1 Change log file for Exim from version 4.21
2 -------------------------------------------
7 PP/01 Solaris build fix for Oracle's LDAP libraries.
8 Bugzilla 1109, patch from Stephen Usher.
10 TF/01 HP/UX build fix: avoid arithmetic on a void pointer.
12 TK/01 DKIM Verification: Fix relaxed canon for empty headers w/o
15 TF/02 Fix a couple more cases where we did not log the error message
16 when unlink() failed. See also change 4.74-TF/03.
18 TF/03 Make the exiwhat support code safe for signals. Previously Exim might
19 lock up or crash if it happened to be inside a call to libc when it
20 got a SIGUSR1 from exiwhat.
22 The SIGUSR1 handler appends the current process status to the process
23 log which is later printed by exiwhat. It used to use the general
24 purpose logging code to do this, but several functions it calls are
27 The new output code in the SIGUSR1 handler is specific to the process
28 log, and simple enough that it's easy to inspect for signal safety.
29 Removing some special cases also simplifies the general logging code.
30 Removing the spurious timestamps from the process log simplifies
33 TF/04 Improved ratelimit ACL condition.
35 The /noupdate option has been deprecated in favour of /readonly which
36 has clearer semantics. The /leaky, /strict, and /readonly update modes
37 are mutually exclusive. The update mode is no longer included in the
38 database key; it just determines when the database is updated. (This
39 means that when you upgrde Exim will forget old rate measurements.)
41 Exim now checks that the per_* options are used with an update mode that
42 makes sense for the current ACL. For example, when Exim is processing a
43 message (e.g. acl_smtp_rcpt or acl_smtp_data, etc.) you can specify
44 per_mail/leaky or per_mail/strict; otherwise (e.g. in acl_smtp_helo) you
45 must specify per_mail/readonly. If you omit the update mode it defaults to
46 /leaky where that makes sense (as before) or /readonly where required.
48 The /noupdate option is now undocumented but still supported for
49 backwards compatibility. It is equivalent to /readonly except that in
50 ACLs where /readonly is required you may specify /leaky/noupdate or
51 /strict/noupdate which are treated the same as /readonly.
53 A useful new feature is the /count= option. This is a generalization
54 of the per_byte option, so that you can measure the throughput of other
55 aggregate values. For example, the per_byte option is now equivalent
56 to per_mail/count=${if >{0}{$message_size} {0} {$message_size} }.
58 The per_rcpt option has been generalized using the /count= mechanism
59 (though it's more complicated than the per_byte equivalence). When it is
60 used in acl_smtp_rcpt, the per_rcpt option adds recipients to the
61 measured rate one at a time; if it is used later (e.g. in acl_smtp_data)
62 or in a non-SMTP ACL it adds all the recipients in one go. (The latter
63 /count=$recipients_count behaviour used to work only in non-SMTP ACLs.)
64 Note that using per_rcpt with a non-readonly update mode in more than
65 one ACL will cause the recipients to be double-counted. (The per_mail
66 and per_byte options don't have this problem.)
68 The handling of very low rates has changed slightly. If the computed rate
69 is less than the event's count (usually one) then this event is the first
70 after a long gap. In this case the rate is set to the same as this event's
71 count, so that the first message of a spam run is counted properly.
73 The major new feature is a mechanism for counting the rate of unique
74 events. The new per_addr option counts the number of different
75 recipients that someone has sent messages to in the last time period. It
76 behaves like per_rcpt if all the recipient addresses are different, but
77 duplicate recipient addresses do not increase the measured rate. Like
78 the /count= option this is a general mechanism, so the per_addr option
79 is equivalent to per_rcpt/unique=$local_part@$domain. You can, for
80 example, measure the rate that a client uses different sender addresses
81 with the options per_mail/unique=$sender_address. There are further
82 details in the main documentation.
84 TF/05 Removed obsolete $Cambridge$ CVS revision strings.
86 TF/06 Removed a few PCRE remnants.
88 TF/07 Automatically extract Exim's version number from tags in the git
89 repository when doing development or release builds.
91 PP/02 Raise smtp_cmd_buffer_size to 16kB.
92 Bugzilla 879. Patch from Paul Fisher.
94 PP/03 Implement SSL-on-connect outbound with protocol=smtps on smtp transport.
95 Heavily based on revision 40f9a89a from Simon Arlott's tree.
98 PP/04 Use .dylib instead of .so for dynamic library loading on MacOS.
100 PP/05 Variable $av_failed, true if the AV scanner deferred.
101 Bugzilla 1078. Patch from John Horne.
103 PP/06 Stop make process more reliably on build failure.
104 Bugzilla 1087. Patch from Heiko Schlittermann.
106 PP/07 Make maildir_use_size_file an _expandable_ boolean.
107 Bugzilla 1089. Patch from Heiko Schlittermann.
109 PP/08 Handle ${run} returning more data than OS pipe buffer size.
110 Bugzilla 1131. Patch from Holger Weiß.
112 PP/09 Handle IPv6 addresses with SPF.
113 Bugzilla 860. Patch from Wolfgang Breyha.
115 PP/10 GnuTLS: support TLS 1.2 & 1.1.
117 Use gnutls_certificate_verify_peers2() [patch from Andreas Metzler].
120 PP/11 match_* no longer expand right-hand-side by default.
121 New compile-time build option, EXPAND_LISTMATCH_RHS.
122 New expansion conditions, "inlist", "inlisti".
124 PP/12 fix uninitialised greeting string from PP/03 (smtps client support).
130 PP/01 The new ldap_require_cert option would segfault if used. Fixed.
132 PP/02 Harmonised TLS library version reporting; only show if debugging.
133 Layout now matches that introduced for other libraries in 4.74 PP/03.
135 PP/03 New openssl_options items: no_sslv2 no_sslv3 no_ticket no_tlsv1
137 PP/04 New "dns_use_edns0" global option.
139 PP/05 Don't segfault on misconfiguration of ref:name exim-user as uid.
142 PP/06 Extra paranoia around buffer usage at the STARTTLS transition.
143 nb: Exim is not vulnerable to http://www.kb.cert.org/vuls/id/555316
145 TK/01 Updated PolarSSL code to 0.14.2.
146 Bugzilla 1097. Patch from Andreas Metzler.
148 PP/07 Catch divide-by-zero in ${eval:...}.
151 PP/08 Condition negation of bool{}/bool_lax{} did not negate. Fixed.
154 TK/02 Bugzilla 1106: CVE-2011-1764 - DKIM log line was subject to a
155 format-string attack -- SECURITY: remote arbitrary code execution.
157 TK/03 SECURITY - DKIM signature header parsing was double-expanded, second
158 time unintentionally subject to list matching rules, letting the header
159 cause arbitrary Exim lookups (of items which can occur in lists, *not*
160 arbitrary string expansion). This allowed for information disclosure.
162 PP/09 Fix another SIGFPE (x86) in ${eval:...} expansion, this time related to
163 INT_MIN/-1 -- value coerced to INT_MAX.
169 NM/01 Workround for PCRE version dependency in version reporting
172 TF/01 Update valgrind.h and memcheck.h to copies from valgrind-3.6.0.
173 This fixes portability to compilers other than gcc, notably
174 Solaris CC and HP-UX CC. Fixes Bugzilla 1050.
176 TF/02 Bugzilla 139: Avoid using the += operator in the modular lookup
177 makefiles for portability to HP-UX and POSIX correctness.
179 PP/01 Permit LOOKUP_foo enabling on the make command-line.
180 Also via indented variable definition in the Makefile.
181 (Debugging by Oliver Heesakkers).
183 PP/02 Restore caching of spamd results with expanded spamd_address.
184 Patch from author of expandable spamd_address patch, Wolfgang Breyha.
186 PP/03 Build issue: lookups-Makefile now exports LC_ALL=C
187 Improves build reliability. Fix from: Frank Elsner
189 NM/02 Fix wide character breakage in the rfc2047 coding
190 Fixes bug 1064. Patch from Andrey N. Oktyabrski
192 NM/03 Allow underscore in dnslist lookups
193 Fixes bug 1026. Patch from Graeme Fowler
195 PP/04 Bugzilla 230: Support TLS-enabled LDAP (in addition to ldaps).
196 Code patches from Adam Ciarcinski of NetBSD.
198 NM/04 Fixed exiqgrep to cope with mailq missing size issue
201 PP/05 Bugzilla 1083: when lookup expansion defers, escape the output which
202 is logged, to avoid truncation. Patch from John Horne.
204 PP/06 Bugzilla 1042: implement freeze_signal on pipe transports.
205 Patch from Jakob Hirsch.
207 PP/07 Bugzilla 1061: restrict error messages sent over SMTP to not reveal
208 SQL string expansion failure details.
209 Patch from Andrey Oktyabrski.
211 PP/08 Bugzilla 486: implement %M datestamping in log filenames.
212 Patch from Simon Arlott.
214 PP/09 New lookups functionality failed to compile on old gcc which rejects
215 extern declarations in function scope.
216 Patch from Oliver Fleischmann
218 PP/10 Use sig_atomic_t for flags set from signal handlers.
219 Check getgroups() return and improve debugging.
220 Fixed developed for diagnosis in bug 927 (which turned out to be
223 PP/11 Bugzilla 1055: Update $message_linecount for maildir_tag.
224 Patch from Mark Zealey.
226 PP/12 Bugzilla 1056: Improved spamd server selection.
227 Patch from Mark Zealey.
229 PP/13 Bugzilla 1086: Deal with maildir quota file races.
230 Based on patch from Heiko Schlittermann.
232 PP/14 Bugzilla 1019: DKIM multiple signature generation fix.
233 Patch from Uwe Doering, sign-off by Michael Haardt.
235 NM/05 Fix to spam.c to accommodate older gcc versions which dislike
236 variable declaration deep within a block. Bug and patch from
239 PP/15 lookups-Makefile IRIX compatibilty coercion.
241 PP/16 Make DISABLE_DKIM build knob functional.
243 NM/06 Bugzilla 968: child_open_uid: restore default SIGPIPE handler
244 Patch by Simon Arlott
246 TF/03 Fix valgrind.h portability to C89 compilers that do not support
247 variable argument macros. Our copy now differs from upstream.
253 TF/01 Failure to get a lock on a hints database can have serious
254 consequences so log it to the panic log.
256 TF/02 Log LMTP confirmation messages in the same way as SMTP,
257 controlled using the smtp_confirmation log selector.
259 TF/03 Include the error message when we fail to unlink a spool file.
261 DW/01 Bugzilla 139: Support dynamically loaded lookups as modules.
262 With thanks to Steve Haslam, Johannes Berg & Serge Demonchaux
263 for maintaining out-of-tree patches for some time.
265 PP/01 Bugzilla 139: Documentation and portability issues.
266 Avoid GNU Makefile-isms, let Exim continue to build on BSD.
267 Handle per-OS dynamic-module compilation flags.
269 PP/02 Let /dev/null have normal permissions.
270 The 4.73 fixes were a little too stringent and complained about the
271 permissions on /dev/null. Exempt it from some checks.
272 Reported by Andreas M. Kirchwitz.
274 PP/03 Report version information for many libraries, including
275 Exim version information for dynamically loaded libraries. Created
276 version.h, now support a version extension string for distributors
277 who patch heavily. Dynamic module ABI change.
279 PP/04 CVE-2011-0017 - check return value of setuid/setgid. This is a
280 privilege escalation vulnerability whereby the Exim run-time user
281 can cause root to append content of the attacker's choosing to
284 PP/05 Bugzilla 1041: merged DCC maintainer's fixes for return code.
287 PP/06 Bugzilla 1071: fix delivery logging with untrusted macros.
288 If dropping privileges for untrusted macros, we disabled normal logging
289 on the basis that it would fail; for the Exim run-time user, this is not
290 the case, and it resulted in successful deliveries going unlogged.
291 Fixed. Reported by Andreas Metzler.
297 PP/01 Date: & Message-Id: revert to normally being appended to a message,
298 only prepend for the Resent-* case. Fixes regression introduced in
299 Exim 4.70 by NM/22 for Bugzilla 607.
301 PP/02 Include check_rfc2047_length in configure.default because we're seeing
302 increasing numbers of administrators be bitten by this.
304 JJ/01 Added DISABLE_DKIM and comment to src/EDITME
306 PP/03 Bugzilla 994: added openssl_options main configuration option.
308 PP/04 Bugzilla 995: provide better SSL diagnostics on failed reads.
310 PP/05 Bugzilla 834: provide a permit_coredump option for pipe transports.
312 PP/06 Adjust NTLM authentication to handle SASL Initial Response.
314 PP/07 If TLS negotiated an anonymous cipher, we could end up with SSL but
315 without a peer certificate, leading to a segfault because of an
316 assumption that peers always have certificates. Be a little more
317 paranoid. Problem reported by Martin Tscholak.
319 PP/08 Bugzilla 926: switch ClamAV to use the new zINSTREAM API for content
320 filtering; old API available if built with WITH_OLD_CLAMAV_STREAM=yes
321 NB: ClamAV planning to remove STREAM in "middle of 2010".
322 CL also introduces -bmalware, various -d+acl logging additions and
323 more caution in buffer sizes.
325 PP/09 Implemented reverse_ip expansion operator.
327 PP/10 Bugzilla 937: provide a "debug" ACL control.
329 PP/11 Bugzilla 922: Documentation dusting, patch provided by John Horne.
331 PP/12 Bugzilla 973: Implement --version.
333 PP/13 Bugzilla 752: Refuse to build/run if Exim user is root/0.
335 PP/14 Build without WITH_CONTENT_SCAN. Path from Andreas Metzler.
337 PP/15 Bugzilla 816: support multiple condition rules on Routers.
339 PP/16 Add bool_lax{} expansion operator and use that for combining multiple
340 condition rules, instead of bool{}. Make both bool{} and bool_lax{}
341 ignore trailing whitespace.
343 JJ/02 prevent non-panic DKIM error from being sent to paniclog
345 JJ/03 added tcp_wrappers_daemon_name to allow host entries other than
348 PP/17 Fix malware regression for cmdline scanner introduced in PP/08.
349 Notification from Dr Andrew Aitchison.
351 PP/18 Change ClamAV response parsing to be more robust and to handle ClamAV's
352 ExtendedDetectionInfo response format.
353 Notification from John Horne.
355 PP/19 OpenSSL 1.0.0a compatibility const-ness change, should be backwards
358 PP/20 Added a CONTRIBUTING file. Fixed the documentation build to use http:
359 XSL and documented dependency on system catalogs, with examples of how
362 DW/21 Added Valgrind hooks in store.c to help it capture out-of-bounds store
365 DW/22 Bugzilla 1044: CVE-2010-4345 - partial fix: restrict default behaviour
366 of CONFIGURE_OWNER and CONFIGURE_GROUP options to no longer allow a
367 configuration file which is writeable by the Exim user or group.
369 DW/23 Bugzilla 1044: CVE-2010-4345 - part two: extend checks for writeability
370 of configuration files to cover files specified with the -C option if
371 they are going to be used with root privileges, not just the default
374 DW/24 Bugzilla 1044: CVE-2010-4345 - part three: remove ALT_CONFIG_ROOT_ONLY
375 option (effectively making it always true).
377 DW/25 Add TRUSTED_CONFIG_PREFIX_FILE option to allow alternative configuration
378 files to be used while preserving root privileges.
380 DW/26 Set FD_CLOEXEC on SMTP sockets after forking in the daemon, to ensure
381 that rogue child processes cannot use them.
383 PP/27 Bugzilla 1047: change the default for system_filter_user to be the Exim
384 run-time user, instead of root.
386 PP/28 Add WHITELIST_D_MACROS option to let some macros be overridden by the
387 Exim run-time user without dropping privileges.
389 DW/29 Remove use of va_copy() which breaks pre-C99 systems. Duplicate the
390 result string, instead of calling string_vformat() twice with the same
393 DW/30 Allow TRUSTED_CONFIG_PREFIX_FILE only for Exim or CONFIGURE_OWNER, not
394 for other users. Others should always drop root privileges if they use
395 -C on the command line, even for a whitelisted configure file.
397 DW/31 Turn TRUSTED_CONFIG_PREFIX_FILE into TRUSTED_CONFIG_FILE. No prefixes.
399 NM/01 Fixed bug #1002 - Message loss when using multiple deliveries
405 JJ/01 installed exipick 20100104.1, adding $max_received_linelength,
406 $data_path, and $header_path variables; fixed documentation bugs and
409 JJ/02 installed exipick 20100222.0, added --input-dir and --finput to allow
410 exipick to access non-standard spools, including the "frozen" queue
413 NM/01 Bugzilla 965: Support mysql stored procedures.
414 Patch from Alain Williams
416 NM/02 Bugzilla 961: Spacing fix (syntax error) on Makefile directives for NetBSD
418 NM/03 Bugzilla 955: Documentation fix for max_rcpts.
419 Patch from Andreas Metzler
421 NM/04 Bugzilla 954: Fix for unknown responses from Dovecot authenticator.
422 Patch from Kirill Miazine
424 NM/05 Bugzilla 671: Added umask to procmail example.
426 JJ/03 installed exipick 20100323.0, fixing doc bug
428 NM/06 Bugzilla 988: CVE-2010-2023 - prevent hardlink attack on sticky mail
429 directory. Notification and patch from Dan Rosenberg.
431 TK/01 PDKIM: Upgrade PolarSSL files to upstream version 0.12.1.
433 TK/02 Improve log output when DKIM signing operation fails.
435 MH/01 Treat the transport option dkim_domain as a colon separated
436 list, not as a single string, and sign the message with each element,
437 omitting multiple occurences of the same signer.
439 NM/07 Null terminate DKIM strings, Null initialise DKIM variable
440 Bugzilla 985, 986. Patch by Simon Arlott
442 NM/08 Bugzilla 967. dnsdb DNS TXT record bug fix (DKIM-related)
443 Patch by Simon Arlott
445 PP/01 Bugzilla 989: CVE-2010-2024 - work round race condition on
446 MBX locking. Notification from Dan Rosenberg.
452 TK/01 Bugzilla 912: Fix DKIM segfault on empty headers/body.
454 NM/01 Bugzilla 913: Documentation fix for gnutls_* options.
456 NM/02 Bugzilla 722: Documentation for randint. Better randomness defaults.
458 NM/03 Bugzilla 847: Enable DNSDB lookup by default.
460 NM/04 Bugzilla 915: Flag broken perl installation during build.
466 TK/01 Added patch by Johannes Berg that expands the main option
467 "spamd_address" if it starts with a dollar sign.
469 TK/02 Write list of recipients to X-Envelope-Sender header when building
470 the mbox-format spool file for content scanning (suggested by Jakob
473 TK/03 Added patch by Wolfgang Breyha that adds experimental DCC
474 (http://www.dcc-servers.net/) support via dccifd. Activated by
475 setting EXPERIMENTAL_DCC=yes in Local/Makefile.
477 TK/04 Bugzilla 673: Add f-protd malware scanner support. Patch submitted
478 by Mark Daniel Reidel <mr@df.eu>.
480 NM/01 Bugzilla 657: Embedded PCRE removed from the exim source tree.
481 When building exim an external PCRE library is now needed -
482 PCRE is a system library on the majority of modern systems.
483 See entry on PCRE_LIBS in EDITME file.
485 NM/02 Bugzilla 646: Removed unwanted C/R in Dovecot authenticator
486 conversation. Added nologin parameter to request.
487 Patch contributed by Kirill Miazine.
489 TF/01 Do not log submission mode rewrites if they do not change the address.
491 TF/02 Bugzilla 662: Fix stack corruption before exec() in daemon.c.
493 NM/03 Bugzilla 602: exicyclog now handles panic log, and creates empty
494 log files in place. Contributed by Roberto Lima.
496 NM/04 Bugzilla 667: Close socket used by dovecot authenticator.
498 TF/03 Bugzilla 615: When checking the local_parts router precondition
499 after a local_part_suffix or local_part_prefix option, Exim now
500 does not use the address's named list lookup cache, since this
501 contains cached lookups for the whole local part.
503 NM/05 Bugzilla 521: Integrated SPF Best Guess support contributed by
504 Robert Millan. Documentation is in experimental-spec.txt.
506 TF/04 Bugzilla 668: Fix parallel build (make -j).
508 NM/05.2 Bugzilla 437: Prevent Maildir aux files being created with mode 000.
510 NM/05.3 Bugzilla 598: Improvement to Dovecot authenticator handling.
511 Patch provided by Jan Srzednicki.
513 TF/05 Leading white space used to be stripped from $spam_report which
514 wrecked the formatting. Now it is preserved.
516 TF/06 Save $spam_score, $spam_bar, and $spam_report in spool files, so
517 that they are available at delivery time.
519 TF/07 Fix the way ${extract is skipped in the untaken branch of a conditional.
521 TF/08 TLS error reporting now respects the incoming_interface and
522 incoming_port log selectors.
524 TF/09 Produce a more useful error message if an SMTP transport's hosts
525 setting expands to an empty string.
527 NM/06 Bugzilla 744: EXPN did not work under TLS.
528 Patch provided by Phil Pennock.
530 NM/07 Bugzilla 769: Extraneous comma in usage fprintf
531 Patch provided by Richard Godbee.
533 NM/08 Fixed erroneous documentation references to smtp_notquit_acl to be
534 acl_smtp_notquit, added index entry.
536 NM/09 Bugzilla 787: Potential buffer overflow in string_format.
537 Patch provided by Eugene Bujak.
539 NM/10 Bugzilla 770: Problem on some platforms modifying the len parameter to
540 accept(). Patch provided by Maxim Dounin.
542 NM/11 Bugzilla 749: Preserve old behaviour of blanks comparing equal to zero.
543 Patch provided by Phil Pennock.
545 NM/12 Bugzilla 497: Correct behaviour of exiwhat when no config exists.
547 NM/13 Bugzilla 590: Correct handling of Resent-Date headers.
548 Patch provided by Brad "anomie" Jorsch.
550 NM/14 Bugzilla 622: Added timeout setting to transport filter.
551 Patch provided by Dean Brooks.
553 TK/05 Add native DKIM support (does not depend on external libraries).
555 NM/15 Bugzilla 854: Removed code that symlinks to pcre as its no longer useful.
556 Patch provided by Graeme Fowler.
558 NM/16 Bugzilla 851: Documentation example syntax fix.
560 NM/17 Changed NOTICE file to remove references to embedded PCRE.
562 NM/18 Bugzilla 894: Fix issue with very long lines including comments in
565 NM/19 Bugzilla 745: TLS version reporting.
566 Patch provided by Phil Pennock.
568 NM/20 Bugzilla 167: bool: condition support.
569 Patch provided by Phil Pennock.
571 NM/21 Bugzilla 665: gnutls_compat_mode to allow compatibility with broken
572 clients. Patch provided by Phil Pennock.
574 NM/22 Bugzilla 607: prepend (not append) Resent-Message-ID and Resent-Date.
575 Patch provided by Brad "anomie" Jorsch.
577 NM/23 Bugzilla 687: Fix misparses in eximstats.
578 Patch provided by Heiko Schlittermann.
580 NM/24 Bugzilla 688: Fix exiwhat to handle log_selector = +pid.
581 Patch provided by Heiko Schlittermann.
583 NM/25 Bugzilla 727: Use transport mode as default mode for maildirsize file.
584 plus update to original patch.
586 NM/26 Bugzilla 799: Documentation correction for ratelimit.
588 NM/27 Bugzilla 802: Improvements to local interface IP addr detection.
589 Patch provided by David Brownlee.
591 NM/28 Bugzilla 807: Improvements to LMTP delivery logging.
593 NM/29 Bugzilla 862, 866, 875: Documentation bugfixes.
595 NM/30 Bugzilla 888: TLS documentation bugfixes.
597 NM/31 Bugzilla 896: Dovecot buffer overrun fix.
599 NM/32 Bugzilla 889: Change all instances of "expr" in shell scripts to "expr --"
600 Unlike the original bugzilla I have changed all shell scripts in src tree.
602 NM/33 Bugzilla 898: Transport filter timeout fix.
603 Patch by Todd Rinaldo.
605 NM/34 Bugzilla 901: Fix sign/unsigned and UTF mistmatches.
606 Patch by Serge Demonchaux.
608 NM/35 Bugzilla 39: Base64 decode bug fixes.
609 Patch by Jakob Hirsch.
611 NM/36 Bugzilla 909: Correct connect() call in dcc code.
613 NM/37 Bugzilla 910: Correct issue with relaxed/simple handling.
615 NM/38 Bugzilla 908: Removed NetBSD3 support as no longer needed.
617 NM/39 Bugzilla 911: Fixed MakeLinks build script.
623 TK/01 Add preliminary DKIM support. Currently requires a forked version of
624 ALT-N's libdkim that I have put here:
625 http://duncanthrax.net/exim-experimental/
627 Note to Michael Haardt: I had to rename some vars in sieve.c. They
628 were called 'true' and it seems that C99 defines that as a reserved
629 keyword to be used with 'bool' variable types. That means you could
630 not include C99-style headers which use bools without triggering
631 build errors in sieve.c.
633 NM/01 Bugzilla 592: --help option is handled incorrectly if exim is invoked
634 as mailq or other aliases. Changed the --help handling significantly
635 to do whats expected. exim_usage() emits usage/help information.
637 SC/01 Added the -bylocaldomain option to eximstats.
639 NM/02 Bugzilla 619: Defended against bad data coming back from gethostbyaddr.
641 NM/03 Bugzilla 613: Documentation fix for acl_not_smtp.
643 NM/04 Bugzilla 628: PCRE update to 7.4 (work done by John Hall).
649 PH/01 Another patch from the Sieve maintainer.
651 PH/02 When an IPv6 address is converted to a string for single-key lookup
652 in an address list (e.g. for an item such as "net24-dbm;/net/works"),
653 dots are used instead of colons so that keys in lsearch files need not
654 contain colons. This was done some time before quoting was made available
655 in lsearch files. However, iplsearch files do require colons in IPv6 keys
656 (notated using the quote facility) so as to distinguish them from IPv4
657 keys. This meant that lookups for IP addresses in host lists did not work
658 for iplsearch lookups.
660 This has been fixed by arranging for IPv6 addresses to be expressed with
661 colons if the lookup type is iplsearch. This is not incompatible, because
662 previously such lookups could never work.
664 The situation is now rather anomolous, since one *can* have colons in
665 ordinary lsearch keys. However, making the change in all cases is
666 incompatible and would probably break a number of configurations.
668 TK/01 Change PRVS address formatting scheme to reflect latests BATV draft
671 MH/01 The "spam" ACL condition code contained a sscanf() call with a %s
672 conversion specification without a maximum field width, thereby enabling
673 a rogue spamd server to cause a buffer overflow. While nobody in their
674 right mind would setup Exim to query an untrusted spamd server, an
675 attacker that gains access to a server running spamd could potentially
676 exploit this vulnerability to run arbitrary code as the Exim user.
678 TK/02 Bugzilla 502: Apply patch to make the SPF-Received: header use
679 $primary_hostname instead of what libspf2 thinks the hosts name is.
681 MH/02 The dsearch lookup now uses lstat(2) instead of stat(2) to look for
682 a directory entry by the name of the lookup key. Previously, if a
683 symlink pointed to a non-existing file or a file in a directory that
684 Exim lacked permissions to read, a lookup for a key matching that
685 symlink would fail. Now it is enough that a matching directory entry
686 exists, symlink or not. (Bugzilla 503.)
688 PH/03 The body_linecount and body_zerocount variables are now exported in the
691 PH/04 Added the $dnslist_matched variable.
693 PH/05 Unset $tls_cipher and $tls_peerdn before making a connection as a client.
694 This means they are set thereafter only if the connection becomes
697 PH/06 Added the client_condition to authenticators so that some can be skipped
698 by clients under certain conditions.
700 PH/07 The error message for a badly-placed control=no_multiline_responses left
701 "_responses" off the end of the name.
703 PH/08 Added -Mvc to output a copy of a message in RFC 2822 format.
705 PH/09 Tidied the code for creating ratelimiting keys, creating them explicitly
706 (without spaces) instead of just copying the configuration text.
708 PH/10 Added the /noupdate option to the ratelimit ACL condition.
710 PH/11 Added $max_received_linelength.
712 PH/12 Added +ignore_defer and +include_defer to host lists.
714 PH/13 Installed PCRE version 7.2. This needed some changes because of the new
715 way in which PCRE > 7.0 is built.
717 PH/14 Implemented queue_only_load_latch.
719 PH/15 Removed an incorrect (int) cast when reading the value of SIZE in a
720 MAIL command. The effect was to mangle the value on 64-bit systems.
722 PH/16 Another patch from the Sieve maintainer.
724 PH/17 Added the NOTQUIT ACL, based on a patch from Ted Cooper.
726 PH/18 If a system quota error occurred while trying to create the file for
727 a maildir delivery, the message "Mailbox is full" was not appended to the
728 bounce if the delivery eventually timed out. Change 4.67/27 below applied
729 only to a quota excession during the actual writing of the file.
731 PH/19 It seems that peer DN values may contain newlines (and other non-printing
732 characters?) which causes problems in log lines. The DN values are now
733 passed through string_printing() before being added to log lines.
735 PH/20 Added the "servers=" facility to MySQL and PostgreSQL lookups. (Oracle
736 and InterBase are left for another time.)
738 PH/21 Added message_body_newlines option.
740 PH/22 Guard against possible overflow in moan_check_errorcopy().
742 PH/23 POSIX allows open() to be a macro; guard against that.
744 PH/24 If the recipient of an error message contained an @ in the local part
745 (suitably quoted, of course), incorrect values were put in $domain and
746 $local_part during the evaluation of errors_copy.
752 MH/01 Fix for bug #448, segfault in Dovecot authenticator when interface_address
753 is unset (happens when testing with -bh and -oMi isn't used). Thanks to
756 PH/01 Added a new log selector smtp_no_mail, to log SMTP sessions that do not
757 issue a MAIL command.
759 PH/02 In an ACL statement such as
761 deny dnslists = X!=127.0.0.2 : X=127.0.0.2
763 if a client was not listed at all, or was listed with a value other than
764 127.0.0.2, in the X list, but was listed with 127.0.0.2 in the Y list,
765 the condition was not true (as it should be), so access was not denied.
766 The bug was that the ! inversion was incorrectly passed on to the second
767 item. This has been fixed.
769 PH/03 Added additional dnslists conditions == and =& which are different from
770 = and & when the dns lookup returns more than one IP address.
772 PH/04 Added gnutls_require_{kx,mac,protocols} to give more control over the
773 cipher suites used by GnuTLS. These options are ignored by OpenSSL.
775 PH/05 After discussion on the list, added a compile time option ENABLE_DISABLE_
776 FSYNC, which compiles an option called disable_fsync that allows for
777 bypassing fsync(). The documentation is heavily laced with warnings.
779 SC/01 Updated eximstats to collate all SpamAssassin rejects into one bucket.
781 PH/06 Some tidies to the infrastructure of the Test Suite that is concerned
782 with the auxiliary C programs that it uses: (1) Arrange for BIND_8_COMPAT
783 to be defined when compiling on OSX (Darwin); (2) Tidies to the Makefile,
784 including adding "make clean"; (3) Added -fPIC when compiling the test
785 dynamically loaded module, to get rid of a warning.
787 MH/02 Fix for bug #451, causing paniclog entries to be written if a bounce
788 message fails, move_frozen_messages = true and ignore_bounce_errors_after
789 = 0s. The bug is otherwise harmless.
791 PH/07 There was a bug in the dovecot authenticator such that the value of
792 $auth1 could be overwritten, and so not correctly preserved, after a
793 successful authentication. This usually meant that the value preserved by
794 the server_setid option was incorrect.
796 PH/08 Added $smtp_count_at_connection_start, deliberately with a long name.
798 PH/09 Installed PCRE release 7.0.
800 PH/10 The acl_not_smtp_start ACL was, contrary to the documentation, not being
801 run for batched SMTP input. It is now run at the start of every message
802 in the batch. While fixing this I discovered that the process information
803 (output by running exiwhat) was not always getting set for -bs and -bS
804 input. This is fixed, and it now also says "batched" for BSMTP.
806 PH/11 Added control=no_pipelining.
808 PH/12 Added $sending_ip_address and $sending_port (mostly Magnus Holmgren's
809 patch, slightly modified), and move the expansion of helo_data till after
810 the connection is made in the smtp transport (so it can use these
813 PH/13 Added ${rfc2047d: to decoded RFC 2047 strings.
815 PH/14 Added log_selector = +pid.
817 PH/15 Flush SMTP output before delaying, unless control=no_delay_flush is set.
819 PH/16 Add ${if forany and ${if forall.
821 PH/17 Added dsn_from option to vary the From: line in DSNs.
823 PH/18 Flush SMTP output before performing a callout, unless control =
824 no_callout_flush is set.
826 PH/19 Change 4.64/PH/36 introduced a bug: when address_retry_include_sender
827 was true (the default) a successful delivery failed to delete the retry
828 item, thus causing premature timeout of the address. The bug is now
831 PH/20 Added hosts_avoid_pipelining to the smtp transport.
833 PH/21 Long custom messages for fakedefer and fakereject are now split up
834 into multiline reponses in the same way that messages for "deny" and
835 other ACL rejections are.
837 PH/22 Applied Jori Hamalainen's speed-up changes and typo fixes to exigrep,
838 with slight modification.
840 PH/23 Applied sieve patches from the maintainer "tracking the latest notify
841 draft, changing the syntax and factoring some duplicate code".
843 PH/24 When the log selector "outgoing_port" was set, the port was shown as -1
844 for deliveries of the second and subsequent messages over the same SMTP
847 PH/25 Applied Magnus Holmgren's patch for ${addresses, ${map, ${filter, and
848 ${reduce, with only minor "tidies".
850 SC/02 Applied Daniel Tiefnig's patch to improve the '($parent) =' pattern match.
852 PH/26 Added a "continue" ACL modifier that does nothing, for the benefit of its
853 expansion side effects.
855 PH/27 When a message times out after an over-quota error from an Exim-imposed
856 quota, the bounce message says "mailbox is full". This message was not
857 being given when it was a system quota that was exceeded. It now should
860 MH/03 Made $recipients available in local_scan(). local_scan() already has
861 better access to the recipient list through recipients_list[], but
862 $recipients can be useful in postmaster-provided expansion strings.
864 PH/28 The $smtp_command and $smtp_command_argument variables were not correct
865 in the case of a MAIL command with additional options following the
866 address, for example: MAIL FROM:<foo@bar> SIZE=1234. The option settings
867 were accidentally chopped off.
869 PH/29 SMTP synchronization checks are implemented when a command is read -
870 there is a check that no more input is waiting when there shouldn't be
871 any. However, for some commands, a delay in an ACL can mean that it is
872 some time before the response is written. In this time, more input might
873 arrive, invalidly. So now there are extra checks after an ACL has run for
874 HELO/EHLO and after the predata ACL, and likewise for MAIL and RCPT when
875 pipelining has not been advertised.
877 PH/30 MH's patch to allow iscntrl() characters to be list separators.
879 PH/31 Unlike :fail:, a custom message specified with :defer: was not being
880 returned in the SMTP response when smtp_return_error_details was false.
883 PH/32 Change the Dovecot authenticator to use read() and write() on the socket
884 instead of the C I/O that was originally supplied, because problems were
887 PH/33 Compile failed with OpenSSL 0.9.8e. This was due to a coding error in
888 Exim which did not show up earlier: it was assuming that a call to
889 SSL_CTX_set_info_callback() might give an error value. In fact, there is
890 no error. In previous releases of OpenSSL, SSL_CTX_set_info_callback()
891 was a macro that became an assignment, so it seemed to work. This has
892 changed to a proper function call with a void return, hence the compile
893 error. Exim's code has been fixed.
895 PH/34 Change HDA_SIZE in oracle.c from 256 to 512. This is needed for 64-bit
898 PH/35 Applied a patch from the Sieve maintainer which fixes a bug in "notify".
900 PH/36 Applied John Jetmore's patch to add -v functionality to exigrep.
902 PH/37 If a message is not accepted after it has had an id assigned (e.g.
903 because it turns out to be too big or there is a timeout) there is no
904 "Completed" line in the log. When some messages of this type were
905 selected by exigrep, they were listed as "not completed". Others were
906 picked up by some special patterns. I have improved the selection
907 criteria to be more general.
909 PH/38 The host_find_failed option in the manualroute router can now be set
910 to "ignore", to completely ignore a host whose IP address cannot be
911 found. If all hosts are ignored, the behaviour is controlled by the new
912 host_all_ignored option.
914 PH/39 In a list of hosts for manualroute, if one item (either because of multi-
915 homing or because of multiple MX records with /mx) generated more than
916 one IP address, and the following item turned out to be the local host,
917 all the secondary addresses of the first item were incorrectly removed
918 from the list, along with the local host and any following hosts (which
919 is what is supposed to happen).
921 PH/40 When Exim receives a message, it writes the login name, uid, and gid of
922 whoever called Exim into the -H file. In the case of the daemon it was
923 behaving confusingly. When first started, it used values for whoever
924 started the daemon, but after a SIGHUP it used the Exim user (because it
925 calls itself on a restart). I have changed the code so that it now always
928 PH/41 (Following a suggestion from Tony Finch) If all the RCPT commands in a
929 message are rejected with the same error (e.g. no authentication or bad
930 sender address), and a DATA command is nevertheless sent (as can happen
931 with PIPELINING or a stupid MUA), the error message that was given to the
932 RCPT commands is included in the rejection of the DATA command. This is
933 intended to be helpful for MUAs that show only the final error to their
936 PH/42 Another patch from the Sieve maintainer.
938 SC/02 Eximstats - Differentiate between permanent and temporary rejects.
939 Eximstats - Fixed some broken HTML links and added missing column headers
941 Eximstats - Fixed Grand Total Summary Domains, Edomains, and Email
942 columns for Rejects, Temp Rejects, Ham, and Spam rows.
944 SC/03 Eximstats - V1.58 Fix to get <> and blackhole to show in edomain tables.
946 PH/43 Yet another patch from the Sieve maintainer.
948 PH/44 I found a way to check for a TCP/IP connection going away before sending
949 the response to the final '.' that terminates a message, but only in the
950 case where the client has not sent further data following the '.'
951 (unfortunately, this is allowed). However, in many cases there won't be
952 any further data because there won't be any more messages to send. A call
953 to select() can be used: if it shows that the input is "ready", there is
954 either input waiting, or the socket has been closed. An attempt to read
955 the next input character can distinguish the two cases. Previously, Exim
956 would have sent an OK response which the client would never have see.
957 This could lead to message repetition. This fix should cure that, at
958 least in a lot of common cases.
960 PH/45 Do not advertise STARTTLS in response to HELP unless it would be
961 advertised in response to EHLO.
967 PH/01 Two more bugs that were introduced by 4.64/PH/07, in addition to the one
968 fixed by 4.65/MH/01 (is this a record?) are fixed:
970 (i) An empty string was always treated as zero by the numeric comparison
971 operators. This behaviour has been restored.
973 (ii) It is documented that the numeric comparison operators always treat
974 their arguments as decimal numbers. This was broken in that numbers
975 starting with 0 were being interpreted as octal.
977 While fixing these problems I realized that there was another issue that
978 hadn't been noticed. Values of message_size_limit (both the global option
979 and the transport option) were treated as octal if they started with 0.
980 The documentation was vague. These values are now always treated as
981 decimal, and I will make that clear in the documentation.
987 TK/01 Disable default definition of HAVE_LINUX_SENDFILE. Clashes with
988 Linux large file support (_FILE_OFFSET_BITS=64) on older glibc
991 MH/01 Don't check that the operands of numeric comparison operators are
992 integers when their expansion is in "skipping" mode (fixes bug
993 introduced by 4.64-PH/07).
995 PH/01 If a system filter or a router generates more than SHRT_MAX (32767)
996 child addresses, Exim now panics and dies. Previously, because the count
997 is held in a short int, deliveries were likely to be lost. As such a
998 large number of recipients for a single message is ridiculous
999 (performance will be very, very poor), I have chosen to impose a limit
1000 rather than extend the field.
1006 TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
1007 leftover -K file (the existence of which was triggered by #402).
1008 While we were at it, introduced process PID as part of the -K
1009 filename. This should rule out race conditions when creating
1012 TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
1013 processing considerably. Previous code took too long for large mails,
1014 triggering a timeout which in turn triggers #401.
1016 TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
1017 in the DK code in transports.c. sendfile() is not really portable,
1018 hence the _LINUX specificness.
1020 TF/01 In the add_headers option to the mail command in an Exim filter,
1021 there was a bug that Exim would claim a syntax error in any
1022 header after the first one which had an odd number of characters
1025 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
1026 callout verification, Exim cached a "reject" for the entire domain. This
1027 is correct for most verifications, but it is not correct for a recipient
1028 verification with use_sender or use_postmaster set, because in that case
1029 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
1030 case of MAIL FROM:<> rejection from other early rejections (e.g.
1031 rejection of HELO). When verifying a recipient using a non-null MAIL
1032 address, the cache is ignored if it shows MAIL FROM:<> rejection.
1033 Whatever the result of the callout, the value of the domain cache is
1034 left unchanged (for any other kind of callout, getting as far as trying
1035 RCPT means that the domain itself is ok).
1037 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
1040 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
1041 manifest itself as EPIPE rather than ECONNECT. When tidying away a
1042 session, the daemon ignores ECONNECT errors and logs others; it now
1043 ignores EPIPE as well.
1045 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
1046 (quoted-printable decoding).
1048 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
1049 later the small subsequent patch to fix an introduced bug.
1051 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
1053 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
1055 PH/08 An error is now given if message_size_limit is specified negative.
1057 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
1058 to be given (somewhat) arbitrary names.
1060 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
1063 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
1064 miscellaneous code fixes
1066 PH/10 Added the log_reject_target ACL modifier to specify where to log
1069 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
1070 hostname. This is wrong, because it relates to the incoming message (and
1071 probably the interface on which it is arriving) and not to the outgoing
1072 callout (which could be using a different interface). This has been
1073 changed to use the value of the helo_data option from the smtp transport
1074 instead - this is what is used when a message is actually being sent. If
1075 there is no remote transport (possible with a router that sets up host
1076 addresses), $smtp_active_hostname is used.
1078 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
1079 tweaks were necessary in order to get it to work (see also 21 below):
1080 (a) The code assumed that strncpy() returns a negative number on buffer
1081 overflow, which isn't the case. Replaced with Exim's string_format()
1083 (b) There were several signed/unsigned issues. I just did the minimum
1084 hacking in of casts. There is scope for a larger refactoring.
1085 (c) The code used strcasecmp() which is not a standard C function.
1086 Replaced with Exim's strcmpic() function.
1087 (d) The code set only $1; it now sets $auth1 as well.
1088 (e) A simple test gave the error "authentication client didn't specify
1089 service in request". It would seem that Dovecot has changed its
1090 interface. Fortunately there's a specification; I followed it and
1091 changed what the client sends and it appears to be working now.
1093 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
1096 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
1097 address A is aliased to B and C, where B exists and C does not. Without
1098 -v the output is "A verified" because verification stops after a
1099 successful redirection if more than one address is generated. However,
1100 with -v the child addresses are also verified. Exim was outputting "A
1101 failed to verify" and then showing the successful verification for C,
1102 with its parentage. It now outputs "B failed to verify", showing B's
1103 parentage before showing the successful verification of C.
1105 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
1106 look up a TXT record in a specific list after matching in a combined
1109 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
1110 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
1111 they consult the DNS. I had assumed they would set it the way they
1112 wanted; and indeed my experiments on Linux seem to show that in some
1113 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
1114 To be on the safe side, however, I have now made the interface to
1115 host_find_byname() similar to host_find_bydns(), with an argument
1116 containing the DNS resolver options. The host_find_byname() function now
1117 sets these options at its start, just as host_find_bydns() does. The smtp
1118 transport options dns_qualify_single and dns_search_parents are passed to
1119 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
1120 of host_find_byname() use the default settings of RES_DEFNAMES
1121 (qualify_single) but not RES_DNSRCH (search_parents).
1123 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
1124 spool_read_header() do less string testing, by means of a preliminary
1125 switch on the second character of optional "-foo" lines. (This is
1126 overdue, caused by the large number of possibilities that now exist.
1127 Originally there were few.) While I was there, I also converted the
1128 str(n)cmp tests so they don't re-test the leading "-" and the first
1129 character, in the hope this might squeeze out yet more improvement.
1131 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
1132 flag allowing group syntax was set by the header_syntax check but not
1133 turned off, possible causing trouble later; (2) The flag was not being
1134 set at all for the header_verify test, causing "group"-style headers to
1135 be rejected. I have now set it in this case, and also caused header_
1136 verify to ignore an empty address taken from a group. While doing this, I
1137 came across some other cases where the code for allowing group syntax
1138 while scanning a header line wasn't quite right (mostly, not resetting
1139 the flag correctly in the right place). These bugs could have caused
1140 trouble for malformed header lines. I hope it is now all correct.
1142 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
1143 with the "reply" argument non-NULL. The code, however (which originally
1144 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
1145 but it didn't always do it. This confused somebody who was copying the
1146 code for some other use. I have removed all the tests.
1148 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
1149 feature that was used to support insecure browsers during the U.S. crypto
1150 embargo. It requires special client support, and Exim is probably the
1151 only MTA that supported it -- and would never use it because real RSA is
1152 always available. This code has been removed, because it had the bad
1153 effect of slowing Exim down by computing (never used) parameters for the
1154 RSA_EXPORT functionality.
1156 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
1157 authenticator to fail if there's a tab character in the incoming data
1158 (there should never be unless someone is messing about, as it's supposed
1159 to be base64-encoded). Also added, on Timo's advice, the "secured" option
1160 if the connection is using TLS or if the remote IP is the same as the
1161 local IP, and the "valid-client-cert option" if a client certificate has
1164 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
1165 authenticators. This can be used for authorization after authentication
1166 succeeds. (In the case of plaintext, it servers for both authentication
1169 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
1170 if any retry times were supplied.
1172 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
1173 connection, where there is no client IP address to check. In this
1174 situation, the verify now always succeeds.
1176 PH/25 Applied John Jetmore's -Mset patch.
1178 PH/26 Added -bem to be like -Mset, but loading a message from a file.
1180 PH/27 In a string expansion for a processed (not raw) header when multiple
1181 headers of the same name were present, leading whitespace was being
1182 removed from all of them, but trailing whitespace was being removed only
1183 from the last one. Now trailing whitespace is removed from each header
1184 before concatenation. Completely empty headers in a concatenation (as
1185 before) are ignored.
1187 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
1188 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
1190 PH/29 [Removed. This was a change that I later backed out, and forgot to
1191 correct the ChangeLog entry (that I had efficiently created) before
1192 committing the later change.]
1194 PH/30 Exim was sometimes attempting to deliver messages that had suffered
1195 address errors (4xx response to RCPT) over the same connection as other
1196 messages routed to the same hosts. Such deliveries are always "forced",
1197 so retry times are not inspected. This resulted in far too many retries
1198 for the affected addresses. The effect occurred only when there were more
1199 hosts than the hosts_max_try setting in the smtp transport when it had
1200 the 4xx errors. Those hosts that it had tried were not added to the list
1201 of hosts for which the message was waiting, so if all were tried, there
1202 was no problem. Two fixes have been applied:
1204 (i) If there are any address or message errors in an SMTP delivery, none
1205 of the hosts (tried or untried) are now added to the list of hosts
1206 for which the message is waiting, so the message should not be a
1207 candidate for sending over the same connection that was used for a
1208 successful delivery of some other message. This seems entirely
1209 reasonable: after all the message is NOT "waiting for some host".
1210 This is so "obvious" that I'm not sure why it wasn't done
1211 previously. Hope I haven't missed anything, but it can't do any
1212 harm, as the worst effect is to miss an optimization.
1214 (ii) If, despite (i), such a delivery is accidentally attempted, the
1215 routing retry time is respected, so at least it doesn't keep
1216 hammering the server.
1218 PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
1219 in ${readsocket because some servers need this prod.
1221 PH/32 Added some extra debug output when updating a wait-xxx database.
1223 PH/33 The hint "could be header name not terminated by colon", which has been
1224 given for certain expansion errors for a long time, was not being given
1225 for the ${if def:h_colon_omitted{... case.
1227 PH/34 The spec says: "With one important exception, whenever a domain list is
1228 being scanned, $domain contains the subject domain." There was at least
1229 one case where this was not true.
1231 PH/35 The error "getsockname() failed: connection reset by peer" was being
1232 written to the panic log as well as the main log, but it isn't really
1233 panic-worthy as it just means the connection died rather early on. I have
1234 removed the panic log writing for the ECONNRESET error when getsockname()
1237 PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
1238 runs only) independently of the message's sender address. This meant
1239 that, if the 4xx error was in fact related to the sender, a different
1240 message to the same recipient with a different sender could confuse
1241 things. In particualar, this can happen when sending to a greylisting
1242 server, but other circumstances could also provoke similar problems.
1243 I have changed the default so that the retry time for these errors is now
1244 based a combination of the sender and recipient addresses. This change
1245 can be overridden by setting address_retry_include_sender=false in the
1248 PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
1249 remote server are returned as part of bounce messages. This was not
1250 happening for LMTP over a pipe (the lmtp transport), but now it is the
1251 same for both kinds of LMTP.
1253 PH/38 Despite being documented as not happening, Exim was rewriting addresses
1254 in header lines that were in fact CNAMEs. This is no longer the case.
1256 PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
1257 and queue runs started by the daemon processed all messages. This has
1258 been fixed so that -R and -S can now usefully be given with -q<time>.
1260 PH/40 Import PCRE release 6.7 (fixes some bugs).
1262 PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
1264 PH/42 Give an error if -q is specified more than once.
1266 PH/43 Renamed the variables $interface_address and $interface_port as
1267 $received_ip_address and $received_port, to make it clear that these
1268 values apply to message reception, and not to the outgoing interface when
1269 a message is delivered. (The old names remain recognized, of course.)
1271 PH/44 There was no timeout on the connect() call when using a Unix domain
1272 socket in the ${readsocket expansion. There now is.
1274 PH/45 Applied a modified version of Brad Jorsch's patch to allow "message" to
1275 be meaningful with "accept".
1277 SC/01 Eximstats V1.43
1278 Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
1280 SC/02 Eximstats V1.44
1281 Use a glob alias rather than an array ref in the generated
1282 parser. This improves both readability and performance.
1284 SC/03 Eximstats V1.45 (Marco Gaiarin / Steve Campbell)
1285 Collect SpamAssassin and rejection statistics.
1286 Don't display local sender or destination tables unless
1287 there is data to show.
1288 Added average volumes into the top table text output.
1290 SC/04 Eximstats V1.46
1291 Collect data on the number of addresses (recipients)
1292 as well as the number of messages.
1294 SC/05 Eximstats V1.47
1295 Added 'Message too big' to the list of mail rejection
1296 reasons (thanks to Marco Gaiarin).
1298 SC/06 Eximstats V1.48
1299 Mainlog lines which have GMT offsets and are too short to
1300 have a flag are now skipped.
1302 SC/07 Eximstats V1.49 (Alain Williams)
1303 Added the -emptyok flag.
1305 SC/08 Eximstats V1.50
1306 Fixes for obtaining the IP address from reject messages.
1308 JJ/03 exipick.20061117.2, made header handling as similar to exim as possible
1309 (added [br]h_ prefixes, implemented RFC2047 decoding. Fixed
1310 whitesspace changes from 4.64-PH/27
1312 JJ/04 exipick.20061117.2, fixed format and added $message_headers_raw to
1315 JJ/05 exipick.20061117.2, bug fixes (error out sooner when invalid criteria
1316 are found, allow negative numbers in numeric criteria)
1318 JJ/06 exipick.20061117.2, added new $message_body_missing variable
1320 JJ/07 exipick.20061117.2, added $received_ip_address and $received_port
1321 to match changes made in 4.64-PH/43
1323 PH/46 Applied Jori Hamalainen's patch to add features to exiqsumm.
1325 PH/47 Put in an explicit test for a DNS lookup of an address record where the
1326 "domain" is actually an IP address, and force a failure. This locks out
1327 those revolvers/nameservers that support "A-for-A" lookups, in
1328 contravention of the specifications.
1330 PH/48 When a host name was looked up from an IP address, and the subsequent
1331 forward lookup of the name timed out, the host name was left in
1332 $sender_host_name, contrary to the specification.
1334 PH/49 Although default lookup types such as lsearch* or cdb*@ have always been
1335 restricted to single-key lookups, Exim was not diagnosing an error if
1336 * or *@ was used with a query-style lookup.
1338 PH/50 Increased the value of DH_BITS in tls-gnu.c from 768 to 1024.
1340 MH/01 local_scan ABI version incremented to 1.1. It should have been updated
1341 long ago, but noone interested enough thought of it. Let's just say that
1342 the "1.1" means that there are some new functions that weren't there at
1343 some point in the past.
1345 PH/51 Error processing for expansion failure of helo_data from an smtp
1346 transport during callout processing was broken.
1348 PH/52 Applied John Jetmore's patch to allow tls-on-connect and STARTTLS to be
1349 tested/used via the -bh/-bhc/-bs options.
1351 PH/53 Added missing "#include <time.h>" to pcre/pcretest.c (this was a PCRE
1352 bug, fixed in subsequent PCRE releases).
1354 PH/54 Applied Robert Bannocks' patch to avoid a problem with references that
1355 arises when using the Solaris LDAP libraries (but not with OpenLDAP).
1357 PH/55 Check for a ridiculously long file name in exim_dbmbuild.
1363 SC/01 Use a glob alias rather than an array ref in eximstats generated
1364 parser. This improves both readability and performance.
1366 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
1367 Don't display local sender or destination tables in eximstats unless
1368 there is data to show.
1369 Added average volumes into the eximstats top table text output.
1371 SC/03 Collect data on the number of addresses (recipients) as well
1372 as the number of messages in eximstats.
1374 TF/01 Correct an error in the documentation for the redirect router. Exim
1375 does (usually) call initgroups() when daemonizing.
1377 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
1378 with consistent privilege compared to when running as a daemon.
1380 TF/03 Note in the spec that $authenticated_id is not set for local
1381 submissions from trusted users.
1383 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
1384 Thanks to Dean Brooks <dean@iglou.com> for the patch.
1386 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
1387 by adding some example configuration directives to the default
1388 configuration file. A little bit of work is required to uncomment the
1389 directives and define how usernames and passwords are checked, but
1390 there is now a framework to start from.
1392 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
1393 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
1394 without this. I don't know how relevant this is to other LDAP libraries.
1396 PH/02 Add the verb name to the "unknown ACL verb" error.
1398 PH/03 Magnus Holmgren's patch for filter_prepend_home.
1400 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
1402 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
1403 directory not expanded when it should be if an expanded home directory
1404 was set for the address (which is overridden by the transport).
1406 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
1409 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
1410 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
1411 because it is too late at that time, and has no effect.
1413 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
1414 security issue with \' (bugzilla #107). I could not use the
1415 PQescapeStringConn() function, because it needs a PGconn value as one of
1418 PH/08 When testing addresses using -bt, indicate those final addresses that
1419 are duplicates that would not cause an additional delivery. At least one
1420 person was confused, thinking that -bt output corresponded to deliveries.
1421 (Suppressing duplicates isn't a good idea as you lose the information
1422 about possibly different redirections that led to the duplicates.)
1424 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
1425 systems where poll() doesn't work, in particular OS X.
1427 PH/10 Added more information to debugging output for retry time not reached.
1429 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
1430 operations in malware.c.
1432 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
1435 PH/13 If write_rejectlog was set false when logging was sent to syslog with
1436 syslog_duplication set false, log lines that would normally be written
1437 both the the main log and to the reject log were not written to syslog at
1440 PH/14 In the default configuration, change the use of "message" in ACL warn
1441 statements to "add_header".
1443 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
1444 not followed by a command (e.g. "seen endif").
1446 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
1447 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
1450 PH/17 Added extra conditions to the default value of delay_warning_condition
1454 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
1455 { match{$h_precedence:}{(?i)bulk|list|junk} } \
1456 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
1459 The Auto-Submitted: and various List- headers are standardised, whereas I
1460 don't think Precedence: ever was.
1462 PH/18 Refactored debugging code in route_finduser() to show more information,
1463 in particular, the error code if getpwnam() issues one.
1465 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
1466 This is apparently needed in addition to the PH/07 change above to avoid
1467 any possible encoding problems.
1469 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
1470 but not after initializing Perl.
1472 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
1473 output them only if debugging. By default they are written stderr,
1474 apparently, which is not desirable.
1476 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
1479 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
1482 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
1484 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
1485 authenticated or an ident call has been made. Suppress the default
1486 values for $authenticated_id and $authenticated_sender (but permit -oMai
1487 and -oMas) when testing with -bh.
1489 PH/24 Re-jigged the order of the tests in the default configuration so that the
1490 tests for valid domains and recipients precede the DNS black list and CSA
1491 tests, on the grounds that those ones are more expensive.
1493 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
1494 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
1495 This bug exists in every version of Exim that I still have, right back to
1498 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
1499 However, an attempt to turn on case-sensitivity in a regex key by
1500 including (?-i) didn't work because the subject string was already
1501 lowercased, and the effects were non-intuitive. It turns out that a
1502 one-line patch can be used to allow (?-i) to work as expected.
1508 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
1509 other effects) broke the use of negated acl sub-conditions.
1511 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
1514 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
1515 "Deny" causes Exim to reject the incoming connection with a 554 error.
1516 Unfortunately, if there is a major crisis, such as a disk failure,
1517 tcp-wrappers gives "deny", whereas what one would like would be some
1518 kind of temporary error. A kludge has been added to help with this.
1519 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
1520 554 error is used if errno is still zero or contains ENOENT (which occurs
1521 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
1524 PH/03 Add -lutil to the default FreeBSD LIBS setting.
1526 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
1527 errors. Otherwise a message that provokes a temporary error (when other
1528 messages do not) can cause a whole host to time out.
1530 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
1531 addresses were routed directly to files or pipes from a redirect router.
1532 File deliveries just didn't batch; pipe deliveries might have suffered
1535 PH/06 A failure to get a lock for a hints database would erroneously always say
1536 "Failed to get write lock", even when it was really a read lock.
1538 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
1539 of 0600. This has been changed to use the value of the lockfile_mode
1540 option (which defaults to 0600).
1542 PH/08 Applied small patch from the Sieve maintainer.
1544 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
1545 folder from quota calculations, a direct delivery into this folder messed
1546 up the contents of the maildirsize file. This was because the regex was
1547 used only to exclude .Trash (or whatever) when the size of the mailbox
1548 was calculated. There was no check that a delivery was happening into an
1549 excluded directory. This bug has been fixed by ignoring all quota
1550 processing for deliveries into excluded directories.
1552 PH/10 Added the maildirfolder_create_regex option to appendfile.
1558 PH/01 The code for finding all the local interface addresses on a FreeBSD
1559 system running IPv6 was broken. This may well have applied to all BSD
1560 systems, as well as to others that have similar system calls. The broken
1561 code found IPv4 interfaces correctly, but gave incorrect values for the
1562 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
1563 that it would not match correctly against @[] and not recognize the IPv6
1566 PH/02 The ipliteral router was not recognizing addresses of the form user@
1567 [ipv6:....] because it didn't know about the "ipv6:" prefix.
1569 PH/03 Added disable_ipv6.
1571 PH/04 Changed $reply_address to use the raw form of the headers instead of the
1572 decoded form, because it is most often used to construct To: headers
1573 lines in autoreplies, and the decoded form may well be syntactically
1574 invalid. However, $reply_address has leading white space removed, and all
1575 newlines turned into spaces so that the autoreply transport does not
1578 PH/05 If group was specified without a user on a router, and no group or user
1579 was specified on a transport, the group from the router was ignored.
1581 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
1582 for visible compile-time settings that can be used to change these
1583 numbers, for those that want even more. Backwards compatibility with old
1584 spool files has been maintained. However, going back to a previous Exim
1585 release will lost any variables that are in spool files.
1587 PH/07 Two small changes when running in the test harness: increase delay when
1588 passing a TCP/IP connection to a new process, in case the original
1589 process has to generate a bounce, and remove special handling of
1590 127.0.0.2 (sic), which is no longer necessary.
1592 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
1593 be the same on different OS.
1595 PH/09 Moved a debug statement in filter processing to avoid a race problem when
1598 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
1599 whether --show-vars was specified or not
1601 JJ/02 exipick: Added support for new ACL variable spool format introduced
1604 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
1605 syntactically invalid From: or Reply-to: line, and a filter used this to
1606 generate an autoreply, and therefore failed to obtain an address for the
1607 autoreply, Exim could try to deliver to a non-existent relative file
1608 name, causing unrelated and misleading errors. What now happens is that
1609 it logs this as a hard delivery error, but does not attempt to create a
1612 PH/11 The exinext utility has a -C option for testing purposes, but although
1613 the given file was scanned by exinext itself; it wasn't being passed on
1614 when Exim was called.
1616 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
1617 an end-of-file indication when reading a command response.
1619 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
1620 compiled. In many other places in Exim, IPv6 addresses are always
1621 recognized, so I have changed this. It also means that IPv4 domain
1622 literals of the form [IPV4:n.n.n.n] are now always recognized.
1624 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
1625 used if the router is not running as root, for example, when verifying at
1626 ACL time, or when using -bh. The debugging output from this situation was
1627 non-existent - all you got was a failure to exec. I have made two
1630 (a) Failures to set uid/gid, the current directory, or a process leader
1631 in a subprocess such as that created by queryprogram now generate
1632 suitable debugging ouput when -d is set.
1634 (b) The queryprogram router detects when it is not running as root,
1635 outputs suitable debugging information if -d is set, and then runs
1636 the subprocess without attempting to change uid/gid.
1638 PH/15 Minor change to Makefile for building test_host (undocumented testing
1641 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
1642 additional section of a DNS packet that returns MX or SRV records.
1643 Instead, it always explicitly searches for A/AAAA records. This avoids
1644 major problems that occur when a DNS server includes only records of one
1645 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
1646 fixed another bug: if SRV records were looked up and the corresponding
1647 address records were *not* found in the additional section, the port
1648 values from the SRV records were lost.
1650 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
1651 using the correct key (the original address) when searching the retry
1652 rules in order to find which one to use for generating the retry hint.
1654 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
1655 adding the default one. Similarly, if it contains a Reply-To: header, the
1656 errors_reply_to option, if set, is not used.
1658 PH/19 When calculating a retry time, Exim used to measure the "time since
1659 failure" by looking at the "first failed" field in the retry record. Now
1660 it does not use this if it is later than than the arrival time of the
1661 message. Instead it uses the arrival time. This makes for better
1662 behaviour in cases where some deliveries succeed, thus re-setting the
1663 "first failed" field. An example is a quota failure for a huge message
1664 when small messages continue to be delivered. Without this change, the
1665 "time since failure" will always be short, possible causing more frequent
1666 delivery attempts for the huge message than are intended.
1667 [Note: This change was subsequently modified - see PH/04 for 4.62.]
1669 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
1670 $1, $2, $3) because the numerical variables can be reset during some
1671 expansion items (e.g. "match"), thereby losing the authentication data.
1673 PH/21 Make -bV show the size of off_t variables so that the test suite can
1674 decide whether to run tests for quotas > 2G.
1676 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
1677 mailbox_size, and mailbox_filecount in the appendfile transport. If a
1678 filecount value is greater than 2G or if a quota value is greater than 2G
1679 on a system where the size of off_t is not greater than 4, a panic error
1682 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
1683 never match. The debug and -bh output now contains an explicit error
1684 message indicating a malformed IPv4 address or mask.
1686 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
1687 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
1688 PH/23 above applies.
1690 PH/25 Do not write to syslog when running in the test harness. The only
1691 occasion when this arises is a failure to open the main or panic logs
1692 (for which there is an explicit test).
1694 PH/26 Added the /no_tell option to "control=freeze".
1696 PH/27 If a host name lookup failed very early in a connection, for example, if
1697 the IP address matched host_lookup and the reverse lookup yielded a name
1698 that did not have a forward lookup, an error message of the form "no IP
1699 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
1700 could be logged. Now it outputs the IP address instead of "NULL".
1702 PH/28 An enabling patch from MH: add new function child_open_exim2() which
1703 allows the sender and the authenticated sender to be set when
1704 submitting a message from within Exim. Since child_open_exim() is
1705 documented for local_scan(), the new function should be too.
1707 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
1708 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
1709 results in an empty string is now treated as unset.
1711 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
1713 PH/31 Added sender_verify_fail logging option.
1715 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
1716 needed by RFC 822 but not by RFC 2822 was commented out. I have now
1717 tidied the source and removed it altogether.
1719 PH/33 When a queue run was abandoned because the load average was too high, a
1720 log line was always written; now it is written only if the queue_run log
1721 selector is set. In addition, the log line for abandonment now contains
1722 information about the queue run such as the pid. This is always present
1723 in "start" and "stop" lines but was omitted from the "abandon" line.
1725 PH/34 Omit spaces between a header name and the colon in the error message that
1726 is given when verify = headers_syntax fails (if there are lots of them,
1727 the message gets confusing).
1729 PH/35 Change the default for dns_check_names_pattern to allow slashes within
1730 names, as there are now some PTR records that contain slashes. This check
1731 is only to protect against broken name servers that fall over on strange
1732 characters, so the fact that it applies to all lookups doesn't matter.
1734 PH/36 Now that the new test suite is complete, we can remove some of the
1735 special code in Exim that was needed for the old test suite. For example,
1736 sorting DNS records because real resolvers return them in an arbitrary
1737 order. The new test suite's fake resolver always returns records in the
1740 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
1741 bounces) except when queue_only is set, to avoid logging races between
1742 the different processes.
1744 PH/38 Panic-die if .include specifies a non-absolute path.
1746 PH/39 A tweak to the "H" retry rule from its user.
1748 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
1749 a label. They prevented compilation on older perls.
1751 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
1752 a warning to be raised on newish perls.
1754 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
1755 on queue. Changes to match documented behaviour of showing count of
1756 messages matching specified criteria.
1758 PH/40 Changed the default ident timeout from 30s to 5s.
1760 PH/41 Added support for the use of login_cap features, on those BSD systems
1761 that have them, for controlling the resources used by pipe deliveries.
1763 PH/42 The content-scanning code uses fopen() to create files in which to put
1764 message data. Previously it was not paying any attention to the mode of
1765 the files. Exim runs with umask(0) because the rest of the code creates
1766 files with open(), and sets the required mode explicitly. Thus, these
1767 files were ending up world-writeable. This was not a big issue, because,
1768 being within the spool directory, they were not world-accessible. I have
1769 created a function called modefopen, which takes an additional mode
1770 argument. It sets umask(777), creates the file, chmods it to the required
1771 mode, then resets the umask. All the relevant calls to fopen() in the
1772 content scanning code have been changed to use this function.
1774 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
1775 to 24 hours. This avoids potential overflow problems when processing G
1776 and H retry rules. I suspect nobody ever tinkers with this value.
1778 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
1780 PH/45 When the plaintext authenticator is running as a client, the server's
1781 challenges are checked to ensure they are valid base64 strings. By
1782 default, the authentication attempt is cancelled if an invalid string is
1783 received. Setting client_ignore_invalid_base64 true ignores these errors.
1784 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
1785 they are received. Thus, the responses can be made to depend on the
1786 challenges. If an invalid string is ignored, an empty string is placed in
1789 PH/46 Messages that are created by the autoreply transport now contains a
1790 References: header, in accordance with RFCs 2822 and 3834.
1792 PH/47 Added authenticated_sender_force to the smtp transport.
1794 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
1796 PH/49 Installed latest patch from the Sieve maintainer.
1798 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
1799 was also set, the appendfile transport was unnecessarily scanning a
1800 directory of message files (e.g. for maildir delivery) to find the count
1801 of files (along with the size), even though it did not need this
1802 information. It now does the scan only if it needs to find either the
1803 size of the count of files.
1805 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
1807 PH/52 Two bugs concerned with error handling when the smtp transport is
1810 (i) Exim was not creating retry information for temporary errors given
1811 for individual recipients after the DATA command when the smtp transport
1812 was used in LMTP mode. This meant that they could be retried too
1813 frequently, and not timed out correctly.
1815 (ii) Exim was setting the flag that allows error details to be returned
1816 for LMTP errors on RCPT commands, but not for LMTP errors for individual
1817 recipients that were returned after the DATA command.
1819 PH/53 This is related to PH/52, but is more general: for any failing address,
1820 when detailed error information was permitted to be returned to the
1821 sender, but the error was temporary, then after the final timeout, only
1822 "retry timeout exceeded" was returned. Now it returns the full error as
1823 well as "retry timeout exceeded".
1825 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
1826 do this, and (what is worse) MTAs that accept it.
1828 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
1829 will now be deprecated.
1831 PH/56 New os.c-cygwin from the Cygwin maintainer.
1833 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
1834 formats (previously only available in exim formats via -bpr, -bpru,
1835 and -bpra. Now also available in native and exiqgrep formats)
1837 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
1838 with very large, slow to parse queues
1840 JJ/08 exipick: added ! as generic prefix to negate any criteria format
1842 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
1844 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
1845 responses to authentication challenges, though it was showing the
1846 challenges; (ii) I've removed the CR characters from the debug output for
1849 PH/58 Allow for the insertion of a newline as well as a space when a string
1850 is turned into more than one encoded-word during RFC 2047 encoding. The
1851 Sieve code now uses this.
1853 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
1854 data_4xx, lost_connection, tls_required.
1856 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
1857 message was being sent as an SMTP response.
1859 PH/61 Add -l and -k options to exicyclog.
1861 PH/62 When verifying, if an address was redirected to one new address, so that
1862 verification continued, and the new address failed or deferred after
1863 having set something in $address_data, the value of $address_data was not
1864 passed back to the ACL. This was different to the case when no
1865 redirection occurred. The value is now passed back in both cases.
1867 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
1868 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
1869 use login_cap.h, so on its own it isn't the distinguishing feature. The
1870 new name refers directly to the setclassresources() function.
1872 PH/65 Added configuration files for NetBSD3.
1874 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
1876 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
1877 is preferred over IPv4.
1879 PH/68 The bounce_return_message and bounce_return_body options were not being
1880 honoured for bounces generated during the reception of non-SMTP messages.
1881 In particular, this applied to messages rejected by the ACL. This bug has
1882 been fixed. However, if bounce_return_message is true and bounce_return_
1883 body is false, the headers that are returned for a non-SMTP message
1884 include only those that have been read before the error was detected.
1885 (In the case of an ACL rejection, they have all been read.)
1887 PH/69 The HTML version of the specification is now built in a directory called
1888 spec_html instead of spec.html, because the latter looks like a path with
1889 a MIME-type, and this confuses some software.
1891 PH/70 Catch two compiler warnings in sieve.c.
1893 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
1894 function verify_get_ident() calls ip_connect() to connect a socket, but
1895 if the "connect()" function timed out, ip_connect() used to close the
1896 socket. However, verify_get_ident() also closes the socket later, and in
1897 between Exim writes to the log, which may get opened at this point. When
1898 the socket was closed in ip_connect(), the log could get the same file
1899 descriptor number as the socket. This naturally causes chaos. The fix is
1900 not to close the socket in ip_connect(); the socket should be closed by
1901 the function that creates it. There was only one place in the code where
1902 this was missing, in the iplookup router, which I don't think anybody now
1903 uses, but I've fixed it anyway.
1905 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
1906 well as to direct DNS lookups. Otherwise the handling of names in host
1907 lists is inconsistent and therefore confusing.
1913 PH/01 Two changes to the default runtime configuration:
1915 (1) Move the checks for relay_from_hosts and authenticated clients from
1916 after to before the (commented out) DNS black list checks.
1918 (2) Add control=submission to the relay_from_hosts and authenticated
1919 clients checks, on the grounds that messages accepted by these
1920 statements are most likely to be submissions.
1922 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
1924 (1) Generate an error if the third argument for the ${prvs expansion is
1927 (2) Treat a missing third argument of ${prvscheck as if it were an empty
1930 (3) Reset the variables that are obtained from the first argument of
1931 ${prvscheck and used in the second argument before leaving the code,
1932 because their memory is reclaimed, so using them afterwards may do
1935 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
1936 one (it's much easier than Tom thought :-).
1938 (5) Because of (4), we can now allow for the use of $prvscheck_result
1939 inside the third argument.
1941 PH/03 For some reason, the default setting of PATH when running a command from
1942 a pipe transport was just "/usr/bin". I have changed it to
1945 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
1946 anything to be listed in the output from -bV.
1948 PH/05 When a filter generated an autoreply, the entire To: header line was
1949 quoted in the delivery log line, like this:
1951 => >A.N.Other <ano@some.domain> <original@ddress> ...
1953 This has been changed so that it extracts the operative address. There
1954 may be more than one such address. If so, they are comma-separated, like
1957 => >ano@some.domain,ona@other.domain <original@ddress> ...
1959 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
1960 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
1961 not being looked up in the rDNS to get a host name, Exim was showing the
1962 IP address twice in Received: lines, even though the IP addresses were
1963 identical. For example:
1965 Received: from [1.2.3.4] (helo=[1.2.3.4])
1967 However, if the real host name was known, it was omitting the HELO data
1968 if it matched the actual IP address. This has been tidied up so that it
1969 doesn't show the same IP address twice.
1971 PH/07 When both +timestamp and +memory debugging was on, the value given by
1972 $tod_xxx expansions could be wrong, because the tod_stamp() function was
1973 called by the debug printing, thereby overwriting the timestamp buffer.
1974 Debugging no longer uses the tod_stamp() function when +timestamp is set.
1976 PH/08 When the original message was included in an autoreply transport, it
1977 always said "this is a copy of the message, including all the headers",
1978 even if body_only or headers_only was set. It now gives an appropriate
1981 PH/09 Applied a patch from the Sieve maintainer which:
1983 o fixes some comments
1984 o adds the (disabled) notify extension core
1985 o adds some debug output for the result of if/elsif tests
1986 o points to the current vacation draft in the documentation
1987 and documents the missing references header update
1991 o fixes a bug in processing the envelope test (when testing
1992 multiple envelope elements, the last element determinted the
1995 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
1996 Electronic Mail") by including:
1998 Auto-submitted: auto-generated
2000 in the messages that it generates (bounce messages and others, such as
2001 warnings). In the case of bounce messages for non-SMTP mesages, there was
2002 also a typo: it was using "Auto_submitted" (underscore instead of
2003 hyphen). Since every message generated by Exim is necessarily in response
2004 to another message, thes have all been changed to:
2006 Auto-Submitted: auto-replied
2008 in accordance with these statements in the RFC:
2010 The auto-replied keyword:
2012 - SHOULD be used on messages sent in direct response to another
2013 message by an automatic process,
2015 - MUST NOT be used on manually-generated messages,
2017 - MAY be used on Delivery Status Notifications (DSNs) and Message
2018 Disposition Notifications (MDNs),
2020 - MUST NOT be used on messages generated by automatic or periodic
2021 processes, except for messages which are automatic responses to
2024 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
2025 to the default Received: header definition.
2027 PH/12 Added log selector acl_warn_skipped (default on).
2029 PH/13 After a successful wildlsearch lookup, discard the values of numeric
2030 variables because (a) they are in the wrong storage pool and (b) even if
2031 they were copied, it wouldn't work properly because of the caching.
2033 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
2034 checking when decoding. Apparently there are clients that generate
2035 overlong encoded strings. Why am I not surprised?
2037 PH/15 If the first argument of "${if match_address" was not empty, but did not
2038 contain an "@" character, Exim crashed. Now it writes a panic log message
2039 and treats the condition as false.
2041 PH/16 In autoreply, treat an empty string for "once" the same as unset.
2043 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
2044 extension "envelope-auth". The code is finished and in agreement with
2045 other implementations, but there is no documentation so far and in fact,
2046 nobody wrote the draft yet. This extension is currently #undef'ed, thus
2047 not changing the active code.
2049 Print executed "if" and "elsif" statements when debugging is used. This
2050 helps a great deal to understand what a filter does.
2052 Document more things not specified clearly in RFC3028. I had all this
2053 sorted out, when out of a sudden new issues came to my mind. Oops."
2055 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
2058 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
2059 canonical form (as documented). However, after a host name lookup from
2060 the IP address, check_host() was doing a simple string comparison with
2061 addresses acquired from the DNS when checking that the found name did
2062 have the original IP as one of its addresses. Since any found IPv6
2063 addresses are likely to be in abbreviated form, the comparison could
2064 fail. Luckily, there already exists a function for doing the comparison
2065 by converting both addresses to binary, so now that is used instead of
2066 the text comparison.
2068 PH/20 There was another similar case to PH/19, when a complete host name was
2069 given in a host list; looking up its IP address could give an abbreviated
2070 form, whereas the current host's name might or might not be abbreviated.
2071 The same fix has been applied.
2077 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
2078 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
2081 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
2082 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
2084 PH/03 Typo: missing ".o" in src/pcre/Makefile.
2086 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
2087 header line, restrict the check to what is listed in RFCs 2369 and 2929.
2088 Also, for "Auto-Submitted", treat anything other than "no" as
2089 non-personal, in accordance with RFC 3834. (Previously it treated
2090 anything starting "auto-" as non-personal.)
2092 TF/01 The control=submission/name=... option had a problem with syntax
2093 errors if the name included a slash character. The /name= option
2094 now slurps the rest of the string, so it can include any characters
2095 but it must come last in the list of options (after /sender_retain
2098 PH/05 Some modifications to the interface to the fake nameserver for the new
2106 TK/01 Added the "success_on_redirect" address verification option. See
2107 NewStuff for rationale and an example.
2109 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
2111 PH/02 Patch to exigrep to allow it to work on syslog lines.
2113 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
2114 fread() to skip over the body file's header line, because in Cygwin the
2115 header line is locked and is inaccessible.
2117 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
2118 co-exist for some time) to make it clear that it is the Exim ID that is
2119 referenced, not the Message-ID: header line.
2121 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
2122 string_format() function, because snprintf() does not exist on all
2125 PH/06 The use of forbid_filter_existstest now also locks out the use of the
2126 ${stat: expansion item.
2128 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
2129 protocol synchronization error", to keep the pedants happy.
2131 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
2132 well as for IRIX systems, when gcc is being used. See the host.c source
2135 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
2137 PH/10 Named domain lists were not working if used in a queue_smtp_domains
2140 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
2141 transport and to the smtp transport in LMTP mode.
2143 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
2145 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
2146 run a filter in a subprocess. This could lead to confusion in subsequent
2147 lookups in the parent process. There should also be a search_tidyup() at
2148 the end of the subprocess.
2150 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
2151 only if the host matched helo_try_verify_hosts, which caused the
2152 verification to occur when the EHLO/HELO command was issued. The ACL just
2153 tested the remembered result. Now, if a previous verification attempt has
2154 not happened, "verify = helo" does it there and then.
2156 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
2158 TK/03 Fix log output including CR from clamd.
2160 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
2161 exist provoked a memory error which could cause a segfault.
2163 PH/15 Installed PCRE 6.2
2165 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
2167 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
2168 of the problem. Specifically, suggested +O2 rather than +O1 for the
2171 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
2173 PH/20 If a delivery was routed to a non-standard port by means of an SRV
2174 record, the port was not correctly logged when the outgoing_port log
2175 selector was set (it logged the transort's default port).
2177 PH/21 Added support for host-specific ports to manualroute, queryprogram,
2178 fallback_hosts, and "hosts" in the smtp transport.
2180 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
2181 host errors such as "Connection refused".
2183 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
2184 authentication with radiusclient 0.4.9:
2186 - Error returned from rc_read_config was caught wrongly
2187 - Username/password not passed on to radius server due to wrong length.
2189 The presumption is that some radiusclient API changes for 4.51/PH/17
2190 were not taken care of correctly. The code is still untested by me (my
2191 Linux distribution still has 0.3.2 of radiusclient), but it was
2192 contributed by a Radius user.
2194 PH/24 When doing a callout, the value of $domain wasn't set correctly when
2195 expanding the "port" option of the smtp transport.
2197 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
2198 while reading a MIME header. Thanks to Tom Hughes for a patch.
2200 PH/24 Include config.h inside local_scan.h so that configuration settings are
2203 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
2204 that in an ACL for RCPT (for example), you can examine exactly what was
2207 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
2208 commands, but it was not correctly comparing the address with the actual
2209 client host address. Thus, it would show the EHLO address in Received:
2210 header lines when this was not necessary.
2212 PH/27 Added the % operator to ${eval:}.
2214 PH/28 Exim tries to create and chdir to its spool directory when it starts;
2215 it should be ignoring failures (because with -C, for example, it has lost
2216 privilege). It wasn't ignoring creation failures other than "already
2219 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
2222 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
2223 because an input line was too long, either on its own, or by virtue of
2224 too many continuations, the temporary file was not being removed, and the
2225 return code was incorrect.
2227 PH/31 Missing "BOOL" in function definition in filtertest.c.
2229 PH/32 Applied Sieve patches from the maintainer.
2231 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
2233 PH/33 Added "verify = not_blind".
2235 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
2236 Local/Makefile (with some defaults set). These are used in built scripts
2237 such as exicyclog, but they have never been used in the exim_install
2238 script (though there are many overriding facilities there). I have
2239 arranged that the exim_install script now takes note of these two
2242 PH/35 Installed configuration files for Dragonfly.
2244 PH/36 When a locally submitted message by a trusted user did not contain a
2245 From: header, and the sender address was obtained from -f or from an SMTP
2246 MAIL command, and the trusted user did not use -F to supply a sender
2247 name, $originator_name was incorrectly used when constructing a From:
2248 header. Furthermore, $originator_name was used for submission mode
2249 messages from external hosts without From: headers in a similar way,
2250 which is clearly wrong.
2252 PH/37 Added control=suppress_local_fixups.
2254 PH/38 When log_selector = +received_sender was set, and the addition of the
2255 sender made the log line's construction buffer exactly full, or one byte
2256 less than full, an overflow happened when the terminating "\n" was
2259 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
2260 when the result of a list match is failure because a DNS lookup failed.
2262 PH/40 RM_COMMAND is now used in the building process.
2264 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
2265 the "build-* directories that it finds.
2267 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
2268 address in a domain literal was a prefix of an interface address.
2270 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
2271 when verifying a sender address, unless rewrite_headers is false.
2273 PH/44 Wrote a long comment about why errors_to addresses are verified as
2274 recipients, not senders.
2276 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
2277 the ratelimit ACL was added.
2279 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
2281 PH/46 Added extra information about PostgreSQL errors to the error string.
2283 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
2284 avoiding the need to install special zones in a real server. This is
2285 backwards compatible; if it can't find the fake resolver, it drops back.
2286 Thus, both old and new test suites can be run.
2288 TF/02 Added util/ratelimit.pl
2290 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
2291 clock is set back in time.
2293 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
2294 Candler <B.Candler@pobox.com>.
2296 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
2297 OK for addresses that are the result of redirections.
2299 PH/48 A number of further additions for the benefit of the new test suite,
2300 including a fake gethostbyname() that interfaces to the fake DNS resolver
2303 TF/06 The fix for widen_domains has also been applied to qualify_single and
2304 search_parents which are the other dnslookup options that can cause
2307 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
2310 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
2312 TF/07 Exim produced the error message "an SRV record indicated no SMTP
2313 service" if it encountered an MX record with an empty target hostname.
2314 The message is now "an MX or SRV record indicated no SMTP service".
2316 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
2317 if the DNS of the sending site is misconfigured. This is quite a
2318 common situation. This change restores the behaviour of treating a
2319 helo verification defer as a failure.
2321 PH/51 If self=fail was set on a router, the bounce message did not include the
2322 actual error message.
2328 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
2330 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
2331 command itself ended in error, the underlying message about the transport
2332 filter timeout was being overwritten with the pipe command error. Now the
2333 underlying error message should be appended to the second error message.
2335 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
2337 PH/02 Reduce the amount of output that "make" produces by default. Full output
2338 can still be requested.
2340 PH/03 The warning log line about a condition test deferring for a "warn" verb
2341 was being output only once per connection, rather than after each
2342 occurrence (because it was using the same function as for successful
2343 "warn" verbs). This seems wrong, so I have changed it.
2345 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
2346 it should not have, which might have caused a crash in the right
2347 circumstances, but probably never did.
2349 PH/04 Installed a modified version of Tony Finch's patch to make submission
2350 mode fix the return path as well as the Sender: header line, and to
2351 add a /name= option so that you can make the user's friendly name appear
2354 TF/03 Added the control = fakedefer ACL modifier.
2356 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
2357 Mark Lowes for thorough testing.
2359 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
2361 TK/03 Merged latest SRS patch from Miles Wilton.
2363 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
2364 with the definition in sysexits.h (which is #included earlier).
2365 Fortunately, Exim does not actually use EX_OK. The code used to try to
2366 preserve the sysexits.h value, by assumimg that macro definitions were
2367 scanned for macro replacements. I have been disabused of this notion,
2368 so now the code just undefines EX_OK before #including unistd.h.
2370 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
2371 in the smtp transport. When a block could not be written in a single
2372 write() function, the timeout was being re-applied to each part-write.
2373 This seems wrong - if the receiver was accepting one byte at a time it
2374 would take for ever. The timeout is now adjusted when this happens. It
2375 doesn't have to be particularly precise.
2377 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
2378 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
2380 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
2381 without a domain if the check to <postmaster@domain> fails.
2383 SC/01 Eximstats: added -xls and the ability to specify output files
2384 (patch written by Frank Heydlauf).
2386 SC/02 Eximstats: use FileHandles for outputing results.
2388 SC/03 Eximstats: allow any combination of xls, txt, and html output.
2390 SC/04 Eximstats: fixed display of large numbers with -nvr option
2392 SC/05 Eximstats: fixed merging of reports with empty tables.
2394 SC/06 Eximstats: added the -include_original_destination flag
2396 SC/07 Eximstats: removed tabs and trailing whitespace.
2398 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
2400 TK/06 MBOX spool code: Add real "From " MBOX separator line
2401 so the .eml file is really in mbox format (even though
2402 most programs do not really care). Patch from Alex Miller.
2404 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
2405 The latter is generated from $received_to and is only set if the
2406 message has one envelope recipient. SA can use these headers,
2407 obviously out-of-the-box. Patch from Alex Miller.
2409 PH/08 The ${def test on a variable was returning false if the variable's
2410 value was "0", contrary to what the specification has always said!
2411 The result should be true unless the variable is empty.
2413 PH/09 The syntax error of a character other than { following "${if
2414 def:variable_name" (after optional whitespace) was not being diagnosed.
2415 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
2416 accidental colon was present, for example, could give incorrect results.
2418 PH/10 Tidied the code in a number of places where the st_size field of a stat()
2419 result is used (not including appendfile, where other changes are about
2422 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
2423 This involved changing a lot of size variables from int to off_t. It
2424 should work with maildirs and everything.
2426 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
2427 spamd dying while we are connected to it.
2429 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
2432 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
2433 function for BDB 4.3.
2435 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
2437 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
2438 the sources for PCRE have been re-arranged and more files are now
2441 PH/15 The code I had for printing potentially long long variables in PH/11
2442 above was not the best (it lost precision). The length of off_t variables
2443 is now inspected at build time, and an appropriate printing format (%ld
2444 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
2445 to be "long long int" or "long int". This is needed for the internal
2446 formatting function string_vformat().
2448 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
2449 the configuration file to be ":syslog", then the script "guesses" where
2450 the logs files are, rather than using the compiled in default. In our
2451 case the guess is not the same as the compiled default, so the script
2452 suddenly stopped working when I started to use syslog. The patch checks
2453 to see if log_file_path is "". If so, it attempts to read it from exim
2454 with no configuration file to get the compiled in version, before it
2455 falls back to the previous guessing code."
2457 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
2458 implementing BATV in an Exim configuration. See NewStuff for the gory
2461 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
2462 Makefile that are specific to HP-UX.
2464 PH/18 If the "use_postmaster" option was set for a recipient callout together
2465 with the "random" option, the postmaster address was used as the MAIL
2466 FROM address for the random test, but not for the subsequent recipient
2467 test. It is now used for both.
2469 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
2470 patch removes a few documentation additions to RFC 3028, because the
2471 latest draft now contains them. It adds the new en;ascii-case comparator
2472 and a new error check for 8bit text in MIME parts. Comparator and
2473 require names are now matched exactly. I enabled the subaddress
2474 extension, but it is not well tested yet (read: it works for me)."
2476 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
2477 rework some of the code of TK/09 above to avoid the hardwired use of
2478 "%lld" and "long long". Replaced the call to snprintf() with a call to
2481 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
2482 records point to non-existent hosts", "retry timeout exceeded", and
2483 "retry time not reached for any host after a long failure period".
2485 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
2486 experimental DomainKeys support:
2488 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
2489 (2) On an error such as an illegally used "control", the wrong name for
2490 the control was given.
2492 These problems did NOT occur unless DomainKeys support was compiled.
2494 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
2496 PH/24 Added ${if match_ip condition.
2498 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
2499 databases so that it will be absolutely obvious if a crash occurs in the
2500 DB library. This is a regular occurrence (often caused by mis-matched
2503 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
2504 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
2505 picked up on a user's system that detects such things. There doesn't seem
2506 to be a gcc warning option for this - only an attribute that has to be
2507 put on the function's prototype. It seems that in Fedora Core 4 they have
2508 set this on a number of new functions. No doubt there will be more in due
2511 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
2512 specify a transport. However, if an address that was verified by such a
2513 router was the subject of a callout, Exim crashed because it tried to
2514 read the rcpt_include_affixes from the non-existent transport. Now it
2515 just assumes that the setting of that option is false. This bug was
2516 introduced by 4.51/PH/31.
2518 PH/28 Changed -d+all to exclude +memory, because that information is very
2519 rarely of interest, but it makes the output a lot bigger. People tend to
2520 do -d+all out of habit.
2522 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
2523 code in os-type was giving problems when libc.so lives in lib64, like on
2526 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
2527 aren't the modern standard, and it seems that some systems' include files
2528 don't always have them. Exim was already checking for some of the newer
2529 ones like T_AAAA, and defining it itself. I've added checks for all the
2530 record types that Exim uses.
2532 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
2533 not automatically generating a new one, as it is supposed to. This
2534 prevented TLS from working. If the file did exist, but contained invalid
2535 data, a new version was generated, as expected. It was only the case of a
2536 non-existent file that was broken.
2538 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
2539 with a change in libdomainkeys > 0.64.
2541 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
2542 from DNS. If the selector record carries the flag, it now has
2543 precedence over the domain-wide flag.
2545 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
2547 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
2548 the use of a port name that isn't defined in /etc/services) occurred, the
2549 message was deferred as in a normal delivery, and thus remained on the
2550 spool, instead of being failed because of the mua_wrapper setting. This
2551 is now fixed, and I tidied up some of the mua_wrapper messages at the
2554 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
2555 the messages in a hash of arrays rather than using individual hashes.
2556 This is a bit cleaner and results in dramatic memory savings, albeit
2557 at a slight CPU cost.
2559 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
2560 as requested by Marc Sherman.
2562 SC/10 Eximstats: added histograms for user specified patterns as requested
2565 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
2567 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
2568 fopen() in the content-scanning modules that did not already have it.
2574 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
2575 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
2577 TK/02 Fix ACL "control" statement not being available in MIME ACL.
2579 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
2581 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
2582 to test Sieve filters that use "vacation".
2584 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
2585 that changes the way the GnuTLS parameters are stored in the cache file.
2586 The new format can be generated externally. For backward compatibility,
2587 if the data in the cache doesn't make sense, Exim assumes it has read an
2588 old-format file, and it generates new data and writes a new file. This
2589 means that you can't go back to an older release without removing the
2592 PH/03 A redirect router that has both "unseen" and "one_time" set does not
2593 work if there are any delivery delays because "one_time" forces the
2594 parent to be marked "delivered", so its unseen clone is never tried
2595 again. For this reason, Exim now forbids the simultaneous setting of
2598 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
2599 redirected to themselves ("homonym" addresses). Read the long ChangeLog
2600 entry if you want to know the details. The fix, however, neglected to
2601 consider the case when local delivery batching is involved. The test for
2602 "previously delivered" was not happening when checking to see if an
2603 address could be batched with a previous (undelivered) one; under
2604 certain circumstances this could lead to multiple deliveries to the same
2607 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
2608 in its include files, and this causes problems building Exim.
2610 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
2611 header_syntax) but Exim was just ignoring anything given after a slash.
2612 In particular, this caused confusion with an attempt to use "verify =
2613 reverse_host_lookup/defer_ok". An error is now given when options are
2614 supplied for verify items that do not have them. (Maybe reverse_host_
2615 lookup should have a defer_ok option, but that's a different point.)
2617 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
2618 defined by RFC 821) to 2048, because there were problems with some AUTH
2619 commands, and RFC 1869 says the size should be increased for extended
2620 SMTP commands that take arguments.
2622 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
2625 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
2626 "unknown" error; now it says that the functionality isn't in the binary.
2628 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
2629 an address' error message when a string expansion fails (syntax or
2630 whatever). Otherwise the password may appear in the log. Following change
2631 PH/42 below, there is no longer a chance of it appearing in a bounce
2634 PH/11 Installed exipick version 20050225.0 from John Jetmore.
2636 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
2637 of its addresses was ever tried. (Bugzilla bug #2.)
2639 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
2640 the result incorrectly in the debug output. (It correctly added a newline
2641 to what was transported.)
2643 TF/01 Added $received_time.
2645 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
2646 commented out examples of how to interface to a virus scanner and to
2647 SpamAssassin. Also added commented examples of av_scanner and
2648 spamd_address settings.
2650 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
2651 and controls are allowed in which ACLs. There were a couple of minor
2652 errors. Some of the entries in the conditions table (which is a table of
2653 where they are NOT allowed) were getting very unwieldy; rewrote them as a
2654 negation of where the condition IS allowed.
2656 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
2658 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
2659 header file does not have a version number, so I've had to invent a new
2660 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
2661 API. The code is untested by me (my Linux distribution still has 0.3.2 of
2662 radiusclient), but it was contributed by a Radius user.
2664 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
2665 files or directories, for OpenSSL.
2667 PH/19 When an Exim process that is running as root has to create an Exim log
2668 file, it does so in a subprocess that runs as exim:exim so as to get the
2669 ownership right at creation (otherwise, other Exim processes might see
2670 the file with the wrong ownership). There was no test for failure of this
2671 fork() call, which would lead to the process getting stuck as it waited
2672 for a non-existent subprocess. Forks do occasionally fail when resources
2673 run out. I reviewed all the other calls to fork(); they all seem to check
2676 PH/20 When checking for unexpected SMTP input at connect time (before writing
2677 the banner), Exim was not dealing correctly with a non-positive return
2678 from the read() function. If the client had disconnected by this time,
2679 the result was a log entry for a synchronization error with an empty
2680 string after "input=" when read() returned zero. If read() returned -1
2681 (an event I could not check), uninitialized data bytes were printed.
2682 There were reports of junk text (parts of files, etc) appearing after
2685 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
2687 PH/22 Added support for macro redefinition, and (re)definition in between
2688 driver and ACL definitions.
2690 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
2691 forgetting to use the resulting value; it was using the unexpanded value.
2693 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
2694 hadn't been configured. The fix is from Juergen Kreileder, who
2695 understands it better than I do:
2697 "Here's what I see happening with three configured cyrus_sasl
2698 authenticators configured (plain, login, cram-md5):
2700 On startup auth_cyrus_sasl_init() gets called for each of these.
2701 This means three calls to sasl_listmech() without a specified mech_list.
2702 => SASL tests which mechs of all available mechs actually work
2703 => three warnings about OTP not working
2704 => the returned list contains: plain, login, cram-md5, digest-md5, ...
2706 With the patch, sasl_listmech() also gets called three times. But now
2707 SASL's mech_list option is set to the server_mech specified in the the
2708 authenticator. Or in other words, the answer from sasl_listmech()
2709 gets limited to just the mech you're testing for (which is different
2711 => the return list contains just 'plain' or 'login', 'cram-md5' or
2712 nothing depending on the value of ob->server_mech.
2714 I've just tested the patch: Authentication still works fine,
2715 unavailable mechs specified in the exim configuration are still
2716 caught, and the auth.log warnings about OTP are gone."
2718 PH/25 When debugging is enabled, the contents of the command line are added
2719 to the debugging output, even when log_selector=+arguments is not
2722 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
2723 answer is "GNU", and only if the return is "GNU/something" is the answer
2726 PH/27 $acl_verify_message is now set immediately after the failure of a
2727 verification in an ACL, and so is available in subsequent modifiers. In
2728 particular, the message can be preserved by coding like this:
2730 warn !verify = sender
2731 set acl_m0 = $acl_verify_message
2733 Previously, $acl_verify_message was set only while expanding "message"
2734 and "log_message" when a very denied access.
2736 PH/28 Modified OS/os.c-Linux with
2738 -#ifndef OS_LOAD_AVERAGE
2739 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
2741 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
2742 nomenclature these days.)
2744 PH/29 Installed patch from the Sieve maintainer that adds the options
2745 sieve_useraddress and sieve_subaddress to the redirect router.
2747 PH/30 In these circumstances:
2748 . Two addresses routed to the same list of hosts;
2749 . First host does not offer TLS;
2750 . First host accepts first address;
2751 . First host gives temporary error to second address;
2752 . Second host offers TLS and a TLS session is established;
2753 . Second host accepts second address.
2754 Exim incorrectly logged both deliveries with the TLS parameters (cipher
2755 and peerdn, if requested) that were in fact used only for the second
2758 PH/31 When doing a callout as part of verifying an address, Exim was not paying
2759 attention to any local part prefix or suffix that was matched by the
2760 router that accepted the address. It now behaves in the same way as it
2761 does for delivery: the affixes are removed from the local part unless
2762 rcpt_include_affixes is set on the transport.
2764 PH/32 Add the sender address, as F=<...>, to the log line when logging a
2765 timeout during the DATA phase of an incoming message.
2767 PH/33 Sieve envelope tests were broken for match types other than :is. I have
2768 applied a patch sanctioned by the Sieve maintainer.
2770 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
2771 the uid or gid is negative. A case of a negative gid caused this to be
2772 noticed. The fix allows for either to be negative.
2774 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
2775 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
2778 PH/36 The change PH/12 above was broken. Fixed it.
2780 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
2781 the grounds that routing the same address twice would always produce the
2782 same answer. This might have been true once, but it is certainly no
2783 longer true now. Routing a child address may depend on the previous
2784 routing that produced that child. Some complicated redirection strategies
2785 went wrong when messages had multiple recipients, and made Exim's
2786 behaviour dependent on the order in which the addresses were given.
2788 I have moved the duplicate checking until after the routing is complete.
2789 Exim scans the addresses that are assigned to local and remote
2790 transports, and removes any duplicates. This means that more work will be
2791 done, as duplicates will always all be routed, but duplicates are
2792 presumably rare, so I don't expect this is of any significance.
2794 For deliveries to pipes, files, and autoreplies, the duplicate checking
2795 still happens during the routing process, since they are not going to be
2798 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
2799 It corrects a timeout issue with spamd. This is Ian's comment: "The
2800 background is that sometimes spamd either never reads data from a
2801 connection it has accepted, or it never writes response data. The exiscan
2802 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
2803 blindly assumes that writes won't block so it may never time out."
2805 PH/39 Allow G after quota size as well as K and M.
2807 PH/40 The value set for $authenticated_id in an authenticator may not contain
2808 binary zeroes or newlines because the value is written to log lines and
2809 to spool files. There was no check on this. Now the value is run through
2810 the string_printing() function so that such characters are converted to
2811 printable escape sequences.
2813 PH/41 $message_linecount is a new variable that contains the total number of
2814 lines in the message. Compare $body_linecount, which is the count for the
2817 PH/42 Exim no longer gives details of delivery errors for specific addresses in
2818 bounce and delay warning messages, except in certain special cases, which
2821 (a) An SMTP error message from a remote host;
2822 (b) A message specified in a :fail: redirection;
2823 (c) A message specified in a "fail" command in a system filter;
2824 (d) A message specified in a FAIL return from the queryprogram router;
2825 (e) A message specified by the cannot_route_message router option.
2827 In these cases only, Exim does include the error details in bounce and
2828 warning messages. There are also a few cases where bland messages such
2829 as "unrouteable address" or "local delivery error" are given.
2831 PH/43 $value is now also set for the "else" part of a ${run expansion.
2833 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
2834 being worked on, but at least Exim now implements the latest version to
2837 PH/45 In a pipe transport, although a timeout while waiting for the pipe
2838 process to complete was treated as a delivery failure, a timeout while
2839 writing the message to the pipe was logged, but erroneously treated as a
2840 successful delivery. Such timeouts include transport filter timeouts. For
2841 consistency with the overall process timeout, these timeouts are now
2842 treated as errors, giving rise to delivery failures by default. However,
2843 there is now a new Boolean option for the pipe transport called
2844 timeout_defer, which, if set TRUE, converts the failures into defers for
2845 both kinds of timeout. A transport filter timeout is now identified in
2848 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
2849 systems where "make" and "gmake" are different, calling "gmake" at top
2850 level broke things. I've arranged for the value of $(MAKE) to be passed
2851 from the Makefile to this script so that it can call the same version of
2855 A note about Exim versions 4.44 and 4.50
2856 ----------------------------------------
2858 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
2859 changes of various kinds. As a consequence, a big documentation update was
2860 needed. This delayed the release for rather longer than seemed good, especially
2861 in the light of a couple of (minor) security issues. Therefore, the changes
2862 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
2863 release. So 4.44 and 4.50 are in effect two different branches that both start
2866 I have left the 4.50 change log unchanged; it contains all the changes since
2867 4.43. The change log for 4.44 is below; many of its items are identical to
2868 those for 4.50. This seems to be the most sensible way to preserve the
2869 historical information.
2875 1. Minor wording change to the doc/README.SIEVE file.
2877 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
2878 computation of the current number of files was incorrect.
2880 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
2881 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
2884 4. Give more explanation in the error message when the command for a transport
2885 filter fails to execute.
2887 5. There are several places where Exim runs a non-Exim command in a
2888 subprocess. The SIGUSR1 signal should be disabled for these processes. This
2889 was being done only for the command run by the queryprogram router. It is
2890 now done for all such subprocesses. The other cases are: ${run, transport
2891 filters, and the commands run by the lmtp and pipe transports.
2893 6. Added CONFIGURE_GROUP build-time option.
2895 7. Some older OS have a limit of 256 on the maximum number of file
2896 descriptors. Exim was using setrlimit() to set 1000 as a large value
2897 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
2898 systems. I've change it so that if it can't get 1000, it tries for 256.
2900 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
2901 was an oversight, and furthermore, ever since the addition of extra
2902 controls (e.g. 4.43/32), the checks on when to allow different forms of
2903 "control" were broken. There should now be diagnostics for all cases when a
2904 control that does not make sense is encountered.
2906 9. Added the /retain_sender option to "control=submission".
2908 10. $recipients is now available in the predata ACL (oversight).
2910 11. Tidy the search cache before the fork to do a delivery from a message
2911 received from the command line. Otherwise the child will trigger a lookup
2912 failure and thereby defer the delivery if it tries to use (for example) a
2913 cached ldap connection that the parent has called unbind on.
2915 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
2916 of $address_data from the recipient verification was clobbered by the
2917 sender verification.
2919 13. The value of address_data from a sender verification is now available in
2920 $sender_address_data in subsequent conditions in the ACL statement.
2922 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
2924 15. Added a new option "connect=<time>" to callout options, to set a different
2927 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
2928 was its contents. (It was OK if the option was not defined at all.)
2930 17. A "Completed" log line is now written for messages that are removed from
2931 the spool by the -Mrm option.
2933 18. New variables $sender_verify_failure and $recipient_verify_failure contain
2934 information about exactly what failed.
2936 19. Added -dd to debug only the daemon process.
2938 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
2939 handles timeouts, both on the server side and network timeouts. Renamed the
2940 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
2942 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
2943 It is now set to "smtps".
2945 22. $host_address is now set to the target address during the checking of
2946 ignore_target_hosts.
2948 23. When checking ignore_target_hosts for an ipliteral router, no host name was
2949 being passed; this would have caused $sender_host_name to have been used if
2950 matching the list had actually called for a host name (not very likely,
2951 since this list is usually IP addresses). A host name is now passed as
2954 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
2955 code that specifies a non-restarting handler (typically sigaction() in
2956 modern systems) in an attempt to fix a rare and obscure crash bug.
2958 25. Narrowed the window for a race in the daemon that could cause it to ignore
2959 SIGCHLD signals. This is not a major problem, because they are used only to
2960 wake it up if nothing else does.
2962 26. A malformed maildirsize file could cause Exim to calculate negative values
2963 for the mailbox size or file count. Odd effects could occur as a result.
2964 The maildirsize information is now recalculated if the size or filecount
2967 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
2968 support for a long time. Removed HAVE_SYS_VFS_H.
2970 28. Installed the latest version of exipick from John Jetmore.
2972 29. In an address list, if the pattern was not a regular expression, an empty
2973 subject address (from a bounce message) matched only if the pattern was an
2974 empty string. Non-empty patterns were not even tested. This was the wrong
2975 because it is perfectly reasonable to use an empty address as part of a
2976 database query. An empty address is now tested by patterns that are
2977 lookups. However, all the other forms of pattern expect the subject to
2978 contain a local part and a domain, and therefore, for them, an empty
2979 address still always fails if the pattern is not itself empty.
2981 30. Exim went into a mad DNS loop when attempting to do a callout where the
2982 host was specified on an smtp transport, and looking it up yielded more
2983 than one IP address.
2985 31. Re-factored the code for checking spool and log partition space into a
2986 function that finds that data and another that does the check. The former
2987 is then used to implement four new variables: $spool_space, $log_space,
2988 $spool_inodes, and $log_inodes.
2990 32. The RFC2047 encoding function was originally intended for short strings
2991 such as real names; it was not keeping to the 75-character limit for
2992 encoded words that the RFC imposes. It now respects the limit, and
2993 generates multiple encoded words if necessary. To be on the safe side, I
2994 have increased the buffer size for the ${rfc2047: expansion operator from
2997 33. It is now permitted to omit both strings after an "if" condition; if the
2998 condition is true, the result is "true". As before, when the second string
2999 is omitted, a false condition yields an empty string. This makes it less
3000 cumbersome to write custom ACL and router conditions.
3002 34. Failure to deliver a bounce message always caused it to be frozen, even if
3003 there was an errors_to setting on the router. The errors_to setting is now
3006 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
3007 canonical form (fully expanded) before being placed in
3008 $sender_host_address.
3010 36. The table in the code that translates DNS record types into text (T_A to
3011 "A" for instance) was missing entries for NS and CNAME. It is just possible
3012 that this could have caused confusion if both these types were looked up
3013 for the same domain, because the text type is used as part of Exim's
3014 per-process caching. But the chance of anyone hitting this buglet seems
3017 37. The dnsdb lookup has been extended in a number of ways.
3019 (1) There is a new type, "zns", which walks up the domain tree until it
3020 finds some nameserver records. It should be used with care.
3022 (2) There is a new type, "mxh", which is like "mx" except that it returns
3023 just the host names, not the priorities.
3025 (3) It is now possible to give a list of domains (or IP addresses) to be
3026 looked up. The behaviour when one of the lookups defers can be
3027 controlled by a keyword.
3029 (4) It is now possible to specify the separator character for use when
3030 multiple records are returned.
3032 38. The dnslists ACL condition has been extended: it is now possible to supply
3033 a list of IP addresses and/or domains to be looked up in a particular DNS
3036 39. Added log_selector=+queue_time_overall.
3038 40. When running the queue in the test harness, wait just a tad after forking a
3039 delivery process, to get repeatability of debugging output.
3041 41. Include certificate and key file names in error message when GnuTLS fails
3042 to set them up, because the GnuTLS error message doesn't include the name
3043 of the failing file when there is a problem reading it.
3045 42. Allow both -bf and -bF in the same test run.
3047 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
3049 44. The "Exiscan patch" is now merged into the mainline Exim source.
3051 45. Sometimes the final signoff response after QUIT could fail to get
3052 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
3053 < 0 before doing a fflush(). This bug looks as though it goes back to the
3054 introduction of TLS in release 3.20, but "sometimes" must have been rare
3055 because the tests only now provoked it.
3057 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
3058 (this can affect the format of dates).
3060 47. exim_tidydb, when checking for the continued existence of a message for
3061 which it has found a message-specific retry record, was not finding
3062 messages that were in split spool directories. Consequently, it was
3063 deleting retry records that should have stayed in existence.
3065 48. Steve fixed some bugs in eximstats.
3067 49. The SPA authentication driver was not abandoning authentication and moving
3068 on to the next authenticator when an expansion was forced to fail,
3069 contradicting the general specification for all authenticators. Instead it
3070 was generating a temporary error. It now behaves as specified.
3072 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
3073 (the order specifies the preference for clients). The order is now AES256,
3074 AES128, 3DES, ARCFOUR128.
3076 51. Small patch to Sieve code - explicitly set From: when generating an
3079 52. Exim crashed if a remote delivery caused a very long error message to be
3080 recorded - for instance if somebody sent an entire SpamAssassin report back
3081 as a large number of 550 error lines. This bug was coincidentally fixed by
3082 increasing the size of one of Exim's internal buffers (big_buffer) that
3083 happened as part of the Exiscan merge. However, to be on the safe side, I
3084 have made the code more robust (and fixed the comments that describe what
3087 53. Now that there can be additional text after "Completed" in log lines (if
3088 the queue_time_overall log selector is set), a one-byte patch to exigrep
3089 was needed to allow it to recognize "Completed" as not the last thing in
3092 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
3093 patch that reportedly fixes this has been added. I am not expert enough to
3094 create a test for it. This is what the patch creator wrote:
3096 "I found a little strange behaviour of ldap code when working with
3097 Windows 2003 AD Domain, where users was placed in more than one
3098 Organization Units. When I tried to give exim partial DN, the exit code
3099 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
3100 But simultaneously result of request was absolutely normal ldap result,
3101 so I produce this patch..."
3103 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
3104 REFERENCE, so I have modified the code to exclude the patch when that macro
3107 55. Some experimental protocols are using DNS PTR records for new purposes. The
3108 keys for these records are domain names, not reversed IP addresses. The
3109 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
3110 leaves it alone. Component reversal etc. now happens only for IP addresses.
3113 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
3115 57. Double the size of the debug message buffer (to 2048) so that more of very
3116 long debug lines gets shown.
3118 58. The exicyclog utility now does better if the number of log files to keep
3119 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
3121 59. Two changes related to the smtp_active_hostname option:
3123 (1) $smtp_active_hostname is now available as a variable.
3124 (2) The default for smtp_banner uses $smtp_active_hostname instead
3125 of $primary_hostname.
3127 60. The host_aton() function is supposed to be passed a string that is known
3128 to be a valid IP address. However, in the case of IPv6 addresses, it was
3129 not checking this. This is a hostage to fortune. Exim now panics and dies
3130 if the condition is not met. A case was found where this could be provoked
3131 from a dnsdb PTR lookup with an IPv6 address that had more than 8
3132 components; fortuitously, this particular loophole had already been fixed
3133 by change 4.50/55 above.
3135 If there are any other similar loopholes, the new check in host_aton()
3136 itself should stop them being exploited. The report I received stated that
3137 data on the command line could provoke the exploit when Exim was running as
3138 exim, but did not say which command line option was involved. All I could
3139 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
3140 running as the user.
3143 61. There was a buffer overflow vulnerability in the SPA authentication code
3144 (which came originally from the Samba project). I have added a test to the
3145 spa_base64_to_bits() function which I hope fixes it.
3148 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
3149 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
3151 63. The daemon start-up calls getloadavg() while still root for those OS that
3152 need the first call to be done as root, but it missed one case: when
3153 deliver_queue_load_max is set with deliver_drop_privilege. This is
3154 necessary for the benefit of the queue runner, because there is no re-exec
3155 when deliver_drop_privilege is set.
3157 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
3158 This has been fixed.
3160 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
3161 was in use, was not putting the data itself into the right store pool;
3162 consequently, it could be overwritten for a subsequent message in the same
3163 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
3166 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
3168 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
3169 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
3170 the function were treating the return as a boolean value, which happened to
3171 work because 0=false and not-0=true, but is not correct code.
3173 68. The host_aton() function was not handling scoped IPv6 addresses (those
3174 with, for example, "%eth0" on the end) correctly.
3176 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
3177 negated items (that is, ~something) in unsigned ints. Some compilers
3178 apparently mutter when there is no cast.
3180 70. If an address verification called from an ACL failed, and did not produce a
3181 user-specific message (i.e. there was only a "system" message), nothing was
3182 put in $acl_verify_message. In this situation, it now puts the system
3185 71. Change 4.23/11 added synchronization checking at the start of an SMTP
3186 session; change 4.31/43 added the unwanted input to the log line - except
3187 that it did not do this in the start of session case. It now does.
3189 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
3190 This is wrong and can cause the other end to generate a synchronization
3191 error if it is another Exim or anything else that does the synchronization
3192 check. A QUIT command is no longer sent after a timeout.
3194 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
3195 during host lookups.
3197 74. The defer_ok option of callout verification was not working if it was used
3198 when verifying addresses in header lines, that is, for this case:
3200 verify = header_sender/callout=defer_ok
3202 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
3203 those file descriptors could be used for SMTP connections. If anything
3204 wrote to stderr (the example that came up was "warn" in embedded Perl), it
3205 could be sent to the SMTP client, causing chaos. The daemon now opens
3206 stdin, stdout, and stderr to /dev/null when it puts itself into the
3209 76. Arrange for output from Perl's "warn" command to be written to Exim's main
3210 log by default. The user can override this with suitable Perl magic.
3212 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
3213 the log message when discard triggers, was not working for the DATA ACL or
3214 for the non-SMTP ACL.
3216 78. Error message wording change in sieve.c.
3218 79. If smtp_accept_max_per_host was set, the number of connections could be
3219 restricted to fewer than expected, because the daemon was trying to set up
3220 a new connection before checking whether the processes handling previous
3221 connections had finished. The check for completed processes is now done
3222 earlier. On busy systems, this bug wouldn't be noticed because something
3223 else would have woken the daemon, and it would have reaped the completed
3226 80. If a message was submitted locally by a user whose login name contained one
3227 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
3228 It caused a spool format error. I have fixed the spool reading code. A
3229 related problem was that the "from" clause in the Received: line became
3230 illegal because of the space(s). It is now covered by ${quote_local_part.
3232 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
3235 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
3237 83. Patch from Sieve maintainer to fix unterminated string problem in
3238 "vacation" handling.
3240 84. Some minor changes to the Linux configuration files to help with other
3241 OS variants using glibc.
3243 85. One more patch for Sieve to update vacation handling to latest spec.
3246 ----------------------------------------------------
3247 See the note above about the 4.44 and 4.50 releases.
3248 ----------------------------------------------------
3254 1. Change 4.43/35 introduced a bug that caused file counts to be
3255 incorrectly computed when quota_filecount was set in an appendfile
3258 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
3259 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
3262 3. Give more explanation in the error message when the command for a transport
3263 filter fails to execute.
3265 4. There are several places where Exim runs a non-Exim command in a
3266 subprocess. The SIGUSR1 signal should be disabled for these processes. This
3267 was being done only for the command run by the queryprogram router. It is
3268 now done for all such subprocesses. The other cases are: ${run, transport
3269 filters, and the commands run by the lmtp and pipe transports.
3271 5. Some older OS have a limit of 256 on the maximum number of file
3272 descriptors. Exim was using setrlimit() to set 1000 as a large value
3273 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
3274 systems. I've change it so that if it can't get 1000, it tries for 256.
3276 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
3277 was an oversight, and furthermore, ever since the addition of extra
3278 controls (e.g. 4.43/32), the checks on when to allow different forms of
3279 "control" were broken. There should now be diagnostics for all cases when a
3280 control that does not make sense is encountered.
3282 7. $recipients is now available in the predata ACL (oversight).
3284 8. Tidy the search cache before the fork to do a delivery from a message
3285 received from the command line. Otherwise the child will trigger a lookup
3286 failure and thereby defer the delivery if it tries to use (for example) a
3287 cached ldap connection that the parent has called unbind on.
3289 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
3290 of $address_data from the recipient verification was clobbered by the
3291 sender verification.
3293 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
3294 was its contents. (It was OK if the option was not defined at all.)
3296 11. A "Completed" log line is now written for messages that are removed from
3297 the spool by the -Mrm option.
3299 12. $host_address is now set to the target address during the checking of
3300 ignore_target_hosts.
3302 13. When checking ignore_target_hosts for an ipliteral router, no host name was
3303 being passed; this would have caused $sender_host_name to have been used if
3304 matching the list had actually called for a host name (not very likely,
3305 since this list is usually IP addresses). A host name is now passed as
3308 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
3309 code that specifies a non-restarting handler (typically sigaction() in
3310 modern systems) in an attempt to fix a rare and obscure crash bug.
3312 15. Narrowed the window for a race in the daemon that could cause it to ignore
3313 SIGCHLD signals. This is not a major problem, because they are used only to
3314 wake it up if nothing else does.
3316 16. A malformed maildirsize file could cause Exim to calculate negative values
3317 for the mailbox size or file count. Odd effects could occur as a result.
3318 The maildirsize information is now recalculated if the size or filecount
3321 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
3322 support for a long time. Removed HAVE_SYS_VFS_H.
3324 18. Updated exipick to current release from John Jetmore.
3326 19. Allow an empty sender to be matched against a lookup in an address list.
3327 Previously the only cases considered were a regular expression, or an
3330 20. Exim went into a mad DNS lookup loop when doing a callout where the
3331 host was specified on the transport, if the DNS lookup yielded more than
3334 21. The RFC2047 encoding function was originally intended for short strings
3335 such as real names; it was not keeping to the 75-character limit for
3336 encoded words that the RFC imposes. It now respects the limit, and
3337 generates multiple encoded words if necessary. To be on the safe side, I
3338 have increased the buffer size for the ${rfc2047: expansion operator from
3341 22. Failure to deliver a bounce message always caused it to be frozen, even if
3342 there was an errors_to setting on the router. The errors_to setting is now
3345 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
3346 canonical form (fully expanded) before being placed in
3347 $sender_host_address.
3349 24. Updated eximstats to version 1.33
3351 25. Include certificate and key file names in error message when GnuTLS fails
3352 to set them up, because the GnuTLS error message doesn't include the name
3353 of the failing file when there is a problem reading it.
3355 26. Expand error message when OpenSSL has problems setting up cert/key files.
3358 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
3359 (this can affect the format of dates).
3361 28. exim_tidydb, when checking for the continued existence of a message for
3362 which it has found a message-specific retry record, was not finding
3363 messages that were in split spool directories. Consequently, it was
3364 deleting retry records that should have stayed in existence.
3366 29. eximstats updated to version 1.35
3367 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
3368 1.35 - bugfix such that pie charts by volume are generated correctly
3370 30. The SPA authentication driver was not abandoning authentication and moving
3371 on to the next authenticator when an expansion was forced to fail,
3372 contradicting the general specification for all authenticators. Instead it
3373 was generating a temporary error. It now behaves as specified.
3375 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
3376 (the order specifies the preference for clients). The order is now AES256,
3377 AES128, 3DES, ARCFOUR128.
3379 31. Small patch to Sieve code - explicitly set From: when generating an
3382 32. Exim crashed if a remote delivery caused a very long error message to be
3383 recorded - for instance if somebody sent an entire SpamAssassin report back
3384 as a large number of 550 error lines. This bug was coincidentally fixed by
3385 increasing the size of one of Exim's internal buffers (big_buffer) that
3386 happened as part of the Exiscan merge. However, to be on the safe side, I
3387 have made the code more robust (and fixed the comments that describe what
3390 33. Some experimental protocols are using DNS PTR records for new purposes. The
3391 keys for these records are domain names, not reversed IP addresses. The
3392 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
3393 leaves it alone. Component reversal etc. now happens only for IP addresses.
3396 34. The host_aton() function is supposed to be passed a string that is known
3397 to be a valid IP address. However, in the case of IPv6 addresses, it was
3398 not checking this. This is a hostage to fortune. Exim now panics and dies
3399 if the condition is not met. A case was found where this could be provoked
3400 from a dnsdb PTR lookup with an IPv6 address that had more than 8
3401 components; fortuitously, this particular loophole had already been fixed
3402 by change 4.50/55 or 4.44/33 above.
3404 If there are any other similar loopholes, the new check in host_aton()
3405 itself should stop them being exploited. The report I received stated that
3406 data on the command line could provoke the exploit when Exim was running as
3407 exim, but did not say which command line option was involved. All I could
3408 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
3409 running as the user.
3412 35. There was a buffer overflow vulnerability in the SPA authentication code
3413 (which came originally from the Samba project). I have added a test to the
3414 spa_base64_to_bits() function which I hope fixes it.
3417 36. The daemon start-up calls getloadavg() while still root for those OS that
3418 need the first call to be done as root, but it missed one case: when
3419 deliver_queue_load_max is set with deliver_drop_privilege. This is
3420 necessary for the benefit of the queue runner, because there is no re-exec
3421 when deliver_drop_privilege is set.
3423 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
3424 was in use, was not putting the data itself into the right store pool;
3425 consequently, it could be overwritten for a subsequent message in the same
3426 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
3429 38. Sometimes the final signoff response after QUIT could fail to get
3430 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
3431 < 0 before doing a fflush(). This bug looks as though it goes back to the
3432 introduction of TLS in release 3.20, but "sometimes" must have been rare
3433 because the tests only now provoked it.
3439 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
3440 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
3441 Now it returns an integer. A number of calls were still expecting a T/F
3442 return. Fortuitously, in all cases, the tests worked in OK situations,
3443 which is the norm. However, things would have gone wrong on any write
3444 failures on the smtp file descriptor. This function is used when sending
3445 messages over SMTP and also when doing verify callouts.
3447 2. When Exim is called to do synchronous delivery of a locally submitted
3448 message (the -odf or -odi options), it no longer closes stderr before doing
3451 3. Implemented the mua_wrapper option.
3453 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
3455 5. Implemented the functions header_remove(), header_testname(),
3456 header_add_at_position(), and receive_remove_recipient(), and exported them
3459 6. If an ACL "warn" statement specified the addition of headers, Exim already
3460 inserted X-ACL-Warn: at the start if there was no header name. However, it
3461 was not making this test for the second and subsequent header lines if
3462 there were newlines in the string. This meant that an invalid header could
3463 be inserted if Exim was badly configured.
3465 7. Allow an ACL "warn" statement to add header lines at the start or after all
3466 the Received: headers, as well as at the end.
3468 8. Added the rcpt_4xx retry error code.
3470 9. Added postmaster_mailfrom=xxx to callout verification option.
3472 10. Added mailfrom=xxxx to the callout verification option, for verify=
3475 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
3476 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
3478 12. Inserted some casts to stop certain compilers warning when using pointer
3479 differences as field lengths or precisions in printf-type calls (mostly
3480 affecting debugging statements).
3482 13. Added optional readline() support for -be (dynamically loaded).
3484 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
3485 same clock tick as a message's arrival, so that its received time was the
3486 same as the "first fail" time on the retry record, and that message
3487 remained on the queue past the ultimate address timeout, every queue runner
3488 would try a delivery (because it was past the ultimate address timeout) but
3489 after another failure, the ultimate address timeout, which should have then
3490 bounced the address, did not kick in. This was a "< instead of <=" error;
3491 in most cases the first failure would have been in the next clock tick
3492 after the received time, and all would be well.
3494 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
3495 being recognized when the domain list was tested by the match_domain
3496 condition in an expansion string.
3498 16. Added the ${str2b64: operator.
3500 17. Exim was always calling setrlimit() to set a large limit for the number of
3501 processes, without checking whether the existing limit was already
3502 adequate. (It did check for the limit on file descriptors.) Furthermore,
3503 errors from getrlimit() and setrlimit() were being ignored. Now they are
3504 logged to the main and panic logs, but Exim does carry on, to try to do its
3505 job under whatever limits there are.
3507 18. Imported PCRE 5.0.
3509 19. Trivial typo in log message " temporarily refused connection" (the leading
3512 20. If the log selector return_path_on_delivery was set and an address was
3513 redirected to /dev/null, the delivery process crashed because it assumed
3514 that a return path would always be set for a "successful" delivery. In this
3515 case, the whole delivery is bypassed as an optimization, and therefore no
3518 21. Internal re-arrangement: the function for sending a challenge and reading
3519 a response while authentication was assuming a zero-terminated challenge
3520 string. It's now changed to take a pointer and a length, to allow for
3521 binary data in such strings.
3523 22. Added the cyrus_sasl authenticator (code supplied by MBM).
3525 23. Exim was not respecting finduser_retries when seeking the login of the
3526 uid under which it was called; it was always trying 10 times. (The default
3527 setting of finduser_retries is zero.) Also, it was sleeping after the final
3528 failure, which is pointless.
3530 24. Implemented tls_on_connect_ports.
3532 25. Implemented acl_smtp_predata.
3534 26. If the domain in control=submission is set empty, Exim assumes that the
3535 authenticated id is a complete email address when it generates From: or
3536 Sender: header lines.
3538 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
3539 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
3540 chown and chgrp in /bin and hostname in /usr/bin.
3542 28. Exim was keeping the "process log" file open after each use, just as it
3543 does for the main log. This opens the possibility of it remaining open for
3544 long periods when the USR1 signal hits a daemon. Occasional processlog
3545 errors were reported, that could have been caused by this. Anyway, it seems
3546 much more sensible not to leave this file open at all, so that is what now
3549 29. The long-running daemon process does not normally write to the log once it
3550 has entered its main loop, and it closes the log before doing so. This is
3551 so that log files can straightforwardly be renamed and moved. However,
3552 there are a couple of unusual error situations where the daemon does write
3553 log entries, and I had neglected to close the log afterwards.
3555 30. The text of an SMTP error response that was received during a remote
3556 delivery was being truncated at 512 bytes. This is too short for some of
3557 the long messages that one sometimes sees. I've increased the limit to
3560 31. It is now possible to make retry rules that apply only when a message has a
3561 specific sender, in particular, an empty sender.
3563 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
3564 it possible to be selective about when SMTP synchronization is enforced.
3566 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
3568 32. Implemented hosts_connection_nolog.
3570 33. Added an ACL for QUIT.
3572 34. Setting "delay_warning=" to disable warnings was not working; it gave a
3575 35. Added mailbox_size and mailbox_filecount to appendfile.
3577 36. Added control = no_multiline_responses to ACLs.
3579 37. There was a bug in the logic of the code that waits for the clock to tick
3580 in the case where the clock went backwards by a substantial amount such
3581 that the microsecond fraction of "now" was more than the microsecond
3582 fraction of "then" (but the whole seconds number was less).
3584 38. Added support for the libradius Radius client library this is found on
3585 FreeBSD (previously only the radiusclient library was supported).
3591 1. When certain lookups returned multiple values in the form name=value, the
3592 quoting of the values was not always being done properly. Specifically:
3593 (a) If the value started with a double quote, but contained no whitespace,
3595 (b) If the value contained whitespace other than a space character (i.e.
3596 tabs or newlines or carriage returns) it was not quoted.
3597 This fix has been applied to the mysql and pgsql lookups by writing a
3598 separate quoting function and calling it from the lookup code. The fix
3599 should probably also be applied to nisplus, ibase and oracle lookups, but
3600 since I cannot test any of those, I have not disturbed their existing code.
3602 2. A hit in the callout cache for a specific address caused a log line with no
3603 reason for rejecting RCPT. Now it says "Previous (cached) callout
3604 verification failure".
3606 3. There was an off-by-one bug in the queryprogram router. An over-long
3607 return line was truncated at 256 instead of 255 characters, thereby
3608 overflowing its buffer with the terminating zero. As well as fixing this, I
3609 have increased the buffer size to 1024 (and made a note to document this).
3611 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
3612 when Exim is waiting for an SMTP response from a remote server, Exim
3613 restarts its select() call on the socket, thereby resetting its timeout.
3614 This is not a problem when such interrupts are rare. Somebody set up a cron
3615 job to run exiwhat every 2 minutes, which is less than the normal select()
3616 timeout (5 or 10 minutes). This meant that the select() timeout never
3617 kicked in because it was always reset. I have fixed this by comparing the
3618 time when an interrupt arrives with the time at the start of the first call
3619 to select(). If more time than the timeout has elapsed, the interrupt is
3620 treated as a timeout.
3622 5. Some internal re-factoring in preparation for the addition of Sieve
3623 extensions (by MH). In particular, the "personal" test is moved to a
3624 separate function, and given an option for scanning Cc: and Bcc: (which is
3625 not set for Exim filters).
3627 6. When Exim created an email address using the login of the caller as the
3628 local part (e.g. when creating a From: or Sender: header line), it was not
3629 quoting the local part when it contained special characters such as @.
3631 7. Installed new OpenBSD configuration files.
3633 8. Reworded some messages for syntax errors in "and" and "or" conditions to
3634 try to make them clearer.
3636 9. Callout options, other than the timeout value, were being ignored when
3637 verifying sender addresses in header lines. For example, when using
3639 verify = header_sender/callout=no_cache
3641 the cache was (incorrectly) being used.
3643 10. Added a missing instance of ${EXE} to the exim_install script; this affects
3644 only the Cygwin environment.
3646 11. When return_path_on_delivery was set as a log selector, if different remote
3647 addresses in the same message used different return paths and parallel
3648 remote delivery occurred, the wrong values would sometimes be logged.
3649 (Whenever a remote delivery process finished, the return path value from
3650 the most recently started remote delivery process was logged.)
3652 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
3653 lines when AUTH and/or TLS are in use. This is the "received protocol"
3654 field. Exim used to use "asmtp" for authenticated SMTP, without any
3655 indication (in the protocol name) for TLS use. Now it follows the RFC and
3656 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
3657 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
3658 names appear in log lines as well as in Received: header lines.
3660 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
3661 extensions, and comparison tests, and to fix some bugs.
3663 14. Changes to the "personal" filter test:
3665 (1) The test was buggy in that it was just doing the equivalent of
3666 "contains" tests on header lines. For example, if a user's address was
3667 anne@some.where, the "personal" test would incorrectly be true for
3669 To: susanne@some.where
3671 This test is now done by extracting each address from the header in turn,
3672 and checking the entire address. Other tests that are part of "personal"
3673 are now done using regular expressions (for example, to check local parts
3674 of addresses in From: header lines).
3676 (2) The list of non-personal local parts in From: addresses has been
3677 extended to include "listserv", "majordomo", "*-request", and "owner-*",
3678 taken from the Sieve specification recommendations.
3680 (3) If the message contains any header line starting with "List-" it is
3681 treated as non-personal.
3683 (4) The test for "circular" in the Subject: header line has been removed
3684 because it now seems ill-conceived.
3686 15. Minor typos in src/EDITME comments corrected.
3688 16. Installed latest exipick from John Jetmore.
3690 17. If headers_add on a router specified a text string that was too long for
3691 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
3692 of string_sprintf() is now avoided.
3694 18. $message_body_size was not set (it was always zero) when running the DATA
3695 ACL and the local_scan() function.
3697 19. For the "mail" command in an Exim filter, no default was being set for
3698 the once_repeat time, causing a random time value to be used if "once" was
3699 specified. (If the value happened to be <= 0, no repeat happened.) The
3700 default is now 0s, meaning "never repeat". The "vacation" command was OK
3701 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
3702 (I found it when inspecting the code).
3704 20. There is now an overall timeout for performing a callout verification. It
3705 defaults to 4 times the callout timeout, which applies to individual SMTP
3706 commands during the callout. The overall timeout applies when there is more
3707 than one host that can be tried. The timeout is checked before trying the
3708 next host. This prevents very long delays if there are a large number of
3709 hosts and all are timing out (e.g. when the network connections are timing
3710 out). The value of the overall timeout can be changed by specifying an
3711 additional sub-option for "callout", called "maxwait". For example:
3713 verify = sender/callout=5s,maxwait=20s
3715 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
3716 to the end before writing, but this should make it even safer).
3718 22. Exim was forgetting that it had advertised PIPELINING for the second and
3719 subsequent messages on an SMTP connection. It was also not resetting its
3720 memory on STARTTLS and an internal HELO.
3722 23. When Exim logs an SMTP synchronization error within a session, it now
3723 records whether PIPELINING has been advertised or not.
3725 24. Added 3 instances of "(long int)" casts to time_t variables that were being
3726 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
3727 rather than long int.
3729 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
3731 26. Added the never_mail option to autoreply.
3737 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
3738 crash if the getsockname() call failed; this can happen if a connection is
3739 closed very soon after it is established. The problem was simply in the
3740 order in which certain operations were done, causing Exim to try to write
3741 to the SMTP stream before it had set up the file descriptor. The bug has
3742 been fixed by making things happen in the correct order.
3748 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
3749 before the connection was closed, thus losing the rejection response.
3751 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
3752 some early Solaris releases, but causes trouble in current releases where
3753 socklen_t is defined.
3755 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
3758 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
3761 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
3762 that releasing the top of it at the end releases what was used for sub-
3763 expansions (unless the block got too big). However, discard this block if
3764 the first thing is a variable or header, so that we can use its block when
3765 it is dynamic (useful for very large $message_headers, for example).
3767 6. Lookups now cache *every* query, not just the most recent. A new, separate
3768 store pool is used for this. It can be recovered when all lookup caches are
3769 flushed. Lookups now release memory at the end of their result strings.
3770 This has involved some general refactoring of the lookup sources.
3772 7. Some code has been added to the store_xxx() functions to reduce the amount
3773 of flapping under certain conditions.
3775 8. log_incoming_interface used to affect only the <= reception log lines. Now
3776 it causes the local interface and port to be added to several more SMTP log
3777 lines, for example "SMTP connection from", and rejection lines.
3779 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
3781 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
3783 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
3784 could be overwritten at the end of the current message (or the start of a
3785 new message if it was set in a HELO ACL). The value is now preserved for
3786 the duration of the SMTP connection.
3788 12. If a transport had a headers_rewrite setting, and a matching header line
3789 contained an unqualified address, that address was qualified, even if it
3790 did not match any rewriting rules. The underlying bug was that the values
3791 of the flags that permit the existence of unqualified sender and recipient
3792 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
3793 non-local messages, and by -bnq for local messages) were not being
3794 preserved with the message after it was received.
3796 13. When Exim was logging an SMTP synchronization error, it could sometimes log
3797 "next input=" as part of the text comprising the host identity instead of
3798 the correct text. The code was using the same buffer for two different
3799 strings. However, depending on which order the printing function evaluated
3800 its arguments, the bug did not always show up. Under Linux, for example, my
3801 test suite worked just fine.
3803 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
3804 doesn't work with some older versions of Perl. It has been changed to "my",
3805 which in any case is probably the better facility to use.
3807 15. A really picky compiler found some instances of statements for creating
3808 error messages that either had too many or two few arguments for the format
3811 16. The size of the buffer for calls to the DNS resolver has been increased
3812 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
3813 for addresses that have a lot of PTR records. This alleviates a problem; it
3814 does not fully solve it.
3816 17. A dnsdb lookup for PTR records that receives more data than will fit in the
3817 buffer now truncates the list and logs the incident, which is the same
3818 action as happens when Exim is looking up a host name and its aliases.
3819 Previously in this situation something unpredictable would happen;
3820 sometimes it was "internal error: store_reset failed".
3822 18. If a server dropped the connection unexpectedly when an Exim client was
3823 using GnuTLS and trying to read a response, the client delivery process
3824 crashed while trying to generate an error log message.
3826 19. If a "warn" verb in an ACL added multiple headers to a message in a single
3827 string, for example:
3829 warn message = H1: something\nH2: something
3831 the text was added as a single header line from Exim's point of view
3832 though it ended up OK in the delivered message. However, searching for the
3833 second and subsequent header lines using $h_h2: did not work. This has been
3834 fixed. Similarly, if a system filter added multiple headers in this way,
3835 the routers could not see them.
3837 20. Expanded the error message when iplsearch is called with an invalid key to
3838 suggest using net-iplsearch in a host list.
3840 21. When running tests using -bh, any delays imposed by "delay" modifiers in
3841 ACLs are no longer actually imposed (and a message to that effect is
3844 22. If a "gecos" field in a passwd entry contained escaped characters, in
3845 particular, if it contained a \" sequence, Exim got it wrong when building
3846 a From: or a Sender: header from that name. A second bug also caused
3847 incorrect handling when an unquoted " was present following a character
3848 that needed quoting.
3850 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
3851 was not being matched caselessly.
3853 24. Arranged for all hyphens in the exim.8 source to be escaped with
3856 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
3857 properly. Recipient callout cache records were still being keyed to include
3858 the sender, even when use_sender was set false. This led to far more
3859 callouts that were necessary. The sender is no longer included in the key
3860 when use_sender is false.
3862 26. Added "control = submission" modifier to ACLs.
3864 27. Added the ${base62d: operator to decode base 62 numbers.
3866 28. dnsdb lookups can now access SRV records.
3868 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
3869 the configuration file.
3871 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
3872 (-v) mode. This makes the output for a verbose queue run more intelligible.
3874 31. Added a use_postmaster feature to recipient callouts.
3876 32. Added the $body_zerocount variable, containing the number of binary zero
3877 bytes in the message body.
3879 33. The time of last modification of the "new" subdirectory is now used as the
3880 "mailbox time last read" when there is a quota error for a maildir
3883 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
3885 35. Added +ignore_unknown as a special item in host lists.
3887 36. Code for decoding IPv6 addresses in host lists is now included, even if
3888 IPv6 support is not being compiled. This fixes a bug in which an IPv6
3889 address was recognized as an IP address, but was then not correctly decoded
3890 into binary, causing unexpected and incorrect effects when compared with
3897 1. Very minor rewording of debugging text in manualroute to say "list of
3898 hosts" instead of "hostlist".
3900 2. If verify=header_syntax was set, and a header line with an unqualified
3901 address (no domain) and a large number of spaces between the end of the
3902 name and the colon was received, the reception process suffered a buffer
3903 overflow, and (when I tested it) crashed. This was caused by some obsolete
3904 code that should have been removed. The fix is to remove it!
3906 3. When running in the test harness, delay a bit after writing a bounce
3907 message to get a bit more predictability in the log output.
3909 4. Added a call to search_tidyup() just before forking a reception process. In
3910 theory, someone could use a lookup in the expansion of smtp_accept_max_
3911 per_host which, without the tidyup, could leave open a database connection.
3913 5. Added the variables $recipient_data and $sender_data which get set from a
3914 lookup success in an ACL "recipients" or "senders" condition, or a router
3915 "senders" option, similar to $domain_data and $local_part_data.
3917 6. Moved the writing of debug_print from before to after the "senders" test
3920 7. Change 4.31/66 (moving the time when the Received: is generated) caused
3921 problems for message scanning, either using a data ACL, or using
3922 local_scan() because the Received: header was not generated till after they
3923 were called (in order to set the time as the time of reception completion).
3924 I have revised the way this works. The header is now generated after the
3925 body is received, but before the ACL or local_scan() are called. After they
3926 are run, the timestamp in the header is updated.
3932 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
3933 before starting a queue runner without re-exec. This happened only when
3934 deliver_drop_privilege was set or when the Exim user was set to root. The
3935 effect of the bug was that timeouts during subsequent deliveries caused
3936 crashes instead of being properly handled. The handler is now left at its
3937 default (and expected) setting.
3939 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
3940 message, again when deliver_drop_privilege is set or Exim is run as root.
3941 The bug described in (1) was not present in this case, but the tidying up
3942 of the other signals was missing. I have made the two cases consistent.
3944 3. The ignore_target_hosts setting on a manualroute router was being ignored
3945 for hosts that were looked up using the /MX notation.
3947 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
3950 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
3951 operated on the sender address. After changing the $sender_address to <>
3952 for the sender address verify, Exim was re-instated it as the original
3953 (before rewriting) address, but remembering that it had rewritten it, so it
3954 wasn't rewriting it again. This bug also had the effect of breaking the
3955 sender address verification caching when the sender address was rewritten.
3957 6. The ignore_target_hosts option was being ignored by the ipliteral router.
3958 This has been changed so that if the ip literal address matches
3959 ignore_target_hosts, the router declines.
3961 7. Added expansion conditions match_domain, match_address, and match_local_
3962 part (NOT match_host).
3964 8. The placeholder for the Received: header didn't have a length field set.
3966 9. Added code to Exim itself and to exim_lock to test for a specific race
3967 condition that could lead to file corruption when using MBX delivery. The
3968 issue is with the lockfile that is created in /tmp. If this file is removed
3969 after a process has opened it but before that process has acquired a lock,
3970 there is the potential for a second process to recreate the file and also
3971 acquire a lock. This could lead to two Exim processes writing to the file
3972 at the same time. The added code performs the same test as UW imapd; it
3973 checks after acquiring the lock that its file descriptor still refers to
3974 the same named file.
3976 10. The buffer for building added header lines was of fixed size, 8192 bytes.
3977 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
3980 11. Added the smtp_active_hostname option. If used, this will typically be made
3981 to depend on the incoming interface address. Because $interface_address is
3982 not set up until the daemon has forked a reception process, error responses
3983 that can happen earlier (such as "too many connections") no longer contain
3986 12. If an expansion in a condition on a "warn" statement fails because a lookup
3987 defers, the "warn" statement is abandoned, and the next ACL statement is
3988 processed. Previously this caused the whole ACL to be aborted.
3990 13. Added the iplsearch lookup type.
3992 14. Added ident_timeout as a log selector.
3994 15. Added tls_certificate_verified as a log selector.
3996 16. Added a global option tls_require_ciphers (compare the smtp transport
3997 option of the same name). This controls incoming TLS connections.
3999 17. I finally figured out how to make tls_require_ciphers do a similar thing
4000 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
4001 before starting the TLS session.
4003 18. Tabs are now shown as \t in -bP output.
4005 19. If the log selector return_path_on_delivery was set, Exim crashed when
4006 bouncing a message because it had too many Received: header lines.
4008 20. If two routers both had headers_remove settings, and the first one included
4009 a superfluous trailing colon, the final name in the first list and the
4010 first name in the second list were incorrectly joined into one item (with a
4011 colon in the middle).
4017 1. Added -C and -D options to the exinext utility, mainly to make it easier
4018 to include in the automated testing, but these could be helpful when
4019 multiple configurations are in use.
4021 2. The exinext utility was not formatting the output nicely when there was
4022 an alternate port involved in the retry record key, nor when there was a
4023 message id as well (for retries that were specific to a specific message
4024 and a specific host). It was also confused by IPv6 addresses, because of
4025 the additional colons they contain. I have fixed the IPv4 problem, and
4026 patched it up to do a reasonable job for IPv6.
4028 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
4029 delivery, the log line now contains "pipelined" if PIPELINING was used.
4031 4. An SMTP transport process used to panic and die if the bind() call to set
4032 an explicit outgoing interface failed. This has been changed; it is now
4033 treated in the same way as a connect() failure.
4035 5. A reference to $sender_host_name in the part of a conditional expansion
4036 that was being skipped was still causing a DNS lookup. This no longer
4039 6. The def: expansion condition was not recognizing references to header lines
4040 that used bh_ and bheader_.
4042 7. Added the _cache feature to named lists.
4044 8. The code for checking quota_filecount in the appendfile transport was
4045 allowing one more file than it should have been.
4047 9. For compatibility with Sendmail, the command line option
4055 and sets the incoming protocol and host name (for trusted callers). The
4056 host name and its colon can be omitted when only the protocol is to be set.
4057 Note the Exim already has two private options, -pd and -ps, that refer to
4058 embedded Perl. It is therefore impossible to set a protocol value of "d" or
4059 "s", but I don't think that's a major issue.
4061 10. A number of refactoring changes to the code, none of which should affect
4064 (a) The number of logging options was getting close to filling up the
4065 32-bit word that was used as a bit map. I have split them into two classes:
4066 those that are passed in the argument to log_write(), and those that are
4067 only ever tested independently outside of that function. These are now in
4068 separate 32-bit words, so there is plenty of room for expansion again.
4069 There is no change in the user interface or the logging behaviour.
4071 (b) When building, for example, log lines, the code previously used a
4072 macro that called string_cat() twice, in order to add two strings. This is
4073 not really sufficiently general. Furthermore, there was one instance where
4074 it was actually wrong because one of the argument was used twice, and in
4075 one call a function was used. (As it happened, calling the function twice
4076 did not affect the overall behaviour.) The macro has been replaced by a
4077 function that can join an arbitrary number of extra strings onto a growing
4080 (c) The code for expansion conditions now uses a table and a binary chop
4081 instead of a serial search (which was left over from when there were very
4082 few conditions). Also, it now recognizes conditions like "pam" even when
4083 the relevant support is not compiled in: a suitably worded error message is
4084 given if an attempt is made to use such a condition.
4086 11. Added ${time_interval:xxxxx}.
4088 12. A bug was causing one of the ddress fields not to be passed back correctly
4089 from remote delivery subprocesses. The field in question was not being
4090 subsequently used, so this caused to problems in practice.
4092 13. Added new log selectors queue_time and deliver_time.
4094 14. Might have fixed a bug in maildirsizefile handling that threw up
4095 "unexpected character" debug warnings, and recalculated the data
4096 unnecessarily. In any case, I expanded the warning message to give more
4099 15. Added the message "Restricted characters in address" to the statements in
4100 the default ACL that block characters like @ and % in local parts.
4102 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
4103 Three changes have been made:
4105 (a) There was a serious bug; a negative response to MAIL caused the whole
4106 recipient domain to be cached as invalid, thereby blocking all messages
4107 to all local parts at the same domain, from all senders. This bug has
4108 been fixed. The domain is no longer cached after a negative response to
4109 MAIL if the sender used is not empty.
4111 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
4114 (c) A new callout option, "use_sender" has been added for people who want
4115 the modified behaviour.
4121 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
4124 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
4125 indeed breaks things for older releases.
4127 3. Added additional logging to the case where there is a problem reading data
4128 from a filter that is running in a subprocess using a pipe, in order to
4129 try to track down a specific problem.
4131 4. Testing facility fudge: when running in the test harness and attempting
4132 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
4133 getting "No route to host". Convert this to a timeout.
4135 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
4138 6. Some OS don't have socklen_t but use size_t instead. This affects the
4139 fifth argument of getsockopt() amongst other things. This is now
4140 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
4141 can be set for individual OS. I have set it for SunOS5, OSF1, and
4142 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
4143 some earlier ones do not.
4145 7. Change 4.30/15 was not doing the test caselessly.
4147 8. The standard form for an IPv6 address literal was being rejected by address
4148 parsing in, for example, MAIL and RCPT commands. An example of this kind of
4149 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
4150 this, as well as the form without the "IPv6" on the front (but only when
4151 address literals are enabled, of course).
4153 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
4155 10. Exim crashed if a message with an empty sender address specified by -f
4156 encountered a router with an errors_to setting. This could be provoked only
4157 by a command such as
4161 where an empty string was supplied; "<>" did not hit this bug.
4163 11. Installed PCRE release 4.5.
4165 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
4166 remained set. It is now erased.
4168 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
4169 times from message ids (which are base 36 rather than the normal 62).
4171 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
4172 were being counted as actual protocol errors, and logged if the log
4173 selector +smtp_protocol_error was set. One cannot be perfect in this test,
4174 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
4175 and DATA following a set of rejected RCPTs do not count as protocol errors.
4176 In other words, Exim assumes they were pipelined, though this may not
4177 actually be the case. Of course, in all cases the client gets an
4178 appropriate error code.
4180 15. If a lookup fails in an ACL condition, a message about the failure may
4181 be available; it is used if testing the ACL cannot continue, because most
4182 such messages specify what the cause of the deferral is. However, some
4183 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
4184 that caused an old message to be retained and used if a later statement
4185 caused a defer, replacing the real cause of the deferral.
4187 16. If an IP address had so many PTR records that the DNS lookup buffer
4188 was not large enough to hold them, Exim could crash while trying to process
4189 the truncated data. It now detects and logs this case.
4191 17. Further to 4.21/58, another change has been made: if (and only if) the
4192 first line of a message (the first header line) ends with CRLF, a bare LF
4193 in a subsequent header line has a space inserted after it, so as not to
4194 terminate the header.
4196 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
4197 unnecessarily, used atoi() instead of strtol(), and didn't check the
4198 termination when getting file sizes from file names by regex.
4200 19. Completely re-implemented the support for maildirsize files, in the light
4201 of a number of problems with the previous contributed implementation
4202 (4.30/29). In particular:
4204 . If the quota is zero, the maildirsize file is maintained, but no quota is
4207 . If the maildir directory does not exist, it is created before any attempt
4208 to write a maildirsize file.
4210 . The quota value in the file is just a cache; if the quota is changed in
4211 the transport, the new value overrides.
4213 . A regular expression is available for excluding directories from the
4216 20. The autoreply transport checks the characters in options that define the
4217 message's headers; it allows continued headers, but it was checking with
4218 isspace() after an embedded newline instead of explicitly looking for a
4221 21. If all the "regular" hosts to which an address was routed had passed their
4222 expiry times, and had not reached their retry times, the address was
4223 bounced, even if fallback hosts were defined. Now Exim should go on to try
4226 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
4227 equivalent code in the SMTP transport. Some hosts send humungous responses
4228 to HELO/EHLO, more than 1024 it seems.
4230 23. Refactoring: code in filter.c used (void *) for "any old type" but this
4231 gives compiler warnings in some environments. I've now done it "properly",
4234 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
4235 (because of problems with the built-in one) was declared to return uschar *
4236 instead of char *, causing compiler failure.
4238 25. Fixed a file descriptor leak when processing alias/forward files.
4240 26. Fixed a minor format string issue in dbfn.c.
4242 27. Typo in exim.c: ("dmbnz" for "dbmnz").
4244 28. If a filter file refered to $h_xxx or $message_headers, and the headers
4245 contained RFC 2047 "words", Exim's memory could, under certain conditions,
4248 29. When a sender address is verified, it is cached, to save repeating the test
4249 when there is more than one recipient in a message. However, when the
4250 verification involves a callout, it is possible for different callout
4251 options to be set for different recipients. It is too complicated to keep
4252 track of this in the cache, so now Exim always runs a verification when a
4253 callout is required, relying on the callout cache for the optimization.
4254 The overhead is duplication of the address routing, but this should not be
4257 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
4258 to be verified with callout=postmaster, and the main callout worked but the
4259 postmaster check failed, the verification correctly failed. However, if a
4260 subsequent RCPT command asked for sender verification *without* the
4261 postmaster check, incorrect caching caused this verification also to fail,
4264 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
4265 it was not caching the DNS options (qualify_single, search_parents) that
4266 were used when the lookup failed. A subsequent lookup with different
4267 options therefore always gave the same answer, though there were cases
4268 where it should not have. (Example: a "domains = !$mx_any" option on a
4269 dnslookup router: the "domains" option is always processed without any
4270 widening, but the router might have qualify_single set.) Now Exim uses the
4271 cached value only when the same options are set.
4273 32. Added John Jetmore's "exipick" utility to the distribution.
4275 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
4276 than a timeout (e.g. a certificate is required, and is not provided), an
4277 Exim server now closes the connection immediately. Previously it waited for
4278 the client to close - but if the client is SSL, it seems that they each
4279 wait for each other, leading to a delay before one of them times out.
4281 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
4282 maintained 0.8.x compatibility because I don't think many are using it, and
4283 it is clearly obsolete.
4285 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
4288 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
4289 client certificate was expired. A simple patch fixes this, though I don't
4290 understand the full logic of why the verify callback is called multiple
4293 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
4294 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
4295 which causes problems with some clients (such as the Certicom SSL Plus
4296 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
4297 disables the coutermeasure allowing Eudora to connect."
4299 38. Exim was not checking that a write() to a log file succeeded. This could
4300 lead to Bad Things if a log got too big, in particular if it hit a file
4301 size limit. Exim now panics and dies if it cannot write to a log file, just
4302 as it does if it cannot open a log file.
4304 39. Modified OS/Makefile-Linux so that it now contains
4306 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
4308 The two -D definitions ensure that Exim is compiled with large file
4309 support, which makes it possible to handle log files that are bigger than
4312 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
4313 instance) a domain was checked against a named list that involved a lookup,
4314 causing $domain_data to be set, then another domain was checked against the
4315 same list, then the first domain was re-checked, the value of $domain_data
4316 after the final check could be wrong. In particular, if the second check
4317 failed, it could be set empty. This bug probably also applied to
4320 41. The strip_trailing_dot option was not being applied to the address given
4321 with the -f command-line option.
4323 42. The code for reading a message's header from the spool was incrementing
4324 $received_count, but never initializing it. This meant that the value was
4325 incorrect (doubled) while delivering a message in the same process in which
4326 it was received. In the most common configuration of Exim, this never
4327 happens - a fresh exec is done - but it can happen when
4328 deliver_drop_privilege is set.
4330 43. When Exim logs an SMTP synchronization error - client data sent too soon -
4331 it now includes up to 150 characters of the unexpected data in the log
4334 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
4335 and building data strings. The size of both of these buffers was 10 000
4336 bytes - far larger than anybody would *ever* want, thought I. Needless to
4337 say, somebody hit the limit. I have increased the maximum line length to
4338 20 000 and the maximum data length of concatenated lines to 100 000. I have
4339 also fixed two bugs, because there was no checking on these buffers. Tsk,
4340 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
4341 buffer is too small.
4343 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
4344 lsearch lookups. Now it does.
4346 46. When parsing a route_list item in a manualroute router, a fixed-length
4347 buffer was used for the list of hosts. I made this 1024 bytes long,
4348 thinking that nobody would ever have a list of hosts that long. Wrong.
4349 Somebody had a whole pile of complicated expansion conditions, and the
4350 string was silently truncated, leading to an expansion error. It turns out
4351 that it is easier to change to an unlimited length (owing to other changes
4352 that have happened since this code was originally written) than to build
4353 structure for giving a limitation error. The length of the item that
4354 expands into the list of hosts is now unlimited.
4356 47. The lsearch lookup could not handle data where the length of text line was
4357 more than 4095 characters. Such lines were truncated, leading to shortened
4358 data being returned. It should now handle lines of any length.
4360 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
4361 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
4364 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
4365 replacing strings such as BIN_DIRECTORY in a source file: the replacement
4366 no longer happens in comment lines. A list of replacements is now placed
4367 at the head of all of the source files, except those whose only change is
4368 to replace PERL_COMMAND in the very first #! line.
4370 50. Replaced the slow insertion sort in queue.c, for sorting the list of
4371 messages on the queue, with a bottom-up merge sort, using code contributed
4372 by Michael Haardt. This should make operations like -bp somewhat faster on
4373 large queues. It won't affect queue runners, except when queue_run_in_order
4376 51. Installed eximstats 1.31 in the distribution.
4378 52. Added support for SRV lookups to the dnslookup router.
4380 53. If an ACL referred to $message_body or $message_body_end, the value was not
4381 reset for any messages that followed in the same SMTP session.
4383 54. The store-handling optimization for building very long strings was not
4384 differentiating between the different store pools. I don't think this
4385 actually made any difference in practice, but I've tidied it.
4387 55. While running the routers to verify a sender address, $sender_address
4388 was still set to the sender address. This is wrong, because when routing to
4389 send a bounce to the sender, it would be empty. Therefore, I have changed
4390 it so that, while verifying a sender address, $sender_address is set to <>.
4391 (There is no change to what happens when verifying a recipient address.)
4393 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
4394 target A or AAAA records (if not already returned) without resetting the
4395 qualify_single or search_parents options of the DNS resolver. These are
4396 inappropriate in this case because the targets of MX and SRV records must
4397 be FQDNs. A broken DNS record could cause trouble if it happened to have a
4398 target that, when qualified, matched something in the local domain. These
4399 two options are now turned off when doing these lookups.
4401 57. It seems that at least some releases of Reiserfs (which does not have the
4402 concept of a fixed number of inodes) returns zero and not -1 for the
4403 number of available inodes. This interacted badly with check_spool_inodes,
4404 which assumed that -1 was the "no such thing" setting. What I have done is
4405 to check that the total number of inodes is greater than zero before doing
4406 the test of how many are available.
4408 58. When a "warn" ACL statement has a log_message modifier, the message is
4409 remembered, and not repeated. This is to avoid a lot of repetition when a
4410 message has many recipients that cause the same warning to be written.
4411 Howewer, Exim was preserving the list of already written lines for an
4412 entire SMTP session, which doesn't seem right. The memory is now reset if a
4413 new message is started.
4415 59. The "rewrite" debugging flag was not showing the result of rewriting in the
4416 debugging output unless log_rewrite was also set.
4418 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
4419 of (int)(handle) when we know that handle contains (void *)(-1).
4421 61. The Exim daemon panic-logs an error return when it closes the incoming
4422 connection. However "connection reset by peer" seems to be common, and
4423 isn't really an error worthy of noting specially, so that particular error
4426 62. When Exim is trying to find all the local interfaces, it used to panic and
4427 die if the ioctl to get the interface flags failed. However, it seems that
4428 on at least one OS (Solaris 9) it is possible to have an interface that is
4429 included in the list of interfaces, but for which you get a failure error
4430 for this call. This happens when the interface is not "plumbed" into a
4431 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
4432 failure of the "get flags" call assumes that the interface is down.
4434 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
4435 makes life easier for people who are doing arithmetic on fields extracted
4436 from dates, where you often get leading zeros that should not be
4437 interpreted as octal.
4439 64. Added qualify_domain to the redirect router, to override the global
4442 65. If a pathologically long header line contained very many addresses (the
4443 report of this problem mentioned 10 000) and each of them was rewritten,
4444 Exim could use up a very large amount of memory. (It kept on making new
4445 copies of the header line as it rewrote, and never released the old ones.)
4446 At the expense of a bit more processing, the header rewriting function has
4447 been changed so that it no longer eats memory in this way.
4449 66. The generation of the Received: header has been moved from the time that a
4450 message starts to be received, to the time that it finishes. The timestamp
4451 in the Received: header should now be very close to that of the <= log
4452 line. There are two side-effects of this change:
4454 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
4455 logged header lines no longer include the local Received: line, because
4456 it has not yet been created. The same applies to a copy of the message
4457 that is returned to a non-SMTP sender when a message is rejected.
4459 (b) When a filter file is tested using -bf, no additional Received: header
4460 is added to the test message. After some thought, I decided that this
4463 This change does not affect the value of $received_for. It is still set
4464 after address rewriting, but before local_scan() is called.
4466 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
4468 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
4469 gave an unhelpful panic error message, and a defer error. I have managed to
4470 change this behaviour so that it now rejects any supplied certificate,
4471 which seems right, as the list of acceptable certificates is empty.
4473 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
4474 gave an unhelpful defer error. I have not managed to make this reject any
4475 supplied certificates, but the error message it gives is "no certificate
4476 supplied", which is not helpful.
4478 70. exigrep's output now also includes lines that are not associated with any
4479 message, but which match the given pattern. Implemented by a patch from
4480 Martin Sluka, which also tidied up the Perl a bit.
4482 71. Recipient callout verification, like sender verification, was using <> in
4483 the MAIL FROM command. This isn't really the right thing, since the actual
4484 sender may affect whether the remote host accepts the recipient or not. I
4485 have changed it to use the actual sender in the callout; this means that
4486 the cache record is now keyed on a recipient/sender pair, not just the
4487 recipient address. There doesn't seem to be a real danger of callout loops,
4488 since a callout by the remote host to check the sender would use <>.
4489 [SEE ABOVE: changed after hitting problems.]
4491 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
4492 temporary errors. However, in the case of such a code being given after
4493 the end of a data transmission (i.e. after ".") Exim was failing to write
4494 a retry record for the message. (Yes, there was some broken host that was
4495 actually sending 8xx at this point.)
4497 73. An unknown lookup type in a host list could cause Exim to panic-die when
4498 the list was checked. (An example that provoked this was putting <; in the
4499 middle of a list instead of at the start.) If this happened during a DATA
4500 ACL check, a -D file could be left lying around. This kind of configuration
4501 error no longer causes Exim to die; instead it causes a defer errror. The
4502 incident is still logged to the main and panic logs.
4504 74. Buglet left over from Exim 3 conversion. The message "too many messages
4505 in one connection" was written to the rejectlog but not the mainlog, except
4506 when address rewriting (yes!) was being logged.
4508 75. Added write_rejectlog option.
4510 76. When a system filter was run not as root (that is, when system_filter_user
4511 was set), the values of the $n variables were not being returned to the
4512 main process; thus, they were not subsequently available in the $sn
4515 77. Added +return_path_on_delivery log selector.
4517 78. A connection timeout was being treated differently from recipients deferred
4518 when testing hosts_max_try with a message that was older than the host's
4519 retry timeout. (The host should not be counted, thus allowing all hosts to
4520 be tried at least once before bouncing.) This may have been the cause of an
4521 occasionally reported bug whereby a message would remain on the queue
4522 longer than the retry timeout, but would be bounced if a delivery was
4523 forced. I say "may" because I never totally pinned down the problem;
4524 setting up timeout/retry tests is difficult. See also the next item.
4526 79. The ultimate address timeout was not being applied to errors that involved
4527 a combination of host plus message (for example, a timeout on a MAIL
4528 command). When an address resolved to a number of possible hosts, and they
4529 were not all tried for each delivery (e.g. because of hosts_max_try), a
4530 message could remain on the queue longer than the retry timeout.
4532 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
4535 81. Fixed an obscure SMTP outgoing bug which required at least the following
4536 conditions: (a) there was another message waiting for the same server;
4537 (b) the server returned 5xx to all RCPT commands in the first message so
4538 that the message was not completed; (c) the server dropped the connection
4539 or gave a negative response to the RSET that Exim sends to abort the
4540 transaction. The observed case was a dropped connection after DATA that had
4541 been sent in pipelining mode. That is, the server had advertised PIPELINING
4542 but was not implementing it correctly. The effect of the bug was incorrect
4543 behaviour, such as trying another host, and this could lead to a crash.
4549 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
4550 and daemon.c were passed as pointers to ints; they should have been
4551 pointers to socklen_t variables (which are typically unsigned ints).
4553 2. Some signed/unsigned type warnings in the os.c file for Linux have been
4556 3. Fixed a really odd bug that affected only the testing scheme; patching a
4557 certain fixed string in the binary changed the value of another string that
4558 happened to be identical to the end of the original first string.
4560 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
4561 name", it returns that address as the IP address. On some operating
4562 systems (e.g. Solaris), it also passes back the IP address string as the
4563 "host name". However, on others (e.g. Linux), it passes back an empty
4564 string. Exim wasn't checking for this, and was changing the host name to an
4565 empty string, assuming it had been canonicized.
4567 5. Although rare, it is permitted to have more than one PTR record for a given
4568 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
4569 all the names associated with an address, because they do in Solaris.
4570 However, it seems that they do not in Linux for data that comes from the
4571 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
4572 I found this out when I moved to a new Linux workstation and tried to run
4573 the Exim test suite.
4575 To get round this problem I have changed the code so that it now does its
4576 own call to the DNS to look up PTR records when searching for a host name.
4577 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
4578 addresses that are only in /etc/hosts are still found.
4580 This behaviour is, however, controlled by an option called host_lookup_
4581 order, which defaults to "bydns:byaddr". If people want to use the other
4582 order, or indeed, just use one or the other means of lookup, they can
4583 specify it in this variable.
4585 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
4586 some operating systems, this comes back from gethostbyaddr() as an empty
4587 string, and this is what Exim used to test for. However, it seems that in
4588 other systems, "." is yielded. Exim now tests for this case too.
4590 7. The values of check_spool_space and check_log_space are now held internally
4591 as a number of kilobytes instead of an absolute number of bytes. If a
4592 numbers is specified without 'K' or 'M', it is rounded up to the nearest
4593 kilobyte. This means that much larger values can be stored.
4595 8. Exim monitor: an attempt to get the action menu when not actually pointing
4596 at a message produces an empty menu entitled "No message selected". This
4597 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
4598 no entries in it ("Shell widget menu has zero width and/or height"). So I
4599 have added a single, blank menu entry in this case.
4601 9. Added ${quote_local_part.
4603 10. MIME decoding is now applied to the contents of Subject: header lines when
4606 11. Now that a reference to $sender_host_address automatically causes a reverse
4607 lookup to occur if necessary (4.13/18), there is no need to arrange for a
4608 host lookup before query-style lookups in lists that might use this
4609 variable. This has therefore been abolished, and the "net-" prefix is no
4610 longer necessary for query-style lookups.
4612 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
4613 have been a typo for LFLAGS, so it has been changed.
4615 13. The install script calls Exim with "-C /dev/null" in order to find the
4616 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
4617 to be output. Howeve, since Exim outputs its version number before the
4618 error, it didn't break the script. It just looked ugly. I fixed this by
4619 always allowing "-C /dev/null" if the caller is root.
4621 14. Ignore overlarge ACL variable number when reading spool file - insurance
4622 against a later release with more variables having written the file.
4624 15. The standard form for an IPv6 address literal was being rejected by EHLO.
4625 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
4626 this, as well as the form without the "IPv6" on the front.
4628 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
4629 OS/Makefile-Darwin file.
4631 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
4632 only to LDAP libraries that do not have LDAP_OPT_DEREF.
4634 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
4635 variable. However, some OS use ints for inodes. I've added cast to long int
4636 to get rid of the compiler warning.
4638 19. I had forgotten to lock out "/../" in configuration file names when
4639 ALT_CONFIG_PREFIX was set.
4641 20. Routers used for verification do not need to specify transports. However,
4642 if such a router generated a host list, and callout was configured, Exim
4643 crashed, because it could not find a port number from the (non-existent)
4644 transport. It now assumes port 25 in this circumstance.
4646 21. Added the -t option to exigrep.
4648 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
4649 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
4650 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
4651 them is set, LOOKUP_LSEARCH is forced.
4653 23. "exim -bV" now outputs a list of lookups that are included in the binary.
4655 24. Added sender and host information to the "rejected by local_scan()" log
4656 line; previously there was no indication of these.
4658 25. Added .include_if_exists.
4660 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
4661 Linux. It turns out that not all file systems support this. Apparently some
4662 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
4663 do it.) To cope with this, the error EINVAL, which means that sync-ing is
4664 not supported on the file descriptor, is now ignored when Exim is trying to
4665 sync a directory. This applies only to Linux.
4667 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
4669 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
4670 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
4673 29. Added support for maildirsize files from supplied patch (modified a bit).
4675 30. The use of :fail: followed by an empty string could lead Exim to respond to
4676 sender verification failures with (e.g.):
4678 550 Verification failed for <xxx>
4679 550 Sender verify failed
4681 where the first response line was missing the '-' that indicates it is not
4682 the final line of the response.
4684 31. The loop for finding the name of the user that called Exim had a hardwired
4685 limit of 10; it now uses the value of finduser_retries, which is used for
4686 all other user lookups.
4688 32. Added $received_count variable, available in data and not_smtp ACLs, and at
4691 33. Exim was neglecting to zero errno before one call of strtol() when
4692 expanding a string and expecting an integer value. On some systems this
4693 resulted in spurious "integer overflow" errors. Also, it was casting the
4694 result into an int without checking.
4696 34. Testing for a connection timeout using "timeout_connect" in the retry rules
4697 did not work. The code looks as if it has *never* worked, though it appears
4698 to have been documented since at least releast 1.62. I have made it work.
4700 35. The "timeout_DNS" error in retry rules, also documented since at least
4701 1.62, also never worked. As it isn't clear exactly what this means, and
4702 clearly it isn't a major issue, I have abolished the feature by treating it
4703 as "timeout", and writing a warning to the main and panic logs.
4705 36. The display of retry rules for -brt wasn't always showing the error code
4708 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
4709 timeout_connect_A, timeout_connect_MX.
4711 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
4712 to the empty sender.
4714 39. The daemon was not analysing the content of -oX till after it had closed
4715 stderr and disconnected from the controlling terminal. This meant that any
4716 syntax errors were only noted on the panic log, and the return code from
4717 the command was 0. By re-arranging the code a little, I've made the
4718 decoding happen first, so such errors now appear on stderr, and the return
4719 code is 1. However, the actual setting up of the sockets still happens in
4720 the disconnected process, so errors there are still only recorded on the
4723 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
4724 connections (as happens on some IP stacks) was logged at start up time as
4725 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
4726 it from "IPv6 and IPv4", which means that two separate sockets are being
4729 41. The debug output for gethostbyname2() or getipnodebyname() failures now
4730 says whether AF_INET or AF_INET6 was passed as an argument.
4732 42. Exiwhat output was messed up when time zones were included in log
4735 43. Exiwhat now gives more information about the daemon's listening ports,
4736 and whether -tls-on-connect was used.
4738 44. The "port" option of the smtp transport is now expanded.
4740 45. A "message" modifier in a "warn" statement in a non-message ACL was being
4741 silently ignored. Now an error message is written to the main and panic
4744 46. There's a new ACL modifier called "logwrite" which writes to a log file
4745 as soon as it is encountered.
4747 47. Added $local_user_uid and $local_user_gid at routing time.
4749 48. Exim crashed when trying to verify a sender address that was being
4752 49. Exim was recognizing only a space character after ".include". It now also
4753 recognizes a tab character.
4755 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
4756 extracting the relevant information from the specification. The man page no
4757 longer contains scrambled data for the -d option, and I've added a section
4758 at the front about calling Exim under different names.
4760 51. Added "extra_headers" argument to the "mail" command in filter files.
4762 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
4765 53. Installed eximstats 1.29.
4767 54. Added transport_filter_timeout as a generic transport option.
4769 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
4770 Cc: header lines. This was required by RFC 822, but it not required by RFC
4773 56. Exim used to add From:, Date:, and Message-Id: header lines to any
4774 incoming messages that did not have them. Now it does so only if the
4775 message originates locally, that is, if there is no associated remote host
4776 address. When Resent- header lines are present, this applies to the Resent-
4777 lines rather than the non-Resent- lines.
4779 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
4780 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
4782 58. Messages for configuration errors now include the name of the main
4783 configuration file - useful now that there may be more than one file in a
4784 list (.included file names were always shown).
4786 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
4787 for those rare installations that do not start the daemon as root or run it
4788 setuid root. I've cut out the call to initgroups() if the daemon is not
4791 60. The Exim user and group can now be bound into the binary as text strings
4792 that are looked up at the start of Exim's processing.
4794 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
4796 62. Added $mailstore_basename variable.
4798 63. Installed patch to sieve.c from Michael Haardt.
4800 64. When Exim failed to open the panic log after failing to open the main log,
4801 the original message it was trying to log was written to stderr and debug
4802 output, but if they were not available (the usual case in production), it
4803 was lost. Now it is written to syslog before the two lines that record the
4804 failures to open the logs.
4806 65. Users' Exim filters run in subprocesses under the user's uid. It is
4807 possible for a "deliver" command or an alias in a "personal" command to
4808 provoke an address rewrite. If logging of address rewriting is configured,
4809 this fails because the process is not running as root or exim. There may be
4810 a better way of dealing with this, but for the moment (because 4.30 needs
4811 to be released), I have disabled address rewrite logging when running a
4812 filter in a non-root, non-exim process.
4818 1. The buildconfig auxiliary program wasn't quoting the value set for
4819 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
4820 not defined. This bug was masked in 4.22 by the effect that was fixed in
4823 2. Some messages that were rejected after a message id was allocated were
4824 shown as "incomplete" by exigrep. It no longer does this for messages that
4825 are rejected by local_scan() or the DATA or non-SMTP ACLs.
4827 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
4828 have allow_domain_literals set, the ID did not get logged in the <= line.
4829 Domain literals are now always recognized in Message-ID: header lines.
4831 4. The first argument for a ${extract expansion item is the key name or field
4832 number. Leading and trailing spaces in this item were not being ignored,
4833 causing some misleading effects.
4835 5. When deliver_drop_privilege was set, single queue runner processes started
4836 manually (i.e. by the command "exim -q") or by the daemon (which uses the
4837 same command in the process it spins off) were not dropping privilege.
4839 6. When the daemon running as "exim" started a queue runner, it always
4840 re-executed Exim in the spun-off process. This is a waste of effort when
4841 deliver_drop_privilege is set. The new process now just calls the
4842 queue-runner function directly.
4848 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
4851 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
4852 ignored. Though the use of -oP was forcing the writing of a pid file, it
4853 was always written to the default place.
4855 3. If the message "no IP address found for host xxxx" is generated during
4856 incoming verification, it is now followed by identification of the incoming
4857 connection (so you can more easily find what provoked it).
4859 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
4861 5. Added some features to "harden" Exim a bit more against certain attacks:
4863 (a) There is now a build-time option called FIXED_NEVER_USERS that can
4864 be put in Local/Makefile. This is like the never_users runtime option,
4865 but it cannot be overridden. The default setting is "root".
4867 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
4868 prefix string with which any file named in a -C command line option
4871 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
4872 is retained for -C and -D only if the caller of Exim is root. Without
4873 it, the exim user may also use -C and -D and retain privilege.
4875 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
4876 command line option is disabled.
4878 6. Macro names set by the -D option must start with an upper case letter, just
4879 like macro names defined in the configuration file.
4881 7. Added "dereference=" facility to LDAP.
4883 8. Two instances of the typo "uknown" in the source files are fixed.
4885 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
4886 the Configure-Makefile script screwed up while processing it.
4888 10. Incorporated PCRE 4.4.
4890 11. The SMTP synchronization check was not operating right at the start of an
4891 SMTP session. For example, it could not catch a HELO sent before the client
4892 waited for the greeting. There is now a check for outstanding input at the
4893 point when the greeting is written. Because of the duplex, asynchronous
4894 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
4895 way, but not yet received, when the check is performed.
4897 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
4898 on TCP/IP sockets, because this apparently causes some broken clients to
4901 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
4902 unchanged) from the Cygwin maintainer.
4904 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
4905 was supported instead of testing for mbx. Effectively a typo.
4907 15. The spa authenticator server code was not checking that the input it
4908 received was valid base64.
4910 16. The debug output line for the "set" modifier in ACLs was not showing the
4911 name of the variable that was being set.
4913 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
4915 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
4916 lookup on its own. Something else was needed to trigger the lookup. For
4917 example, a match in host_lookup or the need for a host name in a host list.
4918 Now, if $sender_host_name is referenced and the host name has not yet been
4919 looked up, a lookup is performed. If the lookup fails, the variable remains
4920 empty, and $host_lookup_failed is set to "1".
4922 19. Added "eqi" as a case-independent comparison operator.
4924 20. The saslauthd authentication condition could segfault if neither service
4925 nor realm was specified.
4927 21. If an overflowing value such as "2048M" was set for message_size_limit, the
4928 error message that was logged was misleading, and incoming SMTP
4929 connections were dropped. The message is now more accurate, and temporary
4930 errors are given to SMTP connections.
4932 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
4933 (except RSET) with a 421 error, until QUIT is received. However, it was
4934 failing to send a response to QUIT.
4936 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
4937 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
4938 ACL is now run after the helo_try_verify_hosts code.
4940 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
4941 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
4942 case-independent, so other case variants are also recognized). Apparently
4943 some systems use these upper case variants.
4945 25. If more than two messages were waiting for the same host, and a transport
4946 filter was specified for the transport, Exim sent two messages over the
4947 same TCP/IP connection, and then failed with "socket operation on non-
4948 socket" when it tried to send the third.
4950 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
4952 27. The extern definition of crypt16() in expand.c was not being excluded when
4953 the OS had its own crypt16() function.
4955 28. Added bounce_return_body as a new option, and bounce_return_size_limit
4956 as a preferred synonym for return_size_limit, both as an option and as an
4959 29. Added LIBS=-liconv to OS/Makefile-OSF1.
4961 30. Changed the default configuration ACL to relax the local part checking rule
4962 for addresses that are not in any local domains. For these addresses,
4963 slashes and pipe symbols are allowed within local parts, but the sequence
4964 /../ is explicitly forbidden.
4966 31. SPA server authentication was not clearing the challenge buffer before
4969 32. log_message in a "warn" ACL statement was writing to the reject log as
4970 well as to the main log, which contradicts the documentation and doesn't
4971 seem right (because no rejection is happening). So I have stopped it.
4973 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
4974 However, I am unable to do any testing of this.
4976 34. Fixed an infelicity in the appendfile transport. When checking directories
4977 for a mailbox, to see if any needed to be created, it was accidentally
4978 using path names with one or more superfluous leading slashes; tracing
4979 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
4981 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
4982 contents are added to the log line that is written for every discarded
4983 recipient. (Previously a log_message setting was ignored.)
4985 36. The ${quote: operator now quotes the string if it is empty.
4987 37. The install script runs exim in order to find its version number. If for
4988 some reason other than non-existence or emptiness, which it checks, it
4989 could not run './exim', it was installing it with an empty version number,
4990 i.e. as "exim-". This error state is now caught, and the installation is
4993 38. An argument was missing from the function that creates an error message
4994 when Exim fails to connect to the socket for saslauthd authentication.
4995 This could cause Exim to crash, or give a corrupted message.
4997 39. Added isip, isip4, and isip6 to ${if conditions.
4999 40. The ACL variables $acl_xx are now saved with the message, and can be
5000 accessed later in routers, transports, and filters.
5002 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
5003 strings in the file are not string-expanded.
5005 42. If a MAIL command specified a SIZE value that was too large to fit into an
5006 int variable, the check against message_size_limit failed. Such values are
5007 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
5008 day this will have to be increased, but I don't think I want to be around
5009 when emails are that large.
5016 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
5017 iconv() is not standard in FreeBSD.
5019 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
5020 IPv6 enabled. The observed symptom was a segmentation fault on return from
5021 the function os_common_find_running_interfaces() in src/os.c.
5023 3. In the check_special_case() function in daemon.c I had used "errno" as an
5024 argument name, which causes warnings on some systems. This was basically a
5025 typo, since it was named "eno" in the comments!
5027 4. The code that waits for the clock to tick (at a resolution of some fraction
5028 of a second) so as to ensure message-id uniqueness was always waiting for
5029 at least one whole tick, when it could have waited for less. [This is
5030 almost certainly not relevant at current processor speeds, where it is
5031 unlikely to ever wait at all. But we try to future-proof.]
5033 5. The function that sleeps for a time interval that includes fractions of a
5034 second contained a race. It did not block SIGALRM between setting the
5035 timer, and suspending (a couple of lines later). If the interval was short
5036 and the sigsuspend() was delayed until after it had expired, the suspension
5037 never ended. On busy systems this could lead to processes getting stuck for
5040 6. Some uncommon configurations may cause a lookup to happen in a queue runner
5041 process, before it forks any delivery processes. The open lookup caching
5042 mechanism meant that the open file or database connection was passed into
5043 the delivery process. The problem was that delivery processes always tidy
5044 up cached lookup data. This could cause a problem for the next delivery
5045 process started by the queue runner, because the external queue runner
5046 process does not know about the closure. So the next delivery process
5047 still has data in the lookup cache. In the case of a file lookup, there was
5048 no problem because closing a file descriptor in a subprocess doesn't affect
5049 the parent. However, if the lookup was caching a connection to a database,
5050 the connection was closed, and the second delivery process was likely to
5051 see errors such as "PGSQL: query failed: server closed the connection
5052 unexpectedly". The problem has been fixed by closing all cached lookups
5053 in a queue runner before running a delivery process.
5055 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
5056 seem to have the "const" qualifier which it has on other OS. I've
5059 8. Change 4.21/2 was too strict. It is only if there are two authenticators
5060 *of the same type* (client or server) with the same public name that an
5061 error should be diagnosed.
5063 9. When Exim looked up a host name for an IP address, but failed to find the
5064 original IP address when looking up the host name (a safety check), it
5065 output the message "<ip address> does not match any IP for NULL", which was
5066 confusing, to say the least. The bug was that the host name should have
5067 appeared instead of "NULL".
5069 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
5070 user that is built into the binary, and the -C or -D options is used, root
5071 privilege is dropped before the configuration file is read. In addition,
5072 logging is switched to stderr instead of the normal log files. If the
5073 configuration then re-defines the Exim user, the unprivileged environment
5074 is probably not what is expected, so Exim logs a panic warning message (but
5077 However, if deliver_drop_privilege is set, the unprivileged state may well
5078 be exactly what is intended, so the warning has been cut out in that case,
5079 and Exim is allowed to try to write to its normal log files.
5085 1. smtp_return_error_details was not giving details for temporary sender
5086 or receiver verification errors.
5088 2. Diagnose a configuration error if two authenticators have the same public
5091 3. Exim used not to create the message log file for a message until the first
5092 delivery attempt. This could be confusing when incoming messages were held
5093 for policy or load reasons. The message log file is now created at the time
5094 the message is received, and an initial "Received" line is written to it.
5096 4. The automatically generated man page for command line options had a minor
5097 bug that caused no ill effects; however, a more serious problem was that
5098 the procedure for building the man page automatically didn't always
5099 operate. Consequently, release 4.20 contains an out-of-date version. This
5100 shouldn't happen again.
5102 5. When building Exim with embedded Perl support, the script that builds the
5103 Makefile was calling 'perl' to find its compile-time parameters, ignoring
5104 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
5106 6. The freeze_tell option was not being used for messages that were frozen on
5107 arrival, either by an ACL or by local_scan().
5109 7. Added the smtp_incomplete_transaction log selector.
5111 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
5112 was accepting AUTH without a new EHLO.
5114 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
5115 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
5118 10. Logging of TCP/IP connections (when configured) now happens in the main
5119 daemon process instead of the child process, so that the TCP/IP connection
5120 count is more accurate (but it can never be perfect).
5122 11. The use of "drop" in a nested ACL was not being handled correctly in the
5123 outer ACL. Now, if condition failure induced by the nested "drop" causes
5124 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
5125 or "require"), the connection is dropped.
5127 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
5128 that yield "discard" can now be used with an "accept" or a "discard" verb,
5129 but an error is generated for any others (because I can't see a useful way
5130 to define what should happen).
5132 13. When an ACL is read dynamically from a file (or anywhere else), the lines
5133 are now processed in the same way as lines in the Exim configuration file.
5134 In particular, continuation lines are supported.
5136 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
5138 15. Added -ti meaning -t -i.
5140 16. Check for letters, digits, hyphens, and dots in the names of dnslist
5141 domains, and warn by logging if others are found.
5143 17. At least on BSD, alignment is not guarenteed for the array of ifreq's
5144 returned from GIFCONF when Exim is trying to find the list of interfaces on
5145 a host. The code in os.c has been modified to copy each ifreq to an aligned
5146 structure in all cases.
5148 Also, in some cases, the returned ifreq's were being copied to a 'struct
5149 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
5150 means the last couple of bytes of an IPv6 address could be chopped if the
5151 ifreq contained only a normal sockaddr (14 bytes storage).
5153 18. Named domain lists were not supported in the hosts_treat_as_local option.
5154 An entry such as +xxxx was not recognized, and was treated as a literal
5157 19. Ensure that header lines added by a DATA ACL are included in the reject log
5158 if the ACL subsequently rejects the message.
5160 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
5161 MD5 (which is deprecated).
5163 21. When testing a filter file using -bf, Exim was writing a message when it
5164 took the sender from a "From " line in the message, but it was not doing so
5165 when it took $return_path from a Return-Path: header line. It now does.
5167 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
5168 with a valid header line field name (a series of printing characters
5169 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
5171 23. Changed "disc" in the source to "disk" to conform to the documentation and
5172 the book and for uniformity.
5174 24. Ignore Sendmail's -Ooption=value command line item.
5176 25. When execve() failed while trying to run a command in a pipe transport,
5177 Exim was returning EX_UNAVAILBLE (69) from the subprocess. However, this
5178 could be confused with a return value of 69 from the command itself. This
5179 has been changed to 127, the value the shell returns if it is asked to run
5180 a non-existent command. The wording for the related log line suggests a
5181 non-existent command as the problem.
5183 26. If received_header_text expands to an empty string, do not add a Received:
5184 header line to the message. (Well, it adds a token one on the spool, but
5185 marks it "old" so that it doesn't get used or transmitted.)
5187 27. Installed eximstats 1.28 (addition of -nt option).
5189 28. There was no check for failure on the call to getsockname() in the daemon
5190 code. This can fail if there is a shortage of resources on the system, with
5191 ENOMEM, for example. A temporary error is now given on failure.
5193 29. Contrary to the C standard, it seems that in some environments, the
5194 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
5195 program. Exim now does this explicitly; it affects the formatting of
5196 timestamps using strftime().
5198 30. If exiqsumm was given junk data, it threw up some uninitialized variable
5199 complaints. I've now initialized all the variables, to avoid this.
5201 32. Header lines added by a system filter were not being "seen" during
5202 transport-time rewrites.
5204 33. The info_callback() function passed to OpenSSL is set up with type void
5205 (*)(SSL *, int, int), as described somewhere. However, when calling the
5206 function (actually a macro) that sets it up, the type void(*)() is
5207 expected. I've put in a cast to prevent warnings from picky compilers.
5209 34. If a DNS black list lookup found a CNAME record, but there were no A
5210 records associated with the domain it pointed at, Exim crashed.
5212 35. If a DNS black list lookup returned more than one A record, Exim ignored
5213 all but the first. It now scans all returned addresses if a particular IP
5214 value is being sought. In this situation, the contents of the
5215 $dnslist_value variable are a list of all the addresses, separated by a
5218 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
5219 to accept a host name and all its aliases if the forward lookup for any of
5220 them yielded the IP address of the incoming connection. Now it accepts only
5221 those names whose forward lookup yields the correct IP address. Any other
5222 names are discarded. This closes a loophole whereby a rogue DNS
5223 administrator could create reverse DNS records to break through a
5224 wildcarded host restriction in an ACL.
5226 37. If a user filter or a system filter that ran in a subprocess used any of
5227 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
5228 the wrong values were passed to the pipe command ($thisaddress had the
5229 value of $0, $0 had the value of $1, etc). This bug was introduced by
5230 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
5232 38. Improved the line breaking for long SMTP error messages from ACLs.
5233 Previously, if there was no break point between 40 and 75 characters, Exim
5234 left the rest of the message alone. Two changes have been made: (a) I've
5235 reduced the minimum length to 35 characters; (b) if it can't find a break
5236 point between 35 and 75 characters, it looks ahead and uses the first one
5237 that it finds. This may give the occasional overlong line, but at least the
5238 remaining text gets split now.
5240 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
5241 file descriptors might be low, and that setting 1000 would always raise it.
5242 It turns out that in some environments, the limit is already over 1000 and
5243 that lowering it causes trouble. So now Exim takes care not to decrease it.
5245 40. When delivering a message, the value of $return_path is set to $sender_
5246 address at the start of routing (routers may change the value). By an
5247 oversight, this default was not being set up when an address was tested by
5248 -bt or -bv, which affected the outcome if any router or filter referred to
5251 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
5252 log only when "message" or "log_message" are set. However, if one of the
5253 conditions was an address verification, or a call to a nested ACL, the
5254 messages generated by the underlying test were being passed through. This
5255 no longer happens. The underlying message is available in $acl_verify_
5256 message for both "message" and "log_message" expansions, so it can be
5257 passed through if needed.
5259 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
5260 new expansion $bh_ to give the encoded byte string without charset
5261 translation. Translation happens only if iconv() is available; HAVE_ICONV
5262 indicates this at build time. HEADERS_CHARSET gives the charset to
5263 translate to; headers_charset can change it in the configuration, and
5264 "headers charset" can change it in an individual filter file.
5266 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
5267 that creates RFC 2047 encoded "words" labels them as that charset instead
5268 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
5269 expansion operator; (ii) when Exim creates a From: line for a local
5270 message; (iii) when a header line is rewritten to include a "phrase" part.
5272 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
5273 buggy, causing it to skip the first lines of messages whose message ID
5274 ended in 'D'. This would not have bitten before Exim release 4.14, because
5275 message IDs were unlikely to end in 'D' before then. The effect was to have
5276 incorrect size information for certain domains.
5278 45. #include "config.h" was missing at the start of the crypt16.c module. This
5279 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
5282 46. If there was a timeout during a "random" callout check, Exim treated it as
5283 a failure of the random address, and carried on sending RSET and the real
5284 address. If the delay was just some slowness somewhere, the response to the
5285 original RCPT would be taken as a response to RSET and so on, causing
5286 mayhem of various kinds.
5288 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
5289 when I implemented it. It didn't allow for the fact that some option values
5290 may legitimatetly be negative (e.g. size_addition), and it didn't even do
5291 the right test for positive values.
5293 48. Domain names in DNS records are case-independent. Exim always looks them up
5294 in lower case. Some resolvers return domain names in exactly the case they
5295 appear in the zone file, that is, they may contain uppercase letters. Not
5296 all resolvers do this - some return always lower case. Exim was treating a
5297 change of case by a resolver as a change of domain, similar to a widening
5298 of a domain abbreviation. This triggered its re-routing code and so it was
5299 trying to route what was effectively the same domain again. This normally
5300 caused routing to fail (because the router wouldn't handle the domain
5301 twice). Now Exim checks for this case specially, and just changes the
5302 casing of the domain that it ultimately uses when it transmits the message
5305 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
5308 50. If a filter generated a file delivery with a non-absolute name (possible if
5309 no home directory exists for the router), the forbid_file option was not
5312 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
5313 the existing equality matching.
5315 52. Exim was using ints instead of ino_t variables in some places where it was
5316 dealing with inode numbers.
5318 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
5319 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
5320 called TMPDIR, and if it finds it is different, it changes its value.
5322 54. The smtp_printf() function is now made available to local_scan() so
5323 additional output lines can be written before returning. There is also an
5324 smtp_fflush() function to enable the detection of a dropped connection.
5325 The variables smtp_input and smtp_batched_input are exported to
5328 55. Changed the default runtime configuration: the message "Unknown user"
5329 has been removed from the ACL, and instead placed on the localuser router,
5330 using the cannot_route_message feature. This means that any verification
5331 failures that generate their own messages won't get overridden. Similarly,
5332 the "Unrouteable address" message that was in the ACL for unverifiable
5333 relay addresses has also been removed.
5335 56. Added hosts_avoid_esmtp to the smtp transport.
5337 57. The exicyclog script was not checking for the esoteric option
5338 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
5339 will work only if exicyclog is run under the appropriate euid.
5341 58. Following a discussion on the list, the rules by which Exim recognises line
5342 endings on incoming messages have been changed. The -dropcr and drop_cr
5343 options are now no-ops, retained only for backwards compatibility. The
5344 following line terminators are recognized: LF CRLF CR. However, special
5345 processing applies to CR:
5347 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
5348 nor a local message in the state where . is a terminator.
5350 (ii) If a bare CR is encountered in a header line, an extra space is added
5351 after the line terminator so as not to end the header. The reasoning
5352 behind this is that bare CRs in header lines are most likely either
5353 to be mistakes, or people trying to play silly games.
5355 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
5356 was being incorrectly given as 18 bytes larger than it should have been.
5357 This is a VOB (very old bug).
5359 60. This may never have affected anything current, but just in case it has:
5360 When the local host is found other than at the start of a list of hosts,
5361 the local host, those with the same MX, and any that follow, are discarded.
5362 When the list in question was part of a longer list of hosts, the following
5363 hosts (not currently being processed) were also being discarded. This no
5364 longer happens. I'm not sure if this situation could ever has previously
5367 61. Added the "/MX" feature to lists of hosts in the manualroute and query
5370 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
5371 header. This is something that is recommended in a new Internet Draft, and
5372 is something that is documented as being done by Sendmail. There are two
5373 possible values. For messages generated by the autoreply transport, Exim
5376 Auto-Submitted: auto-replied
5378 whereas for all other generated messages (e.g. bounces) it adds
5380 Auto-Submitted: auto-generated
5382 63. The "personal" condition in filters now includes a test for the
5383 Auto-Submitted: header. If it contains the string "auto-" the message it
5384 not considered personal.
5386 64. Added rcpt_include_affixes as a generic transport option.
5388 65. Added queue_only_override (default true).
5390 66. Added the syslog_duplication option.
5392 67. If what should have been the first header line of a message consisted of
5393 a space followed by a colon, Exim was mis-interpreting it as a header line.
5394 It isn't of course - it is syntactically invalid and should therefore be
5395 treated as the start of the message body. The misbehaviour could have
5396 caused a number of strange effects, including loss of data in subsequent
5397 header lines, and spool format errors.
5399 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
5400 client host had authenticated. This control can now be exercised by an ACL
5401 for more flexibility.
5403 69. By default, callouts do not happen when testing with -bh. There is now a
5404 variant, -bhc, which does actually run the callout code, including
5405 consulting and updating the callout cache.
5407 70. Added support for saslauthd authentication, courtesy of Alexander
5410 71. If statvfs() failed on the spool or log directories while checking their
5411 size for availability, Exim confusingly gave the error "space shortage".
5412 Furthermore, in debugging mode it crashed with a floating point exception.
5413 These checks are done if check_{spool,log}_{space,inodes} are set, and when
5414 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
5415 serious problem, Exim now writes to the main and panic logs when this
5416 happens, with details of the failure. It then refuses to accept the
5417 incoming message, giving the message "spool directory problem" or "log
5418 directory problem" with a 421 code for SMTP messages.
5420 72. When Exim is about to re-exec itself, it ensures that the file descriptors
5421 0, 1, and 2 exist, because some OS complain for execs without them (see
5422 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
5423 descriptors. However, the code omitted to check that the open succeeded,
5424 causing mysterious errors if for some reason the permissions on /dev/null
5425 got screwed. Now Exim writes a message to the main and panic logs, and
5426 bombs out if it can't open /dev/null.
5428 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
5429 interact so that it is all more flexible. It is supposed to remain
5430 backwards compatible. Also added extra_local_interfaces.
5432 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
5433 to bomb out with an assertion failure - to the client this appears as a
5434 connection drop. This problem occurs in the part of the code that was taken
5435 from the Samba project. Fortunately, the assertion is in a very simple
5436 function, so I have fixed this by reproducing the function inline in the
5437 one place where it is called, and arranging for authentication to fail
5438 instead of killing the process with assert().
5440 75. The SPA client code was not working when the server requested OEM rather
5441 than Unicode encoding.
5443 76. Added code to make require_files with a specific uid setting more usable in
5444 the case where statting the file as root fails - usually a non-root-mounted
5445 NFS file system. When this happens and the failure is EACCES, Exim now
5446 forks a subprocess and does the per-uid checking as the relevant uid.
5448 77. Added process_log_path.
5450 78. If log_file_path was not explicitly set, a setting of check_log_space or
5451 check_log_inodes was ignored.
5453 79. If a space check for the spool or log partitions fails, the incident is now
5454 logged. Of course, in the latter case the data may get lost...
5456 80. Added the %p formatting code to string_format() so that it can be used to
5457 print addresses in debug_print(). Adjusted all the address printing in the
5458 debugging in store.c to use %p rather than %d.
5460 81. There was a concern that a line of code in smtp_in.c could overflow a
5461 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
5462 initially expressed, the concern was not well-founded, because trailing
5463 spaces are removed early. However, if the trailing spaces were followed by
5464 a NULL, they did not get removed, so the overflow was possible. Two fixes
5467 (a) I re-wrote the offending code in a cleaner fashion.
5468 (b) If an incoming SMTP command contains a NULL character, it is rejected
5471 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
5472 runs initgroups(), so that if the Exim user is in any additional groups,
5473 they will be used during message reception.
5479 The change log for 4.20 and earlier releases has been archived.
git://git.exim.org / exim.git / blob