[exim.git] / doc / doc-txt / ChangeLog
1 $Cambridge: exim/doc/doc-txt/ChangeLog,v 1.319 2006/03/01 11:40:51 ph10 Exp $
3 Change log file for Exim from version 4.21
4 -------------------------------------------
6 Exim version 4.61
7 -----------------
9 PH/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.
17 PH/02 The ipliteral router was not recognizing addresses of the form user@
18 [ipv6:....] because it didn't know about the "ipv6:" prefix.
20 PH/03 Added disable_ipv6.
22 PH/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.
29 PH/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.
32 PH/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.
38 PH/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 (sic), which is no longer necessary.
43 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
44 be the same on different OS.
46 PH/09 Moved a debug statement in filter processing to avoid a race problem when
47 testing.
49 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
50 whether --show-vars was specified or not
52 JJ/02 exipick: Added support for new ACL variable spool format introduced
53 in 4.61-PH/06
55 PH/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.
63 PH/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.
67 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
68 an end-of-file indication when reading a command response.
70 PH/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.
75 PH/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:
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.
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.
89 PH/15 Minor change to Makefile for building test_host (undocumented testing
90 feature).
92 PH/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.
101 PH/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.
105 PH/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.
109 PH/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.
119 PH/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.
123 PH/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.
126 PH/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.
132 PH/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.
136 PH/24 An host item such as was being treated as the IP address
137 without a mask. Now it is not recognized as an IP address, and
138 PH/23 above applies.
140 PH/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).
144 PH/26 Added the /no_tell option to "control=freeze".
146 PH/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".
152 PH/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.
157 PH/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.
161 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
163 PH/31 Added sender_verify_fail logging option.
165 PH/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.
169 PH/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.
175 PH/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).
179 PH/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.
184 PH/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.
190 PH/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.
194 PH/38 Panic-die if .include specifies a non-absolute path.
196 PH/39 A tweak to the "H" retry rule from its user.
198 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
199 a label. They prevented compilation on older perls.
201 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
202 a warning to be raised on newish perls.
204 JJ/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.
208 PH/40 Changed the default ident timeout from 30s to 5s.
210 PH/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.
213 PH/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.
224 PH/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.
228 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
230 PH/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.
239 PH/46 Messages that are created by the autoreply transport now contains a
240 References: header, in accordance with RFCs 2822 and 3834.
242 PH/47 Added authenticated_sender_force to the smtp transport.
244 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
246 PH/49 Installed latest patch from the Sieve maintainer.
248 PH/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.
255 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
258 Exim version 4.60
259 -----------------
261 PH/01 Two changes to the default runtime configuration:
263 (1) Move the checks for relay_from_hosts and authenticated clients from
264 after to before the (commented out) DNS black list checks.
266 (2) Add control=submission to the relay_from_hosts and authenticated
267 clients checks, on the grounds that messages accepted by these
268 statements are most likely to be submissions.
270 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
272 (1) Generate an error if the third argument for the ${prvs expansion is
273 not a single digit.
275 (2) Treat a missing third argument of ${prvscheck as if it were an empty
276 string.
278 (3) Reset the variables that are obtained from the first argument of
279 ${prvscheck and used in the second argument before leaving the code,
280 because their memory is reclaimed, so using them afterwards may do
281 silly things.
283 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
284 one (it's much easier than Tom thought :-).
286 (5) Because of (4), we can now allow for the use of $prvscheck_result
287 inside the third argument.
289 PH/03 For some reason, the default setting of PATH when running a command from
290 a pipe transport was just "/usr/bin". I have changed it to
291 "/bin:/usr/bin".
294 anything to be listed in the output from -bV.
296 PH/05 When a filter generated an autoreply, the entire To: header line was
297 quoted in the delivery log line, like this:
299 => >A.N.Other <ano@some.domain> <original@ddress> ...
301 This has been changed so that it extracts the operative address. There
302 may be more than one such address. If so, they are comma-separated, like
303 this:
305 => >ano@some.domain,ona@other.domain <original@ddress> ...
307 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
308 command, (for example, EHLO []) and the client's IP address was
309 not being looked up in the rDNS to get a host name, Exim was showing the
310 IP address twice in Received: lines, even though the IP addresses were
311 identical. For example:
313 Received: from [] (helo=[])
315 However, if the real host name was known, it was omitting the HELO data
316 if it matched the actual IP address. This has been tidied up so that it
317 doesn't show the same IP address twice.
319 PH/07 When both +timestamp and +memory debugging was on, the value given by
320 $tod_xxx expansions could be wrong, because the tod_stamp() function was
321 called by the debug printing, thereby overwriting the timestamp buffer.
322 Debugging no longer uses the tod_stamp() function when +timestamp is set.
324 PH/08 When the original message was included in an autoreply transport, it
325 always said "this is a copy of the message, including all the headers",
326 even if body_only or headers_only was set. It now gives an appropriate
327 message.
329 PH/09 Applied a patch from the Sieve maintainer which:
331 o fixes some comments
332 o adds the (disabled) notify extension core
333 o adds some debug output for the result of if/elsif tests
334 o points to the current vacation draft in the documentation
335 and documents the missing references header update
337 and most important:
339 o fixes a bug in processing the envelope test (when testing
340 multiple envelope elements, the last element determinted the
341 result)
343 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
344 Electronic Mail") by including:
346 Auto-submitted: auto-generated
348 in the messages that it generates (bounce messages and others, such as
349 warnings). In the case of bounce messages for non-SMTP mesages, there was
350 also a typo: it was using "Auto_submitted" (underscore instead of
351 hyphen). Since every message generated by Exim is necessarily in response
352 to another message, thes have all been changed to:
354 Auto-Submitted: auto-replied
356 in accordance with these statements in the RFC:
358 The auto-replied keyword:
360 - SHOULD be used on messages sent in direct response to another
361 message by an automatic process,
363 - MUST NOT be used on manually-generated messages,
365 - MAY be used on Delivery Status Notifications (DSNs) and Message
366 Disposition Notifications (MDNs),
368 - MUST NOT be used on messages generated by automatic or periodic
369 processes, except for messages which are automatic responses to
370 other messages.
372 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
373 to the default Received: header definition.
375 PH/12 Added log selector acl_warn_skipped (default on).
377 PH/13 After a successful wildlsearch lookup, discard the values of numeric
378 variables because (a) they are in the wrong storage pool and (b) even if
379 they were copied, it wouldn't work properly because of the caching.
381 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
382 checking when decoding. Apparently there are clients that generate
383 overlong encoded strings. Why am I not surprised?
385 PH/15 If the first argument of "${if match_address" was not empty, but did not
386 contain an "@" character, Exim crashed. Now it writes a panic log message
387 and treats the condition as false.
389 PH/16 In autoreply, treat an empty string for "once" the same as unset.
391 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
392 extension "envelope-auth". The code is finished and in agreement with
393 other implementations, but there is no documentation so far and in fact,
394 nobody wrote the draft yet. This extension is currently #undef'ed, thus
395 not changing the active code.
397 Print executed "if" and "elsif" statements when debugging is used. This
398 helps a great deal to understand what a filter does.
400 Document more things not specified clearly in RFC3028. I had all this
401 sorted out, when out of a sudden new issues came to my mind. Oops."
403 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
404 (Bugzilla #53).
406 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
407 canonical form (as documented). However, after a host name lookup from
408 the IP address, check_host() was doing a simple string comparison with
409 addresses acquired from the DNS when checking that the found name did
410 have the original IP as one of its addresses. Since any found IPv6
411 addresses are likely to be in abbreviated form, the comparison could
412 fail. Luckily, there already exists a function for doing the comparison
413 by converting both addresses to binary, so now that is used instead of
414 the text comparison.
416 PH/20 There was another similar case to PH/19, when a complete host name was
417 given in a host list; looking up its IP address could give an abbreviated
418 form, whereas the current host's name might or might not be abbreviated.
419 The same fix has been applied.
422 Exim version 4.54
423 -----------------
425 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
426 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
427 It now does.
429 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
430 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
432 PH/03 Typo: missing ".o" in src/pcre/Makefile.
434 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
435 header line, restrict the check to what is listed in RFCs 2369 and 2929.
436 Also, for "Auto-Submitted", treat anything other than "no" as
437 non-personal, in accordance with RFC 3834. (Previously it treated
438 anything starting "auto-" as non-personal.)
440 TF/01 The control=submission/name=... option had a problem with syntax
441 errors if the name included a slash character. The /name= option
442 now slurps the rest of the string, so it can include any characters
443 but it must come last in the list of options (after /sender_retain
444 or /domain=).
446 PH/05 Some modifications to the interface to the fake nameserver for the new
447 testing suite.
451 Exim version 4.53
452 -----------------
454 TK/01 Added the "success_on_redirect" address verification option. See
455 NewStuff for rationale and an example.
457 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
459 PH/02 Patch to exigrep to allow it to work on syslog lines.
461 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
462 fread() to skip over the body file's header line, because in Cygwin the
463 header line is locked and is inaccessible.
465 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
466 co-exist for some time) to make it clear that it is the Exim ID that is
467 referenced, not the Message-ID: header line.
469 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
470 string_format() function, because snprintf() does not exist on all
471 operating systems.
473 PH/06 The use of forbid_filter_existstest now also locks out the use of the
474 ${stat: expansion item.
476 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
477 protocol synchronization error", to keep the pedants happy.
479 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
480 well as for IRIX systems, when gcc is being used. See the host.c source
481 file for comments.
483 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
485 PH/10 Named domain lists were not working if used in a queue_smtp_domains
486 setting.
488 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
489 transport and to the smtp transport in LMTP mode.
491 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
493 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
494 run a filter in a subprocess. This could lead to confusion in subsequent
495 lookups in the parent process. There should also be a search_tidyup() at
496 the end of the subprocess.
498 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
499 only if the host matched helo_try_verify_hosts, which caused the
500 verification to occur when the EHLO/HELO command was issued. The ACL just
501 tested the remembered result. Now, if a previous verification attempt has
502 not happened, "verify = helo" does it there and then.
504 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
506 TK/03 Fix log output including CR from clamd.
508 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
509 exist provoked a memory error which could cause a segfault.
511 PH/15 Installed PCRE 6.2
513 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
515 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
516 of the problem. Specifically, suggested +O2 rather than +O1 for the
517 HP-UX compiler.
519 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
521 PH/20 If a delivery was routed to a non-standard port by means of an SRV
522 record, the port was not correctly logged when the outgoing_port log
523 selector was set (it logged the transort's default port).
525 PH/21 Added support for host-specific ports to manualroute, queryprogram,
526 fallback_hosts, and "hosts" in the smtp transport.
528 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
529 host errors such as "Connection refused".
531 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
532 authentication with radiusclient 0.4.9:
534 - Error returned from rc_read_config was caught wrongly
535 - Username/password not passed on to radius server due to wrong length.
537 The presumption is that some radiusclient API changes for 4.51/PH/17
538 were not taken care of correctly. The code is still untested by me (my
539 Linux distribution still has 0.3.2 of radiusclient), but it was
540 contributed by a Radius user.
542 PH/24 When doing a callout, the value of $domain wasn't set correctly when
543 expanding the "port" option of the smtp transport.
545 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
546 while reading a MIME header. Thanks to Tom Hughes for a patch.
548 PH/24 Include config.h inside local_scan.h so that configuration settings are
549 available.
551 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
552 that in an ACL for RCPT (for example), you can examine exactly what was
553 received.
555 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
556 commands, but it was not correctly comparing the address with the actual
557 client host address. Thus, it would show the EHLO address in Received:
558 header lines when this was not necessary.
560 PH/27 Added the % operator to ${eval:}.
562 PH/28 Exim tries to create and chdir to its spool directory when it starts;
563 it should be ignoring failures (because with -C, for example, it has lost
564 privilege). It wasn't ignoring creation failures other than "already
565 exists".
567 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
568 -bV or -d is used.
570 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
571 because an input line was too long, either on its own, or by virtue of
572 too many continuations, the temporary file was not being removed, and the
573 return code was incorrect.
575 PH/31 Missing "BOOL" in function definition in filtertest.c.
577 PH/32 Applied Sieve patches from the maintainer.
579 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
581 PH/33 Added "verify = not_blind".
583 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
584 Local/Makefile (with some defaults set). These are used in built scripts
585 such as exicyclog, but they have never been used in the exim_install
586 script (though there are many overriding facilities there). I have
587 arranged that the exim_install script now takes note of these two
588 settings.
590 PH/35 Installed configuration files for Dragonfly.
592 PH/36 When a locally submitted message by a trusted user did not contain a
593 From: header, and the sender address was obtained from -f or from an SMTP
594 MAIL command, and the trusted user did not use -F to supply a sender
595 name, $originator_name was incorrectly used when constructing a From:
596 header. Furthermore, $originator_name was used for submission mode
597 messages from external hosts without From: headers in a similar way,
598 which is clearly wrong.
600 PH/37 Added control=suppress_local_fixups.
602 PH/38 When log_selector = +received_sender was set, and the addition of the
603 sender made the log line's construction buffer exactly full, or one byte
604 less than full, an overflow happened when the terminating "\n" was
605 subsequently added.
607 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
608 when the result of a list match is failure because a DNS lookup failed.
610 PH/40 RM_COMMAND is now used in the building process.
612 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
613 the "build-* directories that it finds.
615 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
616 address in a domain literal was a prefix of an interface address.
618 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
619 when verifying a sender address, unless rewrite_headers is false.
621 PH/44 Wrote a long comment about why errors_to addresses are verified as
622 recipients, not senders.
624 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
625 the ratelimit ACL was added.
627 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
629 PH/46 Added extra information about PostgreSQL errors to the error string.
631 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
632 avoiding the need to install special zones in a real server. This is
633 backwards compatible; if it can't find the fake resolver, it drops back.
634 Thus, both old and new test suites can be run.
636 TF/02 Added util/ratelimit.pl
638 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
639 clock is set back in time.
641 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
642 Candler <B.Candler@pobox.com>.
644 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
645 OK for addresses that are the result of redirections.
647 PH/48 A number of further additions for the benefit of the new test suite,
648 including a fake gethostbyname() that interfaces to the fake DNS resolver
649 (see PH/47 above).
651 TF/06 The fix for widen_domains has also been applied to qualify_single and
652 search_parents which are the other dnslookup options that can cause
653 header rewrites.
655 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
656 type ("H").
658 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
660 TF/07 Exim produced the error message "an SRV record indicated no SMTP
661 service" if it encountered an MX record with an empty target hostname.
662 The message is now "an MX or SRV record indicated no SMTP service".
664 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
665 if the DNS of the sending site is misconfigured. This is quite a
666 common situation. This change restores the behaviour of treating a
667 helo verification defer as a failure.
669 PH/51 If self=fail was set on a router, the bounce message did not include the
670 actual error message.
673 Exim version 4.52
674 -----------------
676 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
678 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
679 command itself ended in error, the underlying message about the transport
680 filter timeout was being overwritten with the pipe command error. Now the
681 underlying error message should be appended to the second error message.
683 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
685 PH/02 Reduce the amount of output that "make" produces by default. Full output
686 can still be requested.
688 PH/03 The warning log line about a condition test deferring for a "warn" verb
689 was being output only once per connection, rather than after each
690 occurrence (because it was using the same function as for successful
691 "warn" verbs). This seems wrong, so I have changed it.
693 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
694 it should not have, which might have caused a crash in the right
695 circumstances, but probably never did.
697 PH/04 Installed a modified version of Tony Finch's patch to make submission
698 mode fix the return path as well as the Sender: header line, and to
699 add a /name= option so that you can make the user's friendly name appear
700 in the header line.
702 TF/03 Added the control = fakedefer ACL modifier.
704 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
705 Mark Lowes for thorough testing.
707 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
709 TK/03 Merged latest SRS patch from Miles Wilton.
711 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
712 with the definition in sysexits.h (which is #included earlier).
713 Fortunately, Exim does not actually use EX_OK. The code used to try to
714 preserve the sysexits.h value, by assumimg that macro definitions were
715 scanned for macro replacements. I have been disabused of this notion,
716 so now the code just undefines EX_OK before #including unistd.h.
718 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
719 in the smtp transport. When a block could not be written in a single
720 write() function, the timeout was being re-applied to each part-write.
721 This seems wrong - if the receiver was accepting one byte at a time it
722 would take for ever. The timeout is now adjusted when this happens. It
723 doesn't have to be particularly precise.
725 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
726 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
728 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
729 without a domain if the check to <postmaster@domain> fails.
731 SC/01 Eximstats: added -xls and the ability to specify output files
732 (patch written by Frank Heydlauf).
734 SC/02 Eximstats: use FileHandles for outputing results.
736 SC/03 Eximstats: allow any combination of xls, txt, and html output.
738 SC/04 Eximstats: fixed display of large numbers with -nvr option
740 SC/05 Eximstats: fixed merging of reports with empty tables.
742 SC/06 Eximstats: added the -include_original_destination flag
744 SC/07 Eximstats: removed tabs and trailing whitespace.
746 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
748 TK/06 MBOX spool code: Add real "From " MBOX separator line
749 so the .eml file is really in mbox format (even though
750 most programs do not really care). Patch from Alex Miller.
752 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
753 The latter is generated from $received_to and is only set if the
754 message has one envelope recipient. SA can use these headers,
755 obviously out-of-the-box. Patch from Alex Miller.
757 PH/08 The ${def test on a variable was returning false if the variable's
758 value was "0", contrary to what the specification has always said!
759 The result should be true unless the variable is empty.
761 PH/09 The syntax error of a character other than { following "${if
762 def:variable_name" (after optional whitespace) was not being diagnosed.
763 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
764 accidental colon was present, for example, could give incorrect results.
766 PH/10 Tidied the code in a number of places where the st_size field of a stat()
767 result is used (not including appendfile, where other changes are about
768 to be made).
770 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
771 This involved changing a lot of size variables from int to off_t. It
772 should work with maildirs and everything.
774 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
775 spamd dying while we are connected to it.
777 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
778 <jgh@wizmail.org>
780 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
781 function for BDB 4.3.
783 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
785 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
786 the sources for PCRE have been re-arranged and more files are now
787 involved.
789 PH/15 The code I had for printing potentially long long variables in PH/11
790 above was not the best (it lost precision). The length of off_t variables
791 is now inspected at build time, and an appropriate printing format (%ld
792 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
793 to be "long long int" or "long int". This is needed for the internal
794 formatting function string_vformat().
796 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
797 the configuration file to be ":syslog", then the script "guesses" where
798 the logs files are, rather than using the compiled in default. In our
799 case the guess is not the same as the compiled default, so the script
800 suddenly stopped working when I started to use syslog. The patch checks
801 to see if log_file_path is "". If so, it attempts to read it from exim
802 with no configuration file to get the compiled in version, before it
803 falls back to the previous guessing code."
805 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
806 implementing BATV in an Exim configuration. See NewStuff for the gory
807 details.
809 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
810 Makefile that are specific to HP-UX.
812 PH/18 If the "use_postmaster" option was set for a recipient callout together
813 with the "random" option, the postmaster address was used as the MAIL
814 FROM address for the random test, but not for the subsequent recipient
815 test. It is now used for both.
817 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
818 patch removes a few documentation additions to RFC 3028, because the
819 latest draft now contains them. It adds the new en;ascii-case comparator
820 and a new error check for 8bit text in MIME parts. Comparator and
821 require names are now matched exactly. I enabled the subaddress
822 extension, but it is not well tested yet (read: it works for me)."
824 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
825 rework some of the code of TK/09 above to avoid the hardwired use of
826 "%lld" and "long long". Replaced the call to snprintf() with a call to
827 string_vformat().
829 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
830 records point to non-existent hosts", "retry timeout exceeded", and
831 "retry time not reached for any host after a long failure period".
833 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
834 experimental DomainKeys support:
836 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
837 (2) On an error such as an illegally used "control", the wrong name for
838 the control was given.
840 These problems did NOT occur unless DomainKeys support was compiled.
842 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
844 PH/24 Added ${if match_ip condition.
846 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
847 databases so that it will be absolutely obvious if a crash occurs in the
848 DB library. This is a regular occurrence (often caused by mis-matched
849 db.h files).
851 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
852 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
853 picked up on a user's system that detects such things. There doesn't seem
854 to be a gcc warning option for this - only an attribute that has to be
855 put on the function's prototype. It seems that in Fedora Core 4 they have
856 set this on a number of new functions. No doubt there will be more in due
857 course.
859 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
860 specify a transport. However, if an address that was verified by such a
861 router was the subject of a callout, Exim crashed because it tried to
862 read the rcpt_include_affixes from the non-existent transport. Now it
863 just assumes that the setting of that option is false. This bug was
864 introduced by 4.51/PH/31.
866 PH/28 Changed -d+all to exclude +memory, because that information is very
867 rarely of interest, but it makes the output a lot bigger. People tend to
868 do -d+all out of habit.
870 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
871 code in os-type was giving problems when libc.so lives in lib64, like on
872 x86_64 Fedora Core.
874 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
875 aren't the modern standard, and it seems that some systems' include files
876 don't always have them. Exim was already checking for some of the newer
877 ones like T_AAAA, and defining it itself. I've added checks for all the
878 record types that Exim uses.
880 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
881 not automatically generating a new one, as it is supposed to. This
882 prevented TLS from working. If the file did exist, but contained invalid
883 data, a new version was generated, as expected. It was only the case of a
884 non-existent file that was broken.
886 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
887 with a change in libdomainkeys > 0.64.
889 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
890 from DNS. If the selector record carries the flag, it now has
891 precedence over the domain-wide flag.
893 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
895 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
896 the use of a port name that isn't defined in /etc/services) occurred, the
897 message was deferred as in a normal delivery, and thus remained on the
898 spool, instead of being failed because of the mua_wrapper setting. This
899 is now fixed, and I tidied up some of the mua_wrapper messages at the
900 same time.
902 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
903 the messages in a hash of arrays rather than using individual hashes.
904 This is a bit cleaner and results in dramatic memory savings, albeit
905 at a slight CPU cost.
907 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
908 as requested by Marc Sherman.
910 SC/10 Eximstats: added histograms for user specified patterns as requested
911 by Marc Sherman.
913 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
915 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
916 fopen() in the content-scanning modules that did not already have it.
919 Exim version 4.51
920 -----------------
922 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
923 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
925 TK/02 Fix ACL "control" statement not being available in MIME ACL.
927 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
929 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
930 to test Sieve filters that use "vacation".
932 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
933 that changes the way the GnuTLS parameters are stored in the cache file.
934 The new format can be generated externally. For backward compatibility,
935 if the data in the cache doesn't make sense, Exim assumes it has read an
936 old-format file, and it generates new data and writes a new file. This
937 means that you can't go back to an older release without removing the
938 file.
940 PH/03 A redirect router that has both "unseen" and "one_time" set does not
941 work if there are any delivery delays because "one_time" forces the
942 parent to be marked "delivered", so its unseen clone is never tried
943 again. For this reason, Exim now forbids the simultaneous setting of
944 these two options.
946 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
947 redirected to themselves ("homonym" addresses). Read the long ChangeLog
948 entry if you want to know the details. The fix, however, neglected to
949 consider the case when local delivery batching is involved. The test for
950 "previously delivered" was not happening when checking to see if an
951 address could be batched with a previous (undelivered) one; under
952 certain circumstances this could lead to multiple deliveries to the same
953 address.
955 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
956 in its include files, and this causes problems building Exim.
958 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
959 header_syntax) but Exim was just ignoring anything given after a slash.
960 In particular, this caused confusion with an attempt to use "verify =
961 reverse_host_lookup/defer_ok". An error is now given when options are
962 supplied for verify items that do not have them. (Maybe reverse_host_
963 lookup should have a defer_ok option, but that's a different point.)
965 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
966 defined by RFC 821) to 2048, because there were problems with some AUTH
967 commands, and RFC 1869 says the size should be increased for extended
968 SMTP commands that take arguments.
970 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
971 Finch).
973 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
974 "unknown" error; now it says that the functionality isn't in the binary.
976 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
977 an address' error message when a string expansion fails (syntax or
978 whatever). Otherwise the password may appear in the log. Following change
979 PH/42 below, there is no longer a chance of it appearing in a bounce
980 message.
982 PH/11 Installed exipick version 20050225.0 from John Jetmore.
984 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
985 of its addresses was ever tried. (Bugzilla bug #2.)
987 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
988 the result incorrectly in the debug output. (It correctly added a newline
989 to what was transported.)
991 TF/01 Added $received_time.
993 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
994 commented out examples of how to interface to a virus scanner and to
995 SpamAssassin. Also added commented examples of av_scanner and
996 spamd_address settings.
998 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
999 and controls are allowed in which ACLs. There were a couple of minor
1000 errors. Some of the entries in the conditions table (which is a table of
1001 where they are NOT allowed) were getting very unwieldy; rewrote them as a
1002 negation of where the condition IS allowed.
1004 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
1006 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
1007 header file does not have a version number, so I've had to invent a new
1008 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
1009 API. The code is untested by me (my Linux distribution still has 0.3.2 of
1010 radiusclient), but it was contributed by a Radius user.
1012 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
1013 files or directories, for OpenSSL.
1015 PH/19 When an Exim process that is running as root has to create an Exim log
1016 file, it does so in a subprocess that runs as exim:exim so as to get the
1017 ownership right at creation (otherwise, other Exim processes might see
1018 the file with the wrong ownership). There was no test for failure of this
1019 fork() call, which would lead to the process getting stuck as it waited
1020 for a non-existent subprocess. Forks do occasionally fail when resources
1021 run out. I reviewed all the other calls to fork(); they all seem to check
1022 for failure.
1024 PH/20 When checking for unexpected SMTP input at connect time (before writing
1025 the banner), Exim was not dealing correctly with a non-positive return
1026 from the read() function. If the client had disconnected by this time,
1027 the result was a log entry for a synchronization error with an empty
1028 string after "input=" when read() returned zero. If read() returned -1
1029 (an event I could not check), uninitialized data bytes were printed.
1030 There were reports of junk text (parts of files, etc) appearing after
1031 "input=".
1033 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
1035 PH/22 Added support for macro redefinition, and (re)definition in between
1036 driver and ACL definitions.
1038 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
1039 forgetting to use the resulting value; it was using the unexpanded value.
1041 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
1042 hadn't been configured. The fix is from Juergen Kreileder, who
1043 understands it better than I do:
1045 "Here's what I see happening with three configured cyrus_sasl
1046 authenticators configured (plain, login, cram-md5):
1048 On startup auth_cyrus_sasl_init() gets called for each of these.
1049 This means three calls to sasl_listmech() without a specified mech_list.
1050 => SASL tests which mechs of all available mechs actually work
1051 => three warnings about OTP not working
1052 => the returned list contains: plain, login, cram-md5, digest-md5, ...
1054 With the patch, sasl_listmech() also gets called three times. But now
1055 SASL's mech_list option is set to the server_mech specified in the the
1056 authenticator. Or in other words, the answer from sasl_listmech()
1057 gets limited to just the mech you're testing for (which is different
1058 for each call.)
1059 => the return list contains just 'plain' or 'login', 'cram-md5' or
1060 nothing depending on the value of ob->server_mech.
1062 I've just tested the patch: Authentication still works fine,
1063 unavailable mechs specified in the exim configuration are still
1064 caught, and the auth.log warnings about OTP are gone."
1066 PH/25 When debugging is enabled, the contents of the command line are added
1067 to the debugging output, even when log_selector=+arguments is not
1068 specified.
1070 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
1071 answer is "GNU", and only if the return is "GNU/something" is the answer
1072 "Linux".
1074 PH/27 $acl_verify_message is now set immediately after the failure of a
1075 verification in an ACL, and so is available in subsequent modifiers. In
1076 particular, the message can be preserved by coding like this:
1078 warn !verify = sender
1079 set acl_m0 = $acl_verify_message
1081 Previously, $acl_verify_message was set only while expanding "message"
1082 and "log_message" when a very denied access.
1084 PH/28 Modified OS/os.c-Linux with
1086 -#ifndef OS_LOAD_AVERAGE
1087 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
1089 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
1090 nomenclature these days.)
1092 PH/29 Installed patch from the Sieve maintainer that adds the options
1093 sieve_useraddress and sieve_subaddress to the redirect router.
1095 PH/30 In these circumstances:
1096 . Two addresses routed to the same list of hosts;
1097 . First host does not offer TLS;
1098 . First host accepts first address;
1099 . First host gives temporary error to second address;
1100 . Second host offers TLS and a TLS session is established;
1101 . Second host accepts second address.
1102 Exim incorrectly logged both deliveries with the TLS parameters (cipher
1103 and peerdn, if requested) that were in fact used only for the second
1104 address.
1106 PH/31 When doing a callout as part of verifying an address, Exim was not paying
1107 attention to any local part prefix or suffix that was matched by the
1108 router that accepted the address. It now behaves in the same way as it
1109 does for delivery: the affixes are removed from the local part unless
1110 rcpt_include_affixes is set on the transport.
1112 PH/32 Add the sender address, as F=<...>, to the log line when logging a
1113 timeout during the DATA phase of an incoming message.
1115 PH/33 Sieve envelope tests were broken for match types other than :is. I have
1116 applied a patch sanctioned by the Sieve maintainer.
1118 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
1119 the uid or gid is negative. A case of a negative gid caused this to be
1120 noticed. The fix allows for either to be negative.
1122 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
1123 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
1124 overlooked.
1126 PH/36 The change PH/12 above was broken. Fixed it.
1128 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
1129 the grounds that routing the same address twice would always produce the
1130 same answer. This might have been true once, but it is certainly no
1131 longer true now. Routing a child address may depend on the previous
1132 routing that produced that child. Some complicated redirection strategies
1133 went wrong when messages had multiple recipients, and made Exim's
1134 behaviour dependent on the order in which the addresses were given.
1136 I have moved the duplicate checking until after the routing is complete.
1137 Exim scans the addresses that are assigned to local and remote
1138 transports, and removes any duplicates. This means that more work will be
1139 done, as duplicates will always all be routed, but duplicates are
1140 presumably rare, so I don't expect this is of any significance.
1142 For deliveries to pipes, files, and autoreplies, the duplicate checking
1143 still happens during the routing process, since they are not going to be
1144 routed further.
1146 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
1147 It corrects a timeout issue with spamd. This is Ian's comment: "The
1148 background is that sometimes spamd either never reads data from a
1149 connection it has accepted, or it never writes response data. The exiscan
1150 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
1151 blindly assumes that writes won't block so it may never time out."
1153 PH/39 Allow G after quota size as well as K and M.
1155 PH/40 The value set for $authenticated_id in an authenticator may not contain
1156 binary zeroes or newlines because the value is written to log lines and
1157 to spool files. There was no check on this. Now the value is run through
1158 the string_printing() function so that such characters are converted to
1159 printable escape sequences.
1161 PH/41 $message_linecount is a new variable that contains the total number of
1162 lines in the message. Compare $body_linecount, which is the count for the
1163 body only.
1165 PH/42 Exim no longer gives details of delivery errors for specific addresses in
1166 bounce and delay warning messages, except in certain special cases, which
1167 are as follows:
1169 (a) An SMTP error message from a remote host;
1170 (b) A message specified in a :fail: redirection;
1171 (c) A message specified in a "fail" command in a system filter;
1172 (d) A message specified in a FAIL return from the queryprogram router;
1173 (e) A message specified by the cannot_route_message router option.
1175 In these cases only, Exim does include the error details in bounce and
1176 warning messages. There are also a few cases where bland messages such
1177 as "unrouteable address" or "local delivery error" are given.
1179 PH/43 $value is now also set for the "else" part of a ${run expansion.
1181 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
1182 being worked on, but at least Exim now implements the latest version to
1183 play with."
1185 PH/45 In a pipe transport, although a timeout while waiting for the pipe
1186 process to complete was treated as a delivery failure, a timeout while
1187 writing the message to the pipe was logged, but erroneously treated as a
1188 successful delivery. Such timeouts include transport filter timeouts. For
1189 consistency with the overall process timeout, these timeouts are now
1190 treated as errors, giving rise to delivery failures by default. However,
1191 there is now a new Boolean option for the pipe transport called
1192 timeout_defer, which, if set TRUE, converts the failures into defers for
1193 both kinds of timeout. A transport filter timeout is now identified in
1194 the log output.
1196 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
1197 systems where "make" and "gmake" are different, calling "gmake" at top
1198 level broke things. I've arranged for the value of $(MAKE) to be passed
1199 from the Makefile to this script so that it can call the same version of
1200 "make".
1203 A note about Exim versions 4.44 and 4.50
1204 ----------------------------------------
1206 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
1207 changes of various kinds. As a consequence, a big documentation update was
1208 needed. This delayed the release for rather longer than seemed good, especially
1209 in the light of a couple of (minor) security issues. Therefore, the changes
1210 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
1211 release. So 4.44 and 4.50 are in effect two different branches that both start
1212 from 4.43.
1214 I have left the 4.50 change log unchanged; it contains all the changes since
1215 4.43. The change log for 4.44 is below; many of its items are identical to
1216 those for 4.50. This seems to be the most sensible way to preserve the
1217 historical information.
1220 Exim version 4.50
1221 -----------------
1223 1. Minor wording change to the doc/README.SIEVE file.
1225 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
1226 computation of the current number of files was incorrect.
1228 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
1229 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
1230 place.
1232 4. Give more explanation in the error message when the command for a transport
1233 filter fails to execute.
1235 5. There are several places where Exim runs a non-Exim command in a
1236 subprocess. The SIGUSR1 signal should be disabled for these processes. This
1237 was being done only for the command run by the queryprogram router. It is
1238 now done for all such subprocesses. The other cases are: ${run, transport
1239 filters, and the commands run by the lmtp and pipe transports.
1241 6. Added CONFIGURE_GROUP build-time option.
1243 7. Some older OS have a limit of 256 on the maximum number of file
1244 descriptors. Exim was using setrlimit() to set 1000 as a large value
1245 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
1246 systems. I've change it so that if it can't get 1000, it tries for 256.
1248 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
1249 was an oversight, and furthermore, ever since the addition of extra
1250 controls (e.g. 4.43/32), the checks on when to allow different forms of
1251 "control" were broken. There should now be diagnostics for all cases when a
1252 control that does not make sense is encountered.
1254 9. Added the /retain_sender option to "control=submission".
1256 10. $recipients is now available in the predata ACL (oversight).
1258 11. Tidy the search cache before the fork to do a delivery from a message
1259 received from the command line. Otherwise the child will trigger a lookup
1260 failure and thereby defer the delivery if it tries to use (for example) a
1261 cached ldap connection that the parent has called unbind on.
1263 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
1264 of $address_data from the recipient verification was clobbered by the
1265 sender verification.
1267 13. The value of address_data from a sender verification is now available in
1268 $sender_address_data in subsequent conditions in the ACL statement.
1270 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
1272 15. Added a new option "connect=<time>" to callout options, to set a different
1273 connection timeout.
1275 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
1276 was its contents. (It was OK if the option was not defined at all.)
1278 17. A "Completed" log line is now written for messages that are removed from
1279 the spool by the -Mrm option.
1281 18. New variables $sender_verify_failure and $recipient_verify_failure contain
1282 information about exactly what failed.
1284 19. Added -dd to debug only the daemon process.
1286 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
1287 handles timeouts, both on the server side and network timeouts. Renamed the
1288 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
1290 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
1291 It is now set to "smtps".
1293 22. $host_address is now set to the target address during the checking of
1294 ignore_target_hosts.
1296 23. When checking ignore_target_hosts for an ipliteral router, no host name was
1297 being passed; this would have caused $sender_host_name to have been used if
1298 matching the list had actually called for a host name (not very likely,
1299 since this list is usually IP addresses). A host name is now passed as
1300 "[x.x.x.x]".
1302 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
1303 code that specifies a non-restarting handler (typically sigaction() in
1304 modern systems) in an attempt to fix a rare and obscure crash bug.
1306 25. Narrowed the window for a race in the daemon that could cause it to ignore
1307 SIGCHLD signals. This is not a major problem, because they are used only to
1308 wake it up if nothing else does.
1310 26. A malformed maildirsize file could cause Exim to calculate negative values
1311 for the mailbox size or file count. Odd effects could occur as a result.
1312 The maildirsize information is now recalculated if the size or filecount
1313 end up negative.
1315 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
1316 support for a long time. Removed HAVE_SYS_VFS_H.
1318 28. Installed the latest version of exipick from John Jetmore.
1320 29. In an address list, if the pattern was not a regular expression, an empty
1321 subject address (from a bounce message) matched only if the pattern was an
1322 empty string. Non-empty patterns were not even tested. This was the wrong
1323 because it is perfectly reasonable to use an empty address as part of a
1324 database query. An empty address is now tested by patterns that are
1325 lookups. However, all the other forms of pattern expect the subject to
1326 contain a local part and a domain, and therefore, for them, an empty
1327 address still always fails if the pattern is not itself empty.
1329 30. Exim went into a mad DNS loop when attempting to do a callout where the
1330 host was specified on an smtp transport, and looking it up yielded more
1331 than one IP address.
1333 31. Re-factored the code for checking spool and log partition space into a
1334 function that finds that data and another that does the check. The former
1335 is then used to implement four new variables: $spool_space, $log_space,
1336 $spool_inodes, and $log_inodes.
1338 32. The RFC2047 encoding function was originally intended for short strings
1339 such as real names; it was not keeping to the 75-character limit for
1340 encoded words that the RFC imposes. It now respects the limit, and
1341 generates multiple encoded words if necessary. To be on the safe side, I
1342 have increased the buffer size for the ${rfc2047: expansion operator from
1343 1024 to 2048 bytes.
1345 33. It is now permitted to omit both strings after an "if" condition; if the
1346 condition is true, the result is "true". As before, when the second string
1347 is omitted, a false condition yields an empty string. This makes it less
1348 cumbersome to write custom ACL and router conditions.
1350 34. Failure to deliver a bounce message always caused it to be frozen, even if
1351 there was an errors_to setting on the router. The errors_to setting is now
1352 respected.
1354 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
1355 canonical form (fully expanded) before being placed in
1356 $sender_host_address.
1358 36. The table in the code that translates DNS record types into text (T_A to
1359 "A" for instance) was missing entries for NS and CNAME. It is just possible
1360 that this could have caused confusion if both these types were looked up
1361 for the same domain, because the text type is used as part of Exim's
1362 per-process caching. But the chance of anyone hitting this buglet seems
1363 very small.
1365 37. The dnsdb lookup has been extended in a number of ways.
1367 (1) There is a new type, "zns", which walks up the domain tree until it
1368 finds some nameserver records. It should be used with care.
1370 (2) There is a new type, "mxh", which is like "mx" except that it returns
1371 just the host names, not the priorities.
1373 (3) It is now possible to give a list of domains (or IP addresses) to be
1374 looked up. The behaviour when one of the lookups defers can be
1375 controlled by a keyword.
1377 (4) It is now possible to specify the separator character for use when
1378 multiple records are returned.
1380 38. The dnslists ACL condition has been extended: it is now possible to supply
1381 a list of IP addresses and/or domains to be looked up in a particular DNS
1382 domain.
1384 39. Added log_selector=+queue_time_overall.
1386 40. When running the queue in the test harness, wait just a tad after forking a
1387 delivery process, to get repeatability of debugging output.
1389 41. Include certificate and key file names in error message when GnuTLS fails
1390 to set them up, because the GnuTLS error message doesn't include the name
1391 of the failing file when there is a problem reading it.
1393 42. Allow both -bf and -bF in the same test run.
1395 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
1397 44. The "Exiscan patch" is now merged into the mainline Exim source.
1399 45. Sometimes the final signoff response after QUIT could fail to get
1400 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
1401 < 0 before doing a fflush(). This bug looks as though it goes back to the
1402 introduction of TLS in release 3.20, but "sometimes" must have been rare
1403 because the tests only now provoked it.
1405 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
1406 (this can affect the format of dates).
1408 47. exim_tidydb, when checking for the continued existence of a message for
1409 which it has found a message-specific retry record, was not finding
1410 messages that were in split spool directories. Consequently, it was
1411 deleting retry records that should have stayed in existence.
1413 48. Steve fixed some bugs in eximstats.
1415 49. The SPA authentication driver was not abandoning authentication and moving
1416 on to the next authenticator when an expansion was forced to fail,
1417 contradicting the general specification for all authenticators. Instead it
1418 was generating a temporary error. It now behaves as specified.
1420 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
1421 (the order specifies the preference for clients). The order is now AES256,
1422 AES128, 3DES, ARCFOUR128.
1424 51. Small patch to Sieve code - explicitly set From: when generating an
1425 autoreply.
1427 52. Exim crashed if a remote delivery caused a very long error message to be
1428 recorded - for instance if somebody sent an entire SpamAssassin report back
1429 as a large number of 550 error lines. This bug was coincidentally fixed by
1430 increasing the size of one of Exim's internal buffers (big_buffer) that
1431 happened as part of the Exiscan merge. However, to be on the safe side, I
1432 have made the code more robust (and fixed the comments that describe what
1433 is going on).
1435 53. Now that there can be additional text after "Completed" in log lines (if
1436 the queue_time_overall log selector is set), a one-byte patch to exigrep
1437 was needed to allow it to recognize "Completed" as not the last thing in
1438 the line.
1440 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
1441 patch that reportedly fixes this has been added. I am not expert enough to
1442 create a test for it. This is what the patch creator wrote:
1444 "I found a little strange behaviour of ldap code when working with
1445 Windows 2003 AD Domain, where users was placed in more than one
1446 Organization Units. When I tried to give exim partial DN, the exit code
1447 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
1448 But simultaneously result of request was absolutely normal ldap result,
1449 so I produce this patch..."
1451 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
1452 REFERENCE, so I have modified the code to exclude the patch when that macro
1453 is not defined.
1455 55. Some experimental protocols are using DNS PTR records for new purposes. The
1456 keys for these records are domain names, not reversed IP addresses. The
1457 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
1458 leaves it alone. Component reversal etc. now happens only for IP addresses.
1459 CAN-2005-0021
1461 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
1463 57. Double the size of the debug message buffer (to 2048) so that more of very
1464 long debug lines gets shown.
1466 58. The exicyclog utility now does better if the number of log files to keep
1467 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
1469 59. Two changes related to the smtp_active_hostname option:
1471 (1) $smtp_active_hostname is now available as a variable.
1472 (2) The default for smtp_banner uses $smtp_active_hostname instead
1473 of $primary_hostname.
1475 60. The host_aton() function is supposed to be passed a string that is known
1476 to be a valid IP address. However, in the case of IPv6 addresses, it was
1477 not checking this. This is a hostage to fortune. Exim now panics and dies
1478 if the condition is not met. A case was found where this could be provoked
1479 from a dnsdb PTR lookup with an IPv6 address that had more than 8
1480 components; fortuitously, this particular loophole had already been fixed
1481 by change 4.50/55 above.
1483 If there are any other similar loopholes, the new check in host_aton()
1484 itself should stop them being exploited. The report I received stated that
1485 data on the command line could provoke the exploit when Exim was running as
1486 exim, but did not say which command line option was involved. All I could
1487 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
1488 running as the user.
1489 CAN-2005-0021
1491 61. There was a buffer overflow vulnerability in the SPA authentication code
1492 (which came originally from the Samba project). I have added a test to the
1493 spa_base64_to_bits() function which I hope fixes it.
1494 CAN-2005-0022
1496 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
1497 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
1499 63. The daemon start-up calls getloadavg() while still root for those OS that
1500 need the first call to be done as root, but it missed one case: when
1501 deliver_queue_load_max is set with deliver_drop_privilege. This is
1502 necessary for the benefit of the queue runner, because there is no re-exec
1503 when deliver_drop_privilege is set.
1505 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
1506 This has been fixed.
1508 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
1509 was in use, was not putting the data itself into the right store pool;
1510 consequently, it could be overwritten for a subsequent message in the same
1511 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
1512 the caching.)
1514 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
1516 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
1517 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
1518 the function were treating the return as a boolean value, which happened to
1519 work because 0=false and not-0=true, but is not correct code.
1521 68. The host_aton() function was not handling scoped IPv6 addresses (those
1522 with, for example, "%eth0" on the end) correctly.
1524 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
1525 negated items (that is, ~something) in unsigned ints. Some compilers
1526 apparently mutter when there is no cast.
1528 70. If an address verification called from an ACL failed, and did not produce a
1529 user-specific message (i.e. there was only a "system" message), nothing was
1530 put in $acl_verify_message. In this situation, it now puts the system
1531 message there.
1533 71. Change 4.23/11 added synchronization checking at the start of an SMTP
1534 session; change 4.31/43 added the unwanted input to the log line - except
1535 that it did not do this in the start of session case. It now does.
1537 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
1538 This is wrong and can cause the other end to generate a synchronization
1539 error if it is another Exim or anything else that does the synchronization
1540 check. A QUIT command is no longer sent after a timeout.
1542 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
1543 during host lookups.
1545 74. The defer_ok option of callout verification was not working if it was used
1546 when verifying addresses in header lines, that is, for this case:
1548 verify = header_sender/callout=defer_ok
1550 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
1551 those file descriptors could be used for SMTP connections. If anything
1552 wrote to stderr (the example that came up was "warn" in embedded Perl), it
1553 could be sent to the SMTP client, causing chaos. The daemon now opens
1554 stdin, stdout, and stderr to /dev/null when it puts itself into the
1555 background.
1557 76. Arrange for output from Perl's "warn" command to be written to Exim's main
1558 log by default. The user can override this with suitable Perl magic.
1560 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
1561 the log message when discard triggers, was not working for the DATA ACL or
1562 for the non-SMTP ACL.
1564 78. Error message wording change in sieve.c.
1566 79. If smtp_accept_max_per_host was set, the number of connections could be
1567 restricted to fewer than expected, because the daemon was trying to set up
1568 a new connection before checking whether the processes handling previous
1569 connections had finished. The check for completed processes is now done
1570 earlier. On busy systems, this bug wouldn't be noticed because something
1571 else would have woken the daemon, and it would have reaped the completed
1572 process earlier.
1574 80. If a message was submitted locally by a user whose login name contained one
1575 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
1576 It caused a spool format error. I have fixed the spool reading code. A
1577 related problem was that the "from" clause in the Received: line became
1578 illegal because of the space(s). It is now covered by ${quote_local_part.
1580 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
1581 tables).
1583 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
1585 83. Patch from Sieve maintainer to fix unterminated string problem in
1586 "vacation" handling.
1588 84. Some minor changes to the Linux configuration files to help with other
1589 OS variants using glibc.
1591 85. One more patch for Sieve to update vacation handling to latest spec.
1594 ----------------------------------------------------
1595 See the note above about the 4.44 and 4.50 releases.
1596 ----------------------------------------------------
1599 Exim version 4.44
1600 -----------------
1602 1. Change 4.43/35 introduced a bug that caused file counts to be
1603 incorrectly computed when quota_filecount was set in an appendfile
1604 transport
1606 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
1607 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
1608 place.
1610 3. Give more explanation in the error message when the command for a transport
1611 filter fails to execute.
1613 4. There are several places where Exim runs a non-Exim command in a
1614 subprocess. The SIGUSR1 signal should be disabled for these processes. This
1615 was being done only for the command run by the queryprogram router. It is
1616 now done for all such subprocesses. The other cases are: ${run, transport
1617 filters, and the commands run by the lmtp and pipe transports.
1619 5. Some older OS have a limit of 256 on the maximum number of file
1620 descriptors. Exim was using setrlimit() to set 1000 as a large value
1621 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
1622 systems. I've change it so that if it can't get 1000, it tries for 256.
1624 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
1625 was an oversight, and furthermore, ever since the addition of extra
1626 controls (e.g. 4.43/32), the checks on when to allow different forms of
1627 "control" were broken. There should now be diagnostics for all cases when a
1628 control that does not make sense is encountered.
1630 7. $recipients is now available in the predata ACL (oversight).
1632 8. Tidy the search cache before the fork to do a delivery from a message
1633 received from the command line. Otherwise the child will trigger a lookup
1634 failure and thereby defer the delivery if it tries to use (for example) a
1635 cached ldap connection that the parent has called unbind on.
1637 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
1638 of $address_data from the recipient verification was clobbered by the
1639 sender verification.
1641 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
1642 was its contents. (It was OK if the option was not defined at all.)
1644 11. A "Completed" log line is now written for messages that are removed from
1645 the spool by the -Mrm option.
1647 12. $host_address is now set to the target address during the checking of
1648 ignore_target_hosts.
1650 13. When checking ignore_target_hosts for an ipliteral router, no host name was
1651 being passed; this would have caused $sender_host_name to have been used if
1652 matching the list had actually called for a host name (not very likely,
1653 since this list is usually IP addresses). A host name is now passed as
1654 "[x.x.x.x]".
1656 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
1657 code that specifies a non-restarting handler (typically sigaction() in
1658 modern systems) in an attempt to fix a rare and obscure crash bug.
1660 15. Narrowed the window for a race in the daemon that could cause it to ignore
1661 SIGCHLD signals. This is not a major problem, because they are used only to
1662 wake it up if nothing else does.
1664 16. A malformed maildirsize file could cause Exim to calculate negative values
1665 for the mailbox size or file count. Odd effects could occur as a result.
1666 The maildirsize information is now recalculated if the size or filecount
1667 end up negative.
1669 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
1670 support for a long time. Removed HAVE_SYS_VFS_H.
1672 18. Updated exipick to current release from John Jetmore.
1674 19. Allow an empty sender to be matched against a lookup in an address list.
1675 Previously the only cases considered were a regular expression, or an
1676 empty pattern.
1678 20. Exim went into a mad DNS lookup loop when doing a callout where the
1679 host was specified on the transport, if the DNS lookup yielded more than
1680 one IP address.
1682 21. The RFC2047 encoding function was originally intended for short strings
1683 such as real names; it was not keeping to the 75-character limit for
1684 encoded words that the RFC imposes. It now respects the limit, and
1685 generates multiple encoded words if necessary. To be on the safe side, I
1686 have increased the buffer size for the ${rfc2047: expansion operator from
1687 1024 to 2048 bytes.
1689 22. Failure to deliver a bounce message always caused it to be frozen, even if
1690 there was an errors_to setting on the router. The errors_to setting is now
1691 respected.
1693 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
1694 canonical form (fully expanded) before being placed in
1695 $sender_host_address.
1697 24. Updated eximstats to version 1.33
1699 25. Include certificate and key file names in error message when GnuTLS fails
1700 to set them up, because the GnuTLS error message doesn't include the name
1701 of the failing file when there is a problem reading it.
1703 26. Expand error message when OpenSSL has problems setting up cert/key files.
1704 As per change 25.
1706 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
1707 (this can affect the format of dates).
1709 28. exim_tidydb, when checking for the continued existence of a message for
1710 which it has found a message-specific retry record, was not finding
1711 messages that were in split spool directories. Consequently, it was
1712 deleting retry records that should have stayed in existence.
1714 29. eximstats updated to version 1.35
1715 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
1716 1.35 - bugfix such that pie charts by volume are generated correctly
1718 30. The SPA authentication driver was not abandoning authentication and moving
1719 on to the next authenticator when an expansion was forced to fail,
1720 contradicting the general specification for all authenticators. Instead it
1721 was generating a temporary error. It now behaves as specified.
1723 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
1724 (the order specifies the preference for clients). The order is now AES256,
1725 AES128, 3DES, ARCFOUR128.
1727 31. Small patch to Sieve code - explicitly set From: when generating an
1728 autoreply.
1730 32. Exim crashed if a remote delivery caused a very long error message to be
1731 recorded - for instance if somebody sent an entire SpamAssassin report back
1732 as a large number of 550 error lines. This bug was coincidentally fixed by
1733 increasing the size of one of Exim's internal buffers (big_buffer) that
1734 happened as part of the Exiscan merge. However, to be on the safe side, I
1735 have made the code more robust (and fixed the comments that describe what
1736 is going on).
1738 33. Some experimental protocols are using DNS PTR records for new purposes. The
1739 keys for these records are domain names, not reversed IP addresses. The
1740 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
1741 leaves it alone. Component reversal etc. now happens only for IP addresses.
1742 CAN-2005-0021
1744 34. The host_aton() function is supposed to be passed a string that is known
1745 to be a valid IP address. However, in the case of IPv6 addresses, it was
1746 not checking this. This is a hostage to fortune. Exim now panics and dies
1747 if the condition is not met. A case was found where this could be provoked
1748 from a dnsdb PTR lookup with an IPv6 address that had more than 8
1749 components; fortuitously, this particular loophole had already been fixed
1750 by change 4.50/55 or 4.44/33 above.
1752 If there are any other similar loopholes, the new check in host_aton()
1753 itself should stop them being exploited. The report I received stated that
1754 data on the command line could provoke the exploit when Exim was running as
1755 exim, but did not say which command line option was involved. All I could
1756 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
1757 running as the user.
1758 CAN-2005-0021
1760 35. There was a buffer overflow vulnerability in the SPA authentication code
1761 (which came originally from the Samba project). I have added a test to the
1762 spa_base64_to_bits() function which I hope fixes it.
1763 CAN-2005-0022
1765 36. The daemon start-up calls getloadavg() while still root for those OS that
1766 need the first call to be done as root, but it missed one case: when
1767 deliver_queue_load_max is set with deliver_drop_privilege. This is
1768 necessary for the benefit of the queue runner, because there is no re-exec
1769 when deliver_drop_privilege is set.
1771 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
1772 was in use, was not putting the data itself into the right store pool;
1773 consequently, it could be overwritten for a subsequent message in the same
1774 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
1775 the caching.)
1777 38. Sometimes the final signoff response after QUIT could fail to get
1778 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
1779 < 0 before doing a fflush(). This bug looks as though it goes back to the
1780 introduction of TLS in release 3.20, but "sometimes" must have been rare
1781 because the tests only now provoked it.
1784 Exim version 4.43
1785 -----------------
1787 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
1788 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
1789 Now it returns an integer. A number of calls were still expecting a T/F
1790 return. Fortuitously, in all cases, the tests worked in OK situations,
1791 which is the norm. However, things would have gone wrong on any write
1792 failures on the smtp file descriptor. This function is used when sending
1793 messages over SMTP and also when doing verify callouts.
1795 2. When Exim is called to do synchronous delivery of a locally submitted
1796 message (the -odf or -odi options), it no longer closes stderr before doing
1797 the delivery.
1799 3. Implemented the mua_wrapper option.
1801 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
1803 5. Implemented the functions header_remove(), header_testname(),
1804 header_add_at_position(), and receive_remove_recipient(), and exported them
1805 to local_scan().
1807 6. If an ACL "warn" statement specified the addition of headers, Exim already
1808 inserted X-ACL-Warn: at the start if there was no header name. However, it
1809 was not making this test for the second and subsequent header lines if
1810 there were newlines in the string. This meant that an invalid header could
1811 be inserted if Exim was badly configured.
1813 7. Allow an ACL "warn" statement to add header lines at the start or after all
1814 the Received: headers, as well as at the end.
1816 8. Added the rcpt_4xx retry error code.
1818 9. Added postmaster_mailfrom=xxx to callout verification option.
1820 10. Added mailfrom=xxxx to the callout verification option, for verify=
1821 header_sender only.
1823 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
1824 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
1826 12. Inserted some casts to stop certain compilers warning when using pointer
1827 differences as field lengths or precisions in printf-type calls (mostly
1828 affecting debugging statements).
1830 13. Added optional readline() support for -be (dynamically loaded).
1832 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
1833 same clock tick as a message's arrival, so that its received time was the
1834 same as the "first fail" time on the retry record, and that message
1835 remained on the queue past the ultimate address timeout, every queue runner
1836 would try a delivery (because it was past the ultimate address timeout) but
1837 after another failure, the ultimate address timeout, which should have then
1838 bounced the address, did not kick in. This was a "< instead of <=" error;
1839 in most cases the first failure would have been in the next clock tick
1840 after the received time, and all would be well.
1842 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
1843 being recognized when the domain list was tested by the match_domain
1844 condition in an expansion string.
1846 16. Added the ${str2b64: operator.
1848 17. Exim was always calling setrlimit() to set a large limit for the number of
1849 processes, without checking whether the existing limit was already
1850 adequate. (It did check for the limit on file descriptors.) Furthermore,
1851 errors from getrlimit() and setrlimit() were being ignored. Now they are
1852 logged to the main and panic logs, but Exim does carry on, to try to do its
1853 job under whatever limits there are.
1855 18. Imported PCRE 5.0.
1857 19. Trivial typo in log message " temporarily refused connection" (the leading
1858 space).
1860 20. If the log selector return_path_on_delivery was set and an address was
1861 redirected to /dev/null, the delivery process crashed because it assumed
1862 that a return path would always be set for a "successful" delivery. In this
1863 case, the whole delivery is bypassed as an optimization, and therefore no
1864 return path is set.
1866 21. Internal re-arrangement: the function for sending a challenge and reading
1867 a response while authentication was assuming a zero-terminated challenge
1868 string. It's now changed to take a pointer and a length, to allow for
1869 binary data in such strings.
1871 22. Added the cyrus_sasl authenticator (code supplied by MBM).
1873 23. Exim was not respecting finduser_retries when seeking the login of the
1874 uid under which it was called; it was always trying 10 times. (The default
1875 setting of finduser_retries is zero.) Also, it was sleeping after the final
1876 failure, which is pointless.
1878 24. Implemented tls_on_connect_ports.
1880 25. Implemented acl_smtp_predata.
1882 26. If the domain in control=submission is set empty, Exim assumes that the
1883 authenticated id is a complete email address when it generates From: or
1884 Sender: header lines.
1886 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
1887 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
1888 chown and chgrp in /bin and hostname in /usr/bin.
1890 28. Exim was keeping the "process log" file open after each use, just as it
1891 does for the main log. This opens the possibility of it remaining open for
1892 long periods when the USR1 signal hits a daemon. Occasional processlog
1893 errors were reported, that could have been caused by this. Anyway, it seems
1894 much more sensible not to leave this file open at all, so that is what now
1895 happens.
1897 29. The long-running daemon process does not normally write to the log once it
1898 has entered its main loop, and it closes the log before doing so. This is
1899 so that log files can straightforwardly be renamed and moved. However,
1900 there are a couple of unusual error situations where the daemon does write
1901 log entries, and I had neglected to close the log afterwards.
1903 30. The text of an SMTP error response that was received during a remote
1904 delivery was being truncated at 512 bytes. This is too short for some of
1905 the long messages that one sometimes sees. I've increased the limit to
1906 1024.
1908 31. It is now possible to make retry rules that apply only when a message has a
1909 specific sender, in particular, an empty sender.
1911 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
1912 it possible to be selective about when SMTP synchronization is enforced.
1914 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
1916 32. Implemented hosts_connection_nolog.
1918 33. Added an ACL for QUIT.
1920 34. Setting "delay_warning=" to disable warnings was not working; it gave a
1921 syntax error.
1923 35. Added mailbox_size and mailbox_filecount to appendfile.
1925 36. Added control = no_multiline_responses to ACLs.
1927 37. There was a bug in the logic of the code that waits for the clock to tick
1928 in the case where the clock went backwards by a substantial amount such
1929 that the microsecond fraction of "now" was more than the microsecond
1930 fraction of "then" (but the whole seconds number was less).
1932 38. Added support for the libradius Radius client library this is found on
1933 FreeBSD (previously only the radiusclient library was supported).
1936 Exim version 4.42
1937 -----------------
1939 1. When certain lookups returned multiple values in the form name=value, the
1940 quoting of the values was not always being done properly. Specifically:
1941 (a) If the value started with a double quote, but contained no whitespace,
1942 it was not quoted.
1943 (b) If the value contained whitespace other than a space character (i.e.
1944 tabs or newlines or carriage returns) it was not quoted.
1945 This fix has been applied to the mysql and pgsql lookups by writing a
1946 separate quoting function and calling it from the lookup code. The fix
1947 should probably also be applied to nisplus, ibase and oracle lookups, but
1948 since I cannot test any of those, I have not disturbed their existing code.
1950 2. A hit in the callout cache for a specific address caused a log line with no
1951 reason for rejecting RCPT. Now it says "Previous (cached) callout
1952 verification failure".
1954 3. There was an off-by-one bug in the queryprogram router. An over-long
1955 return line was truncated at 256 instead of 255 characters, thereby
1956 overflowing its buffer with the terminating zero. As well as fixing this, I
1957 have increased the buffer size to 1024 (and made a note to document this).
1959 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
1960 when Exim is waiting for an SMTP response from a remote server, Exim
1961 restarts its select() call on the socket, thereby resetting its timeout.
1962 This is not a problem when such interrupts are rare. Somebody set up a cron
1963 job to run exiwhat every 2 minutes, which is less than the normal select()
1964 timeout (5 or 10 minutes). This meant that the select() timeout never
1965 kicked in because it was always reset. I have fixed this by comparing the
1966 time when an interrupt arrives with the time at the start of the first call
1967 to select(). If more time than the timeout has elapsed, the interrupt is
1968 treated as a timeout.
1970 5. Some internal re-factoring in preparation for the addition of Sieve
1971 extensions (by MH). In particular, the "personal" test is moved to a
1972 separate function, and given an option for scanning Cc: and Bcc: (which is
1973 not set for Exim filters).
1975 6. When Exim created an email address using the login of the caller as the
1976 local part (e.g. when creating a From: or Sender: header line), it was not
1977 quoting the local part when it contained special characters such as @.
1979 7. Installed new OpenBSD configuration files.
1981 8. Reworded some messages for syntax errors in "and" and "or" conditions to
1982 try to make them clearer.
1984 9. Callout options, other than the timeout value, were being ignored when
1985 verifying sender addresses in header lines. For example, when using
1987 verify = header_sender/callout=no_cache
1989 the cache was (incorrectly) being used.
1991 10. Added a missing instance of ${EXE} to the exim_install script; this affects
1992 only the Cygwin environment.
1994 11. When return_path_on_delivery was set as a log selector, if different remote
1995 addresses in the same message used different return paths and parallel
1996 remote delivery occurred, the wrong values would sometimes be logged.
1997 (Whenever a remote delivery process finished, the return path value from
1998 the most recently started remote delivery process was logged.)
2000 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
2001 lines when AUTH and/or TLS are in use. This is the "received protocol"
2002 field. Exim used to use "asmtp" for authenticated SMTP, without any
2003 indication (in the protocol name) for TLS use. Now it follows the RFC and
2004 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
2005 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
2006 names appear in log lines as well as in Received: header lines.
2008 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
2009 extensions, and comparison tests, and to fix some bugs.
2011 14. Changes to the "personal" filter test:
2013 (1) The test was buggy in that it was just doing the equivalent of
2014 "contains" tests on header lines. For example, if a user's address was
2015 anne@some.where, the "personal" test would incorrectly be true for
2017 To: susanne@some.where
2019 This test is now done by extracting each address from the header in turn,
2020 and checking the entire address. Other tests that are part of "personal"
2021 are now done using regular expressions (for example, to check local parts
2022 of addresses in From: header lines).
2024 (2) The list of non-personal local parts in From: addresses has been
2025 extended to include "listserv", "majordomo", "*-request", and "owner-*",
2026 taken from the Sieve specification recommendations.
2028 (3) If the message contains any header line starting with "List-" it is
2029 treated as non-personal.
2031 (4) The test for "circular" in the Subject: header line has been removed
2032 because it now seems ill-conceived.
2034 15. Minor typos in src/EDITME comments corrected.
2036 16. Installed latest exipick from John Jetmore.
2038 17. If headers_add on a router specified a text string that was too long for
2039 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
2040 of string_sprintf() is now avoided.
2042 18. $message_body_size was not set (it was always zero) when running the DATA
2043 ACL and the local_scan() function.
2045 19. For the "mail" command in an Exim filter, no default was being set for
2046 the once_repeat time, causing a random time value to be used if "once" was
2047 specified. (If the value happened to be <= 0, no repeat happened.) The
2048 default is now 0s, meaning "never repeat". The "vacation" command was OK
2049 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
2050 (I found it when inspecting the code).
2052 20. There is now an overall timeout for performing a callout verification. It
2053 defaults to 4 times the callout timeout, which applies to individual SMTP
2054 commands during the callout. The overall timeout applies when there is more
2055 than one host that can be tried. The timeout is checked before trying the
2056 next host. This prevents very long delays if there are a large number of
2057 hosts and all are timing out (e.g. when the network connections are timing
2058 out). The value of the overall timeout can be changed by specifying an
2059 additional sub-option for "callout", called "maxwait". For example:
2061 verify = sender/callout=5s,maxwait=20s
2063 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
2064 to the end before writing, but this should make it even safer).
2066 22. Exim was forgetting that it had advertised PIPELINING for the second and
2067 subsequent messages on an SMTP connection. It was also not resetting its
2068 memory on STARTTLS and an internal HELO.
2070 23. When Exim logs an SMTP synchronization error within a session, it now
2071 records whether PIPELINING has been advertised or not.
2073 24. Added 3 instances of "(long int)" casts to time_t variables that were being
2074 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
2075 rather than long int.
2077 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
2079 26. Added the never_mail option to autoreply.
2082 Exim version 4.41
2083 -----------------
2085 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
2086 crash if the getsockname() call failed; this can happen if a connection is
2087 closed very soon after it is established. The problem was simply in the
2088 order in which certain operations were done, causing Exim to try to write
2089 to the SMTP stream before it had set up the file descriptor. The bug has
2090 been fixed by making things happen in the correct order.
2093 Exim version 4.40
2094 -----------------
2096 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
2097 before the connection was closed, thus losing the rejection response.
2099 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
2100 some early Solaris releases, but causes trouble in current releases where
2101 socklen_t is defined.
2103 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
2104 always exist.
2106 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
2107 configured.
2109 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
2110 that releasing the top of it at the end releases what was used for sub-
2111 expansions (unless the block got too big). However, discard this block if
2112 the first thing is a variable or header, so that we can use its block when
2113 it is dynamic (useful for very large $message_headers, for example).
2115 6. Lookups now cache *every* query, not just the most recent. A new, separate
2116 store pool is used for this. It can be recovered when all lookup caches are
2117 flushed. Lookups now release memory at the end of their result strings.
2118 This has involved some general refactoring of the lookup sources.
2120 7. Some code has been added to the store_xxx() functions to reduce the amount
2121 of flapping under certain conditions.
2123 8. log_incoming_interface used to affect only the <= reception log lines. Now
2124 it causes the local interface and port to be added to several more SMTP log
2125 lines, for example "SMTP connection from", and rejection lines.
2127 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
2129 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
2131 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
2132 could be overwritten at the end of the current message (or the start of a
2133 new message if it was set in a HELO ACL). The value is now preserved for
2134 the duration of the SMTP connection.
2136 12. If a transport had a headers_rewrite setting, and a matching header line
2137 contained an unqualified address, that address was qualified, even if it
2138 did not match any rewriting rules. The underlying bug was that the values
2139 of the flags that permit the existence of unqualified sender and recipient
2140 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
2141 non-local messages, and by -bnq for local messages) were not being
2142 preserved with the message after it was received.
2144 13. When Exim was logging an SMTP synchronization error, it could sometimes log
2145 "next input=" as part of the text comprising the host identity instead of
2146 the correct text. The code was using the same buffer for two different
2147 strings. However, depending on which order the printing function evaluated
2148 its arguments, the bug did not always show up. Under Linux, for example, my
2149 test suite worked just fine.
2151 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
2152 doesn't work with some older versions of Perl. It has been changed to "my",
2153 which in any case is probably the better facility to use.
2155 15. A really picky compiler found some instances of statements for creating
2156 error messages that either had too many or two few arguments for the format
2157 string.
2159 16. The size of the buffer for calls to the DNS resolver has been increased
2160 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
2161 for addresses that have a lot of PTR records. This alleviates a problem; it
2162 does not fully solve it.
2164 17. A dnsdb lookup for PTR records that receives more data than will fit in the
2165 buffer now truncates the list and logs the incident, which is the same
2166 action as happens when Exim is looking up a host name and its aliases.
2167 Previously in this situation something unpredictable would happen;
2168 sometimes it was "internal error: store_reset failed".
2170 18. If a server dropped the connection unexpectedly when an Exim client was
2171 using GnuTLS and trying to read a response, the client delivery process
2172 crashed while trying to generate an error log message.
2174 19. If a "warn" verb in an ACL added multiple headers to a message in a single
2175 string, for example:
2177 warn message = H1: something\nH2: something
2179 the text was added as a single header line from Exim's point of view
2180 though it ended up OK in the delivered message. However, searching for the
2181 second and subsequent header lines using $h_h2: did not work. This has been
2182 fixed. Similarly, if a system filter added multiple headers in this way,
2183 the routers could not see them.
2185 20. Expanded the error message when iplsearch is called with an invalid key to
2186 suggest using net-iplsearch in a host list.
2188 21. When running tests using -bh, any delays imposed by "delay" modifiers in
2189 ACLs are no longer actually imposed (and a message to that effect is
2190 output).
2192 22. If a "gecos" field in a passwd entry contained escaped characters, in
2193 particular, if it contained a \" sequence, Exim got it wrong when building
2194 a From: or a Sender: header from that name. A second bug also caused
2195 incorrect handling when an unquoted " was present following a character
2196 that needed quoting.
2198 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
2199 was not being matched caselessly.
2201 24. Arranged for all hyphens in the exim.8 source to be escaped with
2202 backslashes.
2204 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
2205 properly. Recipient callout cache records were still being keyed to include
2206 the sender, even when use_sender was set false. This led to far more
2207 callouts that were necessary. The sender is no longer included in the key
2208 when use_sender is false.
2210 26. Added "control = submission" modifier to ACLs.
2212 27. Added the ${base62d: operator to decode base 62 numbers.
2214 28. dnsdb lookups can now access SRV records.
2216 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
2217 the configuration file.
2219 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
2220 (-v) mode. This makes the output for a verbose queue run more intelligible.
2222 31. Added a use_postmaster feature to recipient callouts.
2224 32. Added the $body_zerocount variable, containing the number of binary zero
2225 bytes in the message body.
2227 33. The time of last modification of the "new" subdirectory is now used as the
2228 "mailbox time last read" when there is a quota error for a maildir
2229 delivery.
2231 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
2233 35. Added +ignore_unknown as a special item in host lists.
2235 36. Code for decoding IPv6 addresses in host lists is now included, even if
2236 IPv6 support is not being compiled. This fixes a bug in which an IPv6
2237 address was recognized as an IP address, but was then not correctly decoded
2238 into binary, causing unexpected and incorrect effects when compared with
2239 another IP address.
2242 Exim version 4.34
2243 -----------------
2245 1. Very minor rewording of debugging text in manualroute to say "list of
2246 hosts" instead of "hostlist".
2248 2. If verify=header_syntax was set, and a header line with an unqualified
2249 address (no domain) and a large number of spaces between the end of the
2250 name and the colon was received, the reception process suffered a buffer
2251 overflow, and (when I tested it) crashed. This was caused by some obsolete
2252 code that should have been removed. The fix is to remove it!
2254 3. When running in the test harness, delay a bit after writing a bounce
2255 message to get a bit more predictability in the log output.
2257 4. Added a call to search_tidyup() just before forking a reception process. In
2258 theory, someone could use a lookup in the expansion of smtp_accept_max_
2259 per_host which, without the tidyup, could leave open a database connection.
2261 5. Added the variables $recipient_data and $sender_data which get set from a
2262 lookup success in an ACL "recipients" or "senders" condition, or a router
2263 "senders" option, similar to $domain_data and $local_part_data.
2265 6. Moved the writing of debug_print from before to after the "senders" test
2266 for routers.
2268 7. Change 4.31/66 (moving the time when the Received: is generated) caused
2269 problems for message scanning, either using a data ACL, or using
2270 local_scan() because the Received: header was not generated till after they
2271 were called (in order to set the time as the time of reception completion).
2272 I have revised the way this works. The header is now generated after the
2273 body is received, but before the ACL or local_scan() are called. After they
2274 are run, the timestamp in the header is updated.
2277 Exim version 4.33
2278 -----------------
2280 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
2281 before starting a queue runner without re-exec. This happened only when
2282 deliver_drop_privilege was set or when the Exim user was set to root. The
2283 effect of the bug was that timeouts during subsequent deliveries caused
2284 crashes instead of being properly handled. The handler is now left at its
2285 default (and expected) setting.
2287 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
2288 message, again when deliver_drop_privilege is set or Exim is run as root.
2289 The bug described in (1) was not present in this case, but the tidying up
2290 of the other signals was missing. I have made the two cases consistent.
2292 3. The ignore_target_hosts setting on a manualroute router was being ignored
2293 for hosts that were looked up using the /MX notation.
2295 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
2296 in domain lists.
2298 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
2299 operated on the sender address. After changing the $sender_address to <>
2300 for the sender address verify, Exim was re-instated it as the original
2301 (before rewriting) address, but remembering that it had rewritten it, so it
2302 wasn't rewriting it again. This bug also had the effect of breaking the
2303 sender address verification caching when the sender address was rewritten.
2305 6. The ignore_target_hosts option was being ignored by the ipliteral router.
2306 This has been changed so that if the ip literal address matches
2307 ignore_target_hosts, the router declines.
2309 7. Added expansion conditions match_domain, match_address, and match_local_
2310 part (NOT match_host).
2312 8. The placeholder for the Received: header didn't have a length field set.
2314 9. Added code to Exim itself and to exim_lock to test for a specific race
2315 condition that could lead to file corruption when using MBX delivery. The
2316 issue is with the lockfile that is created in /tmp. If this file is removed
2317 after a process has opened it but before that process has acquired a lock,
2318 there is the potential for a second process to recreate the file and also
2319 acquire a lock. This could lead to two Exim processes writing to the file
2320 at the same time. The added code performs the same test as UW imapd; it
2321 checks after acquiring the lock that its file descriptor still refers to
2322 the same named file.
2324 10. The buffer for building added header lines was of fixed size, 8192 bytes.
2325 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
2326 when Exim is built.
2328 11. Added the smtp_active_hostname option. If used, this will typically be made
2329 to depend on the incoming interface address. Because $interface_address is
2330 not set up until the daemon has forked a reception process, error responses
2331 that can happen earlier (such as "too many connections") no longer contain
2332 a host name.
2334 12. If an expansion in a condition on a "warn" statement fails because a lookup
2335 defers, the "warn" statement is abandoned, and the next ACL statement is
2336 processed. Previously this caused the whole ACL to be aborted.
2338 13. Added the iplsearch lookup type.
2340 14. Added ident_timeout as a log selector.
2342 15. Added tls_certificate_verified as a log selector.
2344 16. Added a global option tls_require_ciphers (compare the smtp transport
2345 option of the same name). This controls incoming TLS connections.
2347 17. I finally figured out how to make tls_require_ciphers do a similar thing
2348 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
2349 before starting the TLS session.
2351 18. Tabs are now shown as \t in -bP output.
2353 19. If the log selector return_path_on_delivery was set, Exim crashed when
2354 bouncing a message because it had too many Received: header lines.
2356 20. If two routers both had headers_remove settings, and the first one included
2357 a superfluous trailing colon, the final name in the first list and the
2358 first name in the second list were incorrectly joined into one item (with a
2359 colon in the middle).
2362 Exim version 4.32
2363 -----------------
2365 1. Added -C and -D options to the exinext utility, mainly to make it easier
2366 to include in the automated testing, but these could be helpful when
2367 multiple configurations are in use.
2369 2. The exinext utility was not formatting the output nicely when there was
2370 an alternate port involved in the retry record key, nor when there was a
2371 message id as well (for retries that were specific to a specific message
2372 and a specific host). It was also confused by IPv6 addresses, because of
2373 the additional colons they contain. I have fixed the IPv4 problem, and
2374 patched it up to do a reasonable job for IPv6.
2376 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
2377 delivery, the log line now contains "pipelined" if PIPELINING was used.
2379 4. An SMTP transport process used to panic and die if the bind() call to set
2380 an explicit outgoing interface failed. This has been changed; it is now
2381 treated in the same way as a connect() failure.
2383 5. A reference to $sender_host_name in the part of a conditional expansion
2384 that was being skipped was still causing a DNS lookup. This no longer
2385 occurs.
2387 6. The def: expansion condition was not recognizing references to header lines
2388 that used bh_ and bheader_.
2390 7. Added the _cache feature to named lists.
2392 8. The code for checking quota_filecount in the appendfile transport was
2393 allowing one more file than it should have been.
2395 9. For compatibility with Sendmail, the command line option
2397 -prval:sval
2399 is equivalent to
2401 -oMr rval -oMs sval
2403 and sets the incoming protocol and host name (for trusted callers). The
2404 host name and its colon can be omitted when only the protocol is to be set.
2405 Note the Exim already has two private options, -pd and -ps, that refer to
2406 embedded Perl. It is therefore impossible to set a protocol value of "d" or
2407 "s", but I don't think that's a major issue.
2409 10. A number of refactoring changes to the code, none of which should affect
2410 Exim's behaviour:
2412 (a) The number of logging options was getting close to filling up the
2413 32-bit word that was used as a bit map. I have split them into two classes:
2414 those that are passed in the argument to log_write(), and those that are
2415 only ever tested independently outside of that function. These are now in
2416 separate 32-bit words, so there is plenty of room for expansion again.
2417 There is no change in the user interface or the logging behaviour.
2419 (b) When building, for example, log lines, the code previously used a
2420 macro that called string_cat() twice, in order to add two strings. This is
2421 not really sufficiently general. Furthermore, there was one instance where
2422 it was actually wrong because one of the argument was used twice, and in
2423 one call a function was used. (As it happened, calling the function twice
2424 did not affect the overall behaviour.) The macro has been replaced by a
2425 function that can join an arbitrary number of extra strings onto a growing
2426 string.
2428 (c) The code for expansion conditions now uses a table and a binary chop
2429 instead of a serial search (which was left over from when there were very
2430 few conditions). Also, it now recognizes conditions like "pam" even when
2431 the relevant support is not compiled in: a suitably worded error message is
2432 given if an attempt is made to use such a condition.
2434 11. Added ${time_interval:xxxxx}.
2436 12. A bug was causing one of the ddress fields not to be passed back correctly
2437 from remote delivery subprocesses. The field in question was not being
2438 subsequently used, so this caused to problems in practice.
2440 13. Added new log selectors queue_time and deliver_time.
2442 14. Might have fixed a bug in maildirsizefile handling that threw up
2443 "unexpected character" debug warnings, and recalculated the data
2444 unnecessarily. In any case, I expanded the warning message to give more
2445 information.
2447 15. Added the message "Restricted characters in address" to the statements in
2448 the default ACL that block characters like @ and % in local parts.
2450 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
2451 Three changes have been made:
2453 (a) There was a serious bug; a negative response to MAIL caused the whole
2454 recipient domain to be cached as invalid, thereby blocking all messages
2455 to all local parts at the same domain, from all senders. This bug has
2456 been fixed. The domain is no longer cached after a negative response to
2457 MAIL if the sender used is not empty.
2459 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
2460 been restored.
2462 (c) A new callout option, "use_sender" has been added for people who want
2463 the modified behaviour.
2466 Exim version 4.31
2467 -----------------
2469 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
2470 Larry Rosenman.
2472 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
2473 indeed breaks things for older releases.
2475 3. Added additional logging to the case where there is a problem reading data
2476 from a filter that is running in a subprocess using a pipe, in order to
2477 try to track down a specific problem.
2479 4. Testing facility fudge: when running in the test harness and attempting
2480 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
2481 getting "No route to host". Convert this to a timeout.
2483 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
2484 warning.
2486 6. Some OS don't have socklen_t but use size_t instead. This affects the
2487 fifth argument of getsockopt() amongst other things. This is now
2488 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
2489 can be set for individual OS. I have set it for SunOS5, OSF1, and
2490 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
2491 some earlier ones do not.
2493 7. Change 4.30/15 was not doing the test caselessly.
2495 8. The standard form for an IPv6 address literal was being rejected by address
2496 parsing in, for example, MAIL and RCPT commands. An example of this kind of
2497 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
2498 this, as well as the form without the "IPv6" on the front (but only when
2499 address literals are enabled, of course).
2501 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
2503 10. Exim crashed if a message with an empty sender address specified by -f
2504 encountered a router with an errors_to setting. This could be provoked only
2505 by a command such as
2507 exim -f "" ...
2509 where an empty string was supplied; "<>" did not hit this bug.
2511 11. Installed PCRE release 4.5.
2513 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
2514 remained set. It is now erased.
2516 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
2517 times from message ids (which are base 36 rather than the normal 62).
2519 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
2520 were being counted as actual protocol errors, and logged if the log
2521 selector +smtp_protocol_error was set. One cannot be perfect in this test,
2522 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
2523 and DATA following a set of rejected RCPTs do not count as protocol errors.
2524 In other words, Exim assumes they were pipelined, though this may not
2525 actually be the case. Of course, in all cases the client gets an
2526 appropriate error code.
2528 15. If a lookup fails in an ACL condition, a message about the failure may
2529 be available; it is used if testing the ACL cannot continue, because most
2530 such messages specify what the cause of the deferral is. However, some
2531 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
2532 that caused an old message to be retained and used if a later statement
2533 caused a defer, replacing the real cause of the deferral.
2535 16. If an IP address had so many PTR records that the DNS lookup buffer
2536 was not large enough to hold them, Exim could crash while trying to process
2537 the truncated data. It now detects and logs this case.
2539 17. Further to 4.21/58, another change has been made: if (and only if) the
2540 first line of a message (the first header line) ends with CRLF, a bare LF
2541 in a subsequent header line has a space inserted after it, so as not to
2542 terminate the header.
2544 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
2545 unnecessarily, used atoi() instead of strtol(), and didn't check the
2546 termination when getting file sizes from file names by regex.
2548 19. Completely re-implemented the support for maildirsize files, in the light
2549 of a number of problems with the previous contributed implementation
2550 (4.30/29). In particular:
2552 . If the quota is zero, the maildirsize file is maintained, but no quota is
2553 imposed.
2555 . If the maildir directory does not exist, it is created before any attempt
2556 to write a maildirsize file.
2558 . The quota value in the file is just a cache; if the quota is changed in
2559 the transport, the new value overrides.
2561 . A regular expression is available for excluding directories from the
2562 count.
2564 20. The autoreply transport checks the characters in options that define the
2565 message's headers; it allows continued headers, but it was checking with
2566 isspace() after an embedded newline instead of explicitly looking for a
2567 space or a tab.
2569 21. If all the "regular" hosts to which an address was routed had passed their
2570 expiry times, and had not reached their retry times, the address was
2571 bounced, even if fallback hosts were defined. Now Exim should go on to try
2572 the fallback hosts.
2574 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
2575 equivalent code in the SMTP transport. Some hosts send humungous responses
2576 to HELO/EHLO, more than 1024 it seems.
2578 23. Refactoring: code in filter.c used (void *) for "any old type" but this
2579 gives compiler warnings in some environments. I've now done it "properly",
2580 using a union.
2582 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
2583 (because of problems with the built-in one) was declared to return uschar *
2584 instead of char *, causing compiler failure.
2586 25. Fixed a file descriptor leak when processing alias/forward files.
2588 26. Fixed a minor format string issue in dbfn.c.
2590 27. Typo in exim.c: ("dmbnz" for "dbmnz").
2592 28. If a filter file refered to $h_xxx or $message_headers, and the headers
2593 contained RFC 2047 "words", Exim's memory could, under certain conditions,
2594 become corrupted.
2596 29. When a sender address is verified, it is cached, to save repeating the test
2597 when there is more than one recipient in a message. However, when the
2598 verification involves a callout, it is possible for different callout
2599 options to be set for different recipients. It is too complicated to keep
2600 track of this in the cache, so now Exim always runs a verification when a
2601 callout is required, relying on the callout cache for the optimization.
2602 The overhead is duplication of the address routing, but this should not be
2603 too great.
2605 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
2606 to be verified with callout=postmaster, and the main callout worked but the
2607 postmaster check failed, the verification correctly failed. However, if a
2608 subsequent RCPT command asked for sender verification *without* the
2609 postmaster check, incorrect caching caused this verification also to fail,
2610 incorrectly.
2612 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
2613 it was not caching the DNS options (qualify_single, search_parents) that
2614 were used when the lookup failed. A subsequent lookup with different
2615 options therefore always gave the same answer, though there were cases
2616 where it should not have. (Example: a "domains = !$mx_any" option on a
2617 dnslookup router: the "domains" option is always processed without any
2618 widening, but the router might have qualify_single set.) Now Exim uses the
2619 cached value only when the same options are set.
2621 32. Added John Jetmore's "exipick" utility to the distribution.
2623 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
2624 than a timeout (e.g. a certificate is required, and is not provided), an
2625 Exim server now closes the connection immediately. Previously it waited for
2626 the client to close - but if the client is SSL, it seems that they each
2627 wait for each other, leading to a delay before one of them times out.
2629 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
2630 maintained 0.8.x compatibility because I don't think many are using it, and
2631 it is clearly obsolete.
2633 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
2634 transport.
2636 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
2637 client certificate was expired. A simple patch fixes this, though I don't
2638 understand the full logic of why the verify callback is called multiple
2639 times.
2641 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
2642 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
2643 which causes problems with some clients (such as the Certicom SSL Plus
2644 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
2645 disables the coutermeasure allowing Eudora to connect."
2647 38. Exim was not checking that a write() to a log file succeeded. This could
2648 lead to Bad Things if a log got too big, in particular if it hit a file
2649 size limit. Exim now panics and dies if it cannot write to a log file, just
2650 as it does if it cannot open a log file.
2652 39. Modified OS/Makefile-Linux so that it now contains
2656 The two -D definitions ensure that Exim is compiled with large file
2657 support, which makes it possible to handle log files that are bigger than
2658 2^31.
2660 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
2661 instance) a domain was checked against a named list that involved a lookup,
2662 causing $domain_data to be set, then another domain was checked against the
2663 same list, then the first domain was re-checked, the value of $domain_data
2664 after the final check could be wrong. In particular, if the second check
2665 failed, it could be set empty. This bug probably also applied to
2666 $localpart_data.
2668 41. The strip_trailing_dot option was not being applied to the address given
2669 with the -f command-line option.
2671 42. The code for reading a message's header from the spool was incrementing
2672 $received_count, but never initializing it. This meant that the value was
2673 incorrect (doubled) while delivering a message in the same process in which
2674 it was received. In the most common configuration of Exim, this never
2675 happens - a fresh exec is done - but it can happen when
2676 deliver_drop_privilege is set.
2678 43. When Exim logs an SMTP synchronization error - client data sent too soon -
2679 it now includes up to 150 characters of the unexpected data in the log
2680 line.
2682 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
2683 and building data strings. The size of both of these buffers was 10 000
2684 bytes - far larger than anybody would *ever* want, thought I. Needless to
2685 say, somebody hit the limit. I have increased the maximum line length to
2686 20 000 and the maximum data length of concatenated lines to 100 000. I have
2687 also fixed two bugs, because there was no checking on these buffers. Tsk,
2688 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
2689 buffer is too small.
2691 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
2692 lsearch lookups. Now it does.
2694 46. When parsing a route_list item in a manualroute router, a fixed-length
2695 buffer was used for the list of hosts. I made this 1024 bytes long,
2696 thinking that nobody would ever have a list of hosts that long. Wrong.
2697 Somebody had a whole pile of complicated expansion conditions, and the
2698 string was silently truncated, leading to an expansion error. It turns out
2699 that it is easier to change to an unlimited length (owing to other changes
2700 that have happened since this code was originally written) than to build
2701 structure for giving a limitation error. The length of the item that
2702 expands into the list of hosts is now unlimited.
2704 47. The lsearch lookup could not handle data where the length of text line was
2705 more than 4095 characters. Such lines were truncated, leading to shortened
2706 data being returned. It should now handle lines of any length.
2708 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
2709 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
2710 ACL").
2712 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
2713 replacing strings such as BIN_DIRECTORY in a source file: the replacement
2714 no longer happens in comment lines. A list of replacements is now placed
2715 at the head of all of the source files, except those whose only change is
2716 to replace PERL_COMMAND in the very first #! line.
2718 50. Replaced the slow insertion sort in queue.c, for sorting the list of
2719 messages on the queue, with a bottom-up merge sort, using code contributed
2720 by Michael Haardt. This should make operations like -bp somewhat faster on
2721 large queues. It won't affect queue runners, except when queue_run_in_order
2722 is set.
2724 51. Installed eximstats 1.31 in the distribution.
2726 52. Added support for SRV lookups to the dnslookup router.
2728 53. If an ACL referred to $message_body or $message_body_end, the value was not
2729 reset for any messages that followed in the same SMTP session.
2731 54. The store-handling optimization for building very long strings was not
2732 differentiating between the different store pools. I don't think this
2733 actually made any difference in practice, but I've tidied it.
2735 55. While running the routers to verify a sender address, $sender_address
2736 was still set to the sender address. This is wrong, because when routing to
2737 send a bounce to the sender, it would be empty. Therefore, I have changed
2738 it so that, while verifying a sender address, $sender_address is set to <>.
2739 (There is no change to what happens when verifying a recipient address.)
2741 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
2742 target A or AAAA records (if not already returned) without resetting the
2743 qualify_single or search_parents options of the DNS resolver. These are
2744 inappropriate in this case because the targets of MX and SRV records must
2745 be FQDNs. A broken DNS record could cause trouble if it happened to have a
2746 target that, when qualified, matched something in the local domain. These
2747 two options are now turned off when doing these lookups.
2749 57. It seems that at least some releases of Reiserfs (which does not have the
2750 concept of a fixed number of inodes) returns zero and not -1 for the
2751 number of available inodes. This interacted badly with check_spool_inodes,
2752 which assumed that -1 was the "no such thing" setting. What I have done is
2753 to check that the total number of inodes is greater than zero before doing
2754 the test of how many are available.
2756 58. When a "warn" ACL statement has a log_message modifier, the message is
2757 remembered, and not repeated. This is to avoid a lot of repetition when a
2758 message has many recipients that cause the same warning to be written.
2759 Howewer, Exim was preserving the list of already written lines for an
2760 entire SMTP session, which doesn't seem right. The memory is now reset if a
2761 new message is started.
2763 59. The "rewrite" debugging flag was not showing the result of rewriting in the
2764 debugging output unless log_rewrite was also set.
2766 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
2767 of (int)(handle) when we know that handle contains (void *)(-1).
2769 61. The Exim daemon panic-logs an error return when it closes the incoming
2770 connection. However "connection reset by peer" seems to be common, and
2771 isn't really an error worthy of noting specially, so that particular error
2772 is no long logged.
2774 62. When Exim is trying to find all the local interfaces, it used to panic and
2775 die if the ioctl to get the interface flags failed. However, it seems that
2776 on at least one OS (Solaris 9) it is possible to have an interface that is
2777 included in the list of interfaces, but for which you get a failure error
2778 for this call. This happens when the interface is not "plumbed" into a
2779 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
2780 failure of the "get flags" call assumes that the interface is down.
2782 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
2783 makes life easier for people who are doing arithmetic on fields extracted
2784 from dates, where you often get leading zeros that should not be
2785 interpreted as octal.
2787 64. Added qualify_domain to the redirect router, to override the global
2788 setting.
2790 65. If a pathologically long header line contained very many addresses (the
2791 report of this problem mentioned 10 000) and each of them was rewritten,
2792 Exim could use up a very large amount of memory. (It kept on making new
2793 copies of the header line as it rewrote, and never released the old ones.)
2794 At the expense of a bit more processing, the header rewriting function has
2795 been changed so that it no longer eats memory in this way.
2797 66. The generation of the Received: header has been moved from the time that a
2798 message starts to be received, to the time that it finishes. The timestamp
2799 in the Received: header should now be very close to that of the <= log
2800 line. There are two side-effects of this change:
2802 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
2803 logged header lines no longer include the local Received: line, because
2804 it has not yet been created. The same applies to a copy of the message
2805 that is returned to a non-SMTP sender when a message is rejected.
2807 (b) When a filter file is tested using -bf, no additional Received: header
2808 is added to the test message. After some thought, I decided that this
2809 is a bug fix.
2811 This change does not affect the value of $received_for. It is still set
2812 after address rewriting, but before local_scan() is called.
2814 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
2816 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
2817 gave an unhelpful panic error message, and a defer error. I have managed to
2818 change this behaviour so that it now rejects any supplied certificate,
2819 which seems right, as the list of acceptable certificates is empty.
2821 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
2822 gave an unhelpful defer error. I have not managed to make this reject any
2823 supplied certificates, but the error message it gives is "no certificate
2824 supplied", which is not helpful.
2826 70. exigrep's output now also includes lines that are not associated with any
2827 message, but which match the given pattern. Implemented by a patch from
2828 Martin Sluka, which also tidied up the Perl a bit.
2830 71. Recipient callout verification, like sender verification, was using <> in
2831 the MAIL FROM command. This isn't really the right thing, since the actual
2832 sender may affect whether the remote host accepts the recipient or not. I
2833 have changed it to use the actual sender in the callout; this means that
2834 the cache record is now keyed on a recipient/sender pair, not just the
2835 recipient address. There doesn't seem to be a real danger of callout loops,
2836 since a callout by the remote host to check the sender would use <>.
2837 [SEE ABOVE: changed after hitting problems.]
2839 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
2840 temporary errors. However, in the case of such a code being given after
2841 the end of a data transmission (i.e. after ".") Exim was failing to write
2842 a retry record for the message. (Yes, there was some broken host that was
2843 actually sending 8xx at this point.)
2845 73. An unknown lookup type in a host list could cause Exim to panic-die when
2846 the list was checked. (An example that provoked this was putting <; in the
2847 middle of a list instead of at the start.) If this happened during a DATA
2848 ACL check, a -D file could be left lying around. This kind of configuration
2849 error no longer causes Exim to die; instead it causes a defer errror. The
2850 incident is still logged to the main and panic logs.
2852 74. Buglet left over from Exim 3 conversion. The message "too many messages
2853 in one connection" was written to the rejectlog but not the mainlog, except
2854 when address rewriting (yes!) was being logged.
2856 75. Added write_rejectlog option.
2858 76. When a system filter was run not as root (that is, when system_filter_user
2859 was set), the values of the $n variables were not being returned to the
2860 main process; thus, they were not subsequently available in the $sn
2861 variables.
2863 77. Added +return_path_on_delivery log selector.
2865 78. A connection timeout was being treated differently from recipients deferred
2866 when testing hosts_max_try with a message that was older than the host's
2867 retry timeout. (The host should not be counted, thus allowing all hosts to
2868 be tried at least once before bouncing.) This may have been the cause of an
2869 occasionally reported bug whereby a message would remain on the queue
2870 longer than the retry timeout, but would be bounced if a delivery was
2871 forced. I say "may" because I never totally pinned down the problem;
2872 setting up timeout/retry tests is difficult. See also the next item.
2874 79. The ultimate address timeout was not being applied to errors that involved
2875 a combination of host plus message (for example, a timeout on a MAIL
2876 command). When an address resolved to a number of possible hosts, and they
2877 were not all tried for each delivery (e.g. because of hosts_max_try), a
2878 message could remain on the queue longer than the retry timeout.
2880 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
2881 Haardt.
2883 81. Fixed an obscure SMTP outgoing bug which required at least the following
2884 conditions: (a) there was another message waiting for the same server;
2885 (b) the server returned 5xx to all RCPT commands in the first message so
2886 that the message was not completed; (c) the server dropped the connection
2887 or gave a negative response to the RSET that Exim sends to abort the
2888 transaction. The observed case was a dropped connection after DATA that had
2889 been sent in pipelining mode. That is, the server had advertised PIPELINING
2890 but was not implementing it correctly. The effect of the bug was incorrect
2891 behaviour, such as trying another host, and this could lead to a crash.
2894 Exim version 4.30
2895 -----------------
2897 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
2898 and daemon.c were passed as pointers to ints; they should have been
2899 pointers to socklen_t variables (which are typically unsigned ints).
2901 2. Some signed/unsigned type warnings in the os.c file for Linux have been
2902 fixed.
2904 3. Fixed a really odd bug that affected only the testing scheme; patching a
2905 certain fixed string in the binary changed the value of another string that
2906 happened to be identical to the end of the original first string.
2908 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
2909 name", it returns that address as the IP address. On some operating
2910 systems (e.g. Solaris), it also passes back the IP address string as the
2911 "host name". However, on others (e.g. Linux), it passes back an empty
2912 string. Exim wasn't checking for this, and was changing the host name to an
2913 empty string, assuming it had been canonicized.
2915 5. Although rare, it is permitted to have more than one PTR record for a given
2916 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
2917 all the names associated with an address, because they do in Solaris.
2918 However, it seems that they do not in Linux for data that comes from the
2919 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
2920 I found this out when I moved to a new Linux workstation and tried to run
2921 the Exim test suite.
2923 To get round this problem I have changed the code so that it now does its
2924 own call to the DNS to look up PTR records when searching for a host name.
2925 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
2926 addresses that are only in /etc/hosts are still found.
2928 This behaviour is, however, controlled by an option called host_lookup_
2929 order, which defaults to "bydns:byaddr". If people want to use the other
2930 order, or indeed, just use one or the other means of lookup, they can
2931 specify it in this variable.
2933 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
2934 some operating systems, this comes back from gethostbyaddr() as an empty
2935 string, and this is what Exim used to test for. However, it seems that in
2936 other systems, "." is yielded. Exim now tests for this case too.
2938 7. The values of check_spool_space and check_log_space are now held internally
2939 as a number of kilobytes instead of an absolute number of bytes. If a
2940 numbers is specified without 'K' or 'M', it is rounded up to the nearest
2941 kilobyte. This means that much larger values can be stored.
2943 8. Exim monitor: an attempt to get the action menu when not actually pointing
2944 at a message produces an empty menu entitled "No message selected". This
2945 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
2946 no entries in it ("Shell widget menu has zero width and/or height"). So I
2947 have added a single, blank menu entry in this case.
2949 9. Added ${quote_local_part.
2951 10. MIME decoding is now applied to the contents of Subject: header lines when
2952 they are logged.
2954 11. Now that a reference to $sender_host_address automatically causes a reverse
2955 lookup to occur if necessary (4.13/18), there is no need to arrange for a
2956 host lookup before query-style lookups in lists that might use this
2957 variable. This has therefore been abolished, and the "net-" prefix is no
2958 longer necessary for query-style lookups.
2960 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
2961 have been a typo for LFLAGS, so it has been changed.
2963 13. The install script calls Exim with "-C /dev/null" in order to find the
2964 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
2965 to be output. Howeve, since Exim outputs its version number before the
2966 error, it didn't break the script. It just looked ugly. I fixed this by
2967 always allowing "-C /dev/null" if the caller is root.
2969 14. Ignore overlarge ACL variable number when reading spool file - insurance
2970 against a later release with more variables having written the file.
2972 15. The standard form for an IPv6 address literal was being rejected by EHLO.
2973 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
2974 this, as well as the form without the "IPv6" on the front.
2976 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
2977 OS/Makefile-Darwin file.
2979 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
2980 only to LDAP libraries that do not have LDAP_OPT_DEREF.
2982 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
2983 variable. However, some OS use ints for inodes. I've added cast to long int
2984 to get rid of the compiler warning.
2986 19. I had forgotten to lock out "/../" in configuration file names when
2987 ALT_CONFIG_PREFIX was set.
2989 20. Routers used for verification do not need to specify transports. However,
2990 if such a router generated a host list, and callout was configured, Exim
2991 crashed, because it could not find a port number from the (non-existent)
2992 transport. It now assumes port 25 in this circumstance.
2994 21. Added the -t option to exigrep.
2996 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
2997 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
2998 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
2999 them is set, LOOKUP_LSEARCH is forced.
3001 23. "exim -bV" now outputs a list of lookups that are included in the binary.
3003 24. Added sender and host information to the "rejected by local_scan()" log
3004 line; previously there was no indication of these.
3006 25. Added .include_if_exists.
3008 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
3009 Linux. It turns out that not all file systems support this. Apparently some
3010 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
3011 do it.) To cope with this, the error EINVAL, which means that sync-ing is
3012 not supported on the file descriptor, is now ignored when Exim is trying to
3013 sync a directory. This applies only to Linux.
3015 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
3017 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
3018 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
3019 this.
3021 29. Added support for maildirsize files from supplied patch (modified a bit).
3023 30. The use of :fail: followed by an empty string could lead Exim to respond to
3024 sender verification failures with (e.g.):
3026 550 Verification failed for <xxx>
3027 550 Sender verify failed
3029 where the first response line was missing the '-' that indicates it is not
3030 the final line of the response.
3032 31. The loop for finding the name of the user that called Exim had a hardwired
3033 limit of 10; it now uses the value of finduser_retries, which is used for
3034 all other user lookups.
3036 32. Added $received_count variable, available in data and not_smtp ACLs, and at
3037 delivery time.
3039 33. Exim was neglecting to zero errno before one call of strtol() when
3040 expanding a string and expecting an integer value. On some systems this
3041 resulted in spurious "integer overflow" errors. Also, it was casting the
3042 result into an int without checking.
3044 34. Testing for a connection timeout using "timeout_connect" in the retry rules
3045 did not work. The code looks as if it has *never* worked, though it appears
3046 to have been documented since at least releast 1.62. I have made it work.
3048 35. The "timeout_DNS" error in retry rules, also documented since at least
3049 1.62, also never worked. As it isn't clear exactly what this means, and
3050 clearly it isn't a major issue, I have abolished the feature by treating it
3051 as "timeout", and writing a warning to the main and panic logs.
3053 36. The display of retry rules for -brt wasn't always showing the error code
3054 correctly.
3056 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
3057 timeout_connect_A, timeout_connect_MX.
3059 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
3060 to the empty sender.
3062 39. The daemon was not analysing the content of -oX till after it had closed
3063 stderr and disconnected from the controlling terminal. This meant that any
3064 syntax errors were only noted on the panic log, and the return code from
3065 the command was 0. By re-arranging the code a little, I've made the
3066 decoding happen first, so such errors now appear on stderr, and the return
3067 code is 1. However, the actual setting up of the sockets still happens in
3068 the disconnected process, so errors there are still only recorded on the
3069 panic log.
3071 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
3072 connections (as happens on some IP stacks) was logged at start up time as
3073 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
3074 it from "IPv6 and IPv4", which means that two separate sockets are being
3075 used.
3077 41. The debug output for gethostbyname2() or getipnodebyname() failures now
3078 says whether AF_INET or AF_INET6 was passed as an argument.
3080 42. Exiwhat output was messed up when time zones were included in log
3081 timestamps.
3083 43. Exiwhat now gives more information about the daemon's listening ports,
3084 and whether -tls-on-connect was used.
3086 44. The "port" option of the smtp transport is now expanded.
3088 45. A "message" modifier in a "warn" statement in a non-message ACL was being
3089 silently ignored. Now an error message is written to the main and panic
3090 logs.
3092 46. There's a new ACL modifier called "logwrite" which writes to a log file
3093 as soon as it is encountered.
3095 47. Added $local_user_uid and $local_user_gid at routing time.
3097 48. Exim crashed when trying to verify a sender address that was being
3098 rewritten to "<>".
3100 49. Exim was recognizing only a space character after ".include". It now also
3101 recognizes a tab character.
3103 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
3104 extracting the relevant information from the specification. The man page no
3105 longer contains scrambled data for the -d option, and I've added a section
3106 at the front about calling Exim under different names.
3108 51. Added "extra_headers" argument to the "mail" command in filter files.
3110 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
3111 crash.
3113 53. Installed eximstats 1.29.
3115 54. Added transport_filter_timeout as a generic transport option.
3117 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
3118 Cc: header lines. This was required by RFC 822, but it not required by RFC
3119 2822.
3121 56. Exim used to add From:, Date:, and Message-Id: header lines to any
3122 incoming messages that did not have them. Now it does so only if the
3123 message originates locally, that is, if there is no associated remote host
3124 address. When Resent- header lines are present, this applies to the Resent-
3125 lines rather than the non-Resent- lines.
3127 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
3128 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
3130 58. Messages for configuration errors now include the name of the main
3131 configuration file - useful now that there may be more than one file in a
3132 list (.included file names were always shown).
3134 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
3135 for those rare installations that do not start the daemon as root or run it
3136 setuid root. I've cut out the call to initgroups() if the daemon is not
3137 root at that time.
3139 60. The Exim user and group can now be bound into the binary as text strings
3140 that are looked up at the start of Exim's processing.
3142 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
3144 62. Added $mailstore_basename variable.
3146 63. Installed patch to sieve.c from Michael Haardt.
3148 64. When Exim failed to open the panic log after failing to open the main log,
3149 the original message it was trying to log was written to stderr and debug
3150 output, but if they were not available (the usual case in production), it
3151 was lost. Now it is written to syslog before the two lines that record the
3152 failures to open the logs.
3154 65. Users' Exim filters run in subprocesses under the user's uid. It is
3155 possible for a "deliver" command or an alias in a "personal" command to
3156 provoke an address rewrite. If logging of address rewriting is configured,
3157 this fails because the process is not running as root or exim. There may be
3158 a better way of dealing with this, but for the moment (because 4.30 needs
3159 to be released), I have disabled address rewrite logging when running a
3160 filter in a non-root, non-exim process.
3163 Exim version 4.24
3164 -----------------
3166 1. The buildconfig auxiliary program wasn't quoting the value set for
3167 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
3168 not defined. This bug was masked in 4.22 by the effect that was fixed in
3169 change 4.23/1.
3171 2. Some messages that were rejected after a message id was allocated were
3172 shown as "incomplete" by exigrep. It no longer does this for messages that
3173 are rejected by local_scan() or the DATA or non-SMTP ACLs.
3175 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
3176 have allow_domain_literals set, the ID did not get logged in the <= line.
3177 Domain literals are now always recognized in Message-ID: header lines.
3179 4. The first argument for a ${extract expansion item is the key name or field
3180 number. Leading and trailing spaces in this item were not being ignored,
3181 causing some misleading effects.
3183 5. When deliver_drop_privilege was set, single queue runner processes started
3184 manually (i.e. by the command "exim -q") or by the daemon (which uses the
3185 same command in the process it spins off) were not dropping privilege.
3187 6. When the daemon running as "exim" started a queue runner, it always
3188 re-executed Exim in the spun-off process. This is a waste of effort when
3189 deliver_drop_privilege is set. The new process now just calls the
3190 queue-runner function directly.
3193 Exim version 4.23
3194 -----------------
3196 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
3199 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
3200 ignored. Though the use of -oP was forcing the writing of a pid file, it
3201 was always written to the default place.
3203 3. If the message "no IP address found for host xxxx" is generated during
3204 incoming verification, it is now followed by identification of the incoming
3205 connection (so you can more easily find what provoked it).
3207 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
3209 5. Added some features to "harden" Exim a bit more against certain attacks:
3211 (a) There is now a build-time option called FIXED_NEVER_USERS that can
3212 be put in Local/Makefile. This is like the never_users runtime option,
3213 but it cannot be overridden. The default setting is "root".
3215 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
3216 prefix string with which any file named in a -C command line option
3217 must start.
3219 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
3220 is retained for -C and -D only if the caller of Exim is root. Without
3221 it, the exim user may also use -C and -D and retain privilege.
3223 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
3224 command line option is disabled.
3226 6. Macro names set by the -D option must start with an upper case letter, just
3227 like macro names defined in the configuration file.
3229 7. Added "dereference=" facility to LDAP.
3231 8. Two instances of the typo "uknown" in the source files are fixed.
3233 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
3234 the Configure-Makefile script screwed up while processing it.
3236 10. Incorporated PCRE 4.4.
3238 11. The SMTP synchronization check was not operating right at the start of an
3239 SMTP session. For example, it could not catch a HELO sent before the client
3240 waited for the greeting. There is now a check for outstanding input at the
3241 point when the greeting is written. Because of the duplex, asynchronous
3242 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
3243 way, but not yet received, when the check is performed.
3245 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
3246 on TCP/IP sockets, because this apparently causes some broken clients to
3247 timeout.
3249 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
3250 unchanged) from the Cygwin maintainer.
3252 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
3253 was supported instead of testing for mbx. Effectively a typo.
3255 15. The spa authenticator server code was not checking that the input it
3256 received was valid base64.
3258 16. The debug output line for the "set" modifier in ACLs was not showing the
3259 name of the variable that was being set.
3261 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
3263 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
3264 lookup on its own. Something else was needed to trigger the lookup. For
3265 example, a match in host_lookup or the need for a host name in a host list.
3266 Now, if $sender_host_name is referenced and the host name has not yet been
3267 looked up, a lookup is performed. If the lookup fails, the variable remains
3268 empty, and $host_lookup_failed is set to "1".
3270 19. Added "eqi" as a case-independent comparison operator.
3272 20. The saslauthd authentication condition could segfault if neither service
3273 nor realm was specified.
3275 21. If an overflowing value such as "2048M" was set for message_size_limit, the
3276 error message that was logged was misleading, and incoming SMTP
3277 connections were dropped. The message is now more accurate, and temporary
3278 errors are given to SMTP connections.
3280 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
3281 (except RSET) with a 421 error, until QUIT is received. However, it was
3282 failing to send a response to QUIT.
3284 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
3285 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
3286 ACL is now run after the helo_try_verify_hosts code.
3288 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
3289 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
3290 case-independent, so other case variants are also recognized). Apparently
3291 some systems use these upper case variants.
3293 25. If more than two messages were waiting for the same host, and a transport
3294 filter was specified for the transport, Exim sent two messages over the
3295 same TCP/IP connection, and then failed with "socket operation on non-
3296 socket" when it tried to send the third.
3298 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
3300 27. The extern definition of crypt16() in expand.c was not being excluded when
3301 the OS had its own crypt16() function.
3303 28. Added bounce_return_body as a new option, and bounce_return_size_limit
3304 as a preferred synonym for return_size_limit, both as an option and as an
3305 expansion variable.
3307 29. Added LIBS=-liconv to OS/Makefile-OSF1.
3309 30. Changed the default configuration ACL to relax the local part checking rule
3310 for addresses that are not in any local domains. For these addresses,
3311 slashes and pipe symbols are allowed within local parts, but the sequence
3312 /../ is explicitly forbidden.
3314 31. SPA server authentication was not clearing the challenge buffer before
3315 using it.
3317 32. log_message in a "warn" ACL statement was writing to the reject log as
3318 well as to the main log, which contradicts the documentation and doesn't
3319 seem right (because no rejection is happening). So I have stopped it.
3321 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
3322 However, I am unable to do any testing of this.
3324 34. Fixed an infelicity in the appendfile transport. When checking directories
3325 for a mailbox, to see if any needed to be created, it was accidentally
3326 using path names with one or more superfluous leading slashes; tracing
3327 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
3329 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
3330 contents are added to the log line that is written for every discarded
3331 recipient. (Previously a log_message setting was ignored.)
3333 36. The ${quote: operator now quotes the string if it is empty.
3335 37. The install script runs exim in order to find its version number. If for
3336 some reason other than non-existence or emptiness, which it checks, it
3337 could not run './exim', it was installing it with an empty version number,
3338 i.e. as "exim-". This error state is now caught, and the installation is
3339 aborted.
3341 38. An argument was missing from the function that creates an error message
3342 when Exim fails to connect to the socket for saslauthd authentication.
3343 This could cause Exim to crash, or give a corrupted message.
3345 39. Added isip, isip4, and isip6 to ${if conditions.
3347 40. The ACL variables $acl_xx are now saved with the message, and can be
3348 accessed later in routers, transports, and filters.
3350 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
3351 strings in the file are not string-expanded.
3353 42. If a MAIL command specified a SIZE value that was too large to fit into an
3354 int variable, the check against message_size_limit failed. Such values are
3355 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
3356 day this will have to be increased, but I don't think I want to be around
3357 when emails are that large.
3361 Exim version 4.22
3362 -----------------
3364 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
3365 iconv() is not standard in FreeBSD.
3367 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
3368 IPv6 enabled. The observed symptom was a segmentation fault on return from
3369 the function os_common_find_running_interfaces() in src/os.c.
3371 3. In the check_special_case() function in daemon.c I had used "errno" as an
3372 argument name, which causes warnings on some systems. This was basically a
3373 typo, since it was named "eno" in the comments!
3375 4. The code that waits for the clock to tick (at a resolution of some fraction
3376 of a second) so as to ensure message-id uniqueness was always waiting for
3377 at least one whole tick, when it could have waited for less. [This is
3378 almost certainly not relevant at current processor speeds, where it is
3379 unlikely to ever wait at all. But we try to future-proof.]
3381 5. The function that sleeps for a time interval that includes fractions of a
3382 second contained a race. It did not block SIGALRM between setting the
3383 timer, and suspending (a couple of lines later). If the interval was short
3384 and the sigsuspend() was delayed until after it had expired, the suspension
3385 never ended. On busy systems this could lead to processes getting stuck for
3386 ever.
3388 6. Some uncommon configurations may cause a lookup to happen in a queue runner
3389 process, before it forks any delivery processes. The open lookup caching
3390 mechanism meant that the open file or database connection was passed into
3391 the delivery process. The problem was that delivery processes always tidy
3392 up cached lookup data. This could cause a problem for the next delivery
3393 process started by the queue runner, because the external queue runner
3394 process does not know about the closure. So the next delivery process
3395 still has data in the lookup cache. In the case of a file lookup, there was
3396 no problem because closing a file descriptor in a subprocess doesn't affect
3397 the parent. However, if the lookup was caching a connection to a database,
3398 the connection was closed, and the second delivery process was likely to
3399 see errors such as "PGSQL: query failed: server closed the connection
3400 unexpectedly". The problem has been fixed by closing all cached lookups
3401 in a queue runner before running a delivery process.
3403 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
3404 seem to have the "const" qualifier which it has on other OS. I've
3405 parameterised it.
3407 8. Change 4.21/2 was too strict. It is only if there are two authenticators
3408 *of the same type* (client or server) with the same public name that an
3409 error should be diagnosed.
3411 9. When Exim looked up a host name for an IP address, but failed to find the
3412 original IP address when looking up the host name (a safety check), it
3413 output the message "<ip address> does not match any IP for NULL", which was
3414 confusing, to say the least. The bug was that the host name should have
3415 appeared instead of "NULL".
3417 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
3418 user that is built into the binary, and the -C or -D options is used, root
3419 privilege is dropped before the configuration file is read. In addition,
3420 logging is switched to stderr instead of the normal log files. If the
3421 configuration then re-defines the Exim user, the unprivileged environment
3422 is probably not what is expected, so Exim logs a panic warning message (but
3423 proceeds).
3425 However, if deliver_drop_privilege is set, the unprivileged state may well
3426 be exactly what is intended, so the warning has been cut out in that case,
3427 and Exim is allowed to try to write to its normal log files.
3430 Exim version 4.21
3431 -----------------
3433 1. smtp_return_error_details was not giving details for temporary sender
3434 or receiver verification errors.
3436 2. Diagnose a configuration error if two authenticators have the same public
3437 name.
3439 3. Exim used not to create the message log file for a message until the first
3440 delivery attempt. This could be confusing when incoming messages were held
3441 for policy or load reasons. The message log file is now created at the time
3442 the message is received, and an initial "Received" line is written to it.
3444 4. The automatically generated man page for command line options had a minor
3445 bug that caused no ill effects; however, a more serious problem was that
3446 the procedure for building the man page automatically didn't always
3447 operate. Consequently, release 4.20 contains an out-of-date version. This
3448 shouldn't happen again.
3450 5. When building Exim with embedded Perl support, the script that builds the
3451 Makefile was calling 'perl' to find its compile-time parameters, ignoring
3452 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
3454 6. The freeze_tell option was not being used for messages that were frozen on
3455 arrival, either by an ACL or by local_scan().
3457 7. Added the smtp_incomplete_transaction log selector.
3459 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
3460 was accepting AUTH without a new EHLO.
3462 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
3463 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
3464 RFC.
3466 10. Logging of TCP/IP connections (when configured) now happens in the main
3467 daemon process instead of the child process, so that the TCP/IP connection
3468 count is more accurate (but it can never be perfect).
3470 11. The use of "drop" in a nested ACL was not being handled correctly in the
3471 outer ACL. Now, if condition failure induced by the nested "drop" causes
3472 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
3473 or "require"), the connection is dropped.
3475 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
3476 that yield "discard" can now be used with an "accept" or a "discard" verb,
3477 but an error is generated for any others (because I can't see a useful way
3478 to define what should happen).
3480 13. When an ACL is read dynamically from a file (or anywhere else), the lines
3481 are now processed in the same way as lines in the Exim configuration file.
3482 In particular, continuation lines are supported.
3484 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
3486 15. Added -ti meaning -t -i.
3488 16. Check for letters, digits, hyphens, and dots in the names of dnslist
3489 domains, and warn by logging if others are found.
3491 17. At least on BSD, alignment is not guarenteed for the array of ifreq's
3492 returned from GIFCONF when Exim is trying to find the list of interfaces on
3493 a host. The code in os.c has been modified to copy each ifreq to an aligned
3494 structure in all cases.
3496 Also, in some cases, the returned ifreq's were being copied to a 'struct
3497 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
3498 means the last couple of bytes of an IPv6 address could be chopped if the
3499 ifreq contained only a normal sockaddr (14 bytes storage).
3501 18. Named domain lists were not supported in the hosts_treat_as_local option.
3502 An entry such as +xxxx was not recognized, and was treated as a literal
3503 domain name.
3505 19. Ensure that header lines added by a DATA ACL are included in the reject log
3506 if the ACL subsequently rejects the message.
3508 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
3509 MD5 (which is deprecated).
3511 21. When testing a filter file using -bf, Exim was writing a message when it
3512 took the sender from a "From " line in the message, but it was not doing so
3513 when it took $return_path from a Return-Path: header line. It now does.
3515 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
3516 with a valid header line field name (a series of printing characters
3517 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
3519 23. Changed "disc" in the source to "disk" to conform to the documentation and
3520 the book and for uniformity.
3522 24. Ignore Sendmail's -Ooption=value command line item.
3524 25. When execve() failed while trying to run a command in a pipe transport,
3525 Exim was returning EX_UNAVAILBLE (69) from the subprocess. However, this
3526 could be confused with a return value of 69 from the command itself. This
3527 has been changed to 127, the value the shell returns if it is asked to run
3528 a non-existent command. The wording for the related log line suggests a
3529 non-existent command as the problem.
3531 26. If received_header_text expands to an empty string, do not add a Received:
3532 header line to the message. (Well, it adds a token one on the spool, but
3533 marks it "old" so that it doesn't get used or transmitted.)
3535 27. Installed eximstats 1.28 (addition of -nt option).
3537 28. There was no check for failure on the call to getsockname() in the daemon
3538 code. This can fail if there is a shortage of resources on the system, with
3539 ENOMEM, for example. A temporary error is now given on failure.
3541 29. Contrary to the C standard, it seems that in some environments, the
3542 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
3543 program. Exim now does this explicitly; it affects the formatting of
3544 timestamps using strftime().
3546 30. If exiqsumm was given junk data, it threw up some uninitialized variable
3547 complaints. I've now initialized all the variables, to avoid this.
3549 32. Header lines added by a system filter were not being "seen" during
3550 transport-time rewrites.
3552 33. The info_callback() function passed to OpenSSL is set up with type void
3553 (*)(SSL *, int, int), as described somewhere. However, when calling the
3554 function (actually a macro) that sets it up, the type void(*)() is
3555 expected. I've put in a cast to prevent warnings from picky compilers.
3557 34. If a DNS black list lookup found a CNAME record, but there were no A
3558 records associated with the domain it pointed at, Exim crashed.
3560 35. If a DNS black list lookup returned more than one A record, Exim ignored
3561 all but the first. It now scans all returned addresses if a particular IP
3562 value is being sought. In this situation, the contents of the
3563 $dnslist_value variable are a list of all the addresses, separated by a
3564 comma and a space.
3566 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
3567 to accept a host name and all its aliases if the forward lookup for any of
3568 them yielded the IP address of the incoming connection. Now it accepts only
3569 those names whose forward lookup yields the correct IP address. Any other
3570 names are discarded. This closes a loophole whereby a rogue DNS
3571 administrator could create reverse DNS records to break through a
3572 wildcarded host restriction in an ACL.
3574 37. If a user filter or a system filter that ran in a subprocess used any of
3575 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
3576 the wrong values were passed to the pipe command ($thisaddress had the
3577 value of $0, $0 had the value of $1, etc). This bug was introduced by
3578 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
3580 38. Improved the line breaking for long SMTP error messages from ACLs.
3581 Previously, if there was no break point between 40 and 75 characters, Exim
3582 left the rest of the message alone. Two changes have been made: (a) I've
3583 reduced the minimum length to 35 characters; (b) if it can't find a break
3584 point between 35 and 75 characters, it looks ahead and uses the first one
3585 that it finds. This may give the occasional overlong line, but at least the
3586 remaining text gets split now.
3588 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
3589 file descriptors might be low, and that setting 1000 would always raise it.
3590 It turns out that in some environments, the limit is already over 1000 and
3591 that lowering it causes trouble. So now Exim takes care not to decrease it.
3593 40. When delivering a message, the value of $return_path is set to $sender_
3594 address at the start of routing (routers may change the value). By an
3595 oversight, this default was not being set up when an address was tested by
3596 -bt or -bv, which affected the outcome if any router or filter referred to
3597 $return_path.
3599 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
3600 log only when "message" or "log_message" are set. However, if one of the
3601 conditions was an address verification, or a call to a nested ACL, the
3602 messages generated by the underlying test were being passed through. This
3603 no longer happens. The underlying message is available in $acl_verify_
3604 message for both "message" and "log_message" expansions, so it can be
3605 passed through if needed.
3607 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
3608 new expansion $bh_ to give the encoded byte string without charset
3609 translation. Translation happens only if iconv() is available; HAVE_ICONV
3610 indicates this at build time. HEADERS_CHARSET gives the charset to
3611 translate to; headers_charset can change it in the configuration, and
3612 "headers charset" can change it in an individual filter file.
3614 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
3615 that creates RFC 2047 encoded "words" labels them as that charset instead
3616 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
3617 expansion operator; (ii) when Exim creates a From: line for a local
3618 message; (iii) when a header line is rewritten to include a "phrase" part.
3620 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
3621 buggy, causing it to skip the first lines of messages whose message ID
3622 ended in 'D'. This would not have bitten before Exim release 4.14, because
3623 message IDs were unlikely to end in 'D' before then. The effect was to have
3624 incorrect size information for certain domains.
3626 45. #include "config.h" was missing at the start of the crypt16.c module. This
3627 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
3628 noticed.
3630 46. If there was a timeout during a "random" callout check, Exim treated it as
3631 a failure of the random address, and carried on sending RSET and the real
3632 address. If the delay was just some slowness somewhere, the response to the
3633 original RCPT would be taken as a response to RSET and so on, causing
3634 mayhem of various kinds.
3636 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
3637 when I implemented it. It didn't allow for the fact that some option values
3638 may legitimatetly be negative (e.g. size_addition), and it didn't even do
3639 the right test for positive values.
3641 48. Domain names in DNS records are case-independent. Exim always looks them up
3642 in lower case. Some resolvers return domain names in exactly the case they
3643 appear in the zone file, that is, they may contain uppercase letters. Not
3644 all resolvers do this - some return always lower case. Exim was treating a
3645 change of case by a resolver as a change of domain, similar to a widening
3646 of a domain abbreviation. This triggered its re-routing code and so it was
3647 trying to route what was effectively the same domain again. This normally
3648 caused routing to fail (because the router wouldn't handle the domain
3649 twice). Now Exim checks for this case specially, and just changes the
3650 casing of the domain that it ultimately uses when it transmits the message
3651 envelope.
3653 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
3654 module.
3656 50. If a filter generated a file delivery with a non-absolute name (possible if
3657 no home directory exists for the router), the forbid_file option was not
3658 forbidding it.
3660 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
3661 the existing equality matching.
3663 52. Exim was using ints instead of ino_t variables in some places where it was
3664 dealing with inode numbers.
3666 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
3667 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
3668 called TMPDIR, and if it finds it is different, it changes its value.
3670 54. The smtp_printf() function is now made available to local_scan() so
3671 additional output lines can be written before returning. There is also an
3672 smtp_fflush() function to enable the detection of a dropped connection.
3673 The variables smtp_input and smtp_batched_input are exported to
3674 local_scan().
3676 55. Changed the default runtime configuration: the message "Unknown user"
3677 has been removed from the ACL, and instead placed on the localuser router,
3678 using the cannot_route_message feature. This means that any verification
3679 failures that generate their own messages won't get overridden. Similarly,
3680 the "Unrouteable address" message that was in the ACL for unverifiable
3681 relay addresses has also been removed.
3683 56. Added hosts_avoid_esmtp to the smtp transport.
3685 57. The exicyclog script was not checking for the esoteric option
3686 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
3687 will work only if exicyclog is run under the appropriate euid.
3689 58. Following a discussion on the list, the rules by which Exim recognises line
3690 endings on incoming messages have been changed. The -dropcr and drop_cr
3691 options are now no-ops, retained only for backwards compatibility. The
3692 following line terminators are recognized: LF CRLF CR. However, special
3693 processing applies to CR:
3695 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
3696 nor a local message in the state where . is a terminator.
3698 (ii) If a bare CR is encountered in a header line, an extra space is added
3699 after the line terminator so as not to end the header. The reasoning
3700 behind this is that bare CRs in header lines are most likely either
3701 to be mistakes, or people trying to play silly games.
3703 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
3704 was being incorrectly given as 18 bytes larger than it should have been.
3705 This is a VOB (very old bug).
3707 60. This may never have affected anything current, but just in case it has:
3708 When the local host is found other than at the start of a list of hosts,
3709 the local host, those with the same MX, and any that follow, are discarded.
3710 When the list in question was part of a longer list of hosts, the following
3711 hosts (not currently being processed) were also being discarded. This no
3712 longer happens. I'm not sure if this situation could ever has previously
3713 arisen.
3715 61. Added the "/MX" feature to lists of hosts in the manualroute and query
3716 program routers.
3718 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
3719 header. This is something that is recommended in a new Internet Draft, and
3720 is something that is documented as being done by Sendmail. There are two
3721 possible values. For messages generated by the autoreply transport, Exim
3722 adds:
3724 Auto-Submitted: auto-replied
3726 whereas for all other generated messages (e.g. bounces) it adds
3728 Auto-Submitted: auto-generated
3730 63. The "personal" condition in filters now includes a test for the
3731 Auto-Submitted: header. If it contains the string "auto-" the message it
3732 not considered personal.
3734 64. Added rcpt_include_affixes as a generic transport option.
3736 65. Added queue_only_override (default true).
3738 66. Added the syslog_duplication option.
3740 67. If what should have been the first header line of a message consisted of
3741 a space followed by a colon, Exim was mis-interpreting it as a header line.
3742 It isn't of course - it is syntactically invalid and should therefore be
3743 treated as the start of the message body. The misbehaviour could have
3744 caused a number of strange effects, including loss of data in subsequent
3745 header lines, and spool format errors.
3747 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
3748 client host had authenticated. This control can now be exercised by an ACL
3749 for more flexibility.
3751 69. By default, callouts do not happen when testing with -bh. There is now a
3752 variant, -bhc, which does actually run the callout code, including
3753 consulting and updating the callout cache.
3755 70. Added support for saslauthd authentication, courtesy of Alexander
3756 Sabourenkov.
3758 71. If statvfs() failed on the spool or log directories while checking their
3759 size for availability, Exim confusingly gave the error "space shortage".
3760 Furthermore, in debugging mode it crashed with a floating point exception.
3761 These checks are done if check_{spool,log}_{space,inodes} are set, and when
3762 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
3763 serious problem, Exim now writes to the main and panic logs when this
3764 happens, with details of the failure. It then refuses to accept the
3765 incoming message, giving the message "spool directory problem" or "log
3766 directory problem" with a 421 code for SMTP messages.
3768 72. When Exim is about to re-exec itself, it ensures that the file descriptors
3769 0, 1, and 2 exist, because some OS complain for execs without them (see
3770 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
3771 descriptors. However, the code omitted to check that the open succeeded,
3772 causing mysterious errors if for some reason the permissions on /dev/null
3773 got screwed. Now Exim writes a message to the main and panic logs, and
3774 bombs out if it can't open /dev/null.
3776 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
3777 interact so that it is all more flexible. It is supposed to remain
3778 backwards compatible. Also added extra_local_interfaces.
3780 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
3781 to bomb out with an assertion failure - to the client this appears as a
3782 connection drop. This problem occurs in the part of the code that was taken
3783 from the Samba project. Fortunately, the assertion is in a very simple
3784 function, so I have fixed this by reproducing the function inline in the
3785 one place where it is called, and arranging for authentication to fail
3786 instead of killing the process with assert().
3788 75. The SPA client code was not working when the server requested OEM rather
3789 than Unicode encoding.
3791 76. Added code to make require_files with a specific uid setting more usable in
3792 the case where statting the file as root fails - usually a non-root-mounted
3793 NFS file system. When this happens and the failure is EACCES, Exim now
3794 forks a subprocess and does the per-uid checking as the relevant uid.
3796 77. Added process_log_path.
3798 78. If log_file_path was not explicitly set, a setting of check_log_space or
3799 check_log_inodes was ignored.
3801 79. If a space check for the spool or log partitions fails, the incident is now
3802 logged. Of course, in the latter case the data may get lost...
3804 80. Added the %p formatting code to string_format() so that it can be used to
3805 print addresses in debug_print(). Adjusted all the address printing in the
3806 debugging in store.c to use %p rather than %d.
3808 81. There was a concern that a line of code in smtp_in.c could overflow a
3809 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
3810 initially expressed, the concern was not well-founded, because trailing
3811 spaces are removed early. However, if the trailing spaces were followed by
3812 a NULL, they did not get removed, so the overflow was possible. Two fixes
3813 were applied:
3815 (a) I re-wrote the offending code in a cleaner fashion.
3816 (b) If an incoming SMTP command contains a NULL character, it is rejected
3817 as invalid.
3819 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
3820 runs initgroups(), so that if the Exim user is in any additional groups,
3821 they will be used during message reception.
3824 Exim version 4.20
3825 -----------------
3827 The change log for 4.20 and earlier releases has been archived.
3829 ****