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 Heiko Schlittermann.
113 PP/01 The new ldap_require_cert option would segfault if used. Fixed.
115 PP/02 Harmonised TLS library version reporting; only show if debugging.
116 Layout now matches that introduced for other libraries in 4.74 PP/03.
118 PP/03 New openssl_options items: no_sslv2 no_sslv3 no_ticket no_tlsv1
120 PP/04 New "dns_use_edns0" global option.
122 PP/05 Don't segfault on misconfiguration of ref:name exim-user as uid.
125 PP/06 Extra paranoia around buffer usage at the STARTTLS transition.
126 nb: Exim is not vulnerable to http://www.kb.cert.org/vuls/id/555316
128 TK/01 Updated PolarSSL code to 0.14.2.
129 Bugzilla 1097. Patch from Andreas Metzler.
131 PP/07 Catch divide-by-zero in ${eval:...}.
134 PP/08 Condition negation of bool{}/bool_lax{} did not negate. Fixed.
137 TK/02 Bugzilla 1106: CVE-2011-1764 - DKIM log line was subject to a
138 format-string attack -- SECURITY: remote arbitrary code execution.
140 TK/03 SECURITY - DKIM signature header parsing was double-expanded, second
141 time unintentionally subject to list matching rules, letting the header
142 cause arbitrary Exim lookups (of items which can occur in lists, *not*
143 arbitrary string expansion). This allowed for information disclosure.
145 PP/09 Fix another SIGFPE (x86) in ${eval:...} expansion, this time related to
146 INT_MIN/-1 -- value coerced to INT_MAX.
152 NM/01 Workround for PCRE version dependency in version reporting
155 TF/01 Update valgrind.h and memcheck.h to copies from valgrind-3.6.0.
156 This fixes portability to compilers other than gcc, notably
157 Solaris CC and HP-UX CC. Fixes Bugzilla 1050.
159 TF/02 Bugzilla 139: Avoid using the += operator in the modular lookup
160 makefiles for portability to HP-UX and POSIX correctness.
162 PP/01 Permit LOOKUP_foo enabling on the make command-line.
163 Also via indented variable definition in the Makefile.
164 (Debugging by Oliver Heesakkers).
166 PP/02 Restore caching of spamd results with expanded spamd_address.
167 Patch from author of expandable spamd_address patch, Wolfgang Breyha.
169 PP/03 Build issue: lookups-Makefile now exports LC_ALL=C
170 Improves build reliability. Fix from: Frank Elsner
172 NM/02 Fix wide character breakage in the rfc2047 coding
173 Fixes bug 1064. Patch from Andrey N. Oktyabrski
175 NM/03 Allow underscore in dnslist lookups
176 Fixes bug 1026. Patch from Graeme Fowler
178 PP/04 Bugzilla 230: Support TLS-enabled LDAP (in addition to ldaps).
179 Code patches from Adam Ciarcinski of NetBSD.
181 NM/04 Fixed exiqgrep to cope with mailq missing size issue
184 PP/05 Bugzilla 1083: when lookup expansion defers, escape the output which
185 is logged, to avoid truncation. Patch from John Horne.
187 PP/06 Bugzilla 1042: implement freeze_signal on pipe transports.
188 Patch from Jakob Hirsch.
190 PP/07 Bugzilla 1061: restrict error messages sent over SMTP to not reveal
191 SQL string expansion failure details.
192 Patch from Andrey Oktyabrski.
194 PP/08 Bugzilla 486: implement %M datestamping in log filenames.
195 Patch from Simon Arlott.
197 PP/09 New lookups functionality failed to compile on old gcc which rejects
198 extern declarations in function scope.
199 Patch from Oliver Fleischmann
201 PP/10 Use sig_atomic_t for flags set from signal handlers.
202 Check getgroups() return and improve debugging.
203 Fixed developed for diagnosis in bug 927 (which turned out to be
206 PP/11 Bugzilla 1055: Update $message_linecount for maildir_tag.
207 Patch from Mark Zealey.
209 PP/12 Bugzilla 1056: Improved spamd server selection.
210 Patch from Mark Zealey.
212 PP/13 Bugzilla 1086: Deal with maildir quota file races.
213 Based on patch from Heiko Schlittermann.
215 PP/14 Bugzilla 1019: DKIM multiple signature generation fix.
216 Patch from Uwe Doering, sign-off by Michael Haardt.
218 NM/05 Fix to spam.c to accommodate older gcc versions which dislike
219 variable declaration deep within a block. Bug and patch from
222 PP/15 lookups-Makefile IRIX compatibilty coercion.
224 PP/16 Make DISABLE_DKIM build knob functional.
226 NM/06 Bugzilla 968: child_open_uid: restore default SIGPIPE handler
227 Patch by Simon Arlott
229 TF/03 Fix valgrind.h portability to C89 compilers that do not support
230 variable argument macros. Our copy now differs from upstream.
236 TF/01 Failure to get a lock on a hints database can have serious
237 consequences so log it to the panic log.
239 TF/02 Log LMTP confirmation messages in the same way as SMTP,
240 controlled using the smtp_confirmation log selector.
242 TF/03 Include the error message when we fail to unlink a spool file.
244 DW/01 Bugzilla 139: Support dynamically loaded lookups as modules.
245 With thanks to Steve Haslam, Johannes Berg & Serge Demonchaux
246 for maintaining out-of-tree patches for some time.
248 PP/01 Bugzilla 139: Documentation and portability issues.
249 Avoid GNU Makefile-isms, let Exim continue to build on BSD.
250 Handle per-OS dynamic-module compilation flags.
252 PP/02 Let /dev/null have normal permissions.
253 The 4.73 fixes were a little too stringent and complained about the
254 permissions on /dev/null. Exempt it from some checks.
255 Reported by Andreas M. Kirchwitz.
257 PP/03 Report version information for many libraries, including
258 Exim version information for dynamically loaded libraries. Created
259 version.h, now support a version extension string for distributors
260 who patch heavily. Dynamic module ABI change.
262 PP/04 CVE-2011-0017 - check return value of setuid/setgid. This is a
263 privilege escalation vulnerability whereby the Exim run-time user
264 can cause root to append content of the attacker's choosing to
267 PP/05 Bugzilla 1041: merged DCC maintainer's fixes for return code.
270 PP/06 Bugzilla 1071: fix delivery logging with untrusted macros.
271 If dropping privileges for untrusted macros, we disabled normal logging
272 on the basis that it would fail; for the Exim run-time user, this is not
273 the case, and it resulted in successful deliveries going unlogged.
274 Fixed. Reported by Andreas Metzler.
280 PP/01 Date: & Message-Id: revert to normally being appended to a message,
281 only prepend for the Resent-* case. Fixes regression introduced in
282 Exim 4.70 by NM/22 for Bugzilla 607.
284 PP/02 Include check_rfc2047_length in configure.default because we're seeing
285 increasing numbers of administrators be bitten by this.
287 JJ/01 Added DISABLE_DKIM and comment to src/EDITME
289 PP/03 Bugzilla 994: added openssl_options main configuration option.
291 PP/04 Bugzilla 995: provide better SSL diagnostics on failed reads.
293 PP/05 Bugzilla 834: provide a permit_coredump option for pipe transports.
295 PP/06 Adjust NTLM authentication to handle SASL Initial Response.
297 PP/07 If TLS negotiated an anonymous cipher, we could end up with SSL but
298 without a peer certificate, leading to a segfault because of an
299 assumption that peers always have certificates. Be a little more
300 paranoid. Problem reported by Martin Tscholak.
302 PP/08 Bugzilla 926: switch ClamAV to use the new zINSTREAM API for content
303 filtering; old API available if built with WITH_OLD_CLAMAV_STREAM=yes
304 NB: ClamAV planning to remove STREAM in "middle of 2010".
305 CL also introduces -bmalware, various -d+acl logging additions and
306 more caution in buffer sizes.
308 PP/09 Implemented reverse_ip expansion operator.
310 PP/10 Bugzilla 937: provide a "debug" ACL control.
312 PP/11 Bugzilla 922: Documentation dusting, patch provided by John Horne.
314 PP/12 Bugzilla 973: Implement --version.
316 PP/13 Bugzilla 752: Refuse to build/run if Exim user is root/0.
318 PP/14 Build without WITH_CONTENT_SCAN. Path from Andreas Metzler.
320 PP/15 Bugzilla 816: support multiple condition rules on Routers.
322 PP/16 Add bool_lax{} expansion operator and use that for combining multiple
323 condition rules, instead of bool{}. Make both bool{} and bool_lax{}
324 ignore trailing whitespace.
326 JJ/02 prevent non-panic DKIM error from being sent to paniclog
328 JJ/03 added tcp_wrappers_daemon_name to allow host entries other than
331 PP/17 Fix malware regression for cmdline scanner introduced in PP/08.
332 Notification from Dr Andrew Aitchison.
334 PP/18 Change ClamAV response parsing to be more robust and to handle ClamAV's
335 ExtendedDetectionInfo response format.
336 Notification from John Horne.
338 PP/19 OpenSSL 1.0.0a compatibility const-ness change, should be backwards
341 PP/20 Added a CONTRIBUTING file. Fixed the documentation build to use http:
342 XSL and documented dependency on system catalogs, with examples of how
345 DW/21 Added Valgrind hooks in store.c to help it capture out-of-bounds store
348 DW/22 Bugzilla 1044: CVE-2010-4345 - partial fix: restrict default behaviour
349 of CONFIGURE_OWNER and CONFIGURE_GROUP options to no longer allow a
350 configuration file which is writeable by the Exim user or group.
352 DW/23 Bugzilla 1044: CVE-2010-4345 - part two: extend checks for writeability
353 of configuration files to cover files specified with the -C option if
354 they are going to be used with root privileges, not just the default
357 DW/24 Bugzilla 1044: CVE-2010-4345 - part three: remove ALT_CONFIG_ROOT_ONLY
358 option (effectively making it always true).
360 DW/25 Add TRUSTED_CONFIG_PREFIX_FILE option to allow alternative configuration
361 files to be used while preserving root privileges.
363 DW/26 Set FD_CLOEXEC on SMTP sockets after forking in the daemon, to ensure
364 that rogue child processes cannot use them.
366 PP/27 Bugzilla 1047: change the default for system_filter_user to be the Exim
367 run-time user, instead of root.
369 PP/28 Add WHITELIST_D_MACROS option to let some macros be overridden by the
370 Exim run-time user without dropping privileges.
372 DW/29 Remove use of va_copy() which breaks pre-C99 systems. Duplicate the
373 result string, instead of calling string_vformat() twice with the same
376 DW/30 Allow TRUSTED_CONFIG_PREFIX_FILE only for Exim or CONFIGURE_OWNER, not
377 for other users. Others should always drop root privileges if they use
378 -C on the command line, even for a whitelisted configure file.
380 DW/31 Turn TRUSTED_CONFIG_PREFIX_FILE into TRUSTED_CONFIG_FILE. No prefixes.
382 NM/01 Fixed bug #1002 - Message loss when using multiple deliveries
388 JJ/01 installed exipick 20100104.1, adding $max_received_linelength,
389 $data_path, and $header_path variables; fixed documentation bugs and
392 JJ/02 installed exipick 20100222.0, added --input-dir and --finput to allow
393 exipick to access non-standard spools, including the "frozen" queue
396 NM/01 Bugzilla 965: Support mysql stored procedures.
397 Patch from Alain Williams
399 NM/02 Bugzilla 961: Spacing fix (syntax error) on Makefile directives for NetBSD
401 NM/03 Bugzilla 955: Documentation fix for max_rcpts.
402 Patch from Andreas Metzler
404 NM/04 Bugzilla 954: Fix for unknown responses from Dovecot authenticator.
405 Patch from Kirill Miazine
407 NM/05 Bugzilla 671: Added umask to procmail example.
409 JJ/03 installed exipick 20100323.0, fixing doc bug
411 NM/06 Bugzilla 988: CVE-2010-2023 - prevent hardlink attack on sticky mail
412 directory. Notification and patch from Dan Rosenberg.
414 TK/01 PDKIM: Upgrade PolarSSL files to upstream version 0.12.1.
416 TK/02 Improve log output when DKIM signing operation fails.
418 MH/01 Treat the transport option dkim_domain as a colon separated
419 list, not as a single string, and sign the message with each element,
420 omitting multiple occurences of the same signer.
422 NM/07 Null terminate DKIM strings, Null initialise DKIM variable
423 Bugzilla 985, 986. Patch by Simon Arlott
425 NM/08 Bugzilla 967. dnsdb DNS TXT record bug fix (DKIM-related)
426 Patch by Simon Arlott
428 PP/01 Bugzilla 989: CVE-2010-2024 - work round race condition on
429 MBX locking. Notification from Dan Rosenberg.
435 TK/01 Bugzilla 912: Fix DKIM segfault on empty headers/body.
437 NM/01 Bugzilla 913: Documentation fix for gnutls_* options.
439 NM/02 Bugzilla 722: Documentation for randint. Better randomness defaults.
441 NM/03 Bugzilla 847: Enable DNSDB lookup by default.
443 NM/04 Bugzilla 915: Flag broken perl installation during build.
449 TK/01 Added patch by Johannes Berg that expands the main option
450 "spamd_address" if it starts with a dollar sign.
452 TK/02 Write list of recipients to X-Envelope-Sender header when building
453 the mbox-format spool file for content scanning (suggested by Jakob
456 TK/03 Added patch by Wolfgang Breyha that adds experimental DCC
457 (http://www.dcc-servers.net/) support via dccifd. Activated by
458 setting EXPERIMENTAL_DCC=yes in Local/Makefile.
460 TK/04 Bugzilla 673: Add f-protd malware scanner support. Patch submitted
461 by Mark Daniel Reidel <mr@df.eu>.
463 NM/01 Bugzilla 657: Embedded PCRE removed from the exim source tree.
464 When building exim an external PCRE library is now needed -
465 PCRE is a system library on the majority of modern systems.
466 See entry on PCRE_LIBS in EDITME file.
468 NM/02 Bugzilla 646: Removed unwanted C/R in Dovecot authenticator
469 conversation. Added nologin parameter to request.
470 Patch contributed by Kirill Miazine.
472 TF/01 Do not log submission mode rewrites if they do not change the address.
474 TF/02 Bugzilla 662: Fix stack corruption before exec() in daemon.c.
476 NM/03 Bugzilla 602: exicyclog now handles panic log, and creates empty
477 log files in place. Contributed by Roberto Lima.
479 NM/04 Bugzilla 667: Close socket used by dovecot authenticator.
481 TF/03 Bugzilla 615: When checking the local_parts router precondition
482 after a local_part_suffix or local_part_prefix option, Exim now
483 does not use the address's named list lookup cache, since this
484 contains cached lookups for the whole local part.
486 NM/05 Bugzilla 521: Integrated SPF Best Guess support contributed by
487 Robert Millan. Documentation is in experimental-spec.txt.
489 TF/04 Bugzilla 668: Fix parallel build (make -j).
491 NM/05.2 Bugzilla 437: Prevent Maildir aux files being created with mode 000.
493 NM/05.3 Bugzilla 598: Improvement to Dovecot authenticator handling.
494 Patch provided by Jan Srzednicki.
496 TF/05 Leading white space used to be stripped from $spam_report which
497 wrecked the formatting. Now it is preserved.
499 TF/06 Save $spam_score, $spam_bar, and $spam_report in spool files, so
500 that they are available at delivery time.
502 TF/07 Fix the way ${extract is skipped in the untaken branch of a conditional.
504 TF/08 TLS error reporting now respects the incoming_interface and
505 incoming_port log selectors.
507 TF/09 Produce a more useful error message if an SMTP transport's hosts
508 setting expands to an empty string.
510 NM/06 Bugzilla 744: EXPN did not work under TLS.
511 Patch provided by Phil Pennock.
513 NM/07 Bugzilla 769: Extraneous comma in usage fprintf
514 Patch provided by Richard Godbee.
516 NM/08 Fixed erroneous documentation references to smtp_notquit_acl to be
517 acl_smtp_notquit, added index entry.
519 NM/09 Bugzilla 787: Potential buffer overflow in string_format.
520 Patch provided by Eugene Bujak.
522 NM/10 Bugzilla 770: Problem on some platforms modifying the len parameter to
523 accept(). Patch provided by Maxim Dounin.
525 NM/11 Bugzilla 749: Preserve old behaviour of blanks comparing equal to zero.
526 Patch provided by Phil Pennock.
528 NM/12 Bugzilla 497: Correct behaviour of exiwhat when no config exists.
530 NM/13 Bugzilla 590: Correct handling of Resent-Date headers.
531 Patch provided by Brad "anomie" Jorsch.
533 NM/14 Bugzilla 622: Added timeout setting to transport filter.
534 Patch provided by Dean Brooks.
536 TK/05 Add native DKIM support (does not depend on external libraries).
538 NM/15 Bugzilla 854: Removed code that symlinks to pcre as its no longer useful.
539 Patch provided by Graeme Fowler.
541 NM/16 Bugzilla 851: Documentation example syntax fix.
543 NM/17 Changed NOTICE file to remove references to embedded PCRE.
545 NM/18 Bugzilla 894: Fix issue with very long lines including comments in
548 NM/19 Bugzilla 745: TLS version reporting.
549 Patch provided by Phil Pennock.
551 NM/20 Bugzilla 167: bool: condition support.
552 Patch provided by Phil Pennock.
554 NM/21 Bugzilla 665: gnutls_compat_mode to allow compatibility with broken
555 clients. Patch provided by Phil Pennock.
557 NM/22 Bugzilla 607: prepend (not append) Resent-Message-ID and Resent-Date.
558 Patch provided by Brad "anomie" Jorsch.
560 NM/23 Bugzilla 687: Fix misparses in eximstats.
561 Patch provided by Heiko Schlittermann.
563 NM/24 Bugzilla 688: Fix exiwhat to handle log_selector = +pid.
564 Patch provided by Heiko Schlittermann.
566 NM/25 Bugzilla 727: Use transport mode as default mode for maildirsize file.
567 plus update to original patch.
569 NM/26 Bugzilla 799: Documentation correction for ratelimit.
571 NM/27 Bugzilla 802: Improvements to local interface IP addr detection.
572 Patch provided by David Brownlee.
574 NM/28 Bugzilla 807: Improvements to LMTP delivery logging.
576 NM/29 Bugzilla 862, 866, 875: Documentation bugfixes.
578 NM/30 Bugzilla 888: TLS documentation bugfixes.
580 NM/31 Bugzilla 896: Dovecot buffer overrun fix.
582 NM/32 Bugzilla 889: Change all instances of "expr" in shell scripts to "expr --"
583 Unlike the original bugzilla I have changed all shell scripts in src tree.
585 NM/33 Bugzilla 898: Transport filter timeout fix.
586 Patch by Todd Rinaldo.
588 NM/34 Bugzilla 901: Fix sign/unsigned and UTF mistmatches.
589 Patch by Serge Demonchaux.
591 NM/35 Bugzilla 39: Base64 decode bug fixes.
592 Patch by Jakob Hirsch.
594 NM/36 Bugzilla 909: Correct connect() call in dcc code.
596 NM/37 Bugzilla 910: Correct issue with relaxed/simple handling.
598 NM/38 Bugzilla 908: Removed NetBSD3 support as no longer needed.
600 NM/39 Bugzilla 911: Fixed MakeLinks build script.
606 TK/01 Add preliminary DKIM support. Currently requires a forked version of
607 ALT-N's libdkim that I have put here:
608 http://duncanthrax.net/exim-experimental/
610 Note to Michael Haardt: I had to rename some vars in sieve.c. They
611 were called 'true' and it seems that C99 defines that as a reserved
612 keyword to be used with 'bool' variable types. That means you could
613 not include C99-style headers which use bools without triggering
614 build errors in sieve.c.
616 NM/01 Bugzilla 592: --help option is handled incorrectly if exim is invoked
617 as mailq or other aliases. Changed the --help handling significantly
618 to do whats expected. exim_usage() emits usage/help information.
620 SC/01 Added the -bylocaldomain option to eximstats.
622 NM/02 Bugzilla 619: Defended against bad data coming back from gethostbyaddr.
624 NM/03 Bugzilla 613: Documentation fix for acl_not_smtp.
626 NM/04 Bugzilla 628: PCRE update to 7.4 (work done by John Hall).
632 PH/01 Another patch from the Sieve maintainer.
634 PH/02 When an IPv6 address is converted to a string for single-key lookup
635 in an address list (e.g. for an item such as "net24-dbm;/net/works"),
636 dots are used instead of colons so that keys in lsearch files need not
637 contain colons. This was done some time before quoting was made available
638 in lsearch files. However, iplsearch files do require colons in IPv6 keys
639 (notated using the quote facility) so as to distinguish them from IPv4
640 keys. This meant that lookups for IP addresses in host lists did not work
641 for iplsearch lookups.
643 This has been fixed by arranging for IPv6 addresses to be expressed with
644 colons if the lookup type is iplsearch. This is not incompatible, because
645 previously such lookups could never work.
647 The situation is now rather anomolous, since one *can* have colons in
648 ordinary lsearch keys. However, making the change in all cases is
649 incompatible and would probably break a number of configurations.
651 TK/01 Change PRVS address formatting scheme to reflect latests BATV draft
654 MH/01 The "spam" ACL condition code contained a sscanf() call with a %s
655 conversion specification without a maximum field width, thereby enabling
656 a rogue spamd server to cause a buffer overflow. While nobody in their
657 right mind would setup Exim to query an untrusted spamd server, an
658 attacker that gains access to a server running spamd could potentially
659 exploit this vulnerability to run arbitrary code as the Exim user.
661 TK/02 Bugzilla 502: Apply patch to make the SPF-Received: header use
662 $primary_hostname instead of what libspf2 thinks the hosts name is.
664 MH/02 The dsearch lookup now uses lstat(2) instead of stat(2) to look for
665 a directory entry by the name of the lookup key. Previously, if a
666 symlink pointed to a non-existing file or a file in a directory that
667 Exim lacked permissions to read, a lookup for a key matching that
668 symlink would fail. Now it is enough that a matching directory entry
669 exists, symlink or not. (Bugzilla 503.)
671 PH/03 The body_linecount and body_zerocount variables are now exported in the
674 PH/04 Added the $dnslist_matched variable.
676 PH/05 Unset $tls_cipher and $tls_peerdn before making a connection as a client.
677 This means they are set thereafter only if the connection becomes
680 PH/06 Added the client_condition to authenticators so that some can be skipped
681 by clients under certain conditions.
683 PH/07 The error message for a badly-placed control=no_multiline_responses left
684 "_responses" off the end of the name.
686 PH/08 Added -Mvc to output a copy of a message in RFC 2822 format.
688 PH/09 Tidied the code for creating ratelimiting keys, creating them explicitly
689 (without spaces) instead of just copying the configuration text.
691 PH/10 Added the /noupdate option to the ratelimit ACL condition.
693 PH/11 Added $max_received_linelength.
695 PH/12 Added +ignore_defer and +include_defer to host lists.
697 PH/13 Installed PCRE version 7.2. This needed some changes because of the new
698 way in which PCRE > 7.0 is built.
700 PH/14 Implemented queue_only_load_latch.
702 PH/15 Removed an incorrect (int) cast when reading the value of SIZE in a
703 MAIL command. The effect was to mangle the value on 64-bit systems.
705 PH/16 Another patch from the Sieve maintainer.
707 PH/17 Added the NOTQUIT ACL, based on a patch from Ted Cooper.
709 PH/18 If a system quota error occurred while trying to create the file for
710 a maildir delivery, the message "Mailbox is full" was not appended to the
711 bounce if the delivery eventually timed out. Change 4.67/27 below applied
712 only to a quota excession during the actual writing of the file.
714 PH/19 It seems that peer DN values may contain newlines (and other non-printing
715 characters?) which causes problems in log lines. The DN values are now
716 passed through string_printing() before being added to log lines.
718 PH/20 Added the "servers=" facility to MySQL and PostgreSQL lookups. (Oracle
719 and InterBase are left for another time.)
721 PH/21 Added message_body_newlines option.
723 PH/22 Guard against possible overflow in moan_check_errorcopy().
725 PH/23 POSIX allows open() to be a macro; guard against that.
727 PH/24 If the recipient of an error message contained an @ in the local part
728 (suitably quoted, of course), incorrect values were put in $domain and
729 $local_part during the evaluation of errors_copy.
735 MH/01 Fix for bug #448, segfault in Dovecot authenticator when interface_address
736 is unset (happens when testing with -bh and -oMi isn't used). Thanks to
739 PH/01 Added a new log selector smtp_no_mail, to log SMTP sessions that do not
740 issue a MAIL command.
742 PH/02 In an ACL statement such as
744 deny dnslists = X!=127.0.0.2 : X=127.0.0.2
746 if a client was not listed at all, or was listed with a value other than
747 127.0.0.2, in the X list, but was listed with 127.0.0.2 in the Y list,
748 the condition was not true (as it should be), so access was not denied.
749 The bug was that the ! inversion was incorrectly passed on to the second
750 item. This has been fixed.
752 PH/03 Added additional dnslists conditions == and =& which are different from
753 = and & when the dns lookup returns more than one IP address.
755 PH/04 Added gnutls_require_{kx,mac,protocols} to give more control over the
756 cipher suites used by GnuTLS. These options are ignored by OpenSSL.
758 PH/05 After discussion on the list, added a compile time option ENABLE_DISABLE_
759 FSYNC, which compiles an option called disable_fsync that allows for
760 bypassing fsync(). The documentation is heavily laced with warnings.
762 SC/01 Updated eximstats to collate all SpamAssassin rejects into one bucket.
764 PH/06 Some tidies to the infrastructure of the Test Suite that is concerned
765 with the auxiliary C programs that it uses: (1) Arrange for BIND_8_COMPAT
766 to be defined when compiling on OSX (Darwin); (2) Tidies to the Makefile,
767 including adding "make clean"; (3) Added -fPIC when compiling the test
768 dynamically loaded module, to get rid of a warning.
770 MH/02 Fix for bug #451, causing paniclog entries to be written if a bounce
771 message fails, move_frozen_messages = true and ignore_bounce_errors_after
772 = 0s. The bug is otherwise harmless.
774 PH/07 There was a bug in the dovecot authenticator such that the value of
775 $auth1 could be overwritten, and so not correctly preserved, after a
776 successful authentication. This usually meant that the value preserved by
777 the server_setid option was incorrect.
779 PH/08 Added $smtp_count_at_connection_start, deliberately with a long name.
781 PH/09 Installed PCRE release 7.0.
783 PH/10 The acl_not_smtp_start ACL was, contrary to the documentation, not being
784 run for batched SMTP input. It is now run at the start of every message
785 in the batch. While fixing this I discovered that the process information
786 (output by running exiwhat) was not always getting set for -bs and -bS
787 input. This is fixed, and it now also says "batched" for BSMTP.
789 PH/11 Added control=no_pipelining.
791 PH/12 Added $sending_ip_address and $sending_port (mostly Magnus Holmgren's
792 patch, slightly modified), and move the expansion of helo_data till after
793 the connection is made in the smtp transport (so it can use these
796 PH/13 Added ${rfc2047d: to decoded RFC 2047 strings.
798 PH/14 Added log_selector = +pid.
800 PH/15 Flush SMTP output before delaying, unless control=no_delay_flush is set.
802 PH/16 Add ${if forany and ${if forall.
804 PH/17 Added dsn_from option to vary the From: line in DSNs.
806 PH/18 Flush SMTP output before performing a callout, unless control =
807 no_callout_flush is set.
809 PH/19 Change 4.64/PH/36 introduced a bug: when address_retry_include_sender
810 was true (the default) a successful delivery failed to delete the retry
811 item, thus causing premature timeout of the address. The bug is now
814 PH/20 Added hosts_avoid_pipelining to the smtp transport.
816 PH/21 Long custom messages for fakedefer and fakereject are now split up
817 into multiline reponses in the same way that messages for "deny" and
818 other ACL rejections are.
820 PH/22 Applied Jori Hamalainen's speed-up changes and typo fixes to exigrep,
821 with slight modification.
823 PH/23 Applied sieve patches from the maintainer "tracking the latest notify
824 draft, changing the syntax and factoring some duplicate code".
826 PH/24 When the log selector "outgoing_port" was set, the port was shown as -1
827 for deliveries of the second and subsequent messages over the same SMTP
830 PH/25 Applied Magnus Holmgren's patch for ${addresses, ${map, ${filter, and
831 ${reduce, with only minor "tidies".
833 SC/02 Applied Daniel Tiefnig's patch to improve the '($parent) =' pattern match.
835 PH/26 Added a "continue" ACL modifier that does nothing, for the benefit of its
836 expansion side effects.
838 PH/27 When a message times out after an over-quota error from an Exim-imposed
839 quota, the bounce message says "mailbox is full". This message was not
840 being given when it was a system quota that was exceeded. It now should
843 MH/03 Made $recipients available in local_scan(). local_scan() already has
844 better access to the recipient list through recipients_list[], but
845 $recipients can be useful in postmaster-provided expansion strings.
847 PH/28 The $smtp_command and $smtp_command_argument variables were not correct
848 in the case of a MAIL command with additional options following the
849 address, for example: MAIL FROM:<foo@bar> SIZE=1234. The option settings
850 were accidentally chopped off.
852 PH/29 SMTP synchronization checks are implemented when a command is read -
853 there is a check that no more input is waiting when there shouldn't be
854 any. However, for some commands, a delay in an ACL can mean that it is
855 some time before the response is written. In this time, more input might
856 arrive, invalidly. So now there are extra checks after an ACL has run for
857 HELO/EHLO and after the predata ACL, and likewise for MAIL and RCPT when
858 pipelining has not been advertised.
860 PH/30 MH's patch to allow iscntrl() characters to be list separators.
862 PH/31 Unlike :fail:, a custom message specified with :defer: was not being
863 returned in the SMTP response when smtp_return_error_details was false.
866 PH/32 Change the Dovecot authenticator to use read() and write() on the socket
867 instead of the C I/O that was originally supplied, because problems were
870 PH/33 Compile failed with OpenSSL 0.9.8e. This was due to a coding error in
871 Exim which did not show up earlier: it was assuming that a call to
872 SSL_CTX_set_info_callback() might give an error value. In fact, there is
873 no error. In previous releases of OpenSSL, SSL_CTX_set_info_callback()
874 was a macro that became an assignment, so it seemed to work. This has
875 changed to a proper function call with a void return, hence the compile
876 error. Exim's code has been fixed.
878 PH/34 Change HDA_SIZE in oracle.c from 256 to 512. This is needed for 64-bit
881 PH/35 Applied a patch from the Sieve maintainer which fixes a bug in "notify".
883 PH/36 Applied John Jetmore's patch to add -v functionality to exigrep.
885 PH/37 If a message is not accepted after it has had an id assigned (e.g.
886 because it turns out to be too big or there is a timeout) there is no
887 "Completed" line in the log. When some messages of this type were
888 selected by exigrep, they were listed as "not completed". Others were
889 picked up by some special patterns. I have improved the selection
890 criteria to be more general.
892 PH/38 The host_find_failed option in the manualroute router can now be set
893 to "ignore", to completely ignore a host whose IP address cannot be
894 found. If all hosts are ignored, the behaviour is controlled by the new
895 host_all_ignored option.
897 PH/39 In a list of hosts for manualroute, if one item (either because of multi-
898 homing or because of multiple MX records with /mx) generated more than
899 one IP address, and the following item turned out to be the local host,
900 all the secondary addresses of the first item were incorrectly removed
901 from the list, along with the local host and any following hosts (which
902 is what is supposed to happen).
904 PH/40 When Exim receives a message, it writes the login name, uid, and gid of
905 whoever called Exim into the -H file. In the case of the daemon it was
906 behaving confusingly. When first started, it used values for whoever
907 started the daemon, but after a SIGHUP it used the Exim user (because it
908 calls itself on a restart). I have changed the code so that it now always
911 PH/41 (Following a suggestion from Tony Finch) If all the RCPT commands in a
912 message are rejected with the same error (e.g. no authentication or bad
913 sender address), and a DATA command is nevertheless sent (as can happen
914 with PIPELINING or a stupid MUA), the error message that was given to the
915 RCPT commands is included in the rejection of the DATA command. This is
916 intended to be helpful for MUAs that show only the final error to their
919 PH/42 Another patch from the Sieve maintainer.
921 SC/02 Eximstats - Differentiate between permanent and temporary rejects.
922 Eximstats - Fixed some broken HTML links and added missing column headers
924 Eximstats - Fixed Grand Total Summary Domains, Edomains, and Email
925 columns for Rejects, Temp Rejects, Ham, and Spam rows.
927 SC/03 Eximstats - V1.58 Fix to get <> and blackhole to show in edomain tables.
929 PH/43 Yet another patch from the Sieve maintainer.
931 PH/44 I found a way to check for a TCP/IP connection going away before sending
932 the response to the final '.' that terminates a message, but only in the
933 case where the client has not sent further data following the '.'
934 (unfortunately, this is allowed). However, in many cases there won't be
935 any further data because there won't be any more messages to send. A call
936 to select() can be used: if it shows that the input is "ready", there is
937 either input waiting, or the socket has been closed. An attempt to read
938 the next input character can distinguish the two cases. Previously, Exim
939 would have sent an OK response which the client would never have see.
940 This could lead to message repetition. This fix should cure that, at
941 least in a lot of common cases.
943 PH/45 Do not advertise STARTTLS in response to HELP unless it would be
944 advertised in response to EHLO.
950 PH/01 Two more bugs that were introduced by 4.64/PH/07, in addition to the one
951 fixed by 4.65/MH/01 (is this a record?) are fixed:
953 (i) An empty string was always treated as zero by the numeric comparison
954 operators. This behaviour has been restored.
956 (ii) It is documented that the numeric comparison operators always treat
957 their arguments as decimal numbers. This was broken in that numbers
958 starting with 0 were being interpreted as octal.
960 While fixing these problems I realized that there was another issue that
961 hadn't been noticed. Values of message_size_limit (both the global option
962 and the transport option) were treated as octal if they started with 0.
963 The documentation was vague. These values are now always treated as
964 decimal, and I will make that clear in the documentation.
970 TK/01 Disable default definition of HAVE_LINUX_SENDFILE. Clashes with
971 Linux large file support (_FILE_OFFSET_BITS=64) on older glibc
974 MH/01 Don't check that the operands of numeric comparison operators are
975 integers when their expansion is in "skipping" mode (fixes bug
976 introduced by 4.64-PH/07).
978 PH/01 If a system filter or a router generates more than SHRT_MAX (32767)
979 child addresses, Exim now panics and dies. Previously, because the count
980 is held in a short int, deliveries were likely to be lost. As such a
981 large number of recipients for a single message is ridiculous
982 (performance will be very, very poor), I have chosen to impose a limit
983 rather than extend the field.
989 TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
990 leftover -K file (the existence of which was triggered by #402).
991 While we were at it, introduced process PID as part of the -K
992 filename. This should rule out race conditions when creating
995 TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
996 processing considerably. Previous code took too long for large mails,
997 triggering a timeout which in turn triggers #401.
999 TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
1000 in the DK code in transports.c. sendfile() is not really portable,
1001 hence the _LINUX specificness.
1003 TF/01 In the add_headers option to the mail command in an Exim filter,
1004 there was a bug that Exim would claim a syntax error in any
1005 header after the first one which had an odd number of characters
1008 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
1009 callout verification, Exim cached a "reject" for the entire domain. This
1010 is correct for most verifications, but it is not correct for a recipient
1011 verification with use_sender or use_postmaster set, because in that case
1012 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
1013 case of MAIL FROM:<> rejection from other early rejections (e.g.
1014 rejection of HELO). When verifying a recipient using a non-null MAIL
1015 address, the cache is ignored if it shows MAIL FROM:<> rejection.
1016 Whatever the result of the callout, the value of the domain cache is
1017 left unchanged (for any other kind of callout, getting as far as trying
1018 RCPT means that the domain itself is ok).
1020 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
1023 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
1024 manifest itself as EPIPE rather than ECONNECT. When tidying away a
1025 session, the daemon ignores ECONNECT errors and logs others; it now
1026 ignores EPIPE as well.
1028 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
1029 (quoted-printable decoding).
1031 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
1032 later the small subsequent patch to fix an introduced bug.
1034 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
1036 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
1038 PH/08 An error is now given if message_size_limit is specified negative.
1040 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
1041 to be given (somewhat) arbitrary names.
1043 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
1046 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
1047 miscellaneous code fixes
1049 PH/10 Added the log_reject_target ACL modifier to specify where to log
1052 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
1053 hostname. This is wrong, because it relates to the incoming message (and
1054 probably the interface on which it is arriving) and not to the outgoing
1055 callout (which could be using a different interface). This has been
1056 changed to use the value of the helo_data option from the smtp transport
1057 instead - this is what is used when a message is actually being sent. If
1058 there is no remote transport (possible with a router that sets up host
1059 addresses), $smtp_active_hostname is used.
1061 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
1062 tweaks were necessary in order to get it to work (see also 21 below):
1063 (a) The code assumed that strncpy() returns a negative number on buffer
1064 overflow, which isn't the case. Replaced with Exim's string_format()
1066 (b) There were several signed/unsigned issues. I just did the minimum
1067 hacking in of casts. There is scope for a larger refactoring.
1068 (c) The code used strcasecmp() which is not a standard C function.
1069 Replaced with Exim's strcmpic() function.
1070 (d) The code set only $1; it now sets $auth1 as well.
1071 (e) A simple test gave the error "authentication client didn't specify
1072 service in request". It would seem that Dovecot has changed its
1073 interface. Fortunately there's a specification; I followed it and
1074 changed what the client sends and it appears to be working now.
1076 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
1079 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
1080 address A is aliased to B and C, where B exists and C does not. Without
1081 -v the output is "A verified" because verification stops after a
1082 successful redirection if more than one address is generated. However,
1083 with -v the child addresses are also verified. Exim was outputting "A
1084 failed to verify" and then showing the successful verification for C,
1085 with its parentage. It now outputs "B failed to verify", showing B's
1086 parentage before showing the successful verification of C.
1088 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
1089 look up a TXT record in a specific list after matching in a combined
1092 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
1093 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
1094 they consult the DNS. I had assumed they would set it the way they
1095 wanted; and indeed my experiments on Linux seem to show that in some
1096 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
1097 To be on the safe side, however, I have now made the interface to
1098 host_find_byname() similar to host_find_bydns(), with an argument
1099 containing the DNS resolver options. The host_find_byname() function now
1100 sets these options at its start, just as host_find_bydns() does. The smtp
1101 transport options dns_qualify_single and dns_search_parents are passed to
1102 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
1103 of host_find_byname() use the default settings of RES_DEFNAMES
1104 (qualify_single) but not RES_DNSRCH (search_parents).
1106 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
1107 spool_read_header() do less string testing, by means of a preliminary
1108 switch on the second character of optional "-foo" lines. (This is
1109 overdue, caused by the large number of possibilities that now exist.
1110 Originally there were few.) While I was there, I also converted the
1111 str(n)cmp tests so they don't re-test the leading "-" and the first
1112 character, in the hope this might squeeze out yet more improvement.
1114 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
1115 flag allowing group syntax was set by the header_syntax check but not
1116 turned off, possible causing trouble later; (2) The flag was not being
1117 set at all for the header_verify test, causing "group"-style headers to
1118 be rejected. I have now set it in this case, and also caused header_
1119 verify to ignore an empty address taken from a group. While doing this, I
1120 came across some other cases where the code for allowing group syntax
1121 while scanning a header line wasn't quite right (mostly, not resetting
1122 the flag correctly in the right place). These bugs could have caused
1123 trouble for malformed header lines. I hope it is now all correct.
1125 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
1126 with the "reply" argument non-NULL. The code, however (which originally
1127 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
1128 but it didn't always do it. This confused somebody who was copying the
1129 code for some other use. I have removed all the tests.
1131 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
1132 feature that was used to support insecure browsers during the U.S. crypto
1133 embargo. It requires special client support, and Exim is probably the
1134 only MTA that supported it -- and would never use it because real RSA is
1135 always available. This code has been removed, because it had the bad
1136 effect of slowing Exim down by computing (never used) parameters for the
1137 RSA_EXPORT functionality.
1139 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
1140 authenticator to fail if there's a tab character in the incoming data
1141 (there should never be unless someone is messing about, as it's supposed
1142 to be base64-encoded). Also added, on Timo's advice, the "secured" option
1143 if the connection is using TLS or if the remote IP is the same as the
1144 local IP, and the "valid-client-cert option" if a client certificate has
1147 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
1148 authenticators. This can be used for authorization after authentication
1149 succeeds. (In the case of plaintext, it servers for both authentication
1152 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
1153 if any retry times were supplied.
1155 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
1156 connection, where there is no client IP address to check. In this
1157 situation, the verify now always succeeds.
1159 PH/25 Applied John Jetmore's -Mset patch.
1161 PH/26 Added -bem to be like -Mset, but loading a message from a file.
1163 PH/27 In a string expansion for a processed (not raw) header when multiple
1164 headers of the same name were present, leading whitespace was being
1165 removed from all of them, but trailing whitespace was being removed only
1166 from the last one. Now trailing whitespace is removed from each header
1167 before concatenation. Completely empty headers in a concatenation (as
1168 before) are ignored.
1170 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
1171 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
1173 PH/29 [Removed. This was a change that I later backed out, and forgot to
1174 correct the ChangeLog entry (that I had efficiently created) before
1175 committing the later change.]
1177 PH/30 Exim was sometimes attempting to deliver messages that had suffered
1178 address errors (4xx response to RCPT) over the same connection as other
1179 messages routed to the same hosts. Such deliveries are always "forced",
1180 so retry times are not inspected. This resulted in far too many retries
1181 for the affected addresses. The effect occurred only when there were more
1182 hosts than the hosts_max_try setting in the smtp transport when it had
1183 the 4xx errors. Those hosts that it had tried were not added to the list
1184 of hosts for which the message was waiting, so if all were tried, there
1185 was no problem. Two fixes have been applied:
1187 (i) If there are any address or message errors in an SMTP delivery, none
1188 of the hosts (tried or untried) are now added to the list of hosts
1189 for which the message is waiting, so the message should not be a
1190 candidate for sending over the same connection that was used for a
1191 successful delivery of some other message. This seems entirely
1192 reasonable: after all the message is NOT "waiting for some host".
1193 This is so "obvious" that I'm not sure why it wasn't done
1194 previously. Hope I haven't missed anything, but it can't do any
1195 harm, as the worst effect is to miss an optimization.
1197 (ii) If, despite (i), such a delivery is accidentally attempted, the
1198 routing retry time is respected, so at least it doesn't keep
1199 hammering the server.
1201 PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
1202 in ${readsocket because some servers need this prod.
1204 PH/32 Added some extra debug output when updating a wait-xxx database.
1206 PH/33 The hint "could be header name not terminated by colon", which has been
1207 given for certain expansion errors for a long time, was not being given
1208 for the ${if def:h_colon_omitted{... case.
1210 PH/34 The spec says: "With one important exception, whenever a domain list is
1211 being scanned, $domain contains the subject domain." There was at least
1212 one case where this was not true.
1214 PH/35 The error "getsockname() failed: connection reset by peer" was being
1215 written to the panic log as well as the main log, but it isn't really
1216 panic-worthy as it just means the connection died rather early on. I have
1217 removed the panic log writing for the ECONNRESET error when getsockname()
1220 PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
1221 runs only) independently of the message's sender address. This meant
1222 that, if the 4xx error was in fact related to the sender, a different
1223 message to the same recipient with a different sender could confuse
1224 things. In particualar, this can happen when sending to a greylisting
1225 server, but other circumstances could also provoke similar problems.
1226 I have changed the default so that the retry time for these errors is now
1227 based a combination of the sender and recipient addresses. This change
1228 can be overridden by setting address_retry_include_sender=false in the
1231 PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
1232 remote server are returned as part of bounce messages. This was not
1233 happening for LMTP over a pipe (the lmtp transport), but now it is the
1234 same for both kinds of LMTP.
1236 PH/38 Despite being documented as not happening, Exim was rewriting addresses
1237 in header lines that were in fact CNAMEs. This is no longer the case.
1239 PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
1240 and queue runs started by the daemon processed all messages. This has
1241 been fixed so that -R and -S can now usefully be given with -q<time>.
1243 PH/40 Import PCRE release 6.7 (fixes some bugs).
1245 PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
1247 PH/42 Give an error if -q is specified more than once.
1249 PH/43 Renamed the variables $interface_address and $interface_port as
1250 $received_ip_address and $received_port, to make it clear that these
1251 values apply to message reception, and not to the outgoing interface when
1252 a message is delivered. (The old names remain recognized, of course.)
1254 PH/44 There was no timeout on the connect() call when using a Unix domain
1255 socket in the ${readsocket expansion. There now is.
1257 PH/45 Applied a modified version of Brad Jorsch's patch to allow "message" to
1258 be meaningful with "accept".
1260 SC/01 Eximstats V1.43
1261 Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
1263 SC/02 Eximstats V1.44
1264 Use a glob alias rather than an array ref in the generated
1265 parser. This improves both readability and performance.
1267 SC/03 Eximstats V1.45 (Marco Gaiarin / Steve Campbell)
1268 Collect SpamAssassin and rejection statistics.
1269 Don't display local sender or destination tables unless
1270 there is data to show.
1271 Added average volumes into the top table text output.
1273 SC/04 Eximstats V1.46
1274 Collect data on the number of addresses (recipients)
1275 as well as the number of messages.
1277 SC/05 Eximstats V1.47
1278 Added 'Message too big' to the list of mail rejection
1279 reasons (thanks to Marco Gaiarin).
1281 SC/06 Eximstats V1.48
1282 Mainlog lines which have GMT offsets and are too short to
1283 have a flag are now skipped.
1285 SC/07 Eximstats V1.49 (Alain Williams)
1286 Added the -emptyok flag.
1288 SC/08 Eximstats V1.50
1289 Fixes for obtaining the IP address from reject messages.
1291 JJ/03 exipick.20061117.2, made header handling as similar to exim as possible
1292 (added [br]h_ prefixes, implemented RFC2047 decoding. Fixed
1293 whitesspace changes from 4.64-PH/27
1295 JJ/04 exipick.20061117.2, fixed format and added $message_headers_raw to
1298 JJ/05 exipick.20061117.2, bug fixes (error out sooner when invalid criteria
1299 are found, allow negative numbers in numeric criteria)
1301 JJ/06 exipick.20061117.2, added new $message_body_missing variable
1303 JJ/07 exipick.20061117.2, added $received_ip_address and $received_port
1304 to match changes made in 4.64-PH/43
1306 PH/46 Applied Jori Hamalainen's patch to add features to exiqsumm.
1308 PH/47 Put in an explicit test for a DNS lookup of an address record where the
1309 "domain" is actually an IP address, and force a failure. This locks out
1310 those revolvers/nameservers that support "A-for-A" lookups, in
1311 contravention of the specifications.
1313 PH/48 When a host name was looked up from an IP address, and the subsequent
1314 forward lookup of the name timed out, the host name was left in
1315 $sender_host_name, contrary to the specification.
1317 PH/49 Although default lookup types such as lsearch* or cdb*@ have always been
1318 restricted to single-key lookups, Exim was not diagnosing an error if
1319 * or *@ was used with a query-style lookup.
1321 PH/50 Increased the value of DH_BITS in tls-gnu.c from 768 to 1024.
1323 MH/01 local_scan ABI version incremented to 1.1. It should have been updated
1324 long ago, but noone interested enough thought of it. Let's just say that
1325 the "1.1" means that there are some new functions that weren't there at
1326 some point in the past.
1328 PH/51 Error processing for expansion failure of helo_data from an smtp
1329 transport during callout processing was broken.
1331 PH/52 Applied John Jetmore's patch to allow tls-on-connect and STARTTLS to be
1332 tested/used via the -bh/-bhc/-bs options.
1334 PH/53 Added missing "#include <time.h>" to pcre/pcretest.c (this was a PCRE
1335 bug, fixed in subsequent PCRE releases).
1337 PH/54 Applied Robert Bannocks' patch to avoid a problem with references that
1338 arises when using the Solaris LDAP libraries (but not with OpenLDAP).
1340 PH/55 Check for a ridiculously long file name in exim_dbmbuild.
1346 SC/01 Use a glob alias rather than an array ref in eximstats generated
1347 parser. This improves both readability and performance.
1349 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
1350 Don't display local sender or destination tables in eximstats unless
1351 there is data to show.
1352 Added average volumes into the eximstats top table text output.
1354 SC/03 Collect data on the number of addresses (recipients) as well
1355 as the number of messages in eximstats.
1357 TF/01 Correct an error in the documentation for the redirect router. Exim
1358 does (usually) call initgroups() when daemonizing.
1360 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
1361 with consistent privilege compared to when running as a daemon.
1363 TF/03 Note in the spec that $authenticated_id is not set for local
1364 submissions from trusted users.
1366 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
1367 Thanks to Dean Brooks <dean@iglou.com> for the patch.
1369 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
1370 by adding some example configuration directives to the default
1371 configuration file. A little bit of work is required to uncomment the
1372 directives and define how usernames and passwords are checked, but
1373 there is now a framework to start from.
1375 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
1376 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
1377 without this. I don't know how relevant this is to other LDAP libraries.
1379 PH/02 Add the verb name to the "unknown ACL verb" error.
1381 PH/03 Magnus Holmgren's patch for filter_prepend_home.
1383 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
1385 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
1386 directory not expanded when it should be if an expanded home directory
1387 was set for the address (which is overridden by the transport).
1389 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
1392 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
1393 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
1394 because it is too late at that time, and has no effect.
1396 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
1397 security issue with \' (bugzilla #107). I could not use the
1398 PQescapeStringConn() function, because it needs a PGconn value as one of
1401 PH/08 When testing addresses using -bt, indicate those final addresses that
1402 are duplicates that would not cause an additional delivery. At least one
1403 person was confused, thinking that -bt output corresponded to deliveries.
1404 (Suppressing duplicates isn't a good idea as you lose the information
1405 about possibly different redirections that led to the duplicates.)
1407 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
1408 systems where poll() doesn't work, in particular OS X.
1410 PH/10 Added more information to debugging output for retry time not reached.
1412 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
1413 operations in malware.c.
1415 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
1418 PH/13 If write_rejectlog was set false when logging was sent to syslog with
1419 syslog_duplication set false, log lines that would normally be written
1420 both the the main log and to the reject log were not written to syslog at
1423 PH/14 In the default configuration, change the use of "message" in ACL warn
1424 statements to "add_header".
1426 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
1427 not followed by a command (e.g. "seen endif").
1429 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
1430 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
1433 PH/17 Added extra conditions to the default value of delay_warning_condition
1437 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
1438 { match{$h_precedence:}{(?i)bulk|list|junk} } \
1439 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
1442 The Auto-Submitted: and various List- headers are standardised, whereas I
1443 don't think Precedence: ever was.
1445 PH/18 Refactored debugging code in route_finduser() to show more information,
1446 in particular, the error code if getpwnam() issues one.
1448 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
1449 This is apparently needed in addition to the PH/07 change above to avoid
1450 any possible encoding problems.
1452 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
1453 but not after initializing Perl.
1455 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
1456 output them only if debugging. By default they are written stderr,
1457 apparently, which is not desirable.
1459 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
1462 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
1465 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
1467 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
1468 authenticated or an ident call has been made. Suppress the default
1469 values for $authenticated_id and $authenticated_sender (but permit -oMai
1470 and -oMas) when testing with -bh.
1472 PH/24 Re-jigged the order of the tests in the default configuration so that the
1473 tests for valid domains and recipients precede the DNS black list and CSA
1474 tests, on the grounds that those ones are more expensive.
1476 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
1477 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
1478 This bug exists in every version of Exim that I still have, right back to
1481 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
1482 However, an attempt to turn on case-sensitivity in a regex key by
1483 including (?-i) didn't work because the subject string was already
1484 lowercased, and the effects were non-intuitive. It turns out that a
1485 one-line patch can be used to allow (?-i) to work as expected.
1491 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
1492 other effects) broke the use of negated acl sub-conditions.
1494 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
1497 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
1498 "Deny" causes Exim to reject the incoming connection with a 554 error.
1499 Unfortunately, if there is a major crisis, such as a disk failure,
1500 tcp-wrappers gives "deny", whereas what one would like would be some
1501 kind of temporary error. A kludge has been added to help with this.
1502 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
1503 554 error is used if errno is still zero or contains ENOENT (which occurs
1504 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
1507 PH/03 Add -lutil to the default FreeBSD LIBS setting.
1509 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
1510 errors. Otherwise a message that provokes a temporary error (when other
1511 messages do not) can cause a whole host to time out.
1513 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
1514 addresses were routed directly to files or pipes from a redirect router.
1515 File deliveries just didn't batch; pipe deliveries might have suffered
1518 PH/06 A failure to get a lock for a hints database would erroneously always say
1519 "Failed to get write lock", even when it was really a read lock.
1521 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
1522 of 0600. This has been changed to use the value of the lockfile_mode
1523 option (which defaults to 0600).
1525 PH/08 Applied small patch from the Sieve maintainer.
1527 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
1528 folder from quota calculations, a direct delivery into this folder messed
1529 up the contents of the maildirsize file. This was because the regex was
1530 used only to exclude .Trash (or whatever) when the size of the mailbox
1531 was calculated. There was no check that a delivery was happening into an
1532 excluded directory. This bug has been fixed by ignoring all quota
1533 processing for deliveries into excluded directories.
1535 PH/10 Added the maildirfolder_create_regex option to appendfile.
1541 PH/01 The code for finding all the local interface addresses on a FreeBSD
1542 system running IPv6 was broken. This may well have applied to all BSD
1543 systems, as well as to others that have similar system calls. The broken
1544 code found IPv4 interfaces correctly, but gave incorrect values for the
1545 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
1546 that it would not match correctly against @[] and not recognize the IPv6
1549 PH/02 The ipliteral router was not recognizing addresses of the form user@
1550 [ipv6:....] because it didn't know about the "ipv6:" prefix.
1552 PH/03 Added disable_ipv6.
1554 PH/04 Changed $reply_address to use the raw form of the headers instead of the
1555 decoded form, because it is most often used to construct To: headers
1556 lines in autoreplies, and the decoded form may well be syntactically
1557 invalid. However, $reply_address has leading white space removed, and all
1558 newlines turned into spaces so that the autoreply transport does not
1561 PH/05 If group was specified without a user on a router, and no group or user
1562 was specified on a transport, the group from the router was ignored.
1564 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
1565 for visible compile-time settings that can be used to change these
1566 numbers, for those that want even more. Backwards compatibility with old
1567 spool files has been maintained. However, going back to a previous Exim
1568 release will lost any variables that are in spool files.
1570 PH/07 Two small changes when running in the test harness: increase delay when
1571 passing a TCP/IP connection to a new process, in case the original
1572 process has to generate a bounce, and remove special handling of
1573 127.0.0.2 (sic), which is no longer necessary.
1575 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
1576 be the same on different OS.
1578 PH/09 Moved a debug statement in filter processing to avoid a race problem when
1581 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
1582 whether --show-vars was specified or not
1584 JJ/02 exipick: Added support for new ACL variable spool format introduced
1587 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
1588 syntactically invalid From: or Reply-to: line, and a filter used this to
1589 generate an autoreply, and therefore failed to obtain an address for the
1590 autoreply, Exim could try to deliver to a non-existent relative file
1591 name, causing unrelated and misleading errors. What now happens is that
1592 it logs this as a hard delivery error, but does not attempt to create a
1595 PH/11 The exinext utility has a -C option for testing purposes, but although
1596 the given file was scanned by exinext itself; it wasn't being passed on
1597 when Exim was called.
1599 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
1600 an end-of-file indication when reading a command response.
1602 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
1603 compiled. In many other places in Exim, IPv6 addresses are always
1604 recognized, so I have changed this. It also means that IPv4 domain
1605 literals of the form [IPV4:n.n.n.n] are now always recognized.
1607 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
1608 used if the router is not running as root, for example, when verifying at
1609 ACL time, or when using -bh. The debugging output from this situation was
1610 non-existent - all you got was a failure to exec. I have made two
1613 (a) Failures to set uid/gid, the current directory, or a process leader
1614 in a subprocess such as that created by queryprogram now generate
1615 suitable debugging ouput when -d is set.
1617 (b) The queryprogram router detects when it is not running as root,
1618 outputs suitable debugging information if -d is set, and then runs
1619 the subprocess without attempting to change uid/gid.
1621 PH/15 Minor change to Makefile for building test_host (undocumented testing
1624 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
1625 additional section of a DNS packet that returns MX or SRV records.
1626 Instead, it always explicitly searches for A/AAAA records. This avoids
1627 major problems that occur when a DNS server includes only records of one
1628 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
1629 fixed another bug: if SRV records were looked up and the corresponding
1630 address records were *not* found in the additional section, the port
1631 values from the SRV records were lost.
1633 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
1634 using the correct key (the original address) when searching the retry
1635 rules in order to find which one to use for generating the retry hint.
1637 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
1638 adding the default one. Similarly, if it contains a Reply-To: header, the
1639 errors_reply_to option, if set, is not used.
1641 PH/19 When calculating a retry time, Exim used to measure the "time since
1642 failure" by looking at the "first failed" field in the retry record. Now
1643 it does not use this if it is later than than the arrival time of the
1644 message. Instead it uses the arrival time. This makes for better
1645 behaviour in cases where some deliveries succeed, thus re-setting the
1646 "first failed" field. An example is a quota failure for a huge message
1647 when small messages continue to be delivered. Without this change, the
1648 "time since failure" will always be short, possible causing more frequent
1649 delivery attempts for the huge message than are intended.
1650 [Note: This change was subsequently modified - see PH/04 for 4.62.]
1652 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
1653 $1, $2, $3) because the numerical variables can be reset during some
1654 expansion items (e.g. "match"), thereby losing the authentication data.
1656 PH/21 Make -bV show the size of off_t variables so that the test suite can
1657 decide whether to run tests for quotas > 2G.
1659 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
1660 mailbox_size, and mailbox_filecount in the appendfile transport. If a
1661 filecount value is greater than 2G or if a quota value is greater than 2G
1662 on a system where the size of off_t is not greater than 4, a panic error
1665 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
1666 never match. The debug and -bh output now contains an explicit error
1667 message indicating a malformed IPv4 address or mask.
1669 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
1670 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
1671 PH/23 above applies.
1673 PH/25 Do not write to syslog when running in the test harness. The only
1674 occasion when this arises is a failure to open the main or panic logs
1675 (for which there is an explicit test).
1677 PH/26 Added the /no_tell option to "control=freeze".
1679 PH/27 If a host name lookup failed very early in a connection, for example, if
1680 the IP address matched host_lookup and the reverse lookup yielded a name
1681 that did not have a forward lookup, an error message of the form "no IP
1682 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
1683 could be logged. Now it outputs the IP address instead of "NULL".
1685 PH/28 An enabling patch from MH: add new function child_open_exim2() which
1686 allows the sender and the authenticated sender to be set when
1687 submitting a message from within Exim. Since child_open_exim() is
1688 documented for local_scan(), the new function should be too.
1690 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
1691 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
1692 results in an empty string is now treated as unset.
1694 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
1696 PH/31 Added sender_verify_fail logging option.
1698 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
1699 needed by RFC 822 but not by RFC 2822 was commented out. I have now
1700 tidied the source and removed it altogether.
1702 PH/33 When a queue run was abandoned because the load average was too high, a
1703 log line was always written; now it is written only if the queue_run log
1704 selector is set. In addition, the log line for abandonment now contains
1705 information about the queue run such as the pid. This is always present
1706 in "start" and "stop" lines but was omitted from the "abandon" line.
1708 PH/34 Omit spaces between a header name and the colon in the error message that
1709 is given when verify = headers_syntax fails (if there are lots of them,
1710 the message gets confusing).
1712 PH/35 Change the default for dns_check_names_pattern to allow slashes within
1713 names, as there are now some PTR records that contain slashes. This check
1714 is only to protect against broken name servers that fall over on strange
1715 characters, so the fact that it applies to all lookups doesn't matter.
1717 PH/36 Now that the new test suite is complete, we can remove some of the
1718 special code in Exim that was needed for the old test suite. For example,
1719 sorting DNS records because real resolvers return them in an arbitrary
1720 order. The new test suite's fake resolver always returns records in the
1723 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
1724 bounces) except when queue_only is set, to avoid logging races between
1725 the different processes.
1727 PH/38 Panic-die if .include specifies a non-absolute path.
1729 PH/39 A tweak to the "H" retry rule from its user.
1731 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
1732 a label. They prevented compilation on older perls.
1734 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
1735 a warning to be raised on newish perls.
1737 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
1738 on queue. Changes to match documented behaviour of showing count of
1739 messages matching specified criteria.
1741 PH/40 Changed the default ident timeout from 30s to 5s.
1743 PH/41 Added support for the use of login_cap features, on those BSD systems
1744 that have them, for controlling the resources used by pipe deliveries.
1746 PH/42 The content-scanning code uses fopen() to create files in which to put
1747 message data. Previously it was not paying any attention to the mode of
1748 the files. Exim runs with umask(0) because the rest of the code creates
1749 files with open(), and sets the required mode explicitly. Thus, these
1750 files were ending up world-writeable. This was not a big issue, because,
1751 being within the spool directory, they were not world-accessible. I have
1752 created a function called modefopen, which takes an additional mode
1753 argument. It sets umask(777), creates the file, chmods it to the required
1754 mode, then resets the umask. All the relevant calls to fopen() in the
1755 content scanning code have been changed to use this function.
1757 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
1758 to 24 hours. This avoids potential overflow problems when processing G
1759 and H retry rules. I suspect nobody ever tinkers with this value.
1761 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
1763 PH/45 When the plaintext authenticator is running as a client, the server's
1764 challenges are checked to ensure they are valid base64 strings. By
1765 default, the authentication attempt is cancelled if an invalid string is
1766 received. Setting client_ignore_invalid_base64 true ignores these errors.
1767 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
1768 they are received. Thus, the responses can be made to depend on the
1769 challenges. If an invalid string is ignored, an empty string is placed in
1772 PH/46 Messages that are created by the autoreply transport now contains a
1773 References: header, in accordance with RFCs 2822 and 3834.
1775 PH/47 Added authenticated_sender_force to the smtp transport.
1777 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
1779 PH/49 Installed latest patch from the Sieve maintainer.
1781 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
1782 was also set, the appendfile transport was unnecessarily scanning a
1783 directory of message files (e.g. for maildir delivery) to find the count
1784 of files (along with the size), even though it did not need this
1785 information. It now does the scan only if it needs to find either the
1786 size of the count of files.
1788 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
1790 PH/52 Two bugs concerned with error handling when the smtp transport is
1793 (i) Exim was not creating retry information for temporary errors given
1794 for individual recipients after the DATA command when the smtp transport
1795 was used in LMTP mode. This meant that they could be retried too
1796 frequently, and not timed out correctly.
1798 (ii) Exim was setting the flag that allows error details to be returned
1799 for LMTP errors on RCPT commands, but not for LMTP errors for individual
1800 recipients that were returned after the DATA command.
1802 PH/53 This is related to PH/52, but is more general: for any failing address,
1803 when detailed error information was permitted to be returned to the
1804 sender, but the error was temporary, then after the final timeout, only
1805 "retry timeout exceeded" was returned. Now it returns the full error as
1806 well as "retry timeout exceeded".
1808 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
1809 do this, and (what is worse) MTAs that accept it.
1811 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
1812 will now be deprecated.
1814 PH/56 New os.c-cygwin from the Cygwin maintainer.
1816 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
1817 formats (previously only available in exim formats via -bpr, -bpru,
1818 and -bpra. Now also available in native and exiqgrep formats)
1820 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
1821 with very large, slow to parse queues
1823 JJ/08 exipick: added ! as generic prefix to negate any criteria format
1825 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
1827 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
1828 responses to authentication challenges, though it was showing the
1829 challenges; (ii) I've removed the CR characters from the debug output for
1832 PH/58 Allow for the insertion of a newline as well as a space when a string
1833 is turned into more than one encoded-word during RFC 2047 encoding. The
1834 Sieve code now uses this.
1836 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
1837 data_4xx, lost_connection, tls_required.
1839 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
1840 message was being sent as an SMTP response.
1842 PH/61 Add -l and -k options to exicyclog.
1844 PH/62 When verifying, if an address was redirected to one new address, so that
1845 verification continued, and the new address failed or deferred after
1846 having set something in $address_data, the value of $address_data was not
1847 passed back to the ACL. This was different to the case when no
1848 redirection occurred. The value is now passed back in both cases.
1850 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
1851 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
1852 use login_cap.h, so on its own it isn't the distinguishing feature. The
1853 new name refers directly to the setclassresources() function.
1855 PH/65 Added configuration files for NetBSD3.
1857 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
1859 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
1860 is preferred over IPv4.
1862 PH/68 The bounce_return_message and bounce_return_body options were not being
1863 honoured for bounces generated during the reception of non-SMTP messages.
1864 In particular, this applied to messages rejected by the ACL. This bug has
1865 been fixed. However, if bounce_return_message is true and bounce_return_
1866 body is false, the headers that are returned for a non-SMTP message
1867 include only those that have been read before the error was detected.
1868 (In the case of an ACL rejection, they have all been read.)
1870 PH/69 The HTML version of the specification is now built in a directory called
1871 spec_html instead of spec.html, because the latter looks like a path with
1872 a MIME-type, and this confuses some software.
1874 PH/70 Catch two compiler warnings in sieve.c.
1876 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
1877 function verify_get_ident() calls ip_connect() to connect a socket, but
1878 if the "connect()" function timed out, ip_connect() used to close the
1879 socket. However, verify_get_ident() also closes the socket later, and in
1880 between Exim writes to the log, which may get opened at this point. When
1881 the socket was closed in ip_connect(), the log could get the same file
1882 descriptor number as the socket. This naturally causes chaos. The fix is
1883 not to close the socket in ip_connect(); the socket should be closed by
1884 the function that creates it. There was only one place in the code where
1885 this was missing, in the iplookup router, which I don't think anybody now
1886 uses, but I've fixed it anyway.
1888 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
1889 well as to direct DNS lookups. Otherwise the handling of names in host
1890 lists is inconsistent and therefore confusing.
1896 PH/01 Two changes to the default runtime configuration:
1898 (1) Move the checks for relay_from_hosts and authenticated clients from
1899 after to before the (commented out) DNS black list checks.
1901 (2) Add control=submission to the relay_from_hosts and authenticated
1902 clients checks, on the grounds that messages accepted by these
1903 statements are most likely to be submissions.
1905 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
1907 (1) Generate an error if the third argument for the ${prvs expansion is
1910 (2) Treat a missing third argument of ${prvscheck as if it were an empty
1913 (3) Reset the variables that are obtained from the first argument of
1914 ${prvscheck and used in the second argument before leaving the code,
1915 because their memory is reclaimed, so using them afterwards may do
1918 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
1919 one (it's much easier than Tom thought :-).
1921 (5) Because of (4), we can now allow for the use of $prvscheck_result
1922 inside the third argument.
1924 PH/03 For some reason, the default setting of PATH when running a command from
1925 a pipe transport was just "/usr/bin". I have changed it to
1928 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
1929 anything to be listed in the output from -bV.
1931 PH/05 When a filter generated an autoreply, the entire To: header line was
1932 quoted in the delivery log line, like this:
1934 => >A.N.Other <ano@some.domain> <original@ddress> ...
1936 This has been changed so that it extracts the operative address. There
1937 may be more than one such address. If so, they are comma-separated, like
1940 => >ano@some.domain,ona@other.domain <original@ddress> ...
1942 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
1943 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
1944 not being looked up in the rDNS to get a host name, Exim was showing the
1945 IP address twice in Received: lines, even though the IP addresses were
1946 identical. For example:
1948 Received: from [1.2.3.4] (helo=[1.2.3.4])
1950 However, if the real host name was known, it was omitting the HELO data
1951 if it matched the actual IP address. This has been tidied up so that it
1952 doesn't show the same IP address twice.
1954 PH/07 When both +timestamp and +memory debugging was on, the value given by
1955 $tod_xxx expansions could be wrong, because the tod_stamp() function was
1956 called by the debug printing, thereby overwriting the timestamp buffer.
1957 Debugging no longer uses the tod_stamp() function when +timestamp is set.
1959 PH/08 When the original message was included in an autoreply transport, it
1960 always said "this is a copy of the message, including all the headers",
1961 even if body_only or headers_only was set. It now gives an appropriate
1964 PH/09 Applied a patch from the Sieve maintainer which:
1966 o fixes some comments
1967 o adds the (disabled) notify extension core
1968 o adds some debug output for the result of if/elsif tests
1969 o points to the current vacation draft in the documentation
1970 and documents the missing references header update
1974 o fixes a bug in processing the envelope test (when testing
1975 multiple envelope elements, the last element determinted the
1978 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
1979 Electronic Mail") by including:
1981 Auto-submitted: auto-generated
1983 in the messages that it generates (bounce messages and others, such as
1984 warnings). In the case of bounce messages for non-SMTP mesages, there was
1985 also a typo: it was using "Auto_submitted" (underscore instead of
1986 hyphen). Since every message generated by Exim is necessarily in response
1987 to another message, thes have all been changed to:
1989 Auto-Submitted: auto-replied
1991 in accordance with these statements in the RFC:
1993 The auto-replied keyword:
1995 - SHOULD be used on messages sent in direct response to another
1996 message by an automatic process,
1998 - MUST NOT be used on manually-generated messages,
2000 - MAY be used on Delivery Status Notifications (DSNs) and Message
2001 Disposition Notifications (MDNs),
2003 - MUST NOT be used on messages generated by automatic or periodic
2004 processes, except for messages which are automatic responses to
2007 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
2008 to the default Received: header definition.
2010 PH/12 Added log selector acl_warn_skipped (default on).
2012 PH/13 After a successful wildlsearch lookup, discard the values of numeric
2013 variables because (a) they are in the wrong storage pool and (b) even if
2014 they were copied, it wouldn't work properly because of the caching.
2016 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
2017 checking when decoding. Apparently there are clients that generate
2018 overlong encoded strings. Why am I not surprised?
2020 PH/15 If the first argument of "${if match_address" was not empty, but did not
2021 contain an "@" character, Exim crashed. Now it writes a panic log message
2022 and treats the condition as false.
2024 PH/16 In autoreply, treat an empty string for "once" the same as unset.
2026 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
2027 extension "envelope-auth". The code is finished and in agreement with
2028 other implementations, but there is no documentation so far and in fact,
2029 nobody wrote the draft yet. This extension is currently #undef'ed, thus
2030 not changing the active code.
2032 Print executed "if" and "elsif" statements when debugging is used. This
2033 helps a great deal to understand what a filter does.
2035 Document more things not specified clearly in RFC3028. I had all this
2036 sorted out, when out of a sudden new issues came to my mind. Oops."
2038 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
2041 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
2042 canonical form (as documented). However, after a host name lookup from
2043 the IP address, check_host() was doing a simple string comparison with
2044 addresses acquired from the DNS when checking that the found name did
2045 have the original IP as one of its addresses. Since any found IPv6
2046 addresses are likely to be in abbreviated form, the comparison could
2047 fail. Luckily, there already exists a function for doing the comparison
2048 by converting both addresses to binary, so now that is used instead of
2049 the text comparison.
2051 PH/20 There was another similar case to PH/19, when a complete host name was
2052 given in a host list; looking up its IP address could give an abbreviated
2053 form, whereas the current host's name might or might not be abbreviated.
2054 The same fix has been applied.
2060 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
2061 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
2064 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
2065 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
2067 PH/03 Typo: missing ".o" in src/pcre/Makefile.
2069 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
2070 header line, restrict the check to what is listed in RFCs 2369 and 2929.
2071 Also, for "Auto-Submitted", treat anything other than "no" as
2072 non-personal, in accordance with RFC 3834. (Previously it treated
2073 anything starting "auto-" as non-personal.)
2075 TF/01 The control=submission/name=... option had a problem with syntax
2076 errors if the name included a slash character. The /name= option
2077 now slurps the rest of the string, so it can include any characters
2078 but it must come last in the list of options (after /sender_retain
2081 PH/05 Some modifications to the interface to the fake nameserver for the new
2089 TK/01 Added the "success_on_redirect" address verification option. See
2090 NewStuff for rationale and an example.
2092 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
2094 PH/02 Patch to exigrep to allow it to work on syslog lines.
2096 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
2097 fread() to skip over the body file's header line, because in Cygwin the
2098 header line is locked and is inaccessible.
2100 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
2101 co-exist for some time) to make it clear that it is the Exim ID that is
2102 referenced, not the Message-ID: header line.
2104 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
2105 string_format() function, because snprintf() does not exist on all
2108 PH/06 The use of forbid_filter_existstest now also locks out the use of the
2109 ${stat: expansion item.
2111 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
2112 protocol synchronization error", to keep the pedants happy.
2114 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
2115 well as for IRIX systems, when gcc is being used. See the host.c source
2118 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
2120 PH/10 Named domain lists were not working if used in a queue_smtp_domains
2123 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
2124 transport and to the smtp transport in LMTP mode.
2126 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
2128 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
2129 run a filter in a subprocess. This could lead to confusion in subsequent
2130 lookups in the parent process. There should also be a search_tidyup() at
2131 the end of the subprocess.
2133 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
2134 only if the host matched helo_try_verify_hosts, which caused the
2135 verification to occur when the EHLO/HELO command was issued. The ACL just
2136 tested the remembered result. Now, if a previous verification attempt has
2137 not happened, "verify = helo" does it there and then.
2139 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
2141 TK/03 Fix log output including CR from clamd.
2143 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
2144 exist provoked a memory error which could cause a segfault.
2146 PH/15 Installed PCRE 6.2
2148 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
2150 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
2151 of the problem. Specifically, suggested +O2 rather than +O1 for the
2154 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
2156 PH/20 If a delivery was routed to a non-standard port by means of an SRV
2157 record, the port was not correctly logged when the outgoing_port log
2158 selector was set (it logged the transort's default port).
2160 PH/21 Added support for host-specific ports to manualroute, queryprogram,
2161 fallback_hosts, and "hosts" in the smtp transport.
2163 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
2164 host errors such as "Connection refused".
2166 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
2167 authentication with radiusclient 0.4.9:
2169 - Error returned from rc_read_config was caught wrongly
2170 - Username/password not passed on to radius server due to wrong length.
2172 The presumption is that some radiusclient API changes for 4.51/PH/17
2173 were not taken care of correctly. The code is still untested by me (my
2174 Linux distribution still has 0.3.2 of radiusclient), but it was
2175 contributed by a Radius user.
2177 PH/24 When doing a callout, the value of $domain wasn't set correctly when
2178 expanding the "port" option of the smtp transport.
2180 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
2181 while reading a MIME header. Thanks to Tom Hughes for a patch.
2183 PH/24 Include config.h inside local_scan.h so that configuration settings are
2186 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
2187 that in an ACL for RCPT (for example), you can examine exactly what was
2190 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
2191 commands, but it was not correctly comparing the address with the actual
2192 client host address. Thus, it would show the EHLO address in Received:
2193 header lines when this was not necessary.
2195 PH/27 Added the % operator to ${eval:}.
2197 PH/28 Exim tries to create and chdir to its spool directory when it starts;
2198 it should be ignoring failures (because with -C, for example, it has lost
2199 privilege). It wasn't ignoring creation failures other than "already
2202 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
2205 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
2206 because an input line was too long, either on its own, or by virtue of
2207 too many continuations, the temporary file was not being removed, and the
2208 return code was incorrect.
2210 PH/31 Missing "BOOL" in function definition in filtertest.c.
2212 PH/32 Applied Sieve patches from the maintainer.
2214 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
2216 PH/33 Added "verify = not_blind".
2218 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
2219 Local/Makefile (with some defaults set). These are used in built scripts
2220 such as exicyclog, but they have never been used in the exim_install
2221 script (though there are many overriding facilities there). I have
2222 arranged that the exim_install script now takes note of these two
2225 PH/35 Installed configuration files for Dragonfly.
2227 PH/36 When a locally submitted message by a trusted user did not contain a
2228 From: header, and the sender address was obtained from -f or from an SMTP
2229 MAIL command, and the trusted user did not use -F to supply a sender
2230 name, $originator_name was incorrectly used when constructing a From:
2231 header. Furthermore, $originator_name was used for submission mode
2232 messages from external hosts without From: headers in a similar way,
2233 which is clearly wrong.
2235 PH/37 Added control=suppress_local_fixups.
2237 PH/38 When log_selector = +received_sender was set, and the addition of the
2238 sender made the log line's construction buffer exactly full, or one byte
2239 less than full, an overflow happened when the terminating "\n" was
2242 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
2243 when the result of a list match is failure because a DNS lookup failed.
2245 PH/40 RM_COMMAND is now used in the building process.
2247 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
2248 the "build-* directories that it finds.
2250 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
2251 address in a domain literal was a prefix of an interface address.
2253 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
2254 when verifying a sender address, unless rewrite_headers is false.
2256 PH/44 Wrote a long comment about why errors_to addresses are verified as
2257 recipients, not senders.
2259 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
2260 the ratelimit ACL was added.
2262 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
2264 PH/46 Added extra information about PostgreSQL errors to the error string.
2266 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
2267 avoiding the need to install special zones in a real server. This is
2268 backwards compatible; if it can't find the fake resolver, it drops back.
2269 Thus, both old and new test suites can be run.
2271 TF/02 Added util/ratelimit.pl
2273 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
2274 clock is set back in time.
2276 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
2277 Candler <B.Candler@pobox.com>.
2279 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
2280 OK for addresses that are the result of redirections.
2282 PH/48 A number of further additions for the benefit of the new test suite,
2283 including a fake gethostbyname() that interfaces to the fake DNS resolver
2286 TF/06 The fix for widen_domains has also been applied to qualify_single and
2287 search_parents which are the other dnslookup options that can cause
2290 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
2293 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
2295 TF/07 Exim produced the error message "an SRV record indicated no SMTP
2296 service" if it encountered an MX record with an empty target hostname.
2297 The message is now "an MX or SRV record indicated no SMTP service".
2299 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
2300 if the DNS of the sending site is misconfigured. This is quite a
2301 common situation. This change restores the behaviour of treating a
2302 helo verification defer as a failure.
2304 PH/51 If self=fail was set on a router, the bounce message did not include the
2305 actual error message.
2311 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
2313 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
2314 command itself ended in error, the underlying message about the transport
2315 filter timeout was being overwritten with the pipe command error. Now the
2316 underlying error message should be appended to the second error message.
2318 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
2320 PH/02 Reduce the amount of output that "make" produces by default. Full output
2321 can still be requested.
2323 PH/03 The warning log line about a condition test deferring for a "warn" verb
2324 was being output only once per connection, rather than after each
2325 occurrence (because it was using the same function as for successful
2326 "warn" verbs). This seems wrong, so I have changed it.
2328 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
2329 it should not have, which might have caused a crash in the right
2330 circumstances, but probably never did.
2332 PH/04 Installed a modified version of Tony Finch's patch to make submission
2333 mode fix the return path as well as the Sender: header line, and to
2334 add a /name= option so that you can make the user's friendly name appear
2337 TF/03 Added the control = fakedefer ACL modifier.
2339 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
2340 Mark Lowes for thorough testing.
2342 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
2344 TK/03 Merged latest SRS patch from Miles Wilton.
2346 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
2347 with the definition in sysexits.h (which is #included earlier).
2348 Fortunately, Exim does not actually use EX_OK. The code used to try to
2349 preserve the sysexits.h value, by assumimg that macro definitions were
2350 scanned for macro replacements. I have been disabused of this notion,
2351 so now the code just undefines EX_OK before #including unistd.h.
2353 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
2354 in the smtp transport. When a block could not be written in a single
2355 write() function, the timeout was being re-applied to each part-write.
2356 This seems wrong - if the receiver was accepting one byte at a time it
2357 would take for ever. The timeout is now adjusted when this happens. It
2358 doesn't have to be particularly precise.
2360 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
2361 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
2363 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
2364 without a domain if the check to <postmaster@domain> fails.
2366 SC/01 Eximstats: added -xls and the ability to specify output files
2367 (patch written by Frank Heydlauf).
2369 SC/02 Eximstats: use FileHandles for outputing results.
2371 SC/03 Eximstats: allow any combination of xls, txt, and html output.
2373 SC/04 Eximstats: fixed display of large numbers with -nvr option
2375 SC/05 Eximstats: fixed merging of reports with empty tables.
2377 SC/06 Eximstats: added the -include_original_destination flag
2379 SC/07 Eximstats: removed tabs and trailing whitespace.
2381 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
2383 TK/06 MBOX spool code: Add real "From " MBOX separator line
2384 so the .eml file is really in mbox format (even though
2385 most programs do not really care). Patch from Alex Miller.
2387 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
2388 The latter is generated from $received_to and is only set if the
2389 message has one envelope recipient. SA can use these headers,
2390 obviously out-of-the-box. Patch from Alex Miller.
2392 PH/08 The ${def test on a variable was returning false if the variable's
2393 value was "0", contrary to what the specification has always said!
2394 The result should be true unless the variable is empty.
2396 PH/09 The syntax error of a character other than { following "${if
2397 def:variable_name" (after optional whitespace) was not being diagnosed.
2398 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
2399 accidental colon was present, for example, could give incorrect results.
2401 PH/10 Tidied the code in a number of places where the st_size field of a stat()
2402 result is used (not including appendfile, where other changes are about
2405 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
2406 This involved changing a lot of size variables from int to off_t. It
2407 should work with maildirs and everything.
2409 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
2410 spamd dying while we are connected to it.
2412 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
2415 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
2416 function for BDB 4.3.
2418 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
2420 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
2421 the sources for PCRE have been re-arranged and more files are now
2424 PH/15 The code I had for printing potentially long long variables in PH/11
2425 above was not the best (it lost precision). The length of off_t variables
2426 is now inspected at build time, and an appropriate printing format (%ld
2427 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
2428 to be "long long int" or "long int". This is needed for the internal
2429 formatting function string_vformat().
2431 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
2432 the configuration file to be ":syslog", then the script "guesses" where
2433 the logs files are, rather than using the compiled in default. In our
2434 case the guess is not the same as the compiled default, so the script
2435 suddenly stopped working when I started to use syslog. The patch checks
2436 to see if log_file_path is "". If so, it attempts to read it from exim
2437 with no configuration file to get the compiled in version, before it
2438 falls back to the previous guessing code."
2440 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
2441 implementing BATV in an Exim configuration. See NewStuff for the gory
2444 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
2445 Makefile that are specific to HP-UX.
2447 PH/18 If the "use_postmaster" option was set for a recipient callout together
2448 with the "random" option, the postmaster address was used as the MAIL
2449 FROM address for the random test, but not for the subsequent recipient
2450 test. It is now used for both.
2452 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
2453 patch removes a few documentation additions to RFC 3028, because the
2454 latest draft now contains them. It adds the new en;ascii-case comparator
2455 and a new error check for 8bit text in MIME parts. Comparator and
2456 require names are now matched exactly. I enabled the subaddress
2457 extension, but it is not well tested yet (read: it works for me)."
2459 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
2460 rework some of the code of TK/09 above to avoid the hardwired use of
2461 "%lld" and "long long". Replaced the call to snprintf() with a call to
2464 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
2465 records point to non-existent hosts", "retry timeout exceeded", and
2466 "retry time not reached for any host after a long failure period".
2468 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
2469 experimental DomainKeys support:
2471 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
2472 (2) On an error such as an illegally used "control", the wrong name for
2473 the control was given.
2475 These problems did NOT occur unless DomainKeys support was compiled.
2477 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
2479 PH/24 Added ${if match_ip condition.
2481 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
2482 databases so that it will be absolutely obvious if a crash occurs in the
2483 DB library. This is a regular occurrence (often caused by mis-matched
2486 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
2487 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
2488 picked up on a user's system that detects such things. There doesn't seem
2489 to be a gcc warning option for this - only an attribute that has to be
2490 put on the function's prototype. It seems that in Fedora Core 4 they have
2491 set this on a number of new functions. No doubt there will be more in due
2494 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
2495 specify a transport. However, if an address that was verified by such a
2496 router was the subject of a callout, Exim crashed because it tried to
2497 read the rcpt_include_affixes from the non-existent transport. Now it
2498 just assumes that the setting of that option is false. This bug was
2499 introduced by 4.51/PH/31.
2501 PH/28 Changed -d+all to exclude +memory, because that information is very
2502 rarely of interest, but it makes the output a lot bigger. People tend to
2503 do -d+all out of habit.
2505 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
2506 code in os-type was giving problems when libc.so lives in lib64, like on
2509 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
2510 aren't the modern standard, and it seems that some systems' include files
2511 don't always have them. Exim was already checking for some of the newer
2512 ones like T_AAAA, and defining it itself. I've added checks for all the
2513 record types that Exim uses.
2515 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
2516 not automatically generating a new one, as it is supposed to. This
2517 prevented TLS from working. If the file did exist, but contained invalid
2518 data, a new version was generated, as expected. It was only the case of a
2519 non-existent file that was broken.
2521 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
2522 with a change in libdomainkeys > 0.64.
2524 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
2525 from DNS. If the selector record carries the flag, it now has
2526 precedence over the domain-wide flag.
2528 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
2530 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
2531 the use of a port name that isn't defined in /etc/services) occurred, the
2532 message was deferred as in a normal delivery, and thus remained on the
2533 spool, instead of being failed because of the mua_wrapper setting. This
2534 is now fixed, and I tidied up some of the mua_wrapper messages at the
2537 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
2538 the messages in a hash of arrays rather than using individual hashes.
2539 This is a bit cleaner and results in dramatic memory savings, albeit
2540 at a slight CPU cost.
2542 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
2543 as requested by Marc Sherman.
2545 SC/10 Eximstats: added histograms for user specified patterns as requested
2548 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
2550 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
2551 fopen() in the content-scanning modules that did not already have it.
2557 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
2558 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
2560 TK/02 Fix ACL "control" statement not being available in MIME ACL.
2562 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
2564 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
2565 to test Sieve filters that use "vacation".
2567 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
2568 that changes the way the GnuTLS parameters are stored in the cache file.
2569 The new format can be generated externally. For backward compatibility,
2570 if the data in the cache doesn't make sense, Exim assumes it has read an
2571 old-format file, and it generates new data and writes a new file. This
2572 means that you can't go back to an older release without removing the
2575 PH/03 A redirect router that has both "unseen" and "one_time" set does not
2576 work if there are any delivery delays because "one_time" forces the
2577 parent to be marked "delivered", so its unseen clone is never tried
2578 again. For this reason, Exim now forbids the simultaneous setting of
2581 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
2582 redirected to themselves ("homonym" addresses). Read the long ChangeLog
2583 entry if you want to know the details. The fix, however, neglected to
2584 consider the case when local delivery batching is involved. The test for
2585 "previously delivered" was not happening when checking to see if an
2586 address could be batched with a previous (undelivered) one; under
2587 certain circumstances this could lead to multiple deliveries to the same
2590 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
2591 in its include files, and this causes problems building Exim.
2593 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
2594 header_syntax) but Exim was just ignoring anything given after a slash.
2595 In particular, this caused confusion with an attempt to use "verify =
2596 reverse_host_lookup/defer_ok". An error is now given when options are
2597 supplied for verify items that do not have them. (Maybe reverse_host_
2598 lookup should have a defer_ok option, but that's a different point.)
2600 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
2601 defined by RFC 821) to 2048, because there were problems with some AUTH
2602 commands, and RFC 1869 says the size should be increased for extended
2603 SMTP commands that take arguments.
2605 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
2608 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
2609 "unknown" error; now it says that the functionality isn't in the binary.
2611 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
2612 an address' error message when a string expansion fails (syntax or
2613 whatever). Otherwise the password may appear in the log. Following change
2614 PH/42 below, there is no longer a chance of it appearing in a bounce
2617 PH/11 Installed exipick version 20050225.0 from John Jetmore.
2619 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
2620 of its addresses was ever tried. (Bugzilla bug #2.)
2622 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
2623 the result incorrectly in the debug output. (It correctly added a newline
2624 to what was transported.)
2626 TF/01 Added $received_time.
2628 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
2629 commented out examples of how to interface to a virus scanner and to
2630 SpamAssassin. Also added commented examples of av_scanner and
2631 spamd_address settings.
2633 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
2634 and controls are allowed in which ACLs. There were a couple of minor
2635 errors. Some of the entries in the conditions table (which is a table of
2636 where they are NOT allowed) were getting very unwieldy; rewrote them as a
2637 negation of where the condition IS allowed.
2639 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
2641 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
2642 header file does not have a version number, so I've had to invent a new
2643 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
2644 API. The code is untested by me (my Linux distribution still has 0.3.2 of
2645 radiusclient), but it was contributed by a Radius user.
2647 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
2648 files or directories, for OpenSSL.
2650 PH/19 When an Exim process that is running as root has to create an Exim log
2651 file, it does so in a subprocess that runs as exim:exim so as to get the
2652 ownership right at creation (otherwise, other Exim processes might see
2653 the file with the wrong ownership). There was no test for failure of this
2654 fork() call, which would lead to the process getting stuck as it waited
2655 for a non-existent subprocess. Forks do occasionally fail when resources
2656 run out. I reviewed all the other calls to fork(); they all seem to check
2659 PH/20 When checking for unexpected SMTP input at connect time (before writing
2660 the banner), Exim was not dealing correctly with a non-positive return
2661 from the read() function. If the client had disconnected by this time,
2662 the result was a log entry for a synchronization error with an empty
2663 string after "input=" when read() returned zero. If read() returned -1
2664 (an event I could not check), uninitialized data bytes were printed.
2665 There were reports of junk text (parts of files, etc) appearing after
2668 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
2670 PH/22 Added support for macro redefinition, and (re)definition in between
2671 driver and ACL definitions.
2673 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
2674 forgetting to use the resulting value; it was using the unexpanded value.
2676 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
2677 hadn't been configured. The fix is from Juergen Kreileder, who
2678 understands it better than I do:
2680 "Here's what I see happening with three configured cyrus_sasl
2681 authenticators configured (plain, login, cram-md5):
2683 On startup auth_cyrus_sasl_init() gets called for each of these.
2684 This means three calls to sasl_listmech() without a specified mech_list.
2685 => SASL tests which mechs of all available mechs actually work
2686 => three warnings about OTP not working
2687 => the returned list contains: plain, login, cram-md5, digest-md5, ...
2689 With the patch, sasl_listmech() also gets called three times. But now
2690 SASL's mech_list option is set to the server_mech specified in the the
2691 authenticator. Or in other words, the answer from sasl_listmech()
2692 gets limited to just the mech you're testing for (which is different
2694 => the return list contains just 'plain' or 'login', 'cram-md5' or
2695 nothing depending on the value of ob->server_mech.
2697 I've just tested the patch: Authentication still works fine,
2698 unavailable mechs specified in the exim configuration are still
2699 caught, and the auth.log warnings about OTP are gone."
2701 PH/25 When debugging is enabled, the contents of the command line are added
2702 to the debugging output, even when log_selector=+arguments is not
2705 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
2706 answer is "GNU", and only if the return is "GNU/something" is the answer
2709 PH/27 $acl_verify_message is now set immediately after the failure of a
2710 verification in an ACL, and so is available in subsequent modifiers. In
2711 particular, the message can be preserved by coding like this:
2713 warn !verify = sender
2714 set acl_m0 = $acl_verify_message
2716 Previously, $acl_verify_message was set only while expanding "message"
2717 and "log_message" when a very denied access.
2719 PH/28 Modified OS/os.c-Linux with
2721 -#ifndef OS_LOAD_AVERAGE
2722 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
2724 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
2725 nomenclature these days.)
2727 PH/29 Installed patch from the Sieve maintainer that adds the options
2728 sieve_useraddress and sieve_subaddress to the redirect router.
2730 PH/30 In these circumstances:
2731 . Two addresses routed to the same list of hosts;
2732 . First host does not offer TLS;
2733 . First host accepts first address;
2734 . First host gives temporary error to second address;
2735 . Second host offers TLS and a TLS session is established;
2736 . Second host accepts second address.
2737 Exim incorrectly logged both deliveries with the TLS parameters (cipher
2738 and peerdn, if requested) that were in fact used only for the second
2741 PH/31 When doing a callout as part of verifying an address, Exim was not paying
2742 attention to any local part prefix or suffix that was matched by the
2743 router that accepted the address. It now behaves in the same way as it
2744 does for delivery: the affixes are removed from the local part unless
2745 rcpt_include_affixes is set on the transport.
2747 PH/32 Add the sender address, as F=<...>, to the log line when logging a
2748 timeout during the DATA phase of an incoming message.
2750 PH/33 Sieve envelope tests were broken for match types other than :is. I have
2751 applied a patch sanctioned by the Sieve maintainer.
2753 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
2754 the uid or gid is negative. A case of a negative gid caused this to be
2755 noticed. The fix allows for either to be negative.
2757 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
2758 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
2761 PH/36 The change PH/12 above was broken. Fixed it.
2763 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
2764 the grounds that routing the same address twice would always produce the
2765 same answer. This might have been true once, but it is certainly no
2766 longer true now. Routing a child address may depend on the previous
2767 routing that produced that child. Some complicated redirection strategies
2768 went wrong when messages had multiple recipients, and made Exim's
2769 behaviour dependent on the order in which the addresses were given.
2771 I have moved the duplicate checking until after the routing is complete.
2772 Exim scans the addresses that are assigned to local and remote
2773 transports, and removes any duplicates. This means that more work will be
2774 done, as duplicates will always all be routed, but duplicates are
2775 presumably rare, so I don't expect this is of any significance.
2777 For deliveries to pipes, files, and autoreplies, the duplicate checking
2778 still happens during the routing process, since they are not going to be
2781 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
2782 It corrects a timeout issue with spamd. This is Ian's comment: "The
2783 background is that sometimes spamd either never reads data from a
2784 connection it has accepted, or it never writes response data. The exiscan
2785 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
2786 blindly assumes that writes won't block so it may never time out."
2788 PH/39 Allow G after quota size as well as K and M.
2790 PH/40 The value set for $authenticated_id in an authenticator may not contain
2791 binary zeroes or newlines because the value is written to log lines and
2792 to spool files. There was no check on this. Now the value is run through
2793 the string_printing() function so that such characters are converted to
2794 printable escape sequences.
2796 PH/41 $message_linecount is a new variable that contains the total number of
2797 lines in the message. Compare $body_linecount, which is the count for the
2800 PH/42 Exim no longer gives details of delivery errors for specific addresses in
2801 bounce and delay warning messages, except in certain special cases, which
2804 (a) An SMTP error message from a remote host;
2805 (b) A message specified in a :fail: redirection;
2806 (c) A message specified in a "fail" command in a system filter;
2807 (d) A message specified in a FAIL return from the queryprogram router;
2808 (e) A message specified by the cannot_route_message router option.
2810 In these cases only, Exim does include the error details in bounce and
2811 warning messages. There are also a few cases where bland messages such
2812 as "unrouteable address" or "local delivery error" are given.
2814 PH/43 $value is now also set for the "else" part of a ${run expansion.
2816 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
2817 being worked on, but at least Exim now implements the latest version to
2820 PH/45 In a pipe transport, although a timeout while waiting for the pipe
2821 process to complete was treated as a delivery failure, a timeout while
2822 writing the message to the pipe was logged, but erroneously treated as a
2823 successful delivery. Such timeouts include transport filter timeouts. For
2824 consistency with the overall process timeout, these timeouts are now
2825 treated as errors, giving rise to delivery failures by default. However,
2826 there is now a new Boolean option for the pipe transport called
2827 timeout_defer, which, if set TRUE, converts the failures into defers for
2828 both kinds of timeout. A transport filter timeout is now identified in
2831 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
2832 systems where "make" and "gmake" are different, calling "gmake" at top
2833 level broke things. I've arranged for the value of $(MAKE) to be passed
2834 from the Makefile to this script so that it can call the same version of
2838 A note about Exim versions 4.44 and 4.50
2839 ----------------------------------------
2841 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
2842 changes of various kinds. As a consequence, a big documentation update was
2843 needed. This delayed the release for rather longer than seemed good, especially
2844 in the light of a couple of (minor) security issues. Therefore, the changes
2845 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
2846 release. So 4.44 and 4.50 are in effect two different branches that both start
2849 I have left the 4.50 change log unchanged; it contains all the changes since
2850 4.43. The change log for 4.44 is below; many of its items are identical to
2851 those for 4.50. This seems to be the most sensible way to preserve the
2852 historical information.
2858 1. Minor wording change to the doc/README.SIEVE file.
2860 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
2861 computation of the current number of files was incorrect.
2863 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
2864 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
2867 4. Give more explanation in the error message when the command for a transport
2868 filter fails to execute.
2870 5. There are several places where Exim runs a non-Exim command in a
2871 subprocess. The SIGUSR1 signal should be disabled for these processes. This
2872 was being done only for the command run by the queryprogram router. It is
2873 now done for all such subprocesses. The other cases are: ${run, transport
2874 filters, and the commands run by the lmtp and pipe transports.
2876 6. Added CONFIGURE_GROUP build-time option.
2878 7. Some older OS have a limit of 256 on the maximum number of file
2879 descriptors. Exim was using setrlimit() to set 1000 as a large value
2880 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
2881 systems. I've change it so that if it can't get 1000, it tries for 256.
2883 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
2884 was an oversight, and furthermore, ever since the addition of extra
2885 controls (e.g. 4.43/32), the checks on when to allow different forms of
2886 "control" were broken. There should now be diagnostics for all cases when a
2887 control that does not make sense is encountered.
2889 9. Added the /retain_sender option to "control=submission".
2891 10. $recipients is now available in the predata ACL (oversight).
2893 11. Tidy the search cache before the fork to do a delivery from a message
2894 received from the command line. Otherwise the child will trigger a lookup
2895 failure and thereby defer the delivery if it tries to use (for example) a
2896 cached ldap connection that the parent has called unbind on.
2898 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
2899 of $address_data from the recipient verification was clobbered by the
2900 sender verification.
2902 13. The value of address_data from a sender verification is now available in
2903 $sender_address_data in subsequent conditions in the ACL statement.
2905 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
2907 15. Added a new option "connect=<time>" to callout options, to set a different
2910 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
2911 was its contents. (It was OK if the option was not defined at all.)
2913 17. A "Completed" log line is now written for messages that are removed from
2914 the spool by the -Mrm option.
2916 18. New variables $sender_verify_failure and $recipient_verify_failure contain
2917 information about exactly what failed.
2919 19. Added -dd to debug only the daemon process.
2921 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
2922 handles timeouts, both on the server side and network timeouts. Renamed the
2923 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
2925 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
2926 It is now set to "smtps".
2928 22. $host_address is now set to the target address during the checking of
2929 ignore_target_hosts.
2931 23. When checking ignore_target_hosts for an ipliteral router, no host name was
2932 being passed; this would have caused $sender_host_name to have been used if
2933 matching the list had actually called for a host name (not very likely,
2934 since this list is usually IP addresses). A host name is now passed as
2937 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
2938 code that specifies a non-restarting handler (typically sigaction() in
2939 modern systems) in an attempt to fix a rare and obscure crash bug.
2941 25. Narrowed the window for a race in the daemon that could cause it to ignore
2942 SIGCHLD signals. This is not a major problem, because they are used only to
2943 wake it up if nothing else does.
2945 26. A malformed maildirsize file could cause Exim to calculate negative values
2946 for the mailbox size or file count. Odd effects could occur as a result.
2947 The maildirsize information is now recalculated if the size or filecount
2950 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
2951 support for a long time. Removed HAVE_SYS_VFS_H.
2953 28. Installed the latest version of exipick from John Jetmore.
2955 29. In an address list, if the pattern was not a regular expression, an empty
2956 subject address (from a bounce message) matched only if the pattern was an
2957 empty string. Non-empty patterns were not even tested. This was the wrong
2958 because it is perfectly reasonable to use an empty address as part of a
2959 database query. An empty address is now tested by patterns that are
2960 lookups. However, all the other forms of pattern expect the subject to
2961 contain a local part and a domain, and therefore, for them, an empty
2962 address still always fails if the pattern is not itself empty.
2964 30. Exim went into a mad DNS loop when attempting to do a callout where the
2965 host was specified on an smtp transport, and looking it up yielded more
2966 than one IP address.
2968 31. Re-factored the code for checking spool and log partition space into a
2969 function that finds that data and another that does the check. The former
2970 is then used to implement four new variables: $spool_space, $log_space,
2971 $spool_inodes, and $log_inodes.
2973 32. The RFC2047 encoding function was originally intended for short strings
2974 such as real names; it was not keeping to the 75-character limit for
2975 encoded words that the RFC imposes. It now respects the limit, and
2976 generates multiple encoded words if necessary. To be on the safe side, I
2977 have increased the buffer size for the ${rfc2047: expansion operator from
2980 33. It is now permitted to omit both strings after an "if" condition; if the
2981 condition is true, the result is "true". As before, when the second string
2982 is omitted, a false condition yields an empty string. This makes it less
2983 cumbersome to write custom ACL and router conditions.
2985 34. Failure to deliver a bounce message always caused it to be frozen, even if
2986 there was an errors_to setting on the router. The errors_to setting is now
2989 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
2990 canonical form (fully expanded) before being placed in
2991 $sender_host_address.
2993 36. The table in the code that translates DNS record types into text (T_A to
2994 "A" for instance) was missing entries for NS and CNAME. It is just possible
2995 that this could have caused confusion if both these types were looked up
2996 for the same domain, because the text type is used as part of Exim's
2997 per-process caching. But the chance of anyone hitting this buglet seems
3000 37. The dnsdb lookup has been extended in a number of ways.
3002 (1) There is a new type, "zns", which walks up the domain tree until it
3003 finds some nameserver records. It should be used with care.
3005 (2) There is a new type, "mxh", which is like "mx" except that it returns
3006 just the host names, not the priorities.
3008 (3) It is now possible to give a list of domains (or IP addresses) to be
3009 looked up. The behaviour when one of the lookups defers can be
3010 controlled by a keyword.
3012 (4) It is now possible to specify the separator character for use when
3013 multiple records are returned.
3015 38. The dnslists ACL condition has been extended: it is now possible to supply
3016 a list of IP addresses and/or domains to be looked up in a particular DNS
3019 39. Added log_selector=+queue_time_overall.
3021 40. When running the queue in the test harness, wait just a tad after forking a
3022 delivery process, to get repeatability of debugging output.
3024 41. Include certificate and key file names in error message when GnuTLS fails
3025 to set them up, because the GnuTLS error message doesn't include the name
3026 of the failing file when there is a problem reading it.
3028 42. Allow both -bf and -bF in the same test run.
3030 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
3032 44. The "Exiscan patch" is now merged into the mainline Exim source.
3034 45. Sometimes the final signoff response after QUIT could fail to get
3035 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
3036 < 0 before doing a fflush(). This bug looks as though it goes back to the
3037 introduction of TLS in release 3.20, but "sometimes" must have been rare
3038 because the tests only now provoked it.
3040 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
3041 (this can affect the format of dates).
3043 47. exim_tidydb, when checking for the continued existence of a message for
3044 which it has found a message-specific retry record, was not finding
3045 messages that were in split spool directories. Consequently, it was
3046 deleting retry records that should have stayed in existence.
3048 48. Steve fixed some bugs in eximstats.
3050 49. The SPA authentication driver was not abandoning authentication and moving
3051 on to the next authenticator when an expansion was forced to fail,
3052 contradicting the general specification for all authenticators. Instead it
3053 was generating a temporary error. It now behaves as specified.
3055 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
3056 (the order specifies the preference for clients). The order is now AES256,
3057 AES128, 3DES, ARCFOUR128.
3059 51. Small patch to Sieve code - explicitly set From: when generating an
3062 52. Exim crashed if a remote delivery caused a very long error message to be
3063 recorded - for instance if somebody sent an entire SpamAssassin report back
3064 as a large number of 550 error lines. This bug was coincidentally fixed by
3065 increasing the size of one of Exim's internal buffers (big_buffer) that
3066 happened as part of the Exiscan merge. However, to be on the safe side, I
3067 have made the code more robust (and fixed the comments that describe what
3070 53. Now that there can be additional text after "Completed" in log lines (if
3071 the queue_time_overall log selector is set), a one-byte patch to exigrep
3072 was needed to allow it to recognize "Completed" as not the last thing in
3075 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
3076 patch that reportedly fixes this has been added. I am not expert enough to
3077 create a test for it. This is what the patch creator wrote:
3079 "I found a little strange behaviour of ldap code when working with
3080 Windows 2003 AD Domain, where users was placed in more than one
3081 Organization Units. When I tried to give exim partial DN, the exit code
3082 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
3083 But simultaneously result of request was absolutely normal ldap result,
3084 so I produce this patch..."
3086 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
3087 REFERENCE, so I have modified the code to exclude the patch when that macro
3090 55. Some experimental protocols are using DNS PTR records for new purposes. The
3091 keys for these records are domain names, not reversed IP addresses. The
3092 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
3093 leaves it alone. Component reversal etc. now happens only for IP addresses.
3096 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
3098 57. Double the size of the debug message buffer (to 2048) so that more of very
3099 long debug lines gets shown.
3101 58. The exicyclog utility now does better if the number of log files to keep
3102 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
3104 59. Two changes related to the smtp_active_hostname option:
3106 (1) $smtp_active_hostname is now available as a variable.
3107 (2) The default for smtp_banner uses $smtp_active_hostname instead
3108 of $primary_hostname.
3110 60. The host_aton() function is supposed to be passed a string that is known
3111 to be a valid IP address. However, in the case of IPv6 addresses, it was
3112 not checking this. This is a hostage to fortune. Exim now panics and dies
3113 if the condition is not met. A case was found where this could be provoked
3114 from a dnsdb PTR lookup with an IPv6 address that had more than 8
3115 components; fortuitously, this particular loophole had already been fixed
3116 by change 4.50/55 above.
3118 If there are any other similar loopholes, the new check in host_aton()
3119 itself should stop them being exploited. The report I received stated that
3120 data on the command line could provoke the exploit when Exim was running as
3121 exim, but did not say which command line option was involved. All I could
3122 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
3123 running as the user.
3126 61. There was a buffer overflow vulnerability in the SPA authentication code
3127 (which came originally from the Samba project). I have added a test to the
3128 spa_base64_to_bits() function which I hope fixes it.
3131 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
3132 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
3134 63. The daemon start-up calls getloadavg() while still root for those OS that
3135 need the first call to be done as root, but it missed one case: when
3136 deliver_queue_load_max is set with deliver_drop_privilege. This is
3137 necessary for the benefit of the queue runner, because there is no re-exec
3138 when deliver_drop_privilege is set.
3140 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
3141 This has been fixed.
3143 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
3144 was in use, was not putting the data itself into the right store pool;
3145 consequently, it could be overwritten for a subsequent message in the same
3146 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
3149 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
3151 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
3152 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
3153 the function were treating the return as a boolean value, which happened to
3154 work because 0=false and not-0=true, but is not correct code.
3156 68. The host_aton() function was not handling scoped IPv6 addresses (those
3157 with, for example, "%eth0" on the end) correctly.
3159 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
3160 negated items (that is, ~something) in unsigned ints. Some compilers
3161 apparently mutter when there is no cast.
3163 70. If an address verification called from an ACL failed, and did not produce a
3164 user-specific message (i.e. there was only a "system" message), nothing was
3165 put in $acl_verify_message. In this situation, it now puts the system
3168 71. Change 4.23/11 added synchronization checking at the start of an SMTP
3169 session; change 4.31/43 added the unwanted input to the log line - except
3170 that it did not do this in the start of session case. It now does.
3172 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
3173 This is wrong and can cause the other end to generate a synchronization
3174 error if it is another Exim or anything else that does the synchronization
3175 check. A QUIT command is no longer sent after a timeout.
3177 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
3178 during host lookups.
3180 74. The defer_ok option of callout verification was not working if it was used
3181 when verifying addresses in header lines, that is, for this case:
3183 verify = header_sender/callout=defer_ok
3185 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
3186 those file descriptors could be used for SMTP connections. If anything
3187 wrote to stderr (the example that came up was "warn" in embedded Perl), it
3188 could be sent to the SMTP client, causing chaos. The daemon now opens
3189 stdin, stdout, and stderr to /dev/null when it puts itself into the
3192 76. Arrange for output from Perl's "warn" command to be written to Exim's main
3193 log by default. The user can override this with suitable Perl magic.
3195 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
3196 the log message when discard triggers, was not working for the DATA ACL or
3197 for the non-SMTP ACL.
3199 78. Error message wording change in sieve.c.
3201 79. If smtp_accept_max_per_host was set, the number of connections could be
3202 restricted to fewer than expected, because the daemon was trying to set up
3203 a new connection before checking whether the processes handling previous
3204 connections had finished. The check for completed processes is now done
3205 earlier. On busy systems, this bug wouldn't be noticed because something
3206 else would have woken the daemon, and it would have reaped the completed
3209 80. If a message was submitted locally by a user whose login name contained one
3210 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
3211 It caused a spool format error. I have fixed the spool reading code. A
3212 related problem was that the "from" clause in the Received: line became
3213 illegal because of the space(s). It is now covered by ${quote_local_part.
3215 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
3218 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
3220 83. Patch from Sieve maintainer to fix unterminated string problem in
3221 "vacation" handling.
3223 84. Some minor changes to the Linux configuration files to help with other
3224 OS variants using glibc.
3226 85. One more patch for Sieve to update vacation handling to latest spec.
3229 ----------------------------------------------------
3230 See the note above about the 4.44 and 4.50 releases.
3231 ----------------------------------------------------
3237 1. Change 4.43/35 introduced a bug that caused file counts to be
3238 incorrectly computed when quota_filecount was set in an appendfile
3241 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
3242 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
3245 3. Give more explanation in the error message when the command for a transport
3246 filter fails to execute.
3248 4. There are several places where Exim runs a non-Exim command in a
3249 subprocess. The SIGUSR1 signal should be disabled for these processes. This
3250 was being done only for the command run by the queryprogram router. It is
3251 now done for all such subprocesses. The other cases are: ${run, transport
3252 filters, and the commands run by the lmtp and pipe transports.
3254 5. Some older OS have a limit of 256 on the maximum number of file
3255 descriptors. Exim was using setrlimit() to set 1000 as a large value
3256 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
3257 systems. I've change it so that if it can't get 1000, it tries for 256.
3259 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
3260 was an oversight, and furthermore, ever since the addition of extra
3261 controls (e.g. 4.43/32), the checks on when to allow different forms of
3262 "control" were broken. There should now be diagnostics for all cases when a
3263 control that does not make sense is encountered.
3265 7. $recipients is now available in the predata ACL (oversight).
3267 8. Tidy the search cache before the fork to do a delivery from a message
3268 received from the command line. Otherwise the child will trigger a lookup
3269 failure and thereby defer the delivery if it tries to use (for example) a
3270 cached ldap connection that the parent has called unbind on.
3272 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
3273 of $address_data from the recipient verification was clobbered by the
3274 sender verification.
3276 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
3277 was its contents. (It was OK if the option was not defined at all.)
3279 11. A "Completed" log line is now written for messages that are removed from
3280 the spool by the -Mrm option.
3282 12. $host_address is now set to the target address during the checking of
3283 ignore_target_hosts.
3285 13. When checking ignore_target_hosts for an ipliteral router, no host name was
3286 being passed; this would have caused $sender_host_name to have been used if
3287 matching the list had actually called for a host name (not very likely,
3288 since this list is usually IP addresses). A host name is now passed as
3291 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
3292 code that specifies a non-restarting handler (typically sigaction() in
3293 modern systems) in an attempt to fix a rare and obscure crash bug.
3295 15. Narrowed the window for a race in the daemon that could cause it to ignore
3296 SIGCHLD signals. This is not a major problem, because they are used only to
3297 wake it up if nothing else does.
3299 16. A malformed maildirsize file could cause Exim to calculate negative values
3300 for the mailbox size or file count. Odd effects could occur as a result.
3301 The maildirsize information is now recalculated if the size or filecount
3304 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
3305 support for a long time. Removed HAVE_SYS_VFS_H.
3307 18. Updated exipick to current release from John Jetmore.
3309 19. Allow an empty sender to be matched against a lookup in an address list.
3310 Previously the only cases considered were a regular expression, or an
3313 20. Exim went into a mad DNS lookup loop when doing a callout where the
3314 host was specified on the transport, if the DNS lookup yielded more than
3317 21. The RFC2047 encoding function was originally intended for short strings
3318 such as real names; it was not keeping to the 75-character limit for
3319 encoded words that the RFC imposes. It now respects the limit, and
3320 generates multiple encoded words if necessary. To be on the safe side, I
3321 have increased the buffer size for the ${rfc2047: expansion operator from
3324 22. Failure to deliver a bounce message always caused it to be frozen, even if
3325 there was an errors_to setting on the router. The errors_to setting is now
3328 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
3329 canonical form (fully expanded) before being placed in
3330 $sender_host_address.
3332 24. Updated eximstats to version 1.33
3334 25. Include certificate and key file names in error message when GnuTLS fails
3335 to set them up, because the GnuTLS error message doesn't include the name
3336 of the failing file when there is a problem reading it.
3338 26. Expand error message when OpenSSL has problems setting up cert/key files.
3341 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
3342 (this can affect the format of dates).
3344 28. exim_tidydb, when checking for the continued existence of a message for
3345 which it has found a message-specific retry record, was not finding
3346 messages that were in split spool directories. Consequently, it was
3347 deleting retry records that should have stayed in existence.
3349 29. eximstats updated to version 1.35
3350 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
3351 1.35 - bugfix such that pie charts by volume are generated correctly
3353 30. The SPA authentication driver was not abandoning authentication and moving
3354 on to the next authenticator when an expansion was forced to fail,
3355 contradicting the general specification for all authenticators. Instead it
3356 was generating a temporary error. It now behaves as specified.
3358 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
3359 (the order specifies the preference for clients). The order is now AES256,
3360 AES128, 3DES, ARCFOUR128.
3362 31. Small patch to Sieve code - explicitly set From: when generating an
3365 32. Exim crashed if a remote delivery caused a very long error message to be
3366 recorded - for instance if somebody sent an entire SpamAssassin report back
3367 as a large number of 550 error lines. This bug was coincidentally fixed by
3368 increasing the size of one of Exim's internal buffers (big_buffer) that
3369 happened as part of the Exiscan merge. However, to be on the safe side, I
3370 have made the code more robust (and fixed the comments that describe what
3373 33. Some experimental protocols are using DNS PTR records for new purposes. The
3374 keys for these records are domain names, not reversed IP addresses. The
3375 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
3376 leaves it alone. Component reversal etc. now happens only for IP addresses.
3379 34. The host_aton() function is supposed to be passed a string that is known
3380 to be a valid IP address. However, in the case of IPv6 addresses, it was
3381 not checking this. This is a hostage to fortune. Exim now panics and dies
3382 if the condition is not met. A case was found where this could be provoked
3383 from a dnsdb PTR lookup with an IPv6 address that had more than 8
3384 components; fortuitously, this particular loophole had already been fixed
3385 by change 4.50/55 or 4.44/33 above.
3387 If there are any other similar loopholes, the new check in host_aton()
3388 itself should stop them being exploited. The report I received stated that
3389 data on the command line could provoke the exploit when Exim was running as
3390 exim, but did not say which command line option was involved. All I could
3391 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
3392 running as the user.
3395 35. There was a buffer overflow vulnerability in the SPA authentication code
3396 (which came originally from the Samba project). I have added a test to the
3397 spa_base64_to_bits() function which I hope fixes it.
3400 36. The daemon start-up calls getloadavg() while still root for those OS that
3401 need the first call to be done as root, but it missed one case: when
3402 deliver_queue_load_max is set with deliver_drop_privilege. This is
3403 necessary for the benefit of the queue runner, because there is no re-exec
3404 when deliver_drop_privilege is set.
3406 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
3407 was in use, was not putting the data itself into the right store pool;
3408 consequently, it could be overwritten for a subsequent message in the same
3409 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
3412 38. Sometimes the final signoff response after QUIT could fail to get
3413 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
3414 < 0 before doing a fflush(). This bug looks as though it goes back to the
3415 introduction of TLS in release 3.20, but "sometimes" must have been rare
3416 because the tests only now provoked it.
3422 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
3423 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
3424 Now it returns an integer. A number of calls were still expecting a T/F
3425 return. Fortuitously, in all cases, the tests worked in OK situations,
3426 which is the norm. However, things would have gone wrong on any write
3427 failures on the smtp file descriptor. This function is used when sending
3428 messages over SMTP and also when doing verify callouts.
3430 2. When Exim is called to do synchronous delivery of a locally submitted
3431 message (the -odf or -odi options), it no longer closes stderr before doing
3434 3. Implemented the mua_wrapper option.
3436 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
3438 5. Implemented the functions header_remove(), header_testname(),
3439 header_add_at_position(), and receive_remove_recipient(), and exported them
3442 6. If an ACL "warn" statement specified the addition of headers, Exim already
3443 inserted X-ACL-Warn: at the start if there was no header name. However, it
3444 was not making this test for the second and subsequent header lines if
3445 there were newlines in the string. This meant that an invalid header could
3446 be inserted if Exim was badly configured.
3448 7. Allow an ACL "warn" statement to add header lines at the start or after all
3449 the Received: headers, as well as at the end.
3451 8. Added the rcpt_4xx retry error code.
3453 9. Added postmaster_mailfrom=xxx to callout verification option.
3455 10. Added mailfrom=xxxx to the callout verification option, for verify=
3458 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
3459 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
3461 12. Inserted some casts to stop certain compilers warning when using pointer
3462 differences as field lengths or precisions in printf-type calls (mostly
3463 affecting debugging statements).
3465 13. Added optional readline() support for -be (dynamically loaded).
3467 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
3468 same clock tick as a message's arrival, so that its received time was the
3469 same as the "first fail" time on the retry record, and that message
3470 remained on the queue past the ultimate address timeout, every queue runner
3471 would try a delivery (because it was past the ultimate address timeout) but
3472 after another failure, the ultimate address timeout, which should have then
3473 bounced the address, did not kick in. This was a "< instead of <=" error;
3474 in most cases the first failure would have been in the next clock tick
3475 after the received time, and all would be well.
3477 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
3478 being recognized when the domain list was tested by the match_domain
3479 condition in an expansion string.
3481 16. Added the ${str2b64: operator.
3483 17. Exim was always calling setrlimit() to set a large limit for the number of
3484 processes, without checking whether the existing limit was already
3485 adequate. (It did check for the limit on file descriptors.) Furthermore,
3486 errors from getrlimit() and setrlimit() were being ignored. Now they are
3487 logged to the main and panic logs, but Exim does carry on, to try to do its
3488 job under whatever limits there are.
3490 18. Imported PCRE 5.0.
3492 19. Trivial typo in log message " temporarily refused connection" (the leading
3495 20. If the log selector return_path_on_delivery was set and an address was
3496 redirected to /dev/null, the delivery process crashed because it assumed
3497 that a return path would always be set for a "successful" delivery. In this
3498 case, the whole delivery is bypassed as an optimization, and therefore no
3501 21. Internal re-arrangement: the function for sending a challenge and reading
3502 a response while authentication was assuming a zero-terminated challenge
3503 string. It's now changed to take a pointer and a length, to allow for
3504 binary data in such strings.
3506 22. Added the cyrus_sasl authenticator (code supplied by MBM).
3508 23. Exim was not respecting finduser_retries when seeking the login of the
3509 uid under which it was called; it was always trying 10 times. (The default
3510 setting of finduser_retries is zero.) Also, it was sleeping after the final
3511 failure, which is pointless.
3513 24. Implemented tls_on_connect_ports.
3515 25. Implemented acl_smtp_predata.
3517 26. If the domain in control=submission is set empty, Exim assumes that the
3518 authenticated id is a complete email address when it generates From: or
3519 Sender: header lines.
3521 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
3522 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
3523 chown and chgrp in /bin and hostname in /usr/bin.
3525 28. Exim was keeping the "process log" file open after each use, just as it
3526 does for the main log. This opens the possibility of it remaining open for
3527 long periods when the USR1 signal hits a daemon. Occasional processlog
3528 errors were reported, that could have been caused by this. Anyway, it seems
3529 much more sensible not to leave this file open at all, so that is what now
3532 29. The long-running daemon process does not normally write to the log once it
3533 has entered its main loop, and it closes the log before doing so. This is
3534 so that log files can straightforwardly be renamed and moved. However,
3535 there are a couple of unusual error situations where the daemon does write
3536 log entries, and I had neglected to close the log afterwards.
3538 30. The text of an SMTP error response that was received during a remote
3539 delivery was being truncated at 512 bytes. This is too short for some of
3540 the long messages that one sometimes sees. I've increased the limit to
3543 31. It is now possible to make retry rules that apply only when a message has a
3544 specific sender, in particular, an empty sender.
3546 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
3547 it possible to be selective about when SMTP synchronization is enforced.
3549 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
3551 32. Implemented hosts_connection_nolog.
3553 33. Added an ACL for QUIT.
3555 34. Setting "delay_warning=" to disable warnings was not working; it gave a
3558 35. Added mailbox_size and mailbox_filecount to appendfile.
3560 36. Added control = no_multiline_responses to ACLs.
3562 37. There was a bug in the logic of the code that waits for the clock to tick
3563 in the case where the clock went backwards by a substantial amount such
3564 that the microsecond fraction of "now" was more than the microsecond
3565 fraction of "then" (but the whole seconds number was less).
3567 38. Added support for the libradius Radius client library this is found on
3568 FreeBSD (previously only the radiusclient library was supported).
3574 1. When certain lookups returned multiple values in the form name=value, the
3575 quoting of the values was not always being done properly. Specifically:
3576 (a) If the value started with a double quote, but contained no whitespace,
3578 (b) If the value contained whitespace other than a space character (i.e.
3579 tabs or newlines or carriage returns) it was not quoted.
3580 This fix has been applied to the mysql and pgsql lookups by writing a
3581 separate quoting function and calling it from the lookup code. The fix
3582 should probably also be applied to nisplus, ibase and oracle lookups, but
3583 since I cannot test any of those, I have not disturbed their existing code.
3585 2. A hit in the callout cache for a specific address caused a log line with no
3586 reason for rejecting RCPT. Now it says "Previous (cached) callout
3587 verification failure".
3589 3. There was an off-by-one bug in the queryprogram router. An over-long
3590 return line was truncated at 256 instead of 255 characters, thereby
3591 overflowing its buffer with the terminating zero. As well as fixing this, I
3592 have increased the buffer size to 1024 (and made a note to document this).
3594 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
3595 when Exim is waiting for an SMTP response from a remote server, Exim
3596 restarts its select() call on the socket, thereby resetting its timeout.
3597 This is not a problem when such interrupts are rare. Somebody set up a cron
3598 job to run exiwhat every 2 minutes, which is less than the normal select()
3599 timeout (5 or 10 minutes). This meant that the select() timeout never
3600 kicked in because it was always reset. I have fixed this by comparing the
3601 time when an interrupt arrives with the time at the start of the first call
3602 to select(). If more time than the timeout has elapsed, the interrupt is
3603 treated as a timeout.
3605 5. Some internal re-factoring in preparation for the addition of Sieve
3606 extensions (by MH). In particular, the "personal" test is moved to a
3607 separate function, and given an option for scanning Cc: and Bcc: (which is
3608 not set for Exim filters).
3610 6. When Exim created an email address using the login of the caller as the
3611 local part (e.g. when creating a From: or Sender: header line), it was not
3612 quoting the local part when it contained special characters such as @.
3614 7. Installed new OpenBSD configuration files.
3616 8. Reworded some messages for syntax errors in "and" and "or" conditions to
3617 try to make them clearer.
3619 9. Callout options, other than the timeout value, were being ignored when
3620 verifying sender addresses in header lines. For example, when using
3622 verify = header_sender/callout=no_cache
3624 the cache was (incorrectly) being used.
3626 10. Added a missing instance of ${EXE} to the exim_install script; this affects
3627 only the Cygwin environment.
3629 11. When return_path_on_delivery was set as a log selector, if different remote
3630 addresses in the same message used different return paths and parallel
3631 remote delivery occurred, the wrong values would sometimes be logged.
3632 (Whenever a remote delivery process finished, the return path value from
3633 the most recently started remote delivery process was logged.)
3635 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
3636 lines when AUTH and/or TLS are in use. This is the "received protocol"
3637 field. Exim used to use "asmtp" for authenticated SMTP, without any
3638 indication (in the protocol name) for TLS use. Now it follows the RFC and
3639 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
3640 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
3641 names appear in log lines as well as in Received: header lines.
3643 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
3644 extensions, and comparison tests, and to fix some bugs.
3646 14. Changes to the "personal" filter test:
3648 (1) The test was buggy in that it was just doing the equivalent of
3649 "contains" tests on header lines. For example, if a user's address was
3650 anne@some.where, the "personal" test would incorrectly be true for
3652 To: susanne@some.where
3654 This test is now done by extracting each address from the header in turn,
3655 and checking the entire address. Other tests that are part of "personal"
3656 are now done using regular expressions (for example, to check local parts
3657 of addresses in From: header lines).
3659 (2) The list of non-personal local parts in From: addresses has been
3660 extended to include "listserv", "majordomo", "*-request", and "owner-*",
3661 taken from the Sieve specification recommendations.
3663 (3) If the message contains any header line starting with "List-" it is
3664 treated as non-personal.
3666 (4) The test for "circular" in the Subject: header line has been removed
3667 because it now seems ill-conceived.
3669 15. Minor typos in src/EDITME comments corrected.
3671 16. Installed latest exipick from John Jetmore.
3673 17. If headers_add on a router specified a text string that was too long for
3674 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
3675 of string_sprintf() is now avoided.
3677 18. $message_body_size was not set (it was always zero) when running the DATA
3678 ACL and the local_scan() function.
3680 19. For the "mail" command in an Exim filter, no default was being set for
3681 the once_repeat time, causing a random time value to be used if "once" was
3682 specified. (If the value happened to be <= 0, no repeat happened.) The
3683 default is now 0s, meaning "never repeat". The "vacation" command was OK
3684 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
3685 (I found it when inspecting the code).
3687 20. There is now an overall timeout for performing a callout verification. It
3688 defaults to 4 times the callout timeout, which applies to individual SMTP
3689 commands during the callout. The overall timeout applies when there is more
3690 than one host that can be tried. The timeout is checked before trying the
3691 next host. This prevents very long delays if there are a large number of
3692 hosts and all are timing out (e.g. when the network connections are timing
3693 out). The value of the overall timeout can be changed by specifying an
3694 additional sub-option for "callout", called "maxwait". For example:
3696 verify = sender/callout=5s,maxwait=20s
3698 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
3699 to the end before writing, but this should make it even safer).
3701 22. Exim was forgetting that it had advertised PIPELINING for the second and
3702 subsequent messages on an SMTP connection. It was also not resetting its
3703 memory on STARTTLS and an internal HELO.
3705 23. When Exim logs an SMTP synchronization error within a session, it now
3706 records whether PIPELINING has been advertised or not.
3708 24. Added 3 instances of "(long int)" casts to time_t variables that were being
3709 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
3710 rather than long int.
3712 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
3714 26. Added the never_mail option to autoreply.
3720 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
3721 crash if the getsockname() call failed; this can happen if a connection is
3722 closed very soon after it is established. The problem was simply in the
3723 order in which certain operations were done, causing Exim to try to write
3724 to the SMTP stream before it had set up the file descriptor. The bug has
3725 been fixed by making things happen in the correct order.
3731 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
3732 before the connection was closed, thus losing the rejection response.
3734 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
3735 some early Solaris releases, but causes trouble in current releases where
3736 socklen_t is defined.
3738 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
3741 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
3744 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
3745 that releasing the top of it at the end releases what was used for sub-
3746 expansions (unless the block got too big). However, discard this block if
3747 the first thing is a variable or header, so that we can use its block when
3748 it is dynamic (useful for very large $message_headers, for example).
3750 6. Lookups now cache *every* query, not just the most recent. A new, separate
3751 store pool is used for this. It can be recovered when all lookup caches are
3752 flushed. Lookups now release memory at the end of their result strings.
3753 This has involved some general refactoring of the lookup sources.
3755 7. Some code has been added to the store_xxx() functions to reduce the amount
3756 of flapping under certain conditions.
3758 8. log_incoming_interface used to affect only the <= reception log lines. Now
3759 it causes the local interface and port to be added to several more SMTP log
3760 lines, for example "SMTP connection from", and rejection lines.
3762 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
3764 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
3766 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
3767 could be overwritten at the end of the current message (or the start of a
3768 new message if it was set in a HELO ACL). The value is now preserved for
3769 the duration of the SMTP connection.
3771 12. If a transport had a headers_rewrite setting, and a matching header line
3772 contained an unqualified address, that address was qualified, even if it
3773 did not match any rewriting rules. The underlying bug was that the values
3774 of the flags that permit the existence of unqualified sender and recipient
3775 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
3776 non-local messages, and by -bnq for local messages) were not being
3777 preserved with the message after it was received.
3779 13. When Exim was logging an SMTP synchronization error, it could sometimes log
3780 "next input=" as part of the text comprising the host identity instead of
3781 the correct text. The code was using the same buffer for two different
3782 strings. However, depending on which order the printing function evaluated
3783 its arguments, the bug did not always show up. Under Linux, for example, my
3784 test suite worked just fine.
3786 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
3787 doesn't work with some older versions of Perl. It has been changed to "my",
3788 which in any case is probably the better facility to use.
3790 15. A really picky compiler found some instances of statements for creating
3791 error messages that either had too many or two few arguments for the format
3794 16. The size of the buffer for calls to the DNS resolver has been increased
3795 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
3796 for addresses that have a lot of PTR records. This alleviates a problem; it
3797 does not fully solve it.
3799 17. A dnsdb lookup for PTR records that receives more data than will fit in the
3800 buffer now truncates the list and logs the incident, which is the same
3801 action as happens when Exim is looking up a host name and its aliases.
3802 Previously in this situation something unpredictable would happen;
3803 sometimes it was "internal error: store_reset failed".
3805 18. If a server dropped the connection unexpectedly when an Exim client was
3806 using GnuTLS and trying to read a response, the client delivery process
3807 crashed while trying to generate an error log message.
3809 19. If a "warn" verb in an ACL added multiple headers to a message in a single
3810 string, for example:
3812 warn message = H1: something\nH2: something
3814 the text was added as a single header line from Exim's point of view
3815 though it ended up OK in the delivered message. However, searching for the
3816 second and subsequent header lines using $h_h2: did not work. This has been
3817 fixed. Similarly, if a system filter added multiple headers in this way,
3818 the routers could not see them.
3820 20. Expanded the error message when iplsearch is called with an invalid key to
3821 suggest using net-iplsearch in a host list.
3823 21. When running tests using -bh, any delays imposed by "delay" modifiers in
3824 ACLs are no longer actually imposed (and a message to that effect is
3827 22. If a "gecos" field in a passwd entry contained escaped characters, in
3828 particular, if it contained a \" sequence, Exim got it wrong when building
3829 a From: or a Sender: header from that name. A second bug also caused
3830 incorrect handling when an unquoted " was present following a character
3831 that needed quoting.
3833 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
3834 was not being matched caselessly.
3836 24. Arranged for all hyphens in the exim.8 source to be escaped with
3839 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
3840 properly. Recipient callout cache records were still being keyed to include
3841 the sender, even when use_sender was set false. This led to far more
3842 callouts that were necessary. The sender is no longer included in the key
3843 when use_sender is false.
3845 26. Added "control = submission" modifier to ACLs.
3847 27. Added the ${base62d: operator to decode base 62 numbers.
3849 28. dnsdb lookups can now access SRV records.
3851 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
3852 the configuration file.
3854 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
3855 (-v) mode. This makes the output for a verbose queue run more intelligible.
3857 31. Added a use_postmaster feature to recipient callouts.
3859 32. Added the $body_zerocount variable, containing the number of binary zero
3860 bytes in the message body.
3862 33. The time of last modification of the "new" subdirectory is now used as the
3863 "mailbox time last read" when there is a quota error for a maildir
3866 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
3868 35. Added +ignore_unknown as a special item in host lists.
3870 36. Code for decoding IPv6 addresses in host lists is now included, even if
3871 IPv6 support is not being compiled. This fixes a bug in which an IPv6
3872 address was recognized as an IP address, but was then not correctly decoded
3873 into binary, causing unexpected and incorrect effects when compared with
3880 1. Very minor rewording of debugging text in manualroute to say "list of
3881 hosts" instead of "hostlist".
3883 2. If verify=header_syntax was set, and a header line with an unqualified
3884 address (no domain) and a large number of spaces between the end of the
3885 name and the colon was received, the reception process suffered a buffer
3886 overflow, and (when I tested it) crashed. This was caused by some obsolete
3887 code that should have been removed. The fix is to remove it!
3889 3. When running in the test harness, delay a bit after writing a bounce
3890 message to get a bit more predictability in the log output.
3892 4. Added a call to search_tidyup() just before forking a reception process. In
3893 theory, someone could use a lookup in the expansion of smtp_accept_max_
3894 per_host which, without the tidyup, could leave open a database connection.
3896 5. Added the variables $recipient_data and $sender_data which get set from a
3897 lookup success in an ACL "recipients" or "senders" condition, or a router
3898 "senders" option, similar to $domain_data and $local_part_data.
3900 6. Moved the writing of debug_print from before to after the "senders" test
3903 7. Change 4.31/66 (moving the time when the Received: is generated) caused
3904 problems for message scanning, either using a data ACL, or using
3905 local_scan() because the Received: header was not generated till after they
3906 were called (in order to set the time as the time of reception completion).
3907 I have revised the way this works. The header is now generated after the
3908 body is received, but before the ACL or local_scan() are called. After they
3909 are run, the timestamp in the header is updated.
3915 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
3916 before starting a queue runner without re-exec. This happened only when
3917 deliver_drop_privilege was set or when the Exim user was set to root. The
3918 effect of the bug was that timeouts during subsequent deliveries caused
3919 crashes instead of being properly handled. The handler is now left at its
3920 default (and expected) setting.
3922 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
3923 message, again when deliver_drop_privilege is set or Exim is run as root.
3924 The bug described in (1) was not present in this case, but the tidying up
3925 of the other signals was missing. I have made the two cases consistent.
3927 3. The ignore_target_hosts setting on a manualroute router was being ignored
3928 for hosts that were looked up using the /MX notation.
3930 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
3933 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
3934 operated on the sender address. After changing the $sender_address to <>
3935 for the sender address verify, Exim was re-instated it as the original
3936 (before rewriting) address, but remembering that it had rewritten it, so it
3937 wasn't rewriting it again. This bug also had the effect of breaking the
3938 sender address verification caching when the sender address was rewritten.
3940 6. The ignore_target_hosts option was being ignored by the ipliteral router.
3941 This has been changed so that if the ip literal address matches
3942 ignore_target_hosts, the router declines.
3944 7. Added expansion conditions match_domain, match_address, and match_local_
3945 part (NOT match_host).
3947 8. The placeholder for the Received: header didn't have a length field set.
3949 9. Added code to Exim itself and to exim_lock to test for a specific race
3950 condition that could lead to file corruption when using MBX delivery. The
3951 issue is with the lockfile that is created in /tmp. If this file is removed
3952 after a process has opened it but before that process has acquired a lock,
3953 there is the potential for a second process to recreate the file and also
3954 acquire a lock. This could lead to two Exim processes writing to the file
3955 at the same time. The added code performs the same test as UW imapd; it
3956 checks after acquiring the lock that its file descriptor still refers to
3957 the same named file.
3959 10. The buffer for building added header lines was of fixed size, 8192 bytes.
3960 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
3963 11. Added the smtp_active_hostname option. If used, this will typically be made
3964 to depend on the incoming interface address. Because $interface_address is
3965 not set up until the daemon has forked a reception process, error responses
3966 that can happen earlier (such as "too many connections") no longer contain
3969 12. If an expansion in a condition on a "warn" statement fails because a lookup
3970 defers, the "warn" statement is abandoned, and the next ACL statement is
3971 processed. Previously this caused the whole ACL to be aborted.
3973 13. Added the iplsearch lookup type.
3975 14. Added ident_timeout as a log selector.
3977 15. Added tls_certificate_verified as a log selector.
3979 16. Added a global option tls_require_ciphers (compare the smtp transport
3980 option of the same name). This controls incoming TLS connections.
3982 17. I finally figured out how to make tls_require_ciphers do a similar thing
3983 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
3984 before starting the TLS session.
3986 18. Tabs are now shown as \t in -bP output.
3988 19. If the log selector return_path_on_delivery was set, Exim crashed when
3989 bouncing a message because it had too many Received: header lines.
3991 20. If two routers both had headers_remove settings, and the first one included
3992 a superfluous trailing colon, the final name in the first list and the
3993 first name in the second list were incorrectly joined into one item (with a
3994 colon in the middle).
4000 1. Added -C and -D options to the exinext utility, mainly to make it easier
4001 to include in the automated testing, but these could be helpful when
4002 multiple configurations are in use.
4004 2. The exinext utility was not formatting the output nicely when there was
4005 an alternate port involved in the retry record key, nor when there was a
4006 message id as well (for retries that were specific to a specific message
4007 and a specific host). It was also confused by IPv6 addresses, because of
4008 the additional colons they contain. I have fixed the IPv4 problem, and
4009 patched it up to do a reasonable job for IPv6.
4011 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
4012 delivery, the log line now contains "pipelined" if PIPELINING was used.
4014 4. An SMTP transport process used to panic and die if the bind() call to set
4015 an explicit outgoing interface failed. This has been changed; it is now
4016 treated in the same way as a connect() failure.
4018 5. A reference to $sender_host_name in the part of a conditional expansion
4019 that was being skipped was still causing a DNS lookup. This no longer
4022 6. The def: expansion condition was not recognizing references to header lines
4023 that used bh_ and bheader_.
4025 7. Added the _cache feature to named lists.
4027 8. The code for checking quota_filecount in the appendfile transport was
4028 allowing one more file than it should have been.
4030 9. For compatibility with Sendmail, the command line option
4038 and sets the incoming protocol and host name (for trusted callers). The
4039 host name and its colon can be omitted when only the protocol is to be set.
4040 Note the Exim already has two private options, -pd and -ps, that refer to
4041 embedded Perl. It is therefore impossible to set a protocol value of "d" or
4042 "s", but I don't think that's a major issue.
4044 10. A number of refactoring changes to the code, none of which should affect
4047 (a) The number of logging options was getting close to filling up the
4048 32-bit word that was used as a bit map. I have split them into two classes:
4049 those that are passed in the argument to log_write(), and those that are
4050 only ever tested independently outside of that function. These are now in
4051 separate 32-bit words, so there is plenty of room for expansion again.
4052 There is no change in the user interface or the logging behaviour.
4054 (b) When building, for example, log lines, the code previously used a
4055 macro that called string_cat() twice, in order to add two strings. This is
4056 not really sufficiently general. Furthermore, there was one instance where
4057 it was actually wrong because one of the argument was used twice, and in
4058 one call a function was used. (As it happened, calling the function twice
4059 did not affect the overall behaviour.) The macro has been replaced by a
4060 function that can join an arbitrary number of extra strings onto a growing
4063 (c) The code for expansion conditions now uses a table and a binary chop
4064 instead of a serial search (which was left over from when there were very
4065 few conditions). Also, it now recognizes conditions like "pam" even when
4066 the relevant support is not compiled in: a suitably worded error message is
4067 given if an attempt is made to use such a condition.
4069 11. Added ${time_interval:xxxxx}.
4071 12. A bug was causing one of the ddress fields not to be passed back correctly
4072 from remote delivery subprocesses. The field in question was not being
4073 subsequently used, so this caused to problems in practice.
4075 13. Added new log selectors queue_time and deliver_time.
4077 14. Might have fixed a bug in maildirsizefile handling that threw up
4078 "unexpected character" debug warnings, and recalculated the data
4079 unnecessarily. In any case, I expanded the warning message to give more
4082 15. Added the message "Restricted characters in address" to the statements in
4083 the default ACL that block characters like @ and % in local parts.
4085 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
4086 Three changes have been made:
4088 (a) There was a serious bug; a negative response to MAIL caused the whole
4089 recipient domain to be cached as invalid, thereby blocking all messages
4090 to all local parts at the same domain, from all senders. This bug has
4091 been fixed. The domain is no longer cached after a negative response to
4092 MAIL if the sender used is not empty.
4094 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
4097 (c) A new callout option, "use_sender" has been added for people who want
4098 the modified behaviour.
4104 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
4107 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
4108 indeed breaks things for older releases.
4110 3. Added additional logging to the case where there is a problem reading data
4111 from a filter that is running in a subprocess using a pipe, in order to
4112 try to track down a specific problem.
4114 4. Testing facility fudge: when running in the test harness and attempting
4115 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
4116 getting "No route to host". Convert this to a timeout.
4118 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
4121 6. Some OS don't have socklen_t but use size_t instead. This affects the
4122 fifth argument of getsockopt() amongst other things. This is now
4123 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
4124 can be set for individual OS. I have set it for SunOS5, OSF1, and
4125 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
4126 some earlier ones do not.
4128 7. Change 4.30/15 was not doing the test caselessly.
4130 8. The standard form for an IPv6 address literal was being rejected by address
4131 parsing in, for example, MAIL and RCPT commands. An example of this kind of
4132 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
4133 this, as well as the form without the "IPv6" on the front (but only when
4134 address literals are enabled, of course).
4136 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
4138 10. Exim crashed if a message with an empty sender address specified by -f
4139 encountered a router with an errors_to setting. This could be provoked only
4140 by a command such as
4144 where an empty string was supplied; "<>" did not hit this bug.
4146 11. Installed PCRE release 4.5.
4148 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
4149 remained set. It is now erased.
4151 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
4152 times from message ids (which are base 36 rather than the normal 62).
4154 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
4155 were being counted as actual protocol errors, and logged if the log
4156 selector +smtp_protocol_error was set. One cannot be perfect in this test,
4157 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
4158 and DATA following a set of rejected RCPTs do not count as protocol errors.
4159 In other words, Exim assumes they were pipelined, though this may not
4160 actually be the case. Of course, in all cases the client gets an
4161 appropriate error code.
4163 15. If a lookup fails in an ACL condition, a message about the failure may
4164 be available; it is used if testing the ACL cannot continue, because most
4165 such messages specify what the cause of the deferral is. However, some
4166 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
4167 that caused an old message to be retained and used if a later statement
4168 caused a defer, replacing the real cause of the deferral.
4170 16. If an IP address had so many PTR records that the DNS lookup buffer
4171 was not large enough to hold them, Exim could crash while trying to process
4172 the truncated data. It now detects and logs this case.
4174 17. Further to 4.21/58, another change has been made: if (and only if) the
4175 first line of a message (the first header line) ends with CRLF, a bare LF
4176 in a subsequent header line has a space inserted after it, so as not to
4177 terminate the header.
4179 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
4180 unnecessarily, used atoi() instead of strtol(), and didn't check the
4181 termination when getting file sizes from file names by regex.
4183 19. Completely re-implemented the support for maildirsize files, in the light
4184 of a number of problems with the previous contributed implementation
4185 (4.30/29). In particular:
4187 . If the quota is zero, the maildirsize file is maintained, but no quota is
4190 . If the maildir directory does not exist, it is created before any attempt
4191 to write a maildirsize file.
4193 . The quota value in the file is just a cache; if the quota is changed in
4194 the transport, the new value overrides.
4196 . A regular expression is available for excluding directories from the
4199 20. The autoreply transport checks the characters in options that define the
4200 message's headers; it allows continued headers, but it was checking with
4201 isspace() after an embedded newline instead of explicitly looking for a
4204 21. If all the "regular" hosts to which an address was routed had passed their
4205 expiry times, and had not reached their retry times, the address was
4206 bounced, even if fallback hosts were defined. Now Exim should go on to try
4209 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
4210 equivalent code in the SMTP transport. Some hosts send humungous responses
4211 to HELO/EHLO, more than 1024 it seems.
4213 23. Refactoring: code in filter.c used (void *) for "any old type" but this
4214 gives compiler warnings in some environments. I've now done it "properly",
4217 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
4218 (because of problems with the built-in one) was declared to return uschar *
4219 instead of char *, causing compiler failure.
4221 25. Fixed a file descriptor leak when processing alias/forward files.
4223 26. Fixed a minor format string issue in dbfn.c.
4225 27. Typo in exim.c: ("dmbnz" for "dbmnz").
4227 28. If a filter file refered to $h_xxx or $message_headers, and the headers
4228 contained RFC 2047 "words", Exim's memory could, under certain conditions,
4231 29. When a sender address is verified, it is cached, to save repeating the test
4232 when there is more than one recipient in a message. However, when the
4233 verification involves a callout, it is possible for different callout
4234 options to be set for different recipients. It is too complicated to keep
4235 track of this in the cache, so now Exim always runs a verification when a
4236 callout is required, relying on the callout cache for the optimization.
4237 The overhead is duplication of the address routing, but this should not be
4240 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
4241 to be verified with callout=postmaster, and the main callout worked but the
4242 postmaster check failed, the verification correctly failed. However, if a
4243 subsequent RCPT command asked for sender verification *without* the
4244 postmaster check, incorrect caching caused this verification also to fail,
4247 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
4248 it was not caching the DNS options (qualify_single, search_parents) that
4249 were used when the lookup failed. A subsequent lookup with different
4250 options therefore always gave the same answer, though there were cases
4251 where it should not have. (Example: a "domains = !$mx_any" option on a
4252 dnslookup router: the "domains" option is always processed without any
4253 widening, but the router might have qualify_single set.) Now Exim uses the
4254 cached value only when the same options are set.
4256 32. Added John Jetmore's "exipick" utility to the distribution.
4258 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
4259 than a timeout (e.g. a certificate is required, and is not provided), an
4260 Exim server now closes the connection immediately. Previously it waited for
4261 the client to close - but if the client is SSL, it seems that they each
4262 wait for each other, leading to a delay before one of them times out.
4264 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
4265 maintained 0.8.x compatibility because I don't think many are using it, and
4266 it is clearly obsolete.
4268 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
4271 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
4272 client certificate was expired. A simple patch fixes this, though I don't
4273 understand the full logic of why the verify callback is called multiple
4276 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
4277 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
4278 which causes problems with some clients (such as the Certicom SSL Plus
4279 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
4280 disables the coutermeasure allowing Eudora to connect."
4282 38. Exim was not checking that a write() to a log file succeeded. This could
4283 lead to Bad Things if a log got too big, in particular if it hit a file
4284 size limit. Exim now panics and dies if it cannot write to a log file, just
4285 as it does if it cannot open a log file.
4287 39. Modified OS/Makefile-Linux so that it now contains
4289 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
4291 The two -D definitions ensure that Exim is compiled with large file
4292 support, which makes it possible to handle log files that are bigger than
4295 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
4296 instance) a domain was checked against a named list that involved a lookup,
4297 causing $domain_data to be set, then another domain was checked against the
4298 same list, then the first domain was re-checked, the value of $domain_data
4299 after the final check could be wrong. In particular, if the second check
4300 failed, it could be set empty. This bug probably also applied to
4303 41. The strip_trailing_dot option was not being applied to the address given
4304 with the -f command-line option.
4306 42. The code for reading a message's header from the spool was incrementing
4307 $received_count, but never initializing it. This meant that the value was
4308 incorrect (doubled) while delivering a message in the same process in which
4309 it was received. In the most common configuration of Exim, this never
4310 happens - a fresh exec is done - but it can happen when
4311 deliver_drop_privilege is set.
4313 43. When Exim logs an SMTP synchronization error - client data sent too soon -
4314 it now includes up to 150 characters of the unexpected data in the log
4317 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
4318 and building data strings. The size of both of these buffers was 10 000
4319 bytes - far larger than anybody would *ever* want, thought I. Needless to
4320 say, somebody hit the limit. I have increased the maximum line length to
4321 20 000 and the maximum data length of concatenated lines to 100 000. I have
4322 also fixed two bugs, because there was no checking on these buffers. Tsk,
4323 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
4324 buffer is too small.
4326 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
4327 lsearch lookups. Now it does.
4329 46. When parsing a route_list item in a manualroute router, a fixed-length
4330 buffer was used for the list of hosts. I made this 1024 bytes long,
4331 thinking that nobody would ever have a list of hosts that long. Wrong.
4332 Somebody had a whole pile of complicated expansion conditions, and the
4333 string was silently truncated, leading to an expansion error. It turns out
4334 that it is easier to change to an unlimited length (owing to other changes
4335 that have happened since this code was originally written) than to build
4336 structure for giving a limitation error. The length of the item that
4337 expands into the list of hosts is now unlimited.
4339 47. The lsearch lookup could not handle data where the length of text line was
4340 more than 4095 characters. Such lines were truncated, leading to shortened
4341 data being returned. It should now handle lines of any length.
4343 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
4344 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
4347 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
4348 replacing strings such as BIN_DIRECTORY in a source file: the replacement
4349 no longer happens in comment lines. A list of replacements is now placed
4350 at the head of all of the source files, except those whose only change is
4351 to replace PERL_COMMAND in the very first #! line.
4353 50. Replaced the slow insertion sort in queue.c, for sorting the list of
4354 messages on the queue, with a bottom-up merge sort, using code contributed
4355 by Michael Haardt. This should make operations like -bp somewhat faster on
4356 large queues. It won't affect queue runners, except when queue_run_in_order
4359 51. Installed eximstats 1.31 in the distribution.
4361 52. Added support for SRV lookups to the dnslookup router.
4363 53. If an ACL referred to $message_body or $message_body_end, the value was not
4364 reset for any messages that followed in the same SMTP session.
4366 54. The store-handling optimization for building very long strings was not
4367 differentiating between the different store pools. I don't think this
4368 actually made any difference in practice, but I've tidied it.
4370 55. While running the routers to verify a sender address, $sender_address
4371 was still set to the sender address. This is wrong, because when routing to
4372 send a bounce to the sender, it would be empty. Therefore, I have changed
4373 it so that, while verifying a sender address, $sender_address is set to <>.
4374 (There is no change to what happens when verifying a recipient address.)
4376 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
4377 target A or AAAA records (if not already returned) without resetting the
4378 qualify_single or search_parents options of the DNS resolver. These are
4379 inappropriate in this case because the targets of MX and SRV records must
4380 be FQDNs. A broken DNS record could cause trouble if it happened to have a
4381 target that, when qualified, matched something in the local domain. These
4382 two options are now turned off when doing these lookups.
4384 57. It seems that at least some releases of Reiserfs (which does not have the
4385 concept of a fixed number of inodes) returns zero and not -1 for the
4386 number of available inodes. This interacted badly with check_spool_inodes,
4387 which assumed that -1 was the "no such thing" setting. What I have done is
4388 to check that the total number of inodes is greater than zero before doing
4389 the test of how many are available.
4391 58. When a "warn" ACL statement has a log_message modifier, the message is
4392 remembered, and not repeated. This is to avoid a lot of repetition when a
4393 message has many recipients that cause the same warning to be written.
4394 Howewer, Exim was preserving the list of already written lines for an
4395 entire SMTP session, which doesn't seem right. The memory is now reset if a
4396 new message is started.
4398 59. The "rewrite" debugging flag was not showing the result of rewriting in the
4399 debugging output unless log_rewrite was also set.
4401 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
4402 of (int)(handle) when we know that handle contains (void *)(-1).
4404 61. The Exim daemon panic-logs an error return when it closes the incoming
4405 connection. However "connection reset by peer" seems to be common, and
4406 isn't really an error worthy of noting specially, so that particular error
4409 62. When Exim is trying to find all the local interfaces, it used to panic and
4410 die if the ioctl to get the interface flags failed. However, it seems that
4411 on at least one OS (Solaris 9) it is possible to have an interface that is
4412 included in the list of interfaces, but for which you get a failure error
4413 for this call. This happens when the interface is not "plumbed" into a
4414 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
4415 failure of the "get flags" call assumes that the interface is down.
4417 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
4418 makes life easier for people who are doing arithmetic on fields extracted
4419 from dates, where you often get leading zeros that should not be
4420 interpreted as octal.
4422 64. Added qualify_domain to the redirect router, to override the global
4425 65. If a pathologically long header line contained very many addresses (the
4426 report of this problem mentioned 10 000) and each of them was rewritten,
4427 Exim could use up a very large amount of memory. (It kept on making new
4428 copies of the header line as it rewrote, and never released the old ones.)
4429 At the expense of a bit more processing, the header rewriting function has
4430 been changed so that it no longer eats memory in this way.
4432 66. The generation of the Received: header has been moved from the time that a
4433 message starts to be received, to the time that it finishes. The timestamp
4434 in the Received: header should now be very close to that of the <= log
4435 line. There are two side-effects of this change:
4437 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
4438 logged header lines no longer include the local Received: line, because
4439 it has not yet been created. The same applies to a copy of the message
4440 that is returned to a non-SMTP sender when a message is rejected.
4442 (b) When a filter file is tested using -bf, no additional Received: header
4443 is added to the test message. After some thought, I decided that this
4446 This change does not affect the value of $received_for. It is still set
4447 after address rewriting, but before local_scan() is called.
4449 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
4451 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
4452 gave an unhelpful panic error message, and a defer error. I have managed to
4453 change this behaviour so that it now rejects any supplied certificate,
4454 which seems right, as the list of acceptable certificates is empty.
4456 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
4457 gave an unhelpful defer error. I have not managed to make this reject any
4458 supplied certificates, but the error message it gives is "no certificate
4459 supplied", which is not helpful.
4461 70. exigrep's output now also includes lines that are not associated with any
4462 message, but which match the given pattern. Implemented by a patch from
4463 Martin Sluka, which also tidied up the Perl a bit.
4465 71. Recipient callout verification, like sender verification, was using <> in
4466 the MAIL FROM command. This isn't really the right thing, since the actual
4467 sender may affect whether the remote host accepts the recipient or not. I
4468 have changed it to use the actual sender in the callout; this means that
4469 the cache record is now keyed on a recipient/sender pair, not just the
4470 recipient address. There doesn't seem to be a real danger of callout loops,
4471 since a callout by the remote host to check the sender would use <>.
4472 [SEE ABOVE: changed after hitting problems.]
4474 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
4475 temporary errors. However, in the case of such a code being given after
4476 the end of a data transmission (i.e. after ".") Exim was failing to write
4477 a retry record for the message. (Yes, there was some broken host that was
4478 actually sending 8xx at this point.)
4480 73. An unknown lookup type in a host list could cause Exim to panic-die when
4481 the list was checked. (An example that provoked this was putting <; in the
4482 middle of a list instead of at the start.) If this happened during a DATA
4483 ACL check, a -D file could be left lying around. This kind of configuration
4484 error no longer causes Exim to die; instead it causes a defer errror. The
4485 incident is still logged to the main and panic logs.
4487 74. Buglet left over from Exim 3 conversion. The message "too many messages
4488 in one connection" was written to the rejectlog but not the mainlog, except
4489 when address rewriting (yes!) was being logged.
4491 75. Added write_rejectlog option.
4493 76. When a system filter was run not as root (that is, when system_filter_user
4494 was set), the values of the $n variables were not being returned to the
4495 main process; thus, they were not subsequently available in the $sn
4498 77. Added +return_path_on_delivery log selector.
4500 78. A connection timeout was being treated differently from recipients deferred
4501 when testing hosts_max_try with a message that was older than the host's
4502 retry timeout. (The host should not be counted, thus allowing all hosts to
4503 be tried at least once before bouncing.) This may have been the cause of an
4504 occasionally reported bug whereby a message would remain on the queue
4505 longer than the retry timeout, but would be bounced if a delivery was
4506 forced. I say "may" because I never totally pinned down the problem;
4507 setting up timeout/retry tests is difficult. See also the next item.
4509 79. The ultimate address timeout was not being applied to errors that involved
4510 a combination of host plus message (for example, a timeout on a MAIL
4511 command). When an address resolved to a number of possible hosts, and they
4512 were not all tried for each delivery (e.g. because of hosts_max_try), a
4513 message could remain on the queue longer than the retry timeout.
4515 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
4518 81. Fixed an obscure SMTP outgoing bug which required at least the following
4519 conditions: (a) there was another message waiting for the same server;
4520 (b) the server returned 5xx to all RCPT commands in the first message so
4521 that the message was not completed; (c) the server dropped the connection
4522 or gave a negative response to the RSET that Exim sends to abort the
4523 transaction. The observed case was a dropped connection after DATA that had
4524 been sent in pipelining mode. That is, the server had advertised PIPELINING
4525 but was not implementing it correctly. The effect of the bug was incorrect
4526 behaviour, such as trying another host, and this could lead to a crash.
4532 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
4533 and daemon.c were passed as pointers to ints; they should have been
4534 pointers to socklen_t variables (which are typically unsigned ints).
4536 2. Some signed/unsigned type warnings in the os.c file for Linux have been
4539 3. Fixed a really odd bug that affected only the testing scheme; patching a
4540 certain fixed string in the binary changed the value of another string that
4541 happened to be identical to the end of the original first string.
4543 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
4544 name", it returns that address as the IP address. On some operating
4545 systems (e.g. Solaris), it also passes back the IP address string as the
4546 "host name". However, on others (e.g. Linux), it passes back an empty
4547 string. Exim wasn't checking for this, and was changing the host name to an
4548 empty string, assuming it had been canonicized.
4550 5. Although rare, it is permitted to have more than one PTR record for a given
4551 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
4552 all the names associated with an address, because they do in Solaris.
4553 However, it seems that they do not in Linux for data that comes from the
4554 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
4555 I found this out when I moved to a new Linux workstation and tried to run
4556 the Exim test suite.
4558 To get round this problem I have changed the code so that it now does its
4559 own call to the DNS to look up PTR records when searching for a host name.
4560 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
4561 addresses that are only in /etc/hosts are still found.
4563 This behaviour is, however, controlled by an option called host_lookup_
4564 order, which defaults to "bydns:byaddr". If people want to use the other
4565 order, or indeed, just use one or the other means of lookup, they can
4566 specify it in this variable.
4568 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
4569 some operating systems, this comes back from gethostbyaddr() as an empty
4570 string, and this is what Exim used to test for. However, it seems that in
4571 other systems, "." is yielded. Exim now tests for this case too.
4573 7. The values of check_spool_space and check_log_space are now held internally
4574 as a number of kilobytes instead of an absolute number of bytes. If a
4575 numbers is specified without 'K' or 'M', it is rounded up to the nearest
4576 kilobyte. This means that much larger values can be stored.
4578 8. Exim monitor: an attempt to get the action menu when not actually pointing
4579 at a message produces an empty menu entitled "No message selected". This
4580 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
4581 no entries in it ("Shell widget menu has zero width and/or height"). So I
4582 have added a single, blank menu entry in this case.
4584 9. Added ${quote_local_part.
4586 10. MIME decoding is now applied to the contents of Subject: header lines when
4589 11. Now that a reference to $sender_host_address automatically causes a reverse
4590 lookup to occur if necessary (4.13/18), there is no need to arrange for a
4591 host lookup before query-style lookups in lists that might use this
4592 variable. This has therefore been abolished, and the "net-" prefix is no
4593 longer necessary for query-style lookups.
4595 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
4596 have been a typo for LFLAGS, so it has been changed.
4598 13. The install script calls Exim with "-C /dev/null" in order to find the
4599 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
4600 to be output. Howeve, since Exim outputs its version number before the
4601 error, it didn't break the script. It just looked ugly. I fixed this by
4602 always allowing "-C /dev/null" if the caller is root.
4604 14. Ignore overlarge ACL variable number when reading spool file - insurance
4605 against a later release with more variables having written the file.
4607 15. The standard form for an IPv6 address literal was being rejected by EHLO.
4608 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
4609 this, as well as the form without the "IPv6" on the front.
4611 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
4612 OS/Makefile-Darwin file.
4614 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
4615 only to LDAP libraries that do not have LDAP_OPT_DEREF.
4617 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
4618 variable. However, some OS use ints for inodes. I've added cast to long int
4619 to get rid of the compiler warning.
4621 19. I had forgotten to lock out "/../" in configuration file names when
4622 ALT_CONFIG_PREFIX was set.
4624 20. Routers used for verification do not need to specify transports. However,
4625 if such a router generated a host list, and callout was configured, Exim
4626 crashed, because it could not find a port number from the (non-existent)
4627 transport. It now assumes port 25 in this circumstance.
4629 21. Added the -t option to exigrep.
4631 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
4632 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
4633 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
4634 them is set, LOOKUP_LSEARCH is forced.
4636 23. "exim -bV" now outputs a list of lookups that are included in the binary.
4638 24. Added sender and host information to the "rejected by local_scan()" log
4639 line; previously there was no indication of these.
4641 25. Added .include_if_exists.
4643 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
4644 Linux. It turns out that not all file systems support this. Apparently some
4645 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
4646 do it.) To cope with this, the error EINVAL, which means that sync-ing is
4647 not supported on the file descriptor, is now ignored when Exim is trying to
4648 sync a directory. This applies only to Linux.
4650 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
4652 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
4653 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
4656 29. Added support for maildirsize files from supplied patch (modified a bit).
4658 30. The use of :fail: followed by an empty string could lead Exim to respond to
4659 sender verification failures with (e.g.):
4661 550 Verification failed for <xxx>
4662 550 Sender verify failed
4664 where the first response line was missing the '-' that indicates it is not
4665 the final line of the response.
4667 31. The loop for finding the name of the user that called Exim had a hardwired
4668 limit of 10; it now uses the value of finduser_retries, which is used for
4669 all other user lookups.
4671 32. Added $received_count variable, available in data and not_smtp ACLs, and at
4674 33. Exim was neglecting to zero errno before one call of strtol() when
4675 expanding a string and expecting an integer value. On some systems this
4676 resulted in spurious "integer overflow" errors. Also, it was casting the
4677 result into an int without checking.
4679 34. Testing for a connection timeout using "timeout_connect" in the retry rules
4680 did not work. The code looks as if it has *never* worked, though it appears
4681 to have been documented since at least releast 1.62. I have made it work.
4683 35. The "timeout_DNS" error in retry rules, also documented since at least
4684 1.62, also never worked. As it isn't clear exactly what this means, and
4685 clearly it isn't a major issue, I have abolished the feature by treating it
4686 as "timeout", and writing a warning to the main and panic logs.
4688 36. The display of retry rules for -brt wasn't always showing the error code
4691 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
4692 timeout_connect_A, timeout_connect_MX.
4694 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
4695 to the empty sender.
4697 39. The daemon was not analysing the content of -oX till after it had closed
4698 stderr and disconnected from the controlling terminal. This meant that any
4699 syntax errors were only noted on the panic log, and the return code from
4700 the command was 0. By re-arranging the code a little, I've made the
4701 decoding happen first, so such errors now appear on stderr, and the return
4702 code is 1. However, the actual setting up of the sockets still happens in
4703 the disconnected process, so errors there are still only recorded on the
4706 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
4707 connections (as happens on some IP stacks) was logged at start up time as
4708 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
4709 it from "IPv6 and IPv4", which means that two separate sockets are being
4712 41. The debug output for gethostbyname2() or getipnodebyname() failures now
4713 says whether AF_INET or AF_INET6 was passed as an argument.
4715 42. Exiwhat output was messed up when time zones were included in log
4718 43. Exiwhat now gives more information about the daemon's listening ports,
4719 and whether -tls-on-connect was used.
4721 44. The "port" option of the smtp transport is now expanded.
4723 45. A "message" modifier in a "warn" statement in a non-message ACL was being
4724 silently ignored. Now an error message is written to the main and panic
4727 46. There's a new ACL modifier called "logwrite" which writes to a log file
4728 as soon as it is encountered.
4730 47. Added $local_user_uid and $local_user_gid at routing time.
4732 48. Exim crashed when trying to verify a sender address that was being
4735 49. Exim was recognizing only a space character after ".include". It now also
4736 recognizes a tab character.
4738 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
4739 extracting the relevant information from the specification. The man page no
4740 longer contains scrambled data for the -d option, and I've added a section
4741 at the front about calling Exim under different names.
4743 51. Added "extra_headers" argument to the "mail" command in filter files.
4745 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
4748 53. Installed eximstats 1.29.
4750 54. Added transport_filter_timeout as a generic transport option.
4752 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
4753 Cc: header lines. This was required by RFC 822, but it not required by RFC
4756 56. Exim used to add From:, Date:, and Message-Id: header lines to any
4757 incoming messages that did not have them. Now it does so only if the
4758 message originates locally, that is, if there is no associated remote host
4759 address. When Resent- header lines are present, this applies to the Resent-
4760 lines rather than the non-Resent- lines.
4762 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
4763 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
4765 58. Messages for configuration errors now include the name of the main
4766 configuration file - useful now that there may be more than one file in a
4767 list (.included file names were always shown).
4769 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
4770 for those rare installations that do not start the daemon as root or run it
4771 setuid root. I've cut out the call to initgroups() if the daemon is not
4774 60. The Exim user and group can now be bound into the binary as text strings
4775 that are looked up at the start of Exim's processing.
4777 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
4779 62. Added $mailstore_basename variable.
4781 63. Installed patch to sieve.c from Michael Haardt.
4783 64. When Exim failed to open the panic log after failing to open the main log,
4784 the original message it was trying to log was written to stderr and debug
4785 output, but if they were not available (the usual case in production), it
4786 was lost. Now it is written to syslog before the two lines that record the
4787 failures to open the logs.
4789 65. Users' Exim filters run in subprocesses under the user's uid. It is
4790 possible for a "deliver" command or an alias in a "personal" command to
4791 provoke an address rewrite. If logging of address rewriting is configured,
4792 this fails because the process is not running as root or exim. There may be
4793 a better way of dealing with this, but for the moment (because 4.30 needs
4794 to be released), I have disabled address rewrite logging when running a
4795 filter in a non-root, non-exim process.
4801 1. The buildconfig auxiliary program wasn't quoting the value set for
4802 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
4803 not defined. This bug was masked in 4.22 by the effect that was fixed in
4806 2. Some messages that were rejected after a message id was allocated were
4807 shown as "incomplete" by exigrep. It no longer does this for messages that
4808 are rejected by local_scan() or the DATA or non-SMTP ACLs.
4810 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
4811 have allow_domain_literals set, the ID did not get logged in the <= line.
4812 Domain literals are now always recognized in Message-ID: header lines.
4814 4. The first argument for a ${extract expansion item is the key name or field
4815 number. Leading and trailing spaces in this item were not being ignored,
4816 causing some misleading effects.
4818 5. When deliver_drop_privilege was set, single queue runner processes started
4819 manually (i.e. by the command "exim -q") or by the daemon (which uses the
4820 same command in the process it spins off) were not dropping privilege.
4822 6. When the daemon running as "exim" started a queue runner, it always
4823 re-executed Exim in the spun-off process. This is a waste of effort when
4824 deliver_drop_privilege is set. The new process now just calls the
4825 queue-runner function directly.
4831 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
4834 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
4835 ignored. Though the use of -oP was forcing the writing of a pid file, it
4836 was always written to the default place.
4838 3. If the message "no IP address found for host xxxx" is generated during
4839 incoming verification, it is now followed by identification of the incoming
4840 connection (so you can more easily find what provoked it).
4842 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
4844 5. Added some features to "harden" Exim a bit more against certain attacks:
4846 (a) There is now a build-time option called FIXED_NEVER_USERS that can
4847 be put in Local/Makefile. This is like the never_users runtime option,
4848 but it cannot be overridden. The default setting is "root".
4850 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
4851 prefix string with which any file named in a -C command line option
4854 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
4855 is retained for -C and -D only if the caller of Exim is root. Without
4856 it, the exim user may also use -C and -D and retain privilege.
4858 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
4859 command line option is disabled.
4861 6. Macro names set by the -D option must start with an upper case letter, just
4862 like macro names defined in the configuration file.
4864 7. Added "dereference=" facility to LDAP.
4866 8. Two instances of the typo "uknown" in the source files are fixed.
4868 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
4869 the Configure-Makefile script screwed up while processing it.
4871 10. Incorporated PCRE 4.4.
4873 11. The SMTP synchronization check was not operating right at the start of an
4874 SMTP session. For example, it could not catch a HELO sent before the client
4875 waited for the greeting. There is now a check for outstanding input at the
4876 point when the greeting is written. Because of the duplex, asynchronous
4877 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
4878 way, but not yet received, when the check is performed.
4880 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
4881 on TCP/IP sockets, because this apparently causes some broken clients to
4884 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
4885 unchanged) from the Cygwin maintainer.
4887 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
4888 was supported instead of testing for mbx. Effectively a typo.
4890 15. The spa authenticator server code was not checking that the input it
4891 received was valid base64.
4893 16. The debug output line for the "set" modifier in ACLs was not showing the
4894 name of the variable that was being set.
4896 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
4898 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
4899 lookup on its own. Something else was needed to trigger the lookup. For
4900 example, a match in host_lookup or the need for a host name in a host list.
4901 Now, if $sender_host_name is referenced and the host name has not yet been
4902 looked up, a lookup is performed. If the lookup fails, the variable remains
4903 empty, and $host_lookup_failed is set to "1".
4905 19. Added "eqi" as a case-independent comparison operator.
4907 20. The saslauthd authentication condition could segfault if neither service
4908 nor realm was specified.
4910 21. If an overflowing value such as "2048M" was set for message_size_limit, the
4911 error message that was logged was misleading, and incoming SMTP
4912 connections were dropped. The message is now more accurate, and temporary
4913 errors are given to SMTP connections.
4915 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
4916 (except RSET) with a 421 error, until QUIT is received. However, it was
4917 failing to send a response to QUIT.
4919 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
4920 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
4921 ACL is now run after the helo_try_verify_hosts code.
4923 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
4924 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
4925 case-independent, so other case variants are also recognized). Apparently
4926 some systems use these upper case variants.
4928 25. If more than two messages were waiting for the same host, and a transport
4929 filter was specified for the transport, Exim sent two messages over the
4930 same TCP/IP connection, and then failed with "socket operation on non-
4931 socket" when it tried to send the third.
4933 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
4935 27. The extern definition of crypt16() in expand.c was not being excluded when
4936 the OS had its own crypt16() function.
4938 28. Added bounce_return_body as a new option, and bounce_return_size_limit
4939 as a preferred synonym for return_size_limit, both as an option and as an
4942 29. Added LIBS=-liconv to OS/Makefile-OSF1.
4944 30. Changed the default configuration ACL to relax the local part checking rule
4945 for addresses that are not in any local domains. For these addresses,
4946 slashes and pipe symbols are allowed within local parts, but the sequence
4947 /../ is explicitly forbidden.
4949 31. SPA server authentication was not clearing the challenge buffer before
4952 32. log_message in a "warn" ACL statement was writing to the reject log as
4953 well as to the main log, which contradicts the documentation and doesn't
4954 seem right (because no rejection is happening). So I have stopped it.
4956 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
4957 However, I am unable to do any testing of this.
4959 34. Fixed an infelicity in the appendfile transport. When checking directories
4960 for a mailbox, to see if any needed to be created, it was accidentally
4961 using path names with one or more superfluous leading slashes; tracing
4962 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
4964 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
4965 contents are added to the log line that is written for every discarded
4966 recipient. (Previously a log_message setting was ignored.)
4968 36. The ${quote: operator now quotes the string if it is empty.
4970 37. The install script runs exim in order to find its version number. If for
4971 some reason other than non-existence or emptiness, which it checks, it
4972 could not run './exim', it was installing it with an empty version number,
4973 i.e. as "exim-". This error state is now caught, and the installation is
4976 38. An argument was missing from the function that creates an error message
4977 when Exim fails to connect to the socket for saslauthd authentication.
4978 This could cause Exim to crash, or give a corrupted message.
4980 39. Added isip, isip4, and isip6 to ${if conditions.
4982 40. The ACL variables $acl_xx are now saved with the message, and can be
4983 accessed later in routers, transports, and filters.
4985 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
4986 strings in the file are not string-expanded.
4988 42. If a MAIL command specified a SIZE value that was too large to fit into an
4989 int variable, the check against message_size_limit failed. Such values are
4990 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
4991 day this will have to be increased, but I don't think I want to be around
4992 when emails are that large.
4999 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
5000 iconv() is not standard in FreeBSD.
5002 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
5003 IPv6 enabled. The observed symptom was a segmentation fault on return from
5004 the function os_common_find_running_interfaces() in src/os.c.
5006 3. In the check_special_case() function in daemon.c I had used "errno" as an
5007 argument name, which causes warnings on some systems. This was basically a
5008 typo, since it was named "eno" in the comments!
5010 4. The code that waits for the clock to tick (at a resolution of some fraction
5011 of a second) so as to ensure message-id uniqueness was always waiting for
5012 at least one whole tick, when it could have waited for less. [This is
5013 almost certainly not relevant at current processor speeds, where it is
5014 unlikely to ever wait at all. But we try to future-proof.]
5016 5. The function that sleeps for a time interval that includes fractions of a
5017 second contained a race. It did not block SIGALRM between setting the
5018 timer, and suspending (a couple of lines later). If the interval was short
5019 and the sigsuspend() was delayed until after it had expired, the suspension
5020 never ended. On busy systems this could lead to processes getting stuck for
5023 6. Some uncommon configurations may cause a lookup to happen in a queue runner
5024 process, before it forks any delivery processes. The open lookup caching
5025 mechanism meant that the open file or database connection was passed into
5026 the delivery process. The problem was that delivery processes always tidy
5027 up cached lookup data. This could cause a problem for the next delivery
5028 process started by the queue runner, because the external queue runner
5029 process does not know about the closure. So the next delivery process
5030 still has data in the lookup cache. In the case of a file lookup, there was
5031 no problem because closing a file descriptor in a subprocess doesn't affect
5032 the parent. However, if the lookup was caching a connection to a database,
5033 the connection was closed, and the second delivery process was likely to
5034 see errors such as "PGSQL: query failed: server closed the connection
5035 unexpectedly". The problem has been fixed by closing all cached lookups
5036 in a queue runner before running a delivery process.
5038 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
5039 seem to have the "const" qualifier which it has on other OS. I've
5042 8. Change 4.21/2 was too strict. It is only if there are two authenticators
5043 *of the same type* (client or server) with the same public name that an
5044 error should be diagnosed.
5046 9. When Exim looked up a host name for an IP address, but failed to find the
5047 original IP address when looking up the host name (a safety check), it
5048 output the message "<ip address> does not match any IP for NULL", which was
5049 confusing, to say the least. The bug was that the host name should have
5050 appeared instead of "NULL".
5052 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
5053 user that is built into the binary, and the -C or -D options is used, root
5054 privilege is dropped before the configuration file is read. In addition,
5055 logging is switched to stderr instead of the normal log files. If the
5056 configuration then re-defines the Exim user, the unprivileged environment
5057 is probably not what is expected, so Exim logs a panic warning message (but
5060 However, if deliver_drop_privilege is set, the unprivileged state may well
5061 be exactly what is intended, so the warning has been cut out in that case,
5062 and Exim is allowed to try to write to its normal log files.
5068 1. smtp_return_error_details was not giving details for temporary sender
5069 or receiver verification errors.
5071 2. Diagnose a configuration error if two authenticators have the same public
5074 3. Exim used not to create the message log file for a message until the first
5075 delivery attempt. This could be confusing when incoming messages were held
5076 for policy or load reasons. The message log file is now created at the time
5077 the message is received, and an initial "Received" line is written to it.
5079 4. The automatically generated man page for command line options had a minor
5080 bug that caused no ill effects; however, a more serious problem was that
5081 the procedure for building the man page automatically didn't always
5082 operate. Consequently, release 4.20 contains an out-of-date version. This
5083 shouldn't happen again.
5085 5. When building Exim with embedded Perl support, the script that builds the
5086 Makefile was calling 'perl' to find its compile-time parameters, ignoring
5087 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
5089 6. The freeze_tell option was not being used for messages that were frozen on
5090 arrival, either by an ACL or by local_scan().
5092 7. Added the smtp_incomplete_transaction log selector.
5094 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
5095 was accepting AUTH without a new EHLO.
5097 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
5098 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
5101 10. Logging of TCP/IP connections (when configured) now happens in the main
5102 daemon process instead of the child process, so that the TCP/IP connection
5103 count is more accurate (but it can never be perfect).
5105 11. The use of "drop" in a nested ACL was not being handled correctly in the
5106 outer ACL. Now, if condition failure induced by the nested "drop" causes
5107 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
5108 or "require"), the connection is dropped.
5110 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
5111 that yield "discard" can now be used with an "accept" or a "discard" verb,
5112 but an error is generated for any others (because I can't see a useful way
5113 to define what should happen).
5115 13. When an ACL is read dynamically from a file (or anywhere else), the lines
5116 are now processed in the same way as lines in the Exim configuration file.
5117 In particular, continuation lines are supported.
5119 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
5121 15. Added -ti meaning -t -i.
5123 16. Check for letters, digits, hyphens, and dots in the names of dnslist
5124 domains, and warn by logging if others are found.
5126 17. At least on BSD, alignment is not guarenteed for the array of ifreq's
5127 returned from GIFCONF when Exim is trying to find the list of interfaces on
5128 a host. The code in os.c has been modified to copy each ifreq to an aligned
5129 structure in all cases.
5131 Also, in some cases, the returned ifreq's were being copied to a 'struct
5132 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
5133 means the last couple of bytes of an IPv6 address could be chopped if the
5134 ifreq contained only a normal sockaddr (14 bytes storage).
5136 18. Named domain lists were not supported in the hosts_treat_as_local option.
5137 An entry such as +xxxx was not recognized, and was treated as a literal
5140 19. Ensure that header lines added by a DATA ACL are included in the reject log
5141 if the ACL subsequently rejects the message.
5143 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
5144 MD5 (which is deprecated).
5146 21. When testing a filter file using -bf, Exim was writing a message when it
5147 took the sender from a "From " line in the message, but it was not doing so
5148 when it took $return_path from a Return-Path: header line. It now does.
5150 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
5151 with a valid header line field name (a series of printing characters
5152 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
5154 23. Changed "disc" in the source to "disk" to conform to the documentation and
5155 the book and for uniformity.
5157 24. Ignore Sendmail's -Ooption=value command line item.
5159 25. When execve() failed while trying to run a command in a pipe transport,
5160 Exim was returning EX_UNAVAILBLE (69) from the subprocess. However, this
5161 could be confused with a return value of 69 from the command itself. This
5162 has been changed to 127, the value the shell returns if it is asked to run
5163 a non-existent command. The wording for the related log line suggests a
5164 non-existent command as the problem.
5166 26. If received_header_text expands to an empty string, do not add a Received:
5167 header line to the message. (Well, it adds a token one on the spool, but
5168 marks it "old" so that it doesn't get used or transmitted.)
5170 27. Installed eximstats 1.28 (addition of -nt option).
5172 28. There was no check for failure on the call to getsockname() in the daemon
5173 code. This can fail if there is a shortage of resources on the system, with
5174 ENOMEM, for example. A temporary error is now given on failure.
5176 29. Contrary to the C standard, it seems that in some environments, the
5177 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
5178 program. Exim now does this explicitly; it affects the formatting of
5179 timestamps using strftime().
5181 30. If exiqsumm was given junk data, it threw up some uninitialized variable
5182 complaints. I've now initialized all the variables, to avoid this.
5184 32. Header lines added by a system filter were not being "seen" during
5185 transport-time rewrites.
5187 33. The info_callback() function passed to OpenSSL is set up with type void
5188 (*)(SSL *, int, int), as described somewhere. However, when calling the
5189 function (actually a macro) that sets it up, the type void(*)() is
5190 expected. I've put in a cast to prevent warnings from picky compilers.
5192 34. If a DNS black list lookup found a CNAME record, but there were no A
5193 records associated with the domain it pointed at, Exim crashed.
5195 35. If a DNS black list lookup returned more than one A record, Exim ignored
5196 all but the first. It now scans all returned addresses if a particular IP
5197 value is being sought. In this situation, the contents of the
5198 $dnslist_value variable are a list of all the addresses, separated by a
5201 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
5202 to accept a host name and all its aliases if the forward lookup for any of
5203 them yielded the IP address of the incoming connection. Now it accepts only
5204 those names whose forward lookup yields the correct IP address. Any other
5205 names are discarded. This closes a loophole whereby a rogue DNS
5206 administrator could create reverse DNS records to break through a
5207 wildcarded host restriction in an ACL.
5209 37. If a user filter or a system filter that ran in a subprocess used any of
5210 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
5211 the wrong values were passed to the pipe command ($thisaddress had the
5212 value of $0, $0 had the value of $1, etc). This bug was introduced by
5213 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
5215 38. Improved the line breaking for long SMTP error messages from ACLs.
5216 Previously, if there was no break point between 40 and 75 characters, Exim
5217 left the rest of the message alone. Two changes have been made: (a) I've
5218 reduced the minimum length to 35 characters; (b) if it can't find a break
5219 point between 35 and 75 characters, it looks ahead and uses the first one
5220 that it finds. This may give the occasional overlong line, but at least the
5221 remaining text gets split now.
5223 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
5224 file descriptors might be low, and that setting 1000 would always raise it.
5225 It turns out that in some environments, the limit is already over 1000 and
5226 that lowering it causes trouble. So now Exim takes care not to decrease it.
5228 40. When delivering a message, the value of $return_path is set to $sender_
5229 address at the start of routing (routers may change the value). By an
5230 oversight, this default was not being set up when an address was tested by
5231 -bt or -bv, which affected the outcome if any router or filter referred to
5234 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
5235 log only when "message" or "log_message" are set. However, if one of the
5236 conditions was an address verification, or a call to a nested ACL, the
5237 messages generated by the underlying test were being passed through. This
5238 no longer happens. The underlying message is available in $acl_verify_
5239 message for both "message" and "log_message" expansions, so it can be
5240 passed through if needed.
5242 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
5243 new expansion $bh_ to give the encoded byte string without charset
5244 translation. Translation happens only if iconv() is available; HAVE_ICONV
5245 indicates this at build time. HEADERS_CHARSET gives the charset to
5246 translate to; headers_charset can change it in the configuration, and
5247 "headers charset" can change it in an individual filter file.
5249 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
5250 that creates RFC 2047 encoded "words" labels them as that charset instead
5251 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
5252 expansion operator; (ii) when Exim creates a From: line for a local
5253 message; (iii) when a header line is rewritten to include a "phrase" part.
5255 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
5256 buggy, causing it to skip the first lines of messages whose message ID
5257 ended in 'D'. This would not have bitten before Exim release 4.14, because
5258 message IDs were unlikely to end in 'D' before then. The effect was to have
5259 incorrect size information for certain domains.
5261 45. #include "config.h" was missing at the start of the crypt16.c module. This
5262 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
5265 46. If there was a timeout during a "random" callout check, Exim treated it as
5266 a failure of the random address, and carried on sending RSET and the real
5267 address. If the delay was just some slowness somewhere, the response to the
5268 original RCPT would be taken as a response to RSET and so on, causing
5269 mayhem of various kinds.
5271 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
5272 when I implemented it. It didn't allow for the fact that some option values
5273 may legitimatetly be negative (e.g. size_addition), and it didn't even do
5274 the right test for positive values.
5276 48. Domain names in DNS records are case-independent. Exim always looks them up
5277 in lower case. Some resolvers return domain names in exactly the case they
5278 appear in the zone file, that is, they may contain uppercase letters. Not
5279 all resolvers do this - some return always lower case. Exim was treating a
5280 change of case by a resolver as a change of domain, similar to a widening
5281 of a domain abbreviation. This triggered its re-routing code and so it was
5282 trying to route what was effectively the same domain again. This normally
5283 caused routing to fail (because the router wouldn't handle the domain
5284 twice). Now Exim checks for this case specially, and just changes the
5285 casing of the domain that it ultimately uses when it transmits the message
5288 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
5291 50. If a filter generated a file delivery with a non-absolute name (possible if
5292 no home directory exists for the router), the forbid_file option was not
5295 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
5296 the existing equality matching.
5298 52. Exim was using ints instead of ino_t variables in some places where it was
5299 dealing with inode numbers.
5301 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
5302 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
5303 called TMPDIR, and if it finds it is different, it changes its value.
5305 54. The smtp_printf() function is now made available to local_scan() so
5306 additional output lines can be written before returning. There is also an
5307 smtp_fflush() function to enable the detection of a dropped connection.
5308 The variables smtp_input and smtp_batched_input are exported to
5311 55. Changed the default runtime configuration: the message "Unknown user"
5312 has been removed from the ACL, and instead placed on the localuser router,
5313 using the cannot_route_message feature. This means that any verification
5314 failures that generate their own messages won't get overridden. Similarly,
5315 the "Unrouteable address" message that was in the ACL for unverifiable
5316 relay addresses has also been removed.
5318 56. Added hosts_avoid_esmtp to the smtp transport.
5320 57. The exicyclog script was not checking for the esoteric option
5321 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
5322 will work only if exicyclog is run under the appropriate euid.
5324 58. Following a discussion on the list, the rules by which Exim recognises line
5325 endings on incoming messages have been changed. The -dropcr and drop_cr
5326 options are now no-ops, retained only for backwards compatibility. The
5327 following line terminators are recognized: LF CRLF CR. However, special
5328 processing applies to CR:
5330 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
5331 nor a local message in the state where . is a terminator.
5333 (ii) If a bare CR is encountered in a header line, an extra space is added
5334 after the line terminator so as not to end the header. The reasoning
5335 behind this is that bare CRs in header lines are most likely either
5336 to be mistakes, or people trying to play silly games.
5338 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
5339 was being incorrectly given as 18 bytes larger than it should have been.
5340 This is a VOB (very old bug).
5342 60. This may never have affected anything current, but just in case it has:
5343 When the local host is found other than at the start of a list of hosts,
5344 the local host, those with the same MX, and any that follow, are discarded.
5345 When the list in question was part of a longer list of hosts, the following
5346 hosts (not currently being processed) were also being discarded. This no
5347 longer happens. I'm not sure if this situation could ever has previously
5350 61. Added the "/MX" feature to lists of hosts in the manualroute and query
5353 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
5354 header. This is something that is recommended in a new Internet Draft, and
5355 is something that is documented as being done by Sendmail. There are two
5356 possible values. For messages generated by the autoreply transport, Exim
5359 Auto-Submitted: auto-replied
5361 whereas for all other generated messages (e.g. bounces) it adds
5363 Auto-Submitted: auto-generated
5365 63. The "personal" condition in filters now includes a test for the
5366 Auto-Submitted: header. If it contains the string "auto-" the message it
5367 not considered personal.
5369 64. Added rcpt_include_affixes as a generic transport option.
5371 65. Added queue_only_override (default true).
5373 66. Added the syslog_duplication option.
5375 67. If what should have been the first header line of a message consisted of
5376 a space followed by a colon, Exim was mis-interpreting it as a header line.
5377 It isn't of course - it is syntactically invalid and should therefore be
5378 treated as the start of the message body. The misbehaviour could have
5379 caused a number of strange effects, including loss of data in subsequent
5380 header lines, and spool format errors.
5382 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
5383 client host had authenticated. This control can now be exercised by an ACL
5384 for more flexibility.
5386 69. By default, callouts do not happen when testing with -bh. There is now a
5387 variant, -bhc, which does actually run the callout code, including
5388 consulting and updating the callout cache.
5390 70. Added support for saslauthd authentication, courtesy of Alexander
5393 71. If statvfs() failed on the spool or log directories while checking their
5394 size for availability, Exim confusingly gave the error "space shortage".
5395 Furthermore, in debugging mode it crashed with a floating point exception.
5396 These checks are done if check_{spool,log}_{space,inodes} are set, and when
5397 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
5398 serious problem, Exim now writes to the main and panic logs when this
5399 happens, with details of the failure. It then refuses to accept the
5400 incoming message, giving the message "spool directory problem" or "log
5401 directory problem" with a 421 code for SMTP messages.
5403 72. When Exim is about to re-exec itself, it ensures that the file descriptors
5404 0, 1, and 2 exist, because some OS complain for execs without them (see
5405 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
5406 descriptors. However, the code omitted to check that the open succeeded,
5407 causing mysterious errors if for some reason the permissions on /dev/null
5408 got screwed. Now Exim writes a message to the main and panic logs, and
5409 bombs out if it can't open /dev/null.
5411 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
5412 interact so that it is all more flexible. It is supposed to remain
5413 backwards compatible. Also added extra_local_interfaces.
5415 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
5416 to bomb out with an assertion failure - to the client this appears as a
5417 connection drop. This problem occurs in the part of the code that was taken
5418 from the Samba project. Fortunately, the assertion is in a very simple
5419 function, so I have fixed this by reproducing the function inline in the
5420 one place where it is called, and arranging for authentication to fail
5421 instead of killing the process with assert().
5423 75. The SPA client code was not working when the server requested OEM rather
5424 than Unicode encoding.
5426 76. Added code to make require_files with a specific uid setting more usable in
5427 the case where statting the file as root fails - usually a non-root-mounted
5428 NFS file system. When this happens and the failure is EACCES, Exim now
5429 forks a subprocess and does the per-uid checking as the relevant uid.
5431 77. Added process_log_path.
5433 78. If log_file_path was not explicitly set, a setting of check_log_space or
5434 check_log_inodes was ignored.
5436 79. If a space check for the spool or log partitions fails, the incident is now
5437 logged. Of course, in the latter case the data may get lost...
5439 80. Added the %p formatting code to string_format() so that it can be used to
5440 print addresses in debug_print(). Adjusted all the address printing in the
5441 debugging in store.c to use %p rather than %d.
5443 81. There was a concern that a line of code in smtp_in.c could overflow a
5444 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
5445 initially expressed, the concern was not well-founded, because trailing
5446 spaces are removed early. However, if the trailing spaces were followed by
5447 a NULL, they did not get removed, so the overflow was possible. Two fixes
5450 (a) I re-wrote the offending code in a cleaner fashion.
5451 (b) If an incoming SMTP command contains a NULL character, it is rejected
5454 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
5455 runs initgroups(), so that if the Exim user is in any additional groups,
5456 they will be used during message reception.
5462 The change log for 4.20 and earlier releases has been archived.
git://git.exim.org / exim.git / blob