1 Change log file for Exim from version 4.21
2 -------------------------------------------
7 PP/01 Solaris build fix for Oracle's LDAP libraries.
8 Bugzilla 1109, patch from Stephen Usher.
10 TF/01 HP/UX build fix: avoid arithmetic on a void pointer.
12 TK/01 DKIM Verification: Fix relaxed canon for empty headers w/o
15 TF/02 Fix a couple more cases where we did not log the error message
16 when unlink() failed. See also change 4.74-TF/03.
18 TF/03 Make the exiwhat support code safe for signals. Previously Exim might
19 lock up or crash if it happened to be inside a call to libc when it
20 got a SIGUSR1 from exiwhat.
22 The SIGUSR1 handler appends the current process status to the process
23 log which is later printed by exiwhat. It used to use the general
24 purpose logging code to do this, but several functions it calls are
27 The new output code in the SIGUSR1 handler is specific to the process
28 log, and simple enough that it's easy to inspect for signal safety.
29 Removing some special cases also simplifies the general logging code.
30 Removing the spurious timestamps from the process log simplifies
33 TF/04 Improved ratelimit ACL condition.
35 The /noupdate option has been deprecated in favour of /readonly which
36 has clearer semantics. The /leaky, /strict, and /readonly update modes
37 are mutually exclusive. The update mode is no longer included in the
38 database key; it just determines when the database is updated. (This
39 means that when you upgrde Exim will forget old rate measurements.)
41 Exim now checks that the per_* options are used with an update mode that
42 makes sense for the current ACL. For example, when Exim is processing a
43 message (e.g. acl_smtp_rcpt or acl_smtp_data, etc.) you can specify
44 per_mail/leaky or per_mail/strict; otherwise (e.g. in acl_smtp_helo) you
45 must specify per_mail/readonly. If you omit the update mode it defaults to
46 /leaky where that makes sense (as before) or /readonly where required.
48 The /noupdate option is now undocumented but still supported for
49 backwards compatibility. It is equivalent to /readonly except that in
50 ACLs where /readonly is required you may specify /leaky/noupdate or
51 /strict/noupdate which are treated the same as /readonly.
53 A useful new feature is the /count= option. This is a generalization
54 of the per_byte option, so that you can measure the throughput of other
55 aggregate values. For example, the per_byte option is now equivalent
56 to per_mail/count=${if >{0}{$message_size} {0} {$message_size} }.
58 The per_rcpt option has been generalized using the /count= mechanism
59 (though it's more complicated than the per_byte equivalence). When it is
60 used in acl_smtp_rcpt, the per_rcpt option adds recipients to the
61 measured rate one at a time; if it is used later (e.g. in acl_smtp_data)
62 or in a non-SMTP ACL it adds all the recipients in one go. (The latter
63 /count=$recipients_count behaviour used to work only in non-SMTP ACLs.)
64 Note that using per_rcpt with a non-readonly update mode in more than
65 one ACL will cause the recipients to be double-counted. (The per_mail
66 and per_byte options don't have this problem.)
68 The handling of very low rates has changed slightly. If the computed rate
69 is less than the event's count (usually one) then this event is the first
70 after a long gap. In this case the rate is set to the same as this event's
71 count, so that the first message of a spam run is counted properly.
73 The major new feature is a mechanism for counting the rate of unique
74 events. The new per_addr option counts the number of different
75 recipients that someone has sent messages to in the last time period. It
76 behaves like per_rcpt if all the recipient addresses are different, but
77 duplicate recipient addresses do not increase the measured rate. Like
78 the /count= option this is a general mechanism, so the per_addr option
79 is equivalent to per_rcpt/unique=$local_part@$domain. You can, for
80 example, measure the rate that a client uses different sender addresses
81 with the options per_mail/unique=$sender_address. There are further
82 details in the main documentation.
84 TF/05 Removed obsolete $Cambridge$ CVS revision strings.
86 TF/06 Removed a few PCRE remnants.
88 TF/07 Automatically extract Exim's version number from tags in the git
89 repository when doing development or release builds.
91 PP/02 Raise smtp_cmd_buffer_size to 16kB.
92 Bugzilla 879. Patch from Paul Fisher.
94 PP/03 Implement SSL-on-connect outbound with protocol=smtps on smtp transport.
95 Heavily based on revision 40f9a89a from Simon Arlott's tree.
98 PP/04 Use .dylib instead of .so for dynamic library loading on MacOS.
100 PP/05 Variable $av_failed, true if the AV scanner deferred.
101 Bugzilla 1078. Patch from John Horne.
103 PP/06 Stop make process more reliably on build failure.
104 Bugzilla 1087. Patch from Heiko Schlittermann.
106 PP/07 Make maildir_use_size_file an _expandable_ boolean.
107 Bugzilla 1089. Patch from Heiko Schlittermann.
109 PP/08 Handle ${run} returning more data than OS pipe buffer size.
110 Bugzilla 1131. Patch from Holger Weiß.
112 PP/09 Handle IPv6 addresses with SPF.
113 Bugzilla 860. Patch from Wolfgang Breyha.
115 PP/10 GnuTLS: support TLS 1.2 & 1.1.
117 Use gnutls_certificate_verify_peers2() [patch from Andreas Metzler].
120 PP/11 match_* no longer expand right-hand-side by default.
121 New compile-time build option, EXPAND_LISTMATCH_RHS.
122 New expansion conditions, "inlist", "inlisti".
128 PP/01 The new ldap_require_cert option would segfault if used. Fixed.
130 PP/02 Harmonised TLS library version reporting; only show if debugging.
131 Layout now matches that introduced for other libraries in 4.74 PP/03.
133 PP/03 New openssl_options items: no_sslv2 no_sslv3 no_ticket no_tlsv1
135 PP/04 New "dns_use_edns0" global option.
137 PP/05 Don't segfault on misconfiguration of ref:name exim-user as uid.
140 PP/06 Extra paranoia around buffer usage at the STARTTLS transition.
141 nb: Exim is not vulnerable to http://www.kb.cert.org/vuls/id/555316
143 TK/01 Updated PolarSSL code to 0.14.2.
144 Bugzilla 1097. Patch from Andreas Metzler.
146 PP/07 Catch divide-by-zero in ${eval:...}.
149 PP/08 Condition negation of bool{}/bool_lax{} did not negate. Fixed.
152 TK/02 Bugzilla 1106: CVE-2011-1764 - DKIM log line was subject to a
153 format-string attack -- SECURITY: remote arbitrary code execution.
155 TK/03 SECURITY - DKIM signature header parsing was double-expanded, second
156 time unintentionally subject to list matching rules, letting the header
157 cause arbitrary Exim lookups (of items which can occur in lists, *not*
158 arbitrary string expansion). This allowed for information disclosure.
160 PP/09 Fix another SIGFPE (x86) in ${eval:...} expansion, this time related to
161 INT_MIN/-1 -- value coerced to INT_MAX.
167 NM/01 Workround for PCRE version dependency in version reporting
170 TF/01 Update valgrind.h and memcheck.h to copies from valgrind-3.6.0.
171 This fixes portability to compilers other than gcc, notably
172 Solaris CC and HP-UX CC. Fixes Bugzilla 1050.
174 TF/02 Bugzilla 139: Avoid using the += operator in the modular lookup
175 makefiles for portability to HP-UX and POSIX correctness.
177 PP/01 Permit LOOKUP_foo enabling on the make command-line.
178 Also via indented variable definition in the Makefile.
179 (Debugging by Oliver Heesakkers).
181 PP/02 Restore caching of spamd results with expanded spamd_address.
182 Patch from author of expandable spamd_address patch, Wolfgang Breyha.
184 PP/03 Build issue: lookups-Makefile now exports LC_ALL=C
185 Improves build reliability. Fix from: Frank Elsner
187 NM/02 Fix wide character breakage in the rfc2047 coding
188 Fixes bug 1064. Patch from Andrey N. Oktyabrski
190 NM/03 Allow underscore in dnslist lookups
191 Fixes bug 1026. Patch from Graeme Fowler
193 PP/04 Bugzilla 230: Support TLS-enabled LDAP (in addition to ldaps).
194 Code patches from Adam Ciarcinski of NetBSD.
196 NM/04 Fixed exiqgrep to cope with mailq missing size issue
199 PP/05 Bugzilla 1083: when lookup expansion defers, escape the output which
200 is logged, to avoid truncation. Patch from John Horne.
202 PP/06 Bugzilla 1042: implement freeze_signal on pipe transports.
203 Patch from Jakob Hirsch.
205 PP/07 Bugzilla 1061: restrict error messages sent over SMTP to not reveal
206 SQL string expansion failure details.
207 Patch from Andrey Oktyabrski.
209 PP/08 Bugzilla 486: implement %M datestamping in log filenames.
210 Patch from Simon Arlott.
212 PP/09 New lookups functionality failed to compile on old gcc which rejects
213 extern declarations in function scope.
214 Patch from Oliver Fleischmann
216 PP/10 Use sig_atomic_t for flags set from signal handlers.
217 Check getgroups() return and improve debugging.
218 Fixed developed for diagnosis in bug 927 (which turned out to be
221 PP/11 Bugzilla 1055: Update $message_linecount for maildir_tag.
222 Patch from Mark Zealey.
224 PP/12 Bugzilla 1056: Improved spamd server selection.
225 Patch from Mark Zealey.
227 PP/13 Bugzilla 1086: Deal with maildir quota file races.
228 Based on patch from Heiko Schlittermann.
230 PP/14 Bugzilla 1019: DKIM multiple signature generation fix.
231 Patch from Uwe Doering, sign-off by Michael Haardt.
233 NM/05 Fix to spam.c to accommodate older gcc versions which dislike
234 variable declaration deep within a block. Bug and patch from
237 PP/15 lookups-Makefile IRIX compatibilty coercion.
239 PP/16 Make DISABLE_DKIM build knob functional.
241 NM/06 Bugzilla 968: child_open_uid: restore default SIGPIPE handler
242 Patch by Simon Arlott
244 TF/03 Fix valgrind.h portability to C89 compilers that do not support
245 variable argument macros. Our copy now differs from upstream.
251 TF/01 Failure to get a lock on a hints database can have serious
252 consequences so log it to the panic log.
254 TF/02 Log LMTP confirmation messages in the same way as SMTP,
255 controlled using the smtp_confirmation log selector.
257 TF/03 Include the error message when we fail to unlink a spool file.
259 DW/01 Bugzilla 139: Support dynamically loaded lookups as modules.
260 With thanks to Steve Haslam, Johannes Berg & Serge Demonchaux
261 for maintaining out-of-tree patches for some time.
263 PP/01 Bugzilla 139: Documentation and portability issues.
264 Avoid GNU Makefile-isms, let Exim continue to build on BSD.
265 Handle per-OS dynamic-module compilation flags.
267 PP/02 Let /dev/null have normal permissions.
268 The 4.73 fixes were a little too stringent and complained about the
269 permissions on /dev/null. Exempt it from some checks.
270 Reported by Andreas M. Kirchwitz.
272 PP/03 Report version information for many libraries, including
273 Exim version information for dynamically loaded libraries. Created
274 version.h, now support a version extension string for distributors
275 who patch heavily. Dynamic module ABI change.
277 PP/04 CVE-2011-0017 - check return value of setuid/setgid. This is a
278 privilege escalation vulnerability whereby the Exim run-time user
279 can cause root to append content of the attacker's choosing to
282 PP/05 Bugzilla 1041: merged DCC maintainer's fixes for return code.
285 PP/06 Bugzilla 1071: fix delivery logging with untrusted macros.
286 If dropping privileges for untrusted macros, we disabled normal logging
287 on the basis that it would fail; for the Exim run-time user, this is not
288 the case, and it resulted in successful deliveries going unlogged.
289 Fixed. Reported by Andreas Metzler.
295 PP/01 Date: & Message-Id: revert to normally being appended to a message,
296 only prepend for the Resent-* case. Fixes regression introduced in
297 Exim 4.70 by NM/22 for Bugzilla 607.
299 PP/02 Include check_rfc2047_length in configure.default because we're seeing
300 increasing numbers of administrators be bitten by this.
302 JJ/01 Added DISABLE_DKIM and comment to src/EDITME
304 PP/03 Bugzilla 994: added openssl_options main configuration option.
306 PP/04 Bugzilla 995: provide better SSL diagnostics on failed reads.
308 PP/05 Bugzilla 834: provide a permit_coredump option for pipe transports.
310 PP/06 Adjust NTLM authentication to handle SASL Initial Response.
312 PP/07 If TLS negotiated an anonymous cipher, we could end up with SSL but
313 without a peer certificate, leading to a segfault because of an
314 assumption that peers always have certificates. Be a little more
315 paranoid. Problem reported by Martin Tscholak.
317 PP/08 Bugzilla 926: switch ClamAV to use the new zINSTREAM API for content
318 filtering; old API available if built with WITH_OLD_CLAMAV_STREAM=yes
319 NB: ClamAV planning to remove STREAM in "middle of 2010".
320 CL also introduces -bmalware, various -d+acl logging additions and
321 more caution in buffer sizes.
323 PP/09 Implemented reverse_ip expansion operator.
325 PP/10 Bugzilla 937: provide a "debug" ACL control.
327 PP/11 Bugzilla 922: Documentation dusting, patch provided by John Horne.
329 PP/12 Bugzilla 973: Implement --version.
331 PP/13 Bugzilla 752: Refuse to build/run if Exim user is root/0.
333 PP/14 Build without WITH_CONTENT_SCAN. Path from Andreas Metzler.
335 PP/15 Bugzilla 816: support multiple condition rules on Routers.
337 PP/16 Add bool_lax{} expansion operator and use that for combining multiple
338 condition rules, instead of bool{}. Make both bool{} and bool_lax{}
339 ignore trailing whitespace.
341 JJ/02 prevent non-panic DKIM error from being sent to paniclog
343 JJ/03 added tcp_wrappers_daemon_name to allow host entries other than
346 PP/17 Fix malware regression for cmdline scanner introduced in PP/08.
347 Notification from Dr Andrew Aitchison.
349 PP/18 Change ClamAV response parsing to be more robust and to handle ClamAV's
350 ExtendedDetectionInfo response format.
351 Notification from John Horne.
353 PP/19 OpenSSL 1.0.0a compatibility const-ness change, should be backwards
356 PP/20 Added a CONTRIBUTING file. Fixed the documentation build to use http:
357 XSL and documented dependency on system catalogs, with examples of how
360 DW/21 Added Valgrind hooks in store.c to help it capture out-of-bounds store
363 DW/22 Bugzilla 1044: CVE-2010-4345 - partial fix: restrict default behaviour
364 of CONFIGURE_OWNER and CONFIGURE_GROUP options to no longer allow a
365 configuration file which is writeable by the Exim user or group.
367 DW/23 Bugzilla 1044: CVE-2010-4345 - part two: extend checks for writeability
368 of configuration files to cover files specified with the -C option if
369 they are going to be used with root privileges, not just the default
372 DW/24 Bugzilla 1044: CVE-2010-4345 - part three: remove ALT_CONFIG_ROOT_ONLY
373 option (effectively making it always true).
375 DW/25 Add TRUSTED_CONFIG_PREFIX_FILE option to allow alternative configuration
376 files to be used while preserving root privileges.
378 DW/26 Set FD_CLOEXEC on SMTP sockets after forking in the daemon, to ensure
379 that rogue child processes cannot use them.
381 PP/27 Bugzilla 1047: change the default for system_filter_user to be the Exim
382 run-time user, instead of root.
384 PP/28 Add WHITELIST_D_MACROS option to let some macros be overridden by the
385 Exim run-time user without dropping privileges.
387 DW/29 Remove use of va_copy() which breaks pre-C99 systems. Duplicate the
388 result string, instead of calling string_vformat() twice with the same
391 DW/30 Allow TRUSTED_CONFIG_PREFIX_FILE only for Exim or CONFIGURE_OWNER, not
392 for other users. Others should always drop root privileges if they use
393 -C on the command line, even for a whitelisted configure file.
395 DW/31 Turn TRUSTED_CONFIG_PREFIX_FILE into TRUSTED_CONFIG_FILE. No prefixes.
397 NM/01 Fixed bug #1002 - Message loss when using multiple deliveries
403 JJ/01 installed exipick 20100104.1, adding $max_received_linelength,
404 $data_path, and $header_path variables; fixed documentation bugs and
407 JJ/02 installed exipick 20100222.0, added --input-dir and --finput to allow
408 exipick to access non-standard spools, including the "frozen" queue
411 NM/01 Bugzilla 965: Support mysql stored procedures.
412 Patch from Alain Williams
414 NM/02 Bugzilla 961: Spacing fix (syntax error) on Makefile directives for NetBSD
416 NM/03 Bugzilla 955: Documentation fix for max_rcpts.
417 Patch from Andreas Metzler
419 NM/04 Bugzilla 954: Fix for unknown responses from Dovecot authenticator.
420 Patch from Kirill Miazine
422 NM/05 Bugzilla 671: Added umask to procmail example.
424 JJ/03 installed exipick 20100323.0, fixing doc bug
426 NM/06 Bugzilla 988: CVE-2010-2023 - prevent hardlink attack on sticky mail
427 directory. Notification and patch from Dan Rosenberg.
429 TK/01 PDKIM: Upgrade PolarSSL files to upstream version 0.12.1.
431 TK/02 Improve log output when DKIM signing operation fails.
433 MH/01 Treat the transport option dkim_domain as a colon separated
434 list, not as a single string, and sign the message with each element,
435 omitting multiple occurences of the same signer.
437 NM/07 Null terminate DKIM strings, Null initialise DKIM variable
438 Bugzilla 985, 986. Patch by Simon Arlott
440 NM/08 Bugzilla 967. dnsdb DNS TXT record bug fix (DKIM-related)
441 Patch by Simon Arlott
443 PP/01 Bugzilla 989: CVE-2010-2024 - work round race condition on
444 MBX locking. Notification from Dan Rosenberg.
450 TK/01 Bugzilla 912: Fix DKIM segfault on empty headers/body.
452 NM/01 Bugzilla 913: Documentation fix for gnutls_* options.
454 NM/02 Bugzilla 722: Documentation for randint. Better randomness defaults.
456 NM/03 Bugzilla 847: Enable DNSDB lookup by default.
458 NM/04 Bugzilla 915: Flag broken perl installation during build.
464 TK/01 Added patch by Johannes Berg that expands the main option
465 "spamd_address" if it starts with a dollar sign.
467 TK/02 Write list of recipients to X-Envelope-Sender header when building
468 the mbox-format spool file for content scanning (suggested by Jakob
471 TK/03 Added patch by Wolfgang Breyha that adds experimental DCC
472 (http://www.dcc-servers.net/) support via dccifd. Activated by
473 setting EXPERIMENTAL_DCC=yes in Local/Makefile.
475 TK/04 Bugzilla 673: Add f-protd malware scanner support. Patch submitted
476 by Mark Daniel Reidel <mr@df.eu>.
478 NM/01 Bugzilla 657: Embedded PCRE removed from the exim source tree.
479 When building exim an external PCRE library is now needed -
480 PCRE is a system library on the majority of modern systems.
481 See entry on PCRE_LIBS in EDITME file.
483 NM/02 Bugzilla 646: Removed unwanted C/R in Dovecot authenticator
484 conversation. Added nologin parameter to request.
485 Patch contributed by Kirill Miazine.
487 TF/01 Do not log submission mode rewrites if they do not change the address.
489 TF/02 Bugzilla 662: Fix stack corruption before exec() in daemon.c.
491 NM/03 Bugzilla 602: exicyclog now handles panic log, and creates empty
492 log files in place. Contributed by Roberto Lima.
494 NM/04 Bugzilla 667: Close socket used by dovecot authenticator.
496 TF/03 Bugzilla 615: When checking the local_parts router precondition
497 after a local_part_suffix or local_part_prefix option, Exim now
498 does not use the address's named list lookup cache, since this
499 contains cached lookups for the whole local part.
501 NM/05 Bugzilla 521: Integrated SPF Best Guess support contributed by
502 Robert Millan. Documentation is in experimental-spec.txt.
504 TF/04 Bugzilla 668: Fix parallel build (make -j).
506 NM/05.2 Bugzilla 437: Prevent Maildir aux files being created with mode 000.
508 NM/05.3 Bugzilla 598: Improvement to Dovecot authenticator handling.
509 Patch provided by Jan Srzednicki.
511 TF/05 Leading white space used to be stripped from $spam_report which
512 wrecked the formatting. Now it is preserved.
514 TF/06 Save $spam_score, $spam_bar, and $spam_report in spool files, so
515 that they are available at delivery time.
517 TF/07 Fix the way ${extract is skipped in the untaken branch of a conditional.
519 TF/08 TLS error reporting now respects the incoming_interface and
520 incoming_port log selectors.
522 TF/09 Produce a more useful error message if an SMTP transport's hosts
523 setting expands to an empty string.
525 NM/06 Bugzilla 744: EXPN did not work under TLS.
526 Patch provided by Phil Pennock.
528 NM/07 Bugzilla 769: Extraneous comma in usage fprintf
529 Patch provided by Richard Godbee.
531 NM/08 Fixed erroneous documentation references to smtp_notquit_acl to be
532 acl_smtp_notquit, added index entry.
534 NM/09 Bugzilla 787: Potential buffer overflow in string_format.
535 Patch provided by Eugene Bujak.
537 NM/10 Bugzilla 770: Problem on some platforms modifying the len parameter to
538 accept(). Patch provided by Maxim Dounin.
540 NM/11 Bugzilla 749: Preserve old behaviour of blanks comparing equal to zero.
541 Patch provided by Phil Pennock.
543 NM/12 Bugzilla 497: Correct behaviour of exiwhat when no config exists.
545 NM/13 Bugzilla 590: Correct handling of Resent-Date headers.
546 Patch provided by Brad "anomie" Jorsch.
548 NM/14 Bugzilla 622: Added timeout setting to transport filter.
549 Patch provided by Dean Brooks.
551 TK/05 Add native DKIM support (does not depend on external libraries).
553 NM/15 Bugzilla 854: Removed code that symlinks to pcre as its no longer useful.
554 Patch provided by Graeme Fowler.
556 NM/16 Bugzilla 851: Documentation example syntax fix.
558 NM/17 Changed NOTICE file to remove references to embedded PCRE.
560 NM/18 Bugzilla 894: Fix issue with very long lines including comments in
563 NM/19 Bugzilla 745: TLS version reporting.
564 Patch provided by Phil Pennock.
566 NM/20 Bugzilla 167: bool: condition support.
567 Patch provided by Phil Pennock.
569 NM/21 Bugzilla 665: gnutls_compat_mode to allow compatibility with broken
570 clients. Patch provided by Phil Pennock.
572 NM/22 Bugzilla 607: prepend (not append) Resent-Message-ID and Resent-Date.
573 Patch provided by Brad "anomie" Jorsch.
575 NM/23 Bugzilla 687: Fix misparses in eximstats.
576 Patch provided by Heiko Schlittermann.
578 NM/24 Bugzilla 688: Fix exiwhat to handle log_selector = +pid.
579 Patch provided by Heiko Schlittermann.
581 NM/25 Bugzilla 727: Use transport mode as default mode for maildirsize file.
582 plus update to original patch.
584 NM/26 Bugzilla 799: Documentation correction for ratelimit.
586 NM/27 Bugzilla 802: Improvements to local interface IP addr detection.
587 Patch provided by David Brownlee.
589 NM/28 Bugzilla 807: Improvements to LMTP delivery logging.
591 NM/29 Bugzilla 862, 866, 875: Documentation bugfixes.
593 NM/30 Bugzilla 888: TLS documentation bugfixes.
595 NM/31 Bugzilla 896: Dovecot buffer overrun fix.
597 NM/32 Bugzilla 889: Change all instances of "expr" in shell scripts to "expr --"
598 Unlike the original bugzilla I have changed all shell scripts in src tree.
600 NM/33 Bugzilla 898: Transport filter timeout fix.
601 Patch by Todd Rinaldo.
603 NM/34 Bugzilla 901: Fix sign/unsigned and UTF mistmatches.
604 Patch by Serge Demonchaux.
606 NM/35 Bugzilla 39: Base64 decode bug fixes.
607 Patch by Jakob Hirsch.
609 NM/36 Bugzilla 909: Correct connect() call in dcc code.
611 NM/37 Bugzilla 910: Correct issue with relaxed/simple handling.
613 NM/38 Bugzilla 908: Removed NetBSD3 support as no longer needed.
615 NM/39 Bugzilla 911: Fixed MakeLinks build script.
621 TK/01 Add preliminary DKIM support. Currently requires a forked version of
622 ALT-N's libdkim that I have put here:
623 http://duncanthrax.net/exim-experimental/
625 Note to Michael Haardt: I had to rename some vars in sieve.c. They
626 were called 'true' and it seems that C99 defines that as a reserved
627 keyword to be used with 'bool' variable types. That means you could
628 not include C99-style headers which use bools without triggering
629 build errors in sieve.c.
631 NM/01 Bugzilla 592: --help option is handled incorrectly if exim is invoked
632 as mailq or other aliases. Changed the --help handling significantly
633 to do whats expected. exim_usage() emits usage/help information.
635 SC/01 Added the -bylocaldomain option to eximstats.
637 NM/02 Bugzilla 619: Defended against bad data coming back from gethostbyaddr.
639 NM/03 Bugzilla 613: Documentation fix for acl_not_smtp.
641 NM/04 Bugzilla 628: PCRE update to 7.4 (work done by John Hall).
647 PH/01 Another patch from the Sieve maintainer.
649 PH/02 When an IPv6 address is converted to a string for single-key lookup
650 in an address list (e.g. for an item such as "net24-dbm;/net/works"),
651 dots are used instead of colons so that keys in lsearch files need not
652 contain colons. This was done some time before quoting was made available
653 in lsearch files. However, iplsearch files do require colons in IPv6 keys
654 (notated using the quote facility) so as to distinguish them from IPv4
655 keys. This meant that lookups for IP addresses in host lists did not work
656 for iplsearch lookups.
658 This has been fixed by arranging for IPv6 addresses to be expressed with
659 colons if the lookup type is iplsearch. This is not incompatible, because
660 previously such lookups could never work.
662 The situation is now rather anomolous, since one *can* have colons in
663 ordinary lsearch keys. However, making the change in all cases is
664 incompatible and would probably break a number of configurations.
666 TK/01 Change PRVS address formatting scheme to reflect latests BATV draft
669 MH/01 The "spam" ACL condition code contained a sscanf() call with a %s
670 conversion specification without a maximum field width, thereby enabling
671 a rogue spamd server to cause a buffer overflow. While nobody in their
672 right mind would setup Exim to query an untrusted spamd server, an
673 attacker that gains access to a server running spamd could potentially
674 exploit this vulnerability to run arbitrary code as the Exim user.
676 TK/02 Bugzilla 502: Apply patch to make the SPF-Received: header use
677 $primary_hostname instead of what libspf2 thinks the hosts name is.
679 MH/02 The dsearch lookup now uses lstat(2) instead of stat(2) to look for
680 a directory entry by the name of the lookup key. Previously, if a
681 symlink pointed to a non-existing file or a file in a directory that
682 Exim lacked permissions to read, a lookup for a key matching that
683 symlink would fail. Now it is enough that a matching directory entry
684 exists, symlink or not. (Bugzilla 503.)
686 PH/03 The body_linecount and body_zerocount variables are now exported in the
689 PH/04 Added the $dnslist_matched variable.
691 PH/05 Unset $tls_cipher and $tls_peerdn before making a connection as a client.
692 This means they are set thereafter only if the connection becomes
695 PH/06 Added the client_condition to authenticators so that some can be skipped
696 by clients under certain conditions.
698 PH/07 The error message for a badly-placed control=no_multiline_responses left
699 "_responses" off the end of the name.
701 PH/08 Added -Mvc to output a copy of a message in RFC 2822 format.
703 PH/09 Tidied the code for creating ratelimiting keys, creating them explicitly
704 (without spaces) instead of just copying the configuration text.
706 PH/10 Added the /noupdate option to the ratelimit ACL condition.
708 PH/11 Added $max_received_linelength.
710 PH/12 Added +ignore_defer and +include_defer to host lists.
712 PH/13 Installed PCRE version 7.2. This needed some changes because of the new
713 way in which PCRE > 7.0 is built.
715 PH/14 Implemented queue_only_load_latch.
717 PH/15 Removed an incorrect (int) cast when reading the value of SIZE in a
718 MAIL command. The effect was to mangle the value on 64-bit systems.
720 PH/16 Another patch from the Sieve maintainer.
722 PH/17 Added the NOTQUIT ACL, based on a patch from Ted Cooper.
724 PH/18 If a system quota error occurred while trying to create the file for
725 a maildir delivery, the message "Mailbox is full" was not appended to the
726 bounce if the delivery eventually timed out. Change 4.67/27 below applied
727 only to a quota excession during the actual writing of the file.
729 PH/19 It seems that peer DN values may contain newlines (and other non-printing
730 characters?) which causes problems in log lines. The DN values are now
731 passed through string_printing() before being added to log lines.
733 PH/20 Added the "servers=" facility to MySQL and PostgreSQL lookups. (Oracle
734 and InterBase are left for another time.)
736 PH/21 Added message_body_newlines option.
738 PH/22 Guard against possible overflow in moan_check_errorcopy().
740 PH/23 POSIX allows open() to be a macro; guard against that.
742 PH/24 If the recipient of an error message contained an @ in the local part
743 (suitably quoted, of course), incorrect values were put in $domain and
744 $local_part during the evaluation of errors_copy.
750 MH/01 Fix for bug #448, segfault in Dovecot authenticator when interface_address
751 is unset (happens when testing with -bh and -oMi isn't used). Thanks to
754 PH/01 Added a new log selector smtp_no_mail, to log SMTP sessions that do not
755 issue a MAIL command.
757 PH/02 In an ACL statement such as
759 deny dnslists = X!=127.0.0.2 : X=127.0.0.2
761 if a client was not listed at all, or was listed with a value other than
762 127.0.0.2, in the X list, but was listed with 127.0.0.2 in the Y list,
763 the condition was not true (as it should be), so access was not denied.
764 The bug was that the ! inversion was incorrectly passed on to the second
765 item. This has been fixed.
767 PH/03 Added additional dnslists conditions == and =& which are different from
768 = and & when the dns lookup returns more than one IP address.
770 PH/04 Added gnutls_require_{kx,mac,protocols} to give more control over the
771 cipher suites used by GnuTLS. These options are ignored by OpenSSL.
773 PH/05 After discussion on the list, added a compile time option ENABLE_DISABLE_
774 FSYNC, which compiles an option called disable_fsync that allows for
775 bypassing fsync(). The documentation is heavily laced with warnings.
777 SC/01 Updated eximstats to collate all SpamAssassin rejects into one bucket.
779 PH/06 Some tidies to the infrastructure of the Test Suite that is concerned
780 with the auxiliary C programs that it uses: (1) Arrange for BIND_8_COMPAT
781 to be defined when compiling on OSX (Darwin); (2) Tidies to the Makefile,
782 including adding "make clean"; (3) Added -fPIC when compiling the test
783 dynamically loaded module, to get rid of a warning.
785 MH/02 Fix for bug #451, causing paniclog entries to be written if a bounce
786 message fails, move_frozen_messages = true and ignore_bounce_errors_after
787 = 0s. The bug is otherwise harmless.
789 PH/07 There was a bug in the dovecot authenticator such that the value of
790 $auth1 could be overwritten, and so not correctly preserved, after a
791 successful authentication. This usually meant that the value preserved by
792 the server_setid option was incorrect.
794 PH/08 Added $smtp_count_at_connection_start, deliberately with a long name.
796 PH/09 Installed PCRE release 7.0.
798 PH/10 The acl_not_smtp_start ACL was, contrary to the documentation, not being
799 run for batched SMTP input. It is now run at the start of every message
800 in the batch. While fixing this I discovered that the process information
801 (output by running exiwhat) was not always getting set for -bs and -bS
802 input. This is fixed, and it now also says "batched" for BSMTP.
804 PH/11 Added control=no_pipelining.
806 PH/12 Added $sending_ip_address and $sending_port (mostly Magnus Holmgren's
807 patch, slightly modified), and move the expansion of helo_data till after
808 the connection is made in the smtp transport (so it can use these
811 PH/13 Added ${rfc2047d: to decoded RFC 2047 strings.
813 PH/14 Added log_selector = +pid.
815 PH/15 Flush SMTP output before delaying, unless control=no_delay_flush is set.
817 PH/16 Add ${if forany and ${if forall.
819 PH/17 Added dsn_from option to vary the From: line in DSNs.
821 PH/18 Flush SMTP output before performing a callout, unless control =
822 no_callout_flush is set.
824 PH/19 Change 4.64/PH/36 introduced a bug: when address_retry_include_sender
825 was true (the default) a successful delivery failed to delete the retry
826 item, thus causing premature timeout of the address. The bug is now
829 PH/20 Added hosts_avoid_pipelining to the smtp transport.
831 PH/21 Long custom messages for fakedefer and fakereject are now split up
832 into multiline reponses in the same way that messages for "deny" and
833 other ACL rejections are.
835 PH/22 Applied Jori Hamalainen's speed-up changes and typo fixes to exigrep,
836 with slight modification.
838 PH/23 Applied sieve patches from the maintainer "tracking the latest notify
839 draft, changing the syntax and factoring some duplicate code".
841 PH/24 When the log selector "outgoing_port" was set, the port was shown as -1
842 for deliveries of the second and subsequent messages over the same SMTP
845 PH/25 Applied Magnus Holmgren's patch for ${addresses, ${map, ${filter, and
846 ${reduce, with only minor "tidies".
848 SC/02 Applied Daniel Tiefnig's patch to improve the '($parent) =' pattern match.
850 PH/26 Added a "continue" ACL modifier that does nothing, for the benefit of its
851 expansion side effects.
853 PH/27 When a message times out after an over-quota error from an Exim-imposed
854 quota, the bounce message says "mailbox is full". This message was not
855 being given when it was a system quota that was exceeded. It now should
858 MH/03 Made $recipients available in local_scan(). local_scan() already has
859 better access to the recipient list through recipients_list[], but
860 $recipients can be useful in postmaster-provided expansion strings.
862 PH/28 The $smtp_command and $smtp_command_argument variables were not correct
863 in the case of a MAIL command with additional options following the
864 address, for example: MAIL FROM:<foo@bar> SIZE=1234. The option settings
865 were accidentally chopped off.
867 PH/29 SMTP synchronization checks are implemented when a command is read -
868 there is a check that no more input is waiting when there shouldn't be
869 any. However, for some commands, a delay in an ACL can mean that it is
870 some time before the response is written. In this time, more input might
871 arrive, invalidly. So now there are extra checks after an ACL has run for
872 HELO/EHLO and after the predata ACL, and likewise for MAIL and RCPT when
873 pipelining has not been advertised.
875 PH/30 MH's patch to allow iscntrl() characters to be list separators.
877 PH/31 Unlike :fail:, a custom message specified with :defer: was not being
878 returned in the SMTP response when smtp_return_error_details was false.
881 PH/32 Change the Dovecot authenticator to use read() and write() on the socket
882 instead of the C I/O that was originally supplied, because problems were
885 PH/33 Compile failed with OpenSSL 0.9.8e. This was due to a coding error in
886 Exim which did not show up earlier: it was assuming that a call to
887 SSL_CTX_set_info_callback() might give an error value. In fact, there is
888 no error. In previous releases of OpenSSL, SSL_CTX_set_info_callback()
889 was a macro that became an assignment, so it seemed to work. This has
890 changed to a proper function call with a void return, hence the compile
891 error. Exim's code has been fixed.
893 PH/34 Change HDA_SIZE in oracle.c from 256 to 512. This is needed for 64-bit
896 PH/35 Applied a patch from the Sieve maintainer which fixes a bug in "notify".
898 PH/36 Applied John Jetmore's patch to add -v functionality to exigrep.
900 PH/37 If a message is not accepted after it has had an id assigned (e.g.
901 because it turns out to be too big or there is a timeout) there is no
902 "Completed" line in the log. When some messages of this type were
903 selected by exigrep, they were listed as "not completed". Others were
904 picked up by some special patterns. I have improved the selection
905 criteria to be more general.
907 PH/38 The host_find_failed option in the manualroute router can now be set
908 to "ignore", to completely ignore a host whose IP address cannot be
909 found. If all hosts are ignored, the behaviour is controlled by the new
910 host_all_ignored option.
912 PH/39 In a list of hosts for manualroute, if one item (either because of multi-
913 homing or because of multiple MX records with /mx) generated more than
914 one IP address, and the following item turned out to be the local host,
915 all the secondary addresses of the first item were incorrectly removed
916 from the list, along with the local host and any following hosts (which
917 is what is supposed to happen).
919 PH/40 When Exim receives a message, it writes the login name, uid, and gid of
920 whoever called Exim into the -H file. In the case of the daemon it was
921 behaving confusingly. When first started, it used values for whoever
922 started the daemon, but after a SIGHUP it used the Exim user (because it
923 calls itself on a restart). I have changed the code so that it now always
926 PH/41 (Following a suggestion from Tony Finch) If all the RCPT commands in a
927 message are rejected with the same error (e.g. no authentication or bad
928 sender address), and a DATA command is nevertheless sent (as can happen
929 with PIPELINING or a stupid MUA), the error message that was given to the
930 RCPT commands is included in the rejection of the DATA command. This is
931 intended to be helpful for MUAs that show only the final error to their
934 PH/42 Another patch from the Sieve maintainer.
936 SC/02 Eximstats - Differentiate between permanent and temporary rejects.
937 Eximstats - Fixed some broken HTML links and added missing column headers
939 Eximstats - Fixed Grand Total Summary Domains, Edomains, and Email
940 columns for Rejects, Temp Rejects, Ham, and Spam rows.
942 SC/03 Eximstats - V1.58 Fix to get <> and blackhole to show in edomain tables.
944 PH/43 Yet another patch from the Sieve maintainer.
946 PH/44 I found a way to check for a TCP/IP connection going away before sending
947 the response to the final '.' that terminates a message, but only in the
948 case where the client has not sent further data following the '.'
949 (unfortunately, this is allowed). However, in many cases there won't be
950 any further data because there won't be any more messages to send. A call
951 to select() can be used: if it shows that the input is "ready", there is
952 either input waiting, or the socket has been closed. An attempt to read
953 the next input character can distinguish the two cases. Previously, Exim
954 would have sent an OK response which the client would never have see.
955 This could lead to message repetition. This fix should cure that, at
956 least in a lot of common cases.
958 PH/45 Do not advertise STARTTLS in response to HELP unless it would be
959 advertised in response to EHLO.
965 PH/01 Two more bugs that were introduced by 4.64/PH/07, in addition to the one
966 fixed by 4.65/MH/01 (is this a record?) are fixed:
968 (i) An empty string was always treated as zero by the numeric comparison
969 operators. This behaviour has been restored.
971 (ii) It is documented that the numeric comparison operators always treat
972 their arguments as decimal numbers. This was broken in that numbers
973 starting with 0 were being interpreted as octal.
975 While fixing these problems I realized that there was another issue that
976 hadn't been noticed. Values of message_size_limit (both the global option
977 and the transport option) were treated as octal if they started with 0.
978 The documentation was vague. These values are now always treated as
979 decimal, and I will make that clear in the documentation.
985 TK/01 Disable default definition of HAVE_LINUX_SENDFILE. Clashes with
986 Linux large file support (_FILE_OFFSET_BITS=64) on older glibc
989 MH/01 Don't check that the operands of numeric comparison operators are
990 integers when their expansion is in "skipping" mode (fixes bug
991 introduced by 4.64-PH/07).
993 PH/01 If a system filter or a router generates more than SHRT_MAX (32767)
994 child addresses, Exim now panics and dies. Previously, because the count
995 is held in a short int, deliveries were likely to be lost. As such a
996 large number of recipients for a single message is ridiculous
997 (performance will be very, very poor), I have chosen to impose a limit
998 rather than extend the field.
1004 TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
1005 leftover -K file (the existence of which was triggered by #402).
1006 While we were at it, introduced process PID as part of the -K
1007 filename. This should rule out race conditions when creating
1010 TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
1011 processing considerably. Previous code took too long for large mails,
1012 triggering a timeout which in turn triggers #401.
1014 TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
1015 in the DK code in transports.c. sendfile() is not really portable,
1016 hence the _LINUX specificness.
1018 TF/01 In the add_headers option to the mail command in an Exim filter,
1019 there was a bug that Exim would claim a syntax error in any
1020 header after the first one which had an odd number of characters
1023 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
1024 callout verification, Exim cached a "reject" for the entire domain. This
1025 is correct for most verifications, but it is not correct for a recipient
1026 verification with use_sender or use_postmaster set, because in that case
1027 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
1028 case of MAIL FROM:<> rejection from other early rejections (e.g.
1029 rejection of HELO). When verifying a recipient using a non-null MAIL
1030 address, the cache is ignored if it shows MAIL FROM:<> rejection.
1031 Whatever the result of the callout, the value of the domain cache is
1032 left unchanged (for any other kind of callout, getting as far as trying
1033 RCPT means that the domain itself is ok).
1035 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
1038 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
1039 manifest itself as EPIPE rather than ECONNECT. When tidying away a
1040 session, the daemon ignores ECONNECT errors and logs others; it now
1041 ignores EPIPE as well.
1043 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
1044 (quoted-printable decoding).
1046 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
1047 later the small subsequent patch to fix an introduced bug.
1049 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
1051 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
1053 PH/08 An error is now given if message_size_limit is specified negative.
1055 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
1056 to be given (somewhat) arbitrary names.
1058 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
1061 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
1062 miscellaneous code fixes
1064 PH/10 Added the log_reject_target ACL modifier to specify where to log
1067 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
1068 hostname. This is wrong, because it relates to the incoming message (and
1069 probably the interface on which it is arriving) and not to the outgoing
1070 callout (which could be using a different interface). This has been
1071 changed to use the value of the helo_data option from the smtp transport
1072 instead - this is what is used when a message is actually being sent. If
1073 there is no remote transport (possible with a router that sets up host
1074 addresses), $smtp_active_hostname is used.
1076 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
1077 tweaks were necessary in order to get it to work (see also 21 below):
1078 (a) The code assumed that strncpy() returns a negative number on buffer
1079 overflow, which isn't the case. Replaced with Exim's string_format()
1081 (b) There were several signed/unsigned issues. I just did the minimum
1082 hacking in of casts. There is scope for a larger refactoring.
1083 (c) The code used strcasecmp() which is not a standard C function.
1084 Replaced with Exim's strcmpic() function.
1085 (d) The code set only $1; it now sets $auth1 as well.
1086 (e) A simple test gave the error "authentication client didn't specify
1087 service in request". It would seem that Dovecot has changed its
1088 interface. Fortunately there's a specification; I followed it and
1089 changed what the client sends and it appears to be working now.
1091 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
1094 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
1095 address A is aliased to B and C, where B exists and C does not. Without
1096 -v the output is "A verified" because verification stops after a
1097 successful redirection if more than one address is generated. However,
1098 with -v the child addresses are also verified. Exim was outputting "A
1099 failed to verify" and then showing the successful verification for C,
1100 with its parentage. It now outputs "B failed to verify", showing B's
1101 parentage before showing the successful verification of C.
1103 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
1104 look up a TXT record in a specific list after matching in a combined
1107 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
1108 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
1109 they consult the DNS. I had assumed they would set it the way they
1110 wanted; and indeed my experiments on Linux seem to show that in some
1111 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
1112 To be on the safe side, however, I have now made the interface to
1113 host_find_byname() similar to host_find_bydns(), with an argument
1114 containing the DNS resolver options. The host_find_byname() function now
1115 sets these options at its start, just as host_find_bydns() does. The smtp
1116 transport options dns_qualify_single and dns_search_parents are passed to
1117 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
1118 of host_find_byname() use the default settings of RES_DEFNAMES
1119 (qualify_single) but not RES_DNSRCH (search_parents).
1121 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
1122 spool_read_header() do less string testing, by means of a preliminary
1123 switch on the second character of optional "-foo" lines. (This is
1124 overdue, caused by the large number of possibilities that now exist.
1125 Originally there were few.) While I was there, I also converted the
1126 str(n)cmp tests so they don't re-test the leading "-" and the first
1127 character, in the hope this might squeeze out yet more improvement.
1129 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
1130 flag allowing group syntax was set by the header_syntax check but not
1131 turned off, possible causing trouble later; (2) The flag was not being
1132 set at all for the header_verify test, causing "group"-style headers to
1133 be rejected. I have now set it in this case, and also caused header_
1134 verify to ignore an empty address taken from a group. While doing this, I
1135 came across some other cases where the code for allowing group syntax
1136 while scanning a header line wasn't quite right (mostly, not resetting
1137 the flag correctly in the right place). These bugs could have caused
1138 trouble for malformed header lines. I hope it is now all correct.
1140 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
1141 with the "reply" argument non-NULL. The code, however (which originally
1142 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
1143 but it didn't always do it. This confused somebody who was copying the
1144 code for some other use. I have removed all the tests.
1146 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
1147 feature that was used to support insecure browsers during the U.S. crypto
1148 embargo. It requires special client support, and Exim is probably the
1149 only MTA that supported it -- and would never use it because real RSA is
1150 always available. This code has been removed, because it had the bad
1151 effect of slowing Exim down by computing (never used) parameters for the
1152 RSA_EXPORT functionality.
1154 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
1155 authenticator to fail if there's a tab character in the incoming data
1156 (there should never be unless someone is messing about, as it's supposed
1157 to be base64-encoded). Also added, on Timo's advice, the "secured" option
1158 if the connection is using TLS or if the remote IP is the same as the
1159 local IP, and the "valid-client-cert option" if a client certificate has
1162 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
1163 authenticators. This can be used for authorization after authentication
1164 succeeds. (In the case of plaintext, it servers for both authentication
1167 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
1168 if any retry times were supplied.
1170 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
1171 connection, where there is no client IP address to check. In this
1172 situation, the verify now always succeeds.
1174 PH/25 Applied John Jetmore's -Mset patch.
1176 PH/26 Added -bem to be like -Mset, but loading a message from a file.
1178 PH/27 In a string expansion for a processed (not raw) header when multiple
1179 headers of the same name were present, leading whitespace was being
1180 removed from all of them, but trailing whitespace was being removed only
1181 from the last one. Now trailing whitespace is removed from each header
1182 before concatenation. Completely empty headers in a concatenation (as
1183 before) are ignored.
1185 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
1186 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
1188 PH/29 [Removed. This was a change that I later backed out, and forgot to
1189 correct the ChangeLog entry (that I had efficiently created) before
1190 committing the later change.]
1192 PH/30 Exim was sometimes attempting to deliver messages that had suffered
1193 address errors (4xx response to RCPT) over the same connection as other
1194 messages routed to the same hosts. Such deliveries are always "forced",
1195 so retry times are not inspected. This resulted in far too many retries
1196 for the affected addresses. The effect occurred only when there were more
1197 hosts than the hosts_max_try setting in the smtp transport when it had
1198 the 4xx errors. Those hosts that it had tried were not added to the list
1199 of hosts for which the message was waiting, so if all were tried, there
1200 was no problem. Two fixes have been applied:
1202 (i) If there are any address or message errors in an SMTP delivery, none
1203 of the hosts (tried or untried) are now added to the list of hosts
1204 for which the message is waiting, so the message should not be a
1205 candidate for sending over the same connection that was used for a
1206 successful delivery of some other message. This seems entirely
1207 reasonable: after all the message is NOT "waiting for some host".
1208 This is so "obvious" that I'm not sure why it wasn't done
1209 previously. Hope I haven't missed anything, but it can't do any
1210 harm, as the worst effect is to miss an optimization.
1212 (ii) If, despite (i), such a delivery is accidentally attempted, the
1213 routing retry time is respected, so at least it doesn't keep
1214 hammering the server.
1216 PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
1217 in ${readsocket because some servers need this prod.
1219 PH/32 Added some extra debug output when updating a wait-xxx database.
1221 PH/33 The hint "could be header name not terminated by colon", which has been
1222 given for certain expansion errors for a long time, was not being given
1223 for the ${if def:h_colon_omitted{... case.
1225 PH/34 The spec says: "With one important exception, whenever a domain list is
1226 being scanned, $domain contains the subject domain." There was at least
1227 one case where this was not true.
1229 PH/35 The error "getsockname() failed: connection reset by peer" was being
1230 written to the panic log as well as the main log, but it isn't really
1231 panic-worthy as it just means the connection died rather early on. I have
1232 removed the panic log writing for the ECONNRESET error when getsockname()
1235 PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
1236 runs only) independently of the message's sender address. This meant
1237 that, if the 4xx error was in fact related to the sender, a different
1238 message to the same recipient with a different sender could confuse
1239 things. In particualar, this can happen when sending to a greylisting
1240 server, but other circumstances could also provoke similar problems.
1241 I have changed the default so that the retry time for these errors is now
1242 based a combination of the sender and recipient addresses. This change
1243 can be overridden by setting address_retry_include_sender=false in the
1246 PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
1247 remote server are returned as part of bounce messages. This was not
1248 happening for LMTP over a pipe (the lmtp transport), but now it is the
1249 same for both kinds of LMTP.
1251 PH/38 Despite being documented as not happening, Exim was rewriting addresses
1252 in header lines that were in fact CNAMEs. This is no longer the case.
1254 PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
1255 and queue runs started by the daemon processed all messages. This has
1256 been fixed so that -R and -S can now usefully be given with -q<time>.
1258 PH/40 Import PCRE release 6.7 (fixes some bugs).
1260 PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
1262 PH/42 Give an error if -q is specified more than once.
1264 PH/43 Renamed the variables $interface_address and $interface_port as
1265 $received_ip_address and $received_port, to make it clear that these
1266 values apply to message reception, and not to the outgoing interface when
1267 a message is delivered. (The old names remain recognized, of course.)
1269 PH/44 There was no timeout on the connect() call when using a Unix domain
1270 socket in the ${readsocket expansion. There now is.
1272 PH/45 Applied a modified version of Brad Jorsch's patch to allow "message" to
1273 be meaningful with "accept".
1275 SC/01 Eximstats V1.43
1276 Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
1278 SC/02 Eximstats V1.44
1279 Use a glob alias rather than an array ref in the generated
1280 parser. This improves both readability and performance.
1282 SC/03 Eximstats V1.45 (Marco Gaiarin / Steve Campbell)
1283 Collect SpamAssassin and rejection statistics.
1284 Don't display local sender or destination tables unless
1285 there is data to show.
1286 Added average volumes into the top table text output.
1288 SC/04 Eximstats V1.46
1289 Collect data on the number of addresses (recipients)
1290 as well as the number of messages.
1292 SC/05 Eximstats V1.47
1293 Added 'Message too big' to the list of mail rejection
1294 reasons (thanks to Marco Gaiarin).
1296 SC/06 Eximstats V1.48
1297 Mainlog lines which have GMT offsets and are too short to
1298 have a flag are now skipped.
1300 SC/07 Eximstats V1.49 (Alain Williams)
1301 Added the -emptyok flag.
1303 SC/08 Eximstats V1.50
1304 Fixes for obtaining the IP address from reject messages.
1306 JJ/03 exipick.20061117.2, made header handling as similar to exim as possible
1307 (added [br]h_ prefixes, implemented RFC2047 decoding. Fixed
1308 whitesspace changes from 4.64-PH/27
1310 JJ/04 exipick.20061117.2, fixed format and added $message_headers_raw to
1313 JJ/05 exipick.20061117.2, bug fixes (error out sooner when invalid criteria
1314 are found, allow negative numbers in numeric criteria)
1316 JJ/06 exipick.20061117.2, added new $message_body_missing variable
1318 JJ/07 exipick.20061117.2, added $received_ip_address and $received_port
1319 to match changes made in 4.64-PH/43
1321 PH/46 Applied Jori Hamalainen's patch to add features to exiqsumm.
1323 PH/47 Put in an explicit test for a DNS lookup of an address record where the
1324 "domain" is actually an IP address, and force a failure. This locks out
1325 those revolvers/nameservers that support "A-for-A" lookups, in
1326 contravention of the specifications.
1328 PH/48 When a host name was looked up from an IP address, and the subsequent
1329 forward lookup of the name timed out, the host name was left in
1330 $sender_host_name, contrary to the specification.
1332 PH/49 Although default lookup types such as lsearch* or cdb*@ have always been
1333 restricted to single-key lookups, Exim was not diagnosing an error if
1334 * or *@ was used with a query-style lookup.
1336 PH/50 Increased the value of DH_BITS in tls-gnu.c from 768 to 1024.
1338 MH/01 local_scan ABI version incremented to 1.1. It should have been updated
1339 long ago, but noone interested enough thought of it. Let's just say that
1340 the "1.1" means that there are some new functions that weren't there at
1341 some point in the past.
1343 PH/51 Error processing for expansion failure of helo_data from an smtp
1344 transport during callout processing was broken.
1346 PH/52 Applied John Jetmore's patch to allow tls-on-connect and STARTTLS to be
1347 tested/used via the -bh/-bhc/-bs options.
1349 PH/53 Added missing "#include <time.h>" to pcre/pcretest.c (this was a PCRE
1350 bug, fixed in subsequent PCRE releases).
1352 PH/54 Applied Robert Bannocks' patch to avoid a problem with references that
1353 arises when using the Solaris LDAP libraries (but not with OpenLDAP).
1355 PH/55 Check for a ridiculously long file name in exim_dbmbuild.
1361 SC/01 Use a glob alias rather than an array ref in eximstats generated
1362 parser. This improves both readability and performance.
1364 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
1365 Don't display local sender or destination tables in eximstats unless
1366 there is data to show.
1367 Added average volumes into the eximstats top table text output.
1369 SC/03 Collect data on the number of addresses (recipients) as well
1370 as the number of messages in eximstats.
1372 TF/01 Correct an error in the documentation for the redirect router. Exim
1373 does (usually) call initgroups() when daemonizing.
1375 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
1376 with consistent privilege compared to when running as a daemon.
1378 TF/03 Note in the spec that $authenticated_id is not set for local
1379 submissions from trusted users.
1381 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
1382 Thanks to Dean Brooks <dean@iglou.com> for the patch.
1384 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
1385 by adding some example configuration directives to the default
1386 configuration file. A little bit of work is required to uncomment the
1387 directives and define how usernames and passwords are checked, but
1388 there is now a framework to start from.
1390 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
1391 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
1392 without this. I don't know how relevant this is to other LDAP libraries.
1394 PH/02 Add the verb name to the "unknown ACL verb" error.
1396 PH/03 Magnus Holmgren's patch for filter_prepend_home.
1398 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
1400 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
1401 directory not expanded when it should be if an expanded home directory
1402 was set for the address (which is overridden by the transport).
1404 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
1407 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
1408 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
1409 because it is too late at that time, and has no effect.
1411 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
1412 security issue with \' (bugzilla #107). I could not use the
1413 PQescapeStringConn() function, because it needs a PGconn value as one of
1416 PH/08 When testing addresses using -bt, indicate those final addresses that
1417 are duplicates that would not cause an additional delivery. At least one
1418 person was confused, thinking that -bt output corresponded to deliveries.
1419 (Suppressing duplicates isn't a good idea as you lose the information
1420 about possibly different redirections that led to the duplicates.)
1422 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
1423 systems where poll() doesn't work, in particular OS X.
1425 PH/10 Added more information to debugging output for retry time not reached.
1427 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
1428 operations in malware.c.
1430 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
1433 PH/13 If write_rejectlog was set false when logging was sent to syslog with
1434 syslog_duplication set false, log lines that would normally be written
1435 both the the main log and to the reject log were not written to syslog at
1438 PH/14 In the default configuration, change the use of "message" in ACL warn
1439 statements to "add_header".
1441 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
1442 not followed by a command (e.g. "seen endif").
1444 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
1445 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
1448 PH/17 Added extra conditions to the default value of delay_warning_condition
1452 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
1453 { match{$h_precedence:}{(?i)bulk|list|junk} } \
1454 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
1457 The Auto-Submitted: and various List- headers are standardised, whereas I
1458 don't think Precedence: ever was.
1460 PH/18 Refactored debugging code in route_finduser() to show more information,
1461 in particular, the error code if getpwnam() issues one.
1463 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
1464 This is apparently needed in addition to the PH/07 change above to avoid
1465 any possible encoding problems.
1467 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
1468 but not after initializing Perl.
1470 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
1471 output them only if debugging. By default they are written stderr,
1472 apparently, which is not desirable.
1474 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
1477 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
1480 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
1482 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
1483 authenticated or an ident call has been made. Suppress the default
1484 values for $authenticated_id and $authenticated_sender (but permit -oMai
1485 and -oMas) when testing with -bh.
1487 PH/24 Re-jigged the order of the tests in the default configuration so that the
1488 tests for valid domains and recipients precede the DNS black list and CSA
1489 tests, on the grounds that those ones are more expensive.
1491 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
1492 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
1493 This bug exists in every version of Exim that I still have, right back to
1496 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
1497 However, an attempt to turn on case-sensitivity in a regex key by
1498 including (?-i) didn't work because the subject string was already
1499 lowercased, and the effects were non-intuitive. It turns out that a
1500 one-line patch can be used to allow (?-i) to work as expected.
1506 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
1507 other effects) broke the use of negated acl sub-conditions.
1509 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
1512 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
1513 "Deny" causes Exim to reject the incoming connection with a 554 error.
1514 Unfortunately, if there is a major crisis, such as a disk failure,
1515 tcp-wrappers gives "deny", whereas what one would like would be some
1516 kind of temporary error. A kludge has been added to help with this.
1517 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
1518 554 error is used if errno is still zero or contains ENOENT (which occurs
1519 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
1522 PH/03 Add -lutil to the default FreeBSD LIBS setting.
1524 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
1525 errors. Otherwise a message that provokes a temporary error (when other
1526 messages do not) can cause a whole host to time out.
1528 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
1529 addresses were routed directly to files or pipes from a redirect router.
1530 File deliveries just didn't batch; pipe deliveries might have suffered
1533 PH/06 A failure to get a lock for a hints database would erroneously always say
1534 "Failed to get write lock", even when it was really a read lock.
1536 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
1537 of 0600. This has been changed to use the value of the lockfile_mode
1538 option (which defaults to 0600).
1540 PH/08 Applied small patch from the Sieve maintainer.
1542 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
1543 folder from quota calculations, a direct delivery into this folder messed
1544 up the contents of the maildirsize file. This was because the regex was
1545 used only to exclude .Trash (or whatever) when the size of the mailbox
1546 was calculated. There was no check that a delivery was happening into an
1547 excluded directory. This bug has been fixed by ignoring all quota
1548 processing for deliveries into excluded directories.
1550 PH/10 Added the maildirfolder_create_regex option to appendfile.
1556 PH/01 The code for finding all the local interface addresses on a FreeBSD
1557 system running IPv6 was broken. This may well have applied to all BSD
1558 systems, as well as to others that have similar system calls. The broken
1559 code found IPv4 interfaces correctly, but gave incorrect values for the
1560 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
1561 that it would not match correctly against @[] and not recognize the IPv6
1564 PH/02 The ipliteral router was not recognizing addresses of the form user@
1565 [ipv6:....] because it didn't know about the "ipv6:" prefix.
1567 PH/03 Added disable_ipv6.
1569 PH/04 Changed $reply_address to use the raw form of the headers instead of the
1570 decoded form, because it is most often used to construct To: headers
1571 lines in autoreplies, and the decoded form may well be syntactically
1572 invalid. However, $reply_address has leading white space removed, and all
1573 newlines turned into spaces so that the autoreply transport does not
1576 PH/05 If group was specified without a user on a router, and no group or user
1577 was specified on a transport, the group from the router was ignored.
1579 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
1580 for visible compile-time settings that can be used to change these
1581 numbers, for those that want even more. Backwards compatibility with old
1582 spool files has been maintained. However, going back to a previous Exim
1583 release will lost any variables that are in spool files.
1585 PH/07 Two small changes when running in the test harness: increase delay when
1586 passing a TCP/IP connection to a new process, in case the original
1587 process has to generate a bounce, and remove special handling of
1588 127.0.0.2 (sic), which is no longer necessary.
1590 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
1591 be the same on different OS.
1593 PH/09 Moved a debug statement in filter processing to avoid a race problem when
1596 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
1597 whether --show-vars was specified or not
1599 JJ/02 exipick: Added support for new ACL variable spool format introduced
1602 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
1603 syntactically invalid From: or Reply-to: line, and a filter used this to
1604 generate an autoreply, and therefore failed to obtain an address for the
1605 autoreply, Exim could try to deliver to a non-existent relative file
1606 name, causing unrelated and misleading errors. What now happens is that
1607 it logs this as a hard delivery error, but does not attempt to create a
1610 PH/11 The exinext utility has a -C option for testing purposes, but although
1611 the given file was scanned by exinext itself; it wasn't being passed on
1612 when Exim was called.
1614 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
1615 an end-of-file indication when reading a command response.
1617 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
1618 compiled. In many other places in Exim, IPv6 addresses are always
1619 recognized, so I have changed this. It also means that IPv4 domain
1620 literals of the form [IPV4:n.n.n.n] are now always recognized.
1622 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
1623 used if the router is not running as root, for example, when verifying at
1624 ACL time, or when using -bh. The debugging output from this situation was
1625 non-existent - all you got was a failure to exec. I have made two
1628 (a) Failures to set uid/gid, the current directory, or a process leader
1629 in a subprocess such as that created by queryprogram now generate
1630 suitable debugging ouput when -d is set.
1632 (b) The queryprogram router detects when it is not running as root,
1633 outputs suitable debugging information if -d is set, and then runs
1634 the subprocess without attempting to change uid/gid.
1636 PH/15 Minor change to Makefile for building test_host (undocumented testing
1639 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
1640 additional section of a DNS packet that returns MX or SRV records.
1641 Instead, it always explicitly searches for A/AAAA records. This avoids
1642 major problems that occur when a DNS server includes only records of one
1643 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
1644 fixed another bug: if SRV records were looked up and the corresponding
1645 address records were *not* found in the additional section, the port
1646 values from the SRV records were lost.
1648 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
1649 using the correct key (the original address) when searching the retry
1650 rules in order to find which one to use for generating the retry hint.
1652 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
1653 adding the default one. Similarly, if it contains a Reply-To: header, the
1654 errors_reply_to option, if set, is not used.
1656 PH/19 When calculating a retry time, Exim used to measure the "time since
1657 failure" by looking at the "first failed" field in the retry record. Now
1658 it does not use this if it is later than than the arrival time of the
1659 message. Instead it uses the arrival time. This makes for better
1660 behaviour in cases where some deliveries succeed, thus re-setting the
1661 "first failed" field. An example is a quota failure for a huge message
1662 when small messages continue to be delivered. Without this change, the
1663 "time since failure" will always be short, possible causing more frequent
1664 delivery attempts for the huge message than are intended.
1665 [Note: This change was subsequently modified - see PH/04 for 4.62.]
1667 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
1668 $1, $2, $3) because the numerical variables can be reset during some
1669 expansion items (e.g. "match"), thereby losing the authentication data.
1671 PH/21 Make -bV show the size of off_t variables so that the test suite can
1672 decide whether to run tests for quotas > 2G.
1674 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
1675 mailbox_size, and mailbox_filecount in the appendfile transport. If a
1676 filecount value is greater than 2G or if a quota value is greater than 2G
1677 on a system where the size of off_t is not greater than 4, a panic error
1680 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
1681 never match. The debug and -bh output now contains an explicit error
1682 message indicating a malformed IPv4 address or mask.
1684 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
1685 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
1686 PH/23 above applies.
1688 PH/25 Do not write to syslog when running in the test harness. The only
1689 occasion when this arises is a failure to open the main or panic logs
1690 (for which there is an explicit test).
1692 PH/26 Added the /no_tell option to "control=freeze".
1694 PH/27 If a host name lookup failed very early in a connection, for example, if
1695 the IP address matched host_lookup and the reverse lookup yielded a name
1696 that did not have a forward lookup, an error message of the form "no IP
1697 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
1698 could be logged. Now it outputs the IP address instead of "NULL".
1700 PH/28 An enabling patch from MH: add new function child_open_exim2() which
1701 allows the sender and the authenticated sender to be set when
1702 submitting a message from within Exim. Since child_open_exim() is
1703 documented for local_scan(), the new function should be too.
1705 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
1706 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
1707 results in an empty string is now treated as unset.
1709 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
1711 PH/31 Added sender_verify_fail logging option.
1713 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
1714 needed by RFC 822 but not by RFC 2822 was commented out. I have now
1715 tidied the source and removed it altogether.
1717 PH/33 When a queue run was abandoned because the load average was too high, a
1718 log line was always written; now it is written only if the queue_run log
1719 selector is set. In addition, the log line for abandonment now contains
1720 information about the queue run such as the pid. This is always present
1721 in "start" and "stop" lines but was omitted from the "abandon" line.
1723 PH/34 Omit spaces between a header name and the colon in the error message that
1724 is given when verify = headers_syntax fails (if there are lots of them,
1725 the message gets confusing).
1727 PH/35 Change the default for dns_check_names_pattern to allow slashes within
1728 names, as there are now some PTR records that contain slashes. This check
1729 is only to protect against broken name servers that fall over on strange
1730 characters, so the fact that it applies to all lookups doesn't matter.
1732 PH/36 Now that the new test suite is complete, we can remove some of the
1733 special code in Exim that was needed for the old test suite. For example,
1734 sorting DNS records because real resolvers return them in an arbitrary
1735 order. The new test suite's fake resolver always returns records in the
1738 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
1739 bounces) except when queue_only is set, to avoid logging races between
1740 the different processes.
1742 PH/38 Panic-die if .include specifies a non-absolute path.
1744 PH/39 A tweak to the "H" retry rule from its user.
1746 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
1747 a label. They prevented compilation on older perls.
1749 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
1750 a warning to be raised on newish perls.
1752 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
1753 on queue. Changes to match documented behaviour of showing count of
1754 messages matching specified criteria.
1756 PH/40 Changed the default ident timeout from 30s to 5s.
1758 PH/41 Added support for the use of login_cap features, on those BSD systems
1759 that have them, for controlling the resources used by pipe deliveries.
1761 PH/42 The content-scanning code uses fopen() to create files in which to put
1762 message data. Previously it was not paying any attention to the mode of
1763 the files. Exim runs with umask(0) because the rest of the code creates
1764 files with open(), and sets the required mode explicitly. Thus, these
1765 files were ending up world-writeable. This was not a big issue, because,
1766 being within the spool directory, they were not world-accessible. I have
1767 created a function called modefopen, which takes an additional mode
1768 argument. It sets umask(777), creates the file, chmods it to the required
1769 mode, then resets the umask. All the relevant calls to fopen() in the
1770 content scanning code have been changed to use this function.
1772 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
1773 to 24 hours. This avoids potential overflow problems when processing G
1774 and H retry rules. I suspect nobody ever tinkers with this value.
1776 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
1778 PH/45 When the plaintext authenticator is running as a client, the server's
1779 challenges are checked to ensure they are valid base64 strings. By
1780 default, the authentication attempt is cancelled if an invalid string is
1781 received. Setting client_ignore_invalid_base64 true ignores these errors.
1782 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
1783 they are received. Thus, the responses can be made to depend on the
1784 challenges. If an invalid string is ignored, an empty string is placed in
1787 PH/46 Messages that are created by the autoreply transport now contains a
1788 References: header, in accordance with RFCs 2822 and 3834.
1790 PH/47 Added authenticated_sender_force to the smtp transport.
1792 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
1794 PH/49 Installed latest patch from the Sieve maintainer.
1796 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
1797 was also set, the appendfile transport was unnecessarily scanning a
1798 directory of message files (e.g. for maildir delivery) to find the count
1799 of files (along with the size), even though it did not need this
1800 information. It now does the scan only if it needs to find either the
1801 size of the count of files.
1803 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
1805 PH/52 Two bugs concerned with error handling when the smtp transport is
1808 (i) Exim was not creating retry information for temporary errors given
1809 for individual recipients after the DATA command when the smtp transport
1810 was used in LMTP mode. This meant that they could be retried too
1811 frequently, and not timed out correctly.
1813 (ii) Exim was setting the flag that allows error details to be returned
1814 for LMTP errors on RCPT commands, but not for LMTP errors for individual
1815 recipients that were returned after the DATA command.
1817 PH/53 This is related to PH/52, but is more general: for any failing address,
1818 when detailed error information was permitted to be returned to the
1819 sender, but the error was temporary, then after the final timeout, only
1820 "retry timeout exceeded" was returned. Now it returns the full error as
1821 well as "retry timeout exceeded".
1823 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
1824 do this, and (what is worse) MTAs that accept it.
1826 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
1827 will now be deprecated.
1829 PH/56 New os.c-cygwin from the Cygwin maintainer.
1831 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
1832 formats (previously only available in exim formats via -bpr, -bpru,
1833 and -bpra. Now also available in native and exiqgrep formats)
1835 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
1836 with very large, slow to parse queues
1838 JJ/08 exipick: added ! as generic prefix to negate any criteria format
1840 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
1842 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
1843 responses to authentication challenges, though it was showing the
1844 challenges; (ii) I've removed the CR characters from the debug output for
1847 PH/58 Allow for the insertion of a newline as well as a space when a string
1848 is turned into more than one encoded-word during RFC 2047 encoding. The
1849 Sieve code now uses this.
1851 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
1852 data_4xx, lost_connection, tls_required.
1854 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
1855 message was being sent as an SMTP response.
1857 PH/61 Add -l and -k options to exicyclog.
1859 PH/62 When verifying, if an address was redirected to one new address, so that
1860 verification continued, and the new address failed or deferred after
1861 having set something in $address_data, the value of $address_data was not
1862 passed back to the ACL. This was different to the case when no
1863 redirection occurred. The value is now passed back in both cases.
1865 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
1866 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
1867 use login_cap.h, so on its own it isn't the distinguishing feature. The
1868 new name refers directly to the setclassresources() function.
1870 PH/65 Added configuration files for NetBSD3.
1872 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
1874 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
1875 is preferred over IPv4.
1877 PH/68 The bounce_return_message and bounce_return_body options were not being
1878 honoured for bounces generated during the reception of non-SMTP messages.
1879 In particular, this applied to messages rejected by the ACL. This bug has
1880 been fixed. However, if bounce_return_message is true and bounce_return_
1881 body is false, the headers that are returned for a non-SMTP message
1882 include only those that have been read before the error was detected.
1883 (In the case of an ACL rejection, they have all been read.)
1885 PH/69 The HTML version of the specification is now built in a directory called
1886 spec_html instead of spec.html, because the latter looks like a path with
1887 a MIME-type, and this confuses some software.
1889 PH/70 Catch two compiler warnings in sieve.c.
1891 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
1892 function verify_get_ident() calls ip_connect() to connect a socket, but
1893 if the "connect()" function timed out, ip_connect() used to close the
1894 socket. However, verify_get_ident() also closes the socket later, and in
1895 between Exim writes to the log, which may get opened at this point. When
1896 the socket was closed in ip_connect(), the log could get the same file
1897 descriptor number as the socket. This naturally causes chaos. The fix is
1898 not to close the socket in ip_connect(); the socket should be closed by
1899 the function that creates it. There was only one place in the code where
1900 this was missing, in the iplookup router, which I don't think anybody now
1901 uses, but I've fixed it anyway.
1903 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
1904 well as to direct DNS lookups. Otherwise the handling of names in host
1905 lists is inconsistent and therefore confusing.
1911 PH/01 Two changes to the default runtime configuration:
1913 (1) Move the checks for relay_from_hosts and authenticated clients from
1914 after to before the (commented out) DNS black list checks.
1916 (2) Add control=submission to the relay_from_hosts and authenticated
1917 clients checks, on the grounds that messages accepted by these
1918 statements are most likely to be submissions.
1920 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
1922 (1) Generate an error if the third argument for the ${prvs expansion is
1925 (2) Treat a missing third argument of ${prvscheck as if it were an empty
1928 (3) Reset the variables that are obtained from the first argument of
1929 ${prvscheck and used in the second argument before leaving the code,
1930 because their memory is reclaimed, so using them afterwards may do
1933 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
1934 one (it's much easier than Tom thought :-).
1936 (5) Because of (4), we can now allow for the use of $prvscheck_result
1937 inside the third argument.
1939 PH/03 For some reason, the default setting of PATH when running a command from
1940 a pipe transport was just "/usr/bin". I have changed it to
1943 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
1944 anything to be listed in the output from -bV.
1946 PH/05 When a filter generated an autoreply, the entire To: header line was
1947 quoted in the delivery log line, like this:
1949 => >A.N.Other <ano@some.domain> <original@ddress> ...
1951 This has been changed so that it extracts the operative address. There
1952 may be more than one such address. If so, they are comma-separated, like
1955 => >ano@some.domain,ona@other.domain <original@ddress> ...
1957 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
1958 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
1959 not being looked up in the rDNS to get a host name, Exim was showing the
1960 IP address twice in Received: lines, even though the IP addresses were
1961 identical. For example:
1963 Received: from [1.2.3.4] (helo=[1.2.3.4])
1965 However, if the real host name was known, it was omitting the HELO data
1966 if it matched the actual IP address. This has been tidied up so that it
1967 doesn't show the same IP address twice.
1969 PH/07 When both +timestamp and +memory debugging was on, the value given by
1970 $tod_xxx expansions could be wrong, because the tod_stamp() function was
1971 called by the debug printing, thereby overwriting the timestamp buffer.
1972 Debugging no longer uses the tod_stamp() function when +timestamp is set.
1974 PH/08 When the original message was included in an autoreply transport, it
1975 always said "this is a copy of the message, including all the headers",
1976 even if body_only or headers_only was set. It now gives an appropriate
1979 PH/09 Applied a patch from the Sieve maintainer which:
1981 o fixes some comments
1982 o adds the (disabled) notify extension core
1983 o adds some debug output for the result of if/elsif tests
1984 o points to the current vacation draft in the documentation
1985 and documents the missing references header update
1989 o fixes a bug in processing the envelope test (when testing
1990 multiple envelope elements, the last element determinted the
1993 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
1994 Electronic Mail") by including:
1996 Auto-submitted: auto-generated
1998 in the messages that it generates (bounce messages and others, such as
1999 warnings). In the case of bounce messages for non-SMTP mesages, there was
2000 also a typo: it was using "Auto_submitted" (underscore instead of
2001 hyphen). Since every message generated by Exim is necessarily in response
2002 to another message, thes have all been changed to:
2004 Auto-Submitted: auto-replied
2006 in accordance with these statements in the RFC:
2008 The auto-replied keyword:
2010 - SHOULD be used on messages sent in direct response to another
2011 message by an automatic process,
2013 - MUST NOT be used on manually-generated messages,
2015 - MAY be used on Delivery Status Notifications (DSNs) and Message
2016 Disposition Notifications (MDNs),
2018 - MUST NOT be used on messages generated by automatic or periodic
2019 processes, except for messages which are automatic responses to
2022 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
2023 to the default Received: header definition.
2025 PH/12 Added log selector acl_warn_skipped (default on).
2027 PH/13 After a successful wildlsearch lookup, discard the values of numeric
2028 variables because (a) they are in the wrong storage pool and (b) even if
2029 they were copied, it wouldn't work properly because of the caching.
2031 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
2032 checking when decoding. Apparently there are clients that generate
2033 overlong encoded strings. Why am I not surprised?
2035 PH/15 If the first argument of "${if match_address" was not empty, but did not
2036 contain an "@" character, Exim crashed. Now it writes a panic log message
2037 and treats the condition as false.
2039 PH/16 In autoreply, treat an empty string for "once" the same as unset.
2041 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
2042 extension "envelope-auth". The code is finished and in agreement with
2043 other implementations, but there is no documentation so far and in fact,
2044 nobody wrote the draft yet. This extension is currently #undef'ed, thus
2045 not changing the active code.
2047 Print executed "if" and "elsif" statements when debugging is used. This
2048 helps a great deal to understand what a filter does.
2050 Document more things not specified clearly in RFC3028. I had all this
2051 sorted out, when out of a sudden new issues came to my mind. Oops."
2053 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
2056 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
2057 canonical form (as documented). However, after a host name lookup from
2058 the IP address, check_host() was doing a simple string comparison with
2059 addresses acquired from the DNS when checking that the found name did
2060 have the original IP as one of its addresses. Since any found IPv6
2061 addresses are likely to be in abbreviated form, the comparison could
2062 fail. Luckily, there already exists a function for doing the comparison
2063 by converting both addresses to binary, so now that is used instead of
2064 the text comparison.
2066 PH/20 There was another similar case to PH/19, when a complete host name was
2067 given in a host list; looking up its IP address could give an abbreviated
2068 form, whereas the current host's name might or might not be abbreviated.
2069 The same fix has been applied.
2075 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
2076 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
2079 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
2080 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
2082 PH/03 Typo: missing ".o" in src/pcre/Makefile.
2084 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
2085 header line, restrict the check to what is listed in RFCs 2369 and 2929.
2086 Also, for "Auto-Submitted", treat anything other than "no" as
2087 non-personal, in accordance with RFC 3834. (Previously it treated
2088 anything starting "auto-" as non-personal.)
2090 TF/01 The control=submission/name=... option had a problem with syntax
2091 errors if the name included a slash character. The /name= option
2092 now slurps the rest of the string, so it can include any characters
2093 but it must come last in the list of options (after /sender_retain
2096 PH/05 Some modifications to the interface to the fake nameserver for the new
2104 TK/01 Added the "success_on_redirect" address verification option. See
2105 NewStuff for rationale and an example.
2107 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
2109 PH/02 Patch to exigrep to allow it to work on syslog lines.
2111 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
2112 fread() to skip over the body file's header line, because in Cygwin the
2113 header line is locked and is inaccessible.
2115 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
2116 co-exist for some time) to make it clear that it is the Exim ID that is
2117 referenced, not the Message-ID: header line.
2119 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
2120 string_format() function, because snprintf() does not exist on all
2123 PH/06 The use of forbid_filter_existstest now also locks out the use of the
2124 ${stat: expansion item.
2126 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
2127 protocol synchronization error", to keep the pedants happy.
2129 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
2130 well as for IRIX systems, when gcc is being used. See the host.c source
2133 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
2135 PH/10 Named domain lists were not working if used in a queue_smtp_domains
2138 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
2139 transport and to the smtp transport in LMTP mode.
2141 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
2143 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
2144 run a filter in a subprocess. This could lead to confusion in subsequent
2145 lookups in the parent process. There should also be a search_tidyup() at
2146 the end of the subprocess.
2148 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
2149 only if the host matched helo_try_verify_hosts, which caused the
2150 verification to occur when the EHLO/HELO command was issued. The ACL just
2151 tested the remembered result. Now, if a previous verification attempt has
2152 not happened, "verify = helo" does it there and then.
2154 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
2156 TK/03 Fix log output including CR from clamd.
2158 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
2159 exist provoked a memory error which could cause a segfault.
2161 PH/15 Installed PCRE 6.2
2163 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
2165 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
2166 of the problem. Specifically, suggested +O2 rather than +O1 for the
2169 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
2171 PH/20 If a delivery was routed to a non-standard port by means of an SRV
2172 record, the port was not correctly logged when the outgoing_port log
2173 selector was set (it logged the transort's default port).
2175 PH/21 Added support for host-specific ports to manualroute, queryprogram,
2176 fallback_hosts, and "hosts" in the smtp transport.
2178 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
2179 host errors such as "Connection refused".
2181 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
2182 authentication with radiusclient 0.4.9:
2184 - Error returned from rc_read_config was caught wrongly
2185 - Username/password not passed on to radius server due to wrong length.
2187 The presumption is that some radiusclient API changes for 4.51/PH/17
2188 were not taken care of correctly. The code is still untested by me (my
2189 Linux distribution still has 0.3.2 of radiusclient), but it was
2190 contributed by a Radius user.
2192 PH/24 When doing a callout, the value of $domain wasn't set correctly when
2193 expanding the "port" option of the smtp transport.
2195 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
2196 while reading a MIME header. Thanks to Tom Hughes for a patch.
2198 PH/24 Include config.h inside local_scan.h so that configuration settings are
2201 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
2202 that in an ACL for RCPT (for example), you can examine exactly what was
2205 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
2206 commands, but it was not correctly comparing the address with the actual
2207 client host address. Thus, it would show the EHLO address in Received:
2208 header lines when this was not necessary.
2210 PH/27 Added the % operator to ${eval:}.
2212 PH/28 Exim tries to create and chdir to its spool directory when it starts;
2213 it should be ignoring failures (because with -C, for example, it has lost
2214 privilege). It wasn't ignoring creation failures other than "already
2217 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
2220 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
2221 because an input line was too long, either on its own, or by virtue of
2222 too many continuations, the temporary file was not being removed, and the
2223 return code was incorrect.
2225 PH/31 Missing "BOOL" in function definition in filtertest.c.
2227 PH/32 Applied Sieve patches from the maintainer.
2229 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
2231 PH/33 Added "verify = not_blind".
2233 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
2234 Local/Makefile (with some defaults set). These are used in built scripts
2235 such as exicyclog, but they have never been used in the exim_install
2236 script (though there are many overriding facilities there). I have
2237 arranged that the exim_install script now takes note of these two
2240 PH/35 Installed configuration files for Dragonfly.
2242 PH/36 When a locally submitted message by a trusted user did not contain a
2243 From: header, and the sender address was obtained from -f or from an SMTP
2244 MAIL command, and the trusted user did not use -F to supply a sender
2245 name, $originator_name was incorrectly used when constructing a From:
2246 header. Furthermore, $originator_name was used for submission mode
2247 messages from external hosts without From: headers in a similar way,
2248 which is clearly wrong.
2250 PH/37 Added control=suppress_local_fixups.
2252 PH/38 When log_selector = +received_sender was set, and the addition of the
2253 sender made the log line's construction buffer exactly full, or one byte
2254 less than full, an overflow happened when the terminating "\n" was
2257 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
2258 when the result of a list match is failure because a DNS lookup failed.
2260 PH/40 RM_COMMAND is now used in the building process.
2262 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
2263 the "build-* directories that it finds.
2265 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
2266 address in a domain literal was a prefix of an interface address.
2268 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
2269 when verifying a sender address, unless rewrite_headers is false.
2271 PH/44 Wrote a long comment about why errors_to addresses are verified as
2272 recipients, not senders.
2274 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
2275 the ratelimit ACL was added.
2277 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
2279 PH/46 Added extra information about PostgreSQL errors to the error string.
2281 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
2282 avoiding the need to install special zones in a real server. This is
2283 backwards compatible; if it can't find the fake resolver, it drops back.
2284 Thus, both old and new test suites can be run.
2286 TF/02 Added util/ratelimit.pl
2288 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
2289 clock is set back in time.
2291 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
2292 Candler <B.Candler@pobox.com>.
2294 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
2295 OK for addresses that are the result of redirections.
2297 PH/48 A number of further additions for the benefit of the new test suite,
2298 including a fake gethostbyname() that interfaces to the fake DNS resolver
2301 TF/06 The fix for widen_domains has also been applied to qualify_single and
2302 search_parents which are the other dnslookup options that can cause
2305 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
2308 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
2310 TF/07 Exim produced the error message "an SRV record indicated no SMTP
2311 service" if it encountered an MX record with an empty target hostname.
2312 The message is now "an MX or SRV record indicated no SMTP service".
2314 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
2315 if the DNS of the sending site is misconfigured. This is quite a
2316 common situation. This change restores the behaviour of treating a
2317 helo verification defer as a failure.
2319 PH/51 If self=fail was set on a router, the bounce message did not include the
2320 actual error message.
2326 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
2328 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
2329 command itself ended in error, the underlying message about the transport
2330 filter timeout was being overwritten with the pipe command error. Now the
2331 underlying error message should be appended to the second error message.
2333 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
2335 PH/02 Reduce the amount of output that "make" produces by default. Full output
2336 can still be requested.
2338 PH/03 The warning log line about a condition test deferring for a "warn" verb
2339 was being output only once per connection, rather than after each
2340 occurrence (because it was using the same function as for successful
2341 "warn" verbs). This seems wrong, so I have changed it.
2343 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
2344 it should not have, which might have caused a crash in the right
2345 circumstances, but probably never did.
2347 PH/04 Installed a modified version of Tony Finch's patch to make submission
2348 mode fix the return path as well as the Sender: header line, and to
2349 add a /name= option so that you can make the user's friendly name appear
2352 TF/03 Added the control = fakedefer ACL modifier.
2354 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
2355 Mark Lowes for thorough testing.
2357 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
2359 TK/03 Merged latest SRS patch from Miles Wilton.
2361 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
2362 with the definition in sysexits.h (which is #included earlier).
2363 Fortunately, Exim does not actually use EX_OK. The code used to try to
2364 preserve the sysexits.h value, by assumimg that macro definitions were
2365 scanned for macro replacements. I have been disabused of this notion,
2366 so now the code just undefines EX_OK before #including unistd.h.
2368 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
2369 in the smtp transport. When a block could not be written in a single
2370 write() function, the timeout was being re-applied to each part-write.
2371 This seems wrong - if the receiver was accepting one byte at a time it
2372 would take for ever. The timeout is now adjusted when this happens. It
2373 doesn't have to be particularly precise.
2375 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
2376 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
2378 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
2379 without a domain if the check to <postmaster@domain> fails.
2381 SC/01 Eximstats: added -xls and the ability to specify output files
2382 (patch written by Frank Heydlauf).
2384 SC/02 Eximstats: use FileHandles for outputing results.
2386 SC/03 Eximstats: allow any combination of xls, txt, and html output.
2388 SC/04 Eximstats: fixed display of large numbers with -nvr option
2390 SC/05 Eximstats: fixed merging of reports with empty tables.
2392 SC/06 Eximstats: added the -include_original_destination flag
2394 SC/07 Eximstats: removed tabs and trailing whitespace.
2396 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
2398 TK/06 MBOX spool code: Add real "From " MBOX separator line
2399 so the .eml file is really in mbox format (even though
2400 most programs do not really care). Patch from Alex Miller.
2402 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
2403 The latter is generated from $received_to and is only set if the
2404 message has one envelope recipient. SA can use these headers,
2405 obviously out-of-the-box. Patch from Alex Miller.
2407 PH/08 The ${def test on a variable was returning false if the variable's
2408 value was "0", contrary to what the specification has always said!
2409 The result should be true unless the variable is empty.
2411 PH/09 The syntax error of a character other than { following "${if
2412 def:variable_name" (after optional whitespace) was not being diagnosed.
2413 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
2414 accidental colon was present, for example, could give incorrect results.
2416 PH/10 Tidied the code in a number of places where the st_size field of a stat()
2417 result is used (not including appendfile, where other changes are about
2420 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
2421 This involved changing a lot of size variables from int to off_t. It
2422 should work with maildirs and everything.
2424 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
2425 spamd dying while we are connected to it.
2427 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
2430 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
2431 function for BDB 4.3.
2433 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
2435 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
2436 the sources for PCRE have been re-arranged and more files are now
2439 PH/15 The code I had for printing potentially long long variables in PH/11
2440 above was not the best (it lost precision). The length of off_t variables
2441 is now inspected at build time, and an appropriate printing format (%ld
2442 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
2443 to be "long long int" or "long int". This is needed for the internal
2444 formatting function string_vformat().
2446 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
2447 the configuration file to be ":syslog", then the script "guesses" where
2448 the logs files are, rather than using the compiled in default. In our
2449 case the guess is not the same as the compiled default, so the script
2450 suddenly stopped working when I started to use syslog. The patch checks
2451 to see if log_file_path is "". If so, it attempts to read it from exim
2452 with no configuration file to get the compiled in version, before it
2453 falls back to the previous guessing code."
2455 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
2456 implementing BATV in an Exim configuration. See NewStuff for the gory
2459 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
2460 Makefile that are specific to HP-UX.
2462 PH/18 If the "use_postmaster" option was set for a recipient callout together
2463 with the "random" option, the postmaster address was used as the MAIL
2464 FROM address for the random test, but not for the subsequent recipient
2465 test. It is now used for both.
2467 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
2468 patch removes a few documentation additions to RFC 3028, because the
2469 latest draft now contains them. It adds the new en;ascii-case comparator
2470 and a new error check for 8bit text in MIME parts. Comparator and
2471 require names are now matched exactly. I enabled the subaddress
2472 extension, but it is not well tested yet (read: it works for me)."
2474 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
2475 rework some of the code of TK/09 above to avoid the hardwired use of
2476 "%lld" and "long long". Replaced the call to snprintf() with a call to
2479 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
2480 records point to non-existent hosts", "retry timeout exceeded", and
2481 "retry time not reached for any host after a long failure period".
2483 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
2484 experimental DomainKeys support:
2486 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
2487 (2) On an error such as an illegally used "control", the wrong name for
2488 the control was given.
2490 These problems did NOT occur unless DomainKeys support was compiled.
2492 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
2494 PH/24 Added ${if match_ip condition.
2496 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
2497 databases so that it will be absolutely obvious if a crash occurs in the
2498 DB library. This is a regular occurrence (often caused by mis-matched
2501 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
2502 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
2503 picked up on a user's system that detects such things. There doesn't seem
2504 to be a gcc warning option for this - only an attribute that has to be
2505 put on the function's prototype. It seems that in Fedora Core 4 they have
2506 set this on a number of new functions. No doubt there will be more in due
2509 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
2510 specify a transport. However, if an address that was verified by such a
2511 router was the subject of a callout, Exim crashed because it tried to
2512 read the rcpt_include_affixes from the non-existent transport. Now it
2513 just assumes that the setting of that option is false. This bug was
2514 introduced by 4.51/PH/31.
2516 PH/28 Changed -d+all to exclude +memory, because that information is very
2517 rarely of interest, but it makes the output a lot bigger. People tend to
2518 do -d+all out of habit.
2520 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
2521 code in os-type was giving problems when libc.so lives in lib64, like on
2524 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
2525 aren't the modern standard, and it seems that some systems' include files
2526 don't always have them. Exim was already checking for some of the newer
2527 ones like T_AAAA, and defining it itself. I've added checks for all the
2528 record types that Exim uses.
2530 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
2531 not automatically generating a new one, as it is supposed to. This
2532 prevented TLS from working. If the file did exist, but contained invalid
2533 data, a new version was generated, as expected. It was only the case of a
2534 non-existent file that was broken.
2536 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
2537 with a change in libdomainkeys > 0.64.
2539 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
2540 from DNS. If the selector record carries the flag, it now has
2541 precedence over the domain-wide flag.
2543 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
2545 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
2546 the use of a port name that isn't defined in /etc/services) occurred, the
2547 message was deferred as in a normal delivery, and thus remained on the
2548 spool, instead of being failed because of the mua_wrapper setting. This
2549 is now fixed, and I tidied up some of the mua_wrapper messages at the
2552 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
2553 the messages in a hash of arrays rather than using individual hashes.
2554 This is a bit cleaner and results in dramatic memory savings, albeit
2555 at a slight CPU cost.
2557 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
2558 as requested by Marc Sherman.
2560 SC/10 Eximstats: added histograms for user specified patterns as requested
2563 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
2565 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
2566 fopen() in the content-scanning modules that did not already have it.
2572 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
2573 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
2575 TK/02 Fix ACL "control" statement not being available in MIME ACL.
2577 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
2579 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
2580 to test Sieve filters that use "vacation".
2582 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
2583 that changes the way the GnuTLS parameters are stored in the cache file.
2584 The new format can be generated externally. For backward compatibility,
2585 if the data in the cache doesn't make sense, Exim assumes it has read an
2586 old-format file, and it generates new data and writes a new file. This
2587 means that you can't go back to an older release without removing the
2590 PH/03 A redirect router that has both "unseen" and "one_time" set does not
2591 work if there are any delivery delays because "one_time" forces the
2592 parent to be marked "delivered", so its unseen clone is never tried
2593 again. For this reason, Exim now forbids the simultaneous setting of
2596 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
2597 redirected to themselves ("homonym" addresses). Read the long ChangeLog
2598 entry if you want to know the details. The fix, however, neglected to
2599 consider the case when local delivery batching is involved. The test for
2600 "previously delivered" was not happening when checking to see if an
2601 address could be batched with a previous (undelivered) one; under
2602 certain circumstances this could lead to multiple deliveries to the same
2605 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
2606 in its include files, and this causes problems building Exim.
2608 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
2609 header_syntax) but Exim was just ignoring anything given after a slash.
2610 In particular, this caused confusion with an attempt to use "verify =
2611 reverse_host_lookup/defer_ok". An error is now given when options are
2612 supplied for verify items that do not have them. (Maybe reverse_host_
2613 lookup should have a defer_ok option, but that's a different point.)
2615 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
2616 defined by RFC 821) to 2048, because there were problems with some AUTH
2617 commands, and RFC 1869 says the size should be increased for extended
2618 SMTP commands that take arguments.
2620 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
2623 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
2624 "unknown" error; now it says that the functionality isn't in the binary.
2626 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
2627 an address' error message when a string expansion fails (syntax or
2628 whatever). Otherwise the password may appear in the log. Following change
2629 PH/42 below, there is no longer a chance of it appearing in a bounce
2632 PH/11 Installed exipick version 20050225.0 from John Jetmore.
2634 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
2635 of its addresses was ever tried. (Bugzilla bug #2.)
2637 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
2638 the result incorrectly in the debug output. (It correctly added a newline
2639 to what was transported.)
2641 TF/01 Added $received_time.
2643 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
2644 commented out examples of how to interface to a virus scanner and to
2645 SpamAssassin. Also added commented examples of av_scanner and
2646 spamd_address settings.
2648 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
2649 and controls are allowed in which ACLs. There were a couple of minor
2650 errors. Some of the entries in the conditions table (which is a table of
2651 where they are NOT allowed) were getting very unwieldy; rewrote them as a
2652 negation of where the condition IS allowed.
2654 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
2656 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
2657 header file does not have a version number, so I've had to invent a new
2658 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
2659 API. The code is untested by me (my Linux distribution still has 0.3.2 of
2660 radiusclient), but it was contributed by a Radius user.
2662 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
2663 files or directories, for OpenSSL.
2665 PH/19 When an Exim process that is running as root has to create an Exim log
2666 file, it does so in a subprocess that runs as exim:exim so as to get the
2667 ownership right at creation (otherwise, other Exim processes might see
2668 the file with the wrong ownership). There was no test for failure of this
2669 fork() call, which would lead to the process getting stuck as it waited
2670 for a non-existent subprocess. Forks do occasionally fail when resources
2671 run out. I reviewed all the other calls to fork(); they all seem to check
2674 PH/20 When checking for unexpected SMTP input at connect time (before writing
2675 the banner), Exim was not dealing correctly with a non-positive return
2676 from the read() function. If the client had disconnected by this time,
2677 the result was a log entry for a synchronization error with an empty
2678 string after "input=" when read() returned zero. If read() returned -1
2679 (an event I could not check), uninitialized data bytes were printed.
2680 There were reports of junk text (parts of files, etc) appearing after
2683 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
2685 PH/22 Added support for macro redefinition, and (re)definition in between
2686 driver and ACL definitions.
2688 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
2689 forgetting to use the resulting value; it was using the unexpanded value.
2691 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
2692 hadn't been configured. The fix is from Juergen Kreileder, who
2693 understands it better than I do:
2695 "Here's what I see happening with three configured cyrus_sasl
2696 authenticators configured (plain, login, cram-md5):
2698 On startup auth_cyrus_sasl_init() gets called for each of these.
2699 This means three calls to sasl_listmech() without a specified mech_list.
2700 => SASL tests which mechs of all available mechs actually work
2701 => three warnings about OTP not working
2702 => the returned list contains: plain, login, cram-md5, digest-md5, ...
2704 With the patch, sasl_listmech() also gets called three times. But now
2705 SASL's mech_list option is set to the server_mech specified in the the
2706 authenticator. Or in other words, the answer from sasl_listmech()
2707 gets limited to just the mech you're testing for (which is different
2709 => the return list contains just 'plain' or 'login', 'cram-md5' or
2710 nothing depending on the value of ob->server_mech.
2712 I've just tested the patch: Authentication still works fine,
2713 unavailable mechs specified in the exim configuration are still
2714 caught, and the auth.log warnings about OTP are gone."
2716 PH/25 When debugging is enabled, the contents of the command line are added
2717 to the debugging output, even when log_selector=+arguments is not
2720 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
2721 answer is "GNU", and only if the return is "GNU/something" is the answer
2724 PH/27 $acl_verify_message is now set immediately after the failure of a
2725 verification in an ACL, and so is available in subsequent modifiers. In
2726 particular, the message can be preserved by coding like this:
2728 warn !verify = sender
2729 set acl_m0 = $acl_verify_message
2731 Previously, $acl_verify_message was set only while expanding "message"
2732 and "log_message" when a very denied access.
2734 PH/28 Modified OS/os.c-Linux with
2736 -#ifndef OS_LOAD_AVERAGE
2737 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
2739 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
2740 nomenclature these days.)
2742 PH/29 Installed patch from the Sieve maintainer that adds the options
2743 sieve_useraddress and sieve_subaddress to the redirect router.
2745 PH/30 In these circumstances:
2746 . Two addresses routed to the same list of hosts;
2747 . First host does not offer TLS;
2748 . First host accepts first address;
2749 . First host gives temporary error to second address;
2750 . Second host offers TLS and a TLS session is established;
2751 . Second host accepts second address.
2752 Exim incorrectly logged both deliveries with the TLS parameters (cipher
2753 and peerdn, if requested) that were in fact used only for the second
2756 PH/31 When doing a callout as part of verifying an address, Exim was not paying
2757 attention to any local part prefix or suffix that was matched by the
2758 router that accepted the address. It now behaves in the same way as it
2759 does for delivery: the affixes are removed from the local part unless
2760 rcpt_include_affixes is set on the transport.
2762 PH/32 Add the sender address, as F=<...>, to the log line when logging a
2763 timeout during the DATA phase of an incoming message.
2765 PH/33 Sieve envelope tests were broken for match types other than :is. I have
2766 applied a patch sanctioned by the Sieve maintainer.
2768 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
2769 the uid or gid is negative. A case of a negative gid caused this to be
2770 noticed. The fix allows for either to be negative.
2772 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
2773 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
2776 PH/36 The change PH/12 above was broken. Fixed it.
2778 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
2779 the grounds that routing the same address twice would always produce the
2780 same answer. This might have been true once, but it is certainly no
2781 longer true now. Routing a child address may depend on the previous
2782 routing that produced that child. Some complicated redirection strategies
2783 went wrong when messages had multiple recipients, and made Exim's
2784 behaviour dependent on the order in which the addresses were given.
2786 I have moved the duplicate checking until after the routing is complete.
2787 Exim scans the addresses that are assigned to local and remote
2788 transports, and removes any duplicates. This means that more work will be
2789 done, as duplicates will always all be routed, but duplicates are
2790 presumably rare, so I don't expect this is of any significance.
2792 For deliveries to pipes, files, and autoreplies, the duplicate checking
2793 still happens during the routing process, since they are not going to be
2796 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
2797 It corrects a timeout issue with spamd. This is Ian's comment: "The
2798 background is that sometimes spamd either never reads data from a
2799 connection it has accepted, or it never writes response data. The exiscan
2800 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
2801 blindly assumes that writes won't block so it may never time out."
2803 PH/39 Allow G after quota size as well as K and M.
2805 PH/40 The value set for $authenticated_id in an authenticator may not contain
2806 binary zeroes or newlines because the value is written to log lines and
2807 to spool files. There was no check on this. Now the value is run through
2808 the string_printing() function so that such characters are converted to
2809 printable escape sequences.
2811 PH/41 $message_linecount is a new variable that contains the total number of
2812 lines in the message. Compare $body_linecount, which is the count for the
2815 PH/42 Exim no longer gives details of delivery errors for specific addresses in
2816 bounce and delay warning messages, except in certain special cases, which
2819 (a) An SMTP error message from a remote host;
2820 (b) A message specified in a :fail: redirection;
2821 (c) A message specified in a "fail" command in a system filter;
2822 (d) A message specified in a FAIL return from the queryprogram router;
2823 (e) A message specified by the cannot_route_message router option.
2825 In these cases only, Exim does include the error details in bounce and
2826 warning messages. There are also a few cases where bland messages such
2827 as "unrouteable address" or "local delivery error" are given.
2829 PH/43 $value is now also set for the "else" part of a ${run expansion.
2831 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
2832 being worked on, but at least Exim now implements the latest version to
2835 PH/45 In a pipe transport, although a timeout while waiting for the pipe
2836 process to complete was treated as a delivery failure, a timeout while
2837 writing the message to the pipe was logged, but erroneously treated as a
2838 successful delivery. Such timeouts include transport filter timeouts. For
2839 consistency with the overall process timeout, these timeouts are now
2840 treated as errors, giving rise to delivery failures by default. However,
2841 there is now a new Boolean option for the pipe transport called
2842 timeout_defer, which, if set TRUE, converts the failures into defers for
2843 both kinds of timeout. A transport filter timeout is now identified in
2846 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
2847 systems where "make" and "gmake" are different, calling "gmake" at top
2848 level broke things. I've arranged for the value of $(MAKE) to be passed
2849 from the Makefile to this script so that it can call the same version of
2853 A note about Exim versions 4.44 and 4.50
2854 ----------------------------------------
2856 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
2857 changes of various kinds. As a consequence, a big documentation update was
2858 needed. This delayed the release for rather longer than seemed good, especially
2859 in the light of a couple of (minor) security issues. Therefore, the changes
2860 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
2861 release. So 4.44 and 4.50 are in effect two different branches that both start
2864 I have left the 4.50 change log unchanged; it contains all the changes since
2865 4.43. The change log for 4.44 is below; many of its items are identical to
2866 those for 4.50. This seems to be the most sensible way to preserve the
2867 historical information.
2873 1. Minor wording change to the doc/README.SIEVE file.
2875 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
2876 computation of the current number of files was incorrect.
2878 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
2879 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
2882 4. Give more explanation in the error message when the command for a transport
2883 filter fails to execute.
2885 5. There are several places where Exim runs a non-Exim command in a
2886 subprocess. The SIGUSR1 signal should be disabled for these processes. This
2887 was being done only for the command run by the queryprogram router. It is
2888 now done for all such subprocesses. The other cases are: ${run, transport
2889 filters, and the commands run by the lmtp and pipe transports.
2891 6. Added CONFIGURE_GROUP build-time option.
2893 7. Some older OS have a limit of 256 on the maximum number of file
2894 descriptors. Exim was using setrlimit() to set 1000 as a large value
2895 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
2896 systems. I've change it so that if it can't get 1000, it tries for 256.
2898 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
2899 was an oversight, and furthermore, ever since the addition of extra
2900 controls (e.g. 4.43/32), the checks on when to allow different forms of
2901 "control" were broken. There should now be diagnostics for all cases when a
2902 control that does not make sense is encountered.
2904 9. Added the /retain_sender option to "control=submission".
2906 10. $recipients is now available in the predata ACL (oversight).
2908 11. Tidy the search cache before the fork to do a delivery from a message
2909 received from the command line. Otherwise the child will trigger a lookup
2910 failure and thereby defer the delivery if it tries to use (for example) a
2911 cached ldap connection that the parent has called unbind on.
2913 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
2914 of $address_data from the recipient verification was clobbered by the
2915 sender verification.
2917 13. The value of address_data from a sender verification is now available in
2918 $sender_address_data in subsequent conditions in the ACL statement.
2920 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
2922 15. Added a new option "connect=<time>" to callout options, to set a different
2925 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
2926 was its contents. (It was OK if the option was not defined at all.)
2928 17. A "Completed" log line is now written for messages that are removed from
2929 the spool by the -Mrm option.
2931 18. New variables $sender_verify_failure and $recipient_verify_failure contain
2932 information about exactly what failed.
2934 19. Added -dd to debug only the daemon process.
2936 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
2937 handles timeouts, both on the server side and network timeouts. Renamed the
2938 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
2940 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
2941 It is now set to "smtps".
2943 22. $host_address is now set to the target address during the checking of
2944 ignore_target_hosts.
2946 23. When checking ignore_target_hosts for an ipliteral router, no host name was
2947 being passed; this would have caused $sender_host_name to have been used if
2948 matching the list had actually called for a host name (not very likely,
2949 since this list is usually IP addresses). A host name is now passed as
2952 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
2953 code that specifies a non-restarting handler (typically sigaction() in
2954 modern systems) in an attempt to fix a rare and obscure crash bug.
2956 25. Narrowed the window for a race in the daemon that could cause it to ignore
2957 SIGCHLD signals. This is not a major problem, because they are used only to
2958 wake it up if nothing else does.
2960 26. A malformed maildirsize file could cause Exim to calculate negative values
2961 for the mailbox size or file count. Odd effects could occur as a result.
2962 The maildirsize information is now recalculated if the size or filecount
2965 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
2966 support for a long time. Removed HAVE_SYS_VFS_H.
2968 28. Installed the latest version of exipick from John Jetmore.
2970 29. In an address list, if the pattern was not a regular expression, an empty
2971 subject address (from a bounce message) matched only if the pattern was an
2972 empty string. Non-empty patterns were not even tested. This was the wrong
2973 because it is perfectly reasonable to use an empty address as part of a
2974 database query. An empty address is now tested by patterns that are
2975 lookups. However, all the other forms of pattern expect the subject to
2976 contain a local part and a domain, and therefore, for them, an empty
2977 address still always fails if the pattern is not itself empty.
2979 30. Exim went into a mad DNS loop when attempting to do a callout where the
2980 host was specified on an smtp transport, and looking it up yielded more
2981 than one IP address.
2983 31. Re-factored the code for checking spool and log partition space into a
2984 function that finds that data and another that does the check. The former
2985 is then used to implement four new variables: $spool_space, $log_space,
2986 $spool_inodes, and $log_inodes.
2988 32. The RFC2047 encoding function was originally intended for short strings
2989 such as real names; it was not keeping to the 75-character limit for
2990 encoded words that the RFC imposes. It now respects the limit, and
2991 generates multiple encoded words if necessary. To be on the safe side, I
2992 have increased the buffer size for the ${rfc2047: expansion operator from
2995 33. It is now permitted to omit both strings after an "if" condition; if the
2996 condition is true, the result is "true". As before, when the second string
2997 is omitted, a false condition yields an empty string. This makes it less
2998 cumbersome to write custom ACL and router conditions.
3000 34. Failure to deliver a bounce message always caused it to be frozen, even if
3001 there was an errors_to setting on the router. The errors_to setting is now
3004 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
3005 canonical form (fully expanded) before being placed in
3006 $sender_host_address.
3008 36. The table in the code that translates DNS record types into text (T_A to
3009 "A" for instance) was missing entries for NS and CNAME. It is just possible
3010 that this could have caused confusion if both these types were looked up
3011 for the same domain, because the text type is used as part of Exim's
3012 per-process caching. But the chance of anyone hitting this buglet seems
3015 37. The dnsdb lookup has been extended in a number of ways.
3017 (1) There is a new type, "zns", which walks up the domain tree until it
3018 finds some nameserver records. It should be used with care.
3020 (2) There is a new type, "mxh", which is like "mx" except that it returns
3021 just the host names, not the priorities.
3023 (3) It is now possible to give a list of domains (or IP addresses) to be
3024 looked up. The behaviour when one of the lookups defers can be
3025 controlled by a keyword.
3027 (4) It is now possible to specify the separator character for use when
3028 multiple records are returned.
3030 38. The dnslists ACL condition has been extended: it is now possible to supply
3031 a list of IP addresses and/or domains to be looked up in a particular DNS
3034 39. Added log_selector=+queue_time_overall.
3036 40. When running the queue in the test harness, wait just a tad after forking a
3037 delivery process, to get repeatability of debugging output.
3039 41. Include certificate and key file names in error message when GnuTLS fails
3040 to set them up, because the GnuTLS error message doesn't include the name
3041 of the failing file when there is a problem reading it.
3043 42. Allow both -bf and -bF in the same test run.
3045 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
3047 44. The "Exiscan patch" is now merged into the mainline Exim source.
3049 45. Sometimes the final signoff response after QUIT could fail to get
3050 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
3051 < 0 before doing a fflush(). This bug looks as though it goes back to the
3052 introduction of TLS in release 3.20, but "sometimes" must have been rare
3053 because the tests only now provoked it.
3055 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
3056 (this can affect the format of dates).
3058 47. exim_tidydb, when checking for the continued existence of a message for
3059 which it has found a message-specific retry record, was not finding
3060 messages that were in split spool directories. Consequently, it was
3061 deleting retry records that should have stayed in existence.
3063 48. Steve fixed some bugs in eximstats.
3065 49. The SPA authentication driver was not abandoning authentication and moving
3066 on to the next authenticator when an expansion was forced to fail,
3067 contradicting the general specification for all authenticators. Instead it
3068 was generating a temporary error. It now behaves as specified.
3070 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
3071 (the order specifies the preference for clients). The order is now AES256,
3072 AES128, 3DES, ARCFOUR128.
3074 51. Small patch to Sieve code - explicitly set From: when generating an
3077 52. Exim crashed if a remote delivery caused a very long error message to be
3078 recorded - for instance if somebody sent an entire SpamAssassin report back
3079 as a large number of 550 error lines. This bug was coincidentally fixed by
3080 increasing the size of one of Exim's internal buffers (big_buffer) that
3081 happened as part of the Exiscan merge. However, to be on the safe side, I
3082 have made the code more robust (and fixed the comments that describe what
3085 53. Now that there can be additional text after "Completed" in log lines (if
3086 the queue_time_overall log selector is set), a one-byte patch to exigrep
3087 was needed to allow it to recognize "Completed" as not the last thing in
3090 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
3091 patch that reportedly fixes this has been added. I am not expert enough to
3092 create a test for it. This is what the patch creator wrote:
3094 "I found a little strange behaviour of ldap code when working with
3095 Windows 2003 AD Domain, where users was placed in more than one
3096 Organization Units. When I tried to give exim partial DN, the exit code
3097 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
3098 But simultaneously result of request was absolutely normal ldap result,
3099 so I produce this patch..."
3101 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
3102 REFERENCE, so I have modified the code to exclude the patch when that macro
3105 55. Some experimental protocols are using DNS PTR records for new purposes. The
3106 keys for these records are domain names, not reversed IP addresses. The
3107 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
3108 leaves it alone. Component reversal etc. now happens only for IP addresses.
3111 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
3113 57. Double the size of the debug message buffer (to 2048) so that more of very
3114 long debug lines gets shown.
3116 58. The exicyclog utility now does better if the number of log files to keep
3117 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
3119 59. Two changes related to the smtp_active_hostname option:
3121 (1) $smtp_active_hostname is now available as a variable.
3122 (2) The default for smtp_banner uses $smtp_active_hostname instead
3123 of $primary_hostname.
3125 60. The host_aton() function is supposed to be passed a string that is known
3126 to be a valid IP address. However, in the case of IPv6 addresses, it was
3127 not checking this. This is a hostage to fortune. Exim now panics and dies
3128 if the condition is not met. A case was found where this could be provoked
3129 from a dnsdb PTR lookup with an IPv6 address that had more than 8
3130 components; fortuitously, this particular loophole had already been fixed
3131 by change 4.50/55 above.
3133 If there are any other similar loopholes, the new check in host_aton()
3134 itself should stop them being exploited. The report I received stated that
3135 data on the command line could provoke the exploit when Exim was running as
3136 exim, but did not say which command line option was involved. All I could
3137 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
3138 running as the user.
3141 61. There was a buffer overflow vulnerability in the SPA authentication code
3142 (which came originally from the Samba project). I have added a test to the
3143 spa_base64_to_bits() function which I hope fixes it.
3146 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
3147 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
3149 63. The daemon start-up calls getloadavg() while still root for those OS that
3150 need the first call to be done as root, but it missed one case: when
3151 deliver_queue_load_max is set with deliver_drop_privilege. This is
3152 necessary for the benefit of the queue runner, because there is no re-exec
3153 when deliver_drop_privilege is set.
3155 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
3156 This has been fixed.
3158 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
3159 was in use, was not putting the data itself into the right store pool;
3160 consequently, it could be overwritten for a subsequent message in the same
3161 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
3164 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
3166 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
3167 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
3168 the function were treating the return as a boolean value, which happened to
3169 work because 0=false and not-0=true, but is not correct code.
3171 68. The host_aton() function was not handling scoped IPv6 addresses (those
3172 with, for example, "%eth0" on the end) correctly.
3174 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
3175 negated items (that is, ~something) in unsigned ints. Some compilers
3176 apparently mutter when there is no cast.
3178 70. If an address verification called from an ACL failed, and did not produce a
3179 user-specific message (i.e. there was only a "system" message), nothing was
3180 put in $acl_verify_message. In this situation, it now puts the system
3183 71. Change 4.23/11 added synchronization checking at the start of an SMTP
3184 session; change 4.31/43 added the unwanted input to the log line - except
3185 that it did not do this in the start of session case. It now does.
3187 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
3188 This is wrong and can cause the other end to generate a synchronization
3189 error if it is another Exim or anything else that does the synchronization
3190 check. A QUIT command is no longer sent after a timeout.
3192 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
3193 during host lookups.
3195 74. The defer_ok option of callout verification was not working if it was used
3196 when verifying addresses in header lines, that is, for this case:
3198 verify = header_sender/callout=defer_ok
3200 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
3201 those file descriptors could be used for SMTP connections. If anything
3202 wrote to stderr (the example that came up was "warn" in embedded Perl), it
3203 could be sent to the SMTP client, causing chaos. The daemon now opens
3204 stdin, stdout, and stderr to /dev/null when it puts itself into the
3207 76. Arrange for output from Perl's "warn" command to be written to Exim's main
3208 log by default. The user can override this with suitable Perl magic.
3210 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
3211 the log message when discard triggers, was not working for the DATA ACL or
3212 for the non-SMTP ACL.
3214 78. Error message wording change in sieve.c.
3216 79. If smtp_accept_max_per_host was set, the number of connections could be
3217 restricted to fewer than expected, because the daemon was trying to set up
3218 a new connection before checking whether the processes handling previous
3219 connections had finished. The check for completed processes is now done
3220 earlier. On busy systems, this bug wouldn't be noticed because something
3221 else would have woken the daemon, and it would have reaped the completed
3224 80. If a message was submitted locally by a user whose login name contained one
3225 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
3226 It caused a spool format error. I have fixed the spool reading code. A
3227 related problem was that the "from" clause in the Received: line became
3228 illegal because of the space(s). It is now covered by ${quote_local_part.
3230 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
3233 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
3235 83. Patch from Sieve maintainer to fix unterminated string problem in
3236 "vacation" handling.
3238 84. Some minor changes to the Linux configuration files to help with other
3239 OS variants using glibc.
3241 85. One more patch for Sieve to update vacation handling to latest spec.
3244 ----------------------------------------------------
3245 See the note above about the 4.44 and 4.50 releases.
3246 ----------------------------------------------------
3252 1. Change 4.43/35 introduced a bug that caused file counts to be
3253 incorrectly computed when quota_filecount was set in an appendfile
3256 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
3257 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
3260 3. Give more explanation in the error message when the command for a transport
3261 filter fails to execute.
3263 4. There are several places where Exim runs a non-Exim command in a
3264 subprocess. The SIGUSR1 signal should be disabled for these processes. This
3265 was being done only for the command run by the queryprogram router. It is
3266 now done for all such subprocesses. The other cases are: ${run, transport
3267 filters, and the commands run by the lmtp and pipe transports.
3269 5. Some older OS have a limit of 256 on the maximum number of file
3270 descriptors. Exim was using setrlimit() to set 1000 as a large value
3271 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
3272 systems. I've change it so that if it can't get 1000, it tries for 256.
3274 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
3275 was an oversight, and furthermore, ever since the addition of extra
3276 controls (e.g. 4.43/32), the checks on when to allow different forms of
3277 "control" were broken. There should now be diagnostics for all cases when a
3278 control that does not make sense is encountered.
3280 7. $recipients is now available in the predata ACL (oversight).
3282 8. Tidy the search cache before the fork to do a delivery from a message
3283 received from the command line. Otherwise the child will trigger a lookup
3284 failure and thereby defer the delivery if it tries to use (for example) a
3285 cached ldap connection that the parent has called unbind on.
3287 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
3288 of $address_data from the recipient verification was clobbered by the
3289 sender verification.
3291 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
3292 was its contents. (It was OK if the option was not defined at all.)
3294 11. A "Completed" log line is now written for messages that are removed from
3295 the spool by the -Mrm option.
3297 12. $host_address is now set to the target address during the checking of
3298 ignore_target_hosts.
3300 13. When checking ignore_target_hosts for an ipliteral router, no host name was
3301 being passed; this would have caused $sender_host_name to have been used if
3302 matching the list had actually called for a host name (not very likely,
3303 since this list is usually IP addresses). A host name is now passed as
3306 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
3307 code that specifies a non-restarting handler (typically sigaction() in
3308 modern systems) in an attempt to fix a rare and obscure crash bug.
3310 15. Narrowed the window for a race in the daemon that could cause it to ignore
3311 SIGCHLD signals. This is not a major problem, because they are used only to
3312 wake it up if nothing else does.
3314 16. A malformed maildirsize file could cause Exim to calculate negative values
3315 for the mailbox size or file count. Odd effects could occur as a result.
3316 The maildirsize information is now recalculated if the size or filecount
3319 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
3320 support for a long time. Removed HAVE_SYS_VFS_H.
3322 18. Updated exipick to current release from John Jetmore.
3324 19. Allow an empty sender to be matched against a lookup in an address list.
3325 Previously the only cases considered were a regular expression, or an
3328 20. Exim went into a mad DNS lookup loop when doing a callout where the
3329 host was specified on the transport, if the DNS lookup yielded more than
3332 21. The RFC2047 encoding function was originally intended for short strings
3333 such as real names; it was not keeping to the 75-character limit for
3334 encoded words that the RFC imposes. It now respects the limit, and
3335 generates multiple encoded words if necessary. To be on the safe side, I
3336 have increased the buffer size for the ${rfc2047: expansion operator from
3339 22. Failure to deliver a bounce message always caused it to be frozen, even if
3340 there was an errors_to setting on the router. The errors_to setting is now
3343 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
3344 canonical form (fully expanded) before being placed in
3345 $sender_host_address.
3347 24. Updated eximstats to version 1.33
3349 25. Include certificate and key file names in error message when GnuTLS fails
3350 to set them up, because the GnuTLS error message doesn't include the name
3351 of the failing file when there is a problem reading it.
3353 26. Expand error message when OpenSSL has problems setting up cert/key files.
3356 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
3357 (this can affect the format of dates).
3359 28. exim_tidydb, when checking for the continued existence of a message for
3360 which it has found a message-specific retry record, was not finding
3361 messages that were in split spool directories. Consequently, it was
3362 deleting retry records that should have stayed in existence.
3364 29. eximstats updated to version 1.35
3365 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
3366 1.35 - bugfix such that pie charts by volume are generated correctly
3368 30. The SPA authentication driver was not abandoning authentication and moving
3369 on to the next authenticator when an expansion was forced to fail,
3370 contradicting the general specification for all authenticators. Instead it
3371 was generating a temporary error. It now behaves as specified.
3373 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
3374 (the order specifies the preference for clients). The order is now AES256,
3375 AES128, 3DES, ARCFOUR128.
3377 31. Small patch to Sieve code - explicitly set From: when generating an
3380 32. Exim crashed if a remote delivery caused a very long error message to be
3381 recorded - for instance if somebody sent an entire SpamAssassin report back
3382 as a large number of 550 error lines. This bug was coincidentally fixed by
3383 increasing the size of one of Exim's internal buffers (big_buffer) that
3384 happened as part of the Exiscan merge. However, to be on the safe side, I
3385 have made the code more robust (and fixed the comments that describe what
3388 33. Some experimental protocols are using DNS PTR records for new purposes. The
3389 keys for these records are domain names, not reversed IP addresses. The
3390 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
3391 leaves it alone. Component reversal etc. now happens only for IP addresses.
3394 34. The host_aton() function is supposed to be passed a string that is known
3395 to be a valid IP address. However, in the case of IPv6 addresses, it was
3396 not checking this. This is a hostage to fortune. Exim now panics and dies
3397 if the condition is not met. A case was found where this could be provoked
3398 from a dnsdb PTR lookup with an IPv6 address that had more than 8
3399 components; fortuitously, this particular loophole had already been fixed
3400 by change 4.50/55 or 4.44/33 above.
3402 If there are any other similar loopholes, the new check in host_aton()
3403 itself should stop them being exploited. The report I received stated that
3404 data on the command line could provoke the exploit when Exim was running as
3405 exim, but did not say which command line option was involved. All I could
3406 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
3407 running as the user.
3410 35. There was a buffer overflow vulnerability in the SPA authentication code
3411 (which came originally from the Samba project). I have added a test to the
3412 spa_base64_to_bits() function which I hope fixes it.
3415 36. The daemon start-up calls getloadavg() while still root for those OS that
3416 need the first call to be done as root, but it missed one case: when
3417 deliver_queue_load_max is set with deliver_drop_privilege. This is
3418 necessary for the benefit of the queue runner, because there is no re-exec
3419 when deliver_drop_privilege is set.
3421 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
3422 was in use, was not putting the data itself into the right store pool;
3423 consequently, it could be overwritten for a subsequent message in the same
3424 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
3427 38. Sometimes the final signoff response after QUIT could fail to get
3428 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
3429 < 0 before doing a fflush(). This bug looks as though it goes back to the
3430 introduction of TLS in release 3.20, but "sometimes" must have been rare
3431 because the tests only now provoked it.
3437 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
3438 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
3439 Now it returns an integer. A number of calls were still expecting a T/F
3440 return. Fortuitously, in all cases, the tests worked in OK situations,
3441 which is the norm. However, things would have gone wrong on any write
3442 failures on the smtp file descriptor. This function is used when sending
3443 messages over SMTP and also when doing verify callouts.
3445 2. When Exim is called to do synchronous delivery of a locally submitted
3446 message (the -odf or -odi options), it no longer closes stderr before doing
3449 3. Implemented the mua_wrapper option.
3451 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
3453 5. Implemented the functions header_remove(), header_testname(),
3454 header_add_at_position(), and receive_remove_recipient(), and exported them
3457 6. If an ACL "warn" statement specified the addition of headers, Exim already
3458 inserted X-ACL-Warn: at the start if there was no header name. However, it
3459 was not making this test for the second and subsequent header lines if
3460 there were newlines in the string. This meant that an invalid header could
3461 be inserted if Exim was badly configured.
3463 7. Allow an ACL "warn" statement to add header lines at the start or after all
3464 the Received: headers, as well as at the end.
3466 8. Added the rcpt_4xx retry error code.
3468 9. Added postmaster_mailfrom=xxx to callout verification option.
3470 10. Added mailfrom=xxxx to the callout verification option, for verify=
3473 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
3474 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
3476 12. Inserted some casts to stop certain compilers warning when using pointer
3477 differences as field lengths or precisions in printf-type calls (mostly
3478 affecting debugging statements).
3480 13. Added optional readline() support for -be (dynamically loaded).
3482 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
3483 same clock tick as a message's arrival, so that its received time was the
3484 same as the "first fail" time on the retry record, and that message
3485 remained on the queue past the ultimate address timeout, every queue runner
3486 would try a delivery (because it was past the ultimate address timeout) but
3487 after another failure, the ultimate address timeout, which should have then
3488 bounced the address, did not kick in. This was a "< instead of <=" error;
3489 in most cases the first failure would have been in the next clock tick
3490 after the received time, and all would be well.
3492 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
3493 being recognized when the domain list was tested by the match_domain
3494 condition in an expansion string.
3496 16. Added the ${str2b64: operator.
3498 17. Exim was always calling setrlimit() to set a large limit for the number of
3499 processes, without checking whether the existing limit was already
3500 adequate. (It did check for the limit on file descriptors.) Furthermore,
3501 errors from getrlimit() and setrlimit() were being ignored. Now they are
3502 logged to the main and panic logs, but Exim does carry on, to try to do its
3503 job under whatever limits there are.
3505 18. Imported PCRE 5.0.
3507 19. Trivial typo in log message " temporarily refused connection" (the leading
3510 20. If the log selector return_path_on_delivery was set and an address was
3511 redirected to /dev/null, the delivery process crashed because it assumed
3512 that a return path would always be set for a "successful" delivery. In this
3513 case, the whole delivery is bypassed as an optimization, and therefore no
3516 21. Internal re-arrangement: the function for sending a challenge and reading
3517 a response while authentication was assuming a zero-terminated challenge
3518 string. It's now changed to take a pointer and a length, to allow for
3519 binary data in such strings.
3521 22. Added the cyrus_sasl authenticator (code supplied by MBM).
3523 23. Exim was not respecting finduser_retries when seeking the login of the
3524 uid under which it was called; it was always trying 10 times. (The default
3525 setting of finduser_retries is zero.) Also, it was sleeping after the final
3526 failure, which is pointless.
3528 24. Implemented tls_on_connect_ports.
3530 25. Implemented acl_smtp_predata.
3532 26. If the domain in control=submission is set empty, Exim assumes that the
3533 authenticated id is a complete email address when it generates From: or
3534 Sender: header lines.
3536 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
3537 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
3538 chown and chgrp in /bin and hostname in /usr/bin.
3540 28. Exim was keeping the "process log" file open after each use, just as it
3541 does for the main log. This opens the possibility of it remaining open for
3542 long periods when the USR1 signal hits a daemon. Occasional processlog
3543 errors were reported, that could have been caused by this. Anyway, it seems
3544 much more sensible not to leave this file open at all, so that is what now
3547 29. The long-running daemon process does not normally write to the log once it
3548 has entered its main loop, and it closes the log before doing so. This is
3549 so that log files can straightforwardly be renamed and moved. However,
3550 there are a couple of unusual error situations where the daemon does write
3551 log entries, and I had neglected to close the log afterwards.
3553 30. The text of an SMTP error response that was received during a remote
3554 delivery was being truncated at 512 bytes. This is too short for some of
3555 the long messages that one sometimes sees. I've increased the limit to
3558 31. It is now possible to make retry rules that apply only when a message has a
3559 specific sender, in particular, an empty sender.
3561 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
3562 it possible to be selective about when SMTP synchronization is enforced.
3564 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
3566 32. Implemented hosts_connection_nolog.
3568 33. Added an ACL for QUIT.
3570 34. Setting "delay_warning=" to disable warnings was not working; it gave a
3573 35. Added mailbox_size and mailbox_filecount to appendfile.
3575 36. Added control = no_multiline_responses to ACLs.
3577 37. There was a bug in the logic of the code that waits for the clock to tick
3578 in the case where the clock went backwards by a substantial amount such
3579 that the microsecond fraction of "now" was more than the microsecond
3580 fraction of "then" (but the whole seconds number was less).
3582 38. Added support for the libradius Radius client library this is found on
3583 FreeBSD (previously only the radiusclient library was supported).
3589 1. When certain lookups returned multiple values in the form name=value, the
3590 quoting of the values was not always being done properly. Specifically:
3591 (a) If the value started with a double quote, but contained no whitespace,
3593 (b) If the value contained whitespace other than a space character (i.e.
3594 tabs or newlines or carriage returns) it was not quoted.
3595 This fix has been applied to the mysql and pgsql lookups by writing a
3596 separate quoting function and calling it from the lookup code. The fix
3597 should probably also be applied to nisplus, ibase and oracle lookups, but
3598 since I cannot test any of those, I have not disturbed their existing code.
3600 2. A hit in the callout cache for a specific address caused a log line with no
3601 reason for rejecting RCPT. Now it says "Previous (cached) callout
3602 verification failure".
3604 3. There was an off-by-one bug in the queryprogram router. An over-long
3605 return line was truncated at 256 instead of 255 characters, thereby
3606 overflowing its buffer with the terminating zero. As well as fixing this, I
3607 have increased the buffer size to 1024 (and made a note to document this).
3609 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
3610 when Exim is waiting for an SMTP response from a remote server, Exim
3611 restarts its select() call on the socket, thereby resetting its timeout.
3612 This is not a problem when such interrupts are rare. Somebody set up a cron
3613 job to run exiwhat every 2 minutes, which is less than the normal select()
3614 timeout (5 or 10 minutes). This meant that the select() timeout never
3615 kicked in because it was always reset. I have fixed this by comparing the
3616 time when an interrupt arrives with the time at the start of the first call
3617 to select(). If more time than the timeout has elapsed, the interrupt is
3618 treated as a timeout.
3620 5. Some internal re-factoring in preparation for the addition of Sieve
3621 extensions (by MH). In particular, the "personal" test is moved to a
3622 separate function, and given an option for scanning Cc: and Bcc: (which is
3623 not set for Exim filters).
3625 6. When Exim created an email address using the login of the caller as the
3626 local part (e.g. when creating a From: or Sender: header line), it was not
3627 quoting the local part when it contained special characters such as @.
3629 7. Installed new OpenBSD configuration files.
3631 8. Reworded some messages for syntax errors in "and" and "or" conditions to
3632 try to make them clearer.
3634 9. Callout options, other than the timeout value, were being ignored when
3635 verifying sender addresses in header lines. For example, when using
3637 verify = header_sender/callout=no_cache
3639 the cache was (incorrectly) being used.
3641 10. Added a missing instance of ${EXE} to the exim_install script; this affects
3642 only the Cygwin environment.
3644 11. When return_path_on_delivery was set as a log selector, if different remote
3645 addresses in the same message used different return paths and parallel
3646 remote delivery occurred, the wrong values would sometimes be logged.
3647 (Whenever a remote delivery process finished, the return path value from
3648 the most recently started remote delivery process was logged.)
3650 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
3651 lines when AUTH and/or TLS are in use. This is the "received protocol"
3652 field. Exim used to use "asmtp" for authenticated SMTP, without any
3653 indication (in the protocol name) for TLS use. Now it follows the RFC and
3654 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
3655 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
3656 names appear in log lines as well as in Received: header lines.
3658 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
3659 extensions, and comparison tests, and to fix some bugs.
3661 14. Changes to the "personal" filter test:
3663 (1) The test was buggy in that it was just doing the equivalent of
3664 "contains" tests on header lines. For example, if a user's address was
3665 anne@some.where, the "personal" test would incorrectly be true for
3667 To: susanne@some.where
3669 This test is now done by extracting each address from the header in turn,
3670 and checking the entire address. Other tests that are part of "personal"
3671 are now done using regular expressions (for example, to check local parts
3672 of addresses in From: header lines).
3674 (2) The list of non-personal local parts in From: addresses has been
3675 extended to include "listserv", "majordomo", "*-request", and "owner-*",
3676 taken from the Sieve specification recommendations.
3678 (3) If the message contains any header line starting with "List-" it is
3679 treated as non-personal.
3681 (4) The test for "circular" in the Subject: header line has been removed
3682 because it now seems ill-conceived.
3684 15. Minor typos in src/EDITME comments corrected.
3686 16. Installed latest exipick from John Jetmore.
3688 17. If headers_add on a router specified a text string that was too long for
3689 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
3690 of string_sprintf() is now avoided.
3692 18. $message_body_size was not set (it was always zero) when running the DATA
3693 ACL and the local_scan() function.
3695 19. For the "mail" command in an Exim filter, no default was being set for
3696 the once_repeat time, causing a random time value to be used if "once" was
3697 specified. (If the value happened to be <= 0, no repeat happened.) The
3698 default is now 0s, meaning "never repeat". The "vacation" command was OK
3699 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
3700 (I found it when inspecting the code).
3702 20. There is now an overall timeout for performing a callout verification. It
3703 defaults to 4 times the callout timeout, which applies to individual SMTP
3704 commands during the callout. The overall timeout applies when there is more
3705 than one host that can be tried. The timeout is checked before trying the
3706 next host. This prevents very long delays if there are a large number of
3707 hosts and all are timing out (e.g. when the network connections are timing
3708 out). The value of the overall timeout can be changed by specifying an
3709 additional sub-option for "callout", called "maxwait". For example:
3711 verify = sender/callout=5s,maxwait=20s
3713 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
3714 to the end before writing, but this should make it even safer).
3716 22. Exim was forgetting that it had advertised PIPELINING for the second and
3717 subsequent messages on an SMTP connection. It was also not resetting its
3718 memory on STARTTLS and an internal HELO.
3720 23. When Exim logs an SMTP synchronization error within a session, it now
3721 records whether PIPELINING has been advertised or not.
3723 24. Added 3 instances of "(long int)" casts to time_t variables that were being
3724 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
3725 rather than long int.
3727 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
3729 26. Added the never_mail option to autoreply.
3735 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
3736 crash if the getsockname() call failed; this can happen if a connection is
3737 closed very soon after it is established. The problem was simply in the
3738 order in which certain operations were done, causing Exim to try to write
3739 to the SMTP stream before it had set up the file descriptor. The bug has
3740 been fixed by making things happen in the correct order.
3746 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
3747 before the connection was closed, thus losing the rejection response.
3749 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
3750 some early Solaris releases, but causes trouble in current releases where
3751 socklen_t is defined.
3753 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
3756 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
3759 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
3760 that releasing the top of it at the end releases what was used for sub-
3761 expansions (unless the block got too big). However, discard this block if
3762 the first thing is a variable or header, so that we can use its block when
3763 it is dynamic (useful for very large $message_headers, for example).
3765 6. Lookups now cache *every* query, not just the most recent. A new, separate
3766 store pool is used for this. It can be recovered when all lookup caches are
3767 flushed. Lookups now release memory at the end of their result strings.
3768 This has involved some general refactoring of the lookup sources.
3770 7. Some code has been added to the store_xxx() functions to reduce the amount
3771 of flapping under certain conditions.
3773 8. log_incoming_interface used to affect only the <= reception log lines. Now
3774 it causes the local interface and port to be added to several more SMTP log
3775 lines, for example "SMTP connection from", and rejection lines.
3777 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
3779 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
3781 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
3782 could be overwritten at the end of the current message (or the start of a
3783 new message if it was set in a HELO ACL). The value is now preserved for
3784 the duration of the SMTP connection.
3786 12. If a transport had a headers_rewrite setting, and a matching header line
3787 contained an unqualified address, that address was qualified, even if it
3788 did not match any rewriting rules. The underlying bug was that the values
3789 of the flags that permit the existence of unqualified sender and recipient
3790 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
3791 non-local messages, and by -bnq for local messages) were not being
3792 preserved with the message after it was received.
3794 13. When Exim was logging an SMTP synchronization error, it could sometimes log
3795 "next input=" as part of the text comprising the host identity instead of
3796 the correct text. The code was using the same buffer for two different
3797 strings. However, depending on which order the printing function evaluated
3798 its arguments, the bug did not always show up. Under Linux, for example, my
3799 test suite worked just fine.
3801 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
3802 doesn't work with some older versions of Perl. It has been changed to "my",
3803 which in any case is probably the better facility to use.
3805 15. A really picky compiler found some instances of statements for creating
3806 error messages that either had too many or two few arguments for the format
3809 16. The size of the buffer for calls to the DNS resolver has been increased
3810 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
3811 for addresses that have a lot of PTR records. This alleviates a problem; it
3812 does not fully solve it.
3814 17. A dnsdb lookup for PTR records that receives more data than will fit in the
3815 buffer now truncates the list and logs the incident, which is the same
3816 action as happens when Exim is looking up a host name and its aliases.
3817 Previously in this situation something unpredictable would happen;
3818 sometimes it was "internal error: store_reset failed".
3820 18. If a server dropped the connection unexpectedly when an Exim client was
3821 using GnuTLS and trying to read a response, the client delivery process
3822 crashed while trying to generate an error log message.
3824 19. If a "warn" verb in an ACL added multiple headers to a message in a single
3825 string, for example:
3827 warn message = H1: something\nH2: something
3829 the text was added as a single header line from Exim's point of view
3830 though it ended up OK in the delivered message. However, searching for the
3831 second and subsequent header lines using $h_h2: did not work. This has been
3832 fixed. Similarly, if a system filter added multiple headers in this way,
3833 the routers could not see them.
3835 20. Expanded the error message when iplsearch is called with an invalid key to
3836 suggest using net-iplsearch in a host list.
3838 21. When running tests using -bh, any delays imposed by "delay" modifiers in
3839 ACLs are no longer actually imposed (and a message to that effect is
3842 22. If a "gecos" field in a passwd entry contained escaped characters, in
3843 particular, if it contained a \" sequence, Exim got it wrong when building
3844 a From: or a Sender: header from that name. A second bug also caused
3845 incorrect handling when an unquoted " was present following a character
3846 that needed quoting.
3848 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
3849 was not being matched caselessly.
3851 24. Arranged for all hyphens in the exim.8 source to be escaped with
3854 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
3855 properly. Recipient callout cache records were still being keyed to include
3856 the sender, even when use_sender was set false. This led to far more
3857 callouts that were necessary. The sender is no longer included in the key
3858 when use_sender is false.
3860 26. Added "control = submission" modifier to ACLs.
3862 27. Added the ${base62d: operator to decode base 62 numbers.
3864 28. dnsdb lookups can now access SRV records.
3866 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
3867 the configuration file.
3869 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
3870 (-v) mode. This makes the output for a verbose queue run more intelligible.
3872 31. Added a use_postmaster feature to recipient callouts.
3874 32. Added the $body_zerocount variable, containing the number of binary zero
3875 bytes in the message body.
3877 33. The time of last modification of the "new" subdirectory is now used as the
3878 "mailbox time last read" when there is a quota error for a maildir
3881 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
3883 35. Added +ignore_unknown as a special item in host lists.
3885 36. Code for decoding IPv6 addresses in host lists is now included, even if
3886 IPv6 support is not being compiled. This fixes a bug in which an IPv6
3887 address was recognized as an IP address, but was then not correctly decoded
3888 into binary, causing unexpected and incorrect effects when compared with
3895 1. Very minor rewording of debugging text in manualroute to say "list of
3896 hosts" instead of "hostlist".
3898 2. If verify=header_syntax was set, and a header line with an unqualified
3899 address (no domain) and a large number of spaces between the end of the
3900 name and the colon was received, the reception process suffered a buffer
3901 overflow, and (when I tested it) crashed. This was caused by some obsolete
3902 code that should have been removed. The fix is to remove it!
3904 3. When running in the test harness, delay a bit after writing a bounce
3905 message to get a bit more predictability in the log output.
3907 4. Added a call to search_tidyup() just before forking a reception process. In
3908 theory, someone could use a lookup in the expansion of smtp_accept_max_
3909 per_host which, without the tidyup, could leave open a database connection.
3911 5. Added the variables $recipient_data and $sender_data which get set from a
3912 lookup success in an ACL "recipients" or "senders" condition, or a router
3913 "senders" option, similar to $domain_data and $local_part_data.
3915 6. Moved the writing of debug_print from before to after the "senders" test
3918 7. Change 4.31/66 (moving the time when the Received: is generated) caused
3919 problems for message scanning, either using a data ACL, or using
3920 local_scan() because the Received: header was not generated till after they
3921 were called (in order to set the time as the time of reception completion).
3922 I have revised the way this works. The header is now generated after the
3923 body is received, but before the ACL or local_scan() are called. After they
3924 are run, the timestamp in the header is updated.
3930 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
3931 before starting a queue runner without re-exec. This happened only when
3932 deliver_drop_privilege was set or when the Exim user was set to root. The
3933 effect of the bug was that timeouts during subsequent deliveries caused
3934 crashes instead of being properly handled. The handler is now left at its
3935 default (and expected) setting.
3937 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
3938 message, again when deliver_drop_privilege is set or Exim is run as root.
3939 The bug described in (1) was not present in this case, but the tidying up
3940 of the other signals was missing. I have made the two cases consistent.
3942 3. The ignore_target_hosts setting on a manualroute router was being ignored
3943 for hosts that were looked up using the /MX notation.
3945 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
3948 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
3949 operated on the sender address. After changing the $sender_address to <>
3950 for the sender address verify, Exim was re-instated it as the original
3951 (before rewriting) address, but remembering that it had rewritten it, so it
3952 wasn't rewriting it again. This bug also had the effect of breaking the
3953 sender address verification caching when the sender address was rewritten.
3955 6. The ignore_target_hosts option was being ignored by the ipliteral router.
3956 This has been changed so that if the ip literal address matches
3957 ignore_target_hosts, the router declines.
3959 7. Added expansion conditions match_domain, match_address, and match_local_
3960 part (NOT match_host).
3962 8. The placeholder for the Received: header didn't have a length field set.
3964 9. Added code to Exim itself and to exim_lock to test for a specific race
3965 condition that could lead to file corruption when using MBX delivery. The
3966 issue is with the lockfile that is created in /tmp. If this file is removed
3967 after a process has opened it but before that process has acquired a lock,
3968 there is the potential for a second process to recreate the file and also
3969 acquire a lock. This could lead to two Exim processes writing to the file
3970 at the same time. The added code performs the same test as UW imapd; it
3971 checks after acquiring the lock that its file descriptor still refers to
3972 the same named file.
3974 10. The buffer for building added header lines was of fixed size, 8192 bytes.
3975 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
3978 11. Added the smtp_active_hostname option. If used, this will typically be made
3979 to depend on the incoming interface address. Because $interface_address is
3980 not set up until the daemon has forked a reception process, error responses
3981 that can happen earlier (such as "too many connections") no longer contain
3984 12. If an expansion in a condition on a "warn" statement fails because a lookup
3985 defers, the "warn" statement is abandoned, and the next ACL statement is
3986 processed. Previously this caused the whole ACL to be aborted.
3988 13. Added the iplsearch lookup type.
3990 14. Added ident_timeout as a log selector.
3992 15. Added tls_certificate_verified as a log selector.
3994 16. Added a global option tls_require_ciphers (compare the smtp transport
3995 option of the same name). This controls incoming TLS connections.
3997 17. I finally figured out how to make tls_require_ciphers do a similar thing
3998 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
3999 before starting the TLS session.
4001 18. Tabs are now shown as \t in -bP output.
4003 19. If the log selector return_path_on_delivery was set, Exim crashed when
4004 bouncing a message because it had too many Received: header lines.
4006 20. If two routers both had headers_remove settings, and the first one included
4007 a superfluous trailing colon, the final name in the first list and the
4008 first name in the second list were incorrectly joined into one item (with a
4009 colon in the middle).
4015 1. Added -C and -D options to the exinext utility, mainly to make it easier
4016 to include in the automated testing, but these could be helpful when
4017 multiple configurations are in use.
4019 2. The exinext utility was not formatting the output nicely when there was
4020 an alternate port involved in the retry record key, nor when there was a
4021 message id as well (for retries that were specific to a specific message
4022 and a specific host). It was also confused by IPv6 addresses, because of
4023 the additional colons they contain. I have fixed the IPv4 problem, and
4024 patched it up to do a reasonable job for IPv6.
4026 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
4027 delivery, the log line now contains "pipelined" if PIPELINING was used.
4029 4. An SMTP transport process used to panic and die if the bind() call to set
4030 an explicit outgoing interface failed. This has been changed; it is now
4031 treated in the same way as a connect() failure.
4033 5. A reference to $sender_host_name in the part of a conditional expansion
4034 that was being skipped was still causing a DNS lookup. This no longer
4037 6. The def: expansion condition was not recognizing references to header lines
4038 that used bh_ and bheader_.
4040 7. Added the _cache feature to named lists.
4042 8. The code for checking quota_filecount in the appendfile transport was
4043 allowing one more file than it should have been.
4045 9. For compatibility with Sendmail, the command line option
4053 and sets the incoming protocol and host name (for trusted callers). The
4054 host name and its colon can be omitted when only the protocol is to be set.
4055 Note the Exim already has two private options, -pd and -ps, that refer to
4056 embedded Perl. It is therefore impossible to set a protocol value of "d" or
4057 "s", but I don't think that's a major issue.
4059 10. A number of refactoring changes to the code, none of which should affect
4062 (a) The number of logging options was getting close to filling up the
4063 32-bit word that was used as a bit map. I have split them into two classes:
4064 those that are passed in the argument to log_write(), and those that are
4065 only ever tested independently outside of that function. These are now in
4066 separate 32-bit words, so there is plenty of room for expansion again.
4067 There is no change in the user interface or the logging behaviour.
4069 (b) When building, for example, log lines, the code previously used a
4070 macro that called string_cat() twice, in order to add two strings. This is
4071 not really sufficiently general. Furthermore, there was one instance where
4072 it was actually wrong because one of the argument was used twice, and in
4073 one call a function was used. (As it happened, calling the function twice
4074 did not affect the overall behaviour.) The macro has been replaced by a
4075 function that can join an arbitrary number of extra strings onto a growing
4078 (c) The code for expansion conditions now uses a table and a binary chop
4079 instead of a serial search (which was left over from when there were very
4080 few conditions). Also, it now recognizes conditions like "pam" even when
4081 the relevant support is not compiled in: a suitably worded error message is
4082 given if an attempt is made to use such a condition.
4084 11. Added ${time_interval:xxxxx}.
4086 12. A bug was causing one of the ddress fields not to be passed back correctly
4087 from remote delivery subprocesses. The field in question was not being
4088 subsequently used, so this caused to problems in practice.
4090 13. Added new log selectors queue_time and deliver_time.
4092 14. Might have fixed a bug in maildirsizefile handling that threw up
4093 "unexpected character" debug warnings, and recalculated the data
4094 unnecessarily. In any case, I expanded the warning message to give more
4097 15. Added the message "Restricted characters in address" to the statements in
4098 the default ACL that block characters like @ and % in local parts.
4100 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
4101 Three changes have been made:
4103 (a) There was a serious bug; a negative response to MAIL caused the whole
4104 recipient domain to be cached as invalid, thereby blocking all messages
4105 to all local parts at the same domain, from all senders. This bug has
4106 been fixed. The domain is no longer cached after a negative response to
4107 MAIL if the sender used is not empty.
4109 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
4112 (c) A new callout option, "use_sender" has been added for people who want
4113 the modified behaviour.
4119 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
4122 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
4123 indeed breaks things for older releases.
4125 3. Added additional logging to the case where there is a problem reading data
4126 from a filter that is running in a subprocess using a pipe, in order to
4127 try to track down a specific problem.
4129 4. Testing facility fudge: when running in the test harness and attempting
4130 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
4131 getting "No route to host". Convert this to a timeout.
4133 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
4136 6. Some OS don't have socklen_t but use size_t instead. This affects the
4137 fifth argument of getsockopt() amongst other things. This is now
4138 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
4139 can be set for individual OS. I have set it for SunOS5, OSF1, and
4140 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
4141 some earlier ones do not.
4143 7. Change 4.30/15 was not doing the test caselessly.
4145 8. The standard form for an IPv6 address literal was being rejected by address
4146 parsing in, for example, MAIL and RCPT commands. An example of this kind of
4147 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
4148 this, as well as the form without the "IPv6" on the front (but only when
4149 address literals are enabled, of course).
4151 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
4153 10. Exim crashed if a message with an empty sender address specified by -f
4154 encountered a router with an errors_to setting. This could be provoked only
4155 by a command such as
4159 where an empty string was supplied; "<>" did not hit this bug.
4161 11. Installed PCRE release 4.5.
4163 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
4164 remained set. It is now erased.
4166 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
4167 times from message ids (which are base 36 rather than the normal 62).
4169 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
4170 were being counted as actual protocol errors, and logged if the log
4171 selector +smtp_protocol_error was set. One cannot be perfect in this test,
4172 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
4173 and DATA following a set of rejected RCPTs do not count as protocol errors.
4174 In other words, Exim assumes they were pipelined, though this may not
4175 actually be the case. Of course, in all cases the client gets an
4176 appropriate error code.
4178 15. If a lookup fails in an ACL condition, a message about the failure may
4179 be available; it is used if testing the ACL cannot continue, because most
4180 such messages specify what the cause of the deferral is. However, some
4181 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
4182 that caused an old message to be retained and used if a later statement
4183 caused a defer, replacing the real cause of the deferral.
4185 16. If an IP address had so many PTR records that the DNS lookup buffer
4186 was not large enough to hold them, Exim could crash while trying to process
4187 the truncated data. It now detects and logs this case.
4189 17. Further to 4.21/58, another change has been made: if (and only if) the
4190 first line of a message (the first header line) ends with CRLF, a bare LF
4191 in a subsequent header line has a space inserted after it, so as not to
4192 terminate the header.
4194 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
4195 unnecessarily, used atoi() instead of strtol(), and didn't check the
4196 termination when getting file sizes from file names by regex.
4198 19. Completely re-implemented the support for maildirsize files, in the light
4199 of a number of problems with the previous contributed implementation
4200 (4.30/29). In particular:
4202 . If the quota is zero, the maildirsize file is maintained, but no quota is
4205 . If the maildir directory does not exist, it is created before any attempt
4206 to write a maildirsize file.
4208 . The quota value in the file is just a cache; if the quota is changed in
4209 the transport, the new value overrides.
4211 . A regular expression is available for excluding directories from the
4214 20. The autoreply transport checks the characters in options that define the
4215 message's headers; it allows continued headers, but it was checking with
4216 isspace() after an embedded newline instead of explicitly looking for a
4219 21. If all the "regular" hosts to which an address was routed had passed their
4220 expiry times, and had not reached their retry times, the address was
4221 bounced, even if fallback hosts were defined. Now Exim should go on to try
4224 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
4225 equivalent code in the SMTP transport. Some hosts send humungous responses
4226 to HELO/EHLO, more than 1024 it seems.
4228 23. Refactoring: code in filter.c used (void *) for "any old type" but this
4229 gives compiler warnings in some environments. I've now done it "properly",
4232 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
4233 (because of problems with the built-in one) was declared to return uschar *
4234 instead of char *, causing compiler failure.
4236 25. Fixed a file descriptor leak when processing alias/forward files.
4238 26. Fixed a minor format string issue in dbfn.c.
4240 27. Typo in exim.c: ("dmbnz" for "dbmnz").
4242 28. If a filter file refered to $h_xxx or $message_headers, and the headers
4243 contained RFC 2047 "words", Exim's memory could, under certain conditions,
4246 29. When a sender address is verified, it is cached, to save repeating the test
4247 when there is more than one recipient in a message. However, when the
4248 verification involves a callout, it is possible for different callout
4249 options to be set for different recipients. It is too complicated to keep
4250 track of this in the cache, so now Exim always runs a verification when a
4251 callout is required, relying on the callout cache for the optimization.
4252 The overhead is duplication of the address routing, but this should not be
4255 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
4256 to be verified with callout=postmaster, and the main callout worked but the
4257 postmaster check failed, the verification correctly failed. However, if a
4258 subsequent RCPT command asked for sender verification *without* the
4259 postmaster check, incorrect caching caused this verification also to fail,
4262 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
4263 it was not caching the DNS options (qualify_single, search_parents) that
4264 were used when the lookup failed. A subsequent lookup with different
4265 options therefore always gave the same answer, though there were cases
4266 where it should not have. (Example: a "domains = !$mx_any" option on a
4267 dnslookup router: the "domains" option is always processed without any
4268 widening, but the router might have qualify_single set.) Now Exim uses the
4269 cached value only when the same options are set.
4271 32. Added John Jetmore's "exipick" utility to the distribution.
4273 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
4274 than a timeout (e.g. a certificate is required, and is not provided), an
4275 Exim server now closes the connection immediately. Previously it waited for
4276 the client to close - but if the client is SSL, it seems that they each
4277 wait for each other, leading to a delay before one of them times out.
4279 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
4280 maintained 0.8.x compatibility because I don't think many are using it, and
4281 it is clearly obsolete.
4283 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
4286 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
4287 client certificate was expired. A simple patch fixes this, though I don't
4288 understand the full logic of why the verify callback is called multiple
4291 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
4292 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
4293 which causes problems with some clients (such as the Certicom SSL Plus
4294 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
4295 disables the coutermeasure allowing Eudora to connect."
4297 38. Exim was not checking that a write() to a log file succeeded. This could
4298 lead to Bad Things if a log got too big, in particular if it hit a file
4299 size limit. Exim now panics and dies if it cannot write to a log file, just
4300 as it does if it cannot open a log file.
4302 39. Modified OS/Makefile-Linux so that it now contains
4304 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
4306 The two -D definitions ensure that Exim is compiled with large file
4307 support, which makes it possible to handle log files that are bigger than
4310 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
4311 instance) a domain was checked against a named list that involved a lookup,
4312 causing $domain_data to be set, then another domain was checked against the
4313 same list, then the first domain was re-checked, the value of $domain_data
4314 after the final check could be wrong. In particular, if the second check
4315 failed, it could be set empty. This bug probably also applied to
4318 41. The strip_trailing_dot option was not being applied to the address given
4319 with the -f command-line option.
4321 42. The code for reading a message's header from the spool was incrementing
4322 $received_count, but never initializing it. This meant that the value was
4323 incorrect (doubled) while delivering a message in the same process in which
4324 it was received. In the most common configuration of Exim, this never
4325 happens - a fresh exec is done - but it can happen when
4326 deliver_drop_privilege is set.
4328 43. When Exim logs an SMTP synchronization error - client data sent too soon -
4329 it now includes up to 150 characters of the unexpected data in the log
4332 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
4333 and building data strings. The size of both of these buffers was 10 000
4334 bytes - far larger than anybody would *ever* want, thought I. Needless to
4335 say, somebody hit the limit. I have increased the maximum line length to
4336 20 000 and the maximum data length of concatenated lines to 100 000. I have
4337 also fixed two bugs, because there was no checking on these buffers. Tsk,
4338 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
4339 buffer is too small.
4341 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
4342 lsearch lookups. Now it does.
4344 46. When parsing a route_list item in a manualroute router, a fixed-length
4345 buffer was used for the list of hosts. I made this 1024 bytes long,
4346 thinking that nobody would ever have a list of hosts that long. Wrong.
4347 Somebody had a whole pile of complicated expansion conditions, and the
4348 string was silently truncated, leading to an expansion error. It turns out
4349 that it is easier to change to an unlimited length (owing to other changes
4350 that have happened since this code was originally written) than to build
4351 structure for giving a limitation error. The length of the item that
4352 expands into the list of hosts is now unlimited.
4354 47. The lsearch lookup could not handle data where the length of text line was
4355 more than 4095 characters. Such lines were truncated, leading to shortened
4356 data being returned. It should now handle lines of any length.
4358 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
4359 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
4362 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
4363 replacing strings such as BIN_DIRECTORY in a source file: the replacement
4364 no longer happens in comment lines. A list of replacements is now placed
4365 at the head of all of the source files, except those whose only change is
4366 to replace PERL_COMMAND in the very first #! line.
4368 50. Replaced the slow insertion sort in queue.c, for sorting the list of
4369 messages on the queue, with a bottom-up merge sort, using code contributed
4370 by Michael Haardt. This should make operations like -bp somewhat faster on
4371 large queues. It won't affect queue runners, except when queue_run_in_order
4374 51. Installed eximstats 1.31 in the distribution.
4376 52. Added support for SRV lookups to the dnslookup router.
4378 53. If an ACL referred to $message_body or $message_body_end, the value was not
4379 reset for any messages that followed in the same SMTP session.
4381 54. The store-handling optimization for building very long strings was not
4382 differentiating between the different store pools. I don't think this
4383 actually made any difference in practice, but I've tidied it.
4385 55. While running the routers to verify a sender address, $sender_address
4386 was still set to the sender address. This is wrong, because when routing to
4387 send a bounce to the sender, it would be empty. Therefore, I have changed
4388 it so that, while verifying a sender address, $sender_address is set to <>.
4389 (There is no change to what happens when verifying a recipient address.)
4391 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
4392 target A or AAAA records (if not already returned) without resetting the
4393 qualify_single or search_parents options of the DNS resolver. These are
4394 inappropriate in this case because the targets of MX and SRV records must
4395 be FQDNs. A broken DNS record could cause trouble if it happened to have a
4396 target that, when qualified, matched something in the local domain. These
4397 two options are now turned off when doing these lookups.
4399 57. It seems that at least some releases of Reiserfs (which does not have the
4400 concept of a fixed number of inodes) returns zero and not -1 for the
4401 number of available inodes. This interacted badly with check_spool_inodes,
4402 which assumed that -1 was the "no such thing" setting. What I have done is
4403 to check that the total number of inodes is greater than zero before doing
4404 the test of how many are available.
4406 58. When a "warn" ACL statement has a log_message modifier, the message is
4407 remembered, and not repeated. This is to avoid a lot of repetition when a
4408 message has many recipients that cause the same warning to be written.
4409 Howewer, Exim was preserving the list of already written lines for an
4410 entire SMTP session, which doesn't seem right. The memory is now reset if a
4411 new message is started.
4413 59. The "rewrite" debugging flag was not showing the result of rewriting in the
4414 debugging output unless log_rewrite was also set.
4416 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
4417 of (int)(handle) when we know that handle contains (void *)(-1).
4419 61. The Exim daemon panic-logs an error return when it closes the incoming
4420 connection. However "connection reset by peer" seems to be common, and
4421 isn't really an error worthy of noting specially, so that particular error
4424 62. When Exim is trying to find all the local interfaces, it used to panic and
4425 die if the ioctl to get the interface flags failed. However, it seems that
4426 on at least one OS (Solaris 9) it is possible to have an interface that is
4427 included in the list of interfaces, but for which you get a failure error
4428 for this call. This happens when the interface is not "plumbed" into a
4429 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
4430 failure of the "get flags" call assumes that the interface is down.
4432 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
4433 makes life easier for people who are doing arithmetic on fields extracted
4434 from dates, where you often get leading zeros that should not be
4435 interpreted as octal.
4437 64. Added qualify_domain to the redirect router, to override the global
4440 65. If a pathologically long header line contained very many addresses (the
4441 report of this problem mentioned 10 000) and each of them was rewritten,
4442 Exim could use up a very large amount of memory. (It kept on making new
4443 copies of the header line as it rewrote, and never released the old ones.)
4444 At the expense of a bit more processing, the header rewriting function has
4445 been changed so that it no longer eats memory in this way.
4447 66. The generation of the Received: header has been moved from the time that a
4448 message starts to be received, to the time that it finishes. The timestamp
4449 in the Received: header should now be very close to that of the <= log
4450 line. There are two side-effects of this change:
4452 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
4453 logged header lines no longer include the local Received: line, because
4454 it has not yet been created. The same applies to a copy of the message
4455 that is returned to a non-SMTP sender when a message is rejected.
4457 (b) When a filter file is tested using -bf, no additional Received: header
4458 is added to the test message. After some thought, I decided that this
4461 This change does not affect the value of $received_for. It is still set
4462 after address rewriting, but before local_scan() is called.
4464 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
4466 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
4467 gave an unhelpful panic error message, and a defer error. I have managed to
4468 change this behaviour so that it now rejects any supplied certificate,
4469 which seems right, as the list of acceptable certificates is empty.
4471 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
4472 gave an unhelpful defer error. I have not managed to make this reject any
4473 supplied certificates, but the error message it gives is "no certificate
4474 supplied", which is not helpful.
4476 70. exigrep's output now also includes lines that are not associated with any
4477 message, but which match the given pattern. Implemented by a patch from
4478 Martin Sluka, which also tidied up the Perl a bit.
4480 71. Recipient callout verification, like sender verification, was using <> in
4481 the MAIL FROM command. This isn't really the right thing, since the actual
4482 sender may affect whether the remote host accepts the recipient or not. I
4483 have changed it to use the actual sender in the callout; this means that
4484 the cache record is now keyed on a recipient/sender pair, not just the
4485 recipient address. There doesn't seem to be a real danger of callout loops,
4486 since a callout by the remote host to check the sender would use <>.
4487 [SEE ABOVE: changed after hitting problems.]
4489 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
4490 temporary errors. However, in the case of such a code being given after
4491 the end of a data transmission (i.e. after ".") Exim was failing to write
4492 a retry record for the message. (Yes, there was some broken host that was
4493 actually sending 8xx at this point.)
4495 73. An unknown lookup type in a host list could cause Exim to panic-die when
4496 the list was checked. (An example that provoked this was putting <; in the
4497 middle of a list instead of at the start.) If this happened during a DATA
4498 ACL check, a -D file could be left lying around. This kind of configuration
4499 error no longer causes Exim to die; instead it causes a defer errror. The
4500 incident is still logged to the main and panic logs.
4502 74. Buglet left over from Exim 3 conversion. The message "too many messages
4503 in one connection" was written to the rejectlog but not the mainlog, except
4504 when address rewriting (yes!) was being logged.
4506 75. Added write_rejectlog option.
4508 76. When a system filter was run not as root (that is, when system_filter_user
4509 was set), the values of the $n variables were not being returned to the
4510 main process; thus, they were not subsequently available in the $sn
4513 77. Added +return_path_on_delivery log selector.
4515 78. A connection timeout was being treated differently from recipients deferred
4516 when testing hosts_max_try with a message that was older than the host's
4517 retry timeout. (The host should not be counted, thus allowing all hosts to
4518 be tried at least once before bouncing.) This may have been the cause of an
4519 occasionally reported bug whereby a message would remain on the queue
4520 longer than the retry timeout, but would be bounced if a delivery was
4521 forced. I say "may" because I never totally pinned down the problem;
4522 setting up timeout/retry tests is difficult. See also the next item.
4524 79. The ultimate address timeout was not being applied to errors that involved
4525 a combination of host plus message (for example, a timeout on a MAIL
4526 command). When an address resolved to a number of possible hosts, and they
4527 were not all tried for each delivery (e.g. because of hosts_max_try), a
4528 message could remain on the queue longer than the retry timeout.
4530 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
4533 81. Fixed an obscure SMTP outgoing bug which required at least the following
4534 conditions: (a) there was another message waiting for the same server;
4535 (b) the server returned 5xx to all RCPT commands in the first message so
4536 that the message was not completed; (c) the server dropped the connection
4537 or gave a negative response to the RSET that Exim sends to abort the
4538 transaction. The observed case was a dropped connection after DATA that had
4539 been sent in pipelining mode. That is, the server had advertised PIPELINING
4540 but was not implementing it correctly. The effect of the bug was incorrect
4541 behaviour, such as trying another host, and this could lead to a crash.
4547 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
4548 and daemon.c were passed as pointers to ints; they should have been
4549 pointers to socklen_t variables (which are typically unsigned ints).
4551 2. Some signed/unsigned type warnings in the os.c file for Linux have been
4554 3. Fixed a really odd bug that affected only the testing scheme; patching a
4555 certain fixed string in the binary changed the value of another string that
4556 happened to be identical to the end of the original first string.
4558 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
4559 name", it returns that address as the IP address. On some operating
4560 systems (e.g. Solaris), it also passes back the IP address string as the
4561 "host name". However, on others (e.g. Linux), it passes back an empty
4562 string. Exim wasn't checking for this, and was changing the host name to an
4563 empty string, assuming it had been canonicized.
4565 5. Although rare, it is permitted to have more than one PTR record for a given
4566 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
4567 all the names associated with an address, because they do in Solaris.
4568 However, it seems that they do not in Linux for data that comes from the
4569 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
4570 I found this out when I moved to a new Linux workstation and tried to run
4571 the Exim test suite.
4573 To get round this problem I have changed the code so that it now does its
4574 own call to the DNS to look up PTR records when searching for a host name.
4575 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
4576 addresses that are only in /etc/hosts are still found.
4578 This behaviour is, however, controlled by an option called host_lookup_
4579 order, which defaults to "bydns:byaddr". If people want to use the other
4580 order, or indeed, just use one or the other means of lookup, they can
4581 specify it in this variable.
4583 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
4584 some operating systems, this comes back from gethostbyaddr() as an empty
4585 string, and this is what Exim used to test for. However, it seems that in
4586 other systems, "." is yielded. Exim now tests for this case too.
4588 7. The values of check_spool_space and check_log_space are now held internally
4589 as a number of kilobytes instead of an absolute number of bytes. If a
4590 numbers is specified without 'K' or 'M', it is rounded up to the nearest
4591 kilobyte. This means that much larger values can be stored.
4593 8. Exim monitor: an attempt to get the action menu when not actually pointing
4594 at a message produces an empty menu entitled "No message selected". This
4595 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
4596 no entries in it ("Shell widget menu has zero width and/or height"). So I
4597 have added a single, blank menu entry in this case.
4599 9. Added ${quote_local_part.
4601 10. MIME decoding is now applied to the contents of Subject: header lines when
4604 11. Now that a reference to $sender_host_address automatically causes a reverse
4605 lookup to occur if necessary (4.13/18), there is no need to arrange for a
4606 host lookup before query-style lookups in lists that might use this
4607 variable. This has therefore been abolished, and the "net-" prefix is no
4608 longer necessary for query-style lookups.
4610 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
4611 have been a typo for LFLAGS, so it has been changed.
4613 13. The install script calls Exim with "-C /dev/null" in order to find the
4614 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
4615 to be output. Howeve, since Exim outputs its version number before the
4616 error, it didn't break the script. It just looked ugly. I fixed this by
4617 always allowing "-C /dev/null" if the caller is root.
4619 14. Ignore overlarge ACL variable number when reading spool file - insurance
4620 against a later release with more variables having written the file.
4622 15. The standard form for an IPv6 address literal was being rejected by EHLO.
4623 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
4624 this, as well as the form without the "IPv6" on the front.
4626 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
4627 OS/Makefile-Darwin file.
4629 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
4630 only to LDAP libraries that do not have LDAP_OPT_DEREF.
4632 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
4633 variable. However, some OS use ints for inodes. I've added cast to long int
4634 to get rid of the compiler warning.
4636 19. I had forgotten to lock out "/../" in configuration file names when
4637 ALT_CONFIG_PREFIX was set.
4639 20. Routers used for verification do not need to specify transports. However,
4640 if such a router generated a host list, and callout was configured, Exim
4641 crashed, because it could not find a port number from the (non-existent)
4642 transport. It now assumes port 25 in this circumstance.
4644 21. Added the -t option to exigrep.
4646 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
4647 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
4648 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
4649 them is set, LOOKUP_LSEARCH is forced.
4651 23. "exim -bV" now outputs a list of lookups that are included in the binary.
4653 24. Added sender and host information to the "rejected by local_scan()" log
4654 line; previously there was no indication of these.
4656 25. Added .include_if_exists.
4658 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
4659 Linux. It turns out that not all file systems support this. Apparently some
4660 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
4661 do it.) To cope with this, the error EINVAL, which means that sync-ing is
4662 not supported on the file descriptor, is now ignored when Exim is trying to
4663 sync a directory. This applies only to Linux.
4665 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
4667 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
4668 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
4671 29. Added support for maildirsize files from supplied patch (modified a bit).
4673 30. The use of :fail: followed by an empty string could lead Exim to respond to
4674 sender verification failures with (e.g.):
4676 550 Verification failed for <xxx>
4677 550 Sender verify failed
4679 where the first response line was missing the '-' that indicates it is not
4680 the final line of the response.
4682 31. The loop for finding the name of the user that called Exim had a hardwired
4683 limit of 10; it now uses the value of finduser_retries, which is used for
4684 all other user lookups.
4686 32. Added $received_count variable, available in data and not_smtp ACLs, and at
4689 33. Exim was neglecting to zero errno before one call of strtol() when
4690 expanding a string and expecting an integer value. On some systems this
4691 resulted in spurious "integer overflow" errors. Also, it was casting the
4692 result into an int without checking.
4694 34. Testing for a connection timeout using "timeout_connect" in the retry rules
4695 did not work. The code looks as if it has *never* worked, though it appears
4696 to have been documented since at least releast 1.62. I have made it work.
4698 35. The "timeout_DNS" error in retry rules, also documented since at least
4699 1.62, also never worked. As it isn't clear exactly what this means, and
4700 clearly it isn't a major issue, I have abolished the feature by treating it
4701 as "timeout", and writing a warning to the main and panic logs.
4703 36. The display of retry rules for -brt wasn't always showing the error code
4706 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
4707 timeout_connect_A, timeout_connect_MX.
4709 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
4710 to the empty sender.
4712 39. The daemon was not analysing the content of -oX till after it had closed
4713 stderr and disconnected from the controlling terminal. This meant that any
4714 syntax errors were only noted on the panic log, and the return code from
4715 the command was 0. By re-arranging the code a little, I've made the
4716 decoding happen first, so such errors now appear on stderr, and the return
4717 code is 1. However, the actual setting up of the sockets still happens in
4718 the disconnected process, so errors there are still only recorded on the
4721 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
4722 connections (as happens on some IP stacks) was logged at start up time as
4723 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
4724 it from "IPv6 and IPv4", which means that two separate sockets are being
4727 41. The debug output for gethostbyname2() or getipnodebyname() failures now
4728 says whether AF_INET or AF_INET6 was passed as an argument.
4730 42. Exiwhat output was messed up when time zones were included in log
4733 43. Exiwhat now gives more information about the daemon's listening ports,
4734 and whether -tls-on-connect was used.
4736 44. The "port" option of the smtp transport is now expanded.
4738 45. A "message" modifier in a "warn" statement in a non-message ACL was being
4739 silently ignored. Now an error message is written to the main and panic
4742 46. There's a new ACL modifier called "logwrite" which writes to a log file
4743 as soon as it is encountered.
4745 47. Added $local_user_uid and $local_user_gid at routing time.
4747 48. Exim crashed when trying to verify a sender address that was being
4750 49. Exim was recognizing only a space character after ".include". It now also
4751 recognizes a tab character.
4753 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
4754 extracting the relevant information from the specification. The man page no
4755 longer contains scrambled data for the -d option, and I've added a section
4756 at the front about calling Exim under different names.
4758 51. Added "extra_headers" argument to the "mail" command in filter files.
4760 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
4763 53. Installed eximstats 1.29.
4765 54. Added transport_filter_timeout as a generic transport option.
4767 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
4768 Cc: header lines. This was required by RFC 822, but it not required by RFC
4771 56. Exim used to add From:, Date:, and Message-Id: header lines to any
4772 incoming messages that did not have them. Now it does so only if the
4773 message originates locally, that is, if there is no associated remote host
4774 address. When Resent- header lines are present, this applies to the Resent-
4775 lines rather than the non-Resent- lines.
4777 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
4778 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
4780 58. Messages for configuration errors now include the name of the main
4781 configuration file - useful now that there may be more than one file in a
4782 list (.included file names were always shown).
4784 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
4785 for those rare installations that do not start the daemon as root or run it
4786 setuid root. I've cut out the call to initgroups() if the daemon is not
4789 60. The Exim user and group can now be bound into the binary as text strings
4790 that are looked up at the start of Exim's processing.
4792 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
4794 62. Added $mailstore_basename variable.
4796 63. Installed patch to sieve.c from Michael Haardt.
4798 64. When Exim failed to open the panic log after failing to open the main log,
4799 the original message it was trying to log was written to stderr and debug
4800 output, but if they were not available (the usual case in production), it
4801 was lost. Now it is written to syslog before the two lines that record the
4802 failures to open the logs.
4804 65. Users' Exim filters run in subprocesses under the user's uid. It is
4805 possible for a "deliver" command or an alias in a "personal" command to
4806 provoke an address rewrite. If logging of address rewriting is configured,
4807 this fails because the process is not running as root or exim. There may be
4808 a better way of dealing with this, but for the moment (because 4.30 needs
4809 to be released), I have disabled address rewrite logging when running a
4810 filter in a non-root, non-exim process.
4816 1. The buildconfig auxiliary program wasn't quoting the value set for
4817 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
4818 not defined. This bug was masked in 4.22 by the effect that was fixed in
4821 2. Some messages that were rejected after a message id was allocated were
4822 shown as "incomplete" by exigrep. It no longer does this for messages that
4823 are rejected by local_scan() or the DATA or non-SMTP ACLs.
4825 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
4826 have allow_domain_literals set, the ID did not get logged in the <= line.
4827 Domain literals are now always recognized in Message-ID: header lines.
4829 4. The first argument for a ${extract expansion item is the key name or field
4830 number. Leading and trailing spaces in this item were not being ignored,
4831 causing some misleading effects.
4833 5. When deliver_drop_privilege was set, single queue runner processes started
4834 manually (i.e. by the command "exim -q") or by the daemon (which uses the
4835 same command in the process it spins off) were not dropping privilege.
4837 6. When the daemon running as "exim" started a queue runner, it always
4838 re-executed Exim in the spun-off process. This is a waste of effort when
4839 deliver_drop_privilege is set. The new process now just calls the
4840 queue-runner function directly.
4846 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
4849 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
4850 ignored. Though the use of -oP was forcing the writing of a pid file, it
4851 was always written to the default place.
4853 3. If the message "no IP address found for host xxxx" is generated during
4854 incoming verification, it is now followed by identification of the incoming
4855 connection (so you can more easily find what provoked it).
4857 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
4859 5. Added some features to "harden" Exim a bit more against certain attacks:
4861 (a) There is now a build-time option called FIXED_NEVER_USERS that can
4862 be put in Local/Makefile. This is like the never_users runtime option,
4863 but it cannot be overridden. The default setting is "root".
4865 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
4866 prefix string with which any file named in a -C command line option
4869 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
4870 is retained for -C and -D only if the caller of Exim is root. Without
4871 it, the exim user may also use -C and -D and retain privilege.
4873 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
4874 command line option is disabled.
4876 6. Macro names set by the -D option must start with an upper case letter, just
4877 like macro names defined in the configuration file.
4879 7. Added "dereference=" facility to LDAP.
4881 8. Two instances of the typo "uknown" in the source files are fixed.
4883 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
4884 the Configure-Makefile script screwed up while processing it.
4886 10. Incorporated PCRE 4.4.
4888 11. The SMTP synchronization check was not operating right at the start of an
4889 SMTP session. For example, it could not catch a HELO sent before the client
4890 waited for the greeting. There is now a check for outstanding input at the
4891 point when the greeting is written. Because of the duplex, asynchronous
4892 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
4893 way, but not yet received, when the check is performed.
4895 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
4896 on TCP/IP sockets, because this apparently causes some broken clients to
4899 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
4900 unchanged) from the Cygwin maintainer.
4902 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
4903 was supported instead of testing for mbx. Effectively a typo.
4905 15. The spa authenticator server code was not checking that the input it
4906 received was valid base64.
4908 16. The debug output line for the "set" modifier in ACLs was not showing the
4909 name of the variable that was being set.
4911 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
4913 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
4914 lookup on its own. Something else was needed to trigger the lookup. For
4915 example, a match in host_lookup or the need for a host name in a host list.
4916 Now, if $sender_host_name is referenced and the host name has not yet been
4917 looked up, a lookup is performed. If the lookup fails, the variable remains
4918 empty, and $host_lookup_failed is set to "1".
4920 19. Added "eqi" as a case-independent comparison operator.
4922 20. The saslauthd authentication condition could segfault if neither service
4923 nor realm was specified.
4925 21. If an overflowing value such as "2048M" was set for message_size_limit, the
4926 error message that was logged was misleading, and incoming SMTP
4927 connections were dropped. The message is now more accurate, and temporary
4928 errors are given to SMTP connections.
4930 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
4931 (except RSET) with a 421 error, until QUIT is received. However, it was
4932 failing to send a response to QUIT.
4934 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
4935 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
4936 ACL is now run after the helo_try_verify_hosts code.
4938 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
4939 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
4940 case-independent, so other case variants are also recognized). Apparently
4941 some systems use these upper case variants.
4943 25. If more than two messages were waiting for the same host, and a transport
4944 filter was specified for the transport, Exim sent two messages over the
4945 same TCP/IP connection, and then failed with "socket operation on non-
4946 socket" when it tried to send the third.
4948 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
4950 27. The extern definition of crypt16() in expand.c was not being excluded when
4951 the OS had its own crypt16() function.
4953 28. Added bounce_return_body as a new option, and bounce_return_size_limit
4954 as a preferred synonym for return_size_limit, both as an option and as an
4957 29. Added LIBS=-liconv to OS/Makefile-OSF1.
4959 30. Changed the default configuration ACL to relax the local part checking rule
4960 for addresses that are not in any local domains. For these addresses,
4961 slashes and pipe symbols are allowed within local parts, but the sequence
4962 /../ is explicitly forbidden.
4964 31. SPA server authentication was not clearing the challenge buffer before
4967 32. log_message in a "warn" ACL statement was writing to the reject log as
4968 well as to the main log, which contradicts the documentation and doesn't
4969 seem right (because no rejection is happening). So I have stopped it.
4971 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
4972 However, I am unable to do any testing of this.
4974 34. Fixed an infelicity in the appendfile transport. When checking directories
4975 for a mailbox, to see if any needed to be created, it was accidentally
4976 using path names with one or more superfluous leading slashes; tracing
4977 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
4979 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
4980 contents are added to the log line that is written for every discarded
4981 recipient. (Previously a log_message setting was ignored.)
4983 36. The ${quote: operator now quotes the string if it is empty.
4985 37. The install script runs exim in order to find its version number. If for
4986 some reason other than non-existence or emptiness, which it checks, it
4987 could not run './exim', it was installing it with an empty version number,
4988 i.e. as "exim-". This error state is now caught, and the installation is
4991 38. An argument was missing from the function that creates an error message
4992 when Exim fails to connect to the socket for saslauthd authentication.
4993 This could cause Exim to crash, or give a corrupted message.
4995 39. Added isip, isip4, and isip6 to ${if conditions.
4997 40. The ACL variables $acl_xx are now saved with the message, and can be
4998 accessed later in routers, transports, and filters.
5000 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
5001 strings in the file are not string-expanded.
5003 42. If a MAIL command specified a SIZE value that was too large to fit into an
5004 int variable, the check against message_size_limit failed. Such values are
5005 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
5006 day this will have to be increased, but I don't think I want to be around
5007 when emails are that large.
5014 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
5015 iconv() is not standard in FreeBSD.
5017 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
5018 IPv6 enabled. The observed symptom was a segmentation fault on return from
5019 the function os_common_find_running_interfaces() in src/os.c.
5021 3. In the check_special_case() function in daemon.c I had used "errno" as an
5022 argument name, which causes warnings on some systems. This was basically a
5023 typo, since it was named "eno" in the comments!
5025 4. The code that waits for the clock to tick (at a resolution of some fraction
5026 of a second) so as to ensure message-id uniqueness was always waiting for
5027 at least one whole tick, when it could have waited for less. [This is
5028 almost certainly not relevant at current processor speeds, where it is
5029 unlikely to ever wait at all. But we try to future-proof.]
5031 5. The function that sleeps for a time interval that includes fractions of a
5032 second contained a race. It did not block SIGALRM between setting the
5033 timer, and suspending (a couple of lines later). If the interval was short
5034 and the sigsuspend() was delayed until after it had expired, the suspension
5035 never ended. On busy systems this could lead to processes getting stuck for
5038 6. Some uncommon configurations may cause a lookup to happen in a queue runner
5039 process, before it forks any delivery processes. The open lookup caching
5040 mechanism meant that the open file or database connection was passed into
5041 the delivery process. The problem was that delivery processes always tidy
5042 up cached lookup data. This could cause a problem for the next delivery
5043 process started by the queue runner, because the external queue runner
5044 process does not know about the closure. So the next delivery process
5045 still has data in the lookup cache. In the case of a file lookup, there was
5046 no problem because closing a file descriptor in a subprocess doesn't affect
5047 the parent. However, if the lookup was caching a connection to a database,
5048 the connection was closed, and the second delivery process was likely to
5049 see errors such as "PGSQL: query failed: server closed the connection
5050 unexpectedly". The problem has been fixed by closing all cached lookups
5051 in a queue runner before running a delivery process.
5053 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
5054 seem to have the "const" qualifier which it has on other OS. I've
5057 8. Change 4.21/2 was too strict. It is only if there are two authenticators
5058 *of the same type* (client or server) with the same public name that an
5059 error should be diagnosed.
5061 9. When Exim looked up a host name for an IP address, but failed to find the
5062 original IP address when looking up the host name (a safety check), it
5063 output the message "<ip address> does not match any IP for NULL", which was
5064 confusing, to say the least. The bug was that the host name should have
5065 appeared instead of "NULL".
5067 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
5068 user that is built into the binary, and the -C or -D options is used, root
5069 privilege is dropped before the configuration file is read. In addition,
5070 logging is switched to stderr instead of the normal log files. If the
5071 configuration then re-defines the Exim user, the unprivileged environment
5072 is probably not what is expected, so Exim logs a panic warning message (but
5075 However, if deliver_drop_privilege is set, the unprivileged state may well
5076 be exactly what is intended, so the warning has been cut out in that case,
5077 and Exim is allowed to try to write to its normal log files.
5083 1. smtp_return_error_details was not giving details for temporary sender
5084 or receiver verification errors.
5086 2. Diagnose a configuration error if two authenticators have the same public
5089 3. Exim used not to create the message log file for a message until the first
5090 delivery attempt. This could be confusing when incoming messages were held
5091 for policy or load reasons. The message log file is now created at the time
5092 the message is received, and an initial "Received" line is written to it.
5094 4. The automatically generated man page for command line options had a minor
5095 bug that caused no ill effects; however, a more serious problem was that
5096 the procedure for building the man page automatically didn't always
5097 operate. Consequently, release 4.20 contains an out-of-date version. This
5098 shouldn't happen again.
5100 5. When building Exim with embedded Perl support, the script that builds the
5101 Makefile was calling 'perl' to find its compile-time parameters, ignoring
5102 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
5104 6. The freeze_tell option was not being used for messages that were frozen on
5105 arrival, either by an ACL or by local_scan().
5107 7. Added the smtp_incomplete_transaction log selector.
5109 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
5110 was accepting AUTH without a new EHLO.
5112 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
5113 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
5116 10. Logging of TCP/IP connections (when configured) now happens in the main
5117 daemon process instead of the child process, so that the TCP/IP connection
5118 count is more accurate (but it can never be perfect).
5120 11. The use of "drop" in a nested ACL was not being handled correctly in the
5121 outer ACL. Now, if condition failure induced by the nested "drop" causes
5122 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
5123 or "require"), the connection is dropped.
5125 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
5126 that yield "discard" can now be used with an "accept" or a "discard" verb,
5127 but an error is generated for any others (because I can't see a useful way
5128 to define what should happen).
5130 13. When an ACL is read dynamically from a file (or anywhere else), the lines
5131 are now processed in the same way as lines in the Exim configuration file.
5132 In particular, continuation lines are supported.
5134 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
5136 15. Added -ti meaning -t -i.
5138 16. Check for letters, digits, hyphens, and dots in the names of dnslist
5139 domains, and warn by logging if others are found.
5141 17. At least on BSD, alignment is not guarenteed for the array of ifreq's
5142 returned from GIFCONF when Exim is trying to find the list of interfaces on
5143 a host. The code in os.c has been modified to copy each ifreq to an aligned
5144 structure in all cases.
5146 Also, in some cases, the returned ifreq's were being copied to a 'struct
5147 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
5148 means the last couple of bytes of an IPv6 address could be chopped if the
5149 ifreq contained only a normal sockaddr (14 bytes storage).
5151 18. Named domain lists were not supported in the hosts_treat_as_local option.
5152 An entry such as +xxxx was not recognized, and was treated as a literal
5155 19. Ensure that header lines added by a DATA ACL are included in the reject log
5156 if the ACL subsequently rejects the message.
5158 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
5159 MD5 (which is deprecated).
5161 21. When testing a filter file using -bf, Exim was writing a message when it
5162 took the sender from a "From " line in the message, but it was not doing so
5163 when it took $return_path from a Return-Path: header line. It now does.
5165 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
5166 with a valid header line field name (a series of printing characters
5167 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
5169 23. Changed "disc" in the source to "disk" to conform to the documentation and
5170 the book and for uniformity.
5172 24. Ignore Sendmail's -Ooption=value command line item.
5174 25. When execve() failed while trying to run a command in a pipe transport,
5175 Exim was returning EX_UNAVAILBLE (69) from the subprocess. However, this
5176 could be confused with a return value of 69 from the command itself. This
5177 has been changed to 127, the value the shell returns if it is asked to run
5178 a non-existent command. The wording for the related log line suggests a
5179 non-existent command as the problem.
5181 26. If received_header_text expands to an empty string, do not add a Received:
5182 header line to the message. (Well, it adds a token one on the spool, but
5183 marks it "old" so that it doesn't get used or transmitted.)
5185 27. Installed eximstats 1.28 (addition of -nt option).
5187 28. There was no check for failure on the call to getsockname() in the daemon
5188 code. This can fail if there is a shortage of resources on the system, with
5189 ENOMEM, for example. A temporary error is now given on failure.
5191 29. Contrary to the C standard, it seems that in some environments, the
5192 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
5193 program. Exim now does this explicitly; it affects the formatting of
5194 timestamps using strftime().
5196 30. If exiqsumm was given junk data, it threw up some uninitialized variable
5197 complaints. I've now initialized all the variables, to avoid this.
5199 32. Header lines added by a system filter were not being "seen" during
5200 transport-time rewrites.
5202 33. The info_callback() function passed to OpenSSL is set up with type void
5203 (*)(SSL *, int, int), as described somewhere. However, when calling the
5204 function (actually a macro) that sets it up, the type void(*)() is
5205 expected. I've put in a cast to prevent warnings from picky compilers.
5207 34. If a DNS black list lookup found a CNAME record, but there were no A
5208 records associated with the domain it pointed at, Exim crashed.
5210 35. If a DNS black list lookup returned more than one A record, Exim ignored
5211 all but the first. It now scans all returned addresses if a particular IP
5212 value is being sought. In this situation, the contents of the
5213 $dnslist_value variable are a list of all the addresses, separated by a
5216 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
5217 to accept a host name and all its aliases if the forward lookup for any of
5218 them yielded the IP address of the incoming connection. Now it accepts only
5219 those names whose forward lookup yields the correct IP address. Any other
5220 names are discarded. This closes a loophole whereby a rogue DNS
5221 administrator could create reverse DNS records to break through a
5222 wildcarded host restriction in an ACL.
5224 37. If a user filter or a system filter that ran in a subprocess used any of
5225 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
5226 the wrong values were passed to the pipe command ($thisaddress had the
5227 value of $0, $0 had the value of $1, etc). This bug was introduced by
5228 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
5230 38. Improved the line breaking for long SMTP error messages from ACLs.
5231 Previously, if there was no break point between 40 and 75 characters, Exim
5232 left the rest of the message alone. Two changes have been made: (a) I've
5233 reduced the minimum length to 35 characters; (b) if it can't find a break
5234 point between 35 and 75 characters, it looks ahead and uses the first one
5235 that it finds. This may give the occasional overlong line, but at least the
5236 remaining text gets split now.
5238 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
5239 file descriptors might be low, and that setting 1000 would always raise it.
5240 It turns out that in some environments, the limit is already over 1000 and
5241 that lowering it causes trouble. So now Exim takes care not to decrease it.
5243 40. When delivering a message, the value of $return_path is set to $sender_
5244 address at the start of routing (routers may change the value). By an
5245 oversight, this default was not being set up when an address was tested by
5246 -bt or -bv, which affected the outcome if any router or filter referred to
5249 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
5250 log only when "message" or "log_message" are set. However, if one of the
5251 conditions was an address verification, or a call to a nested ACL, the
5252 messages generated by the underlying test were being passed through. This
5253 no longer happens. The underlying message is available in $acl_verify_
5254 message for both "message" and "log_message" expansions, so it can be
5255 passed through if needed.
5257 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
5258 new expansion $bh_ to give the encoded byte string without charset
5259 translation. Translation happens only if iconv() is available; HAVE_ICONV
5260 indicates this at build time. HEADERS_CHARSET gives the charset to
5261 translate to; headers_charset can change it in the configuration, and
5262 "headers charset" can change it in an individual filter file.
5264 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
5265 that creates RFC 2047 encoded "words" labels them as that charset instead
5266 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
5267 expansion operator; (ii) when Exim creates a From: line for a local
5268 message; (iii) when a header line is rewritten to include a "phrase" part.
5270 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
5271 buggy, causing it to skip the first lines of messages whose message ID
5272 ended in 'D'. This would not have bitten before Exim release 4.14, because
5273 message IDs were unlikely to end in 'D' before then. The effect was to have
5274 incorrect size information for certain domains.
5276 45. #include "config.h" was missing at the start of the crypt16.c module. This
5277 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
5280 46. If there was a timeout during a "random" callout check, Exim treated it as
5281 a failure of the random address, and carried on sending RSET and the real
5282 address. If the delay was just some slowness somewhere, the response to the
5283 original RCPT would be taken as a response to RSET and so on, causing
5284 mayhem of various kinds.
5286 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
5287 when I implemented it. It didn't allow for the fact that some option values
5288 may legitimatetly be negative (e.g. size_addition), and it didn't even do
5289 the right test for positive values.
5291 48. Domain names in DNS records are case-independent. Exim always looks them up
5292 in lower case. Some resolvers return domain names in exactly the case they
5293 appear in the zone file, that is, they may contain uppercase letters. Not
5294 all resolvers do this - some return always lower case. Exim was treating a
5295 change of case by a resolver as a change of domain, similar to a widening
5296 of a domain abbreviation. This triggered its re-routing code and so it was
5297 trying to route what was effectively the same domain again. This normally
5298 caused routing to fail (because the router wouldn't handle the domain
5299 twice). Now Exim checks for this case specially, and just changes the
5300 casing of the domain that it ultimately uses when it transmits the message
5303 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
5306 50. If a filter generated a file delivery with a non-absolute name (possible if
5307 no home directory exists for the router), the forbid_file option was not
5310 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
5311 the existing equality matching.
5313 52. Exim was using ints instead of ino_t variables in some places where it was
5314 dealing with inode numbers.
5316 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
5317 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
5318 called TMPDIR, and if it finds it is different, it changes its value.
5320 54. The smtp_printf() function is now made available to local_scan() so
5321 additional output lines can be written before returning. There is also an
5322 smtp_fflush() function to enable the detection of a dropped connection.
5323 The variables smtp_input and smtp_batched_input are exported to
5326 55. Changed the default runtime configuration: the message "Unknown user"
5327 has been removed from the ACL, and instead placed on the localuser router,
5328 using the cannot_route_message feature. This means that any verification
5329 failures that generate their own messages won't get overridden. Similarly,
5330 the "Unrouteable address" message that was in the ACL for unverifiable
5331 relay addresses has also been removed.
5333 56. Added hosts_avoid_esmtp to the smtp transport.
5335 57. The exicyclog script was not checking for the esoteric option
5336 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
5337 will work only if exicyclog is run under the appropriate euid.
5339 58. Following a discussion on the list, the rules by which Exim recognises line
5340 endings on incoming messages have been changed. The -dropcr and drop_cr
5341 options are now no-ops, retained only for backwards compatibility. The
5342 following line terminators are recognized: LF CRLF CR. However, special
5343 processing applies to CR:
5345 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
5346 nor a local message in the state where . is a terminator.
5348 (ii) If a bare CR is encountered in a header line, an extra space is added
5349 after the line terminator so as not to end the header. The reasoning
5350 behind this is that bare CRs in header lines are most likely either
5351 to be mistakes, or people trying to play silly games.
5353 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
5354 was being incorrectly given as 18 bytes larger than it should have been.
5355 This is a VOB (very old bug).
5357 60. This may never have affected anything current, but just in case it has:
5358 When the local host is found other than at the start of a list of hosts,
5359 the local host, those with the same MX, and any that follow, are discarded.
5360 When the list in question was part of a longer list of hosts, the following
5361 hosts (not currently being processed) were also being discarded. This no
5362 longer happens. I'm not sure if this situation could ever has previously
5365 61. Added the "/MX" feature to lists of hosts in the manualroute and query
5368 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
5369 header. This is something that is recommended in a new Internet Draft, and
5370 is something that is documented as being done by Sendmail. There are two
5371 possible values. For messages generated by the autoreply transport, Exim
5374 Auto-Submitted: auto-replied
5376 whereas for all other generated messages (e.g. bounces) it adds
5378 Auto-Submitted: auto-generated
5380 63. The "personal" condition in filters now includes a test for the
5381 Auto-Submitted: header. If it contains the string "auto-" the message it
5382 not considered personal.
5384 64. Added rcpt_include_affixes as a generic transport option.
5386 65. Added queue_only_override (default true).
5388 66. Added the syslog_duplication option.
5390 67. If what should have been the first header line of a message consisted of
5391 a space followed by a colon, Exim was mis-interpreting it as a header line.
5392 It isn't of course - it is syntactically invalid and should therefore be
5393 treated as the start of the message body. The misbehaviour could have
5394 caused a number of strange effects, including loss of data in subsequent
5395 header lines, and spool format errors.
5397 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
5398 client host had authenticated. This control can now be exercised by an ACL
5399 for more flexibility.
5401 69. By default, callouts do not happen when testing with -bh. There is now a
5402 variant, -bhc, which does actually run the callout code, including
5403 consulting and updating the callout cache.
5405 70. Added support for saslauthd authentication, courtesy of Alexander
5408 71. If statvfs() failed on the spool or log directories while checking their
5409 size for availability, Exim confusingly gave the error "space shortage".
5410 Furthermore, in debugging mode it crashed with a floating point exception.
5411 These checks are done if check_{spool,log}_{space,inodes} are set, and when
5412 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
5413 serious problem, Exim now writes to the main and panic logs when this
5414 happens, with details of the failure. It then refuses to accept the
5415 incoming message, giving the message "spool directory problem" or "log
5416 directory problem" with a 421 code for SMTP messages.
5418 72. When Exim is about to re-exec itself, it ensures that the file descriptors
5419 0, 1, and 2 exist, because some OS complain for execs without them (see
5420 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
5421 descriptors. However, the code omitted to check that the open succeeded,
5422 causing mysterious errors if for some reason the permissions on /dev/null
5423 got screwed. Now Exim writes a message to the main and panic logs, and
5424 bombs out if it can't open /dev/null.
5426 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
5427 interact so that it is all more flexible. It is supposed to remain
5428 backwards compatible. Also added extra_local_interfaces.
5430 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
5431 to bomb out with an assertion failure - to the client this appears as a
5432 connection drop. This problem occurs in the part of the code that was taken
5433 from the Samba project. Fortunately, the assertion is in a very simple
5434 function, so I have fixed this by reproducing the function inline in the
5435 one place where it is called, and arranging for authentication to fail
5436 instead of killing the process with assert().
5438 75. The SPA client code was not working when the server requested OEM rather
5439 than Unicode encoding.
5441 76. Added code to make require_files with a specific uid setting more usable in
5442 the case where statting the file as root fails - usually a non-root-mounted
5443 NFS file system. When this happens and the failure is EACCES, Exim now
5444 forks a subprocess and does the per-uid checking as the relevant uid.
5446 77. Added process_log_path.
5448 78. If log_file_path was not explicitly set, a setting of check_log_space or
5449 check_log_inodes was ignored.
5451 79. If a space check for the spool or log partitions fails, the incident is now
5452 logged. Of course, in the latter case the data may get lost...
5454 80. Added the %p formatting code to string_format() so that it can be used to
5455 print addresses in debug_print(). Adjusted all the address printing in the
5456 debugging in store.c to use %p rather than %d.
5458 81. There was a concern that a line of code in smtp_in.c could overflow a
5459 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
5460 initially expressed, the concern was not well-founded, because trailing
5461 spaces are removed early. However, if the trailing spaces were followed by
5462 a NULL, they did not get removed, so the overflow was possible. Two fixes
5465 (a) I re-wrote the offending code in a cleaner fashion.
5466 (b) If an incoming SMTP command contains a NULL character, it is rejected
5469 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
5470 runs initgroups(), so that if the Exim user is in any additional groups,
5471 they will be used during message reception.
5477 The change log for 4.20 and earlier releases has been archived.
projects / exim.git / blob