze_tell option was not being used for messages that were frozen on
3479 arrival, either by an ACL or by local_scan().
3480
3481 7. Added the smtp_incomplete_transaction log selector.
3482
3483 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
3484 was accepting AUTH without a new EHLO.
3485
3486 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
3487 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
3488 RFC.
3489
349010. Logging of TCP/IP connections (when configured) now happens in the main
3491 daemon process instead of the child process, so that the TCP/IP connection
3492 count is more accurate (but it can never be perfect).
3493
349411. The use of "drop" in a nested ACL was not being handled correctly in the
3495 outer ACL. Now, if condition failure induced by the nested "drop" causes
3496 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
3497 or "require"), the connection is dropped.
3498
349912. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
3500 that yield "discard" can now be used with an "accept" or a "discard" verb,
3501 but an error is generated for any others (because I can't see a useful way
3502 to define what should happen).
3503
350413. When an ACL is read dynamically from a file (or anywhere else), the lines
3505 are now processed in the same way as lines in the Exim configuration file.
3506 In particular, continuation lines are supported.
3507
350814. Added the "dnslists = a.b.c!=n.n.n.n" feature.
3509
351015. Added -ti meaning -t -i.
3511
351216. Check for letters, digits, hyphens, and dots in the names of dnslist
3513 domains, and warn by logging if others are found.
3514
351517. At least on BSD, alignment is not guarenteed for the array of ifreq's
3516 returned from GIFCONF when Exim is trying to find the list of interfaces on
3517 a host. The code in os.c has been modified to copy each ifreq to an aligned
3518 structure in all cases.
3519
3520 Also, in some cases, the returned ifreq's were being copied to a 'struct
3521 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
3522 means the last couple of bytes of an IPv6 address could be chopped if the
3523 ifreq contained only a normal sockaddr (14 bytes storage).
3524
352518. Named domain lists were not supported in the hosts_treat_as_local option.
3526 An entry such as +xxxx was not recognized, and was treated as a literal
3527 domain name.
3528
352919. Ensure that header lines added by a DATA ACL are included in the reject log
3530 if the ACL subsequently rejects the message.
3531
353220. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
3533 MD5 (which is deprecated).
3534
353521. When testing a filter file using -bf, Exim was writing a message when it
3536 took the sender from a "From " line in the message, but it was not doing so
3537 when it took $return_path from a Return-Path: header line. It now does.
3538
353922. If the contents of a "message" modifier for a "warn" ACL verb do not begin
3540 with a valid header line field name (a series of printing characters
3541 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
3542
354323. Changed "disc" in the source to "disk" to conform to the documentation and
3544 the book and for uniformity.
3545
354624. Ignore Sendmail's -Ooption=value command line item.
3547
354825. When execve() failed while trying to run a command in a pipe transport,
3549 Exim was returning EX_UNAVAILBLE (69) from the subprocess. However, this
3550 could be confused with a return value of 69 from the command itself. This
3551 has been changed to 127, the value the shell returns if it is asked to run
3552 a non-existent command. The wording for the related log line suggests a
3553 non-existent command as the problem.
3554
355526. If received_header_text expands to an empty string, do not add a Received:
3556 header line to the message. (Well, it adds a token one on the spool, but
3557 marks it "old" so that it doesn't get used or transmitted.)
3558
355927. Installed eximstats 1.28 (addition of -nt option).
3560
356128. There was no check for failure on the call to getsockname() in the daemon
3562 code. This can fail if there is a shortage of resources on the system, with
3563 ENOMEM, for example. A temporary error is now given on failure.
3564
356529. Contrary to the C standard, it seems that in some environments, the
3566 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
3567 program. Exim now does this explicitly; it affects the formatting of
3568 timestamps using strftime().
3569
357030. If exiqsumm was given junk data, it threw up some uninitialized variable
3571 complaints. I've now initialized all the variables, to avoid this.
3572
357332. Header lines added by a system filter were not being "seen" during
3574 transport-time rewrites.
3575
357633. The info_callback() function passed to OpenSSL is set up with type void
3577 (*)(SSL *, int, int), as described somewhere. However, when calling the
3578 function (actually a macro) that sets it up, the type void(*)() is
3579 expected. I've put in a cast to prevent warnings from picky compilers.
3580
358134. If a DNS black list lookup found a CNAME record, but there were no A
3582 records associated with the domain it pointed at, Exim crashed.
3583
358435. If a DNS black list lookup returned more than one A record, Exim ignored
3585 all but the first. It now scans all returned addresses if a particular IP
3586 value is being sought. In this situation, the contents of the
3587 $dnslist_value variable are a list of all the addresses, separated by a
3588 comma and a space.
3589
359036. Tightened up the rules for host name lookups using reverse DNS. Exim used
3591 to accept a host name and all its aliases if the forward lookup for any of
3592 them yielded the IP address of the incoming connection. Now it accepts only
3593 those names whose forward lookup yields the correct IP address. Any other
3594 names are discarded. This closes a loophole whereby a rogue DNS
3595 administrator could create reverse DNS records to break through a
3596 wildcarded host restriction in an ACL.
3597
359837. If a user filter or a system filter that ran in a subprocess used any of
3599 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
3600 the wrong values were passed to the pipe command ($thisaddress had the
3601 value of $0, $0 had the value of $1, etc). This bug was introduced by
3602 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
3603
360438. Improved the line breaking for long SMTP error messages from ACLs.
3605 Previously, if there was no break point between 40 and 75 characters, Exim
3606 left the rest of the message alone. Two changes have been made: (a) I've
3607 reduced the minimum length to 35 characters; (b) if it can't find a break
3608 point between 35 and 75 characters, it looks ahead and uses the first one
3609 that it finds. This may give the occasional overlong line, but at least the
3610 remaining text gets split now.
3611
361239. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
3613 file descriptors might be low, and that setting 1000 would always raise it.
3614 It turns out that in some environments, the limit is already over 1000 and
3615 that lowering it causes trouble. So now Exim takes care not to decrease it.
3616
361740. When delivering a message, the value of $return_path is set to $sender_
3618 address at the start of routing (routers may change the value). By an
3619 oversight, this default was not being set up when an address was tested by
3620 -bt or -bv, which affected the outcome if any router or filter referred to
3621 $return_path.
3622
362341. The idea of the "warn" ACL verb is that it adds a header or writes to the
3624 log only when "message" or "log_message" are set. However, if one of the
3625 conditions was an address verification, or a call to a nested ACL, the
3626 messages generated by the underlying test were being passed through. This
3627 no longer happens. The underlying message is available in $acl_verify_
3628 message for both "message" and "log_message" expansions, so it can be
3629 passed through if needed.
3630
363142. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
3632 new expansion $bh_ to give the encoded byte string without charset
3633 translation. Translation happens only if iconv() is available; HAVE_ICONV
3634 indicates this at build time. HEADERS_CHARSET gives the charset to
3635 translate to; headers_charset can change it in the configuration, and
3636 "headers charset" can change it in an individual filter file.
3637
363843. Now that we have a default RFC 2047 charset (see above), the code in Exim
3639 that creates RFC 2047 encoded "words" labels them as that charset instead
3640 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
3641 expansion operator; (ii) when Exim creates a From: line for a local
3642 message; (iii) when a header line is rewritten to include a "phrase" part.
3643
364444. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
3645 buggy, causing it to skip the first lines of messages whose message ID
3646 ended in 'D'. This would not have bitten before Exim release 4.14, because
3647 message IDs were unlikely to end in 'D' before then. The effect was to have
3648 incorrect size information for certain domains.
3649
365045. #include "config.h" was missing at the start of the crypt16.c module. This
3651 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
3652 noticed.
3653
365446. If there was a timeout during a "random" callout check, Exim treated it as
3655 a failure of the random address, and carried on sending RSET and the real
3656 address. If the delay was just some slowness somewhere, the response to the
3657 original RCPT would be taken as a response to RSET and so on, causing
3658 mayhem of various kinds.
3659
366047. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
3661 when I implemented it. It didn't allow for the fact that some option values
3662 may legitimatetly be negative (e.g. size_addition), and it didn't even do
3663 the right test for positive values.
3664
366548. Domain names in DNS records are case-independent. Exim always looks them up
3666 in lower case. Some resolvers return domain names in exactly the case they
3667 appear in the zone file, that is, they may contain uppercase letters. Not
3668 all resolvers do this - some return always lower case. Exim was treating a
3669 change of case by a resolver as a change of domain, similar to a widening
3670 of a domain abbreviation. This triggered its re-routing code and so it was
3671 trying to route what was effectively the same domain again. This normally
3672 caused routing to fail (because the router wouldn't handle the domain
3673 twice). Now Exim checks for this case specially, and just changes the
3674 casing of the domain that it ultimately uses when it transmits the message
3675 envelope.
3676