1 $Cambridge: exim/doc/doc-txt/ChangeLog,v 1.279 2005/12/15 18:09:55 jetmore 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
59 PH/01 Two changes to the default runtime configuration:
61 (1) Move the checks for relay_from_hosts and authenticated clients from
62 after to before the (commented out) DNS black list checks.
64 (2) Add control=submission to the relay_from_hosts and authenticated
65 clients checks, on the grounds that messages accepted by these
66 statements are most likely to be submissions.
68 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
70 (1) Generate an error if the third argument for the ${prvs expansion is
73 (2) Treat a missing third argument of ${prvscheck as if it were an empty
76 (3) Reset the variables that are obtained from the first argument of
77 ${prvscheck and used in the second argument before leaving the code,
78 because their memory is reclaimed, so using them afterwards may do
81 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
82 one (it's much easier than Tom thought :-).
84 (5) Because of (4), we can now allow for the use of $prvscheck_result
85 inside the third argument.
87 PH/03 For some reason, the default setting of PATH when running a command from
88 a pipe transport was just "/usr/bin". I have changed it to
91 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
92 anything to be listed in the output from -bV.
94 PH/05 When a filter generated an autoreply, the entire To: header line was
95 quoted in the delivery log line, like this:
97 => >A.N.Other <ano@some.domain> <original@ddress> ...
99 This has been changed so that it extracts the operative address. There
100 may be more than one such address. If so, they are comma-separated, like
103 => >ano@some.domain,ona@other.domain <original@ddress> ...
105 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
106 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
107 not being looked up in the rDNS to get a host name, Exim was showing the
108 IP address twice in Received: lines, even though the IP addresses were
109 identical. For example:
111 Received: from [1.2.3.4] (helo=[1.2.3.4])
113 However, if the real host name was known, it was omitting the HELO data
114 if it matched the actual IP address. This has been tidied up so that it
115 doesn't show the same IP address twice.
117 PH/07 When both +timestamp and +memory debugging was on, the value given by
118 $tod_xxx expansions could be wrong, because the tod_stamp() function was
119 called by the debug printing, thereby overwriting the timestamp buffer.
120 Debugging no longer uses the tod_stamp() function when +timestamp is set.
122 PH/08 When the original message was included in an autoreply transport, it
123 always said "this is a copy of the message, including all the headers",
124 even if body_only or headers_only was set. It now gives an appropriate
127 PH/09 Applied a patch from the Sieve maintainer which:
129 o fixes some comments
130 o adds the (disabled) notify extension core
131 o adds some debug output for the result of if/elsif tests
132 o points to the current vacation draft in the documentation
133 and documents the missing references header update
137 o fixes a bug in processing the envelope test (when testing
138 multiple envelope elements, the last element determinted the
141 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
142 Electronic Mail") by including:
144 Auto-submitted: auto-generated
146 in the messages that it generates (bounce messages and others, such as
147 warnings). In the case of bounce messages for non-SMTP mesages, there was
148 also a typo: it was using "Auto_submitted" (underscore instead of
149 hyphen). Since every message generated by Exim is necessarily in response
150 to another message, thes have all been changed to:
152 Auto-Submitted: auto-replied
154 in accordance with these statements in the RFC:
156 The auto-replied keyword:
158 - SHOULD be used on messages sent in direct response to another
159 message by an automatic process,
161 - MUST NOT be used on manually-generated messages,
163 - MAY be used on Delivery Status Notifications (DSNs) and Message
164 Disposition Notifications (MDNs),
166 - MUST NOT be used on messages generated by automatic or periodic
167 processes, except for messages which are automatic responses to
170 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
171 to the default Received: header definition.
173 PH/12 Added log selector acl_warn_skipped (default on).
175 PH/13 After a successful wildlsearch lookup, discard the values of numeric
176 variables because (a) they are in the wrong storage pool and (b) even if
177 they were copied, it wouldn't work properly because of the caching.
179 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
180 checking when decoding. Apparently there are clients that generate
181 overlong encoded strings. Why am I not surprised?
183 PH/15 If the first argument of "${if match_address" was not empty, but did not
184 contain an "@" character, Exim crashed. Now it writes a panic log message
185 and treats the condition as false.
187 PH/16 In autoreply, treat an empty string for "once" the same as unset.
189 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
190 extension "envelope-auth". The code is finished and in agreement with
191 other implementations, but there is no documentation so far and in fact,
192 nobody wrote the draft yet. This extension is currently #undef'ed, thus
193 not changing the active code.
195 Print executed "if" and "elsif" statements when debugging is used. This
196 helps a great deal to understand what a filter does.
198 Document more things not specified clearly in RFC3028. I had all this
199 sorted out, when out of a sudden new issues came to my mind. Oops."
201 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
204 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
205 canonical form (as documented). However, after a host name lookup from
206 the IP address, check_host() was doing a simple string comparison with
207 addresses acquired from the DNS when checking that the found name did
208 have the original IP as one of its addresses. Since any found IPv6
209 addresses are likely to be in abbreviated form, the comparison could
210 fail. Luckily, there already exists a function for doing the comparison
211 by converting both addresses to binary, so now that is used instead of
214 PH/20 There was another similar case to PH/19, when a complete host name was
215 given in a host list; looking up its IP address could give an abbreviated
216 form, whereas the current host's name might or might not be abbreviated.
217 The same fix has been applied.
223 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
224 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
227 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
228 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
230 PH/03 Typo: missing ".o" in src/pcre/Makefile.
232 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
233 header line, restrict the check to what is listed in RFCs 2369 and 2929.
234 Also, for "Auto-Submitted", treat anything other than "no" as
235 non-personal, in accordance with RFC 3834. (Previously it treated
236 anything starting "auto-" as non-personal.)
238 TF/01 The control=submission/name=... option had a problem with syntax
239 errors if the name included a slash character. The /name= option
240 now slurps the rest of the string, so it can include any characters
241 but it must come last in the list of options (after /sender_retain
244 PH/05 Some modifications to the interface to the fake nameserver for the new
252 TK/01 Added the "success_on_redirect" address verification option. See
253 NewStuff for rationale and an example.
255 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
257 PH/02 Patch to exigrep to allow it to work on syslog lines.
259 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
260 fread() to skip over the body file's header line, because in Cygwin the
261 header line is locked and is inaccessible.
263 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
264 co-exist for some time) to make it clear that it is the Exim ID that is
265 referenced, not the Message-ID: header line.
267 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
268 string_format() function, because snprintf() does not exist on all
271 PH/06 The use of forbid_filter_existstest now also locks out the use of the
272 ${stat: expansion item.
274 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
275 protocol synchronization error", to keep the pedants happy.
277 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
278 well as for IRIX systems, when gcc is being used. See the host.c source
281 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
283 PH/10 Named domain lists were not working if used in a queue_smtp_domains
286 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
287 transport and to the smtp transport in LMTP mode.
289 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
291 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
292 run a filter in a subprocess. This could lead to confusion in subsequent
293 lookups in the parent process. There should also be a search_tidyup() at
294 the end of the subprocess.
296 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
297 only if the host matched helo_try_verify_hosts, which caused the
298 verification to occur when the EHLO/HELO command was issued. The ACL just
299 tested the remembered result. Now, if a previous verification attempt has
300 not happened, "verify = helo" does it there and then.
302 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
304 TK/03 Fix log output including CR from clamd.
306 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
307 exist provoked a memory error which could cause a segfault.
309 PH/15 Installed PCRE 6.2
311 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
313 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
314 of the problem. Specifically, suggested +O2 rather than +O1 for the
317 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
319 PH/20 If a delivery was routed to a non-standard port by means of an SRV
320 record, the port was not correctly logged when the outgoing_port log
321 selector was set (it logged the transort's default port).
323 PH/21 Added support for host-specific ports to manualroute, queryprogram,
324 fallback_hosts, and "hosts" in the smtp transport.
326 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
327 host errors such as "Connection refused".
329 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
330 authentication with radiusclient 0.4.9:
332 - Error returned from rc_read_config was caught wrongly
333 - Username/password not passed on to radius server due to wrong length.
335 The presumption is that some radiusclient API changes for 4.51/PH/17
336 were not taken care of correctly. The code is still untested by me (my
337 Linux distribution still has 0.3.2 of radiusclient), but it was
338 contributed by a Radius user.
340 PH/24 When doing a callout, the value of $domain wasn't set correctly when
341 expanding the "port" option of the smtp transport.
343 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
344 while reading a MIME header. Thanks to Tom Hughes for a patch.
346 PH/24 Include config.h inside local_scan.h so that configuration settings are
349 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
350 that in an ACL for RCPT (for example), you can examine exactly what was
353 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
354 commands, but it was not correctly comparing the address with the actual
355 client host address. Thus, it would show the EHLO address in Received:
356 header lines when this was not necessary.
358 PH/27 Added the % operator to ${eval:}.
360 PH/28 Exim tries to create and chdir to its spool directory when it starts;
361 it should be ignoring failures (because with -C, for example, it has lost
362 privilege). It wasn't ignoring creation failures other than "already
365 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
368 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
369 because an input line was too long, either on its own, or by virtue of
370 too many continuations, the temporary file was not being removed, and the
371 return code was incorrect.
373 PH/31 Missing "BOOL" in function definition in filtertest.c.
375 PH/32 Applied Sieve patches from the maintainer.
377 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
379 PH/33 Added "verify = not_blind".
381 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
382 Local/Makefile (with some defaults set). These are used in built scripts
383 such as exicyclog, but they have never been used in the exim_install
384 script (though there are many overriding facilities there). I have
385 arranged that the exim_install script now takes note of these two
388 PH/35 Installed configuration files for Dragonfly.
390 PH/36 When a locally submitted message by a trusted user did not contain a
391 From: header, and the sender address was obtained from -f or from an SMTP
392 MAIL command, and the trusted user did not use -F to supply a sender
393 name, $originator_name was incorrectly used when constructing a From:
394 header. Furthermore, $originator_name was used for submission mode
395 messages from external hosts without From: headers in a similar way,
396 which is clearly wrong.
398 PH/37 Added control=suppress_local_fixups.
400 PH/38 When log_selector = +received_sender was set, and the addition of the
401 sender made the log line's construction buffer exactly full, or one byte
402 less than full, an overflow happened when the terminating "\n" was
405 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
406 when the result of a list match is failure because a DNS lookup failed.
408 PH/40 RM_COMMAND is now used in the building process.
410 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
411 the "build-* directories that it finds.
413 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
414 address in a domain literal was a prefix of an interface address.
416 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
417 when verifying a sender address, unless rewrite_headers is false.
419 PH/44 Wrote a long comment about why errors_to addresses are verified as
420 recipients, not senders.
422 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
423 the ratelimit ACL was added.
425 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
427 PH/46 Added extra information about PostgreSQL errors to the error string.
429 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
430 avoiding the need to install special zones in a real server. This is
431 backwards compatible; if it can't find the fake resolver, it drops back.
432 Thus, both old and new test suites can be run.
434 TF/02 Added util/ratelimit.pl
436 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
437 clock is set back in time.
439 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
440 Candler <B.Candler@pobox.com>.
442 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
443 OK for addresses that are the result of redirections.
445 PH/48 A number of further additions for the benefit of the new test suite,
446 including a fake gethostbyname() that interfaces to the fake DNS resolver
449 TF/06 The fix for widen_domains has also been applied to qualify_single and
450 search_parents which are the other dnslookup options that can cause
453 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
456 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
458 TF/07 Exim produced the error message "an SRV record indicated no SMTP
459 service" if it encountered an MX record with an empty target hostname.
460 The message is now "an MX or SRV record indicated no SMTP service".
462 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
463 if the DNS of the sending site is misconfigured. This is quite a
464 common situation. This change restores the behaviour of treating a
465 helo verification defer as a failure.
467 PH/51 If self=fail was set on a router, the bounce message did not include the
468 actual error message.
474 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
476 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
477 command itself ended in error, the underlying message about the transport
478 filter timeout was being overwritten with the pipe command error. Now the
479 underlying error message should be appended to the second error message.
481 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
483 PH/02 Reduce the amount of output that "make" produces by default. Full output
484 can still be requested.
486 PH/03 The warning log line about a condition test deferring for a "warn" verb
487 was being output only once per connection, rather than after each
488 occurrence (because it was using the same function as for successful
489 "warn" verbs). This seems wrong, so I have changed it.
491 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
492 it should not have, which might have caused a crash in the right
493 circumstances, but probably never did.
495 PH/04 Installed a modified version of Tony Finch's patch to make submission
496 mode fix the return path as well as the Sender: header line, and to
497 add a /name= option so that you can make the user's friendly name appear
500 TF/03 Added the control = fakedefer ACL modifier.
502 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
503 Mark Lowes for thorough testing.
505 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
507 TK/03 Merged latest SRS patch from Miles Wilton.
509 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
510 with the definition in sysexits.h (which is #included earlier).
511 Fortunately, Exim does not actually use EX_OK. The code used to try to
512 preserve the sysexits.h value, by assumimg that macro definitions were
513 scanned for macro replacements. I have been disabused of this notion,
514 so now the code just undefines EX_OK before #including unistd.h.
516 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
517 in the smtp transport. When a block could not be written in a single
518 write() function, the timeout was being re-applied to each part-write.
519 This seems wrong - if the receiver was accepting one byte at a time it
520 would take for ever. The timeout is now adjusted when this happens. It
521 doesn't have to be particularly precise.
523 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
524 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
526 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
527 without a domain if the check to <postmaster@domain> fails.
529 SC/01 Eximstats: added -xls and the ability to specify output files
530 (patch written by Frank Heydlauf).
532 SC/02 Eximstats: use FileHandles for outputing results.
534 SC/03 Eximstats: allow any combination of xls, txt, and html output.
536 SC/04 Eximstats: fixed display of large numbers with -nvr option
538 SC/05 Eximstats: fixed merging of reports with empty tables.
540 SC/06 Eximstats: added the -include_original_destination flag
542 SC/07 Eximstats: removed tabs and trailing whitespace.
544 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
546 TK/06 MBOX spool code: Add real "From " MBOX separator line
547 so the .eml file is really in mbox format (even though
548 most programs do not really care). Patch from Alex Miller.
550 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
551 The latter is generated from $received_to and is only set if the
552 message has one envelope recipient. SA can use these headers,
553 obviously out-of-the-box. Patch from Alex Miller.
555 PH/08 The ${def test on a variable was returning false if the variable's
556 value was "0", contrary to what the specification has always said!
557 The result should be true unless the variable is empty.
559 PH/09 The syntax error of a character other than { following "${if
560 def:variable_name" (after optional whitespace) was not being diagnosed.
561 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
562 accidental colon was present, for example, could give incorrect results.
564 PH/10 Tidied the code in a number of places where the st_size field of a stat()
565 result is used (not including appendfile, where other changes are about
568 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
569 This involved changing a lot of size variables from int to off_t. It
570 should work with maildirs and everything.
572 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
573 spamd dying while we are connected to it.
575 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
578 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
579 function for BDB 4.3.
581 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
583 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
584 the sources for PCRE have been re-arranged and more files are now
587 PH/15 The code I had for printing potentially long long variables in PH/11
588 above was not the best (it lost precision). The length of off_t variables
589 is now inspected at build time, and an appropriate printing format (%ld
590 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
591 to be "long long int" or "long int". This is needed for the internal
592 formatting function string_vformat().
594 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
595 the configuration file to be ":syslog", then the script "guesses" where
596 the logs files are, rather than using the compiled in default. In our
597 case the guess is not the same as the compiled default, so the script
598 suddenly stopped working when I started to use syslog. The patch checks
599 to see if log_file_path is "". If so, it attempts to read it from exim
600 with no configuration file to get the compiled in version, before it
601 falls back to the previous guessing code."
603 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
604 implementing BATV in an Exim configuration. See NewStuff for the gory
607 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
608 Makefile that are specific to HP-UX.
610 PH/18 If the "use_postmaster" option was set for a recipient callout together
611 with the "random" option, the postmaster address was used as the MAIL
612 FROM address for the random test, but not for the subsequent recipient
613 test. It is now used for both.
615 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
616 patch removes a few documentation additions to RFC 3028, because the
617 latest draft now contains them. It adds the new en;ascii-case comparator
618 and a new error check for 8bit text in MIME parts. Comparator and
619 require names are now matched exactly. I enabled the subaddress
620 extension, but it is not well tested yet (read: it works for me)."
622 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
623 rework some of the code of TK/09 above to avoid the hardwired use of
624 "%lld" and "long long". Replaced the call to snprintf() with a call to
627 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
628 records point to non-existent hosts", "retry timeout exceeded", and
629 "retry time not reached for any host after a long failure period".
631 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
632 experimental DomainKeys support:
634 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
635 (2) On an error such as an illegally used "control", the wrong name for
636 the control was given.
638 These problems did NOT occur unless DomainKeys support was compiled.
640 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
642 PH/24 Added ${if match_ip condition.
644 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
645 databases so that it will be absolutely obvious if a crash occurs in the
646 DB library. This is a regular occurrence (often caused by mis-matched
649 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
650 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
651 picked up on a user's system that detects such things. There doesn't seem
652 to be a gcc warning option for this - only an attribute that has to be
653 put on the function's prototype. It seems that in Fedora Core 4 they have
654 set this on a number of new functions. No doubt there will be more in due
657 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
658 specify a transport. However, if an address that was verified by such a
659 router was the subject of a callout, Exim crashed because it tried to
660 read the rcpt_include_affixes from the non-existent transport. Now it
661 just assumes that the setting of that option is false. This bug was
662 introduced by 4.51/PH/31.
664 PH/28 Changed -d+all to exclude +memory, because that information is very
665 rarely of interest, but it makes the output a lot bigger. People tend to
666 do -d+all out of habit.
668 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
669 code in os-type was giving problems when libc.so lives in lib64, like on
672 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
673 aren't the modern standard, and it seems that some systems' include files
674 don't always have them. Exim was already checking for some of the newer
675 ones like T_AAAA, and defining it itself. I've added checks for all the
676 record types that Exim uses.
678 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
679 not automatically generating a new one, as it is supposed to. This
680 prevented TLS from working. If the file did exist, but contained invalid
681 data, a new version was generated, as expected. It was only the case of a
682 non-existent file that was broken.
684 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
685 with a change in libdomainkeys > 0.64.
687 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
688 from DNS. If the selector record carries the flag, it now has
689 precedence over the domain-wide flag.
691 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
693 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
694 the use of a port name that isn't defined in /etc/services) occurred, the
695 message was deferred as in a normal delivery, and thus remained on the
696 spool, instead of being failed because of the mua_wrapper setting. This
697 is now fixed, and I tidied up some of the mua_wrapper messages at the
700 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
701 the messages in a hash of arrays rather than using individual hashes.
702 This is a bit cleaner and results in dramatic memory savings, albeit
703 at a slight CPU cost.
705 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
706 as requested by Marc Sherman.
708 SC/10 Eximstats: added histograms for user specified patterns as requested
711 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
713 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
714 fopen() in the content-scanning modules that did not already have it.
720 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
721 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
723 TK/02 Fix ACL "control" statement not being available in MIME ACL.
725 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
727 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
728 to test Sieve filters that use "vacation".
730 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
731 that changes the way the GnuTLS parameters are stored in the cache file.
732 The new format can be generated externally. For backward compatibility,
733 if the data in the cache doesn't make sense, Exim assumes it has read an
734 old-format file, and it generates new data and writes a new file. This
735 means that you can't go back to an older release without removing the
738 PH/03 A redirect router that has both "unseen" and "one_time" set does not
739 work if there are any delivery delays because "one_time" forces the
740 parent to be marked "delivered", so its unseen clone is never tried
741 again. For this reason, Exim now forbids the simultaneous setting of
744 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
745 redirected to themselves ("homonym" addresses). Read the long ChangeLog
746 entry if you want to know the details. The fix, however, neglected to
747 consider the case when local delivery batching is involved. The test for
748 "previously delivered" was not happening when checking to see if an
749 address could be batched with a previous (undelivered) one; under
750 certain circumstances this could lead to multiple deliveries to the same
753 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
754 in its include files, and this causes problems building Exim.
756 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
757 header_syntax) but Exim was just ignoring anything given after a slash.
758 In particular, this caused confusion with an attempt to use "verify =
759 reverse_host_lookup/defer_ok". An error is now given when options are
760 supplied for verify items that do not have them. (Maybe reverse_host_
761 lookup should have a defer_ok option, but that's a different point.)
763 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
764 defined by RFC 821) to 2048, because there were problems with some AUTH
765 commands, and RFC 1869 says the size should be increased for extended
766 SMTP commands that take arguments.
768 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
771 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
772 "unknown" error; now it says that the functionality isn't in the binary.
774 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
775 an address' error message when a string expansion fails (syntax or
776 whatever). Otherwise the password may appear in the log. Following change
777 PH/42 below, there is no longer a chance of it appearing in a bounce
780 PH/11 Installed exipick version 20050225.0 from John Jetmore.
782 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
783 of its addresses was ever tried. (Bugzilla bug #2.)
785 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
786 the result incorrectly in the debug output. (It correctly added a newline
787 to what was transported.)
789 TF/01 Added $received_time.
791 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
792 commented out examples of how to interface to a virus scanner and to
793 SpamAssassin. Also added commented examples of av_scanner and
794 spamd_address settings.
796 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
797 and controls are allowed in which ACLs. There were a couple of minor
798 errors. Some of the entries in the conditions table (which is a table of
799 where they are NOT allowed) were getting very unwieldy; rewrote them as a
800 negation of where the condition IS allowed.
802 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
804 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
805 header file does not have a version number, so I've had to invent a new
806 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
807 API. The code is untested by me (my Linux distribution still has 0.3.2 of
808 radiusclient), but it was contributed by a Radius user.
810 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
811 files or directories, for OpenSSL.
813 PH/19 When an Exim process that is running as root has to create an Exim log
814 file, it does so in a subprocess that runs as exim:exim so as to get the
815 ownership right at creation (otherwise, other Exim processes might see
816 the file with the wrong ownership). There was no test for failure of this
817 fork() call, which would lead to the process getting stuck as it waited
818 for a non-existent subprocess. Forks do occasionally fail when resources
819 run out. I reviewed all the other calls to fork(); they all seem to check
822 PH/20 When checking for unexpected SMTP input at connect time (before writing
823 the banner), Exim was not dealing correctly with a non-positive return
824 from the read() function. If the client had disconnected by this time,
825 the result was a log entry for a synchronization error with an empty
826 string after "input=" when read() returned zero. If read() returned -1
827 (an event I could not check), uninitialized data bytes were printed.
828 There were reports of junk text (parts of files, etc) appearing after
831 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
833 PH/22 Added support for macro redefinition, and (re)definition in between
834 driver and ACL definitions.
836 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
837 forgetting to use the resulting value; it was using the unexpanded value.
839 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
840 hadn't been configured. The fix is from Juergen Kreileder, who
841 understands it better than I do:
843 "Here's what I see happening with three configured cyrus_sasl
844 authenticators configured (plain, login, cram-md5):
846 On startup auth_cyrus_sasl_init() gets called for each of these.
847 This means three calls to sasl_listmech() without a specified mech_list.
848 => SASL tests which mechs of all available mechs actually work
849 => three warnings about OTP not working
850 => the returned list contains: plain, login, cram-md5, digest-md5, ...
852 With the patch, sasl_listmech() also gets called three times. But now
853 SASL's mech_list option is set to the server_mech specified in the the
854 authenticator. Or in other words, the answer from sasl_listmech()
855 gets limited to just the mech you're testing for (which is different
857 => the return list contains just 'plain' or 'login', 'cram-md5' or
858 nothing depending on the value of ob->server_mech.
860 I've just tested the patch: Authentication still works fine,
861 unavailable mechs specified in the exim configuration are still
862 caught, and the auth.log warnings about OTP are gone."
864 PH/25 When debugging is enabled, the contents of the command line are added
865 to the debugging output, even when log_selector=+arguments is not
868 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
869 answer is "GNU", and only if the return is "GNU/something" is the answer
872 PH/27 $acl_verify_message is now set immediately after the failure of a
873 verification in an ACL, and so is available in subsequent modifiers. In
874 particular, the message can be preserved by coding like this:
876 warn !verify = sender
877 set acl_m0 = $acl_verify_message
879 Previously, $acl_verify_message was set only while expanding "message"
880 and "log_message" when a very denied access.
882 PH/28 Modified OS/os.c-Linux with
884 -#ifndef OS_LOAD_AVERAGE
885 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
887 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
888 nomenclature these days.)
890 PH/29 Installed patch from the Sieve maintainer that adds the options
891 sieve_useraddress and sieve_subaddress to the redirect router.
893 PH/30 In these circumstances:
894 . Two addresses routed to the same list of hosts;
895 . First host does not offer TLS;
896 . First host accepts first address;
897 . First host gives temporary error to second address;
898 . Second host offers TLS and a TLS session is established;
899 . Second host accepts second address.
900 Exim incorrectly logged both deliveries with the TLS parameters (cipher
901 and peerdn, if requested) that were in fact used only for the second
904 PH/31 When doing a callout as part of verifying an address, Exim was not paying
905 attention to any local part prefix or suffix that was matched by the
906 router that accepted the address. It now behaves in the same way as it
907 does for delivery: the affixes are removed from the local part unless
908 rcpt_include_affixes is set on the transport.
910 PH/32 Add the sender address, as F=<...>, to the log line when logging a
911 timeout during the DATA phase of an incoming message.
913 PH/33 Sieve envelope tests were broken for match types other than :is. I have
914 applied a patch sanctioned by the Sieve maintainer.
916 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
917 the uid or gid is negative. A case of a negative gid caused this to be
918 noticed. The fix allows for either to be negative.
920 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
921 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
924 PH/36 The change PH/12 above was broken. Fixed it.
926 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
927 the grounds that routing the same address twice would always produce the
928 same answer. This might have been true once, but it is certainly no
929 longer true now. Routing a child address may depend on the previous
930 routing that produced that child. Some complicated redirection strategies
931 went wrong when messages had multiple recipients, and made Exim's
932 behaviour dependent on the order in which the addresses were given.
934 I have moved the duplicate checking until after the routing is complete.
935 Exim scans the addresses that are assigned to local and remote
936 transports, and removes any duplicates. This means that more work will be
937 done, as duplicates will always all be routed, but duplicates are
938 presumably rare, so I don't expect this is of any significance.
940 For deliveries to pipes, files, and autoreplies, the duplicate checking
941 still happens during the routing process, since they are not going to be
944 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
945 It corrects a timeout issue with spamd. This is Ian's comment: "The
946 background is that sometimes spamd either never reads data from a
947 connection it has accepted, or it never writes response data. The exiscan
948 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
949 blindly assumes that writes won't block so it may never time out."
951 PH/39 Allow G after quota size as well as K and M.
953 PH/40 The value set for $authenticated_id in an authenticator may not contain
954 binary zeroes or newlines because the value is written to log lines and
955 to spool files. There was no check on this. Now the value is run through
956 the string_printing() function so that such characters are converted to
957 printable escape sequences.
959 PH/41 $message_linecount is a new variable that contains the total number of
960 lines in the message. Compare $body_linecount, which is the count for the
963 PH/42 Exim no longer gives details of delivery errors for specific addresses in
964 bounce and delay warning messages, except in certain special cases, which
967 (a) An SMTP error message from a remote host;
968 (b) A message specified in a :fail: redirection;
969 (c) A message specified in a "fail" command in a system filter;
970 (d) A message specified in a FAIL return from the queryprogram router;
971 (e) A message specified by the cannot_route_message router option.
973 In these cases only, Exim does include the error details in bounce and
974 warning messages. There are also a few cases where bland messages such
975 as "unrouteable address" or "local delivery error" are given.
977 PH/43 $value is now also set for the "else" part of a ${run expansion.
979 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
980 being worked on, but at least Exim now implements the latest version to
983 PH/45 In a pipe transport, although a timeout while waiting for the pipe
984 process to complete was treated as a delivery failure, a timeout while
985 writing the message to the pipe was logged, but erroneously treated as a
986 successful delivery. Such timeouts include transport filter timeouts. For
987 consistency with the overall process timeout, these timeouts are now
988 treated as errors, giving rise to delivery failures by default. However,
989 there is now a new Boolean option for the pipe transport called
990 timeout_defer, which, if set TRUE, converts the failures into defers for
991 both kinds of timeout. A transport filter timeout is now identified in
994 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
995 systems where "make" and "gmake" are different, calling "gmake" at top
996 level broke things. I've arranged for the value of $(MAKE) to be passed
997 from the Makefile to this script so that it can call the same version of
1001 A note about Exim versions 4.44 and 4.50
1002 ----------------------------------------
1004 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
1005 changes of various kinds. As a consequence, a big documentation update was
1006 needed. This delayed the release for rather longer than seemed good, especially
1007 in the light of a couple of (minor) security issues. Therefore, the changes
1008 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
1009 release. So 4.44 and 4.50 are in effect two different branches that both start
1012 I have left the 4.50 change log unchanged; it contains all the changes since
1013 4.43. The change log for 4.44 is below; many of its items are identical to
1014 those for 4.50. This seems to be the most sensible way to preserve the
1015 historical information.
1021 1. Minor wording change to the doc/README.SIEVE file.
1023 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
1024 computation of the current number of files was incorrect.
1026 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
1027 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
1030 4. Give more explanation in the error message when the command for a transport
1031 filter fails to execute.
1033 5. There are several places where Exim runs a non-Exim command in a
1034 subprocess. The SIGUSR1 signal should be disabled for these processes. This
1035 was being done only for the command run by the queryprogram router. It is
1036 now done for all such subprocesses. The other cases are: ${run, transport
1037 filters, and the commands run by the lmtp and pipe transports.
1039 6. Added CONFIGURE_GROUP build-time option.
1041 7. Some older OS have a limit of 256 on the maximum number of file
1042 descriptors. Exim was using setrlimit() to set 1000 as a large value
1043 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
1044 systems. I've change it so that if it can't get 1000, it tries for 256.
1046 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
1047 was an oversight, and furthermore, ever since the addition of extra
1048 controls (e.g. 4.43/32), the checks on when to allow different forms of
1049 "control" were broken. There should now be diagnostics for all cases when a
1050 control that does not make sense is encountered.
1052 9. Added the /retain_sender option to "control=submission".
1054 10. $recipients is now available in the predata ACL (oversight).
1056 11. Tidy the search cache before the fork to do a delivery from a message
1057 received from the command line. Otherwise the child will trigger a lookup
1058 failure and thereby defer the delivery if it tries to use (for example) a
1059 cached ldap connection that the parent has called unbind on.
1061 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
1062 of $address_data from the recipient verification was clobbered by the
1063 sender verification.
1065 13. The value of address_data from a sender verification is now available in
1066 $sender_address_data in subsequent conditions in the ACL statement.
1068 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
1070 15. Added a new option "connect=<time>" to callout options, to set a different
1073 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
1074 was its contents. (It was OK if the option was not defined at all.)
1076 17. A "Completed" log line is now written for messages that are removed from
1077 the spool by the -Mrm option.
1079 18. New variables $sender_verify_failure and $recipient_verify_failure contain
1080 information about exactly what failed.
1082 19. Added -dd to debug only the daemon process.
1084 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
1085 handles timeouts, both on the server side and network timeouts. Renamed the
1086 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
1088 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
1089 It is now set to "smtps".
1091 22. $host_address is now set to the target address during the checking of
1092 ignore_target_hosts.
1094 23. When checking ignore_target_hosts for an ipliteral router, no host name was
1095 being passed; this would have caused $sender_host_name to have been used if
1096 matching the list had actually called for a host name (not very likely,
1097 since this list is usually IP addresses). A host name is now passed as
1100 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
1101 code that specifies a non-restarting handler (typically sigaction() in
1102 modern systems) in an attempt to fix a rare and obscure crash bug.
1104 25. Narrowed the window for a race in the daemon that could cause it to ignore
1105 SIGCHLD signals. This is not a major problem, because they are used only to
1106 wake it up if nothing else does.
1108 26. A malformed maildirsize file could cause Exim to calculate negative values
1109 for the mailbox size or file count. Odd effects could occur as a result.
1110 The maildirsize information is now recalculated if the size or filecount
1113 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
1114 support for a long time. Removed HAVE_SYS_VFS_H.
1116 28. Installed the latest version of exipick from John Jetmore.
1118 29. In an address list, if the pattern was not a regular expression, an empty
1119 subject address (from a bounce message) matched only if the pattern was an
1120 empty string. Non-empty patterns were not even tested. This was the wrong
1121 because it is perfectly reasonable to use an empty address as part of a
1122 database query. An empty address is now tested by patterns that are
1123 lookups. However, all the other forms of pattern expect the subject to
1124 contain a local part and a domain, and therefore, for them, an empty
1125 address still always fails if the pattern is not itself empty.
1127 30. Exim went into a mad DNS loop when attempting to do a callout where the
1128 host was specified on an smtp transport, and looking it up yielded more
1129 than one IP address.
1131 31. Re-factored the code for checking spool and log partition space into a
1132 function that finds that data and another that does the check. The former
1133 is then used to implement four new variables: $spool_space, $log_space,
1134 $spool_inodes, and $log_inodes.
1136 32. The RFC2047 encoding function was originally intended for short strings
1137 such as real names; it was not keeping to the 75-character limit for
1138 encoded words that the RFC imposes. It now respects the limit, and
1139 generates multiple encoded words if necessary. To be on the safe side, I
1140 have increased the buffer size for the ${rfc2047: expansion operator from
1143 33. It is now permitted to omit both strings after an "if" condition; if the
1144 condition is true, the result is "true". As before, when the second string
1145 is omitted, a false condition yields an empty string. This makes it less
1146 cumbersome to write custom ACL and router conditions.
1148 34. Failure to deliver a bounce message always caused it to be frozen, even if
1149 there was an errors_to setting on the router. The errors_to setting is now
1152 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
1153 canonical form (fully expanded) before being placed in
1154 $sender_host_address.
1156 36. The table in the code that translates DNS record types into text (T_A to
1157 "A" for instance) was missing entries for NS and CNAME. It is just possible
1158 that this could have caused confusion if both these types were looked up
1159 for the same domain, because the text type is used as part of Exim's
1160 per-process caching. But the chance of anyone hitting this buglet seems
1163 37. The dnsdb lookup has been extended in a number of ways.
1165 (1) There is a new type, "zns", which walks up the domain tree until it
1166 finds some nameserver records. It should be used with care.
1168 (2) There is a new type, "mxh", which is like "mx" except that it returns
1169 just the host names, not the priorities.
1171 (3) It is now possible to give a list of domains (or IP addresses) to be
1172 looked up. The behaviour when one of the lookups defers can be
1173 controlled by a keyword.
1175 (4) It is now possible to specify the separator character for use when
1176 multiple records are returned.
1178 38. The dnslists ACL condition has been extended: it is now possible to supply
1179 a list of IP addresses and/or domains to be looked up in a particular DNS
1182 39. Added log_selector=+queue_time_overall.
1184 40. When running the queue in the test harness, wait just a tad after forking a
1185 delivery process, to get repeatability of debugging output.
1187 41. Include certificate and key file names in error message when GnuTLS fails
1188 to set them up, because the GnuTLS error message doesn't include the name
1189 of the failing file when there is a problem reading it.
1191 42. Allow both -bf and -bF in the same test run.
1193 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
1195 44. The "Exiscan patch" is now merged into the mainline Exim source.
1197 45. Sometimes the final signoff response after QUIT could fail to get
1198 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
1199 < 0 before doing a fflush(). This bug looks as though it goes back to the
1200 introduction of TLS in release 3.20, but "sometimes" must have been rare
1201 because the tests only now provoked it.
1203 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
1204 (this can affect the format of dates).
1206 47. exim_tidydb, when checking for the continued existence of a message for
1207 which it has found a message-specific retry record, was not finding
1208 messages that were in split spool directories. Consequently, it was
1209 deleting retry records that should have stayed in existence.
1211 48. Steve fixed some bugs in eximstats.
1213 49. The SPA authentication driver was not abandoning authentication and moving
1214 on to the next authenticator when an expansion was forced to fail,
1215 contradicting the general specification for all authenticators. Instead it
1216 was generating a temporary error. It now behaves as specified.
1218 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
1219 (the order specifies the preference for clients). The order is now AES256,
1220 AES128, 3DES, ARCFOUR128.
1222 51. Small patch to Sieve code - explicitly set From: when generating an
1225 52. Exim crashed if a remote delivery caused a very long error message to be
1226 recorded - for instance if somebody sent an entire SpamAssassin report back
1227 as a large number of 550 error lines. This bug was coincidentally fixed by
1228 increasing the size of one of Exim's internal buffers (big_buffer) that
1229 happened as part of the Exiscan merge. However, to be on the safe side, I
1230 have made the code more robust (and fixed the comments that describe what
1233 53. Now that there can be additional text after "Completed" in log lines (if
1234 the queue_time_overall log selector is set), a one-byte patch to exigrep
1235 was needed to allow it to recognize "Completed" as not the last thing in
1238 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
1239 patch that reportedly fixes this has been added. I am not expert enough to
1240 create a test for it. This is what the patch creator wrote:
1242 "I found a little strange behaviour of ldap code when working with
1243 Windows 2003 AD Domain, where users was placed in more than one
1244 Organization Units. When I tried to give exim partial DN, the exit code
1245 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
1246 But simultaneously result of request was absolutely normal ldap result,
1247 so I produce this patch..."
1249 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
1250 REFERENCE, so I have modified the code to exclude the patch when that macro
1253 55. Some experimental protocols are using DNS PTR records for new purposes. The
1254 keys for these records are domain names, not reversed IP addresses. The
1255 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
1256 leaves it alone. Component reversal etc. now happens only for IP addresses.
1259 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
1261 57. Double the size of the debug message buffer (to 2048) so that more of very
1262 long debug lines gets shown.
1264 58. The exicyclog utility now does better if the number of log files to keep
1265 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
1267 59. Two changes related to the smtp_active_hostname option:
1269 (1) $smtp_active_hostname is now available as a variable.
1270 (2) The default for smtp_banner uses $smtp_active_hostname instead
1271 of $primary_hostname.
1273 60. The host_aton() function is supposed to be passed a string that is known
1274 to be a valid IP address. However, in the case of IPv6 addresses, it was
1275 not checking this. This is a hostage to fortune. Exim now panics and dies
1276 if the condition is not met. A case was found where this could be provoked
1277 from a dnsdb PTR lookup with an IPv6 address that had more than 8
1278 components; fortuitously, this particular loophole had already been fixed
1279 by change 4.50/55 above.
1281 If there are any other similar loopholes, the new check in host_aton()
1282 itself should stop them being exploited. The report I received stated that
1283 data on the command line could provoke the exploit when Exim was running as
1284 exim, but did not say which command line option was involved. All I could
1285 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
1286 running as the user.
1289 61. There was a buffer overflow vulnerability in the SPA authentication code
1290 (which came originally from the Samba project). I have added a test to the
1291 spa_base64_to_bits() function which I hope fixes it.
1294 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
1295 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
1297 63. The daemon start-up calls getloadavg() while still root for those OS that
1298 need the first call to be done as root, but it missed one case: when
1299 deliver_queue_load_max is set with deliver_drop_privilege. This is
1300 necessary for the benefit of the queue runner, because there is no re-exec
1301 when deliver_drop_privilege is set.
1303 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
1304 This has been fixed.
1306 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
1307 was in use, was not putting the data itself into the right store pool;
1308 consequently, it could be overwritten for a subsequent message in the same
1309 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
1312 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
1314 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
1315 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
1316 the function were treating the return as a boolean value, which happened to
1317 work because 0=false and not-0=true, but is not correct code.
1319 68. The host_aton() function was not handling scoped IPv6 addresses (those
1320 with, for example, "%eth0" on the end) correctly.
1322 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
1323 negated items (that is, ~something) in unsigned ints. Some compilers
1324 apparently mutter when there is no cast.
1326 70. If an address verification called from an ACL failed, and did not produce a
1327 user-specific message (i.e. there was only a "system" message), nothing was
1328 put in $acl_verify_message. In this situation, it now puts the system
1331 71. Change 4.23/11 added synchronization checking at the start of an SMTP
1332 session; change 4.31/43 added the unwanted input to the log line - except
1333 that it did not do this in the start of session case. It now does.
1335 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
1336 This is wrong and can cause the other end to generate a synchronization
1337 error if it is another Exim or anything else that does the synchronization
1338 check. A QUIT command is no longer sent after a timeout.
1340 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
1341 during host lookups.
1343 74. The defer_ok option of callout verification was not working if it was used
1344 when verifying addresses in header lines, that is, for this case:
1346 verify = header_sender/callout=defer_ok
1348 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
1349 those file descriptors could be used for SMTP connections. If anything
1350 wrote to stderr (the example that came up was "warn" in embedded Perl), it
1351 could be sent to the SMTP client, causing chaos. The daemon now opens
1352 stdin, stdout, and stderr to /dev/null when it puts itself into the
1355 76. Arrange for output from Perl's "warn" command to be written to Exim's main
1356 log by default. The user can override this with suitable Perl magic.
1358 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
1359 the log message when discard triggers, was not working for the DATA ACL or
1360 for the non-SMTP ACL.
1362 78. Error message wording change in sieve.c.
1364 79. If smtp_accept_max_per_host was set, the number of connections could be
1365 restricted to fewer than expected, because the daemon was trying to set up
1366 a new connection before checking whether the processes handling previous
1367 connections had finished. The check for completed processes is now done
1368 earlier. On busy systems, this bug wouldn't be noticed because something
1369 else would have woken the daemon, and it would have reaped the completed
1372 80. If a message was submitted locally by a user whose login name contained one
1373 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
1374 It caused a spool format error. I have fixed the spool reading code. A
1375 related problem was that the "from" clause in the Received: line became
1376 illegal because of the space(s). It is now covered by ${quote_local_part.
1378 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
1381 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
1383 83. Patch from Sieve maintainer to fix unterminated string problem in
1384 "vacation" handling.
1386 84. Some minor changes to the Linux configuration files to help with other
1387 OS variants using glibc.
1389 85. One more patch for Sieve to update vacation handling to latest spec.
1392 ----------------------------------------------------
1393 See the note above about the 4.44 and 4.50 releases.
1394 ----------------------------------------------------
1400 1. Change 4.43/35 introduced a bug that caused file counts to be
1401 incorrectly computed when quota_filecount was set in an appendfile
1404 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
1405 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
1408 3. Give more explanation in the error message when the command for a transport
1409 filter fails to execute.
1411 4. There are several places where Exim runs a non-Exim command in a
1412 subprocess. The SIGUSR1 signal should be disabled for these processes. This
1413 was being done only for the command run by the queryprogram router. It is
1414 now done for all such subprocesses. The other cases are: ${run, transport
1415 filters, and the commands run by the lmtp and pipe transports.
1417 5. Some older OS have a limit of 256 on the maximum number of file
1418 descriptors. Exim was using setrlimit() to set 1000 as a large value
1419 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
1420 systems. I've change it so that if it can't get 1000, it tries for 256.
1422 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
1423 was an oversight, and furthermore, ever since the addition of extra
1424 controls (e.g. 4.43/32), the checks on when to allow different forms of
1425 "control" were broken. There should now be diagnostics for all cases when a
1426 control that does not make sense is encountered.
1428 7. $recipients is now available in the predata ACL (oversight).
1430 8. Tidy the search cache before the fork to do a delivery from a message
1431 received from the command line. Otherwise the child will trigger a lookup
1432 failure and thereby defer the delivery if it tries to use (for example) a
1433 cached ldap connection that the parent has called unbind on.
1435 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
1436 of $address_data from the recipient verification was clobbered by the
1437 sender verification.
1439 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
1440 was its contents. (It was OK if the option was not defined at all.)
1442 11. A "Completed" log line is now written for messages that are removed from
1443 the spool by the -Mrm option.
1445 12. $host_address is now set to the target address during the checking of
1446 ignore_target_hosts.
1448 13. When checking ignore_target_hosts for an ipliteral router, no host name was
1449 being passed; this would have caused $sender_host_name to have been used if
1450 matching the list had actually called for a host name (not very likely,
1451 since this list is usually IP addresses). A host name is now passed as
1454 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
1455 code that specifies a non-restarting handler (typically sigaction() in
1456 modern systems) in an attempt to fix a rare and obscure crash bug.
1458 15. Narrowed the window for a race in the daemon that could cause it to ignore
1459 SIGCHLD signals. This is not a major problem, because they are used only to
1460 wake it up if nothing else does.
1462 16. A malformed maildirsize file could cause Exim to calculate negative values
1463 for the mailbox size or file count. Odd effects could occur as a result.
1464 The maildirsize information is now recalculated if the size or filecount
1467 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
1468 support for a long time. Removed HAVE_SYS_VFS_H.
1470 18. Updated exipick to current release from John Jetmore.
1472 19. Allow an empty sender to be matched against a lookup in an address list.
1473 Previously the only cases considered were a regular expression, or an
1476 20. Exim went into a mad DNS lookup loop when doing a callout where the
1477 host was specified on the transport, if the DNS lookup yielded more than
1480 21. The RFC2047 encoding function was originally intended for short strings
1481 such as real names; it was not keeping to the 75-character limit for
1482 encoded words that the RFC imposes. It now respects the limit, and
1483 generates multiple encoded words if necessary. To be on the safe side, I
1484 have increased the buffer size for the ${rfc2047: expansion operator from
1487 22. Failure to deliver a bounce message always caused it to be frozen, even if
1488 there was an errors_to setting on the router. The errors_to setting is now
1491 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
1492 canonical form (fully expanded) before being placed in
1493 $sender_host_address.
1495 24. Updated eximstats to version 1.33
1497 25. Include certificate and key file names in error message when GnuTLS fails
1498 to set them up, because the GnuTLS error message doesn't include the name
1499 of the failing file when there is a problem reading it.
1501 26. Expand error message when OpenSSL has problems setting up cert/key files.
1504 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
1505 (this can affect the format of dates).
1507 28. exim_tidydb, when checking for the continued existence of a message for
1508 which it has found a message-specific retry record, was not finding
1509 messages that were in split spool directories. Consequently, it was
1510 deleting retry records that should have stayed in existence.
1512 29. eximstats updated to version 1.35
1513 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
1514 1.35 - bugfix such that pie charts by volume are generated correctly
1516 30. The SPA authentication driver was not abandoning authentication and moving
1517 on to the next authenticator when an expansion was forced to fail,
1518 contradicting the general specification for all authenticators. Instead it
1519 was generating a temporary error. It now behaves as specified.
1521 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
1522 (the order specifies the preference for clients). The order is now AES256,
1523 AES128, 3DES, ARCFOUR128.
1525 31. Small patch to Sieve code - explicitly set From: when generating an
1528 32. Exim crashed if a remote delivery caused a very long error message to be
1529 recorded - for instance if somebody sent an entire SpamAssassin report back
1530 as a large number of 550 error lines. This bug was coincidentally fixed by
1531 increasing the size of one of Exim's internal buffers (big_buffer) that
1532 happened as part of the Exiscan merge. However, to be on the safe side, I
1533 have made the code more robust (and fixed the comments that describe what
1536 33. Some experimental protocols are using DNS PTR records for new purposes. The
1537 keys for these records are domain names, not reversed IP addresses. The
1538 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
1539 leaves it alone. Component reversal etc. now happens only for IP addresses.
1542 34. The host_aton() function is supposed to be passed a string that is known
1543 to be a valid IP address. However, in the case of IPv6 addresses, it was
1544 not checking this. This is a hostage to fortune. Exim now panics and dies
1545 if the condition is not met. A case was found where this could be provoked
1546 from a dnsdb PTR lookup with an IPv6 address that had more than 8
1547 components; fortuitously, this particular loophole had already been fixed
1548 by change 4.50/55 or 4.44/33 above.
1550 If there are any other similar loopholes, the new check in host_aton()
1551 itself should stop them being exploited. The report I received stated that
1552 data on the command line could provoke the exploit when Exim was running as
1553 exim, but did not say which command line option was involved. All I could
1554 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
1555 running as the user.
1558 35. There was a buffer overflow vulnerability in the SPA authentication code
1559 (which came originally from the Samba project). I have added a test to the
1560 spa_base64_to_bits() function which I hope fixes it.
1563 36. The daemon start-up calls getloadavg() while still root for those OS that
1564 need the first call to be done as root, but it missed one case: when
1565 deliver_queue_load_max is set with deliver_drop_privilege. This is
1566 necessary for the benefit of the queue runner, because there is no re-exec
1567 when deliver_drop_privilege is set.
1569 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
1570 was in use, was not putting the data itself into the right store pool;
1571 consequently, it could be overwritten for a subsequent message in the same
1572 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
1575 38. Sometimes the final signoff response after QUIT could fail to get
1576 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
1577 < 0 before doing a fflush(). This bug looks as though it goes back to the
1578 introduction of TLS in release 3.20, but "sometimes" must have been rare
1579 because the tests only now provoked it.
1585 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
1586 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
1587 Now it returns an integer. A number of calls were still expecting a T/F
1588 return. Fortuitously, in all cases, the tests worked in OK situations,
1589 which is the norm. However, things would have gone wrong on any write
1590 failures on the smtp file descriptor. This function is used when sending
1591 messages over SMTP and also when doing verify callouts.
1593 2. When Exim is called to do synchronous delivery of a locally submitted
1594 message (the -odf or -odi options), it no longer closes stderr before doing
1597 3. Implemented the mua_wrapper option.
1599 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
1601 5. Implemented the functions header_remove(), header_testname(),
1602 header_add_at_position(), and receive_remove_recipient(), and exported them
1605 6. If an ACL "warn" statement specified the addition of headers, Exim already
1606 inserted X-ACL-Warn: at the start if there was no header name. However, it
1607 was not making this test for the second and subsequent header lines if
1608 there were newlines in the string. This meant that an invalid header could
1609 be inserted if Exim was badly configured.
1611 7. Allow an ACL "warn" statement to add header lines at the start or after all
1612 the Received: headers, as well as at the end.
1614 8. Added the rcpt_4xx retry error code.
1616 9. Added postmaster_mailfrom=xxx to callout verification option.
1618 10. Added mailfrom=xxxx to the callout verification option, for verify=
1621 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
1622 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
1624 12. Inserted some casts to stop certain compilers warning when using pointer
1625 differences as field lengths or precisions in printf-type calls (mostly
1626 affecting debugging statements).
1628 13. Added optional readline() support for -be (dynamically loaded).
1630 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
1631 same clock tick as a message's arrival, so that its received time was the
1632 same as the "first fail" time on the retry record, and that message
1633 remained on the queue past the ultimate address timeout, every queue runner
1634 would try a delivery (because it was past the ultimate address timeout) but
1635 after another failure, the ultimate address timeout, which should have then
1636 bounced the address, did not kick in. This was a "< instead of <=" error;
1637 in most cases the first failure would have been in the next clock tick
1638 after the received time, and all would be well.
1640 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
1641 being recognized when the domain list was tested by the match_domain
1642 condition in an expansion string.
1644 16. Added the ${str2b64: operator.
1646 17. Exim was always calling setrlimit() to set a large limit for the number of
1647 processes, without checking whether the existing limit was already
1648 adequate. (It did check for the limit on file descriptors.) Furthermore,
1649 errors from getrlimit() and setrlimit() were being ignored. Now they are
1650 logged to the main and panic logs, but Exim does carry on, to try to do its
1651 job under whatever limits there are.
1653 18. Imported PCRE 5.0.
1655 19. Trivial typo in log message " temporarily refused connection" (the leading
1658 20. If the log selector return_path_on_delivery was set and an address was
1659 redirected to /dev/null, the delivery process crashed because it assumed
1660 that a return path would always be set for a "successful" delivery. In this
1661 case, the whole delivery is bypassed as an optimization, and therefore no
1664 21. Internal re-arrangement: the function for sending a challenge and reading
1665 a response while authentication was assuming a zero-terminated challenge
1666 string. It's now changed to take a pointer and a length, to allow for
1667 binary data in such strings.
1669 22. Added the cyrus_sasl authenticator (code supplied by MBM).
1671 23. Exim was not respecting finduser_retries when seeking the login of the
1672 uid under which it was called; it was always trying 10 times. (The default
1673 setting of finduser_retries is zero.) Also, it was sleeping after the final
1674 failure, which is pointless.
1676 24. Implemented tls_on_connect_ports.
1678 25. Implemented acl_smtp_predata.
1680 26. If the domain in control=submission is set empty, Exim assumes that the
1681 authenticated id is a complete email address when it generates From: or
1682 Sender: header lines.
1684 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
1685 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
1686 chown and chgrp in /bin and hostname in /usr/bin.
1688 28. Exim was keeping the "process log" file open after each use, just as it
1689 does for the main log. This opens the possibility of it remaining open for
1690 long periods when the USR1 signal hits a daemon. Occasional processlog
1691 errors were reported, that could have been caused by this. Anyway, it seems
1692 much more sensible not to leave this file open at all, so that is what now
1695 29. The long-running daemon process does not normally write to the log once it
1696 has entered its main loop, and it closes the log before doing so. This is
1697 so that log files can straightforwardly be renamed and moved. However,
1698 there are a couple of unusual error situations where the daemon does write
1699 log entries, and I had neglected to close the log afterwards.
1701 30. The text of an SMTP error response that was received during a remote
1702 delivery was being truncated at 512 bytes. This is too short for some of
1703 the long messages that one sometimes sees. I've increased the limit to
1706 31. It is now possible to make retry rules that apply only when a message has a
1707 specific sender, in particular, an empty sender.
1709 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
1710 it possible to be selective about when SMTP synchronization is enforced.
1712 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
1714 32. Implemented hosts_connection_nolog.
1716 33. Added an ACL for QUIT.
1718 34. Setting "delay_warning=" to disable warnings was not working; it gave a
1721 35. Added mailbox_size and mailbox_filecount to appendfile.
1723 36. Added control = no_multiline_responses to ACLs.
1725 37. There was a bug in the logic of the code that waits for the clock to tick
1726 in the case where the clock went backwards by a substantial amount such
1727 that the microsecond fraction of "now" was more than the microsecond
1728 fraction of "then" (but the whole seconds number was less).
1730 38. Added support for the libradius Radius client library this is found on
1731 FreeBSD (previously only the radiusclient library was supported).
1737 1. When certain lookups returned multiple values in the form name=value, the
1738 quoting of the values was not always being done properly. Specifically:
1739 (a) If the value started with a double quote, but contained no whitespace,
1741 (b) If the value contained whitespace other than a space character (i.e.
1742 tabs or newlines or carriage returns) it was not quoted.
1743 This fix has been applied to the mysql and pgsql lookups by writing a
1744 separate quoting function and calling it from the lookup code. The fix
1745 should probably also be applied to nisplus, ibase and oracle lookups, but
1746 since I cannot test any of those, I have not disturbed their existing code.
1748 2. A hit in the callout cache for a specific address caused a log line with no
1749 reason for rejecting RCPT. Now it says "Previous (cached) callout
1750 verification failure".
1752 3. There was an off-by-one bug in the queryprogram router. An over-long
1753 return line was truncated at 256 instead of 255 characters, thereby
1754 overflowing its buffer with the terminating zero. As well as fixing this, I
1755 have increased the buffer size to 1024 (and made a note to document this).
1757 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
1758 when Exim is waiting for an SMTP response from a remote server, Exim
1759 restarts its select() call on the socket, thereby resetting its timeout.
1760 This is not a problem when such interrupts are rare. Somebody set up a cron
1761 job to run exiwhat every 2 minutes, which is less than the normal select()
1762 timeout (5 or 10 minutes). This meant that the select() timeout never
1763 kicked in because it was always reset. I have fixed this by comparing the
1764 time when an interrupt arrives with the time at the start of the first call
1765 to select(). If more time than the timeout has elapsed, the interrupt is
1766 treated as a timeout.
1768 5. Some internal re-factoring in preparation for the addition of Sieve
1769 extensions (by MH). In particular, the "personal" test is moved to a
1770 separate function, and given an option for scanning Cc: and Bcc: (which is
1771 not set for Exim filters).
1773 6. When Exim created an email address using the login of the caller as the
1774 local part (e.g. when creating a From: or Sender: header line), it was not
1775 quoting the local part when it contained special characters such as @.
1777 7. Installed new OpenBSD configuration files.
1779 8. Reworded some messages for syntax errors in "and" and "or" conditions to
1780 try to make them clearer.
1782 9. Callout options, other than the timeout value, were being ignored when
1783 verifying sender addresses in header lines. For example, when using
1785 verify = header_sender/callout=no_cache
1787 the cache was (incorrectly) being used.
1789 10. Added a missing instance of ${EXE} to the exim_install script; this affects
1790 only the Cygwin environment.
1792 11. When return_path_on_delivery was set as a log selector, if different remote
1793 addresses in the same message used different return paths and parallel
1794 remote delivery occurred, the wrong values would sometimes be logged.
1795 (Whenever a remote delivery process finished, the return path value from
1796 the most recently started remote delivery process was logged.)
1798 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
1799 lines when AUTH and/or TLS are in use. This is the "received protocol"
1800 field. Exim used to use "asmtp" for authenticated SMTP, without any
1801 indication (in the protocol name) for TLS use. Now it follows the RFC and
1802 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
1803 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
1804 names appear in log lines as well as in Received: header lines.
1806 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
1807 extensions, and comparison tests, and to fix some bugs.
1809 14. Changes to the "personal" filter test:
1811 (1) The test was buggy in that it was just doing the equivalent of
1812 "contains" tests on header lines. For example, if a user's address was
1813 anne@some.where, the "personal" test would incorrectly be true for
1815 To: susanne@some.where
1817 This test is now done by extracting each address from the header in turn,
1818 and checking the entire address. Other tests that are part of "personal"
1819 are now done using regular expressions (for example, to check local parts
1820 of addresses in From: header lines).
1822 (2) The list of non-personal local parts in From: addresses has been
1823 extended to include "listserv", "majordomo", "*-request", and "owner-*",
1824 taken from the Sieve specification recommendations.
1826 (3) If the message contains any header line starting with "List-" it is
1827 treated as non-personal.
1829 (4) The test for "circular" in the Subject: header line has been removed
1830 because it now seems ill-conceived.
1832 15. Minor typos in src/EDITME comments corrected.
1834 16. Installed latest exipick from John Jetmore.
1836 17. If headers_add on a router specified a text string that was too long for
1837 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
1838 of string_sprintf() is now avoided.
1840 18. $message_body_size was not set (it was always zero) when running the DATA
1841 ACL and the local_scan() function.
1843 19. For the "mail" command in an Exim filter, no default was being set for
1844 the once_repeat time, causing a random time value to be used if "once" was
1845 specified. (If the value happened to be <= 0, no repeat happened.) The
1846 default is now 0s, meaning "never repeat". The "vacation" command was OK
1847 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
1848 (I found it when inspecting the code).
1850 20. There is now an overall timeout for performing a callout verification. It
1851 defaults to 4 times the callout timeout, which applies to individual SMTP
1852 commands during the callout. The overall timeout applies when there is more
1853 than one host that can be tried. The timeout is checked before trying the
1854 next host. This prevents very long delays if there are a large number of
1855 hosts and all are timing out (e.g. when the network connections are timing
1856 out). The value of the overall timeout can be changed by specifying an
1857 additional sub-option for "callout", called "maxwait". For example:
1859 verify = sender/callout=5s,maxwait=20s
1861 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
1862 to the end before writing, but this should make it even safer).
1864 22. Exim was forgetting that it had advertised PIPELINING for the second and
1865 subsequent messages on an SMTP connection. It was also not resetting its
1866 memory on STARTTLS and an internal HELO.
1868 23. When Exim logs an SMTP synchronization error within a session, it now
1869 records whether PIPELINING has been advertised or not.
1871 24. Added 3 instances of "(long int)" casts to time_t variables that were being
1872 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
1873 rather than long int.
1875 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
1877 26. Added the never_mail option to autoreply.
1883 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
1884 crash if the getsockname() call failed; this can happen if a connection is
1885 closed very soon after it is established. The problem was simply in the
1886 order in which certain operations were done, causing Exim to try to write
1887 to the SMTP stream before it had set up the file descriptor. The bug has
1888 been fixed by making things happen in the correct order.
1894 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
1895 before the connection was closed, thus losing the rejection response.
1897 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
1898 some early Solaris releases, but causes trouble in current releases where
1899 socklen_t is defined.
1901 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
1904 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
1907 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
1908 that releasing the top of it at the end releases what was used for sub-
1909 expansions (unless the block got too big). However, discard this block if
1910 the first thing is a variable or header, so that we can use its block when
1911 it is dynamic (useful for very large $message_headers, for example).
1913 6. Lookups now cache *every* query, not just the most recent. A new, separate
1914 store pool is used for this. It can be recovered when all lookup caches are
1915 flushed. Lookups now release memory at the end of their result strings.
1916 This has involved some general refactoring of the lookup sources.
1918 7. Some code has been added to the store_xxx() functions to reduce the amount
1919 of flapping under certain conditions.
1921 8. log_incoming_interface used to affect only the <= reception log lines. Now
1922 it causes the local interface and port to be added to several more SMTP log
1923 lines, for example "SMTP connection from", and rejection lines.
1925 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
1927 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
1929 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
1930 could be overwritten at the end of the current message (or the start of a
1931 new message if it was set in a HELO ACL). The value is now preserved for
1932 the duration of the SMTP connection.
1934 12. If a transport had a headers_rewrite setting, and a matching header line
1935 contained an unqualified address, that address was qualified, even if it
1936 did not match any rewriting rules. The underlying bug was that the values
1937 of the flags that permit the existence of unqualified sender and recipient
1938 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
1939 non-local messages, and by -bnq for local messages) were not being
1940 preserved with the message after it was received.
1942 13. When Exim was logging an SMTP synchronization error, it could sometimes log
1943 "next input=" as part of the text comprising the host identity instead of
1944 the correct text. The code was using the same buffer for two different
1945 strings. However, depending on which order the printing function evaluated
1946 its arguments, the bug did not always show up. Under Linux, for example, my
1947 test suite worked just fine.
1949 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
1950 doesn't work with some older versions of Perl. It has been changed to "my",
1951 which in any case is probably the better facility to use.
1953 15. A really picky compiler found some instances of statements for creating
1954 error messages that either had too many or two few arguments for the format
1957 16. The size of the buffer for calls to the DNS resolver has been increased
1958 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
1959 for addresses that have a lot of PTR records. This alleviates a problem; it
1960 does not fully solve it.
1962 17. A dnsdb lookup for PTR records that receives more data than will fit in the
1963 buffer now truncates the list and logs the incident, which is the same
1964 action as happens when Exim is looking up a host name and its aliases.
1965 Previously in this situation something unpredictable would happen;
1966 sometimes it was "internal error: store_reset failed".
1968 18. If a server dropped the connection unexpectedly when an Exim client was
1969 using GnuTLS and trying to read a response, the client delivery process
1970 crashed while trying to generate an error log message.
1972 19. If a "warn" verb in an ACL added multiple headers to a message in a single
1973 string, for example:
1975 warn message = H1: something\nH2: something
1977 the text was added as a single header line from Exim's point of view
1978 though it ended up OK in the delivered message. However, searching for the
1979 second and subsequent header lines using $h_h2: did not work. This has been
1980 fixed. Similarly, if a system filter added multiple headers in this way,
1981 the routers could not see them.
1983 20. Expanded the error message when iplsearch is called with an invalid key to
1984 suggest using net-iplsearch in a host list.
1986 21. When running tests using -bh, any delays imposed by "delay" modifiers in
1987 ACLs are no longer actually imposed (and a message to that effect is
1990 22. If a "gecos" field in a passwd entry contained escaped characters, in
1991 particular, if it contained a \" sequence, Exim got it wrong when building
1992 a From: or a Sender: header from that name. A second bug also caused
1993 incorrect handling when an unquoted " was present following a character
1994 that needed quoting.
1996 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
1997 was not being matched caselessly.
1999 24. Arranged for all hyphens in the exim.8 source to be escaped with
2002 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
2003 properly. Recipient callout cache records were still being keyed to include
2004 the sender, even when use_sender was set false. This led to far more
2005 callouts that were necessary. The sender is no longer included in the key
2006 when use_sender is false.
2008 26. Added "control = submission" modifier to ACLs.
2010 27. Added the ${base62d: operator to decode base 62 numbers.
2012 28. dnsdb lookups can now access SRV records.
2014 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
2015 the configuration file.
2017 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
2018 (-v) mode. This makes the output for a verbose queue run more intelligible.
2020 31. Added a use_postmaster feature to recipient callouts.
2022 32. Added the $body_zerocount variable, containing the number of binary zero
2023 bytes in the message body.
2025 33. The time of last modification of the "new" subdirectory is now used as the
2026 "mailbox time last read" when there is a quota error for a maildir
2029 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
2031 35. Added +ignore_unknown as a special item in host lists.
2033 36. Code for decoding IPv6 addresses in host lists is now included, even if
2034 IPv6 support is not being compiled. This fixes a bug in which an IPv6
2035 address was recognized as an IP address, but was then not correctly decoded
2036 into binary, causing unexpected and incorrect effects when compared with
2043 1. Very minor rewording of debugging text in manualroute to say "list of
2044 hosts" instead of "hostlist".
2046 2. If verify=header_syntax was set, and a header line with an unqualified
2047 address (no domain) and a large number of spaces between the end of the
2048 name and the colon was received, the reception process suffered a buffer
2049 overflow, and (when I tested it) crashed. This was caused by some obsolete
2050 code that should have been removed. The fix is to remove it!
2052 3. When running in the test harness, delay a bit after writing a bounce
2053 message to get a bit more predictability in the log output.
2055 4. Added a call to search_tidyup() just before forking a reception process. In
2056 theory, someone could use a lookup in the expansion of smtp_accept_max_
2057 per_host which, without the tidyup, could leave open a database connection.
2059 5. Added the variables $recipient_data and $sender_data which get set from a
2060 lookup success in an ACL "recipients" or "senders" condition, or a router
2061 "senders" option, similar to $domain_data and $local_part_data.
2063 6. Moved the writing of debug_print from before to after the "senders" test
2066 7. Change 4.31/66 (moving the time when the Received: is generated) caused
2067 problems for message scanning, either using a data ACL, or using
2068 local_scan() because the Received: header was not generated till after they
2069 were called (in order to set the time as the time of reception completion).
2070 I have revised the way this works. The header is now generated after the
2071 body is received, but before the ACL or local_scan() are called. After they
2072 are run, the timestamp in the header is updated.
2078 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
2079 before starting a queue runner without re-exec. This happened only when
2080 deliver_drop_privilege was set or when the Exim user was set to root. The
2081 effect of the bug was that timeouts during subsequent deliveries caused
2082 crashes instead of being properly handled. The handler is now left at its
2083 default (and expected) setting.
2085 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
2086 message, again when deliver_drop_privilege is set or Exim is run as root.
2087 The bug described in (1) was not present in this case, but the tidying up
2088 of the other signals was missing. I have made the two cases consistent.
2090 3. The ignore_target_hosts setting on a manualroute router was being ignored
2091 for hosts that were looked up using the /MX notation.
2093 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
2096 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
2097 operated on the sender address. After changing the $sender_address to <>
2098 for the sender address verify, Exim was re-instated it as the original
2099 (before rewriting) address, but remembering that it had rewritten it, so it
2100 wasn't rewriting it again. This bug also had the effect of breaking the
2101 sender address verification caching when the sender address was rewritten.
2103 6. The ignore_target_hosts option was being ignored by the ipliteral router.
2104 This has been changed so that if the ip literal address matches
2105 ignore_target_hosts, the router declines.
2107 7. Added expansion conditions match_domain, match_address, and match_local_
2108 part (NOT match_host).
2110 8. The placeholder for the Received: header didn't have a length field set.
2112 9. Added code to Exim itself and to exim_lock to test for a specific race
2113 condition that could lead to file corruption when using MBX delivery. The
2114 issue is with the lockfile that is created in /tmp. If this file is removed
2115 after a process has opened it but before that process has acquired a lock,
2116 there is the potential for a second process to recreate the file and also
2117 acquire a lock. This could lead to two Exim processes writing to the file
2118 at the same time. The added code performs the same test as UW imapd; it
2119 checks after acquiring the lock that its file descriptor still refers to
2120 the same named file.
2122 10. The buffer for building added header lines was of fixed size, 8192 bytes.
2123 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
2126 11. Added the smtp_active_hostname option. If used, this will typically be made
2127 to depend on the incoming interface address. Because $interface_address is
2128 not set up until the daemon has forked a reception process, error responses
2129 that can happen earlier (such as "too many connections") no longer contain
2132 12. If an expansion in a condition on a "warn" statement fails because a lookup
2133 defers, the "warn" statement is abandoned, and the next ACL statement is
2134 processed. Previously this caused the whole ACL to be aborted.
2136 13. Added the iplsearch lookup type.
2138 14. Added ident_timeout as a log selector.
2140 15. Added tls_certificate_verified as a log selector.
2142 16. Added a global option tls_require_ciphers (compare the smtp transport
2143 option of the same name). This controls incoming TLS connections.
2145 17. I finally figured out how to make tls_require_ciphers do a similar thing
2146 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
2147 before starting the TLS session.
2149 18. Tabs are now shown as \t in -bP output.
2151 19. If the log selector return_path_on_delivery was set, Exim crashed when
2152 bouncing a message because it had too many Received: header lines.
2154 20. If two routers both had headers_remove settings, and the first one included
2155 a superfluous trailing colon, the final name in the first list and the
2156 first name in the second list were incorrectly joined into one item (with a
2157 colon in the middle).
2163 1. Added -C and -D options to the exinext utility, mainly to make it easier
2164 to include in the automated testing, but these could be helpful when
2165 multiple configurations are in use.
2167 2. The exinext utility was not formatting the output nicely when there was
2168 an alternate port involved in the retry record key, nor when there was a
2169 message id as well (for retries that were specific to a specific message
2170 and a specific host). It was also confused by IPv6 addresses, because of
2171 the additional colons they contain. I have fixed the IPv4 problem, and
2172 patched it up to do a reasonable job for IPv6.
2174 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
2175 delivery, the log line now contains "pipelined" if PIPELINING was used.
2177 4. An SMTP transport process used to panic and die if the bind() call to set
2178 an explicit outgoing interface failed. This has been changed; it is now
2179 treated in the same way as a connect() failure.
2181 5. A reference to $sender_host_name in the part of a conditional expansion
2182 that was being skipped was still causing a DNS lookup. This no longer
2185 6. The def: expansion condition was not recognizing references to header lines
2186 that used bh_ and bheader_.
2188 7. Added the _cache feature to named lists.
2190 8. The code for checking quota_filecount in the appendfile transport was
2191 allowing one more file than it should have been.
2193 9. For compatibility with Sendmail, the command line option
2201 and sets the incoming protocol and host name (for trusted callers). The
2202 host name and its colon can be omitted when only the protocol is to be set.
2203 Note the Exim already has two private options, -pd and -ps, that refer to
2204 embedded Perl. It is therefore impossible to set a protocol value of "d" or
2205 "s", but I don't think that's a major issue.
2207 10. A number of refactoring changes to the code, none of which should affect
2210 (a) The number of logging options was getting close to filling up the
2211 32-bit word that was used as a bit map. I have split them into two classes:
2212 those that are passed in the argument to log_write(), and those that are
2213 only ever tested independently outside of that function. These are now in
2214 separate 32-bit words, so there is plenty of room for expansion again.
2215 There is no change in the user interface or the logging behaviour.
2217 (b) When building, for example, log lines, the code previously used a
2218 macro that called string_cat() twice, in order to add two strings. This is
2219 not really sufficiently general. Furthermore, there was one instance where
2220 it was actually wrong because one of the argument was used twice, and in
2221 one call a function was used. (As it happened, calling the function twice
2222 did not affect the overall behaviour.) The macro has been replaced by a
2223 function that can join an arbitrary number of extra strings onto a growing
2226 (c) The code for expansion conditions now uses a table and a binary chop
2227 instead of a serial search (which was left over from when there were very
2228 few conditions). Also, it now recognizes conditions like "pam" even when
2229 the relevant support is not compiled in: a suitably worded error message is
2230 given if an attempt is made to use such a condition.
2232 11. Added ${time_interval:xxxxx}.
2234 12. A bug was causing one of the ddress fields not to be passed back correctly
2235 from remote delivery subprocesses. The field in question was not being
2236 subsequently used, so this caused to problems in practice.
2238 13. Added new log selectors queue_time and deliver_time.
2240 14. Might have fixed a bug in maildirsizefile handling that threw up
2241 "unexpected character" debug warnings, and recalculated the data
2242 unnecessarily. In any case, I expanded the warning message to give more
2245 15. Added the message "Restricted characters in address" to the statements in
2246 the default ACL that block characters like @ and % in local parts.
2248 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
2249 Three changes have been made:
2251 (a) There was a serious bug; a negative response to MAIL caused the whole
2252 recipient domain to be cached as invalid, thereby blocking all messages
2253 to all local parts at the same domain, from all senders. This bug has
2254 been fixed. The domain is no longer cached after a negative response to
2255 MAIL if the sender used is not empty.
2257 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
2260 (c) A new callout option, "use_sender" has been added for people who want
2261 the modified behaviour.
2267 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
2270 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
2271 indeed breaks things for older releases.
2273 3. Added additional logging to the case where there is a problem reading data
2274 from a filter that is running in a subprocess using a pipe, in order to
2275 try to track down a specific problem.
2277 4. Testing facility fudge: when running in the test harness and attempting
2278 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
2279 getting "No route to host". Convert this to a timeout.
2281 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
2284 6. Some OS don't have socklen_t but use size_t instead. This affects the
2285 fifth argument of getsockopt() amongst other things. This is now
2286 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
2287 can be set for individual OS. I have set it for SunOS5, OSF1, and
2288 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
2289 some earlier ones do not.
2291 7. Change 4.30/15 was not doing the test caselessly.
2293 8. The standard form for an IPv6 address literal was being rejected by address
2294 parsing in, for example, MAIL and RCPT commands. An example of this kind of
2295 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
2296 this, as well as the form without the "IPv6" on the front (but only when
2297 address literals are enabled, of course).
2299 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
2301 10. Exim crashed if a message with an empty sender address specified by -f
2302 encountered a router with an errors_to setting. This could be provoked only
2303 by a command such as
2307 where an empty string was supplied; "<>" did not hit this bug.
2309 11. Installed PCRE release 4.5.
2311 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
2312 remained set. It is now erased.
2314 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
2315 times from message ids (which are base 36 rather than the normal 62).
2317 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
2318 were being counted as actual protocol errors, and logged if the log
2319 selector +smtp_protocol_error was set. One cannot be perfect in this test,
2320 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
2321 and DATA following a set of rejected RCPTs do not count as protocol errors.
2322 In other words, Exim assumes they were pipelined, though this may not
2323 actually be the case. Of course, in all cases the client gets an
2324 appropriate error code.
2326 15. If a lookup fails in an ACL condition, a message about the failure may
2327 be available; it is used if testing the ACL cannot continue, because most
2328 such messages specify what the cause of the deferral is. However, some
2329 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
2330 that caused an old message to be retained and used if a later statement
2331 caused a defer, replacing the real cause of the deferral.
2333 16. If an IP address had so many PTR records that the DNS lookup buffer
2334 was not large enough to hold them, Exim could crash while trying to process
2335 the truncated data. It now detects and logs this case.
2337 17. Further to 4.21/58, another change has been made: if (and only if) the
2338 first line of a message (the first header line) ends with CRLF, a bare LF
2339 in a subsequent header line has a space inserted after it, so as not to
2340 terminate the header.
2342 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
2343 unnecessarily, used atoi() instead of strtol(), and didn't check the
2344 termination when getting file sizes from file names by regex.
2346 19. Completely re-implemented the support for maildirsize files, in the light
2347 of a number of problems with the previous contributed implementation
2348 (4.30/29). In particular:
2350 . If the quota is zero, the maildirsize file is maintained, but no quota is
2353 . If the maildir directory does not exist, it is created before any attempt
2354 to write a maildirsize file.
2356 . The quota value in the file is just a cache; if the quota is changed in
2357 the transport, the new value overrides.
2359 . A regular expression is available for excluding directories from the
2362 20. The autoreply transport checks the characters in options that define the
2363 message's headers; it allows continued headers, but it was checking with
2364 isspace() after an embedded newline instead of explicitly looking for a
2367 21. If all the "regular" hosts to which an address was routed had passed their
2368 expiry times, and had not reached their retry times, the address was
2369 bounced, even if fallback hosts were defined. Now Exim should go on to try
2372 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
2373 equivalent code in the SMTP transport. Some hosts send humungous responses
2374 to HELO/EHLO, more than 1024 it seems.
2376 23. Refactoring: code in filter.c used (void *) for "any old type" but this
2377 gives compiler warnings in some environments. I've now done it "properly",
2380 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
2381 (because of problems with the built-in one) was declared to return uschar *
2382 instead of char *, causing compiler failure.
2384 25. Fixed a file descriptor leak when processing alias/forward files.
2386 26. Fixed a minor format string issue in dbfn.c.
2388 27. Typo in exim.c: ("dmbnz" for "dbmnz").
2390 28. If a filter file refered to $h_xxx or $message_headers, and the headers
2391 contained RFC 2047 "words", Exim's memory could, under certain conditions,
2394 29. When a sender address is verified, it is cached, to save repeating the test
2395 when there is more than one recipient in a message. However, when the
2396 verification involves a callout, it is possible for different callout
2397 options to be set for different recipients. It is too complicated to keep
2398 track of this in the cache, so now Exim always runs a verification when a
2399 callout is required, relying on the callout cache for the optimization.
2400 The overhead is duplication of the address routing, but this should not be
2403 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
2404 to be verified with callout=postmaster, and the main callout worked but the
2405 postmaster check failed, the verification correctly failed. However, if a
2406 subsequent RCPT command asked for sender verification *without* the
2407 postmaster check, incorrect caching caused this verification also to fail,
2410 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
2411 it was not caching the DNS options (qualify_single, search_parents) that
2412 were used when the lookup failed. A subsequent lookup with different
2413 options therefore always gave the same answer, though there were cases
2414 where it should not have. (Example: a "domains = !$mx_any" option on a
2415 dnslookup router: the "domains" option is always processed without any
2416 widening, but the router might have qualify_single set.) Now Exim uses the
2417 cached value only when the same options are set.
2419 32. Added John Jetmore's "exipick" utility to the distribution.
2421 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
2422 than a timeout (e.g. a certificate is required, and is not provided), an
2423 Exim server now closes the connection immediately. Previously it waited for
2424 the client to close - but if the client is SSL, it seems that they each
2425 wait for each other, leading to a delay before one of them times out.
2427 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
2428 maintained 0.8.x compatibility because I don't think many are using it, and
2429 it is clearly obsolete.
2431 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
2434 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
2435 client certificate was expired. A simple patch fixes this, though I don't
2436 understand the full logic of why the verify callback is called multiple
2439 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
2440 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
2441 which causes problems with some clients (such as the Certicom SSL Plus
2442 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
2443 disables the coutermeasure allowing Eudora to connect."
2445 38. Exim was not checking that a write() to a log file succeeded. This could
2446 lead to Bad Things if a log got too big, in particular if it hit a file
2447 size limit. Exim now panics and dies if it cannot write to a log file, just
2448 as it does if it cannot open a log file.
2450 39. Modified OS/Makefile-Linux so that it now contains
2452 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
2454 The two -D definitions ensure that Exim is compiled with large file
2455 support, which makes it possible to handle log files that are bigger than
2458 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
2459 instance) a domain was checked against a named list that involved a lookup,
2460 causing $domain_data to be set, then another domain was checked against the
2461 same list, then the first domain was re-checked, the value of $domain_data
2462 after the final check could be wrong. In particular, if the second check
2463 failed, it could be set empty. This bug probably also applied to
2466 41. The strip_trailing_dot option was not being applied to the address given
2467 with the -f command-line option.
2469 42. The code for reading a message's header from the spool was incrementing
2470 $received_count, but never initializing it. This meant that the value was
2471 incorrect (doubled) while delivering a message in the same process in which
2472 it was received. In the most common configuration of Exim, this never
2473 happens - a fresh exec is done - but it can happen when
2474 deliver_drop_privilege is set.
2476 43. When Exim logs an SMTP synchronization error - client data sent too soon -
2477 it now includes up to 150 characters of the unexpected data in the log
2480 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
2481 and building data strings. The size of both of these buffers was 10 000
2482 bytes - far larger than anybody would *ever* want, thought I. Needless to
2483 say, somebody hit the limit. I have increased the maximum line length to
2484 20 000 and the maximum data length of concatenated lines to 100 000. I have
2485 also fixed two bugs, because there was no checking on these buffers. Tsk,
2486 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
2487 buffer is too small.
2489 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
2490 lsearch lookups. Now it does.
2492 46. When parsing a route_list item in a manualroute router, a fixed-length
2493 buffer was used for the list of hosts. I made this 1024 bytes long,
2494 thinking that nobody would ever have a list of hosts that long. Wrong.
2495 Somebody had a whole pile of complicated expansion conditions, and the
2496 string was silently truncated, leading to an expansion error. It turns out
2497 that it is easier to change to an unlimited length (owing to other changes
2498 that have happened since this code was originally written) than to build
2499 structure for giving a limitation error. The length of the item that
2500 expands into the list of hosts is now unlimited.
2502 47. The lsearch lookup could not handle data where the length of text line was
2503 more than 4095 characters. Such lines were truncated, leading to shortened
2504 data being returned. It should now handle lines of any length.
2506 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
2507 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
2510 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
2511 replacing strings such as BIN_DIRECTORY in a source file: the replacement
2512 no longer happens in comment lines. A list of replacements is now placed
2513 at the head of all of the source files, except those whose only change is
2514 to replace PERL_COMMAND in the very first #! line.
2516 50. Replaced the slow insertion sort in queue.c, for sorting the list of
2517 messages on the queue, with a bottom-up merge sort, using code contributed
2518 by Michael Haardt. This should make operations like -bp somewhat faster on
2519 large queues. It won't affect queue runners, except when queue_run_in_order
2522 51. Installed eximstats 1.31 in the distribution.
2524 52. Added support for SRV lookups to the dnslookup router.
2526 53. If an ACL referred to $message_body or $message_body_end, the value was not
2527 reset for any messages that followed in the same SMTP session.
2529 54. The store-handling optimization for building very long strings was not
2530 differentiating between the different store pools. I don't think this
2531 actually made any difference in practice, but I've tidied it.
2533 55. While running the routers to verify a sender address, $sender_address
2534 was still set to the sender address. This is wrong, because when routing to
2535 send a bounce to the sender, it would be empty. Therefore, I have changed
2536 it so that, while verifying a sender address, $sender_address is set to <>.
2537 (There is no change to what happens when verifying a recipient address.)
2539 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
2540 target A or AAAA records (if not already returned) without resetting the
2541 qualify_single or search_parents options of the DNS resolver. These are
2542 inappropriate in this case because the targets of MX and SRV records must
2543 be FQDNs. A broken DNS record could cause trouble if it happened to have a
2544 target that, when qualified, matched something in the local domain. These
2545 two options are now turned off when doing these lookups.
2547 57. It seems that at least some releases of Reiserfs (which does not have the
2548 concept of a fixed number of inodes) returns zero and not -1 for the
2549 number of available inodes. This interacted badly with check_spool_inodes,
2550 which assumed that -1 was the "no such thing" setting. What I have done is
2551 to check that the total number of inodes is greater than zero before doing
2552 the test of how many are available.
2554 58. When a "warn" ACL statement has a log_message modifier, the message is
2555 remembered, and not repeated. This is to avoid a lot of repetition when a
2556 message has many recipients that cause the same warning to be written.
2557 Howewer, Exim was preserving the list of already written lines for an
2558 entire SMTP session, which doesn't seem right. The memory is now reset if a
2559 new message is started.
2561 59. The "rewrite" debugging flag was not showing the result of rewriting in the
2562 debugging output unless log_rewrite was also set.
2564 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
2565 of (int)(handle) when we know that handle contains (void *)(-1).
2567 61. The Exim daemon panic-logs an error return when it closes the incoming
2568 connection. However "connection reset by peer" seems to be common, and
2569 isn't really an error worthy of noting specially, so that particular error
2572 62. When Exim is trying to find all the local interfaces, it used to panic and
2573 die if the ioctl to get the interface flags failed. However, it seems that
2574 on at least one OS (Solaris 9) it is possible to have an interface that is
2575 included in the list of interfaces, but for which you get a failure error
2576 for this call. This happens when the interface is not "plumbed" into a
2577 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
2578 failure of the "get flags" call assumes that the interface is down.
2580 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
2581 makes life easier for people who are doing arithmetic on fields extracted
2582 from dates, where you often get leading zeros that should not be
2583 interpreted as octal.
2585 64. Added qualify_domain to the redirect router, to override the global
2588 65. If a pathologically long header line contained very many addresses (the
2589 report of this problem mentioned 10 000) and each of them was rewritten,
2590 Exim could use up a very large amount of memory. (It kept on making new
2591 copies of the header line as it rewrote, and never released the old ones.)
2592 At the expense of a bit more processing, the header rewriting function has
2593 been changed so that it no longer eats memory in this way.
2595 66. The generation of the Received: header has been moved from the time that a
2596 message starts to be received, to the time that it finishes. The timestamp
2597 in the Received: header should now be very close to that of the <= log
2598 line. There are two side-effects of this change:
2600 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
2601 logged header lines no longer include the local Received: line, because
2602 it has not yet been created. The same applies to a copy of the message
2603 that is returned to a non-SMTP sender when a message is rejected.
2605 (b) When a filter file is tested using -bf, no additional Received: header
2606 is added to the test message. After some thought, I decided that this
2609 This change does not affect the value of $received_for. It is still set
2610 after address rewriting, but before local_scan() is called.
2612 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
2614 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
2615 gave an unhelpful panic error message, and a defer error. I have managed to
2616 change this behaviour so that it now rejects any supplied certificate,
2617 which seems right, as the list of acceptable certificates is empty.
2619 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
2620 gave an unhelpful defer error. I have not managed to make this reject any
2621 supplied certificates, but the error message it gives is "no certificate
2622 supplied", which is not helpful.
2624 70. exigrep's output now also includes lines that are not associated with any
2625 message, but which match the given pattern. Implemented by a patch from
2626 Martin Sluka, which also tidied up the Perl a bit.
2628 71. Recipient callout verification, like sender verification, was using <> in
2629 the MAIL FROM command. This isn't really the right thing, since the actual
2630 sender may affect whether the remote host accepts the recipient or not. I
2631 have changed it to use the actual sender in the callout; this means that
2632 the cache record is now keyed on a recipient/sender pair, not just the
2633 recipient address. There doesn't seem to be a real danger of callout loops,
2634 since a callout by the remote host to check the sender would use <>.
2635 [SEE ABOVE: changed after hitting problems.]
2637 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
2638 temporary errors. However, in the case of such a code being given after
2639 the end of a data transmission (i.e. after ".") Exim was failing to write
2640 a retry record for the message. (Yes, there was some broken host that was
2641 actually sending 8xx at this point.)
2643 73. An unknown lookup type in a host list could cause Exim to panic-die when
2644 the list was checked. (An example that provoked this was putting <; in the
2645 middle of a list instead of at the start.) If this happened during a DATA
2646 ACL check, a -D file could be left lying around. This kind of configuration
2647 error no longer causes Exim to die; instead it causes a defer errror. The
2648 incident is still logged to the main and panic logs.
2650 74. Buglet left over from Exim 3 conversion. The message "too many messages
2651 in one connection" was written to the rejectlog but not the mainlog, except
2652 when address rewriting (yes!) was being logged.
2654 75. Added write_rejectlog option.
2656 76. When a system filter was run not as root (that is, when system_filter_user
2657 was set), the values of the $n variables were not being returned to the
2658 main process; thus, they were not subsequently available in the $sn
2661 77. Added +return_path_on_delivery log selector.
2663 78. A connection timeout was being treated differently from recipients deferred
2664 when testing hosts_max_try with a message that was older than the host's
2665 retry timeout. (The host should not be counted, thus allowing all hosts to
2666 be tried at least once before bouncing.) This may have been the cause of an
2667 occasionally reported bug whereby a message would remain on the queue
2668 longer than the retry timeout, but would be bounced if a delivery was
2669 forced. I say "may" because I never totally pinned down the problem;
2670 setting up timeout/retry tests is difficult. See also the next item.
2672 79. The ultimate address timeout was not being applied to errors that involved
2673 a combination of host plus message (for example, a timeout on a MAIL
2674 command). When an address resolved to a number of possible hosts, and they
2675 were not all tried for each delivery (e.g. because of hosts_max_try), a
2676 message could remain on the queue longer than the retry timeout.
2678 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
2681 81. Fixed an obscure SMTP outgoing bug which required at least the following
2682 conditions: (a) there was another message waiting for the same server;
2683 (b) the server returned 5xx to all RCPT commands in the first message so
2684 that the message was not completed; (c) the server dropped the connection
2685 or gave a negative response to the RSET that Exim sends to abort the
2686 transaction. The observed case was a dropped connection after DATA that had
2687 been sent in pipelining mode. That is, the server had advertised PIPELINING
2688 but was not implementing it correctly. The effect of the bug was incorrect
2689 behaviour, such as trying another host, and this could lead to a crash.
2695 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
2696 and daemon.c were passed as pointers to ints; they should have been
2697 pointers to socklen_t variables (which are typically unsigned ints).
2699 2. Some signed/unsigned type warnings in the os.c file for Linux have been
2702 3. Fixed a really odd bug that affected only the testing scheme; patching a
2703 certain fixed string in the binary changed the value of another string that
2704 happened to be identical to the end of the original first string.
2706 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
2707 name", it returns that address as the IP address. On some operating
2708 systems (e.g. Solaris), it also passes back the IP address string as the
2709 "host name". However, on others (e.g. Linux), it passes back an empty
2710 string. Exim wasn't checking for this, and was changing the host name to an
2711 empty string, assuming it had been canonicized.
2713 5. Although rare, it is permitted to have more than one PTR record for a given
2714 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
2715 all the names associated with an address, because they do in Solaris.
2716 However, it seems that they do not in Linux for data that comes from the
2717 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
2718 I found this out when I moved to a new Linux workstation and tried to run
2719 the Exim test suite.
2721 To get round this problem I have changed the code so that it now does its
2722 own call to the DNS to look up PTR records when searching for a host name.
2723 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
2724 addresses that are only in /etc/hosts are still found.
2726 This behaviour is, however, controlled by an option called host_lookup_
2727 order, which defaults to "bydns:byaddr". If people want to use the other
2728 order, or indeed, just use one or the other means of lookup, they can
2729 specify it in this variable.
2731 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
2732 some operating systems, this comes back from gethostbyaddr() as an empty
2733 string, and this is what Exim used to test for. However, it seems that in
2734 other systems, "." is yielded. Exim now tests for this case too.
2736 7. The values of check_spool_space and check_log_space are now held internally
2737 as a number of kilobytes instead of an absolute number of bytes. If a
2738 numbers is specified without 'K' or 'M', it is rounded up to the nearest
2739 kilobyte. This means that much larger values can be stored.
2741 8. Exim monitor: an attempt to get the action menu when not actually pointing
2742 at a message produces an empty menu entitled "No message selected". This
2743 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
2744 no entries in it ("Shell widget menu has zero width and/or height"). So I
2745 have added a single, blank menu entry in this case.
2747 9. Added ${quote_local_part.
2749 10. MIME decoding is now applied to the contents of Subject: header lines when
2752 11. Now that a reference to $sender_host_address automatically causes a reverse
2753 lookup to occur if necessary (4.13/18), there is no need to arrange for a
2754 host lookup before query-style lookups in lists that might use this
2755 variable. This has therefore been abolished, and the "net-" prefix is no
2756 longer necessary for query-style lookups.
2758 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
2759 have been a typo for LFLAGS, so it has been changed.
2761 13. The install script calls Exim with "-C /dev/null" in order to find the
2762 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
2763 to be output. Howeve, since Exim outputs its version number before the
2764 error, it didn't break the script. It just looked ugly. I fixed this by
2765 always allowing "-C /dev/null" if the caller is root.
2767 14. Ignore overlarge ACL variable number when reading spool file - insurance
2768 against a later release with more variables having written the file.
2770 15. The standard form for an IPv6 address literal was being rejected by EHLO.
2771 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
2772 this, as well as the form without the "IPv6" on the front.
2774 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
2775 OS/Makefile-Darwin file.
2777 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
2778 only to LDAP libraries that do not have LDAP_OPT_DEREF.
2780 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
2781 variable. However, some OS use ints for inodes. I've added cast to long int
2782 to get rid of the compiler warning.
2784 19. I had forgotten to lock out "/../" in configuration file names when
2785 ALT_CONFIG_PREFIX was set.
2787 20. Routers used for verification do not need to specify transports. However,
2788 if such a router generated a host list, and callout was configured, Exim
2789 crashed, because it could not find a port number from the (non-existent)
2790 transport. It now assumes port 25 in this circumstance.
2792 21. Added the -t option to exigrep.
2794 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
2795 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
2796 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
2797 them is set, LOOKUP_LSEARCH is forced.
2799 23. "exim -bV" now outputs a list of lookups that are included in the binary.
2801 24. Added sender and host information to the "rejected by local_scan()" log
2802 line; previously there was no indication of these.
2804 25. Added .include_if_exists.
2806 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
2807 Linux. It turns out that not all file systems support this. Apparently some
2808 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
2809 do it.) To cope with this, the error EINVAL, which means that sync-ing is
2810 not supported on the file descriptor, is now ignored when Exim is trying to
2811 sync a directory. This applies only to Linux.
2813 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
2815 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
2816 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
2819 29. Added support for maildirsize files from supplied patch (modified a bit).
2821 30. The use of :fail: followed by an empty string could lead Exim to respond to
2822 sender verification failures with (e.g.):
2824 550 Verification failed for <xxx>
2825 550 Sender verify failed
2827 where the first response line was missing the '-' that indicates it is not
2828 the final line of the response.
2830 31. The loop for finding the name of the user that called Exim had a hardwired
2831 limit of 10; it now uses the value of finduser_retries, which is used for
2832 all other user lookups.
2834 32. Added $received_count variable, available in data and not_smtp ACLs, and at
2837 33. Exim was neglecting to zero errno before one call of strtol() when
2838 expanding a string and expecting an integer value. On some systems this
2839 resulted in spurious "integer overflow" errors. Also, it was casting the
2840 result into an int without checking.
2842 34. Testing for a connection timeout using "timeout_connect" in the retry rules
2843 did not work. The code looks as if it has *never* worked, though it appears
2844 to have been documented since at least releast 1.62. I have made it work.
2846 35. The "timeout_DNS" error in retry rules, also documented since at least
2847 1.62, also never worked. As it isn't clear exactly what this means, and
2848 clearly it isn't a major issue, I have abolished the feature by treating it
2849 as "timeout", and writing a warning to the main and panic logs.
2851 36. The display of retry rules for -brt wasn't always showing the error code
2854 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
2855 timeout_connect_A, timeout_connect_MX.
2857 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
2858 to the empty sender.
2860 39. The daemon was not analysing the content of -oX till after it had closed
2861 stderr and disconnected from the controlling terminal. This meant that any
2862 syntax errors were only noted on the panic log, and the return code from
2863 the command was 0. By re-arranging the code a little, I've made the
2864 decoding happen first, so such errors now appear on stderr, and the return
2865 code is 1. However, the actual setting up of the sockets still happens in
2866 the disconnected process, so errors there are still only recorded on the
2869 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
2870 connections (as happens on some IP stacks) was logged at start up time as
2871 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
2872 it from "IPv6 and IPv4", which means that two separate sockets are being
2875 41. The debug output for gethostbyname2() or getipnodebyname() failures now
2876 says whether AF_INET or AF_INET6 was passed as an argument.
2878 42. Exiwhat output was messed up when time zones were included in log
2881 43. Exiwhat now gives more information about the daemon's listening ports,
2882 and whether -tls-on-connect was used.
2884 44. The "port" option of the smtp transport is now expanded.
2886 45. A "message" modifier in a "warn" statement in a non-message ACL was being
2887 silently ignored. Now an error message is written to the main and panic
2890 46. There's a new ACL modifier called "logwrite" which writes to a log file
2891 as soon as it is encountered.
2893 47. Added $local_user_uid and $local_user_gid at routing time.
2895 48. Exim crashed when trying to verify a sender address that was being
2898 49. Exim was recognizing only a space character after ".include". It now also
2899 recognizes a tab character.
2901 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
2902 extracting the relevant information from the specification. The man page no
2903 longer contains scrambled data for the -d option, and I've added a section
2904 at the front about calling Exim under different names.
2906 51. Added "extra_headers" argument to the "mail" command in filter files.
2908 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
2911 53. Installed eximstats 1.29.
2913 54. Added transport_filter_timeout as a generic transport option.
2915 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
2916 Cc: header lines. This was required by RFC 822, but it not required by RFC
2919 56. Exim used to add From:, Date:, and Message-Id: header lines to any
2920 incoming messages that did not have them. Now it does so only if the
2921 message originates locally, that is, if there is no associated remote host
2922 address. When Resent- header lines are present, this applies to the Resent-
2923 lines rather than the non-Resent- lines.
2925 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
2926 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
2928 58. Messages for configuration errors now include the name of the main
2929 configuration file - useful now that there may be more than one file in a
2930 list (.included file names were always shown).
2932 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
2933 for those rare installations that do not start the daemon as root or run it
2934 setuid root. I've cut out the call to initgroups() if the daemon is not
2937 60. The Exim user and group can now be bound into the binary as text strings
2938 that are looked up at the start of Exim's processing.
2940 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
2942 62. Added $mailstore_basename variable.
2944 63. Installed patch to sieve.c from Michael Haardt.
2946 64. When Exim failed to open the panic log after failing to open the main log,
2947 the original message it was trying to log was written to stderr and debug
2948 output, but if they were not available (the usual case in production), it
2949 was lost. Now it is written to syslog before the two lines that record the
2950 failures to open the logs.
2952 65. Users' Exim filters run in subprocesses under the user's uid. It is
2953 possible for a "deliver" command or an alias in a "personal" command to
2954 provoke an address rewrite. If logging of address rewriting is configured,
2955 this fails because the process is not running as root or exim. There may be
2956 a better way of dealing with this, but for the moment (because 4.30 needs
2957 to be released), I have disabled address rewrite logging when running a
2958 filter in a non-root, non-exim process.
2964 1. The buildconfig auxiliary program wasn't quoting the value set for
2965 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
2966 not defined. This bug was masked in 4.22 by the effect that was fixed in
2969 2. Some messages that were rejected after a message id was allocated were
2970 shown as "incomplete" by exigrep. It no longer does this for messages that
2971 are rejected by local_scan() or the DATA or non-SMTP ACLs.
2973 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
2974 have allow_domain_literals set, the ID did not get logged in the <= line.
2975 Domain literals are now always recognized in Message-ID: header lines.
2977 4. The first argument for a ${extract expansion item is the key name or field
2978 number. Leading and trailing spaces in this item were not being ignored,
2979 causing some misleading effects.
2981 5. When deliver_drop_privilege was set, single queue runner processes started
2982 manually (i.e. by the command "exim -q") or by the daemon (which uses the
2983 same command in the process it spins off) were not dropping privilege.
2985 6. When the daemon running as "exim" started a queue runner, it always
2986 re-executed Exim in the spun-off process. This is a waste of effort when
2987 deliver_drop_privilege is set. The new process now just calls the
2988 queue-runner function directly.
2994 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
2997 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
2998 ignored. Though the use of -oP was forcing the writing of a pid file, it
2999 was always written to the default place.
3001 3. If the message "no IP address found for host xxxx" is generated during
3002 incoming verification, it is now followed by identification of the incoming
3003 connection (so you can more easily find what provoked it).
3005 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
3007 5. Added some features to "harden" Exim a bit more against certain attacks:
3009 (a) There is now a build-time option called FIXED_NEVER_USERS that can
3010 be put in Local/Makefile. This is like the never_users runtime option,
3011 but it cannot be overridden. The default setting is "root".
3013 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
3014 prefix string with which any file named in a -C command line option
3017 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
3018 is retained for -C and -D only if the caller of Exim is root. Without
3019 it, the exim user may also use -C and -D and retain privilege.
3021 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
3022 command line option is disabled.
3024 6. Macro names set by the -D option must start with an upper case letter, just
3025 like macro names defined in the configuration file.
3027 7. Added "dereference=" facility to LDAP.
3029 8. Two instances of the typo "uknown" in the source files are fixed.
3031 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
3032 the Configure-Makefile script screwed up while processing it.
3034 10. Incorporated PCRE 4.4.
3036 11. The SMTP synchronization check was not operating right at the start of an
3037 SMTP session. For example, it could not catch a HELO sent before the client
3038 waited for the greeting. There is now a check for outstanding input at the
3039 point when the greeting is written. Because of the duplex, asynchronous
3040 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
3041 way, but not yet received, when the check is performed.
3043 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
3044 on TCP/IP sockets, because this apparently causes some broken clients to
3047 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
3048 unchanged) from the Cygwin maintainer.
3050 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
3051 was supported instead of testing for mbx. Effectively a typo.
3053 15. The spa authenticator server code was not checking that the input it
3054 received was valid base64.
3056 16. The debug output line for the "set" modifier in ACLs was not showing the
3057 name of the variable that was being set.
3059 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
3061 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
3062 lookup on its own. Something else was needed to trigger the lookup. For
3063 example, a match in host_lookup or the need for a host name in a host list.
3064 Now, if $sender_host_name is referenced and the host name has not yet been
3065 looked up, a lookup is performed. If the lookup fails, the variable remains
3066 empty, and $host_lookup_failed is set to "1".
3068 19. Added "eqi" as a case-independent comparison operator.
3070 20. The saslauthd authentication condition could segfault if neither service
3071 nor realm was specified.
3073 21. If an overflowing value such as "2048M" was set for message_size_limit, the
3074 error message that was logged was misleading, and incoming SMTP
3075 connections were dropped. The message is now more accurate, and temporary
3076 errors are given to SMTP connections.
3078 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
3079 (except RSET) with a 421 error, until QUIT is received. However, it was
3080 failing to send a response to QUIT.
3082 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
3083 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
3084 ACL is now run after the helo_try_verify_hosts code.
3086 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
3087 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
3088 case-independent, so other case variants are also recognized). Apparently
3089 some systems use these upper case variants.
3091 25. If more than two messages were waiting for the same host, and a transport
3092 filter was specified for the transport, Exim sent two messages over the
3093 same TCP/IP connection, and then failed with "socket operation on non-
3094 socket" when it tried to send the third.
3096 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
3098 27. The extern definition of crypt16() in expand.c was not being excluded when
3099 the OS had its own crypt16() function.
3101 28. Added bounce_return_body as a new option, and bounce_return_size_limit
3102 as a preferred synonym for return_size_limit, both as an option and as an
3105 29. Added LIBS=-liconv to OS/Makefile-OSF1.
3107 30. Changed the default configuration ACL to relax the local part checking rule
3108 for addresses that are not in any local domains. For these addresses,
3109 slashes and pipe symbols are allowed within local parts, but the sequence
3110 /../ is explicitly forbidden.
3112 31. SPA server authentication was not clearing the challenge buffer before
3115 32. log_message in a "warn" ACL statement was writing to the reject log as
3116 well as to the main log, which contradicts the documentation and doesn't
3117 seem right (because no rejection is happening). So I have stopped it.
3119 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
3120 However, I am unable to do any testing of this.
3122 34. Fixed an infelicity in the appendfile transport. When checking directories
3123 for a mailbox, to see if any needed to be created, it was accidentally
3124 using path names with one or more superfluous leading slashes; tracing
3125 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
3127 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
3128 contents are added to the log line that is written for every discarded
3129 recipient. (Previously a log_message setting was ignored.)
3131 36. The ${quote: operator now quotes the string if it is empty.
3133 37. The install script runs exim in order to find its version number. If for
3134 some reason other than non-existence or emptiness, which it checks, it
3135 could not run './exim', it was installing it with an empty version number,
3136 i.e. as "exim-". This error state is now caught, and the installation is
3139 38. An argument was missing from the function that creates an error message
3140 when Exim fails to connect to the socket for saslauthd authentication.
3141 This could cause Exim to crash, or give a corrupted message.
3143 39. Added isip, isip4, and isip6 to ${if conditions.
3145 40. The ACL variables $acl_xx are now saved with the message, and can be
3146 accessed later in routers, transports, and filters.
3148 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
3149 strings in the file are not string-expanded.
3151 42. If a MAIL command specified a SIZE value that was too large to fit into an
3152 int variable, the check against message_size_limit failed. Such values are
3153 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
3154 day this will have to be increased, but I don't think I want to be around
3155 when emails are that large.
3162 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
3163 iconv() is not standard in FreeBSD.
3165 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
3166 IPv6 enabled. The observed symptom was a segmentation fault on return from
3167 the function os_common_find_running_interfaces() in src/os.c.
3169 3. In the check_special_case() function in daemon.c I had used "errno" as an
3170 argument name, which causes warnings on some systems. This was basically a
3171 typo, since it was named "eno" in the comments!
3173 4. The code that waits for the clock to tick (at a resolution of some fraction
3174 of a second) so as to ensure message-id uniqueness was always waiting for
3175 at least one whole tick, when it could have waited for less. [This is
3176 almost certainly not relevant at current processor speeds, where it is
3177 unlikely to ever wait at all. But we try to future-proof.]
3179 5. The function that sleeps for a time interval that includes fractions of a
3180 second contained a race. It did not block SIGALRM between setting the
3181 timer, and suspending (a couple of lines later). If the interval was short
3182 and the sigsuspend() was delayed until after it had expired, the suspension
3183 never ended. On busy systems this could lead to processes getting stuck for
3186 6. Some uncommon configurations may cause a lookup to happen in a queue runner
3187 process, before it forks any delivery processes. The open lookup caching
3188 mechanism meant that the open file or database connection was passed into
3189 the delivery process. The problem was that delivery processes always tidy
3190 up cached lookup data. This could cause a problem for the next delivery
3191 process started by the queue runner, because the external queue runner
3192 process does not know about the closure. So the next delivery process
3193 still has data in the lookup cache. In the case of a file lookup, there was
3194 no problem because closing a file descriptor in a subprocess doesn't affect
3195 the parent. However, if the lookup was caching a connection to a database,
3196 the connection was closed, and the second delivery process was likely to
3197 see errors such as "PGSQL: query failed: server closed the connection
3198 unexpectedly". The problem has been fixed by closing all cached lookups
3199 in a queue runner before running a delivery process.
3201 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
3202 seem to have the "const" qualifier which it has on other OS. I've
3205 8. Change 4.21/2 was too strict. It is only if there are two authenticators
3206 *of the same type* (client or server) with the same public name that an
3207 error should be diagnosed.
3209 9. When Exim looked up a host name for an IP address, but failed to find the
3210 original IP address when looking up the host name (a safety check), it
3211 output the message "<ip address> does not match any IP for NULL", which was
3212 confusing, to say the least. The bug was that the host name should have
3213 appeared instead of "NULL".
3215 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
3216 user that is built into the binary, and the -C or -D options is used, root
3217 privilege is dropped before the configuration file is read. In addition,
3218 logging is switched to stderr instead of the normal log files. If the
3219 configuration then re-defines the Exim user, the unprivileged environment
3220 is probably not what is expected, so Exim logs a panic warning message (but
3223 However, if deliver_drop_privilege is set, the unprivileged state may well
3224 be exactly what is intended, so the warning has been cut out in that case,
3225 and Exim is allowed to try to write to its normal log files.
3231 1. smtp_return_error_details was not giving details for temporary sender
3232 or receiver verification errors.
3234 2. Diagnose a configuration error if two authenticators have the same public
3237 3. Exim used not to create the message log file for a message until the first
3238 delivery attempt. This could be confusing when incoming messages were held
3239 for policy or load reasons. The message log file is now created at the time
3240 the message is received, and an initial "Received" line is written to it.
3242 4. The automatically generated man page for command line options had a minor
3243 bug that caused no ill effects; however, a more serious problem was that
3244 the procedure for building the man page automatically didn't always
3245 operate. Consequently, release 4.20 contains an out-of-date version. This
3246 shouldn't happen again.
3248 5. When building Exim with embedded Perl support, the script that builds the
3249 Makefile was calling 'perl' to find its compile-time parameters, ignoring
3250 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
3252 6. The freeze_tell option was not being used for messages that were frozen on
3253 arrival, either by an ACL or by local_scan().
3255 7. Added the smtp_incomplete_transaction log selector.
3257 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
3258 was accepting AUTH without a new EHLO.
3260 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
3261 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
3264 10. Logging of TCP/IP connections (when configured) now happens in the main
3265 daemon process instead of the child process, so that the TCP/IP connection
3266 count is more accurate (but it can never be perfect).
3268 11. The use of "drop" in a nested ACL was not being handled correctly in the
3269 outer ACL. Now, if condition failure induced by the nested "drop" causes
3270 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
3271 or "require"), the connection is dropped.
3273 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
3274 that yield "discard" can now be used with an "accept" or a "discard" verb,
3275 but an error is generated for any others (because I can't see a useful way
3276 to define what should happen).
3278 13. When an ACL is read dynamically from a file (or anywhere else), the lines
3279 are now processed in the same way as lines in the Exim configuration file.
3280 In particular, continuation lines are supported.
3282 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
3284 15. Added -ti meaning -t -i.
3286 16. Check for letters, digits, hyphens, and dots in the names of dnslist
3287 domains, and warn by logging if others are found.
3289 17. At least on BSD, alignment is not guarenteed for the array of ifreq's
3290 returned from GIFCONF when Exim is trying to find the list of interfaces on
3291 a host. The code in os.c has been modified to copy each ifreq to an aligned
3292 structure in all cases.
3294 Also, in some cases, the returned ifreq's were being copied to a 'struct
3295 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
3296 means the last couple of bytes of an IPv6 address could be chopped if the
3297 ifreq contained only a normal sockaddr (14 bytes storage).
3299 18. Named domain lists were not supported in the hosts_treat_as_local option.
3300 An entry such as +xxxx was not recognized, and was treated as a literal
3303 19. Ensure that header lines added by a DATA ACL are included in the reject log
3304 if the ACL subsequently rejects the message.
3306 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
3307 MD5 (which is deprecated).
3309 21. When testing a filter file using -bf, Exim was writing a message when it
3310 took the sender from a "From " line in the message, but it was not doing so
3311 when it took $return_path from a Return-Path: header line. It now does.
3313 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
3314 with a valid header line field name (a series of printing characters
3315 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
3317 23. Changed "disc" in the source to "disk" to conform to the documentation and
3318 the book and for uniformity.
3320 24. Ignore Sendmail's -Ooption=value command line item.
3322 25. When execve() failed while trying to run a command in a pipe transport,
3323 Exim was returning EX_UNAVAILBLE (69) from the subprocess. However, this
3324 could be confused with a return value of 69 from the command itself. This
3325 has been changed to 127, the value the shell returns if it is asked to run
3326 a non-existent command. The wording for the related log line suggests a
3327 non-existent command as the problem.
3329 26. If received_header_text expands to an empty string, do not add a Received:
3330 header line to the message. (Well, it adds a token one on the spool, but
3331 marks it "old" so that it doesn't get used or transmitted.)
3333 27. Installed eximstats 1.28 (addition of -nt option).
3335 28. There was no check for failure on the call to getsockname() in the daemon
3336 code. This can fail if there is a shortage of resources on the system, with
3337 ENOMEM, for example. A temporary error is now given on failure.
3339 29. Contrary to the C standard, it seems that in some environments, the
3340 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
3341 program. Exim now does this explicitly; it affects the formatting of
3342 timestamps using strftime().
3344 30. If exiqsumm was given junk data, it threw up some uninitialized variable
3345 complaints. I've now initialized all the variables, to avoid this.
3347 32. Header lines added by a system filter were not being "seen" during
3348 transport-time rewrites.
3350 33. The info_callback() function passed to OpenSSL is set up with type void
3351 (*)(SSL *, int, int), as described somewhere. However, when calling the
3352 function (actually a macro) that sets it up, the type void(*)() is
3353 expected. I've put in a cast to prevent warnings from picky compilers.
3355 34. If a DNS black list lookup found a CNAME record, but there were no A
3356 records associated with the domain it pointed at, Exim crashed.
3358 35. If a DNS black list lookup returned more than one A record, Exim ignored
3359 all but the first. It now scans all returned addresses if a particular IP
3360 value is being sought. In this situation, the contents of the
3361 $dnslist_value variable are a list of all the addresses, separated by a
3364 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
3365 to accept a host name and all its aliases if the forward lookup for any of
3366 them yielded the IP address of the incoming connection. Now it accepts only
3367 those names whose forward lookup yields the correct IP address. Any other
3368 names are discarded. This closes a loophole whereby a rogue DNS
3369 administrator could create reverse DNS records to break through a
3370 wildcarded host restriction in an ACL.
3372 37. If a user filter or a system filter that ran in a subprocess used any of
3373 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
3374 the wrong values were passed to the pipe command ($thisaddress had the
3375 value of $0, $0 had the value of $1, etc). This bug was introduced by
3376 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
3378 38. Improved the line breaking for long SMTP error messages from ACLs.
3379 Previously, if there was no break point between 40 and 75 characters, Exim
3380 left the rest of the message alone. Two changes have been made: (a) I've
3381 reduced the minimum length to 35 characters; (b) if it can't find a break
3382 point between 35 and 75 characters, it looks ahead and uses the first one
3383 that it finds. This may give the occasional overlong line, but at least the
3384 remaining text gets split now.
3386 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
3387 file descriptors might be low, and that setting 1000 would always raise it.
3388 It turns out that in some environments, the limit is already over 1000 and
3389 that lowering it causes trouble. So now Exim takes care not to decrease it.
3391 40. When delivering a message, the value of $return_path is set to $sender_
3392 address at the start of routing (routers may change the value). By an
3393 oversight, this default was not being set up when an address was tested by
3394 -bt or -bv, which affected the outcome if any router or filter referred to
3397 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
3398 log only when "message" or "log_message" are set. However, if one of the
3399 conditions was an address verification, or a call to a nested ACL, the
3400 messages generated by the underlying test were being passed through. This
3401 no longer happens. The underlying message is available in $acl_verify_
3402 message for both "message" and "log_message" expansions, so it can be
3403 passed through if needed.
3405 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
3406 new expansion $bh_ to give the encoded byte string without charset
3407 translation. Translation happens only if iconv() is available; HAVE_ICONV
3408 indicates this at build time. HEADERS_CHARSET gives the charset to
3409 translate to; headers_charset can change it in the configuration, and
3410 "headers charset" can change it in an individual filter file.
3412 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
3413 that creates RFC 2047 encoded "words" labels them as that charset instead
3414 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
3415 expansion operator; (ii) when Exim creates a From: line for a local
3416 message; (iii) when a header line is rewritten to include a "phrase" part.
3418 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
3419 buggy, causing it to skip the first lines of messages whose message ID
3420 ended in 'D'. This would not have bitten before Exim release 4.14, because
3421 message IDs were unlikely to end in 'D' before then. The effect was to have
3422 incorrect size information for certain domains.
3424 45. #include "config.h" was missing at the start of the crypt16.c module. This
3425 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
3428 46. If there was a timeout during a "random" callout check, Exim treated it as
3429 a failure of the random address, and carried on sending RSET and the real
3430 address. If the delay was just some slowness somewhere, the response to the
3431 original RCPT would be taken as a response to RSET and so on, causing
3432 mayhem of various kinds.
3434 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
3435 when I implemented it. It didn't allow for the fact that some option values
3436 may legitimatetly be negative (e.g. size_addition), and it didn't even do
3437 the right test for positive values.
3439 48. Domain names in DNS records are case-independent. Exim always looks them up
3440 in lower case. Some resolvers return domain names in exactly the case they
3441 appear in the zone file, that is, they may contain uppercase letters. Not
3442 all resolvers do this - some return always lower case. Exim was treating a
3443 change of case by a resolver as a change of domain, similar to a widening
3444 of a domain abbreviation. This triggered its re-routing code and so it was
3445 trying to route what was effectively the same domain again. This normally
3446 caused routing to fail (because the router wouldn't handle the domain
3447 twice). Now Exim checks for this case specially, and just changes the
3448 casing of the domain that it ultimately uses when it transmits the message
3451 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
3454 50. If a filter generated a file delivery with a non-absolute name (possible if
3455 no home directory exists for the router), the forbid_file option was not
3458 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
3459 the existing equality matching.
3461 52. Exim was using ints instead of ino_t variables in some places where it was
3462 dealing with inode numbers.
3464 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
3465 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
3466 called TMPDIR, and if it finds it is different, it changes its value.
3468 54. The smtp_printf() function is now made available to local_scan() so
3469 additional output lines can be written before returning. There is also an
3470 smtp_fflush() function to enable the detection of a dropped connection.
3471 The variables smtp_input and smtp_batched_input are exported to
3474 55. Changed the default runtime configuration: the message "Unknown user"
3475 has been removed from the ACL, and instead placed on the localuser router,
3476 using the cannot_route_message feature. This means that any verification
3477 failures that generate their own messages won't get overridden. Similarly,
3478 the "Unrouteable address" message that was in the ACL for unverifiable
3479 relay addresses has also been removed.
3481 56. Added hosts_avoid_esmtp to the smtp transport.
3483 57. The exicyclog script was not checking for the esoteric option
3484 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
3485 will work only if exicyclog is run under the appropriate euid.
3487 58. Following a discussion on the list, the rules by which Exim recognises line
3488 endings on incoming messages have been changed. The -dropcr and drop_cr
3489 options are now no-ops, retained only for backwards compatibility. The
3490 following line terminators are recognized: LF CRLF CR. However, special
3491 processing applies to CR:
3493 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
3494 nor a local message in the state where . is a terminator.
3496 (ii) If a bare CR is encountered in a header line, an extra space is added
3497 after the line terminator so as not to end the header. The reasoning
3498 behind this is that bare CRs in header lines are most likely either
3499 to be mistakes, or people trying to play silly games.
3501 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
3502 was being incorrectly given as 18 bytes larger than it should have been.
3503 This is a VOB (very old bug).
3505 60. This may never have affected anything current, but just in case it has:
3506 When the local host is found other than at the start of a list of hosts,
3507 the local host, those with the same MX, and any that follow, are discarded.
3508 When the list in question was part of a longer list of hosts, the following
3509 hosts (not currently being processed) were also being discarded. This no
3510 longer happens. I'm not sure if this situation could ever has previously
3513 61. Added the "/MX" feature to lists of hosts in the manualroute and query
3516 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
3517 header. This is something that is recommended in a new Internet Draft, and
3518 is something that is documented as being done by Sendmail. There are two
3519 possible values. For messages generated by the autoreply transport, Exim
3522 Auto-Submitted: auto-replied
3524 whereas for all other generated messages (e.g. bounces) it adds
3526 Auto-Submitted: auto-generated
3528 63. The "personal" condition in filters now includes a test for the
3529 Auto-Submitted: header. If it contains the string "auto-" the message it
3530 not considered personal.
3532 64. Added rcpt_include_affixes as a generic transport option.
3534 65. Added queue_only_override (default true).
3536 66. Added the syslog_duplication option.
3538 67. If what should have been the first header line of a message consisted of
3539 a space followed by a colon, Exim was mis-interpreting it as a header line.
3540 It isn't of course - it is syntactically invalid and should therefore be
3541 treated as the start of the message body. The misbehaviour could have
3542 caused a number of strange effects, including loss of data in subsequent
3543 header lines, and spool format errors.
3545 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
3546 client host had authenticated. This control can now be exercised by an ACL
3547 for more flexibility.
3549 69. By default, callouts do not happen when testing with -bh. There is now a
3550 variant, -bhc, which does actually run the callout code, including
3551 consulting and updating the callout cache.
3553 70. Added support for saslauthd authentication, courtesy of Alexander
3556 71. If statvfs() failed on the spool or log directories while checking their
3557 size for availability, Exim confusingly gave the error "space shortage".
3558 Furthermore, in debugging mode it crashed with a floating point exception.
3559 These checks are done if check_{spool,log}_{space,inodes} are set, and when
3560 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
3561 serious problem, Exim now writes to the main and panic logs when this
3562 happens, with details of the failure. It then refuses to accept the
3563 incoming message, giving the message "spool directory problem" or "log
3564 directory problem" with a 421 code for SMTP messages.
3566 72. When Exim is about to re-exec itself, it ensures that the file descriptors
3567 0, 1, and 2 exist, because some OS complain for execs without them (see
3568 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
3569 descriptors. However, the code omitted to check that the open succeeded,
3570 causing mysterious errors if for some reason the permissions on /dev/null
3571 got screwed. Now Exim writes a message to the main and panic logs, and
3572 bombs out if it can't open /dev/null.
3574 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
3575 interact so that it is all more flexible. It is supposed to remain
3576 backwards compatible. Also added extra_local_interfaces.
3578 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
3579 to bomb out with an assertion failure - to the client this appears as a
3580 connection drop. This problem occurs in the part of the code that was taken
3581 from the Samba project. Fortunately, the assertion is in a very simple
3582 function, so I have fixed this by reproducing the function inline in the
3583 one place where it is called, and arranging for authentication to fail
3584 instead of killing the process with assert().
3586 75. The SPA client code was not working when the server requested OEM rather
3587 than Unicode encoding.
3589 76. Added code to make require_files with a specific uid setting more usable in
3590 the case where statting the file as root fails - usually a non-root-mounted
3591 NFS file system. When this happens and the failure is EACCES, Exim now
3592 forks a subprocess and does the per-uid checking as the relevant uid.
3594 77. Added process_log_path.
3596 78. If log_file_path was not explicitly set, a setting of check_log_space or
3597 check_log_inodes was ignored.
3599 79. If a space check for the spool or log partitions fails, the incident is now
3600 logged. Of course, in the latter case the data may get lost...
3602 80. Added the %p formatting code to string_format() so that it can be used to
3603 print addresses in debug_print(). Adjusted all the address printing in the
3604 debugging in store.c to use %p rather than %d.
3606 81. There was a concern that a line of code in smtp_in.c could overflow a
3607 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
3608 initially expressed, the concern was not well-founded, because trailing
3609 spaces are removed early. However, if the trailing spaces were followed by
3610 a NULL, they did not get removed, so the overflow was possible. Two fixes
3613 (a) I re-wrote the offending code in a cleaner fashion.
3614 (b) If an incoming SMTP command contains a NULL character, it is rejected
3617 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
3618 runs initgroups(), so that if the Exim user is in any additional groups,
3619 they will be used during message reception.
3625 The change log for 4.20 and earlier releases has been archived.
git://git.exim.org / users / heiko / exim.git / blob