1 $Cambridge: exim/doc/doc-txt/ChangeLog,v 1.311 2006/02/22 15:08:20 ph10 Exp $
3 Change log file for Exim from version 4.21
4 -------------------------------------------
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
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
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
41 127.0.0.2 (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
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
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
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
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
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
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
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 1.2.3.4/abc was being treated as the IP address
137 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
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
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.
232 PH/01 Two changes to the default runtime configuration:
234 (1) Move the checks for relay_from_hosts and authenticated clients from
235 after to before the (commented out) DNS black list checks.
237 (2) Add control=submission to the relay_from_hosts and authenticated
238 clients checks, on the grounds that messages accepted by these
239 statements are most likely to be submissions.
241 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
243 (1) Generate an error if the third argument for the ${prvs expansion is
246 (2) Treat a missing third argument of ${prvscheck as if it were an empty
249 (3) Reset the variables that are obtained from the first argument of
250 ${prvscheck and used in the second argument before leaving the code,
251 because their memory is reclaimed, so using them afterwards may do
254 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
255 one (it's much easier than Tom thought :-).
257 (5) Because of (4), we can now allow for the use of $prvscheck_result
258 inside the third argument.
260 PH/03 For some reason, the default setting of PATH when running a command from
261 a pipe transport was just "/usr/bin". I have changed it to
264 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
265 anything to be listed in the output from -bV.
267 PH/05 When a filter generated an autoreply, the entire To: header line was
268 quoted in the delivery log line, like this:
270 => >A.N.Other <ano@some.domain> <original@ddress> ...
272 This has been changed so that it extracts the operative address. There
273 may be more than one such address. If so, they are comma-separated, like
276 => >ano@some.domain,ona@other.domain <original@ddress> ...
278 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
279 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
280 not being looked up in the rDNS to get a host name, Exim was showing the
281 IP address twice in Received: lines, even though the IP addresses were
282 identical. For example:
284 Received: from [1.2.3.4] (helo=[1.2.3.4])
286 However, if the real host name was known, it was omitting the HELO data
287 if it matched the actual IP address. This has been tidied up so that it
288 doesn't show the same IP address twice.
290 PH/07 When both +timestamp and +memory debugging was on, the value given by
291 $tod_xxx expansions could be wrong, because the tod_stamp() function was
292 called by the debug printing, thereby overwriting the timestamp buffer.
293 Debugging no longer uses the tod_stamp() function when +timestamp is set.
295 PH/08 When the original message was included in an autoreply transport, it
296 always said "this is a copy of the message, including all the headers",
297 even if body_only or headers_only was set. It now gives an appropriate
300 PH/09 Applied a patch from the Sieve maintainer which:
302 o fixes some comments
303 o adds the (disabled) notify extension core
304 o adds some debug output for the result of if/elsif tests
305 o points to the current vacation draft in the documentation
306 and documents the missing references header update
310 o fixes a bug in processing the envelope test (when testing
311 multiple envelope elements, the last element determinted the
314 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
315 Electronic Mail") by including:
317 Auto-submitted: auto-generated
319 in the messages that it generates (bounce messages and others, such as
320 warnings). In the case of bounce messages for non-SMTP mesages, there was
321 also a typo: it was using "Auto_submitted" (underscore instead of
322 hyphen). Since every message generated by Exim is necessarily in response
323 to another message, thes have all been changed to:
325 Auto-Submitted: auto-replied
327 in accordance with these statements in the RFC:
329 The auto-replied keyword:
331 - SHOULD be used on messages sent in direct response to another
332 message by an automatic process,
334 - MUST NOT be used on manually-generated messages,
336 - MAY be used on Delivery Status Notifications (DSNs) and Message
337 Disposition Notifications (MDNs),
339 - MUST NOT be used on messages generated by automatic or periodic
340 processes, except for messages which are automatic responses to
343 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
344 to the default Received: header definition.
346 PH/12 Added log selector acl_warn_skipped (default on).
348 PH/13 After a successful wildlsearch lookup, discard the values of numeric
349 variables because (a) they are in the wrong storage pool and (b) even if
350 they were copied, it wouldn't work properly because of the caching.
352 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
353 checking when decoding. Apparently there are clients that generate
354 overlong encoded strings. Why am I not surprised?
356 PH/15 If the first argument of "${if match_address" was not empty, but did not
357 contain an "@" character, Exim crashed. Now it writes a panic log message
358 and treats the condition as false.
360 PH/16 In autoreply, treat an empty string for "once" the same as unset.
362 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
363 extension "envelope-auth". The code is finished and in agreement with
364 other implementations, but there is no documentation so far and in fact,
365 nobody wrote the draft yet. This extension is currently #undef'ed, thus
366 not changing the active code.
368 Print executed "if" and "elsif" statements when debugging is used. This
369 helps a great deal to understand what a filter does.
371 Document more things not specified clearly in RFC3028. I had all this
372 sorted out, when out of a sudden new issues came to my mind. Oops."
374 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
377 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
378 canonical form (as documented). However, after a host name lookup from
379 the IP address, check_host() was doing a simple string comparison with
380 addresses acquired from the DNS when checking that the found name did
381 have the original IP as one of its addresses. Since any found IPv6
382 addresses are likely to be in abbreviated form, the comparison could
383 fail. Luckily, there already exists a function for doing the comparison
384 by converting both addresses to binary, so now that is used instead of
387 PH/20 There was another similar case to PH/19, when a complete host name was
388 given in a host list; looking up its IP address could give an abbreviated
389 form, whereas the current host's name might or might not be abbreviated.
390 The same fix has been applied.
396 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
397 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
400 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
401 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
403 PH/03 Typo: missing ".o" in src/pcre/Makefile.
405 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
406 header line, restrict the check to what is listed in RFCs 2369 and 2929.
407 Also, for "Auto-Submitted", treat anything other than "no" as
408 non-personal, in accordance with RFC 3834. (Previously it treated
409 anything starting "auto-" as non-personal.)
411 TF/01 The control=submission/name=... option had a problem with syntax
412 errors if the name included a slash character. The /name= option
413 now slurps the rest of the string, so it can include any characters
414 but it must come last in the list of options (after /sender_retain
417 PH/05 Some modifications to the interface to the fake nameserver for the new
425 TK/01 Added the "success_on_redirect" address verification option. See
426 NewStuff for rationale and an example.
428 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
430 PH/02 Patch to exigrep to allow it to work on syslog lines.
432 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
433 fread() to skip over the body file's header line, because in Cygwin the
434 header line is locked and is inaccessible.
436 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
437 co-exist for some time) to make it clear that it is the Exim ID that is
438 referenced, not the Message-ID: header line.
440 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
441 string_format() function, because snprintf() does not exist on all
444 PH/06 The use of forbid_filter_existstest now also locks out the use of the
445 ${stat: expansion item.
447 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
448 protocol synchronization error", to keep the pedants happy.
450 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
451 well as for IRIX systems, when gcc is being used. See the host.c source
454 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
456 PH/10 Named domain lists were not working if used in a queue_smtp_domains
459 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
460 transport and to the smtp transport in LMTP mode.
462 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
464 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
465 run a filter in a subprocess. This could lead to confusion in subsequent
466 lookups in the parent process. There should also be a search_tidyup() at
467 the end of the subprocess.
469 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
470 only if the host matched helo_try_verify_hosts, which caused the
471 verification to occur when the EHLO/HELO command was issued. The ACL just
472 tested the remembered result. Now, if a previous verification attempt has
473 not happened, "verify = helo" does it there and then.
475 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
477 TK/03 Fix log output including CR from clamd.
479 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
480 exist provoked a memory error which could cause a segfault.
482 PH/15 Installed PCRE 6.2
484 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
486 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
487 of the problem. Specifically, suggested +O2 rather than +O1 for the
490 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
492 PH/20 If a delivery was routed to a non-standard port by means of an SRV
493 record, the port was not correctly logged when the outgoing_port log
494 selector was set (it logged the transort's default port).
496 PH/21 Added support for host-specific ports to manualroute, queryprogram,
497 fallback_hosts, and "hosts" in the smtp transport.
499 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
500 host errors such as "Connection refused".
502 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
503 authentication with radiusclient 0.4.9:
505 - Error returned from rc_read_config was caught wrongly
506 - Username/password not passed on to radius server due to wrong length.
508 The presumption is that some radiusclient API changes for 4.51/PH/17
509 were not taken care of correctly. The code is still untested by me (my
510 Linux distribution still has 0.3.2 of radiusclient), but it was
511 contributed by a Radius user.
513 PH/24 When doing a callout, the value of $domain wasn't set correctly when
514 expanding the "port" option of the smtp transport.
516 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
517 while reading a MIME header. Thanks to Tom Hughes for a patch.
519 PH/24 Include config.h inside local_scan.h so that configuration settings are
522 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
523 that in an ACL for RCPT (for example), you can examine exactly what was
526 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
527 commands, but it was not correctly comparing the address with the actual
528 client host address. Thus, it would show the EHLO address in Received:
529 header lines when this was not necessary.
531 PH/27 Added the % operator to ${eval:}.
533 PH/28 Exim tries to create and chdir to its spool directory when it starts;
534 it should be ignoring failures (because with -C, for example, it has lost
535 privilege). It wasn't ignoring creation failures other than "already
538 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
541 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
542 because an input line was too long, either on its own, or by virtue of
543 too many continuations, the temporary file was not being removed, and the
544 return code was incorrect.
546 PH/31 Missing "BOOL" in function definition in filtertest.c.
548 PH/32 Applied Sieve patches from the maintainer.
550 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
552 PH/33 Added "verify = not_blind".
554 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
555 Local/Makefile (with some defaults set). These are used in built scripts
556 such as exicyclog, but they have never been used in the exim_install
557 script (though there are many overriding facilities there). I have
558 arranged that the exim_install script now takes note of these two
561 PH/35 Installed configuration files for Dragonfly.
563 PH/36 When a locally submitted message by a trusted user did not contain a
564 From: header, and the sender address was obtained from -f or from an SMTP
565 MAIL command, and the trusted user did not use -F to supply a sender
566 name, $originator_name was incorrectly used when constructing a From:
567 header. Furthermore, $originator_name was used for submission mode
568 messages from external hosts without From: headers in a similar way,
569 which is clearly wrong.
571 PH/37 Added control=suppress_local_fixups.
573 PH/38 When log_selector = +received_sender was set, and the addition of the
574 sender made the log line's construction buffer exactly full, or one byte
575 less than full, an overflow happened when the terminating "\n" was
578 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
579 when the result of a list match is failure because a DNS lookup failed.
581 PH/40 RM_COMMAND is now used in the building process.
583 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
584 the "build-* directories that it finds.
586 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
587 address in a domain literal was a prefix of an interface address.
589 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
590 when verifying a sender address, unless rewrite_headers is false.
592 PH/44 Wrote a long comment about why errors_to addresses are verified as
593 recipients, not senders.
595 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
596 the ratelimit ACL was added.
598 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
600 PH/46 Added extra information about PostgreSQL errors to the error string.
602 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
603 avoiding the need to install special zones in a real server. This is
604 backwards compatible; if it can't find the fake resolver, it drops back.
605 Thus, both old and new test suites can be run.
607 TF/02 Added util/ratelimit.pl
609 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
610 clock is set back in time.
612 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
613 Candler <B.Candler@pobox.com>.
615 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
616 OK for addresses that are the result of redirections.
618 PH/48 A number of further additions for the benefit of the new test suite,
619 including a fake gethostbyname() that interfaces to the fake DNS resolver
622 TF/06 The fix for widen_domains has also been applied to qualify_single and
623 search_parents which are the other dnslookup options that can cause
626 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
629 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
631 TF/07 Exim produced the error message "an SRV record indicated no SMTP
632 service" if it encountered an MX record with an empty target hostname.
633 The message is now "an MX or SRV record indicated no SMTP service".
635 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
636 if the DNS of the sending site is misconfigured. This is quite a
637 common situation. This change restores the behaviour of treating a
638 helo verification defer as a failure.
640 PH/51 If self=fail was set on a router, the bounce message did not include the
641 actual error message.
647 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
649 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
650 command itself ended in error, the underlying message about the transport
651 filter timeout was being overwritten with the pipe command error. Now the
652 underlying error message should be appended to the second error message.
654 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
656 PH/02 Reduce the amount of output that "make" produces by default. Full output
657 can still be requested.
659 PH/03 The warning log line about a condition test deferring for a "warn" verb
660 was being output only once per connection, rather than after each
661 occurrence (because it was using the same function as for successful
662 "warn" verbs). This seems wrong, so I have changed it.
664 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
665 it should not have, which might have caused a crash in the right
666 circumstances, but probably never did.
668 PH/04 Installed a modified version of Tony Finch's patch to make submission
669 mode fix the return path as well as the Sender: header line, and to
670 add a /name= option so that you can make the user's friendly name appear
673 TF/03 Added the control = fakedefer ACL modifier.
675 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
676 Mark Lowes for thorough testing.
678 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
680 TK/03 Merged latest SRS patch from Miles Wilton.
682 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
683 with the definition in sysexits.h (which is #included earlier).
684 Fortunately, Exim does not actually use EX_OK. The code used to try to
685 preserve the sysexits.h value, by assumimg that macro definitions were
686 scanned for macro replacements. I have been disabused of this notion,
687 so now the code just undefines EX_OK before #including unistd.h.
689 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
690 in the smtp transport. When a block could not be written in a single
691 write() function, the timeout was being re-applied to each part-write.
692 This seems wrong - if the receiver was accepting one byte at a time it
693 would take for ever. The timeout is now adjusted when this happens. It
694 doesn't have to be particularly precise.
696 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
697 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
699 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
700 without a domain if the check to <postmaster@domain> fails.
702 SC/01 Eximstats: added -xls and the ability to specify output files
703 (patch written by Frank Heydlauf).
705 SC/02 Eximstats: use FileHandles for outputing results.
707 SC/03 Eximstats: allow any combination of xls, txt, and html output.
709 SC/04 Eximstats: fixed display of large numbers with -nvr option
711 SC/05 Eximstats: fixed merging of reports with empty tables.
713 SC/06 Eximstats: added the -include_original_destination flag
715 SC/07 Eximstats: removed tabs and trailing whitespace.
717 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
719 TK/06 MBOX spool code: Add real "From " MBOX separator line
720 so the .eml file is really in mbox format (even though
721 most programs do not really care). Patch from Alex Miller.
723 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
724 The latter is generated from $received_to and is only set if the
725 message has one envelope recipient. SA can use these headers,
726 obviously out-of-the-box. Patch from Alex Miller.
728 PH/08 The ${def test on a variable was returning false if the variable's
729 value was "0", contrary to what the specification has always said!
730 The result should be true unless the variable is empty.
732 PH/09 The syntax error of a character other than { following "${if
733 def:variable_name" (after optional whitespace) was not being diagnosed.
734 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
735 accidental colon was present, for example, could give incorrect results.
737 PH/10 Tidied the code in a number of places where the st_size field of a stat()
738 result is used (not including appendfile, where other changes are about
741 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
742 This involved changing a lot of size variables from int to off_t. It
743 should work with maildirs and everything.
745 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
746 spamd dying while we are connected to it.
748 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
751 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
752 function for BDB 4.3.
754 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
756 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
757 the sources for PCRE have been re-arranged and more files are now
760 PH/15 The code I had for printing potentially long long variables in PH/11
761 above was not the best (it lost precision). The length of off_t variables
762 is now inspected at build time, and an appropriate printing format (%ld
763 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
764 to be "long long int" or "long int". This is needed for the internal
765 formatting function string_vformat().
767 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
768 the configuration file to be ":syslog", then the script "guesses" where
769 the logs files are, rather than using the compiled in default. In our
770 case the guess is not the same as the compiled default, so the script
771 suddenly stopped working when I started to use syslog. The patch checks
772 to see if log_file_path is "". If so, it attempts to read it from exim
773 with no configuration file to get the compiled in version, before it
774 falls back to the previous guessing code."
776 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
777 implementing BATV in an Exim configuration. See NewStuff for the gory
780 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
781 Makefile that are specific to HP-UX.
783 PH/18 If the "use_postmaster" option was set for a recipient callout together
784 with the "random" option, the postmaster address was used as the MAIL
785 FROM address for the random test, but not for the subsequent recipient
786 test. It is now used for both.
788 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
789 patch removes a few documentation additions to RFC 3028, because the
790 latest draft now contains them. It adds the new en;ascii-case comparator
791 and a new error check for 8bit text in MIME parts. Comparator and
792 require names are now matched exactly. I enabled the subaddress
793 extension, but it is not well tested yet (read: it works for me)."
795 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
796 rework some of the code of TK/09 above to avoid the hardwired use of
797 "%lld" and "long long". Replaced the call to snprintf() with a call to
800 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
801 records point to non-existent hosts", "retry timeout exceeded", and
802 "retry time not reached for any host after a long failure period".
804 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
805 experimental DomainKeys support:
807 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
808 (2) On an error such as an illegally used "control", the wrong name for
809 the control was given.
811 These problems did NOT occur unless DomainKeys support was compiled.
813 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
815 PH/24 Added ${if match_ip condition.
817 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
818 databases so that it will be absolutely obvious if a crash occurs in the
819 DB library. This is a regular occurrence (often caused by mis-matched
822 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
823 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
824 picked up on a user's system that detects such things. There doesn't seem
825 to be a gcc warning option for this - only an attribute that has to be
826 put on the function's prototype. It seems that in Fedora Core 4 they have
827 set this on a number of new functions. No doubt there will be more in due
830 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
831 specify a transport. However, if an address that was verified by such a
832 router was the subject of a callout, Exim crashed because it tried to
833 read the rcpt_include_affixes from the non-existent transport. Now it
834 just assumes that the setting of that option is false. This bug was
835 introduced by 4.51/PH/31.
837 PH/28 Changed -d+all to exclude +memory, because that information is very
838 rarely of interest, but it makes the output a lot bigger. People tend to
839 do -d+all out of habit.
841 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
842 code in os-type was giving problems when libc.so lives in lib64, like on
845 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
846 aren't the modern standard, and it seems that some systems' include files
847 don't always have them. Exim was already checking for some of the newer
848 ones like T_AAAA, and defining it itself. I've added checks for all the
849 record types that Exim uses.
851 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
852 not automatically generating a new one, as it is supposed to. This
853 prevented TLS from working. If the file did exist, but contained invalid
854 data, a new version was generated, as expected. It was only the case of a
855 non-existent file that was broken.
857 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
858 with a change in libdomainkeys > 0.64.
860 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
861 from DNS. If the selector record carries the flag, it now has
862 precedence over the domain-wide flag.
864 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
866 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
867 the use of a port name that isn't defined in /etc/services) occurred, the
868 message was deferred as in a normal delivery, and thus remained on the
869 spool, instead of being failed because of the mua_wrapper setting. This
870 is now fixed, and I tidied up some of the mua_wrapper messages at the
873 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
874 the messages in a hash of arrays rather than using individual hashes.
875 This is a bit cleaner and results in dramatic memory savings, albeit
876 at a slight CPU cost.
878 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
879 as requested by Marc Sherman.
881 SC/10 Eximstats: added histograms for user specified patterns as requested
884 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
886 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
887 fopen() in the content-scanning modules that did not already have it.
893 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
894 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
896 TK/02 Fix ACL "control" statement not being available in MIME ACL.
898 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
900 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
901 to test Sieve filters that use "vacation".
903 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
904 that changes the way the GnuTLS parameters are stored in the cache file.
905 The new format can be generated externally. For backward compatibility,
906 if the data in the cache doesn't make sense, Exim assumes it has read an
907 old-format file, and it generates new data and writes a new file. This
908 means that you can't go back to an older release without removing the
911 PH/03 A redirect router that has both "unseen" and "one_time" set does not
912 work if there are any delivery delays because "one_time" forces the
913 parent to be marked "delivered", so its unseen clone is never tried
914 again. For this reason, Exim now forbids the simultaneous setting of
917 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
918 redirected to themselves ("homonym" addresses). Read the long ChangeLog
919 entry if you want to know the details. The fix, however, neglected to
920 consider the case when local delivery batching is involved. The test for
921 "previously delivered" was not happening when checking to see if an
922 address could be batched with a previous (undelivered) one; under
923 certain circumstances this could lead to multiple deliveries to the same
926 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
927 in its include files, and this causes problems building Exim.
929 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
930 header_syntax) but Exim was just ignoring anything given after a slash.
931 In particular, this caused confusion with an attempt to use "verify =
932 reverse_host_lookup/defer_ok". An error is now given when options are
933 supplied for verify items that do not have them. (Maybe reverse_host_
934 lookup should have a defer_ok option, but that's a different point.)
936 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
937 defined by RFC 821) to 2048, because there were problems with some AUTH
938 commands, and RFC 1869 says the size should be increased for extended
939 SMTP commands that take arguments.
941 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
944 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
945 "unknown" error; now it says that the functionality isn't in the binary.
947 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
948 an address' error message when a string expansion fails (syntax or
949 whatever). Otherwise the password may appear in the log. Following change
950 PH/42 below, there is no longer a chance of it appearing in a bounce
953 PH/11 Installed exipick version 20050225.0 from John Jetmore.
955 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
956 of its addresses was ever tried. (Bugzilla bug #2.)
958 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
959 the result incorrectly in the debug output. (It correctly added a newline
960 to what was transported.)
962 TF/01 Added $received_time.
964 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
965 commented out examples of how to interface to a virus scanner and to
966 SpamAssassin. Also added commented examples of av_scanner and
967 spamd_address settings.
969 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
970 and controls are allowed in which ACLs. There were a couple of minor
971 errors. Some of the entries in the conditions table (which is a table of
972 where they are NOT allowed) were getting very unwieldy; rewrote them as a
973 negation of where the condition IS allowed.
975 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
977 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
978 header file does not have a version number, so I've had to invent a new
979 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
980 API. The code is untested by me (my Linux distribution still has 0.3.2 of
981 radiusclient), but it was contributed by a Radius user.
983 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
984 files or directories, for OpenSSL.
986 PH/19 When an Exim process that is running as root has to create an Exim log
987 file, it does so in a subprocess that runs as exim:exim so as to get the
988 ownership right at creation (otherwise, other Exim processes might see
989 the file with the wrong ownership). There was no test for failure of this
990 fork() call, which would lead to the process getting stuck as it waited
991 for a non-existent subprocess. Forks do occasionally fail when resources
992 run out. I reviewed all the other calls to fork(); they all seem to check
995 PH/20 When checking for unexpected SMTP input at connect time (before writing
996 the banner), Exim was not dealing correctly with a non-positive return
997 from the read() function. If the client had disconnected by this time,
998 the result was a log entry for a synchronization error with an empty
999 string after "input=" when read() returned zero. If read() returned -1
1000 (an event I could not check), uninitialized data bytes were printed.
1001 There were reports of junk text (parts of files, etc) appearing after
1004 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
1006 PH/22 Added support for macro redefinition, and (re)definition in between
1007 driver and ACL definitions.
1009 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
1010 forgetting to use the resulting value; it was using the unexpanded value.
1012 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
1013 hadn't been configured. The fix is from Juergen Kreileder, who
1014 understands it better than I do:
1016 "Here's what I see happening with three configured cyrus_sasl
1017 authenticators configured (plain, login, cram-md5):
1019 On startup auth_cyrus_sasl_init() gets called for each of these.
1020 This means three calls to sasl_listmech() without a specified mech_list.
1021 => SASL tests which mechs of all available mechs actually work
1022 => three warnings about OTP not working
1023 => the returned list contains: plain, login, cram-md5, digest-md5, ...
1025 With the patch, sasl_listmech() also gets called three times. But now
1026 SASL's mech_list option is set to the server_mech specified in the the
1027 authenticator. Or in other words, the answer from sasl_listmech()
1028 gets limited to just the mech you're testing for (which is different
1030 => the return list contains just 'plain' or 'login', 'cram-md5' or
1031 nothing depending on the value of ob->server_mech.
1033 I've just tested the patch: Authentication still works fine,
1034 unavailable mechs specified in the exim configuration are still
1035 caught, and the auth.log warnings about OTP are gone."
1037 PH/25 When debugging is enabled, the contents of the command line are added
1038 to the debugging output, even when log_selector=+arguments is not
1041 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
1042 answer is "GNU", and only if the return is "GNU/something" is the answer
1045 PH/27 $acl_verify_message is now set immediately after the failure of a
1046 verification in an ACL, and so is available in subsequent modifiers. In
1047 particular, the message can be preserved by coding like this:
1049 warn !verify = sender
1050 set acl_m0 = $acl_verify_message
1052 Previously, $acl_verify_message was set only while expanding "message"
1053 and "log_message" when a very denied access.
1055 PH/28 Modified OS/os.c-Linux with
1057 -#ifndef OS_LOAD_AVERAGE
1058 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
1060 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
1061 nomenclature these days.)
1063 PH/29 Installed patch from the Sieve maintainer that adds the options
1064 sieve_useraddress and sieve_subaddress to the redirect router.
1066 PH/30 In these circumstances:
1067 . Two addresses routed to the same list of hosts;
1068 . First host does not offer TLS;
1069 . First host accepts first address;
1070 . First host gives temporary error to second address;
1071 . Second host offers TLS and a TLS session is established;
1072 . Second host accepts second address.
1073 Exim incorrectly logged both deliveries with the TLS parameters (cipher
1074 and peerdn, if requested) that were in fact used only for the second
1077 PH/31 When doing a callout as part of verifying an address, Exim was not paying
1078 attention to any local part prefix or suffix that was matched by the
1079 router that accepted the address. It now behaves in the same way as it
1080 does for delivery: the affixes are removed from the local part unless
1081 rcpt_include_affixes is set on the transport.
1083 PH/32 Add the sender address, as F=<...>, to the log line when logging a
1084 timeout during the DATA phase of an incoming message.
1086 PH/33 Sieve envelope tests were broken for match types other than :is. I have
1087 applied a patch sanctioned by the Sieve maintainer.
1089 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
1090 the uid or gid is negative. A case of a negative gid caused this to be
1091 noticed. The fix allows for either to be negative.
1093 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
1094 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
1097 PH/36 The change PH/12 above was broken. Fixed it.
1099 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
1100 the grounds that routing the same address twice would always produce the
1101 same answer. This might have been true once, but it is certainly no
1102 longer true now. Routing a child address may depend on the previous
1103 routing that produced that child. Some complicated redirection strategies
1104 went wrong when messages had multiple recipients, and made Exim's
1105 behaviour dependent on the order in which the addresses were given.
1107 I have moved the duplicate checking until after the routing is complete.
1108 Exim scans the addresses that are assigned to local and remote
1109 transports, and removes any duplicates. This means that more work will be
1110 done, as duplicates will always all be routed, but duplicates are
1111 presumably rare, so I don't expect this is of any significance.
1113 For deliveries to pipes, files, and autoreplies, the duplicate checking
1114 still happens during the routing process, since they are not going to be
1117 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
1118 It corrects a timeout issue with spamd. This is Ian's comment: "The
1119 background is that sometimes spamd either never reads data from a
1120 connection it has accepted, or it never writes response data. The exiscan
1121 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
1122 blindly assumes that writes won't block so it may never time out."
1124 PH/39 Allow G after quota size as well as K and M.
1126 PH/40 The value set for $authenticated_id in an authenticator may not contain
1127 binary zeroes or newlines because the value is written to log lines and
1128 to spool files. There was no check on this. Now the value is run through
1129 the string_printing() function so that such characters are converted to
1130 printable escape sequences.
1132 PH/41 $message_linecount is a new variable that contains the total number of
1133 lines in the message. Compare $body_linecount, which is the count for the
1136 PH/42 Exim no longer gives details of delivery errors for specific addresses in
1137 bounce and delay warning messages, except in certain special cases, which
1140 (a) An SMTP error message from a remote host;
1141 (b) A message specified in a :fail: redirection;
1142 (c) A message specified in a "fail" command in a system filter;
1143 (d) A message specified in a FAIL return from the queryprogram router;
1144 (e) A message specified by the cannot_route_message router option.
1146 In these cases only, Exim does include the error details in bounce and
1147 warning messages. There are also a few cases where bland messages such
1148 as "unrouteable address" or "local delivery error" are given.
1150 PH/43 $value is now also set for the "else" part of a ${run expansion.
1152 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
1153 being worked on, but at least Exim now implements the latest version to
1156 PH/45 In a pipe transport, although a timeout while waiting for the pipe
1157 process to complete was treated as a delivery failure, a timeout while
1158 writing the message to the pipe was logged, but erroneously treated as a
1159 successful delivery. Such timeouts include transport filter timeouts. For
1160 consistency with the overall process timeout, these timeouts are now
1161 treated as errors, giving rise to delivery failures by default. However,
1162 there is now a new Boolean option for the pipe transport called
1163 timeout_defer, which, if set TRUE, converts the failures into defers for
1164 both kinds of timeout. A transport filter timeout is now identified in
1167 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
1168 systems where "make" and "gmake" are different, calling "gmake" at top
1169 level broke things. I've arranged for the value of $(MAKE) to be passed
1170 from the Makefile to this script so that it can call the same version of
1174 A note about Exim versions 4.44 and 4.50
1175 ----------------------------------------
1177 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
1178 changes of various kinds. As a consequence, a big documentation update was
1179 needed. This delayed the release for rather longer than seemed good, especially
1180 in the light of a couple of (minor) security issues. Therefore, the changes
1181 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
1182 release. So 4.44 and 4.50 are in effect two different branches that both start
1185 I have left the 4.50 change log unchanged; it contains all the changes since
1186 4.43. The change log for 4.44 is below; many of its items are identical to
1187 those for 4.50. This seems to be the most sensible way to preserve the
1188 historical information.
1194 1. Minor wording change to the doc/README.SIEVE file.
1196 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
1197 computation of the current number of files was incorrect.
1199 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
1200 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
1203 4. Give more explanation in the error message when the command for a transport
1204 filter fails to execute.
1206 5. There are several places where Exim runs a non-Exim command in a
1207 subprocess. The SIGUSR1 signal should be disabled for these processes. This
1208 was being done only for the command run by the queryprogram router. It is
1209 now done for all such subprocesses. The other cases are: ${run, transport
1210 filters, and the commands run by the lmtp and pipe transports.
1212 6. Added CONFIGURE_GROUP build-time option.
1214 7. Some older OS have a limit of 256 on the maximum number of file
1215 descriptors. Exim was using setrlimit() to set 1000 as a large value
1216 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
1217 systems. I've change it so that if it can't get 1000, it tries for 256.
1219 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
1220 was an oversight, and furthermore, ever since the addition of extra
1221 controls (e.g. 4.43/32), the checks on when to allow different forms of
1222 "control" were broken. There should now be diagnostics for all cases when a
1223 control that does not make sense is encountered.
1225 9. Added the /retain_sender option to "control=submission".
1227 10. $recipients is now available in the predata ACL (oversight).
1229 11. Tidy the search cache before the fork to do a delivery from a message
1230 received from the command line. Otherwise the child will trigger a lookup
1231 failure and thereby defer the delivery if it tries to use (for example) a
1232 cached ldap connection that the parent has called unbind on.
1234 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
1235 of $address_data from the recipient verification was clobbered by the
1236 sender verification.
1238 13. The value of address_data from a sender verification is now available in
1239 $sender_address_data in subsequent conditions in the ACL statement.
1241 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
1243 15. Added a new option "connect=<time>" to callout options, to set a different
1246 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
1247 was its contents. (It was OK if the option was not defined at all.)
1249 17. A "Completed" log line is now written for messages that are removed from
1250 the spool by the -Mrm option.
1252 18. New variables $sender_verify_failure and $recipient_verify_failure contain
1253 information about exactly what failed.
1255 19. Added -dd to debug only the daemon process.
1257 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
1258 handles timeouts, both on the server side and network timeouts. Renamed the
1259 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
1261 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
1262 It is now set to "smtps".
1264 22. $host_address is now set to the target address during the checking of
1265 ignore_target_hosts.
1267 23. When checking ignore_target_hosts for an ipliteral router, no host name was
1268 being passed; this would have caused $sender_host_name to have been used if
1269 matching the list had actually called for a host name (not very likely,
1270 since this list is usually IP addresses). A host name is now passed as
1273 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
1274 code that specifies a non-restarting handler (typically sigaction() in
1275 modern systems) in an attempt to fix a rare and obscure crash bug.
1277 25. Narrowed the window for a race in the daemon that could cause it to ignore
1278 SIGCHLD signals. This is not a major problem, because they are used only to
1279 wake it up if nothing else does.
1281 26. A malformed maildirsize file could cause Exim to calculate negative values
1282 for the mailbox size or file count. Odd effects could occur as a result.
1283 The maildirsize information is now recalculated if the size or filecount
1286 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
1287 support for a long time. Removed HAVE_SYS_VFS_H.
1289 28. Installed the latest version of exipick from John Jetmore.
1291 29. In an address list, if the pattern was not a regular expression, an empty
1292 subject address (from a bounce message) matched only if the pattern was an
1293 empty string. Non-empty patterns were not even tested. This was the wrong
1294 because it is perfectly reasonable to use an empty address as part of a
1295 database query. An empty address is now tested by patterns that are
1296 lookups. However, all the other forms of pattern expect the subject to
1297 contain a local part and a domain, and therefore, for them, an empty
1298 address still always fails if the pattern is not itself empty.
1300 30. Exim went into a mad DNS loop when attempting to do a callout where the
1301 host was specified on an smtp transport, and looking it up yielded more
1302 than one IP address.
1304 31. Re-factored the code for checking spool and log partition space into a
1305 function that finds that data and another that does the check. The former
1306 is then used to implement four new variables: $spool_space, $log_space,
1307 $spool_inodes, and $log_inodes.
1309 32. The RFC2047 encoding function was originally intended for short strings
1310 such as real names; it was not keeping to the 75-character limit for
1311 encoded words that the RFC imposes. It now respects the limit, and
1312 generates multiple encoded words if necessary. To be on the safe side, I
1313 have increased the buffer size for the ${rfc2047: expansion operator from
1316 33. It is now permitted to omit both strings after an "if" condition; if the
1317 condition is true, the result is "true". As before, when the second string
1318 is omitted, a false condition yields an empty string. This makes it less
1319 cumbersome to write custom ACL and router conditions.
1321 34. Failure to deliver a bounce message always caused it to be frozen, even if
1322 there was an errors_to setting on the router. The errors_to setting is now
1325 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
1326 canonical form (fully expanded) before being placed in
1327 $sender_host_address.
1329 36. The table in the code that translates DNS record types into text (T_A to
1330 "A" for instance) was missing entries for NS and CNAME. It is just possible
1331 that this could have caused confusion if both these types were looked up
1332 for the same domain, because the text type is used as part of Exim's
1333 per-process caching. But the chance of anyone hitting this buglet seems
1336 37. The dnsdb lookup has been extended in a number of ways.
1338 (1) There is a new type, "zns", which walks up the domain tree until it
1339 finds some nameserver records. It should be used with care.
1341 (2) There is a new type, "mxh", which is like "mx" except that it returns
1342 just the host names, not the priorities.
1344 (3) It is now possible to give a list of domains (or IP addresses) to be
1345 looked up. The behaviour when one of the lookups defers can be
1346 controlled by a keyword.
1348 (4) It is now possible to specify the separator character for use when
1349 multiple records are returned.
1351 38. The dnslists ACL condition has been extended: it is now possible to supply
1352 a list of IP addresses and/or domains to be looked up in a particular DNS
1355 39. Added log_selector=+queue_time_overall.
1357 40. When running the queue in the test harness, wait just a tad after forking a
1358 delivery process, to get repeatability of debugging output.
1360 41. Include certificate and key file names in error message when GnuTLS fails
1361 to set them up, because the GnuTLS error message doesn't include the name
1362 of the failing file when there is a problem reading it.
1364 42. Allow both -bf and -bF in the same test run.
1366 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
1368 44. The "Exiscan patch" is now merged into the mainline Exim source.
1370 45. Sometimes the final signoff response after QUIT could fail to get
1371 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
1372 < 0 before doing a fflush(). This bug looks as though it goes back to the
1373 introduction of TLS in release 3.20, but "sometimes" must have been rare
1374 because the tests only now provoked it.
1376 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
1377 (this can affect the format of dates).
1379 47. exim_tidydb, when checking for the continued existence of a message for
1380 which it has found a message-specific retry record, was not finding
1381 messages that were in split spool directories. Consequently, it was
1382 deleting retry records that should have stayed in existence.
1384 48. Steve fixed some bugs in eximstats.
1386 49. The SPA authentication driver was not abandoning authentication and moving
1387 on to the next authenticator when an expansion was forced to fail,
1388 contradicting the general specification for all authenticators. Instead it
1389 was generating a temporary error. It now behaves as specified.
1391 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
1392 (the order specifies the preference for clients). The order is now AES256,
1393 AES128, 3DES, ARCFOUR128.
1395 51. Small patch to Sieve code - explicitly set From: when generating an
1398 52. Exim crashed if a remote delivery caused a very long error message to be
1399 recorded - for instance if somebody sent an entire SpamAssassin report back
1400 as a large number of 550 error lines. This bug was coincidentally fixed by
1401 increasing the size of one of Exim's internal buffers (big_buffer) that
1402 happened as part of the Exiscan merge. However, to be on the safe side, I
1403 have made the code more robust (and fixed the comments that describe what
1406 53. Now that there can be additional text after "Completed" in log lines (if
1407 the queue_time_overall log selector is set), a one-byte patch to exigrep
1408 was needed to allow it to recognize "Completed" as not the last thing in
1411 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
1412 patch that reportedly fixes this has been added. I am not expert enough to
1413 create a test for it. This is what the patch creator wrote:
1415 "I found a little strange behaviour of ldap code when working with
1416 Windows 2003 AD Domain, where users was placed in more than one
1417 Organization Units. When I tried to give exim partial DN, the exit code
1418 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
1419 But simultaneously result of request was absolutely normal ldap result,
1420 so I produce this patch..."
1422 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
1423 REFERENCE, so I have modified the code to exclude the patch when that macro
1426 55. Some experimental protocols are using DNS PTR records for new purposes. The
1427 keys for these records are domain names, not reversed IP addresses. The
1428 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
1429 leaves it alone. Component reversal etc. now happens only for IP addresses.
1432 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
1434 57. Double the size of the debug message buffer (to 2048) so that more of very
1435 long debug lines gets shown.
1437 58. The exicyclog utility now does better if the number of log files to keep
1438 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
1440 59. Two changes related to the smtp_active_hostname option:
1442 (1) $smtp_active_hostname is now available as a variable.
1443 (2) The default for smtp_banner uses $smtp_active_hostname instead
1444 of $primary_hostname.
1446 60. The host_aton() function is supposed to be passed a string that is known
1447 to be a valid IP address. However, in the case of IPv6 addresses, it was
1448 not checking this. This is a hostage to fortune. Exim now panics and dies
1449 if the condition is not met. A case was found where this could be provoked
1450 from a dnsdb PTR lookup with an IPv6 address that had more than 8
1451 components; fortuitously, this particular loophole had already been fixed
1452 by change 4.50/55 above.
1454 If there are any other similar loopholes, the new check in host_aton()
1455 itself should stop them being exploited. The report I received stated that
1456 data on the command line could provoke the exploit when Exim was running as
1457 exim, but did not say which command line option was involved. All I could
1458 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
1459 running as the user.
1462 61. There was a buffer overflow vulnerability in the SPA authentication code
1463 (which came originally from the Samba project). I have added a test to the
1464 spa_base64_to_bits() function which I hope fixes it.
1467 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
1468 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
1470 63. The daemon start-up calls getloadavg() while still root for those OS that
1471 need the first call to be done as root, but it missed one case: when
1472 deliver_queue_load_max is set with deliver_drop_privilege. This is
1473 necessary for the benefit of the queue runner, because there is no re-exec
1474 when deliver_drop_privilege is set.
1476 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
1477 This has been fixed.
1479 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
1480 was in use, was not putting the data itself into the right store pool;
1481 consequently, it could be overwritten for a subsequent message in the same
1482 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
1485 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
1487 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
1488 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
1489 the function were treating the return as a boolean value, which happened to
1490 work because 0=false and not-0=true, but is not correct code.
1492 68. The host_aton() function was not handling scoped IPv6 addresses (those
1493 with, for example, "%eth0" on the end) correctly.
1495 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
1496 negated items (that is, ~something) in unsigned ints. Some compilers
1497 apparently mutter when there is no cast.
1499 70. If an address verification called from an ACL failed, and did not produce a
1500 user-specific message (i.e. there was only a "system" message), nothing was
1501 put in $acl_verify_message. In this situation, it now puts the system
1504 71. Change 4.23/11 added synchronization checking at the start of an SMTP
1505 session; change 4.31/43 added the unwanted input to the log line - except
1506 that it did not do this in the start of session case. It now does.
1508 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
1509 This is wrong and can cause the other end to generate a synchronization
1510 error if it is another Exim or anything else that does the synchronization
1511 check. A QUIT command is no longer sent after a timeout.
1513 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
1514 during host lookups.
1516 74. The defer_ok option of callout verification was not working if it was used
1517 when verifying addresses in header lines, that is, for this case:
1519 verify = header_sender/callout=defer_ok
1521 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
1522 those file descriptors could be used for SMTP connections. If anything
1523 wrote to stderr (the example that came up was "warn" in embedded Perl), it
1524 could be sent to the SMTP client, causing chaos. The daemon now opens
1525 stdin, stdout, and stderr to /dev/null when it puts itself into the
1528 76. Arrange for output from Perl's "warn" command to be written to Exim's main
1529 log by default. The user can override this with suitable Perl magic.
1531 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
1532 the log message when discard triggers, was not working for the DATA ACL or
1533 for the non-SMTP ACL.
1535 78. Error message wording change in sieve.c.
1537 79. If smtp_accept_max_per_host was set, the number of connections could be
1538 restricted to fewer than expected, because the daemon was trying to set up
1539 a new connection before checking whether the processes handling previous
1540 connections had finished. The check for completed processes is now done
1541 earlier. On busy systems, this bug wouldn't be noticed because something
1542 else would have woken the daemon, and it would have reaped the completed
1545 80. If a message was submitted locally by a user whose login name contained one
1546 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
1547 It caused a spool format error. I have fixed the spool reading code. A
1548 related problem was that the "from" clause in the Received: line became
1549 illegal because of the space(s). It is now covered by ${quote_local_part.
1551 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
1554 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
1556 83. Patch from Sieve maintainer to fix unterminated string problem in
1557 "vacation" handling.
1559 84. Some minor changes to the Linux configuration files to help with other
1560 OS variants using glibc.
1562 85. One more patch for Sieve to update vacation handling to latest spec.
1565 ----------------------------------------------------
1566 See the note above about the 4.44 and 4.50 releases.
1567 ----------------------------------------------------
1573 1. Change 4.43/35 introduced a bug that caused file counts to be
1574 incorrectly computed when quota_filecount was set in an appendfile
1577 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
1578 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
1581 3. Give more explanation in the error message when the command for a transport
1582 filter fails to execute.
1584 4. There are several places where Exim runs a non-Exim command in a
1585 subprocess. The SIGUSR1 signal should be disabled for these processes. This
1586 was being done only for the command run by the queryprogram router. It is
1587 now done for all such subprocesses. The other cases are: ${run, transport
1588 filters, and the commands run by the lmtp and pipe transports.
1590 5. Some older OS have a limit of 256 on the maximum number of file
1591 descriptors. Exim was using setrlimit() to set 1000 as a large value
1592 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
1593 systems. I've change it so that if it can't get 1000, it tries for 256.
1595 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
1596 was an oversight, and furthermore, ever since the addition of extra
1597 controls (e.g. 4.43/32), the checks on when to allow different forms of
1598 "control" were broken. There should now be diagnostics for all cases when a
1599 control that does not make sense is encountered.
1601 7. $recipients is now available in the predata ACL (oversight).
1603 8. Tidy the search cache before the fork to do a delivery from a message
1604 received from the command line. Otherwise the child will trigger a lookup
1605 failure and thereby defer the delivery if it tries to use (for example) a
1606 cached ldap connection that the parent has called unbind on.
1608 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
1609 of $address_data from the recipient verification was clobbered by the
1610 sender verification.
1612 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
1613 was its contents. (It was OK if the option was not defined at all.)
1615 11. A "Completed" log line is now written for messages that are removed from
1616 the spool by the -Mrm option.
1618 12. $host_address is now set to the target address during the checking of
1619 ignore_target_hosts.
1621 13. When checking ignore_target_hosts for an ipliteral router, no host name was
1622 being passed; this would have caused $sender_host_name to have been used if
1623 matching the list had actually called for a host name (not very likely,
1624 since this list is usually IP addresses). A host name is now passed as
1627 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
1628 code that specifies a non-restarting handler (typically sigaction() in
1629 modern systems) in an attempt to fix a rare and obscure crash bug.
1631 15. Narrowed the window for a race in the daemon that could cause it to ignore
1632 SIGCHLD signals. This is not a major problem, because they are used only to
1633 wake it up if nothing else does.
1635 16. A malformed maildirsize file could cause Exim to calculate negative values
1636 for the mailbox size or file count. Odd effects could occur as a result.
1637 The maildirsize information is now recalculated if the size or filecount
1640 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
1641 support for a long time. Removed HAVE_SYS_VFS_H.
1643 18. Updated exipick to current release from John Jetmore.
1645 19. Allow an empty sender to be matched against a lookup in an address list.
1646 Previously the only cases considered were a regular expression, or an
1649 20. Exim went into a mad DNS lookup loop when doing a callout where the
1650 host was specified on the transport, if the DNS lookup yielded more than
1653 21. The RFC2047 encoding function was originally intended for short strings
1654 such as real names; it was not keeping to the 75-character limit for
1655 encoded words that the RFC imposes. It now respects the limit, and
1656 generates multiple encoded words if necessary. To be on the safe side, I
1657 have increased the buffer size for the ${rfc2047: expansion operator from
1660 22. Failure to deliver a bounce message always caused it to be frozen, even if
1661 there was an errors_to setting on the router. The errors_to setting is now
1664 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
1665 canonical form (fully expanded) before being placed in
1666 $sender_host_address.
1668 24. Updated eximstats to version 1.33
1670 25. Include certificate and key file names in error message when GnuTLS fails
1671 to set them up, because the GnuTLS error message doesn't include the name
1672 of the failing file when there is a problem reading it.
1674 26. Expand error message when OpenSSL has problems setting up cert/key files.
1677 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
1678 (this can affect the format of dates).
1680 28. exim_tidydb, when checking for the continued existence of a message for
1681 which it has found a message-specific retry record, was not finding
1682 messages that were in split spool directories. Consequently, it was
1683 deleting retry records that should have stayed in existence.
1685 29. eximstats updated to version 1.35
1686 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
1687 1.35 - bugfix such that pie charts by volume are generated correctly
1689 30. The SPA authentication driver was not abandoning authentication and moving
1690 on to the next authenticator when an expansion was forced to fail,
1691 contradicting the general specification for all authenticators. Instead it
1692 was generating a temporary error. It now behaves as specified.
1694 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
1695 (the order specifies the preference for clients). The order is now AES256,
1696 AES128, 3DES, ARCFOUR128.
1698 31. Small patch to Sieve code - explicitly set From: when generating an
1701 32. Exim crashed if a remote delivery caused a very long error message to be
1702 recorded - for instance if somebody sent an entire SpamAssassin report back
1703 as a large number of 550 error lines. This bug was coincidentally fixed by
1704 increasing the size of one of Exim's internal buffers (big_buffer) that
1705 happened as part of the Exiscan merge. However, to be on the safe side, I
1706 have made the code more robust (and fixed the comments that describe what
1709 33. Some experimental protocols are using DNS PTR records for new purposes. The
1710 keys for these records are domain names, not reversed IP addresses. The
1711 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
1712 leaves it alone. Component reversal etc. now happens only for IP addresses.
1715 34. The host_aton() function is supposed to be passed a string that is known
1716 to be a valid IP address. However, in the case of IPv6 addresses, it was
1717 not checking this. This is a hostage to fortune. Exim now panics and dies
1718 if the condition is not met. A case was found where this could be provoked
1719 from a dnsdb PTR lookup with an IPv6 address that had more than 8
1720 components; fortuitously, this particular loophole had already been fixed
1721 by change 4.50/55 or 4.44/33 above.
1723 If there are any other similar loopholes, the new check in host_aton()
1724 itself should stop them being exploited. The report I received stated that
1725 data on the command line could provoke the exploit when Exim was running as
1726 exim, but did not say which command line option was involved. All I could
1727 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
1728 running as the user.
1731 35. There was a buffer overflow vulnerability in the SPA authentication code
1732 (which came originally from the Samba project). I have added a test to the
1733 spa_base64_to_bits() function which I hope fixes it.
1736 36. The daemon start-up calls getloadavg() while still root for those OS that
1737 need the first call to be done as root, but it missed one case: when
1738 deliver_queue_load_max is set with deliver_drop_privilege. This is
1739 necessary for the benefit of the queue runner, because there is no re-exec
1740 when deliver_drop_privilege is set.
1742 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
1743 was in use, was not putting the data itself into the right store pool;
1744 consequently, it could be overwritten for a subsequent message in the same
1745 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
1748 38. Sometimes the final signoff response after QUIT could fail to get
1749 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
1750 < 0 before doing a fflush(). This bug looks as though it goes back to the
1751 introduction of TLS in release 3.20, but "sometimes" must have been rare
1752 because the tests only now provoked it.
1758 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
1759 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
1760 Now it returns an integer. A number of calls were still expecting a T/F
1761 return. Fortuitously, in all cases, the tests worked in OK situations,
1762 which is the norm. However, things would have gone wrong on any write
1763 failures on the smtp file descriptor. This function is used when sending
1764 messages over SMTP and also when doing verify callouts.
1766 2. When Exim is called to do synchronous delivery of a locally submitted
1767 message (the -odf or -odi options), it no longer closes stderr before doing
1770 3. Implemented the mua_wrapper option.
1772 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
1774 5. Implemented the functions header_remove(), header_testname(),
1775 header_add_at_position(), and receive_remove_recipient(), and exported them
1778 6. If an ACL "warn" statement specified the addition of headers, Exim already
1779 inserted X-ACL-Warn: at the start if there was no header name. However, it
1780 was not making this test for the second and subsequent header lines if
1781 there were newlines in the string. This meant that an invalid header could
1782 be inserted if Exim was badly configured.
1784 7. Allow an ACL "warn" statement to add header lines at the start or after all
1785 the Received: headers, as well as at the end.
1787 8. Added the rcpt_4xx retry error code.
1789 9. Added postmaster_mailfrom=xxx to callout verification option.
1791 10. Added mailfrom=xxxx to the callout verification option, for verify=
1794 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
1795 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
1797 12. Inserted some casts to stop certain compilers warning when using pointer
1798 differences as field lengths or precisions in printf-type calls (mostly
1799 affecting debugging statements).
1801 13. Added optional readline() support for -be (dynamically loaded).
1803 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
1804 same clock tick as a message's arrival, so that its received time was the
1805 same as the "first fail" time on the retry record, and that message
1806 remained on the queue past the ultimate address timeout, every queue runner
1807 would try a delivery (because it was past the ultimate address timeout) but
1808 after another failure, the ultimate address timeout, which should have then
1809 bounced the address, did not kick in. This was a "< instead of <=" error;
1810 in most cases the first failure would have been in the next clock tick
1811 after the received time, and all would be well.
1813 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
1814 being recognized when the domain list was tested by the match_domain
1815 condition in an expansion string.
1817 16. Added the ${str2b64: operator.
1819 17. Exim was always calling setrlimit() to set a large limit for the number of
1820 processes, without checking whether the existing limit was already
1821 adequate. (It did check for the limit on file descriptors.) Furthermore,
1822 errors from getrlimit() and setrlimit() were being ignored. Now they are
1823 logged to the main and panic logs, but Exim does carry on, to try to do its
1824 job under whatever limits there are.
1826 18. Imported PCRE 5.0.
1828 19. Trivial typo in log message " temporarily refused connection" (the leading
1831 20. If the log selector return_path_on_delivery was set and an address was
1832 redirected to /dev/null, the delivery process crashed because it assumed
1833 that a return path would always be set for a "successful" delivery. In this
1834 case, the whole delivery is bypassed as an optimization, and therefore no
1837 21. Internal re-arrangement: the function for sending a challenge and reading
1838 a response while authentication was assuming a zero-terminated challenge
1839 string. It's now changed to take a pointer and a length, to allow for
1840 binary data in such strings.
1842 22. Added the cyrus_sasl authenticator (code supplied by MBM).
1844 23. Exim was not respecting finduser_retries when seeking the login of the
1845 uid under which it was called; it was always trying 10 times. (The default
1846 setting of finduser_retries is zero.) Also, it was sleeping after the final
1847 failure, which is pointless.
1849 24. Implemented tls_on_connect_ports.
1851 25. Implemented acl_smtp_predata.
1853 26. If the domain in control=submission is set empty, Exim assumes that the
1854 authenticated id is a complete email address when it generates From: or
1855 Sender: header lines.
1857 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
1858 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
1859 chown and chgrp in /bin and hostname in /usr/bin.
1861 28. Exim was keeping the "process log" file open after each use, just as it
1862 does for the main log. This opens the possibility of it remaining open for
1863 long periods when the USR1 signal hits a daemon. Occasional processlog
1864 errors were reported, that could have been caused by this. Anyway, it seems
1865 much more sensible not to leave this file open at all, so that is what now
1868 29. The long-running daemon process does not normally write to the log once it
1869 has entered its main loop, and it closes the log before doing so. This is
1870 so that log files can straightforwardly be renamed and moved. However,
1871 there are a couple of unusual error situations where the daemon does write
1872 log entries, and I had neglected to close the log afterwards.
1874 30. The text of an SMTP error response that was received during a remote
1875 delivery was being truncated at 512 bytes. This is too short for some of
1876 the long messages that one sometimes sees. I've increased the limit to
1879 31. It is now possible to make retry rules that apply only when a message has a
1880 specific sender, in particular, an empty sender.
1882 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
1883 it possible to be selective about when SMTP synchronization is enforced.
1885 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
1887 32. Implemented hosts_connection_nolog.
1889 33. Added an ACL for QUIT.
1891 34. Setting "delay_warning=" to disable warnings was not working; it gave a
1894 35. Added mailbox_size and mailbox_filecount to appendfile.
1896 36. Added control = no_multiline_responses to ACLs.
1898 37. There was a bug in the logic of the code that waits for the clock to tick
1899 in the case where the clock went backwards by a substantial amount such
1900 that the microsecond fraction of "now" was more than the microsecond
1901 fraction of "then" (but the whole seconds number was less).
1903 38. Added support for the libradius Radius client library this is found on
1904 FreeBSD (previously only the radiusclient library was supported).
1910 1. When certain lookups returned multiple values in the form name=value, the
1911 quoting of the values was not always being done properly. Specifically:
1912 (a) If the value started with a double quote, but contained no whitespace,
1914 (b) If the value contained whitespace other than a space character (i.e.
1915 tabs or newlines or carriage returns) it was not quoted.
1916 This fix has been applied to the mysql and pgsql lookups by writing a
1917 separate quoting function and calling it from the lookup code. The fix
1918 should probably also be applied to nisplus, ibase and oracle lookups, but
1919 since I cannot test any of those, I have not disturbed their existing code.
1921 2. A hit in the callout cache for a specific address caused a log line with no
1922 reason for rejecting RCPT. Now it says "Previous (cached) callout
1923 verification failure".
1925 3. There was an off-by-one bug in the queryprogram router. An over-long
1926 return line was truncated at 256 instead of 255 characters, thereby
1927 overflowing its buffer with the terminating zero. As well as fixing this, I
1928 have increased the buffer size to 1024 (and made a note to document this).
1930 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
1931 when Exim is waiting for an SMTP response from a remote server, Exim
1932 restarts its select() call on the socket, thereby resetting its timeout.
1933 This is not a problem when such interrupts are rare. Somebody set up a cron
1934 job to run exiwhat every 2 minutes, which is less than the normal select()
1935 timeout (5 or 10 minutes). This meant that the select() timeout never
1936 kicked in because it was always reset. I have fixed this by comparing the
1937 time when an interrupt arrives with the time at the start of the first call
1938 to select(). If more time than the timeout has elapsed, the interrupt is
1939 treated as a timeout.
1941 5. Some internal re-factoring in preparation for the addition of Sieve
1942 extensions (by MH). In particular, the "personal" test is moved to a
1943 separate function, and given an option for scanning Cc: and Bcc: (which is
1944 not set for Exim filters).
1946 6. When Exim created an email address using the login of the caller as the
1947 local part (e.g. when creating a From: or Sender: header line), it was not
1948 quoting the local part when it contained special characters such as @.
1950 7. Installed new OpenBSD configuration files.
1952 8. Reworded some messages for syntax errors in "and" and "or" conditions to
1953 try to make them clearer.
1955 9. Callout options, other than the timeout value, were being ignored when
1956 verifying sender addresses in header lines. For example, when using
1958 verify = header_sender/callout=no_cache
1960 the cache was (incorrectly) being used.
1962 10. Added a missing instance of ${EXE} to the exim_install script; this affects
1963 only the Cygwin environment.
1965 11. When return_path_on_delivery was set as a log selector, if different remote
1966 addresses in the same message used different return paths and parallel
1967 remote delivery occurred, the wrong values would sometimes be logged.
1968 (Whenever a remote delivery process finished, the return path value from
1969 the most recently started remote delivery process was logged.)
1971 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
1972 lines when AUTH and/or TLS are in use. This is the "received protocol"
1973 field. Exim used to use "asmtp" for authenticated SMTP, without any
1974 indication (in the protocol name) for TLS use. Now it follows the RFC and
1975 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
1976 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
1977 names appear in log lines as well as in Received: header lines.
1979 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
1980 extensions, and comparison tests, and to fix some bugs.
1982 14. Changes to the "personal" filter test:
1984 (1) The test was buggy in that it was just doing the equivalent of
1985 "contains" tests on header lines. For example, if a user's address was
1986 anne@some.where, the "personal" test would incorrectly be true for
1988 To: susanne@some.where
1990 This test is now done by extracting each address from the header in turn,
1991 and checking the entire address. Other tests that are part of "personal"
1992 are now done using regular expressions (for example, to check local parts
1993 of addresses in From: header lines).
1995 (2) The list of non-personal local parts in From: addresses has been
1996 extended to include "listserv", "majordomo", "*-request", and "owner-*",
1997 taken from the Sieve specification recommendations.
1999 (3) If the message contains any header line starting with "List-" it is
2000 treated as non-personal.
2002 (4) The test for "circular" in the Subject: header line has been removed
2003 because it now seems ill-conceived.
2005 15. Minor typos in src/EDITME comments corrected.
2007 16. Installed latest exipick from John Jetmore.
2009 17. If headers_add on a router specified a text string that was too long for
2010 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
2011 of string_sprintf() is now avoided.
2013 18. $message_body_size was not set (it was always zero) when running the DATA
2014 ACL and the local_scan() function.
2016 19. For the "mail" command in an Exim filter, no default was being set for
2017 the once_repeat time, causing a random time value to be used if "once" was
2018 specified. (If the value happened to be <= 0, no repeat happened.) The
2019 default is now 0s, meaning "never repeat". The "vacation" command was OK
2020 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
2021 (I found it when inspecting the code).
2023 20. There is now an overall timeout for performing a callout verification. It
2024 defaults to 4 times the callout timeout, which applies to individual SMTP
2025 commands during the callout. The overall timeout applies when there is more
2026 than one host that can be tried. The timeout is checked before trying the
2027 next host. This prevents very long delays if there are a large number of
2028 hosts and all are timing out (e.g. when the network connections are timing
2029 out). The value of the overall timeout can be changed by specifying an
2030 additional sub-option for "callout", called "maxwait". For example:
2032 verify = sender/callout=5s,maxwait=20s
2034 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
2035 to the end before writing, but this should make it even safer).
2037 22. Exim was forgetting that it had advertised PIPELINING for the second and
2038 subsequent messages on an SMTP connection. It was also not resetting its
2039 memory on STARTTLS and an internal HELO.
2041 23. When Exim logs an SMTP synchronization error within a session, it now
2042 records whether PIPELINING has been advertised or not.
2044 24. Added 3 instances of "(long int)" casts to time_t variables that were being
2045 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
2046 rather than long int.
2048 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
2050 26. Added the never_mail option to autoreply.
2056 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
2057 crash if the getsockname() call failed; this can happen if a connection is
2058 closed very soon after it is established. The problem was simply in the
2059 order in which certain operations were done, causing Exim to try to write
2060 to the SMTP stream before it had set up the file descriptor. The bug has
2061 been fixed by making things happen in the correct order.
2067 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
2068 before the connection was closed, thus losing the rejection response.
2070 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
2071 some early Solaris releases, but causes trouble in current releases where
2072 socklen_t is defined.
2074 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
2077 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
2080 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
2081 that releasing the top of it at the end releases what was used for sub-
2082 expansions (unless the block got too big). However, discard this block if
2083 the first thing is a variable or header, so that we can use its block when
2084 it is dynamic (useful for very large $message_headers, for example).
2086 6. Lookups now cache *every* query, not just the most recent. A new, separate
2087 store pool is used for this. It can be recovered when all lookup caches are
2088 flushed. Lookups now release memory at the end of their result strings.
2089 This has involved some general refactoring of the lookup sources.
2091 7. Some code has been added to the store_xxx() functions to reduce the amount
2092 of flapping under certain conditions.
2094 8. log_incoming_interface used to affect only the <= reception log lines. Now
2095 it causes the local interface and port to be added to several more SMTP log
2096 lines, for example "SMTP connection from", and rejection lines.
2098 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
2100 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
2102 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
2103 could be overwritten at the end of the current message (or the start of a
2104 new message if it was set in a HELO ACL). The value is now preserved for
2105 the duration of the SMTP connection.
2107 12. If a transport had a headers_rewrite setting, and a matching header line
2108 contained an unqualified address, that address was qualified, even if it
2109 did not match any rewriting rules. The underlying bug was that the values
2110 of the flags that permit the existence of unqualified sender and recipient
2111 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
2112 non-local messages, and by -bnq for local messages) were not being
2113 preserved with the message after it was received.
2115 13. When Exim was logging an SMTP synchronization error, it could sometimes log
2116 "next input=" as part of the text comprising the host identity instead of
2117 the correct text. The code was using the same buffer for two different
2118 strings. However, depending on which order the printing function evaluated
2119 its arguments, the bug did not always show up. Under Linux, for example, my
2120 test suite worked just fine.
2122 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
2123 doesn't work with some older versions of Perl. It has been changed to "my",
2124 which in any case is probably the better facility to use.
2126 15. A really picky compiler found some instances of statements for creating
2127 error messages that either had too many or two few arguments for the format
2130 16. The size of the buffer for calls to the DNS resolver has been increased
2131 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
2132 for addresses that have a lot of PTR records. This alleviates a problem; it
2133 does not fully solve it.
2135 17. A dnsdb lookup for PTR records that receives more data than will fit in the
2136 buffer now truncates the list and logs the incident, which is the same
2137 action as happens when Exim is looking up a host name and its aliases.
2138 Previously in this situation something unpredictable would happen;
2139 sometimes it was "internal error: store_reset failed".
2141 18. If a server dropped the connection unexpectedly when an Exim client was
2142 using GnuTLS and trying to read a response, the client delivery process
2143 crashed while trying to generate an error log message.
2145 19. If a "warn" verb in an ACL added multiple headers to a message in a single
2146 string, for example:
2148 warn message = H1: something\nH2: something
2150 the text was added as a single header line from Exim's point of view
2151 though it ended up OK in the delivered message. However, searching for the
2152 second and subsequent header lines using $h_h2: did not work. This has been
2153 fixed. Similarly, if a system filter added multiple headers in this way,
2154 the routers could not see them.
2156 20. Expanded the error message when iplsearch is called with an invalid key to
2157 suggest using net-iplsearch in a host list.
2159 21. When running tests using -bh, any delays imposed by "delay" modifiers in
2160 ACLs are no longer actually imposed (and a message to that effect is
2163 22. If a "gecos" field in a passwd entry contained escaped characters, in
2164 particular, if it contained a \" sequence, Exim got it wrong when building
2165 a From: or a Sender: header from that name. A second bug also caused
2166 incorrect handling when an unquoted " was present following a character
2167 that needed quoting.
2169 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
2170 was not being matched caselessly.
2172 24. Arranged for all hyphens in the exim.8 source to be escaped with
2175 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
2176 properly. Recipient callout cache records were still being keyed to include
2177 the sender, even when use_sender was set false. This led to far more
2178 callouts that were necessary. The sender is no longer included in the key
2179 when use_sender is false.
2181 26. Added "control = submission" modifier to ACLs.
2183 27. Added the ${base62d: operator to decode base 62 numbers.
2185 28. dnsdb lookups can now access SRV records.
2187 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
2188 the configuration file.
2190 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
2191 (-v) mode. This makes the output for a verbose queue run more intelligible.
2193 31. Added a use_postmaster feature to recipient callouts.
2195 32. Added the $body_zerocount variable, containing the number of binary zero
2196 bytes in the message body.
2198 33. The time of last modification of the "new" subdirectory is now used as the
2199 "mailbox time last read" when there is a quota error for a maildir
2202 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
2204 35. Added +ignore_unknown as a special item in host lists.
2206 36. Code for decoding IPv6 addresses in host lists is now included, even if
2207 IPv6 support is not being compiled. This fixes a bug in which an IPv6
2208 address was recognized as an IP address, but was then not correctly decoded
2209 into binary, causing unexpected and incorrect effects when compared with
2216 1. Very minor rewording of debugging text in manualroute to say "list of
2217 hosts" instead of "hostlist".
2219 2. If verify=header_syntax was set, and a header line with an unqualified
2220 address (no domain) and a large number of spaces between the end of the
2221 name and the colon was received, the reception process suffered a buffer
2222 overflow, and (when I tested it) crashed. This was caused by some obsolete
2223 code that should have been removed. The fix is to remove it!
2225 3. When running in the test harness, delay a bit after writing a bounce
2226 message to get a bit more predictability in the log output.
2228 4. Added a call to search_tidyup() just before forking a reception process. In
2229 theory, someone could use a lookup in the expansion of smtp_accept_max_
2230 per_host which, without the tidyup, could leave open a database connection.
2232 5. Added the variables $recipient_data and $sender_data which get set from a
2233 lookup success in an ACL "recipients" or "senders" condition, or a router
2234 "senders" option, similar to $domain_data and $local_part_data.
2236 6. Moved the writing of debug_print from before to after the "senders" test
2239 7. Change 4.31/66 (moving the time when the Received: is generated) caused
2240 problems for message scanning, either using a data ACL, or using
2241 local_scan() because the Received: header was not generated till after they
2242 were called (in order to set the time as the time of reception completion).
2243 I have revised the way this works. The header is now generated after the
2244 body is received, but before the ACL or local_scan() are called. After they
2245 are run, the timestamp in the header is updated.
2251 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
2252 before starting a queue runner without re-exec. This happened only when
2253 deliver_drop_privilege was set or when the Exim user was set to root. The
2254 effect of the bug was that timeouts during subsequent deliveries caused
2255 crashes instead of being properly handled. The handler is now left at its
2256 default (and expected) setting.
2258 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
2259 message, again when deliver_drop_privilege is set or Exim is run as root.
2260 The bug described in (1) was not present in this case, but the tidying up
2261 of the other signals was missing. I have made the two cases consistent.
2263 3. The ignore_target_hosts setting on a manualroute router was being ignored
2264 for hosts that were looked up using the /MX notation.
2266 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
2269 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
2270 operated on the sender address. After changing the $sender_address to <>
2271 for the sender address verify, Exim was re-instated it as the original
2272 (before rewriting) address, but remembering that it had rewritten it, so it
2273 wasn't rewriting it again. This bug also had the effect of breaking the
2274 sender address verification caching when the sender address was rewritten.
2276 6. The ignore_target_hosts option was being ignored by the ipliteral router.
2277 This has been changed so that if the ip literal address matches
2278 ignore_target_hosts, the router declines.
2280 7. Added expansion conditions match_domain, match_address, and match_local_
2281 part (NOT match_host).
2283 8. The placeholder for the Received: header didn't have a length field set.
2285 9. Added code to Exim itself and to exim_lock to test for a specific race
2286 condition that could lead to file corruption when using MBX delivery. The
2287 issue is with the lockfile that is created in /tmp. If this file is removed
2288 after a process has opened it but before that process has acquired a lock,
2289 there is the potential for a second process to recreate the file and also
2290 acquire a lock. This could lead to two Exim processes writing to the file
2291 at the same time. The added code performs the same test as UW imapd; it
2292 checks after acquiring the lock that its file descriptor still refers to
2293 the same named file.
2295 10. The buffer for building added header lines was of fixed size, 8192 bytes.
2296 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
2299 11. Added the smtp_active_hostname option. If used, this will typically be made
2300 to depend on the incoming interface address. Because $interface_address is
2301 not set up until the daemon has forked a reception process, error responses
2302 that can happen earlier (such as "too many connections") no longer contain
2305 12. If an expansion in a condition on a "warn" statement fails because a lookup
2306 defers, the "warn" statement is abandoned, and the next ACL statement is
2307 processed. Previously this caused the whole ACL to be aborted.
2309 13. Added the iplsearch lookup type.
2311 14. Added ident_timeout as a log selector.
2313 15. Added tls_certificate_verified as a log selector.
2315 16. Added a global option tls_require_ciphers (compare the smtp transport
2316 option of the same name). This controls incoming TLS connections.
2318 17. I finally figured out how to make tls_require_ciphers do a similar thing
2319 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
2320 before starting the TLS session.
2322 18. Tabs are now shown as \t in -bP output.
2324 19. If the log selector return_path_on_delivery was set, Exim crashed when
2325 bouncing a message because it had too many Received: header lines.
2327 20. If two routers both had headers_remove settings, and the first one included
2328 a superfluous trailing colon, the final name in the first list and the
2329 first name in the second list were incorrectly joined into one item (with a
2330 colon in the middle).
2336 1. Added -C and -D options to the exinext utility, mainly to make it easier
2337 to include in the automated testing, but these could be helpful when
2338 multiple configurations are in use.
2340 2. The exinext utility was not formatting the output nicely when there was
2341 an alternate port involved in the retry record key, nor when there was a
2342 message id as well (for retries that were specific to a specific message
2343 and a specific host). It was also confused by IPv6 addresses, because of
2344 the additional colons they contain. I have fixed the IPv4 problem, and
2345 patched it up to do a reasonable job for IPv6.
2347 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
2348 delivery, the log line now contains "pipelined" if PIPELINING was used.
2350 4. An SMTP transport process used to panic and die if the bind() call to set
2351 an explicit outgoing interface failed. This has been changed; it is now
2352 treated in the same way as a connect() failure.
2354 5. A reference to $sender_host_name in the part of a conditional expansion
2355 that was being skipped was still causing a DNS lookup. This no longer
2358 6. The def: expansion condition was not recognizing references to header lines
2359 that used bh_ and bheader_.
2361 7. Added the _cache feature to named lists.
2363 8. The code for checking quota_filecount in the appendfile transport was
2364 allowing one more file than it should have been.
2366 9. For compatibility with Sendmail, the command line option
2374 and sets the incoming protocol and host name (for trusted callers). The
2375 host name and its colon can be omitted when only the protocol is to be set.
2376 Note the Exim already has two private options, -pd and -ps, that refer to
2377 embedded Perl. It is therefore impossible to set a protocol value of "d" or
2378 "s", but I don't think that's a major issue.
2380 10. A number of refactoring changes to the code, none of which should affect
2383 (a) The number of logging options was getting close to filling up the
2384 32-bit word that was used as a bit map. I have split them into two classes:
2385 those that are passed in the argument to log_write(), and those that are
2386 only ever tested independently outside of that function. These are now in
2387 separate 32-bit words, so there is plenty of room for expansion again.
2388 There is no change in the user interface or the logging behaviour.
2390 (b) When building, for example, log lines, the code previously used a
2391 macro that called string_cat() twice, in order to add two strings. This is
2392 not really sufficiently general. Furthermore, there was one instance where
2393 it was actually wrong because one of the argument was used twice, and in
2394 one call a function was used. (As it happened, calling the function twice
2395 did not affect the overall behaviour.) The macro has been replaced by a
2396 function that can join an arbitrary number of extra strings onto a growing
2399 (c) The code for expansion conditions now uses a table and a binary chop
2400 instead of a serial search (which was left over from when there were very
2401 few conditions). Also, it now recognizes conditions like "pam" even when
2402 the relevant support is not compiled in: a suitably worded error message is
2403 given if an attempt is made to use such a condition.
2405 11. Added ${time_interval:xxxxx}.
2407 12. A bug was causing one of the ddress fields not to be passed back correctly
2408 from remote delivery subprocesses. The field in question was not being
2409 subsequently used, so this caused to problems in practice.
2411 13. Added new log selectors queue_time and deliver_time.
2413 14. Might have fixed a bug in maildirsizefile handling that threw up
2414 "unexpected character" debug warnings, and recalculated the data
2415 unnecessarily. In any case, I expanded the warning message to give more
2418 15. Added the message "Restricted characters in address" to the statements in
2419 the default ACL that block characters like @ and % in local parts.
2421 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
2422 Three changes have been made:
2424 (a) There was a serious bug; a negative response to MAIL caused the whole
2425 recipient domain to be cached as invalid, thereby blocking all messages
2426 to all local parts at the same domain, from all senders. This bug has
2427 been fixed. The domain is no longer cached after a negative response to
2428 MAIL if the sender used is not empty.
2430 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
2433 (c) A new callout option, "use_sender" has been added for people who want
2434 the modified behaviour.
2440 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
2443 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
2444 indeed breaks things for older releases.
2446 3. Added additional logging to the case where there is a problem reading data
2447 from a filter that is running in a subprocess using a pipe, in order to
2448 try to track down a specific problem.
2450 4. Testing facility fudge: when running in the test harness and attempting
2451 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
2452 getting "No route to host". Convert this to a timeout.
2454 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
2457 6. Some OS don't have socklen_t but use size_t instead. This affects the
2458 fifth argument of getsockopt() amongst other things. This is now
2459 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
2460 can be set for individual OS. I have set it for SunOS5, OSF1, and
2461 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
2462 some earlier ones do not.
2464 7. Change 4.30/15 was not doing the test caselessly.
2466 8. The standard form for an IPv6 address literal was being rejected by address
2467 parsing in, for example, MAIL and RCPT commands. An example of this kind of
2468 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
2469 this, as well as the form without the "IPv6" on the front (but only when
2470 address literals are enabled, of course).
2472 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
2474 10. Exim crashed if a message with an empty sender address specified by -f
2475 encountered a router with an errors_to setting. This could be provoked only
2476 by a command such as
2480 where an empty string was supplied; "<>" did not hit this bug.
2482 11. Installed PCRE release 4.5.
2484 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
2485 remained set. It is now erased.
2487 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
2488 times from message ids (which are base 36 rather than the normal 62).
2490 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
2491 were being counted as actual protocol errors, and logged if the log
2492 selector +smtp_protocol_error was set. One cannot be perfect in this test,
2493 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
2494 and DATA following a set of rejected RCPTs do not count as protocol errors.
2495 In other words, Exim assumes they were pipelined, though this may not
2496 actually be the case. Of course, in all cases the client gets an
2497 appropriate error code.
2499 15. If a lookup fails in an ACL condition, a message about the failure may
2500 be available; it is used if testing the ACL cannot continue, because most
2501 such messages specify what the cause of the deferral is. However, some
2502 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
2503 that caused an old message to be retained and used if a later statement
2504 caused a defer, replacing the real cause of the deferral.
2506 16. If an IP address had so many PTR records that the DNS lookup buffer
2507 was not large enough to hold them, Exim could crash while trying to process
2508 the truncated data. It now detects and logs this case.
2510 17. Further to 4.21/58, another change has been made: if (and only if) the
2511 first line of a message (the first header line) ends with CRLF, a bare LF
2512 in a subsequent header line has a space inserted after it, so as not to
2513 terminate the header.
2515 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
2516 unnecessarily, used atoi() instead of strtol(), and didn't check the
2517 termination when getting file sizes from file names by regex.
2519 19. Completely re-implemented the support for maildirsize files, in the light
2520 of a number of problems with the previous contributed implementation
2521 (4.30/29). In particular:
2523 . If the quota is zero, the maildirsize file is maintained, but no quota is
2526 . If the maildir directory does not exist, it is created before any attempt
2527 to write a maildirsize file.
2529 . The quota value in the file is just a cache; if the quota is changed in
2530 the transport, the new value overrides.
2532 . A regular expression is available for excluding directories from the
2535 20. The autoreply transport checks the characters in options that define the
2536 message's headers; it allows continued headers, but it was checking with
2537 isspace() after an embedded newline instead of explicitly looking for a
2540 21. If all the "regular" hosts to which an address was routed had passed their
2541 expiry times, and had not reached their retry times, the address was
2542 bounced, even if fallback hosts were defined. Now Exim should go on to try
2545 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
2546 equivalent code in the SMTP transport. Some hosts send humungous responses
2547 to HELO/EHLO, more than 1024 it seems.
2549 23. Refactoring: code in filter.c used (void *) for "any old type" but this
2550 gives compiler warnings in some environments. I've now done it "properly",
2553 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
2554 (because of problems with the built-in one) was declared to return uschar *
2555 instead of char *, causing compiler failure.
2557 25. Fixed a file descriptor leak when processing alias/forward files.
2559 26. Fixed a minor format string issue in dbfn.c.
2561 27. Typo in exim.c: ("dmbnz" for "dbmnz").
2563 28. If a filter file refered to $h_xxx or $message_headers, and the headers
2564 contained RFC 2047 "words", Exim's memory could, under certain conditions,
2567 29. When a sender address is verified, it is cached, to save repeating the test
2568 when there is more than one recipient in a message. However, when the
2569 verification involves a callout, it is possible for different callout
2570 options to be set for different recipients. It is too complicated to keep
2571 track of this in the cache, so now Exim always runs a verification when a
2572 callout is required, relying on the callout cache for the optimization.
2573 The overhead is duplication of the address routing, but this should not be
2576 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
2577 to be verified with callout=postmaster, and the main callout worked but the
2578 postmaster check failed, the verification correctly failed. However, if a
2579 subsequent RCPT command asked for sender verification *without* the
2580 postmaster check, incorrect caching caused this verification also to fail,
2583 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
2584 it was not caching the DNS options (qualify_single, search_parents) that
2585 were used when the lookup failed. A subsequent lookup with different
2586 options therefore always gave the same answer, though there were cases
2587 where it should not have. (Example: a "domains = !$mx_any" option on a
2588 dnslookup router: the "domains" option is always processed without any
2589 widening, but the router might have qualify_single set.) Now Exim uses the
2590 cached value only when the same options are set.
2592 32. Added John Jetmore's "exipick" utility to the distribution.
2594 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
2595 than a timeout (e.g. a certificate is required, and is not provided), an
2596 Exim server now closes the connection immediately. Previously it waited for
2597 the client to close - but if the client is SSL, it seems that they each
2598 wait for each other, leading to a delay before one of them times out.
2600 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
2601 maintained 0.8.x compatibility because I don't think many are using it, and
2602 it is clearly obsolete.
2604 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
2607 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
2608 client certificate was expired. A simple patch fixes this, though I don't
2609 understand the full logic of why the verify callback is called multiple
2612 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
2613 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
2614 which causes problems with some clients (such as the Certicom SSL Plus
2615 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
2616 disables the coutermeasure allowing Eudora to connect."
2618 38. Exim was not checking that a write() to a log file succeeded. This could
2619 lead to Bad Things if a log got too big, in particular if it hit a file
2620 size limit. Exim now panics and dies if it cannot write to a log file, just
2621 as it does if it cannot open a log file.
2623 39. Modified OS/Makefile-Linux so that it now contains
2625 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
2627 The two -D definitions ensure that Exim is compiled with large file
2628 support, which makes it possible to handle log files that are bigger than
2631 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
2632 instance) a domain was checked against a named list that involved a lookup,
2633 causing $domain_data to be set, then another domain was checked against the
2634 same list, then the first domain was re-checked, the value of $domain_data
2635 after the final check could be wrong. In particular, if the second check
2636 failed, it could be set empty. This bug probably also applied to
2639 41. The strip_trailing_dot option was not being applied to the address given
2640 with the -f command-line option.
2642 42. The code for reading a message's header from the spool was incrementing
2643 $received_count, but never initializing it. This meant that the value was
2644 incorrect (doubled) while delivering a message in the same process in which
2645 it was received. In the most common configuration of Exim, this never
2646 happens - a fresh exec is done - but it can happen when
2647 deliver_drop_privilege is set.
2649 43. When Exim logs an SMTP synchronization error - client data sent too soon -
2650 it now includes up to 150 characters of the unexpected data in the log
2653 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
2654 and building data strings. The size of both of these buffers was 10 000
2655 bytes - far larger than anybody would *ever* want, thought I. Needless to
2656 say, somebody hit the limit. I have increased the maximum line length to
2657 20 000 and the maximum data length of concatenated lines to 100 000. I have
2658 also fixed two bugs, because there was no checking on these buffers. Tsk,
2659 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
2660 buffer is too small.
2662 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
2663 lsearch lookups. Now it does.
2665 46. When parsing a route_list item in a manualroute router, a fixed-length
2666 buffer was used for the list of hosts. I made this 1024 bytes long,
2667 thinking that nobody would ever have a list of hosts that long. Wrong.
2668 Somebody had a whole pile of complicated expansion conditions, and the
2669 string was silently truncated, leading to an expansion error. It turns out
2670 that it is easier to change to an unlimited length (owing to other changes
2671 that have happened since this code was originally written) than to build
2672 structure for giving a limitation error. The length of the item that
2673 expands into the list of hosts is now unlimited.
2675 47. The lsearch lookup could not handle data where the length of text line was
2676 more than 4095 characters. Such lines were truncated, leading to shortened
2677 data being returned. It should now handle lines of any length.
2679 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
2680 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
2683 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
2684 replacing strings such as BIN_DIRECTORY in a source file: the replacement
2685 no longer happens in comment lines. A list of replacements is now placed
2686 at the head of all of the source files, except those whose only change is
2687 to replace PERL_COMMAND in the very first #! line.
2689 50. Replaced the slow insertion sort in queue.c, for sorting the list of
2690 messages on the queue, with a bottom-up merge sort, using code contributed
2691 by Michael Haardt. This should make operations like -bp somewhat faster on
2692 large queues. It won't affect queue runners, except when queue_run_in_order
2695 51. Installed eximstats 1.31 in the distribution.
2697 52. Added support for SRV lookups to the dnslookup router.
2699 53. If an ACL referred to $message_body or $message_body_end, the value was not
2700 reset for any messages that followed in the same SMTP session.
2702 54. The store-handling optimization for building very long strings was not
2703 differentiating between the different store pools. I don't think this
2704 actually made any difference in practice, but I've tidied it.
2706 55. While running the routers to verify a sender address, $sender_address
2707 was still set to the sender address. This is wrong, because when routing to
2708 send a bounce to the sender, it would be empty. Therefore, I have changed
2709 it so that, while verifying a sender address, $sender_address is set to <>.
2710 (There is no change to what happens when verifying a recipient address.)
2712 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
2713 target A or AAAA records (if not already returned) without resetting the
2714 qualify_single or search_parents options of the DNS resolver. These are
2715 inappropriate in this case because the targets of MX and SRV records must
2716 be FQDNs. A broken DNS record could cause trouble if it happened to have a
2717 target that, when qualified, matched something in the local domain. These
2718 two options are now turned off when doing these lookups.
2720 57. It seems that at least some releases of Reiserfs (which does not have the
2721 concept of a fixed number of inodes) returns zero and not -1 for the
2722 number of available inodes. This interacted badly with check_spool_inodes,
2723 which assumed that -1 was the "no such thing" setting. What I have done is
2724 to check that the total number of inodes is greater than zero before doing
2725 the test of how many are available.
2727 58. When a "warn" ACL statement has a log_message modifier, the message is
2728 remembered, and not repeated. This is to avoid a lot of repetition when a
2729 message has many recipients that cause the same warning to be written.
2730 Howewer, Exim was preserving the list of already written lines for an
2731 entire SMTP session, which doesn't seem right. The memory is now reset if a
2732 new message is started.
2734 59. The "rewrite" debugging flag was not showing the result of rewriting in the
2735 debugging output unless log_rewrite was also set.
2737 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
2738 of (int)(handle) when we know that handle contains (void *)(-1).
2740 61. The Exim daemon panic-logs an error return when it closes the incoming
2741 connection. However "connection reset by peer" seems to be common, and
2742 isn't really an error worthy of noting specially, so that particular error
2745 62. When Exim is trying to find all the local interfaces, it used to panic and
2746 die if the ioctl to get the interface flags failed. However, it seems that
2747 on at least one OS (Solaris 9) it is possible to have an interface that is
2748 included in the list of interfaces, but for which you get a failure error
2749 for this call. This happens when the interface is not "plumbed" into a
2750 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
2751 failure of the "get flags" call assumes that the interface is down.
2753 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
2754 makes life easier for people who are doing arithmetic on fields extracted
2755 from dates, where you often get leading zeros that should not be
2756 interpreted as octal.
2758 64. Added qualify_domain to the redirect router, to override the global
2761 65. If a pathologically long header line contained very many addresses (the
2762 report of this problem mentioned 10 000) and each of them was rewritten,
2763 Exim could use up a very large amount of memory. (It kept on making new
2764 copies of the header line as it rewrote, and never released the old ones.)
2765 At the expense of a bit more processing, the header rewriting function has
2766 been changed so that it no longer eats memory in this way.
2768 66. The generation of the Received: header has been moved from the time that a
2769 message starts to be received, to the time that it finishes. The timestamp
2770 in the Received: header should now be very close to that of the <= log
2771 line. There are two side-effects of this change:
2773 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
2774 logged header lines no longer include the local Received: line, because
2775 it has not yet been created. The same applies to a copy of the message
2776 that is returned to a non-SMTP sender when a message is rejected.
2778 (b) When a filter file is tested using -bf, no additional Received: header
2779 is added to the test message. After some thought, I decided that this
2782 This change does not affect the value of $received_for. It is still set
2783 after address rewriting, but before local_scan() is called.
2785 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
2787 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
2788 gave an unhelpful panic error message, and a defer error. I have managed to
2789 change this behaviour so that it now rejects any supplied certificate,
2790 which seems right, as the list of acceptable certificates is empty.
2792 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
2793 gave an unhelpful defer error. I have not managed to make this reject any
2794 supplied certificates, but the error message it gives is "no certificate
2795 supplied", which is not helpful.
2797 70. exigrep's output now also includes lines that are not associated with any
2798 message, but which match the given pattern. Implemented by a patch from
2799 Martin Sluka, which also tidied up the Perl a bit.
2801 71. Recipient callout verification, like sender verification, was using <> in
2802 the MAIL FROM command. This isn't really the right thing, since the actual
2803 sender may affect whether the remote host accepts the recipient or not. I
2804 have changed it to use the actual sender in the callout; this means that
2805 the cache record is now keyed on a recipient/sender pair, not just the
2806 recipient address. There doesn't seem to be a real danger of callout loops,
2807 since a callout by the remote host to check the sender would use <>.
2808 [SEE ABOVE: changed after hitting problems.]
2810 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
2811 temporary errors. However, in the case of such a code being given after
2812 the end of a data transmission (i.e. after ".") Exim was failing to write
2813 a retry record for the message. (Yes, there was some broken host that was
2814 actually sending 8xx at this point.)
2816 73. An unknown lookup type in a host list could cause Exim to panic-die when
2817 the list was checked. (An example that provoked this was putting <; in the
2818 middle of a list instead of at the start.) If this happened during a DATA
2819 ACL check, a -D file could be left lying around. This kind of configuration
2820 error no longer causes Exim to die; instead it causes a defer errror. The
2821 incident is still logged to the main and panic logs.
2823 74. Buglet left over from Exim 3 conversion. The message "too many messages
2824 in one connection" was written to the rejectlog but not the mainlog, except
2825 when address rewriting (yes!) was being logged.
2827 75. Added write_rejectlog option.
2829 76. When a system filter was run not as root (that is, when system_filter_user
2830 was set), the values of the $n variables were not being returned to the
2831 main process; thus, they were not subsequently available in the $sn
2834 77. Added +return_path_on_delivery log selector.
2836 78. A connection timeout was being treated differently from recipients deferred
2837 when testing hosts_max_try with a message that was older than the host's
2838 retry timeout. (The host should not be counted, thus allowing all hosts to
2839 be tried at least once before bouncing.) This may have been the cause of an
2840 occasionally reported bug whereby a message would remain on the queue
2841 longer than the retry timeout, but would be bounced if a delivery was
2842 forced. I say "may" because I never totally pinned down the problem;
2843 setting up timeout/retry tests is difficult. See also the next item.
2845 79. The ultimate address timeout was not being applied to errors that involved
2846 a combination of host plus message (for example, a timeout on a MAIL
2847 command). When an address resolved to a number of possible hosts, and they
2848 were not all tried for each delivery (e.g. because of hosts_max_try), a
2849 message could remain on the queue longer than the retry timeout.
2851 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
2854 81. Fixed an obscure SMTP outgoing bug which required at least the following
2855 conditions: (a) there was another message waiting for the same server;
2856 (b) the server returned 5xx to all RCPT commands in the first message so
2857 that the message was not completed; (c) the server dropped the connection
2858 or gave a negative response to the RSET that Exim sends to abort the
2859 transaction. The observed case was a dropped connection after DATA that had
2860 been sent in pipelining mode. That is, the server had advertised PIPELINING
2861 but was not implementing it correctly. The effect of the bug was incorrect
2862 behaviour, such as trying another host, and this could lead to a crash.
2868 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
2869 and daemon.c were passed as pointers to ints; they should have been
2870 pointers to socklen_t variables (which are typically unsigned ints).
2872 2. Some signed/unsigned type warnings in the os.c file for Linux have been
2875 3. Fixed a really odd bug that affected only the testing scheme; patching a
2876 certain fixed string in the binary changed the value of another string that
2877 happened to be identical to the end of the original first string.
2879 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
2880 name", it returns that address as the IP address. On some operating
2881 systems (e.g. Solaris), it also passes back the IP address string as the
2882 "host name". However, on others (e.g. Linux), it passes back an empty
2883 string. Exim wasn't checking for this, and was changing the host name to an
2884 empty string, assuming it had been canonicized.
2886 5. Although rare, it is permitted to have more than one PTR record for a given
2887 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
2888 all the names associated with an address, because they do in Solaris.
2889 However, it seems that they do not in Linux for data that comes from the
2890 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
2891 I found this out when I moved to a new Linux workstation and tried to run
2892 the Exim test suite.
2894 To get round this problem I have changed the code so that it now does its
2895 own call to the DNS to look up PTR records when searching for a host name.
2896 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
2897 addresses that are only in /etc/hosts are still found.
2899 This behaviour is, however, controlled by an option called host_lookup_
2900 order, which defaults to "bydns:byaddr". If people want to use the other
2901 order, or indeed, just use one or the other means of lookup, they can
2902 specify it in this variable.
2904 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
2905 some operating systems, this comes back from gethostbyaddr() as an empty
2906 string, and this is what Exim used to test for. However, it seems that in
2907 other systems, "." is yielded. Exim now tests for this case too.
2909 7. The values of check_spool_space and check_log_space are now held internally
2910 as a number of kilobytes instead of an absolute number of bytes. If a
2911 numbers is specified without 'K' or 'M', it is rounded up to the nearest
2912 kilobyte. This means that much larger values can be stored.
2914 8. Exim monitor: an attempt to get the action menu when not actually pointing
2915 at a message produces an empty menu entitled "No message selected". This
2916 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
2917 no entries in it ("Shell widget menu has zero width and/or height"). So I
2918 have added a single, blank menu entry in this case.
2920 9. Added ${quote_local_part.
2922 10. MIME decoding is now applied to the contents of Subject: header lines when
2925 11. Now that a reference to $sender_host_address automatically causes a reverse
2926 lookup to occur if necessary (4.13/18), there is no need to arrange for a
2927 host lookup before query-style lookups in lists that might use this
2928 variable. This has therefore been abolished, and the "net-" prefix is no
2929 longer necessary for query-style lookups.
2931 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
2932 have been a typo for LFLAGS, so it has been changed.
2934 13. The install script calls Exim with "-C /dev/null" in order to find the
2935 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
2936 to be output. Howeve, since Exim outputs its version number before the
2937 error, it didn't break the script. It just looked ugly. I fixed this by
2938 always allowing "-C /dev/null" if the caller is root.
2940 14. Ignore overlarge ACL variable number when reading spool file - insurance
2941 against a later release with more variables having written the file.
2943 15. The standard form for an IPv6 address literal was being rejected by EHLO.
2944 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
2945 this, as well as the form without the "IPv6" on the front.
2947 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
2948 OS/Makefile-Darwin file.
2950 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
2951 only to LDAP libraries that do not have LDAP_OPT_DEREF.
2953 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
2954 variable. However, some OS use ints for inodes. I've added cast to long int
2955 to get rid of the compiler warning.
2957 19. I had forgotten to lock out "/../" in configuration file names when
2958 ALT_CONFIG_PREFIX was set.
2960 20. Routers used for verification do not need to specify transports. However,
2961 if such a router generated a host list, and callout was configured, Exim
2962 crashed, because it could not find a port number from the (non-existent)
2963 transport. It now assumes port 25 in this circumstance.
2965 21. Added the -t option to exigrep.
2967 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
2968 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
2969 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
2970 them is set, LOOKUP_LSEARCH is forced.
2972 23. "exim -bV" now outputs a list of lookups that are included in the binary.
2974 24. Added sender and host information to the "rejected by local_scan()" log
2975 line; previously there was no indication of these.
2977 25. Added .include_if_exists.
2979 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
2980 Linux. It turns out that not all file systems support this. Apparently some
2981 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
2982 do it.) To cope with this, the error EINVAL, which means that sync-ing is
2983 not supported on the file descriptor, is now ignored when Exim is trying to
2984 sync a directory. This applies only to Linux.
2986 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
2988 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
2989 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
2992 29. Added support for maildirsize files from supplied patch (modified a bit).
2994 30. The use of :fail: followed by an empty string could lead Exim to respond to
2995 sender verification failures with (e.g.):
2997 550 Verification failed for <xxx>
2998 550 Sender verify failed
3000 where the first response line was missing the '-' that indicates it is not
3001 the final line of the response.
3003 31. The loop for finding the name of the user that called Exim had a hardwired
3004 limit of 10; it now uses the value of finduser_retries, which is used for
3005 all other user lookups.
3007 32. Added $received_count variable, available in data and not_smtp ACLs, and at
3010 33. Exim was neglecting to zero errno before one call of strtol() when
3011 expanding a string and expecting an integer value. On some systems this
3012 resulted in spurious "integer overflow" errors. Also, it was casting the
3013 result into an int without checking.
3015 34. Testing for a connection timeout using "timeout_connect" in the retry rules
3016 did not work. The code looks as if it has *never* worked, though it appears
3017 to have been documented since at least releast 1.62. I have made it work.
3019 35. The "timeout_DNS" error in retry rules, also documented since at least
3020 1.62, also never worked. As it isn't clear exactly what this means, and
3021 clearly it isn't a major issue, I have abolished the feature by treating it
3022 as "timeout", and writing a warning to the main and panic logs.
3024 36. The display of retry rules for -brt wasn't always showing the error code
3027 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
3028 timeout_connect_A, timeout_connect_MX.
3030 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
3031 to the empty sender.
3033 39. The daemon was not analysing the content of -oX till after it had closed
3034 stderr and disconnected from the controlling terminal. This meant that any
3035 syntax errors were only noted on the panic log, and the return code from
3036 the command was 0. By re-arranging the code a little, I've made the
3037 decoding happen first, so such errors now appear on stderr, and the return
3038 code is 1. However, the actual setting up of the sockets still happens in
3039 the disconnected process, so errors there are still only recorded on the
3042 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
3043 connections (as happens on some IP stacks) was logged at start up time as
3044 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
3045 it from "IPv6 and IPv4", which means that two separate sockets are being
3048 41. The debug output for gethostbyname2() or getipnodebyname() failures now
3049 says whether AF_INET or AF_INET6 was passed as an argument.
3051 42. Exiwhat output was messed up when time zones were included in log
3054 43. Exiwhat now gives more information about the daemon's listening ports,
3055 and whether -tls-on-connect was used.
3057 44. The "port" option of the smtp transport is now expanded.
3059 45. A "message" modifier in a "warn" statement in a non-message ACL was being
3060 silently ignored. Now an error message is written to the main and panic
3063 46. There's a new ACL modifier called "logwrite" which writes to a log file
3064 as soon as it is encountered.
3066 47. Added $local_user_uid and $local_user_gid at routing time.
3068 48. Exim crashed when trying to verify a sender address that was being
3071 49. Exim was recognizing only a space character after ".include". It now also
3072 recognizes a tab character.
3074 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
3075 extracting the relevant information from the specification. The man page no
3076 longer contains scrambled data for the -d option, and I've added a section
3077 at the front about calling Exim under different names.
3079 51. Added "extra_headers" argument to the "mail" command in filter files.
3081 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
3084 53. Installed eximstats 1.29.
3086 54. Added transport_filter_timeout as a generic transport option.
3088 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
3089 Cc: header lines. This was required by RFC 822, but it not required by RFC
3092 56. Exim used to add From:, Date:, and Message-Id: header lines to any
3093 incoming messages that did not have them. Now it does so only if the
3094 message originates locally, that is, if there is no associated remote host
3095 address. When Resent- header lines are present, this applies to the Resent-
3096 lines rather than the non-Resent- lines.
3098 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
3099 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
3101 58. Messages for configuration errors now include the name of the main
3102 configuration file - useful now that there may be more than one file in a
3103 list (.included file names were always shown).
3105 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
3106 for those rare installations that do not start the daemon as root or run it
3107 setuid root. I've cut out the call to initgroups() if the daemon is not
3110 60. The Exim user and group can now be bound into the binary as text strings
3111 that are looked up at the start of Exim's processing.
3113 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
3115 62. Added $mailstore_basename variable.
3117 63. Installed patch to sieve.c from Michael Haardt.
3119 64. When Exim failed to open the panic log after failing to open the main log,
3120 the original message it was trying to log was written to stderr and debug
3121 output, but if they were not available (the usual case in production), it
3122 was lost. Now it is written to syslog before the two lines that record the
3123 failures to open the logs.
3125 65. Users' Exim filters run in subprocesses under the user's uid. It is
3126 possible for a "deliver" command or an alias in a "personal" command to
3127 provoke an address rewrite. If logging of address rewriting is configured,
3128 this fails because the process is not running as root or exim. There may be
3129 a better way of dealing with this, but for the moment (because 4.30 needs
3130 to be released), I have disabled address rewrite logging when running a
3131 filter in a non-root, non-exim process.
3137 1. The buildconfig auxiliary program wasn't quoting the value set for
3138 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
3139 not defined. This bug was masked in 4.22 by the effect that was fixed in
3142 2. Some messages that were rejected after a message id was allocated were
3143 shown as "incomplete" by exigrep. It no longer does this for messages that
3144 are rejected by local_scan() or the DATA or non-SMTP ACLs.
3146 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
3147 have allow_domain_literals set, the ID did not get logged in the <= line.
3148 Domain literals are now always recognized in Message-ID: header lines.
3150 4. The first argument for a ${extract expansion item is the key name or field
3151 number. Leading and trailing spaces in this item were not being ignored,
3152 causing some misleading effects.
3154 5. When deliver_drop_privilege was set, single queue runner processes started
3155 manually (i.e. by the command "exim -q") or by the daemon (which uses the
3156 same command in the process it spins off) were not dropping privilege.
3158 6. When the daemon running as "exim" started a queue runner, it always
3159 re-executed Exim in the spun-off process. This is a waste of effort when
3160 deliver_drop_privilege is set. The new process now just calls the
3161 queue-runner function directly.
3167 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
3170 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
3171 ignored. Though the use of -oP was forcing the writing of a pid file, it
3172 was always written to the default place.
3174 3. If the message "no IP address found for host xxxx" is generated during
3175 incoming verification, it is now followed by identification of the incoming
3176 connection (so you can more easily find what provoked it).
3178 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
3180 5. Added some features to "harden" Exim a bit more against certain attacks:
3182 (a) There is now a build-time option called FIXED_NEVER_USERS that can
3183 be put in Local/Makefile. This is like the never_users runtime option,
3184 but it cannot be overridden. The default setting is "root".
3186 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
3187 prefix string with which any file named in a -C command line option
3190 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
3191 is retained for -C and -D only if the caller of Exim is root. Without
3192 it, the exim user may also use -C and -D and retain privilege.
3194 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
3195 command line option is disabled.
3197 6. Macro names set by the -D option must start with an upper case letter, just
3198 like macro names defined in the configuration file.
3200 7. Added "dereference=" facility to LDAP.
3202 8. Two instances of the typo "uknown" in the source files are fixed.
3204 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
3205 the Configure-Makefile script screwed up while processing it.
3207 10. Incorporated PCRE 4.4.
3209 11. The SMTP synchronization check was not operating right at the start of an
3210 SMTP session. For example, it could not catch a HELO sent before the client
3211 waited for the greeting. There is now a check for outstanding input at the
3212 point when the greeting is written. Because of the duplex, asynchronous
3213 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
3214 way, but not yet received, when the check is performed.
3216 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
3217 on TCP/IP sockets, because this apparently causes some broken clients to
3220 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
3221 unchanged) from the Cygwin maintainer.
3223 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
3224 was supported instead of testing for mbx. Effectively a typo.
3226 15. The spa authenticator server code was not checking that the input it
3227 received was valid base64.
3229 16. The debug output line for the "set" modifier in ACLs was not showing the
3230 name of the variable that was being set.
3232 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
3234 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
3235 lookup on its own. Something else was needed to trigger the lookup. For
3236 example, a match in host_lookup or the need for a host name in a host list.
3237 Now, if $sender_host_name is referenced and the host name has not yet been
3238 looked up, a lookup is performed. If the lookup fails, the variable remains
3239 empty, and $host_lookup_failed is set to "1".
3241 19. Added "eqi" as a case-independent comparison operator.
3243 20. The saslauthd authentication condition could segfault if neither service
3244 nor realm was specified.
3246 21. If an overflowing value such as "2048M" was set for message_size_limit, the
3247 error message that was logged was misleading, and incoming SMTP
3248 connections were dropped. The message is now more accurate, and temporary
3249 errors are given to SMTP connections.
3251 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
3252 (except RSET) with a 421 error, until QUIT is received. However, it was
3253 failing to send a response to QUIT.
3255 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
3256 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
3257 ACL is now run after the helo_try_verify_hosts code.
3259 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
3260 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
3261 case-independent, so other case variants are also recognized). Apparently
3262 some systems use these upper case variants.
3264 25. If more than two messages were waiting for the same host, and a transport
3265 filter was specified for the transport, Exim sent two messages over the
3266 same TCP/IP connection, and then failed with "socket operation on non-
3267 socket" when it tried to send the third.
3269 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
3271 27. The extern definition of crypt16() in expand.c was not being excluded when
3272 the OS had its own crypt16() function.
3274 28. Added bounce_return_body as a new option, and bounce_return_size_limit
3275 as a preferred synonym for return_size_limit, both as an option and as an
3278 29. Added LIBS=-liconv to OS/Makefile-OSF1.
3280 30. Changed the default configuration ACL to relax the local part checking rule
3281 for addresses that are not in any local domains. For these addresses,
3282 slashes and pipe symbols are allowed within local parts, but the sequence
3283 /../ is explicitly forbidden.
3285 31. SPA server authentication was not clearing the challenge buffer before
3288 32. log_message in a "warn" ACL statement was writing to the reject log as
3289 well as to the main log, which contradicts the documentation and doesn't
3290 seem right (because no rejection is happening). So I have stopped it.
3292 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
3293 However, I am unable to do any testing of this.
3295 34. Fixed an infelicity in the appendfile transport. When checking directories
3296 for a mailbox, to see if any needed to be created, it was accidentally
3297 using path names with one or more superfluous leading slashes; tracing
3298 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
3300 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
3301 contents are added to the log line that is written for every discarded
3302 recipient. (Previously a log_message setting was ignored.)
3304 36. The ${quote: operator now quotes the string if it is empty.
3306 37. The install script runs exim in order to find its version number. If for
3307 some reason other than non-existence or emptiness, which it checks, it
3308 could not run './exim', it was installing it with an empty version number,
3309 i.e. as "exim-". This error state is now caught, and the installation is
3312 38. An argument was missing from the function that creates an error message
3313 when Exim fails to connect to the socket for saslauthd authentication.
3314 This could cause Exim to crash, or give a corrupted message.
3316 39. Added isip, isip4, and isip6 to ${if conditions.
3318 40. The ACL variables $acl_xx are now saved with the message, and can be
3319 accessed later in routers, transports, and filters.
3321 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
3322 strings in the file are not string-expanded.
3324 42. If a MAIL command specified a SIZE value that was too large to fit into an
3325 int variable, the check against message_size_limit failed. Such values are
3326 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
3327 day this will have to be increased, but I don't think I want to be around
3328 when emails are that large.
3335 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
3336 iconv() is not standard in FreeBSD.
3338 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
3339 IPv6 enabled. The observed symptom was a segmentation fault on return from
3340 the function os_common_find_running_interfaces() in src/os.c.
3342 3. In the check_special_case() function in daemon.c I had used "errno" as an
3343 argument name, which causes warnings on some systems. This was basically a
3344 typo, since it was named "eno" in the comments!
3346 4. The code that waits for the clock to tick (at a resolution of some fraction
3347 of a second) so as to ensure message-id uniqueness was always waiting for
3348 at least one whole tick, when it could have waited for less. [This is
3349 almost certainly not relevant at current processor speeds, where it is
3350 unlikely to ever wait at all. But we try to future-proof.]
3352 5. The function that sleeps for a time interval that includes fractions of a
3353 second contained a race. It did not block SIGALRM between setting the
3354 timer, and suspending (a couple of lines later). If the interval was short
3355 and the sigsuspend() was delayed until after it had expired, the suspension
3356 never ended. On busy systems this could lead to processes getting stuck for
3359 6. Some uncommon configurations may cause a lookup to happen in a queue runner
3360 process, before it forks any delivery processes. The open lookup caching
3361 mechanism meant that the open file or database connection was passed into
3362 the delivery process. The problem was that delivery processes always tidy
3363 up cached lookup data. This could cause a problem for the next delivery
3364 process started by the queue runner, because the external queue runner
3365 process does not know about the closure. So the next delivery process
3366 still has data in the lookup cache. In the case of a file lookup, there was
3367 no problem because closing a file descriptor in a subprocess doesn't affect
3368 the parent. However, if the lookup was caching a connection to a database,
3369 the connection was closed, and the second delivery process was likely to
3370 see errors such as "PGSQL: query failed: server closed the connection
3371 unexpectedly". The problem has been fixed by closing all cached lookups
3372 in a queue runner before running a delivery process.
3374 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
3375 seem to have the "const" qualifier which it has on other OS. I've
3378 8. Change 4.21/2 was too strict. It is only if there are two authenticators
3379 *of the same type* (client or server) with the same public name that an
3380 error should be diagnosed.
3382 9. When Exim looked up a host name for an IP address, but failed to find the
3383 original IP address when looking up the host name (a safety check), it
3384 output the message "<ip address> does not match any IP for NULL", which was
3385 confusing, to say the least. The bug was that the host name should have
3386 appeared instead of "NULL".
3388 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
3389 user that is built into the binary, and the -C or -D options is used, root
3390 privilege is dropped before the configuration file is read. In addition,
3391 logging is switched to stderr instead of the normal log files. If the
3392 configuration then re-defines the Exim user, the unprivileged environment
3393 is probably not what is expected, so Exim logs a panic warning message (but
3396 However, if deliver_drop_privilege is set, the unprivileged state may well
3397 be exactly what is intended, so the warning has been cut out in that case,
3398 and Exim is allowed to try to write to its normal log files.
3404 1. smtp_return_error_details was not giving details for temporary sender
3405 or receiver verification errors.
3407 2. Diagnose a configuration error if two authenticators have the same public
3410 3. Exim used not to create the message log file for a message until the first
3411 delivery attempt. This could be confusing when incoming messages were held
3412 for policy or load reasons. The message log file is now created at the time
3413 the message is received, and an initial "Received" line is written to it.
3415 4. The automatically generated man page for command line options had a minor
3416 bug that caused no ill effects; however, a more serious problem was that
3417 the procedure for building the man page automatically didn't always
3418 operate. Consequently, release 4.20 contains an out-of-date version. This
3419 shouldn't happen again.
3421 5. When building Exim with embedded Perl support, the script that builds the
3422 Makefile was calling 'perl' to find its compile-time parameters, ignoring
3423 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
3425 6. The freeze_tell option was not being used for messages that were frozen on
3426 arrival, either by an ACL or by local_scan().
3428 7. Added the smtp_incomplete_transaction log selector.
3430 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
3431 was accepting AUTH without a new EHLO.
3433 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
3434 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
3437 10. Logging of TCP/IP connections (when configured) now happens in the main
3438 daemon process instead of the child process, so that the TCP/IP connection
3439 count is more accurate (but it can never be perfect).
3441 11. The use of "drop" in a nested ACL was not being handled correctly in the
3442 outer ACL. Now, if condition failure induced by the nested "drop" causes
3443 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
3444 or "require"), the connection is dropped.
3446 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
3447 that yield "discard" can now be used with an "accept" or a "discard" verb,
3448 but an error is generated for any others (because I can't see a useful way
3449 to define what should happen).
3451 13. When an ACL is read dynamically from a file (or anywhere else), the lines
3452 are now processed in the same way as lines in the Exim configuration file.
3453 In particular, continuation lines are supported.
3455 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
3457 15. Added -ti meaning -t -i.
3459 16. Check for letters, digits, hyphens, and dots in the names of dnslist
3460 domains, and warn by logging if others are found.
3462 17. At least on BSD, alignment is not guarenteed for the array of ifreq's
3463 returned from GIFCONF when Exim is trying to find the list of interfaces on
3464 a host. The code in os.c has been modified to copy each ifreq to an aligned
3465 structure in all cases.
3467 Also, in some cases, the returned ifreq's were being copied to a 'struct
3468 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
3469 means the last couple of bytes of an IPv6 address could be chopped if the
3470 ifreq contained only a normal sockaddr (14 bytes storage).
3472 18. Named domain lists were not supported in the hosts_treat_as_local option.
3473 An entry such as +xxxx was not recognized, and was treated as a literal
3476 19. Ensure that header lines added by a DATA ACL are included in the reject log
3477 if the ACL subsequently rejects the message.
3479 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
3480 MD5 (which is deprecated).
3482 21. When testing a filter file using -bf, Exim was writing a message when it
3483 took the sender from a "From " line in the message, but it was not doing so
3484 when it took $return_path from a Return-Path: header line. It now does.
3486 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
3487 with a valid header line field name (a series of printing characters
3488 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
3490 23. Changed "disc" in the source to "disk" to conform to the documentation and
3491 the book and for uniformity.
3493 24. Ignore Sendmail's -Ooption=value command line item.
3495 25. When execve() failed while trying to run a command in a pipe transport,
3496 Exim was returning EX_UNAVAILBLE (69) from the subprocess. However, this
3497 could be confused with a return value of 69 from the command itself. This
3498 has been changed to 127, the value the shell returns if it is asked to run
3499 a non-existent command. The wording for the related log line suggests a
3500 non-existent command as the problem.
3502 26. If received_header_text expands to an empty string, do not add a Received:
3503 header line to the message. (Well, it adds a token one on the spool, but
3504 marks it "old" so that it doesn't get used or transmitted.)
3506 27. Installed eximstats 1.28 (addition of -nt option).
3508 28. There was no check for failure on the call to getsockname() in the daemon
3509 code. This can fail if there is a shortage of resources on the system, with
3510 ENOMEM, for example. A temporary error is now given on failure.
3512 29. Contrary to the C standard, it seems that in some environments, the
3513 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
3514 program. Exim now does this explicitly; it affects the formatting of
3515 timestamps using strftime().
3517 30. If exiqsumm was given junk data, it threw up some uninitialized variable
3518 complaints. I've now initialized all the variables, to avoid this.
3520 32. Header lines added by a system filter were not being "seen" during
3521 transport-time rewrites.
3523 33. The info_callback() function passed to OpenSSL is set up with type void
3524 (*)(SSL *, int, int), as described somewhere. However, when calling the
3525 function (actually a macro) that sets it up, the type void(*)() is
3526 expected. I've put in a cast to prevent warnings from picky compilers.
3528 34. If a DNS black list lookup found a CNAME record, but there were no A
3529 records associated with the domain it pointed at, Exim crashed.
3531 35. If a DNS black list lookup returned more than one A record, Exim ignored
3532 all but the first. It now scans all returned addresses if a particular IP
3533 value is being sought. In this situation, the contents of the
3534 $dnslist_value variable are a list of all the addresses, separated by a
3537 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
3538 to accept a host name and all its aliases if the forward lookup for any of
3539 them yielded the IP address of the incoming connection. Now it accepts only
3540 those names whose forward lookup yields the correct IP address. Any other
3541 names are discarded. This closes a loophole whereby a rogue DNS
3542 administrator could create reverse DNS records to break through a
3543 wildcarded host restriction in an ACL.
3545 37. If a user filter or a system filter that ran in a subprocess used any of
3546 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
3547 the wrong values were passed to the pipe command ($thisaddress had the
3548 value of $0, $0 had the value of $1, etc). This bug was introduced by
3549 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
3551 38. Improved the line breaking for long SMTP error messages from ACLs.
3552 Previously, if there was no break point between 40 and 75 characters, Exim
3553 left the rest of the message alone. Two changes have been made: (a) I've
3554 reduced the minimum length to 35 characters; (b) if it can't find a break
3555 point between 35 and 75 characters, it looks ahead and uses the first one
3556 that it finds. This may give the occasional overlong line, but at least the
3557 remaining text gets split now.
3559 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
3560 file descriptors might be low, and that setting 1000 would always raise it.
3561 It turns out that in some environments, the limit is already over 1000 and
3562 that lowering it causes trouble. So now Exim takes care not to decrease it.
3564 40. When delivering a message, the value of $return_path is set to $sender_
3565 address at the start of routing (routers may change the value). By an
3566 oversight, this default was not being set up when an address was tested by
3567 -bt or -bv, which affected the outcome if any router or filter referred to
3570 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
3571 log only when "message" or "log_message" are set. However, if one of the
3572 conditions was an address verification, or a call to a nested ACL, the
3573 messages generated by the underlying test were being passed through. This
3574 no longer happens. The underlying message is available in $acl_verify_
3575 message for both "message" and "log_message" expansions, so it can be
3576 passed through if needed.
3578 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
3579 new expansion $bh_ to give the encoded byte string without charset
3580 translation. Translation happens only if iconv() is available; HAVE_ICONV
3581 indicates this at build time. HEADERS_CHARSET gives the charset to
3582 translate to; headers_charset can change it in the configuration, and
3583 "headers charset" can change it in an individual filter file.
3585 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
3586 that creates RFC 2047 encoded "words" labels them as that charset instead
3587 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
3588 expansion operator; (ii) when Exim creates a From: line for a local
3589 message; (iii) when a header line is rewritten to include a "phrase" part.
3591 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
3592 buggy, causing it to skip the first lines of messages whose message ID
3593 ended in 'D'. This would not have bitten before Exim release 4.14, because
3594 message IDs were unlikely to end in 'D' before then. The effect was to have
3595 incorrect size information for certain domains.
3597 45. #include "config.h" was missing at the start of the crypt16.c module. This
3598 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
3601 46. If there was a timeout during a "random" callout check, Exim treated it as
3602 a failure of the random address, and carried on sending RSET and the real
3603 address. If the delay was just some slowness somewhere, the response to the
3604 original RCPT would be taken as a response to RSET and so on, causing
3605 mayhem of various kinds.
3607 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
3608 when I implemented it. It didn't allow for the fact that some option values
3609 may legitimatetly be negative (e.g. size_addition), and it didn't even do
3610 the right test for positive values.
3612 48. Domain names in DNS records are case-independent. Exim always looks them up
3613 in lower case. Some resolvers return domain names in exactly the case they
3614 appear in the zone file, that is, they may contain uppercase letters. Not
3615 all resolvers do this - some return always lower case. Exim was treating a
3616 change of case by a resolver as a change of domain, similar to a widening
3617 of a domain abbreviation. This triggered its re-routing code and so it was
3618 trying to route what was effectively the same domain again. This normally
3619 caused routing to fail (because the router wouldn't handle the domain
3620 twice). Now Exim checks for this case specially, and just changes the
3621 casing of the domain that it ultimately uses when it transmits the message
3624 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
3627 50. If a filter generated a file delivery with a non-absolute name (possible if
3628 no home directory exists for the router), the forbid_file option was not
3631 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
3632 the existing equality matching.
3634 52. Exim was using ints instead of ino_t variables in some places where it was
3635 dealing with inode numbers.
3637 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
3638 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
3639 called TMPDIR, and if it finds it is different, it changes its value.
3641 54. The smtp_printf() function is now made available to local_scan() so
3642 additional output lines can be written before returning. There is also an
3643 smtp_fflush() function to enable the detection of a dropped connection.
3644 The variables smtp_input and smtp_batched_input are exported to
3647 55. Changed the default runtime configuration: the message "Unknown user"
3648 has been removed from the ACL, and instead placed on the localuser router,
3649 using the cannot_route_message feature. This means that any verification
3650 failures that generate their own messages won't get overridden. Similarly,
3651 the "Unrouteable address" message that was in the ACL for unverifiable
3652 relay addresses has also been removed.
3654 56. Added hosts_avoid_esmtp to the smtp transport.
3656 57. The exicyclog script was not checking for the esoteric option
3657 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
3658 will work only if exicyclog is run under the appropriate euid.
3660 58. Following a discussion on the list, the rules by which Exim recognises line
3661 endings on incoming messages have been changed. The -dropcr and drop_cr
3662 options are now no-ops, retained only for backwards compatibility. The
3663 following line terminators are recognized: LF CRLF CR. However, special
3664 processing applies to CR:
3666 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
3667 nor a local message in the state where . is a terminator.
3669 (ii) If a bare CR is encountered in a header line, an extra space is added
3670 after the line terminator so as not to end the header. The reasoning
3671 behind this is that bare CRs in header lines are most likely either
3672 to be mistakes, or people trying to play silly games.
3674 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
3675 was being incorrectly given as 18 bytes larger than it should have been.
3676 This is a VOB (very old bug).
3678 60. This may never have affected anything current, but just in case it has:
3679 When the local host is found other than at the start of a list of hosts,
3680 the local host, those with the same MX, and any that follow, are discarded.
3681 When the list in question was part of a longer list of hosts, the following
3682 hosts (not currently being processed) were also being discarded. This no
3683 longer happens. I'm not sure if this situation could ever has previously
3686 61. Added the "/MX" feature to lists of hosts in the manualroute and query
3689 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
3690 header. This is something that is recommended in a new Internet Draft, and
3691 is something that is documented as being done by Sendmail. There are two
3692 possible values. For messages generated by the autoreply transport, Exim
3695 Auto-Submitted: auto-replied
3697 whereas for all other generated messages (e.g. bounces) it adds
3699 Auto-Submitted: auto-generated
3701 63. The "personal" condition in filters now includes a test for the
3702 Auto-Submitted: header. If it contains the string "auto-" the message it
3703 not considered personal.
3705 64. Added rcpt_include_affixes as a generic transport option.
3707 65. Added queue_only_override (default true).
3709 66. Added the syslog_duplication option.
3711 67. If what should have been the first header line of a message consisted of
3712 a space followed by a colon, Exim was mis-interpreting it as a header line.
3713 It isn't of course - it is syntactically invalid and should therefore be
3714 treated as the start of the message body. The misbehaviour could have
3715 caused a number of strange effects, including loss of data in subsequent
3716 header lines, and spool format errors.
3718 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
3719 client host had authenticated. This control can now be exercised by an ACL
3720 for more flexibility.
3722 69. By default, callouts do not happen when testing with -bh. There is now a
3723 variant, -bhc, which does actually run the callout code, including
3724 consulting and updating the callout cache.
3726 70. Added support for saslauthd authentication, courtesy of Alexander
3729 71. If statvfs() failed on the spool or log directories while checking their
3730 size for availability, Exim confusingly gave the error "space shortage".
3731 Furthermore, in debugging mode it crashed with a floating point exception.
3732 These checks are done if check_{spool,log}_{space,inodes} are set, and when
3733 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
3734 serious problem, Exim now writes to the main and panic logs when this
3735 happens, with details of the failure. It then refuses to accept the
3736 incoming message, giving the message "spool directory problem" or "log
3737 directory problem" with a 421 code for SMTP messages.
3739 72. When Exim is about to re-exec itself, it ensures that the file descriptors
3740 0, 1, and 2 exist, because some OS complain for execs without them (see
3741 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
3742 descriptors. However, the code omitted to check that the open succeeded,
3743 causing mysterious errors if for some reason the permissions on /dev/null
3744 got screwed. Now Exim writes a message to the main and panic logs, and
3745 bombs out if it can't open /dev/null.
3747 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
3748 interact so that it is all more flexible. It is supposed to remain
3749 backwards compatible. Also added extra_local_interfaces.
3751 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
3752 to bomb out with an assertion failure - to the client this appears as a
3753 connection drop. This problem occurs in the part of the code that was taken
3754 from the Samba project. Fortunately, the assertion is in a very simple
3755 function, so I have fixed this by reproducing the function inline in the
3756 one place where it is called, and arranging for authentication to fail
3757 instead of killing the process with assert().
3759 75. The SPA client code was not working when the server requested OEM rather
3760 than Unicode encoding.
3762 76. Added code to make require_files with a specific uid setting more usable in
3763 the case where statting the file as root fails - usually a non-root-mounted
3764 NFS file system. When this happens and the failure is EACCES, Exim now
3765 forks a subprocess and does the per-uid checking as the relevant uid.
3767 77. Added process_log_path.
3769 78. If log_file_path was not explicitly set, a setting of check_log_space or
3770 check_log_inodes was ignored.
3772 79. If a space check for the spool or log partitions fails, the incident is now
3773 logged. Of course, in the latter case the data may get lost...
3775 80. Added the %p formatting code to string_format() so that it can be used to
3776 print addresses in debug_print(). Adjusted all the address printing in the
3777 debugging in store.c to use %p rather than %d.
3779 81. There was a concern that a line of code in smtp_in.c could overflow a
3780 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
3781 initially expressed, the concern was not well-founded, because trailing
3782 spaces are removed early. However, if the trailing spaces were followed by
3783 a NULL, they did not get removed, so the overflow was possible. Two fixes
3786 (a) I re-wrote the offending code in a cleaner fashion.
3787 (b) If an incoming SMTP command contains a NULL character, it is rejected
3790 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
3791 runs initgroups(), so that if the Exim user is in any additional groups,
3792 they will be used during message reception.
3798 The change log for 4.20 and earlier releases has been archived.