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 GnuTLS: support TLS 1.2 & 1.1.
117 Use gnutls_certificate_verify_peers2() [patch from Andreas Metzler].
124 PP/01 The new ldap_require_cert option would segfault if used. Fixed.
126 PP/02 Harmonised TLS library version reporting; only show if debugging.
127 Layout now matches that introduced for other libraries in 4.74 PP/03.
129 PP/03 New openssl_options items: no_sslv2 no_sslv3 no_ticket no_tlsv1
131 PP/04 New "dns_use_edns0" global option.
133 PP/05 Don't segfault on misconfiguration of ref:name exim-user as uid.
136 PP/06 Extra paranoia around buffer usage at the STARTTLS transition.
137 nb: Exim is not vulnerable to http://www.kb.cert.org/vuls/id/555316
139 TK/01 Updated PolarSSL code to 0.14.2.
140 Bugzilla 1097. Patch from Andreas Metzler.
142 PP/07 Catch divide-by-zero in ${eval:...}.
145 PP/08 Condition negation of bool{}/bool_lax{} did not negate. Fixed.
148 TK/02 Bugzilla 1106: CVE-2011-1764 - DKIM log line was subject to a
149 format-string attack -- SECURITY: remote arbitrary code execution.
151 TK/03 SECURITY - DKIM signature header parsing was double-expanded, second
152 time unintentionally subject to list matching rules, letting the header
153 cause arbitrary Exim lookups (of items which can occur in lists, *not*
154 arbitrary string expansion). This allowed for information disclosure.
156 PP/09 Fix another SIGFPE (x86) in ${eval:...} expansion, this time related to
157 INT_MIN/-1 -- value coerced to INT_MAX.
163 NM/01 Workround for PCRE version dependency in version reporting
166 TF/01 Update valgrind.h and memcheck.h to copies from valgrind-3.6.0.
167 This fixes portability to compilers other than gcc, notably
168 Solaris CC and HP-UX CC. Fixes Bugzilla 1050.
170 TF/02 Bugzilla 139: Avoid using the += operator in the modular lookup
171 makefiles for portability to HP-UX and POSIX correctness.
173 PP/01 Permit LOOKUP_foo enabling on the make command-line.
174 Also via indented variable definition in the Makefile.
175 (Debugging by Oliver Heesakkers).
177 PP/02 Restore caching of spamd results with expanded spamd_address.
178 Patch from author of expandable spamd_address patch, Wolfgang Breyha.
180 PP/03 Build issue: lookups-Makefile now exports LC_ALL=C
181 Improves build reliability. Fix from: Frank Elsner
183 NM/02 Fix wide character breakage in the rfc2047 coding
184 Fixes bug 1064. Patch from Andrey N. Oktyabrski
186 NM/03 Allow underscore in dnslist lookups
187 Fixes bug 1026. Patch from Graeme Fowler
189 PP/04 Bugzilla 230: Support TLS-enabled LDAP (in addition to ldaps).
190 Code patches from Adam Ciarcinski of NetBSD.
192 NM/04 Fixed exiqgrep to cope with mailq missing size issue
195 PP/05 Bugzilla 1083: when lookup expansion defers, escape the output which
196 is logged, to avoid truncation. Patch from John Horne.
198 PP/06 Bugzilla 1042: implement freeze_signal on pipe transports.
199 Patch from Jakob Hirsch.
201 PP/07 Bugzilla 1061: restrict error messages sent over SMTP to not reveal
202 SQL string expansion failure details.
203 Patch from Andrey Oktyabrski.
205 PP/08 Bugzilla 486: implement %M datestamping in log filenames.
206 Patch from Simon Arlott.
208 PP/09 New lookups functionality failed to compile on old gcc which rejects
209 extern declarations in function scope.
210 Patch from Oliver Fleischmann
212 PP/10 Use sig_atomic_t for flags set from signal handlers.
213 Check getgroups() return and improve debugging.
214 Fixed developed for diagnosis in bug 927 (which turned out to be
217 PP/11 Bugzilla 1055: Update $message_linecount for maildir_tag.
218 Patch from Mark Zealey.
220 PP/12 Bugzilla 1056: Improved spamd server selection.
221 Patch from Mark Zealey.
223 PP/13 Bugzilla 1086: Deal with maildir quota file races.
224 Based on patch from Heiko Schlittermann.
226 PP/14 Bugzilla 1019: DKIM multiple signature generation fix.
227 Patch from Uwe Doering, sign-off by Michael Haardt.
229 NM/05 Fix to spam.c to accommodate older gcc versions which dislike
230 variable declaration deep within a block. Bug and patch from
233 PP/15 lookups-Makefile IRIX compatibilty coercion.
235 PP/16 Make DISABLE_DKIM build knob functional.
237 NM/06 Bugzilla 968: child_open_uid: restore default SIGPIPE handler
238 Patch by Simon Arlott
240 TF/03 Fix valgrind.h portability to C89 compilers that do not support
241 variable argument macros. Our copy now differs from upstream.
247 TF/01 Failure to get a lock on a hints database can have serious
248 consequences so log it to the panic log.
250 TF/02 Log LMTP confirmation messages in the same way as SMTP,
251 controlled using the smtp_confirmation log selector.
253 TF/03 Include the error message when we fail to unlink a spool file.
255 DW/01 Bugzilla 139: Support dynamically loaded lookups as modules.
256 With thanks to Steve Haslam, Johannes Berg & Serge Demonchaux
257 for maintaining out-of-tree patches for some time.
259 PP/01 Bugzilla 139: Documentation and portability issues.
260 Avoid GNU Makefile-isms, let Exim continue to build on BSD.
261 Handle per-OS dynamic-module compilation flags.
263 PP/02 Let /dev/null have normal permissions.
264 The 4.73 fixes were a little too stringent and complained about the
265 permissions on /dev/null. Exempt it from some checks.
266 Reported by Andreas M. Kirchwitz.
268 PP/03 Report version information for many libraries, including
269 Exim version information for dynamically loaded libraries. Created
270 version.h, now support a version extension string for distributors
271 who patch heavily. Dynamic module ABI change.
273 PP/04 CVE-2011-0017 - check return value of setuid/setgid. This is a
274 privilege escalation vulnerability whereby the Exim run-time user
275 can cause root to append content of the attacker's choosing to
278 PP/05 Bugzilla 1041: merged DCC maintainer's fixes for return code.
281 PP/06 Bugzilla 1071: fix delivery logging with untrusted macros.
282 If dropping privileges for untrusted macros, we disabled normal logging
283 on the basis that it would fail; for the Exim run-time user, this is not
284 the case, and it resulted in successful deliveries going unlogged.
285 Fixed. Reported by Andreas Metzler.
291 PP/01 Date: & Message-Id: revert to normally being appended to a message,
292 only prepend for the Resent-* case. Fixes regression introduced in
293 Exim 4.70 by NM/22 for Bugzilla 607.
295 PP/02 Include check_rfc2047_length in configure.default because we're seeing
296 increasing numbers of administrators be bitten by this.
298 JJ/01 Added DISABLE_DKIM and comment to src/EDITME
300 PP/03 Bugzilla 994: added openssl_options main configuration option.
302 PP/04 Bugzilla 995: provide better SSL diagnostics on failed reads.
304 PP/05 Bugzilla 834: provide a permit_coredump option for pipe transports.
306 PP/06 Adjust NTLM authentication to handle SASL Initial Response.
308 PP/07 If TLS negotiated an anonymous cipher, we could end up with SSL but
309 without a peer certificate, leading to a segfault because of an
310 assumption that peers always have certificates. Be a little more
311 paranoid. Problem reported by Martin Tscholak.
313 PP/08 Bugzilla 926: switch ClamAV to use the new zINSTREAM API for content
314 filtering; old API available if built with WITH_OLD_CLAMAV_STREAM=yes
315 NB: ClamAV planning to remove STREAM in "middle of 2010".
316 CL also introduces -bmalware, various -d+acl logging additions and
317 more caution in buffer sizes.
319 PP/09 Implemented reverse_ip expansion operator.
321 PP/10 Bugzilla 937: provide a "debug" ACL control.
323 PP/11 Bugzilla 922: Documentation dusting, patch provided by John Horne.
325 PP/12 Bugzilla 973: Implement --version.
327 PP/13 Bugzilla 752: Refuse to build/run if Exim user is root/0.
329 PP/14 Build without WITH_CONTENT_SCAN. Path from Andreas Metzler.
331 PP/15 Bugzilla 816: support multiple condition rules on Routers.
333 PP/16 Add bool_lax{} expansion operator and use that for combining multiple
334 condition rules, instead of bool{}. Make both bool{} and bool_lax{}
335 ignore trailing whitespace.
337 JJ/02 prevent non-panic DKIM error from being sent to paniclog
339 JJ/03 added tcp_wrappers_daemon_name to allow host entries other than
342 PP/17 Fix malware regression for cmdline scanner introduced in PP/08.
343 Notification from Dr Andrew Aitchison.
345 PP/18 Change ClamAV response parsing to be more robust and to handle ClamAV's
346 ExtendedDetectionInfo response format.
347 Notification from John Horne.
349 PP/19 OpenSSL 1.0.0a compatibility const-ness change, should be backwards
352 PP/20 Added a CONTRIBUTING file. Fixed the documentation build to use http:
353 XSL and documented dependency on system catalogs, with examples of how
356 DW/21 Added Valgrind hooks in store.c to help it capture out-of-bounds store
359 DW/22 Bugzilla 1044: CVE-2010-4345 - partial fix: restrict default behaviour
360 of CONFIGURE_OWNER and CONFIGURE_GROUP options to no longer allow a
361 configuration file which is writeable by the Exim user or group.
363 DW/23 Bugzilla 1044: CVE-2010-4345 - part two: extend checks for writeability
364 of configuration files to cover files specified with the -C option if
365 they are going to be used with root privileges, not just the default
368 DW/24 Bugzilla 1044: CVE-2010-4345 - part three: remove ALT_CONFIG_ROOT_ONLY
369 option (effectively making it always true).
371 DW/25 Add TRUSTED_CONFIG_PREFIX_FILE option to allow alternative configuration
372 files to be used while preserving root privileges.
374 DW/26 Set FD_CLOEXEC on SMTP sockets after forking in the daemon, to ensure
375 that rogue child processes cannot use them.
377 PP/27 Bugzilla 1047: change the default for system_filter_user to be the Exim
378 run-time user, instead of root.
380 PP/28 Add WHITELIST_D_MACROS option to let some macros be overridden by the
381 Exim run-time user without dropping privileges.
383 DW/29 Remove use of va_copy() which breaks pre-C99 systems. Duplicate the
384 result string, instead of calling string_vformat() twice with the same
387 DW/30 Allow TRUSTED_CONFIG_PREFIX_FILE only for Exim or CONFIGURE_OWNER, not
388 for other users. Others should always drop root privileges if they use
389 -C on the command line, even for a whitelisted configure file.
391 DW/31 Turn TRUSTED_CONFIG_PREFIX_FILE into TRUSTED_CONFIG_FILE. No prefixes.
393 NM/01 Fixed bug #1002 - Message loss when using multiple deliveries
399 JJ/01 installed exipick 20100104.1, adding $max_received_linelength,
400 $data_path, and $header_path variables; fixed documentation bugs and
403 JJ/02 installed exipick 20100222.0, added --input-dir and --finput to allow
404 exipick to access non-standard spools, including the "frozen" queue
407 NM/01 Bugzilla 965: Support mysql stored procedures.
408 Patch from Alain Williams
410 NM/02 Bugzilla 961: Spacing fix (syntax error) on Makefile directives for NetBSD
412 NM/03 Bugzilla 955: Documentation fix for max_rcpts.
413 Patch from Andreas Metzler
415 NM/04 Bugzilla 954: Fix for unknown responses from Dovecot authenticator.
416 Patch from Kirill Miazine
418 NM/05 Bugzilla 671: Added umask to procmail example.
420 JJ/03 installed exipick 20100323.0, fixing doc bug
422 NM/06 Bugzilla 988: CVE-2010-2023 - prevent hardlink attack on sticky mail
423 directory. Notification and patch from Dan Rosenberg.
425 TK/01 PDKIM: Upgrade PolarSSL files to upstream version 0.12.1.
427 TK/02 Improve log output when DKIM signing operation fails.
429 MH/01 Treat the transport option dkim_domain as a colon separated
430 list, not as a single string, and sign the message with each element,
431 omitting multiple occurences of the same signer.
433 NM/07 Null terminate DKIM strings, Null initialise DKIM variable
434 Bugzilla 985, 986. Patch by Simon Arlott
436 NM/08 Bugzilla 967. dnsdb DNS TXT record bug fix (DKIM-related)
437 Patch by Simon Arlott
439 PP/01 Bugzilla 989: CVE-2010-2024 - work round race condition on
440 MBX locking. Notification from Dan Rosenberg.
446 TK/01 Bugzilla 912: Fix DKIM segfault on empty headers/body.
448 NM/01 Bugzilla 913: Documentation fix for gnutls_* options.
450 NM/02 Bugzilla 722: Documentation for randint. Better randomness defaults.
452 NM/03 Bugzilla 847: Enable DNSDB lookup by default.
454 NM/04 Bugzilla 915: Flag broken perl installation during build.
460 TK/01 Added patch by Johannes Berg that expands the main option
461 "spamd_address" if it starts with a dollar sign.
463 TK/02 Write list of recipients to X-Envelope-Sender header when building
464 the mbox-format spool file for content scanning (suggested by Jakob
467 TK/03 Added patch by Wolfgang Breyha that adds experimental DCC
468 (http://www.dcc-servers.net/) support via dccifd. Activated by
469 setting EXPERIMENTAL_DCC=yes in Local/Makefile.
471 TK/04 Bugzilla 673: Add f-protd malware scanner support. Patch submitted
472 by Mark Daniel Reidel <mr@df.eu>.
474 NM/01 Bugzilla 657: Embedded PCRE removed from the exim source tree.
475 When building exim an external PCRE library is now needed -
476 PCRE is a system library on the majority of modern systems.
477 See entry on PCRE_LIBS in EDITME file.
479 NM/02 Bugzilla 646: Removed unwanted C/R in Dovecot authenticator
480 conversation. Added nologin parameter to request.
481 Patch contributed by Kirill Miazine.
483 TF/01 Do not log submission mode rewrites if they do not change the address.
485 TF/02 Bugzilla 662: Fix stack corruption before exec() in daemon.c.
487 NM/03 Bugzilla 602: exicyclog now handles panic log, and creates empty
488 log files in place. Contributed by Roberto Lima.
490 NM/04 Bugzilla 667: Close socket used by dovecot authenticator.
492 TF/03 Bugzilla 615: When checking the local_parts router precondition
493 after a local_part_suffix or local_part_prefix option, Exim now
494 does not use the address's named list lookup cache, since this
495 contains cached lookups for the whole local part.
497 NM/05 Bugzilla 521: Integrated SPF Best Guess support contributed by
498 Robert Millan. Documentation is in experimental-spec.txt.
500 TF/04 Bugzilla 668: Fix parallel build (make -j).
502 NM/05.2 Bugzilla 437: Prevent Maildir aux files being created with mode 000.
504 NM/05.3 Bugzilla 598: Improvement to Dovecot authenticator handling.
505 Patch provided by Jan Srzednicki.
507 TF/05 Leading white space used to be stripped from $spam_report which
508 wrecked the formatting. Now it is preserved.
510 TF/06 Save $spam_score, $spam_bar, and $spam_report in spool files, so
511 that they are available at delivery time.
513 TF/07 Fix the way ${extract is skipped in the untaken branch of a conditional.
515 TF/08 TLS error reporting now respects the incoming_interface and
516 incoming_port log selectors.
518 TF/09 Produce a more useful error message if an SMTP transport's hosts
519 setting expands to an empty string.
521 NM/06 Bugzilla 744: EXPN did not work under TLS.
522 Patch provided by Phil Pennock.
524 NM/07 Bugzilla 769: Extraneous comma in usage fprintf
525 Patch provided by Richard Godbee.
527 NM/08 Fixed erroneous documentation references to smtp_notquit_acl to be
528 acl_smtp_notquit, added index entry.
530 NM/09 Bugzilla 787: Potential buffer overflow in string_format.
531 Patch provided by Eugene Bujak.
533 NM/10 Bugzilla 770: Problem on some platforms modifying the len parameter to
534 accept(). Patch provided by Maxim Dounin.
536 NM/11 Bugzilla 749: Preserve old behaviour of blanks comparing equal to zero.
537 Patch provided by Phil Pennock.
539 NM/12 Bugzilla 497: Correct behaviour of exiwhat when no config exists.
541 NM/13 Bugzilla 590: Correct handling of Resent-Date headers.
542 Patch provided by Brad "anomie" Jorsch.
544 NM/14 Bugzilla 622: Added timeout setting to transport filter.
545 Patch provided by Dean Brooks.
547 TK/05 Add native DKIM support (does not depend on external libraries).
549 NM/15 Bugzilla 854: Removed code that symlinks to pcre as its no longer useful.
550 Patch provided by Graeme Fowler.
552 NM/16 Bugzilla 851: Documentation example syntax fix.
554 NM/17 Changed NOTICE file to remove references to embedded PCRE.
556 NM/18 Bugzilla 894: Fix issue with very long lines including comments in
559 NM/19 Bugzilla 745: TLS version reporting.
560 Patch provided by Phil Pennock.
562 NM/20 Bugzilla 167: bool: condition support.
563 Patch provided by Phil Pennock.
565 NM/21 Bugzilla 665: gnutls_compat_mode to allow compatibility with broken
566 clients. Patch provided by Phil Pennock.
568 NM/22 Bugzilla 607: prepend (not append) Resent-Message-ID and Resent-Date.
569 Patch provided by Brad "anomie" Jorsch.
571 NM/23 Bugzilla 687: Fix misparses in eximstats.
572 Patch provided by Heiko Schlittermann.
574 NM/24 Bugzilla 688: Fix exiwhat to handle log_selector = +pid.
575 Patch provided by Heiko Schlittermann.
577 NM/25 Bugzilla 727: Use transport mode as default mode for maildirsize file.
578 plus update to original patch.
580 NM/26 Bugzilla 799: Documentation correction for ratelimit.
582 NM/27 Bugzilla 802: Improvements to local interface IP addr detection.
583 Patch provided by David Brownlee.
585 NM/28 Bugzilla 807: Improvements to LMTP delivery logging.
587 NM/29 Bugzilla 862, 866, 875: Documentation bugfixes.
589 NM/30 Bugzilla 888: TLS documentation bugfixes.
591 NM/31 Bugzilla 896: Dovecot buffer overrun fix.
593 NM/32 Bugzilla 889: Change all instances of "expr" in shell scripts to "expr --"
594 Unlike the original bugzilla I have changed all shell scripts in src tree.
596 NM/33 Bugzilla 898: Transport filter timeout fix.
597 Patch by Todd Rinaldo.
599 NM/34 Bugzilla 901: Fix sign/unsigned and UTF mistmatches.
600 Patch by Serge Demonchaux.
602 NM/35 Bugzilla 39: Base64 decode bug fixes.
603 Patch by Jakob Hirsch.
605 NM/36 Bugzilla 909: Correct connect() call in dcc code.
607 NM/37 Bugzilla 910: Correct issue with relaxed/simple handling.
609 NM/38 Bugzilla 908: Removed NetBSD3 support as no longer needed.
611 NM/39 Bugzilla 911: Fixed MakeLinks build script.
617 TK/01 Add preliminary DKIM support. Currently requires a forked version of
618 ALT-N's libdkim that I have put here:
619 http://duncanthrax.net/exim-experimental/
621 Note to Michael Haardt: I had to rename some vars in sieve.c. They
622 were called 'true' and it seems that C99 defines that as a reserved
623 keyword to be used with 'bool' variable types. That means you could
624 not include C99-style headers which use bools without triggering
625 build errors in sieve.c.
627 NM/01 Bugzilla 592: --help option is handled incorrectly if exim is invoked
628 as mailq or other aliases. Changed the --help handling significantly
629 to do whats expected. exim_usage() emits usage/help information.
631 SC/01 Added the -bylocaldomain option to eximstats.
633 NM/02 Bugzilla 619: Defended against bad data coming back from gethostbyaddr.
635 NM/03 Bugzilla 613: Documentation fix for acl_not_smtp.
637 NM/04 Bugzilla 628: PCRE update to 7.4 (work done by John Hall).
643 PH/01 Another patch from the Sieve maintainer.
645 PH/02 When an IPv6 address is converted to a string for single-key lookup
646 in an address list (e.g. for an item such as "net24-dbm;/net/works"),
647 dots are used instead of colons so that keys in lsearch files need not
648 contain colons. This was done some time before quoting was made available
649 in lsearch files. However, iplsearch files do require colons in IPv6 keys
650 (notated using the quote facility) so as to distinguish them from IPv4
651 keys. This meant that lookups for IP addresses in host lists did not work
652 for iplsearch lookups.
654 This has been fixed by arranging for IPv6 addresses to be expressed with
655 colons if the lookup type is iplsearch. This is not incompatible, because
656 previously such lookups could never work.
658 The situation is now rather anomolous, since one *can* have colons in
659 ordinary lsearch keys. However, making the change in all cases is
660 incompatible and would probably break a number of configurations.
662 TK/01 Change PRVS address formatting scheme to reflect latests BATV draft
665 MH/01 The "spam" ACL condition code contained a sscanf() call with a %s
666 conversion specification without a maximum field width, thereby enabling
667 a rogue spamd server to cause a buffer overflow. While nobody in their
668 right mind would setup Exim to query an untrusted spamd server, an
669 attacker that gains access to a server running spamd could potentially
670 exploit this vulnerability to run arbitrary code as the Exim user.
672 TK/02 Bugzilla 502: Apply patch to make the SPF-Received: header use
673 $primary_hostname instead of what libspf2 thinks the hosts name is.
675 MH/02 The dsearch lookup now uses lstat(2) instead of stat(2) to look for
676 a directory entry by the name of the lookup key. Previously, if a
677 symlink pointed to a non-existing file or a file in a directory that
678 Exim lacked permissions to read, a lookup for a key matching that
679 symlink would fail. Now it is enough that a matching directory entry
680 exists, symlink or not. (Bugzilla 503.)
682 PH/03 The body_linecount and body_zerocount variables are now exported in the
685 PH/04 Added the $dnslist_matched variable.
687 PH/05 Unset $tls_cipher and $tls_peerdn before making a connection as a client.
688 This means they are set thereafter only if the connection becomes
691 PH/06 Added the client_condition to authenticators so that some can be skipped
692 by clients under certain conditions.
694 PH/07 The error message for a badly-placed control=no_multiline_responses left
695 "_responses" off the end of the name.
697 PH/08 Added -Mvc to output a copy of a message in RFC 2822 format.
699 PH/09 Tidied the code for creating ratelimiting keys, creating them explicitly
700 (without spaces) instead of just copying the configuration text.
702 PH/10 Added the /noupdate option to the ratelimit ACL condition.
704 PH/11 Added $max_received_linelength.
706 PH/12 Added +ignore_defer and +include_defer to host lists.
708 PH/13 Installed PCRE version 7.2. This needed some changes because of the new
709 way in which PCRE > 7.0 is built.
711 PH/14 Implemented queue_only_load_latch.
713 PH/15 Removed an incorrect (int) cast when reading the value of SIZE in a
714 MAIL command. The effect was to mangle the value on 64-bit systems.
716 PH/16 Another patch from the Sieve maintainer.
718 PH/17 Added the NOTQUIT ACL, based on a patch from Ted Cooper.
720 PH/18 If a system quota error occurred while trying to create the file for
721 a maildir delivery, the message "Mailbox is full" was not appended to the
722 bounce if the delivery eventually timed out. Change 4.67/27 below applied
723 only to a quota excession during the actual writing of the file.
725 PH/19 It seems that peer DN values may contain newlines (and other non-printing
726 characters?) which causes problems in log lines. The DN values are now
727 passed through string_printing() before being added to log lines.
729 PH/20 Added the "servers=" facility to MySQL and PostgreSQL lookups. (Oracle
730 and InterBase are left for another time.)
732 PH/21 Added message_body_newlines option.
734 PH/22 Guard against possible overflow in moan_check_errorcopy().
736 PH/23 POSIX allows open() to be a macro; guard against that.
738 PH/24 If the recipient of an error message contained an @ in the local part
739 (suitably quoted, of course), incorrect values were put in $domain and
740 $local_part during the evaluation of errors_copy.
746 MH/01 Fix for bug #448, segfault in Dovecot authenticator when interface_address
747 is unset (happens when testing with -bh and -oMi isn't used). Thanks to
750 PH/01 Added a new log selector smtp_no_mail, to log SMTP sessions that do not
751 issue a MAIL command.
753 PH/02 In an ACL statement such as
755 deny dnslists = X!=127.0.0.2 : X=127.0.0.2
757 if a client was not listed at all, or was listed with a value other than
758 127.0.0.2, in the X list, but was listed with 127.0.0.2 in the Y list,
759 the condition was not true (as it should be), so access was not denied.
760 The bug was that the ! inversion was incorrectly passed on to the second
761 item. This has been fixed.
763 PH/03 Added additional dnslists conditions == and =& which are different from
764 = and & when the dns lookup returns more than one IP address.
766 PH/04 Added gnutls_require_{kx,mac,protocols} to give more control over the
767 cipher suites used by GnuTLS. These options are ignored by OpenSSL.
769 PH/05 After discussion on the list, added a compile time option ENABLE_DISABLE_
770 FSYNC, which compiles an option called disable_fsync that allows for
771 bypassing fsync(). The documentation is heavily laced with warnings.
773 SC/01 Updated eximstats to collate all SpamAssassin rejects into one bucket.
775 PH/06 Some tidies to the infrastructure of the Test Suite that is concerned
776 with the auxiliary C programs that it uses: (1) Arrange for BIND_8_COMPAT
777 to be defined when compiling on OSX (Darwin); (2) Tidies to the Makefile,
778 including adding "make clean"; (3) Added -fPIC when compiling the test
779 dynamically loaded module, to get rid of a warning.
781 MH/02 Fix for bug #451, causing paniclog entries to be written if a bounce
782 message fails, move_frozen_messages = true and ignore_bounce_errors_after
783 = 0s. The bug is otherwise harmless.
785 PH/07 There was a bug in the dovecot authenticator such that the value of
786 $auth1 could be overwritten, and so not correctly preserved, after a
787 successful authentication. This usually meant that the value preserved by
788 the server_setid option was incorrect.
790 PH/08 Added $smtp_count_at_connection_start, deliberately with a long name.
792 PH/09 Installed PCRE release 7.0.
794 PH/10 The acl_not_smtp_start ACL was, contrary to the documentation, not being
795 run for batched SMTP input. It is now run at the start of every message
796 in the batch. While fixing this I discovered that the process information
797 (output by running exiwhat) was not always getting set for -bs and -bS
798 input. This is fixed, and it now also says "batched" for BSMTP.
800 PH/11 Added control=no_pipelining.
802 PH/12 Added $sending_ip_address and $sending_port (mostly Magnus Holmgren's
803 patch, slightly modified), and move the expansion of helo_data till after
804 the connection is made in the smtp transport (so it can use these
807 PH/13 Added ${rfc2047d: to decoded RFC 2047 strings.
809 PH/14 Added log_selector = +pid.
811 PH/15 Flush SMTP output before delaying, unless control=no_delay_flush is set.
813 PH/16 Add ${if forany and ${if forall.
815 PH/17 Added dsn_from option to vary the From: line in DSNs.
817 PH/18 Flush SMTP output before performing a callout, unless control =
818 no_callout_flush is set.
820 PH/19 Change 4.64/PH/36 introduced a bug: when address_retry_include_sender
821 was true (the default) a successful delivery failed to delete the retry
822 item, thus causing premature timeout of the address. The bug is now
825 PH/20 Added hosts_avoid_pipelining to the smtp transport.
827 PH/21 Long custom messages for fakedefer and fakereject are now split up
828 into multiline reponses in the same way that messages for "deny" and
829 other ACL rejections are.
831 PH/22 Applied Jori Hamalainen's speed-up changes and typo fixes to exigrep,
832 with slight modification.
834 PH/23 Applied sieve patches from the maintainer "tracking the latest notify
835 draft, changing the syntax and factoring some duplicate code".
837 PH/24 When the log selector "outgoing_port" was set, the port was shown as -1
838 for deliveries of the second and subsequent messages over the same SMTP
841 PH/25 Applied Magnus Holmgren's patch for ${addresses, ${map, ${filter, and
842 ${reduce, with only minor "tidies".
844 SC/02 Applied Daniel Tiefnig's patch to improve the '($parent) =' pattern match.
846 PH/26 Added a "continue" ACL modifier that does nothing, for the benefit of its
847 expansion side effects.
849 PH/27 When a message times out after an over-quota error from an Exim-imposed
850 quota, the bounce message says "mailbox is full". This message was not
851 being given when it was a system quota that was exceeded. It now should
854 MH/03 Made $recipients available in local_scan(). local_scan() already has
855 better access to the recipient list through recipients_list[], but
856 $recipients can be useful in postmaster-provided expansion strings.
858 PH/28 The $smtp_command and $smtp_command_argument variables were not correct
859 in the case of a MAIL command with additional options following the
860 address, for example: MAIL FROM:<foo@bar> SIZE=1234. The option settings
861 were accidentally chopped off.
863 PH/29 SMTP synchronization checks are implemented when a command is read -
864 there is a check that no more input is waiting when there shouldn't be
865 any. However, for some commands, a delay in an ACL can mean that it is
866 some time before the response is written. In this time, more input might
867 arrive, invalidly. So now there are extra checks after an ACL has run for
868 HELO/EHLO and after the predata ACL, and likewise for MAIL and RCPT when
869 pipelining has not been advertised.
871 PH/30 MH's patch to allow iscntrl() characters to be list separators.
873 PH/31 Unlike :fail:, a custom message specified with :defer: was not being
874 returned in the SMTP response when smtp_return_error_details was false.
877 PH/32 Change the Dovecot authenticator to use read() and write() on the socket
878 instead of the C I/O that was originally supplied, because problems were
881 PH/33 Compile failed with OpenSSL 0.9.8e. This was due to a coding error in
882 Exim which did not show up earlier: it was assuming that a call to
883 SSL_CTX_set_info_callback() might give an error value. In fact, there is
884 no error. In previous releases of OpenSSL, SSL_CTX_set_info_callback()
885 was a macro that became an assignment, so it seemed to work. This has
886 changed to a proper function call with a void return, hence the compile
887 error. Exim's code has been fixed.
889 PH/34 Change HDA_SIZE in oracle.c from 256 to 512. This is needed for 64-bit
892 PH/35 Applied a patch from the Sieve maintainer which fixes a bug in "notify".
894 PH/36 Applied John Jetmore's patch to add -v functionality to exigrep.
896 PH/37 If a message is not accepted after it has had an id assigned (e.g.
897 because it turns out to be too big or there is a timeout) there is no
898 "Completed" line in the log. When some messages of this type were
899 selected by exigrep, they were listed as "not completed". Others were
900 picked up by some special patterns. I have improved the selection
901 criteria to be more general.
903 PH/38 The host_find_failed option in the manualroute router can now be set
904 to "ignore", to completely ignore a host whose IP address cannot be
905 found. If all hosts are ignored, the behaviour is controlled by the new
906 host_all_ignored option.
908 PH/39 In a list of hosts for manualroute, if one item (either because of multi-
909 homing or because of multiple MX records with /mx) generated more than
910 one IP address, and the following item turned out to be the local host,
911 all the secondary addresses of the first item were incorrectly removed
912 from the list, along with the local host and any following hosts (which
913 is what is supposed to happen).
915 PH/40 When Exim receives a message, it writes the login name, uid, and gid of
916 whoever called Exim into the -H file. In the case of the daemon it was
917 behaving confusingly. When first started, it used values for whoever
918 started the daemon, but after a SIGHUP it used the Exim user (because it
919 calls itself on a restart). I have changed the code so that it now always
922 PH/41 (Following a suggestion from Tony Finch) If all the RCPT commands in a
923 message are rejected with the same error (e.g. no authentication or bad
924 sender address), and a DATA command is nevertheless sent (as can happen
925 with PIPELINING or a stupid MUA), the error message that was given to the
926 RCPT commands is included in the rejection of the DATA command. This is
927 intended to be helpful for MUAs that show only the final error to their
930 PH/42 Another patch from the Sieve maintainer.
932 SC/02 Eximstats - Differentiate between permanent and temporary rejects.
933 Eximstats - Fixed some broken HTML links and added missing column headers
935 Eximstats - Fixed Grand Total Summary Domains, Edomains, and Email
936 columns for Rejects, Temp Rejects, Ham, and Spam rows.
938 SC/03 Eximstats - V1.58 Fix to get <> and blackhole to show in edomain tables.
940 PH/43 Yet another patch from the Sieve maintainer.
942 PH/44 I found a way to check for a TCP/IP connection going away before sending
943 the response to the final '.' that terminates a message, but only in the
944 case where the client has not sent further data following the '.'
945 (unfortunately, this is allowed). However, in many cases there won't be
946 any further data because there won't be any more messages to send. A call
947 to select() can be used: if it shows that the input is "ready", there is
948 either input waiting, or the socket has been closed. An attempt to read
949 the next input character can distinguish the two cases. Previously, Exim
950 would have sent an OK response which the client would never have see.
951 This could lead to message repetition. This fix should cure that, at
952 least in a lot of common cases.
954 PH/45 Do not advertise STARTTLS in response to HELP unless it would be
955 advertised in response to EHLO.
961 PH/01 Two more bugs that were introduced by 4.64/PH/07, in addition to the one
962 fixed by 4.65/MH/01 (is this a record?) are fixed:
964 (i) An empty string was always treated as zero by the numeric comparison
965 operators. This behaviour has been restored.
967 (ii) It is documented that the numeric comparison operators always treat
968 their arguments as decimal numbers. This was broken in that numbers
969 starting with 0 were being interpreted as octal.
971 While fixing these problems I realized that there was another issue that
972 hadn't been noticed. Values of message_size_limit (both the global option
973 and the transport option) were treated as octal if they started with 0.
974 The documentation was vague. These values are now always treated as
975 decimal, and I will make that clear in the documentation.
981 TK/01 Disable default definition of HAVE_LINUX_SENDFILE. Clashes with
982 Linux large file support (_FILE_OFFSET_BITS=64) on older glibc
985 MH/01 Don't check that the operands of numeric comparison operators are
986 integers when their expansion is in "skipping" mode (fixes bug
987 introduced by 4.64-PH/07).
989 PH/01 If a system filter or a router generates more than SHRT_MAX (32767)
990 child addresses, Exim now panics and dies. Previously, because the count
991 is held in a short int, deliveries were likely to be lost. As such a
992 large number of recipients for a single message is ridiculous
993 (performance will be very, very poor), I have chosen to impose a limit
994 rather than extend the field.
1000 TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
1001 leftover -K file (the existence of which was triggered by #402).
1002 While we were at it, introduced process PID as part of the -K
1003 filename. This should rule out race conditions when creating
1006 TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
1007 processing considerably. Previous code took too long for large mails,
1008 triggering a timeout which in turn triggers #401.
1010 TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
1011 in the DK code in transports.c. sendfile() is not really portable,
1012 hence the _LINUX specificness.
1014 TF/01 In the add_headers option to the mail command in an Exim filter,
1015 there was a bug that Exim would claim a syntax error in any
1016 header after the first one which had an odd number of characters
1019 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
1020 callout verification, Exim cached a "reject" for the entire domain. This
1021 is correct for most verifications, but it is not correct for a recipient
1022 verification with use_sender or use_postmaster set, because in that case
1023 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
1024 case of MAIL FROM:<> rejection from other early rejections (e.g.
1025 rejection of HELO). When verifying a recipient using a non-null MAIL
1026 address, the cache is ignored if it shows MAIL FROM:<> rejection.
1027 Whatever the result of the callout, the value of the domain cache is
1028 left unchanged (for any other kind of callout, getting as far as trying
1029 RCPT means that the domain itself is ok).
1031 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
1034 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
1035 manifest itself as EPIPE rather than ECONNECT. When tidying away a
1036 session, the daemon ignores ECONNECT errors and logs others; it now
1037 ignores EPIPE as well.
1039 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
1040 (quoted-printable decoding).
1042 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
1043 later the small subsequent patch to fix an introduced bug.
1045 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
1047 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
1049 PH/08 An error is now given if message_size_limit is specified negative.
1051 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
1052 to be given (somewhat) arbitrary names.
1054 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
1057 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
1058 miscellaneous code fixes
1060 PH/10 Added the log_reject_target ACL modifier to specify where to log
1063 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
1064 hostname. This is wrong, because it relates to the incoming message (and
1065 probably the interface on which it is arriving) and not to the outgoing
1066 callout (which could be using a different interface). This has been
1067 changed to use the value of the helo_data option from the smtp transport
1068 instead - this is what is used when a message is actually being sent. If
1069 there is no remote transport (possible with a router that sets up host
1070 addresses), $smtp_active_hostname is used.
1072 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
1073 tweaks were necessary in order to get it to work (see also 21 below):
1074 (a) The code assumed that strncpy() returns a negative number on buffer
1075 overflow, which isn't the case. Replaced with Exim's string_format()
1077 (b) There were several signed/unsigned issues. I just did the minimum
1078 hacking in of casts. There is scope for a larger refactoring.
1079 (c) The code used strcasecmp() which is not a standard C function.
1080 Replaced with Exim's strcmpic() function.
1081 (d) The code set only $1; it now sets $auth1 as well.
1082 (e) A simple test gave the error "authentication client didn't specify
1083 service in request". It would seem that Dovecot has changed its
1084 interface. Fortunately there's a specification; I followed it and
1085 changed what the client sends and it appears to be working now.
1087 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
1090 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
1091 address A is aliased to B and C, where B exists and C does not. Without
1092 -v the output is "A verified" because verification stops after a
1093 successful redirection if more than one address is generated. However,
1094 with -v the child addresses are also verified. Exim was outputting "A
1095 failed to verify" and then showing the successful verification for C,
1096 with its parentage. It now outputs "B failed to verify", showing B's
1097 parentage before showing the successful verification of C.
1099 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
1100 look up a TXT record in a specific list after matching in a combined
1103 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
1104 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
1105 they consult the DNS. I had assumed they would set it the way they
1106 wanted; and indeed my experiments on Linux seem to show that in some
1107 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
1108 To be on the safe side, however, I have now made the interface to
1109 host_find_byname() similar to host_find_bydns(), with an argument
1110 containing the DNS resolver options. The host_find_byname() function now
1111 sets these options at its start, just as host_find_bydns() does. The smtp
1112 transport options dns_qualify_single and dns_search_parents are passed to
1113 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
1114 of host_find_byname() use the default settings of RES_DEFNAMES
1115 (qualify_single) but not RES_DNSRCH (search_parents).
1117 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
1118 spool_read_header() do less string testing, by means of a preliminary
1119 switch on the second character of optional "-foo" lines. (This is
1120 overdue, caused by the large number of possibilities that now exist.
1121 Originally there were few.) While I was there, I also converted the
1122 str(n)cmp tests so they don't re-test the leading "-" and the first
1123 character, in the hope this might squeeze out yet more improvement.
1125 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
1126 flag allowing group syntax was set by the header_syntax check but not
1127 turned off, possible causing trouble later; (2) The flag was not being
1128 set at all for the header_verify test, causing "group"-style headers to
1129 be rejected. I have now set it in this case, and also caused header_
1130 verify to ignore an empty address taken from a group. While doing this, I
1131 came across some other cases where the code for allowing group syntax
1132 while scanning a header line wasn't quite right (mostly, not resetting
1133 the flag correctly in the right place). These bugs could have caused
1134 trouble for malformed header lines. I hope it is now all correct.
1136 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
1137 with the "reply" argument non-NULL. The code, however (which originally
1138 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
1139 but it didn't always do it. This confused somebody who was copying the
1140 code for some other use. I have removed all the tests.
1142 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
1143 feature that was used to support insecure browsers during the U.S. crypto
1144 embargo. It requires special client support, and Exim is probably the
1145 only MTA that supported it -- and would never use it because real RSA is
1146 always available. This code has been removed, because it had the bad
1147 effect of slowing Exim down by computing (never used) parameters for the
1148 RSA_EXPORT functionality.
1150 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
1151 authenticator to fail if there's a tab character in the incoming data
1152 (there should never be unless someone is messing about, as it's supposed
1153 to be base64-encoded). Also added, on Timo's advice, the "secured" option
1154 if the connection is using TLS or if the remote IP is the same as the
1155 local IP, and the "valid-client-cert option" if a client certificate has
1158 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
1159 authenticators. This can be used for authorization after authentication
1160 succeeds. (In the case of plaintext, it servers for both authentication
1163 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
1164 if any retry times were supplied.
1166 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
1167 connection, where there is no client IP address to check. In this
1168 situation, the verify now always succeeds.
1170 PH/25 Applied John Jetmore's -Mset patch.
1172 PH/26 Added -bem to be like -Mset, but loading a message from a file.
1174 PH/27 In a string expansion for a processed (not raw) header when multiple
1175 headers of the same name were present, leading whitespace was being
1176 removed from all of them, but trailing whitespace was being removed only
1177 from the last one. Now trailing whitespace is removed from each header
1178 before concatenation. Completely empty headers in a concatenation (as
1179 before) are ignored.
1181 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
1182 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
1184 PH/29 [Removed. This was a change that I later backed out, and forgot to
1185 correct the ChangeLog entry (that I had efficiently created) before
1186 committing the later change.]
1188 PH/30 Exim was sometimes attempting to deliver messages that had suffered
1189 address errors (4xx response to RCPT) over the same connection as other
1190 messages routed to the same hosts. Such deliveries are always "forced",
1191 so retry times are not inspected. This resulted in far too many retries
1192 for the affected addresses. The effect occurred only when there were more
1193 hosts than the hosts_max_try setting in the smtp transport when it had
1194 the 4xx errors. Those hosts that it had tried were not added to the list
1195 of hosts for which the message was waiting, so if all were tried, there
1196 was no problem. Two fixes have been applied:
1198 (i) If there are any address or message errors in an SMTP delivery, none
1199 of the hosts (tried or untried) are now added to the list of hosts
1200 for which the message is waiting, so the message should not be a
1201 candidate for sending over the same connection that was used for a
1202 successful delivery of some other message. This seems entirely
1203 reasonable: after all the message is NOT "waiting for some host".
1204 This is so "obvious" that I'm not sure why it wasn't done
1205 previously. Hope I haven't missed anything, but it can't do any
1206 harm, as the worst effect is to miss an optimization.
1208 (ii) If, despite (i), such a delivery is accidentally attempted, the
1209 routing retry time is respected, so at least it doesn't keep
1210 hammering the server.
1212 PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
1213 in ${readsocket because some servers need this prod.
1215 PH/32 Added some extra debug output when updating a wait-xxx database.
1217 PH/33 The hint "could be header name not terminated by colon", which has been
1218 given for certain expansion errors for a long time, was not being given
1219 for the ${if def:h_colon_omitted{... case.
1221 PH/34 The spec says: "With one important exception, whenever a domain list is
1222 being scanned, $domain contains the subject domain." There was at least
1223 one case where this was not true.
1225 PH/35 The error "getsockname() failed: connection reset by peer" was being
1226 written to the panic log as well as the main log, but it isn't really
1227 panic-worthy as it just means the connection died rather early on. I have
1228 removed the panic log writing for the ECONNRESET error when getsockname()
1231 PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
1232 runs only) independently of the message's sender address. This meant
1233 that, if the 4xx error was in fact related to the sender, a different
1234 message to the same recipient with a different sender could confuse
1235 things. In particualar, this can happen when sending to a greylisting
1236 server, but other circumstances could also provoke similar problems.
1237 I have changed the default so that the retry time for these errors is now
1238 based a combination of the sender and recipient addresses. This change
1239 can be overridden by setting address_retry_include_sender=false in the
1242 PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
1243 remote server are returned as part of bounce messages. This was not
1244 happening for LMTP over a pipe (the lmtp transport), but now it is the
1245 same for both kinds of LMTP.
1247 PH/38 Despite being documented as not happening, Exim was rewriting addresses
1248 in header lines that were in fact CNAMEs. This is no longer the case.
1250 PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
1251 and queue runs started by the daemon processed all messages. This has
1252 been fixed so that -R and -S can now usefully be given with -q<time>.
1254 PH/40 Import PCRE release 6.7 (fixes some bugs).
1256 PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
1258 PH/42 Give an error if -q is specified more than once.
1260 PH/43 Renamed the variables $interface_address and $interface_port as
1261 $received_ip_address and $received_port, to make it clear that these
1262 values apply to message reception, and not to the outgoing interface when
1263 a message is delivered. (The old names remain recognized, of course.)
1265 PH/44 There was no timeout on the connect() call when using a Unix domain
1266 socket in the ${readsocket expansion. There now is.
1268 PH/45 Applied a modified version of Brad Jorsch's patch to allow "message" to
1269 be meaningful with "accept".
1271 SC/01 Eximstats V1.43
1272 Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
1274 SC/02 Eximstats V1.44
1275 Use a glob alias rather than an array ref in the generated
1276 parser. This improves both readability and performance.
1278 SC/03 Eximstats V1.45 (Marco Gaiarin / Steve Campbell)
1279 Collect SpamAssassin and rejection statistics.
1280 Don't display local sender or destination tables unless
1281 there is data to show.
1282 Added average volumes into the top table text output.
1284 SC/04 Eximstats V1.46
1285 Collect data on the number of addresses (recipients)
1286 as well as the number of messages.
1288 SC/05 Eximstats V1.47
1289 Added 'Message too big' to the list of mail rejection
1290 reasons (thanks to Marco Gaiarin).
1292 SC/06 Eximstats V1.48
1293 Mainlog lines which have GMT offsets and are too short to
1294 have a flag are now skipped.
1296 SC/07 Eximstats V1.49 (Alain Williams)
1297 Added the -emptyok flag.
1299 SC/08 Eximstats V1.50
1300 Fixes for obtaining the IP address from reject messages.
1302 JJ/03 exipick.20061117.2, made header handling as similar to exim as possible
1303 (added [br]h_ prefixes, implemented RFC2047 decoding. Fixed
1304 whitesspace changes from 4.64-PH/27
1306 JJ/04 exipick.20061117.2, fixed format and added $message_headers_raw to
1309 JJ/05 exipick.20061117.2, bug fixes (error out sooner when invalid criteria
1310 are found, allow negative numbers in numeric criteria)
1312 JJ/06 exipick.20061117.2, added new $message_body_missing variable
1314 JJ/07 exipick.20061117.2, added $received_ip_address and $received_port
1315 to match changes made in 4.64-PH/43
1317 PH/46 Applied Jori Hamalainen's patch to add features to exiqsumm.
1319 PH/47 Put in an explicit test for a DNS lookup of an address record where the
1320 "domain" is actually an IP address, and force a failure. This locks out
1321 those revolvers/nameservers that support "A-for-A" lookups, in
1322 contravention of the specifications.
1324 PH/48 When a host name was looked up from an IP address, and the subsequent
1325 forward lookup of the name timed out, the host name was left in
1326 $sender_host_name, contrary to the specification.
1328 PH/49 Although default lookup types such as lsearch* or cdb*@ have always been
1329 restricted to single-key lookups, Exim was not diagnosing an error if
1330 * or *@ was used with a query-style lookup.
1332 PH/50 Increased the value of DH_BITS in tls-gnu.c from 768 to 1024.
1334 MH/01 local_scan ABI version incremented to 1.1. It should have been updated
1335 long ago, but noone interested enough thought of it. Let's just say that
1336 the "1.1" means that there are some new functions that weren't there at
1337 some point in the past.
1339 PH/51 Error processing for expansion failure of helo_data from an smtp
1340 transport during callout processing was broken.
1342 PH/52 Applied John Jetmore's patch to allow tls-on-connect and STARTTLS to be
1343 tested/used via the -bh/-bhc/-bs options.
1345 PH/53 Added missing "#include <time.h>" to pcre/pcretest.c (this was a PCRE
1346 bug, fixed in subsequent PCRE releases).
1348 PH/54 Applied Robert Bannocks' patch to avoid a problem with references that
1349 arises when using the Solaris LDAP libraries (but not with OpenLDAP).
1351 PH/55 Check for a ridiculously long file name in exim_dbmbuild.
1357 SC/01 Use a glob alias rather than an array ref in eximstats generated
1358 parser. This improves both readability and performance.
1360 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
1361 Don't display local sender or destination tables in eximstats unless
1362 there is data to show.
1363 Added average volumes into the eximstats top table text output.
1365 SC/03 Collect data on the number of addresses (recipients) as well
1366 as the number of messages in eximstats.
1368 TF/01 Correct an error in the documentation for the redirect router. Exim
1369 does (usually) call initgroups() when daemonizing.
1371 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
1372 with consistent privilege compared to when running as a daemon.
1374 TF/03 Note in the spec that $authenticated_id is not set for local
1375 submissions from trusted users.
1377 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
1378 Thanks to Dean Brooks <dean@iglou.com> for the patch.
1380 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
1381 by adding some example configuration directives to the default
1382 configuration file. A little bit of work is required to uncomment the
1383 directives and define how usernames and passwords are checked, but
1384 there is now a framework to start from.
1386 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
1387 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
1388 without this. I don't know how relevant this is to other LDAP libraries.
1390 PH/02 Add the verb name to the "unknown ACL verb" error.
1392 PH/03 Magnus Holmgren's patch for filter_prepend_home.
1394 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
1396 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
1397 directory not expanded when it should be if an expanded home directory
1398 was set for the address (which is overridden by the transport).
1400 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
1403 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
1404 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
1405 because it is too late at that time, and has no effect.
1407 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
1408 security issue with \' (bugzilla #107). I could not use the
1409 PQescapeStringConn() function, because it needs a PGconn value as one of
1412 PH/08 When testing addresses using -bt, indicate those final addresses that
1413 are duplicates that would not cause an additional delivery. At least one
1414 person was confused, thinking that -bt output corresponded to deliveries.
1415 (Suppressing duplicates isn't a good idea as you lose the information
1416 about possibly different redirections that led to the duplicates.)
1418 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
1419 systems where poll() doesn't work, in particular OS X.
1421 PH/10 Added more information to debugging output for retry time not reached.
1423 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
1424 operations in malware.c.
1426 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
1429 PH/13 If write_rejectlog was set false when logging was sent to syslog with
1430 syslog_duplication set false, log lines that would normally be written
1431 both the the main log and to the reject log were not written to syslog at
1434 PH/14 In the default configuration, change the use of "message" in ACL warn
1435 statements to "add_header".
1437 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
1438 not followed by a command (e.g. "seen endif").
1440 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
1441 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
1444 PH/17 Added extra conditions to the default value of delay_warning_condition
1448 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
1449 { match{$h_precedence:}{(?i)bulk|list|junk} } \
1450 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
1453 The Auto-Submitted: and various List- headers are standardised, whereas I
1454 don't think Precedence: ever was.
1456 PH/18 Refactored debugging code in route_finduser() to show more information,
1457 in particular, the error code if getpwnam() issues one.
1459 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
1460 This is apparently needed in addition to the PH/07 change above to avoid
1461 any possible encoding problems.
1463 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
1464 but not after initializing Perl.
1466 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
1467 output them only if debugging. By default they are written stderr,
1468 apparently, which is not desirable.
1470 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
1473 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
1476 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
1478 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
1479 authenticated or an ident call has been made. Suppress the default
1480 values for $authenticated_id and $authenticated_sender (but permit -oMai
1481 and -oMas) when testing with -bh.
1483 PH/24 Re-jigged the order of the tests in the default configuration so that the
1484 tests for valid domains and recipients precede the DNS black list and CSA
1485 tests, on the grounds that those ones are more expensive.
1487 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
1488 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
1489 This bug exists in every version of Exim that I still have, right back to
1492 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
1493 However, an attempt to turn on case-sensitivity in a regex key by
1494 including (?-i) didn't work because the subject string was already
1495 lowercased, and the effects were non-intuitive. It turns out that a
1496 one-line patch can be used to allow (?-i) to work as expected.
1502 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
1503 other effects) broke the use of negated acl sub-conditions.
1505 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
1508 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
1509 "Deny" causes Exim to reject the incoming connection with a 554 error.
1510 Unfortunately, if there is a major crisis, such as a disk failure,
1511 tcp-wrappers gives "deny", whereas what one would like would be some
1512 kind of temporary error. A kludge has been added to help with this.
1513 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
1514 554 error is used if errno is still zero or contains ENOENT (which occurs
1515 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
1518 PH/03 Add -lutil to the default FreeBSD LIBS setting.
1520 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
1521 errors. Otherwise a message that provokes a temporary error (when other
1522 messages do not) can cause a whole host to time out.
1524 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
1525 addresses were routed directly to files or pipes from a redirect router.
1526 File deliveries just didn't batch; pipe deliveries might have suffered
1529 PH/06 A failure to get a lock for a hints database would erroneously always say
1530 "Failed to get write lock", even when it was really a read lock.
1532 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
1533 of 0600. This has been changed to use the value of the lockfile_mode
1534 option (which defaults to 0600).
1536 PH/08 Applied small patch from the Sieve maintainer.
1538 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
1539 folder from quota calculations, a direct delivery into this folder messed
1540 up the contents of the maildirsize file. This was because the regex was
1541 used only to exclude .Trash (or whatever) when the size of the mailbox
1542 was calculated. There was no check that a delivery was happening into an
1543 excluded directory. This bug has been fixed by ignoring all quota
1544 processing for deliveries into excluded directories.
1546 PH/10 Added the maildirfolder_create_regex option to appendfile.
1552 PH/01 The code for finding all the local interface addresses on a FreeBSD
1553 system running IPv6 was broken. This may well have applied to all BSD
1554 systems, as well as to others that have similar system calls. The broken
1555 code found IPv4 interfaces correctly, but gave incorrect values for the
1556 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
1557 that it would not match correctly against @[] and not recognize the IPv6
1560 PH/02 The ipliteral router was not recognizing addresses of the form user@
1561 [ipv6:....] because it didn't know about the "ipv6:" prefix.
1563 PH/03 Added disable_ipv6.
1565 PH/04 Changed $reply_address to use the raw form of the headers instead of the
1566 decoded form, because it is most often used to construct To: headers
1567 lines in autoreplies, and the decoded form may well be syntactically
1568 invalid. However, $reply_address has leading white space removed, and all
1569 newlines turned into spaces so that the autoreply transport does not
1572 PH/05 If group was specified without a user on a router, and no group or user
1573 was specified on a transport, the group from the router was ignored.
1575 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
1576 for visible compile-time settings that can be used to change these
1577 numbers, for those that want even more. Backwards compatibility with old
1578 spool files has been maintained. However, going back to a previous Exim
1579 release will lost any variables that are in spool files.
1581 PH/07 Two small changes when running in the test harness: increase delay when
1582 passing a TCP/IP connection to a new process, in case the original
1583 process has to generate a bounce, and remove special handling of
1584 127.0.0.2 (sic), which is no longer necessary.
1586 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
1587 be the same on different OS.
1589 PH/09 Moved a debug statement in filter processing to avoid a race problem when
1592 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
1593 whether --show-vars was specified or not
1595 JJ/02 exipick: Added support for new ACL variable spool format introduced
1598 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
1599 syntactically invalid From: or Reply-to: line, and a filter used this to
1600 generate an autoreply, and therefore failed to obtain an address for the
1601 autoreply, Exim could try to deliver to a non-existent relative file
1602 name, causing unrelated and misleading errors. What now happens is that
1603 it logs this as a hard delivery error, but does not attempt to create a
1606 PH/11 The exinext utility has a -C option for testing purposes, but although
1607 the given file was scanned by exinext itself; it wasn't being passed on
1608 when Exim was called.
1610 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
1611 an end-of-file indication when reading a command response.
1613 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
1614 compiled. In many other places in Exim, IPv6 addresses are always
1615 recognized, so I have changed this. It also means that IPv4 domain
1616 literals of the form [IPV4:n.n.n.n] are now always recognized.
1618 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
1619 used if the router is not running as root, for example, when verifying at
1620 ACL time, or when using -bh. The debugging output from this situation was
1621 non-existent - all you got was a failure to exec. I have made two
1624 (a) Failures to set uid/gid, the current directory, or a process leader
1625 in a subprocess such as that created by queryprogram now generate
1626 suitable debugging ouput when -d is set.
1628 (b) The queryprogram router detects when it is not running as root,
1629 outputs suitable debugging information if -d is set, and then runs
1630 the subprocess without attempting to change uid/gid.
1632 PH/15 Minor change to Makefile for building test_host (undocumented testing
1635 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
1636 additional section of a DNS packet that returns MX or SRV records.
1637 Instead, it always explicitly searches for A/AAAA records. This avoids
1638 major problems that occur when a DNS server includes only records of one
1639 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
1640 fixed another bug: if SRV records were looked up and the corresponding
1641 address records were *not* found in the additional section, the port
1642 values from the SRV records were lost.
1644 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
1645 using the correct key (the original address) when searching the retry
1646 rules in order to find which one to use for generating the retry hint.
1648 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
1649 adding the default one. Similarly, if it contains a Reply-To: header, the
1650 errors_reply_to option, if set, is not used.
1652 PH/19 When calculating a retry time, Exim used to measure the "time since
1653 failure" by looking at the "first failed" field in the retry record. Now
1654 it does not use this if it is later than than the arrival time of the
1655 message. Instead it uses the arrival time. This makes for better
1656 behaviour in cases where some deliveries succeed, thus re-setting the
1657 "first failed" field. An example is a quota failure for a huge message
1658 when small messages continue to be delivered. Without this change, the
1659 "time since failure" will always be short, possible causing more frequent
1660 delivery attempts for the huge message than are intended.
1661 [Note: This change was subsequently modified - see PH/04 for 4.62.]
1663 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
1664 $1, $2, $3) because the numerical variables can be reset during some
1665 expansion items (e.g. "match"), thereby losing the authentication data.
1667 PH/21 Make -bV show the size of off_t variables so that the test suite can
1668 decide whether to run tests for quotas > 2G.
1670 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
1671 mailbox_size, and mailbox_filecount in the appendfile transport. If a
1672 filecount value is greater than 2G or if a quota value is greater than 2G
1673 on a system where the size of off_t is not greater than 4, a panic error
1676 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
1677 never match. The debug and -bh output now contains an explicit error
1678 message indicating a malformed IPv4 address or mask.
1680 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
1681 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
1682 PH/23 above applies.
1684 PH/25 Do not write to syslog when running in the test harness. The only
1685 occasion when this arises is a failure to open the main or panic logs
1686 (for which there is an explicit test).
1688 PH/26 Added the /no_tell option to "control=freeze".
1690 PH/27 If a host name lookup failed very early in a connection, for example, if
1691 the IP address matched host_lookup and the reverse lookup yielded a name
1692 that did not have a forward lookup, an error message of the form "no IP
1693 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
1694 could be logged. Now it outputs the IP address instead of "NULL".
1696 PH/28 An enabling patch from MH: add new function child_open_exim2() which
1697 allows the sender and the authenticated sender to be set when
1698 submitting a message from within Exim. Since child_open_exim() is
1699 documented for local_scan(), the new function should be too.
1701 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
1702 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
1703 results in an empty string is now treated as unset.
1705 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
1707 PH/31 Added sender_verify_fail logging option.
1709 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
1710 needed by RFC 822 but not by RFC 2822 was commented out. I have now
1711 tidied the source and removed it altogether.
1713 PH/33 When a queue run was abandoned because the load average was too high, a
1714 log line was always written; now it is written only if the queue_run log
1715 selector is set. In addition, the log line for abandonment now contains
1716 information about the queue run such as the pid. This is always present
1717 in "start" and "stop" lines but was omitted from the "abandon" line.
1719 PH/34 Omit spaces between a header name and the colon in the error message that
1720 is given when verify = headers_syntax fails (if there are lots of them,
1721 the message gets confusing).
1723 PH/35 Change the default for dns_check_names_pattern to allow slashes within
1724 names, as there are now some PTR records that contain slashes. This check
1725 is only to protect against broken name servers that fall over on strange
1726 characters, so the fact that it applies to all lookups doesn't matter.
1728 PH/36 Now that the new test suite is complete, we can remove some of the
1729 special code in Exim that was needed for the old test suite. For example,
1730 sorting DNS records because real resolvers return them in an arbitrary
1731 order. The new test suite's fake resolver always returns records in the
1734 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
1735 bounces) except when queue_only is set, to avoid logging races between
1736 the different processes.
1738 PH/38 Panic-die if .include specifies a non-absolute path.
1740 PH/39 A tweak to the "H" retry rule from its user.
1742 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
1743 a label. They prevented compilation on older perls.
1745 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
1746 a warning to be raised on newish perls.
1748 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
1749 on queue. Changes to match documented behaviour of showing count of
1750 messages matching specified criteria.
1752 PH/40 Changed the default ident timeout from 30s to 5s.
1754 PH/41 Added support for the use of login_cap features, on those BSD systems
1755 that have them, for controlling the resources used by pipe deliveries.
1757 PH/42 The content-scanning code uses fopen() to create files in which to put
1758 message data. Previously it was not paying any attention to the mode of
1759 the files. Exim runs with umask(0) because the rest of the code creates
1760 files with open(), and sets the required mode explicitly. Thus, these
1761 files were ending up world-writeable. This was not a big issue, because,
1762 being within the spool directory, they were not world-accessible. I have
1763 created a function called modefopen, which takes an additional mode
1764 argument. It sets umask(777), creates the file, chmods it to the required
1765 mode, then resets the umask. All the relevant calls to fopen() in the
1766 content scanning code have been changed to use this function.
1768 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
1769 to 24 hours. This avoids potential overflow problems when processing G
1770 and H retry rules. I suspect nobody ever tinkers with this value.
1772 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
1774 PH/45 When the plaintext authenticator is running as a client, the server's
1775 challenges are checked to ensure they are valid base64 strings. By
1776 default, the authentication attempt is cancelled if an invalid string is
1777 received. Setting client_ignore_invalid_base64 true ignores these errors.
1778 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
1779 they are received. Thus, the responses can be made to depend on the
1780 challenges. If an invalid string is ignored, an empty string is placed in
1783 PH/46 Messages that are created by the autoreply transport now contains a
1784 References: header, in accordance with RFCs 2822 and 3834.
1786 PH/47 Added authenticated_sender_force to the smtp transport.
1788 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
1790 PH/49 Installed latest patch from the Sieve maintainer.
1792 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
1793 was also set, the appendfile transport was unnecessarily scanning a
1794 directory of message files (e.g. for maildir delivery) to find the count
1795 of files (along with the size), even though it did not need this
1796 information. It now does the scan only if it needs to find either the
1797 size of the count of files.
1799 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
1801 PH/52 Two bugs concerned with error handling when the smtp transport is
1804 (i) Exim was not creating retry information for temporary errors given
1805 for individual recipients after the DATA command when the smtp transport
1806 was used in LMTP mode. This meant that they could be retried too
1807 frequently, and not timed out correctly.
1809 (ii) Exim was setting the flag that allows error details to be returned
1810 for LMTP errors on RCPT commands, but not for LMTP errors for individual
1811 recipients that were returned after the DATA command.
1813 PH/53 This is related to PH/52, but is more general: for any failing address,
1814 when detailed error information was permitted to be returned to the
1815 sender, but the error was temporary, then after the final timeout, only
1816 "retry timeout exceeded" was returned. Now it returns the full error as
1817 well as "retry timeout exceeded".
1819 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
1820 do this, and (what is worse) MTAs that accept it.
1822 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
1823 will now be deprecated.
1825 PH/56 New os.c-cygwin from the Cygwin maintainer.
1827 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
1828 formats (previously only available in exim formats via -bpr, -bpru,
1829 and -bpra. Now also available in native and exiqgrep formats)
1831 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
1832 with very large, slow to parse queues
1834 JJ/08 exipick: added ! as generic prefix to negate any criteria format
1836 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
1838 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
1839 responses to authentication challenges, though it was showing the
1840 challenges; (ii) I've removed the CR characters from the debug output for
1843 PH/58 Allow for the insertion of a newline as well as a space when a string
1844 is turned into more than one encoded-word during RFC 2047 encoding. The
1845 Sieve code now uses this.
1847 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
1848 data_4xx, lost_connection, tls_required.
1850 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
1851 message was being sent as an SMTP response.
1853 PH/61 Add -l and -k options to exicyclog.
1855 PH/62 When verifying, if an address was redirected to one new address, so that
1856 verification continued, and the new address failed or deferred after
1857 having set something in $address_data, the value of $address_data was not
1858 passed back to the ACL. This was different to the case when no
1859 redirection occurred. The value is now passed back in both cases.
1861 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
1862 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
1863 use login_cap.h, so on its own it isn't the distinguishing feature. The
1864 new name refers directly to the setclassresources() function.
1866 PH/65 Added configuration files for NetBSD3.
1868 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
1870 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
1871 is preferred over IPv4.
1873 PH/68 The bounce_return_message and bounce_return_body options were not being
1874 honoured for bounces generated during the reception of non-SMTP messages.
1875 In particular, this applied to messages rejected by the ACL. This bug has
1876 been fixed. However, if bounce_return_message is true and bounce_return_
1877 body is false, the headers that are returned for a non-SMTP message
1878 include only those that have been read before the error was detected.
1879 (In the case of an ACL rejection, they have all been read.)
1881 PH/69 The HTML version of the specification is now built in a directory called
1882 spec_html instead of spec.html, because the latter looks like a path with
1883 a MIME-type, and this confuses some software.
1885 PH/70 Catch two compiler warnings in sieve.c.
1887 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
1888 function verify_get_ident() calls ip_connect() to connect a socket, but
1889 if the "connect()" function timed out, ip_connect() used to close the
1890 socket. However, verify_get_ident() also closes the socket later, and in
1891 between Exim writes to the log, which may get opened at this point. When
1892 the socket was closed in ip_connect(), the log could get the same file
1893 descriptor number as the socket. This naturally causes chaos. The fix is
1894 not to close the socket in ip_connect(); the socket should be closed by
1895 the function that creates it. There was only one place in the code where
1896 this was missing, in the iplookup router, which I don't think anybody now
1897 uses, but I've fixed it anyway.
1899 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
1900 well as to direct DNS lookups. Otherwise the handling of names in host
1901 lists is inconsistent and therefore confusing.
1907 PH/01 Two changes to the default runtime configuration:
1909 (1) Move the checks for relay_from_hosts and authenticated clients from
1910 after to before the (commented out) DNS black list checks.
1912 (2) Add control=submission to the relay_from_hosts and authenticated
1913 clients checks, on the grounds that messages accepted by these
1914 statements are most likely to be submissions.
1916 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
1918 (1) Generate an error if the third argument for the ${prvs expansion is
1921 (2) Treat a missing third argument of ${prvscheck as if it were an empty
1924 (3) Reset the variables that are obtained from the first argument of
1925 ${prvscheck and used in the second argument before leaving the code,
1926 because their memory is reclaimed, so using them afterwards may do
1929 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
1930 one (it's much easier than Tom thought :-).
1932 (5) Because of (4), we can now allow for the use of $prvscheck_result
1933 inside the third argument.
1935 PH/03 For some reason, the default setting of PATH when running a command from
1936 a pipe transport was just "/usr/bin". I have changed it to
1939 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
1940 anything to be listed in the output from -bV.
1942 PH/05 When a filter generated an autoreply, the entire To: header line was
1943 quoted in the delivery log line, like this:
1945 => >A.N.Other <ano@some.domain> <original@ddress> ...
1947 This has been changed so that it extracts the operative address. There
1948 may be more than one such address. If so, they are comma-separated, like
1951 => >ano@some.domain,ona@other.domain <original@ddress> ...
1953 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
1954 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
1955 not being looked up in the rDNS to get a host name, Exim was showing the
1956 IP address twice in Received: lines, even though the IP addresses were
1957 identical. For example:
1959 Received: from [1.2.3.4] (helo=[1.2.3.4])
1961 However, if the real host name was known, it was omitting the HELO data
1962 if it matched the actual IP address. This has been tidied up so that it
1963 doesn't show the same IP address twice.
1965 PH/07 When both +timestamp and +memory debugging was on, the value given by
1966 $tod_xxx expansions could be wrong, because the tod_stamp() function was
1967 called by the debug printing, thereby overwriting the timestamp buffer.
1968 Debugging no longer uses the tod_stamp() function when +timestamp is set.
1970 PH/08 When the original message was included in an autoreply transport, it
1971 always said "this is a copy of the message, including all the headers",
1972 even if body_only or headers_only was set. It now gives an appropriate
1975 PH/09 Applied a patch from the Sieve maintainer which:
1977 o fixes some comments
1978 o adds the (disabled) notify extension core
1979 o adds some debug output for the result of if/elsif tests
1980 o points to the current vacation draft in the documentation
1981 and documents the missing references header update
1985 o fixes a bug in processing the envelope test (when testing
1986 multiple envelope elements, the last element determinted the
1989 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
1990 Electronic Mail") by including:
1992 Auto-submitted: auto-generated
1994 in the messages that it generates (bounce messages and others, such as
1995 warnings). In the case of bounce messages for non-SMTP mesages, there was
1996 also a typo: it was using "Auto_submitted" (underscore instead of
1997 hyphen). Since every message generated by Exim is necessarily in response
1998 to another message, thes have all been changed to:
2000 Auto-Submitted: auto-replied
2002 in accordance with these statements in the RFC:
2004 The auto-replied keyword:
2006 - SHOULD be used on messages sent in direct response to another
2007 message by an automatic process,
2009 - MUST NOT be used on manually-generated messages,
2011 - MAY be used on Delivery Status Notifications (DSNs) and Message
2012 Disposition Notifications (MDNs),
2014 - MUST NOT be used on messages generated by automatic or periodic
2015 processes, except for messages which are automatic responses to
2018 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
2019 to the default Received: header definition.
2021 PH/12 Added log selector acl_warn_skipped (default on).
2023 PH/13 After a successful wildlsearch lookup, discard the values of numeric
2024 variables because (a) they are in the wrong storage pool and (b) even if
2025 they were copied, it wouldn't work properly because of the caching.
2027 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
2028 checking when decoding. Apparently there are clients that generate
2029 overlong encoded strings. Why am I not surprised?
2031 PH/15 If the first argument of "${if match_address" was not empty, but did not
2032 contain an "@" character, Exim crashed. Now it writes a panic log message
2033 and treats the condition as false.
2035 PH/16 In autoreply, treat an empty string for "once" the same as unset.
2037 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
2038 extension "envelope-auth". The code is finished and in agreement with
2039 other implementations, but there is no documentation so far and in fact,
2040 nobody wrote the draft yet. This extension is currently #undef'ed, thus
2041 not changing the active code.
2043 Print executed "if" and "elsif" statements when debugging is used. This
2044 helps a great deal to understand what a filter does.
2046 Document more things not specified clearly in RFC3028. I had all this
2047 sorted out, when out of a sudden new issues came to my mind. Oops."
2049 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
2052 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
2053 canonical form (as documented). However, after a host name lookup from
2054 the IP address, check_host() was doing a simple string comparison with
2055 addresses acquired from the DNS when checking that the found name did
2056 have the original IP as one of its addresses. Since any found IPv6
2057 addresses are likely to be in abbreviated form, the comparison could
2058 fail. Luckily, there already exists a function for doing the comparison
2059 by converting both addresses to binary, so now that is used instead of
2060 the text comparison.
2062 PH/20 There was another similar case to PH/19, when a complete host name was
2063 given in a host list; looking up its IP address could give an abbreviated
2064 form, whereas the current host's name might or might not be abbreviated.
2065 The same fix has been applied.
2071 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
2072 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
2075 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
2076 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
2078 PH/03 Typo: missing ".o" in src/pcre/Makefile.
2080 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
2081 header line, restrict the check to what is listed in RFCs 2369 and 2929.
2082 Also, for "Auto-Submitted", treat anything other than "no" as
2083 non-personal, in accordance with RFC 3834. (Previously it treated
2084 anything starting "auto-" as non-personal.)
2086 TF/01 The control=submission/name=... option had a problem with syntax
2087 errors if the name included a slash character. The /name= option
2088 now slurps the rest of the string, so it can include any characters
2089 but it must come last in the list of options (after /sender_retain
2092 PH/05 Some modifications to the interface to the fake nameserver for the new
2100 TK/01 Added the "success_on_redirect" address verification option. See
2101 NewStuff for rationale and an example.
2103 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
2105 PH/02 Patch to exigrep to allow it to work on syslog lines.
2107 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
2108 fread() to skip over the body file's header line, because in Cygwin the
2109 header line is locked and is inaccessible.
2111 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
2112 co-exist for some time) to make it clear that it is the Exim ID that is
2113 referenced, not the Message-ID: header line.
2115 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
2116 string_format() function, because snprintf() does not exist on all
2119 PH/06 The use of forbid_filter_existstest now also locks out the use of the
2120 ${stat: expansion item.
2122 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
2123 protocol synchronization error", to keep the pedants happy.
2125 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
2126 well as for IRIX systems, when gcc is being used. See the host.c source
2129 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
2131 PH/10 Named domain lists were not working if used in a queue_smtp_domains
2134 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
2135 transport and to the smtp transport in LMTP mode.
2137 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
2139 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
2140 run a filter in a subprocess. This could lead to confusion in subsequent
2141 lookups in the parent process. There should also be a search_tidyup() at
2142 the end of the subprocess.
2144 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
2145 only if the host matched helo_try_verify_hosts, which caused the
2146 verification to occur when the EHLO/HELO command was issued. The ACL just
2147 tested the remembered result. Now, if a previous verification attempt has
2148 not happened, "verify = helo" does it there and then.
2150 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
2152 TK/03 Fix log output including CR from clamd.
2154 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
2155 exist provoked a memory error which could cause a segfault.
2157 PH/15 Installed PCRE 6.2
2159 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
2161 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
2162 of the problem. Specifically, suggested +O2 rather than +O1 for the
2165 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
2167 PH/20 If a delivery was routed to a non-standard port by means of an SRV
2168 record, the port was not correctly logged when the outgoing_port log
2169 selector was set (it logged the transort's default port).
2171 PH/21 Added support for host-specific ports to manualroute, queryprogram,
2172 fallback_hosts, and "hosts" in the smtp transport.
2174 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
2175 host errors such as "Connection refused".
2177 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
2178 authentication with radiusclient 0.4.9:
2180 - Error returned from rc_read_config was caught wrongly
2181 - Username/password not passed on to radius server due to wrong length.
2183 The presumption is that some radiusclient API changes for 4.51/PH/17
2184 were not taken care of correctly. The code is still untested by me (my
2185 Linux distribution still has 0.3.2 of radiusclient), but it was
2186 contributed by a Radius user.
2188 PH/24 When doing a callout, the value of $domain wasn't set correctly when
2189 expanding the "port" option of the smtp transport.
2191 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
2192 while reading a MIME header. Thanks to Tom Hughes for a patch.
2194 PH/24 Include config.h inside local_scan.h so that configuration settings are
2197 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
2198 that in an ACL for RCPT (for example), you can examine exactly what was
2201 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
2202 commands, but it was not correctly comparing the address with the actual
2203 client host address. Thus, it would show the EHLO address in Received:
2204 header lines when this was not necessary.
2206 PH/27 Added the % operator to ${eval:}.
2208 PH/28 Exim tries to create and chdir to its spool directory when it starts;
2209 it should be ignoring failures (because with -C, for example, it has lost
2210 privilege). It wasn't ignoring creation failures other than "already
2213 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
2216 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
2217 because an input line was too long, either on its own, or by virtue of
2218 too many continuations, the temporary file was not being removed, and the
2219 return code was incorrect.
2221 PH/31 Missing "BOOL" in function definition in filtertest.c.
2223 PH/32 Applied Sieve patches from the maintainer.
2225 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
2227 PH/33 Added "verify = not_blind".
2229 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
2230 Local/Makefile (with some defaults set). These are used in built scripts
2231 such as exicyclog, but they have never been used in the exim_install
2232 script (though there are many overriding facilities there). I have
2233 arranged that the exim_install script now takes note of these two
2236 PH/35 Installed configuration files for Dragonfly.
2238 PH/36 When a locally submitted message by a trusted user did not contain a
2239 From: header, and the sender address was obtained from -f or from an SMTP
2240 MAIL command, and the trusted user did not use -F to supply a sender
2241 name, $originator_name was incorrectly used when constructing a From:
2242 header. Furthermore, $originator_name was used for submission mode
2243 messages from external hosts without From: headers in a similar way,
2244 which is clearly wrong.
2246 PH/37 Added control=suppress_local_fixups.
2248 PH/38 When log_selector = +received_sender was set, and the addition of the
2249 sender made the log line's construction buffer exactly full, or one byte
2250 less than full, an overflow happened when the terminating "\n" was
2253 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
2254 when the result of a list match is failure because a DNS lookup failed.
2256 PH/40 RM_COMMAND is now used in the building process.
2258 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
2259 the "build-* directories that it finds.
2261 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
2262 address in a domain literal was a prefix of an interface address.
2264 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
2265 when verifying a sender address, unless rewrite_headers is false.
2267 PH/44 Wrote a long comment about why errors_to addresses are verified as
2268 recipients, not senders.
2270 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
2271 the ratelimit ACL was added.
2273 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
2275 PH/46 Added extra information about PostgreSQL errors to the error string.
2277 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
2278 avoiding the need to install special zones in a real server. This is
2279 backwards compatible; if it can't find the fake resolver, it drops back.
2280 Thus, both old and new test suites can be run.
2282 TF/02 Added util/ratelimit.pl
2284 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
2285 clock is set back in time.
2287 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
2288 Candler <B.Candler@pobox.com>.
2290 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
2291 OK for addresses that are the result of redirections.
2293 PH/48 A number of further additions for the benefit of the new test suite,
2294 including a fake gethostbyname() that interfaces to the fake DNS resolver
2297 TF/06 The fix for widen_domains has also been applied to qualify_single and
2298 search_parents which are the other dnslookup options that can cause
2301 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
2304 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
2306 TF/07 Exim produced the error message "an SRV record indicated no SMTP
2307 service" if it encountered an MX record with an empty target hostname.
2308 The message is now "an MX or SRV record indicated no SMTP service".
2310 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
2311 if the DNS of the sending site is misconfigured. This is quite a
2312 common situation. This change restores the behaviour of treating a
2313 helo verification defer as a failure.
2315 PH/51 If self=fail was set on a router, the bounce message did not include the
2316 actual error message.
2322 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
2324 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
2325 command itself ended in error, the underlying message about the transport
2326 filter timeout was being overwritten with the pipe command error. Now the
2327 underlying error message should be appended to the second error message.
2329 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
2331 PH/02 Reduce the amount of output that "make" produces by default. Full output
2332 can still be requested.
2334 PH/03 The warning log line about a condition test deferring for a "warn" verb
2335 was being output only once per connection, rather than after each
2336 occurrence (because it was using the same function as for successful
2337 "warn" verbs). This seems wrong, so I have changed it.
2339 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
2340 it should not have, which might have caused a crash in the right
2341 circumstances, but probably never did.
2343 PH/04 Installed a modified version of Tony Finch's patch to make submission
2344 mode fix the return path as well as the Sender: header line, and to
2345 add a /name= option so that you can make the user's friendly name appear
2348 TF/03 Added the control = fakedefer ACL modifier.
2350 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
2351 Mark Lowes for thorough testing.
2353 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
2355 TK/03 Merged latest SRS patch from Miles Wilton.
2357 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
2358 with the definition in sysexits.h (which is #included earlier).
2359 Fortunately, Exim does not actually use EX_OK. The code used to try to
2360 preserve the sysexits.h value, by assumimg that macro definitions were
2361 scanned for macro replacements. I have been disabused of this notion,
2362 so now the code just undefines EX_OK before #including unistd.h.
2364 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
2365 in the smtp transport. When a block could not be written in a single
2366 write() function, the timeout was being re-applied to each part-write.
2367 This seems wrong - if the receiver was accepting one byte at a time it
2368 would take for ever. The timeout is now adjusted when this happens. It
2369 doesn't have to be particularly precise.
2371 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
2372 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
2374 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
2375 without a domain if the check to <postmaster@domain> fails.
2377 SC/01 Eximstats: added -xls and the ability to specify output files
2378 (patch written by Frank Heydlauf).
2380 SC/02 Eximstats: use FileHandles for outputing results.
2382 SC/03 Eximstats: allow any combination of xls, txt, and html output.
2384 SC/04 Eximstats: fixed display of large numbers with -nvr option
2386 SC/05 Eximstats: fixed merging of reports with empty tables.
2388 SC/06 Eximstats: added the -include_original_destination flag
2390 SC/07 Eximstats: removed tabs and trailing whitespace.
2392 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
2394 TK/06 MBOX spool code: Add real "From " MBOX separator line
2395 so the .eml file is really in mbox format (even though
2396 most programs do not really care). Patch from Alex Miller.
2398 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
2399 The latter is generated from $received_to and is only set if the
2400 message has one envelope recipient. SA can use these headers,
2401 obviously out-of-the-box. Patch from Alex Miller.
2403 PH/08 The ${def test on a variable was returning false if the variable's
2404 value was "0", contrary to what the specification has always said!
2405 The result should be true unless the variable is empty.
2407 PH/09 The syntax error of a character other than { following "${if
2408 def:variable_name" (after optional whitespace) was not being diagnosed.
2409 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
2410 accidental colon was present, for example, could give incorrect results.
2412 PH/10 Tidied the code in a number of places where the st_size field of a stat()
2413 result is used (not including appendfile, where other changes are about
2416 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
2417 This involved changing a lot of size variables from int to off_t. It
2418 should work with maildirs and everything.
2420 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
2421 spamd dying while we are connected to it.
2423 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
2426 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
2427 function for BDB 4.3.
2429 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
2431 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
2432 the sources for PCRE have been re-arranged and more files are now
2435 PH/15 The code I had for printing potentially long long variables in PH/11
2436 above was not the best (it lost precision). The length of off_t variables
2437 is now inspected at build time, and an appropriate printing format (%ld
2438 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
2439 to be "long long int" or "long int". This is needed for the internal
2440 formatting function string_vformat().
2442 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
2443 the configuration file to be ":syslog", then the script "guesses" where
2444 the logs files are, rather than using the compiled in default. In our
2445 case the guess is not the same as the compiled default, so the script
2446 suddenly stopped working when I started to use syslog. The patch checks
2447 to see if log_file_path is "". If so, it attempts to read it from exim
2448 with no configuration file to get the compiled in version, before it
2449 falls back to the previous guessing code."
2451 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
2452 implementing BATV in an Exim configuration. See NewStuff for the gory
2455 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
2456 Makefile that are specific to HP-UX.
2458 PH/18 If the "use_postmaster" option was set for a recipient callout together
2459 with the "random" option, the postmaster address was used as the MAIL
2460 FROM address for the random test, but not for the subsequent recipient
2461 test. It is now used for both.
2463 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
2464 patch removes a few documentation additions to RFC 3028, because the
2465 latest draft now contains them. It adds the new en;ascii-case comparator
2466 and a new error check for 8bit text in MIME parts. Comparator and
2467 require names are now matched exactly. I enabled the subaddress
2468 extension, but it is not well tested yet (read: it works for me)."
2470 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
2471 rework some of the code of TK/09 above to avoid the hardwired use of
2472 "%lld" and "long long". Replaced the call to snprintf() with a call to
2475 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
2476 records point to non-existent hosts", "retry timeout exceeded", and
2477 "retry time not reached for any host after a long failure period".
2479 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
2480 experimental DomainKeys support:
2482 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
2483 (2) On an error such as an illegally used "control", the wrong name for
2484 the control was given.
2486 These problems did NOT occur unless DomainKeys support was compiled.
2488 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
2490 PH/24 Added ${if match_ip condition.
2492 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
2493 databases so that it will be absolutely obvious if a crash occurs in the
2494 DB library. This is a regular occurrence (often caused by mis-matched
2497 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
2498 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
2499 picked up on a user's system that detects such things. There doesn't seem
2500 to be a gcc warning option for this - only an attribute that has to be
2501 put on the function's prototype. It seems that in Fedora Core 4 they have
2502 set this on a number of new functions. No doubt there will be more in due
2505 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
2506 specify a transport. However, if an address that was verified by such a
2507 router was the subject of a callout, Exim crashed because it tried to
2508 read the rcpt_include_affixes from the non-existent transport. Now it
2509 just assumes that the setting of that option is false. This bug was
2510 introduced by 4.51/PH/31.
2512 PH/28 Changed -d+all to exclude +memory, because that information is very
2513 rarely of interest, but it makes the output a lot bigger. People tend to
2514 do -d+all out of habit.
2516 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
2517 code in os-type was giving problems when libc.so lives in lib64, like on
2520 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
2521 aren't the modern standard, and it seems that some systems' include files
2522 don't always have them. Exim was already checking for some of the newer
2523 ones like T_AAAA, and defining it itself. I've added checks for all the
2524 record types that Exim uses.
2526 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
2527 not automatically generating a new one, as it is supposed to. This
2528 prevented TLS from working. If the file did exist, but contained invalid
2529 data, a new version was generated, as expected. It was only the case of a
2530 non-existent file that was broken.
2532 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
2533 with a change in libdomainkeys > 0.64.
2535 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
2536 from DNS. If the selector record carries the flag, it now has
2537 precedence over the domain-wide flag.
2539 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
2541 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
2542 the use of a port name that isn't defined in /etc/services) occurred, the
2543 message was deferred as in a normal delivery, and thus remained on the
2544 spool, instead of being failed because of the mua_wrapper setting. This
2545 is now fixed, and I tidied up some of the mua_wrapper messages at the
2548 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
2549 the messages in a hash of arrays rather than using individual hashes.
2550 This is a bit cleaner and results in dramatic memory savings, albeit
2551 at a slight CPU cost.
2553 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
2554 as requested by Marc Sherman.
2556 SC/10 Eximstats: added histograms for user specified patterns as requested
2559 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
2561 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
2562 fopen() in the content-scanning modules that did not already have it.
2568 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
2569 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
2571 TK/02 Fix ACL "control" statement not being available in MIME ACL.
2573 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
2575 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
2576 to test Sieve filters that use "vacation".
2578 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
2579 that changes the way the GnuTLS parameters are stored in the cache file.
2580 The new format can be generated externally. For backward compatibility,
2581 if the data in the cache doesn't make sense, Exim assumes it has read an
2582 old-format file, and it generates new data and writes a new file. This
2583 means that you can't go back to an older release without removing the
2586 PH/03 A redirect router that has both "unseen" and "one_time" set does not
2587 work if there are any delivery delays because "one_time" forces the
2588 parent to be marked "delivered", so its unseen clone is never tried
2589 again. For this reason, Exim now forbids the simultaneous setting of
2592 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
2593 redirected to themselves ("homonym" addresses). Read the long ChangeLog
2594 entry if you want to know the details. The fix, however, neglected to
2595 consider the case when local delivery batching is involved. The test for
2596 "previously delivered" was not happening when checking to see if an
2597 address could be batched with a previous (undelivered) one; under
2598 certain circumstances this could lead to multiple deliveries to the same
2601 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
2602 in its include files, and this causes problems building Exim.
2604 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
2605 header_syntax) but Exim was just ignoring anything given after a slash.
2606 In particular, this caused confusion with an attempt to use "verify =
2607 reverse_host_lookup/defer_ok". An error is now given when options are
2608 supplied for verify items that do not have them. (Maybe reverse_host_
2609 lookup should have a defer_ok option, but that's a different point.)
2611 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
2612 defined by RFC 821) to 2048, because there were problems with some AUTH
2613 commands, and RFC 1869 says the size should be increased for extended
2614 SMTP commands that take arguments.
2616 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
2619 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
2620 "unknown" error; now it says that the functionality isn't in the binary.
2622 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
2623 an address' error message when a string expansion fails (syntax or
2624 whatever). Otherwise the password may appear in the log. Following change
2625 PH/42 below, there is no longer a chance of it appearing in a bounce
2628 PH/11 Installed exipick version 20050225.0 from John Jetmore.
2630 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
2631 of its addresses was ever tried. (Bugzilla bug #2.)
2633 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
2634 the result incorrectly in the debug output. (It correctly added a newline
2635 to what was transported.)
2637 TF/01 Added $received_time.
2639 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
2640 commented out examples of how to interface to a virus scanner and to
2641 SpamAssassin. Also added commented examples of av_scanner and
2642 spamd_address settings.
2644 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
2645 and controls are allowed in which ACLs. There were a couple of minor
2646 errors. Some of the entries in the conditions table (which is a table of
2647 where they are NOT allowed) were getting very unwieldy; rewrote them as a
2648 negation of where the condition IS allowed.
2650 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
2652 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
2653 header file does not have a version number, so I've had to invent a new
2654 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
2655 API. The code is untested by me (my Linux distribution still has 0.3.2 of
2656 radiusclient), but it was contributed by a Radius user.
2658 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
2659 files or directories, for OpenSSL.
2661 PH/19 When an Exim process that is running as root has to create an Exim log
2662 file, it does so in a subprocess that runs as exim:exim so as to get the
2663 ownership right at creation (otherwise, other Exim processes might see
2664 the file with the wrong ownership). There was no test for failure of this
2665 fork() call, which would lead to the process getting stuck as it waited
2666 for a non-existent subprocess. Forks do occasionally fail when resources
2667 run out. I reviewed all the other calls to fork(); they all seem to check
2670 PH/20 When checking for unexpected SMTP input at connect time (before writing
2671 the banner), Exim was not dealing correctly with a non-positive return
2672 from the read() function. If the client had disconnected by this time,
2673 the result was a log entry for a synchronization error with an empty
2674 string after "input=" when read() returned zero. If read() returned -1
2675 (an event I could not check), uninitialized data bytes were printed.
2676 There were reports of junk text (parts of files, etc) appearing after
2679 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
2681 PH/22 Added support for macro redefinition, and (re)definition in between
2682 driver and ACL definitions.
2684 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
2685 forgetting to use the resulting value; it was using the unexpanded value.
2687 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
2688 hadn't been configured. The fix is from Juergen Kreileder, who
2689 understands it better than I do:
2691 "Here's what I see happening with three configured cyrus_sasl
2692 authenticators configured (plain, login, cram-md5):
2694 On startup auth_cyrus_sasl_init() gets called for each of these.
2695 This means three calls to sasl_listmech() without a specified mech_list.
2696 => SASL tests which mechs of all available mechs actually work
2697 => three warnings about OTP not working
2698 => the returned list contains: plain, login, cram-md5, digest-md5, ...
2700 With the patch, sasl_listmech() also gets called three times. But now
2701 SASL's mech_list option is set to the server_mech specified in the the
2702 authenticator. Or in other words, the answer from sasl_listmech()
2703 gets limited to just the mech you're testing for (which is different
2705 => the return list contains just 'plain' or 'login', 'cram-md5' or
2706 nothing depending on the value of ob->server_mech.
2708 I've just tested the patch: Authentication still works fine,
2709 unavailable mechs specified in the exim configuration are still
2710 caught, and the auth.log warnings about OTP are gone."
2712 PH/25 When debugging is enabled, the contents of the command line are added
2713 to the debugging output, even when log_selector=+arguments is not
2716 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
2717 answer is "GNU", and only if the return is "GNU/something" is the answer
2720 PH/27 $acl_verify_message is now set immediately after the failure of a
2721 verification in an ACL, and so is available in subsequent modifiers. In
2722 particular, the message can be preserved by coding like this:
2724 warn !verify = sender
2725 set acl_m0 = $acl_verify_message
2727 Previously, $acl_verify_message was set only while expanding "message"
2728 and "log_message" when a very denied access.
2730 PH/28 Modified OS/os.c-Linux with
2732 -#ifndef OS_LOAD_AVERAGE
2733 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
2735 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
2736 nomenclature these days.)
2738 PH/29 Installed patch from the Sieve maintainer that adds the options
2739 sieve_useraddress and sieve_subaddress to the redirect router.
2741 PH/30 In these circumstances:
2742 . Two addresses routed to the same list of hosts;
2743 . First host does not offer TLS;
2744 . First host accepts first address;
2745 . First host gives temporary error to second address;
2746 . Second host offers TLS and a TLS session is established;
2747 . Second host accepts second address.
2748 Exim incorrectly logged both deliveries with the TLS parameters (cipher
2749 and peerdn, if requested) that were in fact used only for the second
2752 PH/31 When doing a callout as part of verifying an address, Exim was not paying
2753 attention to any local part prefix or suffix that was matched by the
2754 router that accepted the address. It now behaves in the same way as it
2755 does for delivery: the affixes are removed from the local part unless
2756 rcpt_include_affixes is set on the transport.
2758 PH/32 Add the sender address, as F=<...>, to the log line when logging a
2759 timeout during the DATA phase of an incoming message.
2761 PH/33 Sieve envelope tests were broken for match types other than :is. I have
2762 applied a patch sanctioned by the Sieve maintainer.
2764 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
2765 the uid or gid is negative. A case of a negative gid caused this to be
2766 noticed. The fix allows for either to be negative.
2768 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
2769 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
2772 PH/36 The change PH/12 above was broken. Fixed it.
2774 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
2775 the grounds that routing the same address twice would always produce the
2776 same answer. This might have been true once, but it is certainly no
2777 longer true now. Routing a child address may depend on the previous
2778 routing that produced that child. Some complicated redirection strategies
2779 went wrong when messages had multiple recipients, and made Exim's
2780 behaviour dependent on the order in which the addresses were given.
2782 I have moved the duplicate checking until after the routing is complete.
2783 Exim scans the addresses that are assigned to local and remote
2784 transports, and removes any duplicates. This means that more work will be
2785 done, as duplicates will always all be routed, but duplicates are
2786 presumably rare, so I don't expect this is of any significance.
2788 For deliveries to pipes, files, and autoreplies, the duplicate checking
2789 still happens during the routing process, since they are not going to be
2792 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
2793 It corrects a timeout issue with spamd. This is Ian's comment: "The
2794 background is that sometimes spamd either never reads data from a
2795 connection it has accepted, or it never writes response data. The exiscan
2796 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
2797 blindly assumes that writes won't block so it may never time out."
2799 PH/39 Allow G after quota size as well as K and M.
2801 PH/40 The value set for $authenticated_id in an authenticator may not contain
2802 binary zeroes or newlines because the value is written to log lines and
2803 to spool files. There was no check on this. Now the value is run through
2804 the string_printing() function so that such characters are converted to
2805 printable escape sequences.
2807 PH/41 $message_linecount is a new variable that contains the total number of
2808 lines in the message. Compare $body_linecount, which is the count for the
2811 PH/42 Exim no longer gives details of delivery errors for specific addresses in
2812 bounce and delay warning messages, except in certain special cases, which
2815 (a) An SMTP error message from a remote host;
2816 (b) A message specified in a :fail: redirection;
2817 (c) A message specified in a "fail" command in a system filter;
2818 (d) A message specified in a FAIL return from the queryprogram router;
2819 (e) A message specified by the cannot_route_message router option.
2821 In these cases only, Exim does include the error details in bounce and
2822 warning messages. There are also a few cases where bland messages such
2823 as "unrouteable address" or "local delivery error" are given.
2825 PH/43 $value is now also set for the "else" part of a ${run expansion.
2827 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
2828 being worked on, but at least Exim now implements the latest version to
2831 PH/45 In a pipe transport, although a timeout while waiting for the pipe
2832 process to complete was treated as a delivery failure, a timeout while
2833 writing the message to the pipe was logged, but erroneously treated as a
2834 successful delivery. Such timeouts include transport filter timeouts. For
2835 consistency with the overall process timeout, these timeouts are now
2836 treated as errors, giving rise to delivery failures by default. However,
2837 there is now a new Boolean option for the pipe transport called
2838 timeout_defer, which, if set TRUE, converts the failures into defers for
2839 both kinds of timeout. A transport filter timeout is now identified in
2842 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
2843 systems where "make" and "gmake" are different, calling "gmake" at top
2844 level broke things. I've arranged for the value of $(MAKE) to be passed
2845 from the Makefile to this script so that it can call the same version of
2849 A note about Exim versions 4.44 and 4.50
2850 ----------------------------------------
2852 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
2853 changes of various kinds. As a consequence, a big documentation update was
2854 needed. This delayed the release for rather longer than seemed good, especially
2855 in the light of a couple of (minor) security issues. Therefore, the changes
2856 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
2857 release. So 4.44 and 4.50 are in effect two different branches that both start
2860 I have left the 4.50 change log unchanged; it contains all the changes since
2861 4.43. The change log for 4.44 is below; many of its items are identical to
2862 those for 4.50. This seems to be the most sensible way to preserve the
2863 historical information.
2869 1. Minor wording change to the doc/README.SIEVE file.
2871 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
2872 computation of the current number of files was incorrect.
2874 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
2875 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
2878 4. Give more explanation in the error message when the command for a transport
2879 filter fails to execute.
2881 5. There are several places where Exim runs a non-Exim command in a
2882 subprocess. The SIGUSR1 signal should be disabled for these processes. This
2883 was being done only for the command run by the queryprogram router. It is
2884 now done for all such subprocesses. The other cases are: ${run, transport
2885 filters, and the commands run by the lmtp and pipe transports.
2887 6. Added CONFIGURE_GROUP build-time option.
2889 7. Some older OS have a limit of 256 on the maximum number of file
2890 descriptors. Exim was using setrlimit() to set 1000 as a large value
2891 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
2892 systems. I've change it so that if it can't get 1000, it tries for 256.
2894 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
2895 was an oversight, and furthermore, ever since the addition of extra
2896 controls (e.g. 4.43/32), the checks on when to allow different forms of
2897 "control" were broken. There should now be diagnostics for all cases when a
2898 control that does not make sense is encountered.
2900 9. Added the /retain_sender option to "control=submission".
2902 10. $recipients is now available in the predata ACL (oversight).
2904 11. Tidy the search cache before the fork to do a delivery from a message
2905 received from the command line. Otherwise the child will trigger a lookup
2906 failure and thereby defer the delivery if it tries to use (for example) a
2907 cached ldap connection that the parent has called unbind on.
2909 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
2910 of $address_data from the recipient verification was clobbered by the
2911 sender verification.
2913 13. The value of address_data from a sender verification is now available in
2914 $sender_address_data in subsequent conditions in the ACL statement.
2916 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
2918 15. Added a new option "connect=<time>" to callout options, to set a different
2921 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
2922 was its contents. (It was OK if the option was not defined at all.)
2924 17. A "Completed" log line is now written for messages that are removed from
2925 the spool by the -Mrm option.
2927 18. New variables $sender_verify_failure and $recipient_verify_failure contain
2928 information about exactly what failed.
2930 19. Added -dd to debug only the daemon process.
2932 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
2933 handles timeouts, both on the server side and network timeouts. Renamed the
2934 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
2936 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
2937 It is now set to "smtps".
2939 22. $host_address is now set to the target address during the checking of
2940 ignore_target_hosts.
2942 23. When checking ignore_target_hosts for an ipliteral router, no host name was
2943 being passed; this would have caused $sender_host_name to have been used if
2944 matching the list had actually called for a host name (not very likely,
2945 since this list is usually IP addresses). A host name is now passed as
2948 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
2949 code that specifies a non-restarting handler (typically sigaction() in
2950 modern systems) in an attempt to fix a rare and obscure crash bug.
2952 25. Narrowed the window for a race in the daemon that could cause it to ignore
2953 SIGCHLD signals. This is not a major problem, because they are used only to
2954 wake it up if nothing else does.
2956 26. A malformed maildirsize file could cause Exim to calculate negative values
2957 for the mailbox size or file count. Odd effects could occur as a result.
2958 The maildirsize information is now recalculated if the size or filecount
2961 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
2962 support for a long time. Removed HAVE_SYS_VFS_H.
2964 28. Installed the latest version of exipick from John Jetmore.
2966 29. In an address list, if the pattern was not a regular expression, an empty
2967 subject address (from a bounce message) matched only if the pattern was an
2968 empty string. Non-empty patterns were not even tested. This was the wrong
2969 because it is perfectly reasonable to use an empty address as part of a
2970 database query. An empty address is now tested by patterns that are
2971 lookups. However, all the other forms of pattern expect the subject to
2972 contain a local part and a domain, and therefore, for them, an empty
2973 address still always fails if the pattern is not itself empty.
2975 30. Exim went into a mad DNS loop when attempting to do a callout where the
2976 host was specified on an smtp transport, and looking it up yielded more
2977 than one IP address.
2979 31. Re-factored the code for checking spool and log partition space into a
2980 function that finds that data and another that does the check. The former
2981 is then used to implement four new variables: $spool_space, $log_space,
2982 $spool_inodes, and $log_inodes.
2984 32. The RFC2047 encoding function was originally intended for short strings
2985 such as real names; it was not keeping to the 75-character limit for
2986 encoded words that the RFC imposes. It now respects the limit, and
2987 generates multiple encoded words if necessary. To be on the safe side, I
2988 have increased the buffer size for the ${rfc2047: expansion operator from
2991 33. It is now permitted to omit both strings after an "if" condition; if the
2992 condition is true, the result is "true". As before, when the second string
2993 is omitted, a false condition yields an empty string. This makes it less
2994 cumbersome to write custom ACL and router conditions.
2996 34. Failure to deliver a bounce message always caused it to be frozen, even if
2997 there was an errors_to setting on the router. The errors_to setting is now
3000 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
3001 canonical form (fully expanded) before being placed in
3002 $sender_host_address.
3004 36. The table in the code that translates DNS record types into text (T_A to
3005 "A" for instance) was missing entries for NS and CNAME. It is just possible
3006 that this could have caused confusion if both these types were looked up
3007 for the same domain, because the text type is used as part of Exim's
3008 per-process caching. But the chance of anyone hitting this buglet seems
3011 37. The dnsdb lookup has been extended in a number of ways.
3013 (1) There is a new type, "zns", which walks up the domain tree until it
3014 finds some nameserver records. It should be used with care.
3016 (2) There is a new type, "mxh", which is like "mx" except that it returns
3017 just the host names, not the priorities.
3019 (3) It is now possible to give a list of domains (or IP addresses) to be
3020 looked up. The behaviour when one of the lookups defers can be
3021 controlled by a keyword.
3023 (4) It is now possible to specify the separator character for use when
3024 multiple records are returned.
3026 38. The dnslists ACL condition has been extended: it is now possible to supply
3027 a list of IP addresses and/or domains to be looked up in a particular DNS
3030 39. Added log_selector=+queue_time_overall.
3032 40. When running the queue in the test harness, wait just a tad after forking a
3033 delivery process, to get repeatability of debugging output.
3035 41. Include certificate and key file names in error message when GnuTLS fails
3036 to set them up, because the GnuTLS error message doesn't include the name
3037 of the failing file when there is a problem reading it.
3039 42. Allow both -bf and -bF in the same test run.
3041 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
3043 44. The "Exiscan patch" is now merged into the mainline Exim source.
3045 45. Sometimes the final signoff response after QUIT could fail to get
3046 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
3047 < 0 before doing a fflush(). This bug looks as though it goes back to the
3048 introduction of TLS in release 3.20, but "sometimes" must have been rare
3049 because the tests only now provoked it.
3051 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
3052 (this can affect the format of dates).
3054 47. exim_tidydb, when checking for the continued existence of a message for
3055 which it has found a message-specific retry record, was not finding
3056 messages that were in split spool directories. Consequently, it was
3057 deleting retry records that should have stayed in existence.
3059 48. Steve fixed some bugs in eximstats.
3061 49. The SPA authentication driver was not abandoning authentication and moving
3062 on to the next authenticator when an expansion was forced to fail,
3063 contradicting the general specification for all authenticators. Instead it
3064 was generating a temporary error. It now behaves as specified.
3066 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
3067 (the order specifies the preference for clients). The order is now AES256,
3068 AES128, 3DES, ARCFOUR128.
3070 51. Small patch to Sieve code - explicitly set From: when generating an
3073 52. Exim crashed if a remote delivery caused a very long error message to be
3074 recorded - for instance if somebody sent an entire SpamAssassin report back
3075 as a large number of 550 error lines. This bug was coincidentally fixed by
3076 increasing the size of one of Exim's internal buffers (big_buffer) that
3077 happened as part of the Exiscan merge. However, to be on the safe side, I
3078 have made the code more robust (and fixed the comments that describe what
3081 53. Now that there can be additional text after "Completed" in log lines (if
3082 the queue_time_overall log selector is set), a one-byte patch to exigrep
3083 was needed to allow it to recognize "Completed" as not the last thing in
3086 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
3087 patch that reportedly fixes this has been added. I am not expert enough to
3088 create a test for it. This is what the patch creator wrote:
3090 "I found a little strange behaviour of ldap code when working with
3091 Windows 2003 AD Domain, where users was placed in more than one
3092 Organization Units. When I tried to give exim partial DN, the exit code
3093 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
3094 But simultaneously result of request was absolutely normal ldap result,
3095 so I produce this patch..."
3097 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
3098 REFERENCE, so I have modified the code to exclude the patch when that macro
3101 55. Some experimental protocols are using DNS PTR records for new purposes. The
3102 keys for these records are domain names, not reversed IP addresses. The
3103 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
3104 leaves it alone. Component reversal etc. now happens only for IP addresses.
3107 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
3109 57. Double the size of the debug message buffer (to 2048) so that more of very
3110 long debug lines gets shown.
3112 58. The exicyclog utility now does better if the number of log files to keep
3113 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
3115 59. Two changes related to the smtp_active_hostname option:
3117 (1) $smtp_active_hostname is now available as a variable.
3118 (2) The default for smtp_banner uses $smtp_active_hostname instead
3119 of $primary_hostname.
3121 60. The host_aton() function is supposed to be passed a string that is known
3122 to be a valid IP address. However, in the case of IPv6 addresses, it was
3123 not checking this. This is a hostage to fortune. Exim now panics and dies
3124 if the condition is not met. A case was found where this could be provoked
3125 from a dnsdb PTR lookup with an IPv6 address that had more than 8
3126 components; fortuitously, this particular loophole had already been fixed
3127 by change 4.50/55 above.
3129 If there are any other similar loopholes, the new check in host_aton()
3130 itself should stop them being exploited. The report I received stated that
3131 data on the command line could provoke the exploit when Exim was running as
3132 exim, but did not say which command line option was involved. All I could
3133 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
3134 running as the user.
3137 61. There was a buffer overflow vulnerability in the SPA authentication code
3138 (which came originally from the Samba project). I have added a test to the
3139 spa_base64_to_bits() function which I hope fixes it.
3142 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
3143 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
3145 63. The daemon start-up calls getloadavg() while still root for those OS that
3146 need the first call to be done as root, but it missed one case: when
3147 deliver_queue_load_max is set with deliver_drop_privilege. This is
3148 necessary for the benefit of the queue runner, because there is no re-exec
3149 when deliver_drop_privilege is set.
3151 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
3152 This has been fixed.
3154 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
3155 was in use, was not putting the data itself into the right store pool;
3156 consequently, it could be overwritten for a subsequent message in the same
3157 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
3160 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
3162 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
3163 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
3164 the function were treating the return as a boolean value, which happened to
3165 work because 0=false and not-0=true, but is not correct code.
3167 68. The host_aton() function was not handling scoped IPv6 addresses (those
3168 with, for example, "%eth0" on the end) correctly.
3170 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
3171 negated items (that is, ~something) in unsigned ints. Some compilers
3172 apparently mutter when there is no cast.
3174 70. If an address verification called from an ACL failed, and did not produce a
3175 user-specific message (i.e. there was only a "system" message), nothing was
3176 put in $acl_verify_message. In this situation, it now puts the system
3179 71. Change 4.23/11 added synchronization checking at the start of an SMTP
3180 session; change 4.31/43 added the unwanted input to the log line - except
3181 that it did not do this in the start of session case. It now does.
3183 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
3184 This is wrong and can cause the other end to generate a synchronization
3185 error if it is another Exim or anything else that does the synchronization
3186 check. A QUIT command is no longer sent after a timeout.
3188 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
3189 during host lookups.
3191 74. The defer_ok option of callout verification was not working if it was used
3192 when verifying addresses in header lines, that is, for this case:
3194 verify = header_sender/callout=defer_ok
3196 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
3197 those file descriptors could be used for SMTP connections. If anything
3198 wrote to stderr (the example that came up was "warn" in embedded Perl), it
3199 could be sent to the SMTP client, causing chaos. The daemon now opens
3200 stdin, stdout, and stderr to /dev/null when it puts itself into the
3203 76. Arrange for output from Perl's "warn" command to be written to Exim's main
3204 log by default. The user can override this with suitable Perl magic.
3206 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
3207 the log message when discard triggers, was not working for the DATA ACL or
3208 for the non-SMTP ACL.
3210 78. Error message wording change in sieve.c.
3212 79. If smtp_accept_max_per_host was set, the number of connections could be
3213 restricted to fewer than expected, because the daemon was trying to set up
3214 a new connection before checking whether the processes handling previous
3215 connections had finished. The check for completed processes is now done
3216 earlier. On busy systems, this bug wouldn't be noticed because something
3217 else would have woken the daemon, and it would have reaped the completed
3220 80. If a message was submitted locally by a user whose login name contained one
3221 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
3222 It caused a spool format error. I have fixed the spool reading code. A
3223 related problem was that the "from" clause in the Received: line became
3224 illegal because of the space(s). It is now covered by ${quote_local_part.
3226 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
3229 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
3231 83. Patch from Sieve maintainer to fix unterminated string problem in
3232 "vacation" handling.
3234 84. Some minor changes to the Linux configuration files to help with other
3235 OS variants using glibc.
3237 85. One more patch for Sieve to update vacation handling to latest spec.
3240 ----------------------------------------------------
3241 See the note above about the 4.44 and 4.50 releases.
3242 ----------------------------------------------------
3248 1. Change 4.43/35 introduced a bug that caused file counts to be
3249 incorrectly computed when quota_filecount was set in an appendfile
3252 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
3253 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
3256 3. Give more explanation in the error message when the command for a transport
3257 filter fails to execute.
3259 4. There are several places where Exim runs a non-Exim command in a
3260 subprocess. The SIGUSR1 signal should be disabled for these processes. This
3261 was being done only for the command run by the queryprogram router. It is
3262 now done for all such subprocesses. The other cases are: ${run, transport
3263 filters, and the commands run by the lmtp and pipe transports.
3265 5. Some older OS have a limit of 256 on the maximum number of file
3266 descriptors. Exim was using setrlimit() to set 1000 as a large value
3267 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
3268 systems. I've change it so that if it can't get 1000, it tries for 256.
3270 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
3271 was an oversight, and furthermore, ever since the addition of extra
3272 controls (e.g. 4.43/32), the checks on when to allow different forms of
3273 "control" were broken. There should now be diagnostics for all cases when a
3274 control that does not make sense is encountered.
3276 7. $recipients is now available in the predata ACL (oversight).
3278 8. Tidy the search cache before the fork to do a delivery from a message
3279 received from the command line. Otherwise the child will trigger a lookup
3280 failure and thereby defer the delivery if it tries to use (for example) a
3281 cached ldap connection that the parent has called unbind on.
3283 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
3284 of $address_data from the recipient verification was clobbered by the
3285 sender verification.
3287 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
3288 was its contents. (It was OK if the option was not defined at all.)
3290 11. A "Completed" log line is now written for messages that are removed from
3291 the spool by the -Mrm option.
3293 12. $host_address is now set to the target address during the checking of
3294 ignore_target_hosts.
3296 13. When checking ignore_target_hosts for an ipliteral router, no host name was
3297 being passed; this would have caused $sender_host_name to have been used if
3298 matching the list had actually called for a host name (not very likely,
3299 since this list is usually IP addresses). A host name is now passed as
3302 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
3303 code that specifies a non-restarting handler (typically sigaction() in
3304 modern systems) in an attempt to fix a rare and obscure crash bug.
3306 15. Narrowed the window for a race in the daemon that could cause it to ignore
3307 SIGCHLD signals. This is not a major problem, because they are used only to
3308 wake it up if nothing else does.
3310 16. A malformed maildirsize file could cause Exim to calculate negative values
3311 for the mailbox size or file count. Odd effects could occur as a result.
3312 The maildirsize information is now recalculated if the size or filecount
3315 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
3316 support for a long time. Removed HAVE_SYS_VFS_H.
3318 18. Updated exipick to current release from John Jetmore.
3320 19. Allow an empty sender to be matched against a lookup in an address list.
3321 Previously the only cases considered were a regular expression, or an
3324 20. Exim went into a mad DNS lookup loop when doing a callout where the
3325 host was specified on the transport, if the DNS lookup yielded more than
3328 21. The RFC2047 encoding function was originally intended for short strings
3329 such as real names; it was not keeping to the 75-character limit for
3330 encoded words that the RFC imposes. It now respects the limit, and
3331 generates multiple encoded words if necessary. To be on the safe side, I
3332 have increased the buffer size for the ${rfc2047: expansion operator from
3335 22. Failure to deliver a bounce message always caused it to be frozen, even if
3336 there was an errors_to setting on the router. The errors_to setting is now
3339 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
3340 canonical form (fully expanded) before being placed in
3341 $sender_host_address.
3343 24. Updated eximstats to version 1.33
3345 25. Include certificate and key file names in error message when GnuTLS fails
3346 to set them up, because the GnuTLS error message doesn't include the name
3347 of the failing file when there is a problem reading it.
3349 26. Expand error message when OpenSSL has problems setting up cert/key files.
3352 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
3353 (this can affect the format of dates).
3355 28. exim_tidydb, when checking for the continued existence of a message for
3356 which it has found a message-specific retry record, was not finding
3357 messages that were in split spool directories. Consequently, it was
3358 deleting retry records that should have stayed in existence.
3360 29. eximstats updated to version 1.35
3361 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
3362 1.35 - bugfix such that pie charts by volume are generated correctly
3364 30. The SPA authentication driver was not abandoning authentication and moving
3365 on to the next authenticator when an expansion was forced to fail,
3366 contradicting the general specification for all authenticators. Instead it
3367 was generating a temporary error. It now behaves as specified.
3369 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
3370 (the order specifies the preference for clients). The order is now AES256,
3371 AES128, 3DES, ARCFOUR128.
3373 31. Small patch to Sieve code - explicitly set From: when generating an
3376 32. Exim crashed if a remote delivery caused a very long error message to be
3377 recorded - for instance if somebody sent an entire SpamAssassin report back
3378 as a large number of 550 error lines. This bug was coincidentally fixed by
3379 increasing the size of one of Exim's internal buffers (big_buffer) that
3380 happened as part of the Exiscan merge. However, to be on the safe side, I
3381 have made the code more robust (and fixed the comments that describe what
3384 33. Some experimental protocols are using DNS PTR records for new purposes. The
3385 keys for these records are domain names, not reversed IP addresses. The
3386 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
3387 leaves it alone. Component reversal etc. now happens only for IP addresses.
3390 34. The host_aton() function is supposed to be passed a string that is known
3391 to be a valid IP address. However, in the case of IPv6 addresses, it was
3392 not checking this. This is a hostage to fortune. Exim now panics and dies
3393 if the condition is not met. A case was found where this could be provoked
3394 from a dnsdb PTR lookup with an IPv6 address that had more than 8
3395 components; fortuitously, this particular loophole had already been fixed
3396 by change 4.50/55 or 4.44/33 above.
3398 If there are any other similar loopholes, the new check in host_aton()
3399 itself should stop them being exploited. The report I received stated that
3400 data on the command line could provoke the exploit when Exim was running as
3401 exim, but did not say which command line option was involved. All I could
3402 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
3403 running as the user.
3406 35. There was a buffer overflow vulnerability in the SPA authentication code
3407 (which came originally from the Samba project). I have added a test to the
3408 spa_base64_to_bits() function which I hope fixes it.
3411 36. The daemon start-up calls getloadavg() while still root for those OS that
3412 need the first call to be done as root, but it missed one case: when
3413 deliver_queue_load_max is set with deliver_drop_privilege. This is
3414 necessary for the benefit of the queue runner, because there is no re-exec
3415 when deliver_drop_privilege is set.
3417 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
3418 was in use, was not putting the data itself into the right store pool;
3419 consequently, it could be overwritten for a subsequent message in the same
3420 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
3423 38. Sometimes the final signoff response after QUIT could fail to get
3424 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
3425 < 0 before doing a fflush(). This bug looks as though it goes back to the
3426 introduction of TLS in release 3.20, but "sometimes" must have been rare
3427 because the tests only now provoked it.
3433 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
3434 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
3435 Now it returns an integer. A number of calls were still expecting a T/F
3436 return. Fortuitously, in all cases, the tests worked in OK situations,
3437 which is the norm. However, things would have gone wrong on any write
3438 failures on the smtp file descriptor. This function is used when sending
3439 messages over SMTP and also when doing verify callouts.
3441 2. When Exim is called to do synchronous delivery of a locally submitted
3442 message (the -odf or -odi options), it no longer closes stderr before doing
3445 3. Implemented the mua_wrapper option.
3447 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
3449 5. Implemented the functions header_remove(), header_testname(),
3450 header_add_at_position(), and receive_remove_recipient(), and exported them
3453 6. If an ACL "warn" statement specified the addition of headers, Exim already
3454 inserted X-ACL-Warn: at the start if there was no header name. However, it
3455 was not making this test for the second and subsequent header lines if
3456 there were newlines in the string. This meant that an invalid header could
3457 be inserted if Exim was badly configured.
3459 7. Allow an ACL "warn" statement to add header lines at the start or after all
3460 the Received: headers, as well as at the end.
3462 8. Added the rcpt_4xx retry error code.
3464 9. Added postmaster_mailfrom=xxx to callout verification option.
3466 10. Added mailfrom=xxxx to the callout verification option, for verify=
3469 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
3470 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
3472 12. Inserted some casts to stop certain compilers warning when using pointer
3473 differences as field lengths or precisions in printf-type calls (mostly
3474 affecting debugging statements).
3476 13. Added optional readline() support for -be (dynamically loaded).
3478 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
3479 same clock tick as a message's arrival, so that its received time was the
3480 same as the "first fail" time on the retry record, and that message
3481 remained on the queue past the ultimate address timeout, every queue runner
3482 would try a delivery (because it was past the ultimate address timeout) but
3483 after another failure, the ultimate address timeout, which should have then
3484 bounced the address, did not kick in. This was a "< instead of <=" error;
3485 in most cases the first failure would have been in the next clock tick
3486 after the received time, and all would be well.
3488 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
3489 being recognized when the domain list was tested by the match_domain
3490 condition in an expansion string.
3492 16. Added the ${str2b64: operator.
3494 17. Exim was always calling setrlimit() to set a large limit for the number of
3495 processes, without checking whether the existing limit was already
3496 adequate. (It did check for the limit on file descriptors.) Furthermore,
3497 errors from getrlimit() and setrlimit() were being ignored. Now they are
3498 logged to the main and panic logs, but Exim does carry on, to try to do its
3499 job under whatever limits there are.
3501 18. Imported PCRE 5.0.
3503 19. Trivial typo in log message " temporarily refused connection" (the leading
3506 20. If the log selector return_path_on_delivery was set and an address was
3507 redirected to /dev/null, the delivery process crashed because it assumed
3508 that a return path would always be set for a "successful" delivery. In this
3509 case, the whole delivery is bypassed as an optimization, and therefore no
3512 21. Internal re-arrangement: the function for sending a challenge and reading
3513 a response while authentication was assuming a zero-terminated challenge
3514 string. It's now changed to take a pointer and a length, to allow for
3515 binary data in such strings.
3517 22. Added the cyrus_sasl authenticator (code supplied by MBM).
3519 23. Exim was not respecting finduser_retries when seeking the login of the
3520 uid under which it was called; it was always trying 10 times. (The default
3521 setting of finduser_retries is zero.) Also, it was sleeping after the final
3522 failure, which is pointless.
3524 24. Implemented tls_on_connect_ports.
3526 25. Implemented acl_smtp_predata.
3528 26. If the domain in control=submission is set empty, Exim assumes that the
3529 authenticated id is a complete email address when it generates From: or
3530 Sender: header lines.
3532 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
3533 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
3534 chown and chgrp in /bin and hostname in /usr/bin.
3536 28. Exim was keeping the "process log" file open after each use, just as it
3537 does for the main log. This opens the possibility of it remaining open for
3538 long periods when the USR1 signal hits a daemon. Occasional processlog
3539 errors were reported, that could have been caused by this. Anyway, it seems
3540 much more sensible not to leave this file open at all, so that is what now
3543 29. The long-running daemon process does not normally write to the log once it
3544 has entered its main loop, and it closes the log before doing so. This is
3545 so that log files can straightforwardly be renamed and moved. However,
3546 there are a couple of unusual error situations where the daemon does write
3547 log entries, and I had neglected to close the log afterwards.
3549 30. The text of an SMTP error response that was received during a remote
3550 delivery was being truncated at 512 bytes. This is too short for some of
3551 the long messages that one sometimes sees. I've increased the limit to
3554 31. It is now possible to make retry rules that apply only when a message has a
3555 specific sender, in particular, an empty sender.
3557 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
3558 it possible to be selective about when SMTP synchronization is enforced.
3560 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
3562 32. Implemented hosts_connection_nolog.
3564 33. Added an ACL for QUIT.
3566 34. Setting "delay_warning=" to disable warnings was not working; it gave a
3569 35. Added mailbox_size and mailbox_filecount to appendfile.
3571 36. Added control = no_multiline_responses to ACLs.
3573 37. There was a bug in the logic of the code that waits for the clock to tick
3574 in the case where the clock went backwards by a substantial amount such
3575 that the microsecond fraction of "now" was more than the microsecond
3576 fraction of "then" (but the whole seconds number was less).
3578 38. Added support for the libradius Radius client library this is found on
3579 FreeBSD (previously only the radiusclient library was supported).
3585 1. When certain lookups returned multiple values in the form name=value, the
3586 quoting of the values was not always being done properly. Specifically:
3587 (a) If the value started with a double quote, but contained no whitespace,
3589 (b) If the value contained whitespace other than a space character (i.e.
3590 tabs or newlines or carriage returns) it was not quoted.
3591 This fix has been applied to the mysql and pgsql lookups by writing a
3592 separate quoting function and calling it from the lookup code. The fix
3593 should probably also be applied to nisplus, ibase and oracle lookups, but
3594 since I cannot test any of those, I have not disturbed their existing code.
3596 2. A hit in the callout cache for a specific address caused a log line with no
3597 reason for rejecting RCPT. Now it says "Previous (cached) callout
3598 verification failure".
3600 3. There was an off-by-one bug in the queryprogram router. An over-long
3601 return line was truncated at 256 instead of 255 characters, thereby
3602 overflowing its buffer with the terminating zero. As well as fixing this, I
3603 have increased the buffer size to 1024 (and made a note to document this).
3605 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
3606 when Exim is waiting for an SMTP response from a remote server, Exim
3607 restarts its select() call on the socket, thereby resetting its timeout.
3608 This is not a problem when such interrupts are rare. Somebody set up a cron
3609 job to run exiwhat every 2 minutes, which is less than the normal select()
3610 timeout (5 or 10 minutes). This meant that the select() timeout never
3611 kicked in because it was always reset. I have fixed this by comparing the
3612 time when an interrupt arrives with the time at the start of the first call
3613 to select(). If more time than the timeout has elapsed, the interrupt is
3614 treated as a timeout.
3616 5. Some internal re-factoring in preparation for the addition of Sieve
3617 extensions (by MH). In particular, the "personal" test is moved to a
3618 separate function, and given an option for scanning Cc: and Bcc: (which is
3619 not set for Exim filters).
3621 6. When Exim created an email address using the login of the caller as the
3622 local part (e.g. when creating a From: or Sender: header line), it was not
3623 quoting the local part when it contained special characters such as @.
3625 7. Installed new OpenBSD configuration files.
3627 8. Reworded some messages for syntax errors in "and" and "or" conditions to
3628 try to make them clearer.
3630 9. Callout options, other than the timeout value, were being ignored when
3631 verifying sender addresses in header lines. For example, when using
3633 verify = header_sender/callout=no_cache
3635 the cache was (incorrectly) being used.
3637 10. Added a missing instance of ${EXE} to the exim_install script; this affects
3638 only the Cygwin environment.
3640 11. When return_path_on_delivery was set as a log selector, if different remote
3641 addresses in the same message used different return paths and parallel
3642 remote delivery occurred, the wrong values would sometimes be logged.
3643 (Whenever a remote delivery process finished, the return path value from
3644 the most recently started remote delivery process was logged.)
3646 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
3647 lines when AUTH and/or TLS are in use. This is the "received protocol"
3648 field. Exim used to use "asmtp" for authenticated SMTP, without any
3649 indication (in the protocol name) for TLS use. Now it follows the RFC and
3650 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
3651 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
3652 names appear in log lines as well as in Received: header lines.
3654 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
3655 extensions, and comparison tests, and to fix some bugs.
3657 14. Changes to the "personal" filter test:
3659 (1) The test was buggy in that it was just doing the equivalent of
3660 "contains" tests on header lines. For example, if a user's address was
3661 anne@some.where, the "personal" test would incorrectly be true for
3663 To: susanne@some.where
3665 This test is now done by extracting each address from the header in turn,
3666 and checking the entire address. Other tests that are part of "personal"
3667 are now done using regular expressions (for example, to check local parts
3668 of addresses in From: header lines).
3670 (2) The list of non-personal local parts in From: addresses has been
3671 extended to include "listserv", "majordomo", "*-request", and "owner-*",
3672 taken from the Sieve specification recommendations.
3674 (3) If the message contains any header line starting with "List-" it is
3675 treated as non-personal.
3677 (4) The test for "circular" in the Subject: header line has been removed
3678 because it now seems ill-conceived.
3680 15. Minor typos in src/EDITME comments corrected.
3682 16. Installed latest exipick from John Jetmore.
3684 17. If headers_add on a router specified a text string that was too long for
3685 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
3686 of string_sprintf() is now avoided.
3688 18. $message_body_size was not set (it was always zero) when running the DATA
3689 ACL and the local_scan() function.
3691 19. For the "mail" command in an Exim filter, no default was being set for
3692 the once_repeat time, causing a random time value to be used if "once" was
3693 specified. (If the value happened to be <= 0, no repeat happened.) The
3694 default is now 0s, meaning "never repeat". The "vacation" command was OK
3695 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
3696 (I found it when inspecting the code).
3698 20. There is now an overall timeout for performing a callout verification. It
3699 defaults to 4 times the callout timeout, which applies to individual SMTP
3700 commands during the callout. The overall timeout applies when there is more
3701 than one host that can be tried. The timeout is checked before trying the
3702 next host. This prevents very long delays if there are a large number of
3703 hosts and all are timing out (e.g. when the network connections are timing
3704 out). The value of the overall timeout can be changed by specifying an
3705 additional sub-option for "callout", called "maxwait". For example:
3707 verify = sender/callout=5s,maxwait=20s
3709 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
3710 to the end before writing, but this should make it even safer).
3712 22. Exim was forgetting that it had advertised PIPELINING for the second and
3713 subsequent messages on an SMTP connection. It was also not resetting its
3714 memory on STARTTLS and an internal HELO.
3716 23. When Exim logs an SMTP synchronization error within a session, it now
3717 records whether PIPELINING has been advertised or not.
3719 24. Added 3 instances of "(long int)" casts to time_t variables that were being
3720 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
3721 rather than long int.
3723 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
3725 26. Added the never_mail option to autoreply.
3731 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
3732 crash if the getsockname() call failed; this can happen if a connection is
3733 closed very soon after it is established. The problem was simply in the
3734 order in which certain operations were done, causing Exim to try to write
3735 to the SMTP stream before it had set up the file descriptor. The bug has
3736 been fixed by making things happen in the correct order.
3742 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
3743 before the connection was closed, thus losing the rejection response.
3745 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
3746 some early Solaris releases, but causes trouble in current releases where
3747 socklen_t is defined.
3749 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
3752 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
3755 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
3756 that releasing the top of it at the end releases what was used for sub-
3757 expansions (unless the block got too big). However, discard this block if
3758 the first thing is a variable or header, so that we can use its block when
3759 it is dynamic (useful for very large $message_headers, for example).
3761 6. Lookups now cache *every* query, not just the most recent. A new, separate
3762 store pool is used for this. It can be recovered when all lookup caches are
3763 flushed. Lookups now release memory at the end of their result strings.
3764 This has involved some general refactoring of the lookup sources.
3766 7. Some code has been added to the store_xxx() functions to reduce the amount
3767 of flapping under certain conditions.
3769 8. log_incoming_interface used to affect only the <= reception log lines. Now
3770 it causes the local interface and port to be added to several more SMTP log
3771 lines, for example "SMTP connection from", and rejection lines.
3773 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
3775 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
3777 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
3778 could be overwritten at the end of the current message (or the start of a
3779 new message if it was set in a HELO ACL). The value is now preserved for
3780 the duration of the SMTP connection.
3782 12. If a transport had a headers_rewrite setting, and a matching header line
3783 contained an unqualified address, that address was qualified, even if it
3784 did not match any rewriting rules. The underlying bug was that the values
3785 of the flags that permit the existence of unqualified sender and recipient
3786 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
3787 non-local messages, and by -bnq for local messages) were not being
3788 preserved with the message after it was received.
3790 13. When Exim was logging an SMTP synchronization error, it could sometimes log
3791 "next input=" as part of the text comprising the host identity instead of
3792 the correct text. The code was using the same buffer for two different
3793 strings. However, depending on which order the printing function evaluated
3794 its arguments, the bug did not always show up. Under Linux, for example, my
3795 test suite worked just fine.
3797 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
3798 doesn't work with some older versions of Perl. It has been changed to "my",
3799 which in any case is probably the better facility to use.
3801 15. A really picky compiler found some instances of statements for creating
3802 error messages that either had too many or two few arguments for the format
3805 16. The size of the buffer for calls to the DNS resolver has been increased
3806 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
3807 for addresses that have a lot of PTR records. This alleviates a problem; it
3808 does not fully solve it.
3810 17. A dnsdb lookup for PTR records that receives more data than will fit in the
3811 buffer now truncates the list and logs the incident, which is the same
3812 action as happens when Exim is looking up a host name and its aliases.
3813 Previously in this situation something unpredictable would happen;
3814 sometimes it was "internal error: store_reset failed".
3816 18. If a server dropped the connection unexpectedly when an Exim client was
3817 using GnuTLS and trying to read a response, the client delivery process
3818 crashed while trying to generate an error log message.
3820 19. If a "warn" verb in an ACL added multiple headers to a message in a single
3821 string, for example:
3823 warn message = H1: something\nH2: something
3825 the text was added as a single header line from Exim's point of view
3826 though it ended up OK in the delivered message. However, searching for the
3827 second and subsequent header lines using $h_h2: did not work. This has been
3828 fixed. Similarly, if a system filter added multiple headers in this way,
3829 the routers could not see them.
3831 20. Expanded the error message when iplsearch is called with an invalid key to
3832 suggest using net-iplsearch in a host list.
3834 21. When running tests using -bh, any delays imposed by "delay" modifiers in
3835 ACLs are no longer actually imposed (and a message to that effect is
3838 22. If a "gecos" field in a passwd entry contained escaped characters, in
3839 particular, if it contained a \" sequence, Exim got it wrong when building
3840 a From: or a Sender: header from that name. A second bug also caused
3841 incorrect handling when an unquoted " was present following a character
3842 that needed quoting.
3844 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
3845 was not being matched caselessly.
3847 24. Arranged for all hyphens in the exim.8 source to be escaped with
3850 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
3851 properly. Recipient callout cache records were still being keyed to include
3852 the sender, even when use_sender was set false. This led to far more
3853 callouts that were necessary. The sender is no longer included in the key
3854 when use_sender is false.
3856 26. Added "control = submission" modifier to ACLs.
3858 27. Added the ${base62d: operator to decode base 62 numbers.
3860 28. dnsdb lookups can now access SRV records.
3862 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
3863 the configuration file.
3865 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
3866 (-v) mode. This makes the output for a verbose queue run more intelligible.
3868 31. Added a use_postmaster feature to recipient callouts.
3870 32. Added the $body_zerocount variable, containing the number of binary zero
3871 bytes in the message body.
3873 33. The time of last modification of the "new" subdirectory is now used as the
3874 "mailbox time last read" when there is a quota error for a maildir
3877 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
3879 35. Added +ignore_unknown as a special item in host lists.
3881 36. Code for decoding IPv6 addresses in host lists is now included, even if
3882 IPv6 support is not being compiled. This fixes a bug in which an IPv6
3883 address was recognized as an IP address, but was then not correctly decoded
3884 into binary, causing unexpected and incorrect effects when compared with
3891 1. Very minor rewording of debugging text in manualroute to say "list of
3892 hosts" instead of "hostlist".
3894 2. If verify=header_syntax was set, and a header line with an unqualified
3895 address (no domain) and a large number of spaces between the end of the
3896 name and the colon was received, the reception process suffered a buffer
3897 overflow, and (when I tested it) crashed. This was caused by some obsolete
3898 code that should have been removed. The fix is to remove it!
3900 3. When running in the test harness, delay a bit after writing a bounce
3901 message to get a bit more predictability in the log output.
3903 4. Added a call to search_tidyup() just before forking a reception process. In
3904 theory, someone could use a lookup in the expansion of smtp_accept_max_
3905 per_host which, without the tidyup, could leave open a database connection.
3907 5. Added the variables $recipient_data and $sender_data which get set from a
3908 lookup success in an ACL "recipients" or "senders" condition, or a router
3909 "senders" option, similar to $domain_data and $local_part_data.
3911 6. Moved the writing of debug_print from before to after the "senders" test
3914 7. Change 4.31/66 (moving the time when the Received: is generated) caused
3915 problems for message scanning, either using a data ACL, or using
3916 local_scan() because the Received: header was not generated till after they
3917 were called (in order to set the time as the time of reception completion).
3918 I have revised the way this works. The header is now generated after the
3919 body is received, but before the ACL or local_scan() are called. After they
3920 are run, the timestamp in the header is updated.
3926 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
3927 before starting a queue runner without re-exec. This happened only when
3928 deliver_drop_privilege was set or when the Exim user was set to root. The
3929 effect of the bug was that timeouts during subsequent deliveries caused
3930 crashes instead of being properly handled. The handler is now left at its
3931 default (and expected) setting.
3933 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
3934 message, again when deliver_drop_privilege is set or Exim is run as root.
3935 The bug described in (1) was not present in this case, but the tidying up
3936 of the other signals was missing. I have made the two cases consistent.
3938 3. The ignore_target_hosts setting on a manualroute router was being ignored
3939 for hosts that were looked up using the /MX notation.
3941 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
3944 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
3945 operated on the sender address. After changing the $sender_address to <>
3946 for the sender address verify, Exim was re-instated it as the original
3947 (before rewriting) address, but remembering that it had rewritten it, so it
3948 wasn't rewriting it again. This bug also had the effect of breaking the
3949 sender address verification caching when the sender address was rewritten.
3951 6. The ignore_target_hosts option was being ignored by the ipliteral router.
3952 This has been changed so that if the ip literal address matches
3953 ignore_target_hosts, the router declines.
3955 7. Added expansion conditions match_domain, match_address, and match_local_
3956 part (NOT match_host).
3958 8. The placeholder for the Received: header didn't have a length field set.
3960 9. Added code to Exim itself and to exim_lock to test for a specific race
3961 condition that could lead to file corruption when using MBX delivery. The
3962 issue is with the lockfile that is created in /tmp. If this file is removed
3963 after a process has opened it but before that process has acquired a lock,
3964 there is the potential for a second process to recreate the file and also
3965 acquire a lock. This could lead to two Exim processes writing to the file
3966 at the same time. The added code performs the same test as UW imapd; it
3967 checks after acquiring the lock that its file descriptor still refers to
3968 the same named file.
3970 10. The buffer for building added header lines was of fixed size, 8192 bytes.
3971 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
3974 11. Added the smtp_active_hostname option. If used, this will typically be made
3975 to depend on the incoming interface address. Because $interface_address is
3976 not set up until the daemon has forked a reception process, error responses
3977 that can happen earlier (such as "too many connections") no longer contain
3980 12. If an expansion in a condition on a "warn" statement fails because a lookup
3981 defers, the "warn" statement is abandoned, and the next ACL statement is
3982 processed. Previously this caused the whole ACL to be aborted.
3984 13. Added the iplsearch lookup type.
3986 14. Added ident_timeout as a log selector.
3988 15. Added tls_certificate_verified as a log selector.
3990 16. Added a global option tls_require_ciphers (compare the smtp transport
3991 option of the same name). This controls incoming TLS connections.
3993 17. I finally figured out how to make tls_require_ciphers do a similar thing
3994 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
3995 before starting the TLS session.
3997 18. Tabs are now shown as \t in -bP output.
3999 19. If the log selector return_path_on_delivery was set, Exim crashed when
4000 bouncing a message because it had too many Received: header lines.
4002 20. If two routers both had headers_remove settings, and the first one included
4003 a superfluous trailing colon, the final name in the first list and the
4004 first name in the second list were incorrectly joined into one item (with a
4005 colon in the middle).
4011 1. Added -C and -D options to the exinext utility, mainly to make it easier
4012 to include in the automated testing, but these could be helpful when
4013 multiple configurations are in use.
4015 2. The exinext utility was not formatting the output nicely when there was
4016 an alternate port involved in the retry record key, nor when there was a
4017 message id as well (for retries that were specific to a specific message
4018 and a specific host). It was also confused by IPv6 addresses, because of
4019 the additional colons they contain. I have fixed the IPv4 problem, and
4020 patched it up to do a reasonable job for IPv6.
4022 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
4023 delivery, the log line now contains "pipelined" if PIPELINING was used.
4025 4. An SMTP transport process used to panic and die if the bind() call to set
4026 an explicit outgoing interface failed. This has been changed; it is now
4027 treated in the same way as a connect() failure.
4029 5. A reference to $sender_host_name in the part of a conditional expansion
4030 that was being skipped was still causing a DNS lookup. This no longer
4033 6. The def: expansion condition was not recognizing references to header lines
4034 that used bh_ and bheader_.
4036 7. Added the _cache feature to named lists.
4038 8. The code for checking quota_filecount in the appendfile transport was
4039 allowing one more file than it should have been.
4041 9. For compatibility with Sendmail, the command line option
4049 and sets the incoming protocol and host name (for trusted callers). The
4050 host name and its colon can be omitted when only the protocol is to be set.
4051 Note the Exim already has two private options, -pd and -ps, that refer to
4052 embedded Perl. It is therefore impossible to set a protocol value of "d" or
4053 "s", but I don't think that's a major issue.
4055 10. A number of refactoring changes to the code, none of which should affect
4058 (a) The number of logging options was getting close to filling up the
4059 32-bit word that was used as a bit map. I have split them into two classes:
4060 those that are passed in the argument to log_write(), and those that are
4061 only ever tested independently outside of that function. These are now in
4062 separate 32-bit words, so there is plenty of room for expansion again.
4063 There is no change in the user interface or the logging behaviour.
4065 (b) When building, for example, log lines, the code previously used a
4066 macro that called string_cat() twice, in order to add two strings. This is
4067 not really sufficiently general. Furthermore, there was one instance where
4068 it was actually wrong because one of the argument was used twice, and in
4069 one call a function was used. (As it happened, calling the function twice
4070 did not affect the overall behaviour.) The macro has been replaced by a
4071 function that can join an arbitrary number of extra strings onto a growing
4074 (c) The code for expansion conditions now uses a table and a binary chop
4075 instead of a serial search (which was left over from when there were very
4076 few conditions). Also, it now recognizes conditions like "pam" even when
4077 the relevant support is not compiled in: a suitably worded error message is
4078 given if an attempt is made to use such a condition.
4080 11. Added ${time_interval:xxxxx}.
4082 12. A bug was causing one of the ddress fields not to be passed back correctly
4083 from remote delivery subprocesses. The field in question was not being
4084 subsequently used, so this caused to problems in practice.
4086 13. Added new log selectors queue_time and deliver_time.
4088 14. Might have fixed a bug in maildirsizefile handling that threw up
4089 "unexpected character" debug warnings, and recalculated the data
4090 unnecessarily. In any case, I expanded the warning message to give more
4093 15. Added the message "Restricted characters in address" to the statements in
4094 the default ACL that block characters like @ and % in local parts.
4096 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
4097 Three changes have been made:
4099 (a) There was a serious bug; a negative response to MAIL caused the whole
4100 recipient domain to be cached as invalid, thereby blocking all messages
4101 to all local parts at the same domain, from all senders. This bug has
4102 been fixed. The domain is no longer cached after a negative response to
4103 MAIL if the sender used is not empty.
4105 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
4108 (c) A new callout option, "use_sender" has been added for people who want
4109 the modified behaviour.
4115 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
4118 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
4119 indeed breaks things for older releases.
4121 3. Added additional logging to the case where there is a problem reading data
4122 from a filter that is running in a subprocess using a pipe, in order to
4123 try to track down a specific problem.
4125 4. Testing facility fudge: when running in the test harness and attempting
4126 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
4127 getting "No route to host". Convert this to a timeout.
4129 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
4132 6. Some OS don't have socklen_t but use size_t instead. This affects the
4133 fifth argument of getsockopt() amongst other things. This is now
4134 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
4135 can be set for individual OS. I have set it for SunOS5, OSF1, and
4136 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
4137 some earlier ones do not.
4139 7. Change 4.30/15 was not doing the test caselessly.
4141 8. The standard form for an IPv6 address literal was being rejected by address
4142 parsing in, for example, MAIL and RCPT commands. An example of this kind of
4143 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
4144 this, as well as the form without the "IPv6" on the front (but only when
4145 address literals are enabled, of course).
4147 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
4149 10. Exim crashed if a message with an empty sender address specified by -f
4150 encountered a router with an errors_to setting. This could be provoked only
4151 by a command such as
4155 where an empty string was supplied; "<>" did not hit this bug.
4157 11. Installed PCRE release 4.5.
4159 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
4160 remained set. It is now erased.
4162 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
4163 times from message ids (which are base 36 rather than the normal 62).
4165 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
4166 were being counted as actual protocol errors, and logged if the log
4167 selector +smtp_protocol_error was set. One cannot be perfect in this test,
4168 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
4169 and DATA following a set of rejected RCPTs do not count as protocol errors.
4170 In other words, Exim assumes they were pipelined, though this may not
4171 actually be the case. Of course, in all cases the client gets an
4172 appropriate error code.
4174 15. If a lookup fails in an ACL condition, a message about the failure may
4175 be available; it is used if testing the ACL cannot continue, because most
4176 such messages specify what the cause of the deferral is. However, some
4177 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
4178 that caused an old message to be retained and used if a later statement
4179 caused a defer, replacing the real cause of the deferral.
4181 16. If an IP address had so many PTR records that the DNS lookup buffer
4182 was not large enough to hold them, Exim could crash while trying to process
4183 the truncated data. It now detects and logs this case.
4185 17. Further to 4.21/58, another change has been made: if (and only if) the
4186 first line of a message (the first header line) ends with CRLF, a bare LF
4187 in a subsequent header line has a space inserted after it, so as not to
4188 terminate the header.
4190 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
4191 unnecessarily, used atoi() instead of strtol(), and didn't check the
4192 termination when getting file sizes from file names by regex.
4194 19. Completely re-implemented the support for maildirsize files, in the light
4195 of a number of problems with the previous contributed implementation
4196 (4.30/29). In particular:
4198 . If the quota is zero, the maildirsize file is maintained, but no quota is
4201 . If the maildir directory does not exist, it is created before any attempt
4202 to write a maildirsize file.
4204 . The quota value in the file is just a cache; if the quota is changed in
4205 the transport, the new value overrides.
4207 . A regular expression is available for excluding directories from the
4210 20. The autoreply transport checks the characters in options that define the
4211 message's headers; it allows continued headers, but it was checking with
4212 isspace() after an embedded newline instead of explicitly looking for a
4215 21. If all the "regular" hosts to which an address was routed had passed their
4216 expiry times, and had not reached their retry times, the address was
4217 bounced, even if fallback hosts were defined. Now Exim should go on to try
4220 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
4221 equivalent code in the SMTP transport. Some hosts send humungous responses
4222 to HELO/EHLO, more than 1024 it seems.
4224 23. Refactoring: code in filter.c used (void *) for "any old type" but this
4225 gives compiler warnings in some environments. I've now done it "properly",
4228 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
4229 (because of problems with the built-in one) was declared to return uschar *
4230 instead of char *, causing compiler failure.
4232 25. Fixed a file descriptor leak when processing alias/forward files.
4234 26. Fixed a minor format string issue in dbfn.c.
4236 27. Typo in exim.c: ("dmbnz" for "dbmnz").
4238 28. If a filter file refered to $h_xxx or $message_headers, and the headers
4239 contained RFC 2047 "words", Exim's memory could, under certain conditions,
4242 29. When a sender address is verified, it is cached, to save repeating the test
4243 when there is more than one recipient in a message. However, when the
4244 verification involves a callout, it is possible for different callout
4245 options to be set for different recipients. It is too complicated to keep
4246 track of this in the cache, so now Exim always runs a verification when a
4247 callout is required, relying on the callout cache for the optimization.
4248 The overhead is duplication of the address routing, but this should not be
4251 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
4252 to be verified with callout=postmaster, and the main callout worked but the
4253 postmaster check failed, the verification correctly failed. However, if a
4254 subsequent RCPT command asked for sender verification *without* the
4255 postmaster check, incorrect caching caused this verification also to fail,
4258 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
4259 it was not caching the DNS options (qualify_single, search_parents) that
4260 were used when the lookup failed. A subsequent lookup with different
4261 options therefore always gave the same answer, though there were cases
4262 where it should not have. (Example: a "domains = !$mx_any" option on a
4263 dnslookup router: the "domains" option is always processed without any
4264 widening, but the router might have qualify_single set.) Now Exim uses the
4265 cached value only when the same options are set.
4267 32. Added John Jetmore's "exipick" utility to the distribution.
4269 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
4270 than a timeout (e.g. a certificate is required, and is not provided), an
4271 Exim server now closes the connection immediately. Previously it waited for
4272 the client to close - but if the client is SSL, it seems that they each
4273 wait for each other, leading to a delay before one of them times out.
4275 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
4276 maintained 0.8.x compatibility because I don't think many are using it, and
4277 it is clearly obsolete.
4279 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
4282 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
4283 client certificate was expired. A simple patch fixes this, though I don't
4284 understand the full logic of why the verify callback is called multiple
4287 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
4288 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
4289 which causes problems with some clients (such as the Certicom SSL Plus
4290 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
4291 disables the coutermeasure allowing Eudora to connect."
4293 38. Exim was not checking that a write() to a log file succeeded. This could
4294 lead to Bad Things if a log got too big, in particular if it hit a file
4295 size limit. Exim now panics and dies if it cannot write to a log file, just
4296 as it does if it cannot open a log file.
4298 39. Modified OS/Makefile-Linux so that it now contains
4300 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
4302 The two -D definitions ensure that Exim is compiled with large file
4303 support, which makes it possible to handle log files that are bigger than
4306 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
4307 instance) a domain was checked against a named list that involved a lookup,
4308 causing $domain_data to be set, then another domain was checked against the
4309 same list, then the first domain was re-checked, the value of $domain_data
4310 after the final check could be wrong. In particular, if the second check
4311 failed, it could be set empty. This bug probably also applied to
4314 41. The strip_trailing_dot option was not being applied to the address given
4315 with the -f command-line option.
4317 42. The code for reading a message's header from the spool was incrementing
4318 $received_count, but never initializing it. This meant that the value was
4319 incorrect (doubled) while delivering a message in the same process in which
4320 it was received. In the most common configuration of Exim, this never
4321 happens - a fresh exec is done - but it can happen when
4322 deliver_drop_privilege is set.
4324 43. When Exim logs an SMTP synchronization error - client data sent too soon -
4325 it now includes up to 150 characters of the unexpected data in the log
4328 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
4329 and building data strings. The size of both of these buffers was 10 000
4330 bytes - far larger than anybody would *ever* want, thought I. Needless to
4331 say, somebody hit the limit. I have increased the maximum line length to
4332 20 000 and the maximum data length of concatenated lines to 100 000. I have
4333 also fixed two bugs, because there was no checking on these buffers. Tsk,
4334 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
4335 buffer is too small.
4337 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
4338 lsearch lookups. Now it does.
4340 46. When parsing a route_list item in a manualroute router, a fixed-length
4341 buffer was used for the list of hosts. I made this 1024 bytes long,
4342 thinking that nobody would ever have a list of hosts that long. Wrong.
4343 Somebody had a whole pile of complicated expansion conditions, and the
4344 string was silently truncated, leading to an expansion error. It turns out
4345 that it is easier to change to an unlimited length (owing to other changes
4346 that have happened since this code was originally written) than to build
4347 structure for giving a limitation error. The length of the item that
4348 expands into the list of hosts is now unlimited.
4350 47. The lsearch lookup could not handle data where the length of text line was
4351 more than 4095 characters. Such lines were truncated, leading to shortened
4352 data being returned. It should now handle lines of any length.
4354 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
4355 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
4358 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
4359 replacing strings such as BIN_DIRECTORY in a source file: the replacement
4360 no longer happens in comment lines. A list of replacements is now placed
4361 at the head of all of the source files, except those whose only change is
4362 to replace PERL_COMMAND in the very first #! line.
4364 50. Replaced the slow insertion sort in queue.c, for sorting the list of
4365 messages on the queue, with a bottom-up merge sort, using code contributed
4366 by Michael Haardt. This should make operations like -bp somewhat faster on
4367 large queues. It won't affect queue runners, except when queue_run_in_order
4370 51. Installed eximstats 1.31 in the distribution.
4372 52. Added support for SRV lookups to the dnslookup router.
4374 53. If an ACL referred to $message_body or $message_body_end, the value was not
4375 reset for any messages that followed in the same SMTP session.
4377 54. The store-handling optimization for building very long strings was not
4378 differentiating between the different store pools. I don't think this
4379 actually made any difference in practice, but I've tidied it.
4381 55. While running the routers to verify a sender address, $sender_address
4382 was still set to the sender address. This is wrong, because when routing to
4383 send a bounce to the sender, it would be empty. Therefore, I have changed
4384 it so that, while verifying a sender address, $sender_address is set to <>.
4385 (There is no change to what happens when verifying a recipient address.)
4387 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
4388 target A or AAAA records (if not already returned) without resetting the
4389 qualify_single or search_parents options of the DNS resolver. These are
4390 inappropriate in this case because the targets of MX and SRV records must
4391 be FQDNs. A broken DNS record could cause trouble if it happened to have a
4392 target that, when qualified, matched something in the local domain. These
4393 two options are now turned off when doing these lookups.
4395 57. It seems that at least some releases of Reiserfs (which does not have the
4396 concept of a fixed number of inodes) returns zero and not -1 for the
4397 number of available inodes. This interacted badly with check_spool_inodes,
4398 which assumed that -1 was the "no such thing" setting. What I have done is
4399 to check that the total number of inodes is greater than zero before doing
4400 the test of how many are available.
4402 58. When a "warn" ACL statement has a log_message modifier, the message is
4403 remembered, and not repeated. This is to avoid a lot of repetition when a
4404 message has many recipients that cause the same warning to be written.
4405 Howewer, Exim was preserving the list of already written lines for an
4406 entire SMTP session, which doesn't seem right. The memory is now reset if a
4407 new message is started.
4409 59. The "rewrite" debugging flag was not showing the result of rewriting in the
4410 debugging output unless log_rewrite was also set.
4412 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
4413 of (int)(handle) when we know that handle contains (void *)(-1).
4415 61. The Exim daemon panic-logs an error return when it closes the incoming
4416 connection. However "connection reset by peer" seems to be common, and
4417 isn't really an error worthy of noting specially, so that particular error
4420 62. When Exim is trying to find all the local interfaces, it used to panic and
4421 die if the ioctl to get the interface flags failed. However, it seems that
4422 on at least one OS (Solaris 9) it is possible to have an interface that is
4423 included in the list of interfaces, but for which you get a failure error
4424 for this call. This happens when the interface is not "plumbed" into a
4425 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
4426 failure of the "get flags" call assumes that the interface is down.
4428 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
4429 makes life easier for people who are doing arithmetic on fields extracted
4430 from dates, where you often get leading zeros that should not be
4431 interpreted as octal.
4433 64. Added qualify_domain to the redirect router, to override the global
4436 65. If a pathologically long header line contained very many addresses (the
4437 report of this problem mentioned 10 000) and each of them was rewritten,
4438 Exim could use up a very large amount of memory. (It kept on making new
4439 copies of the header line as it rewrote, and never released the old ones.)
4440 At the expense of a bit more processing, the header rewriting function has
4441 been changed so that it no longer eats memory in this way.
4443 66. The generation of the Received: header has been moved from the time that a
4444 message starts to be received, to the time that it finishes. The timestamp
4445 in the Received: header should now be very close to that of the <= log
4446 line. There are two side-effects of this change:
4448 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
4449 logged header lines no longer include the local Received: line, because
4450 it has not yet been created. The same applies to a copy of the message
4451 that is returned to a non-SMTP sender when a message is rejected.
4453 (b) When a filter file is tested using -bf, no additional Received: header
4454 is added to the test message. After some thought, I decided that this
4457 This change does not affect the value of $received_for. It is still set
4458 after address rewriting, but before local_scan() is called.
4460 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
4462 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
4463 gave an unhelpful panic error message, and a defer error. I have managed to
4464 change this behaviour so that it now rejects any supplied certificate,
4465 which seems right, as the list of acceptable certificates is empty.
4467 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
4468 gave an unhelpful defer error. I have not managed to make this reject any
4469 supplied certificates, but the error message it gives is "no certificate
4470 supplied", which is not helpful.
4472 70. exigrep's output now also includes lines that are not associated with any
4473 message, but which match the given pattern. Implemented by a patch from
4474 Martin Sluka, which also tidied up the Perl a bit.
4476 71. Recipient callout verification, like sender verification, was using <> in
4477 the MAIL FROM command. This isn't really the right thing, since the actual
4478 sender may affect whether the remote host accepts the recipient or not. I
4479 have changed it to use the actual sender in the callout; this means that
4480 the cache record is now keyed on a recipient/sender pair, not just the
4481 recipient address. There doesn't seem to be a real danger of callout loops,
4482 since a callout by the remote host to check the sender would use <>.
4483 [SEE ABOVE: changed after hitting problems.]
4485 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
4486 temporary errors. However, in the case of such a code being given after
4487 the end of a data transmission (i.e. after ".") Exim was failing to write
4488 a retry record for the message. (Yes, there was some broken host that was
4489 actually sending 8xx at this point.)
4491 73. An unknown lookup type in a host list could cause Exim to panic-die when
4492 the list was checked. (An example that provoked this was putting <; in the
4493 middle of a list instead of at the start.) If this happened during a DATA
4494 ACL check, a -D file could be left lying around. This kind of configuration
4495 error no longer causes Exim to die; instead it causes a defer errror. The
4496 incident is still logged to the main and panic logs.
4498 74. Buglet left over from Exim 3 conversion. The message "too many messages
4499 in one connection" was written to the rejectlog but not the mainlog, except
4500 when address rewriting (yes!) was being logged.
4502 75. Added write_rejectlog option.
4504 76. When a system filter was run not as root (that is, when system_filter_user
4505 was set), the values of the $n variables were not being returned to the
4506 main process; thus, they were not subsequently available in the $sn
4509 77. Added +return_path_on_delivery log selector.
4511 78. A connection timeout was being treated differently from recipients deferred
4512 when testing hosts_max_try with a message that was older than the host's
4513 retry timeout. (The host should not be counted, thus allowing all hosts to
4514 be tried at least once before bouncing.) This may have been the cause of an
4515 occasionally reported bug whereby a message would remain on the queue
4516 longer than the retry timeout, but would be bounced if a delivery was
4517 forced. I say "may" because I never totally pinned down the problem;
4518 setting up timeout/retry tests is difficult. See also the next item.
4520 79. The ultimate address timeout was not being applied to errors that involved
4521 a combination of host plus message (for example, a timeout on a MAIL
4522 command). When an address resolved to a number of possible hosts, and they
4523 were not all tried for each delivery (e.g. because of hosts_max_try), a
4524 message could remain on the queue longer than the retry timeout.
4526 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
4529 81. Fixed an obscure SMTP outgoing bug which required at least the following
4530 conditions: (a) there was another message waiting for the same server;
4531 (b) the server returned 5xx to all RCPT commands in the first message so
4532 that the message was not completed; (c) the server dropped the connection
4533 or gave a negative response to the RSET that Exim sends to abort the
4534 transaction. The observed case was a dropped connection after DATA that had
4535 been sent in pipelining mode. That is, the server had advertised PIPELINING
4536 but was not implementing it correctly. The effect of the bug was incorrect
4537 behaviour, such as trying another host, and this could lead to a crash.
4543 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
4544 and daemon.c were passed as pointers to ints; they should have been
4545 pointers to socklen_t variables (which are typically unsigned ints).
4547 2. Some signed/unsigned type warnings in the os.c file for Linux have been
4550 3. Fixed a really odd bug that affected only the testing scheme; patching a
4551 certain fixed string in the binary changed the value of another string that
4552 happened to be identical to the end of the original first string.
4554 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
4555 name", it returns that address as the IP address. On some operating
4556 systems (e.g. Solaris), it also passes back the IP address string as the
4557 "host name". However, on others (e.g. Linux), it passes back an empty
4558 string. Exim wasn't checking for this, and was changing the host name to an
4559 empty string, assuming it had been canonicized.
4561 5. Although rare, it is permitted to have more than one PTR record for a given
4562 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
4563 all the names associated with an address, because they do in Solaris.
4564 However, it seems that they do not in Linux for data that comes from the
4565 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
4566 I found this out when I moved to a new Linux workstation and tried to run
4567 the Exim test suite.
4569 To get round this problem I have changed the code so that it now does its
4570 own call to the DNS to look up PTR records when searching for a host name.
4571 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
4572 addresses that are only in /etc/hosts are still found.
4574 This behaviour is, however, controlled by an option called host_lookup_
4575 order, which defaults to "bydns:byaddr". If people want to use the other
4576 order, or indeed, just use one or the other means of lookup, they can
4577 specify it in this variable.
4579 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
4580 some operating systems, this comes back from gethostbyaddr() as an empty
4581 string, and this is what Exim used to test for. However, it seems that in
4582 other systems, "." is yielded. Exim now tests for this case too.
4584 7. The values of check_spool_space and check_log_space are now held internally
4585 as a number of kilobytes instead of an absolute number of bytes. If a
4586 numbers is specified without 'K' or 'M', it is rounded up to the nearest
4587 kilobyte. This means that much larger values can be stored.
4589 8. Exim monitor: an attempt to get the action menu when not actually pointing
4590 at a message produces an empty menu entitled "No message selected". This
4591 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
4592 no entries in it ("Shell widget menu has zero width and/or height"). So I
4593 have added a single, blank menu entry in this case.
4595 9. Added ${quote_local_part.
4597 10. MIME decoding is now applied to the contents of Subject: header lines when
4600 11. Now that a reference to $sender_host_address automatically causes a reverse
4601 lookup to occur if necessary (4.13/18), there is no need to arrange for a
4602 host lookup before query-style lookups in lists that might use this
4603 variable. This has therefore been abolished, and the "net-" prefix is no
4604 longer necessary for query-style lookups.
4606 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
4607 have been a typo for LFLAGS, so it has been changed.
4609 13. The install script calls Exim with "-C /dev/null" in order to find the
4610 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
4611 to be output. Howeve, since Exim outputs its version number before the
4612 error, it didn't break the script. It just looked ugly. I fixed this by
4613 always allowing "-C /dev/null" if the caller is root.
4615 14. Ignore overlarge ACL variable number when reading spool file - insurance
4616 against a later release with more variables having written the file.
4618 15. The standard form for an IPv6 address literal was being rejected by EHLO.
4619 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
4620 this, as well as the form without the "IPv6" on the front.
4622 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
4623 OS/Makefile-Darwin file.
4625 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
4626 only to LDAP libraries that do not have LDAP_OPT_DEREF.
4628 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
4629 variable. However, some OS use ints for inodes. I've added cast to long int
4630 to get rid of the compiler warning.
4632 19. I had forgotten to lock out "/../" in configuration file names when
4633 ALT_CONFIG_PREFIX was set.
4635 20. Routers used for verification do not need to specify transports. However,
4636 if such a router generated a host list, and callout was configured, Exim
4637 crashed, because it could not find a port number from the (non-existent)
4638 transport. It now assumes port 25 in this circumstance.
4640 21. Added the -t option to exigrep.
4642 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
4643 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
4644 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
4645 them is set, LOOKUP_LSEARCH is forced.
4647 23. "exim -bV" now outputs a list of lookups that are included in the binary.
4649 24. Added sender and host information to the "rejected by local_scan()" log
4650 line; previously there was no indication of these.
4652 25. Added .include_if_exists.
4654 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
4655 Linux. It turns out that not all file systems support this. Apparently some
4656 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
4657 do it.) To cope with this, the error EINVAL, which means that sync-ing is
4658 not supported on the file descriptor, is now ignored when Exim is trying to
4659 sync a directory. This applies only to Linux.
4661 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
4663 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
4664 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
4667 29. Added support for maildirsize files from supplied patch (modified a bit).
4669 30. The use of :fail: followed by an empty string could lead Exim to respond to
4670 sender verification failures with (e.g.):
4672 550 Verification failed for <xxx>
4673 550 Sender verify failed
4675 where the first response line was missing the '-' that indicates it is not
4676 the final line of the response.
4678 31. The loop for finding the name of the user that called Exim had a hardwired
4679 limit of 10; it now uses the value of finduser_retries, which is used for
4680 all other user lookups.
4682 32. Added $received_count variable, available in data and not_smtp ACLs, and at
4685 33. Exim was neglecting to zero errno before one call of strtol() when
4686 expanding a string and expecting an integer value. On some systems this
4687 resulted in spurious "integer overflow" errors. Also, it was casting the
4688 result into an int without checking.
4690 34. Testing for a connection timeout using "timeout_connect" in the retry rules
4691 did not work. The code looks as if it has *never* worked, though it appears
4692 to have been documented since at least releast 1.62. I have made it work.
4694 35. The "timeout_DNS" error in retry rules, also documented since at least
4695 1.62, also never worked. As it isn't clear exactly what this means, and
4696 clearly it isn't a major issue, I have abolished the feature by treating it
4697 as "timeout", and writing a warning to the main and panic logs.
4699 36. The display of retry rules for -brt wasn't always showing the error code
4702 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
4703 timeout_connect_A, timeout_connect_MX.
4705 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
4706 to the empty sender.
4708 39. The daemon was not analysing the content of -oX till after it had closed
4709 stderr and disconnected from the controlling terminal. This meant that any
4710 syntax errors were only noted on the panic log, and the return code from
4711 the command was 0. By re-arranging the code a little, I've made the
4712 decoding happen first, so such errors now appear on stderr, and the return
4713 code is 1. However, the actual setting up of the sockets still happens in
4714 the disconnected process, so errors there are still only recorded on the
4717 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
4718 connections (as happens on some IP stacks) was logged at start up time as
4719 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
4720 it from "IPv6 and IPv4", which means that two separate sockets are being
4723 41. The debug output for gethostbyname2() or getipnodebyname() failures now
4724 says whether AF_INET or AF_INET6 was passed as an argument.
4726 42. Exiwhat output was messed up when time zones were included in log
4729 43. Exiwhat now gives more information about the daemon's listening ports,
4730 and whether -tls-on-connect was used.
4732 44. The "port" option of the smtp transport is now expanded.
4734 45. A "message" modifier in a "warn" statement in a non-message ACL was being
4735 silently ignored. Now an error message is written to the main and panic
4738 46. There's a new ACL modifier called "logwrite" which writes to a log file
4739 as soon as it is encountered.
4741 47. Added $local_user_uid and $local_user_gid at routing time.
4743 48. Exim crashed when trying to verify a sender address that was being
4746 49. Exim was recognizing only a space character after ".include". It now also
4747 recognizes a tab character.
4749 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
4750 extracting the relevant information from the specification. The man page no
4751 longer contains scrambled data for the -d option, and I've added a section
4752 at the front about calling Exim under different names.
4754 51. Added "extra_headers" argument to the "mail" command in filter files.
4756 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
4759 53. Installed eximstats 1.29.
4761 54. Added transport_filter_timeout as a generic transport option.
4763 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
4764 Cc: header lines. This was required by RFC 822, but it not required by RFC
4767 56. Exim used to add From:, Date:, and Message-Id: header lines to any
4768 incoming messages that did not have them. Now it does so only if the
4769 message originates locally, that is, if there is no associated remote host
4770 address. When Resent- header lines are present, this applies to the Resent-
4771 lines rather than the non-Resent- lines.
4773 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
4774 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
4776 58. Messages for configuration errors now include the name of the main
4777 configuration file - useful now that there may be more than one file in a
4778 list (.included file names were always shown).
4780 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
4781 for those rare installations that do not start the daemon as root or run it
4782 setuid root. I've cut out the call to initgroups() if the daemon is not
4785 60. The Exim user and group can now be bound into the binary as text strings
4786 that are looked up at the start of Exim's processing.
4788 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
4790 62. Added $mailstore_basename variable.
4792 63. Installed patch to sieve.c from Michael Haardt.
4794 64. When Exim failed to open the panic log after failing to open the main log,
4795 the original message it was trying to log was written to stderr and debug
4796 output, but if they were not available (the usual case in production), it
4797 was lost. Now it is written to syslog before the two lines that record the
4798 failures to open the logs.
4800 65. Users' Exim filters run in subprocesses under the user's uid. It is
4801 possible for a "deliver" command or an alias in a "personal" command to
4802 provoke an address rewrite. If logging of address rewriting is configured,
4803 this fails because the process is not running as root or exim. There may be
4804 a better way of dealing with this, but for the moment (because 4.30 needs
4805 to be released), I have disabled address rewrite logging when running a
4806 filter in a non-root, non-exim process.
4812 1. The buildconfig auxiliary program wasn't quoting the value set for
4813 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
4814 not defined. This bug was masked in 4.22 by the effect that was fixed in
4817 2. Some messages that were rejected after a message id was allocated were
4818 shown as "incomplete" by exigrep. It no longer does this for messages that
4819 are rejected by local_scan() or the DATA or non-SMTP ACLs.
4821 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
4822 have allow_domain_literals set, the ID did not get logged in the <= line.
4823 Domain literals are now always recognized in Message-ID: header lines.
4825 4. The first argument for a ${extract expansion item is the key name or field
4826 number. Leading and trailing spaces in this item were not being ignored,
4827 causing some misleading effects.
4829 5. When deliver_drop_privilege was set, single queue runner processes started
4830 manually (i.e. by the command "exim -q") or by the daemon (which uses the
4831 same command in the process it spins off) were not dropping privilege.
4833 6. When the daemon running as "exim" started a queue runner, it always
4834 re-executed Exim in the spun-off process. This is a waste of effort when
4835 deliver_drop_privilege is set. The new process now just calls the
4836 queue-runner function directly.
4842 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
4845 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
4846 ignored. Though the use of -oP was forcing the writing of a pid file, it
4847 was always written to the default place.
4849 3. If the message "no IP address found for host xxxx" is generated during
4850 incoming verification, it is now followed by identification of the incoming
4851 connection (so you can more easily find what provoked it).
4853 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
4855 5. Added some features to "harden" Exim a bit more against certain attacks:
4857 (a) There is now a build-time option called FIXED_NEVER_USERS that can
4858 be put in Local/Makefile. This is like the never_users runtime option,
4859 but it cannot be overridden. The default setting is "root".
4861 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
4862 prefix string with which any file named in a -C command line option
4865 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
4866 is retained for -C and -D only if the caller of Exim is root. Without
4867 it, the exim user may also use -C and -D and retain privilege.
4869 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
4870 command line option is disabled.
4872 6. Macro names set by the -D option must start with an upper case letter, just
4873 like macro names defined in the configuration file.
4875 7. Added "dereference=" facility to LDAP.
4877 8. Two instances of the typo "uknown" in the source files are fixed.
4879 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
4880 the Configure-Makefile script screwed up while processing it.
4882 10. Incorporated PCRE 4.4.
4884 11. The SMTP synchronization check was not operating right at the start of an
4885 SMTP session. For example, it could not catch a HELO sent before the client
4886 waited for the greeting. There is now a check for outstanding input at the
4887 point when the greeting is written. Because of the duplex, asynchronous
4888 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
4889 way, but not yet received, when the check is performed.
4891 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
4892 on TCP/IP sockets, because this apparently causes some broken clients to
4895 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
4896 unchanged) from the Cygwin maintainer.
4898 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
4899 was supported instead of testing for mbx. Effectively a typo.
4901 15. The spa authenticator server code was not checking that the input it
4902 received was valid base64.
4904 16. The debug output line for the "set" modifier in ACLs was not showing the
4905 name of the variable that was being set.
4907 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
4909 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
4910 lookup on its own. Something else was needed to trigger the lookup. For
4911 example, a match in host_lookup or the need for a host name in a host list.
4912 Now, if $sender_host_name is referenced and the host name has not yet been
4913 looked up, a lookup is performed. If the lookup fails, the variable remains
4914 empty, and $host_lookup_failed is set to "1".
4916 19. Added "eqi" as a case-independent comparison operator.
4918 20. The saslauthd authentication condition could segfault if neither service
4919 nor realm was specified.
4921 21. If an overflowing value such as "2048M" was set for message_size_limit, the
4922 error message that was logged was misleading, and incoming SMTP
4923 connections were dropped. The message is now more accurate, and temporary
4924 errors are given to SMTP connections.
4926 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
4927 (except RSET) with a 421 error, until QUIT is received. However, it was
4928 failing to send a response to QUIT.
4930 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
4931 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
4932 ACL is now run after the helo_try_verify_hosts code.
4934 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
4935 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
4936 case-independent, so other case variants are also recognized). Apparently
4937 some systems use these upper case variants.
4939 25. If more than two messages were waiting for the same host, and a transport
4940 filter was specified for the transport, Exim sent two messages over the
4941 same TCP/IP connection, and then failed with "socket operation on non-
4942 socket" when it tried to send the third.
4944 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
4946 27. The extern definition of crypt16() in expand.c was not being excluded when
4947 the OS had its own crypt16() function.
4949 28. Added bounce_return_body as a new option, and bounce_return_size_limit
4950 as a preferred synonym for return_size_limit, both as an option and as an
4953 29. Added LIBS=-liconv to OS/Makefile-OSF1.
4955 30. Changed the default configuration ACL to relax the local part checking rule
4956 for addresses that are not in any local domains. For these addresses,
4957 slashes and pipe symbols are allowed within local parts, but the sequence
4958 /../ is explicitly forbidden.
4960 31. SPA server authentication was not clearing the challenge buffer before
4963 32. log_message in a "warn" ACL statement was writing to the reject log as
4964 well as to the main log, which contradicts the documentation and doesn't
4965 seem right (because no rejection is happening). So I have stopped it.
4967 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
4968 However, I am unable to do any testing of this.
4970 34. Fixed an infelicity in the appendfile transport. When checking directories
4971 for a mailbox, to see if any needed to be created, it was accidentally
4972 using path names with one or more superfluous leading slashes; tracing
4973 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
4975 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
4976 contents are added to the log line that is written for every discarded
4977 recipient. (Previously a log_message setting was ignored.)
4979 36. The ${quote: operator now quotes the string if it is empty.
4981 37. The install script runs exim in order to find its version number. If for
4982 some reason other than non-existence or emptiness, which it checks, it
4983 could not run './exim', it was installing it with an empty version number,
4984 i.e. as "exim-". This error state is now caught, and the installation is
4987 38. An argument was missing from the function that creates an error message
4988 when Exim fails to connect to the socket for saslauthd authentication.
4989 This could cause Exim to crash, or give a corrupted message.
4991 39. Added isip, isip4, and isip6 to ${if conditions.
4993 40. The ACL variables $acl_xx are now saved with the message, and can be
4994 accessed later in routers, transports, and filters.
4996 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
4997 strings in the file are not string-expanded.
4999 42. If a MAIL command specified a SIZE value that was too large to fit into an
5000 int variable, the check against message_size_limit failed. Such values are
5001 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
5002 day this will have to be increased, but I don't think I want to be around
5003 when emails are that large.
5010 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
5011 iconv() is not standard in FreeBSD.
5013 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
5014 IPv6 enabled. The observed symptom was a segmentation fault on return from
5015 the function os_common_find_running_interfaces() in src/os.c.
5017 3. In the check_special_case() function in daemon.c I had used "errno" as an
5018 argument name, which causes warnings on some systems. This was basically a
5019 typo, since it was named "eno" in the comments!
5021 4. The code that waits for the clock to tick (at a resolution of some fraction
5022 of a second) so as to ensure message-id uniqueness was always waiting for
5023 at least one whole tick, when it could have waited for less. [This is
5024 almost certainly not relevant at current processor speeds, where it is
5025 unlikely to ever wait at all. But we try to future-proof.]
5027 5. The function that sleeps for a time interval that includes fractions of a
5028 second contained a race. It did not block SIGALRM between setting the
5029 timer, and suspending (a couple of lines later). If the interval was short
5030 and the sigsuspend() was delayed until after it had expired, the suspension
5031 never ended. On busy systems this could lead to processes getting stuck for
5034 6. Some uncommon configurations may cause a lookup to happen in a queue runner
5035 process, before it forks any delivery processes. The open lookup caching
5036 mechanism meant that the open file or database connection was passed into
5037 the delivery process. The problem was that delivery processes always tidy
5038 up cached lookup data. This could cause a problem for the next delivery
5039 process started by the queue runner, because the external queue runner
5040 process does not know about the closure. So the next delivery process
5041 still has data in the lookup cache. In the case of a file lookup, there was
5042 no problem because closing a file descriptor in a subprocess doesn't affect
5043 the parent. However, if the lookup was caching a connection to a database,
5044 the connection was closed, and the second delivery process was likely to
5045 see errors such as "PGSQL: query failed: server closed the connection
5046 unexpectedly". The problem has been fixed by closing all cached lookups
5047 in a queue runner before running a delivery process.
5049 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
5050 seem to have the "const" qualifier which it has on other OS. I've
5053 8. Change 4.21/2 was too strict. It is only if there are two authenticators
5054 *of the same type* (client or server) with the same public name that an
5055 error should be diagnosed.
5057 9. When Exim looked up a host name for an IP address, but failed to find the
5058 original IP address when looking up the host name (a safety check), it
5059 output the message "<ip address> does not match any IP for NULL", which was
5060 confusing, to say the least. The bug was that the host name should have
5061 appeared instead of "NULL".
5063 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
5064 user that is built into the binary, and the -C or -D options is used, root
5065 privilege is dropped before the configuration file is read. In addition,
5066 logging is switched to stderr instead of the normal log files. If the
5067 configuration then re-defines the Exim user, the unprivileged environment
5068 is probably not what is expected, so Exim logs a panic warning message (but
5071 However, if deliver_drop_privilege is set, the unprivileged state may well
5072 be exactly what is intended, so the warning has been cut out in that case,
5073 and Exim is allowed to try to write to its normal log files.
5079 1. smtp_return_error_details was not giving details for temporary sender
5080 or receiver verification errors.
5082 2. Diagnose a configuration error if two authenticators have the same public
5085 3. Exim used not to create the message log file for a message until the first
5086 delivery attempt. This could be confusing when incoming messages were held
5087 for policy or load reasons. The message log file is now created at the time
5088 the message is received, and an initial "Received" line is written to it.
5090 4. The automatically generated man page for command line options had a minor
5091 bug that caused no ill effects; however, a more serious problem was that
5092 the procedure for building the man page automatically didn't always
5093 operate. Consequently, release 4.20 contains an out-of-date version. This
5094 shouldn't happen again.
5096 5. When building Exim with embedded Perl support, the script that builds the
5097 Makefile was calling 'perl' to find its compile-time parameters, ignoring
5098 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
5100 6. The freeze_tell option was not being used for messages that were frozen on
5101 arrival, either by an ACL or by local_scan().
5103 7. Added the smtp_incomplete_transaction log selector.
5105 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
5106 was accepting AUTH without a new EHLO.
5108 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
5109 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
5112 10. Logging of TCP/IP connections (when configured) now happens in the main
5113 daemon process instead of the child process, so that the TCP/IP connection
5114 count is more accurate (but it can never be perfect).
5116 11. The use of "drop" in a nested ACL was not being handled correctly in the
5117 outer ACL. Now, if condition failure induced by the nested "drop" causes
5118 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
5119 or "require"), the connection is dropped.
5121 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
5122 that yield "discard" can now be used with an "accept" or a "discard" verb,
5123 but an error is generated for any others (because I can't see a useful way
5124 to define what should happen).
5126 13. When an ACL is read dynamically from a file (or anywhere else), the lines
5127 are now processed in the same way as lines in the Exim configuration file.
5128 In particular, continuation lines are supported.
5130 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
5132 15. Added -ti meaning -t -i.
5134 16. Check for letters, digits, hyphens, and dots in the names of dnslist
5135 domains, and warn by logging if others are found.
5137 17. At least on BSD, alignment is not guarenteed for the array of ifreq's
5138 returned from GIFCONF when Exim is trying to find the list of interfaces on
5139 a host. The code in os.c has been modified to copy each ifreq to an aligned
5140 structure in all cases.
5142 Also, in some cases, the returned ifreq's were being copied to a 'struct
5143 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
5144 means the last couple of bytes of an IPv6 address could be chopped if the
5145 ifreq contained only a normal sockaddr (14 bytes storage).
5147 18. Named domain lists were not supported in the hosts_treat_as_local option.
5148 An entry such as +xxxx was not recognized, and was treated as a literal
5151 19. Ensure that header lines added by a DATA ACL are included in the reject log
5152 if the ACL subsequently rejects the message.
5154 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
5155 MD5 (which is deprecated).
5157 21. When testing a filter file using -bf, Exim was writing a message when it
5158 took the sender from a "From " line in the message, but it was not doing so
5159 when it took $return_path from a Return-Path: header line. It now does.
5161 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
5162 with a valid header line field name (a series of printing characters
5163 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
5165 23. Changed "disc" in the source to "disk" to conform to the documentation and
5166 the book and for uniformity.
5168 24. Ignore Sendmail's -Ooption=value command line item.
5170 25. When execve() failed while trying to run a command in a pipe transport,
5171 Exim was returning EX_UNAVAILBLE (69) from the subprocess. However, this
5172 could be confused with a return value of 69 from the command itself. This
5173 has been changed to 127, the value the shell returns if it is asked to run
5174 a non-existent command. The wording for the related log line suggests a
5175 non-existent command as the problem.
5177 26. If received_header_text expands to an empty string, do not add a Received:
5178 header line to the message. (Well, it adds a token one on the spool, but
5179 marks it "old" so that it doesn't get used or transmitted.)
5181 27. Installed eximstats 1.28 (addition of -nt option).
5183 28. There was no check for failure on the call to getsockname() in the daemon
5184 code. This can fail if there is a shortage of resources on the system, with
5185 ENOMEM, for example. A temporary error is now given on failure.
5187 29. Contrary to the C standard, it seems that in some environments, the
5188 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
5189 program. Exim now does this explicitly; it affects the formatting of
5190 timestamps using strftime().
5192 30. If exiqsumm was given junk data, it threw up some uninitialized variable
5193 complaints. I've now initialized all the variables, to avoid this.
5195 32. Header lines added by a system filter were not being "seen" during
5196 transport-time rewrites.
5198 33. The info_callback() function passed to OpenSSL is set up with type void
5199 (*)(SSL *, int, int), as described somewhere. However, when calling the
5200 function (actually a macro) that sets it up, the type void(*)() is
5201 expected. I've put in a cast to prevent warnings from picky compilers.
5203 34. If a DNS black list lookup found a CNAME record, but there were no A
5204 records associated with the domain it pointed at, Exim crashed.
5206 35. If a DNS black list lookup returned more than one A record, Exim ignored
5207 all but the first. It now scans all returned addresses if a particular IP
5208 value is being sought. In this situation, the contents of the
5209 $dnslist_value variable are a list of all the addresses, separated by a
5212 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
5213 to accept a host name and all its aliases if the forward lookup for any of
5214 them yielded the IP address of the incoming connection. Now it accepts only
5215 those names whose forward lookup yields the correct IP address. Any other
5216 names are discarded. This closes a loophole whereby a rogue DNS
5217 administrator could create reverse DNS records to break through a
5218 wildcarded host restriction in an ACL.
5220 37. If a user filter or a system filter that ran in a subprocess used any of
5221 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
5222 the wrong values were passed to the pipe command ($thisaddress had the
5223 value of $0, $0 had the value of $1, etc). This bug was introduced by
5224 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
5226 38. Improved the line breaking for long SMTP error messages from ACLs.
5227 Previously, if there was no break point between 40 and 75 characters, Exim
5228 left the rest of the message alone. Two changes have been made: (a) I've
5229 reduced the minimum length to 35 characters; (b) if it can't find a break
5230 point between 35 and 75 characters, it looks ahead and uses the first one
5231 that it finds. This may give the occasional overlong line, but at least the
5232 remaining text gets split now.
5234 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
5235 file descriptors might be low, and that setting 1000 would always raise it.
5236 It turns out that in some environments, the limit is already over 1000 and
5237 that lowering it causes trouble. So now Exim takes care not to decrease it.
5239 40. When delivering a message, the value of $return_path is set to $sender_
5240 address at the start of routing (routers may change the value). By an
5241 oversight, this default was not being set up when an address was tested by
5242 -bt or -bv, which affected the outcome if any router or filter referred to
5245 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
5246 log only when "message" or "log_message" are set. However, if one of the
5247 conditions was an address verification, or a call to a nested ACL, the
5248 messages generated by the underlying test were being passed through. This
5249 no longer happens. The underlying message is available in $acl_verify_
5250 message for both "message" and "log_message" expansions, so it can be
5251 passed through if needed.
5253 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
5254 new expansion $bh_ to give the encoded byte string without charset
5255 translation. Translation happens only if iconv() is available; HAVE_ICONV
5256 indicates this at build time. HEADERS_CHARSET gives the charset to
5257 translate to; headers_charset can change it in the configuration, and
5258 "headers charset" can change it in an individual filter file.
5260 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
5261 that creates RFC 2047 encoded "words" labels them as that charset instead
5262 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
5263 expansion operator; (ii) when Exim creates a From: line for a local
5264 message; (iii) when a header line is rewritten to include a "phrase" part.
5266 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
5267 buggy, causing it to skip the first lines of messages whose message ID
5268 ended in 'D'. This would not have bitten before Exim release 4.14, because
5269 message IDs were unlikely to end in 'D' before then. The effect was to have
5270 incorrect size information for certain domains.
5272 45. #include "config.h" was missing at the start of the crypt16.c module. This
5273 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
5276 46. If there was a timeout during a "random" callout check, Exim treated it as
5277 a failure of the random address, and carried on sending RSET and the real
5278 address. If the delay was just some slowness somewhere, the response to the
5279 original RCPT would be taken as a response to RSET and so on, causing
5280 mayhem of various kinds.
5282 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
5283 when I implemented it. It didn't allow for the fact that some option values
5284 may legitimatetly be negative (e.g. size_addition), and it didn't even do
5285 the right test for positive values.
5287 48. Domain names in DNS records are case-independent. Exim always looks them up
5288 in lower case. Some resolvers return domain names in exactly the case they
5289 appear in the zone file, that is, they may contain uppercase letters. Not
5290 all resolvers do this - some return always lower case. Exim was treating a
5291 change of case by a resolver as a change of domain, similar to a widening
5292 of a domain abbreviation. This triggered its re-routing code and so it was
5293 trying to route what was effectively the same domain again. This normally
5294 caused routing to fail (because the router wouldn't handle the domain
5295 twice). Now Exim checks for this case specially, and just changes the
5296 casing of the domain that it ultimately uses when it transmits the message
5299 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
5302 50. If a filter generated a file delivery with a non-absolute name (possible if
5303 no home directory exists for the router), the forbid_file option was not
5306 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
5307 the existing equality matching.
5309 52. Exim was using ints instead of ino_t variables in some places where it was
5310 dealing with inode numbers.
5312 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
5313 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
5314 called TMPDIR, and if it finds it is different, it changes its value.
5316 54. The smtp_printf() function is now made available to local_scan() so
5317 additional output lines can be written before returning. There is also an
5318 smtp_fflush() function to enable the detection of a dropped connection.
5319 The variables smtp_input and smtp_batched_input are exported to
5322 55. Changed the default runtime configuration: the message "Unknown user"
5323 has been removed from the ACL, and instead placed on the localuser router,
5324 using the cannot_route_message feature. This means that any verification
5325 failures that generate their own messages won't get overridden. Similarly,
5326 the "Unrouteable address" message that was in the ACL for unverifiable
5327 relay addresses has also been removed.
5329 56. Added hosts_avoid_esmtp to the smtp transport.
5331 57. The exicyclog script was not checking for the esoteric option
5332 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
5333 will work only if exicyclog is run under the appropriate euid.
5335 58. Following a discussion on the list, the rules by which Exim recognises line
5336 endings on incoming messages have been changed. The -dropcr and drop_cr
5337 options are now no-ops, retained only for backwards compatibility. The
5338 following line terminators are recognized: LF CRLF CR. However, special
5339 processing applies to CR:
5341 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
5342 nor a local message in the state where . is a terminator.
5344 (ii) If a bare CR is encountered in a header line, an extra space is added
5345 after the line terminator so as not to end the header. The reasoning
5346 behind this is that bare CRs in header lines are most likely either
5347 to be mistakes, or people trying to play silly games.
5349 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
5350 was being incorrectly given as 18 bytes larger than it should have been.
5351 This is a VOB (very old bug).
5353 60. This may never have affected anything current, but just in case it has:
5354 When the local host is found other than at the start of a list of hosts,
5355 the local host, those with the same MX, and any that follow, are discarded.
5356 When the list in question was part of a longer list of hosts, the following
5357 hosts (not currently being processed) were also being discarded. This no
5358 longer happens. I'm not sure if this situation could ever has previously
5361 61. Added the "/MX" feature to lists of hosts in the manualroute and query
5364 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
5365 header. This is something that is recommended in a new Internet Draft, and
5366 is something that is documented as being done by Sendmail. There are two
5367 possible values. For messages generated by the autoreply transport, Exim
5370 Auto-Submitted: auto-replied
5372 whereas for all other generated messages (e.g. bounces) it adds
5374 Auto-Submitted: auto-generated
5376 63. The "personal" condition in filters now includes a test for the
5377 Auto-Submitted: header. If it contains the string "auto-" the message it
5378 not considered personal.
5380 64. Added rcpt_include_affixes as a generic transport option.
5382 65. Added queue_only_override (default true).
5384 66. Added the syslog_duplication option.
5386 67. If what should have been the first header line of a message consisted of
5387 a space followed by a colon, Exim was mis-interpreting it as a header line.
5388 It isn't of course - it is syntactically invalid and should therefore be
5389 treated as the start of the message body. The misbehaviour could have
5390 caused a number of strange effects, including loss of data in subsequent
5391 header lines, and spool format errors.
5393 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
5394 client host had authenticated. This control can now be exercised by an ACL
5395 for more flexibility.
5397 69. By default, callouts do not happen when testing with -bh. There is now a
5398 variant, -bhc, which does actually run the callout code, including
5399 consulting and updating the callout cache.
5401 70. Added support for saslauthd authentication, courtesy of Alexander
5404 71. If statvfs() failed on the spool or log directories while checking their
5405 size for availability, Exim confusingly gave the error "space shortage".
5406 Furthermore, in debugging mode it crashed with a floating point exception.
5407 These checks are done if check_{spool,log}_{space,inodes} are set, and when
5408 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
5409 serious problem, Exim now writes to the main and panic logs when this
5410 happens, with details of the failure. It then refuses to accept the
5411 incoming message, giving the message "spool directory problem" or "log
5412 directory problem" with a 421 code for SMTP messages.
5414 72. When Exim is about to re-exec itself, it ensures that the file descriptors
5415 0, 1, and 2 exist, because some OS complain for execs without them (see
5416 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
5417 descriptors. However, the code omitted to check that the open succeeded,
5418 causing mysterious errors if for some reason the permissions on /dev/null
5419 got screwed. Now Exim writes a message to the main and panic logs, and
5420 bombs out if it can't open /dev/null.
5422 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
5423 interact so that it is all more flexible. It is supposed to remain
5424 backwards compatible. Also added extra_local_interfaces.
5426 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
5427 to bomb out with an assertion failure - to the client this appears as a
5428 connection drop. This problem occurs in the part of the code that was taken
5429 from the Samba project. Fortunately, the assertion is in a very simple
5430 function, so I have fixed this by reproducing the function inline in the
5431 one place where it is called, and arranging for authentication to fail
5432 instead of killing the process with assert().
5434 75. The SPA client code was not working when the server requested OEM rather
5435 than Unicode encoding.
5437 76. Added code to make require_files with a specific uid setting more usable in
5438 the case where statting the file as root fails - usually a non-root-mounted
5439 NFS file system. When this happens and the failure is EACCES, Exim now
5440 forks a subprocess and does the per-uid checking as the relevant uid.
5442 77. Added process_log_path.
5444 78. If log_file_path was not explicitly set, a setting of check_log_space or
5445 check_log_inodes was ignored.
5447 79. If a space check for the spool or log partitions fails, the incident is now
5448 logged. Of course, in the latter case the data may get lost...
5450 80. Added the %p formatting code to string_format() so that it can be used to
5451 print addresses in debug_print(). Adjusted all the address printing in the
5452 debugging in store.c to use %p rather than %d.
5454 81. There was a concern that a line of code in smtp_in.c could overflow a
5455 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
5456 initially expressed, the concern was not well-founded, because trailing
5457 spaces are removed early. However, if the trailing spaces were followed by
5458 a NULL, they did not get removed, so the overflow was possible. Two fixes
5461 (a) I re-wrote the offending code in a cleaner fashion.
5462 (b) If an incoming SMTP command contains a NULL character, it is rejected
5465 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
5466 runs initgroups(), so that if the Exim user is in any additional groups,
5467 they will be used during message reception.
5473 The change log for 4.20 and earlier releases has been archived.
git://git.exim.org / users / heiko / exim.git / blob