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.
116 PP/01 The new ldap_require_cert option would segfault if used. Fixed.
118 PP/02 Harmonised TLS library version reporting; only show if debugging.
119 Layout now matches that introduced for other libraries in 4.74 PP/03.
121 PP/03 New openssl_options items: no_sslv2 no_sslv3 no_ticket no_tlsv1
123 PP/04 New "dns_use_edns0" global option.
125 PP/05 Don't segfault on misconfiguration of ref:name exim-user as uid.
128 PP/06 Extra paranoia around buffer usage at the STARTTLS transition.
129 nb: Exim is not vulnerable to http://www.kb.cert.org/vuls/id/555316
131 TK/01 Updated PolarSSL code to 0.14.2.
132 Bugzilla 1097. Patch from Andreas Metzler.
134 PP/07 Catch divide-by-zero in ${eval:...}.
137 PP/08 Condition negation of bool{}/bool_lax{} did not negate. Fixed.
140 TK/02 Bugzilla 1106: CVE-2011-1764 - DKIM log line was subject to a
141 format-string attack -- SECURITY: remote arbitrary code execution.
143 TK/03 SECURITY - DKIM signature header parsing was double-expanded, second
144 time unintentionally subject to list matching rules, letting the header
145 cause arbitrary Exim lookups (of items which can occur in lists, *not*
146 arbitrary string expansion). This allowed for information disclosure.
148 PP/09 Fix another SIGFPE (x86) in ${eval:...} expansion, this time related to
149 INT_MIN/-1 -- value coerced to INT_MAX.
155 NM/01 Workround for PCRE version dependency in version reporting
158 TF/01 Update valgrind.h and memcheck.h to copies from valgrind-3.6.0.
159 This fixes portability to compilers other than gcc, notably
160 Solaris CC and HP-UX CC. Fixes Bugzilla 1050.
162 TF/02 Bugzilla 139: Avoid using the += operator in the modular lookup
163 makefiles for portability to HP-UX and POSIX correctness.
165 PP/01 Permit LOOKUP_foo enabling on the make command-line.
166 Also via indented variable definition in the Makefile.
167 (Debugging by Oliver Heesakkers).
169 PP/02 Restore caching of spamd results with expanded spamd_address.
170 Patch from author of expandable spamd_address patch, Wolfgang Breyha.
172 PP/03 Build issue: lookups-Makefile now exports LC_ALL=C
173 Improves build reliability. Fix from: Frank Elsner
175 NM/02 Fix wide character breakage in the rfc2047 coding
176 Fixes bug 1064. Patch from Andrey N. Oktyabrski
178 NM/03 Allow underscore in dnslist lookups
179 Fixes bug 1026. Patch from Graeme Fowler
181 PP/04 Bugzilla 230: Support TLS-enabled LDAP (in addition to ldaps).
182 Code patches from Adam Ciarcinski of NetBSD.
184 NM/04 Fixed exiqgrep to cope with mailq missing size issue
187 PP/05 Bugzilla 1083: when lookup expansion defers, escape the output which
188 is logged, to avoid truncation. Patch from John Horne.
190 PP/06 Bugzilla 1042: implement freeze_signal on pipe transports.
191 Patch from Jakob Hirsch.
193 PP/07 Bugzilla 1061: restrict error messages sent over SMTP to not reveal
194 SQL string expansion failure details.
195 Patch from Andrey Oktyabrski.
197 PP/08 Bugzilla 486: implement %M datestamping in log filenames.
198 Patch from Simon Arlott.
200 PP/09 New lookups functionality failed to compile on old gcc which rejects
201 extern declarations in function scope.
202 Patch from Oliver Fleischmann
204 PP/10 Use sig_atomic_t for flags set from signal handlers.
205 Check getgroups() return and improve debugging.
206 Fixed developed for diagnosis in bug 927 (which turned out to be
209 PP/11 Bugzilla 1055: Update $message_linecount for maildir_tag.
210 Patch from Mark Zealey.
212 PP/12 Bugzilla 1056: Improved spamd server selection.
213 Patch from Mark Zealey.
215 PP/13 Bugzilla 1086: Deal with maildir quota file races.
216 Based on patch from Heiko Schlittermann.
218 PP/14 Bugzilla 1019: DKIM multiple signature generation fix.
219 Patch from Uwe Doering, sign-off by Michael Haardt.
221 NM/05 Fix to spam.c to accommodate older gcc versions which dislike
222 variable declaration deep within a block. Bug and patch from
225 PP/15 lookups-Makefile IRIX compatibilty coercion.
227 PP/16 Make DISABLE_DKIM build knob functional.
229 NM/06 Bugzilla 968: child_open_uid: restore default SIGPIPE handler
230 Patch by Simon Arlott
232 TF/03 Fix valgrind.h portability to C89 compilers that do not support
233 variable argument macros. Our copy now differs from upstream.
239 TF/01 Failure to get a lock on a hints database can have serious
240 consequences so log it to the panic log.
242 TF/02 Log LMTP confirmation messages in the same way as SMTP,
243 controlled using the smtp_confirmation log selector.
245 TF/03 Include the error message when we fail to unlink a spool file.
247 DW/01 Bugzilla 139: Support dynamically loaded lookups as modules.
248 With thanks to Steve Haslam, Johannes Berg & Serge Demonchaux
249 for maintaining out-of-tree patches for some time.
251 PP/01 Bugzilla 139: Documentation and portability issues.
252 Avoid GNU Makefile-isms, let Exim continue to build on BSD.
253 Handle per-OS dynamic-module compilation flags.
255 PP/02 Let /dev/null have normal permissions.
256 The 4.73 fixes were a little too stringent and complained about the
257 permissions on /dev/null. Exempt it from some checks.
258 Reported by Andreas M. Kirchwitz.
260 PP/03 Report version information for many libraries, including
261 Exim version information for dynamically loaded libraries. Created
262 version.h, now support a version extension string for distributors
263 who patch heavily. Dynamic module ABI change.
265 PP/04 CVE-2011-0017 - check return value of setuid/setgid. This is a
266 privilege escalation vulnerability whereby the Exim run-time user
267 can cause root to append content of the attacker's choosing to
270 PP/05 Bugzilla 1041: merged DCC maintainer's fixes for return code.
273 PP/06 Bugzilla 1071: fix delivery logging with untrusted macros.
274 If dropping privileges for untrusted macros, we disabled normal logging
275 on the basis that it would fail; for the Exim run-time user, this is not
276 the case, and it resulted in successful deliveries going unlogged.
277 Fixed. Reported by Andreas Metzler.
283 PP/01 Date: & Message-Id: revert to normally being appended to a message,
284 only prepend for the Resent-* case. Fixes regression introduced in
285 Exim 4.70 by NM/22 for Bugzilla 607.
287 PP/02 Include check_rfc2047_length in configure.default because we're seeing
288 increasing numbers of administrators be bitten by this.
290 JJ/01 Added DISABLE_DKIM and comment to src/EDITME
292 PP/03 Bugzilla 994: added openssl_options main configuration option.
294 PP/04 Bugzilla 995: provide better SSL diagnostics on failed reads.
296 PP/05 Bugzilla 834: provide a permit_coredump option for pipe transports.
298 PP/06 Adjust NTLM authentication to handle SASL Initial Response.
300 PP/07 If TLS negotiated an anonymous cipher, we could end up with SSL but
301 without a peer certificate, leading to a segfault because of an
302 assumption that peers always have certificates. Be a little more
303 paranoid. Problem reported by Martin Tscholak.
305 PP/08 Bugzilla 926: switch ClamAV to use the new zINSTREAM API for content
306 filtering; old API available if built with WITH_OLD_CLAMAV_STREAM=yes
307 NB: ClamAV planning to remove STREAM in "middle of 2010".
308 CL also introduces -bmalware, various -d+acl logging additions and
309 more caution in buffer sizes.
311 PP/09 Implemented reverse_ip expansion operator.
313 PP/10 Bugzilla 937: provide a "debug" ACL control.
315 PP/11 Bugzilla 922: Documentation dusting, patch provided by John Horne.
317 PP/12 Bugzilla 973: Implement --version.
319 PP/13 Bugzilla 752: Refuse to build/run if Exim user is root/0.
321 PP/14 Build without WITH_CONTENT_SCAN. Path from Andreas Metzler.
323 PP/15 Bugzilla 816: support multiple condition rules on Routers.
325 PP/16 Add bool_lax{} expansion operator and use that for combining multiple
326 condition rules, instead of bool{}. Make both bool{} and bool_lax{}
327 ignore trailing whitespace.
329 JJ/02 prevent non-panic DKIM error from being sent to paniclog
331 JJ/03 added tcp_wrappers_daemon_name to allow host entries other than
334 PP/17 Fix malware regression for cmdline scanner introduced in PP/08.
335 Notification from Dr Andrew Aitchison.
337 PP/18 Change ClamAV response parsing to be more robust and to handle ClamAV's
338 ExtendedDetectionInfo response format.
339 Notification from John Horne.
341 PP/19 OpenSSL 1.0.0a compatibility const-ness change, should be backwards
344 PP/20 Added a CONTRIBUTING file. Fixed the documentation build to use http:
345 XSL and documented dependency on system catalogs, with examples of how
348 DW/21 Added Valgrind hooks in store.c to help it capture out-of-bounds store
351 DW/22 Bugzilla 1044: CVE-2010-4345 - partial fix: restrict default behaviour
352 of CONFIGURE_OWNER and CONFIGURE_GROUP options to no longer allow a
353 configuration file which is writeable by the Exim user or group.
355 DW/23 Bugzilla 1044: CVE-2010-4345 - part two: extend checks for writeability
356 of configuration files to cover files specified with the -C option if
357 they are going to be used with root privileges, not just the default
360 DW/24 Bugzilla 1044: CVE-2010-4345 - part three: remove ALT_CONFIG_ROOT_ONLY
361 option (effectively making it always true).
363 DW/25 Add TRUSTED_CONFIG_PREFIX_FILE option to allow alternative configuration
364 files to be used while preserving root privileges.
366 DW/26 Set FD_CLOEXEC on SMTP sockets after forking in the daemon, to ensure
367 that rogue child processes cannot use them.
369 PP/27 Bugzilla 1047: change the default for system_filter_user to be the Exim
370 run-time user, instead of root.
372 PP/28 Add WHITELIST_D_MACROS option to let some macros be overridden by the
373 Exim run-time user without dropping privileges.
375 DW/29 Remove use of va_copy() which breaks pre-C99 systems. Duplicate the
376 result string, instead of calling string_vformat() twice with the same
379 DW/30 Allow TRUSTED_CONFIG_PREFIX_FILE only for Exim or CONFIGURE_OWNER, not
380 for other users. Others should always drop root privileges if they use
381 -C on the command line, even for a whitelisted configure file.
383 DW/31 Turn TRUSTED_CONFIG_PREFIX_FILE into TRUSTED_CONFIG_FILE. No prefixes.
385 NM/01 Fixed bug #1002 - Message loss when using multiple deliveries
391 JJ/01 installed exipick 20100104.1, adding $max_received_linelength,
392 $data_path, and $header_path variables; fixed documentation bugs and
395 JJ/02 installed exipick 20100222.0, added --input-dir and --finput to allow
396 exipick to access non-standard spools, including the "frozen" queue
399 NM/01 Bugzilla 965: Support mysql stored procedures.
400 Patch from Alain Williams
402 NM/02 Bugzilla 961: Spacing fix (syntax error) on Makefile directives for NetBSD
404 NM/03 Bugzilla 955: Documentation fix for max_rcpts.
405 Patch from Andreas Metzler
407 NM/04 Bugzilla 954: Fix for unknown responses from Dovecot authenticator.
408 Patch from Kirill Miazine
410 NM/05 Bugzilla 671: Added umask to procmail example.
412 JJ/03 installed exipick 20100323.0, fixing doc bug
414 NM/06 Bugzilla 988: CVE-2010-2023 - prevent hardlink attack on sticky mail
415 directory. Notification and patch from Dan Rosenberg.
417 TK/01 PDKIM: Upgrade PolarSSL files to upstream version 0.12.1.
419 TK/02 Improve log output when DKIM signing operation fails.
421 MH/01 Treat the transport option dkim_domain as a colon separated
422 list, not as a single string, and sign the message with each element,
423 omitting multiple occurences of the same signer.
425 NM/07 Null terminate DKIM strings, Null initialise DKIM variable
426 Bugzilla 985, 986. Patch by Simon Arlott
428 NM/08 Bugzilla 967. dnsdb DNS TXT record bug fix (DKIM-related)
429 Patch by Simon Arlott
431 PP/01 Bugzilla 989: CVE-2010-2024 - work round race condition on
432 MBX locking. Notification from Dan Rosenberg.
438 TK/01 Bugzilla 912: Fix DKIM segfault on empty headers/body.
440 NM/01 Bugzilla 913: Documentation fix for gnutls_* options.
442 NM/02 Bugzilla 722: Documentation for randint. Better randomness defaults.
444 NM/03 Bugzilla 847: Enable DNSDB lookup by default.
446 NM/04 Bugzilla 915: Flag broken perl installation during build.
452 TK/01 Added patch by Johannes Berg that expands the main option
453 "spamd_address" if it starts with a dollar sign.
455 TK/02 Write list of recipients to X-Envelope-Sender header when building
456 the mbox-format spool file for content scanning (suggested by Jakob
459 TK/03 Added patch by Wolfgang Breyha that adds experimental DCC
460 (http://www.dcc-servers.net/) support via dccifd. Activated by
461 setting EXPERIMENTAL_DCC=yes in Local/Makefile.
463 TK/04 Bugzilla 673: Add f-protd malware scanner support. Patch submitted
464 by Mark Daniel Reidel <mr@df.eu>.
466 NM/01 Bugzilla 657: Embedded PCRE removed from the exim source tree.
467 When building exim an external PCRE library is now needed -
468 PCRE is a system library on the majority of modern systems.
469 See entry on PCRE_LIBS in EDITME file.
471 NM/02 Bugzilla 646: Removed unwanted C/R in Dovecot authenticator
472 conversation. Added nologin parameter to request.
473 Patch contributed by Kirill Miazine.
475 TF/01 Do not log submission mode rewrites if they do not change the address.
477 TF/02 Bugzilla 662: Fix stack corruption before exec() in daemon.c.
479 NM/03 Bugzilla 602: exicyclog now handles panic log, and creates empty
480 log files in place. Contributed by Roberto Lima.
482 NM/04 Bugzilla 667: Close socket used by dovecot authenticator.
484 TF/03 Bugzilla 615: When checking the local_parts router precondition
485 after a local_part_suffix or local_part_prefix option, Exim now
486 does not use the address's named list lookup cache, since this
487 contains cached lookups for the whole local part.
489 NM/05 Bugzilla 521: Integrated SPF Best Guess support contributed by
490 Robert Millan. Documentation is in experimental-spec.txt.
492 TF/04 Bugzilla 668: Fix parallel build (make -j).
494 NM/05.2 Bugzilla 437: Prevent Maildir aux files being created with mode 000.
496 NM/05.3 Bugzilla 598: Improvement to Dovecot authenticator handling.
497 Patch provided by Jan Srzednicki.
499 TF/05 Leading white space used to be stripped from $spam_report which
500 wrecked the formatting. Now it is preserved.
502 TF/06 Save $spam_score, $spam_bar, and $spam_report in spool files, so
503 that they are available at delivery time.
505 TF/07 Fix the way ${extract is skipped in the untaken branch of a conditional.
507 TF/08 TLS error reporting now respects the incoming_interface and
508 incoming_port log selectors.
510 TF/09 Produce a more useful error message if an SMTP transport's hosts
511 setting expands to an empty string.
513 NM/06 Bugzilla 744: EXPN did not work under TLS.
514 Patch provided by Phil Pennock.
516 NM/07 Bugzilla 769: Extraneous comma in usage fprintf
517 Patch provided by Richard Godbee.
519 NM/08 Fixed erroneous documentation references to smtp_notquit_acl to be
520 acl_smtp_notquit, added index entry.
522 NM/09 Bugzilla 787: Potential buffer overflow in string_format.
523 Patch provided by Eugene Bujak.
525 NM/10 Bugzilla 770: Problem on some platforms modifying the len parameter to
526 accept(). Patch provided by Maxim Dounin.
528 NM/11 Bugzilla 749: Preserve old behaviour of blanks comparing equal to zero.
529 Patch provided by Phil Pennock.
531 NM/12 Bugzilla 497: Correct behaviour of exiwhat when no config exists.
533 NM/13 Bugzilla 590: Correct handling of Resent-Date headers.
534 Patch provided by Brad "anomie" Jorsch.
536 NM/14 Bugzilla 622: Added timeout setting to transport filter.
537 Patch provided by Dean Brooks.
539 TK/05 Add native DKIM support (does not depend on external libraries).
541 NM/15 Bugzilla 854: Removed code that symlinks to pcre as its no longer useful.
542 Patch provided by Graeme Fowler.
544 NM/16 Bugzilla 851: Documentation example syntax fix.
546 NM/17 Changed NOTICE file to remove references to embedded PCRE.
548 NM/18 Bugzilla 894: Fix issue with very long lines including comments in
551 NM/19 Bugzilla 745: TLS version reporting.
552 Patch provided by Phil Pennock.
554 NM/20 Bugzilla 167: bool: condition support.
555 Patch provided by Phil Pennock.
557 NM/21 Bugzilla 665: gnutls_compat_mode to allow compatibility with broken
558 clients. Patch provided by Phil Pennock.
560 NM/22 Bugzilla 607: prepend (not append) Resent-Message-ID and Resent-Date.
561 Patch provided by Brad "anomie" Jorsch.
563 NM/23 Bugzilla 687: Fix misparses in eximstats.
564 Patch provided by Heiko Schlittermann.
566 NM/24 Bugzilla 688: Fix exiwhat to handle log_selector = +pid.
567 Patch provided by Heiko Schlittermann.
569 NM/25 Bugzilla 727: Use transport mode as default mode for maildirsize file.
570 plus update to original patch.
572 NM/26 Bugzilla 799: Documentation correction for ratelimit.
574 NM/27 Bugzilla 802: Improvements to local interface IP addr detection.
575 Patch provided by David Brownlee.
577 NM/28 Bugzilla 807: Improvements to LMTP delivery logging.
579 NM/29 Bugzilla 862, 866, 875: Documentation bugfixes.
581 NM/30 Bugzilla 888: TLS documentation bugfixes.
583 NM/31 Bugzilla 896: Dovecot buffer overrun fix.
585 NM/32 Bugzilla 889: Change all instances of "expr" in shell scripts to "expr --"
586 Unlike the original bugzilla I have changed all shell scripts in src tree.
588 NM/33 Bugzilla 898: Transport filter timeout fix.
589 Patch by Todd Rinaldo.
591 NM/34 Bugzilla 901: Fix sign/unsigned and UTF mistmatches.
592 Patch by Serge Demonchaux.
594 NM/35 Bugzilla 39: Base64 decode bug fixes.
595 Patch by Jakob Hirsch.
597 NM/36 Bugzilla 909: Correct connect() call in dcc code.
599 NM/37 Bugzilla 910: Correct issue with relaxed/simple handling.
601 NM/38 Bugzilla 908: Removed NetBSD3 support as no longer needed.
603 NM/39 Bugzilla 911: Fixed MakeLinks build script.
609 TK/01 Add preliminary DKIM support. Currently requires a forked version of
610 ALT-N's libdkim that I have put here:
611 http://duncanthrax.net/exim-experimental/
613 Note to Michael Haardt: I had to rename some vars in sieve.c. They
614 were called 'true' and it seems that C99 defines that as a reserved
615 keyword to be used with 'bool' variable types. That means you could
616 not include C99-style headers which use bools without triggering
617 build errors in sieve.c.
619 NM/01 Bugzilla 592: --help option is handled incorrectly if exim is invoked
620 as mailq or other aliases. Changed the --help handling significantly
621 to do whats expected. exim_usage() emits usage/help information.
623 SC/01 Added the -bylocaldomain option to eximstats.
625 NM/02 Bugzilla 619: Defended against bad data coming back from gethostbyaddr.
627 NM/03 Bugzilla 613: Documentation fix for acl_not_smtp.
629 NM/04 Bugzilla 628: PCRE update to 7.4 (work done by John Hall).
635 PH/01 Another patch from the Sieve maintainer.
637 PH/02 When an IPv6 address is converted to a string for single-key lookup
638 in an address list (e.g. for an item such as "net24-dbm;/net/works"),
639 dots are used instead of colons so that keys in lsearch files need not
640 contain colons. This was done some time before quoting was made available
641 in lsearch files. However, iplsearch files do require colons in IPv6 keys
642 (notated using the quote facility) so as to distinguish them from IPv4
643 keys. This meant that lookups for IP addresses in host lists did not work
644 for iplsearch lookups.
646 This has been fixed by arranging for IPv6 addresses to be expressed with
647 colons if the lookup type is iplsearch. This is not incompatible, because
648 previously such lookups could never work.
650 The situation is now rather anomolous, since one *can* have colons in
651 ordinary lsearch keys. However, making the change in all cases is
652 incompatible and would probably break a number of configurations.
654 TK/01 Change PRVS address formatting scheme to reflect latests BATV draft
657 MH/01 The "spam" ACL condition code contained a sscanf() call with a %s
658 conversion specification without a maximum field width, thereby enabling
659 a rogue spamd server to cause a buffer overflow. While nobody in their
660 right mind would setup Exim to query an untrusted spamd server, an
661 attacker that gains access to a server running spamd could potentially
662 exploit this vulnerability to run arbitrary code as the Exim user.
664 TK/02 Bugzilla 502: Apply patch to make the SPF-Received: header use
665 $primary_hostname instead of what libspf2 thinks the hosts name is.
667 MH/02 The dsearch lookup now uses lstat(2) instead of stat(2) to look for
668 a directory entry by the name of the lookup key. Previously, if a
669 symlink pointed to a non-existing file or a file in a directory that
670 Exim lacked permissions to read, a lookup for a key matching that
671 symlink would fail. Now it is enough that a matching directory entry
672 exists, symlink or not. (Bugzilla 503.)
674 PH/03 The body_linecount and body_zerocount variables are now exported in the
677 PH/04 Added the $dnslist_matched variable.
679 PH/05 Unset $tls_cipher and $tls_peerdn before making a connection as a client.
680 This means they are set thereafter only if the connection becomes
683 PH/06 Added the client_condition to authenticators so that some can be skipped
684 by clients under certain conditions.
686 PH/07 The error message for a badly-placed control=no_multiline_responses left
687 "_responses" off the end of the name.
689 PH/08 Added -Mvc to output a copy of a message in RFC 2822 format.
691 PH/09 Tidied the code for creating ratelimiting keys, creating them explicitly
692 (without spaces) instead of just copying the configuration text.
694 PH/10 Added the /noupdate option to the ratelimit ACL condition.
696 PH/11 Added $max_received_linelength.
698 PH/12 Added +ignore_defer and +include_defer to host lists.
700 PH/13 Installed PCRE version 7.2. This needed some changes because of the new
701 way in which PCRE > 7.0 is built.
703 PH/14 Implemented queue_only_load_latch.
705 PH/15 Removed an incorrect (int) cast when reading the value of SIZE in a
706 MAIL command. The effect was to mangle the value on 64-bit systems.
708 PH/16 Another patch from the Sieve maintainer.
710 PH/17 Added the NOTQUIT ACL, based on a patch from Ted Cooper.
712 PH/18 If a system quota error occurred while trying to create the file for
713 a maildir delivery, the message "Mailbox is full" was not appended to the
714 bounce if the delivery eventually timed out. Change 4.67/27 below applied
715 only to a quota excession during the actual writing of the file.
717 PH/19 It seems that peer DN values may contain newlines (and other non-printing
718 characters?) which causes problems in log lines. The DN values are now
719 passed through string_printing() before being added to log lines.
721 PH/20 Added the "servers=" facility to MySQL and PostgreSQL lookups. (Oracle
722 and InterBase are left for another time.)
724 PH/21 Added message_body_newlines option.
726 PH/22 Guard against possible overflow in moan_check_errorcopy().
728 PH/23 POSIX allows open() to be a macro; guard against that.
730 PH/24 If the recipient of an error message contained an @ in the local part
731 (suitably quoted, of course), incorrect values were put in $domain and
732 $local_part during the evaluation of errors_copy.
738 MH/01 Fix for bug #448, segfault in Dovecot authenticator when interface_address
739 is unset (happens when testing with -bh and -oMi isn't used). Thanks to
742 PH/01 Added a new log selector smtp_no_mail, to log SMTP sessions that do not
743 issue a MAIL command.
745 PH/02 In an ACL statement such as
747 deny dnslists = X!=127.0.0.2 : X=127.0.0.2
749 if a client was not listed at all, or was listed with a value other than
750 127.0.0.2, in the X list, but was listed with 127.0.0.2 in the Y list,
751 the condition was not true (as it should be), so access was not denied.
752 The bug was that the ! inversion was incorrectly passed on to the second
753 item. This has been fixed.
755 PH/03 Added additional dnslists conditions == and =& which are different from
756 = and & when the dns lookup returns more than one IP address.
758 PH/04 Added gnutls_require_{kx,mac,protocols} to give more control over the
759 cipher suites used by GnuTLS. These options are ignored by OpenSSL.
761 PH/05 After discussion on the list, added a compile time option ENABLE_DISABLE_
762 FSYNC, which compiles an option called disable_fsync that allows for
763 bypassing fsync(). The documentation is heavily laced with warnings.
765 SC/01 Updated eximstats to collate all SpamAssassin rejects into one bucket.
767 PH/06 Some tidies to the infrastructure of the Test Suite that is concerned
768 with the auxiliary C programs that it uses: (1) Arrange for BIND_8_COMPAT
769 to be defined when compiling on OSX (Darwin); (2) Tidies to the Makefile,
770 including adding "make clean"; (3) Added -fPIC when compiling the test
771 dynamically loaded module, to get rid of a warning.
773 MH/02 Fix for bug #451, causing paniclog entries to be written if a bounce
774 message fails, move_frozen_messages = true and ignore_bounce_errors_after
775 = 0s. The bug is otherwise harmless.
777 PH/07 There was a bug in the dovecot authenticator such that the value of
778 $auth1 could be overwritten, and so not correctly preserved, after a
779 successful authentication. This usually meant that the value preserved by
780 the server_setid option was incorrect.
782 PH/08 Added $smtp_count_at_connection_start, deliberately with a long name.
784 PH/09 Installed PCRE release 7.0.
786 PH/10 The acl_not_smtp_start ACL was, contrary to the documentation, not being
787 run for batched SMTP input. It is now run at the start of every message
788 in the batch. While fixing this I discovered that the process information
789 (output by running exiwhat) was not always getting set for -bs and -bS
790 input. This is fixed, and it now also says "batched" for BSMTP.
792 PH/11 Added control=no_pipelining.
794 PH/12 Added $sending_ip_address and $sending_port (mostly Magnus Holmgren's
795 patch, slightly modified), and move the expansion of helo_data till after
796 the connection is made in the smtp transport (so it can use these
799 PH/13 Added ${rfc2047d: to decoded RFC 2047 strings.
801 PH/14 Added log_selector = +pid.
803 PH/15 Flush SMTP output before delaying, unless control=no_delay_flush is set.
805 PH/16 Add ${if forany and ${if forall.
807 PH/17 Added dsn_from option to vary the From: line in DSNs.
809 PH/18 Flush SMTP output before performing a callout, unless control =
810 no_callout_flush is set.
812 PH/19 Change 4.64/PH/36 introduced a bug: when address_retry_include_sender
813 was true (the default) a successful delivery failed to delete the retry
814 item, thus causing premature timeout of the address. The bug is now
817 PH/20 Added hosts_avoid_pipelining to the smtp transport.
819 PH/21 Long custom messages for fakedefer and fakereject are now split up
820 into multiline reponses in the same way that messages for "deny" and
821 other ACL rejections are.
823 PH/22 Applied Jori Hamalainen's speed-up changes and typo fixes to exigrep,
824 with slight modification.
826 PH/23 Applied sieve patches from the maintainer "tracking the latest notify
827 draft, changing the syntax and factoring some duplicate code".
829 PH/24 When the log selector "outgoing_port" was set, the port was shown as -1
830 for deliveries of the second and subsequent messages over the same SMTP
833 PH/25 Applied Magnus Holmgren's patch for ${addresses, ${map, ${filter, and
834 ${reduce, with only minor "tidies".
836 SC/02 Applied Daniel Tiefnig's patch to improve the '($parent) =' pattern match.
838 PH/26 Added a "continue" ACL modifier that does nothing, for the benefit of its
839 expansion side effects.
841 PH/27 When a message times out after an over-quota error from an Exim-imposed
842 quota, the bounce message says "mailbox is full". This message was not
843 being given when it was a system quota that was exceeded. It now should
846 MH/03 Made $recipients available in local_scan(). local_scan() already has
847 better access to the recipient list through recipients_list[], but
848 $recipients can be useful in postmaster-provided expansion strings.
850 PH/28 The $smtp_command and $smtp_command_argument variables were not correct
851 in the case of a MAIL command with additional options following the
852 address, for example: MAIL FROM:<foo@bar> SIZE=1234. The option settings
853 were accidentally chopped off.
855 PH/29 SMTP synchronization checks are implemented when a command is read -
856 there is a check that no more input is waiting when there shouldn't be
857 any. However, for some commands, a delay in an ACL can mean that it is
858 some time before the response is written. In this time, more input might
859 arrive, invalidly. So now there are extra checks after an ACL has run for
860 HELO/EHLO and after the predata ACL, and likewise for MAIL and RCPT when
861 pipelining has not been advertised.
863 PH/30 MH's patch to allow iscntrl() characters to be list separators.
865 PH/31 Unlike :fail:, a custom message specified with :defer: was not being
866 returned in the SMTP response when smtp_return_error_details was false.
869 PH/32 Change the Dovecot authenticator to use read() and write() on the socket
870 instead of the C I/O that was originally supplied, because problems were
873 PH/33 Compile failed with OpenSSL 0.9.8e. This was due to a coding error in
874 Exim which did not show up earlier: it was assuming that a call to
875 SSL_CTX_set_info_callback() might give an error value. In fact, there is
876 no error. In previous releases of OpenSSL, SSL_CTX_set_info_callback()
877 was a macro that became an assignment, so it seemed to work. This has
878 changed to a proper function call with a void return, hence the compile
879 error. Exim's code has been fixed.
881 PH/34 Change HDA_SIZE in oracle.c from 256 to 512. This is needed for 64-bit
884 PH/35 Applied a patch from the Sieve maintainer which fixes a bug in "notify".
886 PH/36 Applied John Jetmore's patch to add -v functionality to exigrep.
888 PH/37 If a message is not accepted after it has had an id assigned (e.g.
889 because it turns out to be too big or there is a timeout) there is no
890 "Completed" line in the log. When some messages of this type were
891 selected by exigrep, they were listed as "not completed". Others were
892 picked up by some special patterns. I have improved the selection
893 criteria to be more general.
895 PH/38 The host_find_failed option in the manualroute router can now be set
896 to "ignore", to completely ignore a host whose IP address cannot be
897 found. If all hosts are ignored, the behaviour is controlled by the new
898 host_all_ignored option.
900 PH/39 In a list of hosts for manualroute, if one item (either because of multi-
901 homing or because of multiple MX records with /mx) generated more than
902 one IP address, and the following item turned out to be the local host,
903 all the secondary addresses of the first item were incorrectly removed
904 from the list, along with the local host and any following hosts (which
905 is what is supposed to happen).
907 PH/40 When Exim receives a message, it writes the login name, uid, and gid of
908 whoever called Exim into the -H file. In the case of the daemon it was
909 behaving confusingly. When first started, it used values for whoever
910 started the daemon, but after a SIGHUP it used the Exim user (because it
911 calls itself on a restart). I have changed the code so that it now always
914 PH/41 (Following a suggestion from Tony Finch) If all the RCPT commands in a
915 message are rejected with the same error (e.g. no authentication or bad
916 sender address), and a DATA command is nevertheless sent (as can happen
917 with PIPELINING or a stupid MUA), the error message that was given to the
918 RCPT commands is included in the rejection of the DATA command. This is
919 intended to be helpful for MUAs that show only the final error to their
922 PH/42 Another patch from the Sieve maintainer.
924 SC/02 Eximstats - Differentiate between permanent and temporary rejects.
925 Eximstats - Fixed some broken HTML links and added missing column headers
927 Eximstats - Fixed Grand Total Summary Domains, Edomains, and Email
928 columns for Rejects, Temp Rejects, Ham, and Spam rows.
930 SC/03 Eximstats - V1.58 Fix to get <> and blackhole to show in edomain tables.
932 PH/43 Yet another patch from the Sieve maintainer.
934 PH/44 I found a way to check for a TCP/IP connection going away before sending
935 the response to the final '.' that terminates a message, but only in the
936 case where the client has not sent further data following the '.'
937 (unfortunately, this is allowed). However, in many cases there won't be
938 any further data because there won't be any more messages to send. A call
939 to select() can be used: if it shows that the input is "ready", there is
940 either input waiting, or the socket has been closed. An attempt to read
941 the next input character can distinguish the two cases. Previously, Exim
942 would have sent an OK response which the client would never have see.
943 This could lead to message repetition. This fix should cure that, at
944 least in a lot of common cases.
946 PH/45 Do not advertise STARTTLS in response to HELP unless it would be
947 advertised in response to EHLO.
953 PH/01 Two more bugs that were introduced by 4.64/PH/07, in addition to the one
954 fixed by 4.65/MH/01 (is this a record?) are fixed:
956 (i) An empty string was always treated as zero by the numeric comparison
957 operators. This behaviour has been restored.
959 (ii) It is documented that the numeric comparison operators always treat
960 their arguments as decimal numbers. This was broken in that numbers
961 starting with 0 were being interpreted as octal.
963 While fixing these problems I realized that there was another issue that
964 hadn't been noticed. Values of message_size_limit (both the global option
965 and the transport option) were treated as octal if they started with 0.
966 The documentation was vague. These values are now always treated as
967 decimal, and I will make that clear in the documentation.
973 TK/01 Disable default definition of HAVE_LINUX_SENDFILE. Clashes with
974 Linux large file support (_FILE_OFFSET_BITS=64) on older glibc
977 MH/01 Don't check that the operands of numeric comparison operators are
978 integers when their expansion is in "skipping" mode (fixes bug
979 introduced by 4.64-PH/07).
981 PH/01 If a system filter or a router generates more than SHRT_MAX (32767)
982 child addresses, Exim now panics and dies. Previously, because the count
983 is held in a short int, deliveries were likely to be lost. As such a
984 large number of recipients for a single message is ridiculous
985 (performance will be very, very poor), I have chosen to impose a limit
986 rather than extend the field.
992 TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
993 leftover -K file (the existence of which was triggered by #402).
994 While we were at it, introduced process PID as part of the -K
995 filename. This should rule out race conditions when creating
998 TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
999 processing considerably. Previous code took too long for large mails,
1000 triggering a timeout which in turn triggers #401.
1002 TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
1003 in the DK code in transports.c. sendfile() is not really portable,
1004 hence the _LINUX specificness.
1006 TF/01 In the add_headers option to the mail command in an Exim filter,
1007 there was a bug that Exim would claim a syntax error in any
1008 header after the first one which had an odd number of characters
1011 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
1012 callout verification, Exim cached a "reject" for the entire domain. This
1013 is correct for most verifications, but it is not correct for a recipient
1014 verification with use_sender or use_postmaster set, because in that case
1015 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
1016 case of MAIL FROM:<> rejection from other early rejections (e.g.
1017 rejection of HELO). When verifying a recipient using a non-null MAIL
1018 address, the cache is ignored if it shows MAIL FROM:<> rejection.
1019 Whatever the result of the callout, the value of the domain cache is
1020 left unchanged (for any other kind of callout, getting as far as trying
1021 RCPT means that the domain itself is ok).
1023 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
1026 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
1027 manifest itself as EPIPE rather than ECONNECT. When tidying away a
1028 session, the daemon ignores ECONNECT errors and logs others; it now
1029 ignores EPIPE as well.
1031 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
1032 (quoted-printable decoding).
1034 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
1035 later the small subsequent patch to fix an introduced bug.
1037 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
1039 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
1041 PH/08 An error is now given if message_size_limit is specified negative.
1043 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
1044 to be given (somewhat) arbitrary names.
1046 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
1049 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
1050 miscellaneous code fixes
1052 PH/10 Added the log_reject_target ACL modifier to specify where to log
1055 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
1056 hostname. This is wrong, because it relates to the incoming message (and
1057 probably the interface on which it is arriving) and not to the outgoing
1058 callout (which could be using a different interface). This has been
1059 changed to use the value of the helo_data option from the smtp transport
1060 instead - this is what is used when a message is actually being sent. If
1061 there is no remote transport (possible with a router that sets up host
1062 addresses), $smtp_active_hostname is used.
1064 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
1065 tweaks were necessary in order to get it to work (see also 21 below):
1066 (a) The code assumed that strncpy() returns a negative number on buffer
1067 overflow, which isn't the case. Replaced with Exim's string_format()
1069 (b) There were several signed/unsigned issues. I just did the minimum
1070 hacking in of casts. There is scope for a larger refactoring.
1071 (c) The code used strcasecmp() which is not a standard C function.
1072 Replaced with Exim's strcmpic() function.
1073 (d) The code set only $1; it now sets $auth1 as well.
1074 (e) A simple test gave the error "authentication client didn't specify
1075 service in request". It would seem that Dovecot has changed its
1076 interface. Fortunately there's a specification; I followed it and
1077 changed what the client sends and it appears to be working now.
1079 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
1082 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
1083 address A is aliased to B and C, where B exists and C does not. Without
1084 -v the output is "A verified" because verification stops after a
1085 successful redirection if more than one address is generated. However,
1086 with -v the child addresses are also verified. Exim was outputting "A
1087 failed to verify" and then showing the successful verification for C,
1088 with its parentage. It now outputs "B failed to verify", showing B's
1089 parentage before showing the successful verification of C.
1091 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
1092 look up a TXT record in a specific list after matching in a combined
1095 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
1096 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
1097 they consult the DNS. I had assumed they would set it the way they
1098 wanted; and indeed my experiments on Linux seem to show that in some
1099 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
1100 To be on the safe side, however, I have now made the interface to
1101 host_find_byname() similar to host_find_bydns(), with an argument
1102 containing the DNS resolver options. The host_find_byname() function now
1103 sets these options at its start, just as host_find_bydns() does. The smtp
1104 transport options dns_qualify_single and dns_search_parents are passed to
1105 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
1106 of host_find_byname() use the default settings of RES_DEFNAMES
1107 (qualify_single) but not RES_DNSRCH (search_parents).
1109 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
1110 spool_read_header() do less string testing, by means of a preliminary
1111 switch on the second character of optional "-foo" lines. (This is
1112 overdue, caused by the large number of possibilities that now exist.
1113 Originally there were few.) While I was there, I also converted the
1114 str(n)cmp tests so they don't re-test the leading "-" and the first
1115 character, in the hope this might squeeze out yet more improvement.
1117 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
1118 flag allowing group syntax was set by the header_syntax check but not
1119 turned off, possible causing trouble later; (2) The flag was not being
1120 set at all for the header_verify test, causing "group"-style headers to
1121 be rejected. I have now set it in this case, and also caused header_
1122 verify to ignore an empty address taken from a group. While doing this, I
1123 came across some other cases where the code for allowing group syntax
1124 while scanning a header line wasn't quite right (mostly, not resetting
1125 the flag correctly in the right place). These bugs could have caused
1126 trouble for malformed header lines. I hope it is now all correct.
1128 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
1129 with the "reply" argument non-NULL. The code, however (which originally
1130 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
1131 but it didn't always do it. This confused somebody who was copying the
1132 code for some other use. I have removed all the tests.
1134 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
1135 feature that was used to support insecure browsers during the U.S. crypto
1136 embargo. It requires special client support, and Exim is probably the
1137 only MTA that supported it -- and would never use it because real RSA is
1138 always available. This code has been removed, because it had the bad
1139 effect of slowing Exim down by computing (never used) parameters for the
1140 RSA_EXPORT functionality.
1142 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
1143 authenticator to fail if there's a tab character in the incoming data
1144 (there should never be unless someone is messing about, as it's supposed
1145 to be base64-encoded). Also added, on Timo's advice, the "secured" option
1146 if the connection is using TLS or if the remote IP is the same as the
1147 local IP, and the "valid-client-cert option" if a client certificate has
1150 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
1151 authenticators. This can be used for authorization after authentication
1152 succeeds. (In the case of plaintext, it servers for both authentication
1155 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
1156 if any retry times were supplied.
1158 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
1159 connection, where there is no client IP address to check. In this
1160 situation, the verify now always succeeds.
1162 PH/25 Applied John Jetmore's -Mset patch.
1164 PH/26 Added -bem to be like -Mset, but loading a message from a file.
1166 PH/27 In a string expansion for a processed (not raw) header when multiple
1167 headers of the same name were present, leading whitespace was being
1168 removed from all of them, but trailing whitespace was being removed only
1169 from the last one. Now trailing whitespace is removed from each header
1170 before concatenation. Completely empty headers in a concatenation (as
1171 before) are ignored.
1173 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
1174 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
1176 PH/29 [Removed. This was a change that I later backed out, and forgot to
1177 correct the ChangeLog entry (that I had efficiently created) before
1178 committing the later change.]
1180 PH/30 Exim was sometimes attempting to deliver messages that had suffered
1181 address errors (4xx response to RCPT) over the same connection as other
1182 messages routed to the same hosts. Such deliveries are always "forced",
1183 so retry times are not inspected. This resulted in far too many retries
1184 for the affected addresses. The effect occurred only when there were more
1185 hosts than the hosts_max_try setting in the smtp transport when it had
1186 the 4xx errors. Those hosts that it had tried were not added to the list
1187 of hosts for which the message was waiting, so if all were tried, there
1188 was no problem. Two fixes have been applied:
1190 (i) If there are any address or message errors in an SMTP delivery, none
1191 of the hosts (tried or untried) are now added to the list of hosts
1192 for which the message is waiting, so the message should not be a
1193 candidate for sending over the same connection that was used for a
1194 successful delivery of some other message. This seems entirely
1195 reasonable: after all the message is NOT "waiting for some host".
1196 This is so "obvious" that I'm not sure why it wasn't done
1197 previously. Hope I haven't missed anything, but it can't do any
1198 harm, as the worst effect is to miss an optimization.
1200 (ii) If, despite (i), such a delivery is accidentally attempted, the
1201 routing retry time is respected, so at least it doesn't keep
1202 hammering the server.
1204 PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
1205 in ${readsocket because some servers need this prod.
1207 PH/32 Added some extra debug output when updating a wait-xxx database.
1209 PH/33 The hint "could be header name not terminated by colon", which has been
1210 given for certain expansion errors for a long time, was not being given
1211 for the ${if def:h_colon_omitted{... case.
1213 PH/34 The spec says: "With one important exception, whenever a domain list is
1214 being scanned, $domain contains the subject domain." There was at least
1215 one case where this was not true.
1217 PH/35 The error "getsockname() failed: connection reset by peer" was being
1218 written to the panic log as well as the main log, but it isn't really
1219 panic-worthy as it just means the connection died rather early on. I have
1220 removed the panic log writing for the ECONNRESET error when getsockname()
1223 PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
1224 runs only) independently of the message's sender address. This meant
1225 that, if the 4xx error was in fact related to the sender, a different
1226 message to the same recipient with a different sender could confuse
1227 things. In particualar, this can happen when sending to a greylisting
1228 server, but other circumstances could also provoke similar problems.
1229 I have changed the default so that the retry time for these errors is now
1230 based a combination of the sender and recipient addresses. This change
1231 can be overridden by setting address_retry_include_sender=false in the
1234 PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
1235 remote server are returned as part of bounce messages. This was not
1236 happening for LMTP over a pipe (the lmtp transport), but now it is the
1237 same for both kinds of LMTP.
1239 PH/38 Despite being documented as not happening, Exim was rewriting addresses
1240 in header lines that were in fact CNAMEs. This is no longer the case.
1242 PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
1243 and queue runs started by the daemon processed all messages. This has
1244 been fixed so that -R and -S can now usefully be given with -q<time>.
1246 PH/40 Import PCRE release 6.7 (fixes some bugs).
1248 PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
1250 PH/42 Give an error if -q is specified more than once.
1252 PH/43 Renamed the variables $interface_address and $interface_port as
1253 $received_ip_address and $received_port, to make it clear that these
1254 values apply to message reception, and not to the outgoing interface when
1255 a message is delivered. (The old names remain recognized, of course.)
1257 PH/44 There was no timeout on the connect() call when using a Unix domain
1258 socket in the ${readsocket expansion. There now is.
1260 PH/45 Applied a modified version of Brad Jorsch's patch to allow "message" to
1261 be meaningful with "accept".
1263 SC/01 Eximstats V1.43
1264 Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
1266 SC/02 Eximstats V1.44
1267 Use a glob alias rather than an array ref in the generated
1268 parser. This improves both readability and performance.
1270 SC/03 Eximstats V1.45 (Marco Gaiarin / Steve Campbell)
1271 Collect SpamAssassin and rejection statistics.
1272 Don't display local sender or destination tables unless
1273 there is data to show.
1274 Added average volumes into the top table text output.
1276 SC/04 Eximstats V1.46
1277 Collect data on the number of addresses (recipients)
1278 as well as the number of messages.
1280 SC/05 Eximstats V1.47
1281 Added 'Message too big' to the list of mail rejection
1282 reasons (thanks to Marco Gaiarin).
1284 SC/06 Eximstats V1.48
1285 Mainlog lines which have GMT offsets and are too short to
1286 have a flag are now skipped.
1288 SC/07 Eximstats V1.49 (Alain Williams)
1289 Added the -emptyok flag.
1291 SC/08 Eximstats V1.50
1292 Fixes for obtaining the IP address from reject messages.
1294 JJ/03 exipick.20061117.2, made header handling as similar to exim as possible
1295 (added [br]h_ prefixes, implemented RFC2047 decoding. Fixed
1296 whitesspace changes from 4.64-PH/27
1298 JJ/04 exipick.20061117.2, fixed format and added $message_headers_raw to
1301 JJ/05 exipick.20061117.2, bug fixes (error out sooner when invalid criteria
1302 are found, allow negative numbers in numeric criteria)
1304 JJ/06 exipick.20061117.2, added new $message_body_missing variable
1306 JJ/07 exipick.20061117.2, added $received_ip_address and $received_port
1307 to match changes made in 4.64-PH/43
1309 PH/46 Applied Jori Hamalainen's patch to add features to exiqsumm.
1311 PH/47 Put in an explicit test for a DNS lookup of an address record where the
1312 "domain" is actually an IP address, and force a failure. This locks out
1313 those revolvers/nameservers that support "A-for-A" lookups, in
1314 contravention of the specifications.
1316 PH/48 When a host name was looked up from an IP address, and the subsequent
1317 forward lookup of the name timed out, the host name was left in
1318 $sender_host_name, contrary to the specification.
1320 PH/49 Although default lookup types such as lsearch* or cdb*@ have always been
1321 restricted to single-key lookups, Exim was not diagnosing an error if
1322 * or *@ was used with a query-style lookup.
1324 PH/50 Increased the value of DH_BITS in tls-gnu.c from 768 to 1024.
1326 MH/01 local_scan ABI version incremented to 1.1. It should have been updated
1327 long ago, but noone interested enough thought of it. Let's just say that
1328 the "1.1" means that there are some new functions that weren't there at
1329 some point in the past.
1331 PH/51 Error processing for expansion failure of helo_data from an smtp
1332 transport during callout processing was broken.
1334 PH/52 Applied John Jetmore's patch to allow tls-on-connect and STARTTLS to be
1335 tested/used via the -bh/-bhc/-bs options.
1337 PH/53 Added missing "#include <time.h>" to pcre/pcretest.c (this was a PCRE
1338 bug, fixed in subsequent PCRE releases).
1340 PH/54 Applied Robert Bannocks' patch to avoid a problem with references that
1341 arises when using the Solaris LDAP libraries (but not with OpenLDAP).
1343 PH/55 Check for a ridiculously long file name in exim_dbmbuild.
1349 SC/01 Use a glob alias rather than an array ref in eximstats generated
1350 parser. This improves both readability and performance.
1352 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
1353 Don't display local sender or destination tables in eximstats unless
1354 there is data to show.
1355 Added average volumes into the eximstats top table text output.
1357 SC/03 Collect data on the number of addresses (recipients) as well
1358 as the number of messages in eximstats.
1360 TF/01 Correct an error in the documentation for the redirect router. Exim
1361 does (usually) call initgroups() when daemonizing.
1363 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
1364 with consistent privilege compared to when running as a daemon.
1366 TF/03 Note in the spec that $authenticated_id is not set for local
1367 submissions from trusted users.
1369 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
1370 Thanks to Dean Brooks <dean@iglou.com> for the patch.
1372 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
1373 by adding some example configuration directives to the default
1374 configuration file. A little bit of work is required to uncomment the
1375 directives and define how usernames and passwords are checked, but
1376 there is now a framework to start from.
1378 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
1379 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
1380 without this. I don't know how relevant this is to other LDAP libraries.
1382 PH/02 Add the verb name to the "unknown ACL verb" error.
1384 PH/03 Magnus Holmgren's patch for filter_prepend_home.
1386 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
1388 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
1389 directory not expanded when it should be if an expanded home directory
1390 was set for the address (which is overridden by the transport).
1392 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
1395 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
1396 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
1397 because it is too late at that time, and has no effect.
1399 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
1400 security issue with \' (bugzilla #107). I could not use the
1401 PQescapeStringConn() function, because it needs a PGconn value as one of
1404 PH/08 When testing addresses using -bt, indicate those final addresses that
1405 are duplicates that would not cause an additional delivery. At least one
1406 person was confused, thinking that -bt output corresponded to deliveries.
1407 (Suppressing duplicates isn't a good idea as you lose the information
1408 about possibly different redirections that led to the duplicates.)
1410 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
1411 systems where poll() doesn't work, in particular OS X.
1413 PH/10 Added more information to debugging output for retry time not reached.
1415 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
1416 operations in malware.c.
1418 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
1421 PH/13 If write_rejectlog was set false when logging was sent to syslog with
1422 syslog_duplication set false, log lines that would normally be written
1423 both the the main log and to the reject log were not written to syslog at
1426 PH/14 In the default configuration, change the use of "message" in ACL warn
1427 statements to "add_header".
1429 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
1430 not followed by a command (e.g. "seen endif").
1432 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
1433 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
1436 PH/17 Added extra conditions to the default value of delay_warning_condition
1440 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
1441 { match{$h_precedence:}{(?i)bulk|list|junk} } \
1442 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
1445 The Auto-Submitted: and various List- headers are standardised, whereas I
1446 don't think Precedence: ever was.
1448 PH/18 Refactored debugging code in route_finduser() to show more information,
1449 in particular, the error code if getpwnam() issues one.
1451 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
1452 This is apparently needed in addition to the PH/07 change above to avoid
1453 any possible encoding problems.
1455 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
1456 but not after initializing Perl.
1458 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
1459 output them only if debugging. By default they are written stderr,
1460 apparently, which is not desirable.
1462 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
1465 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
1468 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
1470 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
1471 authenticated or an ident call has been made. Suppress the default
1472 values for $authenticated_id and $authenticated_sender (but permit -oMai
1473 and -oMas) when testing with -bh.
1475 PH/24 Re-jigged the order of the tests in the default configuration so that the
1476 tests for valid domains and recipients precede the DNS black list and CSA
1477 tests, on the grounds that those ones are more expensive.
1479 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
1480 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
1481 This bug exists in every version of Exim that I still have, right back to
1484 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
1485 However, an attempt to turn on case-sensitivity in a regex key by
1486 including (?-i) didn't work because the subject string was already
1487 lowercased, and the effects were non-intuitive. It turns out that a
1488 one-line patch can be used to allow (?-i) to work as expected.
1494 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
1495 other effects) broke the use of negated acl sub-conditions.
1497 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
1500 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
1501 "Deny" causes Exim to reject the incoming connection with a 554 error.
1502 Unfortunately, if there is a major crisis, such as a disk failure,
1503 tcp-wrappers gives "deny", whereas what one would like would be some
1504 kind of temporary error. A kludge has been added to help with this.
1505 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
1506 554 error is used if errno is still zero or contains ENOENT (which occurs
1507 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
1510 PH/03 Add -lutil to the default FreeBSD LIBS setting.
1512 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
1513 errors. Otherwise a message that provokes a temporary error (when other
1514 messages do not) can cause a whole host to time out.
1516 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
1517 addresses were routed directly to files or pipes from a redirect router.
1518 File deliveries just didn't batch; pipe deliveries might have suffered
1521 PH/06 A failure to get a lock for a hints database would erroneously always say
1522 "Failed to get write lock", even when it was really a read lock.
1524 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
1525 of 0600. This has been changed to use the value of the lockfile_mode
1526 option (which defaults to 0600).
1528 PH/08 Applied small patch from the Sieve maintainer.
1530 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
1531 folder from quota calculations, a direct delivery into this folder messed
1532 up the contents of the maildirsize file. This was because the regex was
1533 used only to exclude .Trash (or whatever) when the size of the mailbox
1534 was calculated. There was no check that a delivery was happening into an
1535 excluded directory. This bug has been fixed by ignoring all quota
1536 processing for deliveries into excluded directories.
1538 PH/10 Added the maildirfolder_create_regex option to appendfile.
1544 PH/01 The code for finding all the local interface addresses on a FreeBSD
1545 system running IPv6 was broken. This may well have applied to all BSD
1546 systems, as well as to others that have similar system calls. The broken
1547 code found IPv4 interfaces correctly, but gave incorrect values for the
1548 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
1549 that it would not match correctly against @[] and not recognize the IPv6
1552 PH/02 The ipliteral router was not recognizing addresses of the form user@
1553 [ipv6:....] because it didn't know about the "ipv6:" prefix.
1555 PH/03 Added disable_ipv6.
1557 PH/04 Changed $reply_address to use the raw form of the headers instead of the
1558 decoded form, because it is most often used to construct To: headers
1559 lines in autoreplies, and the decoded form may well be syntactically
1560 invalid. However, $reply_address has leading white space removed, and all
1561 newlines turned into spaces so that the autoreply transport does not
1564 PH/05 If group was specified without a user on a router, and no group or user
1565 was specified on a transport, the group from the router was ignored.
1567 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
1568 for visible compile-time settings that can be used to change these
1569 numbers, for those that want even more. Backwards compatibility with old
1570 spool files has been maintained. However, going back to a previous Exim
1571 release will lost any variables that are in spool files.
1573 PH/07 Two small changes when running in the test harness: increase delay when
1574 passing a TCP/IP connection to a new process, in case the original
1575 process has to generate a bounce, and remove special handling of
1576 127.0.0.2 (sic), which is no longer necessary.
1578 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
1579 be the same on different OS.
1581 PH/09 Moved a debug statement in filter processing to avoid a race problem when
1584 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
1585 whether --show-vars was specified or not
1587 JJ/02 exipick: Added support for new ACL variable spool format introduced
1590 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
1591 syntactically invalid From: or Reply-to: line, and a filter used this to
1592 generate an autoreply, and therefore failed to obtain an address for the
1593 autoreply, Exim could try to deliver to a non-existent relative file
1594 name, causing unrelated and misleading errors. What now happens is that
1595 it logs this as a hard delivery error, but does not attempt to create a
1598 PH/11 The exinext utility has a -C option for testing purposes, but although
1599 the given file was scanned by exinext itself; it wasn't being passed on
1600 when Exim was called.
1602 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
1603 an end-of-file indication when reading a command response.
1605 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
1606 compiled. In many other places in Exim, IPv6 addresses are always
1607 recognized, so I have changed this. It also means that IPv4 domain
1608 literals of the form [IPV4:n.n.n.n] are now always recognized.
1610 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
1611 used if the router is not running as root, for example, when verifying at
1612 ACL time, or when using -bh. The debugging output from this situation was
1613 non-existent - all you got was a failure to exec. I have made two
1616 (a) Failures to set uid/gid, the current directory, or a process leader
1617 in a subprocess such as that created by queryprogram now generate
1618 suitable debugging ouput when -d is set.
1620 (b) The queryprogram router detects when it is not running as root,
1621 outputs suitable debugging information if -d is set, and then runs
1622 the subprocess without attempting to change uid/gid.
1624 PH/15 Minor change to Makefile for building test_host (undocumented testing
1627 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
1628 additional section of a DNS packet that returns MX or SRV records.
1629 Instead, it always explicitly searches for A/AAAA records. This avoids
1630 major problems that occur when a DNS server includes only records of one
1631 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
1632 fixed another bug: if SRV records were looked up and the corresponding
1633 address records were *not* found in the additional section, the port
1634 values from the SRV records were lost.
1636 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
1637 using the correct key (the original address) when searching the retry
1638 rules in order to find which one to use for generating the retry hint.
1640 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
1641 adding the default one. Similarly, if it contains a Reply-To: header, the
1642 errors_reply_to option, if set, is not used.
1644 PH/19 When calculating a retry time, Exim used to measure the "time since
1645 failure" by looking at the "first failed" field in the retry record. Now
1646 it does not use this if it is later than than the arrival time of the
1647 message. Instead it uses the arrival time. This makes for better
1648 behaviour in cases where some deliveries succeed, thus re-setting the
1649 "first failed" field. An example is a quota failure for a huge message
1650 when small messages continue to be delivered. Without this change, the
1651 "time since failure" will always be short, possible causing more frequent
1652 delivery attempts for the huge message than are intended.
1653 [Note: This change was subsequently modified - see PH/04 for 4.62.]
1655 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
1656 $1, $2, $3) because the numerical variables can be reset during some
1657 expansion items (e.g. "match"), thereby losing the authentication data.
1659 PH/21 Make -bV show the size of off_t variables so that the test suite can
1660 decide whether to run tests for quotas > 2G.
1662 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
1663 mailbox_size, and mailbox_filecount in the appendfile transport. If a
1664 filecount value is greater than 2G or if a quota value is greater than 2G
1665 on a system where the size of off_t is not greater than 4, a panic error
1668 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
1669 never match. The debug and -bh output now contains an explicit error
1670 message indicating a malformed IPv4 address or mask.
1672 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
1673 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
1674 PH/23 above applies.
1676 PH/25 Do not write to syslog when running in the test harness. The only
1677 occasion when this arises is a failure to open the main or panic logs
1678 (for which there is an explicit test).
1680 PH/26 Added the /no_tell option to "control=freeze".
1682 PH/27 If a host name lookup failed very early in a connection, for example, if
1683 the IP address matched host_lookup and the reverse lookup yielded a name
1684 that did not have a forward lookup, an error message of the form "no IP
1685 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
1686 could be logged. Now it outputs the IP address instead of "NULL".
1688 PH/28 An enabling patch from MH: add new function child_open_exim2() which
1689 allows the sender and the authenticated sender to be set when
1690 submitting a message from within Exim. Since child_open_exim() is
1691 documented for local_scan(), the new function should be too.
1693 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
1694 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
1695 results in an empty string is now treated as unset.
1697 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
1699 PH/31 Added sender_verify_fail logging option.
1701 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
1702 needed by RFC 822 but not by RFC 2822 was commented out. I have now
1703 tidied the source and removed it altogether.
1705 PH/33 When a queue run was abandoned because the load average was too high, a
1706 log line was always written; now it is written only if the queue_run log
1707 selector is set. In addition, the log line for abandonment now contains
1708 information about the queue run such as the pid. This is always present
1709 in "start" and "stop" lines but was omitted from the "abandon" line.
1711 PH/34 Omit spaces between a header name and the colon in the error message that
1712 is given when verify = headers_syntax fails (if there are lots of them,
1713 the message gets confusing).
1715 PH/35 Change the default for dns_check_names_pattern to allow slashes within
1716 names, as there are now some PTR records that contain slashes. This check
1717 is only to protect against broken name servers that fall over on strange
1718 characters, so the fact that it applies to all lookups doesn't matter.
1720 PH/36 Now that the new test suite is complete, we can remove some of the
1721 special code in Exim that was needed for the old test suite. For example,
1722 sorting DNS records because real resolvers return them in an arbitrary
1723 order. The new test suite's fake resolver always returns records in the
1726 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
1727 bounces) except when queue_only is set, to avoid logging races between
1728 the different processes.
1730 PH/38 Panic-die if .include specifies a non-absolute path.
1732 PH/39 A tweak to the "H" retry rule from its user.
1734 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
1735 a label. They prevented compilation on older perls.
1737 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
1738 a warning to be raised on newish perls.
1740 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
1741 on queue. Changes to match documented behaviour of showing count of
1742 messages matching specified criteria.
1744 PH/40 Changed the default ident timeout from 30s to 5s.
1746 PH/41 Added support for the use of login_cap features, on those BSD systems
1747 that have them, for controlling the resources used by pipe deliveries.
1749 PH/42 The content-scanning code uses fopen() to create files in which to put
1750 message data. Previously it was not paying any attention to the mode of
1751 the files. Exim runs with umask(0) because the rest of the code creates
1752 files with open(), and sets the required mode explicitly. Thus, these
1753 files were ending up world-writeable. This was not a big issue, because,
1754 being within the spool directory, they were not world-accessible. I have
1755 created a function called modefopen, which takes an additional mode
1756 argument. It sets umask(777), creates the file, chmods it to the required
1757 mode, then resets the umask. All the relevant calls to fopen() in the
1758 content scanning code have been changed to use this function.
1760 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
1761 to 24 hours. This avoids potential overflow problems when processing G
1762 and H retry rules. I suspect nobody ever tinkers with this value.
1764 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
1766 PH/45 When the plaintext authenticator is running as a client, the server's
1767 challenges are checked to ensure they are valid base64 strings. By
1768 default, the authentication attempt is cancelled if an invalid string is
1769 received. Setting client_ignore_invalid_base64 true ignores these errors.
1770 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
1771 they are received. Thus, the responses can be made to depend on the
1772 challenges. If an invalid string is ignored, an empty string is placed in
1775 PH/46 Messages that are created by the autoreply transport now contains a
1776 References: header, in accordance with RFCs 2822 and 3834.
1778 PH/47 Added authenticated_sender_force to the smtp transport.
1780 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
1782 PH/49 Installed latest patch from the Sieve maintainer.
1784 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
1785 was also set, the appendfile transport was unnecessarily scanning a
1786 directory of message files (e.g. for maildir delivery) to find the count
1787 of files (along with the size), even though it did not need this
1788 information. It now does the scan only if it needs to find either the
1789 size of the count of files.
1791 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
1793 PH/52 Two bugs concerned with error handling when the smtp transport is
1796 (i) Exim was not creating retry information for temporary errors given
1797 for individual recipients after the DATA command when the smtp transport
1798 was used in LMTP mode. This meant that they could be retried too
1799 frequently, and not timed out correctly.
1801 (ii) Exim was setting the flag that allows error details to be returned
1802 for LMTP errors on RCPT commands, but not for LMTP errors for individual
1803 recipients that were returned after the DATA command.
1805 PH/53 This is related to PH/52, but is more general: for any failing address,
1806 when detailed error information was permitted to be returned to the
1807 sender, but the error was temporary, then after the final timeout, only
1808 "retry timeout exceeded" was returned. Now it returns the full error as
1809 well as "retry timeout exceeded".
1811 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
1812 do this, and (what is worse) MTAs that accept it.
1814 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
1815 will now be deprecated.
1817 PH/56 New os.c-cygwin from the Cygwin maintainer.
1819 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
1820 formats (previously only available in exim formats via -bpr, -bpru,
1821 and -bpra. Now also available in native and exiqgrep formats)
1823 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
1824 with very large, slow to parse queues
1826 JJ/08 exipick: added ! as generic prefix to negate any criteria format
1828 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
1830 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
1831 responses to authentication challenges, though it was showing the
1832 challenges; (ii) I've removed the CR characters from the debug output for
1835 PH/58 Allow for the insertion of a newline as well as a space when a string
1836 is turned into more than one encoded-word during RFC 2047 encoding. The
1837 Sieve code now uses this.
1839 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
1840 data_4xx, lost_connection, tls_required.
1842 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
1843 message was being sent as an SMTP response.
1845 PH/61 Add -l and -k options to exicyclog.
1847 PH/62 When verifying, if an address was redirected to one new address, so that
1848 verification continued, and the new address failed or deferred after
1849 having set something in $address_data, the value of $address_data was not
1850 passed back to the ACL. This was different to the case when no
1851 redirection occurred. The value is now passed back in both cases.
1853 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
1854 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
1855 use login_cap.h, so on its own it isn't the distinguishing feature. The
1856 new name refers directly to the setclassresources() function.
1858 PH/65 Added configuration files for NetBSD3.
1860 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
1862 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
1863 is preferred over IPv4.
1865 PH/68 The bounce_return_message and bounce_return_body options were not being
1866 honoured for bounces generated during the reception of non-SMTP messages.
1867 In particular, this applied to messages rejected by the ACL. This bug has
1868 been fixed. However, if bounce_return_message is true and bounce_return_
1869 body is false, the headers that are returned for a non-SMTP message
1870 include only those that have been read before the error was detected.
1871 (In the case of an ACL rejection, they have all been read.)
1873 PH/69 The HTML version of the specification is now built in a directory called
1874 spec_html instead of spec.html, because the latter looks like a path with
1875 a MIME-type, and this confuses some software.
1877 PH/70 Catch two compiler warnings in sieve.c.
1879 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
1880 function verify_get_ident() calls ip_connect() to connect a socket, but
1881 if the "connect()" function timed out, ip_connect() used to close the
1882 socket. However, verify_get_ident() also closes the socket later, and in
1883 between Exim writes to the log, which may get opened at this point. When
1884 the socket was closed in ip_connect(), the log could get the same file
1885 descriptor number as the socket. This naturally causes chaos. The fix is
1886 not to close the socket in ip_connect(); the socket should be closed by
1887 the function that creates it. There was only one place in the code where
1888 this was missing, in the iplookup router, which I don't think anybody now
1889 uses, but I've fixed it anyway.
1891 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
1892 well as to direct DNS lookups. Otherwise the handling of names in host
1893 lists is inconsistent and therefore confusing.
1899 PH/01 Two changes to the default runtime configuration:
1901 (1) Move the checks for relay_from_hosts and authenticated clients from
1902 after to before the (commented out) DNS black list checks.
1904 (2) Add control=submission to the relay_from_hosts and authenticated
1905 clients checks, on the grounds that messages accepted by these
1906 statements are most likely to be submissions.
1908 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
1910 (1) Generate an error if the third argument for the ${prvs expansion is
1913 (2) Treat a missing third argument of ${prvscheck as if it were an empty
1916 (3) Reset the variables that are obtained from the first argument of
1917 ${prvscheck and used in the second argument before leaving the code,
1918 because their memory is reclaimed, so using them afterwards may do
1921 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
1922 one (it's much easier than Tom thought :-).
1924 (5) Because of (4), we can now allow for the use of $prvscheck_result
1925 inside the third argument.
1927 PH/03 For some reason, the default setting of PATH when running a command from
1928 a pipe transport was just "/usr/bin". I have changed it to
1931 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
1932 anything to be listed in the output from -bV.
1934 PH/05 When a filter generated an autoreply, the entire To: header line was
1935 quoted in the delivery log line, like this:
1937 => >A.N.Other <ano@some.domain> <original@ddress> ...
1939 This has been changed so that it extracts the operative address. There
1940 may be more than one such address. If so, they are comma-separated, like
1943 => >ano@some.domain,ona@other.domain <original@ddress> ...
1945 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
1946 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
1947 not being looked up in the rDNS to get a host name, Exim was showing the
1948 IP address twice in Received: lines, even though the IP addresses were
1949 identical. For example:
1951 Received: from [1.2.3.4] (helo=[1.2.3.4])
1953 However, if the real host name was known, it was omitting the HELO data
1954 if it matched the actual IP address. This has been tidied up so that it
1955 doesn't show the same IP address twice.
1957 PH/07 When both +timestamp and +memory debugging was on, the value given by
1958 $tod_xxx expansions could be wrong, because the tod_stamp() function was
1959 called by the debug printing, thereby overwriting the timestamp buffer.
1960 Debugging no longer uses the tod_stamp() function when +timestamp is set.
1962 PH/08 When the original message was included in an autoreply transport, it
1963 always said "this is a copy of the message, including all the headers",
1964 even if body_only or headers_only was set. It now gives an appropriate
1967 PH/09 Applied a patch from the Sieve maintainer which:
1969 o fixes some comments
1970 o adds the (disabled) notify extension core
1971 o adds some debug output for the result of if/elsif tests
1972 o points to the current vacation draft in the documentation
1973 and documents the missing references header update
1977 o fixes a bug in processing the envelope test (when testing
1978 multiple envelope elements, the last element determinted the
1981 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
1982 Electronic Mail") by including:
1984 Auto-submitted: auto-generated
1986 in the messages that it generates (bounce messages and others, such as
1987 warnings). In the case of bounce messages for non-SMTP mesages, there was
1988 also a typo: it was using "Auto_submitted" (underscore instead of
1989 hyphen). Since every message generated by Exim is necessarily in response
1990 to another message, thes have all been changed to:
1992 Auto-Submitted: auto-replied
1994 in accordance with these statements in the RFC:
1996 The auto-replied keyword:
1998 - SHOULD be used on messages sent in direct response to another
1999 message by an automatic process,
2001 - MUST NOT be used on manually-generated messages,
2003 - MAY be used on Delivery Status Notifications (DSNs) and Message
2004 Disposition Notifications (MDNs),
2006 - MUST NOT be used on messages generated by automatic or periodic
2007 processes, except for messages which are automatic responses to
2010 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
2011 to the default Received: header definition.
2013 PH/12 Added log selector acl_warn_skipped (default on).
2015 PH/13 After a successful wildlsearch lookup, discard the values of numeric
2016 variables because (a) they are in the wrong storage pool and (b) even if
2017 they were copied, it wouldn't work properly because of the caching.
2019 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
2020 checking when decoding. Apparently there are clients that generate
2021 overlong encoded strings. Why am I not surprised?
2023 PH/15 If the first argument of "${if match_address" was not empty, but did not
2024 contain an "@" character, Exim crashed. Now it writes a panic log message
2025 and treats the condition as false.
2027 PH/16 In autoreply, treat an empty string for "once" the same as unset.
2029 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
2030 extension "envelope-auth". The code is finished and in agreement with
2031 other implementations, but there is no documentation so far and in fact,
2032 nobody wrote the draft yet. This extension is currently #undef'ed, thus
2033 not changing the active code.
2035 Print executed "if" and "elsif" statements when debugging is used. This
2036 helps a great deal to understand what a filter does.
2038 Document more things not specified clearly in RFC3028. I had all this
2039 sorted out, when out of a sudden new issues came to my mind. Oops."
2041 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
2044 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
2045 canonical form (as documented). However, after a host name lookup from
2046 the IP address, check_host() was doing a simple string comparison with
2047 addresses acquired from the DNS when checking that the found name did
2048 have the original IP as one of its addresses. Since any found IPv6
2049 addresses are likely to be in abbreviated form, the comparison could
2050 fail. Luckily, there already exists a function for doing the comparison
2051 by converting both addresses to binary, so now that is used instead of
2052 the text comparison.
2054 PH/20 There was another similar case to PH/19, when a complete host name was
2055 given in a host list; looking up its IP address could give an abbreviated
2056 form, whereas the current host's name might or might not be abbreviated.
2057 The same fix has been applied.
2063 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
2064 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
2067 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
2068 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
2070 PH/03 Typo: missing ".o" in src/pcre/Makefile.
2072 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
2073 header line, restrict the check to what is listed in RFCs 2369 and 2929.
2074 Also, for "Auto-Submitted", treat anything other than "no" as
2075 non-personal, in accordance with RFC 3834. (Previously it treated
2076 anything starting "auto-" as non-personal.)
2078 TF/01 The control=submission/name=... option had a problem with syntax
2079 errors if the name included a slash character. The /name= option
2080 now slurps the rest of the string, so it can include any characters
2081 but it must come last in the list of options (after /sender_retain
2084 PH/05 Some modifications to the interface to the fake nameserver for the new
2092 TK/01 Added the "success_on_redirect" address verification option. See
2093 NewStuff for rationale and an example.
2095 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
2097 PH/02 Patch to exigrep to allow it to work on syslog lines.
2099 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
2100 fread() to skip over the body file's header line, because in Cygwin the
2101 header line is locked and is inaccessible.
2103 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
2104 co-exist for some time) to make it clear that it is the Exim ID that is
2105 referenced, not the Message-ID: header line.
2107 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
2108 string_format() function, because snprintf() does not exist on all
2111 PH/06 The use of forbid_filter_existstest now also locks out the use of the
2112 ${stat: expansion item.
2114 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
2115 protocol synchronization error", to keep the pedants happy.
2117 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
2118 well as for IRIX systems, when gcc is being used. See the host.c source
2121 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
2123 PH/10 Named domain lists were not working if used in a queue_smtp_domains
2126 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
2127 transport and to the smtp transport in LMTP mode.
2129 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
2131 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
2132 run a filter in a subprocess. This could lead to confusion in subsequent
2133 lookups in the parent process. There should also be a search_tidyup() at
2134 the end of the subprocess.
2136 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
2137 only if the host matched helo_try_verify_hosts, which caused the
2138 verification to occur when the EHLO/HELO command was issued. The ACL just
2139 tested the remembered result. Now, if a previous verification attempt has
2140 not happened, "verify = helo" does it there and then.
2142 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
2144 TK/03 Fix log output including CR from clamd.
2146 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
2147 exist provoked a memory error which could cause a segfault.
2149 PH/15 Installed PCRE 6.2
2151 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
2153 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
2154 of the problem. Specifically, suggested +O2 rather than +O1 for the
2157 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
2159 PH/20 If a delivery was routed to a non-standard port by means of an SRV
2160 record, the port was not correctly logged when the outgoing_port log
2161 selector was set (it logged the transort's default port).
2163 PH/21 Added support for host-specific ports to manualroute, queryprogram,
2164 fallback_hosts, and "hosts" in the smtp transport.
2166 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
2167 host errors such as "Connection refused".
2169 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
2170 authentication with radiusclient 0.4.9:
2172 - Error returned from rc_read_config was caught wrongly
2173 - Username/password not passed on to radius server due to wrong length.
2175 The presumption is that some radiusclient API changes for 4.51/PH/17
2176 were not taken care of correctly. The code is still untested by me (my
2177 Linux distribution still has 0.3.2 of radiusclient), but it was
2178 contributed by a Radius user.
2180 PH/24 When doing a callout, the value of $domain wasn't set correctly when
2181 expanding the "port" option of the smtp transport.
2183 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
2184 while reading a MIME header. Thanks to Tom Hughes for a patch.
2186 PH/24 Include config.h inside local_scan.h so that configuration settings are
2189 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
2190 that in an ACL for RCPT (for example), you can examine exactly what was
2193 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
2194 commands, but it was not correctly comparing the address with the actual
2195 client host address. Thus, it would show the EHLO address in Received:
2196 header lines when this was not necessary.
2198 PH/27 Added the % operator to ${eval:}.
2200 PH/28 Exim tries to create and chdir to its spool directory when it starts;
2201 it should be ignoring failures (because with -C, for example, it has lost
2202 privilege). It wasn't ignoring creation failures other than "already
2205 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
2208 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
2209 because an input line was too long, either on its own, or by virtue of
2210 too many continuations, the temporary file was not being removed, and the
2211 return code was incorrect.
2213 PH/31 Missing "BOOL" in function definition in filtertest.c.
2215 PH/32 Applied Sieve patches from the maintainer.
2217 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
2219 PH/33 Added "verify = not_blind".
2221 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
2222 Local/Makefile (with some defaults set). These are used in built scripts
2223 such as exicyclog, but they have never been used in the exim_install
2224 script (though there are many overriding facilities there). I have
2225 arranged that the exim_install script now takes note of these two
2228 PH/35 Installed configuration files for Dragonfly.
2230 PH/36 When a locally submitted message by a trusted user did not contain a
2231 From: header, and the sender address was obtained from -f or from an SMTP
2232 MAIL command, and the trusted user did not use -F to supply a sender
2233 name, $originator_name was incorrectly used when constructing a From:
2234 header. Furthermore, $originator_name was used for submission mode
2235 messages from external hosts without From: headers in a similar way,
2236 which is clearly wrong.
2238 PH/37 Added control=suppress_local_fixups.
2240 PH/38 When log_selector = +received_sender was set, and the addition of the
2241 sender made the log line's construction buffer exactly full, or one byte
2242 less than full, an overflow happened when the terminating "\n" was
2245 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
2246 when the result of a list match is failure because a DNS lookup failed.
2248 PH/40 RM_COMMAND is now used in the building process.
2250 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
2251 the "build-* directories that it finds.
2253 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
2254 address in a domain literal was a prefix of an interface address.
2256 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
2257 when verifying a sender address, unless rewrite_headers is false.
2259 PH/44 Wrote a long comment about why errors_to addresses are verified as
2260 recipients, not senders.
2262 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
2263 the ratelimit ACL was added.
2265 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
2267 PH/46 Added extra information about PostgreSQL errors to the error string.
2269 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
2270 avoiding the need to install special zones in a real server. This is
2271 backwards compatible; if it can't find the fake resolver, it drops back.
2272 Thus, both old and new test suites can be run.
2274 TF/02 Added util/ratelimit.pl
2276 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
2277 clock is set back in time.
2279 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
2280 Candler <B.Candler@pobox.com>.
2282 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
2283 OK for addresses that are the result of redirections.
2285 PH/48 A number of further additions for the benefit of the new test suite,
2286 including a fake gethostbyname() that interfaces to the fake DNS resolver
2289 TF/06 The fix for widen_domains has also been applied to qualify_single and
2290 search_parents which are the other dnslookup options that can cause
2293 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
2296 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
2298 TF/07 Exim produced the error message "an SRV record indicated no SMTP
2299 service" if it encountered an MX record with an empty target hostname.
2300 The message is now "an MX or SRV record indicated no SMTP service".
2302 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
2303 if the DNS of the sending site is misconfigured. This is quite a
2304 common situation. This change restores the behaviour of treating a
2305 helo verification defer as a failure.
2307 PH/51 If self=fail was set on a router, the bounce message did not include the
2308 actual error message.
2314 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
2316 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
2317 command itself ended in error, the underlying message about the transport
2318 filter timeout was being overwritten with the pipe command error. Now the
2319 underlying error message should be appended to the second error message.
2321 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
2323 PH/02 Reduce the amount of output that "make" produces by default. Full output
2324 can still be requested.
2326 PH/03 The warning log line about a condition test deferring for a "warn" verb
2327 was being output only once per connection, rather than after each
2328 occurrence (because it was using the same function as for successful
2329 "warn" verbs). This seems wrong, so I have changed it.
2331 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
2332 it should not have, which might have caused a crash in the right
2333 circumstances, but probably never did.
2335 PH/04 Installed a modified version of Tony Finch's patch to make submission
2336 mode fix the return path as well as the Sender: header line, and to
2337 add a /name= option so that you can make the user's friendly name appear
2340 TF/03 Added the control = fakedefer ACL modifier.
2342 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
2343 Mark Lowes for thorough testing.
2345 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
2347 TK/03 Merged latest SRS patch from Miles Wilton.
2349 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
2350 with the definition in sysexits.h (which is #included earlier).
2351 Fortunately, Exim does not actually use EX_OK. The code used to try to
2352 preserve the sysexits.h value, by assumimg that macro definitions were
2353 scanned for macro replacements. I have been disabused of this notion,
2354 so now the code just undefines EX_OK before #including unistd.h.
2356 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
2357 in the smtp transport. When a block could not be written in a single
2358 write() function, the timeout was being re-applied to each part-write.
2359 This seems wrong - if the receiver was accepting one byte at a time it
2360 would take for ever. The timeout is now adjusted when this happens. It
2361 doesn't have to be particularly precise.
2363 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
2364 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
2366 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
2367 without a domain if the check to <postmaster@domain> fails.
2369 SC/01 Eximstats: added -xls and the ability to specify output files
2370 (patch written by Frank Heydlauf).
2372 SC/02 Eximstats: use FileHandles for outputing results.
2374 SC/03 Eximstats: allow any combination of xls, txt, and html output.
2376 SC/04 Eximstats: fixed display of large numbers with -nvr option
2378 SC/05 Eximstats: fixed merging of reports with empty tables.
2380 SC/06 Eximstats: added the -include_original_destination flag
2382 SC/07 Eximstats: removed tabs and trailing whitespace.
2384 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
2386 TK/06 MBOX spool code: Add real "From " MBOX separator line
2387 so the .eml file is really in mbox format (even though
2388 most programs do not really care). Patch from Alex Miller.
2390 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
2391 The latter is generated from $received_to and is only set if the
2392 message has one envelope recipient. SA can use these headers,
2393 obviously out-of-the-box. Patch from Alex Miller.
2395 PH/08 The ${def test on a variable was returning false if the variable's
2396 value was "0", contrary to what the specification has always said!
2397 The result should be true unless the variable is empty.
2399 PH/09 The syntax error of a character other than { following "${if
2400 def:variable_name" (after optional whitespace) was not being diagnosed.
2401 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
2402 accidental colon was present, for example, could give incorrect results.
2404 PH/10 Tidied the code in a number of places where the st_size field of a stat()
2405 result is used (not including appendfile, where other changes are about
2408 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
2409 This involved changing a lot of size variables from int to off_t. It
2410 should work with maildirs and everything.
2412 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
2413 spamd dying while we are connected to it.
2415 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
2418 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
2419 function for BDB 4.3.
2421 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
2423 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
2424 the sources for PCRE have been re-arranged and more files are now
2427 PH/15 The code I had for printing potentially long long variables in PH/11
2428 above was not the best (it lost precision). The length of off_t variables
2429 is now inspected at build time, and an appropriate printing format (%ld
2430 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
2431 to be "long long int" or "long int". This is needed for the internal
2432 formatting function string_vformat().
2434 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
2435 the configuration file to be ":syslog", then the script "guesses" where
2436 the logs files are, rather than using the compiled in default. In our
2437 case the guess is not the same as the compiled default, so the script
2438 suddenly stopped working when I started to use syslog. The patch checks
2439 to see if log_file_path is "". If so, it attempts to read it from exim
2440 with no configuration file to get the compiled in version, before it
2441 falls back to the previous guessing code."
2443 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
2444 implementing BATV in an Exim configuration. See NewStuff for the gory
2447 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
2448 Makefile that are specific to HP-UX.
2450 PH/18 If the "use_postmaster" option was set for a recipient callout together
2451 with the "random" option, the postmaster address was used as the MAIL
2452 FROM address for the random test, but not for the subsequent recipient
2453 test. It is now used for both.
2455 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
2456 patch removes a few documentation additions to RFC 3028, because the
2457 latest draft now contains them. It adds the new en;ascii-case comparator
2458 and a new error check for 8bit text in MIME parts. Comparator and
2459 require names are now matched exactly. I enabled the subaddress
2460 extension, but it is not well tested yet (read: it works for me)."
2462 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
2463 rework some of the code of TK/09 above to avoid the hardwired use of
2464 "%lld" and "long long". Replaced the call to snprintf() with a call to
2467 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
2468 records point to non-existent hosts", "retry timeout exceeded", and
2469 "retry time not reached for any host after a long failure period".
2471 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
2472 experimental DomainKeys support:
2474 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
2475 (2) On an error such as an illegally used "control", the wrong name for
2476 the control was given.
2478 These problems did NOT occur unless DomainKeys support was compiled.
2480 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
2482 PH/24 Added ${if match_ip condition.
2484 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
2485 databases so that it will be absolutely obvious if a crash occurs in the
2486 DB library. This is a regular occurrence (often caused by mis-matched
2489 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
2490 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
2491 picked up on a user's system that detects such things. There doesn't seem
2492 to be a gcc warning option for this - only an attribute that has to be
2493 put on the function's prototype. It seems that in Fedora Core 4 they have
2494 set this on a number of new functions. No doubt there will be more in due
2497 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
2498 specify a transport. However, if an address that was verified by such a
2499 router was the subject of a callout, Exim crashed because it tried to
2500 read the rcpt_include_affixes from the non-existent transport. Now it
2501 just assumes that the setting of that option is false. This bug was
2502 introduced by 4.51/PH/31.
2504 PH/28 Changed -d+all to exclude +memory, because that information is very
2505 rarely of interest, but it makes the output a lot bigger. People tend to
2506 do -d+all out of habit.
2508 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
2509 code in os-type was giving problems when libc.so lives in lib64, like on
2512 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
2513 aren't the modern standard, and it seems that some systems' include files
2514 don't always have them. Exim was already checking for some of the newer
2515 ones like T_AAAA, and defining it itself. I've added checks for all the
2516 record types that Exim uses.
2518 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
2519 not automatically generating a new one, as it is supposed to. This
2520 prevented TLS from working. If the file did exist, but contained invalid
2521 data, a new version was generated, as expected. It was only the case of a
2522 non-existent file that was broken.
2524 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
2525 with a change in libdomainkeys > 0.64.
2527 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
2528 from DNS. If the selector record carries the flag, it now has
2529 precedence over the domain-wide flag.
2531 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
2533 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
2534 the use of a port name that isn't defined in /etc/services) occurred, the
2535 message was deferred as in a normal delivery, and thus remained on the
2536 spool, instead of being failed because of the mua_wrapper setting. This
2537 is now fixed, and I tidied up some of the mua_wrapper messages at the
2540 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
2541 the messages in a hash of arrays rather than using individual hashes.
2542 This is a bit cleaner and results in dramatic memory savings, albeit
2543 at a slight CPU cost.
2545 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
2546 as requested by Marc Sherman.
2548 SC/10 Eximstats: added histograms for user specified patterns as requested
2551 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
2553 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
2554 fopen() in the content-scanning modules that did not already have it.
2560 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
2561 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
2563 TK/02 Fix ACL "control" statement not being available in MIME ACL.
2565 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
2567 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
2568 to test Sieve filters that use "vacation".
2570 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
2571 that changes the way the GnuTLS parameters are stored in the cache file.
2572 The new format can be generated externally. For backward compatibility,
2573 if the data in the cache doesn't make sense, Exim assumes it has read an
2574 old-format file, and it generates new data and writes a new file. This
2575 means that you can't go back to an older release without removing the
2578 PH/03 A redirect router that has both "unseen" and "one_time" set does not
2579 work if there are any delivery delays because "one_time" forces the
2580 parent to be marked "delivered", so its unseen clone is never tried
2581 again. For this reason, Exim now forbids the simultaneous setting of
2584 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
2585 redirected to themselves ("homonym" addresses). Read the long ChangeLog
2586 entry if you want to know the details. The fix, however, neglected to
2587 consider the case when local delivery batching is involved. The test for
2588 "previously delivered" was not happening when checking to see if an
2589 address could be batched with a previous (undelivered) one; under
2590 certain circumstances this could lead to multiple deliveries to the same
2593 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
2594 in its include files, and this causes problems building Exim.
2596 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
2597 header_syntax) but Exim was just ignoring anything given after a slash.
2598 In particular, this caused confusion with an attempt to use "verify =
2599 reverse_host_lookup/defer_ok". An error is now given when options are
2600 supplied for verify items that do not have them. (Maybe reverse_host_
2601 lookup should have a defer_ok option, but that's a different point.)
2603 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
2604 defined by RFC 821) to 2048, because there were problems with some AUTH
2605 commands, and RFC 1869 says the size should be increased for extended
2606 SMTP commands that take arguments.
2608 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
2611 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
2612 "unknown" error; now it says that the functionality isn't in the binary.
2614 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
2615 an address' error message when a string expansion fails (syntax or
2616 whatever). Otherwise the password may appear in the log. Following change
2617 PH/42 below, there is no longer a chance of it appearing in a bounce
2620 PH/11 Installed exipick version 20050225.0 from John Jetmore.
2622 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
2623 of its addresses was ever tried. (Bugzilla bug #2.)
2625 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
2626 the result incorrectly in the debug output. (It correctly added a newline
2627 to what was transported.)
2629 TF/01 Added $received_time.
2631 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
2632 commented out examples of how to interface to a virus scanner and to
2633 SpamAssassin. Also added commented examples of av_scanner and
2634 spamd_address settings.
2636 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
2637 and controls are allowed in which ACLs. There were a couple of minor
2638 errors. Some of the entries in the conditions table (which is a table of
2639 where they are NOT allowed) were getting very unwieldy; rewrote them as a
2640 negation of where the condition IS allowed.
2642 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
2644 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
2645 header file does not have a version number, so I've had to invent a new
2646 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
2647 API. The code is untested by me (my Linux distribution still has 0.3.2 of
2648 radiusclient), but it was contributed by a Radius user.
2650 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
2651 files or directories, for OpenSSL.
2653 PH/19 When an Exim process that is running as root has to create an Exim log
2654 file, it does so in a subprocess that runs as exim:exim so as to get the
2655 ownership right at creation (otherwise, other Exim processes might see
2656 the file with the wrong ownership). There was no test for failure of this
2657 fork() call, which would lead to the process getting stuck as it waited
2658 for a non-existent subprocess. Forks do occasionally fail when resources
2659 run out. I reviewed all the other calls to fork(); they all seem to check
2662 PH/20 When checking for unexpected SMTP input at connect time (before writing
2663 the banner), Exim was not dealing correctly with a non-positive return
2664 from the read() function. If the client had disconnected by this time,
2665 the result was a log entry for a synchronization error with an empty
2666 string after "input=" when read() returned zero. If read() returned -1
2667 (an event I could not check), uninitialized data bytes were printed.
2668 There were reports of junk text (parts of files, etc) appearing after
2671 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
2673 PH/22 Added support for macro redefinition, and (re)definition in between
2674 driver and ACL definitions.
2676 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
2677 forgetting to use the resulting value; it was using the unexpanded value.
2679 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
2680 hadn't been configured. The fix is from Juergen Kreileder, who
2681 understands it better than I do:
2683 "Here's what I see happening with three configured cyrus_sasl
2684 authenticators configured (plain, login, cram-md5):
2686 On startup auth_cyrus_sasl_init() gets called for each of these.
2687 This means three calls to sasl_listmech() without a specified mech_list.
2688 => SASL tests which mechs of all available mechs actually work
2689 => three warnings about OTP not working
2690 => the returned list contains: plain, login, cram-md5, digest-md5, ...
2692 With the patch, sasl_listmech() also gets called three times. But now
2693 SASL's mech_list option is set to the server_mech specified in the the
2694 authenticator. Or in other words, the answer from sasl_listmech()
2695 gets limited to just the mech you're testing for (which is different
2697 => the return list contains just 'plain' or 'login', 'cram-md5' or
2698 nothing depending on the value of ob->server_mech.
2700 I've just tested the patch: Authentication still works fine,
2701 unavailable mechs specified in the exim configuration are still
2702 caught, and the auth.log warnings about OTP are gone."
2704 PH/25 When debugging is enabled, the contents of the command line are added
2705 to the debugging output, even when log_selector=+arguments is not
2708 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
2709 answer is "GNU", and only if the return is "GNU/something" is the answer
2712 PH/27 $acl_verify_message is now set immediately after the failure of a
2713 verification in an ACL, and so is available in subsequent modifiers. In
2714 particular, the message can be preserved by coding like this:
2716 warn !verify = sender
2717 set acl_m0 = $acl_verify_message
2719 Previously, $acl_verify_message was set only while expanding "message"
2720 and "log_message" when a very denied access.
2722 PH/28 Modified OS/os.c-Linux with
2724 -#ifndef OS_LOAD_AVERAGE
2725 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
2727 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
2728 nomenclature these days.)
2730 PH/29 Installed patch from the Sieve maintainer that adds the options
2731 sieve_useraddress and sieve_subaddress to the redirect router.
2733 PH/30 In these circumstances:
2734 . Two addresses routed to the same list of hosts;
2735 . First host does not offer TLS;
2736 . First host accepts first address;
2737 . First host gives temporary error to second address;
2738 . Second host offers TLS and a TLS session is established;
2739 . Second host accepts second address.
2740 Exim incorrectly logged both deliveries with the TLS parameters (cipher
2741 and peerdn, if requested) that were in fact used only for the second
2744 PH/31 When doing a callout as part of verifying an address, Exim was not paying
2745 attention to any local part prefix or suffix that was matched by the
2746 router that accepted the address. It now behaves in the same way as it
2747 does for delivery: the affixes are removed from the local part unless
2748 rcpt_include_affixes is set on the transport.
2750 PH/32 Add the sender address, as F=<...>, to the log line when logging a
2751 timeout during the DATA phase of an incoming message.
2753 PH/33 Sieve envelope tests were broken for match types other than :is. I have
2754 applied a patch sanctioned by the Sieve maintainer.
2756 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
2757 the uid or gid is negative. A case of a negative gid caused this to be
2758 noticed. The fix allows for either to be negative.
2760 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
2761 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
2764 PH/36 The change PH/12 above was broken. Fixed it.
2766 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
2767 the grounds that routing the same address twice would always produce the
2768 same answer. This might have been true once, but it is certainly no
2769 longer true now. Routing a child address may depend on the previous
2770 routing that produced that child. Some complicated redirection strategies
2771 went wrong when messages had multiple recipients, and made Exim's
2772 behaviour dependent on the order in which the addresses were given.
2774 I have moved the duplicate checking until after the routing is complete.
2775 Exim scans the addresses that are assigned to local and remote
2776 transports, and removes any duplicates. This means that more work will be
2777 done, as duplicates will always all be routed, but duplicates are
2778 presumably rare, so I don't expect this is of any significance.
2780 For deliveries to pipes, files, and autoreplies, the duplicate checking
2781 still happens during the routing process, since they are not going to be
2784 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
2785 It corrects a timeout issue with spamd. This is Ian's comment: "The
2786 background is that sometimes spamd either never reads data from a
2787 connection it has accepted, or it never writes response data. The exiscan
2788 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
2789 blindly assumes that writes won't block so it may never time out."
2791 PH/39 Allow G after quota size as well as K and M.
2793 PH/40 The value set for $authenticated_id in an authenticator may not contain
2794 binary zeroes or newlines because the value is written to log lines and
2795 to spool files. There was no check on this. Now the value is run through
2796 the string_printing() function so that such characters are converted to
2797 printable escape sequences.
2799 PH/41 $message_linecount is a new variable that contains the total number of
2800 lines in the message. Compare $body_linecount, which is the count for the
2803 PH/42 Exim no longer gives details of delivery errors for specific addresses in
2804 bounce and delay warning messages, except in certain special cases, which
2807 (a) An SMTP error message from a remote host;
2808 (b) A message specified in a :fail: redirection;
2809 (c) A message specified in a "fail" command in a system filter;
2810 (d) A message specified in a FAIL return from the queryprogram router;
2811 (e) A message specified by the cannot_route_message router option.
2813 In these cases only, Exim does include the error details in bounce and
2814 warning messages. There are also a few cases where bland messages such
2815 as "unrouteable address" or "local delivery error" are given.
2817 PH/43 $value is now also set for the "else" part of a ${run expansion.
2819 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
2820 being worked on, but at least Exim now implements the latest version to
2823 PH/45 In a pipe transport, although a timeout while waiting for the pipe
2824 process to complete was treated as a delivery failure, a timeout while
2825 writing the message to the pipe was logged, but erroneously treated as a
2826 successful delivery. Such timeouts include transport filter timeouts. For
2827 consistency with the overall process timeout, these timeouts are now
2828 treated as errors, giving rise to delivery failures by default. However,
2829 there is now a new Boolean option for the pipe transport called
2830 timeout_defer, which, if set TRUE, converts the failures into defers for
2831 both kinds of timeout. A transport filter timeout is now identified in
2834 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
2835 systems where "make" and "gmake" are different, calling "gmake" at top
2836 level broke things. I've arranged for the value of $(MAKE) to be passed
2837 from the Makefile to this script so that it can call the same version of
2841 A note about Exim versions 4.44 and 4.50
2842 ----------------------------------------
2844 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
2845 changes of various kinds. As a consequence, a big documentation update was
2846 needed. This delayed the release for rather longer than seemed good, especially
2847 in the light of a couple of (minor) security issues. Therefore, the changes
2848 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
2849 release. So 4.44 and 4.50 are in effect two different branches that both start
2852 I have left the 4.50 change log unchanged; it contains all the changes since
2853 4.43. The change log for 4.44 is below; many of its items are identical to
2854 those for 4.50. This seems to be the most sensible way to preserve the
2855 historical information.
2861 1. Minor wording change to the doc/README.SIEVE file.
2863 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
2864 computation of the current number of files was incorrect.
2866 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
2867 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
2870 4. Give more explanation in the error message when the command for a transport
2871 filter fails to execute.
2873 5. There are several places where Exim runs a non-Exim command in a
2874 subprocess. The SIGUSR1 signal should be disabled for these processes. This
2875 was being done only for the command run by the queryprogram router. It is
2876 now done for all such subprocesses. The other cases are: ${run, transport
2877 filters, and the commands run by the lmtp and pipe transports.
2879 6. Added CONFIGURE_GROUP build-time option.
2881 7. Some older OS have a limit of 256 on the maximum number of file
2882 descriptors. Exim was using setrlimit() to set 1000 as a large value
2883 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
2884 systems. I've change it so that if it can't get 1000, it tries for 256.
2886 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
2887 was an oversight, and furthermore, ever since the addition of extra
2888 controls (e.g. 4.43/32), the checks on when to allow different forms of
2889 "control" were broken. There should now be diagnostics for all cases when a
2890 control that does not make sense is encountered.
2892 9. Added the /retain_sender option to "control=submission".
2894 10. $recipients is now available in the predata ACL (oversight).
2896 11. Tidy the search cache before the fork to do a delivery from a message
2897 received from the command line. Otherwise the child will trigger a lookup
2898 failure and thereby defer the delivery if it tries to use (for example) a
2899 cached ldap connection that the parent has called unbind on.
2901 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
2902 of $address_data from the recipient verification was clobbered by the
2903 sender verification.
2905 13. The value of address_data from a sender verification is now available in
2906 $sender_address_data in subsequent conditions in the ACL statement.
2908 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
2910 15. Added a new option "connect=<time>" to callout options, to set a different
2913 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
2914 was its contents. (It was OK if the option was not defined at all.)
2916 17. A "Completed" log line is now written for messages that are removed from
2917 the spool by the -Mrm option.
2919 18. New variables $sender_verify_failure and $recipient_verify_failure contain
2920 information about exactly what failed.
2922 19. Added -dd to debug only the daemon process.
2924 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
2925 handles timeouts, both on the server side and network timeouts. Renamed the
2926 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
2928 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
2929 It is now set to "smtps".
2931 22. $host_address is now set to the target address during the checking of
2932 ignore_target_hosts.
2934 23. When checking ignore_target_hosts for an ipliteral router, no host name was
2935 being passed; this would have caused $sender_host_name to have been used if
2936 matching the list had actually called for a host name (not very likely,
2937 since this list is usually IP addresses). A host name is now passed as
2940 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
2941 code that specifies a non-restarting handler (typically sigaction() in
2942 modern systems) in an attempt to fix a rare and obscure crash bug.
2944 25. Narrowed the window for a race in the daemon that could cause it to ignore
2945 SIGCHLD signals. This is not a major problem, because they are used only to
2946 wake it up if nothing else does.
2948 26. A malformed maildirsize file could cause Exim to calculate negative values
2949 for the mailbox size or file count. Odd effects could occur as a result.
2950 The maildirsize information is now recalculated if the size or filecount
2953 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
2954 support for a long time. Removed HAVE_SYS_VFS_H.
2956 28. Installed the latest version of exipick from John Jetmore.
2958 29. In an address list, if the pattern was not a regular expression, an empty
2959 subject address (from a bounce message) matched only if the pattern was an
2960 empty string. Non-empty patterns were not even tested. This was the wrong
2961 because it is perfectly reasonable to use an empty address as part of a
2962 database query. An empty address is now tested by patterns that are
2963 lookups. However, all the other forms of pattern expect the subject to
2964 contain a local part and a domain, and therefore, for them, an empty
2965 address still always fails if the pattern is not itself empty.
2967 30. Exim went into a mad DNS loop when attempting to do a callout where the
2968 host was specified on an smtp transport, and looking it up yielded more
2969 than one IP address.
2971 31. Re-factored the code for checking spool and log partition space into a
2972 function that finds that data and another that does the check. The former
2973 is then used to implement four new variables: $spool_space, $log_space,
2974 $spool_inodes, and $log_inodes.
2976 32. The RFC2047 encoding function was originally intended for short strings
2977 such as real names; it was not keeping to the 75-character limit for
2978 encoded words that the RFC imposes. It now respects the limit, and
2979 generates multiple encoded words if necessary. To be on the safe side, I
2980 have increased the buffer size for the ${rfc2047: expansion operator from
2983 33. It is now permitted to omit both strings after an "if" condition; if the
2984 condition is true, the result is "true". As before, when the second string
2985 is omitted, a false condition yields an empty string. This makes it less
2986 cumbersome to write custom ACL and router conditions.
2988 34. Failure to deliver a bounce message always caused it to be frozen, even if
2989 there was an errors_to setting on the router. The errors_to setting is now
2992 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
2993 canonical form (fully expanded) before being placed in
2994 $sender_host_address.
2996 36. The table in the code that translates DNS record types into text (T_A to
2997 "A" for instance) was missing entries for NS and CNAME. It is just possible
2998 that this could have caused confusion if both these types were looked up
2999 for the same domain, because the text type is used as part of Exim's
3000 per-process caching. But the chance of anyone hitting this buglet seems
3003 37. The dnsdb lookup has been extended in a number of ways.
3005 (1) There is a new type, "zns", which walks up the domain tree until it
3006 finds some nameserver records. It should be used with care.
3008 (2) There is a new type, "mxh", which is like "mx" except that it returns
3009 just the host names, not the priorities.
3011 (3) It is now possible to give a list of domains (or IP addresses) to be
3012 looked up. The behaviour when one of the lookups defers can be
3013 controlled by a keyword.
3015 (4) It is now possible to specify the separator character for use when
3016 multiple records are returned.
3018 38. The dnslists ACL condition has been extended: it is now possible to supply
3019 a list of IP addresses and/or domains to be looked up in a particular DNS
3022 39. Added log_selector=+queue_time_overall.
3024 40. When running the queue in the test harness, wait just a tad after forking a
3025 delivery process, to get repeatability of debugging output.
3027 41. Include certificate and key file names in error message when GnuTLS fails
3028 to set them up, because the GnuTLS error message doesn't include the name
3029 of the failing file when there is a problem reading it.
3031 42. Allow both -bf and -bF in the same test run.
3033 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
3035 44. The "Exiscan patch" is now merged into the mainline Exim source.
3037 45. Sometimes the final signoff response after QUIT could fail to get
3038 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
3039 < 0 before doing a fflush(). This bug looks as though it goes back to the
3040 introduction of TLS in release 3.20, but "sometimes" must have been rare
3041 because the tests only now provoked it.
3043 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
3044 (this can affect the format of dates).
3046 47. exim_tidydb, when checking for the continued existence of a message for
3047 which it has found a message-specific retry record, was not finding
3048 messages that were in split spool directories. Consequently, it was
3049 deleting retry records that should have stayed in existence.
3051 48. Steve fixed some bugs in eximstats.
3053 49. The SPA authentication driver was not abandoning authentication and moving
3054 on to the next authenticator when an expansion was forced to fail,
3055 contradicting the general specification for all authenticators. Instead it
3056 was generating a temporary error. It now behaves as specified.
3058 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
3059 (the order specifies the preference for clients). The order is now AES256,
3060 AES128, 3DES, ARCFOUR128.
3062 51. Small patch to Sieve code - explicitly set From: when generating an
3065 52. Exim crashed if a remote delivery caused a very long error message to be
3066 recorded - for instance if somebody sent an entire SpamAssassin report back
3067 as a large number of 550 error lines. This bug was coincidentally fixed by
3068 increasing the size of one of Exim's internal buffers (big_buffer) that
3069 happened as part of the Exiscan merge. However, to be on the safe side, I
3070 have made the code more robust (and fixed the comments that describe what
3073 53. Now that there can be additional text after "Completed" in log lines (if
3074 the queue_time_overall log selector is set), a one-byte patch to exigrep
3075 was needed to allow it to recognize "Completed" as not the last thing in
3078 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
3079 patch that reportedly fixes this has been added. I am not expert enough to
3080 create a test for it. This is what the patch creator wrote:
3082 "I found a little strange behaviour of ldap code when working with
3083 Windows 2003 AD Domain, where users was placed in more than one
3084 Organization Units. When I tried to give exim partial DN, the exit code
3085 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
3086 But simultaneously result of request was absolutely normal ldap result,
3087 so I produce this patch..."
3089 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
3090 REFERENCE, so I have modified the code to exclude the patch when that macro
3093 55. Some experimental protocols are using DNS PTR records for new purposes. The
3094 keys for these records are domain names, not reversed IP addresses. The
3095 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
3096 leaves it alone. Component reversal etc. now happens only for IP addresses.
3099 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
3101 57. Double the size of the debug message buffer (to 2048) so that more of very
3102 long debug lines gets shown.
3104 58. The exicyclog utility now does better if the number of log files to keep
3105 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
3107 59. Two changes related to the smtp_active_hostname option:
3109 (1) $smtp_active_hostname is now available as a variable.
3110 (2) The default for smtp_banner uses $smtp_active_hostname instead
3111 of $primary_hostname.
3113 60. The host_aton() function is supposed to be passed a string that is known
3114 to be a valid IP address. However, in the case of IPv6 addresses, it was
3115 not checking this. This is a hostage to fortune. Exim now panics and dies
3116 if the condition is not met. A case was found where this could be provoked
3117 from a dnsdb PTR lookup with an IPv6 address that had more than 8
3118 components; fortuitously, this particular loophole had already been fixed
3119 by change 4.50/55 above.
3121 If there are any other similar loopholes, the new check in host_aton()
3122 itself should stop them being exploited. The report I received stated that
3123 data on the command line could provoke the exploit when Exim was running as
3124 exim, but did not say which command line option was involved. All I could
3125 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
3126 running as the user.
3129 61. There was a buffer overflow vulnerability in the SPA authentication code
3130 (which came originally from the Samba project). I have added a test to the
3131 spa_base64_to_bits() function which I hope fixes it.
3134 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
3135 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
3137 63. The daemon start-up calls getloadavg() while still root for those OS that
3138 need the first call to be done as root, but it missed one case: when
3139 deliver_queue_load_max is set with deliver_drop_privilege. This is
3140 necessary for the benefit of the queue runner, because there is no re-exec
3141 when deliver_drop_privilege is set.
3143 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
3144 This has been fixed.
3146 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
3147 was in use, was not putting the data itself into the right store pool;
3148 consequently, it could be overwritten for a subsequent message in the same
3149 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
3152 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
3154 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
3155 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
3156 the function were treating the return as a boolean value, which happened to
3157 work because 0=false and not-0=true, but is not correct code.
3159 68. The host_aton() function was not handling scoped IPv6 addresses (those
3160 with, for example, "%eth0" on the end) correctly.
3162 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
3163 negated items (that is, ~something) in unsigned ints. Some compilers
3164 apparently mutter when there is no cast.
3166 70. If an address verification called from an ACL failed, and did not produce a
3167 user-specific message (i.e. there was only a "system" message), nothing was
3168 put in $acl_verify_message. In this situation, it now puts the system
3171 71. Change 4.23/11 added synchronization checking at the start of an SMTP
3172 session; change 4.31/43 added the unwanted input to the log line - except
3173 that it did not do this in the start of session case. It now does.
3175 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
3176 This is wrong and can cause the other end to generate a synchronization
3177 error if it is another Exim or anything else that does the synchronization
3178 check. A QUIT command is no longer sent after a timeout.
3180 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
3181 during host lookups.
3183 74. The defer_ok option of callout verification was not working if it was used
3184 when verifying addresses in header lines, that is, for this case:
3186 verify = header_sender/callout=defer_ok
3188 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
3189 those file descriptors could be used for SMTP connections. If anything
3190 wrote to stderr (the example that came up was "warn" in embedded Perl), it
3191 could be sent to the SMTP client, causing chaos. The daemon now opens
3192 stdin, stdout, and stderr to /dev/null when it puts itself into the
3195 76. Arrange for output from Perl's "warn" command to be written to Exim's main
3196 log by default. The user can override this with suitable Perl magic.
3198 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
3199 the log message when discard triggers, was not working for the DATA ACL or
3200 for the non-SMTP ACL.
3202 78. Error message wording change in sieve.c.
3204 79. If smtp_accept_max_per_host was set, the number of connections could be
3205 restricted to fewer than expected, because the daemon was trying to set up
3206 a new connection before checking whether the processes handling previous
3207 connections had finished. The check for completed processes is now done
3208 earlier. On busy systems, this bug wouldn't be noticed because something
3209 else would have woken the daemon, and it would have reaped the completed
3212 80. If a message was submitted locally by a user whose login name contained one
3213 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
3214 It caused a spool format error. I have fixed the spool reading code. A
3215 related problem was that the "from" clause in the Received: line became
3216 illegal because of the space(s). It is now covered by ${quote_local_part.
3218 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
3221 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
3223 83. Patch from Sieve maintainer to fix unterminated string problem in
3224 "vacation" handling.
3226 84. Some minor changes to the Linux configuration files to help with other
3227 OS variants using glibc.
3229 85. One more patch for Sieve to update vacation handling to latest spec.
3232 ----------------------------------------------------
3233 See the note above about the 4.44 and 4.50 releases.
3234 ----------------------------------------------------
3240 1. Change 4.43/35 introduced a bug that caused file counts to be
3241 incorrectly computed when quota_filecount was set in an appendfile
3244 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
3245 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
3248 3. Give more explanation in the error message when the command for a transport
3249 filter fails to execute.
3251 4. There are several places where Exim runs a non-Exim command in a
3252 subprocess. The SIGUSR1 signal should be disabled for these processes. This
3253 was being done only for the command run by the queryprogram router. It is
3254 now done for all such subprocesses. The other cases are: ${run, transport
3255 filters, and the commands run by the lmtp and pipe transports.
3257 5. Some older OS have a limit of 256 on the maximum number of file
3258 descriptors. Exim was using setrlimit() to set 1000 as a large value
3259 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
3260 systems. I've change it so that if it can't get 1000, it tries for 256.
3262 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
3263 was an oversight, and furthermore, ever since the addition of extra
3264 controls (e.g. 4.43/32), the checks on when to allow different forms of
3265 "control" were broken. There should now be diagnostics for all cases when a
3266 control that does not make sense is encountered.
3268 7. $recipients is now available in the predata ACL (oversight).
3270 8. Tidy the search cache before the fork to do a delivery from a message
3271 received from the command line. Otherwise the child will trigger a lookup
3272 failure and thereby defer the delivery if it tries to use (for example) a
3273 cached ldap connection that the parent has called unbind on.
3275 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
3276 of $address_data from the recipient verification was clobbered by the
3277 sender verification.
3279 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
3280 was its contents. (It was OK if the option was not defined at all.)
3282 11. A "Completed" log line is now written for messages that are removed from
3283 the spool by the -Mrm option.
3285 12. $host_address is now set to the target address during the checking of
3286 ignore_target_hosts.
3288 13. When checking ignore_target_hosts for an ipliteral router, no host name was
3289 being passed; this would have caused $sender_host_name to have been used if
3290 matching the list had actually called for a host name (not very likely,
3291 since this list is usually IP addresses). A host name is now passed as
3294 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
3295 code that specifies a non-restarting handler (typically sigaction() in
3296 modern systems) in an attempt to fix a rare and obscure crash bug.
3298 15. Narrowed the window for a race in the daemon that could cause it to ignore
3299 SIGCHLD signals. This is not a major problem, because they are used only to
3300 wake it up if nothing else does.
3302 16. A malformed maildirsize file could cause Exim to calculate negative values
3303 for the mailbox size or file count. Odd effects could occur as a result.
3304 The maildirsize information is now recalculated if the size or filecount
3307 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
3308 support for a long time. Removed HAVE_SYS_VFS_H.
3310 18. Updated exipick to current release from John Jetmore.
3312 19. Allow an empty sender to be matched against a lookup in an address list.
3313 Previously the only cases considered were a regular expression, or an
3316 20. Exim went into a mad DNS lookup loop when doing a callout where the
3317 host was specified on the transport, if the DNS lookup yielded more than
3320 21. The RFC2047 encoding function was originally intended for short strings
3321 such as real names; it was not keeping to the 75-character limit for
3322 encoded words that the RFC imposes. It now respects the limit, and
3323 generates multiple encoded words if necessary. To be on the safe side, I
3324 have increased the buffer size for the ${rfc2047: expansion operator from
3327 22. Failure to deliver a bounce message always caused it to be frozen, even if
3328 there was an errors_to setting on the router. The errors_to setting is now
3331 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
3332 canonical form (fully expanded) before being placed in
3333 $sender_host_address.
3335 24. Updated eximstats to version 1.33
3337 25. Include certificate and key file names in error message when GnuTLS fails
3338 to set them up, because the GnuTLS error message doesn't include the name
3339 of the failing file when there is a problem reading it.
3341 26. Expand error message when OpenSSL has problems setting up cert/key files.
3344 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
3345 (this can affect the format of dates).
3347 28. exim_tidydb, when checking for the continued existence of a message for
3348 which it has found a message-specific retry record, was not finding
3349 messages that were in split spool directories. Consequently, it was
3350 deleting retry records that should have stayed in existence.
3352 29. eximstats updated to version 1.35
3353 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
3354 1.35 - bugfix such that pie charts by volume are generated correctly
3356 30. The SPA authentication driver was not abandoning authentication and moving
3357 on to the next authenticator when an expansion was forced to fail,
3358 contradicting the general specification for all authenticators. Instead it
3359 was generating a temporary error. It now behaves as specified.
3361 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
3362 (the order specifies the preference for clients). The order is now AES256,
3363 AES128, 3DES, ARCFOUR128.
3365 31. Small patch to Sieve code - explicitly set From: when generating an
3368 32. Exim crashed if a remote delivery caused a very long error message to be
3369 recorded - for instance if somebody sent an entire SpamAssassin report back
3370 as a large number of 550 error lines. This bug was coincidentally fixed by
3371 increasing the size of one of Exim's internal buffers (big_buffer) that
3372 happened as part of the Exiscan merge. However, to be on the safe side, I
3373 have made the code more robust (and fixed the comments that describe what
3376 33. Some experimental protocols are using DNS PTR records for new purposes. The
3377 keys for these records are domain names, not reversed IP addresses. The
3378 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
3379 leaves it alone. Component reversal etc. now happens only for IP addresses.
3382 34. The host_aton() function is supposed to be passed a string that is known
3383 to be a valid IP address. However, in the case of IPv6 addresses, it was
3384 not checking this. This is a hostage to fortune. Exim now panics and dies
3385 if the condition is not met. A case was found where this could be provoked
3386 from a dnsdb PTR lookup with an IPv6 address that had more than 8
3387 components; fortuitously, this particular loophole had already been fixed
3388 by change 4.50/55 or 4.44/33 above.
3390 If there are any other similar loopholes, the new check in host_aton()
3391 itself should stop them being exploited. The report I received stated that
3392 data on the command line could provoke the exploit when Exim was running as
3393 exim, but did not say which command line option was involved. All I could
3394 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
3395 running as the user.
3398 35. There was a buffer overflow vulnerability in the SPA authentication code
3399 (which came originally from the Samba project). I have added a test to the
3400 spa_base64_to_bits() function which I hope fixes it.
3403 36. The daemon start-up calls getloadavg() while still root for those OS that
3404 need the first call to be done as root, but it missed one case: when
3405 deliver_queue_load_max is set with deliver_drop_privilege. This is
3406 necessary for the benefit of the queue runner, because there is no re-exec
3407 when deliver_drop_privilege is set.
3409 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
3410 was in use, was not putting the data itself into the right store pool;
3411 consequently, it could be overwritten for a subsequent message in the same
3412 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
3415 38. Sometimes the final signoff response after QUIT could fail to get
3416 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
3417 < 0 before doing a fflush(). This bug looks as though it goes back to the
3418 introduction of TLS in release 3.20, but "sometimes" must have been rare
3419 because the tests only now provoked it.
3425 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
3426 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
3427 Now it returns an integer. A number of calls were still expecting a T/F
3428 return. Fortuitously, in all cases, the tests worked in OK situations,
3429 which is the norm. However, things would have gone wrong on any write
3430 failures on the smtp file descriptor. This function is used when sending
3431 messages over SMTP and also when doing verify callouts.
3433 2. When Exim is called to do synchronous delivery of a locally submitted
3434 message (the -odf or -odi options), it no longer closes stderr before doing
3437 3. Implemented the mua_wrapper option.
3439 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
3441 5. Implemented the functions header_remove(), header_testname(),
3442 header_add_at_position(), and receive_remove_recipient(), and exported them
3445 6. If an ACL "warn" statement specified the addition of headers, Exim already
3446 inserted X-ACL-Warn: at the start if there was no header name. However, it
3447 was not making this test for the second and subsequent header lines if
3448 there were newlines in the string. This meant that an invalid header could
3449 be inserted if Exim was badly configured.
3451 7. Allow an ACL "warn" statement to add header lines at the start or after all
3452 the Received: headers, as well as at the end.
3454 8. Added the rcpt_4xx retry error code.
3456 9. Added postmaster_mailfrom=xxx to callout verification option.
3458 10. Added mailfrom=xxxx to the callout verification option, for verify=
3461 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
3462 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
3464 12. Inserted some casts to stop certain compilers warning when using pointer
3465 differences as field lengths or precisions in printf-type calls (mostly
3466 affecting debugging statements).
3468 13. Added optional readline() support for -be (dynamically loaded).
3470 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
3471 same clock tick as a message's arrival, so that its received time was the
3472 same as the "first fail" time on the retry record, and that message
3473 remained on the queue past the ultimate address timeout, every queue runner
3474 would try a delivery (because it was past the ultimate address timeout) but
3475 after another failure, the ultimate address timeout, which should have then
3476 bounced the address, did not kick in. This was a "< instead of <=" error;
3477 in most cases the first failure would have been in the next clock tick
3478 after the received time, and all would be well.
3480 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
3481 being recognized when the domain list was tested by the match_domain
3482 condition in an expansion string.
3484 16. Added the ${str2b64: operator.
3486 17. Exim was always calling setrlimit() to set a large limit for the number of
3487 processes, without checking whether the existing limit was already
3488 adequate. (It did check for the limit on file descriptors.) Furthermore,
3489 errors from getrlimit() and setrlimit() were being ignored. Now they are
3490 logged to the main and panic logs, but Exim does carry on, to try to do its
3491 job under whatever limits there are.
3493 18. Imported PCRE 5.0.
3495 19. Trivial typo in log message " temporarily refused connection" (the leading
3498 20. If the log selector return_path_on_delivery was set and an address was
3499 redirected to /dev/null, the delivery process crashed because it assumed
3500 that a return path would always be set for a "successful" delivery. In this
3501 case, the whole delivery is bypassed as an optimization, and therefore no
3504 21. Internal re-arrangement: the function for sending a challenge and reading
3505 a response while authentication was assuming a zero-terminated challenge
3506 string. It's now changed to take a pointer and a length, to allow for
3507 binary data in such strings.
3509 22. Added the cyrus_sasl authenticator (code supplied by MBM).
3511 23. Exim was not respecting finduser_retries when seeking the login of the
3512 uid under which it was called; it was always trying 10 times. (The default
3513 setting of finduser_retries is zero.) Also, it was sleeping after the final
3514 failure, which is pointless.
3516 24. Implemented tls_on_connect_ports.
3518 25. Implemented acl_smtp_predata.
3520 26. If the domain in control=submission is set empty, Exim assumes that the
3521 authenticated id is a complete email address when it generates From: or
3522 Sender: header lines.
3524 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
3525 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
3526 chown and chgrp in /bin and hostname in /usr/bin.
3528 28. Exim was keeping the "process log" file open after each use, just as it
3529 does for the main log. This opens the possibility of it remaining open for
3530 long periods when the USR1 signal hits a daemon. Occasional processlog
3531 errors were reported, that could have been caused by this. Anyway, it seems
3532 much more sensible not to leave this file open at all, so that is what now
3535 29. The long-running daemon process does not normally write to the log once it
3536 has entered its main loop, and it closes the log before doing so. This is
3537 so that log files can straightforwardly be renamed and moved. However,
3538 there are a couple of unusual error situations where the daemon does write
3539 log entries, and I had neglected to close the log afterwards.
3541 30. The text of an SMTP error response that was received during a remote
3542 delivery was being truncated at 512 bytes. This is too short for some of
3543 the long messages that one sometimes sees. I've increased the limit to
3546 31. It is now possible to make retry rules that apply only when a message has a
3547 specific sender, in particular, an empty sender.
3549 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
3550 it possible to be selective about when SMTP synchronization is enforced.
3552 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
3554 32. Implemented hosts_connection_nolog.
3556 33. Added an ACL for QUIT.
3558 34. Setting "delay_warning=" to disable warnings was not working; it gave a
3561 35. Added mailbox_size and mailbox_filecount to appendfile.
3563 36. Added control = no_multiline_responses to ACLs.
3565 37. There was a bug in the logic of the code that waits for the clock to tick
3566 in the case where the clock went backwards by a substantial amount such
3567 that the microsecond fraction of "now" was more than the microsecond
3568 fraction of "then" (but the whole seconds number was less).
3570 38. Added support for the libradius Radius client library this is found on
3571 FreeBSD (previously only the radiusclient library was supported).
3577 1. When certain lookups returned multiple values in the form name=value, the
3578 quoting of the values was not always being done properly. Specifically:
3579 (a) If the value started with a double quote, but contained no whitespace,
3581 (b) If the value contained whitespace other than a space character (i.e.
3582 tabs or newlines or carriage returns) it was not quoted.
3583 This fix has been applied to the mysql and pgsql lookups by writing a
3584 separate quoting function and calling it from the lookup code. The fix
3585 should probably also be applied to nisplus, ibase and oracle lookups, but
3586 since I cannot test any of those, I have not disturbed their existing code.
3588 2. A hit in the callout cache for a specific address caused a log line with no
3589 reason for rejecting RCPT. Now it says "Previous (cached) callout
3590 verification failure".
3592 3. There was an off-by-one bug in the queryprogram router. An over-long
3593 return line was truncated at 256 instead of 255 characters, thereby
3594 overflowing its buffer with the terminating zero. As well as fixing this, I
3595 have increased the buffer size to 1024 (and made a note to document this).
3597 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
3598 when Exim is waiting for an SMTP response from a remote server, Exim
3599 restarts its select() call on the socket, thereby resetting its timeout.
3600 This is not a problem when such interrupts are rare. Somebody set up a cron
3601 job to run exiwhat every 2 minutes, which is less than the normal select()
3602 timeout (5 or 10 minutes). This meant that the select() timeout never
3603 kicked in because it was always reset. I have fixed this by comparing the
3604 time when an interrupt arrives with the time at the start of the first call
3605 to select(). If more time than the timeout has elapsed, the interrupt is
3606 treated as a timeout.
3608 5. Some internal re-factoring in preparation for the addition of Sieve
3609 extensions (by MH). In particular, the "personal" test is moved to a
3610 separate function, and given an option for scanning Cc: and Bcc: (which is
3611 not set for Exim filters).
3613 6. When Exim created an email address using the login of the caller as the
3614 local part (e.g. when creating a From: or Sender: header line), it was not
3615 quoting the local part when it contained special characters such as @.
3617 7. Installed new OpenBSD configuration files.
3619 8. Reworded some messages for syntax errors in "and" and "or" conditions to
3620 try to make them clearer.
3622 9. Callout options, other than the timeout value, were being ignored when
3623 verifying sender addresses in header lines. For example, when using
3625 verify = header_sender/callout=no_cache
3627 the cache was (incorrectly) being used.
3629 10. Added a missing instance of ${EXE} to the exim_install script; this affects
3630 only the Cygwin environment.
3632 11. When return_path_on_delivery was set as a log selector, if different remote
3633 addresses in the same message used different return paths and parallel
3634 remote delivery occurred, the wrong values would sometimes be logged.
3635 (Whenever a remote delivery process finished, the return path value from
3636 the most recently started remote delivery process was logged.)
3638 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
3639 lines when AUTH and/or TLS are in use. This is the "received protocol"
3640 field. Exim used to use "asmtp" for authenticated SMTP, without any
3641 indication (in the protocol name) for TLS use. Now it follows the RFC and
3642 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
3643 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
3644 names appear in log lines as well as in Received: header lines.
3646 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
3647 extensions, and comparison tests, and to fix some bugs.
3649 14. Changes to the "personal" filter test:
3651 (1) The test was buggy in that it was just doing the equivalent of
3652 "contains" tests on header lines. For example, if a user's address was
3653 anne@some.where, the "personal" test would incorrectly be true for
3655 To: susanne@some.where
3657 This test is now done by extracting each address from the header in turn,
3658 and checking the entire address. Other tests that are part of "personal"
3659 are now done using regular expressions (for example, to check local parts
3660 of addresses in From: header lines).
3662 (2) The list of non-personal local parts in From: addresses has been
3663 extended to include "listserv", "majordomo", "*-request", and "owner-*",
3664 taken from the Sieve specification recommendations.
3666 (3) If the message contains any header line starting with "List-" it is
3667 treated as non-personal.
3669 (4) The test for "circular" in the Subject: header line has been removed
3670 because it now seems ill-conceived.
3672 15. Minor typos in src/EDITME comments corrected.
3674 16. Installed latest exipick from John Jetmore.
3676 17. If headers_add on a router specified a text string that was too long for
3677 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
3678 of string_sprintf() is now avoided.
3680 18. $message_body_size was not set (it was always zero) when running the DATA
3681 ACL and the local_scan() function.
3683 19. For the "mail" command in an Exim filter, no default was being set for
3684 the once_repeat time, causing a random time value to be used if "once" was
3685 specified. (If the value happened to be <= 0, no repeat happened.) The
3686 default is now 0s, meaning "never repeat". The "vacation" command was OK
3687 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
3688 (I found it when inspecting the code).
3690 20. There is now an overall timeout for performing a callout verification. It
3691 defaults to 4 times the callout timeout, which applies to individual SMTP
3692 commands during the callout. The overall timeout applies when there is more
3693 than one host that can be tried. The timeout is checked before trying the
3694 next host. This prevents very long delays if there are a large number of
3695 hosts and all are timing out (e.g. when the network connections are timing
3696 out). The value of the overall timeout can be changed by specifying an
3697 additional sub-option for "callout", called "maxwait". For example:
3699 verify = sender/callout=5s,maxwait=20s
3701 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
3702 to the end before writing, but this should make it even safer).
3704 22. Exim was forgetting that it had advertised PIPELINING for the second and
3705 subsequent messages on an SMTP connection. It was also not resetting its
3706 memory on STARTTLS and an internal HELO.
3708 23. When Exim logs an SMTP synchronization error within a session, it now
3709 records whether PIPELINING has been advertised or not.
3711 24. Added 3 instances of "(long int)" casts to time_t variables that were being
3712 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
3713 rather than long int.
3715 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
3717 26. Added the never_mail option to autoreply.
3723 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
3724 crash if the getsockname() call failed; this can happen if a connection is
3725 closed very soon after it is established. The problem was simply in the
3726 order in which certain operations were done, causing Exim to try to write
3727 to the SMTP stream before it had set up the file descriptor. The bug has
3728 been fixed by making things happen in the correct order.
3734 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
3735 before the connection was closed, thus losing the rejection response.
3737 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
3738 some early Solaris releases, but causes trouble in current releases where
3739 socklen_t is defined.
3741 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
3744 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
3747 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
3748 that releasing the top of it at the end releases what was used for sub-
3749 expansions (unless the block got too big). However, discard this block if
3750 the first thing is a variable or header, so that we can use its block when
3751 it is dynamic (useful for very large $message_headers, for example).
3753 6. Lookups now cache *every* query, not just the most recent. A new, separate
3754 store pool is used for this. It can be recovered when all lookup caches are
3755 flushed. Lookups now release memory at the end of their result strings.
3756 This has involved some general refactoring of the lookup sources.
3758 7. Some code has been added to the store_xxx() functions to reduce the amount
3759 of flapping under certain conditions.
3761 8. log_incoming_interface used to affect only the <= reception log lines. Now
3762 it causes the local interface and port to be added to several more SMTP log
3763 lines, for example "SMTP connection from", and rejection lines.
3765 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
3767 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
3769 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
3770 could be overwritten at the end of the current message (or the start of a
3771 new message if it was set in a HELO ACL). The value is now preserved for
3772 the duration of the SMTP connection.
3774 12. If a transport had a headers_rewrite setting, and a matching header line
3775 contained an unqualified address, that address was qualified, even if it
3776 did not match any rewriting rules. The underlying bug was that the values
3777 of the flags that permit the existence of unqualified sender and recipient
3778 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
3779 non-local messages, and by -bnq for local messages) were not being
3780 preserved with the message after it was received.
3782 13. When Exim was logging an SMTP synchronization error, it could sometimes log
3783 "next input=" as part of the text comprising the host identity instead of
3784 the correct text. The code was using the same buffer for two different
3785 strings. However, depending on which order the printing function evaluated
3786 its arguments, the bug did not always show up. Under Linux, for example, my
3787 test suite worked just fine.
3789 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
3790 doesn't work with some older versions of Perl. It has been changed to "my",
3791 which in any case is probably the better facility to use.
3793 15. A really picky compiler found some instances of statements for creating
3794 error messages that either had too many or two few arguments for the format
3797 16. The size of the buffer for calls to the DNS resolver has been increased
3798 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
3799 for addresses that have a lot of PTR records. This alleviates a problem; it
3800 does not fully solve it.
3802 17. A dnsdb lookup for PTR records that receives more data than will fit in the
3803 buffer now truncates the list and logs the incident, which is the same
3804 action as happens when Exim is looking up a host name and its aliases.
3805 Previously in this situation something unpredictable would happen;
3806 sometimes it was "internal error: store_reset failed".
3808 18. If a server dropped the connection unexpectedly when an Exim client was
3809 using GnuTLS and trying to read a response, the client delivery process
3810 crashed while trying to generate an error log message.
3812 19. If a "warn" verb in an ACL added multiple headers to a message in a single
3813 string, for example:
3815 warn message = H1: something\nH2: something
3817 the text was added as a single header line from Exim's point of view
3818 though it ended up OK in the delivered message. However, searching for the
3819 second and subsequent header lines using $h_h2: did not work. This has been
3820 fixed. Similarly, if a system filter added multiple headers in this way,
3821 the routers could not see them.
3823 20. Expanded the error message when iplsearch is called with an invalid key to
3824 suggest using net-iplsearch in a host list.
3826 21. When running tests using -bh, any delays imposed by "delay" modifiers in
3827 ACLs are no longer actually imposed (and a message to that effect is
3830 22. If a "gecos" field in a passwd entry contained escaped characters, in
3831 particular, if it contained a \" sequence, Exim got it wrong when building
3832 a From: or a Sender: header from that name. A second bug also caused
3833 incorrect handling when an unquoted " was present following a character
3834 that needed quoting.
3836 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
3837 was not being matched caselessly.
3839 24. Arranged for all hyphens in the exim.8 source to be escaped with
3842 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
3843 properly. Recipient callout cache records were still being keyed to include
3844 the sender, even when use_sender was set false. This led to far more
3845 callouts that were necessary. The sender is no longer included in the key
3846 when use_sender is false.
3848 26. Added "control = submission" modifier to ACLs.
3850 27. Added the ${base62d: operator to decode base 62 numbers.
3852 28. dnsdb lookups can now access SRV records.
3854 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
3855 the configuration file.
3857 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
3858 (-v) mode. This makes the output for a verbose queue run more intelligible.
3860 31. Added a use_postmaster feature to recipient callouts.
3862 32. Added the $body_zerocount variable, containing the number of binary zero
3863 bytes in the message body.
3865 33. The time of last modification of the "new" subdirectory is now used as the
3866 "mailbox time last read" when there is a quota error for a maildir
3869 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
3871 35. Added +ignore_unknown as a special item in host lists.
3873 36. Code for decoding IPv6 addresses in host lists is now included, even if
3874 IPv6 support is not being compiled. This fixes a bug in which an IPv6
3875 address was recognized as an IP address, but was then not correctly decoded
3876 into binary, causing unexpected and incorrect effects when compared with
3883 1. Very minor rewording of debugging text in manualroute to say "list of
3884 hosts" instead of "hostlist".
3886 2. If verify=header_syntax was set, and a header line with an unqualified
3887 address (no domain) and a large number of spaces between the end of the
3888 name and the colon was received, the reception process suffered a buffer
3889 overflow, and (when I tested it) crashed. This was caused by some obsolete
3890 code that should have been removed. The fix is to remove it!
3892 3. When running in the test harness, delay a bit after writing a bounce
3893 message to get a bit more predictability in the log output.
3895 4. Added a call to search_tidyup() just before forking a reception process. In
3896 theory, someone could use a lookup in the expansion of smtp_accept_max_
3897 per_host which, without the tidyup, could leave open a database connection.
3899 5. Added the variables $recipient_data and $sender_data which get set from a
3900 lookup success in an ACL "recipients" or "senders" condition, or a router
3901 "senders" option, similar to $domain_data and $local_part_data.
3903 6. Moved the writing of debug_print from before to after the "senders" test
3906 7. Change 4.31/66 (moving the time when the Received: is generated) caused
3907 problems for message scanning, either using a data ACL, or using
3908 local_scan() because the Received: header was not generated till after they
3909 were called (in order to set the time as the time of reception completion).
3910 I have revised the way this works. The header is now generated after the
3911 body is received, but before the ACL or local_scan() are called. After they
3912 are run, the timestamp in the header is updated.
3918 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
3919 before starting a queue runner without re-exec. This happened only when
3920 deliver_drop_privilege was set or when the Exim user was set to root. The
3921 effect of the bug was that timeouts during subsequent deliveries caused
3922 crashes instead of being properly handled. The handler is now left at its
3923 default (and expected) setting.
3925 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
3926 message, again when deliver_drop_privilege is set or Exim is run as root.
3927 The bug described in (1) was not present in this case, but the tidying up
3928 of the other signals was missing. I have made the two cases consistent.
3930 3. The ignore_target_hosts setting on a manualroute router was being ignored
3931 for hosts that were looked up using the /MX notation.
3933 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
3936 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
3937 operated on the sender address. After changing the $sender_address to <>
3938 for the sender address verify, Exim was re-instated it as the original
3939 (before rewriting) address, but remembering that it had rewritten it, so it
3940 wasn't rewriting it again. This bug also had the effect of breaking the
3941 sender address verification caching when the sender address was rewritten.
3943 6. The ignore_target_hosts option was being ignored by the ipliteral router.
3944 This has been changed so that if the ip literal address matches
3945 ignore_target_hosts, the router declines.
3947 7. Added expansion conditions match_domain, match_address, and match_local_
3948 part (NOT match_host).
3950 8. The placeholder for the Received: header didn't have a length field set.
3952 9. Added code to Exim itself and to exim_lock to test for a specific race
3953 condition that could lead to file corruption when using MBX delivery. The
3954 issue is with the lockfile that is created in /tmp. If this file is removed
3955 after a process has opened it but before that process has acquired a lock,
3956 there is the potential for a second process to recreate the file and also
3957 acquire a lock. This could lead to two Exim processes writing to the file
3958 at the same time. The added code performs the same test as UW imapd; it
3959 checks after acquiring the lock that its file descriptor still refers to
3960 the same named file.
3962 10. The buffer for building added header lines was of fixed size, 8192 bytes.
3963 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
3966 11. Added the smtp_active_hostname option. If used, this will typically be made
3967 to depend on the incoming interface address. Because $interface_address is
3968 not set up until the daemon has forked a reception process, error responses
3969 that can happen earlier (such as "too many connections") no longer contain
3972 12. If an expansion in a condition on a "warn" statement fails because a lookup
3973 defers, the "warn" statement is abandoned, and the next ACL statement is
3974 processed. Previously this caused the whole ACL to be aborted.
3976 13. Added the iplsearch lookup type.
3978 14. Added ident_timeout as a log selector.
3980 15. Added tls_certificate_verified as a log selector.
3982 16. Added a global option tls_require_ciphers (compare the smtp transport
3983 option of the same name). This controls incoming TLS connections.
3985 17. I finally figured out how to make tls_require_ciphers do a similar thing
3986 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
3987 before starting the TLS session.
3989 18. Tabs are now shown as \t in -bP output.
3991 19. If the log selector return_path_on_delivery was set, Exim crashed when
3992 bouncing a message because it had too many Received: header lines.
3994 20. If two routers both had headers_remove settings, and the first one included
3995 a superfluous trailing colon, the final name in the first list and the
3996 first name in the second list were incorrectly joined into one item (with a
3997 colon in the middle).
4003 1. Added -C and -D options to the exinext utility, mainly to make it easier
4004 to include in the automated testing, but these could be helpful when
4005 multiple configurations are in use.
4007 2. The exinext utility was not formatting the output nicely when there was
4008 an alternate port involved in the retry record key, nor when there was a
4009 message id as well (for retries that were specific to a specific message
4010 and a specific host). It was also confused by IPv6 addresses, because of
4011 the additional colons they contain. I have fixed the IPv4 problem, and
4012 patched it up to do a reasonable job for IPv6.
4014 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
4015 delivery, the log line now contains "pipelined" if PIPELINING was used.
4017 4. An SMTP transport process used to panic and die if the bind() call to set
4018 an explicit outgoing interface failed. This has been changed; it is now
4019 treated in the same way as a connect() failure.
4021 5. A reference to $sender_host_name in the part of a conditional expansion
4022 that was being skipped was still causing a DNS lookup. This no longer
4025 6. The def: expansion condition was not recognizing references to header lines
4026 that used bh_ and bheader_.
4028 7. Added the _cache feature to named lists.
4030 8. The code for checking quota_filecount in the appendfile transport was
4031 allowing one more file than it should have been.
4033 9. For compatibility with Sendmail, the command line option
4041 and sets the incoming protocol and host name (for trusted callers). The
4042 host name and its colon can be omitted when only the protocol is to be set.
4043 Note the Exim already has two private options, -pd and -ps, that refer to
4044 embedded Perl. It is therefore impossible to set a protocol value of "d" or
4045 "s", but I don't think that's a major issue.
4047 10. A number of refactoring changes to the code, none of which should affect
4050 (a) The number of logging options was getting close to filling up the
4051 32-bit word that was used as a bit map. I have split them into two classes:
4052 those that are passed in the argument to log_write(), and those that are
4053 only ever tested independently outside of that function. These are now in
4054 separate 32-bit words, so there is plenty of room for expansion again.
4055 There is no change in the user interface or the logging behaviour.
4057 (b) When building, for example, log lines, the code previously used a
4058 macro that called string_cat() twice, in order to add two strings. This is
4059 not really sufficiently general. Furthermore, there was one instance where
4060 it was actually wrong because one of the argument was used twice, and in
4061 one call a function was used. (As it happened, calling the function twice
4062 did not affect the overall behaviour.) The macro has been replaced by a
4063 function that can join an arbitrary number of extra strings onto a growing
4066 (c) The code for expansion conditions now uses a table and a binary chop
4067 instead of a serial search (which was left over from when there were very
4068 few conditions). Also, it now recognizes conditions like "pam" even when
4069 the relevant support is not compiled in: a suitably worded error message is
4070 given if an attempt is made to use such a condition.
4072 11. Added ${time_interval:xxxxx}.
4074 12. A bug was causing one of the ddress fields not to be passed back correctly
4075 from remote delivery subprocesses. The field in question was not being
4076 subsequently used, so this caused to problems in practice.
4078 13. Added new log selectors queue_time and deliver_time.
4080 14. Might have fixed a bug in maildirsizefile handling that threw up
4081 "unexpected character" debug warnings, and recalculated the data
4082 unnecessarily. In any case, I expanded the warning message to give more
4085 15. Added the message "Restricted characters in address" to the statements in
4086 the default ACL that block characters like @ and % in local parts.
4088 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
4089 Three changes have been made:
4091 (a) There was a serious bug; a negative response to MAIL caused the whole
4092 recipient domain to be cached as invalid, thereby blocking all messages
4093 to all local parts at the same domain, from all senders. This bug has
4094 been fixed. The domain is no longer cached after a negative response to
4095 MAIL if the sender used is not empty.
4097 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
4100 (c) A new callout option, "use_sender" has been added for people who want
4101 the modified behaviour.
4107 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
4110 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
4111 indeed breaks things for older releases.
4113 3. Added additional logging to the case where there is a problem reading data
4114 from a filter that is running in a subprocess using a pipe, in order to
4115 try to track down a specific problem.
4117 4. Testing facility fudge: when running in the test harness and attempting
4118 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
4119 getting "No route to host". Convert this to a timeout.
4121 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
4124 6. Some OS don't have socklen_t but use size_t instead. This affects the
4125 fifth argument of getsockopt() amongst other things. This is now
4126 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
4127 can be set for individual OS. I have set it for SunOS5, OSF1, and
4128 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
4129 some earlier ones do not.
4131 7. Change 4.30/15 was not doing the test caselessly.
4133 8. The standard form for an IPv6 address literal was being rejected by address
4134 parsing in, for example, MAIL and RCPT commands. An example of this kind of
4135 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
4136 this, as well as the form without the "IPv6" on the front (but only when
4137 address literals are enabled, of course).
4139 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
4141 10. Exim crashed if a message with an empty sender address specified by -f
4142 encountered a router with an errors_to setting. This could be provoked only
4143 by a command such as
4147 where an empty string was supplied; "<>" did not hit this bug.
4149 11. Installed PCRE release 4.5.
4151 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
4152 remained set. It is now erased.
4154 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
4155 times from message ids (which are base 36 rather than the normal 62).
4157 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
4158 were being counted as actual protocol errors, and logged if the log
4159 selector +smtp_protocol_error was set. One cannot be perfect in this test,
4160 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
4161 and DATA following a set of rejected RCPTs do not count as protocol errors.
4162 In other words, Exim assumes they were pipelined, though this may not
4163 actually be the case. Of course, in all cases the client gets an
4164 appropriate error code.
4166 15. If a lookup fails in an ACL condition, a message about the failure may
4167 be available; it is used if testing the ACL cannot continue, because most
4168 such messages specify what the cause of the deferral is. However, some
4169 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
4170 that caused an old message to be retained and used if a later statement
4171 caused a defer, replacing the real cause of the deferral.
4173 16. If an IP address had so many PTR records that the DNS lookup buffer
4174 was not large enough to hold them, Exim could crash while trying to process
4175 the truncated data. It now detects and logs this case.
4177 17. Further to 4.21/58, another change has been made: if (and only if) the
4178 first line of a message (the first header line) ends with CRLF, a bare LF
4179 in a subsequent header line has a space inserted after it, so as not to
4180 terminate the header.
4182 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
4183 unnecessarily, used atoi() instead of strtol(), and didn't check the
4184 termination when getting file sizes from file names by regex.
4186 19. Completely re-implemented the support for maildirsize files, in the light
4187 of a number of problems with the previous contributed implementation
4188 (4.30/29). In particular:
4190 . If the quota is zero, the maildirsize file is maintained, but no quota is
4193 . If the maildir directory does not exist, it is created before any attempt
4194 to write a maildirsize file.
4196 . The quota value in the file is just a cache; if the quota is changed in
4197 the transport, the new value overrides.
4199 . A regular expression is available for excluding directories from the
4202 20. The autoreply transport checks the characters in options that define the
4203 message's headers; it allows continued headers, but it was checking with
4204 isspace() after an embedded newline instead of explicitly looking for a
4207 21. If all the "regular" hosts to which an address was routed had passed their
4208 expiry times, and had not reached their retry times, the address was
4209 bounced, even if fallback hosts were defined. Now Exim should go on to try
4212 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
4213 equivalent code in the SMTP transport. Some hosts send humungous responses
4214 to HELO/EHLO, more than 1024 it seems.
4216 23. Refactoring: code in filter.c used (void *) for "any old type" but this
4217 gives compiler warnings in some environments. I've now done it "properly",
4220 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
4221 (because of problems with the built-in one) was declared to return uschar *
4222 instead of char *, causing compiler failure.
4224 25. Fixed a file descriptor leak when processing alias/forward files.
4226 26. Fixed a minor format string issue in dbfn.c.
4228 27. Typo in exim.c: ("dmbnz" for "dbmnz").
4230 28. If a filter file refered to $h_xxx or $message_headers, and the headers
4231 contained RFC 2047 "words", Exim's memory could, under certain conditions,
4234 29. When a sender address is verified, it is cached, to save repeating the test
4235 when there is more than one recipient in a message. However, when the
4236 verification involves a callout, it is possible for different callout
4237 options to be set for different recipients. It is too complicated to keep
4238 track of this in the cache, so now Exim always runs a verification when a
4239 callout is required, relying on the callout cache for the optimization.
4240 The overhead is duplication of the address routing, but this should not be
4243 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
4244 to be verified with callout=postmaster, and the main callout worked but the
4245 postmaster check failed, the verification correctly failed. However, if a
4246 subsequent RCPT command asked for sender verification *without* the
4247 postmaster check, incorrect caching caused this verification also to fail,
4250 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
4251 it was not caching the DNS options (qualify_single, search_parents) that
4252 were used when the lookup failed. A subsequent lookup with different
4253 options therefore always gave the same answer, though there were cases
4254 where it should not have. (Example: a "domains = !$mx_any" option on a
4255 dnslookup router: the "domains" option is always processed without any
4256 widening, but the router might have qualify_single set.) Now Exim uses the
4257 cached value only when the same options are set.
4259 32. Added John Jetmore's "exipick" utility to the distribution.
4261 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
4262 than a timeout (e.g. a certificate is required, and is not provided), an
4263 Exim server now closes the connection immediately. Previously it waited for
4264 the client to close - but if the client is SSL, it seems that they each
4265 wait for each other, leading to a delay before one of them times out.
4267 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
4268 maintained 0.8.x compatibility because I don't think many are using it, and
4269 it is clearly obsolete.
4271 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
4274 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
4275 client certificate was expired. A simple patch fixes this, though I don't
4276 understand the full logic of why the verify callback is called multiple
4279 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
4280 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
4281 which causes problems with some clients (such as the Certicom SSL Plus
4282 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
4283 disables the coutermeasure allowing Eudora to connect."
4285 38. Exim was not checking that a write() to a log file succeeded. This could
4286 lead to Bad Things if a log got too big, in particular if it hit a file
4287 size limit. Exim now panics and dies if it cannot write to a log file, just
4288 as it does if it cannot open a log file.
4290 39. Modified OS/Makefile-Linux so that it now contains
4292 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
4294 The two -D definitions ensure that Exim is compiled with large file
4295 support, which makes it possible to handle log files that are bigger than
4298 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
4299 instance) a domain was checked against a named list that involved a lookup,
4300 causing $domain_data to be set, then another domain was checked against the
4301 same list, then the first domain was re-checked, the value of $domain_data
4302 after the final check could be wrong. In particular, if the second check
4303 failed, it could be set empty. This bug probably also applied to
4306 41. The strip_trailing_dot option was not being applied to the address given
4307 with the -f command-line option.
4309 42. The code for reading a message's header from the spool was incrementing
4310 $received_count, but never initializing it. This meant that the value was
4311 incorrect (doubled) while delivering a message in the same process in which
4312 it was received. In the most common configuration of Exim, this never
4313 happens - a fresh exec is done - but it can happen when
4314 deliver_drop_privilege is set.
4316 43. When Exim logs an SMTP synchronization error - client data sent too soon -
4317 it now includes up to 150 characters of the unexpected data in the log
4320 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
4321 and building data strings. The size of both of these buffers was 10 000
4322 bytes - far larger than anybody would *ever* want, thought I. Needless to
4323 say, somebody hit the limit. I have increased the maximum line length to
4324 20 000 and the maximum data length of concatenated lines to 100 000. I have
4325 also fixed two bugs, because there was no checking on these buffers. Tsk,
4326 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
4327 buffer is too small.
4329 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
4330 lsearch lookups. Now it does.
4332 46. When parsing a route_list item in a manualroute router, a fixed-length
4333 buffer was used for the list of hosts. I made this 1024 bytes long,
4334 thinking that nobody would ever have a list of hosts that long. Wrong.
4335 Somebody had a whole pile of complicated expansion conditions, and the
4336 string was silently truncated, leading to an expansion error. It turns out
4337 that it is easier to change to an unlimited length (owing to other changes
4338 that have happened since this code was originally written) than to build
4339 structure for giving a limitation error. The length of the item that
4340 expands into the list of hosts is now unlimited.
4342 47. The lsearch lookup could not handle data where the length of text line was
4343 more than 4095 characters. Such lines were truncated, leading to shortened
4344 data being returned. It should now handle lines of any length.
4346 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
4347 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
4350 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
4351 replacing strings such as BIN_DIRECTORY in a source file: the replacement
4352 no longer happens in comment lines. A list of replacements is now placed
4353 at the head of all of the source files, except those whose only change is
4354 to replace PERL_COMMAND in the very first #! line.
4356 50. Replaced the slow insertion sort in queue.c, for sorting the list of
4357 messages on the queue, with a bottom-up merge sort, using code contributed
4358 by Michael Haardt. This should make operations like -bp somewhat faster on
4359 large queues. It won't affect queue runners, except when queue_run_in_order
4362 51. Installed eximstats 1.31 in the distribution.
4364 52. Added support for SRV lookups to the dnslookup router.
4366 53. If an ACL referred to $message_body or $message_body_end, the value was not
4367 reset for any messages that followed in the same SMTP session.
4369 54. The store-handling optimization for building very long strings was not
4370 differentiating between the different store pools. I don't think this
4371 actually made any difference in practice, but I've tidied it.
4373 55. While running the routers to verify a sender address, $sender_address
4374 was still set to the sender address. This is wrong, because when routing to
4375 send a bounce to the sender, it would be empty. Therefore, I have changed
4376 it so that, while verifying a sender address, $sender_address is set to <>.
4377 (There is no change to what happens when verifying a recipient address.)
4379 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
4380 target A or AAAA records (if not already returned) without resetting the
4381 qualify_single or search_parents options of the DNS resolver. These are
4382 inappropriate in this case because the targets of MX and SRV records must
4383 be FQDNs. A broken DNS record could cause trouble if it happened to have a
4384 target that, when qualified, matched something in the local domain. These
4385 two options are now turned off when doing these lookups.
4387 57. It seems that at least some releases of Reiserfs (which does not have the
4388 concept of a fixed number of inodes) returns zero and not -1 for the
4389 number of available inodes. This interacted badly with check_spool_inodes,
4390 which assumed that -1 was the "no such thing" setting. What I have done is
4391 to check that the total number of inodes is greater than zero before doing
4392 the test of how many are available.
4394 58. When a "warn" ACL statement has a log_message modifier, the message is
4395 remembered, and not repeated. This is to avoid a lot of repetition when a
4396 message has many recipients that cause the same warning to be written.
4397 Howewer, Exim was preserving the list of already written lines for an
4398 entire SMTP session, which doesn't seem right. The memory is now reset if a
4399 new message is started.
4401 59. The "rewrite" debugging flag was not showing the result of rewriting in the
4402 debugging output unless log_rewrite was also set.
4404 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
4405 of (int)(handle) when we know that handle contains (void *)(-1).
4407 61. The Exim daemon panic-logs an error return when it closes the incoming
4408 connection. However "connection reset by peer" seems to be common, and
4409 isn't really an error worthy of noting specially, so that particular error
4412 62. When Exim is trying to find all the local interfaces, it used to panic and
4413 die if the ioctl to get the interface flags failed. However, it seems that
4414 on at least one OS (Solaris 9) it is possible to have an interface that is
4415 included in the list of interfaces, but for which you get a failure error
4416 for this call. This happens when the interface is not "plumbed" into a
4417 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
4418 failure of the "get flags" call assumes that the interface is down.
4420 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
4421 makes life easier for people who are doing arithmetic on fields extracted
4422 from dates, where you often get leading zeros that should not be
4423 interpreted as octal.
4425 64. Added qualify_domain to the redirect router, to override the global
4428 65. If a pathologically long header line contained very many addresses (the
4429 report of this problem mentioned 10 000) and each of them was rewritten,
4430 Exim could use up a very large amount of memory. (It kept on making new
4431 copies of the header line as it rewrote, and never released the old ones.)
4432 At the expense of a bit more processing, the header rewriting function has
4433 been changed so that it no longer eats memory in this way.
4435 66. The generation of the Received: header has been moved from the time that a
4436 message starts to be received, to the time that it finishes. The timestamp
4437 in the Received: header should now be very close to that of the <= log
4438 line. There are two side-effects of this change:
4440 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
4441 logged header lines no longer include the local Received: line, because
4442 it has not yet been created. The same applies to a copy of the message
4443 that is returned to a non-SMTP sender when a message is rejected.
4445 (b) When a filter file is tested using -bf, no additional Received: header
4446 is added to the test message. After some thought, I decided that this
4449 This change does not affect the value of $received_for. It is still set
4450 after address rewriting, but before local_scan() is called.
4452 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
4454 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
4455 gave an unhelpful panic error message, and a defer error. I have managed to
4456 change this behaviour so that it now rejects any supplied certificate,
4457 which seems right, as the list of acceptable certificates is empty.
4459 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
4460 gave an unhelpful defer error. I have not managed to make this reject any
4461 supplied certificates, but the error message it gives is "no certificate
4462 supplied", which is not helpful.
4464 70. exigrep's output now also includes lines that are not associated with any
4465 message, but which match the given pattern. Implemented by a patch from
4466 Martin Sluka, which also tidied up the Perl a bit.
4468 71. Recipient callout verification, like sender verification, was using <> in
4469 the MAIL FROM command. This isn't really the right thing, since the actual
4470 sender may affect whether the remote host accepts the recipient or not. I
4471 have changed it to use the actual sender in the callout; this means that
4472 the cache record is now keyed on a recipient/sender pair, not just the
4473 recipient address. There doesn't seem to be a real danger of callout loops,
4474 since a callout by the remote host to check the sender would use <>.
4475 [SEE ABOVE: changed after hitting problems.]
4477 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
4478 temporary errors. However, in the case of such a code being given after
4479 the end of a data transmission (i.e. after ".") Exim was failing to write
4480 a retry record for the message. (Yes, there was some broken host that was
4481 actually sending 8xx at this point.)
4483 73. An unknown lookup type in a host list could cause Exim to panic-die when
4484 the list was checked. (An example that provoked this was putting <; in the
4485 middle of a list instead of at the start.) If this happened during a DATA
4486 ACL check, a -D file could be left lying around. This kind of configuration
4487 error no longer causes Exim to die; instead it causes a defer errror. The
4488 incident is still logged to the main and panic logs.
4490 74. Buglet left over from Exim 3 conversion. The message "too many messages
4491 in one connection" was written to the rejectlog but not the mainlog, except
4492 when address rewriting (yes!) was being logged.
4494 75. Added write_rejectlog option.
4496 76. When a system filter was run not as root (that is, when system_filter_user
4497 was set), the values of the $n variables were not being returned to the
4498 main process; thus, they were not subsequently available in the $sn
4501 77. Added +return_path_on_delivery log selector.
4503 78. A connection timeout was being treated differently from recipients deferred
4504 when testing hosts_max_try with a message that was older than the host's
4505 retry timeout. (The host should not be counted, thus allowing all hosts to
4506 be tried at least once before bouncing.) This may have been the cause of an
4507 occasionally reported bug whereby a message would remain on the queue
4508 longer than the retry timeout, but would be bounced if a delivery was
4509 forced. I say "may" because I never totally pinned down the problem;
4510 setting up timeout/retry tests is difficult. See also the next item.
4512 79. The ultimate address timeout was not being applied to errors that involved
4513 a combination of host plus message (for example, a timeout on a MAIL
4514 command). When an address resolved to a number of possible hosts, and they
4515 were not all tried for each delivery (e.g. because of hosts_max_try), a
4516 message could remain on the queue longer than the retry timeout.
4518 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
4521 81. Fixed an obscure SMTP outgoing bug which required at least the following
4522 conditions: (a) there was another message waiting for the same server;
4523 (b) the server returned 5xx to all RCPT commands in the first message so
4524 that the message was not completed; (c) the server dropped the connection
4525 or gave a negative response to the RSET that Exim sends to abort the
4526 transaction. The observed case was a dropped connection after DATA that had
4527 been sent in pipelining mode. That is, the server had advertised PIPELINING
4528 but was not implementing it correctly. The effect of the bug was incorrect
4529 behaviour, such as trying another host, and this could lead to a crash.
4535 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
4536 and daemon.c were passed as pointers to ints; they should have been
4537 pointers to socklen_t variables (which are typically unsigned ints).
4539 2. Some signed/unsigned type warnings in the os.c file for Linux have been
4542 3. Fixed a really odd bug that affected only the testing scheme; patching a
4543 certain fixed string in the binary changed the value of another string that
4544 happened to be identical to the end of the original first string.
4546 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
4547 name", it returns that address as the IP address. On some operating
4548 systems (e.g. Solaris), it also passes back the IP address string as the
4549 "host name". However, on others (e.g. Linux), it passes back an empty
4550 string. Exim wasn't checking for this, and was changing the host name to an
4551 empty string, assuming it had been canonicized.
4553 5. Although rare, it is permitted to have more than one PTR record for a given
4554 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
4555 all the names associated with an address, because they do in Solaris.
4556 However, it seems that they do not in Linux for data that comes from the
4557 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
4558 I found this out when I moved to a new Linux workstation and tried to run
4559 the Exim test suite.
4561 To get round this problem I have changed the code so that it now does its
4562 own call to the DNS to look up PTR records when searching for a host name.
4563 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
4564 addresses that are only in /etc/hosts are still found.
4566 This behaviour is, however, controlled by an option called host_lookup_
4567 order, which defaults to "bydns:byaddr". If people want to use the other
4568 order, or indeed, just use one or the other means of lookup, they can
4569 specify it in this variable.
4571 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
4572 some operating systems, this comes back from gethostbyaddr() as an empty
4573 string, and this is what Exim used to test for. However, it seems that in
4574 other systems, "." is yielded. Exim now tests for this case too.
4576 7. The values of check_spool_space and check_log_space are now held internally
4577 as a number of kilobytes instead of an absolute number of bytes. If a
4578 numbers is specified without 'K' or 'M', it is rounded up to the nearest
4579 kilobyte. This means that much larger values can be stored.
4581 8. Exim monitor: an attempt to get the action menu when not actually pointing
4582 at a message produces an empty menu entitled "No message selected". This
4583 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
4584 no entries in it ("Shell widget menu has zero width and/or height"). So I
4585 have added a single, blank menu entry in this case.
4587 9. Added ${quote_local_part.
4589 10. MIME decoding is now applied to the contents of Subject: header lines when
4592 11. Now that a reference to $sender_host_address automatically causes a reverse
4593 lookup to occur if necessary (4.13/18), there is no need to arrange for a
4594 host lookup before query-style lookups in lists that might use this
4595 variable. This has therefore been abolished, and the "net-" prefix is no
4596 longer necessary for query-style lookups.
4598 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
4599 have been a typo for LFLAGS, so it has been changed.
4601 13. The install script calls Exim with "-C /dev/null" in order to find the
4602 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
4603 to be output. Howeve, since Exim outputs its version number before the
4604 error, it didn't break the script. It just looked ugly. I fixed this by
4605 always allowing "-C /dev/null" if the caller is root.
4607 14. Ignore overlarge ACL variable number when reading spool file - insurance
4608 against a later release with more variables having written the file.
4610 15. The standard form for an IPv6 address literal was being rejected by EHLO.
4611 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
4612 this, as well as the form without the "IPv6" on the front.
4614 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
4615 OS/Makefile-Darwin file.
4617 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
4618 only to LDAP libraries that do not have LDAP_OPT_DEREF.
4620 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
4621 variable. However, some OS use ints for inodes. I've added cast to long int
4622 to get rid of the compiler warning.
4624 19. I had forgotten to lock out "/../" in configuration file names when
4625 ALT_CONFIG_PREFIX was set.
4627 20. Routers used for verification do not need to specify transports. However,
4628 if such a router generated a host list, and callout was configured, Exim
4629 crashed, because it could not find a port number from the (non-existent)
4630 transport. It now assumes port 25 in this circumstance.
4632 21. Added the -t option to exigrep.
4634 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
4635 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
4636 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
4637 them is set, LOOKUP_LSEARCH is forced.
4639 23. "exim -bV" now outputs a list of lookups that are included in the binary.
4641 24. Added sender and host information to the "rejected by local_scan()" log
4642 line; previously there was no indication of these.
4644 25. Added .include_if_exists.
4646 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
4647 Linux. It turns out that not all file systems support this. Apparently some
4648 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
4649 do it.) To cope with this, the error EINVAL, which means that sync-ing is
4650 not supported on the file descriptor, is now ignored when Exim is trying to
4651 sync a directory. This applies only to Linux.
4653 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
4655 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
4656 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
4659 29. Added support for maildirsize files from supplied patch (modified a bit).
4661 30. The use of :fail: followed by an empty string could lead Exim to respond to
4662 sender verification failures with (e.g.):
4664 550 Verification failed for <xxx>
4665 550 Sender verify failed
4667 where the first response line was missing the '-' that indicates it is not
4668 the final line of the response.
4670 31. The loop for finding the name of the user that called Exim had a hardwired
4671 limit of 10; it now uses the value of finduser_retries, which is used for
4672 all other user lookups.
4674 32. Added $received_count variable, available in data and not_smtp ACLs, and at
4677 33. Exim was neglecting to zero errno before one call of strtol() when
4678 expanding a string and expecting an integer value. On some systems this
4679 resulted in spurious "integer overflow" errors. Also, it was casting the
4680 result into an int without checking.
4682 34. Testing for a connection timeout using "timeout_connect" in the retry rules
4683 did not work. The code looks as if it has *never* worked, though it appears
4684 to have been documented since at least releast 1.62. I have made it work.
4686 35. The "timeout_DNS" error in retry rules, also documented since at least
4687 1.62, also never worked. As it isn't clear exactly what this means, and
4688 clearly it isn't a major issue, I have abolished the feature by treating it
4689 as "timeout", and writing a warning to the main and panic logs.
4691 36. The display of retry rules for -brt wasn't always showing the error code
4694 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
4695 timeout_connect_A, timeout_connect_MX.
4697 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
4698 to the empty sender.
4700 39. The daemon was not analysing the content of -oX till after it had closed
4701 stderr and disconnected from the controlling terminal. This meant that any
4702 syntax errors were only noted on the panic log, and the return code from
4703 the command was 0. By re-arranging the code a little, I've made the
4704 decoding happen first, so such errors now appear on stderr, and the return
4705 code is 1. However, the actual setting up of the sockets still happens in
4706 the disconnected process, so errors there are still only recorded on the
4709 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
4710 connections (as happens on some IP stacks) was logged at start up time as
4711 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
4712 it from "IPv6 and IPv4", which means that two separate sockets are being
4715 41. The debug output for gethostbyname2() or getipnodebyname() failures now
4716 says whether AF_INET or AF_INET6 was passed as an argument.
4718 42. Exiwhat output was messed up when time zones were included in log
4721 43. Exiwhat now gives more information about the daemon's listening ports,
4722 and whether -tls-on-connect was used.
4724 44. The "port" option of the smtp transport is now expanded.
4726 45. A "message" modifier in a "warn" statement in a non-message ACL was being
4727 silently ignored. Now an error message is written to the main and panic
4730 46. There's a new ACL modifier called "logwrite" which writes to a log file
4731 as soon as it is encountered.
4733 47. Added $local_user_uid and $local_user_gid at routing time.
4735 48. Exim crashed when trying to verify a sender address that was being
4738 49. Exim was recognizing only a space character after ".include". It now also
4739 recognizes a tab character.
4741 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
4742 extracting the relevant information from the specification. The man page no
4743 longer contains scrambled data for the -d option, and I've added a section
4744 at the front about calling Exim under different names.
4746 51. Added "extra_headers" argument to the "mail" command in filter files.
4748 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
4751 53. Installed eximstats 1.29.
4753 54. Added transport_filter_timeout as a generic transport option.
4755 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
4756 Cc: header lines. This was required by RFC 822, but it not required by RFC
4759 56. Exim used to add From:, Date:, and Message-Id: header lines to any
4760 incoming messages that did not have them. Now it does so only if the
4761 message originates locally, that is, if there is no associated remote host
4762 address. When Resent- header lines are present, this applies to the Resent-
4763 lines rather than the non-Resent- lines.
4765 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
4766 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
4768 58. Messages for configuration errors now include the name of the main
4769 configuration file - useful now that there may be more than one file in a
4770 list (.included file names were always shown).
4772 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
4773 for those rare installations that do not start the daemon as root or run it
4774 setuid root. I've cut out the call to initgroups() if the daemon is not
4777 60. The Exim user and group can now be bound into the binary as text strings
4778 that are looked up at the start of Exim's processing.
4780 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
4782 62. Added $mailstore_basename variable.
4784 63. Installed patch to sieve.c from Michael Haardt.
4786 64. When Exim failed to open the panic log after failing to open the main log,
4787 the original message it was trying to log was written to stderr and debug
4788 output, but if they were not available (the usual case in production), it
4789 was lost. Now it is written to syslog before the two lines that record the
4790 failures to open the logs.
4792 65. Users' Exim filters run in subprocesses under the user's uid. It is
4793 possible for a "deliver" command or an alias in a "personal" command to
4794 provoke an address rewrite. If logging of address rewriting is configured,
4795 this fails because the process is not running as root or exim. There may be
4796 a better way of dealing with this, but for the moment (because 4.30 needs
4797 to be released), I have disabled address rewrite logging when running a
4798 filter in a non-root, non-exim process.
4804 1. The buildconfig auxiliary program wasn't quoting the value set for
4805 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
4806 not defined. This bug was masked in 4.22 by the effect that was fixed in
4809 2. Some messages that were rejected after a message id was allocated were
4810 shown as "incomplete" by exigrep. It no longer does this for messages that
4811 are rejected by local_scan() or the DATA or non-SMTP ACLs.
4813 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
4814 have allow_domain_literals set, the ID did not get logged in the <= line.
4815 Domain literals are now always recognized in Message-ID: header lines.
4817 4. The first argument for a ${extract expansion item is the key name or field
4818 number. Leading and trailing spaces in this item were not being ignored,
4819 causing some misleading effects.
4821 5. When deliver_drop_privilege was set, single queue runner processes started
4822 manually (i.e. by the command "exim -q") or by the daemon (which uses the
4823 same command in the process it spins off) were not dropping privilege.
4825 6. When the daemon running as "exim" started a queue runner, it always
4826 re-executed Exim in the spun-off process. This is a waste of effort when
4827 deliver_drop_privilege is set. The new process now just calls the
4828 queue-runner function directly.
4834 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
4837 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
4838 ignored. Though the use of -oP was forcing the writing of a pid file, it
4839 was always written to the default place.
4841 3. If the message "no IP address found for host xxxx" is generated during
4842 incoming verification, it is now followed by identification of the incoming
4843 connection (so you can more easily find what provoked it).
4845 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
4847 5. Added some features to "harden" Exim a bit more against certain attacks:
4849 (a) There is now a build-time option called FIXED_NEVER_USERS that can
4850 be put in Local/Makefile. This is like the never_users runtime option,
4851 but it cannot be overridden. The default setting is "root".
4853 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
4854 prefix string with which any file named in a -C command line option
4857 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
4858 is retained for -C and -D only if the caller of Exim is root. Without
4859 it, the exim user may also use -C and -D and retain privilege.
4861 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
4862 command line option is disabled.
4864 6. Macro names set by the -D option must start with an upper case letter, just
4865 like macro names defined in the configuration file.
4867 7. Added "dereference=" facility to LDAP.
4869 8. Two instances of the typo "uknown" in the source files are fixed.
4871 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
4872 the Configure-Makefile script screwed up while processing it.
4874 10. Incorporated PCRE 4.4.
4876 11. The SMTP synchronization check was not operating right at the start of an
4877 SMTP session. For example, it could not catch a HELO sent before the client
4878 waited for the greeting. There is now a check for outstanding input at the
4879 point when the greeting is written. Because of the duplex, asynchronous
4880 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
4881 way, but not yet received, when the check is performed.
4883 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
4884 on TCP/IP sockets, because this apparently causes some broken clients to
4887 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
4888 unchanged) from the Cygwin maintainer.
4890 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
4891 was supported instead of testing for mbx. Effectively a typo.
4893 15. The spa authenticator server code was not checking that the input it
4894 received was valid base64.
4896 16. The debug output line for the "set" modifier in ACLs was not showing the
4897 name of the variable that was being set.
4899 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
4901 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
4902 lookup on its own. Something else was needed to trigger the lookup. For
4903 example, a match in host_lookup or the need for a host name in a host list.
4904 Now, if $sender_host_name is referenced and the host name has not yet been
4905 looked up, a lookup is performed. If the lookup fails, the variable remains
4906 empty, and $host_lookup_failed is set to "1".
4908 19. Added "eqi" as a case-independent comparison operator.
4910 20. The saslauthd authentication condition could segfault if neither service
4911 nor realm was specified.
4913 21. If an overflowing value such as "2048M" was set for message_size_limit, the
4914 error message that was logged was misleading, and incoming SMTP
4915 connections were dropped. The message is now more accurate, and temporary
4916 errors are given to SMTP connections.
4918 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
4919 (except RSET) with a 421 error, until QUIT is received. However, it was
4920 failing to send a response to QUIT.
4922 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
4923 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
4924 ACL is now run after the helo_try_verify_hosts code.
4926 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
4927 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
4928 case-independent, so other case variants are also recognized). Apparently
4929 some systems use these upper case variants.
4931 25. If more than two messages were waiting for the same host, and a transport
4932 filter was specified for the transport, Exim sent two messages over the
4933 same TCP/IP connection, and then failed with "socket operation on non-
4934 socket" when it tried to send the third.
4936 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
4938 27. The extern definition of crypt16() in expand.c was not being excluded when
4939 the OS had its own crypt16() function.
4941 28. Added bounce_return_body as a new option, and bounce_return_size_limit
4942 as a preferred synonym for return_size_limit, both as an option and as an
4945 29. Added LIBS=-liconv to OS/Makefile-OSF1.
4947 30. Changed the default configuration ACL to relax the local part checking rule
4948 for addresses that are not in any local domains. For these addresses,
4949 slashes and pipe symbols are allowed within local parts, but the sequence
4950 /../ is explicitly forbidden.
4952 31. SPA server authentication was not clearing the challenge buffer before
4955 32. log_message in a "warn" ACL statement was writing to the reject log as
4956 well as to the main log, which contradicts the documentation and doesn't
4957 seem right (because no rejection is happening). So I have stopped it.
4959 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
4960 However, I am unable to do any testing of this.
4962 34. Fixed an infelicity in the appendfile transport. When checking directories
4963 for a mailbox, to see if any needed to be created, it was accidentally
4964 using path names with one or more superfluous leading slashes; tracing
4965 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
4967 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
4968 contents are added to the log line that is written for every discarded
4969 recipient. (Previously a log_message setting was ignored.)
4971 36. The ${quote: operator now quotes the string if it is empty.
4973 37. The install script runs exim in order to find its version number. If for
4974 some reason other than non-existence or emptiness, which it checks, it
4975 could not run './exim', it was installing it with an empty version number,
4976 i.e. as "exim-". This error state is now caught, and the installation is
4979 38. An argument was missing from the function that creates an error message
4980 when Exim fails to connect to the socket for saslauthd authentication.
4981 This could cause Exim to crash, or give a corrupted message.
4983 39. Added isip, isip4, and isip6 to ${if conditions.
4985 40. The ACL variables $acl_xx are now saved with the message, and can be
4986 accessed later in routers, transports, and filters.
4988 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
4989 strings in the file are not string-expanded.
4991 42. If a MAIL command specified a SIZE value that was too large to fit into an
4992 int variable, the check against message_size_limit failed. Such values are
4993 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
4994 day this will have to be increased, but I don't think I want to be around
4995 when emails are that large.
5002 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
5003 iconv() is not standard in FreeBSD.
5005 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
5006 IPv6 enabled. The observed symptom was a segmentation fault on return from
5007 the function os_common_find_running_interfaces() in src/os.c.
5009 3. In the check_special_case() function in daemon.c I had used "errno" as an
5010 argument name, which causes warnings on some systems. This was basically a
5011 typo, since it was named "eno" in the comments!
5013 4. The code that waits for the clock to tick (at a resolution of some fraction
5014 of a second) so as to ensure message-id uniqueness was always waiting for
5015 at least one whole tick, when it could have waited for less. [This is
5016 almost certainly not relevant at current processor speeds, where it is
5017 unlikely to ever wait at all. But we try to future-proof.]
5019 5. The function that sleeps for a time interval that includes fractions of a
5020 second contained a race. It did not block SIGALRM between setting the
5021 timer, and suspending (a couple of lines later). If the interval was short
5022 and the sigsuspend() was delayed until after it had expired, the suspension
5023 never ended. On busy systems this could lead to processes getting stuck for
5026 6. Some uncommon configurations may cause a lookup to happen in a queue runner
5027 process, before it forks any delivery processes. The open lookup caching
5028 mechanism meant that the open file or database connection was passed into
5029 the delivery process. The problem was that delivery processes always tidy
5030 up cached lookup data. This could cause a problem for the next delivery
5031 process started by the queue runner, because the external queue runner
5032 process does not know about the closure. So the next delivery process
5033 still has data in the lookup cache. In the case of a file lookup, there was
5034 no problem because closing a file descriptor in a subprocess doesn't affect
5035 the parent. However, if the lookup was caching a connection to a database,
5036 the connection was closed, and the second delivery process was likely to
5037 see errors such as "PGSQL: query failed: server closed the connection
5038 unexpectedly". The problem has been fixed by closing all cached lookups
5039 in a queue runner before running a delivery process.
5041 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
5042 seem to have the "const" qualifier which it has on other OS. I've
5045 8. Change 4.21/2 was too strict. It is only if there are two authenticators
5046 *of the same type* (client or server) with the same public name that an
5047 error should be diagnosed.
5049 9. When Exim looked up a host name for an IP address, but failed to find the
5050 original IP address when looking up the host name (a safety check), it
5051 output the message "<ip address> does not match any IP for NULL", which was
5052 confusing, to say the least. The bug was that the host name should have
5053 appeared instead of "NULL".
5055 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
5056 user that is built into the binary, and the -C or -D options is used, root
5057 privilege is dropped before the configuration file is read. In addition,
5058 logging is switched to stderr instead of the normal log files. If the
5059 configuration then re-defines the Exim user, the unprivileged environment
5060 is probably not what is expected, so Exim logs a panic warning message (but
5063 However, if deliver_drop_privilege is set, the unprivileged state may well
5064 be exactly what is intended, so the warning has been cut out in that case,
5065 and Exim is allowed to try to write to its normal log files.
5071 1. smtp_return_error_details was not giving details for temporary sender
5072 or receiver verification errors.
5074 2. Diagnose a configuration error if two authenticators have the same public
5077 3. Exim used not to create the message log file for a message until the first
5078 delivery attempt. This could be confusing when incoming messages were held
5079 for policy or load reasons. The message log file is now created at the time
5080 the message is received, and an initial "Received" line is written to it.
5082 4. The automatically generated man page for command line options had a minor
5083 bug that caused no ill effects; however, a more serious problem was that
5084 the procedure for building the man page automatically didn't always
5085 operate. Consequently, release 4.20 contains an out-of-date version. This
5086 shouldn't happen again.
5088 5. When building Exim with embedded Perl support, the script that builds the
5089 Makefile was calling 'perl' to find its compile-time parameters, ignoring
5090 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
5092 6. The freeze_tell option was not being used for messages that were frozen on
5093 arrival, either by an ACL or by local_scan().
5095 7. Added the smtp_incomplete_transaction log selector.
5097 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
5098 was accepting AUTH without a new EHLO.
5100 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
5101 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
5104 10. Logging of TCP/IP connections (when configured) now happens in the main
5105 daemon process instead of the child process, so that the TCP/IP connection
5106 count is more accurate (but it can never be perfect).
5108 11. The use of "drop" in a nested ACL was not being handled correctly in the
5109 outer ACL. Now, if condition failure induced by the nested "drop" causes
5110 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
5111 or "require"), the connection is dropped.
5113 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
5114 that yield "discard" can now be used with an "accept" or a "discard" verb,
5115 but an error is generated for any others (because I can't see a useful way
5116 to define what should happen).
5118 13. When an ACL is read dynamically from a file (or anywhere else), the lines
5119 are now processed in the same way as lines in the Exim configuration file.
5120 In particular, continuation lines are supported.
5122 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
5124 15. Added -ti meaning -t -i.
5126 16. Check for letters, digits, hyphens, and dots in the names of dnslist
5127 domains, and warn by logging if others are found.
5129 17. At least on BSD, alignment is not guarenteed for the array of ifreq's
5130 returned from GIFCONF when Exim is trying to find the list of interfaces on
5131 a host. The code in os.c has been modified to copy each ifreq to an aligned
5132 structure in all cases.
5134 Also, in some cases, the returned ifreq's were being copied to a 'struct
5135 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
5136 means the last couple of bytes of an IPv6 address could be chopped if the
5137 ifreq contained only a normal sockaddr (14 bytes storage).
5139 18. Named domain lists were not supported in the hosts_treat_as_local option.
5140 An entry such as +xxxx was not recognized, and was treated as a literal
5143 19. Ensure that header lines added by a DATA ACL are included in the reject log
5144 if the ACL subsequently rejects the message.
5146 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
5147 MD5 (which is deprecated).
5149 21. When testing a filter file using -bf, Exim was writing a message when it
5150 took the sender from a "From " line in the message, but it was not doing so
5151 when it took $return_path from a Return-Path: header line. It now does.
5153 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
5154 with a valid header line field name (a series of printing characters
5155 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
5157 23. Changed "disc" in the source to "disk" to conform to the documentation and
5158 the book and for uniformity.
5160 24. Ignore Sendmail's -Ooption=value command line item.
5162 25. When execve() failed while trying to run a command in a pipe transport,
5163 Exim was returning EX_UNAVAILBLE (69) from the subprocess. However, this
5164 could be confused with a return value of 69 from the command itself. This
5165 has been changed to 127, the value the shell returns if it is asked to run
5166 a non-existent command. The wording for the related log line suggests a
5167 non-existent command as the problem.
5169 26. If received_header_text expands to an empty string, do not add a Received:
5170 header line to the message. (Well, it adds a token one on the spool, but
5171 marks it "old" so that it doesn't get used or transmitted.)
5173 27. Installed eximstats 1.28 (addition of -nt option).
5175 28. There was no check for failure on the call to getsockname() in the daemon
5176 code. This can fail if there is a shortage of resources on the system, with
5177 ENOMEM, for example. A temporary error is now given on failure.
5179 29. Contrary to the C standard, it seems that in some environments, the
5180 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
5181 program. Exim now does this explicitly; it affects the formatting of
5182 timestamps using strftime().
5184 30. If exiqsumm was given junk data, it threw up some uninitialized variable
5185 complaints. I've now initialized all the variables, to avoid this.
5187 32. Header lines added by a system filter were not being "seen" during
5188 transport-time rewrites.
5190 33. The info_callback() function passed to OpenSSL is set up with type void
5191 (*)(SSL *, int, int), as described somewhere. However, when calling the
5192 function (actually a macro) that sets it up, the type void(*)() is
5193 expected. I've put in a cast to prevent warnings from picky compilers.
5195 34. If a DNS black list lookup found a CNAME record, but there were no A
5196 records associated with the domain it pointed at, Exim crashed.
5198 35. If a DNS black list lookup returned more than one A record, Exim ignored
5199 all but the first. It now scans all returned addresses if a particular IP
5200 value is being sought. In this situation, the contents of the
5201 $dnslist_value variable are a list of all the addresses, separated by a
5204 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
5205 to accept a host name and all its aliases if the forward lookup for any of
5206 them yielded the IP address of the incoming connection. Now it accepts only
5207 those names whose forward lookup yields the correct IP address. Any other
5208 names are discarded. This closes a loophole whereby a rogue DNS
5209 administrator could create reverse DNS records to break through a
5210 wildcarded host restriction in an ACL.
5212 37. If a user filter or a system filter that ran in a subprocess used any of
5213 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
5214 the wrong values were passed to the pipe command ($thisaddress had the
5215 value of $0, $0 had the value of $1, etc). This bug was introduced by
5216 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
5218 38. Improved the line breaking for long SMTP error messages from ACLs.
5219 Previously, if there was no break point between 40 and 75 characters, Exim
5220 left the rest of the message alone. Two changes have been made: (a) I've
5221 reduced the minimum length to 35 characters; (b) if it can't find a break
5222 point between 35 and 75 characters, it looks ahead and uses the first one
5223 that it finds. This may give the occasional overlong line, but at least the
5224 remaining text gets split now.
5226 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
5227 file descriptors might be low, and that setting 1000 would always raise it.
5228 It turns out that in some environments, the limit is already over 1000 and
5229 that lowering it causes trouble. So now Exim takes care not to decrease it.
5231 40. When delivering a message, the value of $return_path is set to $sender_
5232 address at the start of routing (routers may change the value). By an
5233 oversight, this default was not being set up when an address was tested by
5234 -bt or -bv, which affected the outcome if any router or filter referred to
5237 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
5238 log only when "message" or "log_message" are set. However, if one of the
5239 conditions was an address verification, or a call to a nested ACL, the
5240 messages generated by the underlying test were being passed through. This
5241 no longer happens. The underlying message is available in $acl_verify_
5242 message for both "message" and "log_message" expansions, so it can be
5243 passed through if needed.
5245 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
5246 new expansion $bh_ to give the encoded byte string without charset
5247 translation. Translation happens only if iconv() is available; HAVE_ICONV
5248 indicates this at build time. HEADERS_CHARSET gives the charset to
5249 translate to; headers_charset can change it in the configuration, and
5250 "headers charset" can change it in an individual filter file.
5252 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
5253 that creates RFC 2047 encoded "words" labels them as that charset instead
5254 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
5255 expansion operator; (ii) when Exim creates a From: line for a local
5256 message; (iii) when a header line is rewritten to include a "phrase" part.
5258 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
5259 buggy, causing it to skip the first lines of messages whose message ID
5260 ended in 'D'. This would not have bitten before Exim release 4.14, because
5261 message IDs were unlikely to end in 'D' before then. The effect was to have
5262 incorrect size information for certain domains.
5264 45. #include "config.h" was missing at the start of the crypt16.c module. This
5265 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
5268 46. If there was a timeout during a "random" callout check, Exim treated it as
5269 a failure of the random address, and carried on sending RSET and the real
5270 address. If the delay was just some slowness somewhere, the response to the
5271 original RCPT would be taken as a response to RSET and so on, causing
5272 mayhem of various kinds.
5274 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
5275 when I implemented it. It didn't allow for the fact that some option values
5276 may legitimatetly be negative (e.g. size_addition), and it didn't even do
5277 the right test for positive values.
5279 48. Domain names in DNS records are case-independent. Exim always looks them up
5280 in lower case. Some resolvers return domain names in exactly the case they
5281 appear in the zone file, that is, they may contain uppercase letters. Not
5282 all resolvers do this - some return always lower case. Exim was treating a
5283 change of case by a resolver as a change of domain, similar to a widening
5284 of a domain abbreviation. This triggered its re-routing code and so it was
5285 trying to route what was effectively the same domain again. This normally
5286 caused routing to fail (because the router wouldn't handle the domain
5287 twice). Now Exim checks for this case specially, and just changes the
5288 casing of the domain that it ultimately uses when it transmits the message
5291 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
5294 50. If a filter generated a file delivery with a non-absolute name (possible if
5295 no home directory exists for the router), the forbid_file option was not
5298 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
5299 the existing equality matching.
5301 52. Exim was using ints instead of ino_t variables in some places where it was
5302 dealing with inode numbers.
5304 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
5305 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
5306 called TMPDIR, and if it finds it is different, it changes its value.
5308 54. The smtp_printf() function is now made available to local_scan() so
5309 additional output lines can be written before returning. There is also an
5310 smtp_fflush() function to enable the detection of a dropped connection.
5311 The variables smtp_input and smtp_batched_input are exported to
5314 55. Changed the default runtime configuration: the message "Unknown user"
5315 has been removed from the ACL, and instead placed on the localuser router,
5316 using the cannot_route_message feature. This means that any verification
5317 failures that generate their own messages won't get overridden. Similarly,
5318 the "Unrouteable address" message that was in the ACL for unverifiable
5319 relay addresses has also been removed.
5321 56. Added hosts_avoid_esmtp to the smtp transport.
5323 57. The exicyclog script was not checking for the esoteric option
5324 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
5325 will work only if exicyclog is run under the appropriate euid.
5327 58. Following a discussion on the list, the rules by which Exim recognises line
5328 endings on incoming messages have been changed. The -dropcr and drop_cr
5329 options are now no-ops, retained only for backwards compatibility. The
5330 following line terminators are recognized: LF CRLF CR. However, special
5331 processing applies to CR:
5333 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
5334 nor a local message in the state where . is a terminator.
5336 (ii) If a bare CR is encountered in a header line, an extra space is added
5337 after the line terminator so as not to end the header. The reasoning
5338 behind this is that bare CRs in header lines are most likely either
5339 to be mistakes, or people trying to play silly games.
5341 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
5342 was being incorrectly given as 18 bytes larger than it should have been.
5343 This is a VOB (very old bug).
5345 60. This may never have affected anything current, but just in case it has:
5346 When the local host is found other than at the start of a list of hosts,
5347 the local host, those with the same MX, and any that follow, are discarded.
5348 When the list in question was part of a longer list of hosts, the following
5349 hosts (not currently being processed) were also being discarded. This no
5350 longer happens. I'm not sure if this situation could ever has previously
5353 61. Added the "/MX" feature to lists of hosts in the manualroute and query
5356 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
5357 header. This is something that is recommended in a new Internet Draft, and
5358 is something that is documented as being done by Sendmail. There are two
5359 possible values. For messages generated by the autoreply transport, Exim
5362 Auto-Submitted: auto-replied
5364 whereas for all other generated messages (e.g. bounces) it adds
5366 Auto-Submitted: auto-generated
5368 63. The "personal" condition in filters now includes a test for the
5369 Auto-Submitted: header. If it contains the string "auto-" the message it
5370 not considered personal.
5372 64. Added rcpt_include_affixes as a generic transport option.
5374 65. Added queue_only_override (default true).
5376 66. Added the syslog_duplication option.
5378 67. If what should have been the first header line of a message consisted of
5379 a space followed by a colon, Exim was mis-interpreting it as a header line.
5380 It isn't of course - it is syntactically invalid and should therefore be
5381 treated as the start of the message body. The misbehaviour could have
5382 caused a number of strange effects, including loss of data in subsequent
5383 header lines, and spool format errors.
5385 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
5386 client host had authenticated. This control can now be exercised by an ACL
5387 for more flexibility.
5389 69. By default, callouts do not happen when testing with -bh. There is now a
5390 variant, -bhc, which does actually run the callout code, including
5391 consulting and updating the callout cache.
5393 70. Added support for saslauthd authentication, courtesy of Alexander
5396 71. If statvfs() failed on the spool or log directories while checking their
5397 size for availability, Exim confusingly gave the error "space shortage".
5398 Furthermore, in debugging mode it crashed with a floating point exception.
5399 These checks are done if check_{spool,log}_{space,inodes} are set, and when
5400 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
5401 serious problem, Exim now writes to the main and panic logs when this
5402 happens, with details of the failure. It then refuses to accept the
5403 incoming message, giving the message "spool directory problem" or "log
5404 directory problem" with a 421 code for SMTP messages.
5406 72. When Exim is about to re-exec itself, it ensures that the file descriptors
5407 0, 1, and 2 exist, because some OS complain for execs without them (see
5408 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
5409 descriptors. However, the code omitted to check that the open succeeded,
5410 causing mysterious errors if for some reason the permissions on /dev/null
5411 got screwed. Now Exim writes a message to the main and panic logs, and
5412 bombs out if it can't open /dev/null.
5414 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
5415 interact so that it is all more flexible. It is supposed to remain
5416 backwards compatible. Also added extra_local_interfaces.
5418 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
5419 to bomb out with an assertion failure - to the client this appears as a
5420 connection drop. This problem occurs in the part of the code that was taken
5421 from the Samba project. Fortunately, the assertion is in a very simple
5422 function, so I have fixed this by reproducing the function inline in the
5423 one place where it is called, and arranging for authentication to fail
5424 instead of killing the process with assert().
5426 75. The SPA client code was not working when the server requested OEM rather
5427 than Unicode encoding.
5429 76. Added code to make require_files with a specific uid setting more usable in
5430 the case where statting the file as root fails - usually a non-root-mounted
5431 NFS file system. When this happens and the failure is EACCES, Exim now
5432 forks a subprocess and does the per-uid checking as the relevant uid.
5434 77. Added process_log_path.
5436 78. If log_file_path was not explicitly set, a setting of check_log_space or
5437 check_log_inodes was ignored.
5439 79. If a space check for the spool or log partitions fails, the incident is now
5440 logged. Of course, in the latter case the data may get lost...
5442 80. Added the %p formatting code to string_format() so that it can be used to
5443 print addresses in debug_print(). Adjusted all the address printing in the
5444 debugging in store.c to use %p rather than %d.
5446 81. There was a concern that a line of code in smtp_in.c could overflow a
5447 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
5448 initially expressed, the concern was not well-founded, because trailing
5449 spaces are removed early. However, if the trailing spaces were followed by
5450 a NULL, they did not get removed, so the overflow was possible. Two fixes
5453 (a) I re-wrote the offending code in a cleaner fashion.
5454 (b) If an incoming SMTP command contains a NULL character, it is rejected
5457 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
5458 runs initgroups(), so that if the Exim user is in any additional groups,
5459 they will be used during message reception.
5465 The change log for 4.20 and earlier releases has been archived.
git://git.exim.org / exim.git / blob