1 Change log file for Exim from version 4.21
2 -------------------------------------------
7 PP/01 Solaris build fix for Oracle's LDAP libraries.
8 Bugzilla 1109, patch from Stephen Usher.
10 TF/01 HP/UX build fix: avoid arithmetic on a void pointer.
12 TK/01 DKIM Verification: Fix relaxed canon for empty headers w/o
15 TF/02 Fix a couple more cases where we did not log the error message
16 when unlink() failed. See also change 4.74-TF/03.
18 TF/03 Make the exiwhat support code safe for signals. Previously Exim might
19 lock up or crash if it happened to be inside a call to libc when it
20 got a SIGUSR1 from exiwhat.
22 The SIGUSR1 handler appends the current process status to the process
23 log which is later printed by exiwhat. It used to use the general
24 purpose logging code to do this, but several functions it calls are
27 The new output code in the SIGUSR1 handler is specific to the process
28 log, and simple enough that it's easy to inspect for signal safety.
29 Removing some special cases also simplifies the general logging code.
30 Removing the spurious timestamps from the process log simplifies
33 TF/04 Improved ratelimit ACL condition.
35 The /noupdate option has been deprecated in favour of /readonly which
36 has clearer semantics. The /leaky, /strict, and /readonly update modes
37 are mutually exclusive. The update mode is no longer included in the
38 database key; it just determines when the database is updated. (This
39 means that when you upgrde Exim will forget old rate measurements.)
41 Exim now checks that the per_* options are used with an update mode that
42 makes sense for the current ACL. For example, when Exim is processing a
43 message (e.g. acl_smtp_rcpt or acl_smtp_data, etc.) you can specify
44 per_mail/leaky or per_mail/strict; otherwise (e.g. in acl_smtp_helo) you
45 must specify per_mail/readonly. If you omit the update mode it defaults to
46 /leaky where that makes sense (as before) or /readonly where required.
48 The /noupdate option is now undocumented but still supported for
49 backwards compatibility. It is equivalent to /readonly except that in
50 ACLs where /readonly is required you may specify /leaky/noupdate or
51 /strict/noupdate which are treated the same as /readonly.
53 A useful new feature is the /count= option. This is a generalization
54 of the per_byte option, so that you can measure the throughput of other
55 aggregate values. For example, the per_byte option is now equivalent
56 to per_mail/count=${if >{0}{$message_size} {0} {$message_size} }.
58 The per_rcpt option has been generalized using the /count= mechanism
59 (though it's more complicated than the per_byte equivalence). When it is
60 used in acl_smtp_rcpt, the per_rcpt option adds recipients to the
61 measured rate one at a time; if it is used later (e.g. in acl_smtp_data)
62 or in a non-SMTP ACL it adds all the recipients in one go. (The latter
63 /count=$recipients_count behaviour used to work only in non-SMTP ACLs.)
64 Note that using per_rcpt with a non-readonly update mode in more than
65 one ACL will cause the recipients to be double-counted. (The per_mail
66 and per_byte options don't have this problem.)
68 The handling of very low rates has changed slightly. If the computed rate
69 is less than the event's count (usually one) then this event is the first
70 after a long gap. In this case the rate is set to the same as this event's
71 count, so that the first message of a spam run is counted properly.
73 The major new feature is a mechanism for counting the rate of unique
74 events. The new per_addr option counts the number of different
75 recipients that someone has sent messages to in the last time period. It
76 behaves like per_rcpt if all the recipient addresses are different, but
77 duplicate recipient addresses do not increase the measured rate. Like
78 the /count= option this is a general mechanism, so the per_addr option
79 is equivalent to per_rcpt/unique=$local_part@$domain. You can, for
80 example, measure the rate that a client uses different sender addresses
81 with the options per_mail/unique=$sender_address. There are further
82 details in the main documentation.
84 TF/05 Removed obsolete $Cambridge$ CVS revision strings.
86 TF/06 Removed a few PCRE remnants.
88 TF/07 Automatically extract Exim's version number from tags in the git
89 repository when doing development or release builds.
91 PP/02 Raise smtp_cmd_buffer_size to 16kB.
92 Bugzilla 879. Patch from Paul Fisher.
94 PP/03 Implement SSL-on-connect outbound with protocol=smtps on smtp transport.
95 Heavily based on revision 40f9a89a from Simon Arlott's tree.
98 PP/04 Use .dylib instead of .so for dynamic library loading on MacOS.
100 PP/05 Variable $av_failed, true if the AV scanner deferred.
101 Bugzilla 1078. Patch from John Horne.
103 PP/06 Stop make process more reliably on build failure.
104 Bugzilla 1087. Patch from Heiko Schlittermann.
106 PP/07 Make maildir_use_size_file an _expandable_ boolean.
107 Bugzilla 1089. Patch from Heiko Schlittermann.
109 PP/08 Handle ${run} returning more data than OS pipe buffer size.
110 Bugzilla 1131. Patch from Holger Weiß.
112 PP/09 Handle IPv6 addresses with SPF.
113 Bugzilla 860. Patch from Wolfgang Breyha.
115 PP/10 match_* no longer expand right-hand-side by default.
116 New compile-time build option, EXPAND_LISTMATCH_RHS.
117 New expansion conditions, "inlist", "inlisti".
123 PP/01 The new ldap_require_cert option would segfault if used. Fixed.
125 PP/02 Harmonised TLS library version reporting; only show if debugging.
126 Layout now matches that introduced for other libraries in 4.74 PP/03.
128 PP/03 New openssl_options items: no_sslv2 no_sslv3 no_ticket no_tlsv1
130 PP/04 New "dns_use_edns0" global option.
132 PP/05 Don't segfault on misconfiguration of ref:name exim-user as uid.
135 PP/06 Extra paranoia around buffer usage at the STARTTLS transition.
136 nb: Exim is not vulnerable to http://www.kb.cert.org/vuls/id/555316
138 TK/01 Updated PolarSSL code to 0.14.2.
139 Bugzilla 1097. Patch from Andreas Metzler.
141 PP/07 Catch divide-by-zero in ${eval:...}.
144 PP/08 Condition negation of bool{}/bool_lax{} did not negate. Fixed.
147 TK/02 Bugzilla 1106: CVE-2011-1764 - DKIM log line was subject to a
148 format-string attack -- SECURITY: remote arbitrary code execution.
150 TK/03 SECURITY - DKIM signature header parsing was double-expanded, second
151 time unintentionally subject to list matching rules, letting the header
152 cause arbitrary Exim lookups (of items which can occur in lists, *not*
153 arbitrary string expansion). This allowed for information disclosure.
155 PP/09 Fix another SIGFPE (x86) in ${eval:...} expansion, this time related to
156 INT_MIN/-1 -- value coerced to INT_MAX.
162 NM/01 Workround for PCRE version dependency in version reporting
165 TF/01 Update valgrind.h and memcheck.h to copies from valgrind-3.6.0.
166 This fixes portability to compilers other than gcc, notably
167 Solaris CC and HP-UX CC. Fixes Bugzilla 1050.
169 TF/02 Bugzilla 139: Avoid using the += operator in the modular lookup
170 makefiles for portability to HP-UX and POSIX correctness.
172 PP/01 Permit LOOKUP_foo enabling on the make command-line.
173 Also via indented variable definition in the Makefile.
174 (Debugging by Oliver Heesakkers).
176 PP/02 Restore caching of spamd results with expanded spamd_address.
177 Patch from author of expandable spamd_address patch, Wolfgang Breyha.
179 PP/03 Build issue: lookups-Makefile now exports LC_ALL=C
180 Improves build reliability. Fix from: Frank Elsner
182 NM/02 Fix wide character breakage in the rfc2047 coding
183 Fixes bug 1064. Patch from Andrey N. Oktyabrski
185 NM/03 Allow underscore in dnslist lookups
186 Fixes bug 1026. Patch from Graeme Fowler
188 PP/04 Bugzilla 230: Support TLS-enabled LDAP (in addition to ldaps).
189 Code patches from Adam Ciarcinski of NetBSD.
191 NM/04 Fixed exiqgrep to cope with mailq missing size issue
194 PP/05 Bugzilla 1083: when lookup expansion defers, escape the output which
195 is logged, to avoid truncation. Patch from John Horne.
197 PP/06 Bugzilla 1042: implement freeze_signal on pipe transports.
198 Patch from Jakob Hirsch.
200 PP/07 Bugzilla 1061: restrict error messages sent over SMTP to not reveal
201 SQL string expansion failure details.
202 Patch from Andrey Oktyabrski.
204 PP/08 Bugzilla 486: implement %M datestamping in log filenames.
205 Patch from Simon Arlott.
207 PP/09 New lookups functionality failed to compile on old gcc which rejects
208 extern declarations in function scope.
209 Patch from Oliver Fleischmann
211 PP/10 Use sig_atomic_t for flags set from signal handlers.
212 Check getgroups() return and improve debugging.
213 Fixed developed for diagnosis in bug 927 (which turned out to be
216 PP/11 Bugzilla 1055: Update $message_linecount for maildir_tag.
217 Patch from Mark Zealey.
219 PP/12 Bugzilla 1056: Improved spamd server selection.
220 Patch from Mark Zealey.
222 PP/13 Bugzilla 1086: Deal with maildir quota file races.
223 Based on patch from Heiko Schlittermann.
225 PP/14 Bugzilla 1019: DKIM multiple signature generation fix.
226 Patch from Uwe Doering, sign-off by Michael Haardt.
228 NM/05 Fix to spam.c to accommodate older gcc versions which dislike
229 variable declaration deep within a block. Bug and patch from
232 PP/15 lookups-Makefile IRIX compatibilty coercion.
234 PP/16 Make DISABLE_DKIM build knob functional.
236 NM/06 Bugzilla 968: child_open_uid: restore default SIGPIPE handler
237 Patch by Simon Arlott
239 TF/03 Fix valgrind.h portability to C89 compilers that do not support
240 variable argument macros. Our copy now differs from upstream.
246 TF/01 Failure to get a lock on a hints database can have serious
247 consequences so log it to the panic log.
249 TF/02 Log LMTP confirmation messages in the same way as SMTP,
250 controlled using the smtp_confirmation log selector.
252 TF/03 Include the error message when we fail to unlink a spool file.
254 DW/01 Bugzilla 139: Support dynamically loaded lookups as modules.
255 With thanks to Steve Haslam, Johannes Berg & Serge Demonchaux
256 for maintaining out-of-tree patches for some time.
258 PP/01 Bugzilla 139: Documentation and portability issues.
259 Avoid GNU Makefile-isms, let Exim continue to build on BSD.
260 Handle per-OS dynamic-module compilation flags.
262 PP/02 Let /dev/null have normal permissions.
263 The 4.73 fixes were a little too stringent and complained about the
264 permissions on /dev/null. Exempt it from some checks.
265 Reported by Andreas M. Kirchwitz.
267 PP/03 Report version information for many libraries, including
268 Exim version information for dynamically loaded libraries. Created
269 version.h, now support a version extension string for distributors
270 who patch heavily. Dynamic module ABI change.
272 PP/04 CVE-2011-0017 - check return value of setuid/setgid. This is a
273 privilege escalation vulnerability whereby the Exim run-time user
274 can cause root to append content of the attacker's choosing to
277 PP/05 Bugzilla 1041: merged DCC maintainer's fixes for return code.
280 PP/06 Bugzilla 1071: fix delivery logging with untrusted macros.
281 If dropping privileges for untrusted macros, we disabled normal logging
282 on the basis that it would fail; for the Exim run-time user, this is not
283 the case, and it resulted in successful deliveries going unlogged.
284 Fixed. Reported by Andreas Metzler.
290 PP/01 Date: & Message-Id: revert to normally being appended to a message,
291 only prepend for the Resent-* case. Fixes regression introduced in
292 Exim 4.70 by NM/22 for Bugzilla 607.
294 PP/02 Include check_rfc2047_length in configure.default because we're seeing
295 increasing numbers of administrators be bitten by this.
297 JJ/01 Added DISABLE_DKIM and comment to src/EDITME
299 PP/03 Bugzilla 994: added openssl_options main configuration option.
301 PP/04 Bugzilla 995: provide better SSL diagnostics on failed reads.
303 PP/05 Bugzilla 834: provide a permit_coredump option for pipe transports.
305 PP/06 Adjust NTLM authentication to handle SASL Initial Response.
307 PP/07 If TLS negotiated an anonymous cipher, we could end up with SSL but
308 without a peer certificate, leading to a segfault because of an
309 assumption that peers always have certificates. Be a little more
310 paranoid. Problem reported by Martin Tscholak.
312 PP/08 Bugzilla 926: switch ClamAV to use the new zINSTREAM API for content
313 filtering; old API available if built with WITH_OLD_CLAMAV_STREAM=yes
314 NB: ClamAV planning to remove STREAM in "middle of 2010".
315 CL also introduces -bmalware, various -d+acl logging additions and
316 more caution in buffer sizes.
318 PP/09 Implemented reverse_ip expansion operator.
320 PP/10 Bugzilla 937: provide a "debug" ACL control.
322 PP/11 Bugzilla 922: Documentation dusting, patch provided by John Horne.
324 PP/12 Bugzilla 973: Implement --version.
326 PP/13 Bugzilla 752: Refuse to build/run if Exim user is root/0.
328 PP/14 Build without WITH_CONTENT_SCAN. Path from Andreas Metzler.
330 PP/15 Bugzilla 816: support multiple condition rules on Routers.
332 PP/16 Add bool_lax{} expansion operator and use that for combining multiple
333 condition rules, instead of bool{}. Make both bool{} and bool_lax{}
334 ignore trailing whitespace.
336 JJ/02 prevent non-panic DKIM error from being sent to paniclog
338 JJ/03 added tcp_wrappers_daemon_name to allow host entries other than
341 PP/17 Fix malware regression for cmdline scanner introduced in PP/08.
342 Notification from Dr Andrew Aitchison.
344 PP/18 Change ClamAV response parsing to be more robust and to handle ClamAV's
345 ExtendedDetectionInfo response format.
346 Notification from John Horne.
348 PP/19 OpenSSL 1.0.0a compatibility const-ness change, should be backwards
351 PP/20 Added a CONTRIBUTING file. Fixed the documentation build to use http:
352 XSL and documented dependency on system catalogs, with examples of how
355 DW/21 Added Valgrind hooks in store.c to help it capture out-of-bounds store
358 DW/22 Bugzilla 1044: CVE-2010-4345 - partial fix: restrict default behaviour
359 of CONFIGURE_OWNER and CONFIGURE_GROUP options to no longer allow a
360 configuration file which is writeable by the Exim user or group.
362 DW/23 Bugzilla 1044: CVE-2010-4345 - part two: extend checks for writeability
363 of configuration files to cover files specified with the -C option if
364 they are going to be used with root privileges, not just the default
367 DW/24 Bugzilla 1044: CVE-2010-4345 - part three: remove ALT_CONFIG_ROOT_ONLY
368 option (effectively making it always true).
370 DW/25 Add TRUSTED_CONFIG_PREFIX_FILE option to allow alternative configuration
371 files to be used while preserving root privileges.
373 DW/26 Set FD_CLOEXEC on SMTP sockets after forking in the daemon, to ensure
374 that rogue child processes cannot use them.
376 PP/27 Bugzilla 1047: change the default for system_filter_user to be the Exim
377 run-time user, instead of root.
379 PP/28 Add WHITELIST_D_MACROS option to let some macros be overridden by the
380 Exim run-time user without dropping privileges.
382 DW/29 Remove use of va_copy() which breaks pre-C99 systems. Duplicate the
383 result string, instead of calling string_vformat() twice with the same
386 DW/30 Allow TRUSTED_CONFIG_PREFIX_FILE only for Exim or CONFIGURE_OWNER, not
387 for other users. Others should always drop root privileges if they use
388 -C on the command line, even for a whitelisted configure file.
390 DW/31 Turn TRUSTED_CONFIG_PREFIX_FILE into TRUSTED_CONFIG_FILE. No prefixes.
392 NM/01 Fixed bug #1002 - Message loss when using multiple deliveries
398 JJ/01 installed exipick 20100104.1, adding $max_received_linelength,
399 $data_path, and $header_path variables; fixed documentation bugs and
402 JJ/02 installed exipick 20100222.0, added --input-dir and --finput to allow
403 exipick to access non-standard spools, including the "frozen" queue
406 NM/01 Bugzilla 965: Support mysql stored procedures.
407 Patch from Alain Williams
409 NM/02 Bugzilla 961: Spacing fix (syntax error) on Makefile directives for NetBSD
411 NM/03 Bugzilla 955: Documentation fix for max_rcpts.
412 Patch from Andreas Metzler
414 NM/04 Bugzilla 954: Fix for unknown responses from Dovecot authenticator.
415 Patch from Kirill Miazine
417 NM/05 Bugzilla 671: Added umask to procmail example.
419 JJ/03 installed exipick 20100323.0, fixing doc bug
421 NM/06 Bugzilla 988: CVE-2010-2023 - prevent hardlink attack on sticky mail
422 directory. Notification and patch from Dan Rosenberg.
424 TK/01 PDKIM: Upgrade PolarSSL files to upstream version 0.12.1.
426 TK/02 Improve log output when DKIM signing operation fails.
428 MH/01 Treat the transport option dkim_domain as a colon separated
429 list, not as a single string, and sign the message with each element,
430 omitting multiple occurences of the same signer.
432 NM/07 Null terminate DKIM strings, Null initialise DKIM variable
433 Bugzilla 985, 986. Patch by Simon Arlott
435 NM/08 Bugzilla 967. dnsdb DNS TXT record bug fix (DKIM-related)
436 Patch by Simon Arlott
438 PP/01 Bugzilla 989: CVE-2010-2024 - work round race condition on
439 MBX locking. Notification from Dan Rosenberg.
445 TK/01 Bugzilla 912: Fix DKIM segfault on empty headers/body.
447 NM/01 Bugzilla 913: Documentation fix for gnutls_* options.
449 NM/02 Bugzilla 722: Documentation for randint. Better randomness defaults.
451 NM/03 Bugzilla 847: Enable DNSDB lookup by default.
453 NM/04 Bugzilla 915: Flag broken perl installation during build.
459 TK/01 Added patch by Johannes Berg that expands the main option
460 "spamd_address" if it starts with a dollar sign.
462 TK/02 Write list of recipients to X-Envelope-Sender header when building
463 the mbox-format spool file for content scanning (suggested by Jakob
466 TK/03 Added patch by Wolfgang Breyha that adds experimental DCC
467 (http://www.dcc-servers.net/) support via dccifd. Activated by
468 setting EXPERIMENTAL_DCC=yes in Local/Makefile.
470 TK/04 Bugzilla 673: Add f-protd malware scanner support. Patch submitted
471 by Mark Daniel Reidel <mr@df.eu>.
473 NM/01 Bugzilla 657: Embedded PCRE removed from the exim source tree.
474 When building exim an external PCRE library is now needed -
475 PCRE is a system library on the majority of modern systems.
476 See entry on PCRE_LIBS in EDITME file.
478 NM/02 Bugzilla 646: Removed unwanted C/R in Dovecot authenticator
479 conversation. Added nologin parameter to request.
480 Patch contributed by Kirill Miazine.
482 TF/01 Do not log submission mode rewrites if they do not change the address.
484 TF/02 Bugzilla 662: Fix stack corruption before exec() in daemon.c.
486 NM/03 Bugzilla 602: exicyclog now handles panic log, and creates empty
487 log files in place. Contributed by Roberto Lima.
489 NM/04 Bugzilla 667: Close socket used by dovecot authenticator.
491 TF/03 Bugzilla 615: When checking the local_parts router precondition
492 after a local_part_suffix or local_part_prefix option, Exim now
493 does not use the address's named list lookup cache, since this
494 contains cached lookups for the whole local part.
496 NM/05 Bugzilla 521: Integrated SPF Best Guess support contributed by
497 Robert Millan. Documentation is in experimental-spec.txt.
499 TF/04 Bugzilla 668: Fix parallel build (make -j).
501 NM/05.2 Bugzilla 437: Prevent Maildir aux files being created with mode 000.
503 NM/05.3 Bugzilla 598: Improvement to Dovecot authenticator handling.
504 Patch provided by Jan Srzednicki.
506 TF/05 Leading white space used to be stripped from $spam_report which
507 wrecked the formatting. Now it is preserved.
509 TF/06 Save $spam_score, $spam_bar, and $spam_report in spool files, so
510 that they are available at delivery time.
512 TF/07 Fix the way ${extract is skipped in the untaken branch of a conditional.
514 TF/08 TLS error reporting now respects the incoming_interface and
515 incoming_port log selectors.
517 TF/09 Produce a more useful error message if an SMTP transport's hosts
518 setting expands to an empty string.
520 NM/06 Bugzilla 744: EXPN did not work under TLS.
521 Patch provided by Phil Pennock.
523 NM/07 Bugzilla 769: Extraneous comma in usage fprintf
524 Patch provided by Richard Godbee.
526 NM/08 Fixed erroneous documentation references to smtp_notquit_acl to be
527 acl_smtp_notquit, added index entry.
529 NM/09 Bugzilla 787: Potential buffer overflow in string_format.
530 Patch provided by Eugene Bujak.
532 NM/10 Bugzilla 770: Problem on some platforms modifying the len parameter to
533 accept(). Patch provided by Maxim Dounin.
535 NM/11 Bugzilla 749: Preserve old behaviour of blanks comparing equal to zero.
536 Patch provided by Phil Pennock.
538 NM/12 Bugzilla 497: Correct behaviour of exiwhat when no config exists.
540 NM/13 Bugzilla 590: Correct handling of Resent-Date headers.
541 Patch provided by Brad "anomie" Jorsch.
543 NM/14 Bugzilla 622: Added timeout setting to transport filter.
544 Patch provided by Dean Brooks.
546 TK/05 Add native DKIM support (does not depend on external libraries).
548 NM/15 Bugzilla 854: Removed code that symlinks to pcre as its no longer useful.
549 Patch provided by Graeme Fowler.
551 NM/16 Bugzilla 851: Documentation example syntax fix.
553 NM/17 Changed NOTICE file to remove references to embedded PCRE.
555 NM/18 Bugzilla 894: Fix issue with very long lines including comments in
558 NM/19 Bugzilla 745: TLS version reporting.
559 Patch provided by Phil Pennock.
561 NM/20 Bugzilla 167: bool: condition support.
562 Patch provided by Phil Pennock.
564 NM/21 Bugzilla 665: gnutls_compat_mode to allow compatibility with broken
565 clients. Patch provided by Phil Pennock.
567 NM/22 Bugzilla 607: prepend (not append) Resent-Message-ID and Resent-Date.
568 Patch provided by Brad "anomie" Jorsch.
570 NM/23 Bugzilla 687: Fix misparses in eximstats.
571 Patch provided by Heiko Schlittermann.
573 NM/24 Bugzilla 688: Fix exiwhat to handle log_selector = +pid.
574 Patch provided by Heiko Schlittermann.
576 NM/25 Bugzilla 727: Use transport mode as default mode for maildirsize file.
577 plus update to original patch.
579 NM/26 Bugzilla 799: Documentation correction for ratelimit.
581 NM/27 Bugzilla 802: Improvements to local interface IP addr detection.
582 Patch provided by David Brownlee.
584 NM/28 Bugzilla 807: Improvements to LMTP delivery logging.
586 NM/29 Bugzilla 862, 866, 875: Documentation bugfixes.
588 NM/30 Bugzilla 888: TLS documentation bugfixes.
590 NM/31 Bugzilla 896: Dovecot buffer overrun fix.
592 NM/32 Bugzilla 889: Change all instances of "expr" in shell scripts to "expr --"
593 Unlike the original bugzilla I have changed all shell scripts in src tree.
595 NM/33 Bugzilla 898: Transport filter timeout fix.
596 Patch by Todd Rinaldo.
598 NM/34 Bugzilla 901: Fix sign/unsigned and UTF mistmatches.
599 Patch by Serge Demonchaux.
601 NM/35 Bugzilla 39: Base64 decode bug fixes.
602 Patch by Jakob Hirsch.
604 NM/36 Bugzilla 909: Correct connect() call in dcc code.
606 NM/37 Bugzilla 910: Correct issue with relaxed/simple handling.
608 NM/38 Bugzilla 908: Removed NetBSD3 support as no longer needed.
610 NM/39 Bugzilla 911: Fixed MakeLinks build script.
616 TK/01 Add preliminary DKIM support. Currently requires a forked version of
617 ALT-N's libdkim that I have put here:
618 http://duncanthrax.net/exim-experimental/
620 Note to Michael Haardt: I had to rename some vars in sieve.c. They
621 were called 'true' and it seems that C99 defines that as a reserved
622 keyword to be used with 'bool' variable types. That means you could
623 not include C99-style headers which use bools without triggering
624 build errors in sieve.c.
626 NM/01 Bugzilla 592: --help option is handled incorrectly if exim is invoked
627 as mailq or other aliases. Changed the --help handling significantly
628 to do whats expected. exim_usage() emits usage/help information.
630 SC/01 Added the -bylocaldomain option to eximstats.
632 NM/02 Bugzilla 619: Defended against bad data coming back from gethostbyaddr.
634 NM/03 Bugzilla 613: Documentation fix for acl_not_smtp.
636 NM/04 Bugzilla 628: PCRE update to 7.4 (work done by John Hall).
642 PH/01 Another patch from the Sieve maintainer.
644 PH/02 When an IPv6 address is converted to a string for single-key lookup
645 in an address list (e.g. for an item such as "net24-dbm;/net/works"),
646 dots are used instead of colons so that keys in lsearch files need not
647 contain colons. This was done some time before quoting was made available
648 in lsearch files. However, iplsearch files do require colons in IPv6 keys
649 (notated using the quote facility) so as to distinguish them from IPv4
650 keys. This meant that lookups for IP addresses in host lists did not work
651 for iplsearch lookups.
653 This has been fixed by arranging for IPv6 addresses to be expressed with
654 colons if the lookup type is iplsearch. This is not incompatible, because
655 previously such lookups could never work.
657 The situation is now rather anomolous, since one *can* have colons in
658 ordinary lsearch keys. However, making the change in all cases is
659 incompatible and would probably break a number of configurations.
661 TK/01 Change PRVS address formatting scheme to reflect latests BATV draft
664 MH/01 The "spam" ACL condition code contained a sscanf() call with a %s
665 conversion specification without a maximum field width, thereby enabling
666 a rogue spamd server to cause a buffer overflow. While nobody in their
667 right mind would setup Exim to query an untrusted spamd server, an
668 attacker that gains access to a server running spamd could potentially
669 exploit this vulnerability to run arbitrary code as the Exim user.
671 TK/02 Bugzilla 502: Apply patch to make the SPF-Received: header use
672 $primary_hostname instead of what libspf2 thinks the hosts name is.
674 MH/02 The dsearch lookup now uses lstat(2) instead of stat(2) to look for
675 a directory entry by the name of the lookup key. Previously, if a
676 symlink pointed to a non-existing file or a file in a directory that
677 Exim lacked permissions to read, a lookup for a key matching that
678 symlink would fail. Now it is enough that a matching directory entry
679 exists, symlink or not. (Bugzilla 503.)
681 PH/03 The body_linecount and body_zerocount variables are now exported in the
684 PH/04 Added the $dnslist_matched variable.
686 PH/05 Unset $tls_cipher and $tls_peerdn before making a connection as a client.
687 This means they are set thereafter only if the connection becomes
690 PH/06 Added the client_condition to authenticators so that some can be skipped
691 by clients under certain conditions.
693 PH/07 The error message for a badly-placed control=no_multiline_responses left
694 "_responses" off the end of the name.
696 PH/08 Added -Mvc to output a copy of a message in RFC 2822 format.
698 PH/09 Tidied the code for creating ratelimiting keys, creating them explicitly
699 (without spaces) instead of just copying the configuration text.
701 PH/10 Added the /noupdate option to the ratelimit ACL condition.
703 PH/11 Added $max_received_linelength.
705 PH/12 Added +ignore_defer and +include_defer to host lists.
707 PH/13 Installed PCRE version 7.2. This needed some changes because of the new
708 way in which PCRE > 7.0 is built.
710 PH/14 Implemented queue_only_load_latch.
712 PH/15 Removed an incorrect (int) cast when reading the value of SIZE in a
713 MAIL command. The effect was to mangle the value on 64-bit systems.
715 PH/16 Another patch from the Sieve maintainer.
717 PH/17 Added the NOTQUIT ACL, based on a patch from Ted Cooper.
719 PH/18 If a system quota error occurred while trying to create the file for
720 a maildir delivery, the message "Mailbox is full" was not appended to the
721 bounce if the delivery eventually timed out. Change 4.67/27 below applied
722 only to a quota excession during the actual writing of the file.
724 PH/19 It seems that peer DN values may contain newlines (and other non-printing
725 characters?) which causes problems in log lines. The DN values are now
726 passed through string_printing() before being added to log lines.
728 PH/20 Added the "servers=" facility to MySQL and PostgreSQL lookups. (Oracle
729 and InterBase are left for another time.)
731 PH/21 Added message_body_newlines option.
733 PH/22 Guard against possible overflow in moan_check_errorcopy().
735 PH/23 POSIX allows open() to be a macro; guard against that.
737 PH/24 If the recipient of an error message contained an @ in the local part
738 (suitably quoted, of course), incorrect values were put in $domain and
739 $local_part during the evaluation of errors_copy.
745 MH/01 Fix for bug #448, segfault in Dovecot authenticator when interface_address
746 is unset (happens when testing with -bh and -oMi isn't used). Thanks to
749 PH/01 Added a new log selector smtp_no_mail, to log SMTP sessions that do not
750 issue a MAIL command.
752 PH/02 In an ACL statement such as
754 deny dnslists = X!=127.0.0.2 : X=127.0.0.2
756 if a client was not listed at all, or was listed with a value other than
757 127.0.0.2, in the X list, but was listed with 127.0.0.2 in the Y list,
758 the condition was not true (as it should be), so access was not denied.
759 The bug was that the ! inversion was incorrectly passed on to the second
760 item. This has been fixed.
762 PH/03 Added additional dnslists conditions == and =& which are different from
763 = and & when the dns lookup returns more than one IP address.
765 PH/04 Added gnutls_require_{kx,mac,protocols} to give more control over the
766 cipher suites used by GnuTLS. These options are ignored by OpenSSL.
768 PH/05 After discussion on the list, added a compile time option ENABLE_DISABLE_
769 FSYNC, which compiles an option called disable_fsync that allows for
770 bypassing fsync(). The documentation is heavily laced with warnings.
772 SC/01 Updated eximstats to collate all SpamAssassin rejects into one bucket.
774 PH/06 Some tidies to the infrastructure of the Test Suite that is concerned
775 with the auxiliary C programs that it uses: (1) Arrange for BIND_8_COMPAT
776 to be defined when compiling on OSX (Darwin); (2) Tidies to the Makefile,
777 including adding "make clean"; (3) Added -fPIC when compiling the test
778 dynamically loaded module, to get rid of a warning.
780 MH/02 Fix for bug #451, causing paniclog entries to be written if a bounce
781 message fails, move_frozen_messages = true and ignore_bounce_errors_after
782 = 0s. The bug is otherwise harmless.
784 PH/07 There was a bug in the dovecot authenticator such that the value of
785 $auth1 could be overwritten, and so not correctly preserved, after a
786 successful authentication. This usually meant that the value preserved by
787 the server_setid option was incorrect.
789 PH/08 Added $smtp_count_at_connection_start, deliberately with a long name.
791 PH/09 Installed PCRE release 7.0.
793 PH/10 The acl_not_smtp_start ACL was, contrary to the documentation, not being
794 run for batched SMTP input. It is now run at the start of every message
795 in the batch. While fixing this I discovered that the process information
796 (output by running exiwhat) was not always getting set for -bs and -bS
797 input. This is fixed, and it now also says "batched" for BSMTP.
799 PH/11 Added control=no_pipelining.
801 PH/12 Added $sending_ip_address and $sending_port (mostly Magnus Holmgren's
802 patch, slightly modified), and move the expansion of helo_data till after
803 the connection is made in the smtp transport (so it can use these
806 PH/13 Added ${rfc2047d: to decoded RFC 2047 strings.
808 PH/14 Added log_selector = +pid.
810 PH/15 Flush SMTP output before delaying, unless control=no_delay_flush is set.
812 PH/16 Add ${if forany and ${if forall.
814 PH/17 Added dsn_from option to vary the From: line in DSNs.
816 PH/18 Flush SMTP output before performing a callout, unless control =
817 no_callout_flush is set.
819 PH/19 Change 4.64/PH/36 introduced a bug: when address_retry_include_sender
820 was true (the default) a successful delivery failed to delete the retry
821 item, thus causing premature timeout of the address. The bug is now
824 PH/20 Added hosts_avoid_pipelining to the smtp transport.
826 PH/21 Long custom messages for fakedefer and fakereject are now split up
827 into multiline reponses in the same way that messages for "deny" and
828 other ACL rejections are.
830 PH/22 Applied Jori Hamalainen's speed-up changes and typo fixes to exigrep,
831 with slight modification.
833 PH/23 Applied sieve patches from the maintainer "tracking the latest notify
834 draft, changing the syntax and factoring some duplicate code".
836 PH/24 When the log selector "outgoing_port" was set, the port was shown as -1
837 for deliveries of the second and subsequent messages over the same SMTP
840 PH/25 Applied Magnus Holmgren's patch for ${addresses, ${map, ${filter, and
841 ${reduce, with only minor "tidies".
843 SC/02 Applied Daniel Tiefnig's patch to improve the '($parent) =' pattern match.
845 PH/26 Added a "continue" ACL modifier that does nothing, for the benefit of its
846 expansion side effects.
848 PH/27 When a message times out after an over-quota error from an Exim-imposed
849 quota, the bounce message says "mailbox is full". This message was not
850 being given when it was a system quota that was exceeded. It now should
853 MH/03 Made $recipients available in local_scan(). local_scan() already has
854 better access to the recipient list through recipients_list[], but
855 $recipients can be useful in postmaster-provided expansion strings.
857 PH/28 The $smtp_command and $smtp_command_argument variables were not correct
858 in the case of a MAIL command with additional options following the
859 address, for example: MAIL FROM:<foo@bar> SIZE=1234. The option settings
860 were accidentally chopped off.
862 PH/29 SMTP synchronization checks are implemented when a command is read -
863 there is a check that no more input is waiting when there shouldn't be
864 any. However, for some commands, a delay in an ACL can mean that it is
865 some time before the response is written. In this time, more input might
866 arrive, invalidly. So now there are extra checks after an ACL has run for
867 HELO/EHLO and after the predata ACL, and likewise for MAIL and RCPT when
868 pipelining has not been advertised.
870 PH/30 MH's patch to allow iscntrl() characters to be list separators.
872 PH/31 Unlike :fail:, a custom message specified with :defer: was not being
873 returned in the SMTP response when smtp_return_error_details was false.
876 PH/32 Change the Dovecot authenticator to use read() and write() on the socket
877 instead of the C I/O that was originally supplied, because problems were
880 PH/33 Compile failed with OpenSSL 0.9.8e. This was due to a coding error in
881 Exim which did not show up earlier: it was assuming that a call to
882 SSL_CTX_set_info_callback() might give an error value. In fact, there is
883 no error. In previous releases of OpenSSL, SSL_CTX_set_info_callback()
884 was a macro that became an assignment, so it seemed to work. This has
885 changed to a proper function call with a void return, hence the compile
886 error. Exim's code has been fixed.
888 PH/34 Change HDA_SIZE in oracle.c from 256 to 512. This is needed for 64-bit
891 PH/35 Applied a patch from the Sieve maintainer which fixes a bug in "notify".
893 PH/36 Applied John Jetmore's patch to add -v functionality to exigrep.
895 PH/37 If a message is not accepted after it has had an id assigned (e.g.
896 because it turns out to be too big or there is a timeout) there is no
897 "Completed" line in the log. When some messages of this type were
898 selected by exigrep, they were listed as "not completed". Others were
899 picked up by some special patterns. I have improved the selection
900 criteria to be more general.
902 PH/38 The host_find_failed option in the manualroute router can now be set
903 to "ignore", to completely ignore a host whose IP address cannot be
904 found. If all hosts are ignored, the behaviour is controlled by the new
905 host_all_ignored option.
907 PH/39 In a list of hosts for manualroute, if one item (either because of multi-
908 homing or because of multiple MX records with /mx) generated more than
909 one IP address, and the following item turned out to be the local host,
910 all the secondary addresses of the first item were incorrectly removed
911 from the list, along with the local host and any following hosts (which
912 is what is supposed to happen).
914 PH/40 When Exim receives a message, it writes the login name, uid, and gid of
915 whoever called Exim into the -H file. In the case of the daemon it was
916 behaving confusingly. When first started, it used values for whoever
917 started the daemon, but after a SIGHUP it used the Exim user (because it
918 calls itself on a restart). I have changed the code so that it now always
921 PH/41 (Following a suggestion from Tony Finch) If all the RCPT commands in a
922 message are rejected with the same error (e.g. no authentication or bad
923 sender address), and a DATA command is nevertheless sent (as can happen
924 with PIPELINING or a stupid MUA), the error message that was given to the
925 RCPT commands is included in the rejection of the DATA command. This is
926 intended to be helpful for MUAs that show only the final error to their
929 PH/42 Another patch from the Sieve maintainer.
931 SC/02 Eximstats - Differentiate between permanent and temporary rejects.
932 Eximstats - Fixed some broken HTML links and added missing column headers
934 Eximstats - Fixed Grand Total Summary Domains, Edomains, and Email
935 columns for Rejects, Temp Rejects, Ham, and Spam rows.
937 SC/03 Eximstats - V1.58 Fix to get <> and blackhole to show in edomain tables.
939 PH/43 Yet another patch from the Sieve maintainer.
941 PH/44 I found a way to check for a TCP/IP connection going away before sending
942 the response to the final '.' that terminates a message, but only in the
943 case where the client has not sent further data following the '.'
944 (unfortunately, this is allowed). However, in many cases there won't be
945 any further data because there won't be any more messages to send. A call
946 to select() can be used: if it shows that the input is "ready", there is
947 either input waiting, or the socket has been closed. An attempt to read
948 the next input character can distinguish the two cases. Previously, Exim
949 would have sent an OK response which the client would never have see.
950 This could lead to message repetition. This fix should cure that, at
951 least in a lot of common cases.
953 PH/45 Do not advertise STARTTLS in response to HELP unless it would be
954 advertised in response to EHLO.
960 PH/01 Two more bugs that were introduced by 4.64/PH/07, in addition to the one
961 fixed by 4.65/MH/01 (is this a record?) are fixed:
963 (i) An empty string was always treated as zero by the numeric comparison
964 operators. This behaviour has been restored.
966 (ii) It is documented that the numeric comparison operators always treat
967 their arguments as decimal numbers. This was broken in that numbers
968 starting with 0 were being interpreted as octal.
970 While fixing these problems I realized that there was another issue that
971 hadn't been noticed. Values of message_size_limit (both the global option
972 and the transport option) were treated as octal if they started with 0.
973 The documentation was vague. These values are now always treated as
974 decimal, and I will make that clear in the documentation.
980 TK/01 Disable default definition of HAVE_LINUX_SENDFILE. Clashes with
981 Linux large file support (_FILE_OFFSET_BITS=64) on older glibc
984 MH/01 Don't check that the operands of numeric comparison operators are
985 integers when their expansion is in "skipping" mode (fixes bug
986 introduced by 4.64-PH/07).
988 PH/01 If a system filter or a router generates more than SHRT_MAX (32767)
989 child addresses, Exim now panics and dies. Previously, because the count
990 is held in a short int, deliveries were likely to be lost. As such a
991 large number of recipients for a single message is ridiculous
992 (performance will be very, very poor), I have chosen to impose a limit
993 rather than extend the field.
999 TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
1000 leftover -K file (the existence of which was triggered by #402).
1001 While we were at it, introduced process PID as part of the -K
1002 filename. This should rule out race conditions when creating
1005 TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
1006 processing considerably. Previous code took too long for large mails,
1007 triggering a timeout which in turn triggers #401.
1009 TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
1010 in the DK code in transports.c. sendfile() is not really portable,
1011 hence the _LINUX specificness.
1013 TF/01 In the add_headers option to the mail command in an Exim filter,
1014 there was a bug that Exim would claim a syntax error in any
1015 header after the first one which had an odd number of characters
1018 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
1019 callout verification, Exim cached a "reject" for the entire domain. This
1020 is correct for most verifications, but it is not correct for a recipient
1021 verification with use_sender or use_postmaster set, because in that case
1022 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
1023 case of MAIL FROM:<> rejection from other early rejections (e.g.
1024 rejection of HELO). When verifying a recipient using a non-null MAIL
1025 address, the cache is ignored if it shows MAIL FROM:<> rejection.
1026 Whatever the result of the callout, the value of the domain cache is
1027 left unchanged (for any other kind of callout, getting as far as trying
1028 RCPT means that the domain itself is ok).
1030 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
1033 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
1034 manifest itself as EPIPE rather than ECONNECT. When tidying away a
1035 session, the daemon ignores ECONNECT errors and logs others; it now
1036 ignores EPIPE as well.
1038 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
1039 (quoted-printable decoding).
1041 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
1042 later the small subsequent patch to fix an introduced bug.
1044 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
1046 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
1048 PH/08 An error is now given if message_size_limit is specified negative.
1050 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
1051 to be given (somewhat) arbitrary names.
1053 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
1056 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
1057 miscellaneous code fixes
1059 PH/10 Added the log_reject_target ACL modifier to specify where to log
1062 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
1063 hostname. This is wrong, because it relates to the incoming message (and
1064 probably the interface on which it is arriving) and not to the outgoing
1065 callout (which could be using a different interface). This has been
1066 changed to use the value of the helo_data option from the smtp transport
1067 instead - this is what is used when a message is actually being sent. If
1068 there is no remote transport (possible with a router that sets up host
1069 addresses), $smtp_active_hostname is used.
1071 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
1072 tweaks were necessary in order to get it to work (see also 21 below):
1073 (a) The code assumed that strncpy() returns a negative number on buffer
1074 overflow, which isn't the case. Replaced with Exim's string_format()
1076 (b) There were several signed/unsigned issues. I just did the minimum
1077 hacking in of casts. There is scope for a larger refactoring.
1078 (c) The code used strcasecmp() which is not a standard C function.
1079 Replaced with Exim's strcmpic() function.
1080 (d) The code set only $1; it now sets $auth1 as well.
1081 (e) A simple test gave the error "authentication client didn't specify
1082 service in request". It would seem that Dovecot has changed its
1083 interface. Fortunately there's a specification; I followed it and
1084 changed what the client sends and it appears to be working now.
1086 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
1089 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
1090 address A is aliased to B and C, where B exists and C does not. Without
1091 -v the output is "A verified" because verification stops after a
1092 successful redirection if more than one address is generated. However,
1093 with -v the child addresses are also verified. Exim was outputting "A
1094 failed to verify" and then showing the successful verification for C,
1095 with its parentage. It now outputs "B failed to verify", showing B's
1096 parentage before showing the successful verification of C.
1098 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
1099 look up a TXT record in a specific list after matching in a combined
1102 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
1103 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
1104 they consult the DNS. I had assumed they would set it the way they
1105 wanted; and indeed my experiments on Linux seem to show that in some
1106 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
1107 To be on the safe side, however, I have now made the interface to
1108 host_find_byname() similar to host_find_bydns(), with an argument
1109 containing the DNS resolver options. The host_find_byname() function now
1110 sets these options at its start, just as host_find_bydns() does. The smtp
1111 transport options dns_qualify_single and dns_search_parents are passed to
1112 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
1113 of host_find_byname() use the default settings of RES_DEFNAMES
1114 (qualify_single) but not RES_DNSRCH (search_parents).
1116 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
1117 spool_read_header() do less string testing, by means of a preliminary
1118 switch on the second character of optional "-foo" lines. (This is
1119 overdue, caused by the large number of possibilities that now exist.
1120 Originally there were few.) While I was there, I also converted the
1121 str(n)cmp tests so they don't re-test the leading "-" and the first
1122 character, in the hope this might squeeze out yet more improvement.
1124 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
1125 flag allowing group syntax was set by the header_syntax check but not
1126 turned off, possible causing trouble later; (2) The flag was not being
1127 set at all for the header_verify test, causing "group"-style headers to
1128 be rejected. I have now set it in this case, and also caused header_
1129 verify to ignore an empty address taken from a group. While doing this, I
1130 came across some other cases where the code for allowing group syntax
1131 while scanning a header line wasn't quite right (mostly, not resetting
1132 the flag correctly in the right place). These bugs could have caused
1133 trouble for malformed header lines. I hope it is now all correct.
1135 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
1136 with the "reply" argument non-NULL. The code, however (which originally
1137 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
1138 but it didn't always do it. This confused somebody who was copying the
1139 code for some other use. I have removed all the tests.
1141 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
1142 feature that was used to support insecure browsers during the U.S. crypto
1143 embargo. It requires special client support, and Exim is probably the
1144 only MTA that supported it -- and would never use it because real RSA is
1145 always available. This code has been removed, because it had the bad
1146 effect of slowing Exim down by computing (never used) parameters for the
1147 RSA_EXPORT functionality.
1149 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
1150 authenticator to fail if there's a tab character in the incoming data
1151 (there should never be unless someone is messing about, as it's supposed
1152 to be base64-encoded). Also added, on Timo's advice, the "secured" option
1153 if the connection is using TLS or if the remote IP is the same as the
1154 local IP, and the "valid-client-cert option" if a client certificate has
1157 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
1158 authenticators. This can be used for authorization after authentication
1159 succeeds. (In the case of plaintext, it servers for both authentication
1162 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
1163 if any retry times were supplied.
1165 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
1166 connection, where there is no client IP address to check. In this
1167 situation, the verify now always succeeds.
1169 PH/25 Applied John Jetmore's -Mset patch.
1171 PH/26 Added -bem to be like -Mset, but loading a message from a file.
1173 PH/27 In a string expansion for a processed (not raw) header when multiple
1174 headers of the same name were present, leading whitespace was being
1175 removed from all of them, but trailing whitespace was being removed only
1176 from the last one. Now trailing whitespace is removed from each header
1177 before concatenation. Completely empty headers in a concatenation (as
1178 before) are ignored.
1180 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
1181 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
1183 PH/29 [Removed. This was a change that I later backed out, and forgot to
1184 correct the ChangeLog entry (that I had efficiently created) before
1185 committing the later change.]
1187 PH/30 Exim was sometimes attempting to deliver messages that had suffered
1188 address errors (4xx response to RCPT) over the same connection as other
1189 messages routed to the same hosts. Such deliveries are always "forced",
1190 so retry times are not inspected. This resulted in far too many retries
1191 for the affected addresses. The effect occurred only when there were more
1192 hosts than the hosts_max_try setting in the smtp transport when it had
1193 the 4xx errors. Those hosts that it had tried were not added to the list
1194 of hosts for which the message was waiting, so if all were tried, there
1195 was no problem. Two fixes have been applied:
1197 (i) If there are any address or message errors in an SMTP delivery, none
1198 of the hosts (tried or untried) are now added to the list of hosts
1199 for which the message is waiting, so the message should not be a
1200 candidate for sending over the same connection that was used for a
1201 successful delivery of some other message. This seems entirely
1202 reasonable: after all the message is NOT "waiting for some host".
1203 This is so "obvious" that I'm not sure why it wasn't done
1204 previously. Hope I haven't missed anything, but it can't do any
1205 harm, as the worst effect is to miss an optimization.
1207 (ii) If, despite (i), such a delivery is accidentally attempted, the
1208 routing retry time is respected, so at least it doesn't keep
1209 hammering the server.
1211 PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
1212 in ${readsocket because some servers need this prod.
1214 PH/32 Added some extra debug output when updating a wait-xxx database.
1216 PH/33 The hint "could be header name not terminated by colon", which has been
1217 given for certain expansion errors for a long time, was not being given
1218 for the ${if def:h_colon_omitted{... case.
1220 PH/34 The spec says: "With one important exception, whenever a domain list is
1221 being scanned, $domain contains the subject domain." There was at least
1222 one case where this was not true.
1224 PH/35 The error "getsockname() failed: connection reset by peer" was being
1225 written to the panic log as well as the main log, but it isn't really
1226 panic-worthy as it just means the connection died rather early on. I have
1227 removed the panic log writing for the ECONNRESET error when getsockname()
1230 PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
1231 runs only) independently of the message's sender address. This meant
1232 that, if the 4xx error was in fact related to the sender, a different
1233 message to the same recipient with a different sender could confuse
1234 things. In particualar, this can happen when sending to a greylisting
1235 server, but other circumstances could also provoke similar problems.
1236 I have changed the default so that the retry time for these errors is now
1237 based a combination of the sender and recipient addresses. This change
1238 can be overridden by setting address_retry_include_sender=false in the
1241 PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
1242 remote server are returned as part of bounce messages. This was not
1243 happening for LMTP over a pipe (the lmtp transport), but now it is the
1244 same for both kinds of LMTP.
1246 PH/38 Despite being documented as not happening, Exim was rewriting addresses
1247 in header lines that were in fact CNAMEs. This is no longer the case.
1249 PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
1250 and queue runs started by the daemon processed all messages. This has
1251 been fixed so that -R and -S can now usefully be given with -q<time>.
1253 PH/40 Import PCRE release 6.7 (fixes some bugs).
1255 PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
1257 PH/42 Give an error if -q is specified more than once.
1259 PH/43 Renamed the variables $interface_address and $interface_port as
1260 $received_ip_address and $received_port, to make it clear that these
1261 values apply to message reception, and not to the outgoing interface when
1262 a message is delivered. (The old names remain recognized, of course.)
1264 PH/44 There was no timeout on the connect() call when using a Unix domain
1265 socket in the ${readsocket expansion. There now is.
1267 PH/45 Applied a modified version of Brad Jorsch's patch to allow "message" to
1268 be meaningful with "accept".
1270 SC/01 Eximstats V1.43
1271 Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
1273 SC/02 Eximstats V1.44
1274 Use a glob alias rather than an array ref in the generated
1275 parser. This improves both readability and performance.
1277 SC/03 Eximstats V1.45 (Marco Gaiarin / Steve Campbell)
1278 Collect SpamAssassin and rejection statistics.
1279 Don't display local sender or destination tables unless
1280 there is data to show.
1281 Added average volumes into the top table text output.
1283 SC/04 Eximstats V1.46
1284 Collect data on the number of addresses (recipients)
1285 as well as the number of messages.
1287 SC/05 Eximstats V1.47
1288 Added 'Message too big' to the list of mail rejection
1289 reasons (thanks to Marco Gaiarin).
1291 SC/06 Eximstats V1.48
1292 Mainlog lines which have GMT offsets and are too short to
1293 have a flag are now skipped.
1295 SC/07 Eximstats V1.49 (Alain Williams)
1296 Added the -emptyok flag.
1298 SC/08 Eximstats V1.50
1299 Fixes for obtaining the IP address from reject messages.
1301 JJ/03 exipick.20061117.2, made header handling as similar to exim as possible
1302 (added [br]h_ prefixes, implemented RFC2047 decoding. Fixed
1303 whitesspace changes from 4.64-PH/27
1305 JJ/04 exipick.20061117.2, fixed format and added $message_headers_raw to
1308 JJ/05 exipick.20061117.2, bug fixes (error out sooner when invalid criteria
1309 are found, allow negative numbers in numeric criteria)
1311 JJ/06 exipick.20061117.2, added new $message_body_missing variable
1313 JJ/07 exipick.20061117.2, added $received_ip_address and $received_port
1314 to match changes made in 4.64-PH/43
1316 PH/46 Applied Jori Hamalainen's patch to add features to exiqsumm.
1318 PH/47 Put in an explicit test for a DNS lookup of an address record where the
1319 "domain" is actually an IP address, and force a failure. This locks out
1320 those revolvers/nameservers that support "A-for-A" lookups, in
1321 contravention of the specifications.
1323 PH/48 When a host name was looked up from an IP address, and the subsequent
1324 forward lookup of the name timed out, the host name was left in
1325 $sender_host_name, contrary to the specification.
1327 PH/49 Although default lookup types such as lsearch* or cdb*@ have always been
1328 restricted to single-key lookups, Exim was not diagnosing an error if
1329 * or *@ was used with a query-style lookup.
1331 PH/50 Increased the value of DH_BITS in tls-gnu.c from 768 to 1024.
1333 MH/01 local_scan ABI version incremented to 1.1. It should have been updated
1334 long ago, but noone interested enough thought of it. Let's just say that
1335 the "1.1" means that there are some new functions that weren't there at
1336 some point in the past.
1338 PH/51 Error processing for expansion failure of helo_data from an smtp
1339 transport during callout processing was broken.
1341 PH/52 Applied John Jetmore's patch to allow tls-on-connect and STARTTLS to be
1342 tested/used via the -bh/-bhc/-bs options.
1344 PH/53 Added missing "#include <time.h>" to pcre/pcretest.c (this was a PCRE
1345 bug, fixed in subsequent PCRE releases).
1347 PH/54 Applied Robert Bannocks' patch to avoid a problem with references that
1348 arises when using the Solaris LDAP libraries (but not with OpenLDAP).
1350 PH/55 Check for a ridiculously long file name in exim_dbmbuild.
1356 SC/01 Use a glob alias rather than an array ref in eximstats generated
1357 parser. This improves both readability and performance.
1359 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
1360 Don't display local sender or destination tables in eximstats unless
1361 there is data to show.
1362 Added average volumes into the eximstats top table text output.
1364 SC/03 Collect data on the number of addresses (recipients) as well
1365 as the number of messages in eximstats.
1367 TF/01 Correct an error in the documentation for the redirect router. Exim
1368 does (usually) call initgroups() when daemonizing.
1370 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
1371 with consistent privilege compared to when running as a daemon.
1373 TF/03 Note in the spec that $authenticated_id is not set for local
1374 submissions from trusted users.
1376 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
1377 Thanks to Dean Brooks <dean@iglou.com> for the patch.
1379 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
1380 by adding some example configuration directives to the default
1381 configuration file. A little bit of work is required to uncomment the
1382 directives and define how usernames and passwords are checked, but
1383 there is now a framework to start from.
1385 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
1386 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
1387 without this. I don't know how relevant this is to other LDAP libraries.
1389 PH/02 Add the verb name to the "unknown ACL verb" error.
1391 PH/03 Magnus Holmgren's patch for filter_prepend_home.
1393 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
1395 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
1396 directory not expanded when it should be if an expanded home directory
1397 was set for the address (which is overridden by the transport).
1399 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
1402 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
1403 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
1404 because it is too late at that time, and has no effect.
1406 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
1407 security issue with \' (bugzilla #107). I could not use the
1408 PQescapeStringConn() function, because it needs a PGconn value as one of
1411 PH/08 When testing addresses using -bt, indicate those final addresses that
1412 are duplicates that would not cause an additional delivery. At least one
1413 person was confused, thinking that -bt output corresponded to deliveries.
1414 (Suppressing duplicates isn't a good idea as you lose the information
1415 about possibly different redirections that led to the duplicates.)
1417 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
1418 systems where poll() doesn't work, in particular OS X.
1420 PH/10 Added more information to debugging output for retry time not reached.
1422 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
1423 operations in malware.c.
1425 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
1428 PH/13 If write_rejectlog was set false when logging was sent to syslog with
1429 syslog_duplication set false, log lines that would normally be written
1430 both the the main log and to the reject log were not written to syslog at
1433 PH/14 In the default configuration, change the use of "message" in ACL warn
1434 statements to "add_header".
1436 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
1437 not followed by a command (e.g. "seen endif").
1439 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
1440 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
1443 PH/17 Added extra conditions to the default value of delay_warning_condition
1447 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
1448 { match{$h_precedence:}{(?i)bulk|list|junk} } \
1449 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
1452 The Auto-Submitted: and various List- headers are standardised, whereas I
1453 don't think Precedence: ever was.
1455 PH/18 Refactored debugging code in route_finduser() to show more information,
1456 in particular, the error code if getpwnam() issues one.
1458 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
1459 This is apparently needed in addition to the PH/07 change above to avoid
1460 any possible encoding problems.
1462 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
1463 but not after initializing Perl.
1465 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
1466 output them only if debugging. By default they are written stderr,
1467 apparently, which is not desirable.
1469 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
1472 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
1475 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
1477 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
1478 authenticated or an ident call has been made. Suppress the default
1479 values for $authenticated_id and $authenticated_sender (but permit -oMai
1480 and -oMas) when testing with -bh.
1482 PH/24 Re-jigged the order of the tests in the default configuration so that the
1483 tests for valid domains and recipients precede the DNS black list and CSA
1484 tests, on the grounds that those ones are more expensive.
1486 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
1487 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
1488 This bug exists in every version of Exim that I still have, right back to
1491 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
1492 However, an attempt to turn on case-sensitivity in a regex key by
1493 including (?-i) didn't work because the subject string was already
1494 lowercased, and the effects were non-intuitive. It turns out that a
1495 one-line patch can be used to allow (?-i) to work as expected.
1501 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
1502 other effects) broke the use of negated acl sub-conditions.
1504 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
1507 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
1508 "Deny" causes Exim to reject the incoming connection with a 554 error.
1509 Unfortunately, if there is a major crisis, such as a disk failure,
1510 tcp-wrappers gives "deny", whereas what one would like would be some
1511 kind of temporary error. A kludge has been added to help with this.
1512 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
1513 554 error is used if errno is still zero or contains ENOENT (which occurs
1514 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
1517 PH/03 Add -lutil to the default FreeBSD LIBS setting.
1519 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
1520 errors. Otherwise a message that provokes a temporary error (when other
1521 messages do not) can cause a whole host to time out.
1523 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
1524 addresses were routed directly to files or pipes from a redirect router.
1525 File deliveries just didn't batch; pipe deliveries might have suffered
1528 PH/06 A failure to get a lock for a hints database would erroneously always say
1529 "Failed to get write lock", even when it was really a read lock.
1531 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
1532 of 0600. This has been changed to use the value of the lockfile_mode
1533 option (which defaults to 0600).
1535 PH/08 Applied small patch from the Sieve maintainer.
1537 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
1538 folder from quota calculations, a direct delivery into this folder messed
1539 up the contents of the maildirsize file. This was because the regex was
1540 used only to exclude .Trash (or whatever) when the size of the mailbox
1541 was calculated. There was no check that a delivery was happening into an
1542 excluded directory. This bug has been fixed by ignoring all quota
1543 processing for deliveries into excluded directories.
1545 PH/10 Added the maildirfolder_create_regex option to appendfile.
1551 PH/01 The code for finding all the local interface addresses on a FreeBSD
1552 system running IPv6 was broken. This may well have applied to all BSD
1553 systems, as well as to others that have similar system calls. The broken
1554 code found IPv4 interfaces correctly, but gave incorrect values for the
1555 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
1556 that it would not match correctly against @[] and not recognize the IPv6
1559 PH/02 The ipliteral router was not recognizing addresses of the form user@
1560 [ipv6:....] because it didn't know about the "ipv6:" prefix.
1562 PH/03 Added disable_ipv6.
1564 PH/04 Changed $reply_address to use the raw form of the headers instead of the
1565 decoded form, because it is most often used to construct To: headers
1566 lines in autoreplies, and the decoded form may well be syntactically
1567 invalid. However, $reply_address has leading white space removed, and all
1568 newlines turned into spaces so that the autoreply transport does not
1571 PH/05 If group was specified without a user on a router, and no group or user
1572 was specified on a transport, the group from the router was ignored.
1574 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
1575 for visible compile-time settings that can be used to change these
1576 numbers, for those that want even more. Backwards compatibility with old
1577 spool files has been maintained. However, going back to a previous Exim
1578 release will lost any variables that are in spool files.
1580 PH/07 Two small changes when running in the test harness: increase delay when
1581 passing a TCP/IP connection to a new process, in case the original
1582 process has to generate a bounce, and remove special handling of
1583 127.0.0.2 (sic), which is no longer necessary.
1585 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
1586 be the same on different OS.
1588 PH/09 Moved a debug statement in filter processing to avoid a race problem when
1591 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
1592 whether --show-vars was specified or not
1594 JJ/02 exipick: Added support for new ACL variable spool format introduced
1597 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
1598 syntactically invalid From: or Reply-to: line, and a filter used this to
1599 generate an autoreply, and therefore failed to obtain an address for the
1600 autoreply, Exim could try to deliver to a non-existent relative file
1601 name, causing unrelated and misleading errors. What now happens is that
1602 it logs this as a hard delivery error, but does not attempt to create a
1605 PH/11 The exinext utility has a -C option for testing purposes, but although
1606 the given file was scanned by exinext itself; it wasn't being passed on
1607 when Exim was called.
1609 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
1610 an end-of-file indication when reading a command response.
1612 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
1613 compiled. In many other places in Exim, IPv6 addresses are always
1614 recognized, so I have changed this. It also means that IPv4 domain
1615 literals of the form [IPV4:n.n.n.n] are now always recognized.
1617 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
1618 used if the router is not running as root, for example, when verifying at
1619 ACL time, or when using -bh. The debugging output from this situation was
1620 non-existent - all you got was a failure to exec. I have made two
1623 (a) Failures to set uid/gid, the current directory, or a process leader
1624 in a subprocess such as that created by queryprogram now generate
1625 suitable debugging ouput when -d is set.
1627 (b) The queryprogram router detects when it is not running as root,
1628 outputs suitable debugging information if -d is set, and then runs
1629 the subprocess without attempting to change uid/gid.
1631 PH/15 Minor change to Makefile for building test_host (undocumented testing
1634 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
1635 additional section of a DNS packet that returns MX or SRV records.
1636 Instead, it always explicitly searches for A/AAAA records. This avoids
1637 major problems that occur when a DNS server includes only records of one
1638 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
1639 fixed another bug: if SRV records were looked up and the corresponding
1640 address records were *not* found in the additional section, the port
1641 values from the SRV records were lost.
1643 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
1644 using the correct key (the original address) when searching the retry
1645 rules in order to find which one to use for generating the retry hint.
1647 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
1648 adding the default one. Similarly, if it contains a Reply-To: header, the
1649 errors_reply_to option, if set, is not used.
1651 PH/19 When calculating a retry time, Exim used to measure the "time since
1652 failure" by looking at the "first failed" field in the retry record. Now
1653 it does not use this if it is later than than the arrival time of the
1654 message. Instead it uses the arrival time. This makes for better
1655 behaviour in cases where some deliveries succeed, thus re-setting the
1656 "first failed" field. An example is a quota failure for a huge message
1657 when small messages continue to be delivered. Without this change, the
1658 "time since failure" will always be short, possible causing more frequent
1659 delivery attempts for the huge message than are intended.
1660 [Note: This change was subsequently modified - see PH/04 for 4.62.]
1662 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
1663 $1, $2, $3) because the numerical variables can be reset during some
1664 expansion items (e.g. "match"), thereby losing the authentication data.
1666 PH/21 Make -bV show the size of off_t variables so that the test suite can
1667 decide whether to run tests for quotas > 2G.
1669 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
1670 mailbox_size, and mailbox_filecount in the appendfile transport. If a
1671 filecount value is greater than 2G or if a quota value is greater than 2G
1672 on a system where the size of off_t is not greater than 4, a panic error
1675 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
1676 never match. The debug and -bh output now contains an explicit error
1677 message indicating a malformed IPv4 address or mask.
1679 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
1680 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
1681 PH/23 above applies.
1683 PH/25 Do not write to syslog when running in the test harness. The only
1684 occasion when this arises is a failure to open the main or panic logs
1685 (for which there is an explicit test).
1687 PH/26 Added the /no_tell option to "control=freeze".
1689 PH/27 If a host name lookup failed very early in a connection, for example, if
1690 the IP address matched host_lookup and the reverse lookup yielded a name
1691 that did not have a forward lookup, an error message of the form "no IP
1692 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
1693 could be logged. Now it outputs the IP address instead of "NULL".
1695 PH/28 An enabling patch from MH: add new function child_open_exim2() which
1696 allows the sender and the authenticated sender to be set when
1697 submitting a message from within Exim. Since child_open_exim() is
1698 documented for local_scan(), the new function should be too.
1700 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
1701 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
1702 results in an empty string is now treated as unset.
1704 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
1706 PH/31 Added sender_verify_fail logging option.
1708 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
1709 needed by RFC 822 but not by RFC 2822 was commented out. I have now
1710 tidied the source and removed it altogether.
1712 PH/33 When a queue run was abandoned because the load average was too high, a
1713 log line was always written; now it is written only if the queue_run log
1714 selector is set. In addition, the log line for abandonment now contains
1715 information about the queue run such as the pid. This is always present
1716 in "start" and "stop" lines but was omitted from the "abandon" line.
1718 PH/34 Omit spaces between a header name and the colon in the error message that
1719 is given when verify = headers_syntax fails (if there are lots of them,
1720 the message gets confusing).
1722 PH/35 Change the default for dns_check_names_pattern to allow slashes within
1723 names, as there are now some PTR records that contain slashes. This check
1724 is only to protect against broken name servers that fall over on strange
1725 characters, so the fact that it applies to all lookups doesn't matter.
1727 PH/36 Now that the new test suite is complete, we can remove some of the
1728 special code in Exim that was needed for the old test suite. For example,
1729 sorting DNS records because real resolvers return them in an arbitrary
1730 order. The new test suite's fake resolver always returns records in the
1733 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
1734 bounces) except when queue_only is set, to avoid logging races between
1735 the different processes.
1737 PH/38 Panic-die if .include specifies a non-absolute path.
1739 PH/39 A tweak to the "H" retry rule from its user.
1741 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
1742 a label. They prevented compilation on older perls.
1744 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
1745 a warning to be raised on newish perls.
1747 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
1748 on queue. Changes to match documented behaviour of showing count of
1749 messages matching specified criteria.
1751 PH/40 Changed the default ident timeout from 30s to 5s.
1753 PH/41 Added support for the use of login_cap features, on those BSD systems
1754 that have them, for controlling the resources used by pipe deliveries.
1756 PH/42 The content-scanning code uses fopen() to create files in which to put
1757 message data. Previously it was not paying any attention to the mode of
1758 the files. Exim runs with umask(0) because the rest of the code creates
1759 files with open(), and sets the required mode explicitly. Thus, these
1760 files were ending up world-writeable. This was not a big issue, because,
1761 being within the spool directory, they were not world-accessible. I have
1762 created a function called modefopen, which takes an additional mode
1763 argument. It sets umask(777), creates the file, chmods it to the required
1764 mode, then resets the umask. All the relevant calls to fopen() in the
1765 content scanning code have been changed to use this function.
1767 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
1768 to 24 hours. This avoids potential overflow problems when processing G
1769 and H retry rules. I suspect nobody ever tinkers with this value.
1771 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
1773 PH/45 When the plaintext authenticator is running as a client, the server's
1774 challenges are checked to ensure they are valid base64 strings. By
1775 default, the authentication attempt is cancelled if an invalid string is
1776 received. Setting client_ignore_invalid_base64 true ignores these errors.
1777 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
1778 they are received. Thus, the responses can be made to depend on the
1779 challenges. If an invalid string is ignored, an empty string is placed in
1782 PH/46 Messages that are created by the autoreply transport now contains a
1783 References: header, in accordance with RFCs 2822 and 3834.
1785 PH/47 Added authenticated_sender_force to the smtp transport.
1787 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
1789 PH/49 Installed latest patch from the Sieve maintainer.
1791 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
1792 was also set, the appendfile transport was unnecessarily scanning a
1793 directory of message files (e.g. for maildir delivery) to find the count
1794 of files (along with the size), even though it did not need this
1795 information. It now does the scan only if it needs to find either the
1796 size of the count of files.
1798 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
1800 PH/52 Two bugs concerned with error handling when the smtp transport is
1803 (i) Exim was not creating retry information for temporary errors given
1804 for individual recipients after the DATA command when the smtp transport
1805 was used in LMTP mode. This meant that they could be retried too
1806 frequently, and not timed out correctly.
1808 (ii) Exim was setting the flag that allows error details to be returned
1809 for LMTP errors on RCPT commands, but not for LMTP errors for individual
1810 recipients that were returned after the DATA command.
1812 PH/53 This is related to PH/52, but is more general: for any failing address,
1813 when detailed error information was permitted to be returned to the
1814 sender, but the error was temporary, then after the final timeout, only
1815 "retry timeout exceeded" was returned. Now it returns the full error as
1816 well as "retry timeout exceeded".
1818 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
1819 do this, and (what is worse) MTAs that accept it.
1821 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
1822 will now be deprecated.
1824 PH/56 New os.c-cygwin from the Cygwin maintainer.
1826 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
1827 formats (previously only available in exim formats via -bpr, -bpru,
1828 and -bpra. Now also available in native and exiqgrep formats)
1830 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
1831 with very large, slow to parse queues
1833 JJ/08 exipick: added ! as generic prefix to negate any criteria format
1835 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
1837 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
1838 responses to authentication challenges, though it was showing the
1839 challenges; (ii) I've removed the CR characters from the debug output for
1842 PH/58 Allow for the insertion of a newline as well as a space when a string
1843 is turned into more than one encoded-word during RFC 2047 encoding. The
1844 Sieve code now uses this.
1846 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
1847 data_4xx, lost_connection, tls_required.
1849 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
1850 message was being sent as an SMTP response.
1852 PH/61 Add -l and -k options to exicyclog.
1854 PH/62 When verifying, if an address was redirected to one new address, so that
1855 verification continued, and the new address failed or deferred after
1856 having set something in $address_data, the value of $address_data was not
1857 passed back to the ACL. This was different to the case when no
1858 redirection occurred. The value is now passed back in both cases.
1860 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
1861 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
1862 use login_cap.h, so on its own it isn't the distinguishing feature. The
1863 new name refers directly to the setclassresources() function.
1865 PH/65 Added configuration files for NetBSD3.
1867 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
1869 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
1870 is preferred over IPv4.
1872 PH/68 The bounce_return_message and bounce_return_body options were not being
1873 honoured for bounces generated during the reception of non-SMTP messages.
1874 In particular, this applied to messages rejected by the ACL. This bug has
1875 been fixed. However, if bounce_return_message is true and bounce_return_
1876 body is false, the headers that are returned for a non-SMTP message
1877 include only those that have been read before the error was detected.
1878 (In the case of an ACL rejection, they have all been read.)
1880 PH/69 The HTML version of the specification is now built in a directory called
1881 spec_html instead of spec.html, because the latter looks like a path with
1882 a MIME-type, and this confuses some software.
1884 PH/70 Catch two compiler warnings in sieve.c.
1886 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
1887 function verify_get_ident() calls ip_connect() to connect a socket, but
1888 if the "connect()" function timed out, ip_connect() used to close the
1889 socket. However, verify_get_ident() also closes the socket later, and in
1890 between Exim writes to the log, which may get opened at this point. When
1891 the socket was closed in ip_connect(), the log could get the same file
1892 descriptor number as the socket. This naturally causes chaos. The fix is
1893 not to close the socket in ip_connect(); the socket should be closed by
1894 the function that creates it. There was only one place in the code where
1895 this was missing, in the iplookup router, which I don't think anybody now
1896 uses, but I've fixed it anyway.
1898 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
1899 well as to direct DNS lookups. Otherwise the handling of names in host
1900 lists is inconsistent and therefore confusing.
1906 PH/01 Two changes to the default runtime configuration:
1908 (1) Move the checks for relay_from_hosts and authenticated clients from
1909 after to before the (commented out) DNS black list checks.
1911 (2) Add control=submission to the relay_from_hosts and authenticated
1912 clients checks, on the grounds that messages accepted by these
1913 statements are most likely to be submissions.
1915 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
1917 (1) Generate an error if the third argument for the ${prvs expansion is
1920 (2) Treat a missing third argument of ${prvscheck as if it were an empty
1923 (3) Reset the variables that are obtained from the first argument of
1924 ${prvscheck and used in the second argument before leaving the code,
1925 because their memory is reclaimed, so using them afterwards may do
1928 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
1929 one (it's much easier than Tom thought :-).
1931 (5) Because of (4), we can now allow for the use of $prvscheck_result
1932 inside the third argument.
1934 PH/03 For some reason, the default setting of PATH when running a command from
1935 a pipe transport was just "/usr/bin". I have changed it to
1938 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
1939 anything to be listed in the output from -bV.
1941 PH/05 When a filter generated an autoreply, the entire To: header line was
1942 quoted in the delivery log line, like this:
1944 => >A.N.Other <ano@some.domain> <original@ddress> ...
1946 This has been changed so that it extracts the operative address. There
1947 may be more than one such address. If so, they are comma-separated, like
1950 => >ano@some.domain,ona@other.domain <original@ddress> ...
1952 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
1953 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
1954 not being looked up in the rDNS to get a host name, Exim was showing the
1955 IP address twice in Received: lines, even though the IP addresses were
1956 identical. For example:
1958 Received: from [1.2.3.4] (helo=[1.2.3.4])
1960 However, if the real host name was known, it was omitting the HELO data
1961 if it matched the actual IP address. This has been tidied up so that it
1962 doesn't show the same IP address twice.
1964 PH/07 When both +timestamp and +memory debugging was on, the value given by
1965 $tod_xxx expansions could be wrong, because the tod_stamp() function was
1966 called by the debug printing, thereby overwriting the timestamp buffer.
1967 Debugging no longer uses the tod_stamp() function when +timestamp is set.
1969 PH/08 When the original message was included in an autoreply transport, it
1970 always said "this is a copy of the message, including all the headers",
1971 even if body_only or headers_only was set. It now gives an appropriate
1974 PH/09 Applied a patch from the Sieve maintainer which:
1976 o fixes some comments
1977 o adds the (disabled) notify extension core
1978 o adds some debug output for the result of if/elsif tests
1979 o points to the current vacation draft in the documentation
1980 and documents the missing references header update
1984 o fixes a bug in processing the envelope test (when testing
1985 multiple envelope elements, the last element determinted the
1988 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
1989 Electronic Mail") by including:
1991 Auto-submitted: auto-generated
1993 in the messages that it generates (bounce messages and others, such as
1994 warnings). In the case of bounce messages for non-SMTP mesages, there was
1995 also a typo: it was using "Auto_submitted" (underscore instead of
1996 hyphen). Since every message generated by Exim is necessarily in response
1997 to another message, thes have all been changed to:
1999 Auto-Submitted: auto-replied
2001 in accordance with these statements in the RFC:
2003 The auto-replied keyword:
2005 - SHOULD be used on messages sent in direct response to another
2006 message by an automatic process,
2008 - MUST NOT be used on manually-generated messages,
2010 - MAY be used on Delivery Status Notifications (DSNs) and Message
2011 Disposition Notifications (MDNs),
2013 - MUST NOT be used on messages generated by automatic or periodic
2014 processes, except for messages which are automatic responses to
2017 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
2018 to the default Received: header definition.
2020 PH/12 Added log selector acl_warn_skipped (default on).
2022 PH/13 After a successful wildlsearch lookup, discard the values of numeric
2023 variables because (a) they are in the wrong storage pool and (b) even if
2024 they were copied, it wouldn't work properly because of the caching.
2026 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
2027 checking when decoding. Apparently there are clients that generate
2028 overlong encoded strings. Why am I not surprised?
2030 PH/15 If the first argument of "${if match_address" was not empty, but did not
2031 contain an "@" character, Exim crashed. Now it writes a panic log message
2032 and treats the condition as false.
2034 PH/16 In autoreply, treat an empty string for "once" the same as unset.
2036 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
2037 extension "envelope-auth". The code is finished and in agreement with
2038 other implementations, but there is no documentation so far and in fact,
2039 nobody wrote the draft yet. This extension is currently #undef'ed, thus
2040 not changing the active code.
2042 Print executed "if" and "elsif" statements when debugging is used. This
2043 helps a great deal to understand what a filter does.
2045 Document more things not specified clearly in RFC3028. I had all this
2046 sorted out, when out of a sudden new issues came to my mind. Oops."
2048 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
2051 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
2052 canonical form (as documented). However, after a host name lookup from
2053 the IP address, check_host() was doing a simple string comparison with
2054 addresses acquired from the DNS when checking that the found name did
2055 have the original IP as one of its addresses. Since any found IPv6
2056 addresses are likely to be in abbreviated form, the comparison could
2057 fail. Luckily, there already exists a function for doing the comparison
2058 by converting both addresses to binary, so now that is used instead of
2059 the text comparison.
2061 PH/20 There was another similar case to PH/19, when a complete host name was
2062 given in a host list; looking up its IP address could give an abbreviated
2063 form, whereas the current host's name might or might not be abbreviated.
2064 The same fix has been applied.
2070 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
2071 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
2074 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
2075 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
2077 PH/03 Typo: missing ".o" in src/pcre/Makefile.
2079 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
2080 header line, restrict the check to what is listed in RFCs 2369 and 2929.
2081 Also, for "Auto-Submitted", treat anything other than "no" as
2082 non-personal, in accordance with RFC 3834. (Previously it treated
2083 anything starting "auto-" as non-personal.)
2085 TF/01 The control=submission/name=... option had a problem with syntax
2086 errors if the name included a slash character. The /name= option
2087 now slurps the rest of the string, so it can include any characters
2088 but it must come last in the list of options (after /sender_retain
2091 PH/05 Some modifications to the interface to the fake nameserver for the new
2099 TK/01 Added the "success_on_redirect" address verification option. See
2100 NewStuff for rationale and an example.
2102 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
2104 PH/02 Patch to exigrep to allow it to work on syslog lines.
2106 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
2107 fread() to skip over the body file's header line, because in Cygwin the
2108 header line is locked and is inaccessible.
2110 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
2111 co-exist for some time) to make it clear that it is the Exim ID that is
2112 referenced, not the Message-ID: header line.
2114 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
2115 string_format() function, because snprintf() does not exist on all
2118 PH/06 The use of forbid_filter_existstest now also locks out the use of the
2119 ${stat: expansion item.
2121 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
2122 protocol synchronization error", to keep the pedants happy.
2124 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
2125 well as for IRIX systems, when gcc is being used. See the host.c source
2128 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
2130 PH/10 Named domain lists were not working if used in a queue_smtp_domains
2133 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
2134 transport and to the smtp transport in LMTP mode.
2136 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
2138 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
2139 run a filter in a subprocess. This could lead to confusion in subsequent
2140 lookups in the parent process. There should also be a search_tidyup() at
2141 the end of the subprocess.
2143 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
2144 only if the host matched helo_try_verify_hosts, which caused the
2145 verification to occur when the EHLO/HELO command was issued. The ACL just
2146 tested the remembered result. Now, if a previous verification attempt has
2147 not happened, "verify = helo" does it there and then.
2149 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
2151 TK/03 Fix log output including CR from clamd.
2153 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
2154 exist provoked a memory error which could cause a segfault.
2156 PH/15 Installed PCRE 6.2
2158 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
2160 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
2161 of the problem. Specifically, suggested +O2 rather than +O1 for the
2164 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
2166 PH/20 If a delivery was routed to a non-standard port by means of an SRV
2167 record, the port was not correctly logged when the outgoing_port log
2168 selector was set (it logged the transort's default port).
2170 PH/21 Added support for host-specific ports to manualroute, queryprogram,
2171 fallback_hosts, and "hosts" in the smtp transport.
2173 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
2174 host errors such as "Connection refused".
2176 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
2177 authentication with radiusclient 0.4.9:
2179 - Error returned from rc_read_config was caught wrongly
2180 - Username/password not passed on to radius server due to wrong length.
2182 The presumption is that some radiusclient API changes for 4.51/PH/17
2183 were not taken care of correctly. The code is still untested by me (my
2184 Linux distribution still has 0.3.2 of radiusclient), but it was
2185 contributed by a Radius user.
2187 PH/24 When doing a callout, the value of $domain wasn't set correctly when
2188 expanding the "port" option of the smtp transport.
2190 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
2191 while reading a MIME header. Thanks to Tom Hughes for a patch.
2193 PH/24 Include config.h inside local_scan.h so that configuration settings are
2196 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
2197 that in an ACL for RCPT (for example), you can examine exactly what was
2200 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
2201 commands, but it was not correctly comparing the address with the actual
2202 client host address. Thus, it would show the EHLO address in Received:
2203 header lines when this was not necessary.
2205 PH/27 Added the % operator to ${eval:}.
2207 PH/28 Exim tries to create and chdir to its spool directory when it starts;
2208 it should be ignoring failures (because with -C, for example, it has lost
2209 privilege). It wasn't ignoring creation failures other than "already
2212 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
2215 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
2216 because an input line was too long, either on its own, or by virtue of
2217 too many continuations, the temporary file was not being removed, and the
2218 return code was incorrect.
2220 PH/31 Missing "BOOL" in function definition in filtertest.c.
2222 PH/32 Applied Sieve patches from the maintainer.
2224 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
2226 PH/33 Added "verify = not_blind".
2228 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
2229 Local/Makefile (with some defaults set). These are used in built scripts
2230 such as exicyclog, but they have never been used in the exim_install
2231 script (though there are many overriding facilities there). I have
2232 arranged that the exim_install script now takes note of these two
2235 PH/35 Installed configuration files for Dragonfly.
2237 PH/36 When a locally submitted message by a trusted user did not contain a
2238 From: header, and the sender address was obtained from -f or from an SMTP
2239 MAIL command, and the trusted user did not use -F to supply a sender
2240 name, $originator_name was incorrectly used when constructing a From:
2241 header. Furthermore, $originator_name was used for submission mode
2242 messages from external hosts without From: headers in a similar way,
2243 which is clearly wrong.
2245 PH/37 Added control=suppress_local_fixups.
2247 PH/38 When log_selector = +received_sender was set, and the addition of the
2248 sender made the log line's construction buffer exactly full, or one byte
2249 less than full, an overflow happened when the terminating "\n" was
2252 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
2253 when the result of a list match is failure because a DNS lookup failed.
2255 PH/40 RM_COMMAND is now used in the building process.
2257 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
2258 the "build-* directories that it finds.
2260 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
2261 address in a domain literal was a prefix of an interface address.
2263 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
2264 when verifying a sender address, unless rewrite_headers is false.
2266 PH/44 Wrote a long comment about why errors_to addresses are verified as
2267 recipients, not senders.
2269 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
2270 the ratelimit ACL was added.
2272 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
2274 PH/46 Added extra information about PostgreSQL errors to the error string.
2276 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
2277 avoiding the need to install special zones in a real server. This is
2278 backwards compatible; if it can't find the fake resolver, it drops back.
2279 Thus, both old and new test suites can be run.
2281 TF/02 Added util/ratelimit.pl
2283 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
2284 clock is set back in time.
2286 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
2287 Candler <B.Candler@pobox.com>.
2289 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
2290 OK for addresses that are the result of redirections.
2292 PH/48 A number of further additions for the benefit of the new test suite,
2293 including a fake gethostbyname() that interfaces to the fake DNS resolver
2296 TF/06 The fix for widen_domains has also been applied to qualify_single and
2297 search_parents which are the other dnslookup options that can cause
2300 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
2303 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
2305 TF/07 Exim produced the error message "an SRV record indicated no SMTP
2306 service" if it encountered an MX record with an empty target hostname.
2307 The message is now "an MX or SRV record indicated no SMTP service".
2309 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
2310 if the DNS of the sending site is misconfigured. This is quite a
2311 common situation. This change restores the behaviour of treating a
2312 helo verification defer as a failure.
2314 PH/51 If self=fail was set on a router, the bounce message did not include the
2315 actual error message.
2321 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
2323 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
2324 command itself ended in error, the underlying message about the transport
2325 filter timeout was being overwritten with the pipe command error. Now the
2326 underlying error message should be appended to the second error message.
2328 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
2330 PH/02 Reduce the amount of output that "make" produces by default. Full output
2331 can still be requested.
2333 PH/03 The warning log line about a condition test deferring for a "warn" verb
2334 was being output only once per connection, rather than after each
2335 occurrence (because it was using the same function as for successful
2336 "warn" verbs). This seems wrong, so I have changed it.
2338 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
2339 it should not have, which might have caused a crash in the right
2340 circumstances, but probably never did.
2342 PH/04 Installed a modified version of Tony Finch's patch to make submission
2343 mode fix the return path as well as the Sender: header line, and to
2344 add a /name= option so that you can make the user's friendly name appear
2347 TF/03 Added the control = fakedefer ACL modifier.
2349 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
2350 Mark Lowes for thorough testing.
2352 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
2354 TK/03 Merged latest SRS patch from Miles Wilton.
2356 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
2357 with the definition in sysexits.h (which is #included earlier).
2358 Fortunately, Exim does not actually use EX_OK. The code used to try to
2359 preserve the sysexits.h value, by assumimg that macro definitions were
2360 scanned for macro replacements. I have been disabused of this notion,
2361 so now the code just undefines EX_OK before #including unistd.h.
2363 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
2364 in the smtp transport. When a block could not be written in a single
2365 write() function, the timeout was being re-applied to each part-write.
2366 This seems wrong - if the receiver was accepting one byte at a time it
2367 would take for ever. The timeout is now adjusted when this happens. It
2368 doesn't have to be particularly precise.
2370 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
2371 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
2373 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
2374 without a domain if the check to <postmaster@domain> fails.
2376 SC/01 Eximstats: added -xls and the ability to specify output files
2377 (patch written by Frank Heydlauf).
2379 SC/02 Eximstats: use FileHandles for outputing results.
2381 SC/03 Eximstats: allow any combination of xls, txt, and html output.
2383 SC/04 Eximstats: fixed display of large numbers with -nvr option
2385 SC/05 Eximstats: fixed merging of reports with empty tables.
2387 SC/06 Eximstats: added the -include_original_destination flag
2389 SC/07 Eximstats: removed tabs and trailing whitespace.
2391 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
2393 TK/06 MBOX spool code: Add real "From " MBOX separator line
2394 so the .eml file is really in mbox format (even though
2395 most programs do not really care). Patch from Alex Miller.
2397 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
2398 The latter is generated from $received_to and is only set if the
2399 message has one envelope recipient. SA can use these headers,
2400 obviously out-of-the-box. Patch from Alex Miller.
2402 PH/08 The ${def test on a variable was returning false if the variable's
2403 value was "0", contrary to what the specification has always said!
2404 The result should be true unless the variable is empty.
2406 PH/09 The syntax error of a character other than { following "${if
2407 def:variable_name" (after optional whitespace) was not being diagnosed.
2408 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
2409 accidental colon was present, for example, could give incorrect results.
2411 PH/10 Tidied the code in a number of places where the st_size field of a stat()
2412 result is used (not including appendfile, where other changes are about
2415 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
2416 This involved changing a lot of size variables from int to off_t. It
2417 should work with maildirs and everything.
2419 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
2420 spamd dying while we are connected to it.
2422 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
2425 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
2426 function for BDB 4.3.
2428 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
2430 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
2431 the sources for PCRE have been re-arranged and more files are now
2434 PH/15 The code I had for printing potentially long long variables in PH/11
2435 above was not the best (it lost precision). The length of off_t variables
2436 is now inspected at build time, and an appropriate printing format (%ld
2437 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
2438 to be "long long int" or "long int". This is needed for the internal
2439 formatting function string_vformat().
2441 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
2442 the configuration file to be ":syslog", then the script "guesses" where
2443 the logs files are, rather than using the compiled in default. In our
2444 case the guess is not the same as the compiled default, so the script
2445 suddenly stopped working when I started to use syslog. The patch checks
2446 to see if log_file_path is "". If so, it attempts to read it from exim
2447 with no configuration file to get the compiled in version, before it
2448 falls back to the previous guessing code."
2450 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
2451 implementing BATV in an Exim configuration. See NewStuff for the gory
2454 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
2455 Makefile that are specific to HP-UX.
2457 PH/18 If the "use_postmaster" option was set for a recipient callout together
2458 with the "random" option, the postmaster address was used as the MAIL
2459 FROM address for the random test, but not for the subsequent recipient
2460 test. It is now used for both.
2462 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
2463 patch removes a few documentation additions to RFC 3028, because the
2464 latest draft now contains them. It adds the new en;ascii-case comparator
2465 and a new error check for 8bit text in MIME parts. Comparator and
2466 require names are now matched exactly. I enabled the subaddress
2467 extension, but it is not well tested yet (read: it works for me)."
2469 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
2470 rework some of the code of TK/09 above to avoid the hardwired use of
2471 "%lld" and "long long". Replaced the call to snprintf() with a call to
2474 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
2475 records point to non-existent hosts", "retry timeout exceeded", and
2476 "retry time not reached for any host after a long failure period".
2478 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
2479 experimental DomainKeys support:
2481 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
2482 (2) On an error such as an illegally used "control", the wrong name for
2483 the control was given.
2485 These problems did NOT occur unless DomainKeys support was compiled.
2487 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
2489 PH/24 Added ${if match_ip condition.
2491 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
2492 databases so that it will be absolutely obvious if a crash occurs in the
2493 DB library. This is a regular occurrence (often caused by mis-matched
2496 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
2497 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
2498 picked up on a user's system that detects such things. There doesn't seem
2499 to be a gcc warning option for this - only an attribute that has to be
2500 put on the function's prototype. It seems that in Fedora Core 4 they have
2501 set this on a number of new functions. No doubt there will be more in due
2504 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
2505 specify a transport. However, if an address that was verified by such a
2506 router was the subject of a callout, Exim crashed because it tried to
2507 read the rcpt_include_affixes from the non-existent transport. Now it
2508 just assumes that the setting of that option is false. This bug was
2509 introduced by 4.51/PH/31.
2511 PH/28 Changed -d+all to exclude +memory, because that information is very
2512 rarely of interest, but it makes the output a lot bigger. People tend to
2513 do -d+all out of habit.
2515 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
2516 code in os-type was giving problems when libc.so lives in lib64, like on
2519 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
2520 aren't the modern standard, and it seems that some systems' include files
2521 don't always have them. Exim was already checking for some of the newer
2522 ones like T_AAAA, and defining it itself. I've added checks for all the
2523 record types that Exim uses.
2525 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
2526 not automatically generating a new one, as it is supposed to. This
2527 prevented TLS from working. If the file did exist, but contained invalid
2528 data, a new version was generated, as expected. It was only the case of a
2529 non-existent file that was broken.
2531 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
2532 with a change in libdomainkeys > 0.64.
2534 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
2535 from DNS. If the selector record carries the flag, it now has
2536 precedence over the domain-wide flag.
2538 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
2540 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
2541 the use of a port name that isn't defined in /etc/services) occurred, the
2542 message was deferred as in a normal delivery, and thus remained on the
2543 spool, instead of being failed because of the mua_wrapper setting. This
2544 is now fixed, and I tidied up some of the mua_wrapper messages at the
2547 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
2548 the messages in a hash of arrays rather than using individual hashes.
2549 This is a bit cleaner and results in dramatic memory savings, albeit
2550 at a slight CPU cost.
2552 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
2553 as requested by Marc Sherman.
2555 SC/10 Eximstats: added histograms for user specified patterns as requested
2558 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
2560 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
2561 fopen() in the content-scanning modules that did not already have it.
2567 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
2568 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
2570 TK/02 Fix ACL "control" statement not being available in MIME ACL.
2572 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
2574 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
2575 to test Sieve filters that use "vacation".
2577 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
2578 that changes the way the GnuTLS parameters are stored in the cache file.
2579 The new format can be generated externally. For backward compatibility,
2580 if the data in the cache doesn't make sense, Exim assumes it has read an
2581 old-format file, and it generates new data and writes a new file. This
2582 means that you can't go back to an older release without removing the
2585 PH/03 A redirect router that has both "unseen" and "one_time" set does not
2586 work if there are any delivery delays because "one_time" forces the
2587 parent to be marked "delivered", so its unseen clone is never tried
2588 again. For this reason, Exim now forbids the simultaneous setting of
2591 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
2592 redirected to themselves ("homonym" addresses). Read the long ChangeLog
2593 entry if you want to know the details. The fix, however, neglected to
2594 consider the case when local delivery batching is involved. The test for
2595 "previously delivered" was not happening when checking to see if an
2596 address could be batched with a previous (undelivered) one; under
2597 certain circumstances this could lead to multiple deliveries to the same
2600 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
2601 in its include files, and this causes problems building Exim.
2603 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
2604 header_syntax) but Exim was just ignoring anything given after a slash.
2605 In particular, this caused confusion with an attempt to use "verify =
2606 reverse_host_lookup/defer_ok". An error is now given when options are
2607 supplied for verify items that do not have them. (Maybe reverse_host_
2608 lookup should have a defer_ok option, but that's a different point.)
2610 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
2611 defined by RFC 821) to 2048, because there were problems with some AUTH
2612 commands, and RFC 1869 says the size should be increased for extended
2613 SMTP commands that take arguments.
2615 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
2618 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
2619 "unknown" error; now it says that the functionality isn't in the binary.
2621 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
2622 an address' error message when a string expansion fails (syntax or
2623 whatever). Otherwise the password may appear in the log. Following change
2624 PH/42 below, there is no longer a chance of it appearing in a bounce
2627 PH/11 Installed exipick version 20050225.0 from John Jetmore.
2629 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
2630 of its addresses was ever tried. (Bugzilla bug #2.)
2632 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
2633 the result incorrectly in the debug output. (It correctly added a newline
2634 to what was transported.)
2636 TF/01 Added $received_time.
2638 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
2639 commented out examples of how to interface to a virus scanner and to
2640 SpamAssassin. Also added commented examples of av_scanner and
2641 spamd_address settings.
2643 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
2644 and controls are allowed in which ACLs. There were a couple of minor
2645 errors. Some of the entries in the conditions table (which is a table of
2646 where they are NOT allowed) were getting very unwieldy; rewrote them as a
2647 negation of where the condition IS allowed.
2649 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
2651 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
2652 header file does not have a version number, so I've had to invent a new
2653 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
2654 API. The code is untested by me (my Linux distribution still has 0.3.2 of
2655 radiusclient), but it was contributed by a Radius user.
2657 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
2658 files or directories, for OpenSSL.
2660 PH/19 When an Exim process that is running as root has to create an Exim log
2661 file, it does so in a subprocess that runs as exim:exim so as to get the
2662 ownership right at creation (otherwise, other Exim processes might see
2663 the file with the wrong ownership). There was no test for failure of this
2664 fork() call, which would lead to the process getting stuck as it waited
2665 for a non-existent subprocess. Forks do occasionally fail when resources
2666 run out. I reviewed all the other calls to fork(); they all seem to check
2669 PH/20 When checking for unexpected SMTP input at connect time (before writing
2670 the banner), Exim was not dealing correctly with a non-positive return
2671 from the read() function. If the client had disconnected by this time,
2672 the result was a log entry for a synchronization error with an empty
2673 string after "input=" when read() returned zero. If read() returned -1
2674 (an event I could not check), uninitialized data bytes were printed.
2675 There were reports of junk text (parts of files, etc) appearing after
2678 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
2680 PH/22 Added support for macro redefinition, and (re)definition in between
2681 driver and ACL definitions.
2683 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
2684 forgetting to use the resulting value; it was using the unexpanded value.
2686 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
2687 hadn't been configured. The fix is from Juergen Kreileder, who
2688 understands it better than I do:
2690 "Here's what I see happening with three configured cyrus_sasl
2691 authenticators configured (plain, login, cram-md5):
2693 On startup auth_cyrus_sasl_init() gets called for each of these.
2694 This means three calls to sasl_listmech() without a specified mech_list.
2695 => SASL tests which mechs of all available mechs actually work
2696 => three warnings about OTP not working
2697 => the returned list contains: plain, login, cram-md5, digest-md5, ...
2699 With the patch, sasl_listmech() also gets called three times. But now
2700 SASL's mech_list option is set to the server_mech specified in the the
2701 authenticator. Or in other words, the answer from sasl_listmech()
2702 gets limited to just the mech you're testing for (which is different
2704 => the return list contains just 'plain' or 'login', 'cram-md5' or
2705 nothing depending on the value of ob->server_mech.
2707 I've just tested the patch: Authentication still works fine,
2708 unavailable mechs specified in the exim configuration are still
2709 caught, and the auth.log warnings about OTP are gone."
2711 PH/25 When debugging is enabled, the contents of the command line are added
2712 to the debugging output, even when log_selector=+arguments is not
2715 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
2716 answer is "GNU", and only if the return is "GNU/something" is the answer
2719 PH/27 $acl_verify_message is now set immediately after the failure of a
2720 verification in an ACL, and so is available in subsequent modifiers. In
2721 particular, the message can be preserved by coding like this:
2723 warn !verify = sender
2724 set acl_m0 = $acl_verify_message
2726 Previously, $acl_verify_message was set only while expanding "message"
2727 and "log_message" when a very denied access.
2729 PH/28 Modified OS/os.c-Linux with
2731 -#ifndef OS_LOAD_AVERAGE
2732 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
2734 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
2735 nomenclature these days.)
2737 PH/29 Installed patch from the Sieve maintainer that adds the options
2738 sieve_useraddress and sieve_subaddress to the redirect router.
2740 PH/30 In these circumstances:
2741 . Two addresses routed to the same list of hosts;
2742 . First host does not offer TLS;
2743 . First host accepts first address;
2744 . First host gives temporary error to second address;
2745 . Second host offers TLS and a TLS session is established;
2746 . Second host accepts second address.
2747 Exim incorrectly logged both deliveries with the TLS parameters (cipher
2748 and peerdn, if requested) that were in fact used only for the second
2751 PH/31 When doing a callout as part of verifying an address, Exim was not paying
2752 attention to any local part prefix or suffix that was matched by the
2753 router that accepted the address. It now behaves in the same way as it
2754 does for delivery: the affixes are removed from the local part unless
2755 rcpt_include_affixes is set on the transport.
2757 PH/32 Add the sender address, as F=<...>, to the log line when logging a
2758 timeout during the DATA phase of an incoming message.
2760 PH/33 Sieve envelope tests were broken for match types other than :is. I have
2761 applied a patch sanctioned by the Sieve maintainer.
2763 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
2764 the uid or gid is negative. A case of a negative gid caused this to be
2765 noticed. The fix allows for either to be negative.
2767 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
2768 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
2771 PH/36 The change PH/12 above was broken. Fixed it.
2773 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
2774 the grounds that routing the same address twice would always produce the
2775 same answer. This might have been true once, but it is certainly no
2776 longer true now. Routing a child address may depend on the previous
2777 routing that produced that child. Some complicated redirection strategies
2778 went wrong when messages had multiple recipients, and made Exim's
2779 behaviour dependent on the order in which the addresses were given.
2781 I have moved the duplicate checking until after the routing is complete.
2782 Exim scans the addresses that are assigned to local and remote
2783 transports, and removes any duplicates. This means that more work will be
2784 done, as duplicates will always all be routed, but duplicates are
2785 presumably rare, so I don't expect this is of any significance.
2787 For deliveries to pipes, files, and autoreplies, the duplicate checking
2788 still happens during the routing process, since they are not going to be
2791 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
2792 It corrects a timeout issue with spamd. This is Ian's comment: "The
2793 background is that sometimes spamd either never reads data from a
2794 connection it has accepted, or it never writes response data. The exiscan
2795 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
2796 blindly assumes that writes won't block so it may never time out."
2798 PH/39 Allow G after quota size as well as K and M.
2800 PH/40 The value set for $authenticated_id in an authenticator may not contain
2801 binary zeroes or newlines because the value is written to log lines and
2802 to spool files. There was no check on this. Now the value is run through
2803 the string_printing() function so that such characters are converted to
2804 printable escape sequences.
2806 PH/41 $message_linecount is a new variable that contains the total number of
2807 lines in the message. Compare $body_linecount, which is the count for the
2810 PH/42 Exim no longer gives details of delivery errors for specific addresses in
2811 bounce and delay warning messages, except in certain special cases, which
2814 (a) An SMTP error message from a remote host;
2815 (b) A message specified in a :fail: redirection;
2816 (c) A message specified in a "fail" command in a system filter;
2817 (d) A message specified in a FAIL return from the queryprogram router;
2818 (e) A message specified by the cannot_route_message router option.
2820 In these cases only, Exim does include the error details in bounce and
2821 warning messages. There are also a few cases where bland messages such
2822 as "unrouteable address" or "local delivery error" are given.
2824 PH/43 $value is now also set for the "else" part of a ${run expansion.
2826 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
2827 being worked on, but at least Exim now implements the latest version to
2830 PH/45 In a pipe transport, although a timeout while waiting for the pipe
2831 process to complete was treated as a delivery failure, a timeout while
2832 writing the message to the pipe was logged, but erroneously treated as a
2833 successful delivery. Such timeouts include transport filter timeouts. For
2834 consistency with the overall process timeout, these timeouts are now
2835 treated as errors, giving rise to delivery failures by default. However,
2836 there is now a new Boolean option for the pipe transport called
2837 timeout_defer, which, if set TRUE, converts the failures into defers for
2838 both kinds of timeout. A transport filter timeout is now identified in
2841 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
2842 systems where "make" and "gmake" are different, calling "gmake" at top
2843 level broke things. I've arranged for the value of $(MAKE) to be passed
2844 from the Makefile to this script so that it can call the same version of
2848 A note about Exim versions 4.44 and 4.50
2849 ----------------------------------------
2851 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
2852 changes of various kinds. As a consequence, a big documentation update was
2853 needed. This delayed the release for rather longer than seemed good, especially
2854 in the light of a couple of (minor) security issues. Therefore, the changes
2855 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
2856 release. So 4.44 and 4.50 are in effect two different branches that both start
2859 I have left the 4.50 change log unchanged; it contains all the changes since
2860 4.43. The change log for 4.44 is below; many of its items are identical to
2861 those for 4.50. This seems to be the most sensible way to preserve the
2862 historical information.
2868 1. Minor wording change to the doc/README.SIEVE file.
2870 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
2871 computation of the current number of files was incorrect.
2873 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
2874 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
2877 4. Give more explanation in the error message when the command for a transport
2878 filter fails to execute.
2880 5. There are several places where Exim runs a non-Exim command in a
2881 subprocess. The SIGUSR1 signal should be disabled for these processes. This
2882 was being done only for the command run by the queryprogram router. It is
2883 now done for all such subprocesses. The other cases are: ${run, transport
2884 filters, and the commands run by the lmtp and pipe transports.
2886 6. Added CONFIGURE_GROUP build-time option.
2888 7. Some older OS have a limit of 256 on the maximum number of file
2889 descriptors. Exim was using setrlimit() to set 1000 as a large value
2890 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
2891 systems. I've change it so that if it can't get 1000, it tries for 256.
2893 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
2894 was an oversight, and furthermore, ever since the addition of extra
2895 controls (e.g. 4.43/32), the checks on when to allow different forms of
2896 "control" were broken. There should now be diagnostics for all cases when a
2897 control that does not make sense is encountered.
2899 9. Added the /retain_sender option to "control=submission".
2901 10. $recipients is now available in the predata ACL (oversight).
2903 11. Tidy the search cache before the fork to do a delivery from a message
2904 received from the command line. Otherwise the child will trigger a lookup
2905 failure and thereby defer the delivery if it tries to use (for example) a
2906 cached ldap connection that the parent has called unbind on.
2908 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
2909 of $address_data from the recipient verification was clobbered by the
2910 sender verification.
2912 13. The value of address_data from a sender verification is now available in
2913 $sender_address_data in subsequent conditions in the ACL statement.
2915 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
2917 15. Added a new option "connect=<time>" to callout options, to set a different
2920 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
2921 was its contents. (It was OK if the option was not defined at all.)
2923 17. A "Completed" log line is now written for messages that are removed from
2924 the spool by the -Mrm option.
2926 18. New variables $sender_verify_failure and $recipient_verify_failure contain
2927 information about exactly what failed.
2929 19. Added -dd to debug only the daemon process.
2931 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
2932 handles timeouts, both on the server side and network timeouts. Renamed the
2933 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
2935 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
2936 It is now set to "smtps".
2938 22. $host_address is now set to the target address during the checking of
2939 ignore_target_hosts.
2941 23. When checking ignore_target_hosts for an ipliteral router, no host name was
2942 being passed; this would have caused $sender_host_name to have been used if
2943 matching the list had actually called for a host name (not very likely,
2944 since this list is usually IP addresses). A host name is now passed as
2947 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
2948 code that specifies a non-restarting handler (typically sigaction() in
2949 modern systems) in an attempt to fix a rare and obscure crash bug.
2951 25. Narrowed the window for a race in the daemon that could cause it to ignore
2952 SIGCHLD signals. This is not a major problem, because they are used only to
2953 wake it up if nothing else does.
2955 26. A malformed maildirsize file could cause Exim to calculate negative values
2956 for the mailbox size or file count. Odd effects could occur as a result.
2957 The maildirsize information is now recalculated if the size or filecount
2960 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
2961 support for a long time. Removed HAVE_SYS_VFS_H.
2963 28. Installed the latest version of exipick from John Jetmore.
2965 29. In an address list, if the pattern was not a regular expression, an empty
2966 subject address (from a bounce message) matched only if the pattern was an
2967 empty string. Non-empty patterns were not even tested. This was the wrong
2968 because it is perfectly reasonable to use an empty address as part of a
2969 database query. An empty address is now tested by patterns that are
2970 lookups. However, all the other forms of pattern expect the subject to
2971 contain a local part and a domain, and therefore, for them, an empty
2972 address still always fails if the pattern is not itself empty.
2974 30. Exim went into a mad DNS loop when attempting to do a callout where the
2975 host was specified on an smtp transport, and looking it up yielded more
2976 than one IP address.
2978 31. Re-factored the code for checking spool and log partition space into a
2979 function that finds that data and another that does the check. The former
2980 is then used to implement four new variables: $spool_space, $log_space,
2981 $spool_inodes, and $log_inodes.
2983 32. The RFC2047 encoding function was originally intended for short strings
2984 such as real names; it was not keeping to the 75-character limit for
2985 encoded words that the RFC imposes. It now respects the limit, and
2986 generates multiple encoded words if necessary. To be on the safe side, I
2987 have increased the buffer size for the ${rfc2047: expansion operator from
2990 33. It is now permitted to omit both strings after an "if" condition; if the
2991 condition is true, the result is "true". As before, when the second string
2992 is omitted, a false condition yields an empty string. This makes it less
2993 cumbersome to write custom ACL and router conditions.
2995 34. Failure to deliver a bounce message always caused it to be frozen, even if
2996 there was an errors_to setting on the router. The errors_to setting is now
2999 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
3000 canonical form (fully expanded) before being placed in
3001 $sender_host_address.
3003 36. The table in the code that translates DNS record types into text (T_A to
3004 "A" for instance) was missing entries for NS and CNAME. It is just possible
3005 that this could have caused confusion if both these types were looked up
3006 for the same domain, because the text type is used as part of Exim's
3007 per-process caching. But the chance of anyone hitting this buglet seems
3010 37. The dnsdb lookup has been extended in a number of ways.
3012 (1) There is a new type, "zns", which walks up the domain tree until it
3013 finds some nameserver records. It should be used with care.
3015 (2) There is a new type, "mxh", which is like "mx" except that it returns
3016 just the host names, not the priorities.
3018 (3) It is now possible to give a list of domains (or IP addresses) to be
3019 looked up. The behaviour when one of the lookups defers can be
3020 controlled by a keyword.
3022 (4) It is now possible to specify the separator character for use when
3023 multiple records are returned.
3025 38. The dnslists ACL condition has been extended: it is now possible to supply
3026 a list of IP addresses and/or domains to be looked up in a particular DNS
3029 39. Added log_selector=+queue_time_overall.
3031 40. When running the queue in the test harness, wait just a tad after forking a
3032 delivery process, to get repeatability of debugging output.
3034 41. Include certificate and key file names in error message when GnuTLS fails
3035 to set them up, because the GnuTLS error message doesn't include the name
3036 of the failing file when there is a problem reading it.
3038 42. Allow both -bf and -bF in the same test run.
3040 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
3042 44. The "Exiscan patch" is now merged into the mainline Exim source.
3044 45. Sometimes the final signoff response after QUIT could fail to get
3045 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
3046 < 0 before doing a fflush(). This bug looks as though it goes back to the
3047 introduction of TLS in release 3.20, but "sometimes" must have been rare
3048 because the tests only now provoked it.
3050 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
3051 (this can affect the format of dates).
3053 47. exim_tidydb, when checking for the continued existence of a message for
3054 which it has found a message-specific retry record, was not finding
3055 messages that were in split spool directories. Consequently, it was
3056 deleting retry records that should have stayed in existence.
3058 48. Steve fixed some bugs in eximstats.
3060 49. The SPA authentication driver was not abandoning authentication and moving
3061 on to the next authenticator when an expansion was forced to fail,
3062 contradicting the general specification for all authenticators. Instead it
3063 was generating a temporary error. It now behaves as specified.
3065 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
3066 (the order specifies the preference for clients). The order is now AES256,
3067 AES128, 3DES, ARCFOUR128.
3069 51. Small patch to Sieve code - explicitly set From: when generating an
3072 52. Exim crashed if a remote delivery caused a very long error message to be
3073 recorded - for instance if somebody sent an entire SpamAssassin report back
3074 as a large number of 550 error lines. This bug was coincidentally fixed by
3075 increasing the size of one of Exim's internal buffers (big_buffer) that
3076 happened as part of the Exiscan merge. However, to be on the safe side, I
3077 have made the code more robust (and fixed the comments that describe what
3080 53. Now that there can be additional text after "Completed" in log lines (if
3081 the queue_time_overall log selector is set), a one-byte patch to exigrep
3082 was needed to allow it to recognize "Completed" as not the last thing in
3085 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
3086 patch that reportedly fixes this has been added. I am not expert enough to
3087 create a test for it. This is what the patch creator wrote:
3089 "I found a little strange behaviour of ldap code when working with
3090 Windows 2003 AD Domain, where users was placed in more than one
3091 Organization Units. When I tried to give exim partial DN, the exit code
3092 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
3093 But simultaneously result of request was absolutely normal ldap result,
3094 so I produce this patch..."
3096 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
3097 REFERENCE, so I have modified the code to exclude the patch when that macro
3100 55. Some experimental protocols are using DNS PTR records for new purposes. The
3101 keys for these records are domain names, not reversed IP addresses. The
3102 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
3103 leaves it alone. Component reversal etc. now happens only for IP addresses.
3106 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
3108 57. Double the size of the debug message buffer (to 2048) so that more of very
3109 long debug lines gets shown.
3111 58. The exicyclog utility now does better if the number of log files to keep
3112 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
3114 59. Two changes related to the smtp_active_hostname option:
3116 (1) $smtp_active_hostname is now available as a variable.
3117 (2) The default for smtp_banner uses $smtp_active_hostname instead
3118 of $primary_hostname.
3120 60. The host_aton() function is supposed to be passed a string that is known
3121 to be a valid IP address. However, in the case of IPv6 addresses, it was
3122 not checking this. This is a hostage to fortune. Exim now panics and dies
3123 if the condition is not met. A case was found where this could be provoked
3124 from a dnsdb PTR lookup with an IPv6 address that had more than 8
3125 components; fortuitously, this particular loophole had already been fixed
3126 by change 4.50/55 above.
3128 If there are any other similar loopholes, the new check in host_aton()
3129 itself should stop them being exploited. The report I received stated that
3130 data on the command line could provoke the exploit when Exim was running as
3131 exim, but did not say which command line option was involved. All I could
3132 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
3133 running as the user.
3136 61. There was a buffer overflow vulnerability in the SPA authentication code
3137 (which came originally from the Samba project). I have added a test to the
3138 spa_base64_to_bits() function which I hope fixes it.
3141 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
3142 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
3144 63. The daemon start-up calls getloadavg() while still root for those OS that
3145 need the first call to be done as root, but it missed one case: when
3146 deliver_queue_load_max is set with deliver_drop_privilege. This is
3147 necessary for the benefit of the queue runner, because there is no re-exec
3148 when deliver_drop_privilege is set.
3150 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
3151 This has been fixed.
3153 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
3154 was in use, was not putting the data itself into the right store pool;
3155 consequently, it could be overwritten for a subsequent message in the same
3156 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
3159 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
3161 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
3162 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
3163 the function were treating the return as a boolean value, which happened to
3164 work because 0=false and not-0=true, but is not correct code.
3166 68. The host_aton() function was not handling scoped IPv6 addresses (those
3167 with, for example, "%eth0" on the end) correctly.
3169 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
3170 negated items (that is, ~something) in unsigned ints. Some compilers
3171 apparently mutter when there is no cast.
3173 70. If an address verification called from an ACL failed, and did not produce a
3174 user-specific message (i.e. there was only a "system" message), nothing was
3175 put in $acl_verify_message. In this situation, it now puts the system
3178 71. Change 4.23/11 added synchronization checking at the start of an SMTP
3179 session; change 4.31/43 added the unwanted input to the log line - except
3180 that it did not do this in the start of session case. It now does.
3182 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
3183 This is wrong and can cause the other end to generate a synchronization
3184 error if it is another Exim or anything else that does the synchronization
3185 check. A QUIT command is no longer sent after a timeout.
3187 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
3188 during host lookups.
3190 74. The defer_ok option of callout verification was not working if it was used
3191 when verifying addresses in header lines, that is, for this case:
3193 verify = header_sender/callout=defer_ok
3195 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
3196 those file descriptors could be used for SMTP connections. If anything
3197 wrote to stderr (the example that came up was "warn" in embedded Perl), it
3198 could be sent to the SMTP client, causing chaos. The daemon now opens
3199 stdin, stdout, and stderr to /dev/null when it puts itself into the
3202 76. Arrange for output from Perl's "warn" command to be written to Exim's main
3203 log by default. The user can override this with suitable Perl magic.
3205 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
3206 the log message when discard triggers, was not working for the DATA ACL or
3207 for the non-SMTP ACL.
3209 78. Error message wording change in sieve.c.
3211 79. If smtp_accept_max_per_host was set, the number of connections could be
3212 restricted to fewer than expected, because the daemon was trying to set up
3213 a new connection before checking whether the processes handling previous
3214 connections had finished. The check for completed processes is now done
3215 earlier. On busy systems, this bug wouldn't be noticed because something
3216 else would have woken the daemon, and it would have reaped the completed
3219 80. If a message was submitted locally by a user whose login name contained one
3220 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
3221 It caused a spool format error. I have fixed the spool reading code. A
3222 related problem was that the "from" clause in the Received: line became
3223 illegal because of the space(s). It is now covered by ${quote_local_part.
3225 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
3228 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
3230 83. Patch from Sieve maintainer to fix unterminated string problem in
3231 "vacation" handling.
3233 84. Some minor changes to the Linux configuration files to help with other
3234 OS variants using glibc.
3236 85. One more patch for Sieve to update vacation handling to latest spec.
3239 ----------------------------------------------------
3240 See the note above about the 4.44 and 4.50 releases.
3241 ----------------------------------------------------
3247 1. Change 4.43/35 introduced a bug that caused file counts to be
3248 incorrectly computed when quota_filecount was set in an appendfile
3251 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
3252 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
3255 3. Give more explanation in the error message when the command for a transport
3256 filter fails to execute.
3258 4. There are several places where Exim runs a non-Exim command in a
3259 subprocess. The SIGUSR1 signal should be disabled for these processes. This
3260 was being done only for the command run by the queryprogram router. It is
3261 now done for all such subprocesses. The other cases are: ${run, transport
3262 filters, and the commands run by the lmtp and pipe transports.
3264 5. Some older OS have a limit of 256 on the maximum number of file
3265 descriptors. Exim was using setrlimit() to set 1000 as a large value
3266 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
3267 systems. I've change it so that if it can't get 1000, it tries for 256.
3269 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
3270 was an oversight, and furthermore, ever since the addition of extra
3271 controls (e.g. 4.43/32), the checks on when to allow different forms of
3272 "control" were broken. There should now be diagnostics for all cases when a
3273 control that does not make sense is encountered.
3275 7. $recipients is now available in the predata ACL (oversight).
3277 8. Tidy the search cache before the fork to do a delivery from a message
3278 received from the command line. Otherwise the child will trigger a lookup
3279 failure and thereby defer the delivery if it tries to use (for example) a
3280 cached ldap connection that the parent has called unbind on.
3282 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
3283 of $address_data from the recipient verification was clobbered by the
3284 sender verification.
3286 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
3287 was its contents. (It was OK if the option was not defined at all.)
3289 11. A "Completed" log line is now written for messages that are removed from
3290 the spool by the -Mrm option.
3292 12. $host_address is now set to the target address during the checking of
3293 ignore_target_hosts.
3295 13. When checking ignore_target_hosts for an ipliteral router, no host name was
3296 being passed; this would have caused $sender_host_name to have been used if
3297 matching the list had actually called for a host name (not very likely,
3298 since this list is usually IP addresses). A host name is now passed as
3301 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
3302 code that specifies a non-restarting handler (typically sigaction() in
3303 modern systems) in an attempt to fix a rare and obscure crash bug.
3305 15. Narrowed the window for a race in the daemon that could cause it to ignore
3306 SIGCHLD signals. This is not a major problem, because they are used only to
3307 wake it up if nothing else does.
3309 16. A malformed maildirsize file could cause Exim to calculate negative values
3310 for the mailbox size or file count. Odd effects could occur as a result.
3311 The maildirsize information is now recalculated if the size or filecount
3314 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
3315 support for a long time. Removed HAVE_SYS_VFS_H.
3317 18. Updated exipick to current release from John Jetmore.
3319 19. Allow an empty sender to be matched against a lookup in an address list.
3320 Previously the only cases considered were a regular expression, or an
3323 20. Exim went into a mad DNS lookup loop when doing a callout where the
3324 host was specified on the transport, if the DNS lookup yielded more than
3327 21. The RFC2047 encoding function was originally intended for short strings
3328 such as real names; it was not keeping to the 75-character limit for
3329 encoded words that the RFC imposes. It now respects the limit, and
3330 generates multiple encoded words if necessary. To be on the safe side, I
3331 have increased the buffer size for the ${rfc2047: expansion operator from
3334 22. Failure to deliver a bounce message always caused it to be frozen, even if
3335 there was an errors_to setting on the router. The errors_to setting is now
3338 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
3339 canonical form (fully expanded) before being placed in
3340 $sender_host_address.
3342 24. Updated eximstats to version 1.33
3344 25. Include certificate and key file names in error message when GnuTLS fails
3345 to set them up, because the GnuTLS error message doesn't include the name
3346 of the failing file when there is a problem reading it.
3348 26. Expand error message when OpenSSL has problems setting up cert/key files.
3351 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
3352 (this can affect the format of dates).
3354 28. exim_tidydb, when checking for the continued existence of a message for
3355 which it has found a message-specific retry record, was not finding
3356 messages that were in split spool directories. Consequently, it was
3357 deleting retry records that should have stayed in existence.
3359 29. eximstats updated to version 1.35
3360 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
3361 1.35 - bugfix such that pie charts by volume are generated correctly
3363 30. The SPA authentication driver was not abandoning authentication and moving
3364 on to the next authenticator when an expansion was forced to fail,
3365 contradicting the general specification for all authenticators. Instead it
3366 was generating a temporary error. It now behaves as specified.
3368 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
3369 (the order specifies the preference for clients). The order is now AES256,
3370 AES128, 3DES, ARCFOUR128.
3372 31. Small patch to Sieve code - explicitly set From: when generating an
3375 32. Exim crashed if a remote delivery caused a very long error message to be
3376 recorded - for instance if somebody sent an entire SpamAssassin report back
3377 as a large number of 550 error lines. This bug was coincidentally fixed by
3378 increasing the size of one of Exim's internal buffers (big_buffer) that
3379 happened as part of the Exiscan merge. However, to be on the safe side, I
3380 have made the code more robust (and fixed the comments that describe what
3383 33. Some experimental protocols are using DNS PTR records for new purposes. The
3384 keys for these records are domain names, not reversed IP addresses. The
3385 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
3386 leaves it alone. Component reversal etc. now happens only for IP addresses.
3389 34. The host_aton() function is supposed to be passed a string that is known
3390 to be a valid IP address. However, in the case of IPv6 addresses, it was
3391 not checking this. This is a hostage to fortune. Exim now panics and dies
3392 if the condition is not met. A case was found where this could be provoked
3393 from a dnsdb PTR lookup with an IPv6 address that had more than 8
3394 components; fortuitously, this particular loophole had already been fixed
3395 by change 4.50/55 or 4.44/33 above.
3397 If there are any other similar loopholes, the new check in host_aton()
3398 itself should stop them being exploited. The report I received stated that
3399 data on the command line could provoke the exploit when Exim was running as
3400 exim, but did not say which command line option was involved. All I could
3401 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
3402 running as the user.
3405 35. There was a buffer overflow vulnerability in the SPA authentication code
3406 (which came originally from the Samba project). I have added a test to the
3407 spa_base64_to_bits() function which I hope fixes it.
3410 36. The daemon start-up calls getloadavg() while still root for those OS that
3411 need the first call to be done as root, but it missed one case: when
3412 deliver_queue_load_max is set with deliver_drop_privilege. This is
3413 necessary for the benefit of the queue runner, because there is no re-exec
3414 when deliver_drop_privilege is set.
3416 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
3417 was in use, was not putting the data itself into the right store pool;
3418 consequently, it could be overwritten for a subsequent message in the same
3419 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
3422 38. Sometimes the final signoff response after QUIT could fail to get
3423 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
3424 < 0 before doing a fflush(). This bug looks as though it goes back to the
3425 introduction of TLS in release 3.20, but "sometimes" must have been rare
3426 because the tests only now provoked it.
3432 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
3433 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
3434 Now it returns an integer. A number of calls were still expecting a T/F
3435 return. Fortuitously, in all cases, the tests worked in OK situations,
3436 which is the norm. However, things would have gone wrong on any write
3437 failures on the smtp file descriptor. This function is used when sending
3438 messages over SMTP and also when doing verify callouts.
3440 2. When Exim is called to do synchronous delivery of a locally submitted
3441 message (the -odf or -odi options), it no longer closes stderr before doing
3444 3. Implemented the mua_wrapper option.
3446 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
3448 5. Implemented the functions header_remove(), header_testname(),
3449 header_add_at_position(), and receive_remove_recipient(), and exported them
3452 6. If an ACL "warn" statement specified the addition of headers, Exim already
3453 inserted X-ACL-Warn: at the start if there was no header name. However, it
3454 was not making this test for the second and subsequent header lines if
3455 there were newlines in the string. This meant that an invalid header could
3456 be inserted if Exim was badly configured.
3458 7. Allow an ACL "warn" statement to add header lines at the start or after all
3459 the Received: headers, as well as at the end.
3461 8. Added the rcpt_4xx retry error code.
3463 9. Added postmaster_mailfrom=xxx to callout verification option.
3465 10. Added mailfrom=xxxx to the callout verification option, for verify=
3468 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
3469 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
3471 12. Inserted some casts to stop certain compilers warning when using pointer
3472 differences as field lengths or precisions in printf-type calls (mostly
3473 affecting debugging statements).
3475 13. Added optional readline() support for -be (dynamically loaded).
3477 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
3478 same clock tick as a message's arrival, so that its received time was the
3479 same as the "first fail" time on the retry record, and that message
3480 remained on the queue past the ultimate address timeout, every queue runner
3481 would try a delivery (because it was past the ultimate address timeout) but
3482 after another failure, the ultimate address timeout, which should have then
3483 bounced the address, did not kick in. This was a "< instead of <=" error;
3484 in most cases the first failure would have been in the next clock tick
3485 after the received time, and all would be well.
3487 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
3488 being recognized when the domain list was tested by the match_domain
3489 condition in an expansion string.
3491 16. Added the ${str2b64: operator.
3493 17. Exim was always calling setrlimit() to set a large limit for the number of
3494 processes, without checking whether the existing limit was already
3495 adequate. (It did check for the limit on file descriptors.) Furthermore,
3496 errors from getrlimit() and setrlimit() were being ignored. Now they are
3497 logged to the main and panic logs, but Exim does carry on, to try to do its
3498 job under whatever limits there are.
3500 18. Imported PCRE 5.0.
3502 19. Trivial typo in log message " temporarily refused connection" (the leading
3505 20. If the log selector return_path_on_delivery was set and an address was
3506 redirected to /dev/null, the delivery process crashed because it assumed
3507 that a return path would always be set for a "successful" delivery. In this
3508 case, the whole delivery is bypassed as an optimization, and therefore no
3511 21. Internal re-arrangement: the function for sending a challenge and reading
3512 a response while authentication was assuming a zero-terminated challenge
3513 string. It's now changed to take a pointer and a length, to allow for
3514 binary data in such strings.
3516 22. Added the cyrus_sasl authenticator (code supplied by MBM).
3518 23. Exim was not respecting finduser_retries when seeking the login of the
3519 uid under which it was called; it was always trying 10 times. (The default
3520 setting of finduser_retries is zero.) Also, it was sleeping after the final
3521 failure, which is pointless.
3523 24. Implemented tls_on_connect_ports.
3525 25. Implemented acl_smtp_predata.
3527 26. If the domain in control=submission is set empty, Exim assumes that the
3528 authenticated id is a complete email address when it generates From: or
3529 Sender: header lines.
3531 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
3532 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
3533 chown and chgrp in /bin and hostname in /usr/bin.
3535 28. Exim was keeping the "process log" file open after each use, just as it
3536 does for the main log. This opens the possibility of it remaining open for
3537 long periods when the USR1 signal hits a daemon. Occasional processlog
3538 errors were reported, that could have been caused by this. Anyway, it seems
3539 much more sensible not to leave this file open at all, so that is what now
3542 29. The long-running daemon process does not normally write to the log once it
3543 has entered its main loop, and it closes the log before doing so. This is
3544 so that log files can straightforwardly be renamed and moved. However,
3545 there are a couple of unusual error situations where the daemon does write
3546 log entries, and I had neglected to close the log afterwards.
3548 30. The text of an SMTP error response that was received during a remote
3549 delivery was being truncated at 512 bytes. This is too short for some of
3550 the long messages that one sometimes sees. I've increased the limit to
3553 31. It is now possible to make retry rules that apply only when a message has a
3554 specific sender, in particular, an empty sender.
3556 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
3557 it possible to be selective about when SMTP synchronization is enforced.
3559 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
3561 32. Implemented hosts_connection_nolog.
3563 33. Added an ACL for QUIT.
3565 34. Setting "delay_warning=" to disable warnings was not working; it gave a
3568 35. Added mailbox_size and mailbox_filecount to appendfile.
3570 36. Added control = no_multiline_responses to ACLs.
3572 37. There was a bug in the logic of the code that waits for the clock to tick
3573 in the case where the clock went backwards by a substantial amount such
3574 that the microsecond fraction of "now" was more than the microsecond
3575 fraction of "then" (but the whole seconds number was less).
3577 38. Added support for the libradius Radius client library this is found on
3578 FreeBSD (previously only the radiusclient library was supported).
3584 1. When certain lookups returned multiple values in the form name=value, the
3585 quoting of the values was not always being done properly. Specifically:
3586 (a) If the value started with a double quote, but contained no whitespace,
3588 (b) If the value contained whitespace other than a space character (i.e.
3589 tabs or newlines or carriage returns) it was not quoted.
3590 This fix has been applied to the mysql and pgsql lookups by writing a
3591 separate quoting function and calling it from the lookup code. The fix
3592 should probably also be applied to nisplus, ibase and oracle lookups, but
3593 since I cannot test any of those, I have not disturbed their existing code.
3595 2. A hit in the callout cache for a specific address caused a log line with no
3596 reason for rejecting RCPT. Now it says "Previous (cached) callout
3597 verification failure".
3599 3. There was an off-by-one bug in the queryprogram router. An over-long
3600 return line was truncated at 256 instead of 255 characters, thereby
3601 overflowing its buffer with the terminating zero. As well as fixing this, I
3602 have increased the buffer size to 1024 (and made a note to document this).
3604 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
3605 when Exim is waiting for an SMTP response from a remote server, Exim
3606 restarts its select() call on the socket, thereby resetting its timeout.
3607 This is not a problem when such interrupts are rare. Somebody set up a cron
3608 job to run exiwhat every 2 minutes, which is less than the normal select()
3609 timeout (5 or 10 minutes). This meant that the select() timeout never
3610 kicked in because it was always reset. I have fixed this by comparing the
3611 time when an interrupt arrives with the time at the start of the first call
3612 to select(). If more time than the timeout has elapsed, the interrupt is
3613 treated as a timeout.
3615 5. Some internal re-factoring in preparation for the addition of Sieve
3616 extensions (by MH). In particular, the "personal" test is moved to a
3617 separate function, and given an option for scanning Cc: and Bcc: (which is
3618 not set for Exim filters).
3620 6. When Exim created an email address using the login of the caller as the
3621 local part (e.g. when creating a From: or Sender: header line), it was not
3622 quoting the local part when it contained special characters such as @.
3624 7. Installed new OpenBSD configuration files.
3626 8. Reworded some messages for syntax errors in "and" and "or" conditions to
3627 try to make them clearer.
3629 9. Callout options, other than the timeout value, were being ignored when
3630 verifying sender addresses in header lines. For example, when using
3632 verify = header_sender/callout=no_cache
3634 the cache was (incorrectly) being used.
3636 10. Added a missing instance of ${EXE} to the exim_install script; this affects
3637 only the Cygwin environment.
3639 11. When return_path_on_delivery was set as a log selector, if different remote
3640 addresses in the same message used different return paths and parallel
3641 remote delivery occurred, the wrong values would sometimes be logged.
3642 (Whenever a remote delivery process finished, the return path value from
3643 the most recently started remote delivery process was logged.)
3645 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
3646 lines when AUTH and/or TLS are in use. This is the "received protocol"
3647 field. Exim used to use "asmtp" for authenticated SMTP, without any
3648 indication (in the protocol name) for TLS use. Now it follows the RFC and
3649 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
3650 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
3651 names appear in log lines as well as in Received: header lines.
3653 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
3654 extensions, and comparison tests, and to fix some bugs.
3656 14. Changes to the "personal" filter test:
3658 (1) The test was buggy in that it was just doing the equivalent of
3659 "contains" tests on header lines. For example, if a user's address was
3660 anne@some.where, the "personal" test would incorrectly be true for
3662 To: susanne@some.where
3664 This test is now done by extracting each address from the header in turn,
3665 and checking the entire address. Other tests that are part of "personal"
3666 are now done using regular expressions (for example, to check local parts
3667 of addresses in From: header lines).
3669 (2) The list of non-personal local parts in From: addresses has been
3670 extended to include "listserv", "majordomo", "*-request", and "owner-*",
3671 taken from the Sieve specification recommendations.
3673 (3) If the message contains any header line starting with "List-" it is
3674 treated as non-personal.
3676 (4) The test for "circular" in the Subject: header line has been removed
3677 because it now seems ill-conceived.
3679 15. Minor typos in src/EDITME comments corrected.
3681 16. Installed latest exipick from John Jetmore.
3683 17. If headers_add on a router specified a text string that was too long for
3684 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
3685 of string_sprintf() is now avoided.
3687 18. $message_body_size was not set (it was always zero) when running the DATA
3688 ACL and the local_scan() function.
3690 19. For the "mail" command in an Exim filter, no default was being set for
3691 the once_repeat time, causing a random time value to be used if "once" was
3692 specified. (If the value happened to be <= 0, no repeat happened.) The
3693 default is now 0s, meaning "never repeat". The "vacation" command was OK
3694 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
3695 (I found it when inspecting the code).
3697 20. There is now an overall timeout for performing a callout verification. It
3698 defaults to 4 times the callout timeout, which applies to individual SMTP
3699 commands during the callout. The overall timeout applies when there is more
3700 than one host that can be tried. The timeout is checked before trying the
3701 next host. This prevents very long delays if there are a large number of
3702 hosts and all are timing out (e.g. when the network connections are timing
3703 out). The value of the overall timeout can be changed by specifying an
3704 additional sub-option for "callout", called "maxwait". For example:
3706 verify = sender/callout=5s,maxwait=20s
3708 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
3709 to the end before writing, but this should make it even safer).
3711 22. Exim was forgetting that it had advertised PIPELINING for the second and
3712 subsequent messages on an SMTP connection. It was also not resetting its
3713 memory on STARTTLS and an internal HELO.
3715 23. When Exim logs an SMTP synchronization error within a session, it now
3716 records whether PIPELINING has been advertised or not.
3718 24. Added 3 instances of "(long int)" casts to time_t variables that were being
3719 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
3720 rather than long int.
3722 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
3724 26. Added the never_mail option to autoreply.
3730 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
3731 crash if the getsockname() call failed; this can happen if a connection is
3732 closed very soon after it is established. The problem was simply in the
3733 order in which certain operations were done, causing Exim to try to write
3734 to the SMTP stream before it had set up the file descriptor. The bug has
3735 been fixed by making things happen in the correct order.
3741 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
3742 before the connection was closed, thus losing the rejection response.
3744 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
3745 some early Solaris releases, but causes trouble in current releases where
3746 socklen_t is defined.
3748 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
3751 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
3754 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
3755 that releasing the top of it at the end releases what was used for sub-
3756 expansions (unless the block got too big). However, discard this block if
3757 the first thing is a variable or header, so that we can use its block when
3758 it is dynamic (useful for very large $message_headers, for example).
3760 6. Lookups now cache *every* query, not just the most recent. A new, separate
3761 store pool is used for this. It can be recovered when all lookup caches are
3762 flushed. Lookups now release memory at the end of their result strings.
3763 This has involved some general refactoring of the lookup sources.
3765 7. Some code has been added to the store_xxx() functions to reduce the amount
3766 of flapping under certain conditions.
3768 8. log_incoming_interface used to affect only the <= reception log lines. Now
3769 it causes the local interface and port to be added to several more SMTP log
3770 lines, for example "SMTP connection from", and rejection lines.
3772 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
3774 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
3776 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
3777 could be overwritten at the end of the current message (or the start of a
3778 new message if it was set in a HELO ACL). The value is now preserved for
3779 the duration of the SMTP connection.
3781 12. If a transport had a headers_rewrite setting, and a matching header line
3782 contained an unqualified address, that address was qualified, even if it
3783 did not match any rewriting rules. The underlying bug was that the values
3784 of the flags that permit the existence of unqualified sender and recipient
3785 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
3786 non-local messages, and by -bnq for local messages) were not being
3787 preserved with the message after it was received.
3789 13. When Exim was logging an SMTP synchronization error, it could sometimes log
3790 "next input=" as part of the text comprising the host identity instead of
3791 the correct text. The code was using the same buffer for two different
3792 strings. However, depending on which order the printing function evaluated
3793 its arguments, the bug did not always show up. Under Linux, for example, my
3794 test suite worked just fine.
3796 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
3797 doesn't work with some older versions of Perl. It has been changed to "my",
3798 which in any case is probably the better facility to use.
3800 15. A really picky compiler found some instances of statements for creating
3801 error messages that either had too many or two few arguments for the format
3804 16. The size of the buffer for calls to the DNS resolver has been increased
3805 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
3806 for addresses that have a lot of PTR records. This alleviates a problem; it
3807 does not fully solve it.
3809 17. A dnsdb lookup for PTR records that receives more data than will fit in the
3810 buffer now truncates the list and logs the incident, which is the same
3811 action as happens when Exim is looking up a host name and its aliases.
3812 Previously in this situation something unpredictable would happen;
3813 sometimes it was "internal error: store_reset failed".
3815 18. If a server dropped the connection unexpectedly when an Exim client was
3816 using GnuTLS and trying to read a response, the client delivery process
3817 crashed while trying to generate an error log message.
3819 19. If a "warn" verb in an ACL added multiple headers to a message in a single
3820 string, for example:
3822 warn message = H1: something\nH2: something
3824 the text was added as a single header line from Exim's point of view
3825 though it ended up OK in the delivered message. However, searching for the
3826 second and subsequent header lines using $h_h2: did not work. This has been
3827 fixed. Similarly, if a system filter added multiple headers in this way,
3828 the routers could not see them.
3830 20. Expanded the error message when iplsearch is called with an invalid key to
3831 suggest using net-iplsearch in a host list.
3833 21. When running tests using -bh, any delays imposed by "delay" modifiers in
3834 ACLs are no longer actually imposed (and a message to that effect is
3837 22. If a "gecos" field in a passwd entry contained escaped characters, in
3838 particular, if it contained a \" sequence, Exim got it wrong when building
3839 a From: or a Sender: header from that name. A second bug also caused
3840 incorrect handling when an unquoted " was present following a character
3841 that needed quoting.
3843 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
3844 was not being matched caselessly.
3846 24. Arranged for all hyphens in the exim.8 source to be escaped with
3849 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
3850 properly. Recipient callout cache records were still being keyed to include
3851 the sender, even when use_sender was set false. This led to far more
3852 callouts that were necessary. The sender is no longer included in the key
3853 when use_sender is false.
3855 26. Added "control = submission" modifier to ACLs.
3857 27. Added the ${base62d: operator to decode base 62 numbers.
3859 28. dnsdb lookups can now access SRV records.
3861 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
3862 the configuration file.
3864 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
3865 (-v) mode. This makes the output for a verbose queue run more intelligible.
3867 31. Added a use_postmaster feature to recipient callouts.
3869 32. Added the $body_zerocount variable, containing the number of binary zero
3870 bytes in the message body.
3872 33. The time of last modification of the "new" subdirectory is now used as the
3873 "mailbox time last read" when there is a quota error for a maildir
3876 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
3878 35. Added +ignore_unknown as a special item in host lists.
3880 36. Code for decoding IPv6 addresses in host lists is now included, even if
3881 IPv6 support is not being compiled. This fixes a bug in which an IPv6
3882 address was recognized as an IP address, but was then not correctly decoded
3883 into binary, causing unexpected and incorrect effects when compared with
3890 1. Very minor rewording of debugging text in manualroute to say "list of
3891 hosts" instead of "hostlist".
3893 2. If verify=header_syntax was set, and a header line with an unqualified
3894 address (no domain) and a large number of spaces between the end of the
3895 name and the colon was received, the reception process suffered a buffer
3896 overflow, and (when I tested it) crashed. This was caused by some obsolete
3897 code that should have been removed. The fix is to remove it!
3899 3. When running in the test harness, delay a bit after writing a bounce
3900 message to get a bit more predictability in the log output.
3902 4. Added a call to search_tidyup() just before forking a reception process. In
3903 theory, someone could use a lookup in the expansion of smtp_accept_max_
3904 per_host which, without the tidyup, could leave open a database connection.
3906 5. Added the variables $recipient_data and $sender_data which get set from a
3907 lookup success in an ACL "recipients" or "senders" condition, or a router
3908 "senders" option, similar to $domain_data and $local_part_data.
3910 6. Moved the writing of debug_print from before to after the "senders" test
3913 7. Change 4.31/66 (moving the time when the Received: is generated) caused
3914 problems for message scanning, either using a data ACL, or using
3915 local_scan() because the Received: header was not generated till after they
3916 were called (in order to set the time as the time of reception completion).
3917 I have revised the way this works. The header is now generated after the
3918 body is received, but before the ACL or local_scan() are called. After they
3919 are run, the timestamp in the header is updated.
3925 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
3926 before starting a queue runner without re-exec. This happened only when
3927 deliver_drop_privilege was set or when the Exim user was set to root. The
3928 effect of the bug was that timeouts during subsequent deliveries caused
3929 crashes instead of being properly handled. The handler is now left at its
3930 default (and expected) setting.
3932 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
3933 message, again when deliver_drop_privilege is set or Exim is run as root.
3934 The bug described in (1) was not present in this case, but the tidying up
3935 of the other signals was missing. I have made the two cases consistent.
3937 3. The ignore_target_hosts setting on a manualroute router was being ignored
3938 for hosts that were looked up using the /MX notation.
3940 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
3943 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
3944 operated on the sender address. After changing the $sender_address to <>
3945 for the sender address verify, Exim was re-instated it as the original
3946 (before rewriting) address, but remembering that it had rewritten it, so it
3947 wasn't rewriting it again. This bug also had the effect of breaking the
3948 sender address verification caching when the sender address was rewritten.
3950 6. The ignore_target_hosts option was being ignored by the ipliteral router.
3951 This has been changed so that if the ip literal address matches
3952 ignore_target_hosts, the router declines.
3954 7. Added expansion conditions match_domain, match_address, and match_local_
3955 part (NOT match_host).
3957 8. The placeholder for the Received: header didn't have a length field set.
3959 9. Added code to Exim itself and to exim_lock to test for a specific race
3960 condition that could lead to file corruption when using MBX delivery. The
3961 issue is with the lockfile that is created in /tmp. If this file is removed
3962 after a process has opened it but before that process has acquired a lock,
3963 there is the potential for a second process to recreate the file and also
3964 acquire a lock. This could lead to two Exim processes writing to the file
3965 at the same time. The added code performs the same test as UW imapd; it
3966 checks after acquiring the lock that its file descriptor still refers to
3967 the same named file.
3969 10. The buffer for building added header lines was of fixed size, 8192 bytes.
3970 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
3973 11. Added the smtp_active_hostname option. If used, this will typically be made
3974 to depend on the incoming interface address. Because $interface_address is
3975 not set up until the daemon has forked a reception process, error responses
3976 that can happen earlier (such as "too many connections") no longer contain
3979 12. If an expansion in a condition on a "warn" statement fails because a lookup
3980 defers, the "warn" statement is abandoned, and the next ACL statement is
3981 processed. Previously this caused the whole ACL to be aborted.
3983 13. Added the iplsearch lookup type.
3985 14. Added ident_timeout as a log selector.
3987 15. Added tls_certificate_verified as a log selector.
3989 16. Added a global option tls_require_ciphers (compare the smtp transport
3990 option of the same name). This controls incoming TLS connections.
3992 17. I finally figured out how to make tls_require_ciphers do a similar thing
3993 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
3994 before starting the TLS session.
3996 18. Tabs are now shown as \t in -bP output.
3998 19. If the log selector return_path_on_delivery was set, Exim crashed when
3999 bouncing a message because it had too many Received: header lines.
4001 20. If two routers both had headers_remove settings, and the first one included
4002 a superfluous trailing colon, the final name in the first list and the
4003 first name in the second list were incorrectly joined into one item (with a
4004 colon in the middle).
4010 1. Added -C and -D options to the exinext utility, mainly to make it easier
4011 to include in the automated testing, but these could be helpful when
4012 multiple configurations are in use.
4014 2. The exinext utility was not formatting the output nicely when there was
4015 an alternate port involved in the retry record key, nor when there was a
4016 message id as well (for retries that were specific to a specific message
4017 and a specific host). It was also confused by IPv6 addresses, because of
4018 the additional colons they contain. I have fixed the IPv4 problem, and
4019 patched it up to do a reasonable job for IPv6.
4021 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
4022 delivery, the log line now contains "pipelined" if PIPELINING was used.
4024 4. An SMTP transport process used to panic and die if the bind() call to set
4025 an explicit outgoing interface failed. This has been changed; it is now
4026 treated in the same way as a connect() failure.
4028 5. A reference to $sender_host_name in the part of a conditional expansion
4029 that was being skipped was still causing a DNS lookup. This no longer
4032 6. The def: expansion condition was not recognizing references to header lines
4033 that used bh_ and bheader_.
4035 7. Added the _cache feature to named lists.
4037 8. The code for checking quota_filecount in the appendfile transport was
4038 allowing one more file than it should have been.
4040 9. For compatibility with Sendmail, the command line option
4048 and sets the incoming protocol and host name (for trusted callers). The
4049 host name and its colon can be omitted when only the protocol is to be set.
4050 Note the Exim already has two private options, -pd and -ps, that refer to
4051 embedded Perl. It is therefore impossible to set a protocol value of "d" or
4052 "s", but I don't think that's a major issue.
4054 10. A number of refactoring changes to the code, none of which should affect
4057 (a) The number of logging options was getting close to filling up the
4058 32-bit word that was used as a bit map. I have split them into two classes:
4059 those that are passed in the argument to log_write(), and those that are
4060 only ever tested independently outside of that function. These are now in
4061 separate 32-bit words, so there is plenty of room for expansion again.
4062 There is no change in the user interface or the logging behaviour.
4064 (b) When building, for example, log lines, the code previously used a
4065 macro that called string_cat() twice, in order to add two strings. This is
4066 not really sufficiently general. Furthermore, there was one instance where
4067 it was actually wrong because one of the argument was used twice, and in
4068 one call a function was used. (As it happened, calling the function twice
4069 did not affect the overall behaviour.) The macro has been replaced by a
4070 function that can join an arbitrary number of extra strings onto a growing
4073 (c) The code for expansion conditions now uses a table and a binary chop
4074 instead of a serial search (which was left over from when there were very
4075 few conditions). Also, it now recognizes conditions like "pam" even when
4076 the relevant support is not compiled in: a suitably worded error message is
4077 given if an attempt is made to use such a condition.
4079 11. Added ${time_interval:xxxxx}.
4081 12. A bug was causing one of the ddress fields not to be passed back correctly
4082 from remote delivery subprocesses. The field in question was not being
4083 subsequently used, so this caused to problems in practice.
4085 13. Added new log selectors queue_time and deliver_time.
4087 14. Might have fixed a bug in maildirsizefile handling that threw up
4088 "unexpected character" debug warnings, and recalculated the data
4089 unnecessarily. In any case, I expanded the warning message to give more
4092 15. Added the message "Restricted characters in address" to the statements in
4093 the default ACL that block characters like @ and % in local parts.
4095 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
4096 Three changes have been made:
4098 (a) There was a serious bug; a negative response to MAIL caused the whole
4099 recipient domain to be cached as invalid, thereby blocking all messages
4100 to all local parts at the same domain, from all senders. This bug has
4101 been fixed. The domain is no longer cached after a negative response to
4102 MAIL if the sender used is not empty.
4104 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
4107 (c) A new callout option, "use_sender" has been added for people who want
4108 the modified behaviour.
4114 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
4117 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
4118 indeed breaks things for older releases.
4120 3. Added additional logging to the case where there is a problem reading data
4121 from a filter that is running in a subprocess using a pipe, in order to
4122 try to track down a specific problem.
4124 4. Testing facility fudge: when running in the test harness and attempting
4125 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
4126 getting "No route to host". Convert this to a timeout.
4128 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
4131 6. Some OS don't have socklen_t but use size_t instead. This affects the
4132 fifth argument of getsockopt() amongst other things. This is now
4133 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
4134 can be set for individual OS. I have set it for SunOS5, OSF1, and
4135 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
4136 some earlier ones do not.
4138 7. Change 4.30/15 was not doing the test caselessly.
4140 8. The standard form for an IPv6 address literal was being rejected by address
4141 parsing in, for example, MAIL and RCPT commands. An example of this kind of
4142 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
4143 this, as well as the form without the "IPv6" on the front (but only when
4144 address literals are enabled, of course).
4146 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
4148 10. Exim crashed if a message with an empty sender address specified by -f
4149 encountered a router with an errors_to setting. This could be provoked only
4150 by a command such as
4154 where an empty string was supplied; "<>" did not hit this bug.
4156 11. Installed PCRE release 4.5.
4158 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
4159 remained set. It is now erased.
4161 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
4162 times from message ids (which are base 36 rather than the normal 62).
4164 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
4165 were being counted as actual protocol errors, and logged if the log
4166 selector +smtp_protocol_error was set. One cannot be perfect in this test,
4167 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
4168 and DATA following a set of rejected RCPTs do not count as protocol errors.
4169 In other words, Exim assumes they were pipelined, though this may not
4170 actually be the case. Of course, in all cases the client gets an
4171 appropriate error code.
4173 15. If a lookup fails in an ACL condition, a message about the failure may
4174 be available; it is used if testing the ACL cannot continue, because most
4175 such messages specify what the cause of the deferral is. However, some
4176 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
4177 that caused an old message to be retained and used if a later statement
4178 caused a defer, replacing the real cause of the deferral.
4180 16. If an IP address had so many PTR records that the DNS lookup buffer
4181 was not large enough to hold them, Exim could crash while trying to process
4182 the truncated data. It now detects and logs this case.
4184 17. Further to 4.21/58, another change has been made: if (and only if) the
4185 first line of a message (the first header line) ends with CRLF, a bare LF
4186 in a subsequent header line has a space inserted after it, so as not to
4187 terminate the header.
4189 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
4190 unnecessarily, used atoi() instead of strtol(), and didn't check the
4191 termination when getting file sizes from file names by regex.
4193 19. Completely re-implemented the support for maildirsize files, in the light
4194 of a number of problems with the previous contributed implementation
4195 (4.30/29). In particular:
4197 . If the quota is zero, the maildirsize file is maintained, but no quota is
4200 . If the maildir directory does not exist, it is created before any attempt
4201 to write a maildirsize file.
4203 . The quota value in the file is just a cache; if the quota is changed in
4204 the transport, the new value overrides.
4206 . A regular expression is available for excluding directories from the
4209 20. The autoreply transport checks the characters in options that define the
4210 message's headers; it allows continued headers, but it was checking with
4211 isspace() after an embedded newline instead of explicitly looking for a
4214 21. If all the "regular" hosts to which an address was routed had passed their
4215 expiry times, and had not reached their retry times, the address was
4216 bounced, even if fallback hosts were defined. Now Exim should go on to try
4219 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
4220 equivalent code in the SMTP transport. Some hosts send humungous responses
4221 to HELO/EHLO, more than 1024 it seems.
4223 23. Refactoring: code in filter.c used (void *) for "any old type" but this
4224 gives compiler warnings in some environments. I've now done it "properly",
4227 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
4228 (because of problems with the built-in one) was declared to return uschar *
4229 instead of char *, causing compiler failure.
4231 25. Fixed a file descriptor leak when processing alias/forward files.
4233 26. Fixed a minor format string issue in dbfn.c.
4235 27. Typo in exim.c: ("dmbnz" for "dbmnz").
4237 28. If a filter file refered to $h_xxx or $message_headers, and the headers
4238 contained RFC 2047 "words", Exim's memory could, under certain conditions,
4241 29. When a sender address is verified, it is cached, to save repeating the test
4242 when there is more than one recipient in a message. However, when the
4243 verification involves a callout, it is possible for different callout
4244 options to be set for different recipients. It is too complicated to keep
4245 track of this in the cache, so now Exim always runs a verification when a
4246 callout is required, relying on the callout cache for the optimization.
4247 The overhead is duplication of the address routing, but this should not be
4250 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
4251 to be verified with callout=postmaster, and the main callout worked but the
4252 postmaster check failed, the verification correctly failed. However, if a
4253 subsequent RCPT command asked for sender verification *without* the
4254 postmaster check, incorrect caching caused this verification also to fail,
4257 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
4258 it was not caching the DNS options (qualify_single, search_parents) that
4259 were used when the lookup failed. A subsequent lookup with different
4260 options therefore always gave the same answer, though there were cases
4261 where it should not have. (Example: a "domains = !$mx_any" option on a
4262 dnslookup router: the "domains" option is always processed without any
4263 widening, but the router might have qualify_single set.) Now Exim uses the
4264 cached value only when the same options are set.
4266 32. Added John Jetmore's "exipick" utility to the distribution.
4268 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
4269 than a timeout (e.g. a certificate is required, and is not provided), an
4270 Exim server now closes the connection immediately. Previously it waited for
4271 the client to close - but if the client is SSL, it seems that they each
4272 wait for each other, leading to a delay before one of them times out.
4274 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
4275 maintained 0.8.x compatibility because I don't think many are using it, and
4276 it is clearly obsolete.
4278 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
4281 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
4282 client certificate was expired. A simple patch fixes this, though I don't
4283 understand the full logic of why the verify callback is called multiple
4286 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
4287 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
4288 which causes problems with some clients (such as the Certicom SSL Plus
4289 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
4290 disables the coutermeasure allowing Eudora to connect."
4292 38. Exim was not checking that a write() to a log file succeeded. This could
4293 lead to Bad Things if a log got too big, in particular if it hit a file
4294 size limit. Exim now panics and dies if it cannot write to a log file, just
4295 as it does if it cannot open a log file.
4297 39. Modified OS/Makefile-Linux so that it now contains
4299 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
4301 The two -D definitions ensure that Exim is compiled with large file
4302 support, which makes it possible to handle log files that are bigger than
4305 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
4306 instance) a domain was checked against a named list that involved a lookup,
4307 causing $domain_data to be set, then another domain was checked against the
4308 same list, then the first domain was re-checked, the value of $domain_data
4309 after the final check could be wrong. In particular, if the second check
4310 failed, it could be set empty. This bug probably also applied to
4313 41. The strip_trailing_dot option was not being applied to the address given
4314 with the -f command-line option.
4316 42. The code for reading a message's header from the spool was incrementing
4317 $received_count, but never initializing it. This meant that the value was
4318 incorrect (doubled) while delivering a message in the same process in which
4319 it was received. In the most common configuration of Exim, this never
4320 happens - a fresh exec is done - but it can happen when
4321 deliver_drop_privilege is set.
4323 43. When Exim logs an SMTP synchronization error - client data sent too soon -
4324 it now includes up to 150 characters of the unexpected data in the log
4327 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
4328 and building data strings. The size of both of these buffers was 10 000
4329 bytes - far larger than anybody would *ever* want, thought I. Needless to
4330 say, somebody hit the limit. I have increased the maximum line length to
4331 20 000 and the maximum data length of concatenated lines to 100 000. I have
4332 also fixed two bugs, because there was no checking on these buffers. Tsk,
4333 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
4334 buffer is too small.
4336 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
4337 lsearch lookups. Now it does.
4339 46. When parsing a route_list item in a manualroute router, a fixed-length
4340 buffer was used for the list of hosts. I made this 1024 bytes long,
4341 thinking that nobody would ever have a list of hosts that long. Wrong.
4342 Somebody had a whole pile of complicated expansion conditions, and the
4343 string was silently truncated, leading to an expansion error. It turns out
4344 that it is easier to change to an unlimited length (owing to other changes
4345 that have happened since this code was originally written) than to build
4346 structure for giving a limitation error. The length of the item that
4347 expands into the list of hosts is now unlimited.
4349 47. The lsearch lookup could not handle data where the length of text line was
4350 more than 4095 characters. Such lines were truncated, leading to shortened
4351 data being returned. It should now handle lines of any length.
4353 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
4354 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
4357 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
4358 replacing strings such as BIN_DIRECTORY in a source file: the replacement
4359 no longer happens in comment lines. A list of replacements is now placed
4360 at the head of all of the source files, except those whose only change is
4361 to replace PERL_COMMAND in the very first #! line.
4363 50. Replaced the slow insertion sort in queue.c, for sorting the list of
4364 messages on the queue, with a bottom-up merge sort, using code contributed
4365 by Michael Haardt. This should make operations like -bp somewhat faster on
4366 large queues. It won't affect queue runners, except when queue_run_in_order
4369 51. Installed eximstats 1.31 in the distribution.
4371 52. Added support for SRV lookups to the dnslookup router.
4373 53. If an ACL referred to $message_body or $message_body_end, the value was not
4374 reset for any messages that followed in the same SMTP session.
4376 54. The store-handling optimization for building very long strings was not
4377 differentiating between the different store pools. I don't think this
4378 actually made any difference in practice, but I've tidied it.
4380 55. While running the routers to verify a sender address, $sender_address
4381 was still set to the sender address. This is wrong, because when routing to
4382 send a bounce to the sender, it would be empty. Therefore, I have changed
4383 it so that, while verifying a sender address, $sender_address is set to <>.
4384 (There is no change to what happens when verifying a recipient address.)
4386 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
4387 target A or AAAA records (if not already returned) without resetting the
4388 qualify_single or search_parents options of the DNS resolver. These are
4389 inappropriate in this case because the targets of MX and SRV records must
4390 be FQDNs. A broken DNS record could cause trouble if it happened to have a
4391 target that, when qualified, matched something in the local domain. These
4392 two options are now turned off when doing these lookups.
4394 57. It seems that at least some releases of Reiserfs (which does not have the
4395 concept of a fixed number of inodes) returns zero and not -1 for the
4396 number of available inodes. This interacted badly with check_spool_inodes,
4397 which assumed that -1 was the "no such thing" setting. What I have done is
4398 to check that the total number of inodes is greater than zero before doing
4399 the test of how many are available.
4401 58. When a "warn" ACL statement has a log_message modifier, the message is
4402 remembered, and not repeated. This is to avoid a lot of repetition when a
4403 message has many recipients that cause the same warning to be written.
4404 Howewer, Exim was preserving the list of already written lines for an
4405 entire SMTP session, which doesn't seem right. The memory is now reset if a
4406 new message is started.
4408 59. The "rewrite" debugging flag was not showing the result of rewriting in the
4409 debugging output unless log_rewrite was also set.
4411 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
4412 of (int)(handle) when we know that handle contains (void *)(-1).
4414 61. The Exim daemon panic-logs an error return when it closes the incoming
4415 connection. However "connection reset by peer" seems to be common, and
4416 isn't really an error worthy of noting specially, so that particular error
4419 62. When Exim is trying to find all the local interfaces, it used to panic and
4420 die if the ioctl to get the interface flags failed. However, it seems that
4421 on at least one OS (Solaris 9) it is possible to have an interface that is
4422 included in the list of interfaces, but for which you get a failure error
4423 for this call. This happens when the interface is not "plumbed" into a
4424 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
4425 failure of the "get flags" call assumes that the interface is down.
4427 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
4428 makes life easier for people who are doing arithmetic on fields extracted
4429 from dates, where you often get leading zeros that should not be
4430 interpreted as octal.
4432 64. Added qualify_domain to the redirect router, to override the global
4435 65. If a pathologically long header line contained very many addresses (the
4436 report of this problem mentioned 10 000) and each of them was rewritten,
4437 Exim could use up a very large amount of memory. (It kept on making new
4438 copies of the header line as it rewrote, and never released the old ones.)
4439 At the expense of a bit more processing, the header rewriting function has
4440 been changed so that it no longer eats memory in this way.
4442 66. The generation of the Received: header has been moved from the time that a
4443 message starts to be received, to the time that it finishes. The timestamp
4444 in the Received: header should now be very close to that of the <= log
4445 line. There are two side-effects of this change:
4447 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
4448 logged header lines no longer include the local Received: line, because
4449 it has not yet been created. The same applies to a copy of the message
4450 that is returned to a non-SMTP sender when a message is rejected.
4452 (b) When a filter file is tested using -bf, no additional Received: header
4453 is added to the test message. After some thought, I decided that this
4456 This change does not affect the value of $received_for. It is still set
4457 after address rewriting, but before local_scan() is called.
4459 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
4461 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
4462 gave an unhelpful panic error message, and a defer error. I have managed to
4463 change this behaviour so that it now rejects any supplied certificate,
4464 which seems right, as the list of acceptable certificates is empty.
4466 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
4467 gave an unhelpful defer error. I have not managed to make this reject any
4468 supplied certificates, but the error message it gives is "no certificate
4469 supplied", which is not helpful.
4471 70. exigrep's output now also includes lines that are not associated with any
4472 message, but which match the given pattern. Implemented by a patch from
4473 Martin Sluka, which also tidied up the Perl a bit.
4475 71. Recipient callout verification, like sender verification, was using <> in
4476 the MAIL FROM command. This isn't really the right thing, since the actual
4477 sender may affect whether the remote host accepts the recipient or not. I
4478 have changed it to use the actual sender in the callout; this means that
4479 the cache record is now keyed on a recipient/sender pair, not just the
4480 recipient address. There doesn't seem to be a real danger of callout loops,
4481 since a callout by the remote host to check the sender would use <>.
4482 [SEE ABOVE: changed after hitting problems.]
4484 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
4485 temporary errors. However, in the case of such a code being given after
4486 the end of a data transmission (i.e. after ".") Exim was failing to write
4487 a retry record for the message. (Yes, there was some broken host that was
4488 actually sending 8xx at this point.)
4490 73. An unknown lookup type in a host list could cause Exim to panic-die when
4491 the list was checked. (An example that provoked this was putting <; in the
4492 middle of a list instead of at the start.) If this happened during a DATA
4493 ACL check, a -D file could be left lying around. This kind of configuration
4494 error no longer causes Exim to die; instead it causes a defer errror. The
4495 incident is still logged to the main and panic logs.
4497 74. Buglet left over from Exim 3 conversion. The message "too many messages
4498 in one connection" was written to the rejectlog but not the mainlog, except
4499 when address rewriting (yes!) was being logged.
4501 75. Added write_rejectlog option.
4503 76. When a system filter was run not as root (that is, when system_filter_user
4504 was set), the values of the $n variables were not being returned to the
4505 main process; thus, they were not subsequently available in the $sn
4508 77. Added +return_path_on_delivery log selector.
4510 78. A connection timeout was being treated differently from recipients deferred
4511 when testing hosts_max_try with a message that was older than the host's
4512 retry timeout. (The host should not be counted, thus allowing all hosts to
4513 be tried at least once before bouncing.) This may have been the cause of an
4514 occasionally reported bug whereby a message would remain on the queue
4515 longer than the retry timeout, but would be bounced if a delivery was
4516 forced. I say "may" because I never totally pinned down the problem;
4517 setting up timeout/retry tests is difficult. See also the next item.
4519 79. The ultimate address timeout was not being applied to errors that involved
4520 a combination of host plus message (for example, a timeout on a MAIL
4521 command). When an address resolved to a number of possible hosts, and they
4522 were not all tried for each delivery (e.g. because of hosts_max_try), a
4523 message could remain on the queue longer than the retry timeout.
4525 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
4528 81. Fixed an obscure SMTP outgoing bug which required at least the following
4529 conditions: (a) there was another message waiting for the same server;
4530 (b) the server returned 5xx to all RCPT commands in the first message so
4531 that the message was not completed; (c) the server dropped the connection
4532 or gave a negative response to the RSET that Exim sends to abort the
4533 transaction. The observed case was a dropped connection after DATA that had
4534 been sent in pipelining mode. That is, the server had advertised PIPELINING
4535 but was not implementing it correctly. The effect of the bug was incorrect
4536 behaviour, such as trying another host, and this could lead to a crash.
4542 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
4543 and daemon.c were passed as pointers to ints; they should have been
4544 pointers to socklen_t variables (which are typically unsigned ints).
4546 2. Some signed/unsigned type warnings in the os.c file for Linux have been
4549 3. Fixed a really odd bug that affected only the testing scheme; patching a
4550 certain fixed string in the binary changed the value of another string that
4551 happened to be identical to the end of the original first string.
4553 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
4554 name", it returns that address as the IP address. On some operating
4555 systems (e.g. Solaris), it also passes back the IP address string as the
4556 "host name". However, on others (e.g. Linux), it passes back an empty
4557 string. Exim wasn't checking for this, and was changing the host name to an
4558 empty string, assuming it had been canonicized.
4560 5. Although rare, it is permitted to have more than one PTR record for a given
4561 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
4562 all the names associated with an address, because they do in Solaris.
4563 However, it seems that they do not in Linux for data that comes from the
4564 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
4565 I found this out when I moved to a new Linux workstation and tried to run
4566 the Exim test suite.
4568 To get round this problem I have changed the code so that it now does its
4569 own call to the DNS to look up PTR records when searching for a host name.
4570 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
4571 addresses that are only in /etc/hosts are still found.
4573 This behaviour is, however, controlled by an option called host_lookup_
4574 order, which defaults to "bydns:byaddr". If people want to use the other
4575 order, or indeed, just use one or the other means of lookup, they can
4576 specify it in this variable.
4578 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
4579 some operating systems, this comes back from gethostbyaddr() as an empty
4580 string, and this is what Exim used to test for. However, it seems that in
4581 other systems, "." is yielded. Exim now tests for this case too.
4583 7. The values of check_spool_space and check_log_space are now held internally
4584 as a number of kilobytes instead of an absolute number of bytes. If a
4585 numbers is specified without 'K' or 'M', it is rounded up to the nearest
4586 kilobyte. This means that much larger values can be stored.
4588 8. Exim monitor: an attempt to get the action menu when not actually pointing
4589 at a message produces an empty menu entitled "No message selected". This
4590 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
4591 no entries in it ("Shell widget menu has zero width and/or height"). So I
4592 have added a single, blank menu entry in this case.
4594 9. Added ${quote_local_part.
4596 10. MIME decoding is now applied to the contents of Subject: header lines when
4599 11. Now that a reference to $sender_host_address automatically causes a reverse
4600 lookup to occur if necessary (4.13/18), there is no need to arrange for a
4601 host lookup before query-style lookups in lists that might use this
4602 variable. This has therefore been abolished, and the "net-" prefix is no
4603 longer necessary for query-style lookups.
4605 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
4606 have been a typo for LFLAGS, so it has been changed.
4608 13. The install script calls Exim with "-C /dev/null" in order to find the
4609 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
4610 to be output. Howeve, since Exim outputs its version number before the
4611 error, it didn't break the script. It just looked ugly. I fixed this by
4612 always allowing "-C /dev/null" if the caller is root.
4614 14. Ignore overlarge ACL variable number when reading spool file - insurance
4615 against a later release with more variables having written the file.
4617 15. The standard form for an IPv6 address literal was being rejected by EHLO.
4618 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
4619 this, as well as the form without the "IPv6" on the front.
4621 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
4622 OS/Makefile-Darwin file.
4624 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
4625 only to LDAP libraries that do not have LDAP_OPT_DEREF.
4627 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
4628 variable. However, some OS use ints for inodes. I've added cast to long int
4629 to get rid of the compiler warning.
4631 19. I had forgotten to lock out "/../" in configuration file names when
4632 ALT_CONFIG_PREFIX was set.
4634 20. Routers used for verification do not need to specify transports. However,
4635 if such a router generated a host list, and callout was configured, Exim
4636 crashed, because it could not find a port number from the (non-existent)
4637 transport. It now assumes port 25 in this circumstance.
4639 21. Added the -t option to exigrep.
4641 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
4642 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
4643 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
4644 them is set, LOOKUP_LSEARCH is forced.
4646 23. "exim -bV" now outputs a list of lookups that are included in the binary.
4648 24. Added sender and host information to the "rejected by local_scan()" log
4649 line; previously there was no indication of these.
4651 25. Added .include_if_exists.
4653 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
4654 Linux. It turns out that not all file systems support this. Apparently some
4655 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
4656 do it.) To cope with this, the error EINVAL, which means that sync-ing is
4657 not supported on the file descriptor, is now ignored when Exim is trying to
4658 sync a directory. This applies only to Linux.
4660 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
4662 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
4663 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
4666 29. Added support for maildirsize files from supplied patch (modified a bit).
4668 30. The use of :fail: followed by an empty string could lead Exim to respond to
4669 sender verification failures with (e.g.):
4671 550 Verification failed for <xxx>
4672 550 Sender verify failed
4674 where the first response line was missing the '-' that indicates it is not
4675 the final line of the response.
4677 31. The loop for finding the name of the user that called Exim had a hardwired
4678 limit of 10; it now uses the value of finduser_retries, which is used for
4679 all other user lookups.
4681 32. Added $received_count variable, available in data and not_smtp ACLs, and at
4684 33. Exim was neglecting to zero errno before one call of strtol() when
4685 expanding a string and expecting an integer value. On some systems this
4686 resulted in spurious "integer overflow" errors. Also, it was casting the
4687 result into an int without checking.
4689 34. Testing for a connection timeout using "timeout_connect" in the retry rules
4690 did not work. The code looks as if it has *never* worked, though it appears
4691 to have been documented since at least releast 1.62. I have made it work.
4693 35. The "timeout_DNS" error in retry rules, also documented since at least
4694 1.62, also never worked. As it isn't clear exactly what this means, and
4695 clearly it isn't a major issue, I have abolished the feature by treating it
4696 as "timeout", and writing a warning to the main and panic logs.
4698 36. The display of retry rules for -brt wasn't always showing the error code
4701 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
4702 timeout_connect_A, timeout_connect_MX.
4704 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
4705 to the empty sender.
4707 39. The daemon was not analysing the content of -oX till after it had closed
4708 stderr and disconnected from the controlling terminal. This meant that any
4709 syntax errors were only noted on the panic log, and the return code from
4710 the command was 0. By re-arranging the code a little, I've made the
4711 decoding happen first, so such errors now appear on stderr, and the return
4712 code is 1. However, the actual setting up of the sockets still happens in
4713 the disconnected process, so errors there are still only recorded on the
4716 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
4717 connections (as happens on some IP stacks) was logged at start up time as
4718 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
4719 it from "IPv6 and IPv4", which means that two separate sockets are being
4722 41. The debug output for gethostbyname2() or getipnodebyname() failures now
4723 says whether AF_INET or AF_INET6 was passed as an argument.
4725 42. Exiwhat output was messed up when time zones were included in log
4728 43. Exiwhat now gives more information about the daemon's listening ports,
4729 and whether -tls-on-connect was used.
4731 44. The "port" option of the smtp transport is now expanded.
4733 45. A "message" modifier in a "warn" statement in a non-message ACL was being
4734 silently ignored. Now an error message is written to the main and panic
4737 46. There's a new ACL modifier called "logwrite" which writes to a log file
4738 as soon as it is encountered.
4740 47. Added $local_user_uid and $local_user_gid at routing time.
4742 48. Exim crashed when trying to verify a sender address that was being
4745 49. Exim was recognizing only a space character after ".include". It now also
4746 recognizes a tab character.
4748 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
4749 extracting the relevant information from the specification. The man page no
4750 longer contains scrambled data for the -d option, and I've added a section
4751 at the front about calling Exim under different names.
4753 51. Added "extra_headers" argument to the "mail" command in filter files.
4755 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
4758 53. Installed eximstats 1.29.
4760 54. Added transport_filter_timeout as a generic transport option.
4762 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
4763 Cc: header lines. This was required by RFC 822, but it not required by RFC
4766 56. Exim used to add From:, Date:, and Message-Id: header lines to any
4767 incoming messages that did not have them. Now it does so only if the
4768 message originates locally, that is, if there is no associated remote host
4769 address. When Resent- header lines are present, this applies to the Resent-
4770 lines rather than the non-Resent- lines.
4772 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
4773 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
4775 58. Messages for configuration errors now include the name of the main
4776 configuration file - useful now that there may be more than one file in a
4777 list (.included file names were always shown).
4779 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
4780 for those rare installations that do not start the daemon as root or run it
4781 setuid root. I've cut out the call to initgroups() if the daemon is not
4784 60. The Exim user and group can now be bound into the binary as text strings
4785 that are looked up at the start of Exim's processing.
4787 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
4789 62. Added $mailstore_basename variable.
4791 63. Installed patch to sieve.c from Michael Haardt.
4793 64. When Exim failed to open the panic log after failing to open the main log,
4794 the original message it was trying to log was written to stderr and debug
4795 output, but if they were not available (the usual case in production), it
4796 was lost. Now it is written to syslog before the two lines that record the
4797 failures to open the logs.
4799 65. Users' Exim filters run in subprocesses under the user's uid. It is
4800 possible for a "deliver" command or an alias in a "personal" command to
4801 provoke an address rewrite. If logging of address rewriting is configured,
4802 this fails because the process is not running as root or exim. There may be
4803 a better way of dealing with this, but for the moment (because 4.30 needs
4804 to be released), I have disabled address rewrite logging when running a
4805 filter in a non-root, non-exim process.
4811 1. The buildconfig auxiliary program wasn't quoting the value set for
4812 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
4813 not defined. This bug was masked in 4.22 by the effect that was fixed in
4816 2. Some messages that were rejected after a message id was allocated were
4817 shown as "incomplete" by exigrep. It no longer does this for messages that
4818 are rejected by local_scan() or the DATA or non-SMTP ACLs.
4820 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
4821 have allow_domain_literals set, the ID did not get logged in the <= line.
4822 Domain literals are now always recognized in Message-ID: header lines.
4824 4. The first argument for a ${extract expansion item is the key name or field
4825 number. Leading and trailing spaces in this item were not being ignored,
4826 causing some misleading effects.
4828 5. When deliver_drop_privilege was set, single queue runner processes started
4829 manually (i.e. by the command "exim -q") or by the daemon (which uses the
4830 same command in the process it spins off) were not dropping privilege.
4832 6. When the daemon running as "exim" started a queue runner, it always
4833 re-executed Exim in the spun-off process. This is a waste of effort when
4834 deliver_drop_privilege is set. The new process now just calls the
4835 queue-runner function directly.
4841 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
4844 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
4845 ignored. Though the use of -oP was forcing the writing of a pid file, it
4846 was always written to the default place.
4848 3. If the message "no IP address found for host xxxx" is generated during
4849 incoming verification, it is now followed by identification of the incoming
4850 connection (so you can more easily find what provoked it).
4852 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
4854 5. Added some features to "harden" Exim a bit more against certain attacks:
4856 (a) There is now a build-time option called FIXED_NEVER_USERS that can
4857 be put in Local/Makefile. This is like the never_users runtime option,
4858 but it cannot be overridden. The default setting is "root".
4860 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
4861 prefix string with which any file named in a -C command line option
4864 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
4865 is retained for -C and -D only if the caller of Exim is root. Without
4866 it, the exim user may also use -C and -D and retain privilege.
4868 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
4869 command line option is disabled.
4871 6. Macro names set by the -D option must start with an upper case letter, just
4872 like macro names defined in the configuration file.
4874 7. Added "dereference=" facility to LDAP.
4876 8. Two instances of the typo "uknown" in the source files are fixed.
4878 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
4879 the Configure-Makefile script screwed up while processing it.
4881 10. Incorporated PCRE 4.4.
4883 11. The SMTP synchronization check was not operating right at the start of an
4884 SMTP session. For example, it could not catch a HELO sent before the client
4885 waited for the greeting. There is now a check for outstanding input at the
4886 point when the greeting is written. Because of the duplex, asynchronous
4887 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
4888 way, but not yet received, when the check is performed.
4890 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
4891 on TCP/IP sockets, because this apparently causes some broken clients to
4894 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
4895 unchanged) from the Cygwin maintainer.
4897 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
4898 was supported instead of testing for mbx. Effectively a typo.
4900 15. The spa authenticator server code was not checking that the input it
4901 received was valid base64.
4903 16. The debug output line for the "set" modifier in ACLs was not showing the
4904 name of the variable that was being set.
4906 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
4908 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
4909 lookup on its own. Something else was needed to trigger the lookup. For
4910 example, a match in host_lookup or the need for a host name in a host list.
4911 Now, if $sender_host_name is referenced and the host name has not yet been
4912 looked up, a lookup is performed. If the lookup fails, the variable remains
4913 empty, and $host_lookup_failed is set to "1".
4915 19. Added "eqi" as a case-independent comparison operator.
4917 20. The saslauthd authentication condition could segfault if neither service
4918 nor realm was specified.
4920 21. If an overflowing value such as "2048M" was set for message_size_limit, the
4921 error message that was logged was misleading, and incoming SMTP
4922 connections were dropped. The message is now more accurate, and temporary
4923 errors are given to SMTP connections.
4925 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
4926 (except RSET) with a 421 error, until QUIT is received. However, it was
4927 failing to send a response to QUIT.
4929 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
4930 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
4931 ACL is now run after the helo_try_verify_hosts code.
4933 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
4934 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
4935 case-independent, so other case variants are also recognized). Apparently
4936 some systems use these upper case variants.
4938 25. If more than two messages were waiting for the same host, and a transport
4939 filter was specified for the transport, Exim sent two messages over the
4940 same TCP/IP connection, and then failed with "socket operation on non-
4941 socket" when it tried to send the third.
4943 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
4945 27. The extern definition of crypt16() in expand.c was not being excluded when
4946 the OS had its own crypt16() function.
4948 28. Added bounce_return_body as a new option, and bounce_return_size_limit
4949 as a preferred synonym for return_size_limit, both as an option and as an
4952 29. Added LIBS=-liconv to OS/Makefile-OSF1.
4954 30. Changed the default configuration ACL to relax the local part checking rule
4955 for addresses that are not in any local domains. For these addresses,
4956 slashes and pipe symbols are allowed within local parts, but the sequence
4957 /../ is explicitly forbidden.
4959 31. SPA server authentication was not clearing the challenge buffer before
4962 32. log_message in a "warn" ACL statement was writing to the reject log as
4963 well as to the main log, which contradicts the documentation and doesn't
4964 seem right (because no rejection is happening). So I have stopped it.
4966 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
4967 However, I am unable to do any testing of this.
4969 34. Fixed an infelicity in the appendfile transport. When checking directories
4970 for a mailbox, to see if any needed to be created, it was accidentally
4971 using path names with one or more superfluous leading slashes; tracing
4972 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
4974 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
4975 contents are added to the log line that is written for every discarded
4976 recipient. (Previously a log_message setting was ignored.)
4978 36. The ${quote: operator now quotes the string if it is empty.
4980 37. The install script runs exim in order to find its version number. If for
4981 some reason other than non-existence or emptiness, which it checks, it
4982 could not run './exim', it was installing it with an empty version number,
4983 i.e. as "exim-". This error state is now caught, and the installation is
4986 38. An argument was missing from the function that creates an error message
4987 when Exim fails to connect to the socket for saslauthd authentication.
4988 This could cause Exim to crash, or give a corrupted message.
4990 39. Added isip, isip4, and isip6 to ${if conditions.
4992 40. The ACL variables $acl_xx are now saved with the message, and can be
4993 accessed later in routers, transports, and filters.
4995 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
4996 strings in the file are not string-expanded.
4998 42. If a MAIL command specified a SIZE value that was too large to fit into an
4999 int variable, the check against message_size_limit failed. Such values are
5000 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
5001 day this will have to be increased, but I don't think I want to be around
5002 when emails are that large.
5009 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
5010 iconv() is not standard in FreeBSD.
5012 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
5013 IPv6 enabled. The observed symptom was a segmentation fault on return from
5014 the function os_common_find_running_interfaces() in src/os.c.
5016 3. In the check_special_case() function in daemon.c I had used "errno" as an
5017 argument name, which causes warnings on some systems. This was basically a
5018 typo, since it was named "eno" in the comments!
5020 4. The code that waits for the clock to tick (at a resolution of some fraction
5021 of a second) so as to ensure message-id uniqueness was always waiting for
5022 at least one whole tick, when it could have waited for less. [This is
5023 almost certainly not relevant at current processor speeds, where it is
5024 unlikely to ever wait at all. But we try to future-proof.]
5026 5. The function that sleeps for a time interval that includes fractions of a
5027 second contained a race. It did not block SIGALRM between setting the
5028 timer, and suspending (a couple of lines later). If the interval was short
5029 and the sigsuspend() was delayed until after it had expired, the suspension
5030 never ended. On busy systems this could lead to processes getting stuck for
5033 6. Some uncommon configurations may cause a lookup to happen in a queue runner
5034 process, before it forks any delivery processes. The open lookup caching
5035 mechanism meant that the open file or database connection was passed into
5036 the delivery process. The problem was that delivery processes always tidy
5037 up cached lookup data. This could cause a problem for the next delivery
5038 process started by the queue runner, because the external queue runner
5039 process does not know about the closure. So the next delivery process
5040 still has data in the lookup cache. In the case of a file lookup, there was
5041 no problem because closing a file descriptor in a subprocess doesn't affect
5042 the parent. However, if the lookup was caching a connection to a database,
5043 the connection was closed, and the second delivery process was likely to
5044 see errors such as "PGSQL: query failed: server closed the connection
5045 unexpectedly". The problem has been fixed by closing all cached lookups
5046 in a queue runner before running a delivery process.
5048 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
5049 seem to have the "const" qualifier which it has on other OS. I've
5052 8. Change 4.21/2 was too strict. It is only if there are two authenticators
5053 *of the same type* (client or server) with the same public name that an
5054 error should be diagnosed.
5056 9. When Exim looked up a host name for an IP address, but failed to find the
5057 original IP address when looking up the host name (a safety check), it
5058 output the message "<ip address> does not match any IP for NULL", which was
5059 confusing, to say the least. The bug was that the host name should have
5060 appeared instead of "NULL".
5062 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
5063 user that is built into the binary, and the -C or -D options is used, root
5064 privilege is dropped before the configuration file is read. In addition,
5065 logging is switched to stderr instead of the normal log files. If the
5066 configuration then re-defines the Exim user, the unprivileged environment
5067 is probably not what is expected, so Exim logs a panic warning message (but
5070 However, if deliver_drop_privilege is set, the unprivileged state may well
5071 be exactly what is intended, so the warning has been cut out in that case,
5072 and Exim is allowed to try to write to its normal log files.
5078 1. smtp_return_error_details was not giving details for temporary sender
5079 or receiver verification errors.
5081 2. Diagnose a configuration error if two authenticators have the same public
5084 3. Exim used not to create the message log file for a message until the first
5085 delivery attempt. This could be confusing when incoming messages were held
5086 for policy or load reasons. The message log file is now created at the time
5087 the message is received, and an initial "Received" line is written to it.
5089 4. The automatically generated man page for command line options had a minor
5090 bug that caused no ill effects; however, a more serious problem was that
5091 the procedure for building the man page automatically didn't always
5092 operate. Consequently, release 4.20 contains an out-of-date version. This
5093 shouldn't happen again.
5095 5. When building Exim with embedded Perl support, the script that builds the
5096 Makefile was calling 'perl' to find its compile-time parameters, ignoring
5097 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
5099 6. The freeze_tell option was not being used for messages that were frozen on
5100 arrival, either by an ACL or by local_scan().
5102 7. Added the smtp_incomplete_transaction log selector.
5104 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
5105 was accepting AUTH without a new EHLO.
5107 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
5108 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
5111 10. Logging of TCP/IP connections (when configured) now happens in the main
5112 daemon process instead of the child process, so that the TCP/IP connection
5113 count is more accurate (but it can never be perfect).
5115 11. The use of "drop" in a nested ACL was not being handled correctly in the
5116 outer ACL. Now, if condition failure induced by the nested "drop" causes
5117 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
5118 or "require"), the connection is dropped.
5120 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
5121 that yield "discard" can now be used with an "accept" or a "discard" verb,
5122 but an error is generated for any others (because I can't see a useful way
5123 to define what should happen).
5125 13. When an ACL is read dynamically from a file (or anywhere else), the lines
5126 are now processed in the same way as lines in the Exim configuration file.
5127 In particular, continuation lines are supported.
5129 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
5131 15. Added -ti meaning -t -i.
5133 16. Check for letters, digits, hyphens, and dots in the names of dnslist
5134 domains, and warn by logging if others are found.
5136 17. At least on BSD, alignment is not guarenteed for the array of ifreq's
5137 returned from GIFCONF when Exim is trying to find the list of interfaces on
5138 a host. The code in os.c has been modified to copy each ifreq to an aligned
5139 structure in all cases.
5141 Also, in some cases, the returned ifreq's were being copied to a 'struct
5142 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
5143 means the last couple of bytes of an IPv6 address could be chopped if the
5144 ifreq contained only a normal sockaddr (14 bytes storage).
5146 18. Named domain lists were not supported in the hosts_treat_as_local option.
5147 An entry such as +xxxx was not recognized, and was treated as a literal
5150 19. Ensure that header lines added by a DATA ACL are included in the reject log
5151 if the ACL subsequently rejects the message.
5153 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
5154 MD5 (which is deprecated).
5156 21. When testing a filter file using -bf, Exim was writing a message when it
5157 took the sender from a "From " line in the message, but it was not doing so
5158 when it took $return_path from a Return-Path: header line. It now does.
5160 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
5161 with a valid header line field name (a series of printing characters
5162 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
5164 23. Changed "disc" in the source to "disk" to conform to the documentation and
5165 the book and for uniformity.
5167 24. Ignore Sendmail's -Ooption=value command line item.
5169 25. When execve() failed while trying to run a command in a pipe transport,
5170 Exim was returning EX_UNAVAILBLE (69) from the subprocess. However, this
5171 could be confused with a return value of 69 from the command itself. This
5172 has been changed to 127, the value the shell returns if it is asked to run
5173 a non-existent command. The wording for the related log line suggests a
5174 non-existent command as the problem.
5176 26. If received_header_text expands to an empty string, do not add a Received:
5177 header line to the message. (Well, it adds a token one on the spool, but
5178 marks it "old" so that it doesn't get used or transmitted.)
5180 27. Installed eximstats 1.28 (addition of -nt option).
5182 28. There was no check for failure on the call to getsockname() in the daemon
5183 code. This can fail if there is a shortage of resources on the system, with
5184 ENOMEM, for example. A temporary error is now given on failure.
5186 29. Contrary to the C standard, it seems that in some environments, the
5187 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
5188 program. Exim now does this explicitly; it affects the formatting of
5189 timestamps using strftime().
5191 30. If exiqsumm was given junk data, it threw up some uninitialized variable
5192 complaints. I've now initialized all the variables, to avoid this.
5194 32. Header lines added by a system filter were not being "seen" during
5195 transport-time rewrites.
5197 33. The info_callback() function passed to OpenSSL is set up with type void
5198 (*)(SSL *, int, int), as described somewhere. However, when calling the
5199 function (actually a macro) that sets it up, the type void(*)() is
5200 expected. I've put in a cast to prevent warnings from picky compilers.
5202 34. If a DNS black list lookup found a CNAME record, but there were no A
5203 records associated with the domain it pointed at, Exim crashed.
5205 35. If a DNS black list lookup returned more than one A record, Exim ignored
5206 all but the first. It now scans all returned addresses if a particular IP
5207 value is being sought. In this situation, the contents of the
5208 $dnslist_value variable are a list of all the addresses, separated by a
5211 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
5212 to accept a host name and all its aliases if the forward lookup for any of
5213 them yielded the IP address of the incoming connection. Now it accepts only
5214 those names whose forward lookup yields the correct IP address. Any other
5215 names are discarded. This closes a loophole whereby a rogue DNS
5216 administrator could create reverse DNS records to break through a
5217 wildcarded host restriction in an ACL.
5219 37. If a user filter or a system filter that ran in a subprocess used any of
5220 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
5221 the wrong values were passed to the pipe command ($thisaddress had the
5222 value of $0, $0 had the value of $1, etc). This bug was introduced by
5223 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
5225 38. Improved the line breaking for long SMTP error messages from ACLs.
5226 Previously, if there was no break point between 40 and 75 characters, Exim
5227 left the rest of the message alone. Two changes have been made: (a) I've
5228 reduced the minimum length to 35 characters; (b) if it can't find a break
5229 point between 35 and 75 characters, it looks ahead and uses the first one
5230 that it finds. This may give the occasional overlong line, but at least the
5231 remaining text gets split now.
5233 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
5234 file descriptors might be low, and that setting 1000 would always raise it.
5235 It turns out that in some environments, the limit is already over 1000 and
5236 that lowering it causes trouble. So now Exim takes care not to decrease it.
5238 40. When delivering a message, the value of $return_path is set to $sender_
5239 address at the start of routing (routers may change the value). By an
5240 oversight, this default was not being set up when an address was tested by
5241 -bt or -bv, which affected the outcome if any router or filter referred to
5244 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
5245 log only when "message" or "log_message" are set. However, if one of the
5246 conditions was an address verification, or a call to a nested ACL, the
5247 messages generated by the underlying test were being passed through. This
5248 no longer happens. The underlying message is available in $acl_verify_
5249 message for both "message" and "log_message" expansions, so it can be
5250 passed through if needed.
5252 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
5253 new expansion $bh_ to give the encoded byte string without charset
5254 translation. Translation happens only if iconv() is available; HAVE_ICONV
5255 indicates this at build time. HEADERS_CHARSET gives the charset to
5256 translate to; headers_charset can change it in the configuration, and
5257 "headers charset" can change it in an individual filter file.
5259 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
5260 that creates RFC 2047 encoded "words" labels them as that charset instead
5261 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
5262 expansion operator; (ii) when Exim creates a From: line for a local
5263 message; (iii) when a header line is rewritten to include a "phrase" part.
5265 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
5266 buggy, causing it to skip the first lines of messages whose message ID
5267 ended in 'D'. This would not have bitten before Exim release 4.14, because
5268 message IDs were unlikely to end in 'D' before then. The effect was to have
5269 incorrect size information for certain domains.
5271 45. #include "config.h" was missing at the start of the crypt16.c module. This
5272 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
5275 46. If there was a timeout during a "random" callout check, Exim treated it as
5276 a failure of the random address, and carried on sending RSET and the real
5277 address. If the delay was just some slowness somewhere, the response to the
5278 original RCPT would be taken as a response to RSET and so on, causing
5279 mayhem of various kinds.
5281 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
5282 when I implemented it. It didn't allow for the fact that some option values
5283 may legitimatetly be negative (e.g. size_addition), and it didn't even do
5284 the right test for positive values.
5286 48. Domain names in DNS records are case-independent. Exim always looks them up
5287 in lower case. Some resolvers return domain names in exactly the case they
5288 appear in the zone file, that is, they may contain uppercase letters. Not
5289 all resolvers do this - some return always lower case. Exim was treating a
5290 change of case by a resolver as a change of domain, similar to a widening
5291 of a domain abbreviation. This triggered its re-routing code and so it was
5292 trying to route what was effectively the same domain again. This normally
5293 caused routing to fail (because the router wouldn't handle the domain
5294 twice). Now Exim checks for this case specially, and just changes the
5295 casing of the domain that it ultimately uses when it transmits the message
5298 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
5301 50. If a filter generated a file delivery with a non-absolute name (possible if
5302 no home directory exists for the router), the forbid_file option was not
5305 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
5306 the existing equality matching.
5308 52. Exim was using ints instead of ino_t variables in some places where it was
5309 dealing with inode numbers.
5311 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
5312 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
5313 called TMPDIR, and if it finds it is different, it changes its value.
5315 54. The smtp_printf() function is now made available to local_scan() so
5316 additional output lines can be written before returning. There is also an
5317 smtp_fflush() function to enable the detection of a dropped connection.
5318 The variables smtp_input and smtp_batched_input are exported to
5321 55. Changed the default runtime configuration: the message "Unknown user"
5322 has been removed from the ACL, and instead placed on the localuser router,
5323 using the cannot_route_message feature. This means that any verification
5324 failures that generate their own messages won't get overridden. Similarly,
5325 the "Unrouteable address" message that was in the ACL for unverifiable
5326 relay addresses has also been removed.
5328 56. Added hosts_avoid_esmtp to the smtp transport.
5330 57. The exicyclog script was not checking for the esoteric option
5331 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
5332 will work only if exicyclog is run under the appropriate euid.
5334 58. Following a discussion on the list, the rules by which Exim recognises line
5335 endings on incoming messages have been changed. The -dropcr and drop_cr
5336 options are now no-ops, retained only for backwards compatibility. The
5337 following line terminators are recognized: LF CRLF CR. However, special
5338 processing applies to CR:
5340 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
5341 nor a local message in the state where . is a terminator.
5343 (ii) If a bare CR is encountered in a header line, an extra space is added
5344 after the line terminator so as not to end the header. The reasoning
5345 behind this is that bare CRs in header lines are most likely either
5346 to be mistakes, or people trying to play silly games.
5348 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
5349 was being incorrectly given as 18 bytes larger than it should have been.
5350 This is a VOB (very old bug).
5352 60. This may never have affected anything current, but just in case it has:
5353 When the local host is found other than at the start of a list of hosts,
5354 the local host, those with the same MX, and any that follow, are discarded.
5355 When the list in question was part of a longer list of hosts, the following
5356 hosts (not currently being processed) were also being discarded. This no
5357 longer happens. I'm not sure if this situation could ever has previously
5360 61. Added the "/MX" feature to lists of hosts in the manualroute and query
5363 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
5364 header. This is something that is recommended in a new Internet Draft, and
5365 is something that is documented as being done by Sendmail. There are two
5366 possible values. For messages generated by the autoreply transport, Exim
5369 Auto-Submitted: auto-replied
5371 whereas for all other generated messages (e.g. bounces) it adds
5373 Auto-Submitted: auto-generated
5375 63. The "personal" condition in filters now includes a test for the
5376 Auto-Submitted: header. If it contains the string "auto-" the message it
5377 not considered personal.
5379 64. Added rcpt_include_affixes as a generic transport option.
5381 65. Added queue_only_override (default true).
5383 66. Added the syslog_duplication option.
5385 67. If what should have been the first header line of a message consisted of
5386 a space followed by a colon, Exim was mis-interpreting it as a header line.
5387 It isn't of course - it is syntactically invalid and should therefore be
5388 treated as the start of the message body. The misbehaviour could have
5389 caused a number of strange effects, including loss of data in subsequent
5390 header lines, and spool format errors.
5392 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
5393 client host had authenticated. This control can now be exercised by an ACL
5394 for more flexibility.
5396 69. By default, callouts do not happen when testing with -bh. There is now a
5397 variant, -bhc, which does actually run the callout code, including
5398 consulting and updating the callout cache.
5400 70. Added support for saslauthd authentication, courtesy of Alexander
5403 71. If statvfs() failed on the spool or log directories while checking their
5404 size for availability, Exim confusingly gave the error "space shortage".
5405 Furthermore, in debugging mode it crashed with a floating point exception.
5406 These checks are done if check_{spool,log}_{space,inodes} are set, and when
5407 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
5408 serious problem, Exim now writes to the main and panic logs when this
5409 happens, with details of the failure. It then refuses to accept the
5410 incoming message, giving the message "spool directory problem" or "log
5411 directory problem" with a 421 code for SMTP messages.
5413 72. When Exim is about to re-exec itself, it ensures that the file descriptors
5414 0, 1, and 2 exist, because some OS complain for execs without them (see
5415 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
5416 descriptors. However, the code omitted to check that the open succeeded,
5417 causing mysterious errors if for some reason the permissions on /dev/null
5418 got screwed. Now Exim writes a message to the main and panic logs, and
5419 bombs out if it can't open /dev/null.
5421 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
5422 interact so that it is all more flexible. It is supposed to remain
5423 backwards compatible. Also added extra_local_interfaces.
5425 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
5426 to bomb out with an assertion failure - to the client this appears as a
5427 connection drop. This problem occurs in the part of the code that was taken
5428 from the Samba project. Fortunately, the assertion is in a very simple
5429 function, so I have fixed this by reproducing the function inline in the
5430 one place where it is called, and arranging for authentication to fail
5431 instead of killing the process with assert().
5433 75. The SPA client code was not working when the server requested OEM rather
5434 than Unicode encoding.
5436 76. Added code to make require_files with a specific uid setting more usable in
5437 the case where statting the file as root fails - usually a non-root-mounted
5438 NFS file system. When this happens and the failure is EACCES, Exim now
5439 forks a subprocess and does the per-uid checking as the relevant uid.
5441 77. Added process_log_path.
5443 78. If log_file_path was not explicitly set, a setting of check_log_space or
5444 check_log_inodes was ignored.
5446 79. If a space check for the spool or log partitions fails, the incident is now
5447 logged. Of course, in the latter case the data may get lost...
5449 80. Added the %p formatting code to string_format() so that it can be used to
5450 print addresses in debug_print(). Adjusted all the address printing in the
5451 debugging in store.c to use %p rather than %d.
5453 81. There was a concern that a line of code in smtp_in.c could overflow a
5454 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
5455 initially expressed, the concern was not well-founded, because trailing
5456 spaces are removed early. However, if the trailing spaces were followed by
5457 a NULL, they did not get removed, so the overflow was possible. Two fixes
5460 (a) I re-wrote the offending code in a cleaner fashion.
5461 (b) If an incoming SMTP command contains a NULL character, it is rejected
5464 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
5465 runs initgroups(), so that if the Exim user is in any additional groups,
5466 they will be used during message reception.
5472 The change log for 4.20 and earlier releases has been archived.
git://git.exim.org / users / heiko / exim.git / blob